KR20240015003A - A conjugate, composition and method for noncovalent antibody catenation that increases antigen-binding avidity of an antibody in proportion to the density of a target antigen - Google Patents

Abstract

A conjugate in which a protein capable of forming a homodimer is fused to the C-terminus of the heavy or light chain of an antibody or fragment thereof, a composition for enhancing antibody-antigen binding comprising the same, and a method for enhancing antibody-antigen binding using the same. It's about. Additionally, it relates to treatment or diagnosis of diseases using the conjugate.

Description

{A conjugate, composition and method for noncovalent antibody catenation that increases antigen-binding avidity of an antibody in proportion to the density of the target antigen density of a target antigen}

A conjugate in which a protein capable of forming a homodimer is fused to the C-terminus of the heavy or light chain of an antibody or fragment thereof, a composition for enhancing antibody-antigen binding comprising the same, and a method for enhancing antibody-antigen binding using the same. It's about. Additionally, it relates to treatment or diagnosis of diseases using the conjugate.

Immunoglobulin G (IgG) antibodies are biological agents widely used in diagnosis and treatment. IgG antibodies have two full-length light chains and two full-length heavy chains, and specifically form a heterodimer consisting of one copy of the heavy chain (~50 kDa) and one copy of the light chain (~25 kDa). , It is a homodimer in which two heterodimers are arranged in a Y shape. The C-terminal domains of the two heavy chains constitute the Fc (fragment crystallizable) region in the form of a homodimer, and the N-terminal domains of the two heavy chains each form a Fab (antigen) arm that forms two arms in a Y shape together with the light chain. -binding fragment) Configures the area. Fab contains a complementarity determining region (CDR), which is the region that binds to the antigen, and this antibody-antigen binding prevents the antigen from binding to its cognate partner or removes the antigen molecule from the cell surface by receptor-mediated endocytosis. You can. Fc enables a long half-life through binding to FcRn (neonatal Fc receptor) and exerts its effector function through binding to C1q of complement factor C1 or the Fcγ receptor on effector immune cells, causing the cell to which the antibody molecule is bound to be killed. You can. As such, IgG antibodies have desirable properties for use as diagnostic or therapeutic agents, such as high specificity for target antigens, low immunogenicity, and long serum half-life, so most diagnostic or therapeutic antibodies are in the form of IgG.

In general, antibodies for diagnostic and therapeutic use must exhibit high antigen binding affinity ( K _D < 10 nM) at low nanomolar levels. However, monoclonal antibodies (mAbs) developed for therapeutic use often show limited, moderate effectiveness due to insufficient blocking of target antigens for various reasons, including insufficient antigen binding affinity.

Therefore, the development of technology to increase the level of antigen binding affinity of antibodies is required.

US Patent Publication US 2016-0115214 A1 (2016.04.28)

The present invention fuses a protein capable of forming a homodimer to the two C-termini of the heavy chain or the two C-termini of the light chain of an IgG type antibody or a fragment thereof, to form an antigen of the antibody on the target surface where the antigen is present. This relates to a technology that can dramatically increase the actual binding affinity to , compared to the intrinsic binding affinity.

Accordingly, an example of the present application provides a conjugate in which a protein capable of forming a homodimer is fused to the two heavy chain C-termini or the two light chain C-termini of an antibody or fragment thereof.

Another example of the present application provides a composition for enhancing antibody-antigen binding, including the conjugate.

Another example of the present application provides a method of enhancing antibody-antigen binding, including the step of treating the conjugate on a target surface where antigens to which the antibody included in the conjugate specifically binds exist.

Another example herein provides a biosensor for a sandwich immunoassay, comprising a conjugate as a secondary antibody in the sandwich immunoassay.

Another example of the present application provides a composition for treating or diagnosing a disease, including the conjugate.

Another example of the present application provides a method of treating or diagnosing a disease using the conjugate.

survey

Considering the unique dimeric structure of IgG, the present inventors proposed a homodimer forming protein (herein referred to as a 'catenator') to each of the two C-terminal domains of the heavy chain or to each of the two C-terminal domains of the light chain. ), thereby significantly improving the total antigen-binding avidity of the antibody by inducing a reversible linkage (also called 'catenation' herein) between antibody molecules on a surface rich in target antigen molecules. This improvement in binding affinity arises from the “proximity effect,” in which binding of one subunit of a dimer to a target limits the search space for the target of other subunits.

More specifically, due to the overall dimeric structure of IgG, these antibodies, which have a homodimer-forming protein (catenator) fused to each of the two C-terminal domains of the heavy chain or to each of the two C-terminal domains of the light chain, have one When a homodimer is formed between two antibody molecules rather than within the antibody molecule, they can be linked in an arm-in-arm fashion. Accordingly, in the present invention, if a catheter with very low homodimerization affinity is fused to the two C-termini of the heavy or light chain of an antibody, the catheter remains in a monomeric state in solution. It was assumed that on the cell surface rich in target antigen molecules, catenators would form homodimers due to the proximity effect, forming catenations (links) between fused antibody molecules. The reason why homodimers do not form between catheters intramolecularly in an antibody molecule is because the C-terminal domains of the two heavy chains or the C-terminal domains of the two light chains are physically constrained to be far apart from each other (e.g. For example, see Figure 1a left: the C-terminal domains of the two heavy chains are separated by approximately 23 Å), and when fusion of catheters with very low homodimerization affinities occurs, the binding affinity is weak and they remain dissociated even when they meet. Because it does. On the other hand, since there are no such physical restrictions between antibody molecules (intermolecular), the catheters of adjacent antibody molecules can meet to form a homodimer (see right of Figure 1a). In particular, on target surfaces with a large number of antigens, the local concentration of catheter-fused antibodies increases due to the binding between antigens and antibodies, which also increases the concentration of catheters, thereby increasing the formation of homodimers between catheters due to the proximity effect. As a result, antibody molecules on the target surface can be linked together (catenation) in a long chain like a chain holding hands (see Figures 1b and 1c). In other words, catheter-fused antibodies are not linked in a solution that is not a target surface, but can be linked to a target surface containing many antigens. When connected in this way, the rate at which the catheter-fused antibody dissociates from the antigen can be lowered and the effective binding affinity of the antigen binding site can be increased.

As a result of thermodynamic simulations to verify this herein, catheters with very low homodimerization affinities (e.g., dissociation constants of 0.1 μM to 500 μM) increase the total antigen binding capacity from nanomolar levels to picomolar levels. It was confirmed that the level of increase was greatly dependent on the density of the antigen (see Figures 2 to 8). Additionally, in a proof-of-concept experiment in which antigen molecules were immobilized on a biosensor tip, a catheter with weak homodimerization affinity was fused to the C-terminus of four types of antibodies, and the total antigen binding force was at least 210 times higher than the intrinsic binding force. It was confirmed that it was improved by 5,120 times (see FIGS. 9 to 12). Furthermore, as a result of analyzing the characteristics of catenated antibodies using cancer cell lines, the binding ability of the antibody-catenator to antigen increases in proportion to the density of the antigen, and binds selectively to cancer cells that express more antigen. It was confirmed that it has the ability (see FIGS. 13 to 15).

Therefore, antibody catenation induced by this intermolecular homodimerization significantly improves the overall avidity of the antibody to the antigen on the target surface, thereby enabling many antibodies, including the detection of rare biomarkers and targeted anticancer therapy. The present invention was completed by confirming that it can be a simple, powerful, and general approach that can be applied to application fields.

According to one aspect of the invention, comprising (i) an antibody or fragment thereof, and (ii) a catenator polypeptide fused to either the two heavy chain C-termini or the two light chain C-termini, respectively, of the antibody or fragment. Provided is a conjugate, preferably the catenator polypeptide refers to a polypeptide that satisfies the following conditions:

(a) capable of forming homodimers;

(b) the dissociation constant ( K _D ) of homodimer formation is 0.1 μM to 500 μM;

(c) molecular weight is 3 kDa to 20 kDa; and

(d) On the surface where target antigens to which the antibody specifically binds exist, homodimers may be formed between the catheters included in two or more of the conjugate molecules, thereby linking the conjugate molecules to each other.

According to another aspect of the present invention, a composition for enhancing antibody-antigen binding ability is provided, comprising the conjugate, wherein the composition is provided on a surface where target antigens to which the antibody included in the conjugate specifically binds are present. In this, a homodimer is formed between the catheters included in two or more of the conjugate molecules, thereby linking the conjugate molecules to each other, thereby suppressing dissociation of the antibody.

According to another aspect of the present invention, a method is provided for enhancing antibody-antigen binding, comprising the step of treating the conjugate on a target surface where antigens to which the antibody contained in the conjugate specifically binds exist, , the method is characterized in that a homodimer is formed between the catheters included in two or more of the conjugate molecules, thereby linking the conjugate molecules to each other, thereby suppressing dissociation of the antibody.

According to another aspect of the present invention, a biosensor for sandwich immunoassay is provided, comprising the conjugate as a secondary antibody in sandwich immunoassay.

According to another aspect of the present invention, a composition for treating or diagnosing a disease comprising the conjugate is provided.

According to another aspect of the present invention, a method for treating or diagnosing a disease using the conjugate is provided.

Hereinafter, the present invention will be described in more detail.

When the terms “comprise, comprises,” “comprised,” or “comprising” are used in this specification (including the claims), they refer to the described features, integers, steps or elements. It should be interpreted as specifying the presence of an element, but not excluding the presence of one or more other features, integers, steps, elements or groups thereof.

Descriptions of documents, laws, materials, devices, and articles in this specification are included solely for the purpose of providing context for the present invention. They do not suggest or indicate that all or any part of them forms part of the prior art base or that was common general knowledge in the field to which the present invention pertains prior to the priority date of each claim of this application.

Homodimer-forming protein (catheter)

As used herein, the term “homodimer” refers to two identical monomeric proteins interacting to form a complex. On the other hand, the term “heterodimer” means that two different monomeric proteins interact to form a complex.

The term “homodimer forming protein” refers to a monomeric protein capable of forming homodimers, and herein, as they are used to link antibody molecules by forming homodimers, “catenator”; Also called “catenator polypeptide” or “linker”.

A catheter herein is one that satisfies at least the following conditions:

(a) capable of forming homodimers;

(b) the dissociation constant ( K _D ) of homodimer formation is 0.1 μM to 500 μM;

(c) molecular weight is 3 kDa to 20 kDa; and

(d) On the surface where target antigens to which the antibody specifically binds exist, homodimers may be formed between the catheters included in two or more of the conjugate molecules, thereby linking the conjugate molecules to each other.

Condition (a) is that the main technical idea of the present application is to form a long chain-like catenated antibody by forming a homodimer between the catenator molecules fused to each antibody, thereby forming a catenated antibody that is linked to a long chain. The goal is to increase the binding force, and the catheters herein must be able to form homodimers when present in close proximity to each other under physiological conditions.

Condition (b) means that the catheter must have low homodimerization affinity. The catheter herein is fused to the C-terminus of the heavy or light chain of the antibody. Typically, an IgG-type antibody consists of two heavy chains and two light chains, so the antibody has two heavy chain C-termini and two light chain C-termini. Accordingly, the catheters herein may be fused to the C-termini of two heavy chains, respectively, or to the C-termini of two light chains, respectively, so that at least two catheters may be present in one antibody. However, for the purpose of the present application to connect antibody molecules by forming homodimers of catheters, the formation of homodimers between catheters within one antibody molecule (intramolecular) is prevented and between antibody molecules ( Intermolecular) homodimers need to be formed between catheters. Considering that the two C-terminal domains of the heavy chain or the two C-terminal domains of the light chain are physically constrained to be far apart from each other, if the catenators have sufficiently low homodimerization affinities, the physical constraints will allow them to form a single antibody. It can prevent the formation of homodimers between catheters within the molecule. Therefore, although in solution (e.g., blood) the catheters remain monomeric (i.e., remain in the antibody-cathenator fusion state), on the cell surface rich in target antigen molecules, the proximity effect causes the catheters to By forming homodimers, catenation (linkage) can be formed between fused antibody molecules.

Accordingly, if the dissociation constant ( K _D ) of homodimer formation of the catheter is 0.1 μM or more and 500 μM or less, considering the physical distance between the two C-terminal domains of the heavy chain or the C-terminal domains of the two light chains of the antibody The antibody may be considered to have a sufficiently low homodimerization affinity to prevent catenation within the molecule. For example, the dissociation constant ( K _D ) of homodimer formation of a catheter is 0.1 μM or higher, 1 μM or higher, 10 μM or higher, 20 μM or higher, 30 μM or higher, 40 μM or higher, 50 μM or higher, or 60 μM or higher. , may be 70 μM or more, 80 μM or more, 90 μM or more, or 100 μM or more, and may be 500 μM or less, 400 μM or less, 300 μM or less, or 200 μM or less, but is not limited thereto.

