Method for identifying common l chain for use in production of multispecific antibody

Abstract

The problem addressed by the present invention is to provide a method for identifying a common L chain for use in the production of a multispecific antibody. The present invention provides a method for identifying a common L chain for use in the production of a multispecific antibody, the method comprising: (1) a step for producing a first library containing a plurality of antibodies that recognize a first antigen; (2) a step for producing a second library containing a plurality of antibodies that recognize a second antigen; and (3) a step for selecting, from the first library and the second library, an antibody that recognizes the first antigen and an antibody that recognizes the second antigen that have a common L chain.

Description

Method for identifying common light chains for use in producing multispecific antibodies

[0001] The present invention relates to a method for identifying a common light chain for use in producing a multispecific antibody.The present invention further relates to a method for producing a multispecific antibody.

[0002] An antibody consists of two different genes for an H chain and an L chain, which form a pair of structural units to form an antigen-binding molecule. A multispecific antibody is an artificial protein that simultaneously binds to two or more different antigens. In recent years, there has been an increasing trend in the development of multispecific antibodies as an antibody-related technology.

[0003] One challenge in developing multispecific antibodies is the need for different pairs of heavy and light chains for each specificity, making the expression structure complex. In the case of bispecific antibodies, two types of heavy chains and two types of light chains are required, and obtaining the correct combination requires complex processes during expression and purification. For this reason, efforts are being made to standardize the light chains of antibodies.

[0004] For example, Non-Patent Document 1 describes the discovery of a common L chain by modifying each L chain for two antigens bound by a bispecific antibody. Specifically, the amino acid sequences of each L chain were shuffled to create a chimeric L chain by genetic engineering, which was then co-expressed with each H chain, and an L chain that maintained binding to the target antigen was selected, thereby successfully obtaining a common L chain.

[0005] It is also known to use a light chain library to obtain a common light chain (Non-Patent Document 2 and Non-Patent Document 3). In Non-Patent Document 3, the natural antibody diversification process that occurs in B cells is mimicked, and antibodies with high binding activity are obtained by a step of affinity maturation in which mutations are introduced into light chains directly isolated from the library. In Non-Patent Document 2, the repertoire size of the L chain library is 1.1 x 10 ５ and 1.7 x 10 ６However, in Reference 3, there is no specific information about the size of the light chain library, etc. Furthermore, because both libraries use heavy chains derived from the libraries and attempt to standardize the light chains, there is a drawback in that the versatility of using a common light chain is poor.

[0006] Zenjiro Sampei et al., PLos One 8 e57479 (2013)Nat Biotechnol. 1998 Jul;16(7):677-81.MABS. 2018, VOL. 10, NO. 2, 256-268

[0007] Although efforts have been made to standardize antibody light chains in the production of multispecific antibodies, a highly versatile technology has not yet been established. An object of the present invention is to provide a method for identifying a common light chain to be used in the production of multispecific antibodies. Another object of the present invention is to provide a method for producing multispecific antibodies using the common light chain identified by the above method.

[0008] As a result of studies to solve the above-mentioned problems, the present inventors discovered that a common L chain for use in producing multispecific antibodies can be identified by preparing a first library containing multiple antibodies recognizing a first antigen and a second library containing multiple antibodies recognizing a second antigen, and then selecting antibodies recognizing the first antigen and antibodies recognizing the second antigen, which share a common L chain, from the first and second libraries. One feature of the present invention is that new combinations can be selected by screening the same light chain library, even when using heavy chains derived from mouse immunization or heavy chains derived from human antibodies produced by humanized mice, making the method highly versatile. In the present invention, the binding activity measured using ELISA after screening actually shows a signal intensity equal to or greater than that of the original sequence. In other words, the method of the present invention is highly efficient in that isolated light chains can be used directly. The present invention was completed based on the above findings. According to the present invention, the following inventions are provided:

[0009] <1> A method for identifying a common L chain for use in producing a multispecific antibody, comprising: (1) a step of producing a first library containing a plurality of antibodies that recognize a first antigen, (2) a step of producing a second library containing a plurality of antibodies that recognize a second antigen, and (3) a step of selecting, from the first library and the second library, antibodies that recognize the first antigen and antibodies that recognize the second antigen, which have a common L chain. <2> The method according to <1>, wherein the step (1) is a step of producing a first library by selecting a plurality of antibodies that recognize the first antigen from a library containing a plurality of antibodies that have an H chain and an L chain, and the step (2) is a step of producing a second library by selecting a plurality of antibodies that recognize the second antigen from a library containing a plurality of antibodies that have an H chain and an L chain. <3> A method according to <1>, wherein the library containing a plurality of antibodies that have an H chain and an L chain is selected from 10 ９ <4> The method according to <2>, wherein the library containing a plurality of antibodies having H chains and L chains contains 10 or more types of H chains. ６ <5> The method according to <2>, wherein the number of types of L chains in the library containing a plurality of antibodies having H chains and L chains is 10 or more. ３<6> The method according to <2>, wherein the ratio of the types of H chains to the types of L chains in the library containing a plurality of antibodies having H chains and L chains is 1000 times or more. <7> The method according to <1>, wherein the step (1) comprises: (1-A) a step of incorporating a nucleic acid encoding the H chain of an antibody that recognizes a first antigen into a library of nucleic acids encoding L chains, thereby producing an antibody library containing the H chains and L chains of antibodies that recognize a first antigen, and (1-B) a step of selecting antibodies that recognize the first antigen from the antibody library obtained in step (1-A), and the step (2) comprises: (2-A) a step of incorporating a nucleic acid encoding the H chain of an antibody that recognizes a second antigen into a library of nucleic acids encoding L chains, thereby producing an antibody library containing the H chains and L chains of antibodies that recognize a second antigen, and (2-B) a step of selecting antibodies that recognize the second antigen from the antibody library obtained in step (2-A). <8> The number of types of L chains in the antibody library obtained in the step (1-A) and the antibody library obtained in the step (2-A) is 10, ３ The method according to <7>, wherein: <9> The method according to <1>, wherein the step (1) is a step of producing a first library by selecting a plurality of antibodies that recognize a first antigen from a library containing a plurality of antibodies having an H chain and an L chain, and the step (2) comprises: (2-A) a step of producing an antibody library containing H chains and L chains of antibodies that recognize a second antigen by incorporating a nucleic acid encoding the H chain of an antibody that recognizes a second antigen into a library of nucleic acids encoding the L chain, and (2-B) a step of selecting an antibody that recognizes the second antigen from the antibody library obtained in the step (2-A). <10> The method according to <1>, wherein the first antigen and the second antigen are each selected from a cancer antigen, an antigen of an immune cell, a physiologically active molecule, and an enzyme. <11> A method for producing a multispecific antibody, comprising: a step of identifying an L chain by the method according to any one of <1> to <10>, and a step of producing a multispecific antibody using the L chain identified in the step (1), an H chain that recognizes the first antigen, and an H chain that recognizes the second antigen.

[0010] According to the present invention, it is possible to identify a common light chain for use in producing a multispecific antibody. By standardizing the multiple light chains required for producing a multispecific antibody, it is possible to reduce the number of components of the multispecific antibody and the number of steps required for expression and purification of the multispecific antibody.

[0011] Figure 1 shows a first embodiment of the present invention. Figure 2 shows a second embodiment of the present invention. Figure 3 shows a third embodiment of the present invention. Figure 4 shows the results of identifying a common light chain sequence. Figure 5 shows the results of identifying a common light chain sequence. Figure 6 shows the results of confirming reactivity by ELISA performed on the results of screening from an L chain library into which the H chain of Cetuximab has been inserted. Figure 7 shows the results of confirming reactivity by ELISA performed on the results of screening from an L chain library into which the H chain of Panitumumab has been inserted. Figure 8 shows the results of confirming ELISA reactivity against TfR protein.

[0012] The present invention will now be described in further detail. The present invention relates to a method for identifying a common light chain for use in producing multispecific antibodies, comprising: (1) producing a first library containing a plurality of antibodies that recognize a first antigen, (2) producing a second library containing a plurality of antibodies that recognize a second antigen, and (3) selecting antibodies that recognize the first antigen and antibodies that recognize the second antigen, which have a common light chain, from the first library and the second library.

[0013] A multispecific antibody is an antibody that recognizes two or more antigens. Bispecific antibodies are particularly commonly used. The types of first and second antigens recognized by a multispecific antibody are not particularly limited, and can be selected from, for example, cancer antigens, immune cell antigens, physiologically active molecules, and enzymes, respectively. For example, combinations of first and second antigens include, but are not limited to, a combination of a cancer antigen with a cancer antigen, a combination of a cancer antigen with an immune cell antigen, a combination of a physiologically active molecule with a physiologically active molecule, and a combination of an enzyme with an enzyme.

