RU2295537C2 - Modified antagonistic antibody - Google Patents

Abstract

FIELD: immunology, biotechnology, in particular antibody modified by gene-engineering methods.

SUBSTANCE: obtained antibody contains two or more V-regions of H-chain and two or more V-regions of L-chain of monoclonal antibody and is capable of signal transduction into cells by formation of cross-linked cell surface molecule (molecules). Modifies antibody is useful as transduction signal agonist and as prophylaxis and/or treatment agent for various diseases, such as cancer, inflammation, hormonal disturbances and blood diseases. Modifies antibody also is useful in inducing of antagonistic action in relates to cells. Also disclosed are method for detection of abovementioned antibody and method for measurement of antibody antagonistic action in relates to cells.

EFFECT: antibodies of low molecular mass having antagonistic action due to cross-linking of cell surface molecule or intracell molecule.

31 cl, 96 dwg, 3 tbl, 8 ex

Description

FIELD OF THE INVENTION

The present invention relates to modified antibodies containing two or more V regions of an H chain and two or more V regions of an L chain of a monoclonal antibody that exhibits agonistic activity by crosslinking a cell surface molecule (s) or an intracellular molecule (s). These modified antibodies have agonistic activity of signal transduction into cells through crosslinking of the cell surface molecules (molecules) and are applicable as a medicine for various purposes.

State of the art

Japanese Patent Application JP-A 9-295999 describes the preparation of a specific monoclonal antibody using the spleen stromal cell line as a sensitizing antigen, in order to obtain specific antibodies that can recognize the aforementioned spleen stromal cells, and to obtain new monoclonal antibodies that recognize the murine associated with integrin protein (mouse IAP) as an antigen. Japan Patent Application JP-A 9-295999 also describes that these monoclonal antibodies are capable of inducing apoptosis of myeloid cells.

WO 99/12973 describes monoclonal antibodies for which the human integrin-associated protein (hereinafter referred to as human IAP; its amino acid sequence and nucleotide sequence is described in J. Cell Biol., 123, 485-496, 1993; see also Journal of Cell Science, 108, 3419-3425, 1995), which are capable of inducing apoptosis of human nucleated blood cells (hemocytes) (myeloid cell and lymphocyte) containing said human IAP. These monoclonal antibodies are called the MABL-1 antibody and the MABL-2 antibody, and the hybridomas producing these antibodies are also called MABL-1 (FERM BP-6100) and MABL-2 (FERM BP-6101), respectively.

Japanese Patent Application 11-63557 describes the preparation of single chain Fv regions, which are Fv regions of monoclonal antibodies, the antigen of which is human IAP. Single chain Fvs are capable of inducing apoptosis of nucleated blood cells having human IAPs.

A monoclonal antibody that recognizes IAP as an antigen induces apoptosis of nucleated blood cells containing human IAP, but it also causes hemagglutination in vitro. This indicates that the administration of a large amount of a monoclonal antibody recognizing IAP as an antigen can lead to a side effect, such as hemagglutination.

The inventors conducted an intensive study regarding the use of monoclonal antibodies against human IAP as a therapeutic agent in blood diseases and obtained single-chain Fv having a single-chain Fv fragment capable of inducing apoptosis of nucleated blood cells containing human IAP.

On the other hand, modified antibodies have been developed, in particular, antibodies with reduced molecular size, for example, single-chain Fv, to improve penetration into tissues and tumors by reducing molecular size and to obtain a recombinant method. Recently, single chain Fv dimers, in particular bispecific dimers, have been used to cross-link cells. Typical examples of such dimers are single chain Fv heterodimers that recognize cancer cell antigens and host cell antigens similar to NK cells (natural killers) and neutrophils (Kipriyanov et al., Int. J. Cancer, 77, 9763-9772, 1998) . They were obtained by the method of constructing single-chain Fvs in the form of modified antibodies that are more effective in treating cancers by inducing intercellular crosslinking. It was believed that intercellular crosslinking is induced by antibodies and their fragments (e.g., Fab fragment), bispecific modified antibodies, and even single chain Fv dimers that are monospecific.

As antibodies capable of transmitting a signal by crosslinking the cell surface molecule (s), an anti-EPO receptor involved in the differentiation and proliferation of cells (JP-A 2000-95800), an anti-MuSK receptor antibody (Xie et al., Nature Biotech 15, 768-771, 1997) and others. However, there have been no reports of modified antibodies of reduced molecular size.

Noting that single-chain Fv monomers derived from the MABL-1 antibody and the MABL-2 antibody do not induce cell apoptosis, while single-chain Fv dimers induce apoptosis of cells containing IAP, the inventors found that they cross-link (dimerizate ) IAP receptor on the cell surface, as a result of which the signal is transmitted to these cells and, as a result, apoptosis is induced. This suggests that monospecific single chain Fv dimers crosslink the cell surface molecule (s) (e.g., receptor) and transmit a signal like a ligand, thus acting as an agonist.

By focusing on intercellular crosslinking, the inventors found that the aforementioned single chain Fv dimers do not cause hemagglutination, while the aforementioned monoclonal antibodies cause hemagglutination. The same result was also observed with single chain bivalent antibodies (single chain polypeptides containing two H chain V regions and two L chain V regions). This suggests that monoclonal antibodies can form intercellular cross-links, while modified antibodies, like single-chain Fv dimers and single-chain bivalent antibodies, cross-link the cell surface molecule (s), but do not form intercellular cross-links.

Based on these observations, the inventors first discovered that modified antibodies, such as single chain Fv dimers and single chain bivalent antibodies, crosslink a cell surface molecule (s) or an intracellular molecule (s) of the same cell, along with known intercellular crosslinking, and suitable as a ligand for such a molecule (s) (in particular, as a ligand that mimics the action of a natural ligand).

By detecting further that the antibody molecule (whole IgG) can be modified into single chain Fv dimers, single chain bivalent antibodies and the like, which crosslink the cell surface molecule (s), thereby reducing side effects caused by intercellular crosslinking, and the provision of new drugs that induce only the desired effect on the cell, the authors completed this invention. The modified antibodies of the invention have unusually high activity compared to natural ligands such as TPO (thrombopoietin), EPO (erythropoietin) or G-CSF (granulocyte colony stimulating factor), or whole antibodies (IgG) having the same V region, as modified antibodies. They have improved tissue permeability due to the reduced molecular size compared to antibody molecules and the absence of constant regions.

SUMMARY OF THE INVENTION

The aim of this invention is the provision of agonistic modified antibodies of small molecular size, which contain two or more V-regions of the H-chain and two or more V-regions of the L-chain of monoclonal antibodies and have an agonistic effect by crosslinking the molecule (s) of the cell surface or intracellular molecule (molecules).

Thus, the present invention relates to modified antibodies that contain two or more V-regions of the H chain and two or more V-regions of the L chain, preferably from 2 to 6 each, particularly preferably from 2 to 4 each, most preferably two each of them exhibits agonistic activity by crosslinking the molecule (s) of the cell surface or the intracellular molecule (s).

The term "modified antibodies" in this description means any substance that contains two or more V-regions of the H-chain and two or more V-regions of the L-chain, where these V-regions are combined directly or via a linker via a covalent bond or non-covalent bond for example, polypeptides and compounds obtained by combining each V region of an antibody via a peptide linker or chemical crosslinking agent, and the like. Two or more V-regions of the H chain and two or more V-regions of the L chain used in this invention can be obtained from the same antibody or from different antibodies.

Preferred examples of the modified antibodies of the present invention are multimers, such as dimers, trimers or tetramers of a single chain Fv fragment containing the V region of the H chain and V region of the L chain, or single chain polypeptides containing two or more V regions of the H chain and two or more V-regions of the L chain. When the modified antibodies of the present invention are single chain Fv multimers, such as dimers, trimers, tetramers and the like, containing the H chain V region and the L chain V region, it is preferred that the H chain V region and V the region of the L chain located in the same chain was not associated with the formation of an antigen binding site.

More preferred examples are single chain Fv dimers that contain an H chain V region and an L chain V region, or a single chain polypeptide comprising two H chain V regions and two L chain V regions. The V region of the H chain and the V region of the L chain are preferably connected via a linker in these modified antibodies.

"Agonistic action" in this description means the biological action that occurs in the cell (s) into which the signal is transmitted by crosslinking the molecule (s) of the cell surface or intracellular molecule (s), for example, induction of apoptosis, induction of cell proliferation, induction of cell differentiation, induction of cell division or regulation of the cell cycle.

The agonistic action ED50 in this invention is determined by known methods used to measure agonistic activity. Examples are the detection of agonist-specific cell death or cell proliferation, the detection of expression of proteins specific for cell differentiation (e.g., specific antigens), or the measurement of kinase activity specific for the cell cycle. ED50 is the dose required to achieve 50% of the maximum effect, taken as 100% on the dose-response curve.

Preferred modified antibodies of the present invention have an agonistic effect (ED50) equivalent to or better than the agonistic effect of an antibody having the same antigen binding region as the modified antibody, namely a whole antibody such as IgG (hereinafter referred to as the "parent antibody"), having the same pair of the V region of the H chain and the V region of the L chain as the pair of the V region of the H chain and the V region of the L chain, forming the antigen binding region of the modified antibody. More preferred are antibodies having an agonistic effect (ED50) of more than two times the agonistic effect of the parent antibody, more preferably more than 5 times, most preferably more than 10 times. The present invention includes modified agonist antibodies containing the V region of the H chain and the V region of the L chain, forming the same antigen binding region as the original antibody that binds to the target molecule (target molecules) of the cell surface or intracellular molecule (s), but does not have an agonistic effect on this molecule.

Compounds containing two or more H chain V regions and two or more L chain V regions of the present invention can be any compounds that contain two or more H chain V regions and two or more L chain V regions antibodies and exhibit an agonistic effect (ED50), equivalent to or better than the agonistic effect of a natural ligand that binds to the cell surface molecule (s) or intracellular molecule (s). Preferred are compounds having an agonistic effect (ED50) of more than two times the agonistic effect of a natural ligand, more preferably more than 5 times, most preferably more than 10 times.

The “compounds” referred to here include not only the modified antibodies of the invention, but also any compounds containing two or more, preferably from 2 to 6, more preferably from 2 to 4, most preferably 2 antigen binding regions, such as whole antibodies or F (ab ') ₂ .

Modified antibodies or compounds of this invention containing two or more V-regions of the H chain and two or more V-regions of the L chain of the antibody, preferably do not have significant intercellular adhesive action. When the V region of the H chain and the V region of the L chain of the modified antibodies of the present invention are derived from the same antibody, they preferably have an intercellular adhesion action (ED50) of not more than 1/10 compared to the original antibody.

An intercellular adhesion action ED50 is determined in this invention by known methods for measuring agonistic action, for example, by measuring the agglomerating effect of cells expressing said cell surface molecule, such as a hemagglutination test.

This invention relates to DNA that encodes these modified antibodies.

This invention relates to animal cells or microorganisms that produce these modified antibodies.

This invention relates to the use of modified antibodies as an agonist.

This invention relates to a method for transmitting a signal to cells by crosslinking a cell surface molecule or an intracellular molecule using a modified antibody and thereby inducing an agonistic effect on cells, such as inducing apoptosis, inducing cell proliferation, inducing cell differentiation, inducing cell division or regulating effects cell cycle.

This invention relates to a medicament containing a modified antibody.

This invention relates to the use of a modified antibody as a medicine.

This invention relates to a method for screening or measuring a modified antibody, which contains two or more V-regions of the H chain and two or more V-regions of the L chain of the antibody and exhibits an agonistic effect by crosslinking a cell surface molecule or an intracellular molecule, including 1) obtaining a modified antibody containing two or more V-regions of the H chain and two or more V-regions of the L chain of the antibody and binding a specific molecule, 2) contacting the modified antibody with glue tissues expressing the specified molecule, and 3) measuring the agonistic effect caused by crosslinking of the specified molecule, which takes place in these cells. This measurement method is applicable for quality control in obtaining the modified antibodies of the present invention as a medicine and for other purposes.

The aforementioned single chain Fv dimer includes a dimer formed through a non-covalent bond, a dimer formed through a covalent bond through a crosslinking radical, and a dimer formed through a crosslinking reagent (antibody, antibody fragment or bivalent modified antibody). Conventional crosslinking radicals used to crosslink peptides can be used as crosslinking radicals to form these dimers. Examples are disulfide cross-links involving a cysteine residue, other crosslinking radicals such as C ₄ -C ₁₀ alkylenes (e.g. tetramethylene, pentamethylene, hexamethylene, heptamethylene and octamethylene, etc.) or C ₄ -C ₁₀ alkenylene ( cis / trans-3-butenylene, cis / trans-2-pentenylene, cis / trans-3-pentenylene, cis / trans-3-hexenylene, etc.).

In addition, a crosslinking reagent that can be combined with a single chain Fv is, for example, an amino acid sequence that can optionally be introduced into the Fv, for example, an antibody against the FLAG sequence and the like. or a fragment thereof, or a modified antibody derived from this antibody, for example, single chain Fv.

The present invention also relates to a method for inducing an agonistic effect on cells by introducing a first ligand and a second ligand that combine with a cell surface molecule (s) or an intracellular molecule (s), and introducing a substance that binds to the first and second ligands and forms a transverse the relationship between the first and second ligands. The first and second ligands can be any substances that contain a binding site for the specified molecule and can induce an agonistic effect, being crosslinked. Preferred examples are monovalent modified antibodies, such as the same or a different single chain Fv monomer, antibody fragment, etc. The substance for crosslinking the aforementioned ligand can be any substance that induces an agonistic effect on cells by means of crosslinking between the first ligand and the second ligand. Preferred examples are antibodies, antibody fragments, (Fab) _2, or bivalent modified antibodies. Examples of bivalent antibodies are (Fab) ₂ , single chain Fv dimers containing one H chain V region and one L chain V region, and single chain polypeptides containing two H chain V regions and two L chain V regions . This method is effective for studying receptors that transmit a signal to cells through cross-linking, it is expected that it can be used for DDS to deliver drugs to target cells, and is also useful as a drug delivery system that suppresses side effects. action and allows the drug to become effective at the desired time and for the desired period of time.

The modified antibodies of the present invention can be any substance that contains the V region of the H chain and V region of the L chain of the antibody (for example, MABL-1 antibodies, MABL-2 antibodies, 12 B5 antibodies, 12E10 antibodies, etc.) and which specifically recognize the cell surface molecule (s) or intracellular molecule (s), for example, a protein (receptor or protein involved in signal transduction) or the carbohydrate chain of the aforementioned protein or cell membrane protein and crosslink these cell surface molecules, thereby transmitting a signal VK letki. Modified antibodies are also included in which a portion of the amino acid sequence of the V region has been changed.

Depending on the characteristics of the cell surface molecule or intracellular molecule to be combined, for example, the structure of the molecule or the mechanism of action, the modified antibodies can be monospecific or multispecific, for example, bispecific. When a modified antibody binds to a receptor molecule that homodimerizes and transmits a signal to cells (e.g., erythropoietin receptor, thrombopoietin receptor, G-CSF receptor, SCF receptor, EGF receptor, IAP (CD47), etc.), preferred is a monospecific modified antibody. When it binds to a receptor molecule that heterodimerizes and transmits a signal to cells (e.g., IL-6 receptor, LIF receptor, IL-11 receptor), a bispecific modified antibody is preferred. When it binds to a receptor molecule that heterotrimers and transmits a signal to cells (e.g., IL-2 receptor, CNTF receptor, OSM receptor), a trispecific modified antibody is preferred. A method for producing bispecific single chain Fv dimers is described in W094 / 13804 and the like.

The present invention also relates to modified antibodies, the V region of the H chain and / or the V region of the L chain of which is the V region of the H chain derived from a human antibody, and / or the V region of the L chain derived from the antibody person. The V region of the H chain and / or the V region of the L chain obtained from a human antibody can be obtained by screening a library of human monoclonal antibodies, as described in W099 / 10494. Also included are the V region of the H chain and the V region of the L chain derived from human monoclonal antibodies.

The invention further relates to modified antibodies, the V region of the H chain and / or the V region of the L chain of which are the humanized V regions of the H chain and / or the humanized V regions of the L chain. Specifically, humanized modified antibodies consist of a humanized V region of an L chain that contains framework regions (FR) originating from the V region of the L chain of a human monoclonal antibody and determining complementarity of the (hypervariable) regions (hereinafter referred to as “CDRs”) originating from the V region of the L chain of a non-human mammalian monoclonal antibody (e.g., mouse, rat, cow, sheep, monkey), and / or a humanized V region of the H chain that contains FRs originating from the V region of H monoclonal chains ntitela human and CDR, originating from the V-region of the H chain of a monoclonal antibody of a mammal other than human (e.g., mouse, rat, cow, sheep, monkeys). In this case, the amino acid sequence of CDR and FR can be partially changed, for example, deleted, replaced or supplemented.

The V regions of the H chain and / or the V regions of the L chain of the modified antibodies of the present invention may be the V region of the H chain and / or the V region of the L chain derived from monoclonal antibodies of animals other than humans (such as a mouse , rat, cow, sheep, monkey, chicken, etc.). In this case, the amino acid sequence of CDR and FR can be partially changed, for example, deleted, replaced or supplemented.

The present invention also relates to DNA encoding the various modified antibodies mentioned above, and genetic engineering methods for producing recombinant vectors containing such DNA.

This invention also relates to host cells transformed with these recombinant vectors. Examples of host cells are animal cells, such as human cells, mouse cells or the like, and microorganisms, such as E. coli. Bacillus subtilis, yeast or the like

This invention relates to a method for producing modified antibodies, comprising culturing the aforementioned hosts and extracting the modified antibodies from their culture.

The present invention further relates to a method for producing a single-chain Fv dimer comprising culturing single-chain Fv-producing animal host cells in a serum-free medium to secrete single-chain Fv into this medium and isolating the single-chain Fv dimer formed in the medium.

The present invention also relates to the use of modified antibodies as an agonist, i.e. it relates to a signaling agonist that contains the aforementioned modified antibody as an active ingredient. Since the modified antibodies used in this invention are modified antibodies that form cross-links between the cell surface molecule (s) or intracellular molecule (s) and induce signal transmission, such a molecule can be any molecule that oligomerizes, for example, dimerizes, by compounds with the ligand and thereby transmits a signal to the cells.

Such a cell surface molecule includes hormone receptors and cytokine receptors. Hormone receptors include, for example, an estrogen receptor. Cytokine receptor, etc. includes a hematopoietic factor receptor, a lymphokine receptor, a growth factor receptor, a differentiation regulation receptor, and the like. Examples of cytokine receptors are erythropoietin receptor (EPO), thrombopoietin receptor (TPO), granulocyte colony stimulating factor receptor (G-CSF), macrophage colony stimulating factor receptor (M-CSF), granulocyte macrophage colony stimulating factor (GM) receptor, GM receptor tumor necrosis (TNF), interleukin-1 receptor (IL-1), interleukin-2 receptor (IL-2), interleukin-3 receptor (IL-3), interleukin-4 receptor (IL-4), interleukin-5 receptor (IL-5), interleukin-6 receptor (IL-6), interleukin-7 receptor (IL-7), interleukin-9 receptor (IL-9), interleukin-10 receptor (IL-10), interleukin-11 receptor (IL-11), interleukin-12 receptor (IL-12), interleukin-13 receptor (IL-13), interleukin-15 receptor (IL-15), interferon-alpha receptor (IFN-alpha), interferon-beta receptor (IFN-beta), interferon-gamma receptor (IFN-gamma), growth hormone receptor (GH), insulin receptor, blood stem cell proliferation factor receptor (SCF), vascular endothelial growth factor receptor (VEGF), epidermal growth factor receptor (EGF), nerve growth factor receptor (NGF), fibrobl growth factor receptor astov (FGF), platelet growth factor receptor (PDGF), transforming growth factor beta receptor (TGF-beta), leukocyte migration inhibition factor receptor (LIF), ciliary neurotrophic factor receptor (CNTF), oncostatin M receptor (OSM), Notch receptor -families (families of large transmembrane receptors), etc.

The intracellular surface molecule includes TAK1, TAB1, and the like. TAK1 and TAB1 function in the TGF-β signal transmission pathway, activate MAP kinase by forming a heterodimer, and transmit a series of signals. Many cancer cells have a TGF-β receptor mutation that represses cancer growth and, therefore, no TGF-β signal is transmitted. Modified antibodies, which can transmit a signal through crosslinking between TAK1 and TAB1, can induce this TGF-β signal through an agonist action through the TAK1 / TAB1 compound. Such modified antibodies of the present invention can inhibit the growth of TGF-β-resistant cancer cells and provide a new method for the treatment of cancer. Other examples of an intracellular molecule are the homodimer of transcription factor E2F and the heterodimer E2F / DP1, which have a proliferative effect on cells. The modified antibodies of the present invention can also induce an agonistic effect on these molecules and, therefore, can be used to treat various cell proliferation (cell-proliferative) diseases. The modified antibodies of the present invention can induce an agonistic effect by cross-linking the intracellular factor involved in signal transmission associated with the induction of apoptosis and, therefore, induce apoptotic death of cancer cells or autoimmune disease-related cells.

To allow the modified antibodies of this invention to interact with the intracellular molecule, peptides capable of penetrating the cell membrane (e.g., Pegelin, Penetratin) can be used to transport the modified antibodies to cells (Martine Mazel et al., Doxorubicin-peptide conjugates overcome multidrug resistance Anti-Cancer Drugs 2001, 12, Dccrossi D. et al., The third helix of the antennapedia homeodomain translocates through biological membranes, J. Biol. Chem. 1994, 269, 10444-10450).

Thus, pharmaceutical preparations containing an agonist modified antibody as an active ingredient are useful as prophylactic agents and / or therapeutic agents, etc. for various diseases, such as cancers, inflammation, hormonal disorders, blood diseases and autoimmune diseases.

The oligomers that can be formed by receptor proteins can be homo-oligomers or hetero-oligomers and any oligomers such as dimers, trimers and tetramers. For example, erythropoietin receptor, thrombopoietin receptor, G-CSF receptor, SCF receptor, EGF receptor and the like are known. form homodimers, that IL-6 receptor, LIF receptor and IL-11 receptor form heterodimers and that IL-2 receptor, CNTF receptor, OSM receptor form heterotrimers.

