Method for separating multispecific antibody

Abstract

To provide a method that enables separation of a multispecific antibody contained in a sample.SOLUTION: A method for separating a multispecific antibody comprises the steps of: bringing a sample containing a multispecific antibody into contact with an antibody adsorbent comprising an insoluble carrier and an immunoglobulin-binding domain of Protein L (FpL) derived from a bacterium of the genus Finegoldia immobilized on the insoluble carrier, thereby adsorbing the antibody onto the adsorbent; and eluting the antibody adsorbed on the adsorbent using an eluent. Therein the eluent contains chloride ions, and the elution step elutes the antibody by a gradient that decreases a chloride ion concentration and a pH of the eluent, thereby solving the problem to be solved.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to a method for isolating polyspecific antibodies, and in particular to a method for isolating said antibodies by affinity chromatography using Protein L derived from bacteria of the genus Finegoldia as a ligand. [Background technology]

[0002] Patent Document 1 discloses a conventional method for purifying multispecific antibodies contained in a sample using affinity chromatography, in which the bispecific antibodies are adsorbed onto KappaSelect resin or LambdaFabSelect resin (both manufactured by Cytiva) and then eluted from the resin with a buffer containing 250 mmol / L sodium chloride to purify the bispecific antibodies. Patent Document 2 also discloses a method for purifying bispecific antibodies by adsorbing the bispecific antibodies onto an antibody separating agent comprising an insoluble carrier and an immunoglobulin-binding domain of Protein L derived from bacteria of the genus Finegoldia immobilized on the insoluble carrier, and then eluting the bispecific antibodies from the separating agent under a constant acidic pH condition using a gradient that decreases the conductivity of the buffer (specifically, decreases the sodium chloride concentration).

[0003] However, further improvement in the ability to separate these antibodies from other antibodies was desired. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Special Publication No. 2014-530848 [Patent Document 2] Special Publication No. 2020-508968 Summary of the Invention [Problem to be solved by the invention]

[0005] An object of the present invention is to provide a method capable of separating multispecific antibodies contained in a sample.

[0006] In one aspect, an object of the present invention is to provide a method capable of separating the antibody with high accuracy. [Means for solving the problem]

[0007] As a result of extensive investigations, the present inventors have solved the above-mentioned problems and arrived at the present invention by purifying polyspecific antibodies contained in a sample using an antibody adsorbent comprising an insoluble carrier and an immunoglobulin-binding domain of Protein L derived from bacteria of the genus Finegoldia immobilized on the carrier, and eluting the polyspecific antibodies adsorbed to the adsorbent with an elution solution using a gradient that decreases the ion concentration and pH of the elution solution.

[0008] That is, the present invention includes the following embodiments [1] and [4].

[0009] [1] A method for separating a polyspecific antibody, comprising the steps of: bringing a sample containing the polyspecific antibody into contact with an antibody adsorbent comprising an insoluble carrier and an immunoglobulin-binding domain of Protein L (FpL) derived from bacteria of the genus Finegoldia immobilized on the insoluble carrier, thereby adsorbing the antibody onto the adsorbent; and eluting the antibody adsorbed onto the adsorbent using an elution solution, The method, wherein the elution solution contains chloride ions, and the elution step elutes the antibody using a gradient that decreases the chloride ion concentration and pH of the elution solution.

[0010] [2] The method according to [1], wherein the chloride ion concentration of the eluate is 50 mmol / L or more and 150 mmol / L or less at the start of the gradient.

[0011] [3] The method according to [1] or [2], wherein the immunoglobulin-binding domain of FpL is a polypeptide selected from the group consisting of (a) and (c) below: (a) a polypeptide comprising at least the amino acid residues of the sequence set forth in SEQ ID NO: 1; (b) a polypeptide having an amino acid sequence containing at least the amino acid residues of the sequence set forth in SEQ ID NO: 1, with the proviso that the amino acid sequence contains substitution, deletion, insertion, or addition of one or several amino acid residues at one or several positions, and having antibody-binding activity; (c) A polypeptide comprising at least the amino acid residues of the sequence set forth in SEQ ID NO: 1, which has 70% or more identity to the sequence of amino acid residues, and which has antibody binding activity.

[0012] [4] The method according to any one of [1] to [3], wherein the multispecific antibody is a bispecific antibody. [Effects of the Invention]

[0013] The present invention provides a method for separating multispecific antibodies contained in a sample using an antibody adsorbent comprising an insoluble carrier and an immunoglobulin-binding domain of Protein L derived from bacteria of the genus Finegoldia immobilized on the insoluble carrier, characterized in that when the multispecific antibodies adsorbed to the adsorbent are eluted using an elution solution, the elution is performed using a gradient that decreases the chloride ion concentration and pH of the elution solution. The present invention enables separation of multispecific antibodies contained in a sample. In one embodiment, the antibodies can be separated with high accuracy. [Brief explanation of the drawings]

