Method for selecting high-affinity compounds for target molecules

Abstract

To provide a method for selecting compounds that bind to a target molecule and then slowly dissociate from it (strongly bind to the target molecule) without performing off-rate selection using an excess amount of the target molecule. [Solution] A method for selecting a compound with high affinity for a target molecule, comprising the following steps. (1) A step of mixing a target molecule (B) with a candidate compound (A) that is a high affinity compound for the target molecule (B) to obtain the candidate compound (A) to which the target molecule (B) is bound, wherein the candidate compound (A) has a structural site (A1) that can bind to the target molecule (B) and a structural site (A2) that can bind to a selective auxiliary material (C), and is capable of binding to only one of the target molecule (B) and the selective auxiliary material (C), the above step (2) A step to obtain candidate compound (A) to which the target molecule (B) obtained in step (1) is bound, by mixing with a selective aid (C) to obtain candidate compound (A) to which the target molecule (B) is bound and candidate compound (A) to which the selective aid (C) is bound. (3) A step in which the candidate compound (A) to which the target molecule (B) obtained in step (2) is bound is separated from the candidate compound (A) to which the selection aid (C) is bound, and the candidate compound (A) to which the target molecule (B) is bound is recovered as a high affinity compound for the target molecule.

Description

[Technical Field]

[0001] This invention relates to a method for selecting a compound with high affinity for a target molecule. [Background technology]

[0002] In the manufacturing of antibody drugs, antibodies with a higher affinity for the target molecule, the antigen, are selected. One method for selecting compounds with high affinity for the target molecule is "off-rate selection," which involves trapping and removing compounds that dissociate slowly from the target molecule (i.e., compounds that bind strongly to the target molecule) using a separately prepared target molecule.

[0003] Conventional off-rate selection involves adding an excess amount of unlabeled target molecule to a compound bound to a labeled target molecule to create competition. The compound that remains bound to the labeled target molecule despite the addition of the excess unlabeled target molecule is selected as the compound that dissociates slowly from the target molecule (e.g., Non-Patent Document 1). However, when it is difficult to prepare large quantities of the target molecule, performing off-rate selection using such an excess amount of target molecule becomes challenging. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] J.Biol.Chem.2004 279:18870-7 [Overview of the Initiative] [Problems that the invention aims to solve]

[0005] Therefore, the object of the present invention is to provide a method for selecting a compound that binds to a target molecule and then slowly dissociates from the target molecule (i.e., strongly binds to the target molecule) without performing off-rate selection using an excess amount of the target molecule. [Means for solving the problem]

[0006] As a result of diligent research to achieve the above objectives, the inventors of this invention have discovered that by using a selection aid that has the ability to bind to an affinity tag attached to the target molecule, it is possible to easily and reliably select a compound that dissociates slowly from the target molecule from among the compounds bound to the target molecule, thereby completing the present invention.

[0007] In other words, the present invention encompasses the following: [1] A method for selecting a compound with high affinity for a target molecule, comprising the following steps. (1) A step of mixing a target molecule (B) with a candidate compound (A) that is a high affinity compound for the target molecule (B) to obtain the candidate compound (A) to which the target molecule (B) is bound, wherein the candidate compound (A) has a structural site (A1) that can bind to the target molecule (B) and a structural site (A2) that can bind to a selective auxiliary material (C), and is capable of binding to only one of the target molecule (B) and the selective auxiliary material (C), the above step (2) A step to obtain candidate compound (A) to which the target molecule (B) obtained in step (1) is bound, by mixing with a selective aid (C) to obtain candidate compound (A) to which the target molecule (B) is bound and candidate compound (A) to which the selective aid (C) is bound. (3) A step in which the candidate compound (A) to which the target molecule (B) obtained in step (2) is bound is separated from the candidate compound (A) to which the selection aid (C) is bound, and the candidate compound (A) to which the target molecule (B) is bound is recovered as a high affinity compound for the target molecule. [2] The method according to [1], wherein the target molecule (B) is a target molecule bound to the first bead. [3] The method according to [1], further comprising the step of mixing the first beads with a candidate compound (A) of a high affinity compound for the target molecule (B) before step (1), and removing the candidate compound (A) bound to the first beads. [4] The method according to [2] or [3], wherein the volume average diameter of the first bead is 10 nm or more and 1000 μm or less. [5] The method according to [1], wherein the selective auxiliary material (C) has a portion that connects to a structural portion (A2) that can be connected to the selective auxiliary material (C) and a base portion, the base portion being a second bead. [6] The method according to [5], wherein the volume average diameter of the second bead is 10 nm or more and 1000 μm or less. [7] The method according to [1], wherein the molar amount of the selection aid (C) in step (2) is 10 times or more and 10,000 times or less the molar amount of the target molecule (B). [8] In step (2), the molar amount of the selection aid (C) is 10 times or more the molar amount of the candidate compound (A) 11 The method described in [1], which is less than or equal to [1]. [9] The method according to [1], wherein step (2) is carried out in a buffer solution.

[10] The method according to [1], wherein step (2) is carried out at a temperature of 0°C or higher and 50°C or lower.

[11] The method according to [1], wherein the structural portion (A2) that can bind to the selected auxiliary material (C) has a polypeptide structure.

[12] The method according to [1], wherein the candidate compound (A) is a peptide.

[13] The method according to

[12] , wherein the number of amino acid residues of the peptide is 3 or more and 20 or less.

[14] The method according to

[12] or

[13] , wherein the peptide is a modified peptide.

[15] The method according to

[12] , wherein the peptide is in the form of a ribosome display complex comprising the peptide, a ribosome, and mRNA. [Effects of the Invention]

[0008] According to the present invention, compounds that dissociate slowly from the target molecule can be efficiently selected as high-affinity compounds for the target molecule without using an excess amount of the target molecule, as is the case with conventional off-rate selection. [Brief explanation of the drawing]

[0009] [Figure 1]FIG. 1A shows the structure of candidate compound (A) of a high-affinity compound for a target molecule to be selected in the present invention. FIG. 1B shows the procedure of a method for selecting a high-affinity compound for a target molecule by off-rate selection using a selection aid that binds to an affinity tag. [Figure 2] FIG. 2 is a schematic diagram of the structure of mRNA contained in the RNA library prepared in Production Example 1. [Figure 3] FIG. 3 is a graph showing that a change occurs in the recovery amount of the RD complex when off-rate selection using a selection aid that binds to an affinity tag is performed. [Figure 4] FIG. 4 is a schematic diagram showing the design of a sample used for NGS analysis. [Figure 5] FIG. 5 is a graph showing the recovery rate of the RD complex by off-rate selection using a selection aid that binds to an affinity tag for clone sequences with known and different binding abilities.

