Method for producing (antibody fragment)-substrate conjugate

Abstract

According to the present invention, there is developed and provided a method for efficiently producing an antibody-substrate conjugate which is uniform and has a small molecular size and in which the binding ratio between a substrate molecule and an antibody is controlled in a modification site-specific manner.　Provided is a method for producing an (antibody fragment)-substrate conjugate with high production efficiency, the method comprising fragmenting an antibody to obtain an antibody fragment and then modifying the antibody fragment with a substrate molecule by using an MTG fusion body.

Description

Method for producing antibody fragment substrate conjugate

[0001] The present invention relates to a method for producing an antibody fragment-substrate conjugate.

[0002] Antibody drugs are therapeutic drugs that use antibodies that specifically bind to a given disease-related molecule, and have become a central part of biopharmaceuticals in recent years. However, although antibodies have extremely high specificity for antigens, they generally have lower antitumor activity on their own compared to conventional small molecule drugs. On the other hand, small molecule drugs have high antitumor activity, but are prone to side effects due to their low cell specificity and selectivity, making them difficult to use as drugs on their own.

[0003] As a means to resolve the above-mentioned problems of antibody drugs and small molecule drugs, antibody-drug conjugates (often referred to as "ADCs" in this specification) that combine the specificity of antibodies with the efficacy of small molecule drugs have attracted attention as next-generation antibody drugs. ADCs are drugs in which a small molecule drug is attached to an antibody via a linker. Utilizing the antigen specificity of antibodies, small molecule drugs can be delivered specifically and selectively to target cells, enabling a more powerful and broader therapeutic range than existing chemotherapy agents, while also reducing the side effects of small molecule drugs.

[0004] Chemical and enzymatic methods are known for producing ADCs. Chemical methods involve attaching drugs to lysine or cysteine ​​residues on antibodies via chemically reactive groups. However, the large number of lysine residues in antibodies makes it difficult to control the drug binding site. Therefore, enzymatic methods, which are highly specific and can catalyze reactions under mild conditions, have been actively studied, and various enzyme-based ADC production methods have been reported (Non-Patent Documents 1-3). Among these, the enzymatic method that has attracted the most attention in recent years is the method using microbial transglutaminase (often referred to as "MTG" herein) (Patent Document 1, Non-Patent Document 4).

[0005] Transglutaminase (often referred to as "TGase" in this specification) is an enzyme that specifically crosslinks glutamine and lysine residues. It exists in many organisms, including microorganisms, animals, and plants, and is responsible for various functions such as regulating gene expression, blood coagulation reactions, and skin formation (Non-Patent Document 5). Among these, TGase isolated from the actinomycete Streptomyces mobaraensis (Non-Patent Document 6) is called MTG, can be mass-expressed, and exhibits high crosslinking activity. Furthermore, it does not contain the cofactor Ca, which is required for catalytic activity in other TGases. 2+ However, MTG does not require this step. As a result, MTG has been used in recent years to produce ADCs with uniform structure as an easy-to-operate TGase.

[0006] Regarding the conjugation of IgG and a drug using MTG, a method in which IgG is deglycosylated and the drug is introduced into the exposed glutamine residue at position 295 (Q295) is disclosed in Non-Patent Document 7. This method is currently the most commonly used method for producing ADCs, but it requires labor and cost for the deglycosylation treatment and is therefore far from being a simple method.

[0007] Furthermore, MTG is expressed in microorganisms as an inactive precursor (Figure 26A). The N-terminal propeptide region is cleaved and removed by the protease activity of S. mobaraensis itself, resulting in the mature, active form (Figure 26B). The mature MTG is required for the production of ADCs using MTG. However, direct expression of the mature MTG in an E. coli expression system results in the formation of insoluble inclusion bodies, making it necessary to coexpress a recombinant MTG precursor with a protease. However, this expression system can inactivate MTG through partial hydrolysis by the protease. Furthermore, a purification step to completely remove the propeptide and protease after expression is required. Therefore, the preparation of active MTG also involves additional steps, labor, and cost.

[0008] Patent Document 2 discloses a new approach to solving the above problems: an MTG mutant (active MTG precursor) that, despite being a precursor, has catalytic activity equivalent to that of the mature form and can be mass-produced in an Escherichia coli expression system. However, this MTG mutant still requires deglycosylation of the antibody, and the modification efficiency of the MTG mutant on untreated native antibodies remains extremely low.

[0009] To solve the above problems, the present inventors developed a microbial transglutaminase fusion (often referred to herein as "MTG fusion") in our previous application, PCT / JP2024 / 016451 (filed April 26, 2024), in which antibody-specific protein G was fused with the aforementioned MTG mutant. This MTG fusion can selectively crosslink lysine residues at specific positions in a native antibody with glutamine residues in a substrate molecule without deglycosylation. Furthermore, by changing the order of protein G and the MTG mutant in the MTG fusion, two lysine residues at positions 65 and 225 on the antibody can be selectively and specifically modified. This enabled the production of homogeneous antibody-substrate complexes with site-specific modification and controlled binding ratios between the substrate molecule and the antibody.

[0010] WO2019 / 107288WO2018 / 004014

[0011] Grunewald J., et al., 2015, Bioconjugate Chem., 26: 2554-2562Sato S., et al., 2020, Bioconjugate Chem., 31: 1417-1424Lee MD, et al., 2019, Bioconjugate Chem., 30: 2539-2543Schneider H., et al., 2020, Anal. Biochem., 595: 113615Chen JSK & Mehta K., 1999, Int. J. Biochem. Cell Biol., 31: 817-836Ando H., et al., 1989, J. Agric. Food Chem., 53 (10): 2613-2617Dennler P., et al., 2014, Bioconjugate Chem., 25: 569-578

[0012] The MTG fusion molecule developed by the present inventors has made it possible to produce antibody-drug conjugates that retain the antigen specificity of the antibody and can specifically and selectively deliver small molecule drugs to target cells. However, the resulting antibody-drug conjugates have limited tissue penetration due to their large molecular size, and have also presented concerns such as increased systemic accumulation after administration and slow clearance. Furthermore, the Fc region of the antibody may cause side effects in ADCs.

[0013] To solve the above problems, the present inventors conducted further research and found that fragmenting the antibody in an antibody-substrate complex to form an antibody-fragment-substrate complex solved the above problems. For example, an antibody-fragment-substrate complex using an antibody Fab fragment has a molecular weight that is only about one-third that of a full-length IgG and has high tissue permeability. Furthermore, when producing an antibody-fragment-substrate complex using an MTG fusion, the inventors found that a higher substrate modification rate can be achieved by crosslinking the antibody fragment and substrate molecule with the MTG fusion, rather than by fragmenting the antibody domain in an antibody-substrate complex using an MTG fusion. The present invention is based on the results of this new research and development and provides the following:

[0014] (1) A method for producing an antibody fragment substrate complex, comprising: an enzyme binding step of binding an MTG fusion to an antibody fragment to prepare an enzyme-antibody fragment complex; a crosslinking step of mixing the enzyme-antibody fragment complex with a substrate molecule containing a glutamine residue and crosslinking the substrate molecule to the antibody fragment to prepare an enzyme-antibody fragment-substrate complex; and an enzymatic separation step of separating the MTG fusion from the enzyme-antibody fragment-substrate complex to obtain an antibody fragment substrate complex, wherein the MTG fusion contains an active MTG precursor and an antibody-binding protein, and the active MTG precursor contains a mutant propeptide region in which at least one tyrosine residue and a lysine residue, at least one asparagine residue and a lysine residue, or at least one asparagine residue, at least one tyrosine residue and a lysine residue in the propeptide region of the MTG precursor are substituted with other amino acid residues. (2) The method according to (1), comprising an antibody fragmentation step of cleaving an immunoglobulin at a predetermined site with a protease to obtain antibody fragments. (3) The production method according to (2), which includes an antibody fragment purification step of separating antibody fragments containing the target site modified by the MTG fusion product from the antibody fragments obtained after the antibody fragmentation step. (4) The production method according to (2) or (3), in which the protease is a cysteine ​​protease or an aspartic acid protease. (5) The production method according to (4), in which the protease is an aspartic acid protease, and in the antibody fragmentation step, antibody fragments are obtained by cleaving immunoglobulin at a predetermined site with the aspartic acid protease and then performing a reduction treatment. (6) The production method according to any of (1) to (5), in which the MTG fusion product contains the active MTG precursor at the N-terminus and the antibody-binding protein at the C-terminus. (7) The production method according to any of (1) to (5), in which the MTG fusion product contains the active MTG precursor at the C-terminus and the antibody-binding protein at the N-terminus. (8) The method of any one of (1) to (7), wherein the antibody-binding protein is a mutant antibody-binding protein. (9) The method of (8), wherein the mutant antibody-binding protein is Fab-binding protein G that has lost the ability to bind to Fc in IgG.(10) The method for producing according to (8), wherein the mutant antibody-binding protein is Fc-binding protein G or Fc-binding protein A that has lost the ability to bind to Fab in IgG. (11) The method for producing according to any one of (1) to (10), wherein the active MTG precursor and the antibody-binding protein are fused via a linker consisting of a peptide chain. (12) The method for producing according to any one of (1) to (11), wherein the substrate molecule is a nucleic acid, a polypeptide, a sugar chain, a lipid, a low-molecular-weight compound, or a combination thereof. (13) The method for producing according to (12), wherein the low-molecular-weight compound is any one selected from the group consisting of a drug, a fluorescent substance, a luminescent substance, and an enzyme. (14) The method for producing according to (12) or (13), wherein the substrate molecule comprises a glutamine tag. (15) The method for producing according to any one of (1) to (14), wherein the substrate molecule comprises an azide group, an alkynyl group, or a cycloalkynyl group. (16) A method for producing an enzyme-antibody fragment conjugate for crosslinking a substrate molecule containing a glutamine residue to an antibody fragment, the method comprising an enzyme binding step of binding a microbial transglutaminase fusion to an antibody fragment to prepare an enzyme-antibody fragment conjugate, the microbial transglutaminase fusion comprising an activated microbial transglutaminase precursor and an antibody-binding protein, and the activated microbial transglutaminase precursor comprising a mutant propeptide region in which at least one tyrosine residue and a lysine residue, at least one asparagine residue and a lysine residue, or at least one asparagine residue, at least one tyrosine residue and a lysine residue in the propeptide region of the microbial transglutaminase precursor are substituted with other amino acid residues. (17) An antibody fragment substrate conjugate in which a substrate molecule is crosslinked to the lysine residue at position 65 of the heavy chain of human Fab. (18) An antibody fragment substrate conjugate in which a substrate molecule is crosslinked to the lysine residue at position 225 of the heavy chain of human F(ab')2. This specification includes the disclosure of Japanese Patent Application No. 2024-105751, from which the present application claims priority.

[0015] According to the method of the present invention for producing an antibody fragment substrate conjugate, it is possible to produce a uniform antibody fragment substrate conjugate in which the number and binding positions of the substrate molecules and antibody fragments are controlled.

[0016] According to the method for producing an antibody fragment substrate conjugate of the present invention, an antibody fragment substrate conjugate can be obtained which has a smaller molecular size than an antibody substrate conjugate and is easier to handle.

[0017] Furthermore, according to the method for producing an antibody fragment substrate complex of the present invention, an antibody fragment substrate complex can be obtained in a higher yield than in a production method in which the antibody is fragmented by enzymatic treatment of the antibody substrate complex.

[0018]

[0033] Figure 1 is a schematic diagram of a microbial transglutaminase fusion used in the method for producing an antibody fragment substrate conjugate of the present invention. A and B both represent the microbial transglutaminase fusions used in the Examples herein, with A representing Pro(K10R / Y12A)MTG-(G4S)3-pG and B representing pG-G4S-Pro(K10R / Y12A)MTG.

[0034] Figure 2 is a flow diagram of the method for producing an antibody fragment substrate conjugate of the present invention.

[0035] Figure 3 is a conceptual diagram of the experimental strategy used in the Examples.

[0036] Figure 2 shows the results of pepsin fragmentation of an antibody substrate conjugate obtained using the MTG fusion, Pro(K10R / Y12A)MTG-(G4S)3-pG, in Strategy 1 shown in Figure 3. (A) shows the results of non-reducing SDS-PAGE staining with CBB of Herceptin (IgG) modified with Q substrate (FAM-YPLQMRG-NH2) by Pro(K10R / Y12A)MTG-(G4S)3-pG (represented as ProMTG-pG in the figure), and (B) shows the results of image analysis using a fluorescent imager. Figure 3 shows the results of pepsin fragmentation of the antibody-substrate complex obtained using the MTG fusion construct pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG (represented as pG-ProMTG in the figure) in Strategy 1. (A) shows the results of non-reducing SDS-PAGE staining with CBB of Herceptin (IgG) modified with Q substrate (FAM-YPLQMRG-NH2) by pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG, and (B) shows the results of image analysis using a fluorescent imager. Figure 3 shows the results of crosslinking a substrate molecule (denoted "FAM-Q" in the figure) using the MTG fusion protein Pro(K10R / Y12A)MTG-(G4S)3-pG (denoted "ProMTG-pG" in the figure) to pepsin-fragmented Herceptin Fab', as shown in Strategy 2 in Figure 3. Panel A shows the results of CBB staining of non-reducing SDS-PAGE after the crosslinking reaction, and Panel B shows the results of image analysis using a fluorescent imager. Figure 3 shows the results of crosslinking a substrate molecule (denoted "FAM-Q" in the figure) using the MTG fusion protein pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG (denoted "pG-ProMTG" in the figure) to pepsin-fragmented Herceptin Fab', as shown in Strategy 2 in Figure 3.Panel A shows the results of CBB staining of non-reducing SDS-PAGE after the crosslinking reaction, and panel B shows the image analysis results using a fluorescent imager. Figure 1 shows the results of reverse-phase HPLC to confirm the modification of K65 on the Herceptin heavy chain by the MTG fusion protein Pro(K10R / Y12A)MTG-(G4S)3-pG with the Q substrate. Panel A shows the results for an enzyme-substrate complex in which K65 on IgG (Herceptin) was modified with the Q substrate (Herceptin(K65)-Q), and panel B shows the results for an enzyme-fragment-substrate complex in which K65 on Herceptin Fab' was modified with the Q substrate (Herceptin Fab'(K65)-Q). In the figure, a represents the MTG fusion, b represents the light chain of Herceptin (or Herceptin Fab'), c represents the unmodified heavy chain of Herceptin, c' represents the modified heavy chain, d represents the unmodified heavy chain of Herceptin Fab', and d' and d'' represent the modified heavy chain of Herceptin Fab'. This figure shows the results of reverse-phase HPLC to confirm the modification of K225 of the Herceptin heavy chain with the Q substrate by the MTG fusion pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG. Panel A shows the results for an enzyme-substrate complex in which K65 of IgG (Herceptin) has been modified with the Q substrate (Herceptin(K65)-Q), and panel B shows the results for an enzyme-fragment-substrate complex in which K65 of Herceptin Fab' has been modified with the Q substrate (Herceptin Fab'(K65)-Q). In the figure, a represents the MTG fusion, b represents the light chain of Herceptin (or Herceptin Fab'), c represents the unmodified heavy chain of Herceptin, c' represents the modified heavy chain, d represents the unmodified heavy chain of Herceptin Fab', and d' and d'' represent the modified heavy chain of Herceptin Fab'. The figure shows the modification rate of substrate molecules to K65 on the Herceptin heavy chain calculated from the results of reverse-phase HPLC using the MTG fusion Pro(K10R / Y12A)MTG-(G4S)3-pG. In the figure, "Whole" represents the modification rate of substrate molecules to K65 on IgG, and "Fab'" represents the modification rate of substrate molecules to K65 on the Fab' fragment. The figure shows the modification rate of substrate molecules to K225 on the Herceptin heavy chain calculated from the results of reverse-phase HPLC using the MTG fusion pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG.In the figure, "Whole" indicates the modification rate of the substrate molecule at K225 of IgG, and "Fab'" indicates the modification rate of the substrate molecule at K225 of the Fab' fragment. The results of papain fragmentation of an antibody-substrate complex obtained using the MTG fusion construct Pro(K10R / Y12A)MTG-(G4S)3-pG are shown. A shows the results of non-reducing SDS-PAGE stained with CBB of the Q substrate (FAM-YPLQMRG-NH2) modification of Herceptin (IgG) with Pro(K10R / Y12A)MTG-(G4S)3-pG, and B shows the image analysis results using a fluorescent imager. The results of papain fragmentation of an antibody-substrate complex obtained using the MTG fusion construct pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG are shown. (A) shows the results of CBB staining of non-reducing SDS-PAGE of Herceptin (IgG) modified with Q substrate (FAM-YPLQMRG-NH2) using pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG, and (B) shows the image analysis results using a fluorescent imager. Figure 1 shows the results of crosslinking of substrate molecules to papain fragment Fab using the MTG fusion complex Pro(K10R / Y12A)MTG-(G4S)3-pG. (A) shows the results of CBB staining of non-reducing SDS-PAGE after crosslinking, and (B) shows the image analysis results using a fluorescent imager. Figure 1 shows the results of crosslinking of substrate molecules to papain fragment Fab using the MTG fusion complex pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG. (A) shows the results of CBB staining of non-reducing SDS-PAGE after crosslinking, and (B) shows the image analysis results using a fluorescent imager. 1 shows the results of reverse-phase HPLC to confirm the modification of K65 on the Herceptin heavy chain with the Q substrate by the MTG fusion construct Pro(K10R / Y12A)MTG-(G4S)3-pG. Panel A shows the results for an enzyme-substrate complex in which K65 on IgG (Herceptin) was modified with the Q substrate (Herceptin(K65)-Q), and panel B shows the results for an enzyme-fragment-substrate complex in which K65 on Herceptin Fab was modified with the Q substrate (Herceptin Fab(K65)-Q).In the figure, a represents the MTG fusion, b represents the light chain of Herceptin (or Herceptin Fab), c represents the unmodified heavy chain of Herceptin, c' represents the modified heavy chain, e represents the unmodified heavy chain of Herceptin Fab, and e' and e'' represent the modified heavy chain of Herceptin Fab. This figure shows the results of reverse-phase HPLC to confirm the modification of K225 of the Herceptin heavy chain with the Q substrate by the MTG fusion pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG. A shows the results for an enzyme-substrate complex in which K65 of IgG (Herceptin) has been modified with the Q substrate (Herceptin(K65)-Q), and B shows the results for an enzyme-fragment-substrate complex in which K65 of Herceptin Fab has been modified with the Q substrate (Herceptin Fab'(K65)-Q). In the figure, a represents the MTG fusion, b represents the light chain of Herceptin (or Herceptin Fab), c represents the unmodified heavy chain of Herceptin, c' represents the modified heavy chain, e represents the unmodified heavy chain of Herceptin Fab, and e' represents the modified heavy chain of Herceptin Fab. The figure shows the modification rate of substrate molecules to K65 on the Herceptin heavy chain calculated from the results of reverse-phase HPLC using the MTG fusion Pro(K10R / Y12A)MTG-(G4S)3-pG. In the figure, "Whole" represents the modification rate of substrate molecules to K65 on IgG, and "Fab" represents the modification rate of substrate molecules to K65 on the Fab fragment. The figure shows the modification rate of substrate molecules to K225 on the Herceptin heavy chain calculated from the results of reverse-phase HPLC using the MTG fusion pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG. In the figure, "Whole" indicates the modification rate of the substrate molecule to K225 of IgG, and "Fab" indicates the modification rate of the substrate molecule to K225 of the Fab fragment. This is a flow diagram of the method for producing an enzyme-antibody fragment conjugate of the present invention. This is a chromatogram by cation exchange chromatography showing the purification of an antibody fragment substrate conjugate from the enzyme-antibody fragment-substrate conjugate. The enzyme-antibody fragment-substrate complexes are A, which is the ProMTG-pG-Fab(K65)-Q substrate complex, and B, which is the pG-ProMTG-Fab(K225)-Q substrate complex. In figure A, "a" indicates the MTG fusion protein ProMTG-pG, "b" indicates the ProMTG-pG-Fab(K65)-Q substrate complex, and "c" indicates the desired Fab(K65)-Q substrate complex.In addition, in the figure of B, d indicates the MTG fusion compound pG-ProMTG, and e indicates the target Fab(K225)-Q substrate complex. This is a cation exchange chromatogram showing the purification of an antibody-substrate complex from an enzyme-antibody-substrate complex. The enzyme-antibody-substrate complexes are A, ProMTG-pG-IgG(K65)-Q substrate complex, and B, pG-ProMTG-IgG(K225)-Q substrate complex. In the figure of A, a, b, and c indicate the MTG fusion compound ProMTG-pG, ProMTG-pG-IgG(K65)-Q substrate complex, and IgG(K65)-Q substrate complex. In the figure of B, d, e, indicate the MTG fusion compound pG-ProMTG, and IgG(K225)-Q substrate complex. This figure shows the results of reverse-phase HPLC to confirm the effect of MTG fusions containing linkers of different lengths on bimolecular modification of Herceptin Fab heavy chain. (A) shows the modification rate of the antibody-substrate complex obtained using Pro(K10R / Y12A)MTG-(G4S)3-pG, which contains a long linker (G4S)3, as the MTG fusion complex. (B) shows the modification rate of the antibody-substrate complex obtained using Pro(K10R / Y12A)MTG-G4S-pG, which contains a short linker G4S, as the MTG fusion complex. In the figure, (a) represents the substrate molecule (TAMRA-Q), (b) represents the light chain of Herceptin Fab, (c) represents the unmodified heavy chain of Herceptin Fab, (c') represents the heavy chain of Herceptin Fab modified with one substrate molecule, and (c'') represents the heavy chain of Herceptin Fab modified with two substrate molecules. This figure shows the binding of antibody-substrate complexes or antibody fragment-substrate complexes to the HER2-positive (HER2(+)) breast cancer cell line SK-BR-3 cells. The figures show CLSM images (bright field) and fluorescence images of SK-BR-3 cells when a Herceptin (K65)-Q substrate conjugate (represented as "IgG(K65)-TAMRA" in the figure) or a Herceptin (K225)-Q substrate conjugate (represented as "IgG(K225)-TAMRA" in the figure) was added as an antibody-substrate conjugate, and a Fab(K65)-Q substrate conjugate (represented as "Fab(K65)-TAMRA" in the figure) or a Fab(K225)-Q substrate conjugate (represented as "IgG(K225)-TAMRA" in the figure) was added as an antibody-fragment substrate conjugate.

