WO2021235509A1

WO2021235509A1 - Antigen-binding molecule

Info

Publication number: WO2021235509A1
Application number: PCT/JP2021/019099
Authority: WO
Inventors: 達也野中; 有起角田; 咲妃奥出
Original assignee: 中外製薬株式会社
Priority date: 2020-05-20
Filing date: 2021-05-20
Publication date: 2021-11-25
Also published as: JP2023116826A

Abstract

An antigen-binding molecule that specifically binds to a complex, said complex being composed of a co-suppression molecule localized in an immune synapse and a ligand thereof, and thus targets the immune synapse. Preferably, the antigen-binding molecule has an agonistic activity to the co-suppression molecule. The antigen-binding molecule is usable not only in a medicine having an immunosuppression action based on the agonistic activity to the co-suppression molecule but also for detecting a complex composed of the co-suppression molecule and a ligand thereof.

Description

Antigen binding molecule

The present invention includes an antigen-binding molecule, a co-suppressing molecule agonist, a co-suppressing molecule signal activator, a pharmaceutical composition, a detection agent for a complex consisting of a co-suppressing molecule and its ligand, a method for suppressing immunity, and a co-suppressing molecule. The present invention relates to a method for detecting a complex composed of the ligand.

Pharmacologically targeted drugs that exist at the immunological synapse are attracting attention (Non-Patent Document 1). Among such molecules, therapies targeting co-suppressive molecules are aimed at cancer treatment. For example, antibodies targeting CTLA-4 and PD-1 have been approved by the US Food and Drug Administration (FDA), and clinical trials are underway for treatments that block TIM-3, LAG3, and TIGIT.

Although the efficacy of agonists for co-suppressive molecules has been shown in mouse models of autoimmune diseases, no agonist has been shown to be effective in human diseases (Non-Patent Document 2).

Agonists for PD-1, one of the co-suppressive molecules, have been tested.
For example, an antibody called PD-1 AB-6 is disclosed to bind to a β-sheet with a PD-1 loop composed of PD-1 residues 100-105 (Patent Document 1). The literature discloses that PD-1 AB-6 is designed to provide PD-1 mediated agonists.
Further, it is disclosed that a substance having specificity for human PD-1 is an excellent substance capable of transmitting a human PD-1 signal (Patent Document 2). However, it has not been experimentally shown whether or not the bispecific antibody having reactivity with PD-1 and CD3 used in the examples of the literature has agonist activity against PD-1.

Currently, Phase 1 of a clinical trial for psoriasis using a PD-1 agonist antibody called CC-90006 is in progress (Non-Patent Document 3).
Recently, Phase 1 of a clinical trial targeting autoimmune diseases using ONO-4685, which is a bispecific antibody targeting both PD-1 and CD3, has been started (Non-Patent Documents 4 and 5). ). ONO-4685 acts as a PD-1 agonist by cross-linking PD-1 and CD3 expressed on the same T cell, and cross-links CD3 expressed on a cell expressing PD-1 and another cell. Therefore, it is expected to have an action of damaging T cells (Non-Patent Document 6).

International Publication No. 2018/053405 International Publication No. 2004/072286

It cannot be said that the medicines for treating or preventing diseases including autoimmune diseases are sufficiently satisfied. Therefore, it is expected that further therapeutic agents or candidates for treating or preventing diseases including autoimmune diseases will be created.

The present inventors have created and experimentally confirmed the concept of an antigen-binding molecule targeting an immunological synapse that has potential efficacy against autoimmune diseases. Specifically, the antigen-binding molecule targets the immunological synapse by specifically binding to a complex consisting of a co-suppressing molecule localized at the immunological synapse and its ligand. The antigen-binding molecule preferably has agonistic activity against the co-suppressing molecule. The antigen-binding molecule can be provided for detection of a complex consisting of a co-suppressing molecule and a ligand thereof, in addition to a drug having an immunosuppressive action based on an agonist activity against the co-suppressing molecule.

As one aspect of the present specification, the following invention is provided.
[1] An antigen-binding molecule containing a first antigen-binding domain.
An antigen-binding molecule in which the first antigen-binding domain can specifically bind to a first complex consisting of a first co-suppressing molecule and a first co-suppressing molecular ligand for the first co-suppressing molecule.
[2] The specific binding of the first antigen-binding domain to the first complex is either the first co-suppressing molecule or the first co-suppressing molecular ligand in the first complex. The antigen-binding molecule of [1], which is an intermolecular force bond with one or both.
[3] The antigen-binding molecule of [1] or [2] capable of activating a first intracellular signal downstream of the first co-suppressing molecule at the immunological synapse in which the first complex is present.
[4] The antigen-binding molecule of [3], wherein activation of T cells can be suppressed by activation of the first intracellular signal.
[5] The immunological synapse is formed between the cells expressing the MHC and the co-suppressing molecular ligand and the T cells expressing the co-suppressing molecule, [3] or [ 4] Antigen-binding molecule.
[6] An antigen-binding molecule containing a first antigen-binding domain.
The first antigen-binding domain may bind to a first complex consisting of a first co-suppressing molecule and a first co-suppressing molecular ligand for the first co-suppressing molecule, and said first antigen-binding domain. The binding activity to the first complex of the first co-suppressing molecule that does not form the first complex and the first co-suppressing molecular ligand that does not form the first complex. An antigen-binding molecule that has a higher binding activity to either or both.
[7] The binding of the first antigen-binding domain to the first complex is either one of the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex, or The antigen-binding molecule of [6], which is an intermolecular force bond with both.
[8] The antigen-binding molecule of [6] or [7] capable of activating a first intracellular signal downstream of the first co-suppressing molecule at the immunological synapse in which the first complex is present.
[9] The antigen-binding molecule of [8], wherein the activation of T cells can be suppressed by the activation of the first intracellular signal.
[10] The immunological synapse is formed between the cells expressing the MHC and the co-suppressing molecular ligand and the T cells expressing the co-suppressing molecule, [8] or [ 9] Antigen-binding molecule.
[11] The antigen-binding molecule according to any one of [6] to [10], wherein the binding activity satisfies any one or both of the following (a) and (b):
(a) In flow cytometry using a first cell that forcibly expresses the first co-suppressing molecule and does not forcibly express the first co-suppressing molecular ligand, the cell of the first co-suppressing molecular ligand. The binding activity of the first antigen-binding domain to the first cell in the presence of the first soluble polypeptide containing the outer domain or a part thereof is the non-binding activity of the first soluble polypeptide. Higher than in existence;
(b) In flow cytometry using a second cell forcibly expressing the first co-suppressing molecular ligand and not forcibly expressing the first co-suppressing molecule, extracellular of the first co-suppressing molecule. The binding activity of the first antigen-binding domain to the second cell in the presence of the second soluble polypeptide containing the domain or a part thereof is the absence of the second soluble polypeptide. Higher than below.
[12] The antigen-binding molecule of [10], wherein both the first cell and the second cell are derived from Chinese hamster ovary cells.
[13] The antigen-binding molecule according to any one of [6] to [10], the binding activity of which is measured by biolayer interference.
[14] An antigen-binding molecule containing a first antigen-binding domain.
An antigen capable of specifically binding the first antigen-binding domain to an epitope present in a first complex consisting of a first co-suppressing molecule and a first co-suppressing molecular ligand for the first co-suppressing molecule. Binding molecule.
[15] Only the first co-suppressing molecule that does not form the first complex or only the first co-suppressing molecular ligand that does not form the first complex in the entire region of the epitope. The antigen-binding molecule of [14], which is not satisfied with.
[16] The specific binding of the first antigen-binding domain to the first complex is either the first co-suppressing molecule or the first co-suppressing molecular ligand in the first complex. The antigen-binding molecule of [14] or [15], which is an intermolecular force bond with one or both.
[17] The antigen binding according to any one of [14] to [16] capable of activating a first intracellular signal downstream of the first co-suppressing molecule at the immunological synapse in which the first complex is present. molecule.
[18] The antigen-binding molecule of [17], wherein the activation of T cells can be suppressed by the activation of the first intracellular signal.
[19] The immunological synapse is formed between the cells expressing the MHC and the co-suppressing molecular ligand and the T cells expressing the co-suppressing molecule, [17] or [ 18] antigen-binding molecule.
[20] The epitope to which the first co-suppressing molecule is PD-1, the first co-suppressing molecular ligand is PD-L1, and the first antigen-binding domain is bound is in Example 7. [1] to [1] to [1] to [1] to which any one of the anti-human PD-L1 / human PD-1 complex protein antibodies shown in Table 1 prepared partially overlaps with the epitope in the first complex to which it binds. 19] Antigen-binding molecule.
[21] A third cell expressing the first co-suppressing molecule and capable of measuring the intensity of the first intracellular signal, an antigen-independent T cell receptor activator, and the first co-suppressing molecule. It is used in a state where the fourth cell expressing the ligand can contact each other, and the first intracellular signal in the third cell is partially produced by an inhibitory antibody against the first co-suppressing molecular ligand. In the system for measuring the intensity of the first intracellular signal, which is performed under suppressed conditions, the intensity of the first intracellular signal in the presence of the antigen-binding molecule is not the same as that of the antigen-binding molecule. The antigen-binding molecule of any of [3] to [5], [8] to [10], and [17] to [19], which is higher than in the presence.
[22] The antigen-binding molecule of [21], wherein the third cell is derived from a Jurkat cell and the fourth cell is derived from a Chinese hamster ovary cell.
[23] The intensity of the first intracellular signal directly or directly to the intracellular domain and the intracellular domain of the first co-suppressing molecule, which occurs when the first co-suppressing molecule is activated. The antigen-binding molecule of [21] or [22], as measured based on its proximity to the intracellular protein that interacts indirectly.
[24] The antigen-binding molecule of [23], wherein the intracellular protein is a dephosphorylation enzyme.
[25] The intensity of the second intracellular signal that expresses the T cell receptor and the first co-suppressive molecule and can be activated downstream of the T cell receptor can be measured, and the first one. A fifth cell in which the second intracellular signal can be suppressed by activation of the intracellular signal and a sixth cell expressing the antigen-independent T cell receptor activator and the first co-suppressing molecular ligand. Are used under conditions where they can contact each other, and
Suppression of the second intracellular signal in the fifth cell by the first co-suppressing molecular ligand is carried out under conditions where it is partially suppressed by an inhibitory antibody against the first co-suppressing molecular ligand. In the second intracellular signal intensity measurement system,
The intensity of the second intracellular signal in the presence of the antigen-binding molecule is lower than that in the absence of the antigen-binding molecule, [3] to [5], [8] to [10], [ An antigen-binding molecule according to any one of 17] to [19] and [21] to [24].
[26] The antigen-binding molecule of [25], wherein the fifth cell is derived from a Jurkat cell and the sixth cell is derived from a Chinese hamster ovary cell.
[27] The antigen-binding molecule of [25] or [26], wherein the second intracellular signal is NFAT activity.
[28] The combination of the first co-suppressing molecule and the first co-suppressing molecular ligand is a combination of PD-1 and PD-L1, a combination of PD-1 and PD-L2, a combination of BTLA and HVEM, and TIGIT. And CD155, TIGIT and CD112, LAG-3 and MHC class II molecules, CTLA4 and CD80, CTLA4 and CD86, TIM-3 and galectin-9, TIM-3 and phosphatidylserine. Any one selected from the group consisting of combinations, combinations of TIM-3 and CEACAM-1, and combinations of TIM-3 and HMGB1, [1] to [19], and [21] to [27]. Any of the antigen-binding molecules.
[29] The antigen-binding molecule is a multiple antigen-binding molecule further comprising a second antigen-binding domain.
Any of [1] to [28], wherein the second antigen-binding domain can specifically bind to a second co-suppressing molecule capable of forming a second complex with the second co-suppressing molecular ligand. Antigen-binding molecule.
[30] The antigen-binding molecule of [29] capable of activating a third intracellular signal downstream of the second co-suppressing molecule at the immunological synapse in which the first complex is present.
[31] The antigen-binding molecule of [30], wherein the activation of the third intracellular signal can suppress the second intracellular signal.
[32] From [29], the binding of the second antigen-binding domain to the second co-suppressing molecule does not compete with the binding of the second co-suppressing molecular ligand to the second co-suppressing molecule. Any of the antigen-binding molecules of [31].
[33] The combination of the second co-suppressing molecule and the second co-suppressing molecular ligand is a combination of PD-1 and PD-L1, a combination of PD-1 and PD-L2, a combination of BTLA and HVEM, and TIGIT. And CD155, TIGIT and CD112, LAG-3 and MHC class II molecules, CTLA4 and CD80, CTLA4 and CD86, TIM-3 and galectin-9, TIM-3 and phosphatidylserine. The antigen-binding molecule of any of [29] to [32], which is any combination selected from the group consisting of combinations, combinations of TIM-3 and CEACAM-1, and combinations of TIM-3 and HMGB1.
[34] The antigen binding according to any one of [29] to [33], wherein the first antigen-binding domain and the second antigen-binding domain are a variable region of an antibody or a fragment thereof that binds to a target antigen thereof. molecule.
[35] The antigen-binding molecule according to any one of [29] to [34], wherein the first co-suppressing molecule is the same as or different from the second co-suppressing molecule.
[36] When the first co-suppressing molecule is the same as the second co-suppressing molecule, the epitope in the first co-suppressing molecule to which the first antigen-binding domain binds is the second antigen. The antigen-binding molecule of [35] that does not overlap, or partially or partially overlaps, the epitope in the second co-suppressing molecule to which the binding domain binds.
[37] The antigen-binding molecule according to any one of [29] to [36], wherein the first co-suppressing molecular ligand is the same as or different from the second co-suppressing molecular ligand.
[38] The antigen-binding molecule of either [30] or [31], wherein the first intracellular signal is the same as or different from the third intracellular signal.
[39] The antigen-binding molecule according to any one of [1] to [38], further comprising a constant region of an antibody.
[40] The antigen-binding molecule of [33], wherein at least one residue in the constant region is different from the amino acid residue at the corresponding position in the constant region of the native antibody.
[41] A co-suppressing molecular agonist comprising any of the antigen-binding molecules of [1] to [40].
[42] A co-suppressing molecular signal activator containing the antigen-binding molecule according to any one of [1] to [40].
[43] A pharmaceutical composition containing the antigen-binding molecule according to any one of [1] to [40].
[44] The pharmaceutical composition of [43] for treating or preventing a disease caused by an abnormal increase in immunity.
[45] A first complex comprising a first co-suppressing molecule and a first co-suppressing molecular ligand for the first co-suppressing molecule, which comprises any of the antigen-binding molecules of [1] to [40]. Detection agent.
[46] A method for suppressing immunity, which comprises administering an antigen-binding molecule according to any one of [1] to [40] to a subject.
[47] From the first co-suppressing molecule and the first co-suppressing molecular ligand to the first co-suppressing molecule, which comprises contacting the immune cell with any of the antigen-binding molecules of [1] to [40]. First method for detecting a complex.
[48] A method for producing an antigen-binding molecule, which comprises culturing a cell containing a nucleic acid encoding the antigen-binding molecule according to any one of [1] to [40].
[49] From a non-human animal immunized with a third complex comprising a first co-suppressing molecule or a partial polypeptide thereof and a first co-suppressing molecular ligand for the first co-suppressing molecule or a partial polypeptide thereof. A method for producing an antigen-binding molecule, which comprises collecting plasma or blood containing any of the antigen-binding molecules from [1] to [40].
[50] The third complex binds the first co-suppressing molecule or a partial polypeptide thereof to the first co-suppressing molecular ligand or a partial polypeptide thereof via a linker or directly by a chemical bond. The method for producing [49], which is one molecule.
[51] The method for producing [49] or [50], further comprising obtaining a composition containing the antigen-binding molecule with increased purity from the plasma or the blood.
[52] The production method according to any one of [49] to [51], wherein the antigen-binding molecule is a polyclonal antibody.
[53] From a non-human animal immunized with a third complex comprising a first co-suppressing molecule or a partial polypeptide thereof and a first co-suppressing molecular ligand for the first co-suppressing molecule or a partial polypeptide thereof. A method for producing an antigen-binding molecule, which comprises isolating a cell clone expressing any of the antigen-binding molecules [1] to [40].
[54] The third complex binds the first co-suppressing molecule or a partial polypeptide thereof to the first co-suppressing molecular ligand or a partial polypeptide thereof via a linker or directly by a chemical bond. The method for producing [53], which is one molecule.
[55] The method for producing [53] or [54], further comprising extracting and isolating the nucleic acid encoding the antigen-binding molecule from the cell clone.
[56] The production method of [55], further comprising recombination of the nucleic acid.
[57] The production method according to [56], further comprising introducing the recombinant nucleic acid into another cell and expressing the antigen-binding molecule in the other cell.
[58] The method for producing [57], further comprising obtaining a composition containing the expressed antigen-binding molecule.
[59] The method for producing [58], further comprising increasing the purity of the expressed antigen-binding molecule in the composition.

The antigen-binding molecule in the present invention targets the immunological synapse by specifically binding to a complex consisting of a co-suppressing molecule localized at the immunological synapse and its ligand, and has agonist activity against the co-suppressing molecule. It can be a therapeutic agent or a candidate for the treatment or prevention of diseases including autoimmune diseases.