Mode for Carrying Out the Invention

[0010] The method for selecting a high-affinity compound for a target molecule of the present invention comprises the following steps: (1) A step of mixing a target molecule (B) and a candidate compound (A) of a high-affinity compound for the target molecule (B) to obtain the candidate compound (A) to which the target molecule (B) is bound, wherein the candidate compound (A) has a structural site (A1) capable of binding to the target molecule (B) and a structural site (A2) capable of binding to a selection aid (C), and is capable of binding to only one of the target molecule (B) and the selection aid (C). (2) A step of mixing a selection aid (C) with the candidate compound (A) to which the target molecule (B) obtained in step (1) is bound to obtain the candidate compound (A) to which the target molecule (B) is bound and the candidate compound (A) to which the selection aid (C) is bound. (3) A step of separating the candidate compound (A) to which the target molecule (B) obtained in step (2) is bound and the candidate compound (A) to which the selection aid (C) is bound, and recovering the candidate compound (A) to which the target molecule (B) is bound as a high-affinity compound for the target molecule. Includes.

[0011] As one embodiment of the method of the present invention, Figure 1B shows a process diagram when candidate compound (A) is a modified peptide presented in the form of a ribosome display complex.

[0012] The method of the present invention is a method for removing compounds that have dissociated from a target molecule from among compounds bound to a target molecule using a selective aid that can bind to a structural site in the compound. Therefore, strictly speaking, it differs from general off-rate selection which uses an excess amount of target molecule. However, since it selects compounds with higher affinity for the target molecule based on the difference in the dissociation rate (off-rate) of compounds bound to the target molecule, this method will be referred to as off-rate selection in this specification, and to distinguish it from general methods, it will be called off-rate selection using a selective aid.

[0013] The following explains each process and terminology.

[0014] Process (1) : In step (1), the target molecule (B) and a candidate compound (A) of high affinity for the target molecule (B) are mixed to obtain a candidate compound (A) to which the target molecule (B) is bound. Here, the candidate compound (A) of high affinity (hereinafter sometimes referred to as "candidate compound (A)") has a structural site (A1) that can bind to the target molecule (B) and a structural site (A2) that can bind to the selective auxiliary material (C), and can bind to only one of the target molecule (B) and the selective auxiliary material (C) (Figure 1A).

[0015] The aforementioned bond is not limited as long as the candidate compound (A) can bond to the target molecule (B) or the selected auxiliary material (C), and may be a chemical bond or a physical bond. Examples of chemical bonds include covalent bonds, ionic bonds, and metallic bonds. Examples of physical bonds include hydrogen bonds, polar attractive forces, and bonds due to intermolecular forces such as van der Waals forces. The bond may also be adsorption.

[0016] In a state where only one of the target molecule (B) and the selective auxiliary material (C) can bind, it is preferable that the target molecule (B) and the selective auxiliary material (C) compete and one of them binds to the candidate compound (A), and more preferable that the target molecule (B) and the selective auxiliary material (C) sterically compete and one of them binds to the candidate compound (A).

[0017] As for the mixing method, for example, a method of mixing candidate compound (A) and target molecule (B) and stirring by inversion is preferred. Furthermore, it is preferable that the mixing be carried out in a buffer solution.

[0018] The mixing temperature can be set appropriately depending on the type of candidate compound (A) and target molecule (B), and is not limited to any particular temperature. However, the upper limit is preferably 50°C or lower, more preferably 40°C or lower, even more preferably 20°C or lower, particularly preferably 10°C or lower, and most preferably 5°C or lower. The lower limit is preferably 0°C or higher, and more preferably 4°C or higher.

[0019] The pH of the buffer solution can be set appropriately depending on the type of candidate compound (A) and target molecule (B), and is not limited, but its upper limit is preferably pH 12 or lower, more preferably pH 10 or lower, and even more preferably pH 8 or lower. Its lower limit is preferably pH 2 or higher, more preferably pH 4 or higher, and even more preferably pH 6 or higher.

[0020] <Candidate compound (A)> In the present invention, "candidate compound (A)" varies depending on the type of target molecule (B) and is not particularly limited as long as it is a compound that binds to the target molecule (B). Examples include peptides, polypeptides, low molecular weight compounds, medium molecular weight compounds, and high molecular weight compounds, and may also be chimeric molecules of these, but peptides or polypeptides are preferred.

[0021] There are no particular restrictions on the lower limit of the number of amino acid residues in the peptide or polypeptide, and it can be appropriately selected depending on the purpose, but it is preferably 2 or more, more preferably 3 or more, even more preferably 5 or more, even more preferably 10 or more, especially preferably 15 or more, and most preferably 20 or more. Similarly, there are no particular restrictions on the upper limit of the number of amino acid residues in the peptide or polypeptide, and it can be appropriately selected depending on the purpose, but it is preferably 1000 or less, more preferably 500 or less, and even more preferably 200 or less.

[0022] The peptide may be a peptide-containing substance, a peptide-containing substance containing one type of peptide, or a peptide-containing substance containing two or more types of peptides. Among these, a peptide-containing substance containing two or more peptides is preferred, a peptide library is more preferred, and a ribosome display peptide library, which is a peptide library of ribosome display complexes (RD complexes), is even more preferred.

[0023] Furthermore, the peptide may be a modified peptide obtained by reacting a peptide with a modifying agent.

[0024] There are no particular restrictions on the modifier, and it can be appropriately selected depending on the purpose, but it is preferable that it does not bind to the target molecule (B). Furthermore, it is preferable that the modifier has a leaving group.

[0025] The modifying agent preferably has both a cell membrane permeability-conferring group and a cyclizing group for cyclizing the peptide, or at least a cyclizing group for cyclizing the peptide. Furthermore, the modifying agent may also have a linking group.

[0026] The cell membrane permeability-conferring group is a group that confers cell membrane permeability to the peptide, and it is preferable that it has a basic functional group.

[0027] There are no particular restrictions on the form of the cell membrane permeability-conducting group, and it can be appropriately selected depending on the purpose, but a form having a dendritic structure is preferred. There are no particular restrictions on the number of dendritic structural units in the cell membrane permeability-conducting group, and it can be appropriately selected depending on the purpose, and it may be one or two or more. There are no particular restrictions on the number of branches per dendritic structural unit, and it can be appropriately selected depending on the purpose, but three or more is preferred.