[0014] [Figure 1] 1 shows a map of the plasmid pEFd used for antibody gene transfer. [Figure 2] This is the chromatogram of Trastuzumab analyzed using an antibody separation column. The peak marked "T" in the figure corresponds to Trastuzumab. [Figure 3] This is the chromatogram of an analysis of belimumab using an antibody separation column. The peak marked "B" in the figure corresponds to belimumab. [Figure 4]This is a chromatogram showing the analysis of bispecific antibodies BsAbs of Trastuzumab and Belimumab using an antibody separation column. In the figure, the "B" peak corresponds to Belimumab, the "TB" peak corresponds to BsAbs, and the "T" peak corresponds to Trastuzumab. [Figure 5] The results show the effect of different concentrations of sodium chloride added to the eluent on the chromatogram of BsAbs. In the figure, the "B" peak corresponds to Belimumab, the "TB" peak corresponds to BsAbs, and the "T" peak corresponds to Trastuzumab. The chromatogram at 0 mmol / L is the same as Figure 4. DETAILED DESCRIPTION OF THE INVENTION

[0015] The present invention will be described in detail below.

[0016] As used herein, the term "multispecific antibody" may refer to an antibody that can specifically bind to two or more different antigens or epitopes. Depending on the number of antigens or epitopes that it can bind to, for example, an antibody that can specifically bind to two different antigens or epitopes may be called a bispecific antibody, and an antibody that can specifically bind to three different antigens or epitopes may be called a trispecific antibody. In the method of the present disclosure, the use of a bispecific antibody is more preferable than the use of a multispecific antibody because higher separation ability can be expected.

[0017] The immunoglobulin-binding domain of Protein L (hereinafter referred to as "FpL") derived from the genus Finegoldia is (a) a polypeptide comprising amino acid residues consisting of the sequence of the domain; As long as it has immunoglobulin-binding activity, (b) It may be a polypeptide comprising amino acid residues consisting of a partial sequence of the domain. Furthermore, as long as it has binding activity to immunoglobulins, (c) A polypeptide containing amino acid residues consisting of the sequence of the domain or a partial sequence thereof, wherein the amino acid residues have one or more substitutions, deletions, insertions, and additions of one or more amino acid residues at one or more positions; (d) It may be a polypeptide comprising at least amino acid residues of a sequence having 70% or more identity with the sequence of the domain or a partial sequence thereof.

[0018] Hereinafter, the polypeptides described in (c) and (d) above will also be collectively referred to as "variants."

[0019] Regarding the immunoglobulin-binding domain of FpL (the above (a)), the Finegoldia bacterium from which FpL is derived includes Finegoldia magna. The immunoglobulin-binding domain of Protein L derived from Finegoldia magna is as follows: Domain B1 (amino acid residues 104 to 173 of GenBank No. AAA25612 (SEQ ID NO: 2)), Domain B2 (amino acid residues 176 to 245 of GenBank No. AAA25612), Domain B3 (amino acid residues 248 to 317 of GenBank No. AAA25612), Domain B4 (amino acid residues 320 to 389 of GenBank No. AAA25612), Domain B5 (amino acid residues 393 to 462 of GenBank No. AAA25612), Domain C1 (amino acid residues 249 to 317 of GenBank No. AAA67503 (SEQ ID NO: 3)), Domain C2 (amino acid residues 320 to 389 of GenBank No. AAA67503), Domain C3 (amino acid residues 394 to 463 of GenBank No. AAA67503: SEQ ID NO: 1), and An example is domain C4 (amino acid residues 468 to 537 of GenBank No. AAA67503), but any domain may be selected.

[0020] The immunoglobulin-binding domain of FpL may contain, in addition to the selected immunoglobulin-binding domain, a portion of another immunoglobulin-binding domain. For example, when the amino acid sequence of immunoglobulin-binding domain C3 of Protein L derived from Finegoldia magna is selected as the immunoglobulin-binding domain of FpL, it may further contain a portion of the N-terminal region of said domain (domain C1, domain C2), or a portion of the C-terminal region of said domain (domain C4).

[0021] Regarding (b), the immunoglobulin-binding domain of FpL ((a)) is composed of four β-sheets, one α-helix, a loop connecting them, and an N-terminal loop region. However, amino acid residues in regions unrelated to antibody binding, such as the N-terminal loop region, may be deleted. As a specific example, when the immunoglobulin-binding domain of FpL ((a)) is domain C3 (SEQ ID NO: 1) of Protein L derived from Finegoldia magna, it is known that antibody-binding ability is maintained even when the amino acid residues from the first glutamic acid to the ninth glutamic acid, which correspond to the N-terminal loop region, are deleted (Housden NG et al., Biochemical Society Transactions, 31, 716-718, 2003). In other words, the partial amino acid sequence in (b) may be sufficient as long as it contains at least the amino acid sequence of the antibody-binding site. In other words, the partial amino acid sequence in (b) may be a partial sequence containing the amino acid sequence of the antibody-binding site.

[0022] The term "one or several" in (c) above varies depending on the position of the amino acid residue in the three-dimensional structure of the protein and the type of amino acid residue, but specifically means, for example, 1 to 20, 1 to 15, 1 to 10, 1 to 9, 1 to 8, 1 to 7, 1 to 6, 1 to 5, 1 to 4, 1 to 3, 1 or 2, or 1.

