Imaging capillary isoelectric focusing for detection of mispaired byproducts in asymmetric bispecific antibodies

Abstract

A method for detecting a by-product in a sample of a target asymmetric bispecific antibody (AsBsAb) is provided, comprising subjecting the sample to capillary isoelectric focusing to separate the by-product from the target AsBsAb based on a difference in their pi. In particular, the AsBsAb comprises two half antibodies, and the pi of the first half antibody is different from the pi of the second half antibody.

Description

[0001] Cross-references

[0002] This application claims the benefit of international patent application PCT / CN2024 / 073668, filed on January 23, 2024, the entire contents of which are incorporated herein by reference.

[0003] sequence list

[0004] This application contains a sequence list, the entire contents of which are incorporated herein by reference. Technical Field

[0005] This application generally relates to methods for characterizing antibodies and byproduct proteins based on pI differences, and the uses of said methods. Background Technology

[0006] In recent years, bispecific antibodies (BsAbs) have attracted much attention because they can simultaneously target two different targets or two different epitopes on the same target. Different forms of BsAbs have been developed to meet various biological needs[1]. Among them, asymmetric bispecific antibodies (AsBsAbs) represent a major class, which have advantages such as improved stability, prolonged serum half-life and reduced immunogenicity[2].

[0007] Preparing AsBsAbs by co-expressing two heavy chains and two light chains in a single host cell can be extremely challenging due to the difficulty in removing closely related mismatch byproducts. Several techniques have been developed to reduce the chain mismatch problem [3]. In particular, the KiH technique has been widely used to promote the formation of heterodimers in the two heavy chains, and the Crossmab technique has been widely used to promote homologous heavy chain (HC)-light chain (LC) pairing. Although these methods significantly improve heterodimer formation, they often still result in low levels of homodimerization during upstream production. The resulting homodimers often exhibit similar physical properties (e.g., size, charge, and hydrophobicity) to the heterodimers, making their separation highly challenging [8].

[0008] Currently, developing analytical methods for monitoring and measuring homodimers is very challenging due to the subtle differences between homodimers and heterodimers. Although several studies [5,9,10] have proposed specific methods for the detection and quantification of BsAb mismatch byproducts, there is currently no standardized analytical method for a given BsAb platform, mainly due to the lack of physicochemical differences between homodimers and heterodimers.

[0009] Therefore, in the development of BsAb, especially asymmetric BsAb, drug products, new analytical methods are urgently needed to accurately identify and quantify mismatch byproducts. Invention Overview

[0011] This application provides an analytical method specifically for a particular form of bispecific antibody (BsAb). Specifically, the present invention develops an imaging capillary isoelectric focusing (iCIEF) method to monitor mismatch byproducts in AsBsAbs, and this method may be expected to become a platform method for detecting the content of mismatch byproducts in a class of AsBsAbs, particularly AsBsAbs containing two half-antibodies paired with significant pI differences.

[0012] In one aspect, this article provides a method for detecting byproducts in a sample of a target asymmetric bispecific antibody (AsBsAb), comprising capillary isoelectric focusing (e.g., iCIEF) of the sample to separate the byproducts based on the pI difference between them and the target AsBsAb, wherein the AsBsAb comprises two half-antibodies with significant pI differences, such as pI differences of at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 1.8, at least 2.0, at least 3.0, at least 4.0 or greater.

[0013] In some embodiments, the target AsBsAb comprises two heavy chains and two light chains, wherein the first half-antibody comprises or consists of a first heavy chain and an associated light chain, and the second half-antibody comprises or consists of a second heavy chain and an associated light chain. In some embodiments, the target AsBsAb comprises two heavy chains and one light chain, wherein the first half-antibody comprises or consists of a first heavy chain and an associated light chain, and the second half-antibody comprises or consists of a second heavy chain.

[0014] In some embodiments, the first hapten comprises a first antigen-binding moiety, and the second hapten comprises a second antigen-binding moiety, wherein the significant pI difference between the two haptens is primarily attributed to the significant pI difference between the first and second antigen-binding moieties. In some specific embodiments, the first antigen-binding moiety is a chimeric Fab whose constant domains CH1 and CL are replaced with a pair of TCR constant domains. Since the pI of the immunoglobulin CH1 and CL domains is typically around 8.8, while the pI of the TCR constant domain (whether native or engineered in a small number of amino acid residues) is around 4.8, the resulting chimeric Fab has a significantly lower pI than the non-chimeric Fab.

[0015] Furthermore, the second antigen-binding moiety can be selected from antigen-binding fragments, such as Fab, single-chain Fv (scFv), and VHH. In some embodiments, the second antigen-binding moiety is a non-chimeric Fab, i.e., its constant domains CH1 and CL are not replaced with the TCR constant domain. Therefore, the first half-antibody consists of a pair of first heavy chains and a first light chain containing a chimeric Fab, and the second half-antibody consists of a pair of second heavy chains and a second light chain containing a non-chimeric Fab. In some other embodiments, the second antigen-binding moiety is scFv. Therefore, the first half-antibody consists of a pair of first heavy chains and a first light chain containing a chimeric Fab, and the second half-antibody consists of a second heavy chain containing scFv. In some other embodiments, the second antigen-binding moiety is VHH. Therefore, the first half-antibody consists of a pair of first heavy chains and a first light chain containing a chimeric Fab, and the second half-antibody consists of a second heavy chain containing VHH. The pI of the chimeric Fab is much lower than that of the non-chimeric Fab, scFv or VHH. This is because the chimeric Fab introduces the TCR constant structural domain to replace the CH1 and CL structural domains.

[0016] In some implementations, the TCR constant domain of the chimeric Fab is engineered to introduce one or more non-natural disulfide bonds while still maintaining a low pI.

[0017] In some embodiments, the TCR constant domains are TCRα and TCRβ constant domains, which are natural or engineered to introduce disulfide bonds. The TCRβ constant domain may contain an amino acid sequence having at least 90%, 95%, or 99% identity with SEQ ID NO: 5, 6, or 10. The TCRα constant domain may contain an amino acid sequence having at least 90%, 95%, or 99% identity with SEQ ID NO: 7, 8, 9, or 11. Specifically, the TCRβ and TCRα constant domains contain the amino acid sequences of SEQ ID NO: 6 and 9, or SEQ ID NO: 10 and 9, or SEQ ID NO: 10 and 11, or SEQ ID NO: 6 and 8, respectively.

[0018] In some embodiments, the chimeric Fab includes a TCRβ constant domain in the heavy chain instead of the CH1 domain, and a TCRα constant domain in the light chain instead of the CL domain. In other embodiments, the chimeric Fab includes a TCRα constant domain in the heavy chain instead of the CH1 domain, and a TCRβ constant domain in the light chain instead of the CL domain.

[0019] In some embodiments, AsBsAb comprises a pair of dimerizing domains that associate with each other to anchor two heavy chains together, wherein each dimerizing domain is fused to the C-terminus of a first and a second antigen-binding moiety. When the antigen-binding moiety is Fab, it can be operatively linked to the first or second dimerizing domain via its heavy chain. The operative linking may be direct or indirect via a peptide linker.

[0020] The paired dimerizing domains may be immunoglobulin constant domains, such as IgG1, IgG2, IgG3, or IgG4 constant domains. In some embodiments, a pair of dimerizing domains is an Fc region containing a pair of Fc domains. The Fc domains can be connected to the first and second antigen-binding moieties via hinge regions.

[0021] In some implementations, AsBsAb includes a "knob-into-hole" structure in the Fc region to facilitate heterodimerization of the two heavy chains. The knob structure can be obtained by replacing a smaller amino acid (such as a bulkier W) in the CH3 domain of one heavy chain, and the mortar structure can be constructed by replacing a larger residue (such as A or T) in the CH3 domain of the other heavy chain.

[0022] In some embodiments, the first half-antibody contains a "pestle" structure in its Fc domain, and the second half-antibody contains a "mortar" structure in its Fc domain. Alternatively, the first half-antibody contains a "mortar" structure in its Fc domain, and the second half-antibody contains a "pestle" structure in its Fc domain.

[0023] In some embodiments, according to EU designations, the pestle structure contains the S354C and / or T366W mutation, and the mortar structure contains the Y349C, T366S, L368A, and / or Y407V mutation. In some embodiments, according to EU designations, the pestle structure contains the T366W mutation, and the mortar structure contains the T366S, L368A, and Y407V mutation. The AsBsAb used in the methods described herein may contain other substitutions capable of forming a pestle-mortar structure.

[0024] In some embodiments, the first half-antibody comprising the chimeric Fab contains a club-shaped structure, and the second half-antibody contains a mortar-shaped structure. The resulting byproducts may include an AsBsAb K-type halfmer composed of a first pair of heavy and light chains, a homodimer of the K-type halfmer, an AsBsAb H-type halfmer composed of a second heavy chain and a second light chain (if present), and a homodimer of the H-type halfmer. In this case, due to the introduction of the TCR constant domain, the pI of the K-type halfmer and its homodimer is significantly lower than that of the H-type halfmer and its homodimer. In some further embodiments, the pI of the K-type halfmer is at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 1.8, at least 2.0, at least 3.0, at least 4.0, or more lower than that of the H-type halfmer. In some further embodiments, the pI of the K-type semimer is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 1.0, at least 2.0 or more lower than the pI of the target AsBsAb, and the pI of the target AsBsAb is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 1.0, at least 2.0 or more lower than the pI of the H-type semimer.