[0028] There are no particular restrictions on the branch structure of a dendritic structure; it can be appropriately selected according to the purpose.

[0029] There are no particular restrictions on the basic functional group; for example, guanidino groups, amino groups, and imidazole groups are examples, but guanidino groups are preferred. The basic functional group may be used alone or in combination of two or more.

[0030] There are no particular restrictions on the number of basic functional groups, and they can be appropriately selected depending on the purpose, but two or more are preferred. The basic functional group is preferably one having two or more guanidino groups.

[0031] There are no particular restrictions on the position of the basic functional group in the linker molecule, and it can be appropriately selected depending on the purpose, but it is preferable that it be located at the end of a branch of the dendritic structure.

[0032] Specific examples of cell membrane permeability-conferring groups having a dendritic structure include, for example, those represented by the following general formula (I).

[0033] [ka]

[0034] The general formula (I) above represents a cell membrane permeability-conducting group having a dendritic structure with 3 branches, where "Y" represents a basic functional group. The basic functional group may be formed at the end of all branches or at the end of some branches. Furthermore, the cell membrane permeability-conducting group may have multiple dendritic structures represented by general formula (I), in which case the number of branches may be, for example, 6 or 9.

[0035] Cyclizing groups contribute to the cyclization of peptides by reacting with reactive amino acid residues in the peptide. Reactive amino acid residues are amino acid residues that react with the cyclizing group, and may be amino acid residues that react directly with the cyclizing group, or amino acid residues that have been modified to react with the cyclizing group. There are no particular restrictions on the cyclizing group, and it can be appropriately selected depending on the purpose, but electron-withdrawing groups are preferred.

[0036] There are no particular restrictions on the electron-withdrawing group, and it can be appropriately selected depending on the purpose, but it is preferable that it contains a halogen.

[0037] There are no particular restrictions on the type of halogen; it can be selected appropriately according to the purpose. There are no particular restrictions on the number of halogens; it can be selected appropriately according to the purpose, but two or more are preferred.

[0038] The electron-withdrawing group preferably has two or more chlorine atoms, more preferably a benzyl chloride which may have substituents, and particularly preferably a 3,5-bis(chloromethyl)benzyl group.

[0039] Cyclization of peptides using cycloforming groups is preferably carried out by a reaction between the cycloforming group and at least one group selected from the group consisting of thiol groups, amino groups, and hydroxyl groups contained in the peptide.

[0040] The linking group is a group that connects the cell membrane permeability-conferring group with the cyclizing group, and its structure is not particularly limited and can be appropriately selected depending on the purpose.

[0041] Specific examples of modifiers include 1,3-dibromo-2-propanone, 1,3,5-tris(bromomethyl)benzene, and compounds represented by the following structural formula. The modifier represented by the following structural formula has a 3,5-bis(chloromethyl)benzyl group as a cyclizing group and three guanidino groups at the end of the dendritic structure as a cell membrane permeability-conferring group, and the cyclizing group and the cell membrane permeability-conferring group are linked via a linking group.

[0042] The modifier represented by the following structural formula can be obtained by the method described in International Publication No. 2020 / 195302.

[0043] [ka]

[0044] <Target molecule (B)> In the present invention, "target molecule (B)" refers to a specific molecule for which a compound with high affinity is selected for the purpose of controlling its function. Specific examples of target molecules (B) include, for example, biomolecules such as antigens, receptors, enzymes, ion channels, and transporters.

[0045] The target molecule (B) is preferably a target molecule bound to a substrate (support). By having the target molecule (B) bound to a substrate (support), the target molecule (B) and the selective auxiliary material (C) can easily compete sterically, resulting in one of them binding to the candidate compound (A).

[0046] The aforementioned binding method is not particularly limited as long as it allows the target molecule (B) and the substrate (support) to bind, and can be appropriately selected depending on the purpose, but streptavidin-biotin binding is preferred. Specifically, the target molecule (B) can be biotin-labeled, streptavidin can be chemically bonded to the substrate (support), and the target molecule (B) can be bound to the substrate (support) by streptavidin-biotin binding.

[0047] The substrate (support) to which the target molecule (B) is bound (hereinafter referred to as the "first bead") is not particularly limited as long as it creates steric hindrance with the second bead described below, but magnetic beads are preferred.

[0048] The lower limit of the volume-average diameter of the first beads is preferably 10 nm or more, more preferably 100 nm or more, even more preferably 200 nm or more, even more preferably 300 nm or more, particularly preferably 500 nm or more, and most preferably 800 nm or more. The upper limit of the volume-average diameter of the first beads is preferably 1000 μm or less, more preferably 100 μm or less, even more preferably 10 μm or less, even more preferably 5 μm or less, particularly preferably 2 μm or less, and most preferably 1200 nm or less. The volume-average diameter can be measured using Partica LA-960 (HORIBA).

[0049] <Selection of auxiliary materials (C)> The selected auxiliary material (C) has at least a portion that binds to the structural portion (A2) of the candidate compound (A), and may also have a substrate (support) portion.

[0050] Because the selective auxiliary material (C) has a substrate (support) portion, the target molecule (B) and the selective auxiliary material (C) can easily compete sterically, resulting in a state where one of them binds to the candidate compound (A).

[0051] The portion of candidate compound (A) that binds to structural site (A2) and the substrate (support) portion can be joined by a chemical bond. Examples of the portion that binds to structural site (A2) of candidate compound (A) include an antibody that recognizes structural site (A2) of candidate compound (A).

[0052] The substrate (support) portion of the selected auxiliary material (C) (referred to as the "second bead" in this specification) is not particularly limited as long as it causes steric hindrance with the first bead, but agarose gel beads are preferred.

[0053] The lower limit of the volume-average diameter of the second bead is preferably 10 nm or more, more preferably 100 nm or more, even more preferably 1000 nm or more, even more preferably 10 μm or more, particularly preferably 20 μm or more, and most preferably 40 μm or more. The upper limit of the volume-average diameter of the second bead is preferably 1000 μm or less, more preferably 800 μm or less, even more preferably 500 μm or less, even more preferably 300 μm or less, particularly preferably 200 μm or less, and most preferably 180 μm or less. The volume-average diameter can be measured using Partica LA-960 (HORIBA).