The term “affinity” refers to the equilibrium constant for the reversible association of two agents and is expressed as the dissociation constant ( K _D ). Affinity can be measured using any method known in the art. Such methods include, for example, fluorescence activated cell sorting (FACS), surface plasmon resonance (SPR, e.g. Biacore, Proteon), biolayer interferometry (BLI) , e.g., Octet), kinetics exclusion assay (e.g., KinExA), separable beads (e.g., magnetic beads), antigen panning ), and/or ELISA. It is known in the art that the binding affinity of a specific antibody will vary depending on the method used to analyze binding affinity.

Condition (c) means that the size of the catheter is preferably small so that it does not cause immunogenicity and does not substantially affect the physical properties of the antibody. In this respect, if the molecular weight is 3 kDa or more and 20 kDa or less, it can be considered to have a sufficiently small size not to cause immunogenicity or affect the physical properties of the antibody while still functioning as a catheter. For example, the molecular weight of the catheter may be 3 kDa or greater, 4 kDa or greater, 5 kDa or greater, 6 kDa or greater, or 7 kDa or greater, 20 kDa or less, 19 kDa or less, 18 kDa or less, 17 kDa or less, 16 kDa or less. , may be 15 kDa or less, 14 kDa or less, 13 kDa or less, 12 kDa or less, 11 kDa or less, or 10 kDa or less, but is not limited thereto.

Condition (d) is that on the surface where the target antigens to which the antibody specifically binds exist, the local concentration of the catheter-fused antibody increases due to the binding between the antigen and the antibody, and the concentration of the catheter also increases, resulting in a proximity effect. This increases the formation of homodimers between catheters, resulting in antibody molecules joining hands together on the target surface to form a long chain (catenation), which inhibits dissociation of antibodies and increases the total antigen-binding capacity of antibodies. It is associated with a large increase.

The term “surface” means any surface of any object or article, whether internal or external, vertical or horizontal, and whether in vivo or in vitro. For example, the surface may be the surface of a biological material such as cells or the surface of a support. For example, the surface may be the surface of a cell expressing a target antigen (e.g., the surface of a cancer cell expressing a tumor antigen) or the surface of a cell expressing a target biomarker for disease diagnosis (e.g., a target biomarker protein). It may be the surface of a cell expressing . As another example, the surface may be, but is not limited to, the surface of a reaction vessel or a support, such as a support made of metal, plastic, glass or ceramic components, or a polymeric or elastic support.

In one embodiment of the present application, a catenator that satisfies the above conditions is a polypeptide (SEQ ID NO: 1; molecular weight 8) consisting of amino acid residues 8 to 67 of SDF-1α (Stroma cell-derived factor-1α), a human chemokine protein. kDa, homodimerization K _D 140μM), SAM domain (Sterile Alpha Motif domain) consisting of amino acid residues 254-316 of SLy1 (SH3 domain-containing protein expressed in lymphocytes 1) (SEQ ID NO. 2, molecular weight 7.4 kDa, homologous Dimerization K _D 117μM), a coiled-coil protein designed de novo by the present inventors, a polypeptide named HomoCC (homodimeric coiled-coil) (SEQ ID NO: 3; molecular weight 8.7 kDa, homodimerization K _D 7.6) μM) was used, but this is only a representative example and the scope of the present invention is not limited thereto.

catheter Amino acid sequence (SEQ ID NO) SDF-1α (8-67) RCPCRFFESHVARANVKHLKILNTPACALQIVARLKNNNEQVCIDPKLKWIQEYLEKALN (SEQ ID NO: 1) SLy1 (254-316) KTLHELLERIGLEEHTTSTLLLNGYQTLEDFKELRETHLNELNIMDPQHRAKLLTAAELLLDYD (SEQ ID NO: 2) HomoCC DEEQRKVVEEDLKVLEHLRRVVERKEHLVRDAYEETFDDQQREVVREKLKVLEHLEKVIERDRHLSSRPGL (SEQ ID NO: 3)

Antibody or fragment thereof

As used herein, the term “antibody” is used in the broadest sense to refer to proteins that specifically bind to a specific antigen or epitope, and is a protein produced by stimulation of an antigen within the immune system or produced chemically or recombinantly. It may be one protein, and its type is not particularly limited. Specifically, monoclonal antibodies (including full-length monoclonal antibodies), polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), synthetic antibodies (also called antibody mimetics), chimeric antibodies, and humanized antibodies, as long as they exhibit the desired biological activity. It encompasses antibodies, human antibodies, or antibody fusion proteins (also called antibody conjugates). Additionally, the antibody may be a therapeutic antibody or a diagnostic antibody, but is not limited thereto.

The term “monoclonal antibody” refers to an antibody molecule of single molecular composition obtained from a substantially homogeneous population of antibodies and exhibits a single binding specificity and affinity for a particular epitope. The term “polyclonal antibody” refers to an antibody mixture containing two or more monoclonal antibodies that bind to different epitopes of the same antigen and can react with multiple epitopes. The term “chimeric antibody” refers to an antibody in which the variable region of a non-human antibody and the constant region of a human antibody are recombined, and the immune response is greatly improved compared to a non-human antibody. The term “humanized antibody” refers to an antibody in which all or part of the CDR sequence of a non-human antibody is transplanted into a human antibody. For example, the CDRs of a mouse monoclonal antibody are combined with the Framework region (FR) of a human antibody. A humanized variable region can be produced by recombination and then recombined with the constant region of a desired human antibody, but is not limited thereto. The term “human antibody” refers to an antibody that is completely free of non-human animal derived portions. In general, human antibodies are antibodies in which both the variable and constant regions, including the CDR region and the FR region, are derived from human germline immunoglobulin sequences.

A complete antibody (e.g., IgG type) is structured with two full-length light chains and two full-length heavy chains, and heterodimers, each consisting of one copy of the heavy chain and one copy of the light chain, are arranged in a Y shape. It is a homodimer. Each light chain is connected to the heavy chain by disulfide bonds. The constant region of an antibody is divided into a heavy chain constant region and a light chain constant region, and the heavy chain constant region has gamma (γ), mu (μ), alpha (α), delta (δ), or epsilon (ε) types, and is classified into subclasses. It has gamma 1 (γ1), gamma 2 (γ2), gamma 3 (γ3), gamma 4 (γ4), alpha 1 (α1), or alpha 2 (α2). The constant region of the light chain has kappa (κ) and lambda (λ) types.

The term "heavy chain" refers to a variable region domain V _H containing an amino acid sequence with sufficient variable region sequence to confer specificity to an antigen and three constant region domains C _H1 , C _H2 and C _H3 and a hinge ( It is interpreted to include both the full-length heavy chain including the hinge and fragments thereof.

The term "light chain" includes both a full-length light chain and fragments thereof comprising a variable region domain V _L and a constant region domain C _L comprising an amino acid sequence with sufficient variable region sequence to confer specificity to an antigen. interpreted as meaning.

The term “complementarity-determining regions (CDR)” refers to a region in the variable region of an antibody that confers binding specificity or binding affinity to an antigen. Generally, three CDRs (CDR-H1, CDR-H2, CDR-H3) exist in the heavy chain variable region, and three CDRs (CDR-L1, CDR-L2, CDR-L3) exist in the light chain variable region. The CDR may provide key contact residues for the antibody or fragment thereof to bind to the antigen or epitope. “Framework region (FR)” refers to the non-CDR portion of the variable regions of the heavy and light chains, typically containing four FRs in the heavy chain variable region (FR-H1, FR-H2, FR-H3, and FR- H4), and there are four FRs (FR-L1, FR-L2, FR-L3 and FR-L4) in the light chain variable region. The exact amino acid sequence boundaries of a given CDR or FR can be determined using a number of well-known systems, including the Kabat numbering system, Chothia numbering system, Contact numbering system, IMGT numbering system, Aho numbering system, and AbM numbering system. It can be easily determined using any one of the following.

The term “variable region” refers to the domain of an antibody heavy or light chain that is involved in binding the antibody to antigen. The heavy chain variable (VH) and light chain variable (VL) regions have generally similar structures, with each domain containing four conserved framework regions (FR) and three CDRs.

The term “fragment of an antibody” refers to any fragment of an antibody that lacks at least some of the amino acids present in the full-length chain of the antibody. For the purposes of the present invention, the fragment of an antibody may be a fragment that maintains the original homodimeric form of the antibody, for example, Fc or F(ab') ₂ , but is not limited thereto.

F(ab') ₂ is a fragment containing the antigen-binding site among the fragments generally produced by cleaving an antibody with the proteolytic enzyme pepsin, and is a heterodimer composed of light chain VL and VH domains and heavy chain C _L and C _H1 domains. It is a homodimer in which two types of Fab are connected by a disulfide bond in the hinge region. Fc represents a dimeric form of the heavy chain fragment consisting of heavy chain C _H2 and C _H3 domains, and in some cases also includes a hinge region.

The antibody or fragment thereof of the present application fuses a protein (catenator) capable of forming a homodimer to the two C-termini of the heavy chain or the two C-termini of the light chain, respectively, to form an antibody molecule on the surface where the target antigen is present. By inducing reversible linkage (catenation), the actual binding affinity of the antibody to the antigen can be dramatically increased compared to the intrinsic binding affinity. Therefore, this technical idea is applicable to various antibody fields using antibodies or fragments thereof that maintain all or part of the antibody's unique homodimer structure, some examples of which are as follows.

Antibody-drug conjugate (ADC)

The term “antibody-drug conjugate (ADC)” refers to the covalent attachment of an antibody to a therapeutically active substance or active pharmaceutical ingredient (API), so that the drug exerts its pharmacological function against the binding target targeted by the antibody. It was made to work. Therapeutically active substances or active pharmaceutical ingredients may be cytotoxins capable of killing malignant cells or cancer cells, antibiotics, enzymes such as nucleases, and radionuclides, but are not limited thereto. Covalent attachment of the therapeutically active substance or active pharmaceutical ingredient can be carried out in a non-site-specific manner using standard chemical linkers that couple them to lysine or cysteine residues, or preferably the conjugation is site-specific. This can be performed in a flexible manner, allowing full control of the conjugation site and drug-to-antibody ratio (DAR) of the resulting ADC.

The terms “cytotoxin” and “cytotoxic agent” refer to any molecule that inhibits or prevents the function of cells, and/or causes destruction of cells (apoptosis) and/or exhibits an anti-proliferative effect. It will be appreciated that the cytotoxin or cytotoxic agent of the ADC is also referred to in the art as the “payload” of the ADC. Many classes of cytotoxic agents are known in the art to have potential utility in ADC molecules and can be used in the ADCs disclosed herein. Classes of such cytotoxic agents include, but are not limited to, microtubulin structure formation inhibitors, mitosis inhibitors, topoisomerase inhibitors, or DNA intercalators. Examples of specific cytotoxic agents include maytansinoid, auristatin, dolastatin, trichothecene, CC-1065 drug (NSC 298223), calicheamicin, enediynes, taxane, anthracycline, methotrexate, adriamycin, vindesine, vinca alkaloid, doxorubicin, melphalan ), mitomycin C, chlorambucil, daunorubicin, daunomycin, etoposide, teniposide, carminomycin ), aminopterin, dactinomycin, bleomycin, esperamicin, 5-fluorouracil, melphalan, nitrogen mustard (methyl Including, but not limited to, chlorethamine hydrochloride (nitrogen mustar (mechlorethamine HCL)), cis-platinum and its analogues, cisplatin, CPT-11, doxorubicin, and docetaxel.

As for the “antibody” of the antibody-drug conjugate herein, the configuration for the antibody as described above may be applied, and the two C-termini of the heavy chain or the two C-termini of the light chain of the antibody or fragment thereof may each contain a categorical antibody as described above. Since the catheter can be fused, homodimers are formed between the catheters on the surface where the target antigens exist, thereby enhancing the antibody-antigen binding force by connecting the antibody-drug conjugates to each other, thereby improving the pharmacological properties of the antibody-drug conjugate. Function can be improved (see Figure 1d).

multispecific antibodies

The term “multispecific” antibody refers to an antibody that has two or more antigen-binding sites, at least two of which bind to different antigens or different epitopes of the same antigen. For example, a “bispecific” antibody refers to an antibody with two different antigen-binding specificities, and a “trispecific” antibody refers to an antibody with three different antigen-binding specificities.

