Method for purifying heteromultimeric antibodies

Abstract

The present invention relates to a method for purifying a heteropolymer Fc domain-containing polypeptide, comprising the following steps, in the order shown: preparing a sample comprising the heteropolymer Fc domain-containing polypeptide and one or more mispairing variants thereof; contacting the sample with a protein A chromatography matrix to bind the heteropolymer Fc domain-containing polypeptide to the protein A chromatography matrix; contacting the protein A chromatography matrix with a washing solution containing octanoate; contacting the protein A chromatography matrix with an elution solution; and recovering an eluate comprising the heteropolymer Fc domain-containing polypeptide. The present invention further relates to the use of buffers in the method, a kit for carrying out the method, and an eluate obtained by the method.

Description

[Technical Field]

[0001] The present invention relates to a method for purifying heteropolymerized Fc domain-containing polypeptides from their mispaired variants. [Background technology]

[0002] Multispecific antibodies, such as bispecific antibodies, are highly promising candidates for cancer treatment, and several different formats of bispecific antibodies have been developed. Many of these are heteromultimeric protein complexes produced by the co-expression of different polypeptide chains. With the development of new antibody formats, a new class of impurities called product-related impurities has begun to appear in considerable quantities. These product-related impurities are mispaired variants of heteromultimeric antibodies. In some cases, correctly paired antibodies constitute only a small fraction of all variants. Because correctly paired and mispaired antibodies often have very similar biochemical properties, it is difficult to separate them using conventional antibody purification methods. For example, in the case of heteromultimeric antibodies containing an Fc domain, many mispaired variants also contain an Fc domain, and therefore cannot be easily removed by conventional chromatography processes that involve a capture step requiring the conjugation of the Fc domain using protein A, protein G, or their derivatives.

[0003] In addition, for any antibody production, a key objective is to remove process-related impurities generated during production, such as host cell proteins or nucleic acids. [Overview of the project] [Problems that the invention aims to solve]

[0004] Therefore, there is a need for a purification method that can separate multispecific antibodies from both process-related impurities and product-related impurities. A solution to this problem is provided by the subject matter of the independent claim. [Means for solving the problem]

[0005] In a first aspect, the present invention is a method for purifying a heteromultimeric Fc domain-containing polypeptide, comprising the following steps, namely a) providing a sample comprising a heteromultimeric Fc domain-containing polypeptide and one or more of its mispaired variants; b) contacting a protein A chromatography matrix with the sample to bind the heteromultimeric Fc domain-containing polypeptide to the protein A chromatography matrix; c) contacting the protein A chromatography matrix with a washing solution containing octanoate; d) contacting the protein A chromatography matrix with an elution solution; e) recovering an eluate containing the heteromultimeric Fc domain-containing polypeptide in the order shown.

[0006] In a second aspect, the present invention relates to the use of a washing solution containing octanoate in a method for purifying a heteromultimeric Fc domain-containing polypeptide according to the first aspect.

[0007] In a third aspect, the present invention relates to a kit comprising a protein A chromatography matrix and a washing solution containing octanoate.

[0008] In a fourth aspect, the present invention relates to an eluate obtained by a method for purifying a heteromultimeric Fc domain-containing polypeptide according to the first aspect, the eluate comprising at least 90% purified heteromultimeric Fc domain-containing polypeptide and less than 0.5% mispaired variants.

[0009] Before describing the present invention in detail below, it should be noted that the specific methods, protocols, and reagents described herein are subject to change and are not limited thereto. Similarly, the terms used herein are for the purpose of describing specific embodiments only and do not limit the scope of the invention, which is limited solely by the appended claims. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art. Preferably, terms used herein are defined as they are in "A multilingual glossary of biotechnological terms: (IUPAC Recommendations)," Leuenberger, HGW, Nagel, B., and Kolbl, H. eds. (1995), Helvetica Chimica Acta, CH-4010 Basel, Switzerland.

[0010] Throughout this specification, multiple documents are referenced. Each of the documents referenced herein (including all patents, patent applications, scientific publications, manufacturer specifications, instructions, etc.), whether listed above or below, is incorporated herein by reference in its entirety. Nothing in this specification should be construed as an acknowledgment that the present invention is not granted prior rights by prior art.

[0011] To carry out the present invention, unless otherwise indicated, conventional methods of chemistry, biochemistry, and recombinant DNA technology described in the literature of the art will be used (e.g., Molecular Cloning: A Laboratory Manual, 2 nd See Edition, J. Sambrook et al. eds., Cold Spring Harbor Laboratory Press, Cold Spring Harbor 1989.

[0012] Throughout this specification and the subsequent claims, unless the context requires otherwise, the word “comprise,” and variations such as “comprises” and “comprising,” should be understood to mean that they include the integers or processes or groups of integers or processes described, but not any other integers or processes or groups of integers or processes. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the content explicitly indicates otherwise.

[0013] The term "antibody," as used herein, includes natural antibodies and modified antibodies. The term "modified antibody" includes functional antibody fragments, single-chain antibodies, single-domain antibodies, monospecific or multispecific (e.g., bispecific, trispecific, quadruple specific, quintuple specific, hexaspecific) antibodies, monovalent or polyvalent (e.g., bivalent, trivalent, tetravalent) antibodies, and antibodies having two or more functions (multifunctional antibodies), for example, antibodies comprising one or more variable domains and additional domains such as an Fc domain capable of binding to the Fc receptor FcγRIII (also known as CD16).

[0014] Natural antibodies are Y-shaped molecules containing four polypeptide chains, namely two heavy chains and two light chains. The heavy chains are linked to each other by disulfide bonds, and each heavy chain is linked to a light chain by a disulfide bond. There are five major heavy chain classes (or isotypes) that determine the functional activity of antibody molecules: IgM, IgD, IgG, IgA, and IgE. When referring to IgG, it generally includes IgG1, IgG2, IgG3, and IgG4 unless otherwise defined. Each light chain consists of two domains, and the N-terminal domain is a variable domain (or region) or V L It is known as a domain (or region), and the C-terminal domain is stationary (or C L) It is known as a domain (constant kappa (Cκ) domain or constant lambda (Cλ) domain). Each heavy chain contains four or five domains depending on the antibody isotype, namely, a variable domain (V H ), followed by the first constant domain (CH1), a hinge region, then the constant domains CH2 and CH3, and in some isotypes, CH4. In the assembled antibody, the V L domain and the V H domain associate to form an antigen-binding domain. Also, the C L domain and the CH1 domain associate to hold one heavy chain associated with one light chain. The two heavy-light chain heterodimers associate by the interaction of the CH2 and CH3 domains and the interaction between the hinge regions of the two heavy chains. The constant domains confer important biological properties such as antibody chain association, secretion, transplacental transfer, complement binding, and binding to the F c receptor (F c R).

[0015] Bispecific antibodies occur in multiple formats (Brinkmann and Kontermann, Mabs 2017, Vol.9, No.2, 182-212). Formats containing only variable domains have the advantage of a very low molecular weight and good tumor penetration, which is important for oncological applications. However, the disadvantage is that they lack the constant domain that mediates binding to FcRn, resulting in a short plasma half-life.

[0016] In the following description, amino acid numbers are used, where applicable, with respect to antibodies or antibody domains, but these do not refer to sequence numbers. These numbers refer to amino acid positions in the antibody, according to the version disclosed on August 26, 2016, in the UniProtKB database (www.uniprot.org / uniprot). Unless otherwise specified, these numbers correspond to positions in human IgG, and in particular in human IgG1. The UniProtKB sequences of antibody domains referred to herein (in the version disclosed on August 26, 2016), including those of human IgG1, are incorporated by reference as specific embodiments of the domains described herein and their variants as defined below.

[0017] In this specification, the term “immunoglobulin (Ig) domain” is used to refer to a protein domain consisting of a two-layer sandwich of 7-9 reverse-equilibrium β-strands arranged in a Greek key configuration within two β-sheets. The Ig domain is perhaps the most frequently used “component” in natural proteins. Proteins containing Ig domains are included in the immunoglobulin superfamily. Not only antibodies, but also cell adhesion molecules, T cell receptors, Fcγ receptors, and many more belong to this protein family. The immunoglobulin fold has been well described in the review by Bork et al. ("The immunoglobulin fold. Structural classification, sequence patterns and common core," September 1994; J.Mol.Biol.242(4):309-20).

[0018] "Specific binding" means that the binding is selective to the antigen and can be distinguished from undesirable or nonspecific interactions. In certain embodiments, an antibody is said to bind specifically to an antigen if it preferentially recognizes its target antigen in a complex mixture of proteins and / or macromolecules. The ability of an antigen-binding molecule to bind to a particular antigen can be measured by either enzyme-linked immunosorbent assay (ELISA) or other techniques well known to those skilled in the art, such as surface plasmon resonance (SPR) techniques. According to these methods, the identified ligand binds to its target molecule and does not bind to other molecules present in significant amounts. Generally, an antibody that "specifically binds" to a target molecule will bind to that target molecule in approximately 10°C. -5 (For example, 10 -6 , 10 -7 , 10 -8 , 10 -9 , 10 -10 , 10 -11 , or 10 -12 It has an equilibrium dissociation constant smaller than M.

[0019] The term "antigen" is used to refer to a substance comprising at least one epitope, preferably a B-cell response or a T-cell response, or an epitope that elicits both a B-cell response and a T-cell response, preferably an immunogenic polypeptide.

[0020] An "epitope," also known as an antigenic determinant, is a substance recognized by the immune system, such as the portion of an immunogenic polypeptide. Preferably, this recognition is mediated by the binding of an antibody, B cell, or T cell to the epitope. In this context, the term "binding" preferably refers to specific binding. Epitopes typically consist of a group of chemically active surfaces of molecules, such as amino acids or sugar side chains, and usually have specific three-dimensional structural properties and specific charge properties. The term "epitope" includes both conformal and non-conformal epitopes. Conformal and non-conformal epitopes are distinguished in that binding to the former is lost in the presence of a denaturing solvent, while binding to the latter is not.

[0021] The immunogenic polypeptide according to the present invention is a polypeptide that is specifically bound by the antigen-binding domain of an antibody or a T cell receptor. In the case of a T cell receptor, the immunogenic polypeptide binds to the MHC protein to form a complex. In some embodiments, the immunogenic polypeptide is a tumor antigen, i.e., a polypeptide specifically expressed by tumor cells, preferably a polypeptide expressed on or presented on the surface of tumor cells. In some embodiments, the immunogenic polypeptide is an antigen derived from a cell receptor. In some embodiments, the immunogenic polypeptide is an antigen derived from a cytokine. In some embodiments, the immunogenic polypeptide is an antigen derived from a pathogen selected from the group consisting of viruses, bacteria, and protists. In some embodiments, the immunogenic polypeptide is an immune cell antigen, preferably an immune effector cell antigen, i.e., a polypeptide expressed by immune cells, preferably a polypeptide expressed on the surface of immune cells.