[0054] Candidate compound (A) has a structural site (A1) that can bind to the target molecule (B) and a structural site (A2) that can bind to the selected auxiliary material (C), and also has other structural sites.

[0055] <Structural site (A1) capable of binding to target molecule (B)> In the present invention, the "structural site (A1) capable of binding to the target molecule (B)" is preferably a structural site (A1) that does not bind to the selection aid (C) but is capable of binding to the target molecule (B). The structural site (A1) capable of binding to the target molecule (B) is preferably a (poly)peptide structure. The structural site (A1) capable of binding to the target molecule (B) is preferably one that contains random sequences at specific positions so as to be useful as a (poly)peptide library. From such random sequences, useful amino acid sequences can be identified according to a predetermined purpose.

[0056] The lower limit of the number of amino acid residues in the structural site (A1) capable of binding to the target molecule (B) is preferably 2 or more, more preferably 3 or more, even more preferably 5 or more, and particularly preferably 10 or more. The upper limit of the number of amino acid residues in the structural site (A1) capable of binding to the target molecule (B) is preferably 1000 or less, more preferably 500 or less, even more preferably 200 or less, even more preferably 100 or less, particularly preferably 50 or less, and most preferably 20 or less.

[0057] The RNA encoding the structural site (A1) capable of binding to the target molecule (B) is preferably one that does not have a stop codon.

[0058] The structural site (A1) capable of binding to the target molecule (B) preferably contains a reactive amino acid residue, and more preferably contains an amino acid residue having a nucleophilic side chain. The modified peptide can be obtained by reacting the reactive amino acid residue or the amino acid residue having a nucleophilic side chain with a modifying agent having a leaving group.

[0059] There are no particular restrictions on the structure of the structural site (A1) that can bind to the target molecule (B), and it can be appropriately selected depending on the purpose, but a cyclic structure is preferred. In particular, forming a cyclic structure through modification restricts the conformation of A1. This is expected to improve affinity with the target molecule and chemical and biological stability.

[0060] Examples of reactive amino acid residues include cysteine ​​residues, lysine residues, serine residues, and threonine residues. Reactive amino acid residues may be used individually or in combination of two or more.

[0061] The number of reactive amino acid residues in a peptide is preferably two or more, in terms of cyclizing the peptide. The upper limit for the number of reactive amino acid residues in a peptide is preferably 10 or less, because if there are many reaction sites, the number and position of the linker molecules that bind to the peptide may become unstable, making it difficult to compare the properties of the peptide derived from the amino acid sequence.

[0062] Furthermore, if, for example, cysteine ​​residues in the peptide are involved in stabilizing the peptide's higher-order structure via disulfide bonds, it is preferable to introduce reactive amino acid residues into the peptide separately.

[0063] There are no particular restrictions on the position of reactive amino acid residues in a peptide; they can be appropriately selected depending on the purpose.

[0064] For example, when using a ribosome display complex (hereinafter sometimes referred to as the "RD complex") containing an mRNA molecule, its translated peptide chain (hereinafter sometimes referred to as the "polypeptide chain"), and a ribosome as the peptide (or polypeptide), it is preferable to use the portion that exits the ribosome's exit tunnel, specifically between the 2nd position from the N-terminus and the 30th position from the C-terminus (including the 2nd position from the N-terminus and the 30th position from the C-terminus), in that the modification reaction by the modifying agent may be less likely to be sterically inhibited by the ribosome.

[0065] The position of the reactive amino acid residue from the C-terminus is preferably the 50th position from the C-terminus, and more preferably the 100th position. Furthermore, when counting the positions of the reactive amino acid residues from the N-terminus, the position can be appropriately set according to the peptide chain length, but for example, it is between the 2nd and 1000th positions from the N-terminus, preferably between the 2nd and 100th positions from the N-terminus, and more preferably between the 2nd and 50th positions from the N-terminus.

[0066] There are no particular restrictions on the method for manufacturing the RD complex; known methods can be appropriately selected, such as the method described in International Publication No. 2017 / 213158. It can also be manufactured using commercially available kits.

[0067] There are no particular restrictions on the position of the random sequence in the peptide; it can be appropriately selected depending on the purpose. For example, similar to the position of reactive amino acid residues, when using an RD complex, it is preferable to select a position between the 2nd position from the N-terminus and the 30th position from the C-terminus (including the 2nd position from the N-terminus and the 30th position from the C-terminus). In other words, it is preferable that the reactive amino acid residues are included within the random sequence. Therefore, the preferred position of the random sequence can be set from the same range as the preferred position of the reactive amino acid residues.

[0068] The number of random sequences in a peptide may be one or two or more. There is no particular upper limit to the number of random sequences, and it can be appropriately selected depending on the purpose, but it is preferable to have 10 or fewer.

[0069] There are no particular restrictions on the number of amino acid residues per random sequence; it can be appropriately selected depending on the purpose, for example, it can be between 1 and 30.

[0070] The longer a single random sequence is, and the greater the number of random sequences, the greater the diversity of the peptide library.

[0071] <Selected auxiliary material (C) and structural part (A2) that can be bonded> In the present invention, the "structural site (A2) that can bind to the selective aid (C)" is preferably a structural site (A2) that does not bind to the target molecule (B) but can bind to the selective aid (C). The structural site (A2) that can bind to the selective aid (C) is preferably a (poly)peptide structure. The structural site (A2) that can bind to the selective aid (C) may also be a (poly)polypeptide structure.

[0072] The lower limit of the number of amino acid residues in the structural site (A2) that can bind to the selective aid (C) is preferably 2 or more, more preferably 3 or more, even more preferably 5 or more, and particularly preferably 8 or more. The upper limit of the number of amino acid residues in the structural site (A2) that can bind to the selective aid (C) is preferably 1000 or less, more preferably 500 or less, even more preferably 200 or less, even more preferably 100 or less, particularly preferably 50 or less, and most preferably 20 or less.

[0073] There are no particular restrictions on specific structural sites (A2) that can bind to the selective auxiliary material (C), and they can be appropriately selected depending on the purpose. Examples include peptide tags, polypeptide tags, luminescent enzymes, and fluorescent proteins.

[0074] Examples of the peptide tag include the FLAG® tag, HA tag, and His tag. Among these, the FLAG® tag is preferred. Examples of the polypeptide tag include GST (glutathione S-transferase), MBP (maltose-binding protein), and thioredoxin. Among these, GST (glutathione S-transferase) is preferred. Examples of the luminescent enzyme include luciferase. Examples of the fluorescent protein include GFP and RFP.