[0014] Examples of cancer antigens include CD19, CD20, CD33, CD38, CD123, CD135, EpCAM, CEA, CDH3, Her2, HLAA / gp100, PSMA, BCMA, MAGE-A4, PRAME, GPC3, GD2, EGFR, MUC1, MUC16, MUC17, MSLN, EphA2, ROR1, and DLL3.

[0015] Antigens of immune cells include CD3, CD47, OX40, and the like.

[0016] Examples of physiologically active molecules include cytokines, hormones, neurotransmitters, etc. Specific examples include IL-2, IFN-α, IFN-β, IFN-γ, TNF, HMGB1, HMGB2, etc.

[0017] Examples of the enzyme include CD73, CD39, MMPs such as MMP-2 and MMP-9.

[0018] Known antibody libraries include, but are not limited to, libraries in which single-chain antibodies (scFv) are displayed on the surface of phages using phage display, libraries using cell-free translation systems, libraries in which antibodies are displayed on the surface of cells or viruses, and libraries using emulsions. For example, techniques using cell-free translation systems include ribosome display, which forms a complex between mRNA and the translated protein via ribosomes by removing a stop codon, cDNA display, which covalently binds a gene sequence to a translated protein using a compound such as puromycin, and mRNA display, and CIS display, which forms a complex between a gene and the translated protein using a nucleic acid-binding protein. In addition to phage display, techniques for displaying antigen-binding molecules on the surface of cells or viruses include E. coli display, Gram-positive bacteria display, yeast display, mammalian cell display, and viral display. Techniques using emulsions include in vitro viral display, which involves encapsulating genes and translation-related molecules in an emulsion.

[0019] Antibodies that recognize specific antigens can be selected using known methods, such as enzyme-linked immunosorbent assay (ELISA), electrochemiluminescence (ECL), fluorescence activated cell sorting (FACS), and methods using surface plasmon resonance (SPR).

[0020] According to a first aspect of the present invention, step (1) of preparing a first library containing a plurality of antibodies recognizing a first antigen is a step of preparing the first library by selecting a plurality of antibodies recognizing the first antigen from a library containing a plurality of antibodies having an H chain and an L chain, and step (2) of preparing a second library containing a plurality of antibodies recognizing a second antigen is a step of preparing the second library by selecting a plurality of antibodies recognizing the second antigen from a library containing a plurality of antibodies having an H chain and an L chain. Subsequently, a common L chain can be identified by selecting antibodies recognizing the first antigen and antibodies recognizing the second antigen that share a common L chain from the first and second libraries. The first aspect of the present invention is shown in Figure 1. Antigen A (first antigen) and antigen B (second antigen) are screened from the antibody library using a phage display method. Screening results in the isolation of a plurality of clones that bind to each antigen. From the resulting anti-antigen A antibody group and anti-antigen B antibody group, clones sharing a common L chain are selected.

[0021] In a first aspect of the present invention, a library containing a plurality of antibodies having H chains and L chains of 10 ９ It is preferred that the antibody contains more than 2 x 10 ９ More preferably, the antibody contains 5 x 10 or more species. ９ It is more preferable that the library contains 10 or more kinds of antibodies. ６ It is preferable that the ratio is 2×10 or more. ６ More preferably, it is 5××10 ６ More preferably, it is equal to or greater than this.

[0022] In a library containing a plurality of antibodies having H chains and L chains, the number of types of L chains is 10. ３ Preferably, the ratio is 1:1 or less, more preferably 700 or less, and even more preferably 500 or less. Furthermore, in the library containing a plurality of antibodies having H chains and L chains, the ratio of the types of H chains to the types of L chains is preferably 1:1 or more, more preferably 5:1 or more, and even more preferably 10:1 or more.

[0023] As described above, by using a library with a limited number of L chain types, it is more likely that L chains that also react with antigen A (first antigen) and antigen B (second antigen) can be obtained.

[0024] According to a second aspect of the present invention, the step (1) comprises: (1-A) incorporating a nucleic acid encoding the heavy chain of an antibody recognizing a first antigen into a library of nucleic acids encoding light chains, thereby producing an antibody library containing heavy and light chains of an antibody recognizing a first antigen; and (1-B) selecting an antibody that recognizes the first antigen from the antibody library obtained in the step (1-A). The step (2) comprises: (2-A) incorporating a nucleic acid encoding the heavy chain of an antibody recognizing a second antigen into a library of nucleic acids encoding light chains, thereby producing an antibody library containing heavy and light chains of an antibody that recognizes a second antigen; and (2-B) selecting an antibody that recognizes the second antigen from the antibody library obtained in the step (2-A). The second aspect of the present invention is shown in Figure 2. In the second aspect, the heavy chain portion of an anti-antigen A antibody is introduced into the light chain library. Similarly, the heavy chain portion of an anti-antigen B antibody is introduced into the light chain library. Antigen A (first antigen) and antigen B (second antigen) are screened from respective antibody libraries using phage display. Screening results in the production of multiple clones that bind to each antigen. Clones with a common light chain are selected from the resulting anti-antigen A antibody group and anti-antigen B antibody group.

[0025] In the second aspect of the present invention, the number of types of L chains in the antibody library obtained in step (1-A) and the antibody library obtained in step (2-A) is 10 or more. ３ Preferably, it is 700 or less, more preferably 700 or less, and even more preferably 500 or less.

[0026] According to a third aspect of the present invention, the step (1) is a step of producing a first library by selecting a plurality of antibodies that recognize a first antigen from a library containing a plurality of antibodies having an H chain and an L chain, and the step (2) comprises: (2-A) a step of incorporating nucleic acid encoding the H chain of an antibody that recognizes a second antigen into a library of nucleic acid encoding an L chain to produce an antibody library containing the H chain and L chain of an antibody that recognizes the second antigen, and (2-B) a step of selecting an antibody that recognizes the second antigen from the antibody library obtained in the step (2-A). The third aspect of the present invention is shown in Figure 3.

[0027] The method for preparing the L chain (light chain) library used in the present invention is not particularly limited. Germline genes of human antibodies are published in "THE INTERNATIONAL IMMUNOGENETICS INFORMATION SYSTEM (IMGT: https: / / www.imgt.org / )." When the subgroups of the light chain variable regions were confirmed, 35-40 types of IGKV were found as functional clusters for kappa chains (breakdown: IGKV1: 17 (+2) species, IGKV2: 9 (+1) species, IGKV3: 5 (+2) species, IGKV4: 1 species, IGKV5: 5 species, IGKV6: 1 species, IGKV7: 1 species, IGKV8: 1 species, IGKV9: 1 species, IGKV10: 1 species, IGKV11: 1 species, IGKV12: 1 species, IGKV13: 1 species, IGKV14: 1 species, IGKV15: 1 species, IGKV16: 1 species, IGKV17: 1 species, IGKV18: 1 species, IGKV19: 1 species, IGKV20: 1 species, IGKV21: 1 species, IGKV22: 1 species, IGKV23: 1 species, IGKV24: 1 species, IGKV25: 1 species, IGKV26: 1 species, IGKV27: 1 species, IGKV28: 1 species, IGKV29: 1 species, IGKV30: 1 species, IGKV31: 1 species, IGKV32: 1 species, IGKV33: 1 species, IGKV34: 1 species, IGKV35: 1 species, IGKV36: 1 species, IG In the lambda chain, there are 29-30 IGLV species (IGLV1: 5 species, IGLV2: 5 species, IGLV3: 8 species, IGLV4: 3 species, IGLV5: 3-4 species, IGLV6: 1 species, IGLV7: 1 species, IGLV8: 1 species, IGLV9: 1 species, IGLV10: 1 species, IGLV11: 0 species) (Williams, S.C. et al. J. Mol. Biol., 264, 220-232 (1996), Kawasaki, K. et al. Genome Res., 7, 250-261 (1997)).

[0028] With reference to the above, the sequences in Table 1 (SEQ ID NOS: 1 to 39) below can be designed as kappa chain basic germline amino acid sequences, and the sequences in Table 2 (SEQ ID NOS: 40 to 70) below can be designed as lambda chain basic germline amino acid sequences. The sequences contained in the above-mentioned kappa chain basic germline amino acid sequences and lambda chain basic germline amino acid sequences are V regions and do not include J regions or CL regions. Therefore, the light chain J region can be provided to the light chain variable region with reference to the human J gene sequence (SEQ ID NOS: 71 to 83 in Tables 3 and 4), and the light chain constant region can be provided to the light chain variable region with reference to the CL gene sequence (SEQ ID NOS: 84 and 85).