[0023] The "identity" in (d) refers to the ratio of amino acid residues of the same type in the amino acid sequences (Experimental Medicine, February 2013, Vol. 31, No. 3, Yodosha). "Identity with respect to the amino acid sequence" refers to identity with respect to the entire amino acid sequence. "High identity" may refer to identity of 70% or more, 80% or more, 90% or more, or 95% or more. Amino acid sequence identity can be determined using alignment programs such as BLAST (Basic Local Alignment Search Tool) or FASTA.

[0024] An example of the "substitution of one or several amino acid residues other than amino acid substitution" in (c) above is a conservative substitution, in which a substitution occurs between amino acids with similar physical and / or chemical properties. It is generally known to those skilled in the art that conservative substitutions maintain protein function between substituted and unsubstituted amino acids. Examples of conservative substitutions include substitutions between glycine and alanine, between serine and proline, or between glutamic acid and alanine (Protein Structure and Function, Medical Science International, 9, 2005). Furthermore, the "one or more of substitutions, deletions, insertions, and additions of one or several amino acid residues" in (c) above also includes naturally occurring mutations (mutants or variants), such as those resulting from individual or species differences in the microorganisms from which the protein or the gene encoding it is derived.

[0025] When the immunoglobulin-binding domain of FpL is a polypeptide comprising the entire amino acid sequence of the immunoglobulin-binding domain C3 (SEQ ID NO: 1) of Protein L derived from Finegoldia magna, preferred embodiments of the variants (above (c) and (d)) include the polypeptides shown in (α) or (β) below. (α) a polypeptide having an amino acid sequence comprising at least the amino acid residues of the sequence set forth in SEQ ID NO: 1, with the proviso that the amino acid sequence contains substitution, deletion, insertion, or addition of one or several amino acid residues at one or several positions, and having antibody-binding activity; (β) A polypeptide comprising at least the amino acid residues of the sequence set forth in SEQ ID NO: 1, having 70% or more identity to the sequence of amino acid residues, and having antibody binding activity. Examples of (α) include the polypeptide disclosed in JP 2023-064059 A, the polypeptide disclosed in JP 2023-103953 A, and a polypeptide comprising at least amino acid residues consisting of the sequence set forth in SEQ ID NO: 1, with the proviso that the amino acid residues have at least one or more amino acid substitutions selected from the following (1) to (5), and having immunoglobulin-binding activity: (1) The amino acid residue corresponding to tyrosine at position 42 of SEQ ID NO: 1 is substituted with histidine (2) The amino acid residue corresponding to lysine 22 of SEQ ID NO: 1 is substituted with glutamic acid (3) The amino acid residue corresponding to the 38th lysine in SEQ ID NO: 1 is proline or asparagine. Substituted with acetic acid (4) The amino acid residue corresponding to tyrosine at position 42 of SEQ ID NO: 1 is tryptophan or Substituted with phenylalanine (5) The amino acid residue corresponding to asparagine 44 of SEQ ID NO: 1 is substituted with proline As used herein, "an amino acid residue corresponding to the Xth amino acid in the amino acid sequence set forth in SEQ ID NO: 1" refers to an amino acid residue in the specific amino acid sequence, which is arranged at the same position as the Xth amino acid in the amino acid sequence set forth in SEQ ID NO: 1 in an alignment of the specific amino acid sequence with the amino acid sequence set forth in SEQ ID NO: 1. For example, "an amino acid residue corresponding to the 42nd tyrosine in SEQ ID NO: 1" in a specific amino acid sequence refers to an amino acid residue in the specific amino acid sequence, which is arranged at the same position as the 42nd tyrosine in the amino acid sequence set forth in SEQ ID NO: 1 in an alignment of the specific amino acid sequence with the amino acid sequence set forth in SEQ ID NO: 1.

[0026] The immunoglobulin-binding domain of FpL may be a polypeptide monomer according to any one of the above embodiments (a) to (d), or a multimer formed by linking two or more of these monomers. When using such a multimer, the monomers may be linked directly or via a linker peptide. Furthermore, when the monomers are variants (embodiments (c) and (d)), the amino acid sequences of the monomers may or may not be identical.

[0027] The immunoglobulin-binding domain of FpL may contain, for example, at its N- or C-terminus, an oligopeptide useful for specifically detecting or isolating a target substance. Examples of such oligopeptides include polyhistidine and polyarginine. The domain may also contain, for example, at its N- or C-terminus, an oligopeptide useful for immobilizing the domain on a solid phase such as a chromatographic support. Examples of such oligopeptides include oligopeptides containing lysine or cysteine ​​residues.

[0028] The immunoglobulin-binding domain of FpL can be immobilized on an insoluble carrier by covalent bonding, for example, to prepare an antibody adsorbent. Specifically, the antibody adsorbent can be prepared by covalently bonding the protein to the insoluble carrier via an active group on the insoluble carrier, thereby immobilizing the protein on the insoluble carrier. That is, the insoluble carrier may have an active group on its surface, etc.