FIG. 1 shows PD-1 signal induction (SHP-2) of a bispecific antibody containing an anti-PD-1 arm and an anti-CD3 arm as measured by an SHP-2 recruitment assay system using NanoBRET®. It is a figure which shows the evaluation result of recruitment) activity. "OKT3 // 949" is a bispecific antibody containing an arm derived from OKT3 and an arm derived from clone 949, and "CE115TR // 949" is a bispecific antibody containing an arm derived from CE115TR and an arm derived from clone 949. Means. FIG. 2 is a diagram showing the results of evaluating the effects of bispecific antibodies containing anti-PD-1 arm and anti-human CD3 arm on the proliferation of human CD4-positive T cells by the T cell proliferation assay. "OKT3 / IC17" includes a bispecific antibody containing an arm derived from OKT3 and an arm derived from an anti-KLH antibody, and "OKT3 / PD1-17 (7J13)" includes an arm derived from OKT3 and an arm derived from PD1-17. Bispecific antibody, "OKT3 / clone10" is a bispecific antibody containing an arm derived from OKT3 and an arm derived from clone10, and "OKT3 / clone949" is a double containing an arm derived from OKT3 and an arm derived from clone949. Means specific antibody. Of these, OKT3 / IC17 was used as a negative control. FIG. 3 is a diagram showing the design of the prepared linker-fused human PD-L1 / human PD-1 complex protein. The figure on the left shows the design of Linker complex # 1 with a linker inserted between N131 of human PD-L1 and N33 of human PD-1, and the figure on the right shows the design of E146 of human PD-1 and F19 of human PD-L1. Shows Linker complex # 2 with a linker inserted between them. FIG. 4 is a diagram showing the design of the prepared disulfide bond-introduced human PD-L1 / human PD-1 complex protein. SS complex # 1 which is a combination of Y56C variant of human PD-L1 and A132C variant of human PD-1, and SS complex which is a combination of A18C variant of human PD-L1 and G90C variant of human PD-1. # 2 was created. FIG. 5 is a diagram showing the results of specificity evaluation of the anti-human PD-L1 / human PD-1 complex antibody. 30 clones (LPB0006, LPB0010, LPB0017, LPB0036, LPB0038, LPC0001, LPC0002, LPC0006, LPC0007, LPC0008, LPC0011, LPC0012, LPC0017, LPC0019, LPC0020, LPC0039, LPC0063, LPC0072, LPD unable , LPE0015, LPE0017, LPE0024, LPE0027, LPE0065, LPE0076) were shown to bind to the complex of human PD-L1 bound to human PD-1. It is a figure which shows the continuation of FIG. 5-1. FIG. 6 shows the PD-1 signal-induced (SHP-2 recruitment) activity of anti-human PD-1 antibodies (clone949 and PDA0129) as measured by the SHP-2 recruitment assay system using NanoBRET®. It is a figure which shows the evaluation result. “129 homo” is the anti-human PD-1 antibody PDA0129 prepared in Example 9, and “949 homo” is the anti-human PD-1 antibody clone 949 prepared in Example 1. FIG. 7 shows PD-1 signal induction (SHP) of anti-human PD-L1 / human PD-1 complex antibodies (LPC0039 and LPB0006) measured by the SHP-2 recruitment assay system using NanoBRET®. -2 Recruitment) It is a figure which shows the evaluation result of the activity. "LPC0039homo" and "LPB0006homo" are anti-human PD-L1 / human PD-1 complex antibodies prepared in Example 7. FIG. 8 is a diagram showing the evaluation results of the inhibitory activity of anti-human PD-1 antibodies (clone 949, PDA0129, clone 10, PD1-17) on the binding between human PD-1 and human PD-L1. PD-1 inhibitory antibodies 5C4 and Pembrolizumab were used as positive controls. Anti-KLH (anti-KLH antibody), which does not bind to either PD-1 or PD-L1, was used as a negative control. FIG. 9 shows PD of a bispecific antibody containing an anti-PD-L1 / PD-1 complex arm and an anti-PD-1 arm as measured by an SHP-2 recruitment assay system using NanoBRET®. It is a figure which shows the evaluation result of -1 signal induction (SHP-2 recruitment) activity. “LPE0024”, “LPB0017”, “LPC0039”, “LPC0020”, and “LPC0017” mean arms from the anti-human PD-L1 / human PD-1 complex antibody prepared in Example 7, and “949” and “129”. Means the arms from the anti-human PD-1 antibodies clone 949 and PDA0129, respectively. It is a figure which shows the continuation of FIG. 9A. FIG. 10 is a diagram showing the evaluation results of suppression of NFAT activity by a bispecific antibody containing an anti-PD-L1 / PD-1 complex arm and an anti-PD-1 arm. “LPB0010”, “LPB0017”, “LPC0039”, “LPE0024”, “LPC0011”, “LPC0020”, “LPC0001”, and “LPB0010” are arms from the anti-human PD-L1 / human PD-1 complex antibody prepared in Example 7. Meaning, "Clone 949" and "PDA0129" mean an arm from an anti-human PD-1 antibody. FIG. 11 is a diagram illustrating the design of each molecular type concept and the function of each arm examined in this embodiment. FIG. 12 is a diagram illustrating the mechanism of action of each molecular type concept examined in this embodiment. FIG. 13 shows a Fab fragment of anti-human PD-L1 / human PD-1 complex antibody LPB0010HCb-G1T7P / LPB0010LCb-k0MTC and human PD-L1 extracellular domain / human PD-1 cells as shown in Example 18. It is a figure which shows the whole structure of the X-ray crystal structure of the complex with the extracellular domain. LPB0010Fab is shown as a ribbon (black: heavy chain, light gray: light chain), and hPD-L1 / hPD-1 complex is shown as a surface representation (gray: hPD-L1, white: hPD-1). FIG. 14 is a diagram showing the mapping of LPB0010Fab epitopes onto the human PD-L1 extracellular domain / human PD-1 extracellular domain complex crystal structure as shown in Example 18 (gray:). hPD-L1, white: hPD-1). Black highlights indicate amino acids within 4.5 Å of LPB0010Fab. The figure on the left shows the mapping of the interaction interface seen from the front, and the figure on the right shows the mapping seen from the direction rotated 90 degrees in the X-axis direction. FIG. 15A is a diagram showing the mapping of the epitope of LPB0010Fab onto the amino acid sequence of the human PD-L1 extracellular domain (amino acid number: 19-238) as shown in Example 18. Black highlights indicate amino acids within 4.5 Å of LPB0010Fab. FIG. 15B is a diagram showing the mapping of the epitope of LPB0010Fab onto the amino acid sequence of the human PD-1 extracellular domain (amino acid number: 21-167) as shown in Example 18. Black highlights indicate amino acids within 4.5 Å of LPB0010Fab. Section A) shows the top of the two-dimensional averaging of negatively stained electron micrographs of the complex of LPB0010Fab fragment and human PD-L1 extracellular domain / human PD-1 extracellular domain as shown in Example 19. It is a figure which shows the image of 5 classes. Figure B) shows the X-ray crystal structure and the previously reported PD-L1 structure (PDB) of the complex of LPB0010Fab fragment and human PD-L1 extracellular domain / human PD-1 extracellular domain as shown in Example 18. It is a figure which looked at the model structure created by superimposing with ID = 4Z18) from various directions so as to correspond to the image of each class of two-dimensional averaging of a negative-stained electron microscope image. FIG. 17 is a two-dimensional negatively stained electron microscope image of a complex of LPC0039Fab fragment and human PD-L1 extracellular domain / human PD-1 extracellular domain (SS complex # 2) as shown in Example 19. It is a figure which shows the image of the top 5 classes of averaging.

A. Definitions As used herein, a "co-inhibitory molecule" is a membrane protein expressed on T cells, and is intracellular that suppresses T cell activity or activation by specific binding of a co-inhibitor molecule ligand expressed on antigen-presenting cells. A molecule that produces a signal.
As used herein, the term "co-suppressing molecular ligand" is a membrane protein expressed on an antigen-presenting cell, and the co-suppressing molecule is applied to T cells by specific binding to the co-suppressing molecule expressed on T cells. Mediated by a molecule that produces an intracellular signal that suppresses the activation or activation of the T cell.
As used herein, the term "complex consisting of a co-suppressing molecule and a co-suppressing molecular ligand" means that the co-suppressing molecular ligand is the co-suppressing molecular ligand when the co-suppressing molecule is a combination capable of binding to the co-suppressing molecular ligand. It refers to a complex formed by those molecules in a state of being bound to the molecule.
As used herein, the "intermolecular force" is mainly an electromagnetic force acting between one molecule and another. Examples of intramolecular forces include ionic interactions, hydrogen bonds, dipole interactions, and van der Waals forces. Intramolecular forces do not include covalent bonds. In macromolecules, intermolecular forces may occur between different parts of the molecule rather than covalent bonds.
As used herein, "binding activity" is used to describe the strength of intramolecular force. The binding activity of one molecule to another is determined by the sum of the various intramolecular forces that occur between those molecules. Flow cytometry (FCM), surface plasmon resonance (SPR), biolayer interference (BLI), and enzyme-linked immunosorbent assay (ELISA) can be adopted as common techniques for measuring intermolecular binding activity. ..
As used herein, the term "specifically binding" refers to binding to the target antigen when, for example, the binding activity to the target antigen is higher than the binding activity to the other antigen. In the present specification, "binding activity" is defined by "i) binding activity" in "(b-2) second aspect" described later.
As used herein, an "epitogen" is an antigen-binding molecule, an antigen-binding domain, or an antigen atom that forms an intermolecular force bond with the antigen-binding molecule, antigen-binding domain, or antibody when the antibody binds to the antigen. A region or region group on the surface of a molecule containing a group. Also called an antigenic determinant.
"Intramolecular three-dimensional structure change" as used herein is a change in one or both of the three-dimensional structure caused by the binding of a co-suppressing molecular ligand to a co-suppressing molecule (Structure, 2015, Vol.23, pp. 2341-2348. And J Biol Chem, 2013, Vol.288, pp.11771-11785.). A protein having a higher-order structure formed to have a physiological function may have a plastic region on the surface of the protein that causes a change in its three-dimensional structure by an intramolecular interaction with another protein. A change in the three-dimensional structure of a protein means a change in the relative positional relationship with other amino acid residues with respect to a specific amino acid residue in a protein having a higher-order structure formed to have a physiological function. In the present specification, it is particularly assumed that the intramolecular conformational change caused by the binding of the co-suppressing molecular ligand to the co-suppressing molecule is assumed.
As used herein, the "arm" of an antibody means a portion of a monovalent antibody similar to a naturally occurring antibody. Specifically, the arm refers to a portion of an antibody in which one heavy chain Fab and one light chain Fab are bound by a disulfide bond. The antibody containing the arm is not limited to a divalent antibody, and may be a monovalent antibody or a multivalent antibody.

As used herein, the term "antibody" is used in the broadest sense, and is not limited to, but is not limited to, a monoclonal antibody, a polyclonal antibody, and a multispecific antibody (for example,) as long as it exhibits a desired antigen-binding activity. Includes various antibody structures, including bispecific antibodies) and antibody fragments.

"Antibody fragment" refers to a molecule other than the complete antibody, which comprises a portion of the complete antibody that binds to the antigen to which the complete antibody binds. Examples of antibody fragments are, but are not limited to, Fv, Fab, Fab', Fab'-SH, F (ab') ₂ ; Diabody; Linear antibody; Single chain antibody molecule (eg, scFv). ); And contains a multispecific antibody formed from an antibody fragment.

As used herein, the term "Fc region" is used to define the C-terminal region of an immunoglobulin heavy chain that contains at least a portion of the constant region. The term includes the Fc region of a native sequence and the mutant Fc region. In one embodiment, the human IgG heavy chain Fc region extends from Cys226 or from Pro230 to the carboxyl terminus of the heavy chain. However, the C-terminal lysine (Lys447) or glycine-lysine (Gly446-Lys447) in the Fc region may or may not be present. Unless otherwise specified herein, the numbering of amino acid residues in the Fc region or constant region is Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Follow the EU numbering system (also known as the EU index) described in MD 1991.

"Framework" or "FR" refers to variable domain residues other than hypervariable region (HVR) residues. A variable domain FR usually consists of four FR domains: FR1, FR2, FR3, and FR4. Accordingly, the sequences of HVR and FR usually appear in VH (or VL) in the following order: FR1-H1 (L1) -FR2-H2 (L2) -FR3-H3 (L3) -FR4.

An "isolated" antibody is one that has been isolated from its original environmental components. In some embodiments, the antibody is subjected to, for example, electrophoresis (eg, SDS-PAGE, isoelectric focusing (IEF), capillary electrophoresis) or chromatograph (eg, ion exchange or reverse phase HPLC). Measured and purified to a purity greater than 95% or 99%. See, for example, Flatman et al., J. Chromatogr. B 848: 79-87 (2007) for a review of methods for assessing antibody purity.

The term "monoclonal antibody" as used herein refers to an antibody obtained from a substantially uniform population of antibodies. That is, the individual antibodies that make up the population are possible mutant antibodies (eg, mutant antibodies that contain naturally occurring mutations, or mutant antibodies that occur during the manufacture of monoclonal antibody preparations, such variants are usually slightly smaller. Except for the amount present), they are identical and / or bind to the same epitope. In contrast to polyclonal antibody preparations, which typically contain different antibodies against different epitopes, each monoclonal antibody in the monoclonal antibody preparation is for a single determinant on the antigen. Therefore, the modifier "monoclonal" should not be construed as requiring the production of an antibody by any particular method, indicating the characteristic of the antibody that it is obtained from a substantially homogeneous population of antibodies. For example, the monoclonal antibodies used in accordance with the present invention are, but are not limited to, hybridoma methods, recombinant DNA methods, phage display methods, transgenic animals containing all or part of the human immunoglobulin loci. It may be made by a variety of methods, including methods that utilize, such methods and other exemplary methods for making monoclonal antibodies are described herein.

"Natural antibody" refers to an immunoglobulin molecule with various naturally occurring structures. For example, a native IgG antibody is a heterotetrameric glycoprotein of approximately 150,000 daltons composed of two identical light chains that are disulfide-bonded and two identical heavy chains. From the N-terminus to the C-terminus, each heavy chain has a variable heavy chain domain or variable region (VH), also known as a heavy chain variable domain, followed by three constant domains (CH1, CH2, and CH3). Similarly, from the N-terminus to the C-terminus, each light chain has a variable region (VL), also called a variable light chain domain or a light chain variable domain, followed by a stationary light chain (CL) domain. The light chain of an antibody may be assigned to one of two types, called kappa (κ) and lambda (λ), based on the amino acid sequence of its constant domain.

The term "variable region" or "variable domain" refers to the heavy or light chain domain of an antibody involved in binding the antibody to an antigen. The heavy and light chain variable domains of native antibodies (VH and VL, respectively) are similar, with each domain usually containing four conserved framework regions (FR) and three hypervariable regions (HVR). Has a structure. (See, for example, Kindt al. Kuby Immunology, 6th ed., W.H. Freeman and Co., page 91 (2007).) One VH or VL domain may be sufficient to confer antigen binding specificity. Furthermore, an antibody that binds to a particular antigen may be isolated by screening a complementary library of VL or VH domains using the VH or VL domain from the antibody that binds to that antigen, respectively. See, for example, Portolano et al., J. Immunol. 150: 880-887 (1993); Clarkson et al., Nature 352: 624-628 (1991).

As used herein, the term "vector" refers to a nucleic acid molecule that can augment another nucleic acid to which it is linked. The term includes a vector as a self-replicating nucleic acid structure and a vector incorporated into the genome of the host cell into which it has been introduced. Certain vectors can result in the expression of nucleic acids to which they are operably linked. Such vectors are also referred to herein as "expression vectors."

B. Antigen-binding molecule In one aspect of the invention, the antigen-binding molecule comprises a first antigen-binding domain.
The first antigen-binding domain may have any chemical structure as long as it can bind to the first complex, and may be, for example, a low molecular weight compound or a high molecular weight compound. .. The first complex consists of a first co-suppressing molecule and a first co-suppressing molecular ligand. The first co-suppressing molecular ligand is a ligand for the first co-suppressing molecule. Specific embodiments of the first antigen-binding domain are exemplified in the “(b-10) Other aspects” column described later.

(B-1) First Aspect In one embodiment, the first antigen binding domain may specifically bind to the first complex.

In that embodiment, the specific binding of the first antigen binding domain to the first complex can be an intermolecular force binding to the first co-suppressing molecule in the first complex, the first. It can be an intermolecular force binding to the first co-suppressing molecular ligand in the complex, or between the molecules of both the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex. It can be a force bond. The specific binding of the first antigen-binding domain to the first complex is preferably an intermolecular force between both the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex. Can be a bond by. The presence of these various binding modes is either the first co-suppressing molecule or the first co-suppressing molecular ligand, as long as the first antigen binding domain is specifically bound to the first complex. Or it means that an intermolecular force can be formed in both.

In this embodiment, when the first antigen-binding domain forms an intermolecular force bond with either the first co-suppressing molecule or the first co-suppressing molecular ligand, the first co-suppressing molecular ligand is used. It is assumed that there is an intramolecular conformational change in either or both of the first co-suppressing molecule and the first co-suppressing molecular ligand when bound to the first co-suppressing molecule. In this case, the first antigen-binding domain specifically recognizes the three-dimensional structure changed by the binding between these two molecules.
For example, when the first co-suppressing molecular ligand binds to the first co-suppressing molecule, there is a conformational change in one of the first co-suppressing molecule and the first co-suppressing molecular ligand. The first antigen-binding domain recognizes the three-dimensional structure changed by the binding, and binds by intermolecular force to the first co-suppressing molecule or the first co-suppressing molecular ligand that caused the change in the three-dimensional structure. Can form.
On the other hand, if there is a conformational change in both the first co-suppressing molecule and the first co-suppressing molecular ligand when the first co-suppressing molecular ligand binds to the first co-suppressing molecule, The first antigen-binding domain recognizes the three-dimensional structure of either the first co-suppressing molecule or the first co-suppressing molecular ligand changed by the binding, and causes the change in the three-dimensional structure. Can form an intermolecular force bond with a co-suppressing molecule or a first co-suppressing molecular ligand.