[0022] When used herein, “variable domain” or “immunoglobulin variable domain” refers to each of a pair of domains that form an antigen-binding domain. An immunoglobulin variable domain may be a variable domain of an antibody or a T cell receptor (TCR). The variable domains of antibodies and TCRs have the same general structure. Each variable domain contains four framework (FR) regions whose sequences are widely conserved, and these are linked by three “hypervariable regions” (or complementarity-determining regions, CDRs). The framework regions take the form of a β-sheet three-dimensional structure, and the CDRs can form loops linking the β-sheet structures and are held within their three-dimensional structure by the framework regions. In preferred embodiments, “variable domain” as used herein refers to an “antibody variable domain.”

[0023] The term “Immunoglobulin Monovariable Domain” (ISVD) is used interchangeably with “Single Variable Domain” and refers to a single monomeric immunoglobulin variable domain that can selectively bind to a specific antigen on its own, i.e., without a second immunoglobulin variable domain. For further explanation of ISVDs, see International Publication No. 2021 / 110816. In light of this definition above, the antigen-binding domains of conventional quadruple-chain antibodies (such as IgG, IgM, IgA, IgD, or IgE, which are known in the art), or of Fab fragments, F(ab')2 fragments, disulfide-bonded Fv fragments, or scFv fragments, or of diabodies (known in the art) derived from such conventional quadruple-chain antibodies, are not typically considered ISVDs. This is because, in these cases, the binding of the antigen to each epitope is usually not caused by a single immunoglobulin domain, but rather by a pair of associated immunoglobulin domains, such as the light chain variable domain and the heavy chain variable domain, i.e., the V of immunoglobulin domains that jointly bind to each antigen's epitope. H -V L This is because it is produced by a pair. In contrast, ISVD can specifically bind to the epitope of the antigen without pairing with an additional immunoglobulin variable domain. The binding site of ISVD is a single V H Domain, Single V H H domain, or single V L It is formed by a domain. H The H domain refers to the variable domain contained in heavy chain antibodies found in camelids. Therefore, ISVD refers to the light chain variable domain sequence (e.g., V L (array) or a suitable fragment thereof, or a heavy chain variable domain sequence (e.g., V H Array or V HIt may be an H sequence or a suitable fragment thereof, provided that the ISVD can form a single antigen-binding unit / domain, i.e., a substantial functional antigen-binding unit / domain from the ISVD, and that the single antigen-binding domain does not need to interact with another variable domain to form a functional antigen-binding unit. The ISVD or single variable domain may be, for example, V H , V H It can be made into a heavy chain ISVD such as H, and camelid V H or humanized V H It includes H. In one embodiment, it is V H H is, camelization V H or humanized V H Contains H. Heavy chain ISVDs can be derived from or from conventional quadruple-chain antibodies. For example, an ISVD may be a single-domain antibody (or an amino acid sequence suitable for use as a single-domain antibody), a "dAb" (or an amino acid sequence suitable for use as a dAb), or a Nanobody® molecule (as defined in the prior art, particularly in International Publication No. 2021 / 110816, and V H It may be a single variable domain (containing but not limited to H), another single variable domain, or any fragment of any one of these. In particular, ISVD is a Nanobody® molecule (humanized V H H or Camelized V H V including H It could be H (or a suitable fragment thereof) or a suitable fragment thereof.

[0024] When used in connection with the present invention, the term “Fc domain” encompasses natural Fc and Fc variants, including monomeric, dimeric, and polymeric Fc domains with or without a hinge region. While the boundaries of the Fc region of immunoglobulin heavy chains can vary, the human IgG heavy chain Fc domain is typically defined as extending from the C226 amino acid residue or from the P230 position to its carboxyl terminus. An Fc domain comprises at least one, preferably two, polypeptide chains, each containing a CH2 domain or a functional portion thereof and a CH3 domain or a functional portion thereof (collectively referred to herein as the “CH2-CH3 region”). The two polypeptide chains may be linked by covalent bonds (i.e., disulfide bonds) and / or non-covalent associations. The number of intermolecular disulfide bonds between monomeric subunits in a natural Fc domain ranges from 1 to 4, depending on the class (e.g., IgG, IgA, and IgE) or subclass (e.g., IgG1, IgG2, IgG3, IgA1, and IgGA2). An example of a natural Fc domain is the dimer resulting from the papain digestion of IgG, in which case the two polypeptide chains associate via disulfide bonds.

[0025] The term "CH2-CH3 region" refers to a region containing at least a functional portion of the CH2 domain and at least a functional portion of the CH3 domain. An example of a functional portion of the CH3 domain is a CH3 domain in which the C-terminal amino acid residues, e.g., 5, 4, 3, 2, or 1 C-terminal amino acid residues, have been removed or replaced with a linker. This modification does not impair the binding of the CH3 domain or the dimeric Fc domain containing this CH3 domain to protein A or protein G.

[0026] The two CH2-CH3 regions combine to form a dimer called the "Fc domain".

[0027] "Protein A" is originally a 42 kDa protein isolated from Staphylococcus aureus, which can specifically bind to the Fc domains of many immunoglobulin molecules, including the Fc domains of human IgG1, IgG2, and IgG4, via its five homologous immunoglobulin-binding domains, each containing three helical bundles. In the context of this invention, Protein A refers to a ligand, which may be either natural or modified (i.e., mutated), used in capture chromatography resins to enhance their chromatographic performance (e.g., increased binding efficiency, increased selectivity, increased caustic stability). Therefore, when used in connection with this invention, the term "Protein A" also refers to derivatives from Protein A that retain at least 10% of the ability of natural Protein A to bind to dimeric immunoglobulin Fc domains, such as proteins containing multiple copies of a single Protein A immunoglobulin-binding domain, or proteins containing immunoglobulin-binding domains of both Protein A and Protein G, or proteins containing one or more modified Protein A domains. "Protein G" is a 58-65 kDa protein originally isolated from streptococcal bacteria that can specifically bind to the Fc domains of many immunoglobulin molecules, including the Fc domains of human IgG1, IgG2, and IgG4. When used in connection with the present invention, the term "Protein G" also refers to derivatives from Protein G that retain at least 10% of the ability of native Protein G to bind to the Fc domains of dimeric immunoglobulins.

[0028] The CH2-CH3 region, and in particular the dimeric Fc domain formed by two CH2-CH3 regions, exhibit specific binding to protein A or protein G, particularly protein A, as defined in relation to the present invention, preferably having an affinity of at least 0.01, at least 0.05, at least 0.1, at least 0.2, or at least 0.5 for the CH2-CH3 region or Fc domain of IgG1, IgG2, or IgG4 to natural protein A or G, particularly protein A, and preferably having an affinity of 100× or less or 10× or less for the CH2-CH3 region or Fc domain of IgG1, IgG2, or IgG4 to natural protein A or G, particularly protein A. The CH2-CH3 region, and in particular the Fc domain, exhibit specific binding to commercially available protein A or protein G chromatography resins (including, but not limited to, MabSelect sure protein A resin (Cytiva)), as defined in relation to the present invention.

[0029] Preferably, the CH2 and CH3 domains of the CH2-CH3 region are derived from IgG1, IgG2, or IgG4. If the CH2 and / or CH3 domains contain deletions, substitutions, and / or insertions or other modifications that make it impossible for the CH2-CH3 region to specifically bind to protein A or protein G, particularly protein A, then they are not considered functional parts of the CH2 or CH3 domain and do not form a CH2-CH3 region as defined in the present invention. Natural human IgD, IgE, and IgG3 have been reported to show no binding or only weak binding to protein A. Therefore, the natural CH2-CH3 region or natural Fc domain of IgD, IgE, or IgG3 are neither functional CH2-CH3 regions nor functional Fc domains in relation to the first, second, third, or fourth aspects of the present invention. Natural human IgA, IgD, IgE, and IgM show no binding or only weak binding to protein G. Therefore, in relation to the alternative embodiments of the present invention relating to purification using a protein G chromatography matrix, the natural CH2-CH3 regions or natural Fc domains of IgA, IgD, IgE, and IgM are neither functional CH2-CH3 regions nor functional Fc domains.

[0030] Typically, but not always, the CH2-CH3 region can also bind to Fc receptors (e.g., FcyR or FcRn) and / or be involved in complement activation.

[0031] In relation to various antibody formats, polypeptide chains containing a CH2-CH3 region and at least one variable domain are often called "heavy chains," while polypeptide chains containing at least one variable domain but not a CH2-CH3 region are often called "light chains."

[0032] The term “Fc domain” as used herein also includes “Fc variant domain.” As used herein, an Fc variant domain refers to a domain modified from a natural Fc domain. In particular, an Fc variant domain is an Fc domain that includes modifications that result in novel or enhanced advantageous properties compared to a natural Fc domain. The term Fc variant domain includes Fc domains containing “knob-into-hole” mutations (see below). The term Fc variant domain may also refer to a domain that has been humanized from a non-human natural Fc domain. Furthermore, an Fc variant domain may also refer to an Fc domain from which regions that result in structural features or biological activity not required for the antigen-binding protein of the present invention have been removed. Therefore, the term Fc variant domain refers to an Fc domain modified by one or more amino acid residues that affects or is involved in (1) disulfide bond formation, (2) incompatibility with selected host cells, (3) N-terminal heterogeneity during expression in selected host cells, (4) glycosylation, (5) interaction with complement, (6) binding to Fc receptors other than salvage receptors, or (7) antibody-dependent cell-mediated cytotoxicity (ADCC). In some embodiments, the Fc variant domain is modified, for example, by introducing mutations as defined below herein.

[0033] If the Fc domain is a dimer, the two polypeptide chains preferably originate from the same antibody isotype or subclass of isotype.

[0034] In some embodiments, the Fc domain contains a “knob-into-hole” mutation. The “knob-into-hole” technique refers to a mutation at the CH2-CH3 interface of a dimeric Fc domain to create a “knob” in one CH3 domain and a “hole” in the other CH3 domain, thereby promoting heteromultimerization. As a non-limiting example, specific “knob-into-hole” mutations are T366S, T366Y, L368A, Y407V, and Y407T. These knob-into-hole mutations may be further stabilized by the introduction of additional cysteine ​​amino acid substitutions, Y349C and S354C. In this specification, “knob-into-hole Fc domain” refers to an Fc domain containing a “knob-into-hole” mutation. In this specification, “knob-into-hole CH2-CH3 region” refers to a CH2-CH3 region in which the CH3 domain contains a “knob-into-hole” mutation.

[0035] In some embodiments, the Fc domain includes an "RF mutation." In this specification, the term "RF mutation" refers to the H435R and Y436F (RF mutation) mutations in one of a pair of CH3 domains.

[0036] In some embodiments, the Fc domain includes further amino acid substitutions, such as charge pair substitutions, to improve heterodimerization of the polypeptide chain forming the Fc domain.