[0033] Figure 1 shows the binding of antibody-substrate conjugates or antibody fragment-substrate conjugates to the HER2-negative (HER2(-)) breast cancer cell line MDA-MB-231 cells. CLSM images (bright field) and fluorescence images are shown of MDA-MB-231 cells when a Herceptin(K65)-Q substrate conjugate (represented as "IgG(K65)-TAMRA" in the figure) or a Herceptin(K225)-Q substrate conjugate (represented as "IgG(K225)-TAMRA" in the figure) was added as the antibody-substrate conjugate, and a Fab(K65)-Q substrate conjugate (represented as "Fab(K65)-TAMRA" in the figure) or a Fab(K225)-Q substrate conjugate (represented as "IgG(K225)-TAMRA" in the figure) was added as the antibody-fragment substrate conjugate. 27 shows schematic diagrams of the precursor (A: ProMTG), mature form (B: mature MTG), and active precursor (C: active ProMTG) of microbial transglutaminase. In the precursor (A), the propeptide region inhibits the activity of the mature MTG region. In the active precursor (C), the inhibitory effect of the propeptide region is inhibited by mutations in the propeptide region (represented by two black lines in the figure), resulting in an active precursor. This figure shows the modification of an antibody fragment with a click reaction element-conjugated glutamine tag using an MTG fusion. A is a conceptual diagram of the modification of an Fab with a glutamine tag using an MTG fusion. B is a graph showing the modification rate of the glutamine tag on an antibody fragment over time. This figure shows the results of mass spectrometry to confirm the number of DBCO-PEG4-LLQG modifications per Fab molecule in the Fab-PEG4-DBCO obtained by modifying an antibody fragment with a click reaction element-conjugated glutamine tag using an MTG fusion in Figure 27. The peak indicated by the arrow in A represents Fab, and the peak indicated by the arrow in B represents Fab-PEG4-DEBCO. Figures showing substrate modification of antibody fragments by click reaction. A is a conceptual diagram showing bond modification by click reaction between the antibody fragment Fab-PEG4-DEBCO and the azido substrate molecule Azide-PEG3-FAM. A is a conceptual diagram showing bond modification by click reaction between the antibody fragment Fab-PEG4-DEBCO and the azido substrate molecule Azide-PEG3-TAMRA.Figure C shows the results of reverse-phase HPLC confirming the modification of Fab-PEG4-DEBCO with Azide-PEG3-FAM via click reaction. In the figure, the solid line shows the results of reverse-phase HPLC after the click reaction of a mixture of Fab-PEG4-DEBCO and Azide-PEG3-FAM, while the dashed line shows the results of reverse-phase HPLC of Fab-PEG4-DEBCO before the click reaction. Also, a shows the light chain of Fab, b shows the heavy chain of Fab-PEG4-DEBCO, and c shows the heavy chain of Fab-FAM. Figure A shows the results of non-reducing SDS-PAGE separation of Fab-PEG4-DEBCO modified with Azide-PEG3-FAM or Azide-PEG3-TAMRA via click reaction. Figure A shows the results of CBB staining of SDS-PAGE, while B and C show the results of fluorescent imager analysis of FAM and TAMRA, respectively. Fab-PEG4-DEBCO was modified with muGFP-PEG3-Azide by click reaction, and then separated by non-reducing SDS-PAGE followed by CBB staining.

[0019] 1. Method for Producing Antibody Fragment Substrate Conjugates 1-1. Overview A first aspect of the present invention is a method for producing an antibody fragment substrate conjugate. The production method of the present invention uses the MTG fusion described in the prior application PCT / JP2024 / 016451 as an enzyme to bind a substrate molecule to a predetermined position on an antibody fragment, thereby obtaining the desired antibody fragment substrate conjugate. According to the production method of the present invention, an antibody fragment substrate conjugate that has a smaller molecular size and is easier to handle than an antibody substrate conjugate can be produced in high yield.

[0020] 1-2. Definitions of Terms The following terms used in this specification are defined below.

[0021] As used herein, the term "antibody fragment-substrate complex" refers to a complex in which a substrate molecule is linked to a specific amino acid residue of an antibody fragment.

[0022] As used herein, the term "antibody fragment-drug conjugate (AFDC)" refers to a type of antibody fragment substrate conjugate in which the substrate molecule is a drug, particularly a low-molecular-weight compound with cytocidal effect. The antibody fragments included in the antibody fragment substrate conjugates of the present invention have a smaller molecular size than antibodies, and are therefore expected to improve uptake into cellular tissues and systemic clearance (ease of excretion of the conjugate from the body).

[0023] As used herein, the term "antibody fragment" refers to a portion of an immunoglobulin that contains at least one variable region having antigen-binding activity. In principle, an antibody fragment as used herein contains at least one modification target site. Antibody fragments as used herein are described in detail in "1-4. Structure of antibody fragments."

[0024] As used herein, the term "modification target site" refers to a specific lysine (K) residue contained in the antibody fragment, which is a reactive site at which a substrate molecule is crosslinked by the catalytic activity of a microbial transglutaminase or a microbial transglutaminase fusion protein, as described below.

[0025] As used herein, the term "substrate molecule" refers to a molecule that can serve as a substrate for the cross-linking reaction of microbial transglutaminase together with an antibody fragment. Examples include nucleic acids, (poly)peptides, sugar chains, lipids, low-molecular-weight compounds, or combinations thereof. Like antibody fragments, substrate molecules also contain at least one glutamine residue (Q), which serves as a reactive site in the cross-linking reaction, either by itself or via a peptide tag. Furthermore, substrate molecules may contain at least one azide group, or an alkynyl group or a cycloalkynyl group.

[0026] As used herein, the term "peptide tag" refers to a peptide chain that provides a glutamine residue as a reactive site in a crosslinking reaction on a substrate molecule. When a substrate molecule does not have an amino acid that serves as a reactive site, modification of the substrate molecule with a peptide tag is essential.

[0027] As mentioned above, "Microbial transglutaminase" (MTG) refers to the transglutaminase (TGase) derived from Streptomyces mobaraensis, a type of actinomycete. Unlike TGases from other organisms, it requires the cofactor Ca for activity. 2+ MTG catalyzes an acyl transfer reaction between the γ-carboxyamide group in the side chain of a glutamine residue in a protein and the amino group and primary amine in the side chain of a lysine residue in a protein, without requiring a catalyzed acyl transfer reaction. MTG is produced as an inactive microbial transglutaminase precursor in microorganisms such as Escherichia coli. In this specification, the term "MTG" refers to both active and inactive MTG unless otherwise specified.

[0028] As used herein, a "microbial transglutaminase fusion (often referred to herein as an "MTG fusion")" is a fusion protein comprising a cross-linking catalytic domain and an antibody-binding domain. The MTG fusion of the present invention has catalytic activity that cross-links a substrate molecule to a specific amino acid residue of an antibody. The specific structure of the MTG fusion will be described below in "1-3. Structure of a microbial transglutaminase fusion (MTG fusion)."

[0029] The term "fusion protein" refers to a polypeptide chain containing multiple peptide domains of different origins that are linearly linked together. In this specification, this term refers to an artificial polypeptide encoded by a recombinant gene artificially produced using genetic engineering technology.

[0030] As used herein, the term "cross-linking catalytic domain" refers to a peptide domain having the enzymatic activity of cross-linking the side chains of lysine and glutamine residues via an amide bond in the MTG fusion of the present invention. Specifically, microbial transglutaminase functions as the cross-linking catalytic domain in the MTG fusion of the present invention.

[0031] As used herein, the term "antibody-binding domain" refers to a peptide domain in the MTG fusion of the present invention that has binding affinity to an antibody. In the MTG fusion of the present invention, specifically, an antibody-binding protein functions as the antibody-binding domain.

[0032] As used herein, the term "microbial transglutaminase precursor (often referred to as "ProMTG")" refers to mature MTG that contains a propeptide region at the N-terminus. Wild-type microbial transglutaminase precursor is an inactive MTG.

[0033] As used herein, the "propeptide region" refers to a 46-amino acid residue subunit located at the N-terminus of ProMTG. The propeptide region functions as an intramolecular chaperone, assisting in the folding of MTG, and is an essential domain for soluble expression in microorganisms. On the other hand, the propeptide region also functions as an inhibitor (quencher) that suppresses the MTG activity of mature microbial transglutaminase.

[0034] As used herein, "wild-type microbial transglutaminase precursor" (often referred to as "wild-type ProMTG") refers to an inactivated ProMTG in which the cross-linking activity of the microbial transglutaminase mature product is suppressed by the propeptide region.

[0035] As used herein, the term "mature microbial transglutaminase" (often referred to as "mature MTG") refers to MTG that has been activated by removing the propeptide region from ProMTG with a protease.

[0036] As used herein, "active microbial transglutaminase precursor (active ProMTG)" refers to a mutant ProMTG that has been made active despite being a precursor by artificially introducing a mutation into the propeptide region of wild-type ProMTG to create a mutant propeptide region.

[0037] As used herein, the term "mutant propeptide region" refers to a propeptide region in which a mutation has been introduced into an amino acid residue in the wild-type propeptide region present on the N-terminal side of wild-type ProMTG, and which has lost the quencher function that inhibits MTG activity.

[0038] As used herein, the term "antibody binding protein" refers to a group of proteins that have binding properties to immunoglobulins, particularly fragments thereof that contain a modified target site.

[0039] As used herein, the term "mutant antibody-binding protein" refers to a group of antibody-binding proteins that have amino acid mutations and that maintain their binding activity to immunoglobulins.

[0040] As used herein, "amino acid identity" refers to the percentage (%) of matching amino acid residues in the total number of amino acid residues when two amino acid sequences are compared and aligned by inserting appropriate gaps into one or both sequences as needed to maximize the number of identical amino acid residues. Alignment of two amino acid sequences to calculate amino acid identity can be performed using known programs such as Blast, FASTA, and ClustalW.

[0041] As used herein, "base identity" refers to the percentage (%) of the number of matching bases in the total number of bases when two base sequences are compared and aligned by inserting appropriate gaps into one or both of them as necessary to maximize the number of matching bases.

[0042] As used herein, "plurality" refers to, for example, 2 to 20, 2 to 15, 2 to 10, 2 to 7, 2 to 5, 2 to 4, or 2 to 3. Additionally, "several" refers to 2 to 3.

[0043] As used herein, "(amino acid) substitution" refers to a substitution within a conservative amino acid group that has similar properties, such as charge, side chain, polarity, and aromaticity, among the 20 amino acids that constitute natural proteins. Examples include substitutions within the uncharged polar amino acid group with a low-polarity side chain (Gly, Asn, Gln, Ser, Thr, Cys, Tyr), the branched-chain amino acid group (Leu, Val, Ile), the neutral amino acid group (Gly, Ile, Val, Leu, Ala, Met, Pro), the neutral amino acid group with a hydrophilic side chain (Asn, Gln, Thr, Ser, Tyr, Cys), the acidic amino acid group (Asp, Glu), the basic amino acid group (Arg, Lys, His), and the aromatic amino acid group (Phe, Tyr, Trp). Amino acid substitutions within these groups are preferred because they are known to be less likely to cause changes in the properties of peptides. In other words, amino acid substitutions within a group are preferred because even if multiple amino acid substitutions (e.g., 2 to 20) occur, the properties of the peptide are unlikely to change and the peptide is likely to exhibit the same function.

[0044] As used herein, the term "linker" refers to a peptide chain that crosslinks the bridging catalytic domain and the antibody-binding domain in the MTG fusion of the present invention. In the MTG fusion containing a linker, the peptide chain of the bridging catalytic domain and the peptide chain of the antibody-binding domain are indirectly linked via the linker.

[0045] As used herein, the term "enzyme-antibody fragment complex" refers to a complex of an MTG fusion as an enzyme and an antibody fragment. This complex is one of the intermediates of the antibody fragment-substrate complex formed in the enzyme binding step in the production method described in the first aspect of the present invention. This complex is formed by binding of the MTG fusion to the target antibody fragment to be modified via the antibody-binding domain contained in the MTG fusion.

[0046] As used herein, the term "enzyme-antibody fragment-substrate complex" refers to a complex consisting of three molecules in which a substrate molecule is crosslinked to a specific amino acid residue in the antibody fragment portion of the enzyme-antibody fragment complex. This is one of the intermediates of the antibody fragment substrate complex formed in the crosslinking step in the production method according to the first aspect of the present invention, and is a complex in a state in which an MTG fusion protein, which is ultimately unnecessary, is bound to the antibody fragment substrate complex, which is the final product, during the production process of the antibody fragment substrate complex.

[0047] The term "Fab region" refers to a region present in the heavy and light chains of an antibody, and composed of the variable region and C1 region (constant region 1) of each.