[0029] Examples of the active group include an N-hydroxysuccinimide (NHS)-activated ester group, an epoxy group, a carboxy group, a maleimide group, a haloacetyl group, a tresyl group, a formyl group, and a haloacetamide group. As the insoluble support having an active group, for example, a commercially available insoluble support having an active group may be used as is, or an insoluble support to which an active group has been introduced may be used. Examples of commercially available supports having an active group include TOYOPEARL AF-Epoxy-650M, TOYOPEARL AF-Tresyl-650M, and TOYOPEARL AF-Formyl-650M (all manufactured by Tosoh Corporation), HiTrap NHS-activated HP Columns, NHS-activated Sepharose 4 Fast Flow, and Epoxy-activated Sepharose 6B (all manufactured by Cytiva), and SulfoLink Coupling Resin (manufactured by Thermo Fisher Scientific).

[0030] An example of a method for introducing active groups onto the support surface is to react one of a compound having two or more active sites with a hydroxy group, epoxy group, carboxy group, amino group, etc. present on the support surface.

[0031] Examples of compounds that introduce epoxy groups into hydroxy groups or amino groups present on the surface of a carrier include epichlorohydrin, ethanediol diglycidyl ether, butanediol diglycidyl ether, and hexanediol diglycidyl ether.

[0032] Furthermore, examples of compounds that introduce carboxy groups into epoxy groups present on the surface of the carrier include 2-mercaptoacetic acid, 3-mercaptopropionic acid, 4-mercaptobutyric acid, 6-mercaptobutyric acid, glycine, 3-aminopropionic acid, 4-aminobutyric acid, and 6-aminohexanoic acid.

[0033] Furthermore, compounds that introduce maleimide groups into hydroxy groups, epoxy groups, carboxy groups, or amino groups present on the surface of a support include N-(ε-maleimidocaproic acid) hydrazide, N-(ε-maleimidopropionic acid) hydrazide, 4-(4-N-maleimidophenyl)acetic acid hydrazide, 2-aminomaleimide, 3-aminomaleimide, 4-aminomaleimide, 6-aminomaleimide, 1-(4-aminophenyl)maleimide, 1-(3-aminophenyl)maleimide, 4-(maleimido)phenyl isocyanate, 2-maleimidoacetic acid, and 3-maleimide. Examples include propionic acid, 4-maleimidobutyric acid, 6-maleimidohexanoic acid, N-(α-maleimidoacetoxy)succinimide ester, (m-maleimidobenzoyl)N-hydroxysuccinimide ester, succinimidyl-4-(maleimidomethyl)cyclohexane-1-carbonyl-(6-aminohexanoic acid), succinimidyl-4-(maleimidomethyl)cyclohexane-1-carboxylic acid, (p-maleimidobenzoyl)N-hydroxysuccinimide ester, and (m-maleimidobenzoyl)N-hydroxysuccinimide ester.

[0034] Furthermore, examples of compounds that introduce haloacetyl groups into hydroxy groups or amino groups present on the surface of a carrier include chloroacetic acid, bromoacetic acid, iodoacetic acid, chloroacetic acid chloride, bromoacetic acid chloride, bromoacetic acid bromide, chloroacetic acid anhydride, bromoacetic acid anhydride, iodoacetic acid anhydride, 2-(iodoacetamido)acetic acid-N-hydroxysuccinimide ester, 3-(bromoacetamido)propionic acid-N-hydroxysuccinimide ester, and 4-(iodoacetyl)aminobenzoic acid-N-hydroxysuccinimide ester.

[0035] Another example of a method for introducing active groups onto the support surface is to react hydroxyl or amino groups present on the support surface with an ω-alkenyl alkane glycidyl ether, followed by activation by halogenating the ω-alkenyl moiety with a halogenating agent. Examples of ω-alkenyl alkane glycidyl ethers include allyl glycidyl ether, 3-butenyl glycidyl ether, and 4-pentenyl glycidyl ether. Examples of halogenating agents include N-chlorosuccinimide, N-bromosuccinimide, and N-iodosuccinimide.

[0036] Another example of a method for introducing active groups onto the support surface is to use a condensing agent and an additive to introduce active groups into carboxy groups present on the support surface. Condensing agents include 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), dicyclohexylcarbodiamide, and carbonyldiimidazole. Additives include N-hydroxysuccinimide (NHS), 4-nitrophenol, and 1-hydroxybenzotriazole.

[0037] Immobilization of the immunoglobulin-binding domain of FpL to an insoluble carrier can be carried out, for example, in a buffer solution. Examples of buffer solutions include acetate buffer, phosphate buffer, MES (2-morpholinoethanesulfonic acid) buffer, HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid) buffer, Tris buffer, and borate buffer. The reaction temperature during immobilization can be appropriately set depending on various conditions, such as the reactivity of the active group and the stability of the domain. The reaction temperature during immobilization may be, for example, 5°C or higher and 50°C or lower, and preferably 10°C or higher and 35°C or lower.

[0038] The method of the present invention comprises: (1) a step of adsorbing polyspecific antibodies contained in a sample onto the antibody adsorbent prepared by the method described above (adsorption step); (2) eluting the multispecific antibody adsorbed to the adsorbent using an elution solution (elution step); The above two steps are described in detail below. Since it is preferable to pack the adsorbent in a column, which simplifies the steps, the following description will be given in the form of a column (hereinafter also referred to as an "antibody separation column").