[0037] In some embodiments, the Fc domain includes one or more modifications that inhibit the binding of the Fc gamma receptor (FcyR). Such modifications may be present in one or both, preferably both, of the polypeptide chains forming the Fc domain. These modifications may include, in non-limiting examples, L234A and L235A.

[0038] In some embodiments, the Fc domain contains "N297Q," "N297G," or "N297A" mutations to remove an N-glycosylation site within the Fc portion. Such mutations inhibit interaction with the Fc-gamma receptor. These mutations may be present in one or both, preferably both, of the polypeptide chains forming the Fc domain.

[0039] A "hinge," "hinge region," or "hinge domain" typically refers to a flexible portion of the heavy chain located between the CH1 and CH2 domains. It is approximately 25 amino acids long and is divided into an "upper hinge," a "central hinge" or "core hinge," and a "lower hinge."

[0040] In some embodiments, the Fc domain comprises at least two additional cysteine ​​residues, for example, one in each polypeptide chain forming the dimeric Fc domain to increase heterodimerization, or both in the same polypeptide chain to form intradomain disulfide bonds, or further.

[0041] In some embodiments, the Fc domain-containing polypeptide is fused to the ISVD described above, or to a cytokine that forms an immune cytokine. In particular, the ISVD or cytokine is fused to at least one of the CH2-CH3 regions of the polypeptide chain of the Fc domain-containing polypeptide. In some embodiments, the Fc domain-containing polypeptide comprises two polypeptide chains, each containing a CH2-CH3 region, and the ISVD or cytokine is fused to each of the CH2-CH3 regions.

[0042] In the context of this invention, "fused" or "combined" means that two polypeptides are covalently bonded, particularly via a peptide bond. Preferably, the two fused polypeptides are expressed as a single polypeptide chain.

[0043] In the context of this invention, the term "cytokine" refers to small proteins (approximately 5–25 kDa) that are involved as immunomodulators in autocrine, paracrine, and endocrine signaling. Cytokines cannot cross the lipid bilayer of cells to enter the cytoplasm and act through cell surface receptors. As used herein, the term "cytokine" includes chemokines, interferons, interleukins, lymphokines, and tumor necrosis factors.

[0044] In the context of this invention, the term "immune cytokine" refers to a molecule comprising a cytokine and a fusion protein containing at least an immunoglobulin variable domain.

[0045] Unless otherwise specified, the binding proteins of this disclosure are positioned, in accordance with standard usage and convention, with the amino-terminal direction ("N-terminal end" or "N-terminus") on the left and the carboxyl-terminal direction ("C-terminal end" or "C-terminus") on the right.

[0046] The determination of the percentage of identity between two sequences is performed using the mathematical algorithm of Karlin and Altschul, Proc. Natl. Acad. Sci. USA 90, 5873-5877, 1993. Such algorithms are incorporated into the BLASTN and BLASTP programs of Altschul et al. (1990) J. Mol. Biol. 215, 403-410. To obtain gap alignment for comparison purposes, Gap BLAST is used as described in Altschul et al. (1997) Nucleic Acids Res. 25, 3389-3402. When using the BLAST program and Gap BLAST program, the default parameters of each program are used. Alternatively, variants can be defined as having up to 20, up to 15, up to 10, up to 5, up to 4, up to 3, up to 2, or up to 1 amino acid substitution, especially conservative amino acid substitutions. Conservative substitutions are well known in the art (see, for example, Creighton (1984) Proteins. WH Freeman and Company). Families of amino acid residues with similar side chains are known in the art and include amino acids having basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), polar uncharged side chains (e.g., glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (e.g., alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine), and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan).

[0047] When used herein, the term "chromatographic matrix" refers to a solid-phase material that can selectively bond to one or more components of the applied load fluid, as is well known in the art.

[0048] The phrase "contacting the chromatography matrix with the solution" refers to applying the solution to the chromatography matrix and represents the step in the claimed purification method in which the solution is contacted with the solid phase. This means that the solution is added to the chromatography device on which the solid phase is placed. In a preferred embodiment, the solid phase is a stationary solid phase. Alternatively, in another embodiment, the solid-phase chromatography matrix may be added directly to the solution and, after binding, recovered by centrifugation as an example. A solution containing one or more substances, particularly heteropolymer Fc domain-containing polypeptides and product-related and process-related impurities, passes through the solid phase, allowing for interactions between the solid phase and the substances. Depending on conditions, such as pH, conductivity, salt concentration, temperature, and / or flow rate, some of the substances in the solution bind to the solid phase and are thus removed from the solution. Other substances remain in the solution. Substances remaining in the solution can be found during flow-through.

[0049] "Flow-through" refers to the solution obtained after passing through a chromatographic device, regardless of its origin. A washing step may optionally be applied to flush the column. Then, the application of an elution buffer can be used to induce the elution of one or more substances. The substances can be recovered from the solution by methods well known to those skilled in the art, such as precipitation, salting out, ultrafiltration, diafiltration, lyophilization, affinity chromatography, or solvent volume reduction, and the substance of interest can be obtained in a purified or even substantially homogeneous form.

[0050] The term “bind-and-elute mode” refers to a method for performing a chromatographic purification method. In this specification, a solution containing the protein to be purified and impurities is applied to a stationary phase, particularly a solid phase, so that the protein to be purified interacts with and is retained by the stationary phase. Some impurities are removed along with the flow-through. The protein to be purified is then recovered from the stationary phase in a second step by applying an elution solution (typically a buffer solution), typically stepwise or linearly (or a combination thereof), so that the protein to be purified is separated from any impurities that may be bound to the stationary phase and eluted.

[0051] The term "flow-through mode" refers to an alternative method for performing chromatographic purification. In this specification, a solution containing the protein to be purified and impurities is applied to a stationary phase, particularly a solid phase, so that the impurities, rather than the protein to be purified, interact with and are retained by the stationary phase. The protein of interest is then eluted along with the flow-through.

[0052] As used herein, “buffer solution” refers to a buffer solution that resists pH changes due to the action of its acid-base conjugate components. “Loading buffer” is used to load a mixture of proteins to be purified into a chromatography matrix, and “wash buffer” is used to wash the chromatography matrix to remove unbound material. “Elute buffer” is used to elute the proteins to be purified from the column.

[0053] The term "multimer" (multi-"many" + -mer "part"), as used herein, refers to a molecule consisting of subunits ("monomers") linked by covalent or non-covalent bonds, including, for example, "dimers" (consisting of two monomers) and "tetramers" (consisting of four monomers).

[0054] As used herein, the term "homodimer" refers to a dimer consisting of two identical subunits, and as used herein, the term "heterodimer" refers to a dimer consisting of two different subunits.

[0055] As used herein, the term "heteropolymeric polypeptide" refers to a protein complex comprising one, preferably at least three, polypeptide chains. The polypeptide chains in a heteropolymeric polypeptide are not identical. In other words, each polypeptide chain in a heteropolymeric polypeptide is distinct from at least one other polypeptide chain, preferably all other polypeptide chains, in the heteropolymeric polypeptide.

[0056] The elements of the present invention are described below. These elements are listed in relation to specific embodiments, but it should be understood that they can be combined in any way and in any number to create additional embodiments. The various examples and preferred embodiments described should not be construed as limiting the invention to only the explicitly described embodiments. This description should be understood as supporting and encompassing embodiments that combine the explicitly described embodiments with any number of disclosed elements and / or preferred elements. Furthermore, any permutations and combinations of all elements described in this application should be considered disclosed by the description of this application unless the context indicates otherwise.

[0057] In a first aspect, the present invention relates to a method for purifying a heteropolymer Fc domain-containing polypeptide, comprising the following steps: a) A step of preparing a sample comprising a heteropolymerized Fc domain-containing polypeptide and one or more mispaired variants thereof, b) The step of contacting the protein A chromatography matrix with the sample to bind the heteropolymerized Fc domain-containing polypeptide to the protein A chromatography matrix, c) The step of contacting the protein A chromatography matrix with a washing solution containing octanoate, d) The step of bringing the protein A chromatography matrix into contact with the elution solution, e) A step of recovering the elute containing the Fc domain-containing polypeptide of the heteromultimer. Regarding a method of including them in the order shown.

[0058] In an alternative embodiment, a method for purifying a heteropolymer Fc domain-containing polypeptide according to the first embodiment is provided, wherein steps b) to d) include the use of a protein A or protein G chromatography matrix, particularly the use of a protein G chromatography matrix. All embodiments specified for the first embodiment are also assumed for this alternative embodiment, where each reference to protein A shall refer to protein A or protein G.

[0059] The expression of heteromultimeric Fc domain-containing polypeptides requires the co-expression of different polypeptide chains contained within the heteromultimeric polypeptide. This co-expression causes not only the desired heteromultimeric polypeptide but also its mispairing variants to be secreted into the cell culture supernatant. These mispairing variants are also referred to herein as "product-related impurities." The presence of product-related impurities can affect the activity and safety of the product. If the heteromultimeric Fc domain-containing polypeptide is a bispecific antibody, for example, a mispairing variant that exhibits binding activity to only one target antigen may block the binding of a fully functional bispecific antibody, thereby antagonizing the desired activity of the bispecific molecule. At the very least, mispairing variants, if not isolated, may reduce the efficacy of the final product. In addition, many mispairing variants have exposed regions that normally facilitate peptide-peptide interactions, and therefore may exhibit immunogenicity and a tendency towards aggregation. Several techniques have been developed to enforce correct pairing of polypeptide chains, including the "knob-into-hole" technique (Ridgway JB et al., Protein Eng 1996;9:617-621) and the "CrossMab" technique (Schaefer, W. et al, PNAS, 108(2011)11187-1191). However, none of these techniques can completely prevent the formation of mispaired variants. In addition, in existing formats that do not use "knob-into-hole," "CrossMab," or other techniques to prevent mispairing, it becomes even more important to isolate mispaired variants from the target protein. Therefore, there is a general need to isolate mispaired variants from correctly paired heteromultimeric polypeptides.

[0060] In addition to “product-related impurities,” the sample may include impurities resulting from the recombinant production of heteropolymeric Fc domain-containing polypeptides in host cells, such as nucleic acids, components derived from cell culture media, endotoxins, viruses, host cell lipids or host cell proteins, such as phospholipases, clatherins, serine proteases, elongation factors, and / or any combination thereof. Such impurities are referred to herein as “process-related impurities.” The claimed method has the additional effect of separating these process-related impurities from correctly paired heteropolymeric polypeptides.