[0048] The term "Fc region" refers to a region present only in the heavy chain of an antibody, and is composed of CH2 and CH3 of the heavy chain.

[0049] A "protease" is a hydrolytic enzyme that cleaves peptide bonds, catalyzing the degradation of polypeptides into smaller polypeptides or single amino acids.

[0050] The method for producing an antibody fragment-substrate conjugate of the present invention uses a microbial transglutaminase fusion (MTG fusion) with a special structure as an enzyme for crosslinking an antibody fragment and a substrate molecule. Therefore, this chapter provides a detailed explanation of the structure of the MTG fusion.

[0051] 1-3-1. Components of MTG fusion protein Figure 1 shows a schematic diagram of an MTG fusion protein. As shown in this figure, the fusion protein MTG fusion protein (0100) contains active ProMTG (0101) and an antibody-binding protein (0102) on its polypeptide chain as essential components, and also contains a linker (0103) as an optional component. Each component is explained below.

[0052] (1) Active proMTG (active microbial transglutaminase precursor) Active ProMTG (0101) is a mutant ProMTG that has cross-linking catalytic activity in the precursor state, and catalyzes the cross-linking reaction between target antibody fragments and substrate molecules as a cross-linking catalytic domain in MTG fusion complexes.

[0053] As mentioned above, wild-type ProMTG is in an inactive state, with its activity suppressed by the propeptide region at its N-terminus. However, when a mutation is introduced into the propeptide region of wild-type ProMTG to form the mutant propeptide region (0104), it changes from an inactive form to an active ProMTG.

[0054] The propeptide region consists of 46 amino acid residues located on the N-terminus of wild-type ProMTG. Wild-type ProMTG can consist of any of the following amino acid sequences: (a) the amino acid sequence shown in SEQ ID NO: 1; (b) the amino acid sequence shown in SEQ ID NO: 1 with one or more amino acid residues added, deleted, and / or substituted at positions other than 10, 12, 16, 25, and 27; or (c) an amino acid sequence having 90% or greater amino acid identity with the amino acid sequence shown in SEQ ID NO: 1. Thus, the propeptide region corresponds to the 46 amino acid residues from the methionine residue at position 1 to the proline residue at position 46 in the amino acid sequence shown in SEQ ID NO: 1. Meanwhile, the amino acid sequence from the aspartic acid residue at position 47 onward constitutes the amino acid sequence of mature MTG.

[0055] The introduction of such mutations as addition, deletion, or substitution in the amino acid sequence of the propeptide region or the like can be carried out using a commercially available kit for mutagenesis that utilizes site-directed mutagenesis. TM Site-Directed Mutagenesis System (Thermo Fisher Scientific), Prime STAR® Mutagenesis Basal kit (Takara Bio Inc.), Mutan TM-Super Express Km (Takara Bio Inc.), etc. The introduction of these mutations can be confirmed by various amino acid sequencing methods, and structural analysis methods such as X-ray and NMR.

[0056] Specific examples of mutations of amino acid residues in the mutant propeptide region include substitutions of the lysine residue at position 10, the tyrosine residues at positions 12 and 16, and / or the asparagine residues at positions 25 and 27 in the amino acid sequence of the propeptide region with other amino acid residues (in this specification, such amino acid substitutions, for example, substitutions of the lysine residue at position 10 and the tyrosine residue at position 12, are often represented as "K10 / Y12").

[0057] Other amino acid residues substituted for the lysine residue include any amino acid selected from the group consisting of arginine residue (R), aspartic acid residue (D), glutamic acid residue (E), asparagine residue (N), glutamine residue (Q), serine residue (S), threonine residue (T), glycine residue (G), alanine residue (A), leucine residue (L), isoleucine residue (I), valine residue (V), cysteine ​​residue (C), tryptophan residue (W), phenylalanine residue (F), methionine residue (M), proline residue (P), tyrosine residue (T), and histidine residue (H). Preferred are arginine residue, aspartic acid residue, or glutamic acid residue.

[0058] Other amino acid residues that can be substituted for the tyrosine residue include alanine, glutamine, or histidine residues.

[0059] Other amino acid residues that can be substituted for the asparagine residue include aspartic acid, glutamic acid, or serine residues.

[0060] Examples of mutation positions of the lysine, tyrosine, and asparagine residues in the propeptide region include, for example, the lysine residue at position 10 (often referred to herein as "K10"), the tyrosine residues at positions 12 and 16 (often referred to herein as "Y12" and "Y16," respectively), and the asparagine residues at positions 25 and 27 (often referred to herein as "N25" and "N27," respectively) in the amino acid sequence shown in SEQ ID NO: 1, where the N-terminal aspartic acid residue is positioned 1. Alternatively, examples include the lysine residue at position 10, the tyrosine residues at positions 12 and 16, and the asparagine residues at positions 25 and 27 in SEQ ID NO: 1 in the wild-type ProMTG described in (b) and (c) above. Here, "corresponding position" refers to the position of an amino acid residue on the amino acid sequence shown in SEQ ID NO: 1 in the amino acid sequence shown in (b) or (c) when the amino acid sequence shown in SEQ ID NO: 1 and the amino acid sequence shown in (b) or (c) are aligned so as to maximize the number of identical amino acid residues.

[0061] The specific mechanism by which the substitution of the specific amino acid residues in the propeptide region of ProMTG with other amino acid residues results in the activation of the precursor is unclear. However, it is speculated that the substitution of the amino acid residues disrupts the hydrogen bonds that contribute to the structural stability of the propeptide region, and that the substitution results in electrostatic repulsion between the substituted amino acid residues, resulting in the activation of the precursor.

[0062] Specific examples of mutant propeptide regions include propeptide regions having two substitutions (herein, such amino acid substitutions are often referred to as "K10R / Y12A" and "K10R / Y16A") in the amino acid sequence set forth in SEQ ID NO: 1: a substitution of the lysine residue at position 10 with an arginine residue (herein, such amino acid substitutions are often referred to as "K10R"), and a substitution of the tyrosine residue at position 12 or 16 with an alanine residue (herein, such amino acid substitutions are often referred to as "Y12A" and "Y16A", respectively) (herein, such mutant propeptide regions are often referred to as "Pro(K10R / Y12A)" and "Pro(K10R / Y16A)", respectively). ProMTG containing such mutant propeptide regions (referred to herein as "Pro(K10R / Y12A)MTG" and "Pro(K10R / Y16A)MTG") can be active ProMTG.

[0063] In addition, a propeptide region having two substitutions (K10R and a mutation in which the asparagine residue at position 25 or 27 is replaced with an aspartic acid residue) (such amino acid substitutions are often referred to herein as "K10R / N25D" or "K10R / N27D", respectively) can also be an active form of ProMTG.

[0064] Furthermore, in addition to K10R, for example, two or more substitutions selected from the group consisting of Y12A, Y16A, N25D, and N27D, specifically, K10R / Y12A / Y16A, K10R / Y12A / N25D, K10R / Y12A / N27D, K10R / Y16A / N25D, K10R / Y16A / N27D, K10R / Propeptide regions having N25D / N27D, and K10R / Y12A / Y16A / N25D, K10R / Y12A / Y16A / N27D, K10R / Y12A / N25D / N27D, K10R / Y16A / N25D / N27D, or K10R / Y12A / Y16A / N25D / N27D can also be active ProMTG.

[0065] In the MTG fusion product, the active ProMTG may be either the full-length or a portion of ProMTG, so long as it maintains its catalytic cross-linking activity. The term "part of ProMTG" as used herein refers to a partial fragment of ProMTG in which one or more amino acid residues are deleted from the full-length ProMTG, while maintaining its catalytic cross-linking activity. TG activity can be evaluated and measured, for example, by fluorescence resonance energy transfer (FRET) or other methods.

[0066] (2) Antibody-binding protein The antibody-binding protein (0102) functions as an antibody-binding domain in the MTG fusion. This function allows the MTG fusion to bind to the target antibody fragment, thereby enabling the active MTG, which is the cross-linking catalytic domain, to come into close proximity with the target antibody fragment, thereby enhancing the cross-linking reaction.

[0067] The antibody-binding protein is not limited as long as it has the ability to bind to an immunoglobulin fragment containing the modified target site. Immunoglobulins are known to be of the IgG, IgM, IgA, IgE, and IgD classes, and any class may be used. IgG is preferred.

[0068] Examples of antibody-binding proteins include, but are not limited to, protein G and protein A. Each of these will be explained in detail below.

[0069] (2-1) Protein G "Protein G" (often referred to as "pG" herein) is a protein present in the cell wall of group G streptococci that has affinity for immunoglobulin G (IgG). pG consists of 12 domains, E, A1, B1, A2, B2, A3, S, C1, D1, C2, W, and M from the N-terminus, and has affinity for and can bind to all subclasses of IgG, from human to mouse IgG. Therefore, the target IgG of pG contained in the MTG fusion complex can be of any species. IgG from mammals is preferred, and human IgG is more preferred. pG binds to the interface between the CH2 and CH3 constant domains of the IgG Fc region and to the CH1 constant domain of the Fab region. pG binds more strongly to the Fc region of human IgG than to the Fab region, but more strongly to the Fab region of mouse IgG. pG is a preferred protein as the antibody binding protein in the MTG fusion.

[0070] (2-2) Protein A "Protein A" (often referred to as "pA" herein) is a protein present in the cell wall of Staphylococcus aureus. pA contains five homologous domains, domains E, D, A, B, and C from the N-terminus, and interacts strongly with the Fc region of IgG and weakly with the Fab region. It can bind to IgG from many species, from human to mouse IgG, but its binding strength differs depending on the IgG subclass. pA interacts with the interface between the CH2 and CH3 constant domains of the IgG Fc region, and binds to the IgG Fc region in almost the same region as pG.

[0071] The antibody-binding protein contained in the MTG fusion may be either a wild-type antibody-binding protein or a mutant antibody-binding protein, as long as it maintains antibody-binding ability.

[0072] In the case of the aforementioned pG, a specific example of the amino acid sequence of wild-type pG is the amino acid sequence shown in SEQ ID NO: 3. Furthermore, an example of the amino acid sequence of mutant pG is an amino acid sequence in which one or more amino acid residues have been added, deleted, and / or substituted in the amino acid sequence shown in SEQ ID NO: 3, or an amino acid sequence having 90% or more amino acid identity with the amino acid sequence shown in SEQ ID NO: 3.

[0073] In the case of the aforementioned pA, a specific amino acid sequence of wild-type pA is, for example, the amino acid sequence shown in SEQ ID NO: 5. An example of the amino acid sequence of mutant pA is an amino acid sequence in which one or more amino acid residues have been added, deleted, and / or substituted in the amino acid sequence shown in SEQ ID NO: 5, or an amino acid sequence having 90% or more amino acid identity with the amino acid sequence shown in SEQ ID NO: 5.

[0074] Furthermore, when an antibody-binding protein contains multiple antibody-binding regions, it may be a mutant antibody-binding protein mutated so that it binds only to a specific region of the antibody fragment. For example, wild-type pG has a Fab-binding region and an Fc-binding region that recognize and bind to the Fab and Fc regions of IgG, respectively. However, as described below, an example of an Fc-binding pG is one in which a mutation is introduced into the Fab-binding region to eliminate Fab-binding ability, thereby improving Fc-binding specificity using the remaining Fc antibody-binding region. When an antibody-binding protein contains multiple antibody-binding regions, there is a possibility that the antibody fragment will aggregate upon reaction between the MTG fusion and the antibody fragment. The mutant antibody-binding protein allows the MTG fusion to specifically bind to a specific region of the antibody fragment, thereby making it possible to suppress this aggregation.

[0075] Examples of mutant antibody-binding proteins having such a structure include Fab-binding protein G, Fc-binding protein G, and protein Z (often referred to as "pZ" herein).

[0076] "Fab-binding protein G" (often referred to herein as "pG(Fab)") is a mutant pG in which mutations have been introduced into the Fc-binding region of pG, resulting in loss of Fc-binding ability and enhanced binding specificity to the Fab region. The Fab-binding region of pG consists, for example, of the amino acid sequence shown in SEQ ID NO: 7. The term pG(Fab) used herein primarily refers to a mutant pG(Fab) (often referred to herein as "pG(Fab)-ΔKQ") that contains amino acid substitutions at all lysine and glutamine residues in the Fab-binding region and that can suppress pG self-labeling and antibody fragment aggregation when incorporated into an MTG fusion protein. However, this is not limited to this. A more specific example of pG(Fab)-ΔKQ is pG(Fab)-ΔKQ shown in SEQ ID NO: 8, in which all lysine and glutamine residues in the amino acid sequence shown in SEQ ID NO: 7 have been substituted with other amino acid residues (e.g., arginine and asparagine residues, respectively).

[0077] "Fc-binding protein G" (often referred to herein as "pG(Fc)") is a mutant pG in which mutations have been introduced into the Fab-binding region of pG, resulting in loss of its ability to bind to Fab fragments and thereby enhancing its binding specificity to the Fc region. The Fc-binding region of pG consists, for example, of the amino acid sequence shown in SEQ ID NO: 9. As used herein, pG(Fc) includes mutant pG(Fc) that contains amino acid substitutions at all or some of the lysine and glutamine residues in the Fc-binding region and that can suppress autolabeling when incorporated into an MTG fusion protein. More specifically, for example, there is pG(Fc) shown in SEQ ID NO: 10, in which five lysine residues in the amino acid sequence shown in SEQ ID NO: 9 have been substituted with other amino acid residues (e.g., arginine residues) (in the present specification, such pG(Fc) is often referred to as "pG(Fc)-ΔKx5". Note that a mutant pG containing amino acid substitutions at all lysine and glutamine residues in the Fc binding region is often referred to as "pG(Fc)-ΔKQ").

[0078] "Protein Z" (often referred to herein as "Fc-binding protein A" or "pZ") is a protein in which two mutations have been introduced into the B domain of pA: A1V for cloning purposes and G29A for hydroxylamine resistance. As a result, pZ loses its ability to bind to the Fab region of IgG and binds only to the Fc region. Examples of mutant pZ include pZCa2+ (SEQ ID NO: 33), which has been conferred Ca2+ binding ability and is used in this example; pZCa2+(ΔK), which has been replaced with arginine residues and is shown in SEQ ID NO: 34; and pZ(ΔK, Alkaline stabilized), which is an alkaline-tolerant variant of pZ (pZ(Alkaline stabilized)) and is shown in SEQ ID NO: 35, in which all lysine residues in the sequence have been replaced with arginine residues.

[0079] (3) Linker: A "linker" (0103) refers to a peptide chain that mediates the bond between activated ProMTG and the antibody-binding protein in an MTG fusion. The linker alleviates steric hindrance caused by the direct fusion of activated ProMTG and the antibody-binding protein, and may have the function of improving the cross-linking reaction between the target antibody fragment and the substrate molecule by increasing the flexibility of activated ProMTG, which has cross-linking catalytic activity, when the MTG fusion binds to the target antibody fragment via the antibody-binding protein.

[0080] The length or amino acid sequence of the linker is not limited. For example, it may be composed of 1 to 50, 4 to 40, 6 to 30, or 8 to 20 amino acid residues. The type of linker is also not particularly limited. Various linkers that link two peptide fragments are known in the art, and any of them can be used.

[0081] Specific examples include a peptide linker containing GGGGS (SEQ ID NO: 32) as one unit (often referred to herein as "G4S"), a linker in which 2 to 5 such units are linked (for example, a linker in which three units as shown in SEQ ID NO: 11 are linked is often referred to as "(G4S)3"), the thrombin linker shown in SEQ ID NO: 12 (GGGSLVPRGSGGGS), the polyproline linker shown in SEQ ID NO: 37 (PPPP), the IgA linker shown in SEQ ID NO: 38 (AAASPSTPPTPSPSTPPA), and the HAtag linker shown in SEQ ID NO: 39 (YPYDVPDYA). Note that by increasing the length of the linker to a certain extent, it is possible to facilitate crosslinking of two molecules of the substrate to the target antibody fragment, whereas by shortening the length of the linker, it is possible to more easily inhibit crosslinking of two molecules of the substrate to the target antibody fragment.

[0082] Linkers containing one or more negatively charged amino acid residues (D: aspartic acid or E: glutamic acid), such as the HAtag linker, are particularly preferred because they can enhance the antibody modification rate. Such linkers may also be mutant linkers in which some of the amino acid residues in known linkers have been substituted with negatively charged amino acid residues. Examples include the mutant HAtag linkers shown in SEQ ID NOs: 40 to 46, 22, and 23, in which one or more of the two amino acid residues from the N-terminus of the HAtag linker shown in SEQ ID NO: 39 (the tyrosine residue at position 1 and the proline residue at position 2) have been substituted with aspartic acid or glutamic acid.

[0083] 1-3-2. Structure of MTG fusion protein The MTG fusion protein may have a structure in which active ProMTG and an antibody-binding protein are directly linked by a peptide bond, or may have a structure in which the two proteins are indirectly linked via a linker.

[0084] In either the direct or indirect linkage, the order in which activated ProMTG and the antibody-binding protein are linked is not important. For example, activated ProMTG may be located at the N-terminus of the MTG fusion, and the antibody-binding protein at the C-terminus. Conversely, the antibody-binding protein may be located at the N-terminus of the MTG fusion, and activated ProMTG at the C-terminus. Depending on this order of placement, the orientation of activated ProMTG relative to the antibody fragment in the MTG fusion may change, and as a result, the modification target site on the antibody fragment (the position of the cross-linking amino acid residue) may also change.