Multispecific antibodies can have various formats depending on how two or more antibodies are combined. For the purposes of this application, multispecific antibodies have the Fc region of a homodimer or heterodimer as a common basis. It is preferable that it has an IgG-like structure.

Specifically, the Fc region of a multispecific antibody may be a homodimer or heterodimer. In general, homodimerization of the Fc region is induced by non-covalent bonds between the last domains of the antibody constant region (CH3 domain in the case of IgG) and disulfide bonds between the hinge regions. Heterodimeric Fc regions are prepared by engineering them to have a binding that favors heterodimerization and disfavors or excludes homodimerization through specific non-covalent bonds between the last domains of the constant region, which contribute to naturally occurring antibody homodimerization. can do. For example, by inducing mutations in the CH3 domains of two different Ig antibody heavy chains through genetic manipulation, the two heavy chains have a structure very similar to that of a naturally occurring antibody and have minimal differences in sequence, forming a heterodimer. can be induced to form. Technologies related to this include, but are not limited to, Genentech's knob-into-hole technology, Zymeworks' ZW1, Gencore's HA-TF, and EMD Serono's SEEDbody. .

Fc-based multispecific antibodies may have a bilaterally symmetrical or asymmetric structure. Symmetric Fc-based multispecific antibodies mainly have a variable region (Fv) or scFv portion with different antigen specificity added to the N-terminus or C-terminus of the light or heavy chain of IgG as an antigen-binding site, making them larger than typical IgGs. You lose. Domain antibodies or alternative binding scaffold molecules may be used instead of Fv or scFv. Asymmetric Fc-based multispecific antibodies are based on the Fc region of a heterodimer and are similar in shape and size to typical IgG, but the two Fab arms can recognize different Fc antigens.

As described above, the multispecific antibody herein can be fused to the two C-termini of the heavy chain or the two C-termini of the light chain, respectively, to the catheter as described above, so that the catheter can be separated between the catheters on the surface where the target antigens are present. By forming a homodimer, the antibody-antigen binding force can be improved by linking multispecific antibodies to each other, thereby improving the pharmacological function of the multispecific antibody.

In particular, when the multispecific antibody herein is based on the Fc region of a heterodimer, the catenators fused to the C-terminus of each of the two copies of the Fc region may be the same or different polypeptides. .

Drug-Fc fragment conjugate

As used herein, the term “carrier” refers to a substance that binds to a drug and increases or decreases the physiological activity of a bioactive polypeptide or increases its in vivo stability. It is known in the art that the Fc region of immunoglobulin can be used as a carrier for drugs.

“Drug” means a substance that exhibits therapeutic activity when administered to humans or animals, and includes, but is not limited to, polypeptides, compounds, extracts, nucleic acids, etc. The drug may preferably be a polypeptide drug. Polypeptide drugs that can be linked to the Fc fragment include hormones (intestinal hormones, growth hormone-releasing hormones, growth hormone-releasing peptides, etc.), interferons (interferons-α, -β, -γ, etc.), interleukins, and growth factors. , granulocyte colony stimulating factor (G-CSF), granulocyte-macrophage colony stimulating factor (GM-CSF), glucone-like peptides (GLP-1, etc.), erythropoietin, enzymes, etc., but are not limited to these. No. Additionally, any derivative or derivative is also included in the scope of the polypeptide drug of the present invention, as long as it has substantially the same or increased function, structure, activity or stability as the natural form of the polypeptide drug.

In addition to polypeptide drugs, various drugs can be linked to the Fc fragment, including tetracycline, minocycline, doxycycline, ofloxacin, levofloxacin, ciprofloxacin, clarithromycin, erythromycin, cefaclor, cefotaxime, imipenem, penicillin, and gentamicin. , antibiotics selected from derivatives and mixtures of streptomycin, vancomycin, etc.; Anticancer agents selected from derivatives and mixtures such as methotrexate, carboplatin, taxol, cisplatin, 5-fluorouracil, doxorubicin, etposide, paclitaxel, camptothecin, and cytosine arabinose; Anti-inflammatory agents selected from derivatives and mixtures of indomethacin, ibuprofen, ketoprofen, piroxicam, flubiprofen, diclofenac, etc.; Antiviral agents selected from derivatives and mixtures such as acyclovir, lobavin, etc.; and antibacterial agents selected from derivatives and mixtures such as ketoconazole, itraconazole, fluconazole, amphotericin-B, and griseofulvin, but are not limited thereto.

In general, the Fc region is a dimeric form of a heavy chain fragment consisting of heavy chain C _H2 and C _H3 domains, and in some cases also includes a hinge region. The Fc fragment of the present invention, as long as it has substantially the same or improved effect as the native type, excludes only the heavy and light chain variable regions of immunoglobulin, and includes part or all of the heavy chain constant region 1 (C _H1 ) and/or light chain constant region 1 ( It may be an extended Fc region including C _L1 ). Additionally, it may be a region in which part of the amino acid sequence corresponding to C _H2 and/or C _H3 has been removed. Additionally, in the present invention, the Fc fragment includes not only the native amino acid sequence but also its sequence derivative (mutant). Amino acid sequence derivative means that one or more amino acid residues in the natural amino acid sequence have a different sequence due to deletion, insertion, non-conservative or conservative substitution, or a combination thereof. Additionally, in the present invention, the Fc fragment may be in the form of a native sugar chain, an increased sugar chain compared to the native type, a decreased sugar chain compared to the native type, or a form in which the sugar chain has been removed.

The Fc fragment herein may be fused to a bioactive polypeptide via a peptide bond, or may be linked to a bioactive polypeptide via a peptide linker or a non-peptide linker, but is not limited thereto.

The Fc fragment herein may be a homodimer or a heterodimer. In general, homodimerization of Fc fragments is induced by non-covalent bonds between the last domains of the antibody constant region (CH3 domain for IgG) and disulfide bonds between the hinge regions. Heterodimers of the Fc fragment are engineered to have a binding that favors heterodimerization and disfavors or excludes homodimerization through specific non-covalent bonds between the last domains of the constant region, which contribute to naturally occurring antibody homodimerization. It can be manufactured. For example, by inducing mutations in the CH3 domains of two different Ig antibody heavy chains through genetic manipulation, the two heavy chains have a structure very similar to that of a naturally occurring antibody and have minimal differences in sequence, forming a heterodimer. can be induced to form. Technologies related to this include, but are not limited to, Genentech's knob-into-hole technology, Zymeworks' ZW1, Gencore's HA-TF, and EMD Serono's SEEDbody. .

In the drug-Fc conjugate herein, the drug may be bound to either or both of the N-termini of the two copies of the Fc region.

In addition, the above-mentioned catheters can be fused to the two C-termini of the drug-Fc conjugate Fc, so that homodimers are formed between the catheters on the surface where the target antigens are present, thereby forming the drug-Fc conjugate. By linking them together, the antibody-antigen binding ability can be improved, thereby improving the pharmacological function of the drug-Fc conjugate (see Figure 1e).

In particular, when the drug-Fc fragment conjugate herein is based on the Fc region of a heterodimer, the catenators fused to the C-terminus of each of the two copies of the Fc region may be polypeptides that are the same or different from each other. You can.

therapeutic antibodies

The term “therapeutic antibody” refers to an antibody used in the treatment of a disease. Therapeutic antibodies can have a variety of mechanisms of action. Therapeutic antibodies can bind to their target and neutralize its normal function. For example, monoclonal antibodies that block the activity of proteins necessary for the survival of cancer cells cause cancer cell death. Another therapeutic monoclonal antibody can bind to its target and activate its normal function. For example, monoclonal antibodies can bind to proteins on cells and trigger apoptosis signals. Finally, if the monoclonal antibody binds to a target expressed only in the diseased tissue, a toxic payload (effective agent), such as a chemotherapeutic agent or radioactive agent, may be conjugated to the monoclonal antibody to It is possible to create agents that specifically deliver toxic payloads to reduce harm to healthy tissue.

Therapeutic antibodies may be antibodies used for the treatment of various diseases, such as cancer (eg, hematological cancer and solid cancer), immune disease, neurological disease, vascular disease, or infectious disease. Antigens for these antibodies include tumor antigens, such as tumor-associated antigen (TAA), tumor-specific antigen (TSA), or tumor-derived neoantigens; Antigens of infectious agents, such as those derived from viruses, bacteria, parasites, or fungi; Self-antigens known or suspected to cause autoimmunity; Alternatively, it may be a peptide derived from an allergen (allergen) known or suspected to cause allergy, but is not limited thereto.

For example, such antibodies may be directed to ALK, adhesion-related kinase receptor (e.g., Axl), or ERBB receptor (e.g., EGFR, ERBB2, ERBB3, ERBB4), or erythropoietin-producing hepatocyte (EPH) receptor (e.g., EphA1, EphA2, EphA3, EphA4, EphA5, EphA6, EphA7, EphA8, EphB1, EphB2, EphB3, EphB4, EphB5, EphB6), or fibroblast growth factor (FGF) receptors (e.g., FGFR1, FGFR2, FGFR3, FGFR4, FGFR5), or Fgr, or IGF1R, or insulin receptor, or LTK, or M-CSFR, or MUSK, or platelet-derived growth factor (PDGF) receptor (e.g., PDGFR-A, PDGFR-B), or RET, or ROR1, or ROR2, or ROS, or RYK, or vascular endothelial growth factor (VEGF) receptors (e.g., VEGFR1/FLT1, VEGFR2/FLK1, VEGF3), or tyrosine kinase (TIE) receptors with immunoglobulin-like and EGF-like domains (e.g., TIE-1, TIE-2/TEK), or Tec, or TYRO10, or insulin-like growth factor (IGF) receptor (e.g., INS-R, IGF-IR, IR-R), or Discoidin Domain )(DD) receptors (e.g., DDR1, DDR2), or receptor for c-Met (MET), or receptor d'origine nantais (RON, also known as macrophage stimulating 1 receptor) , or Flt3 (fins-related tyrosine kinase 3), or colony-stimulating factor 1 (CSF 1) receptor, or receptor for c-kit (KIT or SCFR), or insulin receptor-related (IRR) receptor, or CD19 , or CD20, or HLA-DR, or CD33, or CD52, or G250, or GD3, or PSMA, or CD56, or CEA, or Lewis Y antigen, or IL-6 receptor, It is not limited to this.

For the therapeutic antibody herein, the configuration for the antibody as described above may be applied, and the catheter as described above may be fused to the two C-termini of the heavy chain or the two C-termini of the light chain, respectively, of the antibody or its fragment. Therefore, homodimers are formed between catheters on the surface where target antigens are present, thereby linking the therapeutic antibodies to each other, thereby enhancing the antibody-antigen binding force, thereby improving the pharmacological function of the therapeutic antibody.

In addition, the therapeutic antibody herein may be an antibody against a viral antigen, and in viruses with multiple copies of target antigen on the surface, the antibody-cathenator promotes antibody-antigen binding to the target antigen on the surface of the virus, thereby Neutralization activity can be greatly improved.

diagnostic antibodies

The term “diagnostic antibody” refers to an antibody used as a diagnostic reagent for a disease. Diagnostic antibodies may bind to target biomarkers that are specifically associated with a particular disease or show increased expression in a particular disease. Diagnostic antibodies can be used, for example, for target detection in a biological sample from a patient or for diagnostic imaging of a disease site, such as a tumor, within a patient.

For the diagnostic antibody herein, the configuration for the antibody as described above may be applied, and the catheter as described above may be fused to the two heavy chain C-termini or the two light chain C-termini of the antibody or its fragment, respectively. , Homodimers are formed between catheters on the surface where target antigens are present, thereby linking diagnostic antibodies to each other, thereby improving antibody-antigen binding, making it possible to easily detect target biomarkers in biological samples.

In particular, the technology of the present application may be useful in cases where a very small number of biomarkers exist on target cells and a detection antibody with particularly high binding affinity is needed to detect them.

As another example, the technology herein can be applied to the secondary antibody of a sandwich-type biosensor. The term “sandwich assay” refers to an immunological assay that uses two or more antibodies that bind to different sites on an antigen. For example, the assay typically includes a capture antibody and a detection antibody. As used herein, the term “capture antibody” refers to an antigen bound to a support so that it binds well to the support. When processing a biological sample, if the sample contains a target biomarker (antigen), the biomarker (antigen) binds to the capture antibody to form an antigen-antibody complex. Afterwards, if a secondary antibody that binds to another part of the biomarker (antigen) is added, it reacts with the antigen of the antigen-antibody complex, making it possible to easily detect the presence of the biomarker (antigen) in the sample.