Modified antibodies of the present invention contain two or more V regions of the H chain and two or more V regions of the L chain derived from monoclonal antibodies. The structure of these modified antibodies can be represented by a single chain Fv dimer containing one H chain V region and one L chain V region, or a polypeptide containing two H chain V regions and two L chain V regions. In the modified antibodies of the present invention, the V regions of the H chain and L chain are preferably linked via a peptide linker that consists of one or more amino acids. The obtained modified antibodies contain variable regions of antibodies and bind to the antigen with the same specificity as the specificity of the original monoclonal antibodies.

H chain V region

In the present invention, the H chain V region derived from an antibody recognizes a cell surface molecule (s) or an intracellular molecule (s), for example, a protein (receptor or signal transduction protein) or a carbohydrate chain of that protein or on a cell membrane and oligomerizes, for example, dimerizes the specified molecule through the formation of cross-linking and thereby transduces the signal into cells. The H chain V region of the present invention includes H chain V regions derived from a mammal (e.g., human, mouse, rat, cow, sheep, monkey, etc.), and H chain V regions, having partially modified amino acid sequences of H chain V regions. More preferred is a humanized H chain V region containing the FR of the H chain V region of the human monoclonal antibody and the CDR of the H chain V region of the mouse monoclonal antibody. Also preferred is the V region of the H chain having an amino acid sequence derived from a human that can be obtained recombinantly. The V region of the H chain of the present invention may be a fragment of the V region of the H chain that retains antigen binding capacity.

L chain V region

In the present invention, the L chain V region recognizes a cell surface molecule (s) or an intracellular molecule (s), for example, a protein (receptor or signal transduction-related protein) or a carbohydrate chain of this protein or on a cell membrane and oligomerizes, for example, dimerizes the specified molecule through the formation of cross-linking and thereby transduces the signal into cells. The L chain V region of the present invention includes the L chain V regions of the present invention originating from a mammal (e.g., human, mouse, rat, cow, sheep, monkey, etc.), and the L- V region chains having partially modified amino acid sequences of the V regions of the L chain. More preferred is a humanized V region of the L chain containing the FR of the V region of the L chain of a human monoclonal antibody and the CDR of the V region of the L chain of a mouse monoclonal antibody. Also preferred is the V region of the L chain having an amino acid sequence derived from a human antibody, which can be obtained recombinantly. V-regions of the L chain of the present invention can be fragments of the aforementioned V-region of the L chain, which retain antigen binding ability.

Complementarity determining (hypervariable) region (CDR)

Each V region of the L chain and H chain forms an antigen binding site. The variable region of L and H chains consists of relatively conservative four common framework regions connected to three hypervariable regions, or complementarity determining regions (CDRs) (Kabat, EA et al., "Sequence of Protein of Immunological Interest", US Dept. Health and Human Services, 1983).

The main parts of these four framework regions (FR) form β-folded structures and three CDRs, thus forming a loop. In certain instances, CDRs may form part of this β-fold structure. These three CDRs are kept in sterically close position relative to each other through the FR framework region, which contributes to the formation of the antigen binding site together with the three CDRs.

These CDRs can be identified by comparing the amino acid sequence of the V region of the obtained antibody with the known amino acid sequences of the V regions of the known antibodies in accordance with the rule of thumb Kabat, E.A. et al., "Sequence of Protein of Immunological Interest".

Single chain Fv fragment

A single chain Fv fragment is a polypeptide monomer containing H-chain V region and L chain V region that are derived from monoclonal antibodies. The obtained single-chain Fvs contain the variable regions of the original (“parental”) antibodies and retain their complementarity determining region and, therefore, single-chain Fvs bind to the antigen with the same specificity as the specificity of the original monoclonal antibodies (JP-Appl. Patent Application. 11- 63557). Part of the variable region and / or CDR of a single chain Fv of the present invention can be partially changed, for example, deleted, replaced or supplemented. The V region of the H chain and the V region of the L chain constituting the single chain Fv of the present invention are mentioned above and can be linked directly or via a linker, preferably through a peptide linker. The structure of a single-chain Fv can be as follows: [V-region of the H-chain] - [V-region of the L-chain] or [V-region of the L-chain] - [V-region of the H-chain]. In the present invention, it is possible to ensure that a single chain Fv forms a dimer, trimer or tetramer, from which a modified antibody of the invention can be obtained.

Single chain modified antibody

Single chain modified antibodies of the present invention containing two or more V regions of an H chain and two or more V regions of an L chain, preferably two to four each, particularly preferably two of each, contain two or more V regions of an H chain and V -regions of the L chain, as mentioned above. Each region of this peptide must be positioned so that the modified single chain antibody forms a specific steric structure, specifically mimicking the steric structure formed by the single chain Fv dimer.

For example, V-regions are placed in the following order:

[V region of the H chain] - [V region of the L chain] - [V region of the H chain] - [V region of the L chain] or [V region of the L chain] - [V region of H -chains] - [V-region of the L-chain] - [V-region of the H-chain], where these regions are connected via a peptide linker, respectively.

Linker

In the present invention, linkers for connecting between the V region of the H chain and the V region of the L chain can be any peptide linker that can be introduced by genetic engineering, or any chemically synthesized linker. For example, linkers described in the literature, for example, Protein Engineering, 9 (3), 299-305, 1996, can be used in the present invention. These linkers can be the same or different in the same molecule. If peptide linkers are needed, the following are given as examples of linkers:

Ser

Gly-ser

Gly-gly-ser

Ser-gly-gly

Gly-Gly-Gly-Ser

Ser-Gly-Gly-Gly

Gly-Gly-Gly-Gly-Ser

Ser-Gly-Gly-Gly-Gly

Gly-Gly-Gly-Gly-Gly-Ser

Ser-Gly-Gly-Gly-Gly-Gly

Gly-Gly-Gly-Gly-Gly-Gly-Ser

Ser-Gly-Gly-Gly-Gly-Gly-Gly

(Gly-Gly-Gly-Gly-Ser) _n

(Ser-Gly-Gly-Gly-Gly) _n

where n is an integer of at least 1. The preferred length of the linker peptide varies depending on the receptor that serves as an antigen, in the case of single chain Fv, a range of 1-20 amino acids is usually preferred. In the case of single chain modified antibodies containing two or more V regions of the H chain and two or more V regions of the L chain, peptide linkers connecting the regions forming the same antigen binding site containing the [V region of the H chain] - [V-region of the L chain] (or [V-region of the L chain] - [V-region of the H chain]), have lengths of 1-30 amino acids, preferably 1-20 amino acids, more preferably 3-18 amino acids . Peptide linkers connecting regions that do not form the same antigen binding site containing the [V region of the H chain] - [V region of the L chain] (or [V region of the L chain] - [V region of H chains]) have lengths of 1-40 amino acids, preferably 3-30 amino acids, more preferably 5-20 amino acids. The method of introducing these linkers will be described in the explanation for DNA constructs encoding the modified antibodies of the present invention.

Chemically synthesized linkers, i.e. the crosslinking chemical agents of this invention may be any linkers commonly used to link peptides. Examples of these linkers may include N-hydroxysuccinimide (NHS), disuccinimidyl suberate (DSS), bis (sulfosuccinimidyl) suberate (BS ³ ), dithiobis (succinimidyl propionate) (DSP), dithiobis (sulfosuccinimidyl propionate) (DTS) (DTS) (DTS) EGS), ethylene glycolbis (sulfosuktsinimidilsuktsinat) (sulfo-EGS), disuccinimidyl tartarate (DST), disulfosuktsinimidiltartrat (sulfo-DST), bis [2- (suktsinimidooksikarboniloksi) ethyl] sulfone (BSOCOES), bis [2- (sulfosuktsinimidooksikarboniloksi) ethyl] sulfone ( sulfo-BSOCOES) or the like. They are commercially available. In the case of chemical linkers, it is preferred that they have a length equivalent to the length of the peptide linkers.

To form a single-chain Fv dimer, it is preferable to choose a linker suitable for dimerization in solution, such as culture medium, more than 20%, preferably more than 50%, even more preferably more than 80%, most preferably more than 90% of the single-chain Fv produced in host cells. Specifically, a linker consisting of 2-12 amino acids, preferably 3-10 amino acids, or other linkers corresponding to them are preferred.

Obtaining modified antibodies

Modified antibodies can be prepared by coupling through the aforementioned linker, H chain V regions and L chain V regions derived from known or new monoclonal antibodies that specifically bind to a cell surface molecule (s). As examples of single chain Fvs, MABL1-scFv and MABL2-scFv are provided containing the H chain V region and the L chain V region derived from the MABL-1 antibody and the MABL-2 antibody, respectively. As examples of single chain polypeptides containing two V regions of an H chain and two V regions of an L chain, MABL1-sc (Fv) ₂ and MABL2-sc (Fv) ₂ containing the V region of the H chain and V- L chain region derived from the aforementioned antibodies.

To produce a polypeptide, a signal peptide may be attached to the N-terminus of the polypeptide if the polypeptide is desired to be a secreted peptide. A well-known amino acid sequence suitable for purification of a polypeptide, such as a FLAG sequence, can be attached to effectively purify this polypeptide. In this case, a dimer can be formed using an anti-FLAG antibody.

In order to obtain a modified antibody of the invention, it is necessary to obtain DNA, i.e. DNA encoding a single chain Fv or DNA encoding a reconstructed single chain polypeptide. These DNAs, in particular for MABL1-scFv, MABL2-scFv, MABLl-sc (Fv) ₂ and / or MABL2-sc (Fv) ₂ , can be obtained from DNA encoding the V region of the H chain and V region L chains originating from the indicated Fv. They can also be prepared by the polymerase chain reaction (PCR) method using these DNAs as a template and amplifying a portion of the DNA contained therein encoding the desired amino acid sequence using a pair of primers corresponding to both ends of the DNA.

In the case where each V region having a partially modified amino acid sequence is desired, V regions in which one or more amino acids are modified, i.e. deleted, substituted or added, can be obtained by a procedure known in the art using PCR. Part of the amino acid sequence in the V region is preferably modified by PCR known in the art to produce a modified antibody that is sufficiently active against a specific antigen.

To determine the primers for PCR amplification, it is necessary to determine the type of H chain and L chain of the desired antibodies. However, in the case of the MABL-1 antibody and the MABL-2 antibody, it was reported that the MABL-1 antibody has k-type L chains and γ1-type H chains, and the MABL-2 antibody has k-type L chains and H-chains γ2a type (Japanese Patent Application JP-Appl. 11-63557). For PCR amplification of DNA encoding the H chain and L chain of the MABL-1 antibody and / or the MABL-2 antibody, primers described by Jones, S.T. can be used. et al., Bio / Technology, 9, 88-89, 1991.

To amplify the V regions of the L chains of the MABL-1 antibody and the MABL-2 antibody by PCR, the 5'-terminal and 3'-terminal oligonucleotide primers were selected, as previously mentioned. In the same way, 5'-terminal and 3'-terminal oligonucleotide primers were selected to amplify the V regions of the H chains of the MABL-1 antibody and the MABL-2 antibody.

In embodiments of the invention, 5'-terminal primers that contain the "GANTC" sequence providing the Hinfl restriction enzyme recognition site near their 5'-end and 3'-terminal primers that contain the "CCCGGG" nucleotide sequence providing the Xmal restriction enzyme recognition site are used near their 5'-end. Another restriction enzyme recognition site can be used instead of these sites while they are used to subclone the desired DNA fragment into a cloning vector.

Specifically designed PCR primers are used to provide suitable nucleotide sequences at the 5'-end and 3'-end of the cDNA encoding the V regions of the MABL-1 and MABL-2 antibodies so that these cDNAs are easily inserted into the expression vector and function correctly in this expression vector (for example, this invention provides for an increase in translation efficiency by embedding the Kozak sequence). The V regions of the antibodies MABL-1 and MABL-2, obtained by PCR amplification using these primers, are inserted into the HEF expression vector containing the desired human C region (see WO 92/19759). Cloned DNA can be sequenced using a conventional method, for example, using an automatic DNA sequencer (Applied Biosystems).

A linker, such as a peptide linker, can be introduced into the modified antibody of the present invention as follows. Primers are designed that have a partially complementary sequence with primers for the V regions of the H chains and V regions of the L chains described above, and which encode the N-terminus and C-terminus of the linker. A PCR procedure is then performed using these primers to obtain DNA encoding a peptide linker having the desired amino acid sequence and length. DNAs encoding the V region of the H chain and the V region of the L chain can be linked via the obtained DNA to obtain DNA encoding a modified antibody of the invention that has the desired peptide linker. After obtaining a DNA encoding one of the modified antibodies, DNA encoding the modified antibodies with the desired peptide linker or without the peptide linker can be easily obtained by constructing various primers for this linker and then performing PCR using these primers and the aforementioned DNA as a template.

Each V region of the modified antibody of the present invention can be humanized using conventional methods (e.g., Sato, K. et al., Cancer Res., 53, 1-6 (1993)). After obtaining the DNA encoding each of the humanized Fv, the humanized single chain Fv, the fragment of the humanized single chain Fv, the humanized monoclonal antibody, and the fragment of the humanized monoclonal antibody can be easily obtained in accordance with conventional methods. Preferably, the amino acid sequences of their V regions can be partially modified, if desired.

In addition, DNA obtained from another mammal, for example, DNA encoding the V regions of a human antibody, can be obtained in the same way as when obtaining DNA encoding the V region of the H chain and the V region of the L chain derived from mice in conventional ways, as mentioned above. The obtained DNA can be used to obtain the V region of the H chain and the V region of the L chain of another mammal, in particular, derived from a human antibody, single chain Fv derived from humans, and a fragment thereof, and a monoclonal antibody of human origin and its fragment .

In the case where the modified antibodies of the present invention are bispecific modified antibodies, they can be obtained by known methods (for example, by the method described in WO 94/13804).

As mentioned above, when target DNAs encoding the V regions of the modified antibodies and the V regions of the humanized modified antibodies are obtained, expression vectors containing them and hosts transformed with these vectors can be obtained in accordance with conventional methods. Further, these hosts can be cultured in accordance with the generally accepted method for producing reconstructed single chain Fv, reconstructed humanized single chain Fv, humanized monoclonal antibodies and fragments thereof. They can be isolated from cells or the medium and can be purified to a homogeneous mass. For this purpose, any isolation and purification methods commonly used for proteins can be used in combination, for example, chromatography, ultrafiltration, salting out and dialysis, if necessary, without limitation.

In the preparation of the reconstituted single chain Fv of the present invention by culturing animal cells, such as COS7 cells or CHO cells, preferably CHO cells, in a serum free medium, the dimer of said single chain Fv formed in this medium can be stably removed and purified in high yield. The dimer thus purified can be stably maintained for an extended period. The serum-free medium used in this invention can be any medium commonly used to produce a recombinant protein, without limitation.

Any expression system, for example, eukaryotic cells, such as animal cells, for example, adapted mammalian cell lines, filamentous fungi and yeast, and prokaryotic cells, such as bacterial cells, for example, E. coli, can be used to produce the modified antibodies of the present invention. Preferably, the modified antibodies of the invention are expressed in mammalian cells, for example, COS7 cells or CHO cells.

In these cases, conventional promoters useful for expression in mammalian cells can be used. Preferably, the immediate early promoter of human cytomegalovirus (HCMV) is used. Expression vectors containing the HCMV promoter include HCMV-VH-HCγ1, HCMV-VL-HCK and the like, which are derived from pSV2neo (WO 92/19759).

Additionally, other promoters that can be used to express genes in mammalian cells in this invention include viral promoters derived from retrovirus, polyomavirus, adenovirus and monkey virus 40 (SV40), and promoters originating from mammals, such as the promoter factor-1α elongation of the human polypeptide chain (HEF-1α). The SV40 promoter can be easily used in accordance with the method of Mulligan, R.C., et al. (Nature 277, 108-114 (1979)), and the HEF-12 promoter can also be used in accordance with the methods of Mizushima, S. et al. (Nucleic Acids Research, 18, 5322 (1990)).

The origin of replication (origin of replication) (ori) that can be used in this invention includes ori derived from SV40, polyomavirus, adenovirus, bovine papillomavirus (BPV), and the like. The expression vector may contain, as a selective marker, the APH gene II or I (peo) phosphotransferase (3 '), the thymidine kinase (TK) gene, the E. coli xanthine guanine phosphoribosyl transferase (ECOGPT) gene, or the dihydrofolate reductase gene (DHFR).

The antigen binding activity of a modified antibody obtained as described above can be evaluated by a conventional method, such as radioimmunoassay (RIA), enzyme-linked immunosorbent assay (ELISA), or surface plasma resonance. It can also be evaluated using the ability of the original antibodies to bind-inhibition as an index, for example, by the absence or presence of a concentration-dependent inhibition of the binding of the indicated monoclonal antibody to the antigen.

More specifically, animal cells are cultured transformed with an expression vector containing DNA encoding a modified antibody of the invention, for example, COS7 cells or CHO cells. Cultured cells and / or medium supernatant or a modified antibody purified from them are used to determine antigen binding. As a control, the supernatant of the culture medium in which cells transformed only with the expression vector were cultured was used. In the case of an antigen, for example, an MABL-1 antibody and an MABL-2 antibody, a test sample of a modified antibody of the present invention or a control supernatant is added to the mouse leukemia cell line, L1210 cells expressing human IAP, and then an analysis such as flow cytometry is performed to estimates of antigen binding activity.

An in vitro evaluation of the effect of signal transduction (the effect of induction of apoptosis in cases of MABL-1 antibodies and MABL-2 antibodies) is carried out as follows: the analyzed sample modified as described above, the antibodies are added to cells that express the antibody, or to cells into which it was introduced an antibody gene, and changes caused by signal transduction are evaluated, for example, whether cell death is induced in a manner specific for the human IAP antigen using conventional methods.

An in vivo evaluation of the effect of apoptosis induction, for example, in the case when the modified antibody recognizes human IAP (for example, modified antibodies derived from MABL-1 antibodies and MABL-2 antibodies), is carried out as follows: a murine model of human myeloma is obtained. The mice are injected intravenously with a monoclonal antibody or a modified antibody of the invention, which induces apoptosis of nucleated blood cells containing IAP. The mice in the control group are administered only phosphate-buffered saline (PBS).

Apoptosis induction is evaluated by antitumor activity based on changes in the human IgG content in the serum of mice and their survival time.

As mentioned above, the modified antibodies of the present invention can be obtained by creating modified antibodies that contain two or more V regions of the H chain and two or more V regions of the L chain and specifically bind to the target molecule of the cell surface or intracellular molecule, and screening these modified antibodies with in vivo and in vitro scores as described above.

Modified antibodies of the present invention that contain two or more V regions of an H chain and two or more V regions of an L chain, preferably two to four each, more preferably two each, can be a dimer of a single chain Fv containing one V region of H chains and one V region of the L chain, or a single chain polypeptide in which two or more V regions of the H chain and two or more V regions of the L chain are connected. It is believed that due to this design, this peptide mimics the three-dimensional structure of a natural ligand and, therefore, retains excellent antigen binding capacity and agonistic activity.

The modified antibodies of the present invention have a significantly reduced molecular size in comparison with the antibody molecule (whole IgG) and, therefore, better penetration into tissues and tumors and higher activity than the original agonistic monoclonal antibodies. Thus, the right choice of the initial antibody allows you to transduce various signals into cells and induce various actions in these cells, such as induction of apoptosis, induction of cell proliferation, induction of cell differentiation, induction of cell division or the effect on the regulation of the cell cycle. Pharmaceutical preparations containing them are useful for treating diseases that can be treated by inducing signal transduction, for example, cancers, inflammation, hormonal disorders, autoimmune diseases, as well as pathological changes in the blood, for example, leukemia, malignant lymphoma, aplastic anemia, myelodysplasia syndrome and red , or true polycythemia. It is also expected that the antibody of the present invention can be used as a contrast agent by RI-labeling. This effect may be enhanced by attachment to an RI compound or toxin.

List of figures

Figure 1 shows the result of flow cytometry, illustrating that the human IgG antibody does not bind to L1210 cells expressing human IAP (hIAP / L1210).

Figure 2 shows the result of flow cytometry, illustrating that the chimeric antibody MABL-1 specifically binds to L1210 cells expressing human IAP (hIAP / L1210).

Figure 3 shows the result of flow cytometry, illustrating that the chimeric antibody MABL-2 specifically binds to L1210 cells expressing human IAP (hIAP / L1210).

Figure 4 schematically illustrates a method for producing a single chain Fv in accordance with this invention.

Figure 5 illustrates the structure of an expression plasmid that can be used to express DNA encoding the single-chain Fv of the present invention in E. coli.

Fig. 6 illustrates the structure of an expression plasmid that is used to express DNA encoding the single-chain Fv of the present invention in mammalian cells.

7 shows the result of Western blotting of Example 5.4. From left to right, a molecular weight marker (which corresponds to 97.4, 66, 45, 31, 21.5 and 14.5 kDa from top to bottom), the culture supernatant of COS7 cells with introduced pCHO1 and the culture supernatant of COS7 cells with introduced pCHO2. It illustrates that the reconstructed single chain Fv of the MABL-2 antibody (shown by arrow) is contained in the cell culture supernatant with pCOM2 introduced.

Fig. 8 shows a flow cytometry result illustrating that an antibody from the culture supernatant of pCHO1 / COS7 cells as a control does not bind to pCOS1 / L1210 cells as a control.

Figure 9 shows the result of flow cytometry, illustrating that the antibody from the culture supernatant of MABL2-scFv / COS7 cells does not bind to pCOS1 / L1210 cells as a control.

Figure 10 shows the result of flow cytometry, illustrating that the antibody from the culture supernatant of pCOS1 / COS7 cells does not bind to hIAP / L1210 cells.

11 shows the result of flow cytometry, illustrating that an antibody from the culture supernatant of MABL2-scFv / COS7 cells specifically binds to hIAP / L1210 cells.

12 shows the result of a competitive ELISA in Example 5.6, where the binding activity of the single chain Fv of the present invention (MABL2-scFv) against an antigen is detected by inhibition of binding of the murine monoclonal antibody MABL-2 to this antigen as an indicator, in comparison with the pCHO1 cell culture supernatant / COS7 as a control.