[0075] There are no particular restrictions on the position of the structural site (A2) in candidate compound (A) that can bind to the selective aid (C), and it can be appropriately selected depending on the purpose. However, it is preferable that candidate compound (A) has the structural site (A2) that can bind to the selective aid (C) at its terminal end, and it is even more preferable that candidate compound (A) has the structural site (A2) that can bind to the selective aid (C) at its N-terminus.

[0076] Regarding the positions of the structural site (A1) capable of binding to the target molecule (B) and the structural site (A2) capable of binding to the selective auxiliary material (C), it is preferable that the structural site (A1) capable of binding to the target molecule (B) is located at the C-terminal end of the structural site (A2) capable of binding to the selective auxiliary material (C).

[0077] There are no particular restrictions on the lower limit of the number of amino acid residues between the structural site (A1) that can bind to the target molecule (B) and the structural site (A2) that can bind to the selective aid (C). However, 1 or more is preferred. Similarly, there are no particular restrictions on the upper limit, and it can be appropriately selected depending on the purpose. However, from the viewpoint of sterically competing between the target molecule (B) and the selective aid (C), 20 or less is preferred, 15 or less is more preferred, 10 or less is even more preferred, 5 or less is even more preferred, 3 or less is particularly preferred, and 2 or less is most preferred.

[0078] There are no particular restrictions on the lower limit of the distance between the structural site (A1) that can bind to the target molecule (B) and the structural site (A2) that can bind to the selective aid (C) in the three-dimensional structure of the peptide, and it can be appropriately selected depending on the purpose, but it is preferably 1 nm or more. Similarly, there are no particular restrictions on the upper limit, and it can be appropriately selected depending on the purpose, but from the point of sterically competing between the target molecule (B) and the selective aid, it is preferably 100 μm or less, more preferably 10 μm or less, even more preferably 1 μm or less, even more preferably 500 nm or less, especially preferably 100 nm or less, and most preferably 10 nm or less.

[0079] Other structures are not particularly limited and can be appropriately selected depending on the purpose, for example, a protease recognition structure site and a linker structure site. The protease recognition structure site is not particularly limited and can be appropriately selected depending on the purpose, for example, a TEV protease recognition structure site.

[0080] Process (2) : Step (2) is a step in which a selection aid (C) is mixed with candidate compound (A) to which the target molecule (B) obtained in step (1) is bound, thereby obtaining candidate compound (A) to which the target molecule (B) is bound and candidate compound (A) to which the selection aid (C) is bound.

[0081] If the candidate compound (A) to which the target molecule (B) is bound has a strong binding affinity to the target molecule (B), it will still be bound to the target molecule (B) even after mixing in the selective additive (C). However, if the binding affinity to the target molecule (B) is weak, mixing in the selective additive (C) will allow the compound to dissociate from the target molecule (B) and then bind to the selective additive (C).

[0082] There are no particular restrictions on the mixing method, and it can be appropriately selected depending on the purpose, but a method of mixing the candidate compound (A) to which the target molecule (B) is bound with the selection aid (C) and stirring by inversion is preferred. The mixing is preferably carried out in a buffer solution.

[0083] There are no particular restrictions on the upper limit of the mixing temperature, and it can be appropriately selected depending on the purpose, but it is preferably 50°C or lower, more preferably 40°C or lower, even more preferably 20°C or lower, particularly preferably 10°C or lower, and most preferably 5°C or lower. There are no particular restrictions on the lower limit of the mixing temperature, and it can be appropriately selected depending on the purpose, but it is preferably 0°C or higher, and more preferably 4°C or higher.

[0084] There are no particular restrictions on the upper limit of the pH of the buffer solution, and it can be appropriately selected depending on the purpose, but a pH of 12 or less is preferred, a pH of 10 or less is more preferred, and a pH of 8 or less is even more preferred.

[0085] There are no particular restrictions on the lower limit of the pH of the buffer solution, and it can be appropriately selected depending on the purpose, but a pH of 2 or higher is preferred, a pH of 4 or higher is more preferred, and a pH of 6 or higher is even more preferred.

[0086] The lower limit of the molar amount of the selective aid (C) relative to the molar amount of the target molecule (B) is preferably 10 times or more, more preferably 100 times or more, even more preferably 1000 times or more, and particularly preferably 5000 times or more. The upper limit of the molar amount of the selective aid (C) relative to the molar amount of the target molecule (B) is preferably 10000 times or less.

[0087] The lower limit of the molar amount of the selected auxiliary material (C) relative to the molar amount of the candidate compound (A) is preferably 10 times or more, more preferably 1000 times or more, and 10 5 More than twice as much is even more preferable, 10 7 More than twice as much is even more preferable, 10 9 More than double is particularly preferable, 10 10 More than double is most preferable. Furthermore, the upper limit of the molar amount of the selected auxiliary material (C) relative to the molar amount of the candidate compound (A) is 10 11 It is preferable that it be less than double.

[0088] The molar amount of the selection aid (C) is calculated based on the bead binding capacity (catalog value), the molar amount of the target molecule (B) is calculated assuming that the bead binding capacity (catalog value) has reached saturation, and the molar amount of the candidate compound (A) is calculated by quantifying the amount of RNA paired with candidate compound (A) using qPCR (QuantStudio3, ThermoFisher Scientific).

[0089] Process (3) : Step (3) involves separating the candidate compound (A) to which the target molecule (B) obtained in step (2) is bound from the candidate compound (A) to which the selection aid (C) is bound, and recovering the candidate compound (A) to which the target molecule (B) is bound as a high-affinity compound for the target molecule.

[0090] In step (2), the candidate compound (A) to which the target molecule (B) is bound after mixing with the selective aid (C) dissociates slowly from the target molecule (B), i.e., it is a compound that binds strongly to the target molecule (B), while the candidate compound (A) to which the selective aid (C) is bound dissociates quickly from the target molecule (B), i.e., it is a compound that binds weakly to the target molecule (B). Therefore, by separating the candidate compound (A) to which the target molecule (B) is bound from the candidate compound (A) to which the selective aid (C) is bound, the candidate compound (A) to which the target molecule (B) is bound can be selected and recovered as a compound with higher affinity for the target molecule.

[0091] The method for separating the candidate compound (A) to which the target molecule (B) obtained in step (2) is bound from the candidate compound (A) to which the selective aid (C) is bound is not particularly limited, but for example, it can be separated by the difference in magnetism between the first bead bound to the target molecule (B) and the second bead of the substrate of the selective aid (C).