[0029] Furthermore, to provide diversity to the light chain basic sequence, the proportions of subgroups of family genes can be adjusted to match the proportions present in the human body, and mutations or additional sequences can be artificially inserted into FR1, FR2, FR3, and FR4 of the frame region or CDR1, CDR2, and CDR3 of the complementarity-determining regions. Specifically, the proportions of kappa chain and lambda chain gene families present in the body can be determined by referring to previous studies (Griffiths AD, et al. (1994), Prabakaran P, et al. (2012), etc.). For example, for kappa chains (IGKV1-IGKV6), the proportion of each subgroup in the library can be approximately IGKV1: 36%, IGKV2: 11%, IGKV3: 35%, IGKV4: 11%, IGKV5: 4%, and IGKV6: 3%; and for lambda chains (IGLV1-IGLV10), the proportion of each subgroup in the library can be approximately IGLV1: 40%, IGLV2: 16%, IGLV3: 21%, IGLV4: 3%, IGLV5: 6%, IGLV6: 6%, IGLV7: 2%, IGLV8: 2%, IGLV9: 2%, and IGLV10: 2%. Regarding CDR3, which is thought to be most involved in antigen recognition, the length can be set to an average of 9.2 amino acids with a width of 7 to 11 amino acids for kappa chains, and an average of 10.4 amino acids with a width of 8 to 12 amino acids for lambda chains. The GRAVY index (Grand Average of Hydropathicity), which is an index of hydrophobicity, can be set to an average of -1.00 for kappa chains and an average of -0.38 for lambda chains. Furthermore, the number of nonpolar amino acids (A, C, F, G, I, L, M, P, V, W, Y) contained in CDR3 can be designed to be an average of 3.9 for kappa chains and an average of 4.7 for lambda chains.

[0030] As described above, the L chain (light chain) library used in the present invention is preferably one that has the following characteristics: it reflects the naive repertoire, the L chain gene family has a composition ratio that is consistent with that in vivo, and contains artificial mutations.

[0031] According to the present invention, there is provided a method for producing a multispecific antibody, comprising: identifying an L chain by the method of the present invention; and producing a multispecific antibody using the L chain identified above, an H chain that recognizes a first antigen, and an H chain that recognizes a second antigen.

[0032] In the present invention, multispecific antibodies are produced using a common L chain, which reduces the number of combinations of L chains and H chains, making it easier to express and purify multispecific antibodies.

[0033] In the process of producing a multispecific antibody using an L chain, an H chain recognizing a first antigen, and an H chain recognizing a second antigen, DNA encoding the L chain, the H chain recognizing the first antigen, and the H chain recognizing the second antigen is inserted into an expression vector to create an expression vector, which is then expressed in host cells. The antibodies can then be obtained by collecting and purifying the secreted supernatant. To produce bispecific antibodies, such as bispecific antibodies, which contain two different H chain-L chain pairs (an H chain-L chain pair recognizing a first antigen and an H chain-L chain pair recognizing a second antigen), it is necessary to separately purify the antibody consisting of two identical H chain-L chain pairs recognizing the first antigen and the antibody consisting of two identical H chain-L chain pairs recognizing the second antigen. Although separate purification is still required even when a common L chain is used, techniques for efficiently producing bispecific antibodies have been established through previous research ('Knobs-into-holes' Protein Eng. 1996 Jul;9(7):617-21.). These findings suggest that expression and purification efficiency can be improved by combining the 'knobs-into-holes' technology with antibody design using a common L chain.

[0034] Expression vectors include plasmids, retroviruses, adenoviruses, adeno-associated viruses (AAV), plant viruses such as cauliflower mosaic virus and tobacco mosaic virus, cosmids, YACs, and EBV-derived episomes. Expression vectors and expression control sequences are selected to be compatible with the host cells used for expression. The antibody light chain gene and the antibody heavy chain gene can be inserted into separate vectors, or both genes can be inserted into the same expression vector. The antibody gene is inserted into the expression vector by standard methods (e.g., ligation of complementary restriction sites on the antibody gene fragment and vector, or blunt-end ligation if no restriction sites are present).

[0035] A convenient vector encodes a functionally complete human CH or CL immunoglobulin sequence with appropriate restriction sites engineered to allow for easy insertion and expression of any VH or VL sequence, as described above. In such vectors, splicing typically occurs between the splice donor site in the inserted J region and the splice acceptor site preceding the human C domain, as well as at splice regions present within the human CH exon. Polyadenylation and transcription termination occur at native chromosomal sites downstream of the coding region. The recombinant expression vector can also encode a signal peptide that facilitates secretion of the antibody chain from host cells. The antibody chain gene can be cloned into the vector such that the signal peptide is linked in-frame to the amino terminus of the immunoglobulin chain. The signal peptide can be an immunoglobulin signal peptide or a heterologous signal peptide (i.e., a signal peptide from a non-immunoglobulin protein).

[0036] In addition to the antibody genes and control sequences, antibody expression vectors may contain additional sequences, such as sequences that control replication of the vector in host cells (e.g., origins of replication) and a selectable marker gene. The selectable marker gene facilitates selection of host cells into which the vector has been introduced. For example, typically the selectable marker gene confers resistance to drugs, such as G418, hygromycin, or methotrexate, on host cells into which the vector has been introduced. Preferred selectable marker genes include the dehydrofolate reductase (DHFR) gene (for use in dhfr- host cells with methotrexate selection / amplification), the neomycin phosphotransferase gene (for G418 selection), and the glutamate synthetase gene.

[0037] Host cells are transformed with the antibody gene expression vector prepared by the above method. Host cells may be any cell capable of producing antibodies, such as bacteria, yeast, animal cells, insect cells, or plant cells, but animal cells are preferred. Examples of animal cells include Chinese hamster ovary cells (CHO / dhfr(-) cells, CHO / DG44 cells), monkey-derived COS cells (A. Wright & S.L. Morrison, J. Immunol. 160, 3393-3402 (1998)), and SP2 / O cells (mouse myeloma) (K. Motmans et al., Eur. J. Cancer Prev. 5, 512-5199 (1996), R.P. Junghans et al., Cancer Res. 50, 1495-1502 (1990)). For transformation, the lipofectin method (R.W.M. Solone et al., Proc. Natl. Acad. Sci. USA 86, 6007 (1989), P.L. Felgner et al., Proc. Natl. Acad. Sci. USA 84, 7413 (1987)), electroporation, the calcium phosphate method (F.L. Graham & A.J. van der Eb, Virology 52, 456-467 (1973)), the DEAE-Dextran method, etc. are preferably used.

[0038] After culturing the transformant, the antibody is isolated from the cells of the transformant or from the culture medium. The antibody can be isolated and purified by an appropriate combination of methods such as centrifugation, ammonium sulfate fractionation, salting out, ultrafiltration, affinity chromatography, ion exchange chromatography, and gel filtration chromatography.

[0039] <Example of antibody> The present invention provides a CD3ε antibody obtained in the Examples described below. That is, the present invention provides an antibody comprising a heavy chain variable region consisting of the amino acid sequence of SEQ ID NO: 97 and a light chain variable region consisting of the amino acid sequence of SEQ ID NO: 99. The present invention also provides an antibody comprising a heavy chain variable region consisting of the amino acid sequence of SEQ ID NO: 97 and a light chain consisting of the amino acid sequence of SEQ ID NO: 98. SEQ ID NO: 97: CD3ε antibody_0039 heavy chain variable region QVQLVQSGAEVKKPGASVKVSCKASGYTFTGYYMHWVRQAPGQGLEWMGWINPNSGGTNYAQKFQGRVTMTRDTSISTAYMELSRLRSDDTAVYYCAREMSNWDDAFDIWGQGTMVTVSR SEQ ID NO: 98: CD3ε antibody_0039 light chain SYELTQPPSVSVAPGKTARITCGGNNIGSKSVHWYQQKPGQAPVLVVYDDSDRPSGIPERFSGSNSGNTATLTISRVEAGDEADYYCQVWDSSSDHYVFGTGTKVTVLGQPKANPTVTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKADGSPVKAGVETTKPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEKTVAPTECS Additional CD3 light chain variable region sequence SEQ ID NO: 99: CD3ε antibody_0039 light chain variable region SYELTQPPSVSVAPGKTARITCGGNNIGSKSVHWYQQKPGQAPVLVVYDDSDRPSGIPERFSGSNSGNTATLTISRVEAGDEADYYCQVWDSSSDHYVFGTGTKVTVLG

[0040] The present invention will be described in detail with reference to the following examples, but the present invention is not limited to these examples.