13 shows the results of the apoptosis-inducing action in Example 5.7, illustrating that an antibody from the pCHO1 / COS7 cell culture supernatant as a control does not induce apoptosis of pCOS1 / L1210 cells as a control.

Fig. 14 shows the results of the apoptosis-inducing action in Example 5.7, illustrating that an antibody from the MABL2-scFv / COS7 cell culture supernatant does not induce apoptosis of pCOS1 / L1210 cells as a control.

15 shows the results of the apoptosis-inducing action in Example 5.7, illustrating that an antibody from the pCHO1 / COS7 cell culture supernatant does not induce apoptosis of hIAP / L1210 cells as a control.

Fig. 16 shows the results of the apoptosis inducing action in Example 5.7, illustrating that an antibody from the MABL2-scFv / COS7 cell culture supernatant specifically induces apoptosis of hIAP / L1210 cells.

17 shows the results of the apoptosis-inducing action in Example 5.7, illustrating that an antibody from the pCHO1 / COS7 cell culture supernatant does not induce apoptosis of CCRF-CEM cells (at a final concentration of 50%) as a control.

Fig. 18 shows the results of apoptosis-inducing action in Example 5.7, illustrating that an antibody from the MABL2-scFv / COS7 cell culture supernatant specifically induces apoptosis of CCRF-CEM cells (at a final concentration of 50%).

Fig. 19 shows a chromatogram obtained by purifying a single-chain Fv derived from the MABL-2 antibody produced by CHO cells in Example 5.9, illustrating that fraction A and fraction B were obtained as the main peaks when purifying the fraction after the Blue-Sepharose column on hydroxyapatite column.

Fig. 20 shows the gel filtration results of fraction A and fraction B obtained in Example 5.9- (2), illustrating that the main peaks (AI and BI, respectively) were eluted in case of fraction A at an apparent molecular weight of approximately 36 kD, and in the case of fraction B at approximately 76 kD.

Fig.21 is an analysis of SDS-PAGE electrophoresis of fractions obtained by purification of a single chain Fv derived from MABL-2 antibody produced by CHO cells in Example 5.9, illustrating that a single band with a molecular weight of about 35 kD was observed in both fractions.

Figure 22 shows the results of the analysis of the AI and BI fractions obtained by gel filtration by purification of a single chain Fv derived from the MABL-2 antibody produced by CHO cells, where the AI fraction contains a monomer and the BI fraction contains a dimer.

23 illustrates the structure of an expression plasmid that can be used to express DNA encoding the single chain Fv of the present invention in E. coli.

24 shows purification on a gel filtration column of crude products of a single chain Fv polypeptide derived from a MABL-2 antibody produced by E. coli obtained in Example 5.12, where each peak corresponds to a monomer or dimer, respectively, of a single chain Fv produced by E. coli .

Fig. 25 shows the results of the apoptosis-inducing action in Example 5.13, illustrating that the mouse IgG antibody does not induce apoptosis of hIAP / L1210 cells (final concentration of 3 μg / ml) as a control.

Fig. 26 shows the results of the apoptosis-inducing action in Example 5.13, illustrating that the MABL2-scFv dimer produced by CHO cells strongly induces apoptosis of hIAP / L1210 cells (final concentration 3 μg / ml).

27 shows the results of the apoptosis-inducing action in Example 5.13, illustrating that the E. coli-produced MABL2-scFv dimer strongly induces apoptosis of hIAP / L1210 cells (final concentration 3 μg / ml).

Fig. 28 shows the results of the apoptosis-inducing action in Example 5.13, illustrating that the induction of apoptosis on hIAP / L1210 cells by the MABL2-scFv monomer produced by CHO cells is at the same level as in the control (final concentration 3 μg / ml )

Fig. 29 shows the results of the apoptosis-inducing action in Example 5.13, illustrating that the induction of apoptosis on hIAP / L1210 cells with the MABL2-scFv monomer produced by E. coli is at the same level as in the control (final concentration 3 μg / ml).

30 shows the results of the apoptosis inducing action in Example 5.13, illustrating that the mouse IgG antibody used as a control does not induce apoptosis of hIAP / L1210 cells even when anti-FLAG antibody is added (final concentration 3 μg / ml).

Fig. 31 shows the results of the apoptosis inducing action in Example 5.13, illustrating that the MABL2-scFv monomer produced by CHO cells strongly induces apoptosis of hIAP / L1210 cells when anti-FLAG antibody is added (final concentration 3 μg / ml).

Fig. 32 shows the results of a quantitative measurement of human IgG in mouse serum into which human myeloma KPMM2 cell line was transplanted, showing the amounts of human IgG produced by human myeloma cells in this mouse. This figure illustrates that the scFv / CHO dimer strongly inhibits the growth of CRMM2 cells.

Fig. 33 shows the survival time of the mouse after tumor transplantation, illustrating that the survival time in the group of mice with the scFv / CHO dimer introduced was significantly longer.

Fig. 34 illustrates the structure of an expression plasmid that expresses a modified antibody [sc (Fv) ₂ ] containing two H chain V regions and two L chain V regions derived from an MABL-2 antibody.

Fig. 35 illustrates the structure of a plasmid that expresses scFv (HL type).

where the V-regions are connected by the type of [H-chain] - [L-chain] without a peptide linker.

Fig. 36 illustrates the structure of an HL-type polypeptide and the amino acid sequences of peptide linkers.

Fig. 37 illustrates the structure of a plasmid that expresses scFv (LH type), where V regions are connected as [L chain] - [H chain] without a peptide linker.

Fig. 38 illustrates the structure of an LH-type polypeptide and the amino acid sequences of peptide linkers.

Fig. 39 shows the results of Western blotting in Example 6.4, illustrating that a modified sc (Fv) ₂ antibody containing two V regions of an H chain and two V regions of an L chain, and MABL2-scFv containing different peptide linkers are expressed lengths.

40a and 40b show flow cytometry results using the culture supernatants of COS7 cells obtained in Example 6.3 (1), illustrating that MABL2-scFv and sc (Fv) ₂ containing peptide linkers of different lengths have a high affinity for human IAP.

41 a and 41 b show the results of the apoptosis inducing action in Example 6.6, illustrating that scFv <HL3, 4, 6, 7, LH3, 4, 6 and 7> and sc (Fv) ₂ strongly induce hIAP cell death / L1210.

Fig. 42 shows the results of the evaluation of antigen binding capacity in Example 6.10, illustrating that the scFv <HL5> and sc (Fv) ₂ dimers have a high affinity for human IAP.

Fig. 43 shows the results of the in vitro apoptosis-inducing action in Example 6.11, illustrating that the scFv <HL5> dimer and sc (Fv) ₂ dimer induce apoptosis of hIAP / L1210 cells and CCRF-CEM cells in a concentration-dependent manner.

Fig. 44 shows the results of a quantitative measurement of the M protein produced by the human BMPM2 cell line in mouse serum with transplanted human myeloma cells. This figure illustrates that the scFv <HL5> dimer and sc (Fv) ₂ dimer strongly inhibited the growth of CRMM2 cells.

Fig. 45 shows the survival time (days) of mice after tumor transplantation, illustrating that the survival time of a group of mice injected with scFv <HL5> was significantly increased.

Fig. 46 shows the survival time (days) of mice after tumor transplantation, illustrating that the survival time of a group of mice injected with sc (Fv) ₂ was significantly increased.

Fig. 47 is a diagram showing a method of constructing a DNA fragment encoding a reconstructed single chain Fv 12B5 containing a linker sequence of 15 amino acids and its structure.

48 shows the result of purification of each single chain Fv 12B5 gel filtration obtained in Example 7.5 (1), illustrating that sc12B5 was split into two peaks (fractions A and B).

Fig. 49 shows the result of the analysis of each fraction A and B by SDS-PAGE electrophoresis carried out in Example 7.5 (2).

Fig. 50 shows the result of the analysis of each fraction A and B by chromatography on a Superdex 200 column carried out in Example 7.5 (2), illustrating that the main peak of fraction A was eluted at an apparent molecular weight of about 44 kD, as shown in (a), and that the main peak of fraction B was eluted at an apparent molecular weight of about 22 kD in (b).

Fig. 51 shows a measurement result of TPO-like agonistic activity of sc12B5 and antibody 12B5 (IgG, Fab), illustrating that 12B5IgG and monovalent single chain Fv (sc12B5) showed concentration-dependent TPO-like agonistic activity.

Fig. 52 shows a measurement result of TPO-like agonistic activity of sc12B5 monomer and dimer, illustrating that a single-chain Fv (sc12B5 dimer) having a bivalent antigen binding site had an agonistic activity approximately 400 times higher than the monovalent sc12B5, and that this efficiency is equivalent to or higher than the effectiveness of human TPO.

Fig. 53 shows the result of purification of the obtained single-chain antibody sc12E10 by gel filtration chromatography using a Superdex 200HR column, illustrating that sc12E10 was separated into two peaks (fractions A and B).

Fig. 54 shows the result of purification of the obtained db12T10 single chain antibody by gel filtration chromatography using a Superdex 200HR column, illustrating that db12E10 was separated into two peaks (fractions C and D).

Fig. 55 shows electrophoretic analysis in SDS-PAGE of fractions A and B (sc12E10) and fractions C and D (db12E10) under reducing and non-reducing conditions.

56 shows the result of the analysis of fractions A and B by gel filtration chromatography using a Superdex 200HR column, illustrating that (1) the main peak of fraction A was eluted with an apparent molecular weight of about 42 kDa, and (2) the main peak of fraction B was eluted with an apparent molecular weight of about 20 kDa.

Fig. 57 shows the result of the analysis of fractions C and D by gel filtration chromatography using a Superdex 200HR column, illustrating that (1) the main peak of fraction C was eluted with an apparent molecular weight of about 69 kDa, and (2) the main peak of fraction D was eluted with an apparent molecular weight of about 41 kDa.

Fig. 58 is a graph showing agonistic activity of various variants of 12E10 antibody molecules on MPL, illustrating that single chain Fvs (sc12E10, db12E10) showed TPO-like agonistic activity, while 12E10 IgG and 12E10 Fab were not detected.

Fig. 59 is a graph showing agonistic activity of sc12E10 monomer and dimer and db12E10 dimer and trimer on MPL, illustrating that sc12E10 dimer and db12E10 dimer and trimer showed TPO-like agonistic activity higher than TPO activity.

Information confirming the possibility of carrying out the invention

The invention will be illustrated specifically with reference to the following examples, which in no way limit the scope of the invention.

To illustrate the method for producing the modified antibodies of the present invention, examples of the preparation of single chain Fv are given below. In these examples of the preparation of modified antibodies, murine antibodies against human IAP, MABL-1 and MABL-2 were used. Hybridomas MABL-1 and MABL-2 producing them, respectively, were deposited with an authorized international depository of microorganisms (National Institute of Bioscience and Technology, Agency of Industrial Science and Technology, Minister of International Trade and Industry (1-3 Higasi 1-chome, Tsukuba-shi, Ibaraki-ken, Japan)) September 11, 1997 as PERM BP-6100 and FERM BP-6101.

Examples

Example 1 (Cloning of DNA encoding the V region of murine monoclonal antibodies to human IAP)

DNAs encoding the variable regions of murine monoclonal antibodies to human IAP, MABL-1 and MABL-2, were cloned as follows.

1.1 Obtaining messenger RNA (mRNA)

mRNA hybridomas MABL-1 and MABL-2 were isolated using the mRNA Purification Kit (Pharmacia Biotech).

1.2 Synthesis of double-stranded cDNA

Double-stranded cDNA was synthesized from approximately 1 μg mRNA using the Marathon cDNA Amplification Kit (Clontech) and an adapter was attached to it.

1.3 PCR amplification of genes encoding the variable regions of the antibody

PCR was performed using a Thermal Cycler thermal cycler (Perkin Elmer).

(1) Amplification of a gene encoding the V region of the L chain of MABL-1

The primers used for the PCR method are adapter primer-1 (Clontech) represented by SEQ ID NO 1, which hybridizes with part of the adapter sequence, and MKC primer (Mouse Kappa Constant) (Bio / Technology, 9, 88-89, 1991), represented by SEQ ID NO 2, which hybridizes to the Kappa type L chain V region of a mouse kappa.

50 μl of PCR solution contain 5 μl of 10-fold PCR buffer II, 2 mm MgCl ₂ , 0.16 mm dNTP (dATP, dGTP, dCTP and dTTP), 2.5 units of AmpliTaq Gold DNA polymerase (Perkin Elmer), 0.2 μM adapter primer SEQ ID NO 1, 0.2 μM ISS primer SEQ ID NO 2 and 0.1 μg double-stranded cDNA derived from MABL-1. This solution was preheated at an initial temperature of 94 ° C for 9 minutes and then heated at 94 ° C for 1 minute, at 60 ° C for 1 minute, and at 72 ° C for 1 minute 20 seconds in the indicated sequence. This temperature cycle was repeated 35 times and then the reaction mixture was further heated at 72 ° C. for 10 minutes.

(2) Amplification of cDNA Encoding the Y-Region of the H Chain of MABL-1

The adapter primer-1 represented by SEQ ID NO 1 and the MHC-γ1 primer (Mouse Heavy Constant) (Bio / Technology, 9, 88-89, 1991) represented by SEQ ID NO 3 were used as PCR primers.

Amplification of cDNA was performed in accordance with the method of amplification of the V-region gene of the L chain, which is described in Example 1.3- (1), with the exception of using 0.2 μM MHC-γ1 primer instead of 0.2 μM MKS primer.

(3) Amplification of cDNA encoding the MABL-2 L chain V region

The adapter primer-1 represented by SEQ ID NO 1 and the ISS primer represented by SEQ ID NO 2 were used as primers for PCR.

Amplification of cDNA was carried out in accordance with the method of amplification of the V-region gene of the L chain of the MABL-1 gene, which is described in Example 1.3- (1), except for using 0.1 μg of double-stranded cDNA obtained from MABL-2, instead of 0.1 μg double-stranded cDNA from MABL-1.

(4) Amplification of cDNA Encoding the V Region of the H Chain of MABL-2

The adapter primer-1 represented by SEQ ID NO 1 and the MHC-γ2a primer (Bio / Technology, 9, 88-89, 1991) represented by SEQ ID NO 4 were used as PCR primers.

Amplification of cDNA was carried out in accordance with the method of amplification of the V-region gene of the L chain, which is described in Example 1.3- (3), except for using 0.2 μM MHC-γ2 primer instead of 0.2 μM MKS primer.

1.4 Purification of PCR products

The PCR amplified DNA fragment as described above was purified using the QIAquick PCR Purification Kit (Qiagen) and dissolved in 10 mM Tris-HCl (pH 8.0) containing 1 mM EDTA.

1.5 Ligation and transformation

Approximately 140 ng of the DNA fragment containing the gene encoding the kappa type L mouse murine V region obtained from MABL-1 as described above was ligated with 50 ng of the pGEM-T Easy (Promega) vector in a reaction buffer containing 30 mM Tris-HCl (pH 7.8), 10 mM MgCl ₂ , 10 mM dithiothreitol, 1 mM ATP and 3 units of T4 DNA ligase (Promega), at 15 ° C for 3 hours.

Then, 1 μl of this reaction mixture was added to 50 μl of competent E. coli DH5α cells (Toyobo Inc.). and the cells were kept on ice for 30 minutes, incubated at 42 ° C for 1 minute, and again kept on ice for 2 minutes. 100 μl of SOC medium (Gibco BRL) was added. E. coli cells were seeded on (Molecular Cloning: A Laboratory Manual, Sambrook et al., Cold Spring Harbor Press, 1989) LB agar medium containing 100 μg / ml ampicillin (Sigma) and cultured at 37 ° C. overnight to obtain transformants of E. coli.

The transformant was cultured in 3 ml LB medium containing 50 μg / ml ampicillin at 37 ° C. overnight, and plasmid DNA was isolated from this culture using the QIAprep Spin Miniprep Kit (Qiagen).

The resulting plasmid containing the gene encoding the kappa mouse L chain V region derived from the MABL-1 hybridoma was designated pGEM-M1L.

According to the method described above, a plasmid containing a gene encoding the mouse H chain V region derived from the MABL-1 hybridoma was obtained using a purified DNA fragment and designated pGEM-M1H.

A plasmid containing the gene encoding the mouse Kappa type L chain V region derived from the MABL-2 hybridoma was obtained using a purified DNA fragment and designated pGEM-M2L.

A plasmid containing the gene encoding the mouse H chain V region V originating from the MABL-2 hybridoma was obtained using a purified DNA fragment and designated as pGEM-M2H.

Example 2 (DNA Sequencing)

The nucleotide sequence of the cDNA encoding regions in the above plasmids was determined using an automatic DNA sequencer (Applied Biosystems) and an ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems) sequencing kit according to the manufacturer's protocol.

The nucleotide sequence of the gene encoding the V region of the L chain of the murine MABL-1 antibody, which is inserted into the pGEM-M1L plasmid, is shown in SEQ ID NO 5.

The nucleotide sequence of the gene encoding the V region of the H chain of the murine MABL-1 antibody, which is inserted into the plasmid pGEM-M1H, is shown in SEQ ID NO 6.

The nucleotide sequence of the gene encoding the V region of the L chain of the murine MABL-2 antibody, which is inserted into the pGEM-M2L plasmid, is shown in SEQ ID NO 7.

The nucleotide sequence of a gene encoding the V region of the H chain of the murine MABL-2 antibody, which is inserted into the pGEM-M2H plasmid, is shown in SEQ ID NO 8.

Example 3 (Definition of CDR)

The V regions of the L chain and the H chain usually have a similar structure, and every four frame regions in them are connected by three hypervariable regions, i.e. complementarity determining regions (CDRs). The amino acid sequence of this skeleton is relatively conserved, while the amino acid sequence of the CDR has extremely high variability (Kabat, E. A. et al., "Sequence of Protein of Immunological Interest", US Dept. Health and Human Services, 1983).

Based on these facts, the amino acid sequences of the variable regions of murine monoclonal antibodies to human IAP were compared with the database of amino acid sequences of antibodies obtained by Kabat et al., For the identification of homology. CDR regions were determined based on homology, as shown in table 1.

Table 1 Plasmid SEO ID NO CDR (1) CDR (2) CDR (3) pGEM-M1L 5 43-58 74-80 113-121 pGEM-M1H 6 50-54 69-85 118-125 pGEM-M2L 7 43-58 74-80 113-121 pGEM-M2H 8 50-54 69-85 118-125

Example 4 (Identification of the expression of cloned cDNA. Obtaining chimeric antibodies MABL-1 and chimeric antibodies MABL-2)

4.1 Obtaining vectors expressing the chimeric antibody MABL-1

cDNA clones, pGEM-M1L and pGEM-M1H encoding the V regions of the L chain and H chain of the murine MABL-1 antibody, respectively, were modified by PCR and introduced into the HEF expression vector (WO 92/19759) for for preparing vectors expressing the chimeric MABL-1 antibody.

The MLS forward primer (SEQ ID NO 9) for the L chain V region and the MHS forward primer (SEQ ID NO 10) for the H chain V region were designed to hybridize with DNA encoding the start of the leader sequence of each V region , and contained the Kozak consensus sequence (J. Mol. Biol., 196, 947-950, 1987) and the HindIII restriction site. The reverse primer MLAS (SEQ ID NO 11) for the V region of the L chain and the reverse primer MHAS (SEQ ID NO 12) for the V region of the H chain were designed to hybridize with DNA encoding the end of the J segment and contain the sequence of the splicing donor site and the BamHI restriction site.

100 μl of PCR solution containing 10 μl of 10-fold PCR buffer II, 2 mM MgCl ₂ , 0.16 mM dNTP (dATP, dGTP, dCTP and dTTP), 5 units of AmpliTaq Gold DNA polymerase, 0.4 μM each of primers and 8 ng of the DNA template (pGEM-M1L or pGEM-M1H), were pre-heated at an initial temperature of 94 ° C for 9 minutes and then heated at 94 ° C for 1 minute, at 60 ° C for 1 minute and at 72 ° C for 1 minute 20 seconds in the indicated sequence. This temperature cycle was repeated 35 times and then the reaction mixture was further heated at 72 ° C. for 10 minutes.

The PCR product was purified using a QIAquick PCR Purification Kit (Qiagen) and then digested with HindIII and BamHI. The product from the L chain V region was cloned into the HEF, HEF-κ expression vector, and the product from the H chain V region was cloned into the HEF expression, HEF-γ. After DNA sequencing, plasmids containing the DNA fragment with the correct DNA sequence were designated as HEF-M1L and HEF-M1H, respectively.

4.2 Obtaining vectors expressing chimeric antibodies MABL-2

Modification and cloning of cDNA was performed in the same manner as described in Example 4.1, except that pGEM-M2L and pGEM-M2H were used as the DNA template instead of pGEM-M1L and pGEM-M1H. After DNA sequencing, plasmids containing DNA fragments with the correct DNA sequences were designated as HEF-M2L and HEF-M2H, respectively.

4.3 Transfection into COS7 cells

The above expression vectors were tested in COS7 cells to observe the transient expression of chimeric antibodies MABL-1 and MABL-2.

(1) Transfection with genes of the chimeric antibody MABL-1

COS7 cells were co-transformed with the HEF-M1L and HEF-M1H vectors by electroporation using a Gene Pulser (BioRad) device. Each DNA (10 μg) and 0.8 ml of PBS containing 1 × 10 ⁷ cells / ml was added to the cuvette. The mixture was treated with a pulse at 1.5 kV, with an electric capacitance of 25 μF.

After “reconstitution” for 10 minutes at room temperature, the electroporated cells were transferred to the Eagle culture medium in the Dulbecco modification (DMEM, Gibco BRL) containing 10% γ-globulin-free fetal calf serum. After culturing for 72 hours, the supernatant was collected, centrifuged to remove cell fragments, and separated.

(2) Transfection with genes encoding the chimeric antibody MABL-2

The co-transfection of COS7 cells with genes encoding the chimeric MABL-2 antibody was performed in the same manner as described in Example 4.3- (1), except that the vectors HEF-M2L and HEF-M2H were used instead of the vectors HEF-M1L and HEF-M1H . Supernatant was obtained in the same manner.