[0085] For example, when the antibody-binding protein is pG, an MTG fusion in which the sequence of NH2-active ProMTG-pG-COOH is arranged (often referred to herein as "active ProMTG-pG") differs from an MTG fusion in which the sequence of active ProMTG and pG is reversed and the sequence is NH2-pG-active ProMTG-COOH (often referred to herein as "pG-active ProMTG") in which the sequence is reversed.

[0086] Specifically, for example, active ProMTG-pG(Fab-ΔKQ) selectively modifies the lysine residue at position 65 (K65) in the CDR2 (Complementarity Determining Region) in the variable region of the IgG heavy chain Fab region. However, pG(Fab-ΔKQ)-active ProMTG selectively modifies the lysine residue at position 225 (K225) in the heavy chain hinge region. This selective modification enables labeling at specific positions on antibody fragments. Note that both K65 and K225 are modification positions in the native full-length IgG sequence that are modified by the MTG fusion complex. However, the modification positions may vary in the case of antibody fragments such as Fab and (Fab')2 or when modifying antibodies other than IgG. In this specification, even in such cases, the lysine residues corresponding to these two modification positions will be referred to as K65 and K225, respectively.

[0087] The MTG fusion product may be a peptide derived from a natural product or may be obtained by artificial chemical synthesis, and is not limited thereto.

[0088] Proteins derived from natural products may be obtained directly from natural products by known recovery and purification methods, or may be obtained by inserting a gene encoding the protein into various expression vectors or the like using known gene recombination techniques, introducing the vector into cells, allowing the protein to be expressed, and then recovering and purifying the protein by known recovery and purification methods. Alternatively, the protein may be obtained from commercially available kits, such as the PROTEIOS Reagent Kit. TM The protein may be produced in a cell-free protein synthesis system using a system such as TnT (registered trademark) System (Promega) or the like, and then recovered and purified by known methods, without any particular limitation.

[0089] Chemically synthesized proteins can be obtained using known protein synthesis methods. Examples of synthesis methods include the azide method, acid chloride method, acid anhydride method, mixed acid anhydride method, DCC method, active ester method, carboimidazole method, and oxidation-reduction method. Both solid-phase synthesis and liquid-phase synthesis can be used for the synthesis. A commercially available protein synthesizer may also be used. After the synthesis reaction, the protein can be purified by a combination of known purification methods such as chromatography.

[0090] Furthermore, the MTG fusion product may contain its derivatives, which means all derivatives that can be prepared from the protein, such as those in which some of the constituent amino acids have been substituted with unnatural amino acids or those in which some of the constituent amino acids (mainly their side chains) have been chemically modified.

[0091] The MTG fusion may also comprise a salt of the protein and / or derivative. Physiologically acceptable acid addition salts or base salts are preferred. Acid addition salts include salts with inorganic acids such as hydrochloric acid, phosphoric acid, hydrobromic acid, and sulfuric acid, and salts with organic acids such as acetic acid, formic acid, propionic acid, fumaric acid, maleic acid, succinic acid, tartaric acid, citric acid, malic acid, oxalic acid, benzoic acid, methanesulfonic acid, and benzenesulfonic acid. Base salts include salts with inorganic bases such as sodium hydroxide, potassium hydroxide, ammonium hydroxide, and magnesium hydroxide, and salts with organic bases such as caffeine, piperidine, trimethylamine, and pyridine. Salts can be prepared using an appropriate acid such as hydrochloric acid or an appropriate base such as sodium hydroxide. For example, salts can be prepared by treating the protein or derivative in water or in an inert, water-miscible organic solvent such as methanol, ethanol, or dioxane using standard protocols.

[0092] 1-3-3. Nucleic Acid Encoding MTG Fusion A nucleic acid encoding an MTG fusion (often referred to herein as a "transglutaminase fusion gene (TG fusion gene)") is a recombinant nucleic acid molecule constructed by genetic engineering. This TG fusion gene is essentially composed of natural nucleic acids. "Natural nucleic acids" refer to naturally occurring biopolymers whose structural units are nucleotides linked by phosphodiester bonds. Typically, this refers to RNA linked to ribonucleotides containing any of the bases adenine, guanine, cytosine, and uracil, and / or DNA linked to deoxyribonucleotides containing any of the bases adenine, guanine, cytosine, and thymine. The TG fusion gene may be composed of DNA, RNA, or a combination thereof.

[0093] The specific nucleotide sequence of the TG fusion gene is not limited as long as it encodes the MTG fusion described in aspect 1. MTG fusions include active ProMTG and antibody-binding proteins, and examples of nucleotide sequences encoding wild-type ProMTG include: (a) the nucleotide sequence shown in SEQ ID NO: 2, which encodes the amino acid sequence shown in SEQ ID NO: 1; (b) a nucleotide sequence in which one or more nucleotides have been added, deleted, and / or substituted at positions other than positions 27-29 and 33-35 of the nucleotide sequence shown in SEQ ID NO: 2; (c) a nucleotide sequence that shares 95% or more nucleotide identity with the nucleotide sequence shown in SEQ ID NO: 2; and (d) a nucleotide sequence that hybridizes under highly stringent conditions with a nucleic acid fragment consisting of a nucleotide sequence complementary to the nucleotide sequence shown in SEQ ID NO: 2. Examples of nucleotide sequences encoding wild-type pG include (e) the nucleotide sequence shown in SEQ ID NO: 4, which encodes the amino acid sequence shown in SEQ ID NO: 3, (f) the nucleotide sequence shown in SEQ ID NO: 4 in which one or more nucleotides have been added, deleted, and / or substituted, (g) a nucleotide sequence having 95% or more nucleotide identity with the nucleotide sequence shown in SEQ ID NO: 4, and (h) a nucleotide sequence that hybridizes under highly stringent conditions with a nucleic acid fragment consisting of a nucleotide sequence complementary to the nucleotide sequence shown in SEQ ID NO: 4. Examples of nucleotide sequences encoding wild-type pA include (i) the nucleotide sequence shown in SEQ ID NO: 6, which encodes the amino acid sequence shown in SEQ ID NO: 5, (j) the nucleotide sequence shown in SEQ ID NO: 6 in which one or more nucleotides have been added, deleted, and / or substituted, (k) a nucleotide sequence having 95% or more nucleotide identity with the nucleotide sequence shown in SEQ ID NO: 6, and (l) a nucleotide sequence that hybridizes under highly stringent conditions with a nucleic acid fragment consisting of a nucleotide sequence complementary to the nucleotide sequence shown in SEQ ID NO: 6. Here, "highly stringent conditions" refers to high temperature and low salt conditions under which nonspecific hybrids are not formed. For example, post-hybridization washing conditions are 60°C to 68°C and 1×SSC or less, preferably 65°C to 70°C and 0.1×SSC or less. Here, 1×SSC may contain 150 mM sodium chloride and 15 mM sodium citrate.

[0094] The TG fusion gene may be labeled with a nucleic acid labeling substance at the phosphate group, sugar, and / or base, as necessary. Any substance known in the art can be used as the nucleic acid labeling substance.

[0095] The TG fusion gene can be prepared by cloning techniques. For example, but not limited to, a DNA fragment containing the base sequence shown in SEQ ID NO: 2 encoding wild-type ProMTG and a DNA fragment containing the base sequence shown in SEQ ID NO: 4 encoding wild-type pG can be ligated in a cloning vector, and then a mutation can be introduced at a predetermined site using a site-directed mutagenesis method such as the Kunkel method or the Gapped duplex method. The introduction of a base mutation can be performed using a commercially available site-directed mutagenesis kit. Examples of such kits include the commercially available QuickChange Site-Directed Mutagenesis Kit (Agilent Technology) and GeneTailor TM Site-Directed Mutagenesis System (Thermo Fisher Scientific), Mutan TM -Super Express Km (Takara Bio Inc.). Alternatively, the vector can be prepared by a known nucleic acid amplification method such as PCR using primers designed to introduce the desired missense mutation, with DNA comprising the nucleotide sequence shown in SEQ ID NO: 2 encoding wild-type ProMTG and DNA comprising the nucleotide sequence shown in SEQ ID NO: 4 encoding wild-type pG as templates. These cloning techniques are well known in the art, and reference may be made to the techniques described in, for example, Green & Sambrook, Molecular Cloning, A Laboratory Manual 4th ed., Cold Spring Harbor Laboratory Press (2012), Current Protocols in Molecular Biology, Wiley online library (2023), etc.

[0096] 1-4. Constitution of the antibody fragment The specific constitution of the antibody fragment used in the method for producing an antibody fragment substrate complex of the present invention will be explained using examples.

[0097] As described above, the term "antibody fragment" as used herein refers to a portion of an immunoglobulin that contains at least one variable region having antigen-binding activity and that contains at least one modification target site (e.g., K65 or K225). Therefore, the term "antibody fragment" includes fragment antibodies and synthetic antibodies. Furthermore, the term "antibody fragment" as used herein also encompasses single-chain antibodies, such as VHH antibodies, which are natural antibodies but lack L chains and are composed only of the VH region of an H chain.

[0098] The term "fragment antibody" refers to a fragment obtained by digesting immunoglobulins such as IgG or IgM with a protease or reducing agent. For example, of the Fab fragments and Fc fragments produced when IgG is digested with a cysteine ​​protease such as papain, the Fab fragment satisfies the above-mentioned conditions for an antibody fragment. Also included is the (Fab')2 fragment produced when IgG is digested with an aspartic acid protease such as pepsin. Other fragments that fall within this category include Fv fragments and Fab'. Also included is the so-called rIgG fragment produced when disulfide bonds of IgG are cleaved with a reducing agent such as 2-mercaptoethanol.

[0099] The term "synthetic antibody" refers to an antibody consisting of a portion of an immunoglobulin artificially constructed using chemical or recombinant DNA techniques. For example, a monomeric polypeptide molecule in which the VL and VH of a specific antibody are artificially linked via a linker peptide or the like is included. A specific example of such a polypeptide is a single-chain Fv (scFv: single chain fragment of variable region). This scFv is a monovalent antibody fragment. Furthermore, multimeric polypeptide molecules based on the scFv structure are also included. Examples include diabodies, triabodies, and tetrabodies, which have dimeric to tetrameric scFv structures, respectively. These synthetic antibodies are bivalent to tetravalent antibody fragments, respectively.

[0100] A "single-chain antibody" (VHH antibody) is an antibody produced by animals of the Camelidae family, which is composed of a single polypeptide chain consisting of only the VH region of the H chain, without any L chain, and is a monovalent antibody that can recognize and bind to a target substance by itself. Single-chain antibodies are natural antibodies and are full-length, so they are not fragments. However, since single-chain antibodies are composed only of VH, unlike the basic structure of immunoglobulins which have VL and VH, they are considered to be included in the scope of antibody fragments in this specification.

[0101] 1-5. Processes A flow diagram of the antibody fragment-substrate complex production method of the present invention is shown in Figure 2. As shown in this figure, the production method of the present invention includes an antibody fragmentation step (S0201), an antibody fragment purification step (S0202), an enzyme binding step (S0203), a substrate molecule modification step (S0204), a crosslinking step (S0205), and an enzyme separation step (S206). Of these, the enzyme binding step (S0203), the crosslinking step (S0205), and the enzyme separation step (S0206) are essential steps, while the antibody fragmentation step (S0201), the antibody fragment purification step (S0202), and the substrate molecule modification step (S0204) are optional steps. Each step will be described in detail below.

[0102] (1) Antibody Fragmentation Step The "antibody fragmentation step" (S0201) is a step in which immunoglobulin is cleaved at a predetermined site with a protease to obtain antibody fragments containing a modification target site. This step is carried out prior to the subsequent enzyme binding step (S202). As described above, this step is a selection step, and may be carried out as needed when the antibody fragment to be used in the subsequent enzyme binding step (S202) has not yet been prepared.

[0103] In this step, the desired antibody fragment is prepared by cleaving immunoglobulins, etc., using a protease. Any type of immunoglobulin may be cleaved, but IgG is preferred because it is easily available and easy to manipulate.

[0104] There are no limitations on the type of protease used in this step, but cysteine ​​protease or aspartic acid protease is preferred.

[0105] Cysteine ​​proteases are proteases that use the thiol group of cysteine ​​as a nucleophilic group. Specific examples of cysteine ​​proteases include papain, bromelain, caspase, cathepsin B, and cathepsin K. When IgG is treated with a cysteine ​​protease, such as papain, in this step, the IgG hinge region disulfide bond is cleaved to generate two Fab fragments and one Fc fragment. Of these, the Fab fragment contains one variable region with antigen-binding activity and includes a modification target site, and corresponds to the antibody fragment referred to herein.

[0106] Aspartic acid proteases are proteases that catalyze the cleavage of peptide substrates using activated water molecules bound to aspartic acid residues. Specific examples of aspartic acid proteases include pepsin, renin, and cathepsin D. When IgG is treated with an aspartic acid protease, such as pepsin, in this step, it cleaves below the disulfide bond in the hinge region of the IgG, generating a single F(ab')2 fragment. Meanwhile, the Fc region is extensively fragmented into numerous small peptides. In this case, the F(ab')2 fragment contains two variable regions with antigen-binding activity and includes a modification target site, and corresponds to the antibody fragment defined herein.

[0107] The reaction of cleaving immunoglobulins with proteases may be carried out according to methods known in the art. For example, IgG and the protease may be mixed in a buffer with an optimal pH for the protease to be used, and then treated at an appropriate temperature for a certain period of time. Specifically, when pepsin is used as the protease, IgG and pepsin may be mixed in a strongly acidic buffer of pH 2.0 to pH 4.5, for example, and then treated at 37°C for approximately 10 to 60 minutes. When papain is used as the protease, IgG and papain may be mixed in a weakly acidic to neutral buffer of pH 6.0 to pH 7.0 containing cysteine ​​as an activator, for example, and then treated at 37°C for approximately 10 to 60 minutes.

[0108] After the cleavage reaction, treatment may be carried out under conditions that stop the enzyme reaction. For example, in the case of pepsin, the reaction solution may be made neutral to alkaline for reduction, and in the case of papain, a protease inhibitor such as leupeptin may be added.

[0109] (2) Antibody Fragment Purification Step The "antibody fragment purification step" (S0202) is a step of separating and purifying antibody fragments containing the target site modified by the MTG fusion product from the antibody fragments obtained after the antibody fragmentation step (S0201).

[0110] The antibody fragmentation step (S0201) produces at least two types of antibody fragments in the reaction solution. For example, fragmenting IgG with papain produces two Fab fragments and one Fc fragment. Of these, antibody fragments that do not contain the target site for modification by the MTG fusion (Fc fragments in the above example) are contaminants that are not required for subsequent steps and may also act as reaction inhibitors in the enzyme binding step (S0203) described below. Therefore, this step is intended to separate and purify antibody fragments that contain the target site for modification by the MTG fusion (Fab fragments in the above example) from the reaction solution after the antibody fragmentation step (S0201), which are necessary for subsequent steps, and is selected as needed.

[0111] The method for selectively separating and purifying antibody fragments containing the modification target site from the reaction solution after the antibody fragmentation step (S0201) is not limited. Publications known in the art may be used. For example, if the antibody fragment to be purified is a Fab fragment, a Protein L column that specifically binds to Fab can be used to remove Fc and enzymes such as papain used in fragmentation. Further purification can then be performed using size exclusion chromatography (SEC) or other methods. Generally, the molecular weight and isoelectric point of the desired antibody fragment (e.g., Fab fragment) differ from those of the unwanted antibody fragment (e.g., Fc fragment) and enzymes. Therefore, size exclusion chromatography based on molecular weight and ion exchange chromatography based on isoelectric point are effective purification methods.

[0112] (3) Enzyme binding step The "enzyme binding step" (S0203) is a step of binding an MTG fusion molecule to an antibody fragment to prepare an enzyme-antibody fragment complex.

[0113] This step, together with the cross-linking step (S204) described below, is the most characteristic and important step in the production method of the present invention. That is, instead of producing an antibody fragment-substrate conjugate by treating an antibody-substrate conjugate with a protease, the antibody fragment-substrate conjugate is produced by cross-linking a substrate molecule to an antibody fragment, which makes it possible to obtain a homogeneous antibody fragment-substrate conjugate in a higher yield than one in which the number and binding positions of substrate molecules are controlled.

[0114] The antibody fragment used in this step is a polypeptide fragment to be modified with a substrate molecule, as described in detail above in the section "1-4. Construction of antibody fragments." Although not limited thereto, antibody fragments that can specifically recognize and bind to antigens present on the surface of target substances such as cancer cells are particularly preferred.

[0115] The MTG fusion product used in this step is selected appropriately depending on the modification site of the antibody fragment. Specifically, when modifying a specific lysine residue in an antibody fragment, for example, when modifying the lysine residue corresponding to K65 in the heavy chain of IgG in Fab or (Fab')2, an MTG fusion product in which active ProMTG is located at the N-terminus, such as active ProMTG-pG, is used. Specific examples of such MTG fusion products include Pro(K10R / Y12A)MTG-pG, Pro(K10R / Y12A)MTG-pG(Fab-ΔKQ), and Pro(K10R / Y12A)MTG-pG(Fc-ΔKx5).

[0116] Furthermore, when modifying the lysine residue corresponding to K225 in the heavy chain of IgG in Fab or (Fab')2, it is preferable to use an MTG fusion in which active ProMTG is located at the C-terminus, such as pG-active ProMTG. Specific examples of such MTG fusions include pG-Pro(K10R / Y12A)MTG, pG(Fab-ΔKQ)-Pro(K10R / Y12A)MTG, and pG(Fc-ΔKx5)-Pro(K10R / Y12A)MTG.