[0039] (1) Adsorption process A sample containing a multispecific antibody can be brought into contact with the antibody adsorbent by, for example, adding (applying) the sample to an antibody separation column using a liquid delivery means such as a pump. Herein, adding a liquid to a column is also referred to as "delivering (applying) a liquid to a column." The sample containing a multispecific antibody may be solvent-substituted with an appropriate buffer solution before being added to the antibody separation column. Furthermore, before adding the sample to the antibody separation column (i.e., before the adsorption step), the column may be equilibrated with an appropriate buffer solution (equilibration solution). This equilibration can be expected to, for example, result in more accurate purification of multispecific antibodies.

[0040] Examples of buffer solutions used for solvent substitution and equilibration include glycine buffer, phosphate buffer, acetate buffer, succinate buffer, citrate buffer, Tris buffer, HEPES buffer, and MES buffer, which have buffer capacity in the neutral range (referring to a pH range of 5.0 to 9.0 in this specification). The buffer solution used for solvent substitution and the equilibration solution may or may not be the same.

[0041] Before the elution step described below, the antibody adsorbent to which the polyspecific antibodies have been adsorbed may be washed to remove any contaminants remaining in the column that have not been adsorbed to the antibody adsorbent.

[0042] (2) Elution process The present invention is characterized in that when polyspecific antibodies adsorbed to an antibody adsorbent are eluted using an eluent containing chloride ions, the eluent is eluted using a gradient that decreases the chloride ion concentration and pH of the eluent. By lowering the pH of the eluent (i.e., changing it to the acidic side) and also lowering the chloride ion concentration in the eluent, polyspecific antibodies adsorbed to the antibody adsorbent can be separated with high accuracy. The gradient that decreases the chloride ion concentration and pH of the eluent may be changed in two or more steps (stepwise) (step gradient) or may be changed with a linear gradient (linear gradient), with linear gradient elution being preferred. The eluent may be a buffer that contains chloride ions and has a buffering capacity that is more acidic than the buffer used for solvent replacement or equilibration.

[0043] Examples of the eluent include phosphate buffer, glycine-hydrochloric acid buffer, and acetate buffer. The eluent preferably contains chloride ions, and may be prepared by adding salts such as NaCl, MgCl, KCl, and CaCl.

[0044] The chloride ion concentration of the eluate at the start of the gradient may be in the range of 50 mmol / L to 150 mmol / L, and more preferably in the range of 80 mmol / L to 150 mmol / L. In this specification, the "start of the gradient" may mean immediately after the adsorption step is completed.

[0045] The method for changing the pH and chloride ion concentration of the eluate may be determined appropriately depending on the properties of the polyspecific antibodies to be separated.

[0046] The chloride ion concentration of the eluate is preferably decreased from 0 mmol / L to 49 mmol / L, more preferably from 0 mmol / L to 40 mmol / L, and even more preferably from 0 mmol / L to 30 mmol / L, to elute the multispecific antibody.

[0047] The multispecific antibodies may be eluted using a gradient that decreases the pH of the elution solution from pH 2.0 to pH 4.0, more preferably from pH 2.0 to pH 3.5, and even more preferably from pH 2.0 to pH 3.0.

[0048] The eluate may further contain salts or other substances that do not contain chloride ions.

[0049] There is no particular lower limit to the eluate flow rate, and the upper limit depends on the back pressure of the antibody separation column. [Example]

[0050] EXAMPLES The present invention will be described in detail below using examples and comparative examples, but the present invention is not limited to these examples.

[0051] Example 1: Antibody Production (1) Construction of Trastuzumab expression vector (1-1) A polynucleotide (SEQ ID NO: 6) encoding the heavy chain of Trastuzumab consisting of the amino acid sequence set forth in SEQ ID NO: 4, and a polynucleotide (SEQ ID NO: 7) encoding the light chain of Trastuzumab consisting of the amino acid sequence set forth in SEQ ID NO: 5 were synthesized. Of the polynucleotide (SEQ ID NO: 6) encoding the heavy chain of Trastuzumab, the 6 bases at the 5' end are the recognition sequence (GAATTC) for the restriction enzyme EcoRI, the recognition sequence (GCTAGC) for the restriction enzyme NheI is inserted between the variable region and the constant region (corresponding to the nucleotide sequence from positions 424 to 430 of SEQ ID NO: 6), and the 8 bases at the 3' end are the recognition sequence (GCGGCCGC) for the restriction enzyme NotI. In addition, of the polynucleotide (SEQ ID NO: 7) encoding the heavy chain of Trastuzumab, the 6 bases from the 5' end are the recognition sequence (GAATTC) for the restriction enzyme EcoRI, the recognition sequence (CGTACG) for the restriction enzyme BsiWI is inserted between the variable region and the constant region (corresponding to the nucleotide sequence from positions 388 to 393 of SEQ ID NO: 7), and the 8 bases from the 3' end are the recognition sequence (GCGGCCGC) for the restriction enzyme NotI.

[0052] (1-2) Each polynucleotide synthesized in (1-1) was treated with the restriction enzymes EcoRI and NotI, and then subjected to agarose gel electrophoresis to confirm the presence of a band of the desired size. The band of interest was excised and purified using a QIAquick Gel Extraction kit (QIAGEN).