[0061] sample The sample is derived from cultured cells recombinantly expressing a heteromultimeric Fc domain-containing polypeptide. In some embodiments, the sample is untreated cell culture supernatant derived from cultured cells recombinantly expressing a heteromultimeric Fc domain-containing polypeptide. Such a sample is also referred to herein as “bulk harvest.” In non-limiting examples, the sample may be a conditional cell culture supernatant, a clarified conditional cell culture supernatant, or a clarified homogenized / lysed cell culture. As used herein, the terms “clarified” and “clarified” refer to the removal of particulate matter from a solution and include, but are not limited to, filtration, preferably using a 0.2 polyethersulfone (PES) filter, sterilization, and / or centrifugation. Thus, a sample referred to herein as “clarified bulk harvest” is a liquid material containing heteromultimeric Fc domain-containing polypeptide, product-related impurities, and process-related impurities, extracted from a cell culture, e.g., a fermentation bioreactor, after centrifugation to remove larger solid particles from the material, and / or subsequent filtration to remove finer solid particles and impurities. The cultured cells can be any cells suitable for expressing recombinant nucleic acid sequences, such as prokaryotic cells (E. coli cells, Aspergillus niger cells, etc.), eukaryotic cells (yeast cells, plant cells, insect cells (e.g., SI cells), etc.), and / or mammalian cells (mouse, rat, hamster, rabbit, human, non-human primate, etc.) (e.g., CHO cells, HEK cells, BHK cells, NS0 cells, SP2 / 0 cells, YO myeloma cells, P3X63 mouse myeloma cells, PER cells, PER.C6 cells, or hybridoma cells). In preferred embodiments, the cultured cells are CHO cells or HEK293 cells, preferably CHO cells.

[0062] Fc domain-containing polypeptide In some embodiments, the heteropolymer Fc domain-containing polypeptide is a secreted polypeptide. In some embodiments of the first aspect of the present invention, the heteropolymer Fc domain-containing polypeptide is the antibody as defined above. In some embodiments, the antibody is a human antibody, a humanized antibody, or a chimeric antibody. In some embodiments, the antibody is a monoclonal antibody. In preferred embodiments, the antibody is a multispecific antibody, preferably a bispecific, triplicate, quadruplespecific, quintuple, or hexaspecific antibody, more preferably a bispecific or triplicate antibody.

[0063] The heteropolymer Fc domain-containing polypeptide may contain one, two, three, four, five, six, seven, eight, nine, ten, eleven, or twelve immunoglobulin variable domains that form one, two, three, four, five, or six antigen-binding domains. The antigen-binding domain may contain or consist of the ISVD as defined above. In a preferred embodiment, the antigen-binding domain contains or consists of two variable domains, in particular an immunoglobulin light chain variable domain and an immunoglobulin heavy chain variable domain, each containing three complementarity-determining regions (CDR-1 to CDR-3). Preferably, the heteropolymer Fc domain-containing polypeptide contains four or six immunoglobulin variable domains that form two or three antigen-binding domains.

[0064] Heteromultimeric Fc domain-containing polypeptides may include heavy and / or light chains derived from immunized mouse antibodies, chimeric antibodies, humanized antibodies, or human antibodies, as well as combinations of variable and / or constant domains derived from immunized mouse antibodies, chimeric antibodies, humanized antibodies, or human antibodies and their fragments. Heteromultimeric Fc domain-containing polypeptides may also include variable domains derived from TCRs.

[0065] The Fc-domains contained in the heteropolymerized Fc-domain-containing polypeptide are preferably dimeric Fc-domains formed by two polypeptide chains, each containing a CH2-CH3 region. The dimeric Fc-domains may be homodimers or heterodimers. The bispecific antigen-binding molecule of the present invention contains only one Fc-domain. In one embodiment, the Fc-domain is an IgG Fc-domain, preferably an IgG1Fc-domain or an IgG4Fc-domain.

[0066] In preferred embodiments, the Fc domain-containing polypeptide comprises, preferably, first and second polypeptide chains, each containing a CH2-CH3 region, and a third polypeptide chain, each not containing a CH2-CH3 region. The heteropolymer Fc domain-containing polypeptide contains no more than one copy of the polypeptide chain.

[0067] In another embodiment, the Fc domain-containing polypeptide contains one ISVD fused to the CH2-CH3 region of the polypeptide chain of the Fc domain-containing polypeptide. In another embodiment, the Fc domain-containing polypeptide contains one or more ISVDs fused to the CH2-CH3 region of the polypeptide chain of the Fc domain-containing polypeptide. In another embodiment, the Fc domain-containing polypeptide contains two CH2-CH3 regions, with an ISVD fused to each of the CH2-CH3 regions. In another embodiment, the Fc domain-containing polypeptide contains one cytokine fused to the CH2-CH3 region of the polypeptide chain of the Fc domain-containing polypeptide. In another embodiment, the Fc domain-containing polypeptide contains one or more cytokines fused to the CH2-CH3 region of the polypeptide chain of the Fc domain-containing polypeptide. In another embodiment, the Fc domain-containing polypeptide contains two CH2-CH3 regions, with a cytokine fused to each of the CH2-CH3 regions.

[0068] In some embodiments, components of a heteropolymer Fc domain-containing polypeptide (e.g., CH2-CH3 region, constant domain, variable domain) can be directly linked or linked through various linkers described herein or known in the art, particularly through peptide linkers containing one or more amino acids, typically about 2 to 20 amino acids. When two domains are linked by a peptide linker having a length of 0aa, this means that the two domains are directly linked via peptide bonds between them. In this case, the term "fused" may also be used instead of "linked". The peptide linker is particularly a flexible peptide linker, i.e., it provides flexibility between the domains linked together. Such flexibility is generally increased when the amino acids are small and do not have bulky side chains that hinder rotation or bending of the amino acid chain. Therefore, preferably, the peptide linkers of the present invention have a high content of small amino acids, particularly glycine, alanine, serine, threonine, leucine, and isoleucine. Preferably, at least 20%, 30%, 40%, 50%, 60%, 70%, 80%, or 90% or more of the amino acids in the peptide linker are such small amino acids. In certain embodiments, the amino acids of the linker are selected from glycine and serine, i.e., the linker is a polyglycine, polyserine, or polyglycine / serine linker, where "poly" means glycine residues and / or serine residues in the linker in proportion to at least 50%, 60%, 70%, 80%, 90%, or even 100%. An example of a polyglycine / serine linker is, for example, STGS (SEQ ID NO: 7), or [G w S x G y ] zHerein, w is an integer from 0 to 20, and in some embodiments, an integer from 2 to 5; x is an integer from 0 to 10, and in some embodiments, an integer from 0 to 3; y is an integer from 0 to 20, and in some embodiments, an integer from 0 to 5; and z is an integer from 0 to 10, and in some embodiments, an integer from 0 to 4. In the context of this specification, the term polyglycine / serine linker may also refer to a linker consisting of only one amino acid selected from G or S.

[0069] The inventors have surprisingly found that the method according to the present invention can efficiently separate heteropolymer Fc domain-containing polypeptides from mispaired variants. Therefore, the present invention provides a method for purifying heteropolymer Fc domain-containing polypeptides from their mispaired variants. This method separates and purifies heteropolymer Fc domain-containing polypeptides from mispaired variants to a greater extent than corresponding methods that lack the step of washing the matrix with the washing solution according to the present invention. Prior art describes that certain washing buffers for protein A chromatography can be effective in removing process-related impurities, particularly host cell proteins. Such host cell proteins often interact with Fc domain-containing polypeptides bound to the protein A chromatography matrix. To remove these process-related impurities, the interaction between the host cell proteins and the Fc domain-containing polypeptides must be eliminated. This situation is quite different in the case of product-related impurities, which include the Fc domain itself. Therefore, it was a surprising discovery that by using the method of the present invention, which includes the use of a wash buffer containing octanoate, Fc domain-containing mispaired variants can be securely separated from correctly paired heteropolymer Fc domain-containing polypeptides.

[0070] In some embodiments, the sample is a clarified bulk harvest sample obtained from tissue culture, and the purity of the heteropolymer Fc domain-containing polypeptide is - After step e), at least 40% or 45% of the total protein concentration, - After step h), at least 50%, 60%, 70%, 80%, 90%, or 95% of the total protein concentration, - After step j), at least 90%, 92%, 94%, 95%, 96%, or 97% of the total protein concentration Preferably, the concentration is determined by capillary electrophoresis or capillary gel electrophoresis.

[0071] Mismatch variant Co-expression of different polypeptide chains that form Fc domain-containing heteromultimeric polypeptides leads to the assembly of mispaired variants of Fc domain-containing heteromultimeric polypeptides.

[0072] In some embodiments, the mismatch variant is a) monomers of the first, second, or third polypeptide chain, b) Homodimers or homomultimers of polypeptide chains contained in heteropolymerized Fc domain-containing polypeptides, c) A heteropolymer containing more than one copy of the first, second, or third polypeptide chain, d) Heteropolymers that do not contain the first, second, or third polypeptide chain, and e) Aggregates of mispaired variants and / or aggregates of heteropolymer Fc domain-containing polypeptides according to a)~d) It is selected from the group consisting of the following.

[0073] The embodiments described in e) refer, in particular, to heteromultimers containing paired variable domains that are specific to different antigens and do not form an antigen-binding domain that specifically binds to one antigen, for example, a variable domain specific to a tumor antigen paired with a variable domain specific to an immune effector cell. Such mispaired variants may otherwise represent correct pairing of dimeric domains, such as pairing of two Fc domains to form a dimeric Fc domain, and / or pairing of a CH1 domain with a Cκ domain or a Cλ domain.

[0074] In this specification, an aggregate is an aggregate of multiple copies of a heteromultimer Fc domain-containing polypeptide and / or its mispairing variant. For example, an aggregate may contain 2, 3, 4, 5, 10, 25, 50, or 100 copies of a heteromultimer Fc domain-containing polypeptide and / or its mispairing variant.

[0075] In some embodiments, the method results in the separation of a heteropolymer Fc domain-containing polypeptide from at least one mispairing variant selected from a) to e). In some embodiments, the method results in the separation of a heteropolymer Fc domain-containing polypeptide from at least two, more preferably at least three, and even more preferably all of the mispairing variants selected from a) to e).

[0076] format In some embodiments, the heteropolymer Fc domain-containing polypeptide consists of first, second, and third polypeptide chains, each containing an immunoglobulin variable domain. Preferably, the second polypeptide chain contains two immunoglobulin variable domains.

[0077] In some embodiments, the heteropolymer Fc domain-containing polypeptide consists of three polypeptide chains (1), (2), and (3): - V1-C1-CH2-CH3(1), - V2-C2-CH2-CH3-V3-C3(2), and - V4-C4(3) Includes, V1-V4 are immunoglobulin variable domains, and the antigen-binding domain is formed by V1 and V2, and by V3 and V4, and C1-C4 are heterodimerization domains, preferably CH1, C K and C L Selected from each of the domains, preferably one of C1 and C2 is the CH1 domain and the other is C K or C LAnd of C3 and C4, one is a CH1 domain, and the other is C K or C L And more preferably, C2 and C3 are CH1, and C1 and C4 are C K Or C L Either C2 and C3 are CH1 and C K Or C L And C1 and C4 are C K Or C L And CH1.