In this embodiment, when the first antigen-binding domain forms an intermolecular force bond with both the first co-suppressing molecule and the first co-suppressing molecular ligand, the first co-suppressing molecular ligand is the first. There may or may not be a conformational change within the molecule of the first co-suppressing molecule and the first co-suppressing molecular ligand when bound to the co-suppressing molecule. In this case, if a conformational change occurs, the conformation of either or both of the first co-suppressing molecule and the first co-suppressing molecular ligand can occur.
In this case, the first antigen binding domain recognizes the atomic arrangement peculiar to the first complex and binds by intermolecular force to both the first co-suppressing molecule and the first co-suppressing molecular ligand. Can be formed. In this case, the arrangement of all the atoms recognized by the first antigen-binding domain is only the first co-suppressing molecule that does not form the first complex, or the first that does not form the first complex. It does not exist only in the co-suppressing molecular ligand of.

(B-2) Second Aspect In another embodiment than (b-1) above, the first antigen binding domain may bind to the first complex.

In this embodiment, the binding activity of the first antigen-binding domain to the first complex is higher than the binding activity to the first co-suppressing molecule that does not form the first complex, the first complex. Forming a first complex with a binding activity to a first co-suppressing molecule that is higher than the binding activity to a first co-suppressing molecule ligand that does not form a body, or that does not form a first complex Higher than both binding activity to the first co-suppressing molecular ligand that is not. The binding activity of the first antigen-binding domain to the first complex preferably forms the first complex and the binding activity to the first co-suppressing molecule that does not form the first complex. Higher than both binding activity to the first co-suppressing molecular ligand that is not.

In that embodiment, the binding of the first antigen binding domain to the first complex can be an intermolecular force binding to the first co-suppressing molecule in the first complex, the first complex. It can be an intermolecular force binding to the first co-suppressing molecular ligand in, or by an intermolecular force of both the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex. It can be a bond. Binding of the first antigen-binding domain to the first complex is preferably intermolecular force binding of both the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex. Can be. The presence of these various binding embodiments is due to the presence of the first co-suppressing molecule and the first co-suppressing molecule as long as the first antigen binding domain binds to the first complex and has the above-mentioned binding activity. It means that an intermolecular force can be formed on either or both of the ligands.

In this embodiment, when the first antigen-binding domain forms an intermolecular force bond with either the first co-suppressing molecule or the first co-suppressing molecular ligand, the first co-suppressing molecular ligand is used. It is assumed that there is an intramolecular conformational change in either or both of the first co-suppressing molecule and the first co-suppressing molecular ligand when bound to the first co-suppressing molecule. In this case, the first antigen-binding domain is more likely to recognize the three-dimensional structure changed by the binding between these two molecules than the three-dimensional structure before the change due to the binding.
For example, when the first co-suppressing molecular ligand binds to the first co-suppressing molecule, there is a conformational change in one of the first co-suppressing molecule and the first co-suppressing molecular ligand. The first antigen-binding domain is easier to recognize the three-dimensional structure changed by the binding than the three-dimensional structure before the change, and the first co-suppressing molecule or the first co-suppressing molecule that caused the change in the three-dimensional structure. It is easy to form a bond with a ligand by intermolecular force.
On the other hand, if there is a conformational change in both the first co-suppressing molecule and the first co-suppressing molecular ligand when the first co-suppressing molecular ligand binds to the first co-suppressing molecule, The first antigen-binding domain is easier to recognize the three-dimensional structure of either the first co-suppressing molecule or the first co-suppressing molecular ligand altered by the binding, and is easier to recognize than the pre-changed three-dimensional structure. It is easy to form a bond by intermolecular force with the first co-suppressing molecule or the first co-suppressing molecular ligand that has changed the three-dimensional structure.

In that embodiment, if the first antigen-binding domain forms an intermolecular force bond with both the first co-suppressing molecule and the first co-suppressing molecular ligand, the first co-suppressing molecular ligand is the first. There may or may not be a conformational change within the molecule of the first co-suppressing molecule and the first co-suppressing molecular ligand when bound to the co-suppressing molecule. In this case, if a conformational change occurs, the conformation of either or both of the first co-suppressing molecule and the first co-suppressing molecular ligand can occur. In this case, the first antigen-binding domain has an atomic arrangement peculiar to the first complex, an atomic arrangement in the first co-suppressing molecule that does not form the first complex, and a first complex. It is easier to recognize than the arrangement of atoms in the first co-suppressing molecular ligand that does not form, and it is easier to form an intermolecular bond with both the first co-suppressing molecule and the first co-suppressing molecular ligand. In this case, the arrangement of all the atoms recognized by the first antigen-binding domain is only the first co-suppressing molecule that does not form the first complex, or the first that does not form the first complex. It does not exist only in the co-suppressing molecular ligand of.

i) Binding activity The binding activity in the embodiment can be measured using a commonly used method for measuring intermolecular binding activity. Flow cytometry (FCM), surface plasmon resonance (SPR), biolayer interference (BLI), and enzyme-linked immunosorbent assay (ELISA) can be selected as methods for measuring intermolecular binding activity. When the antigen-binding molecule contains only one of the first antigen-binding molecules, the first complex of the antigen-binding molecules when the antigen-binding activity of the antigen-binding molecule to the first complex is measured by these measurement methods. Affinity to is measured. When the antigen-binding molecule contains a plurality of the first antigen-binding molecules, when the antigen-binding activity of the antigen-binding molecule to the first complex is measured by these measurement methods, the antigen-binding molecule to the first complex Avidity is measured.

(Flow cytometry)
In one embodiment, the binding activity is preferably measured by flow cytometry. When the binding activity is measured by flow cytometry, the first complex is formed on the surface of the cell, so that the binding of the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex The embodiment is closer to the in vivo environment than other measurement methods. The binding activity of the antigen-binding molecule to the first complex measured by flow cytometry easily reflects the environment in the living body.

As an example of flow cytometry for measuring the binding activity in this embodiment, a first cell in which the first co-suppressing molecule is forcibly expressed and the first co-suppressing molecular ligand is not forcibly expressed is used. ..
In this example, the binding activity of the first antigen-binding domain to the first cell is measured in the presence and absence of the first soluble polypeptide.
In this example, the binding activity of the first antigen-binding domain to the first cell in the presence of the first soluble polypeptide is higher than in the absence of the first soluble polypeptide. ..

The first cell may be any cell as long as it is an animal-derived cell that forcibly expresses the first co-suppressing molecule and does not forcibly express the first co-suppressing molecular ligand. The first cell preferably does not substantially express the first co-suppressing molecular ligand. The first cell more preferably does not express the first co-suppressing molecular ligand.
The first cell is preferably a cell that can be suspended and cultured. The first cell type includes cells derived from Chinese hamster ovary cells.

The first antigen-binding domain used for the measurement in this example may be contained in the antigen-binding molecule or may be a part of the antigen-binding molecule. The first antigen-binding domain used for the measurement may be added with an amino acid sequence not contained in the antigen-binding molecule.
The first soluble polypeptide used for the measurement in this example may be any polypeptide as long as it contains or a part of the extracellular domain of the first co-suppressing molecular ligand and is soluble. .. The extracellular domain of the first co-suppressing molecular ligand comprises an Ig-V-like domain and an Ig-C-like domain. The first soluble polypeptide used for the measurement may contain at least the Ig-V-like domain of the first co-suppressing molecular ligand. That is, the first soluble polypeptide used for the measurement may be any polypeptide as long as it contains the Ig-V-like domain of the first co-suppressing molecular ligand and is soluble. The first soluble polypeptide used for the measurement preferably contains and is soluble in the Ig-V-like domain of the first co-suppressing molecular ligand.

As another example of flow cytometry for measuring the binding activity in this embodiment, a second cell in which the first co-suppressing molecular ligand is forcibly expressed and the first co-suppressing molecule is not forcibly expressed is used. Is done.
In this example, the binding activity of the first antigen-binding domain to the second cell is measured in the presence and absence of the second soluble polypeptide.
In this example, the binding activity of the first antigen-binding domain to the second cell in the presence of the second soluble polypeptide is higher than in the absence of the second soluble polypeptide. ..

The second cell may be any cell as long as it is an animal-derived cell that forcibly expresses the first co-suppressing molecular ligand and does not forcibly express the first co-suppressing molecule. The second cell preferably does not substantially express the first co-suppressing molecule. The second cell more preferably does not express the first co-suppressing molecule.
The second cell is preferably a cell that can be suspended and cultured. The second type of cell includes cells derived from Chinese hamster ovary cells.

The first antigen-binding domain used for the measurement in this example may be contained in the antigen-binding molecule or may be a part of the antigen-binding molecule. The first antigen-binding domain used for the measurement may be added with an amino acid sequence not contained in the antigen-binding molecule.
The second soluble polypeptide used for the measurement in this example may be any polypeptide as long as it contains or a part of the extracellular domain of the first co-suppressing molecule and is soluble. The extracellular domain of the first co-suppressive molecule comprises an Ig-V-like domain. The second soluble polypeptide used for the measurement may contain at least the Ig-V-like domain of the first co-suppressing molecule. That is, the second soluble polypeptide used for the measurement may be any polypeptide as long as it contains the Ig-V-like domain of the first co-suppressing molecule and is soluble. The second soluble polypeptide used for the measurement preferably contains and is soluble in the Ig-V-like domain of the first co-suppressing molecule.

Specific embodiments of the binding activity include embodiments that satisfy either or both of the following (a) and (b).
(a) In flow cytometry using a first cell that forcibly expresses the first co-suppressing molecule and does not forcibly express the first co-suppressing molecular ligand, the cell of the first co-suppressing molecular ligand. The binding activity of the first antigen-binding domain to the first cell in the presence of the first soluble polypeptide containing the outer domain or a part thereof is the non-binding activity of the first soluble polypeptide. Higher than in existence.
(b) In flow cytometry using a second cell forcibly expressing the first co-suppressing molecular ligand and not forcibly expressing the first co-suppressing molecule, extracellular of the first co-suppressing molecule. The binding activity of the first antigen-binding domain to the second cell in the presence of the second soluble polypeptide containing the domain or a part thereof is the absence of the second soluble polypeptide. Higher than below.

In that particular embodiment, the first cell of (a) may be derived from Chinese hamster ovary cells. The second cell in (b) can be derived from Chinese hamster ovary cells. Both the first and second cells can be derived from Chinese hamster ovary cells.
In that particular embodiment, "forced expression" is the expression of a foreign protein by genetic recombination in a cell.
The flow cytometry may be one commonly used by those skilled in the art, and commercially available flow cytometry may be used.
The first soluble polypeptide may be any polypeptide comprising the extracellular domain of the first co-suppressing molecular ligand or a portion thereof, and is preferably soluble. The extracellular domain of the first co-suppressing molecular ligand or a portion thereof preferably contains and is soluble in the Ig-V-like domain of the first co-suppressing molecular ligand.
The second soluble polypeptide may be any polypeptide comprising the extracellular domain of the first co-suppressing molecule or a portion thereof, and is preferably soluble. The extracellular domain of the first co-suppressing molecule or a portion thereof preferably contains and is soluble in the Ig-V-like domain of the first co-suppressing molecule.

As a mechanism for detecting the binding activity in the aspect (a) above, first, the first soluble polypeptide is forcibly expressed in the first cell in the presence of the first soluble polypeptide. By binding to the first co-suppressing molecule, a first complex is formed on the first cell surface. The polypeptide comprising the first antigen binding domain (the polypeptide may then be an antigen binding molecule) then binds to the formed first complex. This binding is detected by flow cytometry. Since the first antigen-binding domain specifically binds to the first complex, this binding is not detected in the absence of the first soluble polypeptide.
As a mechanism for detecting the binding activity in the aspect (b) above, first, the second soluble polypeptide is forcibly expressed in the second cell in the presence of the second soluble polypeptide. By binding to the first co-suppressing molecular ligand, the first complex is formed on the second cell surface. The polypeptide comprising the first antigen binding domain (the polypeptide may then be an antigen binding molecule) then binds to the formed first complex. This binding is detected by flow cytometry. Since the first antigen-binding domain specifically binds to the first complex, this binding is not detected in the absence of the second soluble polypeptide.

(Biolayer interference)
In another embodiment, the binding activity is preferably measured by biolayer interference (BLI). In the binding activity by BLI, for example, Octet HTX (Molecular Devies) can be used. As the assay buffer at the time of measurement, for example, 50 mM phosphoric acid, 150 mM NaCl, 0.05 w / v% P20, pH 7.4 can be used. The temperature condition is 30 ° C.

As a biosensor for capturing an antibody, for example, Protein A Biosensor (ForteBio, Cat. 18-05010) can be used. In this, the antigen-binding molecule of interest can be captured so as to be about 2.5 nm.
When measuring the binding activity to the first complex, the first co-suppressing molecule and the first co-suppressing molecular ligand are diluted in the assay buffer with an assay buffer to a final concentration of 1000 nM and used. obtain.
When measuring the binding activity to only the first co-suppressing molecule or only the first co-suppressing molecular ligand as a control group, the first co-suppressing molecule or the first co-suppressing molecular ligand is added to the running buffer. It can be diluted with assay buffer to a final concentration of 1000 nM and used.

The antigen-binding activity of the antigen-binding molecule can be calculated by the following formula 3 from the analysis results using Data Analysis HT 12.0.

[Equation 3]
Antigen binding amount per unit amount of antigen-binding molecule = (antigen binding amount (nm) -blank binding amount (nm)) / antigen-binding molecule capture amount (nm)

As a specific embodiment of the binding activity, the amount of the antigen-binding molecule bound to the first complex per unit amount of the antigen-binding molecule is only the first co-suppressing molecule per unit amount of the antigen-binding molecule, and the first co-suppression occurs. It is higher than the amount of binding to the molecular ligand alone, or both the first co-suppressing molecule alone and the first co-suppressing molecular ligand alone.

(B-3) Third Aspect In another aspect from the above-mentioned (b-1) and (b-2), the first antigen-binding domain is a first co-suppressing molecule and the first co-suppressing. It may specifically bind to an epitope present in a first complex consisting of a first co-suppressing molecular ligand for a molecule.
The entire region of the epitope is preferably not filled with only the first co-suppressing molecule that does not form the first complex, or only the first co-suppressing molecular ligand that does not form the first complex. In this embodiment, both a part of the first co-suppressing molecule that does not form the first complex and a part of the first co-suppressing molecular ligand that does not form the first complex are used as epitopes. Is included, and a three-dimensional structure that does not exist in the first co-suppressing molecule that does not form the first complex but exists in the first co-suppressing molecule that forms the first complex. A three-dimensional structure that does not exist in the case of an epitope and in the first co-suppressing molecular ligand that does not form the first complex, but exists in the first co-suppressing molecular ligand that forms the first complex. The case where the three-dimensional structure change described later is used as an epitope, such as the case where is used as an epitope is included.

In that embodiment, the specific binding of the first antigen binding domain to the epitope can be an intermolecular force binding to the first co-suppressing molecule in the first complex and in the first complex. It can be an intermolecular force binding to the first co-suppressing molecular ligand, or an intermolecular force binding to both the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex. could be. Specific binding of the first antigen-binding domain to the epitope is preferably by intermolecular force binding of both the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex. could be. The presence of these various binding embodiments is associated with either or both of the first co-suppressing molecule and the first co-suppressing molecular ligand as long as the first antigen binding domain is specifically bound to the epitope. It means that an intermolecular force can be formed. The intramolecular force can be an intramolecular force that occurs between the first antigen binding domain and the epitope.

In this embodiment, when the first antigen-binding domain forms an intermolecular force bond with either the first co-suppressing molecule or the first co-suppressing molecular ligand, the first co-suppressing molecular ligand is used. It is assumed that there is an intramolecular conformational change in either or both of the first co-suppressing molecule and the first co-suppressing molecular ligand when bound to the first co-suppressing molecule. In this case, the first antigen-binding domain specifically recognizes the three-dimensional structure changed by the binding between these two molecules as an epitope.
For example, when the first co-suppressing molecular ligand binds to the first co-suppressing molecule, there is a conformational change in one of the first co-suppressing molecule and the first co-suppressing molecular ligand. The first antigen-binding domain recognizes the three-dimensional structure changed by the binding as an epitope, and depends on the first co-suppressing molecule or the first co-suppressing molecular ligand that caused the change in the three-dimensional structure and the intermolecular force. Can form a bond.
On the other hand, if there is a conformational change in both the first co-suppressing molecule and the first co-suppressing molecular ligand when the first co-suppressing molecular ligand binds to the first co-suppressing molecule, The first antigen-binding domain recognizes the three-dimensional structure of either the first co-suppressing molecule or the first co-suppressing molecular ligand changed by the binding as an epitope, and causes the change in the three-dimensional structure. It can form an intermolecular force bond with a first co-suppressing molecule or a first co-suppressing molecular ligand.

In this embodiment, when the first antigen-binding domain forms an intermolecular force bond with both the first co-suppressing molecule and the first co-suppressing molecular ligand, the first co-suppressing molecular ligand is the first. There may or may not be a conformational change within the molecule of the first co-suppressing molecule and the first co-suppressing molecular ligand when bound to the co-suppressing molecule. In this case, if a conformational change occurs, the conformation of either or both of the first co-suppressing molecule and the first co-suppressing molecular ligand can occur.
In this case, the first antigen-binding domain recognizes the atomic arrangement peculiar to the first complex as an epitope, and the epitope and molecule in both the first co-suppressing molecule and the first co-suppressing molecular ligand. Intermolecular bonds can be formed. In this case, the arrangement of all atoms in the epitope recognized by the first antigen-binding domain is only the first co-suppressing molecule that does not form the first complex, or does not form the first complex. It does not exist only in the first co-suppressing molecular ligand.