[0025] In some embodiments, the first half-antibody comprising the chimeric Fab contains a mortar structure, and the second half-antibody contains a pestle structure. The resulting byproducts may include an AsBsAb H-type halfmer composed of a first pair of heavy and light chains, a homodimer of the H-type halfmer, an AsBsAb K-type halfmer composed of a second heavy chain and a second light chain (if present), and a homodimer of the K-type halfmer. In this case, due to the introduction of the TCR constant domain, the pI of the H-type halfmer and its homodimer is significantly lower than that of the K-type halfmer and its homodimer. In some further embodiments, the pI of the H-type halfmer is at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 1.8, at least 2.0, at least 3.0, at least 4.0, or more lower than that of the K-type halfmer. In some further embodiments, the pI of the H-type semimer is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 1.0, at least 2.0 or more lower than the pI of the target AsBsAb, and the pI of the target AsBsAb is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 1.0, at least 2.0 or more lower than the pI of the K-type semimer.

[0026] In some embodiments, the method further includes quantifying the byproducts, for example determining the content levels of K-type semimers and their homodimers in the sample, and / or the content levels of H-type semimers and their homodimers.

[0027] In some embodiments, the concentration of K-type semimer and its homodimer in the sample ranges from about 4% to 20%. In some embodiments, the concentration of H-type semimer and its homodimer in the sample ranges from about 1% to 20%.

[0028] In some implementations, the sample is obtained from a bioprocess used to generate the target AsBsAb, including clarification, chromatographic production, virus inactivation, and filtration. Optionally, the sample has been diluted before capillary isoelectric focusing.

[0029] In some implementations, the sample is obtained from cell culture medium, harvested cell culture medium, eluent from affinity chromatography, anion exchange chromatography or mixed-mode chromatography, filtrate, ultrafiltrate, percolate, active pharmaceutical ingredient (DS), or pharmaceutical product (DP) containing the final formulation.

[0030] In some implementations, capillary isoelectric focusing is imaging capillary isoelectric focusing, optionally coupled with chromatography or mass spectrometry.

[0031] In one aspect, this article provides a method for assessing the heterogeneity of samples derived from AsBsAb production, comprising: performing capillary isoelectric focusing on the sample to separate them based on the pI difference between the byproduct and the target AsBsAb, wherein the AsBsAb comprises two half-antibodies with significant pI differences, such as pI differences of at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 1.8, at least 2.0, at least 3.0, at least 4.0 or greater.

[0032] The above is an overview and therefore necessarily includes simplifications, generalizations and omissions of details; therefore, those skilled in the art should understand that this overview is merely illustrative and not intended to be limiting in any way. Attached Figure Description

[0033] Figure 1 An example of an asymmetric bispecific antibody (BsAb) structure is shown, characterized in that the left hemibody (blue) has a higher isoelectric point, while the right hemibody (orange) has a lower isoelectric point. The left hemibody comprises a chimeric Fab, and the right hemibody comprises a conventional Fab. Figure 1 -a) or scFv ( Figure 1 -b) or VHH ( Figure 1-c). BsAb can contain a pair of engineered Fc domains to facilitate heavy chain-heavy chain pairing.

[0034] Figure 2 A schematic diagram of AsWXbsAb and mismatch byproducts is shown. K-type hemimers refer to hemiantibodies containing a club-shaped structure in the Fc domain, while H-type hemimers refer to hemiantibodies containing a mortar-shaped structure in the Fc domain.

[0035] Figure 3 The specificity profile of the iCIEF method is shown. K-related species refer to K-type hemimers and their homodimers, while H-related species refer to H-type hemimers and their homodimers.

[0036] Figure 4 Linear fitting plots for K-related categories (left) and H-related categories (right) are shown.

[0037] Figure 5 An exemplary purification flowchart is shown, and the method described herein is capable of detecting and characterizing byproducts in BsAb samples collected from any step. Invention Details

[0039] Protein therapeutics require careful formulation to ensure product homogeneity and stability. The presence of byproducts in therapeutic proteins raises concerns about drug efficacy, potency, and patient safety, as impurities in therapeutic proteins can affect their binding affinity, bioactivity, and shelf life in certain cases. Numerous analytical methods have been designed specifically to evaluate mismatch byproducts in different AsBsAb forms. However, a significant gap remains in the current literature: the lack of standardized assays tailored to specific bispecific antibody (BsAb) platforms. To further facilitate the purification of heterodimeric bispecific antibodies, pI engineering techniques have been employed in BsAb molecular design. Therefore, by constructing a pI difference between two different half-antibodies of a BsAb through the substitution of charged residues, a heterodimeric species with a significantly different pI value compared to the potential homodimeric byproducts will be generated. As shown in this paper, the pI difference between the heterodimer and homodimer of AsBsAb not only helps to improve the purification efficiency of the heterodimer, but can also be reasonably used to develop iCIEF as an analytical method for monitoring homodimer and hemimer byproducts.

[0040] While this disclosure may be embodied in many different forms, the contents disclosed herein are specific illustrative embodiments that illustrate the principles of this disclosure. It should be emphasized that this disclosure is not limited to the specific embodiments shown. Furthermore, any section headings used herein are for organizational purposes only and should not be construed as limiting the subject matter described.

[0041] Unless otherwise defined herein, scientific and technical terms used in connection with this disclosure shall have the meanings commonly understood by one of ordinary skill in the art. Furthermore, unless the context otherwise requires, singular terms shall include plural terms, and plural terms shall include singular terms. More specifically, as used in this specification and the appended claims, the singular forms “a” and “the” include plural references unless the context explicitly specifies otherwise. Thus, for example, reference to “a protein” includes multiple proteins; reference to “a cell” includes a mixture of cells, and so on. In this application, unless otherwise stated, the use of “or” means “and / or”. Moreover, the use of the term “comprising” and other forms such as “comprises” and “comprised” is not restrictive. Additionally, the scope provided in this specification and the appended claims includes both endpoints and all points between the endpoints.

[0042] In general, the nomenclature and techniques used in conjunction with cell and tissue culture, molecular biology, immunology, microbiology, genetics, and protein and nucleic acid chemistry and hybridization described herein are well-known and commonly used in the art. Unless otherwise stated, the methods and techniques disclosed herein are generally performed according to conventional methods well-known in the art and as described in the various general and more specific references cited and discussed throughout this specification. See, for example, Abbas et al., Cellular and Molecular Immunology, 6th Edition, WB Saunders Company (2010); Sambrook J. & Russell D. Molecular Cloning: A Laboratory Manual , 3rd edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (2000); Ausubel et al., Short Protocols in Molecular Biology: A Compendium of Methods from Current Protocols in Molecular Biology , Wiley, John & Sons, Inc. (2002); Harlow and Lane Using Antibodies: A Laboratory Manual , Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1998); and Coligan et al., Short Protocols in Protein Science Wiley, John & Sons, Inc. (2003). The nomenclature and laboratory procedures and techniques used in conjunction with the analytical chemistry, synthetic organic chemistry, and medical and medicinal chemistry described herein are well-known and commonly used in the field.

[0043] definition

[0044] To better understand this disclosure, the definitions and explanations of relevant terms are provided below.

[0045] As used herein, the term "isoelectric focusing" or "IEF" refers to a technique for separating charged molecules (typically proteins or peptides) based on their isoelectric point (pI), which is the pH value at which the molecule is uncharged. The principle of IEF is that molecules within a pH gradient migrate towards their pI in an electric field. Proteins and peptides contain multiple ionizable carboxyl and amino groups and are inherently amphoteric molecules. Each amphoteric molecule becomes neutral at the pH value where the positive and negative charges in the molecule are balanced. Proteins begin to move according to their charge and become neutral at locations where the local pH matches their pI. Various techniques exist for performing IEF.

[0046] As used herein, the term “capillary isoelectric focusing” or “CIEF” refers to a type of IEF in which a sample moves through a capillary based on an applied electric field. An analyte, such as a protein, can be detected at a point along the capillary by using a UV detector at that point. In a typical CIEF, the sample protein, standards (i.e., pI markers), amphoteric electrolyte, polymer, and optional additives are premixed and packed together in a capillary. After connecting one end of the packed capillary to a low-pH electrolyte (anodic electrolyte) and the other end to a high-pH electrolyte (cathode electrolyte), an electric field is applied. Hydrated hydrogen ions from the anodic electrolyte vial move towards the cathode, while hydroxide ions move from the cathode electrolyte vial towards the anode. In this process, the protein and amphoteric electrolyte are sorted from low pI to high pI within the capillary. Variations of CIEF, such as cIEF-whole-column detection, also known as imaging CIEF (iCIEF), can also be used.