[0078] For the above format containing three polypeptide chains (1), (2), and (3), the following mismatch variants may occur. - "Extremely light" mispairing variants: single polypeptide chains (1), (2), or (3) (i.e., mispairing variants as defined in a) above), or dimers of (3) (i.e., mispairing variants as defined in b) above). - "Mild" mispairing variants: (1) a multimer consisting of two copies (i.e., the mispairing variant defined in b) above) and optionally a multimer consisting of one or two copies of (3) (i.e., the mispairing variant defined in c) or d) above). These "mild" mispairing variants have molecular weights within the same range as, but lower than, those of, the Fc domain-containing polypeptides of correctly paired heteromultimers. - “Heavy” mispairing variants: for example, a multimer containing two copies plus an additional chain of (2), or one copy plus an additional chain of each of (1), (2), and (3), e.g., (3)-(2)-(2)-(3), (1)-(2)-(3)-(3), or (1)-(2)-(3)-(3)-(3) (i.e., mispairing variants as defined in c) or d) above). These “heavy” mispairing variants have molecular weights within the same range as, but higher than, those of, the Fc domain-containing polypeptides of correctly paired heteromultimers. - Aggregates containing multiple polypeptide chains of all three types (i.e., mispairing variants as defined in e above).

[0079] These mispaired variants can be efficiently separated from the Fc domain-containing polypeptide of the target heteropolymer using the method of the present invention.

[0080] In a preferred embodiment, the heteropolymer Fc domain-containing polypeptide comprises three polypeptide chains: (I), (II), and (III): - V1A-C1A-L3-(CH2-CH3) A (I), - V1B-C1B-L4-(CH2-CH3) B -L1-V2A-C2A-L2(II), and - V2B-C2B(III), Includes Here, V1A and V1B form a V1 bond pair. V2A and V2B form a V2 bond pair. C1A and C1B form a C1 pair (CH1 / CL), C2A and C2B form a C2 pair (CH1 / CL), and CH1 is the constant domain 1 of the immunoglobulin heavy chain, C L This is the constant domain of the immunoglobulin light chain, (CH2-CH3) A and (CH2-CH3) B These are either identical or different, and include immunoglobulin heavy chain constant domain 2 (CH2) and immunoglobulin heavy chain constant domain 3 (CH3). L1, L2, L3, and L4 are optional, independent amino acid linkers that may be the same or different.

[0081] For the above format containing three polypeptide chains (I), (II), and (III), the same mismatch variants as those described for the format containing three polypeptide chains (1), (2), and (3) may occur.

[0082] In a preferred embodiment, the V1 binding pair specifically binds to CD123, and the V2 binding pair specifically binds to NKp46.

[0083] In some embodiments, the heteromultimer Fc domain-containing polypeptide comprises a first antigen-binding domain specific to NKp46 and a second antigen-binding domain specific to CD123. In some embodiments, the heteromultimer Fc domain-containing polypeptide, - A polypeptide chain (I) containing the amino acid sequence of SEQ ID NO: 1, a polypeptide chain (II) containing the amino acid sequence of SEQ ID NO: 2, and a polypeptide chain (III) containing the amino acid sequence of SEQ ID NO: 3, or variants of polypeptide chains (I), (II), and (III) having at least 80%, at least 85%, at least 90%, and at least 95% sequence identity with respect to SEQ ID NOs: 1, 2, and 3, respectively, or - A polypeptide chain (I) containing the amino acid sequence of SEQ ID NO: 4, a polypeptide chain (II) containing the amino acid sequence of SEQ ID NO: 5, and a polypeptide chain (III) containing the amino acid sequence of SEQ ID NO: 6, or variants of polypeptide chains (I), (II), and (III) having at least 80%, at least 85%, at least 90%, and at least 95% sequence identity with respect to SEQ ID NOs: 4, 5, and 6, respectively. Includes.

[0084] In some embodiments, the heteromultimer Fc domain-containing polypeptide comprises at least two polypeptide chains linked by at least one disulfide crosslink. In some embodiments, polypeptide chains (I), (II), and (III) are characterized in that polypeptide chain (I) is covalently bonded to polypeptide chain (II), particularly by one or more disulfide bonds. According to some of these specific embodiments, polypeptide chains (I), (II), and (III) are characterized in that polypeptide chain (II) is covalently bonded to polypeptide chain (III) by one or more disulfide bonds. In some embodiments, the heteromultimer Fc domain-containing polypeptide is characterized in that the Fc domain that binds to the human Fc-γ receptor polypeptide comprises a CH2 heavy chain constant domain having an N-linked glycosylation at residue N297 according to EU numbering. In some embodiments, the heteromultimer Fc domain-containing polypeptide is characterized in that the residue N297 according to EU numbering of the Fc region or its variants contains an N-linked glycosylation. In some embodiments, the heteropolymer Fc domain-containing polypeptide is characterized in that all or part of the Fc domain binds to the human Fc-γ receptor polypeptide. In some embodiments, the heteropolymer Fc domain-containing polypeptide is characterized in that all or part of the Fc domain binds to the human CD16A(FcγRIII) polypeptide.

[0085] An exemplary configuration is shown in Figure 3.

[0086] Protein A Schema In this specification, the term “protein A chromatography” refers to a chromatography method useful for purifying a target protein from a protein mixture, wherein the protein mixture preferably includes product-related impurities and / or process-related impurities. Protein A chromatography relies on the specific and reversible binding of a protein containing a CH2 domain (or its functional portion), a CH3 domain (or its functional portion), preferably a CH2-CH3 region, more preferably a dimeric Fc domain, to protein A. In the method according to the present invention, a protein A chromatography matrix is ​​brought into contact with the sample under conditions suitable for the binding of heteromultimeric Fc domain-containing polypeptides in the sample to protein A. For alternative embodiments, see the paragraph below relating to protein G chromatography. Methods and suitable conditions for contacting and binding Fc domain-containing polypeptides to a protein A matrix or resin are readily apparent to those skilled in the art (e.g., methods described in the protocols of manufacturers of commercially available protein A matrices or resins).Any suitable Protein A matrix or resin known in the art may be used in the methods of this disclosure, for example: Mab Select, Mab Select Xtra, Mab Select Sure, Mab Select Sure LX Protein A, Mab Select PCC, Mab Select PrismA, rProtein A Sepharose CL-4B, and nProtein A Sepharose 4FF (Cytiva); EshmunoA, ProSep A, ProSep-vA High Capacity, ProSep-vA Ultra, and ProSep-vA UltraPlus (Millipore); Porns A and Mabcapture A (Porns); IPA-300, IPA-400, and IPA-500 (RepliGen Corp.); Affigel protein A and Affiprep protein A (Bio-Rad); MABsorbent AIPP and MABsorbent A2P (Affinity Chromatography Ltd.); Protein A Ceramic Hyper DF (Pall Corp.); Ultralink Examples include Immobilized protein A and Agarose Protein A (PIERCE); Protein A Cellthru300 and Protein A Ultraflow (Bioseparation); Amsphere A3 (JSR); Fibro PrismA (Cytiva); Praesto Jetted A50, Praesto AP+ and Praesto APc (Purolite); Sartobind Rapid A membrane (Sartorius); and / or Toyopearl AF-rProtein A HC-650F (Tosoh Biosciences). In some embodiments, the protein A chromatography matrix is ​​used in a column chromatography format.In some embodiments, one or more parameters of the protein A chromatography matrix (such as pH, ionic strength, temperature, or addition of other substances) are adjusted before contacting the protein A matrix or resin with the sample. In some embodiments, the protein A matrix or resin is flushed, washed, equilibrated, delaminated, and / or disinfected before and / or after contacting the protein A matrix or resin with the sample. In some embodiments, the protein chromatography matrix is ​​equilibrated and / or washed before contacting the protein A chromatography matrix with the sample.

[0087] In some embodiments, the protein A matrix or resin is disinfected, detrimental, and / or regenerated between uses.

[0088] In the context of this invention, the term "octanoate" refers to the chemical substance with the formula CH3-(CH2)6-COOH. Octanoate is also known as octanoic acid, octic acid, octoic acid, caprylic acid, the ion (1-)1-heptanecarboxylate, n-octanoate, (n-)octylate, n-octoate, (n-)caprylate, or caprilate.

[0089] In a preferred embodiment, the octanoate in the washing solution in step c) is an octane salt. In some embodiments, the concentration of the octanoate, preferably an octane salt, is about 10 mM to about 500 mM, preferably about 25 mM to about 250 mM, and more preferably about 50 mM to about 150 mM. For example, the concentration of the octanoate, preferably an octane salt, may be about 50 mM, 60 mM, 70 mM, 75 mM, 80 mM, 90 mM, 100 mM, 110 mM, 120 mM, 130 mM, 140 mM, or 150 mM. In a preferred embodiment, the concentration is 75 mM. Any suitable source or form of octanoate known in the art (e.g., alkali salts) can be used in the washing solution of the Disclosure, for example, sodium octanoate, potassium octanoate, lithium octanoate, calcium octanoate, magnesium octanoate, beryllium octanoate, barium octanoate, strontium octanoate, rubidium octanoate, cesium octanoate, and / or any combination thereof. In some embodiments, the octanoate is an alkali salt of octanoate. In some embodiments, the octanoate is sodium octanoate or potassium octanoate. In some embodiments, the octanoate is sodium octanoate.

[0090] In some embodiments, the washing solution of step c) further comprises one or more (e.g., one or more, two or more, three or more, four or more, or all) of the following additives: benzenesulfonate, caprylic acid, hexylene glycol, propylene glycol, benzyl alcohol, unbuffered salt, and / or creatine. In some embodiments, the washing solution of step c) further comprises benzyl alcohol, hexylene glycol, and / or propylene glycol. In some embodiments, the washing solution of step c) further comprises hexylene glycol or propylene glycol in a volume / volume concentration of about 5% to about 25%, preferably about 10% to about 20%, more preferably about 15%. In some embodiments, the washing solution of step c) further comprises benzyl alcohol in a volume / volume concentration of about 0.5% to about 4%, preferably about 1% to about 3%, more preferably about 2%. It is preferable that the washing solution of step c) is benzoate-free, for example, benzoate-free.

[0091] In some embodiments, the cleaning solution further comprises a buffer. Any suitable buffer known in the art may be used in the cleaning solution of the present disclosure, for example, phosphate, tris(tris(hydroxymethyl)methylamine), bis-tris, bis-trispropane, arginine, histidine, triethanolamine, diethanolamine, formate, acetate, carbonate, MES(2-(N-morpholino)ethanesulfonic acid), citrate, HEPES(4-2-hydroxyethyl-1-piperazineethanesulfonic acid), MOPS(3-(N-morpholino)propanesulfonic acid), TAPS(3-{[tris(hydroxymethyl)methyl]amino}propanesulfonic acid), bicine(N,N-bis(2-hydroxyethyl)glycine) Examples include tris(N-tris(hydroxymethyl)methylglycine), TES(2-{[tris(hydroxymethyl)methyl]amino}ethanesulfonic acid), PIPES(piperazine-N,N'-bis(2-ethanesulfonic acid), cacodilate(dimethylarsinic acid), SSC(sodium citrate solution), and / or any combination thereof. In some embodiments, the buffer is selected from phosphate, tris, arginine, acetate, and citrate. In some embodiments, the buffer is concentrated to about 10 mM to about 500 mM. In some embodiments, the washing solution in step c) contains tris, preferably at a concentration of about 25 to about 100 mM, more preferably about 50 mM.