[0041] <Construction of a light chain library> (1) Design of light chains Human antibody germline genes are published in "THE INTERNATIONAL IMMUNOGENETICS INFORMATION SYSTEM (IMGT: https: / / www.imgt.org / )." When the subgroups of the light chain variable region were confirmed, For kappa chains, there were 35-40 IGKV types in functional clusters (IGKV1: 17(+2) species, IGKV2: 9(+1) species, IGKV3: 5(+2) species, IGKV4: 1 species, IGKV5: 5 species, IGKV6: 2 species, IGKV7: - species), and for lambda chains, there were 29-30 IGLV types (IGLV1: 5 species, IGLV2: 5 species, IGLV3: 8 species, IGLV4: 3 species, IGLV5: 3-4 species, IGLV6: 1 species, IGLV7: 1 species, IGLV8: 1 species, IGLV9: 1 species, IGLV10: 1 species, IGLV11: - species) (Williams, S.C. et al. J. Mol. Biol. , 264, 220-232 (1996), Kawasaki, K. et al. Genome Res. , 7, 250-261 (1997)).

[0042] Therefore, the basic germline genes were set by referring to this (Tables 1 and 2).

[0043]

[0044]

[0045] This basic amino acid sequence contains the V region, but does not contain the J region or CL region. Therefore, the light chain J region was assigned to the light chain variable region based on the human J gene sequence (Tables 3 and 4), and the light chain constant region was assigned to the light chain variable region based on the CL gene sequence (SEQ ID NOs: 84 and 85).

[0046]

[0047]

[0048] SEQ ID NO: 84 TVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC SEQ ID NO: 85 QPKANPTVTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKADGSPVKAGVETTKPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEKTVAPTECS

[0049] To provide diversity to the light chain basic sequence described above, the proportions of family gene subgroups were adjusted to match the proportions present in the human body, and mutations or additional sequences were artificially inserted into FR1, FR2, FR3, and FR4 of the frame region and CDR1, CDR2, and CDR3 of the complementarity-determining regions. Specifically, the proportions of each subgroup present in the human body for the kappa chain and lambda chain gene families were determined with reference to previous studies (Griffiths AD, et al. (1994), Prabakaran P, et al. (2012), etc.). That is, for the kappa chains (IGKV1-IGKV6), the proportion of each subgroup in the library was approximately IGKV1: 36%, IGKV2: 11%, IGKV3: 35%, IGKV4: 11%, IGKV5: 4%, and IGKV6: 3%. For the lambda chains (IGLV1-IGLV10), the proportion of each subgroup in the library was approximately IGLV1: 40%, IGLV2: 16%, IGLV3: 21%, IGLV4: 3%, IGLV5: 6%, IGLV6: 6%, IGLV7: 2%, IGLV8: 2%, IGLV9: 2%, and IGLV10: 2%. Regarding CDR3, which is thought to be most involved in antigen recognition, the length was set to an average of 9.2 amino acids with a range of 7 to 11 amino acids for kappa chains and an average of 10.4 amino acids with a range of 8 to 12 amino acids for lambda chains. The GRAVY index (Grand Average of Hydropathicity), which is an index of hydrophobicity, was set to an average of -1.00 for kappa chains and -0.38 for lambda chains. Furthermore, the number of nonpolar amino acids (A, C, F, G, I, L, M, P, V, W, Y) contained in CDR3 was designed to be an average of 3.9 for kappa chains and an average of 4.7 for lambda chains.

[0050] <References> Lefranc, M.-P.; Lefranc, G. Immunoglobulins or Antibodies: IMGT TMBridging Genes, Structures and Functions. Biomedicines 2020, 8, 319. Prabakaran P, Chen W, Singarayan MG, Stewart CC, Streaker E, Feng Y, Dimitrov DS. Expressed antibody repertoires in human cord blood cells: 454 sequencing and IMGT / High V-QUEST analysis of germline gene usage, junctional diversity, and somatic mutations. Immunogenetics. 2012 May;64(5):337-50. Foster SJ, Brezinschek HP, Brezinschek RI, Lipsky PE. Molecular mechanisms and selective influences that shape the k gene repertoire of IgM+ B cells. J Clin Invest 1997;99:1614-27. Farner NL, Dorner T, Lipsky PE. Molecular mechanisms and selection influence the generation of the human V lambda J lambda repertoire. J Immunol 1999;162:2137-45 Griffiths AD, et al. EMBO J. 1994;13(14):3245-3260.

[0051] (2) Gene synthesis As described above, with reference to the sequences of light chain germline genes, gene synthesis was carried out for 200 types of kappa chain and 200 types of lambda chain gene sequences containing the light chain variable regions, V region, J region, and light chain constant region, in the ratio set in (1). Restriction enzymes NcoI and AscI were designed at the termini as junction sites for the vector.

[0052] (3) Library Construction To construct a human antibody library, samples containing human B cells, such as normal human bone marrow-derived mononuclear cells, normal human peripheral blood-derived mononuclear cells, and human umbilical cord blood, were prepared.

[0053] RNA extraction: 5-10x10 prepared cells ７ To the cells, 1 mL of TRIzol Reagent (Thermo Fisher Scientific) was added to disrupt the cells, followed by the addition of 50 μL of 4-bromoanisole. After incubation, the mixture was centrifuged to recover 600 μL of the upper aqueous layer containing RNA. 70% ethanol was added, and the sample was then transferred to a spin cartridge. The spin column was washed with 350 μL of Wash Buffer I and 500 μL of Wash Buffer II, and eluted with 100 μL of RNase-free water.

[0054] cDNA Synthesis 5 μL of RNA was precipitated with ethanol, and the RNA precipitate was suspended in a premixed solution (1 μL of 50 μM Oligo d(T)20 primer, 1 μL of 10 mM dNTP Mix, and 11 μL of DEPC-water). Then, a previously prepared solution (5XSSIV RT Buffer 4 μL, 100 mM DTT 1 μL, RNaseOUT Recombinant RNase Inhibitor 1 μL, SuperScript IV Reverse Transcriptase (200 U / μL) 1 μL, Thermo Fisher Scientific) was added, and the mixture was reacted at 50°C for 10 minutes and 80°C for 10 minutes. After the reaction, the mixture was purified using a PCR Purification Kit (QIAGEN), and the cDNA was eluted in 100 μL.

[0055] Amplification of heavy chain fragments: PCR was performed using the cDNA prepared above as a template. 15 ng of template cDNA was used, and Q5 (NEW ENGLAND BioLabs) polymerase was used. Primers used were seven forward primers (VH1-VH7) that specifically amplify the heavy chain variable region, and four reverse primers (JH1-2, 3, 4-5, and 6) that include six families, each at a final concentration of 0.5 μM. (Marks JD et al. J. Mol. Biol. (1991) 222, 581-597; Campbell, M et al. Mol. Immunol. (1992) 29, 193-203.)

[0056] VH1 GTCCTCGCAACTGCGGCCCAGCCGGCCATGGCCCAGGTGCAGCTGGTGCAGTCTGG (SEQ ID NO: 86) VH2 GTCCTCGCAACTGCGGCCCAGCCGGCCATGGCCCAGRTCACCTTGAAGGAGTCTGGTCC (SEQ ID NO: 87) VH3 GTCCTCGCAACTGCGGCCCAGCCGGCCATGGCCGAGGTGCAGCTGGTGGAGTCTGG (SEQ ID NO: 88) VH4 GTCCTCGCAACTGCGGCCCAGCCGGCCATGGCCCAGGTGCAGCTGCAGGAGTCGGG (SEQ ID NO: 89) VH4-2 GTCCTCGCAACTGCGGCCCAGCCGGCCATGGCCCAGGTGCAGCTACAGCAGTGGGG (SEQ ID NO: 90) VH6 GTCCTCGCAACTGCGGCCCAGCCGGCCATGGCCCAGGTACAGCTGCAGCAGTCAGG (SEQ ID NO: 91) VH7 GTCCTCGCAACTGCGGCCCAGCCGGCCATGGCCCAGGTGCAGCTGGTGCAATCTGGGTCTGAGT (SEQ ID NO: 92)

[0057] Human heavy chain J region primers JH1-2 GGTGGAGGCACTCGAGACGGTGACCAGGGTGC (SEQ ID NO: 93) JH3 GGTGGAGGCACTCGAGACGGTGACCATTGTCC (SEQ ID NO: 94) JH4-5 GGTGGAGGCACTCGAGACGGTGACCAGGGTTC (SEQ ID NO: 95) JH6 GGTGGAGGCACTCGAGACGGTGACCGTGGTCC (SEQ ID NO: 96)

[0058] The reaction was carried out at 94°C for 30 seconds, 65°C for 30 seconds, and 74°C for 2 minutes for 30 cycles, and the amplified DNA fragment was collected. In this way, a series of procedures from RNA extraction to cDNA synthesis and heavy chain fragment amplification were repeated, and approximately 5 to 10 x 10 immune cells were prepared. ７ Antibody fragments were prepared from the cells.