4.4 Flow cytometry

Flow cytometry was performed using the aforementioned COS7 cell culture supernatant to measure antigen binding. The culture supernatant of COS7 cells expressing the chimeric MABL-1 antibody, or COS7 cells expressing the chimeric MABL-2 antibody or human IgG antibody (Sigma) was added as a control to 4 × 10 ⁵ cells of the murine leukemia L1210 cell line expressing human IAP , and incubated on ice. After washing, FITC-labeled anti-human IgG antibody (Cappel) was added thereto. After incubation and washing, fluorescence intensity was measured using a FACScan instrument (Becton Dickinson).

Since the chimeric antibodies MABL-1 and MABL-2 specifically bind to L1210 cells expressing human IAP, it was confirmed that these chimeric antibodies had the correct structure of the V regions of the murine monoclonal antibodies MABL-1 and MABL-2, respectively (figures 1-3 )

Example 5 (Obtaining reconstructed single-chain Fv (scFv) antibodies MABL-1 and antibodies MABL-2)

5.1 Obtaining a reconstructed single chain Fv antibody MABL-1

Reconstructed single chain Fv MABL-1 antibodies were prepared as follows. The V region of the H chain and the V region of the L chain of the MABL-1 antibody and linker, respectively, were amplified by the PCR method and coupled to obtain a reconstructed single chain Fv antibody MABL-1. The production method is illustrated in figure 4. Six primers (A-F) were used to obtain a single chain Fv MABL-1 antibody. Primers A, C, and E contain a sense sequence, and primers B, D, and F contain an antisense sequence.

The VHS forward primer for the H chain V region (Primer A, SEQ ID NO 13) was designed to hybridize with DNA encoding the N-terminus of the H chain V region and contain an NcoI restriction enzyme recognition site. The VHAS reverse primer for the H chain V region (Primer B, SEQ ID NO 14) was designed to hybridize with DNA encoding the C-terminus of the H chain V region and overlap with the linker.

The LS direct primer for the linker (Primer C, SEQ ID NO 15) was designed to hybridize with DNA encoding the N-terminus of the linker and overlap with DNA encoding the C-terminus of the V region of the H chain. The LAS reverse linker primer (Primer D, SEQ ID NO 16) was designed to hybridize with DNA encoding the C-terminus of the linker and overlap with DNA encoding the N-terminus of the V region of the L chain.

The forward VLS primer for the V region of the L chain (Primer E, SEQ ID NO 17) was designed to hybridize with DNA encoding the C-terminus of the linker and overlap with DNA encoding the N-terminus of the V region of the L chain. The VLAS-FLAG reverse primer for the L chain V region (Primer F, SEQ ID NO 18) was designed to hybridize with DNA encoding the C-terminus of the L chain V region and contain a sequence encoding the FLAG peptide (Hopp NP et al., Bio / Technology, 6, 1204-1210, 1988), two stop codons and an EcoRI restriction enzyme recognition site.

In the first stage of PCR, three reactions were performed: AB, C-D, and E-F, and the resulting PCR products were purified. The three PCR products obtained from this first step of PCR were docked based on their complementarity. Then primers A and F were added and the full-length DNA encoding the reconstructed single chain Fv antibody MABL-1 (second PCR) was amplified. In the first PCR, plasmid pGEM-M1H encoding the V region of the H chain of the MABL-1 antibody (see Example 2), plasmid pSC-DPI, which contains the DNA sequence encoding the linker region, containing: Gly Gly Gly was used as a template Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser (SEQ ID NO 19) (Huston, JS, et al., Proc. Natl. Acad. Sci. USA, 85, 5879-5883, 1988), and plasmid pGEM- MIL encoding the V region of the L chain of the MABL-1 antibody (see Example 2), respectively.

50 μl of the solution for the first PCR stage contain 5 μl of 10-fold PCR buffer II, 2 mM MgCl ₂ , 0.16 mm dNTP, 2.5 units of AmpliTaq Gold DNA polymerase (Perkin Elmer), 0.4 μM of each of the primers and 5 ng of each of the DNA templates. The PCR solution was preheated at an initial temperature of 94 ° C for 9 minutes and then heated at 94 ° C for 1 minute, at 65 ° C for 1 minute, and at 72 ° C for 1 minute 20 seconds in the indicated sequence. This temperature cycle was repeated 35 times and then the reaction mixture was further heated at 72 ° C. for 7 minutes.

PCR products AB (371 bp), CD (63 bp) and EF (384 bp) were purified using the QIAquick PCR Purification Kit (Qiagen) and assembled into a second PCR . In the second PCR, 98 μl of PCR solution containing 120 ng of the first PCR product A-B, 20 ng of PCR product CD and 120 ng of PCR product EF, 10 μl of 10-fold PCR buffer II, 2 mM MgCl ₂ , 0, 16 mM dNTP, 5 AmpliTaq Gold DNA polymerase units (Perkin Elmer) were preheated at an initial temperature of 94 ° C for 8 minutes and then heated at 94 ° C for 2 minutes, at 65 ° C for 2 minutes, and at 72 ° C for 2 minutes in the indicated sequence. This temperature cycle was repeated twice and then 0.4 μM of each of primers A and F were added to the reaction, respectively. This mixture was preheated at an initial temperature of 94 ° C for 1 minute and then heated at 94 ° C for 1 minute, at 65 ° C for 1 minute, and at 72 ° C for 1 minute 20 seconds in that order. This temperature cycle was repeated 35 times and then the reaction mixture was further heated at 72 ° C. for 7 minutes.

The 843 bp DNA fragment obtained in the second PCR was purified and digested with NcoI and EcoRI. The resulting DNA fragment was cloned into pSCFVT7 vector. The expression vector pSCFVT7 contains a pelB signal sequence suitable for the periplasmic expression system of E. coli (Lei, S.P., et al., J. Bacteriology, 169, 4379-4383, 1987). After DNA sequencing, a plasmid containing a DNA fragment encoding the correct amino acid sequence of the reconstructed single chain Fv antibody MABL-1 was designated as "pscM1" (see FIG. 5). The nucleotide sequence and amino acid sequence of this reconstructed single-chain Fv antibody MABL-1 contained in the plasmid pscM1 are shown in SEQ ID NO 20.

The pscM1 vector was modified using the PCR method to obtain a vector expressing the reconstructed single chain Fv antibody MABL-1 in mammalian cells. The resulting DNA fragment was introduced into the pCHO1 expression vector. This expression vector, pCHO1, was obtained by digestion of DHFR-ΔE-rvH-PM1-f (WO 92/19759) EcoRI and SmaI to exclude the antibody gene and attach the EcoRI-NotI-BamHI adapter (Takara Shuzo) to it.

As a direct PCR primer, the Sal-VHS primer shown in SEQ ID NO 21 was designed to hybridize with DNA encoding the N-terminus of the V region of the H chain and contain a SalI restriction enzyme recognition site. As a reverse primer for PCR, the FRHlanti primer shown in SEQ ID NO 22 was designed to hybridize with DNA encoding the end of the first frame sequence.

100 μl of PCR solution containing 10 μl of 10-fold PCR buffer II, 2 mM MgCl ₂ , 0.16 mM dNTP, 5 units of AmpliTaq Gold DNA polymerase, 0.4 μM of each primer and 8 ng of DNA template (pscMl ), preheated at an initial temperature of 95 ° C for 9 minutes and then heated at 95 ° C for 1 minute, at 60 ° C for 1 minute and at 72 ° C for 1 minute 20 seconds in the indicated sequence. This temperature cycle was repeated 35 times and then the reaction mixture was further heated at 72 ° C. for 7 minutes.

The PCR product was purified using the QIAquick PCR Purification Kit (Qiagen) and then digested with SaiI and MboII to obtain a DNA fragment encoding the N-terminus of the reconstructed single chain Fv antibody MABL-1. The pscMl vector was digested with MboII and EcoRI to obtain a DNA fragment encoding the C-terminus of the reconstructed single chain Fv antibody MABL-1. The SalI-MboII DNA fragment and the MboII-EcoRI DNA fragment were cloned into the pCHO1-Igs vector. After DNA sequencing, a plasmid containing the desired DNA sequence was designated as “pCOM1" (see FIG. 6). The expression vector pCHO1-Igs contains a mouse IgG1 signal sequence suitable for an expression-secretion system in mammalian cells (Nature, 322, 323-327, 1988). The nucleotide sequence and amino acid sequence of the reconstructed single-chain Fv of the MABL-1 antibody contained in the pCHOMI plasmid are shown in SEQ ID NO 23.

5.2 Obtaining a reconstructed single chain Fv antibody MABL-2

Reconstructed single-chain Fv antibodies MABL-2 were obtained in accordance with the above example 5.1. In the first PCR step, the plasmid pGEM-M2H encoding the V region of the MABL-2 H chain (see Example 2) was used instead of pGEM-M1H and the plasmid pGEM-M2L encoding the V region of the MABL-2 L chain (see Example 2), instead of pGEM-M1L, to obtain the plasmid pscM2, which contains a DNA fragment encoding the desired amino acid sequence of the single-chain Fv antibody MABL-2. The nucleotide sequence and the amino acid sequence of the reconstructed single chain Fv of the MABL-2 antibody contained in the plasmid pscM21 are shown in SEQ ID NO 24.

The pscM2 vector was modified using the PCR method to obtain a vector, pCHOM2, for expression in mammalian cells, which contains a DNA fragment encoding the correct amino acid sequence of the reconstructed single-chain Fv antibody MABL-2. The nucleotide sequence and amino acid sequence of the reconstructed single-chain Fv of the MABL-2 antibody contained in the plasmid pCOM2 are shown in SEQ ID NO 25.

5.3 Transfection into COS7 cells

The pCHOM2 vector was tested in COS7 cells to observe the transient expression of the reconstructed single chain Fv antibody MABL-2.

COS7 cells were transformed with the pCHOM2 vector by electroporation using a Gene Pulser device (BioRad). DNA (10 μg) and 0.8 ml of PBS containing 1 × 10 ⁷ cells / ml were added to the cuvette. The mixture was treated with a pulse at 1.5 kV at an electric capacitance of 25 μF.

After "reconstitution" for 10 minutes at room temperature, the electroporated cells were transferred to IMDM culture medium (Gibco BRL) containing 10% fetal calf serum. After culturing for 72 hours, the supernatant was collected, centrifuged to remove cell fragments, and collected.

5.4 Detection of reconstructed single chain Fv antibody MABL-2 in the culture supernatant of COS7 cells

The presence of the single chain Fv antibody MABL-2 in the culture supernatant of COS7 cells that were transfected with the pCHOM2 vector was confirmed by Western blotting.

The culture supernatant of COS7 cells transfected with the pCHOM2 vector and the culture supernatant of COS7 cells transfected with pCHO1 were subjected to SDS electrophoresis as a control and transferred onto a REINFORCED NC membrane (Schleicher & Schuell). The membrane was blocked with 5% skim milk (Morinaga Nyu-guo), washed with 0.05% Tween 20-PBS and mixed with anti-FLAG antibody (Sigma). The membrane was incubated at room temperature, washed and alkaline phosphatase conjugated anti-mouse IgG antibody (Zymed) was added. After incubation and washing at room temperature, a substrate solution (Kirkegaard Perry Laboratories) was added to develop color (Fig. 7).

The protein giving a specific reaction to the FLAG peptide was found only in the culture supernatant of COS7 cells with the introduced pCHOM2 vector, and therefore it was confirmed that the reconstructed single chain Fv of the MABL-2 antibody was secreted into this culture supernatant.

5.5 Flow cytometry

Flow cytometry was performed using the aforementioned COS7 cell culture supernatant to measure antigen binding. The culture supernatant of COS7 cells expressing the reconstructed single-chain Fv of the MABL-2 antibody, or the culture supernatant of COS7 cells transformed with the pCHO1 vector as a control was added to 2 × 10 ⁵ cells of the murine L1210 mouse leukemia cell line expressing human integrin-associated protein (IAP), or pCOSI transformed L1210 cell line as a control. After incubation on ice and washing, a murine anti-FLAG antibody (Sigma) was added. Then the cells were incubated and washed. Then, a FITC-labeled anti-mouse IgG antibody (Becton Dickinson) was added thereto, and the cells were again incubated and washed. Then, the fluorescence intensity was measured using a FACScan instrument (Becton Dickinson).

Since the single-chain Fv of the MABL-2 antibody specifically binds to L1210 cells expressing human IAP, it was confirmed that the reconstructed single-chain Fv of the MABL-2 antibody has an affinity for the human integrin-associated protein (IAP) (see FIGS. 8-11).

5.6 Competitive ELISA

The binding activity of the reconstructed single chain Fv antibody MABL-2 was measured based on the inhibitory activity against binding of murine monoclonal antibodies to the antigen.

An anti-FLAG antibody, adjusted to a concentration of 1 μg / ml, was added to each well of a 96-well plate and incubated at 37 ° C for 2 hours. After washing, 1% BSA-PBS was blocked. After incubation and washing at room temperature, the culture supernatant of COS7 cells, into which the secretory type human IAP antigen gene (SEQ ID NO 26) was introduced, was diluted with PBS to a double volume and added to each well. After incubation and washing at room temperature, a mixture of 50 μl of biotinylated MABL-2 antibody, adjusted to a concentration of 100 ng / ml, and 50 μl of sequentially diluted supernatant of COS7 cells expressing the reconstructed single-chain Fv of MABL-2 antibody was added to each well. After incubation and washing at room temperature, alkaline phosphatase conjugated streptavidin (Zymed) was added to each well. After incubation and washing at room temperature, a substrate solution (Sigma) was added and the optical density of the reaction mixture in each well was measured at 405 nm.

The results showed that the reconstructed single-chain Fv of the MABL-2 antibody (MABL2-scFv) clearly inhibited a concentration dependent binding of the murine MABL-2 antibody to the human IAP antigen compared to the COS7 cell culture supernatant with pCHO1 introduced as a control (Fig. 12) . Thus, it is assumed that the reconstructed single chain Fv of the MABL-2 antibody has the correct structure for each of the V regions of the murine monoclonal antibody MABL-2.

5.7 Inductive apoptosis in vitro

The apoptosis inducing effect of the reconstructed single-chain Fv antibody MABL-2 was investigated by annexin-V staining (Boehringer Mannheim) using L1210 cells transfected with the human IAP gene, L1210 cells transfected with pCOS2 vector as a control, and CCRF-CEM cells.

To 1 × 10 ⁵ cells of each of the above cell lines, a culture supernatant of COS7 cells expressing the reconstructed single chain Fv antibody MABL-2 or culture supernatant of COS7 cells transfected with pCHO1 vector was added as a control at a final concentration of 50% and these mixtures were cultured in within 24 hours. Annexin-V staining was then performed and the fluorescence intensity was measured using a FACScan instrument (Becton Dickinson).

Annexin-V staining results are shown in FIGS. 13-18, respectively. The dots in the lower left quadrant represent living cells, the dots in the lower right quadrant represent cells in the early stage of apoptosis, and the dots in the upper right quadrant represent cells in the early stage of apoptosis. These results show that the reconstructed single-chain Fv of the MABL-2 antibody (MABL2-scFv) significantly induced the death of L1210 cells specific for the human IAP antigen (FIGS. 13-16), and that the reconstructed single-chain Fv also induced significant death of CCRF-β cells. CEM in comparison with control (figures 17-18).

5.8 Expression of MABL-2-derived single chain Fv in CHO cells

CHO cells were transfected with the pCHOM2 vector to obtain a CHO cell line that constant expression of single-chain Fv (polypeptide) derived from the MABL-2 antibody.

CHO cells were transformed with the pCHOM2 vector by electroporation using a Gene Pulser device (BioRad). A mixture of DNA (10 μg) and 0.7 ml of PBS containing CHO cells (1 × 10 ⁷ cells / ml) was added to the cuvette. The mixture was treated with a pulse at 1.5 kV at an electric capacitance of 25 μF. After "reconstitution" for 10 minutes at room temperature, the electroporated cells were transferred to a nucleic acid-free α-MEM culture medium (Gibco BRL) containing 10% fetal calf serum and cultured. The expression of the desired protein in the obtained clones was confirmed by SDS-PAGE electrophoresis, and a high expression clone was selected as a cell line producing single chain Fv derived from MABL-2 antibody. This cell line was cultured in serum-free medium CHO-S-SEM II (Gibco BRL) containing 10 nM methotrexate (Sigma). Then, the culture supernatant was collected, centrifuged to remove cell fragments, and selected.

5.9 Purification of single-chain Fv derived from MABL-2 produced in CHO cells

The culture supernatant of the CHO cell line expressing single chain Fv obtained in Example 5.8 was concentrated twenty times using an artificial dialysis cassette (PAN130SF, ASAHI MEDICALS). The concentrated solution was stored at -20 ° C and thawed before cleaning.

Purification of single chain Fv from the CHO cell culture supernatant was performed using three types of chromatography, i.e. Blue-Sepharose, Hydroxyapatite and Gel Filtration.

(1) Column Chromatography on Blue-Sepharose

The concentrated supernatant was diluted ten times with 20 mM acetate buffer (pH 6.0) and centrifuged to remove insoluble matter (10,000 (rpm, 30 minutes). This supernatant was applied to a Blue-Sepharose column (20 ml) equilibrated with the same After washing the column with the same buffer, the proteins adsorbed on the column were eluted with a stepwise NaCl gradient in the same buffer, 0.1, 0.2, 0.3, 0.5, and up to 1.0 M. Converted in free volume fraction and each of the eluted fractions were analyzed by SDS-PAGE by electrophoresis. the presence of single-chain Fv was detected (these fractions were eluted with 0.1-0.3 M NaCl), combined and concentrated approximately 20 times using CentriPrep 10 (Amicon).

(2) Hydroxyapatite

The concentrated solution obtained in (1) was diluted 10 times with 10 mM phosphate buffer (pH 7.0) and applied to a hydroxyapatite column (20 ml, BioRad). The column was washed with 60 ml of 10 mM phosphate buffer (pH 7.0). Then, the proteins adsorbed on the column were eluted with a linear gradient of sodium phosphate buffer to 200 mM (see FIG. 19). Analysis of each fraction by SDS-PAGE electrophoresis confirmed the presence of single chain Fv in fraction A and fraction B.

(3) Gel Filtration

Each of the fractions A and B obtained in (2) was separately concentrated using CentriPrep-10 and applied to a TSKgel G3000SWG column (21.5 × 600 mm) balanced with 20 mM acetate buffer (pH 6.0) containing 0 , 15 M NaCl. Chromatograms are shown in FIG. Analysis of these fractions by SDS-PAGE electrophoresis confirmed that both major peaks (AI and BI) are the desired single chain Fv. In this gel filtration analysis, fraction A was eluted with an apparent molecular weight of 36 kDa, and fraction B was eluted with an apparent molecular weight of 76 kDa. Purified single chain Fv (AI, BI) was analyzed using 15% SDS-polyacrylamide gel. Samples were processed in the absence or presence of a reducing agent, and electrophoresis was carried out in accordance with the Laemmli method. Then the protein was stained with Coomassie brilliant blue. As shown in FIG. 21, both AI and BI gave a single band with an apparent molecular weight of 35 kDa, regardless of the absence or presence of a reducing agent. From the above, it was concluded that AI is a monomer of a single chain Fv, and BI is a non-covalently linked dimer of a single chain Fv. Gel-filtration analysis of the AI and BI fractions with TSKgel G3000SWG column (7.5 × 60 mm) showed that the peak of the monomer is detected only in the AI fraction, and the peak of the dimer is detected only in the BI fraction (Fig. 22). The dimer fraction (BI fraction) was 4 percent of the total number of single chain Fv. More than 90% of this dimer in the dimer fraction was stably maintained for more than a month at 4 ° C.

5.10 Construction of a vector expressing single-chain Fv derived from MABL-2 antibody in an E. coli cell

The pscM2 vector was modified by PCR to obtain a vector that efficiently expresses the single-chain Fv of the MABL-2 antibody in E. coli cells. The resulting DNA fragment was inserted into the expression vector pSCFVT7.

As a direct PCR primer, the Nde-VHSm02 primer shown in SEQ ID NO 27 was designed to hybridize with DNA encoding the N-terminus of the V region of the H chain and contain a start codon and an NdeI restriction enzyme recognition site. As a reverse primer for PCR, the VLAS primer shown in SEQ ID NO 28 was designed to hybridize with DNA encoding the C-terminus of the V region of the L chain and contain two stop codons and an EcoRI restriction enzyme recognition site. The Nde-VHSm02 forward primer contains five point mutations in the portion hybridizing to DNA encoding the N-terminus of the V region of the H chain for efficient expression in E. coli.

100 μl of a PCR solution containing 10 μl of a 10-fold PCR buffer No. 1.1 mm MgCl ₂ , 0.2 mm dNTP, 5 units of KOD DNA polymerase (all from Toyobo), 1 μm of each primer and 100 ng of DNA matrices (pscM2) were heated at 98 ° C for 15 seconds, at 65 ° C for 2 seconds, and at 72 ° C for 30 seconds in the sequence shown. This temperature cycle was repeated 25 times.

The PCR product was purified using a QIAquick PCR Purification Kit (Qiagen) and then NdeI and EcoRI were digested and then the resulting DNA fragment was cloned into the pSCFVT7 vector from which the pelB signal sequence was removed by digestion with NdeI and EcoRI. After DNA sequencing, the resulting plasmid containing the DNA fragment with the desired DNA sequence was designated as “pscM2DEm02” (see FIG. 23). The nucleotide sequence and the amino acid sequence of a single chain Fv derived from the MABL-2 antibody contained in the plasmid pscM2DEm02 are shown in SEQ ID NO 29.

5.11 Expression of single-chain Fv derived from MABL-2 antibody in E. coli cells

E. coli BL21 (DE3) pLysS (Stratagene) was transformed with the pscM2DEm02 vector to obtain an E. coli strain expressing single chain Fv derived from MABL-2 antibody. The obtained clones were analyzed for expression of the desired protein by SDS-PAGE electrophoresis, and a clone with a high expression level was selected as a strain producing single-chain Fv originating from the MABL-2 antibody.