(B-4) Action of the antigen-binding molecule at the immunological synapse In the embodiments (b-1), (b-2) and (b-3) described above, the antigen-binding molecule is an immune system in which the first complex is present. At the synapse, it can activate a first intracellular signal downstream of the first co-suppressing molecule.
The "first intracellular signal downstream of the first co-suppressing molecule" means an intracellular signal generated when the first co-suppressing molecule is activated in a cell expressing the first co-suppressing molecule. do. In T cells, activation of the first intracellular signal may have the function of suppressing the signal from the T cell receptor. As a result, T cell activation can be suppressed. T cell activation here refers to the phenotype of T cells that appears when immunity is activated, such as T cell proliferation and differentiation into activated cells, as well as increased production of cytokines that activate immunity. Is. Examples of T cells include CD4 (+) FoxP3 (-) T cells and CD8 (+) T cells.

An "immunological synapse" is a ring-shaped structure formed between cells expressing a co-suppressing molecular ligand and cells of the immune system expressing a co-suppressing molecule in close proximity to each other and in contact with each other. Means (Front Immunol 2016 Vol.7 Article 255, Immunology 2011 Vol.133 pp.420-425). Synapses formed between cells expressing a co-suppressing molecular ligand with MHC and T cells expressing a co-suppressing molecule are exemplified as immunological synapses.

(B-5) Measurement of First Intracellular Signal Intensity In one embodiment, the intensity of the first intracellular signal generated by activation of the first co-suppressing molecule can be measured by in vitro experiments. In this measurement system, the antigen-binding molecule activated the first intracellular signal when the intensity of the first intracellular signal in the presence of the antigen-binding molecule was higher than that in the absence of the antigen-binding molecule. It can be said.

In this embodiment, a third cell expressing the first co-suppressing molecule and a fourth cell expressing the antigen-independent T cell receptor activator and the first co-suppressing molecular ligand can contact each other. The measurement system used in the above state is exemplified as the measurement system of the intensity of the first intracellular signal. The third cell is a cell capable of measuring the intensity of the first intracellular signal, which is a downstream intracellular signal generated by the activation of the first co-suppressing molecule.

When the first intracellular signal in the third cell is maximally activated by the first co-suppressing molecular ligand in the fourth cell in the measurement system, the antigen-binding molecule is the first. It cannot be confirmed that the intracellular signal of the cell was activated. Therefore, in the above embodiment, the measurement of the intensity of the first intracellular signal maximizes the activation of the first intracellular signal in the third cell by the first co-suppressing molecular ligand expressed by the fourth cell. It is done under conditions that have not been reached. In order to make such a condition, a method of partially suppressing the first intracellular signal in the third cell by using an inhibitory antibody against the first co-suppressing molecular ligand in the fourth cell is applied. obtain. A known inhibitory antibody may be applied to the inhibitory antibody against the first co-suppressive molecular ligand. For example, as an inhibitory antibody against PD-L1, YW243.55S70 (see US2016 / 0222117 A1; heavy chain variable region SEQ ID NO: 16 and light chain variable region SEQ ID NO: 17) can be used.

In the embodiment, the third cell can be derived from Jurkat cells. The fourth cell can be derived from Chinese hamster ovary cells. In a preferred embodiment, the third cell is derived from Jurkat cells and the fourth cell is derived from Chinese hamster ovary cells. As the fourth cell, an artificial antigen-presenting cell (# J109A) expressing PD-L1 manufactured by Promega is exemplified.

In the embodiment, the intensity of the first intracellular signal is directly or indirectly mutual with the intracellular domain of the first co-suppressing molecule, which occurs when the first co-suppressing molecule is activated. It can be measured based on its proximity to the intracellular protein of action. The proximity detection system may be any detection system as long as the proximity between molecules existing in the cell can be observed and the intensity thereof can be quantified. Specific examples include NanoBRET®.

In the embodiment, the intracellular protein that directly or indirectly interacts with the intracellular domain of the first co-suppressing molecule may be known or newly discovered by an experimenter. good. Dephosphorylation enzyme is exemplified as the intracellular protein. Specifically, when the co-suppressing molecule is PD-1, the dephosphorylating enzyme is exemplified by SHP-2.

(B-6) Measurement of suppression of immune activation signal by first intracellular signal In one embodiment, immune activation signal is suppressed by a first intracellular signal generated by activation of a first co-suppressing molecule. Whether or not it was done can be confirmed by in vitro experiments. In that experiment, it can be said that the antigen-binding molecule suppressed the immune activation signal when the intensity of the immune activation signal in the presence of the antigen-binding molecule was lower than that in the absence of the antigen-binding molecule. .. The immune activation signal is preferably a second intracellular signal that can be activated downstream of the T cell receptor in T cells.

In that embodiment, a fifth cell expressing the T cell receptor and the first co-suppressing molecule and a sixth cell expressing the antigen-independent T cell receptor activator and the first co-suppressing molecule ligand. A measurement system used under conditions where the cells can come into contact with each other is exemplified as a second intracellular signal intensity measurement system. The fifth cell can measure the intensity of the second intracellular signal, which is a downstream intracellular signal generated by the activation of the T cell receptor, and the second intracellular signal can be measured by the activation of the first intracellular signal. A cell in which intracellular signals can be suppressed. As the sixth cell, the fourth cell used in (b-5) above can be utilized.

In the fifth cell of the measurement system, the second intracellular signal is the first co-suppressing molecule of the fifth cell resulting from the binding of the first co-suppressing molecular ligand of the sixth cell. When the state of maximal suppression by activation followed by activation of the first intracellular signal is reached, it cannot be confirmed that the antigen-binding molecule suppressed the second intracellular signal. Therefore, in the above embodiment, the measurement of the intensity of the second intracellular signal is such that the second intracellular signal in the fifth cell is maximally suppressed by the first co-inhibitory molecular ligand expressed by the sixth cell. It is done under no conditions. In order to make such a condition, there is a method of partially suppressing the suppression of the second intracellular signal in the fifth cell by using an inhibitory antibody against the first co-suppressing molecular ligand in the sixth cell. Can be applied. A known inhibitory antibody may be applied to the inhibitory antibody against the first co-suppressive molecular ligand. For example, as an inhibitory antibody against PD-L1, YW243.55S70 (see US2016 / 0222117 A1; heavy chain variable region SEQ ID NO: 16 and light chain variable region SEQ ID NO: 17) can be used.

In the embodiment, the fifth cell can be derived from Jurkat cells. The sixth cell can be derived from Chinese hamster ovary cells. In a preferred embodiment, the fifth cell is derived from Jurkat cells and the sixth cell is derived from Chinese hamster ovary cells. As the sixth cell, an artificial antigen presenting cell (# J109A) expressing PD-L1 of Promega is exemplified.

In the embodiment, the intensity of the second intracellular signal can be measured based on the transcriptional activity of a transcription factor acting downstream of the T cell receptor signal. The system for measuring transcriptional activity may be by modification or binding of a factor known to act downstream of the T cell receptor signal, or by variability in gene expression. The modification may be a variation in the phosphorylation state of a particular protein. The binding may be an interaction between specific proteins. The variation in gene expression may be a reporter gene assay in which a promoter recognized by a particular transcription factor is fused with a gene encoding a protein having some enzymatic activity. Specifically, in an embodiment in which the second intracellular signal is the activity of NFAT, which is a transcription factor, a reporter gene fused so that expression of luciferase can be controlled downstream of the NFAT response element is the first. An assay in which a substrate introduced into 5 cells and reacted with luciferase is luminescenceed is exemplified as a reporter gene assay capable of measuring the intensity of a downstream signal of a T cell receptor signal.

(B-7) Combination of 1st Co-Suppressing Mole and 1st Co-Suppressing Molecular Ligand In one embodiment, the first co-suppressing in a first complex to which the first antigen-binding domain can specifically bind. The molecule can be appropriately selected from a range of co-suppressing molecules known at that time to those seeking to obtain the antigen-binding molecule. The first co-suppressing molecule typically includes PD-1, BTLA, TIGIT, LAG-3, CTLA4, and TIM-3.

In one embodiment, the first co-suppressing molecular ligand in the first complex can be appropriately selected from a range of co-suppressing molecular ligands known at that time to those who seek to obtain the antigen-binding molecule. The first co-suppressing molecular ligands typically include PD-L1, PD-L2, HVEM, CD155, CD112, MHC class II molecules, CD80, CD86, galectin-9, phosphatidylserine, CEACAM-1, and HMGB1. Will be.

In these embodiments, the combination of the first co-suppressing molecule and the first co-suppressing molecular ligand is preferably a combination of PD-1 and PD-L1, a combination of PD-1 and PD-L2, a combination of BTLA and HVEM. , TIGIT and CD155 combination, TIGIT and CD112 combination, LAG-3 and MHC class II molecule combination, CTLA4 and CD80 combination, CTLA4 and CD86 combination, TIM-3 and galectin-9 combination, TIM-3 and One selected from the group consisting of a combination of phosphatidylserine, a combination of TIM-3 and CEACAM-1, and a combination of TIM-3 and HMGB1.

In a particular embodiment, the combination of the first co-suppressing molecule and the first co-suppressing molecular ligand is a combination of PD-1 and PD-L1. In this case, the epitope to which the first antigen-binding domain binds is preferably any one of the anti-human PD-L1 / human PD-1 complex protein antibodies prepared in Table 1 prepared in Example 7. It overlaps in whole or in part with the epitope in the first complex to which it binds. The epitope to which the first antigen-binding domain binds is more preferably an anti-human PD-L1 / human PD-1 complex protein antibody called LPB0010HCb-G1T7P / LPB0010LCb-k0MTC or LPC0039HCd-G1T7P / LPC0039LCd-k0MTC. It completely or partially overlaps with the epitope in the first complex.

(B-8) Multispecific Antigen Binding Molecules In one embodiment, the antigen binding molecule may further comprise a second antigen binding domain. In that case, the antigen-binding molecule is a multispecific antigen-binding molecule that can bind to two or more antigens or epitopes. The second antigen-binding domain can specifically bind to a second co-suppressing molecule that can form a second complex with the second co-suppressing molecular ligand. The second antigen-binding domain may or may not bind to the second complex. The first antigen-binding domain is characterized by binding to the first complex, whereas the second antigen-binding domain may or may not bind to the second complex. In that respect, the two domains are different. The second antigen-binding domain preferably has agonist activity against the second co-suppressing molecule.
The second antigen-binding domain may have any chemical structure as long as it has the above-mentioned characteristics, and may be, for example, a low molecular weight compound or a high molecular weight compound. Specific embodiments of the second antigen-binding domain are exemplified in the “(b-10) Other aspects” column described later.

In this embodiment, the multispecific antigen binding molecule can activate a third intracellular signal downstream of the second co-suppressing molecule at the immunological synapse in which the first complex is present. The intensity of the third intracellular signal can be measured in the same manner as in the above-mentioned "measurement of the intensity of the first intracellular signal" in (b-5). When measuring the intensity of the third intracellular signal, the first co-suppressing molecule is the second co-suppressing molecule and the first co-suppressing molecule in the "measurement of the intensity of the first intracellular signal". The ligand is replaced by a second co-suppressing molecular ligand, respectively. It is the agonist activity of the second antigen-binding domain on the second co-suppressing molecule that the multispecific antigen-binding molecule can activate the third intracellular signal downstream of the second co-suppressing molecule. preferable.

In this embodiment, activation of the third intracellular signal can suppress the second intracellular signal. The intensity of the second intracellular signal can be measured in the same manner as in "Measurement of suppression of immune activation signal by the first intracellular signal" in (b-6) above.

In this embodiment, the binding of the second antigen-binding domain to the second co-suppressing molecule preferably does not compete with the binding of the second co-suppressing molecular ligand to the second co-suppressing molecule.

(B-9) Combination of a second co-suppressing molecule and a second co-suppressing molecular ligand In one embodiment, the second co-suppressing molecule to which the second antigen-binding domain can specifically bind is a multispecific antigen. It can be appropriately selected from the range of co-suppressing molecules that can be known at that time by the person who tried to obtain the bound molecule. The second co-suppressing molecule typically includes PD-1, BTLA, TIGIT, LAG-3, CTLA4, and TIM-3.

In one embodiment, the second co-suppressing molecular ligand capable of forming a second complex with the second co-suppressing molecule is a co-suppressing molecule to the extent known at that time to those seeking to obtain the multispecific antigen binding molecule. It can be appropriately selected from inhibitory molecular ligands. Second co-suppressing molecular ligands typically include PD-L1, PD-L2, HVEM, CD155, CD112, MHC class II molecules, CD80, CD86, galectin-9, phosphatidylserine, CEACAM-1, and HMGB1. Will be.

In these embodiments, the combination of the second co-suppressing molecule and the second co-suppressing molecular ligand is preferably a combination of PD-1 and PD-L1, a combination of PD-1 and PD-L2, a combination of BTLA and HVEM. , TIGIT and CD155 combination, TIGIT and CD112 combination, LAG-3 and MHC class II molecule combination, CTLA4 and CD80 combination, CTLA4 and CD86 combination, TIM-3 and galectin-9 combination, TIM-3 and One selected from the group consisting of a combination of phosphatidylserine, a combination of TIM-3 and CEACAM-1, and a combination of TIM-3 and HMGB1.

(B-10) Other Aspects In another embodiment, one molecule of the first co-suppressing molecule may be present in the first complex, and a plurality of molecules of the first co-suppressing molecule are present. May be. One molecule of the first co-suppressing molecular ligand may be present in the first complex, or a plurality of molecules of the first co-suppressing molecular ligand may be present. The first complex may contain molecules other than the first co-suppressing molecule and the first co-suppressing molecular ligand.

In other embodiments, the first antigen binding domain can be a variable region of an antibody or a fragment thereof that binds to its antigen of interest. The first antigen-binding domain is preferably a variable region of one antibody or a fragment thereof that binds to the antigen of interest. The "variable region of an antibody or a fragment thereof that binds to a target antigen thereof" in the embodiment means one arm. The first antigen-binding domain can specifically bind to the first complex with only one arm. By doing so, the antigen-binding molecule can localize to the immunological synapse when the first complex is present at the immunological synapse.

In other embodiments, the second antigen binding domain can be a variable region of the antibody or a fragment thereof that binds to the antigen of interest. The second antigen-binding domain is preferably a variable region of one antibody or a fragment thereof that binds to the antigen of interest. The "variable region of an antibody or a fragment thereof that binds to a target antigen thereof" in the embodiment means one arm. The second antigen-binding domain can specifically bind to the second co-suppressing molecule with only one arm. By such binding, the second antigen binding domain can activate a third intracellular signal.

It is preferred that both the first antigen-binding domain and the second antigen-binding domain are the variable region of the antibody or a fragment thereof that binds to the target antigen.

In other embodiments, the first co-suppressing molecular ligand may be the same as or different from the second co-suppressing molecular ligand. The first co-suppressing molecule may be the same as or different from the second co-suppressing molecule. Similarly, in those downstream signals, the first intracellular signal may be the same as or different from the third intracellular signal. When the first co-suppressing molecule is the same as the second co-suppressing molecule, preferably the epitopes in the co-suppressing molecule to which the first antigen-binding domain and the second antigen-binding domain bind do not overlap with each other. Or it overlaps in whole or in part.

In other embodiments, the antigen binding molecule may further comprise a constant region of the antibody. At least one residue in the constant region of the antibody may differ from the amino acid residue at the corresponding position in the constant region of the native antibody. In a further embodiment, the antigen binding molecule can be an antibody or antibody fragment. The antibody is preferably an isolated antibody.

In other embodiments, the antigen-binding molecule may comprise a polypeptide, sugar chain, or small molecule compound in addition to the first antigen-binding domain, the second antigen-binding domain, and the constant region of the antibody. These configurations may be linked to each other by a linker or a covalent bond, or may exist as a complex by intramolecular force. It is preferred that all configurations are covalently linked, either directly or via a linker. The type of linker is not particularly limited as long as it can be used in medicine.

(B-11) Production Method In another aspect of the present invention, the above-mentioned method for producing an antigen-binding molecule is provided.
In one embodiment, the method for producing an antigen-binding molecule may include culturing cells containing a nucleic acid encoding the antigen-binding molecule.

In another embodiment, the method for producing an antigen-binding molecule is a third complex comprising a first co-suppressing molecule or a partial polypeptide thereof and a first co-suppressing molecular ligand for the first co-suppressing molecule or a partial polypeptide thereof. It may include immunizing the body against non-human animals. The third complex is obtained by directly binding the first co-suppressing molecule or a partial polypeptide thereof to the first co-suppressing molecular ligand or a partial polypeptide thereof via a linker or by a chemical bond1. It can be a molecule.
The linker is not particularly limited as long as it can chemically bind the first co-suppressing molecule or its partial polypeptide to the first co-suppressing molecular ligand or its partial polypeptide. The chemical bond is not particularly limited as long as the third complex can function as an antigen. An example of the chemical bond is a covalent bond. If the chemical bond is a covalent bond, the stability of the third complex is enhanced. Examples of the covalent bond include a disulfide bond utilizing a thiol group existing in the side chain of a cysteine residue.

The production method may include collecting plasma or blood containing the antigen-binding molecule from the non-human animal immunized with the third complex. The production method may further comprise obtaining a composition containing the antigen-binding molecule with increased purity from the plasma or blood.
In that embodiment, in the production method, the antigen binding molecule can be a polyclonal antibody.

The production method may include isolating a cell clone expressing the antigen-binding molecule from a non-human animal immunized with the third complex. The production method may further comprise extracting and isolating the nucleic acid encoding the antigen-binding molecule from the cell clone. The production method may further comprise recombination of the nucleic acid. The production method may further comprise introducing the recombinant nucleic acid into another cell and expressing the antigen-binding molecule in the other cell. The production method may further comprise obtaining a composition containing the expressed antigen binding molecule. The production method may further comprise increasing the purity of the expressed antigen-binding molecule in the composition.