[0047] As used herein, the term "antibody" encompasses any form of antibody that exhibits the desired biological or binding activity. This includes, but is not limited to, humanized antibodies, fully human antibodies, chimeric antibodies, and single-domain antibodies, as well as fusion proteins and fragments of any of the above, provided they exhibit the desired antigen-binding activity. Conventional antibodies consist of two heavy chains and two light chains. The heavy chains can be divided into μ, δ, γ, α, and ε chains, which define the antibody isotype as IgM, IgD, IgG, IgA, and IgE, respectively. The heavy chains may contain a heavy chain variable region (V... H ) and heavy chain constant region (C H A heavy chain can contain one or more constant regions, for example, three constant regions (C). H 1. C H 2 and CH 3). Light chains can contain light chain variable regions (V). L ) and light chain constant region (C L V H and V L The regions can be further divided into hypervariable regions (called complementarity-determining regions (CDRs)), which are separated by relatively conserved regions (called frame regions (FRWs)). Antibodies can belong to different antibody isotypes, for example, IgG (e.g., IgG1, IgG2, IgG3, or IgG4 subtypes), IgA1, IgA2, IgD, IgE, or IgM antibodies.

[0048] As used herein, the term "bispecific antibody (BsAb)" refers to an IgG-like antibody having two binding sites, targeting two different antigens or two different epitopes on the same antigen. BsAbs can be further categorized into symmetric or asymmetric architectures. The asymmetric BsAb disclosed herein is an artificial heterodimer antibody comprising two distinct heavy chains, and more specifically, it contains an introduced TCR constant domain within a half-antibody. The heavy chain of a BsAb refers to the chain containing the heterodimerization domain, more specifically, the Fc domain. Preferably, the Fc region in the heavy chain is engineered to include a "mortise and tenon" substitution, thereby promoting heterodimerization.

[0049] As used herein, the term "mortar and pestle" refers to the engineering of the CH3 domain of the antibody's Fc region to construct a "mortar" or "pestle" in each heavy chain. This design aims to promote the formation of the target heterodimer BsAb, rather than a mismatch product. For example, a "pestle" is constructed by replacing T366 with a bulkier residue W on one heavy chain, and a corresponding "mortar" is formed by triple mutations T366S, L368A, and Y407V on another heavy chain. The mortar and pestle structure may include other substitutions as well known in the art. In some embodiments, the asymmetric bispecific antibody disclosed herein is a bispecific antibody comprising a chimeric Fab and a mortar and pestle structure.

[0050] As used herein, in the context of antibodies, "Fab" refers to the portion of an antibody consisting of a single light chain (both the variable and constant regions) associated with a single heavy chain via disulfide bonds to a variable region and a first constant region. In some embodiments, Fab is a chimeric Fab, wherein the constant regions (i.e., CH1 and CL) of the two Fab chains are replaced with engineered or modified TCR constant regions.

[0051] As used herein, the term "Fc region" refers to the C-terminal region of the immunoglobulin heavy chain, including, for example, the native sequence Fc region, the recombinant Fc region, and the variant Fc region, and optionally includes part or all of the hinge region. Fc variants contain an amino acid sequence that differs from the amino acid sequence of the native sequence Fc region, differing in that they have at least one amino acid modification (e.g., substitution, addition, or deletion), preferably one or more amino acid substitutions.

[0052] As used herein, the term "WuXiBody" refers to a bispecific antibody comprising a chimeric Fab containing a variable domain derived from a first parent antibody and a constant domain of the TCR (e.g., the α and β domains of the TCR constant region, also referred to as Cα and Cβ). WuXiBody further comprises a second antigen-binding moiety (e.g., Fab, scFv, or VHH) with different antigen-binding specificities, which typically contains a variable domain derived from the second parent antibody. As used herein, a Fab with a constant TCR domain replacing the CH1 and CL domains is referred to as a chimeric Fab, while a Fab with conventional CH1 and CL domains is referred to as a non-chimeric Fab or conventional Fab. The α and β constant domains of the TCR preferentially associate with each other and less readily associate with CL or CH1, thereby preventing and significantly reducing the formation of unwanted pairings such as Cα-CH, Cα-CL, Cβ-CH, and Cβ-CL. The constant TCR domain is not limited to wild-type constant TCR domains and specific constant TCR domains illustrated herein. WuXiBody can employ a variety of modified TCR constant domain sequences (see WO2022156687, the entire contents of which are incorporated herein by reference), as long as they enable stable association between the heavy and light chains in the chimeric Fab. The TCR constant domains associate with each other via one or more disulfide bonds and can be engineered to form more than one pair of disulfide bonds, thereby improving stability and / or expression levels. Detailed descriptions of various forms of WuXiBody can be found in WO2019057122, WO2019057124, and WO2020057610 (the entire contents of all these references are incorporated herein by reference). In one asymmetric form of WuXiBody described herein, BsAb contains a chimeric Fab in one binding arm and a second Fab in the other binding arm, both operatively linked at the C-terminus to one chain of the immunoglobulin Fc region, which contains a "pestle" substitution to promote heterodimerization.

[0053] The natural "T-cell receptor," or natural "TCR," is a heterodimeric T-cell surface protein that associates with the invariant CD3 chain to form a complex capable of mediating signal transduction. The TCR belongs to the immunoglobulin superfamily and is similar to a hapten with a single heavy chain and a single light chain.

[0054] As used herein, the term "half-antibody" or "half-mer" refers to half of an asymmetric BsAb. If the asymmetric BsAb contains two heavy chains and two light chains, the half-antibody consists of a pair of heavy chains and an associated light chain. If the asymmetric BsAb contains two heavy chains and one light chain, the half-antibody consists of a pair of heavy chains and an associated light chain, or it consists of another heavy chain. Depending on whether the heavy chain of the half-antibody contains a club or a mortar structure in the Fc domain, the half-antibody may be referred to as a "K-type half-mer" or a "H-type half-mer." The term K-type half-mer refers to a half-antibody containing a club structure, while the term H-type half-mer refers to a half-antibody containing a mortar structure. In some embodiments, the BsAb used in the methods disclosed herein comprises K-type and H-type half-mers with significant pI differences.

[0055] As used herein, the term "K-type species" refers to the K-type hemimer and its homodimer (i.e., KK homodimer). Similarly, the term "H-type species" refers to the H-type hemimer and its homodimer (i.e., HH homodimer). K-type and H-type species are typically byproducts that accompany the desired asymmetric BsAbs. During the recombination of asymmetric BsAbs, various byproducts are frequently observed due to chain expression imbalance and incorrect chain pairing, with half-antibodies and homodimers occurring most frequently.

[0056] As used herein, the term "isolated antibody" means an antibody that is substantially free of other antibodies with different antigen specificities. Isolated antibodies may be substantially free of other cellular material and / or chemicals.

[0057] As used herein, the term "host cell" refers to a cell into which a vector can be introduced, including but not limited to prokaryotic cells such as Escherichia coli. E. coli ) or Bacillus subtilis ( Bacillus subtilis ), fungal cells such as yeast cells or Aspergillus spp. ( Aspergillus Cells, insect cells such as S2 Drosophila cells or Sf9 cells, and animal cells such as fibroblasts, CHO cells, COS cells, NSO cells, HeLa cells, BHK cells, HEK 293 cells, or human cells.

[0058] Method Principles

[0059] The method disclosed herein provides a reliable and standardized approach for evaluating mismatch byproducts in the production of asymmetric BsAbs containing two hemibosomes with different pIs. This method is particularly applicable to the WuXiBody platform, where a constant TCR domain is introduced into one of the hemibosomes of the BsAb, thereby amplifying the pI difference between the two hemibosomes. The method is also applicable to BsAbs engineered to have pI differences between their two hemibosomes. Therefore, the method disclosed herein provides a novel platform mismatch detection approach that can improve the efficiency and reliability of AsBsAb development, thereby advancing therapeutic modalities in the biopharmaceutical industry.

[0060] The method of this application is based on capillary isoelectric focusing (CIEF), more specifically iCIEF, to characterize byproducts of asymmetric BsAbs. These byproducts can include mismatched homodimers, haptens, light chain mismatches, antibody fragments, and high levels of high molecular weight (HMW) species, all of which present unique challenges to their downstream processing. Using BsAbs containing a constant TCR domain instead of CH1 and CL domains in a hapten and further including a mortar-and-pot (KiH) structure in the Fc region as model molecules, the inventors demonstrated excellent characterization of BsAb byproducts via iCIEF. These byproducts include mortar-and-pot haptened homodimer mismatches, which are physicochemically very similar to the target bsAb and are still present even in the KiH form, albeit at reduced levels. Byproducts may also include mortar-and-pot haptened homodimers, although occurring less frequently due to the inherent steric hindrance of the mortar, and may also include fragments such as haptens (e.g., BsAbs lacking both heavy chain (HC) and light chain (LC)), as well as HC and LC mismatches.