[0092] As for other steps, there are no particular limitations, and they can be appropriately selected according to the purpose. Examples include steps of introducing mutations into high-affinity compounds for the target molecules recovered through the above steps. Methods for introducing mutations include site-specific mutagenesis using mixed bases and methods for introducing mutations into random locations by error-prone PCR, etc.

Examples

[0093] Hereinafter, examples of the present invention will be described, but the present invention is not limited to these examples in any way.

[0094] (Production Example 1) Preparation of RNA In the same manner as the method described in International Publication No. 2017 / 213158, using the NNK method, an RNA having a sequence containing the sequence of (NNK) 10 (where N represents A, U, G or C, K represents G or U, and NNK corresponds to all codons) and containing 10 12 or more RNAs (a molecular number of 10 12 or more composed of about one each of 10 12 or more types of RNAs) was prepared.

[0095] First, an RNA library (base sequence: SEQ ID NO: 2) was prepared in order to construct a polypeptide library (amino acid sequence: SEQ ID NO: 1) capable of binding to each of a target molecule and a selection aid having the structure shown in FIG. 2. The RNA library was prepared by performing a transcription reaction using a DNA library (base sequence: SEQ ID NO: 3) as a template, and the DNA library was prepared by performing a PCR reaction using plasmid DNA (base sequence: SEQ ID NO: 4) as a template. The plasmid DNA was constructed by homologous recombination (using In Fusion HD Cloning Kit (Takara)) using pUC19 (Takara) and the DNA library (DNA shown in SEQ ID NO: 3).

[0096] Specifically, a reaction solution having the composition shown in Table 1 was used, and PCR was performed using the plasmid DNA as template DNA in the PCR cycle shown in Table 2 to prepare the 5' fragment. KOD FX Neo (TOYOBO) was used as the PCR enzyme. The 2× buffer and 2 mM dNTPs were those provided with the KOD FX Neo (TOYOBO).

[0097] In Table 1, 5FFnew_150409 is a forward primer (SEQ ID NO: 5), and Ma5FragR_withoutHis_150409 is a reverse primer (SEQ ID NO: 6). Both primers were purchased from Eurofin Genomics. The same applies to the following primers.

[0098] [Table 1]

[0099] [Table 2]

[0100] Next, the 3' fragment of the template DNA was prepared using a reaction mixture having the composition shown in Table 3 and the PCR cycle shown in Table 4. KOD FX Neo (TOYOBO) was used as the PCR enzyme. The 2× buffer and 2 mM dNTPs were those provided with the KOD FX Neo (TOYOBO).

[0101] In Table 3, A1_Cys5AaCys5AaCys_210301 is a forward primer (SEQ ID NO: 7), and 3F-Rnew_150409 is a reverse primer (SEQ ID NO: 8). The sequence of A1_Cys5AaCys5AaCys_210301 contains random nucleotide sequences, with a diversity of up to 20. 10 Street (approximately 1 x 10 13 (This is the result.) In this reaction solution, 4 × 10 13 Because this uses multiple primers, each sequence will contain approximately four of them.

[0102] [Table 3]

[0103] [Table 4]

[0104] Next, using a reaction mixture having the composition shown in Table 5, overlapping PCR was performed using the PCR cycle shown in Table 6 to ligate the 5' and 3' fragments. KOD FX Neo (TOYOBO) was used as the PCR enzyme. The 2× buffer and 2 mM dNTPs were those provided with the KOD FX Neo (TOYOBO).

[0105] In Table 5, X to Z are 1 × 10 13 This shows that the 5' and 3' fragments were used, and H2O was added to the reaction mixture to adjust the total volume to 100 μL. Specifically, 1 × 10 12 10 μL of 5' fragments / μL, 1 × 10 12 10 μL of 3' fragments at a concentration of 3' particles / μL and 8 μL of H2O were used.

[0106] [Table 5]

[0107] [Table 6]

[0108] Furthermore, using the obtained DNA strand (overlapping PCR product) as a template, PCR was performed using a reaction mixture with the composition shown in Table 7 and the PCR cycle shown in Table 8 to amplify the full-length DNA and obtain template DNA. KOD FX Neo (TOYOBO) was used as the PCR enzyme. The 2× buffer and 2 mM dNTPs were those provided with the KOD FX Neo (TOYOBO).

[0109] In Table 7, 5FFnew_150409 is a forward primer (SEQ ID NO: 5), and 3F-Rnew_150409 is a reverse primer (SEQ ID NO: 8). In Table 7, X to Y are 1 × 10⁻⁶. 12 This shows that 5 × 10 overlapping PCR products were used, and the total volume was adjusted to 200 μL by adding H2O to the reaction mixture. Specifically, 5 × 10 10 20 μL of overlapping PCR product with a concentration of cells / μL and 24 μL of H2O were used.

[0110] [Table 7]

[0111] [Table 8]

[0112] Using the obtained DNA as a template, and reacting it with a reaction solution having the composition shown in Table 9 at 37°C for 2 hours, a 1×10¹⁶ DNA having the base sequence of Sequence ID No. 3 is obtained. 12 An RNA library containing the above mRNAs was obtained.

[0113] The transcriptase used was T7 RNA polymerase (TAKARA). The 10× buffer, 50 mM DTT, was the one provided with the T7 RNA polymerase (TAKARA).

[0114] In Table 9, X to Y are 1 × 10 12This indicates that more than 1 × 10¹⁶ template DNA molecules were used, and H₂O was added to the reaction mixture to adjust the total volume to 100 μL. Specifically, 1 × 10¹⁶ 12 10 μL of template DNA (with a particle / μL concentration) and 45 μL of H2O were used.

[0115] As shown in Figure 2, the mRNA contained in this library has, in order from the 5' end, a FLAG® sequence, a random sequence, a TEV protease recognition sequence, and a spacer sequence (linker sequence), and does not contain a stop codon.

[0116] [Table 9]

[0117] (Manufacturing Example 2) Fabrication of Ribosome Display Complex Ribosome display (RD) complexes were prepared from the RNA library created in Production Example 1 using a reconstituted cell-free protein synthesis kit (Gene Frontier "PURE frexRD®").

[0118] The above RD complex reaction solution was mixed with agarose gel beads, Anti-FLAG M2 Affinity Gel (Merck), and stirred at 4°C for 1 hour. Through this process, agarose gel beads to which the RD complex containing the FLAG sequence was selectively bound were recovered. Subsequently, the RD complex was dissociated from the agarose gel beads by adding FLAG peptide (Merck).