5.12 Purification of Single Chain Fv Derived from MABL-2 Antibody Produced in E. coli

A single E. coli colony obtained by transformation was cultured in 3 ml of LB medium at 28 ° C for 7 hours and then in 70 ml of LB medium at 28 ° C overnight. This pre-culture was transferred to 7 L of LB medium and cultured at 28 ° C. with stirring at 300 rpm using a Jar fermenter. When the absorption of the medium reached O.D. = 1.5, bacteria induced 1 mM IPTG and then cultured for 3 hours.

The culture medium was centrifuged (10,000 × g, 10 minutes), and the precipitated bacteria were collected. To these bacteria, 50 mM Tris-HCl buffer (pH 8.0) containing EDTA, 0.1 M NaCl and 1% Triton X-100 was added and the bacteria were destroyed by sonication (power: 4, duty cycle: 70%, 1 minute × 10 times). A suspension of the destroyed bacteria was centrifuged (12000 × g, 10 minutes) to precipitate inclusion bodies. The isolated inclusion bodies were mixed with 50 mM Tris-HCl buffer (pH 8.0) containing 5 mM EDTA, 0.1 M NaCl and 4% Triton X-100, and again treated with ultrasound (power: 4, duty cycle: 50 %, 30 seconds × 2 times) and centrifuged (12000 × g, 10 minutes) to isolate the desired protein as a precipitate and to remove impurity proteins in the supernatant.

Inclusion bodies containing the desired protein were lysed in 50 mM Tris-HCl buffer (pH 8.0) containing 6 M urea, 5 mM EDTA, 0.1 M NaCl, and loaded onto a Sephacryl S-300 gel filtration column ( 5 × 90 cm, Amersham Pharmacia) equilibrated with 50 mM Tris-HCl buffer (pH 8.0) containing 4 M urea, 5 mM EDTA, 0.1 M NaCl and 10 mM mercaptoethanol, at a flow rate of 5 ml / min to remove associated single-chain Fv with high molecular weight. The obtained fractions were analyzed by SDS-PAGE and high purity fractions of this protein were diluted with the buffer used in gel filtration to OD ₂₈₀ = 0.25. These fractions were then dialyzed three times against 50 mM Tris-HCl buffer (pH 8.0) containing 5 mM EDTA, 0.1 M NaCl, 0.5 M Arg, 2 mM glutathione in reduced form and 0.2 mM glutathione in oxidized form, so that protein folding occurs. Further, this fraction was dialyzed three times against 20 mM acetate buffer (pH 6.0) containing 0.15 M NaCl to change the buffer.

The dialyzed product was applied to a 200 pg Superdex gel filtration column (2.6 × 60 cm, Amersham Pharmacia), equilibrated with 20 mM acetate buffer (pH 6.0) containing 0.15 NaCl to remove a small amount of high molecular weight protein that is intermolecular stitched together SS. As shown in FIG. 24, two peaks, the main and minor peaks, were eluted after broad peaks, which are thought to be of high molecular weight aggregate. Analysis by SDS-PAGE electrophoresis (see FIG. 21) and elution positions of these two peaks by gel filtration analysis suggest that the main peak corresponds to the single chain Fv monomer, and the minor peak corresponds to the non-covalently bound single chain Fv dimer. Non-covalently bound dimer accounts for 4 percent of the total number of single chain Fv.

5.13 Apopotosis-inducing in vitro activity of single chain Fv derived from MABL-2 antibody

The apoptosis-inducing activity of the single-chain Fv antibody MABL-2 (MABL2-scFv) produced by CHO and E. coli cells was examined in accordance with two protocols by the Annexin-V staining method (Boehringer Mannheim) using L1210 cells (hIAP / L1210), in that the human IAP gene was introduced.

In the first protocol, antibody samples at a final concentration of 3 μg / ml were added to 5 × 10 ⁴ hIAP / L1210 cell line cells and cultured for 24 hours. Antibody samples were analyzed, i.e. monomer and dimer of single-chain Fv MABL-2 from CHO cells obtained in 5.9, monomer and dimer of single-chain Fv MABL-2 from E. coli obtained in example 5.12, and mouse IgG antibody as a control. After cultivation, annexin-V staining was performed and the fluorescence intensity was measured using a FACScan instrument (Becton Dickinson).

In the second protocol, antibody samples at a final concentration of 3 μg / ml were added to 5 × 10 ⁴ hIAP / L1210 cell line cells, cultured for 2 hours and mixed with anti-FLAG antibody (Sigma) at a final concentration of 15 μg / ml and additionally cultivated for 22 hours. Analyzed samples of the antibody represented by the monomer single-chain Fv MABL-2 from CHO cells obtained in 5.9, and mouse IgG antibody as a control. After cultivation, annexin-V staining was performed and fluorescence intensity was measured using a FACScan instrument.

The results of the analysis by staining for Annexin-V are shown in figures 25-31. These results show that dimers of the single-chain Fv MABL-2 polypeptide produced in CHO and E. coli cells strongly induced cell death (FIGS. 26, 27) compared to the control (FIG. 25), while apoptosis no inducing effect was observed in the case of monomers of the single-chain Fv MABL-2 polypeptide produced in CHO and E. coli cells (figures 28, 29). When the anti-FLAG antibody was shared, the monomer of the single chain Fv polypeptide derived from the MABL-2 antibody produced in CHO cells strongly induced cell death (FIG. 31) compared to the control (FIG. 30).

5.14 Antitumor effect of the monomer and dimer of the scFv / CHO polypeptide in a murine model of human myeloma

(1) Quantification of Human IgG in Mouse Serum

Measurement of human IgG (M protein) produced by human myeloma cells and contained in mouse serum was performed using ELISA as follows. 100 μl of goat anti-human IgG antibodies (Biosource, Lot # 902) diluted to 1 μg / ml with 0.1% bicarbonate buffer (pH 9.6) was added to each well of a 96-well plate (Nunc) and incubated at 4 ° C during the night, so that these antibodies are immobilized. After blocking, 100 μl of stepwise diluted mouse serum or human IgG (Cappel, Lot # 00915) was added to each well as standard and incubated for 2 hours at room temperature. After washing, 100 μl of alkaline phosphatase-labeled anti-human IgG antibody (Biosource, Lot # 6202) was added which was diluted 5,000 times and incubated for 1 hour at room temperature. After washing, a substrate solution was added. After incubation, absorbance was measured at 405 nm using a Microplate Reader Model 3550 (BioRad). The concentration of human IgG in mouse serum was calculated from a calibration curve based on the absorbance values obtained from human IgG as a standard.

(2) Obtaining antibodies for administration

The monomer and dimer of the scFv / CHO polypeptide on the day of administration were diluted, respectively, to a concentration of 0.4 mg / ml or 0.25 mg / ml by sterile filtration with PBS (-) to obtain samples for administration.

(3) Obtaining a murine model of human myeloma

A murine model of human myeloma was obtained as follows. In vivo KPMM2 cells (Japanese Patent Application (JP-Appl. 7-236475) using SCID mice (Japan Clare) were suspended in RPMI1640 medium (Gibco BRL) containing 10% fetal calf serum (Gibco BRL), and brought to 3 × 10 ⁷ cells / ml, 200 μl of a suspension of CRMM2 cells (6 × 10 ⁶ cells / mouse) were transplanted into a SCID mouse (males, 6 weeks old) through the tail vein, which was injected subcutaneously asio- the day before the transplantation GM1 antibody (Wako Junyaku, 1 vial, dissolved in 5 ml).

(4) Introduction of antibodies

Antibody samples prepared in (2): monomer (250 μl) and dimer (400 μl) were injected with mouse models of human myeloma obtained in (3) through the tail vein. Administration was started three days after transplantation of CRMM2 cells and was performed twice daily for three days. As a control, 200 μl of filter-sterilized PBS (-) was administered in the same manner twice daily for three days through the tail vein. Each group consisted of seven mice.

(5) Evaluation of the antitumor effect of the monomer and dimer of the scFv / CHO polypeptide using a mouse model of human myeloma

The antitumor effect of the monomer and dimer of the scFv / CHO polypeptide using a mouse model of human myeloma was evaluated by changing the concentration of human IgG (M protein) in mouse serum and mouse survival time. The change in the concentration of human IgG was determined by measuring it in mouse serum collected 24 days after transplantation of CRMM2 cells, using the ELISA method described above in (1). The amount of human serum IgG (M-protein) in the serum of the group (control) that was injected with PBS (-) increased to about 8500 μg / ml, while the amount of human IgG in the group that was injected with the scFv / CHO dimer was significantly lower , i.e. 1/10 or less of the amount in the control group. Thus, these results show that the scFv / CHO dimer strongly inhibits the growth of CRMM2 cells (Figure 32). As shown in FIG. 33, a significant increase in survival time was observed in the scFv / CHO dimer-treated group compared to the PBS (-) -received group.

The above confirms that the scFv / CHO dimer has an antitumor effect in a murine model of human myeloma. It is believed that the antitumor effect of the scFv / CHO dimer, a modified antibody of the present invention, is the result of the apoptosis-inducing action of this modified antibody.

5.15 Hemagglutination Test

A hemagglutination test and determination of hemagglutination was carried out in accordance with Immuno-Biochemical Investigation, Zoku-Seikagaku Jikken Koza, edited by the Biochemical Society of Japan, published by Tokio Kagaku Dojin.

Blood was taken from a healthy donor using heparin-treated syringes and washed with PBS (-) three times and then a suspension of red blood cells was prepared with a final concentration of 2% in PBS (-). The test samples were the MABL-2 antibody, the monomer and dimer of the single chain Fv polypeptide produced by CHO cells, and the monomer and dimer of the single chain Fv polypeptide produced by E. coli, and control was mouse IgG (Zymed). To study the effect of hemagglutination, round-bottom 96-well plates, available from Falcon, were used. 50 μl per well of the above antibody samples and 50 μl of a 2% suspension of red blood cells were added and mixed in the wells. After incubation for 2 hours at 37 ° C, the reaction mixtures were kept at 4 ° C overnight and their hemagglutination was determined. As a control, 50 μl per well of PBS (-) was used and the hemagglutination test was carried out in the same way. Mouse IgG and the MABL-2 antibody were used at a final antibody concentration of 0.01, 0.1, 1.0, 10.0, or 100.0 μg / ml. Single chain Fv was used at a final concentration of 0.004, 0.04, 0.4, 4.0, 40.0 or 80.0 μg / ml and an additional 160 μg / ml only in the case of a dimer of this polypeptide produced by E. coli. The results are shown in Table 2. In the case of the MABL-2 antibody, hemagglutination was observed at a concentration of more than 0.1 μg / ml, while hemagglutination was not observed for both the monomer and the single chain Fv dimer.

Example 6 Modified sc (Fv) ₂ antibody containing two H chain V regions and two L chain V regions and scFv antibodies MABL-2 having linkers of different lengths

6.1 Construction of a plasmid expressing sc (Fv) ₂ antibodies MABL-2

To obtain a plasmid expressing the modified antibody [sc (Fv) ₂ ], which contains two V-regions of the H chain and two V-regions of the L chain, derived from the antibody MABL-2, the aforementioned plasmid pCHOM2, which contains DNA encoding scFv derived from MABL-2 described above was modified by PCR, as described below, and the resulting DNA fragment was inserted into pCHOM2.

The primers used for PCR are EF1 primer (SEQ ID NO 30) as a sense primer that is designed to hybridize with DNA encoding EF1α, and an antisense primer (SEQ ID NO 19) that is designed to hybridize with DNA encoding the C-terminus of the V region of the L chain, and contained the DNA sequence encoding the linker region, and the VLLAS primer containing the restriction enzyme recognition site SaIl (SEQ ID NO 31).

100 μl of PCR solution contain 10 μl of 10-fold PCR buffer No. 1, 1 mm MgCl ₂ , 0.2 mm dNTP (dATP, dGTP, dCTP and dTTP), 5 units of KOD DNA polymerase (Toyobo, Inc.), 1 μM of each of the primers and 100 ng of the DNA template (pCOM2). The PCR solution was heated at 94 ° C for 30 seconds, at 50 ° C for 30 seconds, and at 74 ° C for 1 minute in that order. This temperature cycle was repeated 30 times.

The PCR product was purified using a QIAquick PCR Purification Kit (Qiagen) and SalI was digested. The resulting DNA fragment was cloned into the pBlueScript KS ⁺ vector (Toyobo, Inc.). After DNA sequencing, a plasmid containing the desired DNA sequence was digested with SalI and the resulting DNA fragment was coupled using the Rapid DNA Ligation Kit (Boehringer Mannheim) with pCHOM2 digested with SalI. After DNA sequencing, a plasmid containing the desired DNA sequence was designated as “pCHOM2 (Fv) ₂ ” (see FIG. 34). The nucleotide sequence and amino acid sequence of the sc (Fv) _{2 region of the} MABL- ₂ antibody contained in the plasmid pCHOM2 (Fv) ₂ are shown in SEQ ID NO 32.

6.2 Obtaining a plasmid expressing scFv antibodies MABL-2 having linkers of different lengths

ScFv polypeptides containing linkers of various lengths and V regions, which are arranged in the following order [H chain] - [L chain] (hereinafter referred to as “HL”) or [L chain] - [H chain] (hereinafter referred to as “LH”) was prepared using cDNA encoding the H chain and L chain derived from MABL-2 as a template, as mentioned above.

To construct an HL-type scFv, the PCR procedure was performed using pCHOM2 (Fv) ₂ as a template. In the PCR step, a pair of primers CFHL-Fl (SEQ ID NO 33) and CFHL-R2 (SEQ ID NO 34) or a pair of primers CFHL-F2 (SEQ ID NO 35) and CFHL-R1 (SEQ ID NO 36) and DNA were used -polymerase KOD. The PCR procedure was carried out by repeating 30 times the temperature cycle, consisting of 94 ° C for 30 seconds, 60 ° C for 30 seconds and 72 ° C for 1 minute in that order to obtain cDNA for the H chain containing the leader sequence at the 5'-end, or cDNA for the L chain containing the FLAG sequence at its 3'-end. The obtained cDNAs for the H chain and L chain were mixed and PCR was performed by repeating 5 times the temperature cycle consisting of 94 ° C for 30 seconds, 60 ° C for 30 seconds and 72 ° C for 1 minute in that order with using this mixture as a KOD matrix and polymerase. Primers CFHL-F1 and CFHL-R1 were added to this reaction mixture and then a PCR reaction was carried out by repeating the above-mentioned temperature cycle 30 times to obtain cDNA for the HL-0 type without a linker.

To construct an LH-type scFv, the PCR reaction was performed using pGEM-M2L and pGEM-M2H as a template, which contain cDNAs encoding the V region of the L chain and the V region of the H chain of the MABL-2 antibody, respectively (see Japanese Patent Application JP-Appl. 11-63557). Used a pair of primers T7 (SEQ ID NO 37) and CFLH-R2 (SEQ ID NO 38) or a pair of primers CFLH-F2 (SEQ ID NO 39) and CFLH-R1 (SEQ ID NO 40) and KOD polymerase (Toyobo, Inc. ) The PCR reaction was performed by repeating 30 times the temperature cycle, consisting of 94 ° C for 30 seconds, 60 ° C for 30 seconds and 72 ° C for 1 minute in this order to obtain cDNA L-chain containing a leader sequence of 5 '-end, or cDNA of the H chain containing the FLAG sequence at its 3'-end. The resulting cDNA of the L chain and H chain were mixed and PCR was performed using this mixture as a KOD template and KOD polymerase by repeating 5 times the temperature cycle consisting of 94 ° C for 30 seconds, 60 ° C for 30 seconds and 72 ° C within 1 minute in that order. Primers T7 and CFHL-R1 were added to this reaction mixture and the reaction was carried out by repeating 30 times of the above temperature cycle. The reaction product was used as a template and PCR was performed using a pair of primers CFLH-F4 (SEQ ID NO 41) and CFLH-R1 by repeating 30 times the temperature cycle consisting of 94 ° C for 30 seconds, 60 ° C for 30 seconds and 72 ° C for 1 minute in the indicated order to obtain LH-0 type cDNA without a linker.

The obtained cDNAs of types LH-0 and HL-0 were digested with restriction enzymes EcoRI and BamHI (Takara Shuzo) and the cleaved cDNAs were inserted into the mammalian expression plasmid INPEP4 using the Ligation High kit (Toyobo Inc.), respectively. Competent E. coli JM109 cells (Nippon Gene) were transformed with each plasmid and the desired plasmids were isolated from transformed E. coli using the QIAGEN Plasmid Maxi Kit (Qiagen). Thus, plasmids pCF2LH-0 and pCF2HL-0 were obtained.

To construct HL-type expression plasmids containing linkers of various sizes, pCF2LH-0 as a matrix and CFHL-X3 (SEQ ID NO 42), CFHL-X4 (SEQ ID NO 43), CFHL-X5 (SEQ ID NO 44) were used , CFHL-X6 (SEQ ID NO 45) or CFHL-X7 (SEQ ID NO 46) as a sense primer and BGH-1 primer (SEQ ID NO 47) as an antisense primer that is complementary to a vector sequence. PCR reactions were performed using KOD polymerase by repeating 30 times the temperature cycle consisting of 94 ° C for 30 seconds, 60 ° C for 30 seconds and 72 ° C for 1 minute in this order, and the reaction products were digested with restriction enzymes XhoI and BamHI (Takara Shuzo). Cleaved fragments were inserted between the XhoI and BamHI sites in pCF2HL-0 using the Ligation High kit (Toyobo Inc.), respectively. Competent E. coli JM109 cells were transformed with each plasmid and the desired plasmids were isolated from transformed E. coli using the Qiagen Plasmid Maxi Kit. Thus, expression plasmids pCF2HL-3, pCF2HL-4, pCF2HL-5, pCF2HL-6 and pCF2HL-7 were obtained.

To construct an expression plasmid for transient expression in COS7 cells, plasmids pCF2HL-0, pCF2HL-3, pCF2HL-4, pCF2HL-5, pCF2HL-6 and pCF2HL-7 were digested with restriction enzymes EcoRI and BamHI (Takara fragments of size 800 and about .about. purified by agarose gel electrophoresis. The resulting fragments were inserted between the EcoRI and BamHI sites in the pCOSI expression plasmid for expression in mammalian cells using the Ligation High kit (Toyobo Inc.), respectively. Competent E. coli DH5α cells (Toyobo Inc.) were transformed with each plasmid and the desired plasmids were isolated from transformed E. coli using the Qiagen Plasmid Maxi Kit. Thus, plasmids CF2HL-0 / pCOS1, CF2HL-3 / pCOS1, CF2HL-4 / pCOS1, CF2HL-5 / pCOS1, CF2HL-6 / pCOS1 and CF2HL-7 / pCOS1 were obtained.

As a typical example of these plasmids, the construction of the plasmid CF2HL-0 / pCOS1 is shown in FIG. 35, and the nucleotide sequence and amino acid sequence of the MABL2-scFv <HL-0> contained in this plasmid are shown in SEQ ID NO 48. The nucleotide sequences and amino acid the sequence of linker sites in these plasmids is also shown in Fig.36.

To construct LH-type expression plasmids containing linkers of various sizes, pCF2LH-0 as the template and CFLH-X3 (SEQ ID NO 49), CFLH-X4 (SEQ ID NO 50), CFLH-X5 (SEQ ID NO 51) were used , CFLH-X6 (SEQ ID NO 52) or CFLH-X7 (SEQ ID NO 53) as a sense primer and BGH-1 primer as an antisense primer that is complementary to a vector sequence. The PCR reaction was performed using KOD polymerase by repeating 30 times the temperature cycle consisting of 94 ° C for 30 seconds, 60 ° C for 30 seconds and 72 ° C for 1 minute in this order, and the reaction products were digested with restriction enzymes XhoI and BamHI . Cleaved fragments were inserted between the XhoI and BamHI sites in pCF2LH-0 using the Ligation High kit, respectively. Competent E. coli DH5α cells (Toyobo Inc.) were transformed with each plasmid and the desired plasmids were isolated from transformed E. coli using the Qiagen Plasmid Maxi Kit. Thus, plasmids pCF2LH-3, pCF2LH-4, pCF2LH-5, pCF2LH-6 and pCF2LH-7 were obtained.

To construct an expression plasmid for transient expression in COS7 cells, plasmids pCF2LH-0, pCF2LH-3, pCF2LH-4, pCF2LH-5, pCF2LH-6, and pCF2LH-7 were digested with restriction enzymes EcoRI and BamHI (Takara Shuzo fragments of size about 800 and .about. purified by agarose gel electrophoresis. The resulting fragments were introduced between EcoRI and BamHI sites in the expression plasmid pCOS1 for expression in mammalian cells using the Ligation High kit, respectively. Competent E. coli DH5α cells (Toyobo Inc.) were transformed with each plasmid and the desired plasmids were isolated from transformed E. coli using the Qiagen Plasmid Maxi Kit. Thus, plasmids CF2LH-0 / pCOS1, CF2LH-3 / pCOS1, CF2LH-4 / pCOS1, CF2LH-5 / pCOS1, CF2LH-6 / pCOS1 and CF2LH-7 / pCOS1 were obtained.

As a typical example of these plasmids, the construction of the plasmid CF2LH-0 / pCOS1 is shown in FIG. 37, and the nucleotide sequence and amino acid sequence of MABL2-scFv <LH-0> contained in this plasmid are shown in SEQ ID NO 54. The nucleotide sequences and amino acid the sequence of linker sites in these plasmids is also shown in Fig.38.

6.3 Expression of scFv and sc (Fv) ₂ in COS7 cells

(1) Obtaining a culture supernatant using serum culture medium

scFv and sc (Fv) _{2 of} HL type and LH type were temporarily expressed in COS7 cells (JCRB9127, Japan Health Sciences Foundation). COS7 cells were subcultured in DMEM medium (Gibco BRL) containing 10% fetal calf serum (HyClone) at 37 ° C in a CO ₂ incubator. COS7 cells were transfected with CF2HL-0, 3 ~ 7 / pCOS1, or CF2LH-0, 3 ~ 7 / pCOS1, obtained in Example 6.2, or with pCHOM2 (Fv) ₂ vector by electroporation using a Gene Pulser (BioRad) device. DNA (10 μg) and 0.25 ml of a suspension of 2 × 10 ⁷ cells / ml in DMEM culture medium containing 10% fetal calf serum (FCS) and 5 mM BES (Sigma) were added to the cuvette. After holding for 10 minutes, these mixtures were treated with a pulse at 0.17 kV, with an electric capacitance of 950 μF. After "reconstitution" for 10 minutes at room temperature, the electroporated cells were transferred to DMEM culture medium (10% FCS) in a 75 ml flask. After culturing for 72 hours, the culture supernatant was collected and centrifuged to remove cell fragments. The culture supernatant was subjected to filtration using a 0.22 μm (Falcon) filter placed on the neck of the flask to obtain a culture supernatant (hereinafter referred to as "CM").