C. Uses of Antigen-Binding Molecules (c-1) Co-Suppressing Molecular Agonists In another aspect of the present invention, co-suppressing molecular agonists are provided.
In one embodiment, the co-suppressing molecule agonist comprises the antigen-binding molecule described above. The co-suppressive molecule agonist was used unexpectedly as an agonist for a co-suppressive molecule as well as when it was used with the expectation of acting as an agonist for the co-suppressive molecule, but as a result of the use, the co-suppressive molecule It also includes cases where the agonist activity was or was found to be present.

(C-2) Co-suppressing molecular signal activator In another aspect of the present invention, a co-suppressing molecular signal activator is provided.
In one embodiment, the co-suppressing molecular signal activator contains the antigen-binding molecule described above. The co-suppressive molecule signal activator was used with the expectation of acting as an agonist on the co-suppressive molecule, and was unexpectedly used as an agonist on the co-suppressive molecule, but as a result of the use. It also includes cases where the agonist activity was or was found to be possessed against the co-suppressing molecule.

(C-3) Pharmaceutical Composition A pharmaceutical composition is provided in another aspect of the present invention.
In one embodiment, the pharmaceutical composition may contain the antigen-binding molecule described above. The pharmaceutical composition can be used to treat or prevent a disease resulting from an hyperimmunity. Diseases caused by hyperimmunity include autoimmune diseases.

"Autoimmune disease" refers to a non-malignant disease or disorder that arises from the individual's own tissue and is directed at the individual's own tissue. As used herein, autoimmune disorders expressly exclude malignant or cancerous disorders or conditions, especially B-cell lymphoma, acute lymphoblastic leukemia (ALL), chronic lymphocytes. Exclude chronic lymphocytic leukemia (CLL), hairy cell leukemia, and chronic myeloblastic leukemia. Examples of autoimmune diseases or disorders include, but are not limited to: inflammatory reactions such as inflammatory skin diseases including psoriasis and dermatitis (eg, atopic dermatitis); systemic. Lupus erythematosus and sclerosis; Reactions associated with inflammatory bowel disease (eg, Crohn's disease and ulcerative colitis); Lupus erythematosus syndrome (adult respiratory distress syndrome: including ARDS); Dermatitis; Allergic conditions such as meningitis; encephalitis; vegetationitis; colitis; glomerular nephritis; eczema and asthma and other conditions with T cell infiltration and chronic inflammatory reactions; atherosclerosis; leukocyte adhesion failure; joints Rheumatoid; systemic lupus erythematosus (SLE) (including but not limited to lupus erythematosus, skin lupus); diabetes (eg, type I diabetes or insulin-dependent diabetes); polysclerosis; Raynaud's syndrome; autoimmune Immune thyroiditis; Hashimoto thyroiditis; allergic encephalomyelitis; Schegren's syndrome; juvenile-onset diabetes; and cytokines and T lymphocytes typically found in tuberculosis, sarcoidosis, polymyositis, granulomatosis, and vasculitis Immune reactions associated with mediated acute and delayed hypersensitivity; malignant anemia (Azison's disease); diseases associated with leukocyte leakage; central nervous system (CNS) inflammatory disorders; multi-organ injury syndrome; hemolytic Anemia (including, but not limited to, cryoglobulinemia or Coombs-positive anemia); severe myasthenia; antigen-antibody complex-mediated disease; anti-globulous basal membrane disease; antiphospholipid syndrome; allergic neuritis; Graves' disease; Lambert-Eaton myasthenic syndrome; Lupus erythematosus; Lupus erythematosus; Autoimmune polyglandular endocrine disorders; Reiter's disease; Stiffman's syndrome; Bechet's disease; Giant cell arteritis; Immune complex nephritis; IgA Nephropathy; IgM multiple neuropathy; immune thrombocytopenic purpura (ITP) or autoimmune thrombocytopenia.

In that embodiment, the pharmaceutical composition may contain other ingredients. The other ingredients include other active ingredients applied to the disease of interest and pharmaceutically acceptable carriers. The other active ingredient may be appropriately selected depending on the target disease, and is not particularly limited. The pharmaceutically acceptable carrier is not particularly limited, and examples thereof include buffers, excipients, stabilizers, and preservatives. One or more of the other ingredients may be used in the pharmaceutical composition.

(C-4) Complex Detection Agent and Detection Method In another aspect of the present invention, a first complex consisting of a first co-suppressing molecule and a first co-suppressing molecular ligand for the first co-suppressing molecule. Detection agent is provided.
In one embodiment, the detection agent may contain the antigen-binding molecule described above. The presence of the first complex is not visible visually, but its presence is revealed by the detector. The target to which the detection agent is used is not particularly limited as long as it is suspected that the first complex is present, and is a living body, a tissue in the living body, a sample collected from the living body, or something existing in the living body. It can be a mixture or cells.

In another aspect of the invention, there is provided a method for detecting a first complex comprising a first co-suppressing molecule and a first co-suppressing molecular ligand for the first co-suppressing molecule.
In one embodiment, the detection method may include contacting the above-mentioned antigen-binding molecule with an immune cell.
In the detection method, the antigen-presenting cell may be in close proximity to the immune cell. The immune cell may express the first co-suppressing molecule, and the antigen-presenting cell may express the first co-suppressing molecular ligand. The immune cell can be a T cell. The detection method may be applied in vivo or in vitro. When the detection method is applied in vitro, the immune cell may be contacted with the antigen-binding molecule alive or may be immobilized on a slide glass. In the detection method, the immune cells may be present in the tissue or may be mixed with one or more cells other than the immune cells.

(C-5) Method of Suppressing Immunity In another aspect of the present invention, a method of suppressing immunity is provided.
In one embodiment, the method may comprise administering to the subject the antigen-binding molecule described above. When T cells are activated by antigen-presenting cells in the body of a subject, or when T cells are expected to be activated by antigen-presenting cells, the antigen-binding molecule is administered to the subject. As a result, the activation of the T cells is suppressed, and the immunity of the subject is suppressed.
When a T cell is activated by an antigen-presenting cell, the downstream signal of the T cell receptor in the T cell is activated. On the other hand, the first co-suppressing molecular ligand expressed on the antigen-presenting cell binds to the first co-suppressing molecule expressed on the T cell to form the first complex. The antigen-binding molecule binds to the first complex and activates the downstream signal of the first co-suppressing molecule. In this method, the mechanism by which the immunity of the subject is suppressed by suppressing the downstream signal of the activated T cell receptor described above by the downstream signal of the activated first co-suppressing molecule. is assumed.

According to the antigen-binding molecule of the present invention, the immunological synapse is targeted by specifically binding to a complex consisting of a co-suppressing molecule localized at the immunological synapse and its ligand, and the co-suppressing molecule has agonist activity. Therefore, the antigen-binding molecule can be a therapeutic agent or a candidate for treating or preventing a disease including an autoimmune disease. The antigen-binding molecule has agonist activity against the co-suppressing molecule at the immunological synapse by including at least the first antigen-binding domain. The antigen-binding molecule has remarkable agonist activity against the co-suppressing molecule at the immunological synapse by making it a multiple antigen-binding molecule containing a second antigen-binding domain in addition to the first antigen-binding domain.
On the other hand, when an attempt is made to obtain an antigen-binding molecule having an agonistic activity on a costimulatory molecule, the first complex contained in the first complex to which the first antigen-binding domain of the above-mentioned antigen-binding molecule specifically binds. The co-suppressing molecule and the first co-suppressing molecular ligand for the first co-suppressing molecule can be replaced with a first co-stimulating molecule and a first co-stimulating molecular ligand for the first co-stimulating molecule, respectively. The antigen-binding molecule obtained in this case is a third antigen that can specifically bind to a fourth complex consisting of a first co-stimulator molecule and a first co-stimulator molecule ligand for the first co-stimulator molecule. Includes combined domain. Such antigen-binding molecules can have agonistic activity against costimulatory molecules and are therefore potential therapeutic candidates for cancer and infectious diseases. In this case, the antigen-binding molecule can specifically bind to a second co-stimulator molecule capable of forming a fifth complex with the second co-stimulator molecule ligand in addition to the third antigen-binding domain. It is expected that the multi-antigen binding molecule containing 4 antigen-binding domains will have remarkable agonist activity against the co-stimulating molecule.

Example 1 Preparation of a bispecific antibody comprising an anti-human PD-1 antibody, an anti-human CD3 antibody, and an arm derived from the anti-human PD-1 antibody and an arm derived from the anti-CD3 antibody.
It was thought that the T cell receptor and PD-1 needed to be co-localized in order to create an agonist that induces an immunosuppressive signal for PD-1 (J. Exp. Med. Vol. 209 No. 6 1201-1217). The present inventors prepared bispecific antibodies consisting of anti-PD-1 arm and anti-CD3 arm, and evaluated their PD-1 immunosuppressive signal-inducing ability.

Three types of anti-human PD-1 monoclonal antibody PD1-17 (heavy chain variable region SEQ ID NO: 1 and light chain variable region sequence) that have been suggested to bind to human PD-1 and induce its immunosuppressive signal. Number: 2 (see WO2004 / 056875), as well as heavy chain constant region SEQ ID NO: 11 and light chain constant region SEQ ID NO: 12), antibody949 (clone 949) (heavy chain variable region SEQ ID NO: 3 and light chain variable region). SEQ ID NO: 4 (see WO2011 / 110621), as well as heavy chain constant region SEQ ID NO: 11 and light chain constant region SEQ ID NO: 13), clone 10 (heavy chain variable region SEQ ID NO: 5 and light chain variable region SEQ ID NO: 13). : 6 (see WO2010 / 029435), as well as heavy chain constant region SEQ ID NO: 11, light chain constant region SEQ ID NO: 13) were prepared by methods known to those skilled in the art.

As an anti-human CD3 antibody, two types of anti-human CD3 monoclonal antibody OKT3 (heavy chain variable region) known to bind to the CD3ε chain constituting the T cell receptor and induce an activation signal of the T cell receptor. SEQ ID NO: 7 (NCBI Accession: 1SY6_H, GI: 49259178), light chain variable region SEQ ID NO: 8 (NCBI Accession: 1SY6_L, GI: 49259177), heavy chain constant region SEQ ID NO: 14, and light chain constant region SEQ ID NO: : 13), and CE115TR (heavy chain variable region SEQ ID NO: 9, light chain variable region SEQ ID NO: 10, heavy chain constant region SEQ ID NO: 14, light chain constant region SEQ ID NO: 15) to methods known to those skilled in the art. Prepared.
The present inventors have derived the above-mentioned three types of anti-human PD-1 antibody-derived arms (also referred to as "anti-PD-1 arm" in the present specification), and the above-mentioned two types of anti-human CD3 monoclonal antibodies. Arms (hereinafter also referred to as "anti-CD3 arm") were combined according to the purpose of the experiment to prepare a bispecific antibody containing an anti-PD-1 arm and an anti-CD3 arm by a method known to those skilled in the art. In the following examples, the bispecific antibody in which the two types of arms are combined is, for example, an arm derived from clone 949, which is an anti-human PD-1, and an arm derived from OKT3, which is an anti-human CD3 antibody. Bispecific antibodies are referred to as "clone 949 / OKT3", "OKT3 / clone 949", "clone 949 // OKT3" or "OKT3 // clone 949".

(Example 2) Evaluation of PD-1 signal-inducing activity of bispecific antibodies containing anti-PD-1 arm and anti-CD3 arm (SHP-2 recruitment)
The PD-1 signal-inducing activity of bispecific antibodies, including anti-CD3 and anti-PD-1 arms, is a dephosphorylating enzyme that interacts with the intracellular domain of PD-1 during PD-1 signal induction. It was evaluated using the proximity of SHP2 (SHP-2 recruitment) as an index. The NanoBRET® PD-1 / SHP2 Interaction Assay System (Promega) was used for the evaluation. The inducible activity of the PD-1 signal was determined by the ratio of the fluorescence signal (618 nm) by BRET when PD-1 and SHP2 were in close proximity to the emission (460 nm) derived from the donor SHP2.
Antigen-presenting cells (Promega, # J109A) expressing PD-L1 the day before the assay were placed in F-12 medium (Gibco, 11765-054) containing 10% FBS in 4.0 x 10 ⁴ cells / 100 mL / well 96 well plate. It was sown in (Costar, # 3917) and ^{cultured in a CO 2} incubator at 37 ° C for 16 to 24 hours. On the day of the assay, HaloTag® nanoBRET® 618 Ligand (Promega, # G980A) was diluted 250-fold with Opti-MEM (Gibco, # 31985-062). The medium of the cultured PD-L1 expressing antigen-presenting cells was removed, and 25 μL / well of diluted HaloTag® nanoBRET® 618 Ligand was added. PD-L1 inhibitor antibody (YW243.55S70) (see US2016 / 0222117 A1) (may be referred to as "anti-PD-L1" or "PDL1b" in this example and drawing) 0.08 μg / mL An evaluation sample (40, 8, 1.6 μg / mL) diluted with Opti-MEM containing Opti-MEM was added at 25 μL / well. For the amino acid sequence of the variable region of the PD-L1 inhibitory antibody (YW243.55S70), refer to the description of US2016 / 0222117 A1 (heavy chain variable region SEQ ID NO: 16, light chain variable region SEQ ID NO: 17). In the PD-L1 inhibitory antibody, the heavy chain constant region is F1332 (SEQ ID NO: 18), which is a modified Fc body based on human IgG1, and the light chain constant region is k0MT (SEQ ID NO: 19), which is a human κ chain constant region. ) Was used. Add PD-1 / SHP2 Jurkat cell (Promega, # CS2009A01) to the ^{above 96 well plate at 5 x 10 4} cells / 50 μL / well, suspend well and then in a 5% CO ² incubator at 37 ° C for 2.5 hours. It was cultured. nanoBRET® Nano-Glo® substrate (Promega, # N157A) was diluted 100-fold with Opti-MEM, added to a 96 well plate cultured at 25 mL / well and allowed to stand at room temperature for 30 minutes. Later, Em460 mM and Em618 nm were measured with Envision (PerkinElmer, 2104 EnVision).
The obtained values were applied to the following

equations

1 and 2 to calculate the BRET Ratio (mBU).

[Equation 1]

[Equation 2]

The results of the assay are shown in FIG. As shown in the figure, the SHP2 recruitment of bispecific antibodies containing anti-PD-1 and anti-CD3 arms, i.e. the inducing activity of PD-1 signaling, was not clearly shown.

(Example 3) Evaluation of bispecific antibody containing anti-PD-1 arm and anti-CD3 arm in T cell proliferation assay The bispecific antibody containing anti-PD-1 arm and anti-human CD3 arm is primary CD4 positive. In order to evaluate whether it has inhibitory activity on T cells, the inhibitory activity on CD4-positive T cells collected from fresh blood collected from healthy subjects is referred to as anti-human CD3 arm and anti-KLH antibody (referred to as "IC17" in the present specification. ) Derived arm (heavy chain variable region SEQ ID NO: 112, light chain variable region SEQ ID NO: 113, heavy chain constant region SEQ ID NO: 11, and light chain constant region SEQ ID NO: 13) and bispecific. Sexual antibodies (referred to as "OKT3 / IC17" in FIG. 2) were evaluated as negative controls. On the day of the evaluation, blood collected from a healthy person was mixed with an equal amount of RPMI1640 (Nakarai, # 30264-56), and then gently layered in a centrifuge tube filled with 15/50 amount of Ficoll in the blood mixing solution. After centrifuging at 400 g for 35 minutes in a centrifuge set at 20 ° C, the leukocyte fraction was separated into a new centrifuge tube in the upper layer of Ficoll, and the size was increased to 50 mL with RPMI1640 (Nakarai, # 30264-56). The cells (PBMC) pellets were collected by centrifuging at 500 g for 15 minutes in a centrifuge at 20 ° C. Human CD4 + T Cell Isolation Kit (miltenyi, Cat # 130-096-533) was used to collect human CD4 + T Cell Isolation Kit (miltenyi, Cat # 130-096-533) from the collected PBMC according to the procedure of the kit. Then, the obtained CD4 positive T cells were added to a 96 well round bottom plate at 7.6 x 10 ⁵ cells / well, and 10 ng / mL of IL-2 and the evaluation antibody (20, 5, 1.25, 0.31, 0.08 μg /). mL) was cultured in the presence. After culturing for 5 days, cell proliferation was evaluated using the Cell Titer-Glo® Luminescent Cell Viability Assay (promega, G7570) according to the procedure of the kit. The results are shown in FIG.

As shown in FIG. 2, as a result of examining bispecific antibodies containing three types of anti-PD-1 arms and anti-human CD3 arms, these bispecific antibodies were found to be negative control anti-human CD3 arms and anti-human CD3 arms. Rather, it promoted the proliferation of human CD4-positive T cells compared to bispecific antibodies containing arms derived from anti-KLH antibodies. From this result, it is difficult to suppress the proliferation of CD4 + T cells by bispecific antibodies containing anti-PD-1 arm and anti-human CD3 arm, and to suppress the immune response caused by CD4 + T cells. Was suggested.