[0061] The method presented in this paper utilizes the pI difference between the target BsAb and its hemimeric or homodimeric byproducts and is applicable to the development of purification processes for bispecific antibodies, particularly for BsAbs containing a constant TCR domain within a single half-antibody. The constant TCR domain (natural or engineered) has a relatively low pI of approximately 4.8 due to its specific amino acid sequence rich in acidic residues, leading to significant pI differences between the two half-antibodies. As shown in this paper, the constant TCR domain can be engineered to introduce one or more disulfide bonds by replacing a few amino acids with Cys residues, thereby maintaining the pI of the chimeric Fab at a lower level than that of the conventional Fab. The potential pI difference between the variable domains of the two half-antibodies does not offset this effect or further exacerbates this pI imbalance.

[0062] As illustrated in the embodiments, the method described herein employs imaging capillary isoelectric focusing (iCIEF) for the detection and quantification of mismatched byproducts within a target asymmetric BsAb framework, particularly for pairwise haptens with significantly different isoelectric points (pI). In some embodiments, the asymmetric BsAb comprises or consists of two haptens, wherein the pI difference between the two haptens is at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 1.8, at least 2.0, or greater.

[0063] In some embodiments, the target BsAb is an asymmetric BsAb comprising two pairs of heavy and light chains, wherein the first pair of heavy and light chains comprises a chimeric Fab with a TCR constant domain replacing the CH1 and CL constant domains, and the second pair of heavy and light chains comprises a non-chimeric Fab with conventional CH1 and CL constant domains. In some embodiments, the target BsAb is an asymmetric BsAb comprising two heavy chains and a light chain associated with a first heavy chain, wherein the first heavy chain and the associated light chain comprise a chimeric Fab with a TCR constant domain replacing the CH1 and CL constant domains, and the second heavy chain comprises scFv. In some embodiments, the target BsAb is an asymmetric BsAb comprising two heavy chains and a light chain associated with a first heavy chain, wherein the first heavy chain and the associated light chain comprise a chimeric Fab with a TCR constant domain replacing the CH1 and CL constant domains, and the second heavy chain comprises VHH. The positions of the TCR Cα and TCR Cβ domains can be interchanged. In some embodiments, the CH1 domain in the chimeric fabric is replaced with a TCR Cβ domain, and the CL domain is replaced with a TCR Cα domain. In some other embodiments, the CH1 domain in the chimeric fabric is replaced with a TCR Cα domain, and the CL domain is replaced with a TCR Cβ domain.

[0064] In some alternative implementations, the target BsAb includes a "pestle" structure in the Fc region to facilitate heterodimerization. Any engineered approach can be used to facilitate the heterodimerization of asymmetric BsAbs, and it is not limited to the pestle-and-mortar structure.

[0065] In some embodiments, the heavy chain of the first hapten contains a "groove" structure, while the heavy chain of the second hapten contains a "mortar" structure; that is, the first hapten is a groove-containing hemibody or a K-type hapten, and the second hapten is a mortar-containing hemibody or an H-type hapten. When the TCR constant domain with a lower pI is located in the K-type hapten, the resulting K-type hapten will have a lower pI than the H-type hapten. The pI of the target BsAb lies between the pIs of the K-type hapten and the H-type hapten. In some embodiments, the pI of the target BsAb is at least 0.1, at least 0.2, at least 0.3, or at least 0.4 higher than the pI of the K-type hapten, or more. In some embodiments, the pI of the target BsAb is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, or more lower than the pI of the H-type hapten. In some implementations, the pI of the H-type hapten is at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 2, or more higher than that of the K-type hapten. The pI can be a theoretical pI or a measured pI.

[0066] In some alternative embodiments, the heavy chain of the first hapten contains a "mortar" structure, while the heavy chain of the second hapten contains a "gross club" structure; that is, the first hapten is a hemibody containing a mortar or an H-type hapten, and the second hapten is a hemibody containing a gross club or a K-type hapten. When the lower pI TCR constant domain is located in the H-type hapten, the resulting H-type hapten will have a lower pI than the K-type hapten. The pI of the target BsAb lies between the pIs of the K-type hapten and the H-type hapten. In some embodiments, the pI of the target BsAb is at least 0.1, at least 0.2, at least 0.3, or at least 0.4 higher than the pI of the H-type hapten, or more. In some embodiments, the pI of the target BsAb is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, or more lower than the pI of the K-type hapten. In some implementations, the pI of the K-type hapten is at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 2, or more higher than that of the H-type hapten. The pI can be a theoretical pI or a measured pI.

[0067] In some embodiments, the methods described herein are used for the detection and quantification of haptens containing mortar and their homodimers. Alternatively or additionally, the methods described herein are used for the detection and quantification of haptens containing mortar and their homodimers.

[0068] The method described in this application can provide valuable information about the heterogeneity of BsAbs, which can affect clinical pharmacology related to drug administration, such as the pharmacokinetics, efficacy, and safety of biologics.

[0069] Capillary isoelectric focusing (CIEF)

[0070] The benefits of this application include providing a highly sensitive method for the detection and quantification of mismatched byproducts of asymmetric bispecific antibodies. This method is based on the principle of separating the target BsAb and its byproducts based on the pI difference using capillary isoelectric focusing electrophoresis (CIEF), such as imaging capillary isoelectric focusing electrophoresis, chromatography-coupled capillary electrophoresis, chromatography-coupled imaging capillary electrophoresis, or mass spectrometry (MS)-coupled capillary isoelectric focusing electrophoresis, to obtain further information about the protein sample. CIEF is an advantageous method because it enables high-resolution separation of sample components based on pI and can simultaneously consider both surface-exposed and internal amino acids without resolution degradation due to hydrophobic interactions. Furthermore, imaging cIEF, a form of CIEF, has been widely used in protein-based drug development as a tool for product identification, stability monitoring, and characterization. Applications of icIEF technology using the Convergent Bioscience icIEF instrument equipped with full-field imaging technology include, for example, the development of rapid methods for establishing product release identification tests, concentration determinations during intermediate processes in upstream and downstream product development, and stability studies of proteins under accelerated temperature stress conditions related to their charge heterogeneity. The icIEF method serves as a versatile assay because it can screen for superior candidate products in the early clonal selection stage and support characterization analysis of intermediate processes and final products.

[0071] icIEF utilizes whole-capillary imaging technology to detect focused proteins at 280 nm. Compared to cIEF on conventional CE instruments, icIEF offers two major advantages: shorter method development time and higher reproducibility. In conventional CE systems, once isoelectric focusing separation occurs, proteins still need to move to pass through a single-point UV detector, which can lead to additional spectral broadening and increased analysis time (20 minutes in icIEF, compared to 60 minutes in conventional cIEF). Another advantage of icIEF is its ability to monitor the focusing process of proteins after sample introduction, allowing for visualization of separation dynamics such as adsorption and precipitation.

[0072] The isoelectric point of a protein is one of its inherent properties, defined as the pH value at which the protein carries zero charge. The pKa value of each amino acid in the primary sequence is used as the basis for calculating the theoretical pI of a protein. Therefore, the pI of a half-antibody is essentially the same as that of its homodimer, and the average pI of an AsBsAb assembled from two half-antibodies falls between the pI of the first and second half-antibodies. The measured pI may differ slightly from the theoretical pI.

[0073] In CIEF applications, the sample and a carrier ampholyte (CA) are injected into a capillary. The CA creates a pI gradient throughout the capillary, and each component migrates according to its respective pI. Ultimately, each component focuses at its neutral pH. In some embodiments, the conditions used for CIEF of sample components preferably include a carrier ampholyte, a pI marker, urea, and the sample. The carrier ampholyte, such as Pharmalyte 3-10, can create a stable linear pH gradient in the capillary and maintain uniform conductivity, thereby separating charged molecules (proteins or peptides) according to their pI. If significant precipitation and aggregation occur, different additives can be added to the protein sample to improve reproducibility. Urea is a commonly used additive that can improve protein solubility, thereby preventing protein aggregation and precipitation during focusing.

[0074] In some embodiments, the iCIEF method provided herein uses H3PO4 as the anolyte. In some embodiments, the CIEF method provided herein uses NaOH as the catholyte.

[0075] In some implementations, the iCIEF method described herein is performed using the iCE280 or iCE3 system (ProteinSimple, Santa Clara, CA), which operates in a capillary column and utilizes a full-column UV absorption detector to detect focused protein bands, avoiding interference with these focused bands. Both full-column detection and combinations of full-column detection / single pI detection are applicable to the methods described herein. Most CIEF applications are performed using commercial capillary electrophoresis (CE) instruments. These instruments are equipped with 20–60 cm long capillaries and on-column UV absorption detectors. When performing CIEF using these instruments, all protein bands separated by the focusing process move through the detection point of the on-column detector located at one end of the capillary.

[0076] Since the migration time through the capillary is directly related to the charge (pI) of the analyte, the UV signal from a point in the capillary that changes over time can be represented as a UV trace, which represents the different charges (pI) of the sample components. In an exemplary embodiment, the UV trace generated by CIEF represents the charge variants of the protein of interest, where each UV peak represents an important charge variant.

[0077] In some embodiments, the absorbance of the sample is measured at an ultraviolet wavelength. In some embodiments, the absorbance is measured over a range of pH values, such as about 2 to about 12, or about 2 to about 11, or about 3 to about 10. In this regard, multiple absorbance values ​​can be obtained over a certain pI range, rather than a single reading at a single pI value.