[0059] Construction of Heavy Chain Library An antibody library was constructed by incorporating the amplified DNA fragments into a vector (phagemid vector) for phage display. The phagemid was designed to encode a signal sequence, heavy chain antibody sequence, linker sequence, light chain antibody sequence, and M13 phage cp3 protein, and was composed of an origin of replication for E. coli amplification and an ampicillin resistance gene for drug selection. To recombine the heavy chain variable region fragments into the vector, the PCR-amplified fragments were treated with Sfi-I and Xho-I restriction enzymes, while the vector was treated in the same way. The antibody fragments and vector were then ligated, and the ligation product was introduced into E. coli as follows to obtain a transformant. The ligation product was precipitated with ethanol and dissolved in QW. A 2 μL aliquot was suspended in 20 μL of DH12S (Thermo Fisher Scientific) and electroporated using the Gene Pulser Xcell (Bio-Rad) bacterial program (25 μF, 200 Ω, 1.8 kV, cuvette width 0.1 cm). The entire ligation product was then transformed into the transfected E. coli, which was then cultured in 2xYT medium containing ampicillin. After overnight incubation, the cells were collected and the plasmid was purified.

[0060] Insertion of Light Chain Antibody Genes An antibody library was constructed by combining the heavy chain antibody library constructed above with artificially synthesized light chain antibodies. First, the 400 synthesized light chain antibody fragments were treated with NcoI and AscI restriction enzymes and purified using a PCR Purification Kit (QIAGEN). Similarly, the heavy chain antibody library was treated with NcoI and AscI restriction enzymes, and vector-sized DNA fragments were excised by agarose electrophoresis. These fragments were recovered using a GEL Extraction Kit (QIAGEN), purified by ethanol precipitation, and then dissolved in QW. The light chain fragments and heavy chain antibody library were then ligated under the following conditions: 150 μg of light chain fragments and 100 μg of the heavy chain library were reacted overnight at 15°C with 10 μL of T4 Ligase (Takara). The DNA was then purified by ethanol precipitation and dissolved in QW.

[0061] A 2 μL aliquot of the ligation product was suspended in 20 μL of DH12S (Thermo Fisher Scientific) and electroporated using the Gene Pulser Xcell (Bio-Rad) bacterial program (25 μF, 200 Ω, 1.8 kV, cuvette width 0.1 cm). The entire ligation product was then transformed, and the transfected E. coli was cultured in 2xYT medium containing ampicillin. The number of transformants was calculated by diluting the electroporated E. coli and plating it on agar medium, and the number was calculated to be 1 x 10 by the method described above. １０ Transformants were obtained.

[0062] (4) Preparation of Phage Antibody Library The transformed E. coli was diluted into 2 L of 2xYT medium (2xYTGA) supplemented with 0.05% glucose and 100 μg / mL ampicillin and cultured overnight at 30°C. 625 mL of the overnight-cultured E. coli was then added to 5 L of 2xYTGA, and cultured at 37°C. The absorbance at 600 nm was measured and allowed to grow until the absorbance reached 1.0. The culture was infected with M13KO7 helper phage solution (NEW ENGLAND BioLabs) at an MOI of 10 per flask and cultured at 37°C for 2 hours. 2xYTGA was then added to the culture to a total volume of 12 L, and the culture was again cultured overnight at 30°C. The next day, the culture solution was centrifuged at 8,000 rpm for 10 minutes, and 1 / 6 volume of 20% polyethylene glycol / 2.5 M NaCl was added to the recovered culture supernatant and stirred. After that, it was centrifuged at 8,000 rpm for 20 minutes, and the precipitate was suspended in PBS, and this was used as a phage solution.

[0063] <Preparation of antibodies against CD3> (1) Preparation of CD3 antigen Based on the amino acid information of human CD3ε (Genebank accession number NP_000724), a gene encoding 126 amino acids (amino acid residues 1-126) followed by a gene encoding an 8xHis tag was designed and artificially synthesized. This synthetic gene was inserted into the pCXN3 vector using restriction enzymes to prepare a human CD3ε expression vector. The expression vector was transiently transfected into FreeStyle-293F cells (Invitrogen) and then cultured with shaking for one week. The culture supernatant collected from this was used as a sample and purified by passing it through a His-Trap HP column (Cytiva) to prepare the extracellular domain CD3ε-His protein.

[0064] (2) Screening of antibodies against the extracellular domain CD3ε Screening was performed from a phage antibody library using purified human CD3ε extracellular domain protein. Antigen was prepared in PBS to a concentration of 20 μg / mL, added to ImmunoModule / Strip (NUNC) at 100 μL / well, and left overnight at 4°C. The antigen solution was then discarded, and 200 μL / well of blocking solution (2.5% skim milk / PBS) was added, followed by blocking at room temperature for 2 hours.

[0065] Next, 1 × 10 ImmunoModule / Strip prepared with 2.5% skim milk / PBS was applied to the strip from which the blocking solution had been removed. １３ The phage antibody was added and reacted for 2 hours at room temperature. After the reaction was completed, the plate was washed 10 times with PBS, and 100 μL / well of 0.2 M glycine HCl (pH 3.0) was added to recover the phage. The recovered phage was infected with 20 mL of E. coli DH12S at OD 0.5 for 1 hour, and a portion of the infected phage was plated on an ampicillin LBG plate to calculate the titer of the recovered phage. The phage-infected E. coli was cultured overnight at 30°C in 100 mL of 2xYTGA medium (2xYT, 200 μg / mL ampicillin sulfate, 1% glucose). 10 mL of this overnight culture was mixed with 100 mL of 2xYTA medium (2xYT, 200 μg / mL ampicillin sulfate), and after culturing at 37°C for 1.5 hours, helper phage KO7 was added at 1x10 11 After incubation at 37°C for 1 hour, 400 mL of 2xYTGAK (2xYT, 200 μg / mL ampicillin sulfate, 0.05% glucose, 50 μg / mL kanamycin) was added and incubated overnight at 30°C. This was centrifuged at 8000 rpm for 10 minutes to prepare 500 mL of supernatant, which was then mixed with 100 mL of PEG solution (20% polyethyleneglycol 6000, 2.5 M NaCl) and stirred well, followed by centrifugation at 8000 rpm for 10 minutes to precipitate the phage. This was suspended in 10 mL of PBS, and a portion was used to count the number of E. coli infections. This was used as the phage for the first screening.

[0066] Next, a second screening was performed again using purified extracellular domain protein of human CD3ε. As in the first screening, the antigen was immobilized on ImmunoModule / Strip, and after standing overnight at 4°C, the antigen solution was discarded and blocking was performed at room temperature for 2 hours with 200 μL / well of blocking solution (2.5% skim milk / PBS). After that, 1 × 10 ImmunoModule / Strip prepared with 2.5% skim milk / PBS was applied to the ImmunoModule / Strip from which the blocking solution had been removed. １１ The first screening phage antibody solution was added and reacted for 2 hours at room temperature. After the reaction was completed, the plate was washed 20 times with PBS, and 100 μL / well of 0.2 M glycine HCl (pH 3.0) was added to recover the phage. The recovered phage was infected with 10 mL of E. coli DH12S at OD 0.5 for 1 hour, and a portion of the infected phage was plated on an ampicillin LBG plate to calculate the titer of the recovered phage. The phage-infected E. coli was cultured overnight at 30°C in 50 mL of 2xYTGA medium (2xYT, 100 μg / mL ampicillin sulfate, 1% glucose). 5 mL of this overnight culture was mixed with 50 mL of 2xYTA medium (2xYT, 200 μg / mL ampicillin sulfate), and after culturing at 37°C for 1.5 hours, helper phage KO7 was added at 1x10 ９ After incubation at 37°C for 1 hour, 200 mL of 2xYTGAK (2xYT, 100 μg / mL ampicillin sulfate, 0.05% glucose, 50 μg / mL kanamycin) was added and the mixture was incubated overnight at 30°C. This was centrifuged at 8000 rpm for 10 minutes to prepare 250 mL of supernatant, which was then mixed with 50 mL of PEG solution (20% polyethyleneglycol 6000, 2.5 M NaCl) and stirred well, followed by centrifugation at 8000 rpm for 10 minutes to precipitate the phage. This was suspended in 50 mL of PBS, and a portion was used to count the number of E. coli infections. This was used as the phage for the second screening.