[0117] Therefore, the type of MTG fusion product used in this step can be determined appropriately depending on the modification positions and number of modifications on the antibody fragment, and is not particularly limited. When selectively modifying multiple amino acid residues in an antibody fragment, multiple types of MTG fusion products with different target amino acid residue positions can be used in combination.

[0118] The binding of an antibody fragment and an MTG fusion product is not particularly limited as long as it brings them into contact and forms an enzyme-antibody complex. The binding of an antibody fragment and an MTG fusion product is, in principle, carried out in a reaction solution. The composition of the reaction solution is not particularly limited as long as it maintains the activity of active MTG, which is the cross-linking catalytic domain of the MTG fusion product. It is usually sufficient to mix the antibody fragment and the MTG fusion product in a buffer such as PBS (pH 7.4). The reaction conditions are also not particularly limited as long as they allow the antibody fragment to bind to the MTG fusion product and form an antibody-enzyme complex. For example, the antibody fragment and the MTG fusion product may be mixed in a buffer and then allowed to stand at 4 to 40°C, 10 to 39°C, 15 to 39°C, 18 to 38°C, 20 to 38°C, 23 to 38°C, 25 to 37°C, 27 to 37°C, or 30 to 36°C for 5 minutes to 1 hour, 10 to 40 minutes, or 20 to 30 minutes.

[0119] (4) Substrate Molecule Modification Step The "substrate molecule modification step" (S0204) is a step of pre-modifying the substrate molecule prior to the cross-linking step (S0205) described below.

[0120] The substrate molecule must contain a reactive site for the cross-linking reaction using the MTG fusion. However, if the substrate molecule is not a (poly)peptide containing a glutamine residue, which serves as the reactive site, the substrate molecule cannot be cross-linked with the antibody fragment. Therefore, the purpose of this step is to modify the substrate molecule with a chemical reactive group, such as a glutamine residue or a peptide tag (glutamine tag) containing a glutamine residue, to provide a reactive site to the substrate molecule. Therefore, this is a selection step, and is not necessarily required if the substrate molecule already contains a reactive site. Furthermore, since this step can be performed independently of the enzyme binding step (S0203), the order of this step relative to the enzyme binding step (S0203) is not important.

[0121] The substrate molecule used in this step is crosslinked to the antibody by the catalytic activity of MTG during the production of the antibody fragment-substrate complex. The type of substrate molecule is not particularly limited. As described above, it may be appropriately selected from nucleic acids, (poly)peptides, sugar chains, lipids, low-molecular-weight compounds, or combinations thereof, depending on the purpose of the antibody-substrate complex to be produced.

[0122] Specific examples of low molecular weight compounds include, but are not limited to, drugs, fluorescent substances, luminescent substances, enzymes, and the like.

[0123] For example, if the goal is to produce an antibody fragment drug conjugate (AFDC), the substrate molecule can be a drug such as an anti-cancer drug.

[0124] The type of anticancer drug is not limited, and examples thereof include paclitaxel, doxorubicin, daunorubicin, cyclophosphamide, methotrexate, 5-fluorouracil, thiotepa, busulfan, improsulfan, piposulfan, benzodopa, carboquone, meturedopa, uredopa, altretamine, triethylenemelamine, triethylenephosphoramide, and triethylenethiophosphoramide. ramide), trimethylolomelamine, bullatacin, bullatacinone, camptothecin, bryostatin, callystatin, cryptophycin 1, cryptophycin 8, dolastatin, duocarmycin, eleutherobin, pancratistatin, sarcodictyin, spongistatin, chlorambucil, chloRNAphazine, colofsphamide, estramustine , ifosfamide, mechlorethamine, mechlorethamine oxide hydrochloride, melphalan, novembichin, phenesterine, prednimustine, trofosfamide, uracil mustard, carmustine, chlorozotocin, fotemustine, lomustine, nimustine, ranimustine, calicheamicin, dynemicin , clodronate, esperamicin, aclacinomycin, actinomycin, authramycin, azaserine, bleomycin, cactinomycin, carabicin, carminomycin, carzinophilin, chromomycin, dactinomycin, detorbicin, 6-diazo-5-oxo-L-norleucine, adriamycin, epirubicin, esorubicin, idarubicin,Marcellomycin, mitomycin C, mycophenolic acid, nogalamycin, olivomycins, peplomycin, potfiromycin, puromycin, quelamycin, rodorubicin, streptonigrin, streptozocin, tubercidin, ubenimex, zinostatin, zorubicin, denopterin, pteropterin, trimetrexate, fludarabine Fludarabine, 6-mercaptopurine, thiamiprine, thioguanine, ancitabine, azacitidine, 6-azauridine, carmofur, cytarabine, dideoxyuridine, doxifluridine, enocitabine, floxuridine, androgens, calusterone, dromostanolone propionate, epithiostanol, mepitiostane, testolactone, aminoglutethimide, mitotane, trilostane, florinic acid acid), aceglatone, aldophosphamide glycoside, aminolevulinic acid, eniluracil, amsacrine, bestrabucil, bisantrene, edatraxate, defofamine, demecolcine, diaziquone, elfornithine, elliptinium acetate, epothilone, etoglucid, lentinan, lonidamine, maytansine, ansamitocin, mitoguazone, mitoxantrone, mopidanmol,Nitraerine, pentostatin, phenamet, pirarubicin, losoxantrone, podophyllinic acid, 2-ethylhydrazide, procarbazine, razoxane, rhizoxin, sizofiran, spirogermanium, tenuazonic acid acid), triaziquone, roridine A, anguidine, urethane, vindesine, dacarbazine, mannomustine, mitobronitol, mitolactol, pipobroman, gacytosine, doxetaxel, chlorambucil, gemcitabine, 6-thioguanine, mercaptopurine, cisplatin, oxaliplatin, carboplatin, vinbrain Examples of such inhibitors include cyclosporine, etoposide, ifosfamide, mitoxantrone, vincristine, vinorelbine, novantrone, teniposide, edatrexate, daunomycin, aminopterin, xeloda, ibandronate, irinotecan, topoisomerase inhibitors, difluoromethylolnithine (DMFO), retinoic acid, and capecitabine, pyrrolobenzodiazepine (PBD) dimer, ravtansine, monomethyl auristatin E (MMAE), monomethyl auristatin F (MMAF), mertansine (DM1), tesirine, and trastuzumab (Herceptin).

[0125] Furthermore, if the antibody fragment-substrate complex is intended to be used as a sensing tool, the substrate molecule may be a fluorescent substance, a luminescent substance, or an enzyme.

[0126] As used herein, the term "fluorescent substance" refers to a substance that emits fluorescence of a specific wavelength when irradiated with excitation light of a specific wavelength, and examples thereof include fluorescent proteins and fluorescent dyes. Examples of fluorescent proteins include GFP (including derivatives such as EGFP), CFP, RFP, DsRed, YFP, PE, PerCP, APC, Achilles, CU17S, StayGold, oxStayGold, QC2-6, and mStayGold. Examples of "fluorescent dyes" include TAMRA, FITC, Texas, Cy3, Cy5, Cy7, FAM, HEX, VIC, JOE, Rox, TET, Bodipy493, and NBD.

[0127] As used herein, the term "luminescent substance" refers to a substance that emits energy as light when returning from an excited state to a ground state due to a chemical reaction. Specific examples include chemiluminescent substances such as luminol, lophin, and lucigenin, and bioluminescent substances such as luciferin. Bioluminescent substances become excited by a biochemical reaction catalyzed by a luminescent enzyme.

[0128] As used herein, the term "enzyme" is not particularly limited. Various types of enzymes are known, including oxidoreductases, transferases, hydrolases, lyases, isomerases, and synthases, and any of these may be used. When an antibody fragment-substrate complex is used as the sensing tool, a luminescent enzyme that uses the luminescent substance as a substrate is preferred. A "luminescent enzyme" (luciferase) refers to an enzyme that catalyzes the chemiluminescent reaction of luciferin, a luminescent substance in bioluminescence. The type of luminescent enzyme used herein is not important. Various types of luminescent enzymes have been isolated and identified from many luminescent organisms, and many artificial derivatives based on them have also been developed. However, the present invention does not limit the type of luminescent enzyme as long as it can catalyze the luminescent reaction of the luciferin used. Natural luciferases include, for example, firefly luciferase, Renilla luciferase, and Gaussia luciferase. Non-natural luciferases include, for example, AkaLuc, NanoLuc (registered trademark), and ALuc (registered trademark).

[0129] The length of the amino acids in the glutamine tag is not particularly limited as long as the peptide can serve as a reactive substrate, and may be, for example, 2 to 40 amino acids, 3 to 38 amino acids, 4 to 35 amino acids, 5 to 33 amino acids, 6 to 30 amino acids, 7 to 28 amino acids, 8 to 26 amino acids, 9 to 24 amino acids, 10 to 22 amino acids, or 12 to 20 amino acids.

[0130] The specific amino acid sequence of the glutamine tag is not particularly limited as long as it contains a glutamine residue that serves as a reactive site for the MTG fusion and is a peptide that can serve as a reactive substrate. Specific examples include, but are not limited to, glutamine tags having amino acid sequences containing YPLQMRG (SEQ ID NO: 13) and (SEQ ID NO: 47) that contain glutamine residues.

[0131] Furthermore, when a hydrophobic substance is added to the glutamine tag, polyethylene glycol (PEG) can be linked to enhance hydrophilicity. Examples include compounds in which PEG is linked to a glutamine tag containing the amino acid sequence shown in SEQ ID NO: 13 or SEQ ID NO: 47. Specific examples include NH2-(PEG)n-YPLQMRG shown in formula (I) below, NH2-(PEG)n-LLQG shown in formula (II), or salts thereof. (wherein n is an integer of 1 or more). (wherein n is an integer of 1 or more).

[0132] For example, when n=4 in formula (I) and formula (II), examples include NH2-PEG4-YPLQMRG shown in the following formula (III) and NH2-PEG4-LLQG shown in formula (IV), respectively, or salts thereof.

[0133] The glutamine tag may have an azide group, or an alkynyl or cycloalkynyl group attached, such as a click reaction element.

[0134] As used herein, the term "click reaction element" refers to a reaction substrate that can undergo a click reaction with an azide group without the need for a copper catalyst. Examples include OCT (cyclooctyne), MFCO (monofluorinated cyclooctyne), DFCO (difluorinated cyclooctyne), DBCO (dibenzocyclooctyne), BIARC (biarylazacycloocthionone), BC (bicyclononyne), and DACN (diazacyclononyne).

[0135] Examples of glutamine tags to which a click reaction element has been added (click reaction element-conjugated glutamine tags) include DBCO-PEG4-YPLQMRG shown in the following formula (V) and DBCO-PEG4-LLQG shown in formula (VI).

[0136] The method for modifying the substrate molecule with a glutamine tag is not particularly limited and may be based on a method known in the art depending on the type of substrate molecule.

[0137] When the substrate molecule is a low molecular weight compound such as FAM or TAMRA, it can be bound to the N-terminal amino group of the glutamine tag via an amide group. When the substrate molecule is a nucleic acid, the carboxylic acid of the glutamine tag can be bound to a primary amine-modified nucleic acid via an amide bond. When an alkynyl or cycloalkynyl group is added to the glutamine tag, the glutamine tag and the substrate molecule can also be bound by a click reaction with a substrate molecule to which an azide group has been added. Alternatively, when an azide group is added to the glutamine tag, it can be similarly bound by a click reaction with a substrate molecule to which an alkynyl or cycloalkynyl group has been added.

[0138] If the substrate molecule is a (poly)peptide, a glutamine residue can be added to a position (e.g., the N-terminus or C-terminus) that does not inhibit the activity of the (poly)peptide. Modification can also be performed by contract synthesis at a life science manufacturer.

[0139] (5) Crosslinking Step The "crosslinking step" (S0205) is a step of mixing the enzyme-antibody fragment complex with a substrate molecule containing a glutamine residue, and crosslinking the substrate molecule to a predetermined amino acid residue of the antibody fragment by the catalytic activity of the enzyme in the enzyme-antibody fragment complex, i.e., the TMG fusion complex, to prepare an enzyme-antibody fragment-substrate complex.

[0140] The substrate molecule used in this step can be appropriately selected based on the enzyme-antibody fragment complex obtained after the enzyme binding step (S0203). In the enzyme-antibody fragment complex, when lysine residues on the antibody fragment are modified, the position of the lysine residue to be modified is determined by the type of MTG fusion compound bound to the antibody fragment. MTG catalyzes an acyl transfer reaction between glutamine and lysine residues, cross-linking the two amino acid residues. The position of the antibody fragment to be modified can be determined in advance by determining the desired position to cross-link the substrate molecule.

[0141] Specifically, for example, when the antibody fragment is a Fab fragment or a (Fab')2 fragment and the lysine residue corresponding to K65 or K225 in the IgG heavy chain is to be modified, by using a substrate molecule containing the glutamine tag shown in SEQ ID NO: 13, the catalytic activity of the MTG fusion product can crosslink the lysine residue (K65 or K225) of the antibody fragment with the glutamine residue at position 4 (Q4) of the glutamine tag shown in SEQ ID NO: 13.

[0142] In this step, even after the cross-linking reaction between the antibody fragment and the substrate molecule is completed, the MTG fusion molecule remains bound to the antibody fragment via the antibody-binding domain, and an enzyme-antibody fragment-substrate complex is formed in which the three molecules are associated via the antibody fragment.

[0143] (6) Enzyme Separation Step The "enzyme separation step" (S0206) is a step in which the MTG fusion is separated from the enzyme-antibody fragment-substrate complex obtained after the crosslinking step (S0205) to prepare the target antibody fragment substrate complex. As described above, the antibody fragment substrate complex, which is the final product of the present invention, is produced after the crosslinking step. However, the MTG fusion used in the crosslinking reaction between the antibody fragment and the substrate molecule remains bound to the antibody fragment site of the antibody fragment substrate complex. The MTG fusion after the crosslinking step is no longer needed for the complex. Therefore, the purpose of this step is to separate and remove the MTG fusion from the enzyme-antibody fragment substrate complex.

[0144] The method for separating the MTG fusion from the enzyme-antibody fragment-substrate complex is not limited. For example, pG and the antibody fragment may be separated by contacting the antibody fragment with a low- or high-pH solution that induces a reversible structural change in the antibody fragment. Specific examples include exchanging the reaction solution used in the crosslinking step for a glycine buffer or citrate buffer solution with a pH of 2.5 to 4.0, or adding a large amount of citric acid to the reaction solution.

[0145] The antibody fragment substrate complex obtained after this step may be further purified. After this step, the reaction mixture contains contaminants in addition to the antibody fragment substrate complex. The desired antibody fragment substrate complex may be separated and purified from this mixed reaction mixture. The method for purifying the antibody fragment substrate complex from the reaction mixture is not limited. Publications known in the art may be used. For example, the antibody fragment substrate complex can be isolated using chromatography or ultrafiltration. Generally, because the antibody fragment and the MTG fusion product differ in both molecular weight and isoelectric point, size exclusion chromatography based on molecular weight and ion exchange chromatography based on isoelectric point are effective.

[0146] 2. Method for producing an enzyme-antibody fragment conjugate 2-1. Overview A second aspect of the present invention is a method for producing an enzyme-antibody fragment conjugate. In the production method of the present invention, an enzyme-antibody fragment conjugate for crosslinking any substrate molecule containing a glutamine residue to an antibody fragment can be obtained by the steps up to the enzyme binding step of the first aspect.

[0147] 2-2. Steps A flow diagram of the method for producing an enzyme-antibody fragment conjugate of the present invention is shown in Figure 20. As shown in this figure, the production method of the present invention comprises an antibody fragmentation step (S2001), an antibody fragment purification step (S2002), and an enzyme binding step (S2003). Of these, the enzyme binding step (S2003) is an essential step, while the antibody fragmentation step (S2001) and the antibody fragment purification step (S2002) are optional steps. Detailed explanations of each of these steps are the same as those for the corresponding steps in the method for producing an antibody fragment substrate conjugate described in the first aspect. Therefore, explanations of each step here will be omitted.

[0148] 2-3. Effects The method for producing an enzyme-antibody fragment conjugate of the present invention can obtain an enzyme-antibody fragment conjugate formed by binding an MTG fusion product to an antibody fragment. The obtained enzyme-antibody fragment conjugate can crosslink any substrate molecule containing a glutamine residue to a specific lysine residue in the antibody fragment.

[0149] Example 1 Preparation of MTG Fusion Protein (Purpose) An MTG fusion protein to be used in the method for producing an antibody fragment substrate conjugate of the present invention is prepared.

[0150] (Method) 1. Preparation of Pro(K10R / Y12A)MTG-(G4S)3-pG Pro(K10R / Y12A)MTG-(G4S)3-pG was prepared by fusing the antibody-binding protein Protein G(pG) (C2 domain) to the C-terminus of active ProMTG, Pro(K10R / Y12A)MTG, via the linker (G4S)3.

[0151] Plasmid pET22b (+)_Pro(K10R / Y12A)-MTG-(G4S)3-pZ encoding Pro(K10R / Y12A)-MTG-(G4S)3 Ca2+A vector DNA fragment was prepared by PCR using the template pET22b(+)_H6-pG2 with the Fw primer consisting of the nucleotide sequence shown in SEQ ID NO: 14 and the Rv primer consisting of the nucleotide sequence shown in SEQ ID NO: 15. The PCR reaction mixture consisted of 12.5 μL of PrimeSTAR® Max DNA Polymerase (Takara Bio Inc.), 0.75 μL each of 0.3 μM Fw / Rv primers, and 0.50 μL of template DNA, adjusted to 25 μL with MilliQ water. The PCR reaction conditions were (98°C for 10 seconds + 55°C for 5 seconds + 72°C for 35 seconds) x 35 cycles, followed by storage at 4°C. An insert DNA fragment was prepared using the template pET22b(+)_H6-pG2 encoding pG with the Fw primer consisting of the nucleotide sequence shown in SEQ ID NO: 16 and the Rv primer consisting of the nucleotide sequence shown in SEQ ID NO: 17 under the same conditions as the PCR described above.