[0053] (1-3) The polynucleotide purified in (1-2) was ligated to the plasmid pEFd (plasmid map shown in Figure 1), which had been previously digested with the restriction enzymes EcoRI and NotI, using a DNA Ligation Kit (Takara Bio Inc.). The ligation product was used to transform Escherichia coli BL21(DE3) strain, which was then cultured (37°C, 16 hours) on a Luria-Bertani (LB) plate containing 50 μg / mL kanamycin.

[0054] (1-4) The transformants (genetically modified E. coli) obtained in (1-3) were selected and cultured in LB medium containing 50 μg / mL of kanamycin. After that, the vectors pEFd-Trastuzumab_VH capable of expressing the heavy chain of Trastuzumab and pEFd-Trastuzumab_VL capable of expressing the light chain of Trastuzumab were obtained by purification using a QIAprep Spin Miniprep kit (QIAGEN).

[0055] (1-5) Of the vectors obtained in (1-4), the polynucleotide encoding the heavy or light chain of Trastuzumab and its surrounding region were subjected to cycle sequencing using a Big Dye Terminator Cycle Sequencing ready Reaction kit (Thermo Fisher Scientific) based on the chain terminator method, and the nucleotide sequence was analyzed using a fully automated DNA sequencer, ABI Prism 3700 DNA analyzer (Thermo Fisher Scientific). For this analysis, an oligonucleotide consisting of the sequence set forth in SEQ ID NO: 12 (5'-TGAGAATTCGCCACCATGACC-3') or SEQ ID NO: 13 (5'-CGGTTCGGGGAAGTAGTCCTT-3') was used as a sequencing primer.

[0056] Sequence analysis confirmed that a polynucleotide consisting of the sequence set forth in SEQ ID NO: 6 was inserted into the expression vector pEFd-Trastuzumab_VH, and a polynucleotide consisting of the sequence set forth in SEQ ID NO: 7 was inserted into the expression vector pEFd-Trastuzumab_VL.

[0057] (2) Construction of Belimumab expression vector (2-1) A polynucleotide (SEQ ID NO: 10) encoding the heavy chain of belimumab consisting of the amino acid sequence set forth in SEQ ID NO: 8, and a polynucleotide (SEQ ID NO: 11) encoding the light chain of belimumab consisting of the amino acid sequence set forth in SEQ ID NO: 9 were synthesized. In the polynucleotide (SEQ ID NO: 10) encoding the heavy chain of belimumab, the 6 bases on the 5' end are the recognition sequence (GAATTC) of the restriction enzyme EcoRI, and the 6 bases on the 3' end are the recognition sequence (GCTAGC) of the restriction enzyme NheI. In addition, in the polynucleotide (SEQ ID NO: 11) encoding the heavy chain of belimumab, the 6 bases on the 5' end are the recognition sequence (GAATTC) of the restriction enzyme EcoRI, and the 8 bases on the 3' end are the recognition sequence (CGTACG) of the restriction enzyme BsiWI.

[0058] (2-2) Of the polynucleotides synthesized in (2-1), the polynucleotide encoding the heavy chain of belimumab (SEQ ID NO: 10) was treated with the restriction enzymes EcoRI and NheI, and the polynucleotide encoding the light chain of belimumab (SEQ ID NO: 11) was treated with the restriction enzymes EcoRI and BsiWI, and the polynucleotides were purified in the same manner as in (1-2).

[0059] (2-3) Of the polynucleotides purified in (2-2), the polynucleotide (SEQ ID NO: 10) encoding the heavy chain of Belimumab was ligated to the vector pEFd-Trastuzumab_VH, which had been previously digested with the restriction enzymes EcoRI and NheI, and the polynucleotide (SEQ ID NO: 11) encoding the light chain of Belimumab was ligated to the vector pEFd-Trastuzumab_VL, which had been previously digested with the restriction enzymes EcoRI and BsiWI, and transformed in the same manner as in (1-3).

[0060] (2-4) The transformants (genetically modified E. coli) obtained in (2-3) were selected, and cultured and plasmid purified in the same manner as in (1-4) to obtain vectors pEFd-Belimumab_VH capable of expressing the heavy chain of Belimumab and pEFd-Belimumab_VL capable of expressing the light chain of Belimumab.

[0061] (2-5) Of the vectors obtained in (2-4), the nucleotide sequences of the polynucleotides encoding the heavy chain or light chain of Trastuzumab and their surrounding regions were analyzed by the method described in (1-5).

[0062] Sequence analysis confirmed that a polynucleotide consisting of the sequence set forth in SEQ ID NO: 10 was inserted into the expression vector pEFd-Belimumab_VH, and a polynucleotide consisting of the sequence set forth in SEQ ID NO: 11 was inserted into the expression vector pEFd-Belimumab_VL.