[0067] Next, in the third screening, the antigen was immobilized in the same manner as in the second screening, and the second screening phage solution was １０ The series of operations was repeated using

[0068] (3) Identification of Positive Clones Positive clones were identified by ELISA using human CD3ε extracellular domain protein. Specifically, the antigen was adjusted to 10 μg / mL with PBS and added to ImmunoModule / Strip (NUNC) at 100 μL / well and allowed to stand at 37°C for 2 hours. The antigen solution was then discarded, and blocking solution (2.5% skim milk / PBS) was added at 200 μL / well, followed by blocking at 37°C for 2 hours. The blocking solution was removed, and the cells were washed with PBS. The monoclonal culture supernatant prepared from the third screening was added at 100 μL / well, and the cells were incubated at 37°C for 1 hour. After washing five times with PBS, 100 μL / well of 1 μg / mL Moise anti-cp3 antibody diluted with PBS / 0.05% Tween 20 was added and incubated at 37°C for 1 hour. After washing five times with PBS, 100 μL / well of HRP-labeled anti-Mouse IgG (H+L) antibody (MBL) diluted 2000-fold with PBS / 0.05% Tween 20 was added and incubated at 37°C for 1 hour. After washing five times with PBS, 100 μL / well of OPD in 0.1 M citrate phosphate buffer (pH 5.1) + 0.01% H2O2 was added and incubated at room temperature for 5 minutes. The color reaction was stopped by adding 100 μL / well of 2NH2SO4. The absorbance at 492 nm was then measured using an infinite F200 (TECAN). Based on the results, antibody clones showing a positive reaction to the human CD3 antigen protein were selected, and their DNA sequences were analyzed to identify the amino acid sequences of six antibody clones.

[0069] <Preparation of Antibody to CD73> (1) Preparation of CD73 Antigen Based on the amino acid information of human CD73 (Genebank accession number AAH65937), a gene encoding 547 amino acids (amino acid residues 1-547) followed by a gene encoding an 8xHis tag was designed and artificially synthesized. This synthetic gene was inserted into the pCXN3 vector using restriction enzymes to prepare a human CD73 expression vector. The expression vector was transiently transfected into FreeStyle-293F cells (Invitrogen) and then cultured with shaking for one week. The culture supernatant collected from this was used as a sample and purified by passing it through a His-Trap HP column (Cytiva) to prepare CD73-His protein.

[0070] (2) Screening of antibodies against CD73 Screening was performed from a phage antibody library using human CD73 antigen protein. The procedure was the same as for screening against CD3, and this was repeated up to the third screening.

[0071] (3) Identification of Positive Clones Positive clones were identified in the same manner as in the ELISA for CD3 antigen. CD73 antigen was prepared at 10 μg / mL in PBS, and 100 μL / well was immobilized to confirm the ELISA reactivity of monoclonal culture supernatants prepared from the third round of CD73 screening. The resulting positive clones were sequenced, and multiple amino acid sequences of antibody clones were identified.

[0072] <Preparation of Antibodies to AXL> (1) Preparation of AXL Antigen Based on the amino acid information of human AXL (Genebank Accession Number AAH32229), a gene encoding the 451 amino acid residues of the extracellular domain (amino acid residues 1-451) followed by a gene encoding an 8xHis tag was designed and artificially synthesized. This synthetic gene was inserted into the pCXN3 vector using restriction enzymes to prepare a human AXL expression vector. Expression and purification were performed in the same manner as for the CD3 antigen and CD73 antigen, and the AXL-His protein was prepared.

[0073] (2) Screening of Antibodies Against AXL Screening of antibodies against human AXL extracellular domain protein was carried out in the same manner as screening of CD3 antigen protein and CD73 antigen protein, and multiple amino acid sequences of antibody clones reactive with human AXL were identified.

[0074] <Identification of common light chain sequence> (1) Comparison of antibody sequences We examined whether there were any light chains with the same sequence among the antibody clones reactive with the human CD3ε extracellular domain protein and the human CD73 protein. As a result, we found one antibody clone whose light chain amino acid sequence was completely identical (Figure 4).

[0075] Furthermore, we confirmed whether there were any clones with a common light chain among the antibody clones reactive with the CD3ε extracellular domain protein and the antibody clones reactive with the human AXL protein. As a result, we found that two clones matched the anti-CD3ε antibody (Figure 5).

[0076] SEQ ID NO: 97: CD3ε antibody_0039 heavy chain variable region QVQLVQSGAEVKKPGASVKVSCKASGYTFTGYYMHWVRQAPGQGLEWMGWINPNSGGTNYAQKFQGRVTMTRDTSISTAYMELSRLRSDDTAVYYCAREMSNWDDAFDIWGQGTMVTVSR SEQ ID NO: 98: CD3ε antibody_0039 light chain SYELTQPPSVSVAPGKTARITCGGNNIGSKSVHWYQQKPGQAPVLVVYDDSDRPSGIPERFSGSNSGNTATLTISRVEAGDEADYYCQVWDSSSDHYVFGTGTKVTVLGQPKANPTVTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKADGSPVKAGVETTKPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEKTVAPTECS Additional CD3 light chain variable region sequence SEQ ID NO: 99: CD3ε antibody_0039 light chain variable region SYELTQPPSVSVAPGKTARITCGGNNIGSKSVHWYQQKPGQAPVLVVYDDSDRPSGIPERFSGSNSGNTATLTISRVEAGDEADYYCQVWDSSSDHYVFGTGTKVTVLG

[0077] SEQ ID NO: 100: CD73 antibody_3001 heavy chain variable region QVQLVQSGAEMKKPGESLRISCQGSGYSFTSYWISWVRQTPGKGLEWMGRIDPSDSYTNYSPSLQGHVTISADKSTSTAYLQWNSLKASDSGIYYCARLSRYFYGMDVWGRGTTVTVSR SEQ ID NO: 101: CD73 antibody_3001 light chain SYELTQPPSVSVAPGKTARITCGGNNIGSKSVHWYQQKPGQAPVLVVYDDSDRPSGIPERFSGSNSGNTATLTISRVEAGDEADYYCQVWDSSSDHYVFGTGTKVTVLGQPKANPTVTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKADGSPVKAGVETTKPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEKTVAPTECS

[0078] SEQ ID NO: 102: AXL antibody_2079 heavy chain variable region QVQLQESGPGLVKPSGTLSLTCAVSGGSISSSNWWSWVRQPPGKGLEWIGEIYHSGSTNYNPSLKSRVTISVDKSKNQFSLKLSSVTAADTAVYYCARGAHTELLLSVGAFDIWGQGTMVTVSR SEQ ID NO: 103: AXL antibody_2079 light chain SYELTQPPSVSVAPGKTARITCGGNNIGSKSVHWYQQKPGQAPVLVVYDDSDRPSGIPERFSGSNSGNTATLTISRVEAGDEADYYCQVWDSSSDHYVFGTGTKVTVLGQPKANPTVTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKADGSPVKAGVETTKPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEKTVAPTECS SEQ ID NO: 104: AXL antibody_2085 heavy chain variable region EVQLVESGGGLVQPGRSLRLSCAASGFRFDDYGMHWVRLAPGKGLEWVSTITWNGAYTAYADSVKGRFTISRDNAKNSLYLQMNGLRAEDTALYYCAKDGPEISSAGIDSWGQGTLVTVSR SEQ ID NO: 105: AXL antibody_2085 light chain SYELTQPPSVSVAPGKTARITCGGNNIGSKSVHWYQQKPGQAPVLVVYDDSDRPSGIPERFSGSNSGNTATLTISRVEAGDEADYYCQVWDSSSDHYVFGTGTKVTVLGQPKANPTVTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKADGSPVKAGVETTKPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEKTVAPTECS

[0079] <Shuffling of Exogenous Heavy Chains and Light Chain Libraries> (1) Preparation of a Heavy Chain-Fixed Antibody Library (Cetuximab) The heavy chain sequences were identified from the sequence information of the EGFR antibodies Cetuximab and Panitumumab, which are commercially available as antibody drugs. These heavy chain sequences were then used as exogenous heavy chains and combined with an antibody light chain library to search for novel light chains. In a second aspect of the present invention, even heavy chains not derived from an antibody library were incorporated into a nucleic acid library encoding a light chain. As long as the L chain was optimally matched with the inserted heavy chain, an antibody that functions without compromising binding activity could be selected. Specifically, the heavy chain variable region was identified from the antibody amino acid sequence listed in the Common Technical Document (CTD) for drug approval applications, and then artificial synthesis was performed to create nucleic acids corresponding to the amino acids. This heavy chain variable region sequence was inserted into a library containing nucleic acids encoding a light chain, resulting in the creation of a light chain library in which the heavy chain was fixed to the variable region of Cetuximab. More specifically, 117 amino acids (SEQ ID NO: 106) were obtained from the CTD, and the artificially synthesized DNA fragment was ligated to a vector containing an L-chain library via restriction enzyme treatment. Specifically, 5 μg of heavy chain fragment and 5 μg of vector were reacted with 2 μL of T4 Ligase (Takara) at 15°C for 4 hours. The DNA was then purified by ethanol precipitation, and the plasmid DNA was dissolved in QW. SEQ ID NO: 106: Cetuximab heavy chain variable region QVQLKQSGPGLVQPSQSLSITCTVSGFSLTNYGVHWVRQSPGKGLEWLGVIWSGGNTDYNTPFTSRLSINKDNSKSQVFFKMNSLQSNDTAIYYCARALTYYDYEFAYWGQGTLVSS

[0080] Thereafter, gene transfer of the ligation product and preparation of a phage antibody library were carried out in accordance with the procedures described in the above examples, and a phage solution was prepared.

[0081] (2) Antibody Screening (Cetuximab) A screening procedure was performed to select clones reactive to EGFR protein from the prepared light chain library immobilized to the heavy chain of cetuximab. Specifically, a recombinant protein (GenScript) was prepared as an antigen at 20 μg / mL in PBS, and added to ImmunoModule / Strip (NUNC) at 100 μL / well and allowed to stand overnight at 4°C. The antigen solution was then discarded, and 200 μL / well of blocking solution (2.5% skim milk / PBS) was added, followed by blocking at room temperature for 2 hours. Next, 1 x 10 IgG prepared with 2.5% skim milk / PBS was added to the ImmunoModule / Strip from which the blocking solution had been removed. １０ The phage antibody was added and reacted for 2 hours at room temperature. After the reaction was completed, the plate was washed 10 times with PBS, and 100 μL / well of 0.2 M glycine HCl (pH 3.0) was added to recover the phage. The recovered phage was infected with 20 mL of E. coli DH12S at OD 0.5 for 1 hour, and a portion of the infected phage was plated on an ampicillin LBG plate to calculate the titer of the recovered phage. The phage-infected E. coli was cultured overnight at 30°C in 100 mL of 2xYTGA medium (2xYT, 200 μg / mL ampicillin sulfate, 1% glucose). 10 mL of this overnight culture was mixed with 100 mL of 2xYTA medium (2xYT, 200 μg / mL ampicillin sulfate) and cultured at 37°C for 1.5 hours. After that, 1 x 10 helper phage KO7 was added and cultured at 37°C for 1 hour. After that, 400 mL of 2xYTGAK (2xYT, 200 μg / mL ampicillin sulfate, 0.05% glucose, 50 μg / mL kanamycin) was added and cultured overnight at 30°C. This was centrifuged at 8000 rpm for 10 minutes to prepare 500 mL of supernatant, which was then mixed with 100 mL of PEG solution (20% polyethyleneglycol 6000, 2.5 M NaCl) and stirred well, followed by centrifugation at 8000 rpm for 10 minutes to precipitate the phage. This was suspended in 10 mL of PBS, and a portion was used to count the number of E. coli infections. This was the phage for the first screening.

[0082] Similarly, in the second screening, the antigen was immobilized in the same manner as in the first screening, and the first screening phage solution was diluted to 1 x 10 ９ The series of procedures was repeated, with the washing procedure changed to 15 times. Furthermore, in the third screening, the antigen was immobilized in the same manner as before, and the second screening phage solution was added at 1 x 10 ９ The screening procedure was carried out using the same aliquot, with the washing procedure changed to 25 times.

[0083] (3) Identification of Positive Clones (Cetuximab) Positive clones were identified by ELISA using human EGFR extracellular domain protein. Specifically, the antigen was adjusted to 10 μg / mL with PBS and added to ImmunoModule / Strip (NUNC) at 100 μL / well and incubated at 37°C for 2 hours. The antigen solution was then discarded, and blocking solution (2.5% skim milk / PBS) was added at 200 μL / well for 2 hours at 37°C. The blocking solution was removed, and the cells were washed with PBS. The monoclonal culture supernatant prepared from the third screening was added at 100 μL / well and incubated at 37°C for 1 hour. After washing five times with PBS, 100 μL / well of 1 μg / mL Moise anti-cp3 antibody diluted with PBS / 0.05% Tween 20 was added and incubated at 37°C for 1 hour. After washing five times with PBS, 100 μL / well of HRP-labeled anti-Mouse IgG (H+L) antibody (MBL) diluted 2000-fold with PBS / 0.05% Tween 20 was added and incubated at 37°C for 1 hour. After washing five times with PBS, 100 μL / well of OPD in 0.1 M citrate phosphate buffer (pH 5.1) + 0.01% H2O2 was added and incubated at room temperature for 5 minutes. The color reaction was stopped by adding 100 μL / well of 2NH2SO4. The absorbance at 492 nm was then measured using an infinite F200 (TECAN). Phage clones showing a positive reaction to human EGFR antigen protein were selected from the results, and the DNA sequences of the top four clones were analyzed based on the signal intensity of the ELISA reaction. Of the four, the top clone was classified into one and the remaining three, and the amino acid sequences of two antibody clones were finally identified. These results demonstrated that clones that retained reactivity to EGFR protein could be obtained even when a cetuximab-derived heavy chain was combined with a novel sequence selected from the light chain library (Figure 6). Furthermore, one clone was confirmed to have a higher signal intensity than the scFv-type cetuximab used as a control.

[0084] (4) Identification of the Light Chain Sequence (Cetuximab) The light chain sequence confirmed by ELISA is shown below: SEQ ID NO: 107: Ce007 light chain EIVLTQSPATLSLSPGERATLSCRASQSVSGFLAWYQQKPGQAPRLLIYDTSNRATGIPARFSGSGSGADFTLTISSLESEDLAVYYCQQRSAWPITFGQGTRLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC

[0085] SEQ ID NO: 108: Ce016 light chain ETTLTQSPATLSVSPGERATLSCRASQSVSSNLAWYQQKPGQAPRLLIYGASTRATGIPARFSGSGSGKEFTLTISSLQSEDFAVYYCQQYNNWPLTFGQGTRLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKLYACEVTHQGLSSPVTKSFNRGEC

[0086] (5) Preparation of a heavy chain-fixed library (Panitumumab) The panitumumab sequence was obtained from the CTD according to the method described in (1) above, and the sequence of the heavy chain variable region was synthesized. SEQ ID NO: 109: Pamintumumab heavy chain variable region QVQLQESGPGLVKPSETLSLTCTVSGGSVSSGDYYWTWIRQSPGKGLEWIGHIYYSGNTNYNPSLKSRLTISIDTSKTQFSLKLSSVTAADTAIYYCVRDRVTGAFDIWGQGTMVTVSS Furthermore, a light chain library fixed to the heavy chain of panitumumab was prepared according to the method described above.

[0087] (6) Antibody Screening (Panitumumab) Screening was carried out against EGFR protein according to the method described in (2) above.

[0088] (7) Identification of Positive Clones (Panitumumab) Monoclonal antibodies were produced at the completion of the third screening in accordance with (3) above, and positive clones were identified by ELISA (Figure 7). As a result, many of the 36 clones were determined to be positive, and the DNA sequences of the top 8 clones were analyzed.

[0089] (8) Identification of Light Chain Sequence (Panitumumab) Sequence analysis revealed that the top 8 clones were classified into two types of sequences. SEQ ID NO: 110: Pa006 light chain DIQMTQSPSSLSASVGDRVTITCQASQDISRYLNWYQQKPGKAPELLIYDASDLETGVPSRFSGSGSGTGFTFTISSLQPEDIATYYCQQYDTLPLTFGGGTKVEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC

[0090] SEQ ID NO: 111: Pa007 light chain DIQMTQSPSSLSASVGDRVTITCQASQDISNYLNWYQQKPGKAPKLLIYDASNLETGVPSRFSGSGSGTDFTFTISSLQPEDIATYYCQQYDNLPLTFGGGTKVEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC

[0091] <Isolation of bispecific antibodies> An antibody that binds to the transferrin receptor (TfR) is produced as a pair with the antibody that binds to the EGFR protein. That is, a heavy chain having binding activity to TfR is searched for, using the light chain found in the combination of an anti-TfR antibody from the heavy chain library and the heavy chain of Cetuximab as a common light chain, so as to be paired with the combination of a heavy chain derived from Cetuximab and a light chain derived from the light chain library.