[0152] After the PCR reaction, the vector DNA fragment and the insert DNA fragment were extracted and purified by standard methods, and the two fragments were ligated to construct the recombinant plasmid pET22b(+)Pro(K10R / Y12A)MTG-(G4S)3-pG.

[0153] Next, pET22b(+)Pro(K10R / Y12A)MTG-(G4S)3-pG was introduced into E. coli to obtain transformants, which were then cultured. 600 After the pH reached approximately 0.5, IPTG was added to a final concentration of 0.25 mM to induce expression. After overnight culture, the culture supernatant was collected and analyzed by HisTrap™. TM His-tag purification was performed using an FFcrude column (Cytiva). After purification, the yield and purity were confirmed by SDS-PAGE, and the product was purified using a HiLoad 16 / 600 Superdex column. TM Further purification was carried out using size exclusion chromatography (SEC) (Cytiva) using a 200 pg column to obtain the target protein.

[0154] 1a. Preparation of Pro(K10R / Y12A)MTG-G4S-pG Pro(K10R / Y12A)MTG and pG of the aforementioned Pro(K10R / Y12A)MTG-(G4S)3-pG were fused via the linker G4S to prepare Pro(K10R / Y12A)MTG-G4S-pG. The basic procedure was the same as that described in "1. Preparation of Pro(K10R / Y12A)MTG-(G4S)3-pG" above, and was prepared by a known method.

[0155] 2. Preparation of Pro(K10R / Y12A)MTG-(G4S)3-pG (Fab-ΔKQ) All lysine and arginine residues in the Fab-binding region of the pG domain of Pro(K10R / Y12A)MTG-(G4S)3-pG were replaced with arginine and asparagine residues, respectively, to prepare Pro(K10R / Y12A)MTG-(G4S)3-pG (Fab-ΔKQ), which eliminates the Fab-binding ability of the Fab-binding region.

[0156] A vector DNA fragment was prepared by PCR using the plasmid pET22b(+)_Pro(K10R / Y12A)MTG-(G4S)3-pG encoding Pro(K10R / Y12A)MTG-(G4S)3 as a template and an Fw primer consisting of the nucleotide sequence shown in SEQ ID NO: 24 and an Rv primer consisting of the nucleotide sequence shown in SEQ ID NO: 25.

[0157] The PCR reaction mixture consisted of 12.5 μL of PrimeSTAR Max DNA Polymerase (Takara Bio), 0.75 μL each of 0.3 μM Fw / Rv primers, and 0.50 μL of template DNA, adjusted to 25 μL with MilliQ water. The PCR conditions were (98°C for 10 seconds, 55°C for 5 seconds, and 72°C for 35 seconds) x 35 cycles, followed by storage at 4°C. The insert DNA fragment was prepared by PCR using pG(Fab-ΔKQ), synthesized by Integrated DNA Technologies, Inc., as a template, with the Fw primer consisting of the nucleotide sequence shown in SEQ ID NO:26 and the Rv primer consisting of the nucleotide sequence shown in SEQ ID NO:27. The PCR conditions were (98°C for 10 seconds, 55°C for 5 seconds, and 72°C for 5 seconds) x 35 cycles, followed by storage at 4°C.

[0158] After the PCR reaction, the vector DNA fragment and the insert DNA fragment were extracted and purified by standard methods, and the two fragments were ligated to construct the recombinant plasmid pET22b(+)_Pro(K10R / Y12A)MTG-(G4S)3-pG(Fab-ΔKQ).

[0159] Next, pET22b(+)_Pro(K10R / Y12A)MTG-(G4S)3-pG(Fab-ΔKQ) was introduced into E. coli to obtain transformants, which were then cultured and induced by adding IPTG to a final concentration of 0.1 mM. After overnight culture, the culture supernatant was collected and analyzed using HisTrap TM His-tag purification was performed using an FF crude column (Cytiva). After purification, the yield and purity were confirmed by SDS-PAGE, and the product was purified using a HiLoad 16 / 600 Superdex column. TM Further purification was carried out using size exclusion chromatography (SEC) (Cytiva) using a 200 pg column to obtain the target protein.

[0160] 3. Preparation of pG-G4S-Pro(K10R / Y12A)MTG The positions of Pro(K10R / Y12A)MTG and pG in the Pro(K10R / Y12A)MTG-(G4S)3-pG were swapped to prepare pG-G4S-Pro(K10R / Y12A)MTG, which was fused via the linker G4S. The basic procedure followed the method described above in "1. Preparation of Pro(K10R / Y12A)MTG-(G4S)3-pG." The primer pair used to prepare the vector DNA fragment consisted of an Fw primer with the nucleotide sequence shown in SEQ ID NO: 18 and an Rv primer with the nucleotide sequence shown in SEQ ID NO: 19. The primer pair used to prepare the insert DNA fragment consisted of an Fw primer with the nucleotide sequence shown in SEQ ID NO: 20 and an Rv primer with the nucleotide sequence shown in SEQ ID NO: 21.

[0161] 4. Preparation of pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG All lysine residues and arginine residues present in the Fab binding region of the pG domain of pG-G4S-Pro(K10R / Y12A)MTG prepared in 3 above were substituted with arginine residues and asparagine residues, respectively, to prepare pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG in which the Fab binding ability of the Fab binding region was lost.

[0162] A vector DNA fragment was generated by PCR using the plasmid pET22b(+)_pG-G4S-Pro(K10R / Y12A)-MTG encoding G4S-Pro(K10R / Y12A)-MTG as a template, with the Fw primer consisting of the nucleotide sequence shown in SEQ ID NO: 28 and the Rv primer consisting of the nucleotide sequence shown in SEQ ID NO: 29. The PCR reaction mixture consisted of 12.5 μL of PrimeSTAR Max DNA Polymerase (Takara Bio Inc.), 0.75 μL each of 0.3 μM Fw and Rv primers, and 0.50 μL of template DNA, adjusted to 25 μL with MilliQ water. The PCR reaction conditions were 35 cycles of (98°C for 10 seconds, 55°C for 5 seconds, and 72°C for 35 seconds), followed by storage at 4°C. An insert DNA fragment was prepared by PCR using pG(Fab-ΔKQ), synthesized by Integrated DNA Technologies Co., Ltd., as a template, an Fw primer consisting of the nucleotide sequence shown in SEQ ID NO: 30, and an Rv primer consisting of the nucleotide sequence shown in SEQ ID NO: 31. The PCR reaction conditions were (98°C for 10 seconds + 55°C for 5 seconds + 72°C for 5 seconds) x 35 cycles, followed by storage at 4°C.

[0163] After the PCR reaction, the vector DNA fragment and the insert DNA fragment were extracted and purified by standard methods, and the two fragments were ligated to construct the recombinant plasmid pET22b(+)_pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)-MTG.

[0164] Next, pET22b(+)_pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)-MTG was introduced into E. coli to obtain transformants. After culturing the transformants, IPTG was added to a final concentration of 0.1 mM to induce expression. After overnight culture, the culture supernatant was collected and analyzed using HisTrap®. TM His-tag purification was performed using an FF crude column (Cytiva). After purification, the yield and purity were confirmed by SDS-PAGE, and the product was purified using a HiLoad 16 / 600 Superdex column. TM Further purification was carried out using size exclusion chromatography (SEC) (Cytiva) using a 200 pg column to obtain the target protein.

[0165] <Example 2: Preparation of antibody fragment-substrate complex using MTG fusion product and pepsin> (Purpose) The MTG mutant described in Example 1 developed by the present inventors can selectively crosslink lysine residues at specific positions in native antibodies and glutamine residues in substrate molecules, as described in the prior application PCT / JP2024 / 016451 (filed April 26, 2024).

[0166] Furthermore, by changing the order of protein G and MTG mutants in the MTG fusion complex, substrate molecules can be selectively and specifically modified at lysine residue 65 on the antibody with Pro(K10R / Y12A)MTG-(G4S)3-pG (Fab-ΔKQ), and at lysine residue 225 on the antibody with pG-G4S-Pro(K10R / Y12A)MTG.

[0167] As shown in Figure 3, when the antibody fragment substrate conjugate of the present invention is produced using the above-mentioned MTG mutant, two production strategies are possible depending on the timing of antibody modification with the substrate molecule and antibody fragmentation: Strategy 1: A method in which an antibody is modified with a substrate molecule using an MTG mutant, and then the antibody is fragmented by protease treatment to obtain the desired antibody fragment substrate conjugate; and Strategy 2: An antibody is fragmented with a protease, and then the antibody fragment is modified with a substrate molecule using an MTG mutant to obtain the desired antibody fragment substrate conjugate.

[0168] Therefore, in this Example, pepsin, which has a cleavage site in the hinge region further below the target Lys residue K225 of IgG, was used as the protease, and the crosslinking activity of the products produced using both strategies, including the modification site at position 65 or 225 on the antibody, and the modification rate were evaluated by reverse-phase HPLC.

[0169] (Method) 1. Pepsin Fragmentation of Antibody-Substrate Complexes Using Strategy 1 Herceptin (Trastuzumab) as an IgG, a drug model Gln(Q) substrate, the two MTG mutants prepared in Example 1, DTT, and Treharose were mixed in Tris-HCl (pH 8.0) to carry out an antibody fragmentation reaction. The drug model Q substrate used was FAM-YPLQMRG-NH2 (SEQ ID NO: 13). After 120 minutes of reaction at 37°C, 0.1 M sodium acetate buffer (pH 4.5) was added, and the mixture was centrifuged at 3,000 g for 10 minutes at 4°C using an Amicon (50k). The mixture was then centrifuged at 14,000 g for 10 minutes at 4°C using an Amicon (30k), and concentrated to prepare the antibody-substrate complexes.

[0170] Pepsin solution was added to the antibody-substrate complex to a concentration of 0.01 mg / mL, and the complex was incubated at 37°C for 20 hours. After reduction with 0.1 M sodium phosphate buffer (pH 6.0), the complex was centrifuged at 14,000 g for 10 minutes at 4°C using an Amicon (30k). After concentration, 2-MEA solution was added to a concentration of 0.01 M, and the complex was incubated at 37°C for 90 minutes. The complex was diluted to a Herceptin concentration of 0.5 mg / mL.

[0171] After mixing with 2×SDS-PAGE Sample Buffer without DTT at a ratio of 1:1, non-reducing SDS-PAGE was performed and observed using a fluorescent imager (iBright FL1500; Thermo Fisher Scientific).

[0172] 2. Substrate cross-linking activity using MTG fusion to pepsin-fragmented Fab' (Strategy 2) A 0.5 mg / mL Herceptin solution was added to 0.1 M sodium acetate buffer (pH 4.5) and subjected to ultrafiltration at 14,000 g for 10 minutes at 4°C using an Amicon (100k). The resulting solution was then ultrafiltered at 14,000 g for 10 minutes at 4°C, concentrated, and then subjected to inverse centrifugation at 14,000 g for 10 minutes at 4°C.

[0173] Pepsin solution was added to the Herceptin solution to a final concentration of 0.01 mg / mL, and the mixture was incubated at 37°C for 20 hours. After reduction with 0.1 M sodium phosphate buffer (pH 6.0), the mixture was ultrafiltered at 14,000 g for 10 minutes at 4°C using an Amicon (50k). After concentration, 2-MEA solution was added to a final concentration of 0.01 M, and the mixture was incubated at 37°C for 90 minutes to fragment Herceptin.

[0174] 40 mM Tris-HCl buffer (pH 8.0) was added, and the mixture was concentrated by ultrafiltration at 14,000 g for 10 minutes at 4°C using an Amicon (30k).

[0175] Herceptin Fab', a drug model Q substrate, and the two MTG mutants prepared in Example 1 were mixed in Tris-HCl (pH 8.0) to carry out a crosslinking reaction. The drug model Q substrate used was FAM-YPLQMRG-NH2 (SEQ ID NO: 13). After reacting at 37°C for 120 minutes, N-ethylmaleimide (NEM) was added to terminate the reaction.

[0176] After mixing with 2×SDS-PAGE Sample Buffer without DTT at a ratio of 1:1, non-reducing SDS-PAGE was performed and observed using a fluorescent imager (iBright FL1500; Thermo Fisher Scientific).

[0177] 3. Comparison of the substrate modification rate of IgG or Fab' by each MTG fusion. Samples were prepared to calculate the modification rate. IgG and Fab, each MTG fusion, and drug model Q substrate were mixed in Tris-HCl (pH 8.0) (Table 2-11).

[0178] After 60 minutes of reaction at 37°C, 6 M guanidine hydrochloride (100 mM ammonium bicarbonate) and DTT were added, and the reduction reaction was carried out for 30 minutes at 60°C. The denatured and reduced sample was subjected to HPLC using a COSMOSIL 5Ph-AR-300 column, and the modification rate was calculated from the peak areas of the heavy chain modified with Q substrate and the unmodified heavy chain.

[0179] (Results) 1. Strategy 1: Production of antibody fragment-substrate conjugate by pepsin fragmentation of antibody-substrate conjugate Figure 4 shows the results of pepsin fragmentation of the antibody-substrate conjugate obtained using Pro(K10R / Y12A)MTG-(G4S)3-pG, and Figure 5 shows the results of pepsin fragmentation of the antibody-substrate conjugate obtained using pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG.

[0180] As shown in these figures, the generation of F(ab')2 fragments (approximately 110 kDa) from the antibody-substrate complex by pepsin and the generation of Fab' fragments (approximately 55 kDa) by 2-MEA were confirmed. It was revealed that the desired antibody fragment-substrate complex (Herceptin Fab'-Q) can be generated by fragmenting the antibody in the antibody-substrate complex with pepsin in Strategy 1.

[0181] 2. Strategy 2: Preparation of antibody fragment-substrate complex by crosslinking substrate molecules to pepsin fragment Fab' using MTG fusion complex Figure 6 shows the results of the crosslinking reaction of substrate molecules to Fab' using Pro(K10R / Y12A)MTG-(G4S)3-pG, and Figure 7 shows the results of the crosslinking reaction of substrate molecules to Fab' using pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG.

[0182] As shown in the fluorescence images in these figures, the Fab' band seen around 45 kDa was confirmed to be modified with the fluorescent Q substrate. These results demonstrate that in Strategy 2, substrate modification is possible even for Herceptin Fab' fragmented with pepsin, and the desired antibody fragment-substrate complex (Herceptin Fab'-Q) can be produced.

[0183] 3. Results of reverse-phase HPLC and modification rate for substrate modification to Herceptin heavy chain Figure 8 shows the results of reverse-phase HPLC to confirm the modification of the Q substrate to K65 of the Herceptin heavy chain by Pro(K10R / Y12A)MTG-(G4S)3-pG, and Figure 9 shows the results of reverse-phase HPLC to confirm the modification of the Q substrate to K225 of the Herceptin heavy chain by pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG. In both figures, A shows the enzyme-substrate complex in which K65 of IgG (Herceptin) is modified with the Q substrate (Herceptin(K65)-Q), and B shows the enzyme-fragment-substrate complex in which K65 of Herceptin Fab' is modified with the Q substrate (Herceptin Fab'(K65)-Q).

[0184] Reverse-phase HPLC analysis involves reductive sample pretreatment followed by LC separation using a reverse-phase column to calculate the drug-to-antibody ratio (DAR). IgG has four interchain disulfide bonds, which, when cleaved with a reducing agent, produce antibody fragments derived from the light chain and heavy chain. The light chain (approximately 25 kDa) and heavy chain (approximately 50 kDa) have large molecular weight differences and differing hydrophobicities, making them easily separable by reverse-phase HPLC.

[0185] In Figures 8 and 9, 214 nm represents the absorption peak derived from the peptide bond of the protein, and 495 nm represents the detection wavelength for the fluorescent dye FAM of the Q substrate. Because of reduction using DTT, the antibody and Fab' are detected as separate heavy and light chains. In the figures, a represents the MTG fusion, b represents the light chain of Herceptin or Herceptin Fab', c represents the unmodified Herceptin heavy chain, c' represents the modified Herceptin heavy chain, d represents the unmodified Herceptin Fab' heavy chain, and d' and d'' represent the modified heavy chain of Herceptin Fab' (antibody fragment-substrate complex). These results confirmed the modification of the Q substrate to Fab' by all MTG derivatives in reverse-phase HPLC.

[0186] In Figure 8B, a second peak (d'') was observed at a detection wavelength of 214 nm, further delayed than the modified heavy chain. This second peak d'' is thought to represent a two-molecule substrate-modified heavy chain, in which the modified heavy chain is further modified with another Q substrate molecule, because the peak intensity at a detection wavelength of 214 nm is approximately half that of the mono-molecule-modified heavy chain, whereas the peak intensity at a detection wavelength of 495 nm is almost the same.

[0187] Figure 10 shows the modification rate of substrate molecules to K65 in the Herceptin heavy chain calculated from the results of reverse-phase HPLC using Pro(K10R / Y12A)MTG-(G4S)3-pG, and Figure 11 shows the modification rate of substrate molecules to K225 in the Herceptin heavy chain calculated from the results of reverse-phase HPLC using pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG.

[0188] 10 reveals that the rate of substrate modification to unmodified heavy chain (Fab') K65 (94.9%) was higher than that to Herceptin (IgG) K65 (80.3%). Similarly, FIG. 11 reveals that the rate of substrate modification to unmodified heavy chain (Fab') K225 (73.4%) was higher than that to Herceptin (IgG) K225 (30.6%).