(2) Obtaining culture supernatant using serum-free culture medium

Cells transfected as in (1) were transferred to DMEM medium (10% FCS) in a 75 ml flask and cultured overnight. After cultivation, the supernatant was discarded and the cells were washed with PBS and then added to CHO-S-SFM II medium (Gibco BRL). After culturing for 72 hours, the culture supernatant was collected, centrifuged to remove cell fragments and filtered using a 0.22 μm (Falcon) filter-worn flask to obtain CM.

6.4 Detection of scFv and sc (Fv) ₂ in CM COS7 cells

Various MABL2-scFv and sc (Fv) ₂ in the CM of COS7 cells obtained in the above Example 6.3 (2) were detected by Western blotting.

Each COS7 CM (supernatant) was subjected to SDS-PAGE and transferred to a REINFORCED NC membrane (Schleicher & Schuell). The membrane was blocked with 5% skim milk (Morinaga Nyu-guo) and washed with Tris buffered saline (TBS). Then, an anti-FLAG antibody (Sigma) was added to it. The membrane was incubated at room temperature and washed. Peroxidase-labeled murine IgG antibody (Jackson ImmunoResearch) was added. After incubation and washing at room temperature, a substrate solution (Kirkegaard Perry Laboratories) was added to develop color (Fig. 39).

6.5 Flow cytometry

Flow cytometry was performed using the culture supernatants of COS7 cells obtained in Example 6.3 (1) to measure the binding of MABL2-scFv and sc (Fv) 2 to the human IAP antigen (integrin-associated protein). The culture supernatants to be tested, or the culture supernatant of COS7 cells, were added as control to 2 × 10 ⁵ cells of the murine L1210 mouse leukemia cell line expressing human integrin-associated protein (IAP). After incubation on ice and washing, 10 μg / ml of mouse anti-FLAG antibody (Sigma) was added thereto, and then these cells were incubated and washed.

Then, a FITC-labeled anti-mouse IgG antibody (Becton Dickinson) was added thereto, and the cells were again incubated and washed. Then, the fluorescence intensity was measured using a FACScan instrument (Becton Dickinson). Flow cytometry results show that MABL2-scFv having linkers of different lengths and sc (Fv) ₂ in the culture supernatants of COS7 cells have a high affinity for human IAPs (see FIGS. 40a and 40b).

6.6 Apoptosis inducing effect in vitro

The apoptosis-inducing effect of the COS7 culture supernatants obtained in Example 6.3 (1) was investigated by Annexin-V staining (Boehringer Mannheim) using L1210 cells transfected with the human IAP gene (hIAP / L 1210).

To 5 × 10 ⁴ hIAP / L1210 cells, culture supernatants of COS7 cells transfected with each of the vectors or culture supernatant of COS7 cells were added as a control at a final concentration of 10% and the mixture was cultured for 24 hours. Annexin-V / PI staining was then performed and the fluorescence intensity was measured using a FACScan instrument (Becton Dickinson). The results showed that scFv <HL3, 4, 6, 7, LH3, 4, 6, 7> and sc (Fv) ₂ in the CM of COS7 cells strongly induced hIAP / L1210 cell death. These results are shown in FIG.

6.7 Construction of vectors for expression of scFv and sc (Fv) ₂ in CHO cells

To isolate and purify MABL2-scFv and sc (Fv) ₂ from the culture supernatant, expression vectors for expression in CHO cells were constructed as described below.

The EcoRI-BamHI fragments pCF2HL-0, 3 ~ 7 and pCF2LH-0, 3 ~ 7, obtained in Example 6.2, were inserted between the EcoRI and BamHI sites into the pCHO1 expression vector for CHO cells using the Ligation High kit. Competent E. coli DH5α cells were transformed with each plasmid. Plasmids were isolated from transformed E. coli using the Qiagen Plasmid Midi kit (Qiagen) to obtain expression plasmids pCHOM2HL-0, 3 ~ 7 and pCHOM2LH-0, 3 ~ 7.

6.8 Obtaining CHO cells expressing MABL2-scFv <HL-0, 3 ~ 7>, MABL2-scFv <LH-0, 3 ~ 7> and sc (Fv) ₂ , and obtaining their culture supernatants

CHO cells were transformed with each of the pCHOM2HL-0, 3 ~ 7 and pCHOM2LH-0, 3 ~ 7 expression plasmids constructed in Example 6.7 and the pCHOM2 (Fv) ₂ vector to produce CHO cells stably expressing each modified antibody. As a typical example, the preparation of CHO cells stably expressing MABL2-scFv <HL-5> or sc (Fv) ₂ is illustrated below.

Expression plasmids pCHOM2HL-5 and pCHOM2 (Fv) _{2 were} linearized by PvuI restriction enzyme digestion and transfected into CHO cells by electroporation using a Gene Pulser (BioRad) instrument. DNA (10 μg) and 0.75 ml of PBS containing 1 × 10 ⁷ cells / ml were added to the cuvette and treated with a pulse at 1.5 kV, with an electric capacitance of 25 μF. After “reconstitution” for 10 minutes at room temperature, the electroporated cells were transferred to α-MEM culture medium containing nucleic acid (Gibco BRL) containing 10% fetal calf serum and cultured. After culturing overnight, the supernatant was discarded. Cells were washed with PBS and α-MEM culture medium (Gibco BRL) containing 10% fetal calf serum was added to the nucleic acid-free culture medium. After culturing for two weeks, the cells were cultured in a medium containing 10 nM (final concentration) methotrexate (Sigma), then 50 nM and 100 nM methotrexate. The resulting cells were cultured in serum-free medium CHO-S-SFM II (Gibco BRL) in roller bottles. The culture supernatant was collected, centrifuged to remove cell fragments and filtered using a filter with a pore size of 0.22 μm to obtain CM, respectively.

As described above, CHO cells were obtained that stably express MABL2-scFv <HL-0, -3, -4, -6, -7, and <LH-0, -3, -4, -5, -6, -7 >, and SM of them.

6.9 Purification of MABL2-scFv <HL-5> and sc (Fv) ₂ dimer

MABL2-scFv <HL-5> and sc (Fv) _{2 were} purified from CM obtained in Example 6.8 using the two purification methods described below.

<Cleaning Method 1>

HL-5 and sc (Fv) _{2 were} purified by affinity chromatography on an anti-FLAG antibody column using the FLAG sequence localized at the C-terminus of the polypeptides and gel filtration. One liter of CM obtained in 6.8 was applied to a column (7.9 ml) filled with anti-FLAG M2 affinity gel (Sigma), equilibrated with 50 mM Tris-HCl buffer (TBS, pH 7.5) containing 150 mM NaCl. After washing the TBS scFv column, it was eluted with 0.1 M glycine-HCl buffer, pH 3.5. The resulting fractions were analyzed by SDS-PAGE, and scFv elution was confirmed. The scFv fraction was mixed with Tween 20 to a final concentration of 0.01% and concentrated using Centricon-10 (Millipore). The concentrate was applied to a TSKgel G3000SWG column (7.5 × 600 mm) balanced with 20 mM acetate buffer (pH 6.0) containing 150 mM NaCl and 0.01% Tween 20. At a flow rate of 0.4 ml / min, scFv was detected absorption at 280 nm. HL-5 was eluted as the main fraction at the position corresponding to the dimer, and sc (Fv) _{2 was} eluted at the position corresponding to the monomer.

<Cleaning Method 2>

HL-5 and sc (Fv) _{2 were} purified using three steps, including ion exchange chromatography, hydroxyapatite chromatography, and gel filtration. For ion exchange chromatography, a Q sepharose fast flow column (Pharmacia) for HL-5 was used, and a SP-sepharose fast flow column was used for sc (Fv) ₂ . Starting from the second stage, HL-5 and sc (Fv) _{2 were} processed using the same procedure.

Stage One for HL-5

CM (supernatant) HL-5 was diluted twice with 20 mM Tris-HCl buffer (pH 9.0) containing 0.02% Tween 20, and then the pH was adjusted to 9.0 1 M Tris. The solution was applied to a Q sepharose fast flow column balanced with 20 mM Tris-HCl buffer (pH 8.5) containing 0.02% Tween 20. The polypeptide adsorbed on the column was eluted with a linear NaCl gradient in the same buffer, from 0 , 1 to 0.55 M. By analyzing the eluted fractions by SDS-PAGE, fractions containing HL-5 were collected and chromatographed with hydroxyapatite in a second step.

Stage One for sc (Fv) ₂

CM sc (Fv) _{2 was} diluted twice with 20 mM acetate buffer (pH 5.5) containing 0.02% Tween 20, and then the pH was adjusted to 5.5 with 1 M acetic acid. The solution was applied to an SP-Sepharose fast flow column, equilibrated with 20 mM acetate buffer (pH 5.5) containing 0.02% Tween 20. The polypeptide adsorbed on the column was eluted with a linear NaCl gradient in the same buffer, from 0 to 0 , 5 M. By analyzing the eluted fractions by SDS-PAGE electrophoresis, fractions containing sc (Fv) ₂ were collected and chromatographed with hydroxyapatite in a second step.

Second stage: Chromatography of HL-5 and sc (Fv) ₂ on hydroxyapatite

The HL-5 and sc (Fv) ₂ fractions obtained in the first stage were separately applied to a hydroxyapatite column (Type I, BioRad) equilibrated with 10 mM phosphate buffer containing 0.02% Tween 20, pH 7.0. After washing the column with the same buffer, the polypeptides adsorbed on the column were eluted with a linear gradient of phosphate buffer to 0.5 M. By analyzing the eluted fractions by SDS-PAGE, fractions containing the desired polypeptides were collected.

Stage Three: Gel Filtration of HL-5 and sc (Fv) ₂

Each fraction obtained in the second stage was separately concentrated using CentriPrep-10 (Millipore) and applied to a Superdex 200 column (2.6 × 60 cm, Pharmacia) equilibrated with 20 mM acetate buffer (pH 6.0) containing 0 , 02% Tween 20 and 0.15 M NaCl. HL-5 was eluted at the position corresponding to the dimer, and sc (Fv) HL-5 and sc (Fv) _{2 were} eluted as the main peak at the position corresponding to the monomer, respectively.

Since HL-5 monomer was hardly detected in either of the two purification methods, this showed that single chain Fv dimers were formed in high yields when the single chain Fv linker contains about 5 amino acids. In addition, the HL-5 dimer and sc (Fv) _{2 were} stably maintained for a month at 4 ° C after purification.

6.10 Evaluation of the binding activity of purified scFv <HL-5> and sc (Fv) ₂ dimers against antigen

Flow cytometry was performed using purified MABL2-scFv <HL-5> dimer and purified sc (Fv) ₂ to evaluate binding to the human IAP antigen (integrin-associated protein). 10 μg / ml of purified MABL2-scFv <HL-5> dimer, purified sc (Fv) ₂ , MABL-2 antibodies as a positive control or mouse IgG (Zymed) as a negative control were added to 2 × 10 ⁵ mouse cell line cells leukemia L1210 expressing human IAP (hIAP / L1210), or the L1210 cell line transformed with pCOS1 (pCOS1 / L1210) as a control. After incubation on ice and washing, 10 μg / ml of mouse anti-FLAG antibody (Sigma) was added and then the cells were incubated and washed. FITC-labeled anti-mouse IgG antibody (Becton Dickinson) was added and the cells were again incubated and washed. Then, the fluorescence intensity was measured using a FACScan instrument (Becton Dickinson).

Since the purified MABL2-scFv <HL-5> dimer and purified sc (Fv) ₂ specifically bind to hIAP / L1210 cells, it was confirmed that the scFv <HL-5> dimer and sc (Fv) ₂ have a high affinity for human IAP ( see Fig. 42).

6.11 Apoptosis inducing activity of purified scFv <HL-5> dimer and sc (Fv) ₂ in vitro

The apoptosis inducing effect of the purified MABL2-scFv <HL-5> dimer and purified sc (Fv) _{2 was} tested by staining with Annexin-V (Boehringer Mannheim) using L1210 cells (hIAP / L1210) into which the human IAP gene was introduced, and human leukemia cell line cells CCRF-CEM.

Various concentrations of purified MABL2-scFv <HL-5> dimer, purified MABL2-sc (Fv) ₂ , MABL-2 antibodies as a positive control or mouse IgG as a negative control were added to 5 × 10 ⁴ cells of the hIAP / L1210 cell line or 1 × 10 ⁵ cells of the cell line CCRF-CEM. After culturing for 24 hours, annexin-V staining was performed and the fluorescence intensity was measured using a FACScan instrument (Becton Dickinson). As a result, the dimer MABL2-scFv <HL-5> and MABL2-sc (Fv) ₂ strongly induced the death of hIAP / L1210 and CCRF-CEM cells in a concentration-dependent manner (see FIG. 43). As a result, it was shown that the dimer MABL2-scFv <HL-5> and MABL2-sc (Fv) ₂ induced apoptosis with better efficiency compared to the original antibody MABL-2.

6.12 Hemagglutination test using purified scFv <HL-5> and sc (Fv) ₂ dimers

A hemagglutination test was carried out using various concentrations of purified scFv dimer <HL-5> and purified sc (Fv) ₂ in accordance with Example 5.15.

Hemagglutination was observed with the MABL-2 antibody as a positive control, while hemagglutination was not observed with either the single-chain MABL2-sc (Fv) ₂ antibody or MABL2-scFv <HL-5>. In addition, there was no significant difference in hemagglutination between the two buffers used with the MABL-2 antibody. These results are shown in Table 3.

Table 3
Hemagglutination Test Thinner: PBS (μg / ml) the control 28.9 14.45 7.225 3.6125 1.8063 0.9031 0.4516 0.2258 0.1129 0.0564 0,0282 0.0141 0.0071 0.0035 0.0018 MABL2-sc (Fv) ₂ - - - - - - - - - - - - - - - - the control 28.0 14.0 7.0 3.5 1.75 0.875 0.4375 0.2188 0.1094 0.0547 0.0273 0.0137 0.0068 0.0034 0.0017 MABL2-sc (Fv) <HL5> - - - the control 80 40 twenty 10 5 2,5 1.25 0.625 0.3125 0.1563 0.0781 0.0391 0.0195 0.0098 0.0049 MABL2 - + + + + + + + + + ± - - - - - (intact) mcg / ml Thinner: Acetate Buffer (μg / ml) the control 80 40 twenty 10 5 2.5 1.25 0.625 0.3125 0.1563 0.0781 0.0391 0.0195 0.0098 0.0049 MABL2 (intact) - + + + + + + + + + + + - - - -

6.13 Antitumor effect of purified scFv <HL-5> and sc (Fv) ₂ dimers in a mouse model of human myeloma

The antitumor effects were tested for the dimer scFv <HL-5> and sc (Fv) ₂ , obtained and purified in examples 6.8 and 6.9. The test was performed using the mouse model of human myeloma obtained in Example 5.1, by determining the amount of M protein produced by human myeloma cells in mouse serum using ELISA and studying the survival time of mice. Then, the antitumor effects of scFv <HL-5> and sc (Fv) ₂ dimer were evaluated by changing the amount of M protein in mouse serum and the survival time of mice.

In this test, HL-5 and sc (Fv) _{2 were} used as a solution with a concentration of 0.01, 0.1 or 1 mg / ml in a solvent containing 150 mM NaCl, 0.02% Tween and 20 mM acetate buffer, pH 6.0, and was administered to mice at a dose of 0.1, 1 or 10 mg / kg. The control group of mice was administered only solvent.

Mouse serum was collected 26 days after transplantation of human myeloma cells and the amount of M protein in serum was measured by ELISA in accordance with Example 5.14. As a result, the amount of M protein in the serum of both groups of mice injected with HL-5, dimer, and sc (Fv) ₂ decreased in a dose-dependent manner (see FIG. 44). In addition, a significant lengthening of the survival time was observed in both groups that were injected with HL-5 (Fig. 45) and sc (Fv) ₂ (Fig. 46), in comparison with the control group, which was administered with a solvent. These results show that the HL-5 and sc (Fv) _{2 of} this invention exhibit excellent in vivo antitumor activity.

Example 7

Single chain Fv containing the V region of the H chain and the V region of the L chain of human anti-human MPL antibody 12B5

DNA encoding the V region of the human monoclonal antibody 12B5 against human MPL was constructed as follows:

7.1 Construction of a gene encoding the V region of the H-chain 12V

A gene encoding the V region of the H chain of a human 12B5 antibody binding to human MPL was constructed by attaching the nucleotide sequence of this gene (SEQ ID NO 55) at the 5'-end to the leader sequence (SEQ ID NO 56) derived from the human antibody gene ( Eur. J. Immunol. 1996; 26: 63-69). The constructed nucleotide sequence was divided into four oligonucleotides having overlapping 15 bp sequences. each (12B5VH-1, 12B5VH-2, 12B5VH-3, 12B5VH-4). 12B5VH-1 (SEQ ID NO 57) and 12B5VH-3 (SEQ ID NO 59) were synthesized in the sense orientation, and 12B5VH-2 (SEQ ID NO 58) and 12B5VH-4 (SEQ ID NO 60) in the antisense orientation, respectively . After assembly of the synthesized oligonucleotides based on mutual complementarity, external primers 12B5VH-S and 12B5VH-A) were added to amplify the full-sized gene. 12B5VH-S (SEQ ID NO 61) was designed so that it, as a direct primer, hybridized to the 5'-end of the leader sequence and contained the HindIII restriction enzyme recognition site and Kozak sequence, and 12B5VH-A (SEQ ID NO 62) was designed so that it how the reverse primer hybridized to the nucleotide sequence encoding the C-terminus of the V region of the H chain and contained a donor splicing site and a BamHI restriction enzyme recognition site, respectively.

100 μl of a PCR solution containing 10 μl of a 10-fold Gold II PCR buffer, 1.5 mM MgCl ₂ , 0.08 mm dNTP (dATP, dGTP, dCTP and dTTP), 5 units of AmpliTaq Gold DNA polymerase (all from Perkin Elmer) and 2.5 pmol of each of the synthesized oligonucleotides 12B5VH-1-4) were heated at an initial temperature of 94 ° C for 9 minutes, at 94 ° C for 2 minutes, at 55 ° C for 2 minutes, and at 72 ° C for 2 minutes. After repeating this cycle, 100 pmol of external primers 12B5VH-S and 12B5VH-A were added twice. The mixture was subjected to 35 cycles of heating at 94 ° C for 30 seconds, at 55 ° C for 30 seconds and 72 ° C for 1 minute, and heated at 72 ° C for an additional 5 minutes.

The PCR product was purified using a 1.5% gel of low-melting agarose (Sigma) gel, digested with restriction enzymes BamHI and HindIII and cloned into the expression vector HEF-gγ1 for the human H chain. After determining the DNA sequence, a plasmid containing the correct DNA sequence was designated as HEF-12B5H-gγ1.

HEF-12B5H-gγ1 was digested with restriction enzymes EcoRI and BamHI to give a gene encoding 12B5VH, which was then cloned into the pCOS-Fd expression vector for the human Fab N chain to give pFd-12B5H. An expression vector for the human H chain Fab was constructed by amplification of DNA (SEQ ID NO 63) containing an intron portion between genes encoding the H chain V region and the constant region of a human antibody, and a gene encoding a portion of the H chain constant region human, using PCR and embedding this PCR product in the expression vector for animal cells pCOS1. The constant region of the human H chain was amplified for this gene under the conditions described above, using HEF-gγ1 as the template, as the direct primer G1CH1-S (SEQ ID NO 64), which was designed to hybridize with 5'- the end sequence of intron 1 and contained the restriction enzyme recognition site EcoRI and BamHI, and as a reverse primer G1CH1-A (SEQ ID NO 65), which was designed so that it hybridizes with the 3'-end of the DNA of the CHl domain of the constant region of the human H chain and contained the sequence encoding the hinge part region, two stop codons and BglII restriction enzyme recognition site.

The nucleotide sequence and amino acid sequence of the reconstituted variable region of the 12B5 H chain that was inserted into the plasmids HEF-12B5H-gγ1 and pFd-12B5H are shown in SEQ ID NO 66.

7.2 Construction of the gene encoding the V region of the L-chain 12V

A gene encoding the V region of the L chain of the human 12B5 antibody binding to human MPL was constructed by attaching the nucleotide sequence of this gene (SEQ ID NO 67) at the 5'-end to the leader sequence (SEQ ID NO 68) derived from the human antibody gene 3D6 (Nuc. Acid Res. 1990; 18: 4927). In the same manner as described above, this engineered nucleotide sequence was divided into four oligonucleotides having 15 bp overlapping sequences. each (12B5VL-1, 12B5VL-2, 12B5VL-3, 12B5VL-4), and were synthesized, respectively. 12B5VL-1 (SEQ ID NO 69) and 12B5VL-3 (SEQ ID NO 71) had a sense orientation, and 12B5VL-2 (SEQ ID NO 70) and 12B5VL-4 (SEQ ID NO 72) had an antisense orientation, respectively. The synthesized oligonucleotides were docked on the basis of mutual complementarity and mixed with external primers 12B5VL-S and 12B5VL-A) for amplification of the full-sized gene. 12B5VL-S (SEQ ID NO 73) was designed to hybridize as a direct primer to the 5'-end of the leader sequence and contain the HindIII restriction enzyme recognition site and the Kozak sequence. 12B5VL-A (SEQ ID NO 74) was designed to hybridize with the reverse primer to the nucleotide sequence encoding the C-terminus of the V region of the L chain and contain a splicing donor site and a BamHI restriction enzyme recognition site.