[0119] (Production Example 3) Preparation of modified peptides using cross-linking reaction with modifying agents In the suspension, the RD complex recovered in Production Example 2 was selectively bound to agarose gel beads. Tris(2-carboxyethyl)phosphine hydrochloride (Nacalai tesque) was added to a final concentration of 0.5 mM, and 1,3,5-tris(bromomethyl)benzene (Tokyo Chemical Industries) was added as a modifying agent to a final concentration of 2 mM. The modification reaction was carried out by inversion and stirring at 4°C overnight. After washing off the modifying agent, FLAG peptide (Merck) was added to dissociate the RD complex containing the modified peptide from the agarose gel beads.

[0120] (Example 1-1) Selection of RD complexes that bind to antigens (1) Preparation of beads (antigen-immobilized magnetic beads) on which the antigen (corresponding to the target molecule (B) above) is immobilized. 50 μg of IL-6 (R&D Systems) was mixed with 20 μg of NHS-PEG4-Biotin (ThermoFisher Scientific), and the mixture was reacted on ice for 2 hours. Unreacted reagent was then removed by dialysis to obtain the biotinylated antigen.

[0121] To Nanolink Sterptavidin Magnetic Beads (Solulink, volume average diameter 1.0 μm), five times the binding capacity of the magnetic beads, the biotinylated antigen (35 ng) was added, and the beads were washed by inversion and agitation at 4°C for 1 hour to obtain beads to which the antigen had bound and immobilized (antigen-immobilized magnetic beads).

[0122] (2) Recovery of RD complex bound to antigen The RD complex containing the modified peptide obtained in Production Example 3 was mixed with the antigen-immobilized magnetic beads and reacted with the antigen by inversion and stirring at 4°C for 1 hour. Subsequently, the mixture was washed with a buffer containing 15 mM magnesium ions to remove the RD complex that was not bound to the antigen, and only the RD complex bound to the antigen was recovered.

[0123] (Examples 1-2) Off-rate selection using selection aids bound to affinity tags To the suspension of magnetic beads bound to the RD composite obtained in Example 1-1, an agarose gel bead, Anti-FLAG M2 Affinity Gel (Merck volume average diameter 45-165 μm) (corresponding to the selected auxiliary material (C) above), was added and the suspension was stirred by inversion overnight at 4°C.

[0124] As a result, RD complexes that have not dissociated from the antigen remain on the magnetic beads, and because the magnetic beads and agarose gel beads repel each other sterically, the agarose gel beads cannot approach the affinity tag (corresponding to the structural site (A2)), and therefore the RD complexes remain on the magnetic beads. On the other hand, RD complexes that have dissociated from the antigen have their affinity tags exposed and are easily captured on agarose gel beads that are present in excess in the solution.

[0125] To separate and recover RD complexes that have not dissociated from the antigen from RD complexes that have dissociated from the antigen, magnetic beads and agarose gel beads were separated using a magnetic separation stand. After washing the magnetic beads containing RD complexes that have not dissociated from the antigen, the RD complexes were destabilized with 50 mM EDTA, and the RNA from the RD complexes was recovered. Since EDTA affects subsequent procedures, this RNA was purified using an RNA enrichment and purification kit (QIAGEN's "RNeasy MinElute Cleanup Kit"). 1 μL of the purified RNA was taken, and the amount of RNA was quantified by Real-time PCR to determine the amount of recovered complexes. The results are shown in Figure 3.

[0126] In Figure 3, the black bars represent the total amount of RD complexes before the selection experiment, the shaded bars represent the total amount of RD complexes bound to the antigen (total amount after step (1) and before step (2)), and the dotted bars represent the total amount of complexes recovered from the antigen after off-rate selection using a selection aid that binds to the affinity tag. As shown in Figure 3, there is a difference in the amount of RD complexes recovered before and after the off-rate selection process, and it was confirmed that there are complexes that are recovered from the magnetic beads after off-rate selection. From this, it can be inferred that not all RD complexes in the solution are captured by the agarose gel beads, and that RD complexes that have dissociated from the antigen are captured on the agarose gel beads, while complexes that have not dissociated from the antigen remain on the magnetic beads.

[0127] (Example 1) Analysis of enriched sequences (1) Sample preparation for NGS analysis NGS analysis was performed using Miseq (Illumina). The recovered RNA was used as a template and reacted with primers (SEQ ID NO: 9) at the composition shown in Table 10 at 65°C for 5 minutes. The reaction mixture (reaction product) was then reverse transcribed by reacting it at the composition shown in Table 11 at 50°C for 1 hour, and then at 70°C for 15 minutes.

[0128] The 2mM dNTPs used were those included with the KOD FX Neo (TOYOBO) mentioned above. The reverse transcriptase used was Superscript III Reverse Transcriptase (ThermoFisher Scientific). The 5× buffer and 0.1M DTT were those provided with the Superscript III Reverse Transcriptase (ThermoFisher Scientific).

[0129] In Table 10, X to Y indicate that an appropriate amount of template RNA was used, and H2O was added to the reaction mixture to adjust the total volume to 13 μL. Specifically, 1 × 10 6We used 1 μL of template RNA with a concentration of cells / μL and 6.8 μL of H2O.

[0130] [Table 10]

[0131] [Table 11]

[0132] Furthermore, using the cDNA obtained from this reaction as a template, tailed PCR was performed using a reaction solution with the composition shown in Table 12 and the PCR cycle shown in Table 13 to produce fragments of the design shown in Figure 4, to which the sequences necessary for analysis were added.

[0133] After analysis, in order to distinguish the source sample, samples were prepared using reverse primers with different index sequences for each sample: truseq1R(709) (SEQ ID NO: 10) for samples recovered from magnetic beads with antigen, truseq1R(710) (SEQ ID NO: 11) for samples recovered from magnetic beads without antigen, and truseq1R(711) (SEQ ID NO: 12) for samples recovered from magnetic beads after off-rate selection. Note that in truseq1R(709) (SEQ ID NO: 10), truseq1R(710) (SEQ ID NO: 11), and truseq1R(711) (SEQ ID NO: 12), the index sequence is from the 25th to the 30th base.