[0189] Furthermore, compared with Strategy 1, in which the antibody-substrate complex is fragmented with pepsin to obtain the desired antibody fragment-substrate complex (Figure 10: IgG; Figure 11: IgG), Strategy 2, in which the antibody is fragmented with pepsin and then specific lysine residues of Fab' are modified with TMG fusion compounds to obtain the desired antibody fragment-substrate complex (Figure 10: Fab'; Figure 11: Fab'), showed a higher substrate modification rate.

[0190] Example 3: Preparation of antibody fragment substrate complex using MTG fusion product and papain (Objective) In this example, papain, which has a cleavage point two residues below K225, the target Lys residue of IgG, and above the hinge region, was used as the protease, and similarly to Example 2, crosslinking activity of the products produced, including the modification position at position 65 or 225 on the antibody, and modification rate evaluation by reverse-phase HPLC were investigated.

[0191] (Methods) 1. Strategy 1: Preparation of antibody fragment substrate conjugate by papain fragmentation of antibody substrate conjugate Herceptin as IgG, drug model Gln(Q) substrate, the two MTG mutants prepared in Example 1, DTT, and Treharose were mixed in Tris-HCl (pH 8.0) to carry out an antibody fragmentation reaction. The drug model Q substrate used was FAM-YPLQMRG-NH2 (SEQ ID NO: 13).

[0192] After 120 minutes of reaction at 37°C, 50 mM sodium phosphate buffer (pH 7.0) was added, and the mixture was centrifuged at 3,000 g for 10 minutes at 4°C using an Amicon (50k).The mixture was then centrifuged at 14,000 g for 10 minutes at 4°C using an Amicon (30k) and concentrated to prepare the antibody-substrate complex.

[0193] A papain solution containing 0.05 mg / mL papain, 10 mM L-cysteine, and 10 mM EDTA at final concentrations was prepared and added to the antibody-substrate complex. Iodoacetamide was then added to the solution to a final concentration of 0.3 M, and the mixture was incubated at 37°C for 30 minutes.

[0194] Non-reducing SDS-PAGE was performed and observed using a fluorescent imager (iBright FL1500; Thermo Fisher Scientific).

[0195] 2. Strategy 2: Preparation of antibody fragment-substrate complex by crosslinking substrate molecules to papain fragment Fab using MTG fusion 2-1. Papain fragmentation: A papain solution containing 0.1 mg / mL papain, 50 mM L-cysteine, and 10 mM EDTA (final concentrations) was added to a 2 mg / mL Herceptin solution and incubated at 37°C for 90 minutes. Iodoacetamide was then added to a final concentration of 0.3 M and further incubated at 37°C for 30 minutes to fragment Herceptin into Fab and Fc. Tris-HCl (pH 8.0) was added, and the mixture was ultrafiltered at 14,000 xg for 10 minutes at 4°C using an Amicon (30kJ), concentrated, and then centrifuged at 14,000 xg for 10 minutes at 4°C to obtain a Herceptin-derived Fab and Fc fragment solution.

[0196] After mixing with 2×SDS-PAGE Sample Buffer without DTT at a ratio of 1:1, non-reducing SDS-PAGE was performed and observed using a fluorescent imager (iBright FL1500; Thermo Fisher Scientific).

[0197] 2-2. Purification of Fab using a Protein L column After removing Fc and the fragmentation enzyme, Fab was purified using a Protein L column that specifically binds to Fab. TM Size Exclusion Chromatography (SEC) was performed using a 200 Increase 10 / 300 GL column.

[0198] After fragmentation of IgG with papain, 20 mM sodium phosphate buffer (pH 7.0) was added to prepare a papain-fragmented sample.

[0199] 50 mM sodium phosphate buffer (pH 7.0) was passed through a silica monolith column (MonoSpin Protein L) for antibody purification and centrifuged at 2300 × g for 30 seconds. The papain-fragmented sample was then added and centrifuged at 2300 × g for 30 seconds. 50 mM sodium phosphate buffer (pH 7.0) was then added to the column, which was then centrifuged at 2300 × g for 30 seconds to wash the column. Finally, 100 mM glycine-HCl buffer (pH 2.5) was added to the column, which was then centrifuged at 2300 × g for 30 seconds to elute the sample. 1 M Tris-HCl buffer (pH 8.5) was added to the eluate for neutralization.

[0200] Superdex TM Further purification was performed using size exclusion chromatography (SEC) on a 200 Increase 10 / 300 GL column to obtain Herceptin Fab. After purification, the purity and desired fragment size were confirmed by non-reducing SDS-PAGE.

[0201] Herceptin Fab, a drug model Q substrate, and the two MTG mutants prepared in Example 1 were mixed in Tris-HCl (pH 8.0) to carry out a crosslinking reaction. The drug model Q substrate used was FAM-YPLQMRG-NH2 (SEQ ID NO: 13). After 120 minutes of reaction at 37°C, N-ethylmaleimide (NEM) was added to terminate the reaction.

[0202] 2-3 Comparison of the substrate modification rate of IgG or Fab by each MTG fusion. Samples were prepared to calculate the modification rate. IgG and Fab, each MTG fusion, and drug model Q substrate were mixed in Tris-HCl (pH 8.0) (Table 2-11).

[0203] After 60 minutes of reaction at 37°C, 6 M guanidine hydrochloride (100 mM ammonium bicarbonate) and DTT were added, and the reduction reaction was carried out for 30 minutes at 60°C. The denatured and reduced sample was subjected to HPLC using a COSMOSIL 5Ph-AR-300 column, and the modification rate was calculated from the peak areas of the heavy chain modified with Q substrate and the unmodified heavy chain.

[0204] (Results) 1. Strategy 1: Production of antibody fragment-substrate complex by papain fragmentation of antibody-substrate complex Figure 12 shows the results of papain fragmentation of the antibody-substrate complex obtained using Pro(K10R / Y12A)MTG-(G4S)3-pG, and Figure 13 shows the results of papain fragmentation of the antibody-substrate complex obtained using pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG.

[0205] In both figures, bands were observed around 47 kDa and 50 kDa, corresponding to the Fc fragment, confirming the generation of an antibody fragment substrate complex (Herceptin Fab-Q) (47 kDa) by papain in the Herceptin antibody substrate complex. Furthermore, a band was observed around 130 kDa in both Figures 12 and 13. Papain is primarily used to generate Fab fragments, but it can also be used to generate F(ab')2 fragments by removing the cysteine ​​added for activation after activation. It has also been reported that a band similar to the F(ab')2 fragment is generated when Fab fragments are generated by papain. Therefore, this band around 130 kDa is the F(ab')2 fragment, and it is believed that not only Fab fragments but also F(ab')2 fragments were generated under the reaction conditions used in this study.

[0206] 2. Strategy 2: Preparation of antibody fragment-substrate complex by crosslinking substrate molecules to papain fragment Fab using MTG fusion complex Figure 14 shows the results of the crosslinking reaction of substrate molecules to Fab using Pro(K10R / Y12A)MTG-(G4S)3-pG, and Figure 15 shows the results of the crosslinking reaction of substrate molecules to Fab using pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG.

[0207] As shown in the fluorescence images in these figures, the Fab band seen around 45 kDa was confirmed to be modified with the fluorescent Q substrate. These results demonstrate that in Strategy 2, papain-fragmented Herceptin Fab can also be modified with a substrate molecule, and the desired antibody fragment-substrate complex (Herceptin Fab-Q) can be produced.

[0208] 3. Results of reverse-phase HPLC and modification rate for substrate modification to Herceptin heavy chain Figure 16 shows the modification of the Q substrate to K65 of the Herceptin heavy chain by Pro(K10R / Y12A)MTG-(G4S)3-pG, and Figure 17 shows the results of reverse-phase HPLC to confirm the modification of the Q substrate to K225 of the Herceptin heavy chain by pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG. In both figures, A shows the enzyme-substrate complex in which K65 of IgG (Herceptin) is modified with the Q substrate (Herceptin(K65)-Q), and B shows the enzyme-fragment-substrate complex in which K65 of Fab is modified with the Q substrate (Herceptin Fab(K65)-Q).

[0209] In Figures 16 and 17, 214 nm represents the absorption peak derived from the peptide bond of the protein, and 495 nm represents the detection wavelength for the fluorescent dye FAM of the Q substrate. Because of reduction using DTT, the antibody and Fab are detected as separate heavy and light chains. In the figures, a represents the MTG fusion, b represents the light chain of Herceptin or Herceptin Fab, c represents the unmodified Herceptin heavy chain, c' represents the modified Herceptin heavy chain, e represents the modified Herceptin Fab heavy chain, and e' and e'' represent the modified heavy chain (antibody fragment-substrate complex). These results confirmed the modification of the Q substrate to Fab by all MTG derivatives in reverse-phase HPLC.

[0210] In Figure 16B, a second peak (e'') was observed at a detection wavelength of 214 nm, further delayed than the modified heavy chain. This second peak e'' has approximately half the peak intensity of the mono-modified heavy chain at a detection wavelength of 214 nm, whereas it has almost the same intensity at a detection wavelength of 495 nm. Therefore, as in Figure 8 of Example 2, this second peak e'' is likely to be a two-substrate-modified heavy chain in which an additional Q substrate molecule has been added to the modified heavy chain.

[0211] Figure 18 shows the modification rate of substrate molecules to K65 in the Herceptin heavy chain calculated from the results of reverse-phase HPLC using Pro(K10R / Y12A)MTG-(G4S)3-pG, and Figure 19 shows the modification rate of substrate molecules to K225 in the Herceptin heavy chain calculated from the results of reverse-phase HPLC using pG(Fab-ΔKQ)-G4S-Pro(K10R / Y12A)MTG.

[0212] 18 shows that the rate of substrate modification to unmodified heavy chain (Fab) K65 (98.4%) was higher than that to Herceptin (IgG) K65 (93.3%). Similarly, FIG. 19 shows that the rate of substrate modification to unmodified heavy chain (Fab) K225 (80.3%) was higher than that to Herceptin (IgG) K225 (47.8%).

[0213] Furthermore, compared with Strategy 1 (Fig. 18: IgG; Fig. 19: IgG), in which the antibody-substrate complex is fragmented with papain to obtain the desired antibody fragment-substrate complex, Strategy 2 (Fig. 18: Fab; Fig. 19: Fab), in which the antibody is fragmented with papain and then specific lysine residues of Fab' are modified with an MTG fusion compound to obtain the desired antibody fragment-substrate complex, showed a higher substrate modification rate.

[0214] The results of Examples 2 and 3 demonstrate that, in producing the antibody fragment substrate conjugate of the present invention, fragmenting the antibody with a protease and then modifying specific lysine residues of the antibody fragment with a substrate molecule using an MTG fusion product, rather than fragmenting the antibody substrate conjugate with a protease, results in a higher substrate modification rate and an improved yield of the antibody fragment substrate conjugate.

[0215] Example 4 Purification of antibody fragment substrate complex and evaluation of antigen binding ability (Purpose) The results of Examples 2 and 3 demonstrated that more efficient and uniform antibody fragment substrate complexes can be produced by modifying the substrate molecule with an MTG fusion compound (ProMTG-pG or pG-ProMTG) on an antibody fragment rather than on an antibody.

[0216] However, strictly speaking, the antibody fragment substrate complex obtained in Examples 2 or 3 above is in the state of an enzyme-antibody fragment-substrate complex in which the MTG fusion remains bound to the antibody fragment via the antibody-binding protein. Therefore, in this Example, the MTG fusion is separated and purified from the enzyme-antibody fragment-substrate complex to obtain the final target antibody fragment substrate complex.

[0217] (Method) In this example, enzyme-antibody fragment-substrate complexes were prepared using ProMTG-pG and pG-ProMTG with Fab fragments obtained by fragmenting an IgG antibody with papain. That is, the ProMTG-pG-Fab(K65)-Q substrate complex or pG-ProMTG-Fab(K225)-Q substrate complex obtained in Example 3 was treated under acidic conditions to remove the MTG fusion (ProMTG-pG or pG-ProMTG) and purify the complexes. Furthermore, we investigated whether the obtained antibody fragment-substrate complexes were actually capable of binding to the target antigen.

[0218] 1. Removal and purification of MTG fusion from enzyme-antibody fragment-substrate complex Herceptin was treated with papain to prepare Herceptin Fab using the method described in Example 3, and then Herceptin Fab, drug model Q substrate, and the two MTG mutants prepared in Example 1 were mixed in Tris-HCl (pH 8.0) to carry out a cross-linking reaction. As a control, Herceptin that had not been treated with papain was used, and subsequent experiments were carried out under the same conditions.

[0219] In this example, TAMRA-YPLQMRG-NH2 (SEQ ID NO: 36) (Sigma-Aldrich) was used as the drug model Q substrate instead of FAM-YPLQMRG-NH2 (SEQ ID NO: 13). After 120 minutes of reaction at 37°C, citrate buffer (pH 3.0) was added to terminate the MTG reaction and inactivate the antibody binding of Protein G. In the reaction solution, an enzyme-antibody fragment-substrate complex, namely, a ProMTG-pG-Fab(K65)-Q substrate complex or a pG-ProMTG-Fab(K225)-Q substrate complex, was formed.

[0220] The reaction solution was injected into a Protein Ark HiFliQ (registered trademark) S-type column (Caliber Scientific) and purified by cation exchange chromatography under the conditions listed in Table 1. Solvent B (50 mM citrate buffer (pH 3.0), 0.2 M arginine hydrochloride, 2 M NaCl) was used for purification, and detection wavelengths were 280 nm and 565 nm.

[0221]

[0222] 1 M tris-HCl (pH 8.0) was added to each fraction in advance for neutralization, and only the fraction containing the Fab(K65)-Q substrate complex or the Fab(K225)-Q substrate complex derived from the fluorescently modified antibody was collected.

[0223] The purified fraction was placed in an Amicon (10 kDa) and concentrated at 3000 × g and 4 °C with 1 × PBS. 6 M guanidine hydrochloride (100 mM ammonium bicarbonate, 100 mM DTT) was then added and the reduction reaction was carried out at 60 °C for 30 minutes. The denatured and reduced sample was subjected to HPLC on a COSMOSIL 5Ph-AR-300 column.

[0224] (Results) The results of cation exchange chromatography purification of antibody fragment substrate complexes are shown in Figures 21 and 22. Figure 21 shows the results of removing and purifying the MTG fusion from the ProMTG-pG-Fab(K65)-Q substrate complex, and Figure 22 shows the results of removing and purifying the MTG fusion from the pG-ProMTG-Fab(K225)-Q substrate complex. These results demonstrated that the desired antibody fragment substrate complex could be isolated from enzyme-antibody fragment-substrate complexes bound to either MTG fusion.

[0225] <Example 5: Study on modification of two substrate molecules with an antibody fragment-substrate complex> (Purpose) The results of Examples 2 and 3 revealed that the use of Pro(K10R / Y12A)MTG-(G4S)3-pG enables modification of two substrate molecules to an antibody fragment.

[0226] Therefore, in this example, the influence of different lengths of linkers contained in the MTG fusion products on bimolecular modification of the substrate was examined.

[0227] (Method) In this example, Fab fragments obtained by fragmenting an IgG antibody with papain were subjected to denaturation treatment with antibody fragment substrate complexes prepared using ProMTG-(G4S)3-pG and ProMTG-G4S-pG, and the modification rate was calculated by analysis using reverse-phase HPLC.

[0228] Herceptin was treated with papain to prepare Herceptin Fab as described in Example 3. Then, Herceptin Fab, the drug model Q substrate, and the two MTG fusion proteins prepared in Example 1 (ProMTG-(G4S)3-pG or ProMTG-G4S-pG) were mixed in Tris-HCl (pH 8.0) to carry out a cross-linking reaction.

[0229] In this example, TAMRA-YPLQMRG-NH2 (SEQ ID NO: 36) was used as the drug model Q substrate. After 60 minutes of reaction at 37°C, 6 M guanidine hydrochloride (100 mM ammonium bicarbonate) and DTT were added, and a reduction reaction was carried out at 60°C for 30 minutes. The denatured and reduced sample was subjected to HPLC using a COSMOSIL 5Ph-AR-300 column as in Example 3, and the modification rate was calculated from the peak areas of the heavy chain modified with the Q substrate and the unmodified heavy chain.

[0230] (Results) The results are shown in Figure 23. Panel A shows the modification rate of the antibody-substrate complex obtained using Pro(K10R / Y12A)MTG-(G4S)3-pG, and Panel B shows the modification rate of the antibody-substrate complex obtained using Pro(K10R / Y12A)MTG-G4S-pG. Figure 23 revealed that the modification rate of two substrate molecules onto the antibody fragment (Herceptin Fab) was lower (6%) when Pro(K10R / Y12A)MTG-G4S-pG (short linker) was used compared to (23%) when Pro(K10R / Y12A)MTG-(G4S)3-pG (long linker). These results suggest that shortening the linker length can suppress bimolecular modification of the antibody fragment with substrate molecules, whereas increasing the linker length to a certain extent can promote bimolecular modification. Therefore, it was demonstrated that bimolecular modification of substrate molecules can be controlled by adjusting the linker length of the MTG fusion.

[0231] Molecular modeling of the lysine exposure of Herceptin antibody before and after fragmentation revealed that the exposure of K225 in IgG increased from 37.0% (ranked 16th in the lysine exposure ranking) to 86.2% (ranked 1st in the lysine exposure ranking) after fragmentation (data not shown). This result indicates that K225 becomes the most exposed residue in the Fab fragment after IgG fragmentation, making it the most susceptible to modification. Therefore, bimolecular modification of substrate molecules in Fab fragments with Pro(K10R / Y12A)MTG-(G4S)3-pG and Pro(K10R / Y12A)MTG-G4S-pG suggests that in addition to K65 in the Fab heavy chain, K225 is newly modified with substrate molecules.