After PCR, as described above, the PCR product was purified using a 1.5% gel of low melting agarose (Sigma), digested with restriction enzymes BamHI and HindIII and cloned into the human L-chain expression vector HEF-gκ. After determining the DNA sequence, a plasmid containing the correct DNA sequence was designated as HEF-12B5L-gκ. The nucleotide sequence and amino acid sequence of the reconstructed V region of the 12B5 L chain, which was inserted into the HEF-12B5L-gκ plasmid, are shown in SEQ ID NO 75.

7.3 Preparation of Reconstructed Single Chain Fv 12B5 (scFv)

The reconstructed 12B5 single chain Fv antibodies were designed to have a 12B5VH-linker-12B5VL structure and contain a FLAG sequence (SEQ ID NO 76) at the C-terminus for ease of detection and purification. The reconstructed single chain Fv 12B5 (sc12B5) was constructed using the linker sequence consisting of 15 amino acids represented by (Gly ₄ Ser) ₃ .

(1) Obtaining a reconstructed single chain Fv 12B5 using a linker sequence of 15 amino acids

The gene encoding the reconstructed single chain Fv of antibody 12B5, which contained a linker sequence of 15 amino acids, was constructed by connecting the V region of the 12B5 H chain, the linker region and the V region of the 12B5 L chain, amplifying this product by PCR, respectively. This method is depicted schematically in FIG. Six primers (A-F) were used to obtain this reconstructed single chain Fv 12B5. Primers A, C, and E had sense sequences, and primers B, D, and F had antisense sequences.

Direct primer 12B5-S (Primer A, SEQ ID NO 77) for the V region of the H chain was designed to hybridize to the 5'-end of the H chain leader sequence and contain the EcoRI restriction enzyme recognition site. The HuVHJ3 reverse primer (Primer B, SEQ ID NO 78) for the V region of the H chain was designed to hybridize with DNA encoding the C-terminus of the V region of the H chain.

The RHuJH3 direct primer (Primer C, SEQ ID NO 79) for the linker was designed to hybridize with DNA encoding the N-terminus of the linker and overlap with DNA encoding the C-terminus of the V region of the H chain. The reverse primer RHuVK1 (Primer D, SEQ ID NO 80) for the linker was designed to hybridize with DNA encoding the C-terminus of the linker and overlap with DNA encoding the N-terminus of the V region of the L chain.

The HuVK-1.2 forward primer (Primer E, SEQ ID NO 81) for the V region of the L chain was designed to hybridize with DNA encoding the N terminus of the V region of the L chain. The reverse primer 12B5F-A for the V region of the L chain (Primer F, SEQ ID NO 82) was designed to hybridize with DNA encoding the C-terminus of the V region of the L chain and contain a sequence encoding a FLAG peptide (Hopp NP et al., Bio / Technology, 6, 1204-1210, 1988), two transcriptional stop codons and a NotI restriction enzyme recognition site.

In the first PCR step, three AB, C-D, and E-F reactions were performed, and the three PCR products obtained in the first PCR step were docked based on mutual complementarity. After the addition of primers A and F, the full-length DNA encoding the reconstructed single-stranded Fv 12B5 containing a linker consisting of 15 amino acids was amplified (second PCR). In the first-stage PCR, the plasmid HEF-12B5H-gγ1 (see Example 7.1) encoding the reconstructed V region of the 12B5 H chain, pSCFVT-hM21 (ONS-M21 humanized antibody) (Ohtomo et al., Anticancer Res. 18 (1998), 4311-4316) containing DNA (SEQ ID NO 83) encoding a linker region consisting of Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser Gly Gly Gly Gly Ser (Huston et al., Proc. Natl. Acad. Sci. USA, 85, 5879-5883, 1988), and the plasmid HEF-12B5L-gκ (see Example 7.2) encoding the reconstructed V region of the 12B5 L chain, respectively.

50 μg of the PCR solution for the first PCR stage contained 5 μl of 10-fold Gold II PCR buffer, 1.5 mM MgCl ₂ , 0.08 mM dNTP, 5 units of AmpliTaq Gold DNA polymerase (all from Perkin Elmer), No. 100 pmol of each primer and 100 ng of each DNA template. The PCR solution was heated at an initial temperature of 94 ° C for 9 minutes, then at 94 ° C for 30 seconds, at 55 ° C for 30 seconds, and at 72 ° C for 1 minute. After repeating this cycle 35 times, the reaction mixture was further heated at 72 ° C. for 7 minutes.

PCR products AB, CD and EF were docked using the second PCR. A solution of the PCR mixture for the second stage in a volume of 98 μl, containing as a matrix 1 μl of PCR product A-B, 0.5 μl of PCR product CD and 1 μl of PCR product CD and 1 μl of PCR product EF, 10 μl of a 10-fold PCR buffer Gold II, 1.5 mm MgCl ₂ , 0.08 mm dNTP, 5 units of AmpliTaq Gold DNA polymerase (all from Perkin Elmer), was heated at an initial temperature of 94 ° C for 9 minutes, then at 94 ° C for 2 minutes, at 65 ° C for 2 minutes and at 72 ° C for 2 minutes. After repeating this cycle, 100 pmol of each of primers A and F were added twice. After repeating the cycle consisting of from naga yelling at 94 ° C for 30 seconds, at 55 ° C for 30 seconds and at 72 ° C for 1 minute, the reaction mixture was heated 35 times at 72 ° C for 5 minutes.

The DNA fragments obtained by the second PCR were purified using a 1.5% low-melting agarose gel, digested with EcoRI and NotI, and cloned into the pCHO1 vector and pCOS1 vector (Japan Patent Application No. 8-255196). The expression vector pCHO1 was a vector obtained by deletion of the antibody gene from DHFR-ΔE-rv-PM1-f (see WO 92/19759) by digestion of EcoRI and SmaI and attachment of an EcoRI-NotI-BamHI adapter (Takara Shuzo). After determining the DNA sequence, plasmids containing the DNA fragment encoding the correct amino acid sequence of the reconstructed single chain Fv 12B5 were designated pCHO-sc12B5 and pCOS-sc12B5. The nucleotide sequence and amino acid sequence of the reconstructed single chain Fv antibody 12B5 included in the plasmids pCHO-sc12B5 and pCOS-sc12B5 are shown in SEQ ID NO 84.

7.4 Expression of 12B5 Antibody (IgG, Fab) and Single Chain Fv Polypeptide by Animal Cell

Antibody 12B5 (IgG, Fab) and single chain Fv derived from antibody 12B5 were expressed using COS7 cells or CHO cells.

Temporary expression using COS7 cells was carried out as follows. Transfection was performed by electroporation using a Gene Pulser device (BioRad). For the expression of 12B5 antibody (IgG), 10 μg of the aforementioned HEF-12B5H-gγ1 and HEF-12B5L-gκ expression vectors were added; for the expression of the 12B5 Fab fragment, 10 μg of pFd-12B5H and HEF-12B5L-gκ were added for the expression of single-chain Fv 10 μg pCOS-sc12B5 was added to COS7 cells (1 × 10 ⁷ cells / ml) suspended in 0.8 ml of PBS. The mixture in the cuvette was treated with a 1.5 kV pulse, with an electric capacitance of 25 μF. After "reconstitution" for 10 minutes at room temperature, electroporated cells were added to DMEM culture medium (Gibco BRL) containing 10% fetal calf serum and cultured. After culturing overnight, the cells were washed once with PBS, added to serum-free medium CHO-S-SFM II, and cultured for 2 days. The culture medium was centrifuged to remove cell debris (cell debris) and filtered through a 0.22 μm filter to obtain a culture supernatant.

To obtain a CHO cell line stably expressing a single chain Fv (polypeptide) derived from 12B5 antibody, the pCHO-sc12B5 expression vector was introduced into CHO cells as follows.

The expression vector was introduced into CHO cells by electroporation using a Gene Pulser instrument (BioRad). Linearized DNA (100 μg) obtained by digestion with Pvul restrictase and CHO cells (1 × 10 ⁷ cells / ml) suspended in 0.8 ml of PBS were mixed in a cuvette, left without stirring on ice for 10 minutes and treated with pulse 1, 5 kV, with an electrical capacitance of 25 μF. After "reconstitution" for 10 minutes at room temperature, electroporated cells were added to CHO-S-SFM II medium (Gibco BRL) containing 10% fetal calf serum and cultured. After culturing for 2 days, cultivation was continued in CHO-S-SFM II (Gibco BRL) containing 5 nM methotrexate (Sigma) and 10% fetal calf serum. From the clones thus obtained, a clone with a high expression rate was selected as a cell line producer of single-stranded Fv 12B5. After culturing in serum-free CHO-S-SFM II medium (Gibco BRL) containing 5 nM methotrexate (Sigma), a culture supernatant was obtained by centrifugation to separate cell debris.

7.5 Purification of single chain Fv originating from 12B5 produced by CHO cells

The culture supernatant of the CHO cell line expressing single chain Fv 12B5 obtained in 7.4 was purified using an anti-FLAG antibody column and a gel filtration column.

(1) Column with anti-FLAG antibody

The culture supernatant was added to anti-FLAG M2 affinity gel (Sigma) equilibrated with PBS. After washing this column with the same buffer, the proteins adsorbed on the column were eluted with 0.1 M glycine-HCl buffer (pH 3.5). The eluted fractions were immediately neutralized by the addition of 1 M Tris-HCl buffer (pH 8.0). Eluted fractions were analyzed by SDS-PAGE and the fraction that was confirmed to contain single chain Fv was concentrated using Centricon-10 (Millipore).

(2) Gel Filtration

The concentrated solution obtained in (1) was applied onto a Superdex 200 column (10 × 300 mm, Amersham Pharmacia) balanced with PBS containing 0.01% Tween 20. The sc12B5 product was eluted as two peaks (A, B) (see Fig. 48). Fractions A and B were analyzed by electrophoresis on a 14% SDS-polyacrylamide gel. Electrophoresis of the samples was carried out in the presence and in the absence of a reducing agent according to the Laemmli method and Coomassie was stained with diamond blue after electrophoresis. As shown in FIG. 49, fractions A and B, regardless of the presence of a reducing agent, gave a single band having an apparent molecular weight of about 31 kDa. When fraction A and B were analyzed by gel filtration using Superdex 200 PC 3.2 / 30 (3.2 × 300 mm, Amersham Pharmacia), fraction A gave an eluted product with an apparent molecular weight of about 44 kDa, and fraction B gave a product with an apparent molecular weight mass of 22 kDa (see Fig. 50a and b). These results indicate that fraction A is a non-covalently bound single-chain Fv sc12B5 dimer and B is a monomer.

7.6 Measurement of TPO-like agonistic activity of various single chain Fv

The TPO-like activity of the anti-MPL single chain antibody was evaluated by determining the proliferative effect on Ba / F3 cells (BaF / MpI) expressing human TPO receptor (MPL). After washing the BaF / MpI cells twice with RPMI1640 culture medium (Gibco) containing 10% fetal calf serum (Gibco), the cells were suspended in the culture medium at a cell density of 5 × 10 ⁵ cells / ml. Anti-MPL single chain antibody and human TPO (R & D System) were diluted with this culture medium, respectively. 50 μl of cell suspension and 50 μl of diluted human antibody or TPO were added to a 96-well plate (flat-bottomed) (Falcon) and cultured in a CO ₂ incubator (CO ₂ concentration: 5%) for 24 hours. After incubation, 10 μl of WST-8 reagent (SF: Nacalai Tesque live cell reagent) was added and absorption was immediately measured at a measurement wavelength of 450 nm and at a reference wavelength of 620 nm using a SPECTRA Fluor (Tecan) fluorescence-absorption photometer. After incubation in a CO ₂ incubator (CO ₂ concentration: 5%) for 2 hours, the optical density at a measurement wavelength of 450 nm and a reference wavelength of 620 nm was again measured with SPECTRA Fluor. Since the reagent WST-8 developed color at a wavelength of 450 nm depending on the number of living cells, the proliferative activity of BaF / Mp1 based on the change in optical density after 2 hours was evaluated by ED50 calculated as follows. On the proliferation reaction curve, where the optical density was plotted along the ordinate against the concentration of the antibody along the abscissa, the optical density on the plateau was taken as the 100% reaction rate. Having obtained the approximation formula by the linear approximation method on the basis of pending values close to 50% of the reaction rate, the antibody concentration corresponding to 50% of the reaction rate was calculated and taken as ED50.

The results of agonistic activity against MPL, measured using cultural supernatants of COS7 cells expressing different variants of the 12B5 antibody molecule, found, as illustrated in FIG. 51, that the 12B5IgG antibody having a bivalent antigen binding site increased the optical density in a concentration-dependent manner and had TPO -like agonistic activity (ED50; 29 nM), while the agonistic activity of 12B5Fab antibody having a monovalent antigen-binding site was very weak (ED50; 34724 nM). In contrast, a single chain Fv (sc12D5) having a monovalent antigen binding site, like Fab, showed strong agonistic activity at an ED50 of 75 nM. However, it was known that the variable regions of the H chain and L chain of a single chain Fv are linked via a non-covalent bond, and therefore, each variable region is dissociated in solution and can be associated with the variable region of another molecule to form multimers such as dimers. When measuring the molecular weight of sc12B5 purified by gel filtration, the existence of molecules that were recognized as monomers and dimers was confirmed (see FIG. 48). Then, sc12B5 monomer and sc12B5 dimer (see FIG. 50) were isolated and their agonistic activity against MPL was measured. As shown in FIGS. 51 and 52, the ED50 of sc12B5 monomer was 4438.7 nM, which confirmed that the agonistic activity was less compared to the result obtained using the COS7 cell culture supernatant. In contrast, a single chain Fv (sc12B5 dimer) having a bivalent antigen binding site showed approximately 400 times stronger agonistic activity (ED50; 10.1 nM) compared to monovalent sc12B5. In addition, bivalent single-chain Fv showed agonistic activity equivalent to or higher than the agonistic activity of TPO and human 12B5IgG.

Example 8

Construction of the gene encoding the variable region of the human antibody 12E10 against human MPL

DNA encoding the variable region of the human monoclonal antibody 12E10 against human MPL was constructed as follows.

8.1 Construction of the gene encoding the V region of the H-chain 12E10

The nucleotide sequence of SEQ ID NO 86 was constructed as a gene encoding the V region of the H chain of the human 12E10 antibody binding to human MPL based on the amino acid sequence described in WO 99/10494 (SEQ ID NO 85). A full-length nucleotide sequence was constructed by attaching to its 5'end of the leader sequence (SEQ ID NO 87) derived from the human antibody gene (GenBank accession number AF062252). The constructed nucleotide sequence was divided into four oligonucleotides having overlapping 15 bp sequences. each (12E10VH-1, 12E10VH-2, 12E10VH-3, 12E10VH-4). 12E10VH-1 (SEQ ID NO 88) and 12E10VH-3 (SEQ ID NO 90) were synthesized in the sense orientation, and 12E10VH-2 (SEQ ID NO 89) and 12E10VH-4 (SEQ ID NO 91) in the antisense orientation, respectively. After docking the synthesized oligonucleotides based on mutual complementarity, external primers 12E10VH-S and 12E10VH-A) were added to amplify the full-sized gene. 12E10VH-S (SEQ ID NO 92) was designed to hybridize with the 5'-end of the leader sequence as a direct primer and contain the HindIII restriction enzyme recognition site and Kozak sequence, and 12E10VH-A (SEQ ID NO 93) was designed to how the reverse primer hybridized with the nucleotide sequence encoding the C-terminus of the V region of the H chain and contained a donor splicing site and a BamHI restriction enzyme recognition site, respectively.

100 μl of a PCR solution containing 10 μl of a 10-fold Gold II PCR buffer, 1.5 mM MgCl ₂ 0.08 mm dNTP (dATP, dGTP, dCTP and dTTP), 5 units of AmpliTaq Gold DNA polymerase (all from Perkin Elmer) and 2.5 pmol of each of the synthesized oligonucleotides 12E10VH-1-4) were heated at an initial temperature of 94 ° C for 9 minutes, then at 94 ° C for 2 minutes, at 55 ° C for 2 minutes, and at 72 ° C for 2 minutes. After repeating this cycle, 100 pmol of external primers 12E10VH-S and 12E10VH-A were added twice. The mixture was subjected to a cycle consisting of heating at 94 ° C for 30 seconds, at 55 ° C for 30 seconds and 72 ° C for 1 minute, 35 times and heated at 72 ° C for an additional 5 minutes.

The PCR product was purified using a 1.5% gel of low-melting agarose (Sigma) gel, digested with restriction enzymes BamHI and HindIII and cloned into the expression vector HEF-gγ1 for the human H chain. After determining the DNA sequence, a plasmid containing the correct DNA sequence was named HEF-12E10H-gγ1.

HEF-12E10H-gγ1 was digested with restriction enzymes EcoRI and BamHI to give a gene encoding 12E10VH, which was then cloned into the pCOS-Fd expression vector for the human Fab H chain to produce pFd-12E10H. An expression vector for a human H chain Fab was constructed by amplification of DNA (SEQ ID NO 63) containing a portion of an intron between genes encoding the V region of the H chain and the constant region of a human antibody, and a gene encoding a portion of the constant region of the H chain human, using PCR and embedding this PCR product in the expression vector for animal cells pCOS1. The constant region of the human H chain was amplified for this gene under the conditions described above, using HEF-gγ1 as the template, as the direct primer G1CH1-S (SEQ ID NO 64), which was designed to hybridize with 5'- the end sequence of intron 1 and contained restriction enzyme recognition sites EcoRI and BamHI, and as a reverse primer G1CH1-A (SEQ ID NO 65), which was designed to hybridize with the 3'-end of the DNA of the CH1 domain of the constant region of the human H chain and contained a sequence encoding a part of the hinge region sti, two stop codons and the recognition site of the restriction enzyme BgIII.

The nucleotide sequence and amino acid sequence of the reconstructed variable region of the 12E10H H chain, which was included in the plasmids HEF-12E10H-gγ1 and pFd-12E10H, are shown in SEQ ID NO 94.

8.2 Construction of the gene encoding the V region of the L-chain 12E10

The nucleotide sequence of SEQ ID NO 96 was constructed as a gene encoding the V region of the L chain of the human 12E10 antibody binding to human MPL based on the amino acid sequence described in WO 99/10494 (SEQ ID NO 95). Additionally, the design provided for the attachment to its 5'-end of the leader sequence (SEQ ID NO 97), derived from the human antibody gene (Mol. Immunol. 1992; 29: 1515-1518). In the same manner as described above, this engineered nucleotide sequence was divided into four oligonucleotides having 15 bp overlapping sequences. each (12E10VL1, 12E10VL2, 12E10VL3, 12E10VL4), and were synthesized, respectively. 12E10VL1 (SEQ ID NO 98) and 12E10VL3 (SEQ ID NO 100) had a sense orientation, and 12E10VL2 (SEQ ID NO 99) and 12E10VL4 (SEQ ID NO 101) had an antisense orientation, respectively. Each of the synthesized oligonucleotides was docked on the basis of mutual complementarity and mixed with external primers (12E10VLS and 12E10VLA) to amplify the full-sized gene. 12E10VLS (SEQ ID NO 102) was designed to hybridize as the direct primer to the 5'-end of the leader sequence and contain the EcoRI restriction enzyme recognition site and the Kozak sequence. 12E10VLA (SEQ ID NO 103) was designed to hybridize as a reverse primer to a nucleotide sequence encoding the C-terminus of the V region of the L chain and contain a donor splicing site and a BlnI restriction enzyme recognition site.

After PCR, as described above, the PCR product was purified using 1.5% gel of low melting agarose (Sigma), digested with restriction enzymes EcoRI and BlnI, and cloned into pUC19 vector containing the gene for the constant region of the human lambda chain. After determining the DNA sequence, a plasmid containing the correct DNA sequence was digested with EcoRI to generate a gene encoding the 12E10 L chain V region and the human lambda chain constant region, and then inserted into the pCOS1 expression vector. A plasmid containing the 12E10 L chain gene (SEQ ID NO 104) was designated as pCOS-12E10L.

8.3 Obtaining a reconstructed single chain Fv 12E10

The reconstructed 12E10 single chain Fv antibodies were designed so that their structure was represented by a 12E10VH linker-12E10VL and contained the FLAG sequence (SEQ ID NO 105) at the C-terminus for easy detection and purification. Reconstructed single chain Fv 12E10 (sc 12E10 and db 12E10) were constructed using linker sequences consisting of 15 amino acids represented by (Gly ₄ Ser) ₃ , or 5 amino acids represented by (Gly ₄ Ser) ₁ .

(1) Obtaining a reconstructed single chain Fv 12E10 using a linker sequence consisting of 5 amino acids

The gene encoding the reconstructed single-chain Fv 12E10, which contained a 5-amino acid linker sequence, was constructed by introducing the nucleotide sequence for this linker (Gly ₄ Ser) ₁ from the 3'-end of the gene encoding the 12E10 H chain V region and with 5 the 'end of the gene encoding the V region of the 12E10 L chain, amplification of the thus obtained gene by PCR and docking of the amplified genes. Four primers (AD) were used to obtain the reconstructed single chain Fv 12E10. Primers A and C had sense sequences, and primers B and D had antisense sequences.

The direct primer for the V region of the H chain was 12E10S (Primer A, SEQ ID NO 106). The DB2 reverse primer (Primer B, SEQ ID NO 107) for the V region of the H chain was designed to hybridize with DNA encoding the C-terminus of the V region of the H chain and contain the nucleotide sequence encoding the linker (Gly ₄ Ser ) ₁ , and the nucleotide sequence encoding the N-terminus of the V region of the L chain.

The direct primer DB1 (Primer C, SEQ ID NO 108) for the V region of the L chain was designed to hybridize with DNA encoding the N-terminus of the V region of the L chain and contain the nucleotide sequence encoding the linker (Gly ₄ Ser ) ₁ , and the nucleotide sequence encoding the C-terminus of the V region of the H chain. The 12E10FA reverse primer for the L chain V region (Primer D, SEQ ID NO 109) was designed to hybridize with DNA encoding the C-terminus of the L chain V region and contain a FLAG coding sequence and restriction enzyme recognition site NotI.