[0078] Amphoteric electrolytes are used to generate pH gradients. An amphoteric electrolyte is a molecule that simultaneously has at least one basic group and at least one acidic group. Amphoteric electrolytes are typically mixtures of thousands of amphoteric compounds and are used to form stable pH gradients in isoelectric focusing. This invention is not limited to the type of amphoteric electrolyte. For example, Servalyts™ (Serva), Biolytes (Bio-Rad), Zoom® (Invitrogen), Ampholines™ (GE), and Pharmalytes® are all suitable for the methods disclosed herein. Amphoteric electrolytes can be obtained based on the pH range of the composition. Therefore, depending on the desired pH gradient, it may be more preferable to choose one type of amphoteric electrolyte over another, for example, a amphoteric electrolyte with a pH range of 3.5–9.5 versus a amphoteric electrolyte with a pH range of 2–11.

[0079] In some embodiments, the loading composition contains about 3% to about 10% of an amphoteric electrolyte at pH 3-10. In one embodiment, the sample contains about 4% of an amphoteric electrolyte at pH 3-10. In yet another embodiment, the sample contains about 2% of an amphoteric electrolyte at pH 8-10.5. In another embodiment, the sample contains about 5% to about 10% of an amphoteric electrolyte at pH 2-9. In yet another embodiment, the sample contains about 8% of an amphoteric electrolyte at pH 2-9. Depending on the specific sample, those skilled in the art will readily know which load amphoteric electrolyte to use. For example, given that the two half-antibodies of BsAb have significantly different pIs, it may be desirable to focus the sample over a wider pH range.

[0080] A sample mixture (e.g., a mixture containing a protein sample, a carrier amphoteric electrolyte, additives, and a pI marker) is injected to fill the entire capillary column. A separation voltage is applied to the anolyte and catholyte cells. In some embodiments, separation is performed at conditions of about 100 V / cm to about 1000 V / cm, or about 100 V / cm to about 900 V / cm, or about 100 V / cm to about 800 V / cm, or about 100 V / cm to about 700 V / cm, or about 100 V / cm to about 600 V / cm, or about 100 V / cm to about 500 V / cm. In one embodiment, separation is performed at conditions of about 400 V / cm, or about 500 V / cm, or about 600 V / cm, or about 700 V / cm. In yet another embodiment, separation is performed at conditions of about 600 V / cm.

[0081] Under this voltage, a pH gradient is created within the column. Proteins are separated and focused along the capillary column. The IEF process can be monitored online within the separation column using a whole-column detector, and in one implementation, the focusing time is optimized for a single sample run. At the end of the focusing process, all focused protein bands within the column are recorded by the detector without interfering with the separation resolution. Finally, the column is washed and prepared for the next injection.

[0082] The sample for the target BsAb can be obtained from any step of a biological process, such as cell culture medium (CCF), harvested cell culture medium (HCCF), any step in downstream processing, active pharmaceutical ingredient (DS), or finished product (DP) containing the final formulation. In some specific exemplary embodiments, the sample can be selected from any step in a downstream process such as clarification, chromatographic production, virus inactivation, or filtration. In some specific exemplary embodiments, the pharmaceutical product can be selected from pharmaceutical products prepared during clinical, transport, storage, or disposal processes.

[0083] Target bispecific antibody

[0084] In one aspect, this disclosure provides a method for detecting byproducts of specific types of BsAbs. This method is carried out against an asymmetric bispecific antibody comprising two half-antibodies with different pIs. In some embodiments, the difference in pI is introduced by replacing a constant domain of one Fab with a constant TCR domain, known to have a lower pI than the commonly used CH1 and CL domains. Introducing the constant TCR domain also facilitates ensuring homologous heavy-light chain pairing. Antibodies with chimeric constant TCR domains constructed using the WuXiBody® platform are particularly well-suited for this method. Examples of WuXiBody-type antibodies can be found in WO2019057122, WO2019057124, WO2020057610, and WO2022 / 156687, all of which are incorporated herein by reference.

[0085] In some implementations, the target BsAb comprises a chimeric Fab whose CH1 and CL domains are replaced with a pair of T-cell receptor (TCR) constant domains, and these TCR constant domains are capable of forming one or more non-natural interchain disulfide bonds. These disulfide bonds stabilize the dimer formed between the pair of TCR constant domains.

[0086] In some embodiments, the target BsAb comprises first and second antigen-binding portions. The first antigen-binding portion is a chimeric Fab containing a first heavy chain variable domain (VH1) operably linked to a first T cell receptor (TCR) constant region (C1) and a first light chain variable domain (VL1) operably linked to a second TCR constant region (C2). The second antigen-binding portion may be in the form of Fab, scFv, VHH, etc. In some embodiments, the second antigen-binding portion is also in the form of Fab and contains a second VH (VH2) operably linked to an antibody heavy chain CH1 domain and a second VL (VL2) operably linked to an antibody light chain constant (CL) domain. In other words, in the first antigen-binding portion, the normally present CH1 and CL domains are replaced with a pair of TCR constant regions, and thus the native disulfide bond between the CH1 and CL domains is replaced with one or more engineered non-natural disulfide bonds between the TCR C1 and C2 regions. The positions of the C1 and C2 regions can also be interchanged. In some implementations, region C1 is a Cβ domain and region C2 is a Cα domain. Alternatively, region C1 is a Cα domain and region C2 is a Cβ domain.

[0087] The target BsAb used in the method described herein can take various forms. In some exemplary embodiments, the target BsAb includes: (a) A first heavy chain containing structural domains operably connected in the manner of VH1-C1-Fc, a second heavy chain containing structural domains operably connected in the manner of VH2-CH1-Fc, a first light chain containing structural domains operably connected in the manner of VL1-C2, and a second light chain containing structural domains operably connected in the manner of VL2-CL. (b) A first heavy chain comprising structural domains operably connected in a VH1-C1-Fc manner, a second heavy chain comprising structural domains operably connected in a scFv-Fc (VH2-VL2-Fc or VL2-VH2-Fc) manner, and a light chain associated with the first heavy chain and comprising structural domains operably connected in a VL1-C2 manner; or (c) A first heavy chain comprising structural domains operably connected in a VH1-C1-Fc manner, a second heavy chain comprising structural domains operably connected in a VHH-Fc manner, and a light chain associated with the first heavy chain and comprising structural domains operably connected in a VL1-C2 manner.

[0088] In chimeric Fabs, the TCR Cα and Cβ constant domains can associate with each other via one or more non-natural interchain disulfide bonds. There are two distinct variants of the human TCR β-chain constant region, designated TRBC1 and TRBC2 (IMGT nomenclature). In some embodiments, the sequence of the TCR Cβ domain in the target BsAb is based on the following wild-type TCR sequence: LEDLKNVFPPKVAVFEPSEAEISHTQKATLVCLATGFYPDHVELSWWVNGKEVHSGVSTDPQPLKEQPALNDSRYCLSSRLRVSATFWQNPRNHFRCQVQFYGLSENDEWTQDRAKPVTQIVSAEAWGR (SEQ ID NO: 5), its NCBI registration number is A0A5B9 (https: / / www.uniprot.org / uniprot / A0A5B9).

[0089] The constant region of the human TCR α chain is called the TRAC, and its NCBI accession number is P01848 (https: / / www.uniprot.org / uniprot / P01848). In some implementations, the sequence of the TCR Cα domain in the target BsAb is based on the following wild-type TCR sequence: PDIQNPDPAVYQLRDSKSSDKSVCLFTDFDSQTNVSQSKDSDVYITDKTVLDMRSMDFKSNSAVAWSNKSDFACANAFNNSIIPEDTFFPSPESS (SEQ ID NO: 7).

[0090] In some specific implementations, the pair of TCR constant regions comprises an engineered TCR Cβ domain containing one or more mutation sites, as shown below: LEDLKNVFPPEVAVFEPSEAEISHTQKATLVCLATGFYPDHVELSWWVNGKEVHSGVCTDPQPLKEQPALQDSRYALSSRLRVSATFWQNPRNHFRCQVQFYGLSENDEWTQDRAKPVTQIVSAEAWGR (SEQ ID NO: 6);LEDLKNVFPPEVAVFEPSECEISHTQKATLVCLATGFYPDHVELSWWVNGKEVHSGVCTDPQPLKEQPALQDSRYALSSRLRVSATFWQNPRNHFRCQVQFYGLSENDEWTQDRAKPVTQIVSAEAWGR (SEQ ID NO: 10).

[0091] In some specific implementations, the pair of TCR constant regions comprises an engineered TCR Cα domain containing one or more mutation sites, as shown below: PDIQNPDPAVYQLRDSKSSDKSVCLFTDFDSQTQVSQSKDSDVYITDKCVLDMRSMDFKSNSAVAWSQKSDFACANAFQNSIIPEDTFFPS[PESS] (SEQ ID NO: 8, PESS can be removed); or PDIQNPDCAVYQLRDSKSSDKSVCLFTDFDSQTQVSQSKDSDVYITDKCVLDMRSMDFKSNSAVAWSQKSDFACANAFQNSIIPECTFFPS (SEQ ID NO: 9); PDIQNPDPAVYQLRDSKSSDKSVCLFTDFDSQTQVSQSKDSDVYITDKCVLDMRSMDFKSNSAVAWSQKSDFACANAFQNSIIPEDTFFCS (SEQ ID NO: 11).