[0063] (3) Antibody production (3-1) The vectors prepared in (1) and (2) (pEFd-Trastuzumab_VH, pEFd-Trastuzumab_VL, pEFd-Belimumab_VH, and pEFd-Belimumab_VL) were transfected into COS-1 cells using PEI-MAX (Cosmobio Co., Ltd.) and transformed. The vector combinations used for the transformation are as follows: (A) Trastuzumab-expressing cells: pEFd-Trastuzumab_VH and pEFd-Trastuzumab_VL (B) Belimumab-expressing cells: pEFd-Belimumab_VH and pEFd-Belimumab_VL (C) Bispecific antibodies of Trastuzumab and Belimumab (hereinafter also referred to as "BsAbs"): pEFd-Trastuzumab_VH, pEFd-Trastuzumab_VL, pEFd-Belimumab_VH, and pEFd-Belimumab_VL (3-2) Each of the transformed cells obtained in (3-1) was placed in a dish containing a medium prepared by adding 1 / 100th of a penicillin-streptomycin solution (x100) (Fujifilm Wako Pure Chemical Industries, Ltd., Cat. No. 168-23191) to D-MEM medium (high glucose) (containing L-glutamine, phenol red, and sodium pyruvate) (Fujifilm Wako Pure Chemical Industries, Ltd., Cat. No. 043-30085), and the dish was then left to stand in a CO2 incubator (37°C, 5% CO2) to allow the antibody to be secreted and expressed.

[0064] (3-3) The obtained culture supernatant was passed through a filter to clarify it, and then applied to a column packed with TOYOPEARL rProtein A-650F (manufactured by Tosoh Corporation) that had been equilibrated in advance with PBS (phosphate-buffered saline).

[0065] (3-4) After washing the column with PBS in an amount 10 times the volume of the carrier packed in the column, a 100 mmol / L glycine solution adjusted to pH 3.0 with hydrochloric acid was applied to recover fractions corresponding to each expressed antibody.

[0066] (3-5) The purity of each antibody contained in the collected fractions was confirmed by SDS-PAGE (SDS-polyacrylamide gel electrophoresis). The amount of each antibody contained in the fractions was quantified based on the absorbance of the fractions measured with a spectrophotometer, and the fractions were diluted with PBS to a concentration of 0.5 mg / mL to prepare antibody solutions.

[0067] Example 2 Preparation of antibody separation column (1) A polymethacrylate gel (Toyopearl, manufactured by Tosoh Corporation) with formyl groups as active groups on the surface was used as an insoluble carrier. The carrier slurry was filtered through a glass filter and then dried under suction to prepare a suction-dried gel.

[0068] (2) Using the suction-dried gel prepared in (1), the moisture content and bulk density were calculated using the following method. First, the moisture content per 1 g of suction-dried gel was measured using a heat-drying moisture meter, and the weight (g) of the dry gel obtained by removing the moisture from the suction-dried gel was calculated based on this. Next, 1 g of the suction-dried gel was prepared as a water slurry, transferred to a measuring cylinder, and allowed to stand to measure the volume (mL) of the settled gel. The bulk density (g / mL) was calculated by dividing the previously calculated dry gel weight by the settled gel.

[0069] (3) Based on the calculation results in (2), the suction-dried gel prepared in (1) was weighed into an Erlenmeyer flask so that the volume of the insoluble carrier was 2 mL. Appropriate amounts of an immunoglobulin-binding protein solution consisting of the amino acid sequence set forth in SEQ ID NO: 15, 200 mmol / L borate buffer, and 4 mol / L aqueous sodium chloride solution were added to the flask, and the mixture was shaken at 25°C for 4 hours to bind the insoluble carrier and the protein via Schiff base formation. The immunoglobulin-binding protein consisting of the amino acid sequence set forth in SEQ ID NO: 15 is a polypeptide comprising five directly linked immunoglobulin-binding domains of FpL consisting of the amino acid sequence set forth in SEQ ID NO: 14, and an immobilization tag peptide consisting of the amino acid sequence set forth in SEQ ID NO: 16 and an oligopeptide consisting of six histidine residues and three lysine residues attached to the C-terminus. Furthermore, the immunoglobulin-binding domain of FpL consisting of the amino acid sequence set forth in SEQ ID NO: 14 is a polypeptide in which the following amino acid substitutions have occurred in the amino acid residues of the sequence set forth in SEQ ID NO: 1: E4G (this notation indicates that the fourth glutamic acid in SEQ ID NO: 1 has been replaced with glycine; other amino acid substitutions are interpreted similarly hereinafter), P6S, K7A, K13R, K22R, I23R, K29I, K38E, N44R, K48R, E49D, N50Y, Y53F, N62Y, K67R, and A69V.

[0070] (4) After the reaction of (3) was completed, an appropriate amount of 1.1 mol / L dimethylamine borane aqueous solution was added, and the mixture was shaken at 25°C for 2 hours to reduce the remaining Schiff base.

[0071] (5) After the reaction of (4) was completed, an appropriate amount of 3.6 mol / L aqueous 2-aminoethanol solution was added to block the formyl groups remaining in the insoluble carrier, thereby obtaining an antibody adsorbent.

[0072] (6) The obtained gel was packed into a φ4.6×50 mm stainless steel column to prepare an antibody separation column.

[0073] Comparative Example 1 Antibody Separation by pH Gradient (1) The antibody separation column prepared in Example 2 was equilibrated with 100 mmol / L phosphate buffer (pH 7.0) (hereinafter also referred to as "Buffer A"), and then 100 μL of each antibody solution prepared in Example 1 was applied.