[0092] (1) Preparation of TfR Protein Based on the amino acid information of human TfR (Genebank accession number NP_003225.2), a gene encoding amino acid residues 89 to 760 followed by a gene encoding a His tag was designed and artificially synthesized. This synthetic gene was inserted into the pCXN3 vector using restriction enzymes to prepare a human TfR expression vector. The expression vector was transiently transfected into FreeStyle-293F cells (Invitrogen) and then cultured with shaking for one week. The culture supernatant collected from this sample was purified by passing it through a His-Trap HP column (Cytiva) to prepare the extracellular domain TfR-His protein.

[0093] (2) Screening of antibodies against TfR protein Screening against TfR protein was carried out according to the method described for screening against EGFR protein. That is, after three rounds of screening against TfR protein using the antibody library prepared in "(4) Preparation of phage antibody library" above, multiple clones against TfR were identified.

[0094] (3) Identification of a common light chain: From a group of antibody clones reactive with the TfR protein, we selected a sequence identical to the light chain newly combined with the heavy chain of Cetuximab. As a result, the light chain sequence of Ce016 was found to be completely identical to that of TfR672. SEQ ID NO: 112: TfR672 heavy chain variable region EVQLVESGGGLVQPGGSLRLSCSASGFSFSDYAMQWVRQAPGKGLEYVAAISHEGGTTYYADSVKGRFIISRDNSENTVHLQMSGLRRDDTALYYCVQDRFYYGSQSETHNWGQGTLVTVSS SEQ ID NO: 113: TfR672 light chain ETTLTQSPATLSVSPGERATLSCRASQSVSSNLAWYQQKPGQAPRLLIYGASTRATGIPARFSGSGSGKEFTLTISSLQSEDFAVYYCQQYNNWPLTFGQGTRLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKLYACEVTHQGLSSPVTKSFNRGEC

[0095] Furthermore, from a group of antibodies reactive to the TfR protein, a sequence identical to the light chain newly combined with the heavy chain of panitumumab was selected. As a result, the Pa007 light chain sequence was found to be completely identical to the light chain sequence of TfR443. SEQ ID NO: 114: TfR443 heavy chain variable region QVQLVQSGAEVKKPGASVKVSCKASGYTFATYGMHWVRQAPGQSLEWMGWINVGSGDTEYSQKFQGRVTITRDTSASTAYMELSSLRFEDTAVYYCARDNGRWGQGTLVTVSS SEQ ID NO: 115: TfR443 light chain DIQMTQSPSSLSASVGDRVTITCQASQDISNYLNWYQQKPGKAPKLLIYDASNLETGVPSRFSGSGSGTDFTFTISSLQPEDIATYYCQQYDNLPLTFGGGTKVEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC

[0096] (4) Confirmation of reactivity The ELISA activity of TfR672 and TfR443 was measured. Specifically, the ELISA was performed using the TfR protein as the antigen in the same manner as the ELISA for EGFR protein. As a result, it was confirmed that TfR672 and TfR443 have binding activity to the TfR protein (Figure 8).

[0097] That is, by combining two antibody fragments, one having an antibody fragment containing a cetuximab heavy chain and a Ce016 light chain, and the other having an antibody fragment containing a TfR672 heavy chain and a TfR672 light chain (same sequence as the Ce016 light chain), it is possible to create a bispecific antibody (recognizing both EGFR and TfR) that shares a common light chain. Similarly, by combining two antibody fragments, one having an antibody fragment containing a panitumumab heavy chain and a Pa007, and the other having an antibody fragment containing a TfR443 heavy chain and a TfR443 light chain (same sequence as the Pa007 light chain), it is possible to create a bispecific antibody (recognizing both EGFR and TfR) that shares a common light chain.

[0098] (5) Discussion Cetuximab, a commercially available antibody against the EGFR protein, is a chimeric antibody obtained by immunization with mice and genetically modified to replace the Fc portion with human Fc. Therefore, the mouse antibody genes remain in the variable regions. Meanwhile, the light chain library used in this study utilizes naive human sequences. The examples of the present invention suggest that it is possible to discover sequence combinations in which reactivity is not lost when a heavy chain variable region derived from a mouse antibody is combined with a light chain derived from a human antibody, which is believed to be of great significance for the creation of a common light chain for bispecific antibodies.

[0099] Furthermore, panitumumab is an antibody created by Amgen, Inc. in the United States, using Xenomouse technology and isolated from a human antibody-producing mouse into which a human antibody gene had been inserted. Many of the pharmaceuticals marketed as fully human antibodies in recent years are believed to be derived from such human antibody-producing mice. The fact that a common light chain was obtained for the panitumumab heavy chain obtained by immunizing this human antibody-producing mouse and for the light chain derived from the library suggests that the light chain derived from the library has a versatile sequence group. The fact that multiple light chains can be isolated for heavy chains other than those derived from the library may also be useful for producing bispecific antibodies derived from two types of heavy chains other than those derived from the library.

Claims

1. (1) A step of manufacturing a first library containing a plurality of antibodies that recognize a first antigen; (2) A step of manufacturing a second library containing a plurality of antibodies that recognize a second antigen; and (3) A step of selecting, from among the first library and the second library, an antibody that recognizes the first antigen and an antibody that recognizes the second antigen, which have a common L chain. A method for identifying a common L chain for use in manufacturing a multispecific antibody, comprising the steps.

2. The method according to claim 1, wherein the step (1) is a step of manufacturing the first library by selecting a plurality of antibodies that recognize the first antigen from a library containing a plurality of antibodies having an H chain and an L chain; and the step (2) is a step of manufacturing the second library by selecting a plurality of antibodies that recognize the second antigen from a library containing a plurality of antibodies having an H chain and an L chain.

3. The method according to claim 2, wherein the library comprising a plurality of antibodies having the H chains and L chains as described above contains 10 ９ or more kinds of antibodies.

4. The number of types of H chains in a library containing a plurality of antibodies having the above-mentioned H chains and L chains is 10 ６ or more, the method according to claim 2.

5. The method according to claim 2, wherein the number of types of the L chain in the library containing a plurality of antibodies having the H chain and the L chain is 10 ３ or less.

6. The method according to claim 2, wherein the ratio of the types of H chains to the types of L chains in the library containing a plurality of antibodies having the H chain and the L chain is 1000 times or more.

7. The step (1) includes: (1-A) A step of manufacturing an antibody library containing an H chain and an L chain of an antibody that recognizes a first antigen by incorporating a nucleic acid encoding the H chain of the antibody that recognizes the first antigen into a library of nucleic acids encoding the L chain; and (1-B) A step of selecting an antibody that recognizes the first antigen from the antibody library obtained in the step (1-A). The step (2) includes: (2-A) A step of manufacturing an antibody library containing an H chain and an L chain of an antibody that recognizes a second antigen by incorporating a nucleic acid encoding the H chain of the antibody that recognizes the second antigen into a library of nucleic acids encoding the L chain; and (2-B) A step of selecting an antibody that recognizes the second antigen from the antibody library obtained in the step (2-A). The method according to claim 1.

8. The number of types of L chains in the antibody library obtained in the above step (1-A) and the antibody library obtained in step (2-A) is 10 each ３ The method according to claim 7, wherein the number is as follows 9. The step (1) is a step of producing a first library by selecting a plurality of antibodies that recognize a first antigen from a library containing a plurality of antibodies having H chains and L chains, and the step (2) is: (2-A) A step of producing an antibody library containing an H chain and an L chain of an antibody that recognizes a second antigen by incorporating a nucleic acid encoding the H chain of the antibody that recognizes the second antigen into a library of nucleic acids encoding the L chain; and (2-B) A step of selecting an antibody that recognizes a second antigen from the antibody library obtained in the step (2-A). The method according to claim 1.

10. The method according to claim 1, wherein the first antigen and the second antigen are each selected from a cancer antigen, an antigen of an immune cell, a bioactive molecule, and an enzyme.

11. A method for producing a multispecific antibody, comprising a step of identifying an L chain by the method according to any one of claims 1 to 10, and a step of producing a multispecific antibody using the L chain identified above, an H chain that recognizes a first antigen, and an H chain that recognizes a second antigen.

Images