The secondary antibody may be coupled with one or more detectable labels to facilitate detection and/or quantification of bound antigen. Such labels may include, but are not limited to, fluorescent substances, biotin moieties, and/or enzymes. For example, enzymes include peroxidase such as horseradish peroxidase (HRP), alkaline phosphatase, etc., and fluorescent substances include FITC, RITC, etc., and other radioactive substances. Isotope labeling, latex bead labeling, colloid labeling, biotin labeling, etc. can be used, but are not limited thereto.

The technology of the present application can be applied to the secondary antibody of a sandwich type biosensor, and the catheter as described above can be fused to the two heavy chain C-termini or the two light chain C-termini of the antibody or its fragment, respectively, resulting in immunity. The antibody-antigen binding ability of the secondary antibody to the antigen bound to the capture antibody on the surface of the support for medical diagnosis can be improved, thereby greatly improving the detection sensitivity of biomarkers (antigens) in the sample.

linker

Meanwhile, another example of the conjugate herein may be one in which a catheter polypeptide is connected to an antibody or a fragment thereof through a linker.

The term “linker” refers to a molecule or group of atoms that connects, couples, or joins two or more components together. The linker may be a water-soluble and/or flexible linker. Linkers may have additional functions such as increasing or decreasing water solubility, increasing the distance between the two connecting components, providing flexibility, or increasing stability, but they can also cause immunogenicity or affect the activity of the conjugate. It is desirable not to affect .

In a preferred embodiment, the linker may be a peptide linker and contains 1 to 10, or 2 to 10 amino acids, such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more ( It may be, but is not limited to, a length of 20 amino acids or less, such as 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids. For example, the peptide linker may be composed of neutral amino acids (more specifically, Gly, Ser, Ala, Thr, or a combination of these four amino acids). For example, (GS) _n , (G ₂ S) _n , (G ₃ S) _n , (G ₄ S) _n , G _n , LE, SSGG or GGGGSGGGGG (where G is Gly, S is Ser, L may be Leu, E may be Glu, and n may be an integer of at least 1), for example, GS, GGGGS, LE, SSGG, GG, GGGGG, or GGGGSGGGGG, but is not limited thereto.

Manufacturing method

The conjugate of the present application is not limited thereto, but can preferably be obtained by expression and purification by recombinant methods. Accordingly, the present invention also provides an expression vector containing a nucleic acid molecule encoding the conjugate, and cells transformed therewith, for expression and purification of the conjugate.

Another example herein relates to an expression vector containing the above nucleic acid molecule.

The term “expression vector” refers to a nucleic acid construct operably linked to express a gene insert encoding a protein of interest. In one embodiment, the expression vector may be linear or circular, or single-stranded or double-stranded DNA, cDNA, RNA, etc., encoding two or more target proteins. An expression vector may form part of a vector that can be used to transform, transfect or transfect a host, but is not limited thereto, and may itself be transcribed and/or translated in vitro.

The term “operably linked” means that the linkage between nucleic acid sequences is functionally related. For example, a coding sequence (e.g., a sequence encoding a protein of interest) can be operably linked with appropriate regulatory elements to enable replication, transcription and/or translation thereof. For example, a coding sequence is operably linked to a promoter if the promoter is capable of driving transcription of the coding sequence. Regulatory elements do not need to be adjacent to the coding sequence as long as they function correctly. For example, intervening sequences that are not translated but are transcribed may exist between the promoter sequence and the coding sequence, and the promoter sequence may still be considered “operably linked” to the coding sequence.

Each component within the expression vector must be operably linked to each other, and linking of these component sequences can be accomplished by ligation at convenient restriction enzyme sites or, if such sites do not exist, by conventional methods. It can be performed using a synthetic oligonucleotide adapter or linker according to.

Expression vectors may contain transcriptional and translational expression control sequences that allow the gene to be expressed in the selected host. Expression control sequences may include promoters to effect transcription, optional operator sequences to regulate such transcription, and/or sequences to regulate termination of transcription and translation. Initiation codons and stop codons are generally considered to be part of the nucleic acid sequence encoding the protein of interest, must be functional in the subject when the genetic construct is administered, and must be in frame with the coding sequence.

For example, a promoter refers to a DNA sequence region where transcription regulators bind, and for the purpose of the present invention, a promoter capable of inducing strong and stable gene expression can be used to increase the gene expression rate.

Promoters can be constitutive or inducible. Promoters include, but are not limited to, the early and late promoters of adenovirus, simian virus 40 (SV40), mouse mammary tumor virus (MMTV) promoter, long terminal repeat (LTR) promoter of HIV, Moloney virus, cytomegalo. Examples include virus (CMV) promoter, Epstein virus (EBV) promoter, Roux Sarcoma virus (RSV) promoter, RNA polymerase ± promoter, T3 and T7 promoters, main operator and promoter region of phage lambda, etc.

In addition, expression vectors include adapters or linkers, enhancers, selection markers (e.g., antibiotic resistance markers), replicable units, polyA sequences, tags for purification, or those that regulate the expression of genes in prokaryotic cells or eukaryotic cells or their viruses. Other sequences of known composition and derivation and various combinations thereof may be included as appropriate.

Expression vectors can be of various types, such as plasmids, viral vectors, bacteriophage vectors, cosmid vectors, etc.

Another example herein relates to cells transformed with the expression vector.

In the present invention, the transformed cell can be any host cell known in the art that can stably and continuously clone or express the expression vector, and prokaryotic cells include E. coli, for example. Bacillus genus strains such as E. coli JM109, E. coli BL21, E. coli RR1, E. coli LE392, E. coli B, E. coli , and Enterobacteriaceae strains such as Salmonella Typhimurium, Serratia marcescens, and various Pseudomonas species, and when transformed into eukaryotic cells, yeast (Saccharomyce cerevisiae), insect cells, plant cells, and animal cells as host cells. , for example, CHO cell line (Chinese hamster ovary), W138, BHK, COS-7, 293, HepG2, 3T3, RIN and MDCK cell lines, etc. may be used, but are not limited thereto.

Introducing expression vectors into cells can be accomplished using suitable standard techniques, such as electroporation, electroinjection, microinjection, and calcium phosphate co-precipitation, as known in the art. phosphate co-precipitation), calcium chloride/rubidium chloride method, retroviral infection, DEAE-dextran, cationic liposome method, polyethylene glycol-mediated uptake. , gene guns, etc. may be used, but are not limited thereto. At this time, the circular structure can be cut with an appropriate restriction enzyme and introduced into a linear form.

The method of selecting the transformed cells can be easily performed according to methods widely known in the art, using the phenotype expressed by the selection marker. For example, when the selection marker is a specific antibiotic resistance gene, the transformant can be easily selected by culturing the transformant in a medium containing the antibiotic.

Cultivation of transformed cells can be performed through various methods known in the art. For example, transformed cells are inoculated into a culture medium and cultured, and then when the cell density reaches a certain level, IPTG is added to the medium to induce protein expression and cultured, thereby causing protein secretion into the cells or into the medium. You can get protein.

Proteins secreted into cells or into the medium can be obtained in purified form according to various purification methods known in the art, but are preferably purified through affinity chromatography using an affinity tag. For example, if the conjugate is fused to GST, the desired protein can be easily obtained using a resin column to which glutathione is bound, and if the conjugate is fused to His, using IMAC (immobilized Metal Affinity Chromatography).

Compositions and Methods

In another aspect, the present invention provides a composition for enhancing antibody-antigen binding, comprising the conjugate as described above, wherein the antibody included in the conjugate specifically binds to a surface where target antigens are present, A homodimer is formed between the catheters included in two or more conjugate molecules, thereby linking the conjugate molecules to each other, thereby suppressing dissociation of the antibody.

In another aspect, the present invention provides a method for enhancing antibody-antigen binding, comprising the step of treating the target surface where antigens to which the antibody contained in the above-described conjugate specifically binds are present, 2 The conjugate molecules are connected to each other by forming homodimers between the catheters included in the above conjugate molecules, thereby suppressing dissociation of the antibody.

The term “antigen-antibody avidity” is meant to include the total antigen-binding avidity of an antibody. The total avidity of an antibody is the sum of the avidity between multiple antigenic determinants present on one antigen and multiple binding sites present on one antibody during antigen-antibody binding, or the overall strength of the binding between the antigen and the antibody. says The antigen-antibody binding force can be measured using any method known in the art. Such methods include, for example, fluorescence activated cell sorting (FACS), surface plasmon resonance (e.g. Biacore, Proteon), biolayer interferometry (BLI, e.g. Octet) (Octet), kinetics exclusion assay (e.g., KinExA), separable beads (e.g., magnetic beads), antigen panning, and/or ELISA. Includes.

If the antibody herein is a therapeutic antibody, administration of the composition may prevent disease, or inhibit, stop, or delay the onset or progression of a disease state, or improve or beneficially alter symptoms. .

The term “effective amount” refers to an amount sufficient to achieve a desired result when administered to an individual, including humans, for example, an amount effective in treating or preventing a disease. The effective amount may vary depending on various factors such as formulation method, administration method, patient's age, weight, sex, severity of disease, food, administration time, administration route, excretion rate, and response sensitivity. Dosages or treatment regimens may be adjusted to provide optimal therapeutic response, as will be understood by those skilled in the art.

The composition herein may be provided with one or more additives selected from the group consisting of pharmaceutically acceptable carriers, diluents, and excipients.

The pharmaceutically acceptable carriers are those commonly used in formulations, for example, lactose, dextrose, sucrose, sorbitol, mannitol, starch, gum acacia, calcium phosphate, alginate, gelatin, calcium silicate, fines. It may be one or more selected from the group consisting of crystalline cellulose, polyvinylpyrrolidone, cellulose, water, syrup, methyl cellulose, methylhydroxybenzoate, propylhydroxybenzoate, talc, magnesium stearate, mineral oil, etc. , but is not limited to this. In addition to the above ingredients, the ingredient may further include one or more selected from the group consisting of diluents, excipients, lubricants, wetting agents, sweeteners, flavoring agents, emulsifiers, suspending agents, preservatives, etc. commonly used in the manufacture of pharmaceutical compositions. . Pharmaceutically acceptable carriers and formulations suitable for the present invention, including those exemplified above, are described in detail in Remington's Pharmaceutical Sciences, current edition.

The composition can be administered orally or parenterally. In case of parenteral administration, intravenous injection, subcutaneous injection, intramuscular injection, intraperitoneal injection, endothelial administration, topical administration, intranasal administration, intraocular administration, intrathecal administration, intrathecal administration, intracranial administration, intrastriatal administration, etc. It can be administered.

In some embodiments, the composition may be presented as a sterile liquid formulation, for example, as an isotonic aqueous solution, suspension, emulsion, dispersion, or viscous composition, which in some aspects may be buffered to a selected pH. Liquid formulations are generally easier to prepare than gels, other viscous compositions, and solid compositions. Additionally, liquid compositions are particularly convenient to administer by injection. On the other hand, viscous compositions can be formulated within an appropriate viscosity range to provide a longer period of contact with a particular tissue. Liquid or viscous compositions may comprise a carrier, which may be, for example, a solvent or dispersion medium containing water, saline, phosphate buffered saline, polyols (e.g., glycerol, propylene glycol, liquid polyethylene glycol) and suitable mixtures thereof. You can.

Sterile injectable solutions can be prepared by incorporating the binding molecule in a solvent such as a suitable carrier, diluent, or admixture with excipients such as sterile water, saline, glucose, dextrose, etc. The composition may also be freeze-dried. The composition may contain auxiliary substances such as wetting agents, dispersing agents or emulsifying agents (e.g., methylcellulose), pH buffering agents, gelling or viscosity enhancing additives, preservatives, flavoring agents, colorants, etc., depending on the route of administration and desired manufacturing route.

Various additives can be added to improve the stability and sterilization of the composition, including antibacterial preservatives, antioxidants, chelating agents, and buffers. Prevention of the action of microorganisms can be ensured by various antibacterial and antifungal agents, such as parabens, chlorobutanol, phenol, sorbic acid, etc. Prolonged absorption of injectable pharmaceutical forms may be brought about by the use of agents that delay absorption, such as aluminum monostearate and gelatin.