[0074] (2) After adding the antibody solution, the antibody separation column was washed by running buffer A for 5 minutes. From 5 to 45 minutes after addition, each antibody adsorbed to the antibody adsorbent was eluted with a gradient of 30% buffer A: 70%-100 mmol / L citrate buffer (pH 2.2) (hereinafter also referred to as "buffer B") to 0% buffer A: 100% buffer B. The eluted antibodies were detected using a UV detector (absorbance at 280 nm).

[0075] The analytical results (chromatogram) of the Trastuzumab solution are shown in FIG. 2, the analytical results of the Belimumab solution in FIG. 3, and the analytical results of the BsAbs in FIG.

[0076] Trastuzumab showed a single peak around 38 minutes after addition of the antibody solution (Peak T in Figure 2). On the other hand, belimumab showed only a slight peak around 8 minutes after addition of the antibody solution (Peak B in Figure 3), and most of the antibody was eluted during column washing (within 5 minutes after addition of the antibody solution).

[0077] For BsAbs, peaks were observed around 8 minutes (Peak 1), 30 minutes (Peak 2), and 37 minutes (Peak 3) after antibody addition (Fig. 4). From the analysis results of Trastuzumab (Fig. 2) and Belimumab (Fig. 3), Peak 1 was identified as the Belimumab peak (Peak B in Fig. 4), and Peak 3 was identified as the Trastuzumab peak (Peak T in Fig. 4). Therefore, the remaining Peak 2 can be said to correspond to the peak of a heterodimer (bispecific antibody) containing one light chain each of Trastuzumab and Belimumab (Peak TB in Fig. 4).

[0078] Example 3: Antibody Separation Using Chloride Ion Concentration and pH Gradient (Part 1) (1) A BsAbs solution was analyzed in the same manner as in Comparative Example 1, except that sodium chloride was added to buffer solution A at 100 mmol / L, 150 mmol / L, or 200 mmol / L. When converted into the chloride ion concentration in the eluate at the start of the gradient, the concentration was 70 mmol / L when 100 mmol / L of sodium chloride was added, 105 mmol / L when 150 mmol / L was added, and 140 mmol / L when 200 mmol / L was added.

[0079] (2) In the obtained chromatogram, the resolution (Rs value) between Peak TB corresponding to the bispecific antibody and Peak T corresponding to Trastuzumab was calculated according to the following formula.

[0080] [Number 1] Rs value = 1.18 × (elution time of peak T - elution time of peak TB) / (half-width of peak TB + half-width of peak T) Comparative Example 2: Antibody Separation Using Chloride Ion Concentration and pH Gradient (Part 2) The BsAbs solution was analyzed in the same manner as in Comparative Example 1, except that 250 mmol / L of sodium chloride was added to buffer A, and the Rs value was calculated by the method described in Example 3(2). When converted to the chloride ion concentration contained in the eluate at the start of the gradient, it becomes 175 mmol / L.

[0081] The analytical results (chromatograms) of Example 3 and Comparative Example 2 are shown in Figure 5. In Figure 5, the result at 0 mmol / L is the same as the chromatogram shown in Figure 4 (Comparative Example 1).

[0082] Table 1 shows the results of calculating the resolution (Rs value) in Example 3 and Comparative Example 2, including when sodium chloride was not added (Comparative Example 1). By performing elution with a gradient that decreases the pH and chloride ions in the eluent, and by setting the chloride ion concentration in the eluent to between 50 mmol / L and 150 mmol / L at the start of the gradient end (Example 3), the Rs value was improved compared to elution with a gradient that decreases only the pH (Comparative Example 1) and when the chloride ion concentration in the eluent exceeded 150 mmol / L at the start of the gradient end (Comparative Example 2), demonstrating that polyspecific antibodies can be separated with higher accuracy.

[0083] [Table 1]

Claims

1. A method for separating a multispecific antibody, comprising the steps of: contacting a sample containing the multispecific antibody with an antibody adsorbent comprising an insoluble carrier and an immunoglobulin-binding domain of Protein L (FpL) derived from bacteria of the genus Finegoldia immobilized on the insoluble carrier, thereby adsorbing the antibody onto the adsorbent; and eluting the antibody adsorbed onto the adsorbent using an elution solution, A method in which the elution solution contains chloride ions, and the elution step comprises eluting the antibody using a gradient that decreases the chloride ion concentration and pH of the elution solution.

2. The method according to claim 1, wherein the chloride ion concentration of the eluate is 50 mmol / L or more and 150 mmol / L or less at the start of the gradient.

3. The method for isolation according to claim 1 or 2, wherein the immunoglobulin-binding domain of FpL is a polypeptide selected from the group consisting of: (a) a polypeptide comprising at least the amino acid residues of the sequence set forth in SEQ ID NO: 1; (b) a polypeptide having an amino acid sequence comprising at least the amino acid residues of the sequence set forth in SEQ ID NO: 1, with the proviso that the amino acid sequence contains substitution, deletion, insertion, or addition of one or several amino acid residues at one or several positions, and having antibody-binding activity; (c) A polypeptide comprising at least the amino acid residues of the sequence set forth in SEQ ID NO: 1, which has 70% or more identity to the sequence of amino acid residues, and which has antibody binding activity.

4. The method of claim 1 or 2, wherein the multispecific antibody is a bispecific antibody.

Images