[0092] In some embodiments, the washing solution in step c) has a pH of about 7.0 to about 10.0, for example, about 7.0, about 7.5, about 8.0, about 8.5, about 9.0, about 9.5, or about 10.0, preferably about 7.5 or about 9.0.

[0093] In some embodiments, the washing solution of step c) further comprises NaCl, preferably at a concentration of about 0.5 to about 2.0 M, for example, about 0.5 M, about 0.75 M, about 1.0 M, about 1.25 M, about 1.5 M, about 1.75 M, or about 2.0 M, preferably about 1.0 M.

[0094] In some embodiments, the washing solution in step c) is a solution selected from buffer BON (50 mM Tris, 75 mM sodium octanoate, 1000 mM NaCl, 2% v benzyl alcohol, pH 9.0), buffer HON (50 mM Tris, 75 mM sodium octanoate, 1000 mM NaCl, 15% v hexylene glycol, pH 9.0), and buffer PON (50 mM Tris, 75 mM sodium octanoate, 1000 mM NaCl, 15% v propylene glycol, pH 9.0).

[0095] In some embodiments, the protein A chromatography matrix is ​​contacted with an equilibration buffer (or "priming wash solution") between steps b) and c). In some embodiments, the protein A chromatography matrix is ​​contacted with an equilibration buffer (or "subsequent wash solution") between steps c) and d). In some embodiments, the equilibration buffer (or priming / subsequent wash solution) includes a buffer selected from phosphate buffer, Tris buffer, acetate buffer, carbonate buffer, citrate buffer, and any combination thereof, preferably 50 mM Tris, 20 mM NaCl, pH 7.5.

[0096] This method includes a step of contacting a protein A chromatography matrix with an elution solution after one or more washing steps. In some embodiments, the elution solution in step d) contains 25 mM acetic acid. In some embodiments, the elution solution in step d) has a pH between 4.0 and 4.5, preferably 4.2.

[0097] Using the method of the present invention, the heteromultimer Fc domain-containing polypeptide and its mispaired variants are separated and eluted from the protein A chromatography matrix, thereby separating the heteromultimer Fc domain-containing polypeptide from its mispaired variants. In particular, the heteromultimer Fc domain-containing polypeptide is eluted by the elution solution, while the mispaired variants are eluted with the wash buffer or remain bound to the chromatography matrix without being eluted during the wash and elution steps.

[0098] In some embodiments, the method further includes a step of filtering the eluent via deep filtration. In some embodiments, the method further includes a virus inactivation step between steps e) and f), at a pH of about 3.0 to about 4.0, preferably about 3.2 to about 3.8, and more preferably about 3.5.

[0099] After protein A chromatography, any remaining product-related and / or process-related impurities can be removed based on differences in size, charge (e.g., isoelectric point or "IEP"), solubility, and / or degree of hydrophobicity. Therefore, the sample can be further purified using multimodal chromatography and / or ion-exchange chromatography, preferably using multimodal chromatography and ion-exchange chromatography. This method also removes product-related impurities present in the sample. In some embodiments, this method proceeds to the following step, i.e. f) The multimodal chromatography matrix is ​​brought into contact with the eluate from step e) to bind the heteromultimer Fc domain-containing polypeptide to the multimodal chromatography matrix, g) The step of bringing the multimodal chromatography matrix into contact with the elution solution, h) A step of recovering the elute containing the heteromultimer Fc domain-containing polypeptide. It also includes.

[0100] In some embodiments, this method proceeds to the following step, i.e. i) The step of bringing the ion exchange chromatography matrix into contact with the eluate of step e) or h), j) A step of recovering the elute or flow-through containing the heteropolymer Fc domain-containing polypeptide. It also includes.

[0101] Multimodal chromatography In this specification, the term “multimodal chromatography” refers to a chromatographic method useful for purifying a protein of interest from a protein mixture, wherein the protein mixture preferably includes product-related impurities and / or process-related impurities. Multimodal chromatography (also known as mixed-mode chromatography) relies on a combination of at least two modes of interaction between the protein of interest and the chromatography matrix, e.g., two or more of ion exchange, hydroxyapatite, electrostatic force, calcium coordination complex, affinity, and hydrophobic interaction. Multimodal chromatography also includes size exclusion chromatography as one of the at least two modes. In the method according to the present invention, the multimodal chromatography matrix is ​​brought into contact with the sample under conditions suitable for the binding of heteropolymer Fc domain-containing polypeptides in the sample to the multimodal chromatography matrix. Methods and suitable conditions for using a multimodal chromatography matrix or resin to purify Fc domain-containing polypeptides are readily apparent to those skilled in the art (e.g., methods described in the protocols of manufacturers of commercially available multimodal chromatography matrices or resins). The use of any suitable multimodal chromatography matrix or resin known in the art is assumed herein. In some embodiments, the multimodal chromatography matrix is ​​used in a column chromatography configuration. In some embodiments, one or more parameters of the multimodal chromatography matrix (such as pH, ionic strength, temperature, or addition of other substances) are adjusted before contacting the multimodal chromatography matrix or resin with the sample. In some embodiments, the multimodal chromatography matrix or resin is flushed, washed, equilibrated, delaminated, and / or disinfected before and / or after contacting the multimodal chromatography matrix or resin with the sample.

[0102] In some embodiments, multimodal chromatography purification (steps f-h) is performed after protein A chromatography purification (or, in an alternative embodiment, after protein G chromatography purification).

[0103] In some embodiments, the eluate from step e) is adjusted to a pH of about 4.5 to about 5.5 and a salt concentration of about 0 to about 200 mM NaCl before step f).

[0104] In some embodiments, the multimodal chromatography matrix includes one or more ligands comprising at least two, preferably at least three, more preferably at least four different binding sites selected from the group consisting of hydrophobic sites, ionic sites, hydrogen bond donor sites, and sulfur-containing sites.

[0105] In some embodiments, the multimodal chromatography matrix comprises Mep HyperCel, Capto MMC ImpRes, Capto Phenyl ImpRes, Capto Adhere ImpRes (all available from GE Healthcare), HEA HyperCel, PPA HyperCel, CHT ceramic hydroxyapatite, Toyopearl NH2-750F (available from Tosoh), or Nuvia cPrime, preferably comprising Toyopearl NH2-750F or Capto MMC ImpRes.

[0106] In some embodiments, the multimodal chromatography matrix is ​​given by formula (1) [ka] It contains ligands.

[0107] In some embodiments, the multimodal chromatography matrix is ​​brought into contact with a washing solution between steps f) and g), preferably the washing solution having a pH of about 6.0 to about 8.0.

[0108] In some embodiments, the elution solution from step g) has a pH of about 7.0 to about 8.0 and a salt concentration of about 0 to about 200 mM NaCl.

[0109] Ion exchange chromatography In this specification, the term “ion exchange chromatography” refers to a chromatographic method useful for purifying a target protein from a protein mixture, where the protein mixture preferably includes product-related impurities (proteins) and / or process-related impurities (proteins). Ion exchange chromatography relies on electrostatic interactions between the target protein and an ion exchange chromatography matrix containing charged ions. Ion exchange chromatography includes anion exchange chromatography and cation exchange chromatography. In the method according to the present invention, the ion exchange chromatography matrix is ​​brought into contact with the sample under conditions suitable for the heteropolymer Fc domain-containing polypeptide in the sample to flow through the ion exchange chromatography matrix (in other words, the heteropolymer Fc domain-containing polypeptide passes through the ion exchange chromatography matrix with flow-through). Methods and suitable conditions for purifying Fc domain-containing polypeptides using an ion exchange chromatography matrix or resin are readily apparent to those skilled in the art (e.g., methods described in the protocols of manufacturers of commercially available ion exchange chromatography matrices or resins). The use of any suitable ion exchange chromatography matrix or resin known in the art is assumed herein. In some embodiments, the ion exchange chromatography matrix is ​​used in column chromatography form. In some embodiments, one or more parameters of the ion-exchange chromatography matrix (such as pH, ionic strength, temperature, or addition of other substances) are adjusted before contacting the ion-exchange chromatography matrix or resin with the sample. In some embodiments, the ion-exchange chromatography matrix or resin is flushed, washed, equilibrated, delaminated, and / or disinfected before and / or after contacting the ion-exchange chromatography matrix or resin with the sample.

[0110] In some embodiments, ion exchange chromatography purification (steps i-j) is performed after protein A chromatography purification (or, in an alternative embodiment, after protein G chromatography purification).

[0111] In some embodiments, the eluate from step h) is adjusted to a pH of about 6.0 to about 8.0 and a salt concentration of about 0 to about 400 mM NaCl before step i).

[0112] In some embodiments, the ion exchange chromatography matrix is ​​an anion exchange chromatography matrix.

[0113] In some embodiments, step j) includes contacting an ion-exchange chromatography matrix with an elution solution to recover an eluate containing a heteropolymer Fc domain-containing polypeptide, or recovering a flow-through containing a heteropolymer Fc domain-containing polypeptide.

[0114] Heteromultimer Fc domain-containing polypeptides can be purified using the method provided herein alone or in combination with any other suitable separation technique, such as membrane filtration and protein precipitation techniques, as non-limiting and non-exclusive examples. The method may further include a step of determining the purity and proportion of heteromultimer Fc domain-containing polypeptides. This step can be implemented using any of the following techniques, such as non-reducing capillary electrophoresis (CE-NR) using high-throughput microfluidic electrophoresis on a chip or conventional capillary electrophoresis, size exclusion-high-performance liquid chromatography (SEC-HPLC), hydrophobic interaction-high-performance liquid chromatography (HIC-HPLC), ion exchange-high-performance liquid chromatography (IEX-HPLC), or reverse-phase-high-performance liquid chromatography (RP-HPLC), as non-limiting and non-exclusive examples.

[0115] In an alternative embodiment, a method for purifying a heteropolymer Fc domain-containing polypeptide according to the first embodiment is provided, wherein step b) comprises contacting a chromatographic matrix of protein A or protein G with the sample. All embodiments specified for the first embodiment are also assumed for this alternative embodiment, in which case each reference to protein A shall refer to protein A or protein G, and in particular protein G.

[0116] Protein G chromatography In an alternative embodiment, a chromatography matrix of protein A or protein G is brought into contact with the sample under conditions suitable for the binding of heteropolymer Fc domain-containing polypeptides in the sample to protein A or protein G, respectively. Methods and suitable conditions for contacting and binding the Fc domain-containing polypeptides to the protein A or protein G matrix or resin are readily apparent to those skilled in the art (e.g., methods described in the protocols of manufacturers of commercially available protein A or protein G matrices or resins). Any suitable protein A or protein G matrix or resin known in the art can be used.