In the first stage of PCR, two reactions A-B and C-D were carried out, and two PCR products obtained in the first stage of PCR were docked on the basis of mutual complementarity. After the addition of primers A and D, a full-length DNA encoding the reconstructed single-stranded Fv 12E10 containing a linker consisting of 5 amino acids was amplified (second PCR). In the first-stage PCR, plasmids HEF-12E10H-gγ1 (see Example 8.1) encoding the reconstructed V region of the H chain 12E10 and pCOS-12E10L (see Example 8.2) encoding the reconstructed V region of the L chain were used as a matrix 12E10, respectively.

50 μl of the PCR solution for the first stage contained 5 μl of Gold II 10-fold PCR buffer, 1.5 mM MgCl ₂ , 0.08 mM dNTP, 5 AmpliTaq Gold DNA polymerase units (all from Perkin Elmer), no. 100 pmol each primer and 100 ng of each DNA template. The PCR solution was heated at an initial temperature of 94 ° C for 9 minutes, then at 94 ° C for 30 seconds, at 55 ° C for 30 seconds, and at 72 ° C for 1 minute. After repeating this cycle 35 times, the reaction mixture was heated at 72 ° C for an additional 5 minutes.

PCR products AB (429 bp) and CD (395 bp) were docked using the second PCR. The reaction in the PCR solution for the second stage (98 μl) containing 1 μl of the PCR product of the first stage A-B and CD as a matrix, 100 pmol of each of the primers, 10 μl of 10-fold Gold II PCR buffer, 1.5 mM MgCl ₂ , 0.08 mM dNTP and 5 units of AmpliTaq Gold DNA polymerase (all from Perkin Elmer) were carried out under the same conditions as described above.

The 795 bp DNA fragment obtained in the second PCR was purified using a 1.5% agarose gel, digested with EcoRI and NotI and cloned into the pCHO1 vector or pCOS1 vector. The pCHO1 expression vector was a vector constructed by deletion of an antibody gene from DHFR-ΔE-RVH-PM1-f (see WO 92/19759) by digestion of EcoRI and SmaI and attachment to an EcoRI-NotI-BamHI adapter (Takara Shuzo) . After determining the DNA sequence of the plasmid containing the DNA fragment encoding the correct amino acid sequence of the reconstructed single chain Fv 12E10, were designated as pCHO-db 12E10 and pCOS-db 12E10. The nucleotide sequence and amino acid sequence of the reconstructed single chain Fv antibody 12E10 included in the plasmids pCHO-db12E10 and pCOS-db12E10 are shown in SEQ ID NO 110.

(2) Obtaining a reconstructed single chain Fv 12E10 using a linker sequence of 15 amino acids

The gene encoding the reconstructed single-chain Fv of antibody 12E10, which contained a linker sequence of 15 amino acids, was constructed by introducing the nucleotide sequence for the linker (Gly ₄ Ser) ₃ from the 3'-end of the gene encoding the V region of the 12E10 H chain, and c 5'-end of the gene encoding the V region of the L-chain 12E10, amplification of the thus obtained gene by PCR and docking amplified genes. Four primers (AD) were used to obtain this reconstructed single chain Fv 12E10. Primers A and C had sense sequences, and primers B and D had antisense sequences.

The direct primer for the V region of the H chain was 12E10S (Primer A, SEQ ID NO 106). The sc4.3 reverse primer (Primer B, SEQ ID NO 111) for the V region of the H chain was designed to hybridize with DNA encoding the C-terminus of the V region of the H chain and contain the nucleotide sequence encoding the linker (Gly ₄ SeR) ₃ , and a nucleotide sequence encoding the N-terminus of the V region of the L chain.

The forward scl.3 primer (Primer C, SEQ ID NO 112) for the L chain V region was designed to hybridize with DNA encoding the N-terminus of the L chain V region and contain a nucleotide sequence encoding a linker (Gly ₄ Ser) ₃ , and the nucleotide sequence encoding the C-terminus of the V region of the H chain. The 12E10FA reverse primer for the L chain V region (Primer D, SEQ ID NO 109) was designed to hybridize with DNA encoding the C-terminus of the L chain V region and contain a FLAG coding sequence and restriction enzyme recognition site NotI.

In the first stage of PCR, two reactions A-B and C-D were carried out, and two PCR products obtained in the first stage of PCR were docked on the basis of mutual complementarity. After the addition of primers A and D, a full-length DNA encoding the reconstructed single-chain Fv 12E10 containing a linker consisting of 15 amino acids was amplified (second PCR). In the PCR of the first stage, the plasmid pCOS-db12E10 (see Example 8. 3 (1)) encoding the reconstructed single chain Fv 12E10 was used as a template.

50 μg of the PCR solution for the first PCR stage contained 5 μl of 10-fold ExTaq-buffer II, 0.4 mm dNTP, 2.5 units of TaKaRa ExTaq DNA polymerase (TAKARA), 100 pmol of each primer and 10 ng of each DNA matrices. The PCR solution was heated at an initial temperature of 94 ° C for 30 seconds, then at 94 ° C for 15 seconds and 72 ° C for 2 minutes, and this cycle was repeated 5 times. After repeating 28 times the heating cycle at 94 ° C for 15 seconds and at 70 ° C for 2 minutes, the reaction mixture was further heated at 72 ° C for 5 minutes.

PCR products AB (477 bp) and C-D (447 bp) were docked using the second PCR. In the PCR mixture for the second stage (98 μl) containing 1 μl of the PCR products of the first stage AB and CD as a template, 100 pmol of each of primers A and D, 5 μl of a 10-fold ExTaq buffer, 0.4 mM dNTP, 2.5 units of TaKaRa ExTaq DNA polymerase (TAKARA), the reaction was carried out under the same conditions as described above.

The 825 bp DNA fragment obtained in the second PCR was purified using a 1.0% low melting agarose gel, and EcoRI and NotI were digested. The DNA fragment thus obtained was cloned into the pCHO1 vector or the pCOS1 vector. After determining the DNA sequence of the plasmid containing the DNA fragment encoding the correct amino acid sequence of the reconstructed single chain Fv 12E10, were designated as pCHO-sc12E10 and pCOS-sc12E10. The nucleotide sequence and amino acid sequence of this reconstructed single chain Fv antibody 12E10 included in the plasmids pCHO-sc12E10 and pCOS-sc12E10 are shown in SEQ ID NO 113.

8.4 Expression of 12E10 Antibody (IgG, Fab) and Single Chain Fv Polypeptide by Animal Cell

Antibody 12E10 (IgG, Fab) and single chain Fv derived from antibody 12E10 (linker sequence of 5 amino acids, 15 amino acids) were expressed using COS7 cells or CHO cells.

Temporary expression using COS7 cells was performed as follows. Transfection was performed by electroporation using a Gene Pulser device (BioRad). For the expression of antibody 12E10 (IgG), 10 μg of the aforementioned expression vector HEF-12E10H-gγ1 and pCOS-12E10L were added, for the expression of the Fab fragment 12E10, 10 μg of pFd-12E10H and pCOS-12E10L were added and for the expression of single-chain Fv 10 μg pCOS- sc12E10 or 10 μg pCOS-db12E10 was added to COS7 cells (1 × 10 ⁷ cells / ml) suspended in 0.8 ml of PBS. The mixture in the cuvette was treated with a pulse of 1.5 kV, with an electric capacitance of 25 μF. After "reconstitution" for 10 minutes at room temperature, electroporated cells were added to DMEM culture medium (Gibco BRL) containing 10% fetal calf serum and cultured. After culturing overnight, the cells were washed once with PBS, added to serum-free medium CHO-S-SFM II (Gibco BRL) and cultured for 3 days. The culture medium was centrifuged to remove cell debris and filtered through a 0.22 μm filter.

To obtain a CHO cell line stably expressing single-chain Fv (polypeptide) derived from 12E10 antibody, pCHO-sc12E10 or pCOS-ds12E10 expression vectors were introduced into CHO cells, respectively.

Each expression vector was introduced into CHO cells by electroporation using a Gene Pulser (BioRad) instrument. Linearized DNA (100 μg) obtained by digestion with Pvul restrictase and CHO cells (1 × 10 ⁷ cells / ml) suspended in 0.8 ml of PBS were mixed in a cuvette, left without stirring on ice for 10 minutes and treated with pulse 1, 5 kV, with an electrical capacitance of 25 μF. After "reconstitution" for 10 minutes at room temperature, electroporated cells were added to CHO-S-SFM II medium (Gibco BRL) containing 10% dialyzed fetal calf serum and nucleic acids, and cultured. After culturing for 2 days, cultivation was continued in nucleic acid-free CHO-S-SFM II medium (Gibco BRL) containing 10% dialyzed fetal calf serum. From the clones thus obtained, a clone with a high level of expression was selected as a cell line producer of single chain Fv 12E10. After culturing in serum-free CHO-S-SFM II medium (Gibco BRL), the culture supernatant was centrifuged to separate cell debris (cell debris) and filtered through a 0.22 μm filter.

8.5 Purification of single chain Fv originating from 12E10 produced by CHO cells

The culture supernatants produced by CHO cell lines expressing single chain Fv 12E10 (sc 12E10, db 12E10) obtained in Example 8.4 were purified using an anti-FLAG antibody column and a gel filtration column, respectively, to obtain purified single chain Fv.

(1) Purification using an anti-FLAG antibody column

Each culture supernatant (sc 12E10, db 12E10) was applied to a column with anti-FLAG M2 affinity gel (Sigma) equilibrated with 50 mM Tris-HCl buffer (pH 7.4) containing 150 mM NaCl. After washing this column with the same buffer, the proteins adsorbed on the column were eluted with 100 mM glycine-HCl buffer (pH 3.5). The eluted fractions were immediately neutralized by adding 1 M Tris-HCl buffer (pH 8.0) and analyzed by SDS-PAGE. Fractions that were confirmed to contain single chain Fv were pooled and concentrated approximately 20 times using Centricon-10 (Millipore).

(2) Gel Filtration

The concentrated solution obtained in (1) was added to a Superdex 200 HR column (10 × 300 mm, Amersham Pharmacia) balanced with PBS containing 0.01% Tween 20. Chromatograms are shown in FIGS. 53 and 54. The product sc12E10 was eluted as two peaks (A, B) (see Fig. 53). The db12E10 product was eluted as two peaks (C, D) (see FIG. 54). The fraction of each peak was collected, processed in the presence and absence of a reducing agent, subjected to electrophoresis according to the Laemmli method, and Coomassie was stained with diamond blue after electrophoresis. As shown in FIG. 55, all fractions A, B, C, and D, regardless of the presence or absence of a reducing agent, gave a single band having an apparent molecular weight of about 31 kDa. When analyzing these fractions by gel filtration using Superdex 200 HR, fraction A gave a product eluting with an apparent molecular weight of about 42 kDa, fraction B 20 kDa, fraction C 69 kDa and fraction D 41 kDa (see FIG. 57) . These results suggest that fraction A derived from sc12E10 is a non-covalently linked single chain Fv dimer sc12E10 and fraction B is a single-chain Fv monomer, and fraction C derived from db12E10 is a non-covalently linked single-chain Fv trimer and D is a non-covalently linked single-chain Fv dimer.

8.6 Measurement of TPO-like agonistic activity of various single chain Fv

TPO-like activity of a single chain anti-MPL antibody was evaluated by measuring proliferative activity against Ba / F3 cells (BaF / mpl) expressing human TPO receptor (MPL).

After washing the BaF / mpI cells twice with RPMI1640 culture medium (Gibco) containing 1% fetal calf serum (Gibco), these cells were suspended in the medium at a cell density of 5 × 10 ⁵ cells / ml. Single chain anti-MPL antibody and human TPO (R & D Systems) were diluted with this culture medium, respectively. 50 μl of cell suspension and 50 μl of diluted human antibody or TPO were added to a 96-well microplate (flat-bottom) (Corning) and cultured in a CO ₂ incubator (CO ₂ concentration: 5%) for 24 hours. After incubation, 10 μl of WST-8 reagent (SF: Nacalai Tesque live cell reagent) was added and the absorbance was immediately measured at a measurement wavelength of 450 nm and at a reference wavelength of 655 nm using a Benchmark Plus fluorescence absorption photometer (BioRad). After incubation in a CO ₂ incubator (CO ₂ concentration: 5%) for 2 hours, the optical density was again measured at a measurement wavelength of 450 nm and a reference wavelength of 655 nm using Benchmark Plus. Since the reagent WST-8 developed color at a wavelength of 450 nm depending on the number of living cells, the proliferative activity of BaF / rnpl was evaluated by the change in optical density over 2 hours.

MPL agonist activity measured using COS-7 cell culture supernatants expressing different variants of 12E10 antibody molecules is shown in FIG. 58. Single chain Fvs having a 5 amino acid linker (ds12E10) and a 15 amino acid linker (sc12E10) were increased optical density in a concentration-dependent manner, showing TPO-like agonistic activity (ED50; 9 pM and 51 pM, respectively), while 12E10IgG and 12E10Fab had no activity.

It has been known that the H chain and the L chain of single chain Fv associate not only within the molecule, but also between the molecules to form multimers such as dimers. The culture supernatants of CHO cells expressing 12E10 single chain Fv antibodies were gel filtered and tested for MPL agonist activity. These results are shown in FIG. A small amount of dimer contained in sc12E10 showed approximately 5000 times stronger TPO-like agonistic activity (dimer sc12E10, ED50; 1.9 pM) compared to monomer (monomer sc12E10, ED50;> 10 nM). Activity was higher than TPO activity (ED50; 27 pM). The db12E10 dimer (ED50; 2.0 pM) showed strong activity comparable to that of the sc12E10 dimer. The db12E10 trimer (ED50; 7.4 pM), which was supposed to be a trimer based on its molecular weight obtained by gel filtration, showed high activity, which is lower than the activity of the db12E10 dimer. These results suggest that it is important for the activity of the 12E10 agonist antibody that the antigen-binding site is bivalent (dimer). Given the fact that 12E10-IgG was not active, it is assumed that other factors other than bivalence are important, such as the location of the antigen binding site, distance or angle.

INDUSTRIAL APPLICABILITY

The modified antibodies of the present invention have an agonistic effect, capable of transducing the signal into cells through the formation of cross-links between the cell surface molecule (s), and have advantages due to their high permeability in tissue and tumor due to their smaller molecular size compared to the original molecule antibodies (whole IgG). This invention provides modified antibodies that have unusually high agonistic activity compared to natural ligands, such as TPO, and the original antibody (whole IgG). Even if the original antibody does not have agonistic activity, a modified antibody with higher agonistic activity compared to natural ligands can be obtained. This is due to the fact that the modified antibodies have a form closer to the ligand in comparison with the original antibodies. Thus, modified antibodies can be used as signal transducing agonists to achieve induction of apoptosis, induction of cell proliferation, induction of differentiation, induction of cell division or regulatory action on the cell cycle. Modification of an antibody molecule with the formation of a modified antibody in accordance with this invention reduces the side effects caused by the formation of intercellular crosslinks and provides new drugs that cause only the desired effect by crosslinking the molecule (s) of the cell surface. Medications containing the modified antibody of the invention as an active ingredient are useful as prophylactic and / or therapeutic agents for cancers, inflammations, hormonal disorders, autoimmune diseases and blood diseases, for example, leukemia, malignant lymphoma, aplastic anemia, myeloplasia syndrome and red, or true, polycythemia.

Claims (31)

1. A modified antibody containing two or more V-regions of the H chain and two or more V-regions of the L chain of the antibody, combined directly or via a linker with a covalent or non-covalent bond, and exhibiting an agonistic effect on cells through the formation of cross-linking molecules ( molecules) of the cell surface or intracellular molecule (s), where the modified antibody is: (i) a single chain Fv multimer comprising an H chain V region and an L chain V region; or (ii) a single chain polypeptide comprising two or more V regions of an H chain and two or more V regions of an L chain. 2. The modified antibody according to claim 1, in which the V region of the H chain and the V region of the L chain are connected through a linker. 3. The modified antibody according to claim 2, in which the linker is a peptide linker containing at least one amino acid. 4. The modified antibody according to any one of claims 1 to 3, where this modified antibody consists of a tetramer, trimer or single-chain Fv dimer. 5. The modified antibody of claim 4, wherein the modified antibody consists of a single chain Fv dimer. 6. The modified antibody according to any one of claims 4 and 5, in which the V region of the H chain and the V region of the L chain located in the same chain are not associated with the formation of an antigen binding site. 7. The modified antibody according to any one of claims 1 to 3, where the modified antibody is a single chain polypeptide containing two V regions of the H chain and two V regions of the L chain. 8. The modified antibody according to any one of claims 1 to 7, where this modified antibody further comprises an amino acid sequence (s) for its purification. 9. The modified antibody according to any one of claims 1 to 8, where this modified antibody has been purified. 10. The modified antibody according to any one of claims 1 to 9, where the V region of the H chain and / or the V region of the L chain is the V region of the H chain and / or the V region of the L chain derived from a human antibody . 11. The modified antibody according to any one of claims 1 to 10, where the V region of the H chain and / or the V region of the L chain is the humanized V region of the H chain and / or the V region of the L chain. 12. The modified antibody according to any one of claims 1 to 11, where the cell surface molecule or intracellular molecule is a hormone receptor, cytokine receptor, tyrosine kinase receptor or intranuclear receptor. 13. The modified antibody according to any one of claims 1 to 12, where the cell surface or intracellular molecule is an erythropoietin receptor (EPO), thrombopoietin receptor (TPO), granulocyte colony stimulating factor receptor (G-CSF), macrophage colony stimulating factor receptor (M- CSF), granulocyte-macrophage colony stimulating factor receptor (GM-CSF), tumor necrosis factor receptor (TNF), interleukin-1 receptor (IL-1), interleukin-2 receptor (IL-2), interleukin-3 receptor (IL- 3), interleukin 4 receptor (IL-4), p interleukin-5 receptor (IL-5), interleukin-6 receptor (IL-6), interleukin-7 receptor (IL-7), interleukin-9 receptor (IL-9), interleukin-10 receptor (IL-10), interleukin-11 receptor (IL-11), interleukin-12 receptor (IL-12), interleukin-13 receptor (IL-13), interleukin-15 receptor (IL-15), interferon-alpha receptor (IFN-alpha), interferon-beta receptor (IFN-beta), interferon-gamma receptor (IFN-gamma), growth hormone receptor (GH), insulin receptor, blood stem cell proliferation factor receptor (SCF), vascular endothelial growth factor receptor (VEGF), receptor epidermal growth factor (EGF), nerve growth factor receptor (NGF), fibroblast growth factor receptor (FGF), platelet growth factor receptor (PDGF), transforming growth factor beta receptor (TGF-beta), leukocyte migration inhibition receptor (LIF), ciliary neurotrophic factor receptor (CNTF), oncostatin M receptor (OSM), Notch family receptor (large transmembrane receptor family), E2F, E2F / DP1 or TAK1 / TAB1. 14. The modified antibody according to any one of claims 1 to 13, where the agonistic effect is the induction of apoptosis, the induction of cell proliferation, the induction of cell differentiation, the induction of cell division or the regulatory effect on the cell cycle. 15. The modified antibody according to any one of claims 1 to 14, where the modified antibody is a monospecific modified antibody. 16. The modified antibody according to any one of claims 1 to 14, where the modified antibody is a multispecific modified antibody. 17. The modified antibody according to clause 16, where the modified antibody is a bispecific modified antibody. 18. The modified antibody of claim 17, wherein the V region of the L chain and the V region of the H chain are derived from the same monoclonal antibody. 19. The modified antibody according to any one of claims 1 to 18, which exhibits an equivalent or better agonistic effect (ED50) in comparison with the original monoclonal antibody. 20. The modified antibody according to claim 19, which exhibits at least 2 times greater agonistic effect (ED50) in comparison with the original monoclonal antibody. 21. The modified antibody according to claim 20, which exhibits at least 10 times greater agonistic effect (ED50) in comparison with the original monoclonal antibody. 22. The modified antibody according to any one of claims 1 to 18, which is derived from the original antibody, essentially no agonistic effect. 23. The modified antibody according to any one of claims 1 to 22, which essentially has no intercellular adhesive action. 24. The modified antibody according to any one of claims 1 to 22, which has an intercellular adhesive action (ED50) of not more than 1/10 in comparison with the original antibody. 25. The use of a modified antibody according to any one of claims 1 to 22 as an agonist. 26. A method of inducing an agonistic effect on cells by forming cross-links between a cell surface molecule (s) or an intracellular molecule (s) by contacting the cells with a modified antibody according to any one of claims 1 to 22. 27. The method according to p, where the agonistic effect is the induction of apoptosis, the induction of cell proliferation, the induction of cell differentiation, the induction of cell division or the regulatory effect on the cell cycle. 28. A drug that exhibits an agonistic effect on cells, containing a modified antibody according to any one of claims 1 to 23. 29. The use of the modified antibody according to any one of claims 1 to 23 as a medicine exhibiting an agonistic effect by forming cross-links between the cell surface molecule (s) or intracellular molecule (s). 30. A screening method for detecting a modified antibody containing two or more V-regions of the H chain and two or more V-regions of the L chain of the antibody, linked directly or via a linker with a covalent or non-covalent bond, and exhibiting an agonistic effect on cells through education cross-linking the molecule (s) of the cell surface or intracellular molecule (s), where the modified antibody is: (i) a single chain Fv multimer comprising an H chain V region and an L chain V region or (ii) a single chain polypeptide containing two or more V regions of the H chain and two or more V regions of the L chain, comprising stages: (1) obtaining a modified antibody containing two or more V-regions of the H chain and two or more V-regions of the L chain of the antibody and binding specifically to the cell surface molecule (s) or intracellular molecule (s); (2) interactions of cells expressing said cell surface molecule (s) or intracellular molecule (s) with this modified antibody; and (3) measuring the agonistic action in cells caused by the formation of cross-linking of the indicated cell surface molecule (s) or intracellular molecule (s). 31. A method for measuring the agonistic effect of a modified antibody according to any one of claims 1-24 in relation to cells, comprising the steps of: (1) obtaining a modified antibody containing two or more V-regions of the H chain and two or more V-regions of the L chain of the antibody and binding specifically to the cell surface molecule (s) or intracellular molecule (s), (2) interactions of cells expressing said cell surface molecule (s) or intracellular molecule (s) with this modified antibody; and (3) measuring the agonistic action in cells caused by the formation of cross-linking of the indicated cell surface molecule (s) or intracellular molecule (s).

Images