(Example 4) Preparation of human PD-1 extracellular domain protein
Based on the gene sequence information (NM_005018) encoding human PD-1 obtained from GenBank, HMM + 38 (base sequence is SEQ ID NO: 20, amino acid), which is an artificially secreted signal sequence for the extracellular domain (P21-Q167). The sequence was added with SEQ ID NO: 21) at the N-terminal. In addition, a linker sequence, a histidine tag, a linker sequence, and a BAP tag (biotinylated tag, GLNDIFEAQKIEWHE) were added to the C-terminal to prepare an expression construct (base sequence: SEQ ID NO: 22, amino acid sequence: SEQ ID NO: 23). The gene of the above design was synthesized and subcloned into an in-house constructed vector (pBEF-OriP) for mammalian cells to obtain an expression vector. By introducing this expression vector into Expi293® F cells (Thermo Fisher), human PD-1 extracellular domain protein was transiently expressed. The obtained culture supernatant was purified using an AKTA® Avant25 device (GE healthcare) or a Pure25 device (GE healthcare) and a Pure150 device (GE healthcare). The purification method was ion exchange chromatography using HiTrap (registered trademark) Q HP column (GE healthcare) or Q Sepharose FF resin (GE healthcare), and affinity chromatography using HisTrap (registered trademark) HP column (GE healthcare). After concentrating the fraction containing human PD-1 extracellular domain protein with an ultrafiltration membrane, Superdex® 200 increase 10/300 column (GE healthcare) or HiLoad 16/600 Superdex® 200 pg column By further increasing the purification purity by gel filtration chromatography (size exclusion chromatography) using (GE healthcare) and removing multimers, a monomer fraction of high-purity human PD-1 extracellular domain protein can be obtained. Obtained.

In order to obtain a human PD-1 extracellular domain protein in which Biotin is added to a BAP tag, an expression vector of BirA, which is a Biotin ligase, is used together with an expression vector of the human PD-1 extracellular domain protein in Expi293 (registered trademark) F cells. Biotin was added to the target protein by introducing it using ExpiFectamine (registered trademark) 293 Transfection Kit. Purification was performed by ion exchange chromatography using a HiTrap (registered trademark) QHP column (GE Healthcare) and affinity chromatography using a HisTrap (registered trademark) HP column (GE Healthcare) in the same procedure as described above. Then, affinity chromatography using Streptavidin Mutein Matrix (Roche) was performed for the purpose of increasing the purity of the Biotin adduct. After concentrating the fraction containing human PD-1 extracellular domain protein with an ultrafiltration membrane, gel filtration chromatography (size exclusion chromatography) using a HiLoad 16/600 Superdex® 200 pg column (GE Healthcare) ) To further increase the purification purity and remove multimers to obtain a high-purity Biotin adduct human PD-1 extracellular domain protein monomer fraction.

(Example 5) Preparation of human PD-L1 extracellular domain protein
Based on the gene sequence information (NM_014143) encoding human PD-L1 obtained from GenBank, HMM + 38 (base sequence is SEQ ID NO: 20, amino acid), which is an artificially secreted signal sequence for the extracellular domain (F19-R238). SEQ ID NO: 21) was added to the N-terminal of the sequence, and a linker sequence, histidine tag, linker sequence, and BAP tag (biotinylated tag, GLNDIFEAQKIEWHE) were added to the C-terminal to prepare an expression construct (base sequence is SEQ ID NO: SEQ ID NO: 24, the amino acid sequence is SEQ ID NO: 25). The gene of the above design was synthesized and subcloned into a vector for mammalian cells (pBEF-OriP) to obtain an expression vector. By introducing this expression vector into FreeStyle® CHO cells (Thermo Fisher), human PD-L1 extracellular domain protein was transiently expressed. The obtained culture supernatant was purified using an AKTA® Avant25 device (GE healthcare) or a Pure25 device (GE healthcare) and a Pure150 device (GE healthcare). The purification method was ion exchange chromatography using HiTrap (registered trademark) Q HP column (GE healthcare) or Q Sepharose FF resin (GE healthcare), and affinity chromatography using HisTrap (registered trademark) HP column (GE healthcare). After concentrating the fraction containing human PD-L1 extracellular domain protein with an ultrafiltration membrane, HiLoad 16/600 Superdex® 200 pg column (GE healthcare) or HiLoad 26/600 Superdex® 200 pg A monomer fraction of high-purity human PD-L1 extracellular domain protein by further increasing the purification purity and removing multimers by gel filtration chromatography (size exclusion chromatography) using a column (GE healthcare). Got

To prepare the human PD-L1 extracellular domain protein in which Biotin is added to the BAP tag, the above-purified human PD-L1 extracellular domain protein is mixed with the purified protein of BirA enzyme (biotin ligase) for the Biotin addition reaction. I went by doing. The final concentrations were 35 μM for human PD-L1 extracellular domain protein, 10 mM for ATP, 0.825 μM for BirA enzyme, 50 mM Tris-HCl pH 8.3, 10 mM Mg (OAc) ₂ , 0.05 mM Biotin, and mixed. It was reacted by leaving it at 4 degrees for about 16 hours. Unreacted Biotin was removed by replacement with 20 mM NaPhospate, 600 mM NaCl, pH 7.2 by dialysis using Slide-A-Lyzer Dialysis Cassettes, 3.5K (Thermo Fisher). Then, affinity chromatography using Streptavidin Mutein Matrix (Roche) was performed for the purpose of increasing the purity of the Biotin adduct. After concentrating the fraction containing human PD-L1 extracellular domain protein with an ultrafiltration membrane, gel filtration chromatography using a Superdex® 200 Increase 10/300 GL column (GE Healthcare) (size exclusion chromatography) ) To further increase the purification purity and remove multimers to obtain a high-purity Biotin adduct human PD-L1 extracellular domain protein monomer fraction.

(Example 6) Preparation of human PD-L1 / human PD-1 complex protein
(1) Preparation of linker-fused human PD-L1 / human PD-1 complex protein With variable region F19-N131 of human PD-L1 immunoglobulin domain and variable region N33-E146 of human PD-1 immunoglobulin domain By inserting a linker sequence between the two, stabilization as a complex was achieved by bringing the two closer together. The design is shown in FIG. For human PD-1, C93S modification was introduced to remove free cysteine. The version in which a linker was inserted between N131 of human PD-L1 and N33 of human PD-1 was designated as Linker complex # 1 (base sequence: SEQ ID NO: 26, amino acid sequence: SEQ ID NO: 27) (Fig. 3, left). ). In addition, the version in which a linker was inserted between E146 of human PD-1 and F19 of human PD-L1 was designated as Linker complex # 2 (base sequence: SEQ ID NO: 28, amino acid sequence: SEQ ID NO: 29) (Fig.). 3 right). For both complex proteins, a signal sequence (HMM + 38) was added to the N-terminus, and a linker sequence, histidine tag, linker sequence, and BAP tag (GLNDIFEAQKIEWHE) were added to the C-terminus to prepare an expression construct. By introducing this expression vector into FreeStyle® CHO cells (Thermo Fisher), the protein was transiently expressed. The resulting culture supernatant was subjected to ion exchange chromatography by AKTA® Avant25 instrument (GE Healthcare) and HiTrap® Q HP column (GE Healthcare), affinity by HisTrap® HP column (GE Healthcare). Purified by chromatography. After concentrating the fraction containing human PD-L1 / human PD-1 complex protein with an ultrafiltration membrane, gel filtration chromatography using a HiLoad 26/600 Superdex® 200 pg column (GE Healthcare) ( High-purity human PD-L1 / human PD-1 complex protein was obtained by further increasing the purification purity and removing multimers by size exclusion chromatography).

To prepare the human PD-L1 / human PD-1 complex protein in which Biotin is added to the BAP tag, the above-mentioned purified human PD-L1 / human PD-1 complex protein is purified by the BirA enzyme (biotin ligase). This was done by mixing with protein and performing a Biotin addition reaction. The final concentrations are 25 μM or 26 μM for human PD-L1 / human PD-1 complex protein, 10 mM for ATP, 0.825 μM for BirA enzyme, 50 mM Tris-HCl pH 8.3, 10 mM Mg (OAc) 2, 0.05 mM Biotin. It was mixed so that it would react by leaving it at 4 degrees for about 16 hours. Unreacted Biotin was removed by replacement with 20 mM NaPhospate, 600 mM NaCl, pH 7.2 by dialysis using Slide-A-Lyzer G2 Dialysis Cassettes, 10K (Thermo Fisher). Then, affinity chromatography using Streptavidin Mutein Matrix (Roche) was performed for the purpose of increasing the purity of the Biotin adduct. After concentrating the fraction containing human PD-L1 / human PD-1 complex protein with an ultrafiltration membrane, gel filtration chromatography using Superdex® 200 Increase 10/300 GL column (GE Healthcare) ( By further increasing the purification purity and removing multimers by size exclusion chromatography), a fraction of high-purity Biotin adduct human PD-L1 / human PD-1 complex protein was obtained.

(2) Preparation of disulfide bond-introduced human PD-L1 / human PD-1 complex protein Ectopic by substituting cysteine residues for the amino acid residues that interact with human PD-L1 and human PD-1. The formation of a specific disulfide bond was promoted to stabilize the complex.

The residue to be replaced with cysteine was selected from the structural modeling based on the reported structural data of the cocrystal (PDB: 3BIK). As the molecular design software, the functions of Design Protein tools of Discovery Studio (Dassault Systèmes Co., Ltd.) and Protein Design of MOE (Chemical Computing Group) were used. With these software, C atoms that are close to each other between human PD-L1 and human PD-1 are selected, and residues that can form disulfide bonds by substituting with cysteine from modeling using rotamer are selected. bottom. The combination of the Y56C variant of human PD-L1 and the A132C variant of human PD-1 was designated as SS complex # 1, and the A18C variant of human PD-L1 and the G90C variant of human PD-1 were combined. The version is SS complex # 2. FIG. 4 illustrates the design of a disulfide bond-introduced human PD-L1 / human PD-1 complex protein. For the purpose of increasing the efficiency of formation of the heterodimer between human PD-L1 and human PD-1 and facilitating purification during expression purification, human PD-L1 and human PD-1 are used as heterodimers. We proceeded with expression purification by fusing with Fc including modification of Knobs into Hole, which is a technology that promotes formation.

Human IgG1 type containing the hinge region for the variable region F19-A132 of the immunoglobulin domain of human PD-L1 linked to the linker sequence, BAP tag (GLNDIFEAQKIEWHE), linker sequence, PreScission protease recognition sequence (LEVLFQGP), and linker sequence. Fused to the N-terminus of the Fc domain. Modifications to the Fc domain to lack binding activity to Fcγ receptors and complement C1q (L235R, G236R, S239K) and Hole modifications for Knobs into Hole technology to promote heterodimer formation (D356C) , T366S, L368A, Y407V) was introduced. A His tag for purification was added to the C-terminal of the fusion protein containing PD-L1 via a linker sequence.

On the other hand, for the variable region N33-E146 (including C93S modification) of the immunoglobulin domain of human PD-1, the linker sequence, BAP tag (GLNDIFEAQKIEWHE), linker sequence, Precision protease recognition sequence (LEVLFQGP), and linker sequence are linked. It was fused to the N-terminus of the Fc domain of human IgG1 type including the hinge region. Modifications (L235R, G236R, S239K) and Knob modifications (Y349C, T366W) were introduced into the Fc domain to delete the binding activity to the Fcγ receptor and complement C1q. A FLAG tag for purification was added to the C-terminal of the fusion protein containing PD-1 via a linker sequence.

For each protein, HMM + 38, which is an artificial secretory signal sequence, was added to the N-terminal to prepare an expression construct.
The base sequence of hPDL1 (Y56C) _BAP_PreSission_hole_His10 used for SS complex # 1 is shown in SEQ ID NO: 30, the amino acid sequence is shown in SEQ ID NO: 31, and the base sequence of hPD1 (A132C) _BAP_PreSission_knob_FLAG is shown in SEQ ID NO: 32. Is shown in SEQ ID NO: 33. The base sequence of hPDL1 (A18C) _BAP_PreSission_hole_His10 used for SS complex # 2 is shown in SEQ ID NO: 34, the amino acid sequence is shown in SEQ ID NO: 35, the base sequence of hPD1 (G90C) _BAP_PreSission_knob_FLAG is shown in SEQ ID NO: 36, and the amino acid sequence is SEQ ID NO: 37.

The gene of the above design was synthesized and subcloned into a vector for mammalian cells (pBEF-OriP) to obtain an expression vector. By introducing this expression vector into FreeStyle® CHO cells (Thermo Fisher), the protein was transiently expressed.

The target protein was purified from the obtained culture supernatant by chromatography using an AKTA (registered trademark) 10S device (GE Healthcare) or an AKTA (registered trademark) Avant25 device (GE Healthcare). First, affinity chromatography using ANTI-FLAG M2 Affinity Gel (Sigma-Aldrich) is performed, and then affinity chromatography using HisTrap (registered trademark) HP column (GE Healthcare) is performed to obtain a heterodimer. The forming fraction was obtained. By reacting this heterodimer protein with Turbo3C protease (Accelagen) that cleaves the PreScission recognition sequence, the human PD-L1 / human PD-1 complex protein was separated from the Fc domain. 80 units of Turbo3C protease was mixed with 1 mg of human PD-L1 / human PD-1 complex protein, and the mixture was filled in Slide-A-Lyzer G2 Dialysis Cassettes, 3.5K (Thermo Scientific), and D. The cleavage reaction was carried out at 4 ° C. for about 16 hours while substituting with -PBS (-). After that, in order to remove unnecessary substances such as cleaved Fc domain, uncut Fc fusion, and Turbo3C protein (with His tag attached) contained in the sample, HisTrap (registered trademark) HP column (GE Healthcare) ) And HiTrap (registered trademark) Protein A HP column (GE Healthcare) connected in series, and the sample was passed. After concentrating the obtained fraction containing the cleaved human PD-L1 / human PD-1 complex protein with an ultrafiltration membrane, HiLoad 26/600 Superdex (registered trademark) 200 pg (GE Healthcare) was used. High-purity human PD-L1 / human PD-1 complex protein was obtained by further increasing the purification purity and removing multimers by gel filtration chromatography (size exclusion chromatography).

To prepare the human PD-L1 / human PD-1 complex protein in which Biotin is added to the BAP tag, the above-mentioned purified human PD-L1 / human PD-1 complex protein is purified by the BirA enzyme (biotin ligase). This was done by mixing with protein and performing a Biotin addition reaction. The final concentrations are 21 μM or 22 μM for human PD-L1 / human PD-1 complex protein, 10 mM for ATP, 0.825 μM for BirA enzyme, 50 mM Tris-HCl pH 8.3, 10 mM Mg (OAc) 2, 0.05 mM Biotin. It was mixed so that it would react by leaving it at 4 degrees for about 16 hours. After concentrating this sample with an ultrafiltration membrane, it is unnecessary to connect HisTrap (registered trademark) HP (GE Healthcare) and Superdex (registered trademark) 200 Increase 10/300 GL (GE Healthcare) in series and flow the sample. High-purity Biotin adduct human PD-L1 / human PD-1 complex protein by adsorbing BirA, further increasing the purification purity by gel filtration chromatography (size exclusion chromatography), and removing multimers. I got a fraction.

(Example 7) Preparation of anti-human PD-L1 / human PD-1 complex protein antibody from rabbit Linker, which is each of the four types of human PD-L1 / human PD-1 complex proteins prepared as described above. Monoclonal antibodies were produced from rabbits using complex # 1, Linker complex # 2, SS complex # 1, and SS complex # 2 as antigens. 12-13 week old NZW rabbits were immunosensitized with each of the four antigens. The antigenic protein solution was mixed with an equal volume of TiterMax Gold adjuvant and administered intradermally to rabbits. Immunization was repeated 4 times at intervals of 1 week or longer. Linker complex # 1 and Linker complex # 2 were administered at 200 μg / head during the initial immunization and 100 μg / head during the subsequent immunization. SS complex # 1 and SS complex # 2 were administered at 100 μg / head at all times of immunization. One week after the final immunosensitization, spleen and blood were taken from the immunosensitizer.

In order to concentrate antigen-specific B cells contained in spleen and blood cells, negative selection of cells that bind to human PD-L1 extracellular domain protein was performed by MACS using autoMACS Pro Separator (miltenyi Biotec). After that, a positive selection was performed to concentrate B cells expressing rabbit IgG antibody. In the negative selection for PD-L1 extracellular domain protein, Anti-Biotin MicroBeads (Miltenyi Biotech, 130-090-485) is used by binding human PD-L1 extracellular domain protein with Biotin added to the above-mentioned BAP tag. board. In the positive selection of B cells expressing rabbit IgG antibody, Mouse Anti-Rabbit IgG-PE (SouthernBiotech, 4090-9) is bound to Anti-mouse IgG MicroBeads (Miltenyi Biotech, 130-048-401). board.

In order to further concentrate the cells that had undergone the enrichment operation by MACS against the antibody specific to the human PD-L1 / human PD-1 complex protein, sorting was performed using a cell sorter (FACSariaIII, BD). .. Streptavidin-APC (Miltenyi Biotech, 130-106-) by binding each of the four types of immune antigens (human PD-L1 used for immunization / BAP tag of human PD-1 complex protein with biotin added) Staining according to 791) was performed. At the same time, human PD-1 protein (R & D, 1086-PD) fused with human IgG1 type Fc is bound and stained with Goat anti-Human IgG Fc Cross-Adsorbed Secondary Antibody, DyLight 488 (ThermoFisher, SA5-10134). Was done. B cells expressing IgG-type antibody by Anti-Rabbit IgG-PE (SouthernBiotech, 4090-9) were also selected. Using a cell sorter, B cells that bound to the human PD-L1 / human PD-1 complex protein, did not bind to the human PD-1 protein, and expressed IgG-type antibody were recovered.

The resulting B cells were seeded at a density of 1 / well on a 384-well plate with 12,500 / well EL4 cells (European Collection of Cell Cultures) and 20-fold diluted activated rabbit T cell conditioned medium, 6 After culturing for ~ 12 days, the cells were subjected to antibody screening using the secreted antibody in the B cell culture supernatant. EL4 cells were previously subjected to X-ray irradiation with an irradiation dose of 10 Gy using an X-ray irradiation device MX-160Labo (mediXtec) and subjected to culture.
Activated rabbit T cell conditioned medium is RPMI-1640 containing rabbit thymocytes, phytohaemagglutinin M (Roche, Catalog No. 11082132), phorbol 12-millistate 13-acetate (Sigma, Catalog No. P1585) and 2% FBS. Prepared by culturing in medium. After culturing, the culture supernatant of B cells was collected for further evaluation and the cell pellet was cryopreserved.