[0117] All embodiments specified in the paragraph "Protein A Chromatography" may also be assumed to apply to Protein G Chromatography as appropriate.

[0118] All terms used in relation to the second, third, and fourth aspects of the present invention below have the same meanings as defined in relation to the first aspect of the present invention, unless otherwise specifically defined. Furthermore, all embodiments specified in relation to the first aspect that are applicable to the second, third, and fourth aspects are also assumed to apply to the second, third, and fourth aspects.

[0119] In a second embodiment, the present invention relates to the use of a washing solution containing an octanoate in a method for purifying a heteropolymer Fc domain-containing polypeptide according to a first embodiment of the present invention. In some embodiments, the concentration of the octanoate in the washing solution is about 10 mM to about 500 mM, preferably about 25 mM to about 250 mM, and more preferably about 50 mM to about 150 mM. For example, the concentration of the octanoate may be about 50 mM, 60 mM, 70 mM, 75 mM, 80 mM, 90 mM, 100 mM, 110 mM, 120 mM, 130 mM, 140 mM, or 150 mM. In a preferred embodiment, the concentration is 75 mM.

[0120] In a third embodiment, the present invention relates to a kit comprising a protein A chromatography matrix and a washing solution containing an octanoate. In some embodiments, the concentration of the octanoate in the washing solution is about 10 mM to about 500 mM, preferably about 25 mM to about 250 mM, and more preferably about 50 mM to about 150 mM. For example, the concentration of the octanoate may be about 50 mM, 60 mM, 70 mM, 75 mM, 80 mM, 90 mM, 100 mM, 110 mM, 120 mM, 130 mM, 140 mM, or 150 mM. In a preferred embodiment, the concentration is 75 mM. In some embodiments, the kit further comprises a multimodal chromatography matrix and / or an ion-exchange chromatography matrix.

[0121] In a fourth embodiment, the present invention relates to an eluate obtained by a method according to a first aspect of the present invention, wherein the eluate comprises at least 90%, 92%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.6%, 99.7%, 99.8%, or 99.9% of purified heteropolymer Fc domain-containing polypeptides, and less than 0.5%, less than 0.4%, less than 0.3%, less than 0.2%, or less than 0.1% of mispairing variants.

[0122] Furthermore, alternative embodiments to the second, third, and fourth embodiments of the present invention are provided. These alternative embodiments correspond to the second, third, or fourth embodiments, but relate to purification using a chromatography matrix of protein A or protein G, particularly protein G, instead of a protein A chromatography matrix. All embodiments specified in the second, third, or fourth embodiments are also assumed to apply to these alternative embodiments, where each reference to protein A shall refer to protein A or protein G, particularly protein G.

[0123] In another embodiment, the present invention relates to a purified heteropolymer Fc domain-containing polypeptide obtained by a method according to the present invention. [Brief explanation of the drawing]

[0124] [Figure 1] Purity of heteropolymer Fc domain-containing polypeptides obtained using various protein A washing solutions, as determined by microfluidic electrophoresis separation. [Figure 2] The effect of various protein A washing solutions on the removal of product-related impurities, as determined by microfluidic electrophoresis separation. Virtual gel images generated from the LabChip GXII System showing the analysis of protein A samples using ladders and various washing solutions. [Figure 3] An exemplary configuration of a heteromultimerized Fc domain-containing polypeptide. [Modes for carrying out the invention]

[0125] Example 1: Washing solution for improving the removal of product-related impurities Sample preparation Recombinant Chinese hamster ovary (CHO) cells modified to express a heteromultimeric Fc domain-containing polypeptide (referred to as the “target protein” in the Examples section) having the format shown in Figure 3 were cultured in a cell culture bioreactor. The recombinant product was secreted into the culture medium and then clarified by deep filtration for downstream processing. The clarified harvest material was filtered through a 0.2 polyethersulfone (PES) filter before being loaded onto a Protein A column.

[0126] Protein A Schema The Protein A resin was prepared as follows: MabSelect sure Protein A resin (GE Healthcare Life Science) was obtained from the supplier in a pre-packed robocolumn form (200 μl). Before loading the harvest material, the column was flushed with equilibration buffer (50 mM Tris, 20 mM NaCl, pH 7.5) to remove the storage solution. Subsequently, the column was flushed with 0.5 M acetic acid to ensure complete removal of any bound substances, and then cleaned with 0.5 M sodium hydroxide. The column was then equilibrated until the pH was >7. The prepared sample was then loaded onto the Protein A column. A multispecific heteromultimer Fc domain-containing polypeptide in the format shown in Figure 3 was loaded onto the column up to the resin target at 20 G / L. The loaded column was washed with equilibration buffer (also referred to herein as priming wash buffer). The column was then washed with the test wash solution (Table 2), and then with equilibration buffer. Finally, the target protein was eluted from the column using a solution containing 25 mM acetic acid at a pH of 3.5–3.9.

[0127] [Table 1]

[0128] [Table 2]

[0129] Purity determination For the purity determination analytical procedure, a high-throughput microfluidic Labchip® GXII capillary electrophoresis (CE) platform is used with the HT Protein Express200 Reagent kit under non-reducing conditions. This CE technique separates denatured proteins based on size, and fluorescence detection allows for the determination of the relative purity. Reagents, sample buffer, and chips are prepared according to the protocol of the HT Protein Express200 Reagent kit supplier. Before analysis, samples that have undergone at least one purification step (e.g., protein A purification) are diluted in water to approximately 1 g / L. Samples are then prepared by mixing 4 μL of diluted sample in 16.5 μL of sample buffer in a 96-well plate and heat-denatured at 70°C for 10 minutes. After incubation, 35 μL of water is added to each sample. The 96-well plate is then loaded into the analytical instrument along with the pre-prepared chips. The relative purity (primary form, "target protein") is determined by dividing the individual peak areas by the sum of the total integrated peak areas in the measured sample. The values ​​in Figure 1 and Table 3, as well as the image in Figure 2, were obtained by microfluidic electrophoresis separation using the LabChip GXII System and virtual gel images generated from the LabChip GXII System, respectively.

[0130] The results were confirmed by size exclusion chromatography (SEC) or SEC-HPLC.

[0131] result Product-related impurities (e.g., variants of the target protein with mispaired chains) have properties very similar to the target protein. They share affinity with the immobilized ligand of protein A and therefore co-elute normally with the target protein. To identify potential washing additives that can reduce the amount of mispaired molecules co-eluted with the target monoclonal antibody (mAb), a 200 μl protein A chromatography robocolumn was loaded with a sample containing the target secreted protein collected from CHO cells, clarified by deep filtration before downstream processing. The loaded column was first washed with Tris equilibration buffer. Next, the column was washed with one of several test washing solutions containing additives, carbonates, octanoates, sodium benzoate, arginine, and alcohols, either individually or in combination, as shown in Table 2. Finally, the antibody was eluted from the column, and the pH was adjusted to pH > 5.0 using 2 M Tris base before analysis.

[0132] Interestingly, washing solutions containing octanoate ("HON", "PON", "BON") increased the purity of the target protein in the eluted fraction compared to controls and compared to washing solutions containing carbonate alone or arginine alone (Figures 1 and 2, Table 3). Purity was maximized when both octanoate and benzyl alcohol were present in the washing solution. Therefore, these specific washings significantly improve purification performance by removing product-related impurities.

[0133] [Table 3]

[0134] In this experiment, product-related impurities <100 kDa are single polypeptide chains (1), (2), or (3), or dimers of (3). Product-related impurities of approximately 120 kDa and 120-150 kDa are polymers containing two copies of (1) and, optionally, one or two copies of (3). Product-related impurities >155 kDa are polymers containing, for example, two copies of (2) plus an additional chain, or one copy each of (1), (2), and (3) plus an additional chain, such as (3)-(2)-(2)-(3), (1)-(2)-(3)-(3), or (1)-(2)-(3)-(3). This group of product-related impurities also includes aggregates of heteropolymer Fc domain-containing polypeptides and their mispairing variants.

[0135] Example 2: Polishing Screening The following examples describe the use of various polishing resins to assess the possibility of eliminating product-related impurities (mismatches) during purification.

[0136] Sample preparation Human monoclonal antibody harvest material was prepared as described in Example 1. Harvest was generated in a suspension culture of recombinant CHO cells modified to express the target protein. The recombinant product was secreted into the culture medium and then clarified by deep filtration for downstream processing. The clarified harvest material was filtered through a 0.22 μm polyethersulfone (PES) filter before being loaded onto a chromatography column. This material was then captured on a protein A resin and eluted as described in Example 1 to produce material for assessment of polishing resins.

[0137] Mixed-mode chromatography Next, several polishing resins were evaluated to further improve the purity of the target protein by removing product-related impurities (Table 4). The evaluated resins were used in either flow-through mode or binding and elution mode. A range of conditions for the load material pH (between 6 and 8) and salt concentration (0 to 400 mM NaCl) were assessed for all resins used in flow-through mode. For resins used in binding and elution mode, the load pH (4.5 to 5.5), wash buffer pH (6 to 8), and elution buffer pH (7 to 8) were evaluated in the same way as the salt concentration (0 to 200 mM NaCl) of the load and elution buffers. Before loading the protein A eluate material adjusted to the corresponding pH and salt concentration, the column was equilibrated with the corresponding equilibration buffer at least 5 CV (and until the pH and conductivity reached the set point). After loading, the unbound material was pressed in using the equilibration buffer at least 10 CV. Next, for hydrophobic interaction resins, WFI was used, or for other resins, either 50 mM Tris, 1000 mM NaCl, pH 7.5 was used, followed by cleaning of the flow-through resin with 0.5 M sodium hydroxide. For binding and elution mode resins, the column was then washed with the corresponding washing buffer, and the molecules were eluted using at least 5 CV of elution buffer as described above. The column was then cleaned with 50 mM Tris, 1000 mM NaCl, pH 7.5, followed by cleaning of the column with 0.5 M sodium hydroxide.

[0138] [Table 4]

[0139] result. Product-related impurities possess biochemical properties very similar to those of the target protein. To identify polishing resins that could potentially separate the target protein from mispaired fragments, several polishing resins with specific physicochemical properties were assessed using 200 μl of robocolumn. Protein A eluate, prepared as described in Example 1, was loaded onto the resins. Several operating conditions (load pH, load conductivity, washing pH, elution pH, elution conductivity) were tested to assess the operating range of each resin and find the optimal operating conditions. The resulting samples were then analyzed and compared across all different conditions and different resins. It was found that using the appropriate polishing resin under optimized operating conditions could further remove mispaired variants and increase the purity of the target protein.