[0092] Those skilled in the art will readily understand that BsAbs used in the methods disclosed herein can contain a variety of engineered constant TCR α and β domains, provided they are capable of stabilizing the BsAb and providing pI differential. Specifically, the antibody can contain a constant TCR β domain and a constant TCR α domain, wherein the constant β domain has an amino acid sequence as shown in SEQ ID NO: 5, 6, or 10 or at least 90% (e.g., at least 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%) identical variants thereof, and the constant α domain has an amino acid sequence as shown in SEQ ID NO: 7, 8, 9, or 11 or at least 90% (e.g., at least 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%) identical variants thereof. A limited number of amino acid modifications (e.g., 90% identity) in the constant TCR domain do not alter its pI differential with conventional Fab. Detailed descriptions of TCR constant domain variants that can be used to construct WuXiBody antibody forms can be found in WO2019057122, WO2019057124, WO2020057610, and WO2022 / 156687, which are incorporated herein by reference in their entirety. Highly homologous TCR constant domains will maintain relatively low pI. For example, WO2019057122 describes engineered Cβ containing mutant cysteine ​​residues with substituted amino acid residues selected from the following positions: S56C, S16C, F13C, V12C, E14C, L62C, D58C, S76C, and R78C, and / or engineered Cα containing mutant cysteine ​​residues with substituted amino acid residues selected from the following positions: T49C, Y11C, L13C, S16C, V23C, Y44C, T46C, L51C, and S62C; specifically, engineered Cβ and engineered Cα contain a pair of mutant cysteine ​​residues selected from the following group with a pair of substituted amino acid residues: S16C in Cβ The following pairs of C and Cα contain Y11C, Cβ contains F13C and Cα contains L13C, Cβ contains S16C and Cα contains L13C, Cβ contains V12C and Cα contains S16C, Cβ contains E14C and Cα contains S16C, Cβ contains F13C and Cα contains V23C, Cβ contains L62C and Cα contains Y44C, Cβ contains D58C and Cα contains T46C, Cβ contains S76C and Cα contains T46C, Cβ contains S56C and Cα contains T49C, Cβ contains S56C and Cα contains L51C, Cβ contains S56C and Cα contains S62C, and Cβ contains R78C and Cα contains S62C. (See paragraphs

[0032] and

[0033] ), and also provides specific sequences of engineered Cβ and Cα (see SEQ: 32-48 in Table 20 of WO2019057122).WO2022156687 provides more examples of engineered Cβ and Cα domains, such as Cα and / or Cβ, which may contain one or more mutant residues to form one or more non-natural disulfide bonds, selected from: P8C on Cα, A9C on Cα, V10C on Cα, F26C on Cα, F29C on Cα, T33C on Cα, Q34C on Cα, V35C on Cα, S36C on Cα, S38C on Cα, K39C on Cα, F78C on Cα, N80C on Cα, C The Cα domains include S81C, I82C, P84C, D86C, T87C, F88C, F89C, and P90C, as well as A18C on Cβ. Specific sequences of engineered Cβ and Cα domains are also provided (see Tables 1, 10, 14, and 18 of WO2022156687, especially the sequences T8311-57, T8311-61, W329001-U4T4-78, and W329001-U4T4-79). The BsAb of this paper can employ all of these engineered Cβ and Cα domains.

[0093] Furthermore, to achieve heterodimerization of asymmetric BsAbs, a pestle can be introduced into one heavy chain by replacing a smaller amino acid residue with a larger amino acid residue in the first Fc domain, and a mortar can be introduced into another heavy chain by replacing a larger residue with a smaller residue. Detailed information on pestle-mortar mutation sites can be found in Ridgway et al., 1996, Spiess et al., 2015, and Brinkmann et al., 2017. The distribution of pestle and mortar in the two heavy chains is not fixed; for example, a pestle can be introduced into the Fc of a heavy chain containing a TCR constant domain, and a mortar can be introduced into the Fc of a heavy chain containing a TCR constant domain, or vice versa. In some embodiments, a pestle is introduced into the heavy chain containing a TCR β constant domain. In some embodiments, the positions of the TCR β constant domain and the α constant domain are interchanged, and a pestle is introduced into the heavy chain containing a TCR α constant domain. Alternatively, a mortar can be introduced into the heavy chain containing either the TCR β constant domain or the α constant domain.

[0094] The methods described herein can be implemented using BsAbs that have binding specificity to any suitable antigen or epitope, for example, exogenous antigens, endogenous antigens, autoantigens, neoantigens, viral antigens, or tumor antigens, such as tumor-associated antigens (TAAs) or infectious disease-associated antigens, immune checkpoint molecules, tumor microenvironment targets, and targets associated with autoimmune and inflammatory diseases. The term "tumor-associated antigen" refers to an antigen presented on or capable of being presented on the surface of tumor cells and located on or within tumor cells. Illustrative examples of tumor-associated surface antigens are described in PCT / CN2018 / 106766 (WO 2019 / 057122), and are incorporated herein by reference. In some embodiments, a first antigen-binding portion has first specificity to a first TAA antigen, and a second antigen-binding portion has second specificity to a second TAA antigen. In some embodiments, a first antigen-binding portion has first specificity to a TAA antigen, and a second antigen-binding portion has second specificity to a T-cell-associated antigen.

[0095] For example, the BsAb disclosed herein can target any of the following antigens or epitopes, including but not limited to CD3, 4.1BB (CD137), OX40 (CD134), CD16, CD47, CD19, CD20, CD22, CD33, CD38, CD123, CD133, CEA, cdH3, EpCAM, epidermal growth factor receptor (EGFR), EGFRvIII (mutated form of EGFR), HER2, HER3, dLL3, BCMA, sialylated Lea, 5T4, ROR1, melanoma-associated chondroitin sulfate proteoglycan, mesothelin, folate receptor 1, VEGF receptor, EpCAM, HER2 / neu, HER3 / neu, G250, CEA, MAGE, proteoglycan, VEGF, FGFR, αVβ3-integrin, HLA, HLA-DR, ASC, CD1, CD2, CD4, CD5, CD6, CD7. CD8, CD11, CD13, CD14, CD21, CD23, CD24, CD28, CD30, CD37, CD40, CD41, CD44, CD52, CD64, c-erb-2, CALLA, MHCII, CD44v3, CD44v6, p97, gangliosides GM1, GM2, GM3, GD1a, GD1b, GD2, GD3, GT1b, GT3, GQ1, NY-ESO-1, NFX2, SSX2, SSX4 Trp2, gp100, tyrosinase, Muc-1, telomerase, survivin, G250, p53, CA125, MUC, Wu antigen, Lewis Y antigen, HSP-27, HSP-70, HSP-72, HSP-90, Pgp, MCSP, EpHA2, cell surface targets GC182, GT468 or GT512, IL-17, IL-20, IL-13 and IL-4.

[0096] The first and second antigen-binding moieties of the target BsAb used in the methods described herein can be derived from first and second parental antibodies with different antigen-binding specificities. In some embodiments, the first antigen-binding moieties are chimeric Fabs containing VH and VL domains derived from the first parental antibody and a pair of constant TCR domains combined therewith, while the second antigen-binding moieties are non-chimeric Fabs or scFvs containing VH and VL domains derived from the second parental antibody. The first and second parental antibodies can be developed antibodies or novel antibodies. The terms "derived from / derived from / derived from" mean that the CDR, VH, and VL are highly homologous to or identical to the CDR, VH, and VL of the parental antibody.

[0097] Mutations in the Fc region that promote heterodimerization

[0098] In addition to the two antigen-binding regions, the AsBsAb described in this paper may also contain a pair of dimerizing domains that associate with each other to anchor the two heavy chains together. The most commonly used dimerizing domain is the Fc region, such as the Fc region of human IgG. The IgG Fc region can belong to any isotype, including but not limited to IgG1, IgG2, IgG3, or IgG4. Due to the introduction of the TCR constant domain, the AsBsAb contains two distinct heavy chains. Different rational design strategies can be employed for the Fc region to promote the assembly of the two distinct heavy chains and reduce the generation of byproducts.

[0099] As those skilled in the art will recognize, mutants used to promote heterodimerization in the Fc region may include stereo variants (e.g., "mortar and pestle" or "skewed" variants and "charge pair" variants) and "pI variants" that allow the purification of homodimers from heterodimers. As outlined in WO2014 / 145806 (the entire contents of which are incorporated herein by reference), useful mechanisms for heterodimerization include "mortar and pestle" ("KIH"), sometimes referred to herein as "skewed" variants, "electrostatic orientation" or "charge pair" as described in WO2014 / 145806, as well as pI variants as summarized in WO2014 / 145806 and other general Fc variants.