Using the secreted antibody contained in the B cell culture supernatant, screening was performed using the antibody binding characteristics as an index. MagPlex Microspheres on which human PD-1 extracellular domain protein, human PD-L1 extracellular domain protein, the above-mentioned four human PD-L1 / human PD-1 complex proteins, and negative control BSA are immobilized. It was mixed with Luminex) and the binding to various proteins was confirmed with a flow cytometer (iQueScreener, intellicyt). Different proteins were analyzed by solid phase on MagPlex Microspheres having different fluorescence patterns. After streptavidin was solid-phased in MagPlex Microspheres, each biotinylated protein to be evaluated was solid-phased and used for evaluation. Goat anti-Rabbit IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 488 (Thermo Fisher Scientific, A-11034) was used as the secondary antibody. Data analysis was performed using FlowJo software (Tomy Digital-Biolology). From the results of this experiment, it does not bind to the negative control BSA, human PD-1 extracellular domain protein, or human PD-L1 extracellular domain protein, but only to each human PD-L1 / human PD-1 complex protein. B cells that produce antibodies were selected.

Selected B cell RNA was purified from cryopreserved cell pellets using the ZR-96 Quick-RNA kit (ZYMO RESEARCH, Catalog No. R1053). F1332 (SEQ ID NO:) is a human IgG1 heavy chain constant region in which the DNA encoding the antibody heavy chain variable region is amplified by reverse transcription PCR using the prepared RNA and the binding activity to the Fcγ receptor and complement C1q is deleted. : 18) was recombined in-frame with the encoding DNA. In addition, the DNA encoding the antibody light chain variable region was amplified by reverse transcription PCR, and in the case of the antibody light chain variable region derived from rabbits, k0MTC (SEQ ID NO: 38), which is a modified human κ type light chain constant region, was obtained. It was recombined in-frame with the encoding DNA. As described above, an expression vector having the heavy chain and light chain coding sequences of the recombined antibody was prepared, and the mixture was mixed so that the ratio of the heavy chain to the light chain expression vector was 1: 1 (Expi293). The antibody was transiently expressed by introduction into F cells (Thermo Fisher) using the ExpiFectamine (registered trademark) 293 Transfection Kit. Purified antibodies were prepared from the obtained culture supernatant by affinity purification using Protein A and used for subsequent evaluation.
Table 1 shows the variable and constant region SEQ ID NOs of the heavy and light chains of 30 clones of the prepared anti-human PD-L1 / human PD-1 complex protein antibody (signal peptide and variable of antibody sequence). Regarding the boundary with the area, the prediction result using the Signal Peptide prediction function of GENETY X-SV / RC software (Ver.15.0.3) was followed.).

(Example 8) Evaluation of specificity of anti-human PD-L1 / human PD-1 complex antibody The cell surface of 30 clones of the anti-human PD-L1 / human PD-1 complex protein antibody prepared in Example 7 The binding specificity for the human PD-L1 / human PD-1 complex formed above was evaluated. The binding was evaluated using a flow cytometer (iQue Screener, intellicyt). A solution (called FACS buffer) in which BSA was added to D-PBS (-) to a final concentration of 0.1% was used for diluting cells and proteins to be added, and for washing operations. For evaluation, DXB11s cells were used as the parent strain (parent / CHO), and the human PD-L1 gene (base sequence: SEQ ID NO: 110, amino acid sequence: SEQ ID NO: 111) was introduced on the cell membrane surface to introduce human PD-. By introducing a stable expression strain (hPD-L1 / CHO) in which L1 was forcibly expressed, and a full-length human PD-1 gene (base sequence: SEQ ID NO: 108, amino acid sequence: SEQ ID NO: 109), it was introduced onto the cell membrane surface. A stable expression strain (hPD-1 / CHO) in which human PD-1 was forcibly expressed was prepared. To preform the human PD-L1 / human PD-1 complex protein on the cell surface, hPD-L1 / CHO is mixed with 10 μg / mL soluble human PD-1 extracellular domain (ECD) protein. , HPD-1 / CHO was mixed with 10 μg / mL soluble human PD-L1 extracellular domain (ECD) protein and placed on ice for at least 30 minutes. After adding purified antibody prepared at 20 μg / mL to a 384-well V-bottom plate (Greiner) at 10 μL / well, the above-mentioned cell solution in which the human PD-L1 / human PD-1 complex was previously formed was added to 1 Addition was made at 10 μL / well to 30,000 cells per well. After incubating at 4 ° C. for 30 minutes to bind the antibody, the supernatant was removed by suction after centrifugation. Subsequently, a washing operation of adding 50 μL / well FACS buffer and then suction-removing the supernatant after centrifugation was repeated twice. Goat Anti-Human IgG Fc Cross-Adsorbed Secondary Antibody. DyLight 650 (Thermo Fisher Scientific, SA5-10137) was used as the secondary antibody for detecting binding.

The efficiency of formation of human PD-L1 / human PD-1 complex protein on the cell surface was determined by the biotectic acid added to the soluble human PD-1 extracellular domain protein and the soluble human PD-L1 extracellular domain protein. It was confirmed by adding Streptavidin-APC (Miltenyi Biotec, 130-106791) and detecting it.
Data analysis was performed using FlowJo software (Tomy Digital-Biolology).
The obtained combined data are shown in FIG. 5 and Table 2. Multiple antibodies with high binding selectivity to the PD-L1 / PD-1 complex were obtained.

(Example 9) Preparation of anti-human PD-1 antibody A plurality of anti-human PD-1 antibodies capable of binding to human PD-1 and inducing its immunosuppressive signal were prepared. PD1-17, antibody949 (clone 949), clone 10 described in Example 1 to PDA0129 (heavy chain variable region SEQ ID NO: 99, light chain variable region SEQ ID NO: 100, heavy chain constant region SEQ ID NO: 18, light). Four clones of anti-human PD-1 antibody to which the chain constant region SEQ ID NO: 38) was added were prepared by a method known to those skilled in the art.

An anti-human PD-1 monoclonal antibody 5C4 (heavy chain variable) known to inhibit the binding between human PD-1 and human PD-L1 as a positive control for use in the evaluation of the ligand binding inhibitory activity described in Example 11. Region SEQ ID NO: 101 and light chain variable region SEQ ID NO: 102 (see WO2006 / 121168), and heavy chain constant region SEQ ID NO: 105 and light chain constant region SEQ ID NO: 15) and Pembrolizumab (heavy chain variable region SEQ ID NO: 15). : 103 and light chain variable region SEQ ID NO: 104 (see WO2016 / 137850), and heavy chain constant region SEQ ID NO: 105 and light chain constant region SEQ ID NO: 15) were also prepared by methods known to those skilled in the art.

(Example 10) Evaluation of PD-1 signal-inducing activity of anti-human PD-L1 / human PD-1 complex antibody (SHP-2 recruitment)
Whether anti-human PD-1 antibodies (clone949 and PDA0129) have PD-1 signal-inducing activity, interact with the intracellular domain of PD-1 during PD-1 signal induction by the same procedure as in Example 2. The proximity of SHP2, which is a dephosphorylating enzyme, was used as an index for evaluation.
The results of the assay are shown in FIG. It was suggested that the two anti-human PD-1 antibodies do not have clear SHP2 recruitment-inducing activity, i.e., PD-1 signaling-inducing activity.

Whether the anti-human PD-L1 / human PD-1 complex antibody produced in Example 7 has the PD-1 signal-inducing activity was determined by the same procedure as in Example 2 during PD-1 signal induction. The proximity of SHP2, a dephosphorylating enzyme that interacts with the intracellular domain of -1, was evaluated as an index.
The results of the assay are shown in FIG. The anti-human PD-L1 / human PD-1 complex antibody, unlike the anti-human PD-1 antibody (clone949 and PDA0129) described above, has an activity of inducing SHP2 recruitment, that is, an activity of inducing PD-1 signal. It has been shown.

(Example 11) Evaluation of ligand binding inhibitory activity of anti-human PD-1 antibody The human PD- of the anti-human PD-1 antibody (PD1-17, clone 949, clone 10, and PDA0129) prepared in Example 9 The inhibitory activity on the binding between 1 and human PD-L1 was evaluated. Evaluation of inhibitory activity is an indicator of whether the binding strength of soluble human PD-L1 extracellular domain (ECD) protein that binds to human PD-1 expressing cells is attenuated in the presence of anti-human PD-1 antibody. Went to. The binding was evaluated using a flow cytometer (iQue Screener, intellicyt). A solution (called FACS buffer) in which BSA was added to D-PBS (-) to a final concentration of 0.1% was used for diluting cells and proteins to be added, and for washing operations. For evaluation, a stable expression strain in which human PD-1 was forcibly expressed on the cell membrane surface by introducing the above-mentioned full-length human PD-1 gene (base sequence: SEQ ID NO: 108, amino acid sequence: SEQ ID NO: 109) (stable expression strain). hPD-1 / CHO) was used. A series of staining operations were performed using a V-Bottom 96 Well Cell Culture Plate (costar, 3894). An antibody solution (100 μL / well) diluted to a concentration of 20, 4, 0.8 μg / mL with FACS buffer ^{was prepared into 3 x 10 6} cells / mL, and hPD-1 / CHO cells 50 μL / well (150000 cells / well). The antibody was bound by mixing with well) and placing on ice for 30 minutes. It was then mixed with 10 μg / mL biotinylated soluble human PD-L1 extracellular domain (ECD) protein and placed on ice for 15 minutes. After the centrifugation operation, the supernatant was removed by suction, and then 200 μL / well of FACS buffer was added, and then the washing operation of removing the supernatant by suction after the centrifugation operation was repeated twice. Streptavidin-APC for detecting binding was added at 100 μL / well diluted 1000-fold and incubated on ice for 30 minutes. After centrifugation, remove the supernatant by suction, then add 200 μL / well FACS buffer, then remove the supernatant by suction after centrifugation, and finally suspend in 50 μL / well FACS buffer with iQue Screener. Combined data was acquired. The acquired data was analyzed using FlowJo software (Tomy Digital-Biolology).
As shown in FIG. 8, PD1-17 and clone 10 were antibodies that inhibited the binding of human PD-L1 to human PD-1, while clone 949 and PDA0129 were antibodies that did not inhibit the binding.

(Example 12) Preparation of PD-1 agonist antibody using anti-human PD-L1 / human PD-1 complex antibody The present inventors have prepared the anti-human PD-L1 / human PD-1 complex described above. Based on the body antibody, we investigated further antibody molecular types with stronger agonist activity against PD-1. Specifically, PD-L1 induced PD at the local immunological synapses and TCR microclusters where the endogenous PD-L1 / PD-1 complex is formed and exerts its immunosuppressive regulatory function. We investigated a molecular concept that could further enhance the immunosuppressive signal of -1. The present inventors considered that bispecific antibodies containing an anti-PD-L1 / PD-1 complex arm and an anti-PD-1 arm satisfy the molecular type concept, and evaluated them.

As the anti-human PD-1 antibody that is the source of the designed bispecific antibody, the four clones PD1-17, antibody949 (clone949), clone10, and PDA0129 described in Example 9 were used. Table 3 summarizes the presence or absence of inhibitory activity of these anti-human PD-1 antibodies shown in Example 11 against the binding of human PD-L1 to human PD-1.

The anti-PD-L1 / PD-1 complex antibody that is the source of the bispecific antibody was selected from the 30 clones (Table 1) prepared in Example 7 and used. Table 4 shows the combinations of the anti-PD-1 arm and the anti-PD-L1 / PD-1 complex arm in the prepared bispecific antibody, particularly those used in the later examples.

Table 5 shows the variable regions of the heavy and light chains of the antibody that is the source of the bispecific antibody, including the anti-PD-1 arm and anti-PD-L1 / PD-1 complex arm prepared above, and their respective variable regions. The SEQ ID NO: of the constant region is shown (for the boundary between the signal peptide of the antibody sequence and the variable region, the prediction result using the Signal Peptide prediction function of GENETYX-SV / RC software (Ver.15.0.3) was followed.) ..

(Example 13) Evaluation of PD-1 signal-inducing activity of bispecific antibody containing anti-PD-L1 / PD-1 complex arm and anti-PD-1 arm (SHP-2 recruitment)
Whether the bispecific antibody containing the anti-PD-L1 / PD-1 complex arm and the anti-PD-1 arm has the inducing activity of PD-1 signal is determined by the same procedure as in Example 2 of PD-1. The proximity of SHP2, a dephosphorylating enzyme that interacts with the intracellular domain of PD-1 during signal induction, was used as an index.
The results of the assay are shown in FIG. It was shown that bispecific antibodies containing anti-PD-L1 / PD-1 complex arm and anti-PD-1 arm have the activity of inducing SHP2 recruitment, that is, the activity of inducing PD-1 signal. In particular, bispecific antibodies using arms derived from anti-PD-1 antibodies, such as antibody 949 (clone 949) and PDA0129, which do not inhibit the binding of PD-1 to PD-L1, can be used to induce SHP2 recruitment. It was found that the activity was remarkably high.

(Example 14) Evaluation of PD-1 signal-inducing activity using suppression of NFAT activity as an index PD induced by a bispecific antibody containing an anti-PD-L1 / PD-1 complex arm and an anti-PD-1 arm Whether the -1 signal has an immunosuppressive function was evaluated using the activity of the transcription factor NFAT, which acts downstream of the T cell receptor signal responsible for immune activation, as an index. Using the PD-1 / PD-L1 Blockade Bioassay system (Promega), evaluation was performed using the fact that the luminescence reaction of Luciferase occurs downstream of the NFAT response element. In this assay, the TCR stimulator on PD-L1 aAPC / CHO-K1 cells induces the activation signal of TCR on PD-1 Effector Cell, which causes NFAT activation and luminescence of the reporter gene Luciferase is observed. However, the activation of NFAT is almost completely suppressed by the induction of PD-1 signal by PD-L1. However, when PD-1 signal induction is inhibited by the addition of PD-L1 inhibitory antibody, NFAT activation occurs. The present inventors have previously found that in this assay, PD-L1 inhibition of about 30% is applied when the final concentration at which the PD-L1 inhibitory antibody YW243.55S70 is added is 40 ng / mL. In, it was evaluated whether the bispecific antibody induces the suppression of NFAT activation.

The reaction was carried out using a Round well 384 Well White Flat Bottom Polystyrene TC-Treated Microplate (Corning, 3826). First, PD-L1 aAPC / CHO-K1 cells ^{prepared to 2.5 x 10 5} cells / mL were added at 30 μL / well, _{incubated overnight at 37 ° C at 5% CO 2} , and then adhered, and then cultured. After removing the supernatant, an antibody solution mixed with bispecific antibody of each concentration was added at 10 μL / well so that the final concentration of YW243.55S70 was 40 ng / mL. After allowing to stand at room temperature for 15 minutes, PD-1 Effector Cell ^{prepared to 4.6875 x 10 5} cells / mL was added at 10 μL / well and _{incubated at 37 ° C. under 5% CO 2} conditions for 6 hours. The Luciferase substrate solution of the Bio-Glo Luciferase Assay System (Prmega, G7940) was added at 20 μL / well and allowed to stand at room temperature for 5 minutes, and then the luciferase emission was measured with a microplate reader EnVision Xcite (PerkinElmer).

The result of the evaluation is shown in FIG. The generated bispecific antibody containing the anti-PD-L1 / PD-1 complex arm and the anti-PD-1 arm was confirmed in the SHP-2 recruitment assay system using NanoBRET® described above. It was confirmed that not only has SHP2 recruitment activity, but also has a function of suppressing NFAT activity as shown by this assay.

(Example 15) Design of each molecular type concept and function of each arm, and mechanism of action thereof Specific antibody molecule design of PD-1 agonist antibody and function of each arm examined in the above-mentioned Examples. FIG. 11 shows the estimated mechanism of action in FIG.
Both arms of the anti-human PD-L1 / human PD-1 complex antibody (Fig. 11 top and Fig. 12 left) enhance the interaction of PD-L1 with PD-1 in addition to the targeting function to the immunological synapse. By having a function, it was presumed to exert an agonistic action on PD-1 as shown in Example 10. This mechanism of action is not only an anti-complex antibody capable of enhancing the interaction of PD-L1 with PD-1, but also an anti-complex antibody capable of enhancing the interaction of other co-suppressing molecular ligands with other co-suppressing molecules. It can also be applied to complex antibodies.
In a bispecific antibody comprising an anti-PD-L1 / PD-1 complex arm and an anti-PD-1 arm (bottom of FIG. 11 and right of FIG. 12), the anti-PD-L1 / PD-1 complex arm is at least immune. It suffices to have a synapse targeting function, and may or may not have a function of enhancing the interaction of PD-L1 with PD-1. In the bispecific antibody, the anti-PD-1 arm was presumed to exert an agonistic effect on PD-1. This mechanism of action is not only an anti-complex arm that can enhance the interaction of PD-L1 with PD-1, but also an anti that can enhance the interaction of other co-suppressing molecular ligands with other co-suppressing molecules. It can also be applied to complex arms.

(Example 16) Preparation of anti-human PD-L1 / human PD-1 complex antibody G1T7P (sequence) of the heavy chain constant region of the anti-human PD-L1 / human PD-1 complex protein antibody prepared in Example 7 The gene for the antibody replaced with No .: 114) was produced by a method known to those skilled in the art.
The antibody was transiently expressed in mammalian cells by a method known to those skilled in the art using the generated gene encoding the antibody, and was purified by a method known to those skilled in the art as necessary. The combinations of heavy chain variable region, heavy chain constant region, light chain variable region, and light chain constant region of each antibody used for expression are shown in Table 6.