[0140] When using Capto Adhere resin and Capto Phenyl High Sub resin, no conditions were found that could increase purity to a level higher than that observed in the previous chromatography step (capture), with a purity of approximately 30% in the flow-through fraction. Derivatives of the aforementioned resins (Capto Adhere ImpRes and Capto Phenyl ImpRes) functioned similarly to the aforementioned resins, achieving a purity of approximately 30-40%. However, significantly better results were obtained when using Capto MMC ImpRes and Toyopearl NH2-750F, with Toyopearl NH2-750F achieving a purity of 47% and Capto MMC ImpRes achieving a purity of 97%. These two resins were shown to result in a significant reduction in the content of mispairs and a significant improvement in the purity of the target. Subsequently, these two resins were selected and used in combination for a complete process aimed at providing efficient removal of product-related impurities.

[0141] [Table 5]

[0142] Example 3: Pilot-scale execution This method was used for the pilot-scale batch purification of heteromultimeric Fc domain-containing polypeptides (target protein). The goal was to improve the purity of the target protein and remove product-related impurities (mispairings) and process-related impurities (host cell proteins (HCPs), DNA, etc.). The following examples describe the purification of the target protein using the method described above. First, the target protein was captured and purified by protein A chromatography. Intermediate washing containing octanoate and benzyl alcohol was used to improve / enhance the removal of mispairing impurities. Next, the target molecule was further purified using mixed-mode chromatography. Finally, the target molecule was further purified using an anion exchange resin.

[0143] Sample preparation Human monoclonal antibody harvest material was prepared as described in Example 1. Harvest was generated in suspension culture of recombinant CHO cells modified to express the monoclonal antibody. The recombinant protein of interest was secreted into the culture medium and then clarified by deep filtration for downstream processing. The clarified harvest material was filtered through a 0.22 μm polyethersulfone (PES) filter before being loaded onto a chromatography column.

[0144] Protein A Schema Protein A resin was prepared as described in Example 1. Mabselect Sure Protein A chromatography resin (Cytiva) was packed using a 140 mm diameter column (Axichrom 140 / 300 column). The resin was packed to a bed height of 20 cm ± 2 cm. The column efficiency was determined and measured as >6000 theoretical plates per meter and an asymmetry of 1.0. Before loading the harvest material, the column was flushed with 0.5 M sodium hydroxide and then equilibrated with 50 mM Tris, 20 mM NaCl, pH 7.5. The prepared sample was loaded onto the column. The column was loaded with 30 G / L of target and then washed with the equilibration buffer as described above. Next, the column was washed with a washing solution containing 50 mM Tris, 75 mM octanoate, 1 M NaCl, 2% benzyl alcohol, pH 9.0, followed again with the equilibration buffer. Finally, the target protein was eluted from the column using a solution containing 20 mM acetic acid, pH 4.2. Table 1 shows an exemplary chromatography process. Following the Protein A step, the virus inactivation step consists of acidic adjustment to the inactivation pH, static inactivation for a set period of time, and then alkaline adjustment, after which the process proceeds to the next step.

[0145] [Table 6]

[0146] Multimodal chromatography The Capto MMC ImpRes step is a binding and elution step in which the target protein and some impurities are bound, while other impurities enter during the loading and equilibration flow-through. Capto MMC ImpRes chromatography resin (Cytiva) was packed using a 140 mm diameter column (Axichrom 140 / 300 column). The resin was packed to a bed height of 20 cm ± 2 cm. The column efficiency was determined and measured with >14,000 theoretical plates per meter and an asymmetry of 1.1. Before loading the protein A eluate, the column was flushed with 0.5 M sodium hydroxide and then equilibrated with 50 mM acetic acid, pH 5.0. The sample from protein A was loaded onto the column. The column was loaded with 30 G / L of target and then washed with 50 mM Tris, pH 8.0. Finally, the antibody was eluted from the column using a solution containing 50 mM Tris, 100 mM NaCl, pH 8.0.

[0147] [Table 7]

[0148] Anion exchange chromatography. The Toyopearl NH2-750F step is a flow-through step in which impurities are bound (and further removed in purification), but the target protein enters during the flow-through loading and equilibration. Toyopearl NH2-750F chromatography resin (Tosoh Bioscience) was packed using a 50 mm diameter column (Axichrom 50 / 300 column). The resin was packed to a bed height of 20 cm + / - 2 cm. The column efficiency was determined and measured with >6800 theoretical plates and an asymmetry of 1.1 per meter. Before loading Capto MMC eluents, the column was flushed with 0.5 M sodium hydroxide and then equilibrated with 50 mM Tris, 100 mM NaCl, pH 8.0. Samples from Capto MMC were loaded onto the column. The column was loaded with 150 G / L of target and then the load was pushed in using equilibration buffer.

[0149] [Table 8]

[0150] result A pilot-scale proof of concept was successfully conducted, yielding approximately 50 grams of purified target protein. The initial purity of the clarified bulk harvest is unknown, but is estimated to be around 30% in the best-case scenario. Purity increased to 50.4% (SEC: 49.9%) after capture, then to 94.3% (SEC: 97.9%) after mixing mode, and finally to 97.9% (SEC: 100.0%) after anion exchange (Table 9).

[0151] [Table 9]

[0152] array Sequence ID 1-F25 polypeptide chain (I) [ka] Sequence ID 2-F25 polypeptide chain (II) [ka] Sequence ID 3-F25 polypeptide chain (III) [ka] Sequence ID 4-F5 polypeptide chain (I) [ka] Sequence ID 5-F5 polypeptide chain (II) [ka] Sequence ID 6-F5 polypeptide chain (III) [ka] Sequence ID 7 - Linker STGS

Claims

1. A method for purifying a heteropolymer Fc domain-containing polypeptide, comprising the following steps: a) A step of preparing a sample comprising the Fc domain-containing polypeptide of the heteropolymer and one or more mispairing variants thereof, b) The step of contacting the protein A chromatography matrix with the sample and binding the Fc domain-containing polypeptide of the heteropolymer to the protein A chromatography matrix, c) The step of contacting the protein A chromatography matrix with a washing solution containing octanoate, d) The step of bringing the protein A chromatography matrix into contact with the elution solution, e) A step of recovering the elute containing the Fc domain-containing polypeptide of the heteropolymer. A method that includes them in the order shown.

2. The next step, namely f) The step of contacting the multimodal chromatography matrix with the eluted product from step e) to bind the Fc domain-containing polypeptide of the heteropolymer to the multimodal chromatography matrix, g) The step of bringing the multimodal chromatography matrix into contact with the elution solution, h) A step of recovering the elute containing the Fc domain-containing polypeptide of the heteropolymer. The method according to claim 1, further comprising:

3. The next step, namely i) A step of bringing the ion exchange chromatography matrix into contact with the eluted substance of step e) or h), j) A step of recovering the elute or flow-through containing the Fc domain-containing polypeptide of the heteropolymer. The method according to claim 1 or 2, further comprising:

4. The method according to claim 1 for purifying the Fc domain-containing polypeptide of the heteropolymer from its mispaired variant.

5. The method according to claim 1, wherein the Fc domain-containing polypeptide comprises a first and a second polypeptide chain, each containing a CH2-CH3 region, and a third polypeptide chain that does not contain a CH2-CH3 region.

6. The method according to claim 1, further for removing process-related impurities.

7. The method according to any one of claims 1 to 6, wherein the heteropolymer Fc domain-containing polypeptide is a multispecific antibody, preferably a bispecific, triplicate, quadruplespecific, quintic, or hexaspecific antibody, more preferably a bispecific or triplicate antibody, and most preferably the heteropolymer Fc domain-containing polypeptide comprises two, three, four, five, six, seven, eight, nine, ten, eleven, or twelve immunoglobulin variable domains that form two, three, four, five, or six antigen-binding domains, and particularly four or six immunoglobulin variable domains that form two or three antigen-binding domains.

8. The method according to any one of claims 1 to 7, wherein the Fc domain-containing polypeptide of the heteropolymer contains no more than one copy of the polypeptide chain.

9. The aforementioned mismatch variant is, a) monomers of the first, second, or third polypeptide chain, b) Homodimers or homopolymers of polypeptide chains contained in the Fc domain-containing polypeptide of the heteropolymer, c) A heteropolymer containing more than one copy of the first, second, or third polypeptide chain, d) A heteropolymer that does not contain the first, second, or third polypeptide chain, and e) Aggregates of the mispaired variants according to a) to d) and / or aggregates of the heteropolymers containing the Fc domain A method according to any one of claims 1 to 8, selected from the group consisting of the following.

10. The method according to claim 9, which results in the separation of the heteropolymer Fc domain-containing polypeptide from at least one, preferably at least two, more preferably at least three, and even more preferably all of the mispaired variants.

11. The method according to any one of claims 1 to 10, wherein the heteropolymer Fc domain-containing polypeptide comprises first, second, and third polypeptide chains, each containing an immunoglobulin variable domain, and preferably the second polypeptide chain contains two immunoglobulin variable domains.

12. - The first polypeptide chain is given by formula (1): V1-C1-CH2-CH3 (1) Represented by, - The second polypeptide chain is given by formula (2): V2-C2-CH2-CH3-V3-C3 (2) Represented by, and - The third polypeptide chain is given by formula (3): V4-C4(3) Represented by, V1-V4 are immunoglobulin variable domains, and the antigen-binding domain is formed by V1 and V2, and by V3 and V4, and C1 to C4 are heterodimerization domains, preferably selected from each of the domains of CH1, C Ｋ and C Ｌ and preferably, one of C1 and C2 is the CH1 domain, and the other is C Ｋ or C Ｌ and one of C3 and C4 is the CH1 domain, and the other is C Ｋ or C Ｌ and more preferably, C2 and C3 are CH1, and C1 and C4 are C Ｋ or C Ｌ or C2 and C3 are CH1 and C Ｋ or C Ｌ and C1 and C4 are C Ｋ or C Ｌ and CH1, The method according to claim 11.

13. The purity of the Fc Domain-containing polypeptide of the heteromultimer is, - After step e), at least 40%, 45%, 50%, or 55% of the total protein concentration, - After step h), at least 50%, 60%, 70%, 80%, 90%, or 95% of the total protein concentration, - After step j), at least 90%, 92%, 94%, 95%, 96%, or 97% of the total protein concentration Preferably, the concentration is determined by capillary electrophoresis. The method according to any one of claims 1 to 12.

14. The method according to any one of claims 1 to 13, wherein the concentration of the octanoate in the washing solution in step c) is about 50 mM to about 150 mM, preferably 75 mM, and preferably the octanoate is an octanoate salt.

15. The cleaning solution in step c) is - Further comprising benzyl alcohol, hexylene glycol and / or propylene glycol, - Further containing benzyl alcohol at a volume / volume concentration of about 0.5% to about 4%, preferably about 1% to about 3%, more preferably about 2%, - Further containing hexylene glycol or propylene glycol in a volume / volume concentration of about 5% to about 25%, preferably about 10% to about 20%, more preferably about 15%, and / or - Benzoate-free, The method according to any one of claims 1 to 14.

Images