[0100] In some respects, the asymmetric BsAb used in the methods described herein further incorporates mutations in the Fc region (optionally including the hinge region) to promote heterodimerization. A typical example of such mutations is the mortising structure in the Fc region, and more specifically, the mortising structure at the CH3 interface. The “mortising” can be obtained by replacing a smaller amino acid with a larger amino acid (such as the bulkier W) in the CH3 domain of one heavy chain, and the “mortising” can be constructed by replacing a larger residue with a smaller residue (such as A or T) in the CH3 domain of another heavy chain.

[0101] In some embodiments, the pestle structure contains the T366W mutation, and the mortar structure contains the T366S, L368A, and Y407V mutations. In some embodiments, the pestle structure contains the T366Y mutation, and the mortar structure contains the Y407T mutation. In some embodiments, the pestle structure contains the S354C and / or T366W mutation, and the mortar structure contains the Y349C, T366S, L368A, and / or Y407V mutation.

[0102] Fc region mutations are not limited to the club and mortar structures; other different CH3 positions that promote Fc heterodimerization have been identified, such as S364H, Y349T, and T394F. Another heavy-chain heterodimerization technique involves using opposite electrostatic charges, for example, introducing two positively charged lysine residues on one chain and two negatively charged aspartic residues on another. In some embodiments, one chain of the Fc region contains the D399K and E356K mutations, while the other chain (preferably the chain containing the hemimorph of the TCR constant domain) contains the K409D and K392D mutations. Furthermore, amino acid changes can be located not only in the CH3 domain but also in the hinge region.

[0103] In another heavy chain heterodimerization method called the SEED (Separately Engineered Chain Exchange) technique, more substantial changes can be introduced into the Fc region by engineering alternating IgG and IgA segments within the CH3 domain. This can produce two distinct antiparallel chains, named GA and AG, which form an asymmetric heterodimerization interface.

[0104] When referring to residues in the constant region of the immunoglobulin heavy chain, the “EU numbering system” or “EU index” is commonly used (e.g., Kabat et al., EU index as reported above). “According to Kabat’s EU numbering” or “according to Kabat’s EU index” refers to the residue numbering of human IgG1 EU antibodies. Unless otherwise stated herein, references to residue numbers in the constant domain of antibodies mean residue numbers obtained through the EU numbering system.

[0105] Accuracy, linearity, and repeatability of the method

[0106] Accuracy refers to the degree to which a result measured by a certain method closely approximates the true or reference value. Accuracy can be expressed as recovery rate (%). To test whether the method described in this paper can obtain accurate results for byproduct content levels, the recoveries of samples containing high, medium, and low concentrations of K or H type byproducts were tested. The results (Table 2) show that at least 80% recovery was maintained for all tested concentrations of K type byproducts. For H type byproducts, the results (Table 3) show that at concentrations not less than 4%, at least 90% recovery was maintained, while at lower concentrations, at least 60% recovery was maintained.

[0107] The study also verified whether and to what extent the detection method of the present invention is affected by the concentration of byproducts. Linearity, as assessed herein, refers to the degree to which the spiked concentration of byproducts is proportional to the measured concentration within the design range. Correlation coefficient R0 2The larger the value, the stronger the correlation between the two variables. The range refers to the relationship between the upper and lower limits of the concentration or amount applicable to the method, and the range for validation can be determined before starting experimental studies.

[0108] Repeatability can be assessed by performing multiple sampling measurements under different laboratory conditions (such as different instruments, operators, analytical batches, and times), and the results can be expressed as relative standard deviation (RSD). As demonstrated in the examples, the detection results of the method exhibit high consistency in repeated trials.

[0109] Beneficial effects of the method

[0110] 1. This invention provides a method for monitoring and detecting the content of mismatch byproducts in BsAb protein solutions using iCIEF.

[0111] 2. The present invention provides a quantitative method for measuring mismatch byproducts with high sensitivity.

[0112] 3. This invention provides a potential platform method for detecting the content of mismatch byproducts in AsBsAb samples of a class of half-antibodies paired with significant pI differences.

[0113] Example

[0114] The general description of this disclosure will be more readily understood by referring to the following embodiments, which are provided by way of example and are not intended to limit the disclosure. The embodiments are not intended to represent all or only the experiments conducted.

[0115] Example 1: Separating the target BsAb from H-related and K-related categories using iCIEF

[0116] 1.1 Materials

[0117] Anode electrolyte solution, cathode electrolyte solution, methylcellulose solution (0.5% and 1%), pI markers (4.65 and 9.22), and iCE3 were purchased from ProteinSimple (Silicon Valley, CA, USA). Pharmalyte 3-10 was obtained from Cytiva (Washington, USA), and urea powder was obtained from Sigma (St. Louis, MO, USA). CaptureSelect CH1-XL resin was purchased from Thermo Fisher Scientific (Waltham, MA, USA). Praesto Q65 was purchased from Purolite (Wales, UK). BsAb and corresponding H and K-related species were expressed in CHO-K1 cells grown in HyCloneActiPro medium supplemented with Cell Boost 7a and 7b. A schematic diagram of AsBsAb and its mismatch byproducts is shown below. Figure 2 The AsBsAb used in the examples is a bispecific antibody disclosed in WO2023179443, whose heavy and light chain sequences are listed in SEQ ID NO: 1-4.

[0118] Table A: Amino acid sequences of exemplary AsBsAbs used in the examples

[0119] 1.2 iCIEF Parameters

[0120] Samples from the BsAb production step were subjected to capillary isoelectric focusing (CIEF). CIEF separates protein sample components based on charge (pi), including separating charged variants of the protein of interest. UV traces of the sample components passing through the CIEF capillary were generated, where local maxima of protein concentration were considered "peaks" and corresponded to BsAb and byproducts.

[0121] Table 1. Final experimental parameters of iCIEF

[0122] 1.3 iCIEF can monitor mismatch byproducts of AsBsAbs with half-antibody pairs exhibiting significant pI differences.

[0123] KiH technology was introduced into the Fc region to promote HC-HC heterodimerization. TCRs were introduced into the club arm to promote homologous HC-LC pairing. Representative byproducts during cell culture production included club-club (KK) homodimers, mortar-mortar (HH) homodimers, club (K) hemimers, and mortar (H) hemimers. Notably, the theoretical pI was the same between the KK homodimer and the K hemimer, as was the case with the HH homodimer and the H hemimer. The theoretical pI for the target BsAb was 6.46. As previously mentioned, the pI of the TCR constant domain is relatively low. Therefore, the theoretical pI of the resulting KK homodimer (containing two TCRs) and the K hemimer is low, at 6.06; while the theoretical pI of the resulting HH homodimer (without TCRs) and the H hemimer is high, at 7.25.

[0124] The unique pI difference introduced by the TCR region for this target heterodimer and mismatched byproduct provides a powerful catalyst for the development of charge-based analytical methods. To validate the iCIEF method, highly purified intermediate samples of K-related (KK homodimers and / or K-type hemimers) and H-related (HH homodimers and / or H-type hemimers) species are required. K-related byproducts were enriched using CaptureSelect CH1-XL resin that binds to the IgG CH1 domain. Because the CH1 and CL domains of the K-related species are replaced by TCR α and β domains, the K-related species cannot bind to the CaptureSelect CH1-XL resin. Therefore, the K-related species can be enriched in the flow-through fraction during the CaptureSelect CH1-XL affinity chromatography step.

[0125] Furthermore, the theoretical pI for H-related species was 7.25, which was 0.79 units higher than that of the target BsAb (6.46). Anion exchange chromatography was performed at a loading pH of 7.0 to enrich H-related species in the flow fraction.

[0126] Then, iCIEF analysis was performed on the target AsBsAb, the enriched K-related types, the enriched H-related types, and the target BsAb stress test sample (BsAb DS placed at 50℃ for 14 days). Figure 3 As shown, the experimental pI distribution of these samples is consistent with the theoretical pI. Furthermore, electrophoretic patterns demonstrate that iCIEF can clearly separate the target BsAb from H-related and K-related species. These data warrant further development of iCIEF, making it a promising platform analytical method for monitoring mismatch byproducts of AsBsAb.

[0127] 1.4 Evaluation of CIEF as a method for quantifying mismatch byproducts

[0128] Subsequently, a limit of quantitation (LOQ) experiment was conducted to evaluate the sensitivity of the method. Different percentages (w / w%) of pestle / mortar related species were added to the purified drug substance solution (DS), and spiked recoveries were calculated to assess the accuracy and linearity of the method. Acceptable criteria for spiked recoveries were set at 60% to 140%, and the relative standard deviation (RSD) at different spiking levels should be within 20%. The lowest percentage of spikes meeting these acceptable criteria was designated as the LOQ. Tables 2 and 3 summarize the method validation results, and Figure 4 Linearity plots are shown. All recoveries were within the preset limits for K- and H-related species. K-related species achieved excellent linearity in the range of 4–20% (R0). 2 = 0.9919); H-related categories showed good linearity in the range of 1–20% (R 2 = 0.9805). Based on these results, the LOQ for K-related impurities was determined to be 4%; and the LOQ for H-related impurities was determined to be 1%.