(Example 17) Evaluation of binding specificity of anti-human PD-L1 / human PD-1 complex antibody Human PD-1, of the anti-human PD-L1 / human PD-1 complex protein antibody prepared in Example 16 Binding activity to human PD-L1 and mixtures thereof was assessed using Octet HTX (Molecular Devies).
Evaluation was performed at 30 ° C. using 50 mM phosphoric acid, 150 mM NaCl, 0.05 w / v% P20, pH 7.4 as assay buffer. A Protein A Biosensor (ForteBio, Cat. 18-05010, hereinafter referred to as Pro A biosensor) was used as a biosensor for capturing the antibody. By interacting the antibody solution prepared in the assay buffer with this Pro A biosensor, the antibody was captured at approximately 2.5 nm. Human PD-1 and human PD-L1 used in this measurement were prepared by the methods described in Examples 4 and 5. Human PD-1 and human PD- Each L1 was diluted with assay buffer to a final concentration of 1000 nM and bound to the antibody captured on the biosensor. The chips were regenerated with 10 mM Glycine-HCl (pH 1.5) and the antibody was repeatedly captured and measured. _{The dissociation constant (K D} ) for each antigen of each antibody was calculated using Data Analysis HT 10.0. _{Specifically, the binding rate constant k a} (L / mol / s) and the dissociation rate constant k _dis (1 / s) are calculated by fitting the sensorgrams obtained by the measurement with a 1: 1 binding model. The dissociation constant K _D (mol / L) was calculated from that value. For the binding activity of each antibody to each antigen, the capture amount (nm) and the binding amount to the antigen (nm) of each antibody were calculated using Data Analysis HT 12.0. After dividing the blank binding amount (nm) from this binding amount (nm), the antigen binding amount per unit antibody amount of each antibody was calculated by dividing by the capture amount (nm).
The obtained join data is shown in Tables 7 and 8. It was confirmed that each antibody has high binding selectivity for PD-L1 / PD-1 complex.

(Example 18) Epitope identification by X-ray crystal structure analysis of anti-human PD-L1 / human PD-1 complex antibody Fab fragment, human PD-L1 extracellular domain and human PD-1 extracellular domain complex
LPB0010 prepared in Example 16 to prepare a Fab fragment of the anti-human PD-L1 / human PD-1 complex antibody LPB0010HCb-G1T7P / LPB0010LCb-k0MTC (referred to herein as "LPB0010 Fab"). Full-length antibody is digested with Endoproteinase Lys-C (Roche) at 35 ° C for 2 hours, then HiTrap® MabSelect® SuRe Column (GE Healthcare), HiTrap for Fc fragment removal and purification. Purified on an AKTAxpress ™ device (GE Healthcare) using SP HP column (GE Healthcare) and HiLoad 16/600 Superdex® 75 pg (GE Healthcare).

The purified LPB0010Fab is also referred to as a human PD-L1 extracellular domain protein (hereinafter referred to as "hPD-L1_ECD") prepared by the methods of Example 4 and Example 5 in order to prepare a complex sample for crystallization. ) And human PD-1 extracellular domain protein (hereinafter also referred to as “hPD-1_ECD”) in a molar ratio of about 1: 1: 1 and then Superdex® 200 increases 10/300 GL. Purified by AKTApurifier ™ 10 device (GEHealthcare) by gel filtration chromatography using (GEHealthcare). The fractionated complex fraction was desglycanized with Protein Deglycosylation Mix II (New England Biolabs) at 37 ° C for 4 days, and then equilibrated with 20 mM HEPES buffer pH 7.1, 100 mM sodium chloride (Superdex). Purified with AKTApurifier (trademark) 10 device (GE Healthcare) using 200 increases 10/300 GL (GE Healthcare).

The prepared complex was concentrated to 11 mg / mL using an ultrafiltration membrane and crystallized by the sitting drop vapor diffusion method at 20 ° C in combination with the microseeding method. The reservoir solution for crystallization consisted of 20% (w / v) polyethylene glycol 3350 and 0.2 M magnesium sulfate hydrate. Crystals soaked in a reservoir solution supplemented with 25% (v / v) ethylene glycol for a short time were frozen in liquid nitrogen.

Diffraction images were collected at a temperature of 100 K at the beamline BL45XU of the synchrotron radiation facility SPring-8, and diffraction intensity data with a resolution of up to 1.80 Å was obtained by processing with the AutoPROC (Global Phaseing Ltd.) program. Data collection statistics are shown in Table 9.

Using the obtained X-ray diffraction intensity data, a molecular replacement method using the Phaser (J. Appl. Cryst. 40: 658-674 (2007)) program was carried out, and the initial structure was determined. Crystal structure data of another antibody in the company was used as the search model for LPB0010Fab, and PDBID = 4ZQK (Structure23: 2341-2348 (2015)) was used as the search model for hPD-L1_ECD / hPD-1_ECD. After that, model construction using Coot (ActaCryst. D66: 486-501 (2010)), Refmac5 (ActaCryst. D67: 355-467 (2011)) and Buster version 2.11.8 (Global Phaseing Ltd.) program The refinement was repeated and the final refined coordinates were obtained. The refinement statistics are shown in Table 9.

As shown in FIG. 13, LPB0010Fab binds to the hPD-L1_ECD / hPD-1_ECD complex in both heavy and light chains in a 1: 1 ratio. 14 and 15 show the epitope mapping of LPB0010Fab on the crystal structure and amino acid sequence of hPD-L1_ECD / hPD-1_ECD, respectively. Amino acids containing one or more atoms located within 4.5 Å of any atom in LPB0010Fab are mapped as epitopes. From these results, it was found that LPB0010Fab recognizes both hPD-L1_ECD and hPD-1_ECD molecules by multiple regions on the amino acid sequence, and LPB0010Fab recognizes the hPD-L1_ECD / hPD-1_ECD complex. It was clarified that the three-dimensional structure as an epitope is used as an epitope. 13 and 14 were created using PyMOL Version 2.3 (Schrodinger, LLC.) Software.

As a supplement, although the expression region of hPD-L1_ECD used for crystallization is amino acid number F19-R238 as shown in Example 5, the crystal structure continues after K136 (Ig-C-like domain). ) Does not see the electron density map. Therefore, a model of the Ig-C-like domain of hPD-L1 has not been constructed, but since the amino acids after K136 extend in the direction opposite to the binding site of LPB0010Fab, it is considered that they are not included in the epitope of LPB0010Fab.

(Example 19) Epitope evaluation by negative staining electron microscopic analysis of anti-human PD-L1 / human PD-1 complex antibody Fab fragment, human PD-L1 extracellular domain and human PD-1 extracellular domain complex Anti-human PD -L1 / human PD-1 complex antibody LPB0010HCb-G1T7P / LPB0010LCb-k0MTC Fab fragment (LPB0010 Fab) and LPC0039HCd-G1T7P / LPC0039LCd-k0MTC Fab fragment (referred to as "LPC0039 Fab" in the present specification). To prepare, LPB0010HCb-G1T7P / LPB0010LCb-k0MTC and LPC0039HCd-G1T7P / LPC0039LCd-k0MTC full-length antibodies prepared in Example 16 were digested with Endoproteinase Lys-C (Roche) at 35 ° C for 2 hours and then Fc. HiTrap® MabSelect® SuRe Column (GE Healthcare), HiTrap® SP HP Column (GE Healthcare), and HiLoad 16/600 Superdex® 75 for fragment removal and purification. Purified on an AKTAxpress ™ device (GE Healthcare) using pg (GE Healthcare).

The purified LPB0010Fab is mixed with hPD-1_ECD and hPD-L1_ECD prepared by the methods of Example 4 and Example 5 in a molar ratio of about 1: 1: 1, and then 25 mM HEPES buffer pH. It was purified by AKTApurifier (trademark) 10 device (GE Healthcare) by gel filtration chromatography using Superdex (registered trademark) 200 increases 10/300 GL (GE Healthcare) equilibrated with 7.1, 100 mM sodium chloride. A complex fraction of LPB0010Fab and hPD-L1_ECD / hPD-1_ECD was recovered as a sample for negative staining electron microscope analysis.

The purified LPC0039Fab is a disulfide bond-introduced human PD-L1 / human PD-1 complex protein SS complex # 2 (hereinafter referred to as “hPD-L1_ECD / hPD-1_ECD SS complex # 2”) prepared by the method of Example 6. Superdex® 200 increment 10/300 GL (GE Healthcare), which was mixed with 25 mM HEPES buffer pH 7.1, 100 mM sodium chloride and then equilibrated with a molar ratio of about 1: 1. Purified by gel filtration chromatography using AKTApurifier ™ 10 device (GE Healthcare). As a sample for negative staining electron microscope analysis, a complex fraction of LPC0039Fab and hPD-L1_ECD / hPD-1_ECD S-S complex # 2 was recovered.

Purified LPB0010Fab and hPD-L1_ECD / hPD-1_ECD complex (concentration approx. 0.5 mg / mL) and LPC0039Fab and hPD-L1_ECD / hPD-1_ECD SS complex # 2 complex (concentration approx. 1.0 mg / mL) Is diluted 40-fold and 50-fold with 25 mM HEPES buffer pH 7.1 and 100 mM sodium chloride, respectively, and applied on a grid Cu150 with a support film (Nippon Denshi Co., Ltd.) as a sample of a negative-stained electron microscope. Was done. The applied sample was negatively stained with 2% (w / v) uranyl acetate, and was operated at 120 kV at room temperature with a JEM-1400plus (JEOL Ltd.) electron microscope at a magnification of 60,000 times and a pixel size of 3.058 Å / pixel. The photo was taken under the conditions of (JEOL Ltd.).

From the collected images, the EMAN2 (J Struct Biol. 157: 38-46 (2007)) software suite and the RELION (J Mol Biol.415 (2): 406-418. (2012)) program, respectively, of the particles Picking and classification of two-dimensional averaged images were performed.

FIG. 16A shows five typical two-dimensional average images of the complex of LPB0010Fab and hPD-L1_ECD / hPD-1_ECD. In each average image, it was possible to distinguish each molecule of LPB0010Fab, PD-L1 and PD-1 from the shape (the molecule name is labeled in the figure). From this result, it can be seen that LPB0010 is bound to the interaction interface between PD-L1 and PD-1 (the region indicated by the arrow in No. 1 and No. 4). Therefore, LPB0010Fab is considered to recognize both PD-L1 and PD-1 molecules as epitopes, which is consistent with the results of X-ray crystallography as shown in Example 18.

Furthermore, it was confirmed that each average image corresponds to the X-ray crystal structure (Example 18) of the complex of LPB0010Fab and hPD-L1_ECD / hPD-1_ECD. In order to supplement the hPD-L1_Ig-C-like domain lacking in this crystal structure, the previously reported PD-L1 structure (PDB ID = 4Z18, Fedorov, AA et al. (2015)) and amino acid numbers F19 to N135 The model structure created by superimposing was used. FIG. 16B is a view of the created three-dimensional structure model viewed from various directions so as to correspond to the images of each class of the two-dimensional averaging of FIG. 16A. From this, it can be clearly seen that the result of the negative staining electron microscope analysis is in agreement with the result of the X-ray crystal structure analysis.

In addition, the combination of LPB0010Fab and the linker-fused human PD-L1 / human PD-1 complex Linker complex # 1 (hPD-L1_Ig-C-like domain is not included in the construct) prepared by the method of Example 6). Negative staining of the body Electron microscopic analysis also provided results showing that LPB0010Fab recognizes both PD-L1 and PD-1 molecules as epitopes (data omitted).

FIG. 17 shows five typical two-dimensional average images of a complex of LPC0039Fab and hPD-L1_ECD / hPD-1_ECD S-S complex # 2 (hPD-L1_Ig-C-like domain is not included in the construct). In each average image, it was possible to distinguish each molecule of the LPC0039Fab and hPD-L1 / hPD-1 complex from the shape (the molecule name is labeled in the figure, but PD-L1 and PD). It cannot be distinguished from -1.) From this result, it can be seen that LPC0039Fab is bound to the interaction interface between PD-L1 and PD-1 (the region indicated by the arrow in No. 4), similar to LPB0010Fab. Therefore, it is considered that LPC0039Fab also recognizes both PD-L1 and PD-1 molecules as epitopes.

Claims

An antigen-binding molecule containing a first antigen-binding domain.
An antigen-binding molecule in which the first antigen-binding domain can specifically bind to a first complex consisting of a first co-suppressing molecule and a first co-suppressing molecular ligand for the first co-suppressing molecule.
The specific binding of the first antigen-binding domain to the first complex is either or both of the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex. The antigen-binding molecule according to claim 1, which is an intermolecular force binding to the molecule.
An antigen-binding molecule containing a first antigen-binding domain.
The first antigen-binding domain may bind to a first complex consisting of a first co-suppressing molecule and a first co-suppressing molecular ligand for the first co-suppressing molecule, and said first antigen-binding domain. The binding activity to the first complex of the first co-suppressing molecule that does not form the first complex and the first co-suppressing molecular ligand that does not form the first complex. An antigen-binding molecule that has a higher binding activity to either or both.
The binding of the first antigen-binding domain to the first complex is with one or both of the first co-suppressing molecule and the first co-suppressing molecular ligand in the first complex. The antigen-binding molecule according to claim 3, which is an intermolecular force binding.
The antigen-binding molecule according to claim 3 or 4, wherein the binding activity satisfies any one or both of the following (a) and (b):
(a) In flow cytometry using a first cell that forcibly expresses the first co-suppressing molecule and does not forcibly express the first co-suppressing molecular ligand, the cell of the first co-suppressing molecular ligand. The binding activity of the first antigen-binding domain to the first cell in the presence of the first soluble polypeptide containing the outer domain is higher than that in the absence of the first soluble polypeptide. High;
(b) In flow cytometry using a second cell forcibly expressing the first co-suppressing molecular ligand and not forcibly expressing the first co-suppressing molecule, extracellular of the first co-suppressing molecule. The binding activity of the first antigen-binding domain to the second cell in the presence of the second soluble polypeptide containing the domain is higher than that in the absence of the second soluble polypeptide. high.
The antigen-binding molecule according to any one of claims 1 to 5, which can activate a first intracellular signal downstream of the first co-suppressing molecule at the immunological synapse in which the first complex is present. ..
The antigen-binding molecule according to claim 6, wherein the activation of T cells can be suppressed by the activation of the first intracellular signal.
6. Antigen binding molecule.
Expressing the first co-suppressing molecule and measuring the intensity of the first intracellular signal, the third cell, and the antigen-independent T-cell receptor activator and the first co-suppressing molecular ligand. The first intracellular signal in the third cell is partially suppressed by an inhibitory antibody against the first co-suppressing molecular ligand. In the system for measuring the intensity of the first intracellular signal, which is carried out under the above conditions, the intensity of the first intracellular signal in the presence of the antigen-binding molecule is in the absence of the antigen-binding molecule. The antigen-binding molecule according to any one of claims 6 to 8, which is higher than the above.
The intensity of the second intracellular signal that expresses the T cell receptor and the first co-suppressive molecule and can be activated downstream of the T cell receptor can be measured, and the first intracellular signal can be measured. A fifth cell in which the second intracellular signal can be suppressed by activation of the above, and a sixth cell expressing the antigen-independent T cell receptor activator and the first co-suppressing molecular ligand. Used under conditions that allow contact with each other, and
Suppression of the second intracellular signal in the fifth cell by the first co-suppressing molecular ligand is carried out under conditions where it is partially suppressed by an inhibitory antibody against the first co-suppressing molecular ligand. In the second intracellular signal intensity measurement system,
The antigen-binding molecule according to any one of claims 6 to 9, wherein the intensity of the second intracellular signal in the presence of the antigen-binding molecule is lower than that in the absence of the antigen-binding molecule.
The combination of the first co-suppressing molecule and the first co-suppressing molecular ligand is PD-1 and PD-L1, PD-1 and PD-L2, BTLA and HVEM, TIGIT and CD155. Combination, TIGIT and CD112 combination, LAG-3 and MHC class II molecule combination, CTLA4 and CD80 combination, CTLA4 and CD86 combination, TIM-3 and galectin-9 combination, TIM-3 and phosphatidylserine combination, TIM The antigen-binding molecule according to any one of claims 1 to 10, which is any one selected from the group consisting of a combination of -3 and CEACAM-1 and a combination of TIM-3 and HMGB1.
The antigen-binding molecule is a multiple antigen-binding molecule further comprising a second antigen-binding domain.
One of claims 1 to 11, wherein the second antigen-binding domain can specifically bind to a second co-suppressing molecule capable of forming a second complex with the second co-suppressing molecular ligand. The antigen-binding molecule according to.
The antigen according to claim 12, wherein the binding of the second antigen-binding domain to the second co-suppressing molecule does not compete with the binding of the second co-suppressing molecular ligand to the second co-suppressing molecule. Binding molecule.
The combination of the second co-suppressing molecule and the second co-suppressing molecular ligand is PD-1 and PD-L1, PD-1 and PD-L2, BTLA and HVEM, TIGIT and CD155. Combination, TIGIT and CD112 combination, LAG-3 and MHC class II molecule combination, CTLA4 and CD80 combination, CTLA4 and CD86 combination, TIM-3 and galectin-9 combination, TIM-3 and phosphatidylserine combination, TIM The antigen-binding molecule according to claim 12 or 13, which is any one combination selected from the group consisting of a combination of -3 and CEACAM-1 and a combination of TIM-3 and HMGB1.
The antigen-binding molecule according to any one of claims 12 to 14, wherein the first antigen-binding domain and the second antigen-binding domain are a variable region of an antibody or a fragment thereof that binds to a target antigen thereof. ..