Figure 1. Concept of antibody catenation on the surface of target cells by fusion of catheters.
(a) Molecular model of catheter fusion antibody. The flexible linker (Gly-Gly-Ser) between Fc and the catheter was modeled using ROSETTA software. The catheter is an α-helical hairpin structure that forms four helical antiparallel coils (PDB: 1ROP). The structure of Fc was derived from the IgG1 antibody (PDB: 1IGY), and the structure of Fab was referenced to the antibody against the receptor binding site of the SARS-CoV-2 spike protein (PDB: 6XE1).
(b) Reduction in dissociation rate due to antibody ligation when a catheter is fused to each of the C-termini of the two heavy chains of the antibody. Adjacent ^cat Ab-antigen complexes can be linked to each other, and the more ^cat Ab molecules are linked, the less likely they are to dissociate. The total binding capacity to antigen increases as the dissociation rate of ^cat Ab decreases.
(c) Reduction in dissociation rate due to antibody ligation when a catheter is fused to each of the C-termini of the two light chains of the antibody.
(d) When a catenator is fused to each of the C-termini of the two heavy chains of an antibody contained in an antibody-drug conjugate (ADC), the dissociation rate due to antibody ligation is reduced.
(e) Reduced dissociation rate by antibody ligation when a catheter is fused to each of the two C-termini of the Fc fragment covalently linked to the therapeutic protein.
Figure 2. Example of a catheter forming a homodimer.
Three-dimensional modeling structures of SDF-1α(8-67), Sly1(254-316), and HomoCC and ( K _D ) _catenator values of each homodimer.
Figure 3. ABM for simulating the binding dynamics of catheter fusion antibodies.
(a) ( Left ) Each binding site consists of two antigen molecules (2Ag). ( Right ) Gray circles represent spheres sampled by the catheter, and V _overlap means the overlapping volume between adjacent spheres. Connection between two ^cat Ab molecules is possible only at V _overlap .
(b) Three rules of the ABM model. ( Left ) ^cat Ab-2Ag binding is determined by the relative probability equal to [ ^cat Ab]/ K _D. ( Middle ) The connection between adjacent ^cat Ab-2Ag complexes is determined by the relative probability f (d)/( K _D ) _{Catenator ,} which changes with d, the distance between the ( K _D ) Catenator and the complex. ( Right ) It was assumed that bound catAb molecules do not dissociate from the surface.
(c) Simulation requires determining parameters for the binding site, antibody, and catheter. Through MCMC sampling, the state of the binding site on the target surface is repeatedly updated according to the ABM rules and then sampled. Collect a sufficient number of sampling results to determine binding occupancy and effective dissociation constants.
Figure 4. Calculation of f(d) using the uniform local density approximation.
(a) The forward catenation rate at which two catheters dimerize is a function of the volume overlap ( V(d) ) between the effective concentrations of the catheters, which are assumed to be uniformly distributed in a sphere defined by their reach length (L). It is proportional. V(d) depends on the distance (d) between two adjacent ^cat Ab-2Ag complexes and the length of reach.
(b) Shows a plot of f(d) as a function of d calculated for the indicated reach length (L).
Figure 5. Simulation of binding site occupancy and ( K _D ) _eff value according to ( K _D ) _catenator .
( Left ) Binding site occupancy. Simulations were performed for a square arrangement of the binding site. The variable values were K _D = 10 ^-8 M, [ ^cat Ab] = 10 ^-9 M, (reaching distance) = 7 nm, (spacing between binding sites) = 12 nm, (total number of binding sites) = 98, respectively. The mean value and standard deviation of 1024 MCMC simulations for each ( K _D ) _catenator value are shown in blue, and the data is shown as a scatter plot of representative runs (orange).
( Right ) Effective dissociation constant. Data shown on _{the left were converted to (KD} ) _eff values. The dotted line represents the K _D value for the same antibody without catheter. ( K _D ) The maximum increase rate of the effective total bond force, which can be considered equal to the decrease rate of _eff , is 70.6.
Figure 6. Simulation of various configurations of binding sites.
(a) Comparison of uniformly distributed binding sites. Three different configurations of the binding site are shown at the top. Black dots represent binding sites, and gray lines represent pairs that can be connected by catheters. Red circles and blue lines represent the maximum extent of catenation and number of connections for a given binding site, respectively. ( K _D ) Binding site occupancy and ( K _D ) _eff according to _catenator are shown at the bottom. We sampled 1024 trials for each ( K _D ) _catenator value and presented the results. The parameters were K _D = 10 ^-8 M, [ ^cat Ab] = 10 ^-9 M, reach = 7 nm, spacing between binding sites = 12 nm, and the total number of binding sites was 98 for the square arrangement, hexagonal and In the case of a triangular array, there were 102. The number on the right is the maximum increase rate of the effective total bonding force for each arrangement.
(b) Comparison of randomly distributed binding sites. Three random arrangements of binding sites with different binding site densities (ρ) are shown at the top. The surface area for simulation was 5,760 nm ² . Simulation conditions are the same as (A). ( K _D ) Binding site occupancy and ( K _D ) _eff according to _{the catenator} were constructed the same as in (A).
Figure 7. Simulation for randomly distributed high-density binding sites.
1024 trials were sampled for each ( K _D ) _catheter value at the indicated density and the results were plotted. The parameters were K _D = 10 ^-8 M, [ ^cat Ab] = 10 ^-9 M, reach = 7 nm, spacing between binding sites = 12 nm, and surface area = 5,760 nm ² . The maximum fold enhancement of the effective total binding force and ( K _D ) _catheter are tabulated at the bottom.
Figure 8. Effect of native antigen binding potential ([ ^cat Ab]/ K _D ) on (K _D ) _eff and binding site occupancy .
(a) Bond occupancy and (b) effective dissociation constant (K _D ) for [ ^cat Ab]/K _D for [ ^cat Ab]/ K _{D eff} = 1.0, 0.3, 0.1, 0.03, 0.01. Simulations were performed with a square arrangement of binding sites as shown in Figure 5. Set values for the variable parameters were [ ^cat Ab] = 10 ^-9 M, reach length = 7 nm, spacing between binding sites = 12 nm, and total number of binding sites = 98. The average bond occupancy rate of 1024 MCMC simulations was plotted. ( K _D ) _Catheter was set from 3mM to 30nM. The total antigen binding capacity of the antibody was substantially improved over a wide range of [ ^cat Ab]/ K _D.
Figure 9. BLI results show the total binding force due to the catenation effect (I).
Binding dynamics were measured by attaching the indicated target protein to the sensor tip. Performed with various antibody concentrations as indicated. The experimental graph and fitting curve are shown in red and black, respectively. In the case of graph fitting, a 1:1 combination was assumed.
(a) High affinity antibody. N30A/H91A mutant trastuzumab showed a K _D value of 2.1 nM against the ectodomain of HER2.
(b) Low-affinity antibody. glCV30 showed a K _D value of 51.2 nM against the RBD of SARS-CoV-2. All antibodies fused to SDF-1α were expressed as K _D < 10 pM due to measurement limitations of the machine. The experiment was repeated three times and a representative graph is shown.
Figure 10. BLI results show the total binding force due to the catenation effect (II).
Binding kinetics were measured using BLI. Performed with various antibody concentrations as indicated. The experimental graph and fitting curve are shown in red and black, respectively. In the case of graph fitting, a 1:1 combination was assumed.
Trastuzumab (N30A/H91A)-SDF-1α(8-67) ^(L) and Trastuzumab (N30A/H91A)-HomoCC ^(H) were reacted with the HER ectodomain immobilized on the sensor tip. Both types of antibody-cathenators were expressed as K _D < 10 pM due to the measurement limitations of the instrument. The experiment was repeated three times and a representative graph is shown.
Figure 11. BLI results show the total binding force due to the catenation effect (III).
Binding kinetics were measured using BLI. Performed with various antibody concentrations as indicated. The experimental graph and fitting curve are shown in red and black, respectively. In the case of graph fitting, a 1:1 combination was assumed. Obinutuzumab, obinutuzumab (F101B), or obinutuzumab (F101A)-SDF-1α(8-67) ^(H) was reacted with CD20 immobilized on the sensor tip.
Figure 12. BLI results show the total binding force due to the catenation effect (IV).
Binding kinetics were measured using BLI. Performed with various antibody concentrations as indicated. The experimental graph and fitting curve are shown in red and black, respectively. In the case of graph fitting, a 1:1 combination was assumed.
(Top) HetFc_Trastuzumab(N30A/H91A/Y100A) or SLy1(254-316) ^(H) -HetFc_Trastuzumab(N30A/H91A/Y100A)-SDF-1α(8-67) ^(H) on the sensor tip. It was reacted with the HER ectodomain immobilized on .
(Bottom) This is a diagram showing how an antibody with heterodimeric Fc and two different catheters is channelized on the target surface.
Figure 13. Antibody catenation increases binding to breast cancer cells proportional to HER2 expression level.
Fluorescence microscopy was performed by fusing mScarlet fluorescent protein to the light chain C-terminus of the antibody.
(Top left) Western blotting analysis results showing that the four indicated breast cell lines express HER2 at different levels.
(Bottom left) Cells were seeded in two separate wells and incubated with Ab(Trastuzumab(N30A/H91A)) or ^cat Ab(Trastuzumab(N30A/H91A)-SDF-1α(8-67) ^(H) ) was added to the wells immediately before confocal imaging.
(Right) Confocal images were obtained 30 minutes after adding the antibody.
Figure 14. Trastuzumab (N30A/H91A)-SDF-1α(8-67) ^(H) preferentially binds to BT-474 cells with high HER2 expression level rather than MCF-7 cells with low HER2 expression level.
GFP fluorescent protein was fused to the light chain C-terminus of the antibody and fluorescence microscopy was performed.
(Top left) Western blotting analysis results showing that the four indicated breast cell lines express HER2 at different levels. MCF-7 cells express lower levels of HER2 than BT-474 cells.
(Bottom left) The two indicated cell lines were seeded separately and loaded with Ab (trastuzumab(N30A/H91A)) or ^cat Ab (trastuzumab(N30A/H91A)-SDF-1α(8-67) ^(H) ). The medium was added immediately before confocal imaging to allow access to each well of both cell lines.
(Right) Confocal images were obtained 30 minutes after adding the antibody.
Figure 15. Antibody catenation converts the mother antibody that does not bind to su-DHL-5 cells to bind.
FACS (fluorescence assisted cell sorting) analysis was performed by fusing mScarlet fluorescent protein to the light chain C-terminus of the antibody.
(Left) Shows FACS analysis results showing various expression levels of CD20.
(Right) Obinutuzumab or obinutuzumab (F101A)-SDF-1α(8-67) ^(H) was added to su-DHL-5 cells at a concentration of 100 nM. The V1R region represents antibody binding to cells.
Figure 16. Catenator fusion to glCV30 significantly improves virus neutralizing activity.
(Left) Neutralization of vesicular stomatitis virus (VSV) pseudotyped with the SARS-CoV-2 spike protein. Each of the three antibodies (CV30, glCV30, and glCV30-SDF-1α(8-67)) was serially diluted and added to rVSV-ΔG-Luc containing the SARS-CoV-2 spike protein of the Wuhan-Hu-1 strain. . The mixture was incubated with HEK293T-hACE2 cells and luciferase activity was measured.
(Right) Geometric mean titer (GMT) values with 95% confidence intervals are shown on the right, corresponding to IC50 values of 0.25, 3.22, and 0.21 μg/ml for CV30, glCV30, and glCV30-SDF-1α, respectively.

Hereinafter, the present invention will be described in more detail through the following examples. However, these are only for illustrating the present invention, and the scope of the present invention is not limited by these examples. It is obvious to those skilled in the art that the embodiments described below can be modified without departing from the essential gist of the invention.