[0129] Table 2. Summary of iCIEF Method Certification Results for K-Related Categories

[0130] Table 3. Summary of iCIEF method certification results for H-related categories

[0131] Example 2: iCEIF as a release and intermediate process testing method to monitor mismatch byproducts

[0132] The certified method was then used as an intermediate process test and batch release assay. Figure 5 The downstream process for the target BsAb is shown. Intermediate process samples and raw material solutions were collected and subjected to iCIEF analysis. The test results are summarized in Table 4. No H-related species were detected after affinity chromatography. K-related species were gradually removed and were not detected after the mixed-mode chromatography step. Overall, the downstream process was able to remove all mismatch byproducts, and iCIEF as a method for measuring mismatch byproducts of AsBsAb was successfully implemented.

[0133] Table 4. iCIEF results of purified process samples

[0134] a BLOQ: Below the limit of quantitation

[0135] b ND: Not detected

[0136] Following this case study, iCIEF was successfully developed as an analytical mismatch testing method for several other AsBsAbs that exhibit different pI differences between their mismatch byproducts and the target BsAb. iCIEF holds promise as a platform-based mismatch testing method for such AsBsAbs. This platform strategy significantly reduces the time and effort required to develop such AsBsAbs.

[0137] References

[0138] [1]. Spiess C, Zhai Q, Carter PJ. Alternative molecular formats and therapeutic applications for bispecific antibodies. Mol Immunol. 2015;67:95-106.

[0139] [2].Ma J, Mo Y, Tang M, Shen J, Qi Y, Zhao W et al. BispecificAntibodies: From Research to Clinical Application. Front Immunol. 2021;12:626616.

[0140] [3]. Ridgway JB, Presta LG, Carter P. 'Knobs-into-holes' engineering of antibody CH3 domains for heavy chain heterodimerization. Protein Eng. 1996;9:617-21.

[0141] [4].Merchant AM, Zhu Z, Yuan JQ, Goddard A, Adams CW, Presta LG,Carter P. An efficient route to human bispecific IgG. Nat Biotechnol. 1998;16:677-81.

[0142] [5]. Gramer MJ, van den Bremer ET, van Kampen MD, Kundu A, Kopfmann P, Etter E, et al. Production of stable bispecific IgG1 by controlled Fab-arm exchange: scalability from bench to large-scale manufacturing by application of standard approaches. MAbs. 2013;5:962-73.

[0143] [6]. Carter P. Bispecific human IgG by design. J Immunol Methods. 2001;248:7-15.

[0144] [7]. Brinkmann U, Kontermann RE. The making of bispecific antibodies. MAbs. 2017;9:182-212.

[0145] [8]. Chen SW, Zhang W. Current trends and challenges in the downstream purification of bispecific antibodies. Antib Ther. 2021;4:73-88.

[0146] [9]. Wang C, Vemulapalli B, Cao M, Gadre D, Wang J, Hunter A, et al. A systematic approach for analysis and characterization of mispairing in bispecific antibodies with asymmetric architecture. MAbs. 2018;10:1226-35.

[0147]

[10] .Yan Y, Xing T, Wang S, Daly TJ, Li N. Coupling Mixed-Mode SizeExclusion Chromatography with Native Mass Spectrometry for SensitiveDetection and Quantitation of Homodimer Impurities in Bispecific IgG.Analytical Chemistry. 2019;91:11417-24。

Claims

1. A method for detecting byproducts in a target asymmetric bispecific antibody (AsBsAb) sample, comprising capillary isoelectric focusing of the sample to separate the byproducts from the target AsBsAb based on a pI difference between the byproducts and the target AsBsAb, wherein the AsBsAb comprises two halves, the first halves consisting of a first heavy chain and an associated light chain, and the pI of the first halves being different from that of the second halves.

2. The method of claim 1, wherein the first half antibody comprises a first antigen-binding portion, the first antigen-binding portion being a chimeric Fab in which the CH1 and CL constant domains are replaced by a pair of TCR constant domains.

3. The method of claim 1 or 2, wherein the second half-antibody comprises a second antigen-binding moiety selected from Fab, scFv and VHH.

4. The method of claim 2, wherein the TCR constant domain is a TCRα and a TCRβ constant domain, and the TCRα and TCRβ constant domains are natural or engineered to introduce one or more disulfide bonds capable of stabilizing the first half-antibody.

5. The method of claim 4, wherein the TCRβ constant domain comprises an amino acid sequence having at least 90%, 95%, or 99% identity with SEQ ID NO: 5, 6, or 10.

6. The method of claim 4 or 5, wherein the TCRα constant domain comprises an amino acid sequence having at least 90%, 95%, or 99% identity with SEQ ID NO: 7, 8, 9, or 11.

7. The method according to any one of claims 2-6, wherein: The chimeric Fab includes a TCRβ constant domain in the heavy chain instead of the CH1 domain, and a TCRα constant domain in the light chain instead of the CL domain; or The chimeric Fab includes a TCRα constant domain in the heavy chain instead of the CH1 domain, and a TCRβ constant domain in the light chain instead of the CL domain.

8. The method of any one of claims 3-7, wherein the AsBsAb further comprises a pair-dimerizing domain fused to the C-terminus of the first antigen-binding portion and the second antigen-binding portion.

9. The method of claim 8, wherein the pair-dimerizing domain is an Fc domain of an IgG1, IgG2, IgG3 or IgG4 Fc isotype.

10. The method of claim 9, wherein the AsBsAb comprises two different heavy chains, the two heavy chains having a pestle and a mortar structure respectively in their Fc structural domain.

11. The method of claim 10, wherein the heavy chain of the first half-antibody comprises a club structure and the heavy chain of the second half-antibody comprises a mortar structure, or the heavy chain of the first half-antibody comprises a mortar structure and the heavy chain of the second half-antibody comprises a club structure.

12. The method of claim 11, wherein the pestle structure comprises the S354C and / or T366W mutation according to EU number, and the mortar structure comprises the Y349C, T366S, L368A and / or Y407V mutation according to EU number.

13. The method of any one of claims 1-12, wherein the pI of the first half-antibody differs from the pI of the second half-antibody by at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 1.8, at least 2.0, at least 3.0, or at least 4.0 or more.

14. The method of any one of claims 11-13, wherein the first half-antibody comprises a club-shaped structure and the second half-antibody comprises a mortar-shaped structure, and the byproduct comprises an AsBsAb K-type half-mer and a homodimer composed of the first half-antibody, and an AsBsAb H-type half-mer and a homodimer composed of the second half-antibody.

15. The method of claim 14, wherein the pI of the K-type semimer is at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 1.8, at least 2.0, at least 3.0, or at least 4.0 or more lower than the pI of the H-type semimer.

16. The method of any one of claims 14-15, wherein the pI of the K-type semimer is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 1.0, or at least 2.0 or more lower than the pI of the target AsBsAb, and the pI of the target AsBsAb is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 1.0, or at least 2.0 or more lower than the pI of the H-type semimer.

17. The method of any one of claims 11-13, wherein the first half-antibody comprises a mortar structure and the second half-antibody comprises a pestle structure, and the byproduct comprises an AsBsAb H-type half-mer and a homodimer composed of the first half-antibody, and an AsBsAb K-type half-mer and a homodimer composed of the second half-antibody.

18. The method of claim 17, wherein the pI of the H-type semimer is at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0, at least 1.1, at least 1.2, at least 1.5, at least 1.8, at least 2.0, at least 3.0, or at least 4.0 or more lower than the pI of the K-type semimer.

19. The method of any one of claims 17-18, wherein the pI of the H-type semimer is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 1.0, or at least 2.0 or more lower than the pI of the target AsBsAb, and the pI of the target AsBsAb is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 1.0, or at least 2.0 or more lower than the pI of the K-type semimer.

20. The method of any one of claims 1-19, further comprising quantifying the byproducts, for example determining the content levels of K-type semimers and their homodimers in the sample, and / or the content levels of H-type semimers and their homodimers.

21. The method of any one of claims 1-20, wherein the content level of K-type semimer and its homodimer in the sample ranges from 4% to 20%.

22. The method of any one of claims 1-21, wherein the content level of H-type semimer and its homodimer in the sample ranges from 1% to 20%.

23. The method of any one of claims 1-22, wherein the sample is obtained from a bioprocess for producing the target AsBsAb, including clarification, chromatography, virus inactivation and filtration, and optionally, the sample has been diluted prior to capillary isoelectric focusing.

24. The method of any one of claims 1-23, wherein the sample is obtained from cell culture medium, harvested cell culture medium, eluent of affinity chromatography, anion exchange chromatography or mixed-mode chromatography, filtrate, ultrafiltrate, percolate, drug substance solution (DS) or pharmaceutical product (DP) containing the final formulation.

25. The method of any one of claims 1-24, wherein the capillary isoelectric focusing is imaging capillary isoelectric focusing, optionally coupled with chromatography or mass spectrometry.

26. A method for assessing the heterogeneity of a sample obtained from a production process of a target AsBsAb, comprising imaging capillary isoelectric focusing of the sample to separate the byproduct from the target AsBsAb based on a pI difference between the byproduct and the target AsBsAb, wherein the AsBsAb comprises two haptens, the first hapten consisting of a first heavy chain and an associated light chain, and the pI of the first hapten differing from the pI of the second hapten by at least 0.2.

Images