[0232] Example 6: Evaluation of antigen-binding ability of purified antibody fragment substrate complex (Objective) To confirm that the antibody fragment substrate complex obtained in Example 4 can bind to the target antigen, which is the target molecule, and function as an antibody fragment substrate complex.

[0233] (Method) Herceptin, used as the IgG in the Examples herein, is a humanized monoclonal antibody targeting HER2. To evaluate the antigen-binding ability of the antibody fragment substrate conjugate of the present invention, the antibody fragment substrate conjugate obtained in Example 4 was applied to HER2-positive (HER2(+)) and HER2-negative (HER2(-)) breast cancer cells, and the target molecule specificity of the antibody fragment substrate conjugate and the cell permeability of the Fab were evaluated based on the results of intracellular uptake of the fluorescent substrate molecule (TAMRA). The antibody fragment substrate conjugate and antibody fragment substrate conjugate used in this Example were those prepared in Example 4.

[0234] 1.0 × 10 cells were cultured in a 4-well dish using McCoy's 5A medium. 4HER2-positive breast cancer cells SK-BR-3 (HER2(+)) were seeded at 100 cells / well and incubated at 37°C in a 5% CO2 incubator for 24 hours to allow them to adhere to the bottom. After washing with Opti-MEM, the Fab(K65)-Q substrate conjugate or Fab(K225)-Q substrate conjugate obtained in Example 4 was added, and the cells were incubated at 37°C in a 5% CO2 incubator for 2 hours. After washing with Opti-MEM, 4% paraformaldehyde fixative was added and the cells were allowed to stand for 10 minutes to fix them. After washing with Opti-MEM, the fixed cells were observed by confocal laser scanning microscopy (CLSM).

[0235] Using a similar method, experiments were also performed on HER2-negative breast cancer cells MDA-MB-231 (HER2(-)) using DMEM medium.

[0236] (Results) Figure 24 shows the results for SK-BR-3 cells, a HER2-positive (HER2(+)) breast cancer cell line. Figure 24 shows CLSM images (bright field) and fluorescence images of each breast cancer cell when the antibody-substrate conjugate Herceptin(K65)-Q substrate conjugate (represented as "IgG(K65)-TAMRA" in the figure), Herceptin(K225)-Q substrate conjugate (represented as "IgG(K225)-TAMRA" in the figure), or the antibody fragment-substrate conjugate Fab(K65)-Q substrate conjugate (represented as "Fab(K65)-TAMRA" in the figure) or Fab(K225)-Q substrate conjugate (represented as "IgG(K225)-TAMRA" in the figure) was added.

[0237] In SK-BR-3 cells treated with antibody-substrate or antibody-fragment-substrate conjugates, fluorescence was observed on the cell membrane regardless of the modification position (K65, K225) of the fluorescent substrate molecule TAMRA-YPLQMRG-NH2 (SEQ ID NO: 36) (Q substrate). This result indicates that the antibody (Herceptin) or its antibody fragment (Fab) modified with the Q substrate binds to HER2, an antigen on the cell membrane of SK-BR-3 cells.

[0238] On the other hand, as shown in Figure 25, in MDA-MB-231 cells, a HER2-negative breast cancer cell line, no fluorescence was observed on the cell membrane even when the antibody-substrate conjugate or antibody fragment-substrate conjugate was added.

[0239] As controls, we also observed cells containing only the fluorescent substrate molecule (Q substrate), TAMRA-YPLQMRG-NH2 (SEQ ID NO: 36), antibody and Q substrate, and Fab and Q substrate, but no uptake into the cells was confirmed in any of these cases (data not shown).

[0240] Example 7: Verification of the modification rate of a click reaction element-conjugated glutamine tag on an antibody fragment using an MTG fusion product (Objective) To verify the modification rate on an antibody fragment when a click reaction element-conjugated glutamine tag is used as a glutamine tag for binding a substrate molecule to an antibody or antibody fragment.

[0241] (Method) Herceptin-derived Fab (Herceptin Fab) was used as the antibody fragment. The Herceptin Fab used was obtained by the method described in Example 3, "2-1. Fragmentation with papain." The MTG fusion protein used was Pro(K10R / Y12A)MTG-(G4S)3-pG, obtained by the method described in Example 1, "1. Preparation of Pro(K10R / Y12A)MTG-(G4S)3-pG."

[0242] Next, samples were prepared for calculating the modification rate. As shown in Figure 27A, Herceptin Fab was prepared in 40 mM Tris-HCl (pH 8.0) to a final concentration of 5 μM, and then 5 μM Pro(K10R / Y12A)MTG-(G4S)3-pG was added. The mixture was incubated at 37°C for 3 minutes. Subsequently, 125 μM DBCO-PEG4-LLQG was added as a click reaction element-conjugated glutamine tag, and the mixture was incubated at 37°C for 12 hours. The specific procedure followed a conjugation method known in the art (e.g., Nishioka, R., et al., 2024, Chemical Communications, 60(65), 8545-8548).

[0243] To verify the modification rate of DBCO-PEG4-LLQG over time, 30 μL aliquots were taken from each reaction mixture after 0 hours (immediately after addition of DBCO-PEG4-LLQG), followed by 2 hours, 4 hours, 6 hours, 8 hours, 10 hours, and 12 hours, and the modification rate of the antibody fragment was immediately measured by RP-HPLC analysis.

[0244] The modification rate was measured under reducing conditions by HPLC analysis. 30 μL of the reaction mixture was added with 80 μL of reduction buffer (6 M guanidinium chloride containing 0.1 M DTT / 100 mM ammonium bicarbonate), and the reduction reaction was carried out at 60°C for 30 minutes. After reduction, the reaction mixture was detected at 214 nm using a Cosmosil® 5Ph-AR-300 packed column (4.6 ID × 150 mm, Nacalai Tesque). Separation of the reaction mixture was carried out using two elution systems: 0.1% TFA / HO (eluent A) and 0.1% TFA / acetonitrile (eluent B), under the conditions listed in Table 2. The modification rate was calculated using the following formula: Modification rate (%) = modified heavy chain area × 100 / (unmodified heavy chain area - modified heavy chain area).

[0245]

[0246] (Results) Figure 27B shows the modification rate when DBCO-PEG4-LLQG was added to Herceptin Fab using Pro(K10R / Y12A)MTG-(G4S)3-pG. As shown in this figure, approximately 80% of the Herceptin Fab was modified after 2 hours of reaction, and over 90% after 4 hours. These results demonstrate that even a click reaction element-conjugated glutamine tag can be efficiently modified into an antibody fragment using the MTG fusion of the present invention, and that a reaction time of 4 hours is sufficient for modification. Example 8: Verification of the number of modifications of a click reaction element-conjugated glutamine tag per antibody fragment molecule using an MTG fusion. (Objective) To confirm the number of DBCO-PEG4-LLQG modifications per Fab molecule in Fab-PEG4-DBCO obtained by modifying an antibody fragment with a click reaction element-conjugated glutamine tag using the MTG fusion in Example 7. (Method) Fab-PEG4-DBCO was prepared according to the method described in Example 7. The reaction mixture was allowed to react for 4 hours. Subsequently, 0.1 M citrate buffer (pH 3.0) was added to the reaction mixture to terminate the MTG reaction and inactivate the antibody binding of Protein G. The reaction mixture was then loaded onto a 1 mL Protein Ark HifliQ® S-type FPLC column (Caliber Scientific) and purified by cation exchange chromatography. Purification was performed using two buffer systems: Buffer A (50 mM sodium citrate (pH 3.0) / 0.2 M arginine hydrochloride solution) and Buffer B (50 mM sodium citrate (pH 3.0) / 0.5 M arginine hydrochloride solution / 2 M NaCl). The column was pre-equilibrated with 5% Buffer B, and a gradient elution of Buffer A / B (95:5 to 45:5) was performed for 20 minutes at a flow rate of 1.2 mL / min. 1M Tris-HCl (pH 8.0) was added to each fraction for neutralization, and the fraction containing Fab-PEG4-DBCO was collected. The fractions were then loaded into an Amicon® Ultra-15 10,000MW CO column (Merck) and concentrated at 3000 × g and 4°C with 1× PBS, after which the medium was replaced with PBS. The molecular weight of the resulting Fab-PEG4-DBCO was analyzed by MALDI-ToF-MS.The purified Fab-PEG4-DEBCO fraction was placed in an Amicon® Ultra-0.5 30,000 MW CO column (Merck) and concentrated at 3000 × g and 4°C. The fraction was then eluted with Milli-Q water. 1.5 μL of the sample was then mixed with 1.5 μL of sinapinic acid matrix. Analysis was performed using a MALDI-ToF-MS (autoflex® maX; Bruker) in linear positive mode. For reference, the molecular weight of the purified Fab obtained as described in Example 3, "2-1. Papain Fragmentation," was analyzed in the same manner. (Results) Figure 28 shows the mass spectrum. A indicates Fab, and B indicates Fab-PEG4-DEBCO. The m / z value of Fab was 47489.014, while that of Fab-PEG4-DEBCO was 48429.042, with a difference of 940.028. The molecular weight of DEBCO-PEG4-LLQG is 964.3 g / mol. When DEBCO-PEG4-LLQG binds to Fab, a new isopeptide bond is formed and ammonia (molecular weight 17.03 g / mol) is released. Therefore, the molecular weight per molecule of DEBCO-PEG4-LLQG when bound to Fab is 964.3 - 17.03 = 947.27. The linker-to-antibody ratio (LAR) of DEBCO-PEG4-LLQG bound to Fab is 940.028 / 947.27 = 0.992. These results demonstrate that one molecule of DEBCO-PEG4-LLQG binds to each molecule of Fab-PEG4-DEBCO obtained in Example 6. Example 9: Substrate modification of antibody fragments by click reaction (1) (Objective) By click reaction, azide-PEG3-FAM or Azide-PEG3-TAMRA, an azide-modified fluorescent dye molecule, is bound to Fab-PEG4-DEBCO as a substrate molecule for modification. (Method) As shown in Figure 29A and B, 25 μM Azide-PEG3-FAM or 25 μM Azide-PEG3-TAMRA was mixed with 5 μM Fab-PEG4-DEBCO prepared in Example 7, and the click reaction was carried out at 37°C for 2 hours. After reductive sample pretreatment, reverse-phase HPLC analysis was performed.The reverse-phase HPLC analysis was performed according to the method described in Example 2. Each reaction mixture was then mixed 1:1 with 2x SDS-PAGE Sample Buffer containing no DTT, separated by non-reducing SDS-PAGE, and then observed using a fluorescent imager (iBright FL1500; Thermo Fisher Scientific). (Results) The results are shown in Figures 29C and 30. Figure 29C shows the results of reverse-phase HPLC analysis of Fab-PEG4-DEBCO with Azide-PEG3-FAM via click reaction, as confirmed by the absorption peak derived from the peptide bond of the protein at a detection wavelength of 214 nm. In the figure, the solid line represents the results of reverse-phase HPLC analysis of a mixture of Fab-PEG4-DEBCO and Azide-PEG3-FAM after the click reaction, and the dashed line represents the results of reverse-phase HPLC analysis of Fab-PEG4-DEBCO before the click reaction. Due to the reduction using DTT, the Fab is detected as separate heavy and light chains. In the figure, a represents the light chain of Fab, b represents the heavy chain of Fab-PEG4-DEBCO, and c represents the heavy chain of Fab-FAM (Fab-FAM). As is clear from Figure 29C, the click reaction between Fab-PEG4-DEBCO and Azide-PEG3-FAM produced Fab-FAM (Fab-FAM), in which Fab was modified with FAM (as shown in c). Figure 30 shows the results of non-reducing SDS-PAGE separation after modification of Fab-PEG4-DEBCO with Azide-PEG3-FAM or Azide-PEG3-TAMRA via the click reaction. A shows the results of CBB staining of SDS-PAGE, and B and C show the results of image analysis using a fluorescent imager. B is a fluorescent image of FAM, and C is that of TAMRA. As shown in this figure, the click reaction can yield Fab-FAM (Fab-FAM) or Fab-TAMRA (Fab-TAMRA), in which Fab is modified with FAM or TAMRA via the triazole ring, regardless of the type of substrate. Example 10: Substrate modification of antibody fragments by click reaction (2) (Purpose) In Example 9, antibody fragments were modified with the fluorescent dye molecules FAM or TAMRA by click reaction.In this example, we demonstrate that muGFP (monomeric ultra-stable GFP: 26.8 kDa), which has a larger molecular weight than FAM (376.32 Da) or TAMRA (430.45 Da), can be bound to and modified by a click reaction with an antibody fragment. (Method) The basic procedure for the click reaction was performed according to the method described in Example 9, except that 25 μM muGFP-PEG4-Azide, an azido-type muGFP, was used as the substrate molecule. (Results) Figure 31 shows the results of CBB staining of a sample obtained by modifying Fab-PEG4-DEBCO with muGFP-PEG4-Azide via a click reaction, followed by separation by non-reducing SDS-PAGE. As shown in this figure, Fab-muGFP (Fab-muGFP), in which Fab was modified with muGFP, was produced only in lane 4, where the click reaction was performed. These results demonstrate that antibodies or antibody molecules can be modified by a click reaction, even when the substrate molecule has a large molecular weight. All publications, patents, and patent applications cited herein are hereby incorporated by reference in their entirety.

Claims

1. A method for producing an antibody fragment substrate conjugate, comprising: an enzyme binding step of binding a microbial transglutaminase fusion to an antibody fragment to prepare an enzyme-antibody fragment conjugate; a crosslinking step of mixing the enzyme-antibody fragment conjugate with a substrate molecule containing a glutamine residue and crosslinking the substrate molecule to the antibody fragment to prepare an enzyme-antibody fragment-substrate conjugate; and an enzyme separation step of separating the microbial transglutaminase fusion from the enzyme-antibody fragment-substrate conjugate to obtain an antibody fragment substrate conjugate, wherein the microbial transglutaminase fusion comprises an active microbial transglutaminase precursor and an antibody-binding protein, and the active microbial transglutaminase precursor comprises a mutant propeptide region in which at least one tyrosine residue and a lysine residue, at least one asparagine residue and a lysine residue, or at least one asparagine residue, at least one tyrosine residue and a lysine residue in the propeptide region of the microbial transglutaminase precursor are substituted with other amino acid residues.

2. The method of claim 1, which comprises an antibody fragmentation step in which immunoglobulin is cleaved at a predetermined site with a protease to obtain antibody fragments.

3. The method according to claim 2, further comprising a separation step of separating antibody fragments containing the target site for modification by the microbial transglutaminase fusion product from the antibody fragments obtained after the antibody fragmentation step.

4. The method of claim 2 or 3, wherein the protease is a cysteine ​​protease or an aspartic acid protease.

5. The method according to claim 4, wherein the protease is an aspartic acid protease, and the antibody fragmentation step involves cleaving immunoglobulin at a predetermined site with the aspartic acid protease, followed by reduction to obtain antibody fragments.

6. The method of claim 1, wherein the microbial transglutaminase fusion contains the activated microbial transglutaminase precursor at the N-terminus and the antibody-binding protein at the C-terminus.

7. The method of claim 1, wherein the microbial transglutaminase fusion comprises the active microbial transglutaminase precursor at the C-terminus and the antibody-binding protein at the N-terminus.

8. The method of claim 1, wherein the antibody-binding protein is a mutant antibody-binding protein.

9. The method of claim 8, wherein the mutant antibody-binding protein is a Fab-binding protein G that has lost its ability to bind to Fc in IgG.

10. The method of claim 8, wherein the mutant antibody-binding protein is Fc-binding protein G or Fc-binding protein A that has lost its ability to bind to Fab in IgG.

11. The method of claim 1, wherein the activated microbial transglutaminase precursor and the antibody-binding protein are fused via a linker consisting of a peptide chain.

12. The method of claim 1, wherein the substrate molecule is a nucleic acid, a polypeptide, a sugar chain, a lipid, a low molecular weight compound, or a combination thereof.

13. The method according to claim 12, wherein the low molecular weight compound is any one selected from the group consisting of a drug, a fluorescent substance, a luminescent substance, and an enzyme.

14. The method of claim 12 or 13, wherein the substrate molecule comprises a glutamine tag.

15. The method of claim 1, wherein the substrate molecule comprises an azide group, an alkynyl group, or a cycloalkynyl group.

16. A method for producing an enzyme-antibody fragment conjugate for crosslinking a substrate molecule containing a glutamine residue to an antibody fragment, comprising an enzyme binding step of binding a microbial transglutaminase fusion to an antibody fragment to prepare an enzyme-antibody fragment conjugate, wherein the microbial transglutaminase fusion comprises an activated microbial transglutaminase precursor and an antibody-binding protein, and the activated microbial transglutaminase precursor comprises a mutant propeptide region in which at least one tyrosine residue and a lysine residue, at least one asparagine residue and a lysine residue, or at least one asparagine residue, at least one tyrosine residue and a lysine residue in the propeptide region of the microbial transglutaminase precursor are substituted with other amino acid residues.

17. An antibody fragment-substrate complex in which a substrate molecule is cross-linked to the lysine residue at position 65 of the heavy chain of human Fab.

18. An antibody fragment-substrate conjugate in which a substrate molecule is cross-linked to the lysine residue at position 225 of the heavy chain of human F(ab')2.

Images