Example 1. MCMC (Markov Chain Monte-Carlo) simulation

This simulation consists of 'target surface', ' K _D (for antibody-antigen interaction)', '( K _D ) _catenator (for catheter-catheter interaction)', and ' f(d) (of catheter). It was carried out in three steps specified below with specifications such as 'effective local concentration'. A protein that is a fusion of an antibody and a catheter is named antibody-catenator ( ^cat Ab). Because the number of ^cat Abs was much greater than the number of binding sites in all simulations, the concentration of free ^cat Ab was assumed to be the same as the concentration of total ^cat Ab (free ^cat Ab + antigen-bound ^cat Ab). The setting values of simulation parameters are listed in Table 2.

parameter explanation value ^cat Specification of Ab K _D dissociation constant of antibody 10 nM ( K _D ) _catenator Dissociation constant of catheter 10nM-10mM [ ^cat Ab] antibody concentration 1 nM l Length of flexible linker 6 nm c length of catheter 2nm L Reach length ( l+c/2) 7nm Specification of target surface N _{total_binding_sites}
(Figures 4,5) Antibody = number of binding sites 98 - 102 Connectivity number
(Figures 4,5) Number of possible catenations 3 (Hexagonal)
4 (Square)
6 (Triangular) d (Figures 4,5) Distance between adjacent binding sites 12nm L _surface (Figure 5) surface area of target surface 40 nm ² Binding site density (Figure 5) Surface density of binding site 0.5-4.0
(per 12x12 nm ² ) Specifications of Simulation Updates/MCMC step Number of updates in one MCMC step 30,000-100,000 sampling size Number of samples in parameter set 1,024

Stage 1: Initiation Stage

1. A designated 3D target surface is implemented by assigning binding sites to specific locations on the surface.

2. Each binding site is set to be empty.

Step 2: Probabilistic update step

The following substeps (1-3) are repeated for a sufficient time to achieve thermodynamic equilibrium.

1. Any binding site BS ₁ is selected from the target surface.

2. Update the combined status of BS ₁ .

If BS ₁ is empty, set the combined state to max(1, ) changes to ‘occupied status’ with an acceptance probability of .

If BS ₁ is occupied, set the combined state to max(1, ) changes to an ‘empty state’ with an acceptance probability of

3. BS ₁ The occupied binding site BS ₂ located immediately next to it is selected, and ( BS ₁ , BS ₂ ) The pair's catenation status is updated.

If ( BS ₁ , BS ₂ ) is not catheterized, BS ₁ and BS ₂ If you have a catheter that is not all bound, its bound state is max(1, ) is changed to a 'catenation state' with an acceptance probability of ).

If ( BS ₁ , BS ₂ ) is catenated, its combined state is max(1, ) is changed to a ‘non-catenated state’ with an acceptance probability of ).

Step 3. Sample extraction steps

1. At the end of the update phase, the final state of the target surface is recorded.

2. Count the total number of occupied and empty binding sites.

The model system and simulation code can be found in MATLAB and Github, respectively ( https://github.com/JinyeopSong/Antibody_ThermoCalc_JY ). Detailed explanations can be found in Readme.

Example 2. Preparation of antibody conjugated with general antibody and catheter

2-1. Selection of catheters

Three different types of homodimer-forming proteins were used as catheters. One is a polypeptide composed of amino acid residues 8 to 67 of SDF-1α (Stroma cell-derived factor-1α), a human chemokine protein (Ryu et al., 2007), which has a signal transduction function in the full-length protein. -It is a form with the 7 terminals removed, is small in size ( Mr = 8 kDa), and has a very low homodimerization binding force of 140 μM (Veldkamp et al., 2005). The other is the SAM domain (Sterile Alpha Motif domain) composed of amino acid residues 254-316 of SLy1 (SH3 domain-containing protein expressed in lymphocytes 1), is small in size (Mr = 7.4 kDa), and weakly forms homodimers. ( K _D = 117 μM) (Kukuk et al., 2019). The remaining one was a coiled-coil protein designed de novo by the present inventors and was named HomoCC (homodimeric coiled-coil). HomoCC is small in size (Mr = 8.7 kDa) and has a very low homodimerization binding force of 140 μM (Figure 2).

2-2. Antibody-catheter ( ^cat Preparation of Ab)

Six types of antibody-catheter ( ^cat Ab) were prepared by combining four different antibodies and three types of catheters (Table 3)

antigen antibody Antibody - Catheter fusion protein catheter linker
peptide RBD glCV30 glCV30-SDF-1α(8-67) ^(H) HC-C SDF-1α(8-67) GGGGSGGGGS HER2 Trastuzumab (N30A/H91A) Trastuzumab (N30A/H91A)-SDF-1α(8-67) ^(H) HC-C SDF-1α(8-67) GGGGSGGGGS HER2 Trastuzumab (N30A/H91A) Trastuzumab (N30A/H91A)-SDF-1α(8-67) ^(L) LC-C SDF-1α(8-67) GGGGSGGGGS HER2 Trastuzumab (N30A/H91A) Trastuzumab (N30A/H91A)-HomoCC ^(H) HC-C HomoCC GGGS HER2 HetFc_Trastuzumab (N30A/H91A/Y100A) SLy1(254-316) ^(H) - HetFc_Trastuzumab(N30A/H91A/Y100A)-SDF-1α(8-67) ^(H) HC-C Sly1(254-316);
SDF-1α(8-67) GGGGS GGGGS GGGGS CD20 Obinutuzumab (F101A) Obinutuzumab (F101A)-SDF-1α(8-67) ^(H) HC-C SDF-1α(8-67) GGGGSGGGGS - RBD: Receptor-binding domain of SARS-CoV-2 spike protein
- HER2: human ERBB2
- CD20: Cluster of differentiation 20
- Trastuzumab (N30A/H91A): Trastuzumab with N30A and H91A mutations (Slaga et al., 2018)
- HetFc_Trastuzumab (N30A/H91A/Y100A): Trastuzumab (N30A/H91A/Y100A) with heterodimeric Fc
- glCV30: antibody against SARS-CoV-2 RBD (Hurlburt, et al. (2020) Nat Commun 11, 5413)
^{- (H)} : fused to heavy chain; ^(L) : fused to light chain
- SDF-1α(8-67): Stroma cell-derived factor-1α, residues 8-67
- HomoCC: De novo designed homodimeric coiled-coil
- SLy1(254-316): S H3 domain-containing protein expressed in ly mphocytes 1 , residues 254-316
(= SAM domain; S terile A lpha M otif domain) (Kukuk et al., 2019)
- HC-C: C-terminus of the heavy chain; LC-C: C-terminus of light chain
G: glycine; S: serine

Specifically, each DNA fragment encoding the heavy chain variable region (VH) and light chain variable region (VL) of glCV30 was synthesized by IDT and then cloned into the PCEP4 vector obtained by Invitrogen. The DNA fragment of C _H1 -C _H2 -C _H3 of the gamma heavy chain and the light chain constant region of the kappa light chain were inserted into the heavy chain variable region and the light chain variable region, and the cloned vectors were named glCV30 Hc and glCV30 Lc, respectively. The DNA fragment encoding SDF-1 α(8-67) was synthesized by IDT, and the fragment was linked with a (Gly-Gly-Gly-Gly-Ser) ₂ linker sequence after the C _H3 of glCV30 in glCV 30 Hc. The glCV30-SDF-1α(8-67) Hc vector was cloned by insertion. For antibody production, three vectors were amplified using Macherey-Nagel's NucleoBond Xtra Midi kit. The glCV30 Hc and glCV30 Lc vector combination or the glCV30-SDF-1α Hc and glCV30 Lc vector combination was injected into Gibco's ExpiCHO-S cells. After transfection of the vector, the cells were cultured in Gibco's ExpiCHO expression medium for 10 days.

Centrifugation was performed at 4°C to obtain the supernatant, which was then filtered using a 0.45 μm filter from Millipore. Then, a buffer solution consisting of 150mM NaCl, 20mM Na ₂ HPO ₄ , pH 7.0 was mixed and diluted 1:1, and then poured into an open tube column containing Sino Biological's Protein A resin for binding. To elute the antibody bound to the resin, a buffer solution consisting of 0.1 M glycine, pH 3.0 was flowed through the tube column, and the eluted solution was neutralized using a buffer solution consisting of 1M Tris-HCl, pH 8.5. The antibody was purified to a higher purity by performing gel filtration chromatography using Cytiva's HiLoad 26/60 Superdex 200 column. When performing gel filtration chromatography, the column was homogenized using a buffer solution consisting of 20mM Tris-HCl (pH 7.5) and 150mM NaCl.

To prepare CV30, F27V and T28I mutations were added to glCV30 Hc and the CV 30 Hc vector was cloned. Except for using CV30 Hc and glCV30 Lc for transfection, the protein production and purification process is generally the same as that used for glCV30.

Ordinary Trastuzumab and SDF-1α(8-67) or Trastuzumab conjugated with HomoCC or SLy1(254-316) (Trastuzumab-SDF-1α(8-67) ^(H) , Trastuzumab-SDF To prepare -1α(8-67) ^(L) , Trastuzumab-HomoCC ^(H) , and Trastuzumab-Sly1(254-316) ^(H) ), the heavy chain variable region (VH) of trastuzumab was obtained through IDT. DNA fragments encoding the light chain variable region (VL), HomoCC, and Sly1 (254-316) were synthesized. N30A/H91A mutations were added to the light chain of each trastuzumab. The above cloning, protein production, and purification processes are generally the same as those used for glCV30 and glCV30-SDF-1α (8-67). However, in the case of Trastuzumab (N30A/H91A)-SDF-1α(8-67) ^(L) , SDF-1α(8-67) is added to the end of Trastuzumab Lc (Gly-Gly-Gly-Gly-Ser) ₂ Cloned with linker. When HomoCC was used as a catheter, the linker sequence Gly-Gly-Gly-Ser was used.

To prepare HetFc_Trastuzumab (N30A/H91A/Y100A) ^(H), refer to Heteromeric Fc (PDB: 5DI8) and add T366W to CH3 of one antibody heavy chain and T366S, L358V, and Y407A mutations to CH3 of the other heavy chain. It was added. The sequence of the linker connecting the catheters is (Gly-Gly-Gly-Gly-Ser) ₃ . Other cloning, protein production, and purification processes are generally the same as those used for glCV30.

In order to prepare Obinutuzumab-SDF-1α(8-67) ^(H) , which is a conjugate of regular Obinutuzumab and SDF-1α(8-67), Obinutuzumab was obtained through IDT. DNA fragments encoding the heavy chain variable region (VH) and light chain variable region (VL) of binutuzumab were synthesized. The F101A mutation was added to the heavy chain of obinutuzumab. The above cloning, protein production, and purification processes are generally the same as those used for glCV30 and glCV30-SDF-1α(8-67) ^(H) .

Example 3. Bio-layer interferometry (BLI)

To measure the dissociation constant, a BLI experiment was performed using Sartorius' Octet R8. Biotinylated SARS-CoV-2 RBD (Acrobio system), biotinylated Her2/ERBB2 (Sino Biological), and biotinylated CD20 (ACRO biosystems) were attached to Sartorius' streptavidin biosensor tip for 120 seconds. The baseline was determined by immersing the sensor tip in Sartorius' Kinetics Buffer for 60 seconds. Antibody samples of different concentrations went through an association step of 240 seconds and a dissociation step of 720 seconds. All reactions were performed in Kinetics Buffer. The binding reaction was analyzed using Sartorius' Octet Data Analysis 10.0 software, and kinetic parameters were calculated.

Example 4. Cell line analysis

HER expression levels in breast cancer cell lines MDA-MB-231 (ATCC), MCF-7 (ATCC), SK-BR-3 (ATCC), and BT474 (ATCC) were confirmed by Western blotting. Each cell was seeded in two separate wells, and trastuzumab(N30A/H91A) with mScarlet or GFP fluorescent protein fused to the light chain C-terminus and trastuzumab(N30A/H91A)-SDF-1α(8-67) were seeded in two separate wells. ) ^(H) was added at each concentration and analyzed with a confocal microscope 30 minutes later. Because the lymphoma cell line su-DHL-5 (DSMZ) is a suspended cell, the cell-binding activity of the antibody was analyzed using FACS (fluorescence assisted cell sorting) rather than confocal microscopy.

Example 5. Virus neutralization activity analysis

To prepare vesicular stomatitis virus (VSV) pseudotyped with the SARS-CoV-2 spike protein of the Wuhan-Hu-1 strain, HEK293T cells were previously plated overnight at ^3x10 cells in a 10 cm dish, 15 μg plasmid encoding the spike protein of SARS-CoV-2 with the C-terminal 18 residues deleted was transfected using calcium phosphate. 24 hours after transfection, cells expressing the spike protein were infected for 1 hour with recombinant VSV in which the G gene was replaced by the luciferase gene (rVSV-ΔG-Luc). Cells were washed three times with Dulbecco's phosphate-buffered saline, and 7-10 mL of the same medium containing 10% FBS was added. Culture medium was harvested 24 to 48 hours after infection, filtered through a 0.45 μm filter, and VSV (rVSV-ΔG-Luc) pseudotyped with the SARS-CoV-2 spike protein of the Wuhan-Hu-1 strain was obtained. . It was stored at -80°C for neutralization analysis.