[0134] The PCR enzyme used was 2×KAPA HiFi HS Ready Mix (Roche). In Table 12, the forward primer is the sequence of SEQ ID NO: 13, and the reverse primer is the sequence of SEQ ID NO: 14. In Table 12, X to Y indicate that an appropriate amount of cDNA was used, and H2O was added to the reaction mixture to adjust the total volume to 20 μL. Specifically, 1 × 10 5 We used 1 μL of cDNA with a concentration of cells / μL and 7.8 μL of H2O.

[0135] [Table 12]

[0136] [Table 13]

[0137] After denaturation with 0.2N NaOH, the sample solution for analysis was prepared by diluting it with HT1 buffer (Illumina) while cooling it on ice, and then mixing it with equimolar PhiX Control (Illumina).

[0138] (2)NGS analysis The sequences were read using the paired-end method with the samples prepared using the above procedure. As shown in Table 14, the sequence analysis revealed differences in the sequences before and after off-rate selection.

[0139] In Table 14, the sequences in the "Antigen Present" column (left) are samples before off-rate selection with the antigen bound, the sequences in the "Off-rate" column (right) are samples after off-rate selection, and the sequences in the "No Antigen" column (center) are samples before off-rate selection without the addition of the antigen (background). From top to bottom, the sequences shown are the top 20 sequences in terms of frequency of occurrence based on the sequence analysis results. The darkest gray shading (sequences ranked 1-5, 8-9, and 13 in the "Antigen Present" column) indicates sequences included in the top 1-20 of the "No Antigen" column, the second darkest gray shading (sequences ranked 19 and 20 in the "Antigen Present" column, and sequence ranked 14 in the "Off-rate" column) indicates sequences included in the top 21-40 of the "No Antigen" column, and the third darkest gray shading (sequences ranked 11-12 and 15 in the "Antigen Present" column, and sequence ranked 5 in the "Off-rate" column) indicates sequences included in the top 41-60 of the "No Antigen" column.

[0140] [Table 14]

[0141] If all complexes were captured by the agarose gel beads, this difference should not be observed. This result suggests that complexes that have dissociated from the antigen are captured on the agarose gel beads, while complexes that have not dissociated from the antigen remain on the magnetic beads.

[0142] (Test Example 2) Verification of a system using a sequence with known binding ability To verify whether off-rate selection using a selection aid that binds to affinity tags is functioning as intended, we had ribosomes present sequences whose binding ability had already been measured, and then verified whether the results matched the binding ability.

[0143] (1) Preparation of verification samples The validation samples were prepared in the same manner as described in Manufacturing Example 1. By changing A1_Cys5AaCys5AaCys_210301 listed in Table 3 to "3FF_RD60-39_220107" (SEQ ID NO: 66) or "3FF_PN-2729L_220107" (SEQ ID NO: 67), a known sequence (Sequence 1 or Sequence 2) was presented to the ribosome.

[0144] (2) Verification using RD complexes with known sequences An RD complex was prepared using the RNA prepared in (1) in the same manner as in Production Example 2. Off-rate selection was performed using this complex with a selection aid that binds to the affinity tag, and the amount of complex recovered was quantified by Real-time PCR.

[0145] The results of real-time PCR (recovery rate of RD complexes after off-rate selection) are shown in Figure 5. The recovery rate of RD complexes after off-rate selection was calculated as [amount of RD complexes recovered after off-rate selection] / [amount of RD complexes recovered before off-rate selection]. When comparing RD complexes that presented a known sequence with weak binding affinity (Sequence 1: RD60-39_220107) and a known sequence with strong binding affinity (Sequence 2: PN-2729L_220107), Sequence 2 showed a higher recovery rate than Sequence 1 after off-rate selection. This confirms that off-rate selection using a selection aid that binds to the affinity tag can select between sequences that dissociate quickly and sequences that dissociate slowly.

Claims

1. A method for selecting a high-affinity compound for a target molecule, comprising the following steps. (1) A step of mixing a target molecule (B) with a candidate compound (A) that is a high affinity compound for the target molecule (B) to obtain the candidate compound (A) to which the target molecule (B) is bound, wherein the candidate compound (A) has a structural site (A1) that can bind to the target molecule (B) and a structural site (A2) that can bind to a selective auxiliary material (C), and is capable of binding to only one of the target molecule (B) and the selective auxiliary material (C), the above step (2) A step to obtain candidate compound (A) to which the target molecule (B) obtained in step (1) is bound, by mixing with a selective aid (C) to obtain candidate compound (A) to which the target molecule (B) is bound and candidate compound (A) to which the selective aid (C) is bound. (3) A step to separate the candidate compound (A) to which the target molecule (B) obtained in step (2) is bound from the candidate compound (A) to which the selection aid (C) is bound, and recover the candidate compound (A) to which the target molecule (B) is bound as a high affinity compound for the target molecule.

2. The method according to claim 1, wherein the target molecule (B) is a target molecule bound to the first bead.

3. The method according to claim 1, further comprising the step of mixing the first beads with a candidate compound (A) of a high affinity compound for the target molecule (B) before the above step (1), and removing the candidate compound (A) bound to the first beads.

4. The method according to claim 2 or 3, wherein the volume average diameter of the first bead is 10 nm or more and 1000 μm or less.

5. The method according to claim 1, wherein the selection aid (C) has a portion that connects to a structural portion (A2) that can be connected to the selection aid (C) and a base portion, and the base portion is a second bead.

6. The method according to claim 5, wherein the volume average diameter of the second bead is 10 nm or more and 1000 μm or less.

7. The method according to claim 1, wherein the molar amount of the selection aid (C) in step (2) is 10 times or more and 10,000 times or less than the molar amount of the target molecule (B).

8. In step (2) above, the molar amount of the selection aid (C) is 10 times or more the molar amount of the candidate compound (A) １１ The method according to claim 1, wherein the result is less than or equal to the original amount.

9. The method according to claim 1, wherein step (2) is carried out in a buffer solution.

10. The method according to claim 1, wherein step (2) is carried out at a temperature of 0°C or higher and 50°C or lower.

11. The method according to claim 1, wherein the structural portion (A2) that can bind to the selected auxiliary material (C) has a polypeptide structure.

12. The method according to claim 1, wherein the candidate compound (A) is a peptide.

13. The method according to claim 12, wherein the number of amino acid residues in the peptide is 3 or more and 20 or less.

14. The method according to claim 12 or 13, wherein the peptide is a modified peptide.

15. The method according to claim 12, wherein the peptide is in the form of a ribosome display complex comprising the peptide, ribosomes, and mRNA.

Images