Each of the three antibodies (CV30, glCV30, and glCV30-SDF-1α(8-67)) was serially diluted and added to rVSV-ΔG-Luc containing the SARS-CoV-2 spike protein of the Wuhan-Hu-1 strain; , the mixture was incubated with HEK293T-hACE2 cells, the luciferase activity was measured, and the IC ₅₀ value was determined.

Experiment result

1. Concept of antibody catenation on the target surface

This concept was designed from (i) the unique dimeric structure of IgG-type antibodies and (ii) the proximity effect that occurs on the target cell surface. In the structure of IgG, the Fc region consists of the constant regions (C _H2 , C _H3 ) of two heavy chains that form a homodimer, and the two C termini face in opposite directions and are separated by about 23 Å (Figure 1a, left). Based on these structural features, the length of the connecting linker and the affinity of the homodimer can be adjusted to prevent the homodimeric protein fused to the C-terminus of the heavy or light chain from forming a homodimer within the antibody molecule. Instead, homodimers between antibody molecules can form, resulting in catenation between antibody molecules (Figure 1a, right). A protein that is a fusion of an antibody and a homodimeric protein is designated as an antibody-catenator ( ^cat Ab). Because the local concentration of ^cat Ab on the surface increases due to antibody-antigen binding interactions, a proximity effect for ^cat Ab is expected on the surface of cells overexpressing the target antigen. In summary, the formation of catenator intermolecular homodimers forms long chain-like catenated antibodies (Figure 1b, c). The effective antigen-binding affinity of ^cat Ab increases with catenation, and the degree of increase varies depending on the degree of catenation. By fusing homodimeric proteins with weak affinity, the total binding avidity (antigen-binding avidity) of an IgG-type antibody to the target antigen can be increased.

2. ^cat Agent-based modeling (ABM) to predict Ab behavior

ABM is a computational modeling approach used in a variety of research fields, including statistical physics and biological sciences. This technique can be used to understand the macroscopic behavior of complex systems by defining rules that determine the microscopic behavior of the agents that make up the system.

An ABM was constructed to simulate the appearance of ^cat Ab molecules on the target antigen surface. To avoid complexity, we assumed that each binding site is a pair of two antigen molecules (2Ag), and that ^cat Ab interacts 1:1 bivalently with the binding site to form an occupied binding site ( ^cat Ab-2Ag) (Figure 3a, left). For catenation to occur between two adjacent ^cat Ab-2Ag complexes, the distance (d) between the two adjacent complexes is the sum of the reach length (L) and the linker length (l), defined as l+c/2, and the catenator length. It should be closer than half of (c) (Figure 3a, right). Therefore, several parameters affect catenation at the target surface. In this ABM model, all possible binding sites on the target surface are considered as individual agents in the ABM formulation, and each binding site is assigned to a fixed location on a three-dimensional (3D) surface with periodic boundary conditions. ABM determines the behavior of ^cat Ab molecules on the target surface through three rules. The first rule concerns intrinsic antibody-antigen binding. The empty binding site binds to one free ^cat Ab through a bivalent interaction to form an occupied binding site. Bound ^cat Ab can dissociate from the occupied binding site, leaving the binding site unoccupied. The equilibrium state of occupied and unoccupied binding sites is determined by the total intrinsic avidity of the antibody excluding the influence of the catheter. The relative probability of an occupied state to an unoccupied state at the binding site (intrinsic antigen binding probability) can be defined as [ ^cat Ab-2Ag]/[2Ag] and thus can be expressed as [ ^cat Ab]/ K _D. where [ ^cat Ab] is the concentration of ^cat Ab and K _D is the dissociation constant for the bivalent ^cat Ab-2Ag interaction (Figure 3b, left). The second rule is about catenation. A pair of ^cat Ab-2Ag complexes on the target surface can be linked by the formation of intermolecular homodimers between the catheters (Figure 3B, middle). For a pair of ^cat Ab-2Ag complexes separated by d (Figure 3a, right), the relative probability of the catenated state relative to the uncategorized state is the ratio of the forward kinetics (catenation) and the reverse kinetics (decateration). It is the same as the ratio. The formulas for calculating the forward reaction rate ( R _catenation ) and reverse reaction rate ( R _decatenation ) are as follows:

k _f and k _r are the reaction constants for the forward and reverse directions, respectively. is Avogadro's number, V _sphere is the volume of the oval space where the catheter can be located, and V _overlap (d) means the volume of the space where two catheters can contact to form a homodimer (Figure 3a, right) ). When estimating the forward reaction rate, the catheter was assumed to uniformly sample the V _sphere . The relative probability, defined as R _catenation / R _decatenation , is expressed as:

Therefore, the relative probability is a function of d and is inversely proportional to ( K _D ) _catenator , which is the dissociation constant of the catenator in solution. The function f(d) can be viewed as the local effective concentration of the catheter in V _overlap (d) . As expected, f(d) and relative probability are strongly influenced by the length that the catheter can reach and are limited by ^{the cat} Ab- ^cat Ab distance (Figure 4a,b). Finally, the third rule is for limited dissociation, which assumes that the linked antibody is not allowed to dissociate from the binding site. In the case of a catenated antibody, it holds the dissociated antibody near the binding site, allowing the dissociated antibody to immediately rebind to the binding site. According to this assumption, an antibody molecule can dissociate from its binding site only if its catheter is not involved in homodimerization with a nearby ^cat Ab-2Ag complex (Figure 3b, right).

3. Simulation significantly improves the overall avidity for antigen.

Following the assumed rules, we simulated the effect of antibody catenation on the binding interaction between ^cat Ab and 2Ag on a three-dimensional surface using the MCMC sampling method (Hooten and Wikle, 2010). Our sampling procedure consists of three steps (Figure 3c). The first step is the initiation step, which defines the target surface with the antibody binding site by specifying the coordinates for each site. The set of binding sites is located equidistant from each other or randomly, and the distance between binding sites or the number of binding sites is set as a variable. The next step is the MCMC probabilistic update step. In each iteration, a binding site is randomly selected from the target surface and the probability that the selected binding site state (occupied or not) will change is calculated using the Metropolis-Hasting algorithm (Hastings WK (1970) Monte-Carlo Sampling Methods Using Markov Chains and Their Applications). Biometrika 57(1):97-109;Grazzini J, Richiardi MG, & Tsionas M (2017) Bayesian estimation of agent-based models. J Econ Dyn Control 77:26-47. The 'on' or 'off' status of that location is then updated with the calculated probability. Therefore, the catenation state is updated probabilistically for each update step. In the next sampling step, the total number of occupied binding sites is calculated. This is collected through multiple simulation runs to statistically analyze binding site occupancy and total binding capacity to the antigen. Binding site occupancy is the average of the number of occupied binding sites collected for sampling over 1024 MCMCs. For each simulation, the average binding occupancy rate and effective dissociation constant (( K _D ) _eff ) were calculated considering the antibody catenation effect.

Since the homodimerization of the catheter is affected by how the binding sites are distributed on the 3D surface, simulations were performed for various arrangements of the binding sites. In the simulation, the ( KD ) _catenator was _the main variable and other parameters were set as constants. First, the binding site forming a square lattice was simulated. Through this, it was discovered that ^{the cat} Ab binding site occupancy increases in an S-shape, and it was confirmed that it can increase to saturation by a catheter with low binding affinity. For example, ^{a cat} Ab with ~1 μM of ( K _D ) _catenator shows ~70-fold improvement in effective binding to antigen (= reduction in ( K _D ) _eff ) compared to the same antibody without catenator. shown (Figure 5). As a method of comparison between various simulation settings, '( K _D ) _{catenator ,50} ', defined as ( K _D ) catenator, which enables half-maximum improvement of binding site occupancy, was used (Figure 5).

4. Comparison of simulations for different configurations of binding sites

Next, simulations were performed for various regular arrangements of binding sites and randomly distributed binding sites. Depending on the pattern of regularly distributed binding sites, the number of possible catenations for a given binding site varies. The number of connections was designated as 3, 4, and 6 for the hexagonal, square, and triangular arrangement of the binding site, respectively (Figure 6a). Although these three arrangements were different, they showed similar improvement trends in binding site occupancy and antigen effective binding capacity by the catheter (Figure 6a). As expected, as the number of connections increases, low ( K _D ) _catenator,50 arrays appear. ( K _D ) _catenator,50 was 8.0, 9.2, and 12.2 μM for the hexagonal, square, and triangular arrangement of the binding site, respectively. Simulations showed that as the number of connections increased, the effective binding force increased by up to 41-, 73-, and 93-fold for triangular, square, and hexagonal arrays, respectively (Figure 6a). Therefore, regardless of the distribution pattern, under simulation conditions where the target surface contains only 98 binding sites, the effective binding affinity to the antigen can be increased by at least 51-fold.

For binding sites randomly distributed on the 3D surface relative to the target antigen distribution on the cell surface, the binding site density (ρ) was set as the number of binding sites per unit area (square of the reach of 7 nm) (Figure 6b). . In the simulation, the total surface area was 5,760 nm ² and the number of binding sites was 15, 30, 45, 90, and 120, with ρ of 1.47, 2.94, 4.41, 8.82, and 11.76, respectively. A more compact binding site increases the number of connections for a given binding site. As expected, the simulations showed that higher binding site densities resulted in higher binding site saturation and a much larger increase in effective avidity for antigen. The maximum improvement ranges from 15 (ρ = 1.47) to 1,062 (ρ = 11.76). Likewise, various ( K _D ) _catenator,50 values were confirmed. For example, at a ρ of 11.76 it was 4.2x10 ^-6 M and at a ρ of 1.47 it was 74x10 ^-6 M (Figure 6b). The onset ( K _D ) at which _catenators begin to exert their catenating effect and the maximum saturation were also highly variable. Therefore, while the catenation effect is sensitively affected by the binding site density, an all-or-none catenation effect is observed in the general arrangement of the binding sites (Figure 6b). In particular, the improvement effect of antigen binding ability by catenation was sensitively affected by the ( K _D ) _catenator value in situations where the binding site density was high (p > 4.41) (Fig. 6b). Even greater improvements were observed by further increasing the density of randomly distributed binding sites. At a ρ of 58.8, which is about 1/200th of the density of Her2 receptors on the surface of breast cancer cells in which Her2 is overexpressed, it is improved up to 29,000 times (Figure 7). And this simulation shows that randomly distributed binding sites at high density dramatically improve the effective binding ability of ^cat Ab to the antigen. To estimate the effect of K _D on [ ^cat Ab], simulations were performed for different [ ^cat Ab]/ K _D values. Changing [ ^cat Ab]/ K _D from 0.01 to 1.0 improved antigen binding capacity from 85- to 900-fold, suggesting that the catenation effect is effective over a wide range of K _D values (Figure 8a,b).

5. Proof-of-concept experiment

For experimental verification, SDF-1α(8-67), Sly1(254-316), and HomoCC were selected as catheters. Because these are proteins that are small in size (molecular weight 20 kDa or less) and form low-affinity homodimers ( K _D 0.1 μM or more), these proteins fused to antibodies by a ~40 Å long linker are homologous within the molecule. It does not form monomers. To measure and compare how much the actual binding ability of the antibody to the antigen is improved due to fusion with the catheter, that is, the K _D value of the antibody itself to the antigen and the actual dissociation constant (( K _D ) _eff ) value of ^{the cat} Ab. Biolayer interferometry (BLI) experiments were performed. 4 types of antibodies (Trastuzumab (N30A/H91A), HetFc_Trastuzumab (N30A/H91A/Y100A), glCV30, obinutuzumab (F101A)) and 3 types of catheters (SDF-1α(8- 67), SLy1(254-316), HomoCC), and the two fusion sites of the catheter (heavy chain C-terminus or light chain C-terminus). The analysis method is to immobilize the target antigen on a biosensor tip, react with an antibody or antibody-linker, obtain the kinetics curve of the binding and dissociation process between the two, and obtain the binding force (dissociation constant) from this.

5-1. When SDF-1α (8-67) was fused to the heavy chain C-terminus of trastuzumab (N30A/H91A) or glCV30 (Figure 9a,b)

SDF-1α(8-67) was fused to two types of antibodies using a 10-residue linker (GGGGSGGGGS). It is a trastuzumab antibody that binds to glCV30 and Her2 proteins, which bind to the receptor binding domain (RBD) of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein. In the case of trastuzumab, the N30A and H91A mutations are introduced into the light chain, so the Fab region binds to the ectodomain of HER2 with a K _D of 353 nM (Slaga et al., 2018). glCV30 also binds to RBD with similar binding affinity ( K _D =407nM) (Hurlburt, et al. (2020) Nat Commun 11, 5413). Using BLI, N30A/H91A mutant trastuzumab and glCV-SDF-1α(8-67) fused to SDF-1α(8-67) and unmodified parent antibodies were produced. The parent antibodies for each, N30A/H91A mutant trastuzumab and glCV30, had K _D values of 2.1 nM and 51.2 nM, respectively (Fig. 9a,b). SDF-1α(8-67)-fusion antibodies showed binding patterns similar to those of the parent antibody. The binding rate constants ( k _a s) of the parent antibody N30A/H91A mutant trastuzumab and N30A/H91A mutant trastuzumab fused to the heavy chain C-terminus of SDF-1α(8-67) are 1.5x10 ⁵ Ms ^-1, respectively. and 3.1x10 ⁵ Ms ^-1 . Similarly, the k _a s of glCV30 and glCV30-SDF-1α ^(H) were 25.0x10 ⁴ Ms ^-1 and 5.0x10 ⁴ Ms ^-1 , respectively. However, the antibody fused with the parent antibody and SDF-1α showed a significantly different dissociation pattern. In the case of N30A/H91A mutant trastuzumab, the parent antibody was 3.1x10 ^-4 Ms ^-1 , whereas the SDF-1α(8-67) fused antibody was less than 1.0x10 ^-7 Ms ^-1 . In the case of glCV30, the parent antibody was 1.3x10 ^-4 Ms ^-1 , whereas the SDF-1α(8-67) fused antibody was less than 1.0x10 ^-7 Ms ^-1 (FIGS. 9a,b). These observed dynamics are consistent with the expectation that fused SDF-1α will not affect antibody binding and that in high concentration solutions, SDF-1α will inhibit the dissociation of fused antibody molecules due to catenation. As a result, the N30A/H91A mutant trastuzumab fused with SDF-1α showed a K _D of less than 10 pM, showing a total binding capacity more than 210 times higher than that of the parent antibody, and in the case of glCV30, the total binding capacity was higher due to the fusion of SDF-1α. The binding force increased 5120 times. In other words, by fusing homodimer-forming proteins that do not have a strong affinity, the effect of improving the total binding force from double-digit nM to pM was achieved.

5-2. When SDF-1α (8-67) was fused to the light chain C-terminus of trastuzumab (N30A/H91A) (Figure 10a)

In this case as well, dissociation of Trastuzumab (N30A/H91A)-SDF-1α(8-67) ^(L) was not observed in the dissociation phase, and therefore its binding capacity (( K _D ) _eff < 10 pM) It increased at least 210-fold compared to the binding affinity of Zumab (N30A/H91A) ( K _D = 2.1 nM) (FIG. 10a).

5-3. When HomoCC was fused to the heavy chain C-terminus of trastuzumab (N30A/H91A) (Figure 10b)

Even in this case, in the dissociation phase, the binding affinity of Trastuzumab (N30A/H91A)-HomoCC ^(H) (( K _D ) _eff < 10 pM) is lower than that of Trastuzumab (N30A/H91A) ( K _D = 2.1 nM). It increased at least 210 times compared to (FIG. 10b).

5-4. When SDF-1α (8-67) was fused to the heavy chain C-terminus of obinutuzumab (F101A) (Figure 11)

Obinutuzumab, an anti-lymphoma antibody (Kumar et al., 2020, Mϕssner et al., 2010), shows strong binding affinity ( K _D = 0.18 nM) to its target antigen, CD20, whereas obinutuzumab (F101A) shows a binding force so weak that it cannot be measured by BLI. This is because the F101A mutation seriously affected binding ability. However, when this mutant antibody is made in the form of obinutuzumab (F101A)-SDF-1α(8-67) ^(H), its binding affinity increases dramatically (( K _D ) _eff = 2.09 nM) (FIG. 11).

5-5. When SDF-1α (8-67) and SLy1 (254-316) were fused to the heavy chain C-terminus of trastuzumab (N30A/H91A/Y100A) with heterodimeric Fc (Figure 12)

Wild-type Fc forms a homodimer, and attempts were made to create heterodimeric Fc to make a bispecific antibody, and a knobs-into-holes type mutation (knob) was added to the C _H3 domain of the heavy chain. : T366W; hole: T366S, ^L358V , Y407A) was reported to form a heterodimer well (Leaver-Fay et al., 2016). The present inventors prepared an antibody (HetFc_Trastuzumab(N30A/H91A/Y100A)) in which the Fab of Trastuzumab (N30A/H91A/Y100A) is attached to this heterodimeric Fc and also binds SLy1(254-316) to KH Fc. , antibody-catenator (SLy1(254-316) ^(H) -HetFc_Trastuzumab (N30A/H91A/Y100A)-SDF-1α fused to KH ^-1 Fc and SDF-1α(8-67), respectively. (8-67) ^(H) ) was prepared. The binding affinity of HetFc_Trastuzumab(N30A/H91A/Y100A) to HER2 is very weak ( K _D = 243 nM), but in comparison, SLy1(254-316) ^(H) -HetFc_Trastuzumab(N30A/H91A/Y100A)-SDF The binding force of -1α(8-67) ^(H) increased approximately 300 times (( K _D ) _eff = 0.83 nM) (FIG. 12). This result shows that even antibodies in the form of double antibodies can amplify their binding force to antigens by fusing a catheter.

In summary, antibody catenation is a method that can increase the actual binding capacity of an antibody to an antigen hundreds to thousands of times, regardless of the type of antibody and regardless of the fusion position of the catheter. In addition, since the improvement in actual binding ability is shown to be independent of the type of catheter, this method is not limited to specific IgG antibodies and can be said to be applicable to all IgG antibodies.

6. Confirmation of amplification of antibody-catenator binding force in cancer cells

The properties of catenated antibodies were analyzed using cancer cells. The binding force of the antibody-cathenator to the antigen increases in proportion to the density of the antigen. In general, anticancer target molecules located on the cell surface are expressed in relatively greater quantities in cancer cells than in normal cells, so the antibody-catenator is more sensitive to the number of antigens, that is, the surface of cancer cells with a high density of antigens rather than the surface of normal cells with a low density of antigens. will combine better. To demonstrate this, experiments were performed on a breast cancer cell line that grew attached to a surface and a blood cancer cell line that grew in suspension.

6-1. Enhanced binding of catenated antibodies to breast cancer cell lines (Figure 13)

As a result of treating four types of breast cancer cell lines with different expression of the target antigen HER2 with Trastuzumab (N30A/H91A) and Trastuzumab (N30A/H91A)-SDF-1α(8-67) ^(H) , Trastuzumab Trastuzumab (N30A/H91A)-SDF-1α(8-67) ^(H) binds much better than zumab (N30A/H91A), and the degree of binding increases in proportion to the expression level of HER2 (Figure 13) .

6-2. Selective binding ability to breast cancer cell lines by antibody catenation (Figure 14)

When trastuzumab (N30A/H91A) was treated with two breast cancer cell lines, MCF-7 and BT-474 (Figure 14, top left), which have different expression of the target antigen HER2, the expression of HER2 was higher than that of MCF-7, as expected. It was seen that the trastuzumab (N30A/H91A) antibody bound better to BT-474, which had a higher degree of binding. When treated with Trastuzumab (N30A/H91A)-SDF-1α(8-67) ^(H) at the same concentration, this catenated antibody inhibited BT-474 cells significantly better than the non-catenated antibody. It came together strongly and quickly. However, the binding could not be observed in MCF-7 cells (Figure 14). In other words, catenated antibodies have the property of selectively binding to cells with high antigen density even though they can simultaneously access cells with different target antigen densities. This result demonstrates that the antibody-catenator has the ability to bind to cancer cells but not to normal cells.

6-3. Enhanced binding of catenated antibodies to blood cancer cell lines (Figure 15)

Because su-DHL-5, a lymphoma cell line, is a suspended cell, the cell-binding activity of the antibody was analyzed using FACS (fluorescence assisted cell sorting) rather than confocal microscopy. Obinutuzumab (F101A) antibody has very low binding affinity to CD20 and does not show binding to su-DHL-5 cells, whereas obinutuzumab (F101A)-SDF-1α (8-67) catenalated here ) ^(H) Even at a low concentration (100 nM), 37% of the total antibody was shown to bind to DHL-5 cells (FIG. 15).

7. Confirmation of increased virus neutralizing activity of antibody-catheter

The SARS-CoV-2 neutralizing activity of CV30, glCV30, and glCV30-SDF-1α(8-67) was compared using vesicular stomatitis virus (VSV) pseudotyped with the SARS-CoV-2 spike protein. . This virus has multiple copies of the spike protein on its envelope to which glCV30-SDF-1α (8-67) molecules can attach. Each of the three antibodies (CV30, glCV30, and glCV30-SDF-1α(8-67)) was serially diluted and added to rVSV-ΔG-Luc containing the SARS-CoV-2 spike protein of the Wuhan-Hu-1 strain; , the mixture was incubated with HEK293T-hACE2 cells and the luciferase activity was measured.

As a result, as shown in Figure 16, CV30, glCV30, and glCV30-SDF-1α had IC ₅₀ values of 0.25, 3.22, and 0.21 μg/ml, respectively. That is, compared to glCV30 (IC ₅₀ of 3.22 μg/ml), glCV30-SDF-1α(8-67) (IC ₅₀ of 0.21 μg/ml) showed ~15-fold lower IC ₅₀ value. In addition, this neutralizing efficacy of glCV30-SDF-1α(8-67) (IC ₅₀ of 0.21 μg/ml) was similar to the neutralizing ability of CV30 (IC ₅₀ of 0.25 μg/ml), which increased the antigen-binding capacity of glCV30. The dissociation constant of CV30, when measured through BLI, is 10 pM or less, similar to glCV30-SDF-1α. These results demonstrate that an antibody with weak binding to an antigen can be converted into a strong antibody by fusing the catheter.

From the above description, those skilled in the art will understand that the present invention can be implemented in other specific forms without changing its technical idea or essential features. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive. The scope of the present invention should be construed as including the meaning and scope of the patent claims described below rather than the detailed description above, and all changes or modified forms derived from the equivalent concept thereof are included in the scope of the present invention.

Claims (12)

A conjugate comprising (i) an antibody or fragment thereof, and (ii) a catenator polypeptide fused to two heavy chain C-termini or two light chain C-termini, respectively, of the antibody or fragment,
A conjugate wherein the catenator polypeptide satisfies the following conditions:
(a) capable of forming homodimers;
(b) the dissociation constant ( K _D ) of homodimer formation is 0.1 μM to 500 μM;
(c) molecular weight is 3 kDa to 20 kDa; and
(d) On the surface where target antigens to which the antibody specifically binds exist, homodimers may be formed between the catheters included in two or more of the conjugate molecules, thereby linking the conjugate molecules to each other. According to paragraph 1,
A conjugate wherein the catheter polypeptide is connected to an antibody or a fragment thereof through a linker. According to paragraph 1,
A conjugate wherein the catenator polypeptide is a polypeptide of SEQ ID NO: 1, SEQ ID NO: 2, or SEQ ID NO: 3. According to paragraph 1,
A conjugate wherein the antibody is an IgG type antibody. According to paragraph 1,
A conjugate wherein the antibody is a therapeutic antibody, diagnostic antibody, monoclonal antibody, chimeric antibody, humanized antibody, or human antibody. According to paragraph 1,
A conjugate wherein the antibody is an antibody of an antibody-drug conjugate (ADC). According to paragraph 1,
A conjugate wherein the antibody is a multispecific antibody. According to paragraph 1,
A conjugate wherein the fragment of the antibody is Fc or F(ab') ₂ . According to clause 8,
A conjugate wherein the fragment of the antibody is an Fc fragment of a drug-Fc fragment conjugate. According to paragraph 1,
A conjugate wherein the antibody is a secondary antibody in a sandwich immunoassay. A composition for enhancing antibody-antigen binding, comprising the conjugate of any one of claims 1 to 10,
On the surface where the target antigens to which the antibody contained in the conjugate specifically binds exist, a homodimer is formed between the catheters contained in two or more of the conjugate molecules, and the conjugate molecules are connected to each other. , A composition characterized in that dissociation of antibodies is thereby inhibited. A method for enhancing antibody-antigen binding, comprising the step of treating the conjugate of any one of claims 1 to 10 on a target surface where antigens to which the antibody contained in the conjugate specifically binds exist. ,
A method characterized in that a homodimer is formed between the catheters included in two or more conjugate molecules, thereby linking the conjugate molecules to each other, thereby suppressing dissociation of the antibody.