Bispecific antibodies that specifically bind to GPNMB and CD3 and uses thereof

Abstract

The present invention relates to a bispecific anti-GPNMB / anti-CD3 antibody that specifically binds to CD3 and GPNMB, and uses thereof. In particular, the bispecific antibody exhibits high affinity and specificity for CD3 and GPNMB, and is therefore capable of inducing the death of GPNMB-expressing cancer cells and inhibiting their proliferation. Therefore, the bispecific antibody can be used as an effective therapeutic agent for cancers that express GPNMB.

Description

[Technical Field]

[0001] The present invention relates to novel bispecific antibodies that specifically bind to GPNMB and CD3 and uses thereof. [Background technology]

[0002] Among various causes of death, cancer death occurs frequently, and its proportion is the second largest.Various methods for treating cancer are continuously tried, and typical examples include administering anticancer drugs, radiation therapy, or surgery.When cancer is in the early stage, its treatment can be carried out by using these methods alone or in combination; however, when cancer is in the late stage, or when cancer has spread to other tissues through the blood or recurred, such methods do not produce much therapeutic effect.

[0003] Therefore, research into immune cell-based therapeutic techniques is gaining attention. Specifically, a technique is being developed in which immune cells collected from a patient's peripheral blood are subjected to in vitro mass expansion, and then the resulting immune cells are re-administered to the patient, resulting in the elimination of cancer cells by cancer cell-specific toxic T cells present in the immune cells. Furthermore, with the development of recombinant technology, bispecific antibodies have also been developed for use in therapeutic areas requiring T cell-mediated killing, such as cancer, and their effectiveness has been confirmed (Buhler, P. et al., Cancer Immunology, Immunotherapy 57.1, 2008:43-52). Nevertheless, there remains a need for the development of bispecific antibodies that produce better anti-cancer effects with minimized adverse effects.

[0004] [DISCLOSURE OF THE INVENTION] [Technical issue] The present inventors have developed a bispecific antibody that can target CD3-expressing immune cells to GPNMB-expressing cancer cells so as to effectively induce the death of GPNMB-expressing cancer cells, and have identified its excellent anti-cancer effect, thereby completing the present invention.

[0005] It is therefore an object of the present invention to provide a bispecific antibody that specifically binds to CD3 and GPNMB.

[0006] Another object of the present invention is to provide a pharmaceutical composition for preventing or treating cancer, which comprises a bispecific antibody or a fragment thereof as an active ingredient.

[0007] [Problem Resolution] To achieve the above-mentioned objects, the present invention provides a bispecific antibody comprising a first domain that specifically binds to GPNMB and a second domain that specifically binds to CD3.

[0008] In addition, the present invention provides a pharmaceutical composition for preventing or treating cancer, comprising the bispecific antibody or a fragment thereof.

[0009] [Advantageous effects of the invention] Due to their high affinity and specificity for GPNMB and CD3, bispecific antibodies according to the present invention can induce the death of GPNMB-expressing cancer cells or inhibit their proliferation, and therefore can be used as effective therapeutic agents for GPNMB-expressing cancers. [Brief explanation of the drawings]

[0010] [Figure 1] 1 shows the results demonstrating the production of anti-GPNMB / anti-CD3 bispecific antibodies, where the horizontal axis represents mL (buffer flow rate) and the vertical axis represents mAu (OD at 280 nm). [Figure 2] FIG. 1 shows the results obtained by performing HPLC analysis on anti-GPNMB / anti-CD3 bispecific antibodies. [Figure 3] FIG. 1 shows FACS results showing the binding pattern of an anti-GPNMB / anti-CD3 bispecific antibody to cancer cell lines. [Figure 4]FIG. 1 shows the PBMC-mediated killing efficacy against cancer cell lines (SK-MEL-2, U87MG, and T98G) observed in the presence of anti-GPNMB / anti-CD3 bispecific antibody. [Figure 5] FIG. 1 shows T lymphocyte activation in cancer cell lines (SK-MEL-2, U87MG, and T98G) observed in the presence of anti-GPNMB / anti-CD3 bispecific antibody. DETAILED DESCRIPTION OF THE INVENTION

[0011] In an embodiment of the present invention, a bispecific antibody is provided, comprising a first domain that specifically binds to GPNMB and a second domain that specifically binds to CD3.

[0012] As used herein, the term "GPNMB" is an abbreviation for Glycoprotein Non-Metastatic Melanoma Protein B, which refers to a glycoprotein that is overexpressed in patients with various cancers, such as breast cancer and melanoma. Although the function of GPNMB has not yet been clearly elucidated, overexpression of GPNMB has occurred in cancer cells. Additionally, GPNMB refers to GPNMB present in animals, preferably humans and monkeys. That is, the term "human GPNMB" refers to GPNMB derived from humans, and the term "mouse GPNMB" refers to GPNMB derived from mice. For example, human GPNMB may have the amino acid sequence of SEQ ID NO: 37.

[0013] As used herein, the term "cluster of differentiation 3 (CD3)" refers to a homodimeric or heterodimeric protein expressed on T cells, which associates with the T cell receptor complex and is an essential element for T cell activation. Functional CD3 is formed by the dimeric association of two or more of four different chains (ε, ζ, δ, and γ). CD3 dimer configurations include γ / ε, δ / ε, and ζ / ζ. For example, human CD3 protein (ε / δ) may have the amino acid sequence of SEQ ID NO: 38, and human CD3 protein (ε) may have the amino acid sequence of SEQ ID NO: 39. Antibodies against CD3 are known to bind to CD3 present on T cells and induce T cell activation. Additionally, CD3 refers to CD3 present in animals, preferably humans and monkeys. That is, the term "human CD3" refers to CD3 derived from humans, and "monkey CD3" refers to CD3 derived from monkeys.

[0014] As used herein, the term "antibody" refers to an immunoglobulin molecule that is immunologically reactive with a specific antigen, i.e., a protein molecule that acts as a receptor that specifically recognizes the antigen. The term "antibody" can be used as a concept that encompasses whole antibodies and antibody fragments.

[0015] As used herein, the term "bispecific antibody" refers to an antibody that can simultaneously bind to two different antigens. In particular, if the type of antigen that a bispecific antibody binds to is appropriately selected, immune cells such as T cells can be toxic only to specific target cells such as cancer cells, and cannot be toxic to other normal cells. Therefore, bispecific antibodies can maximize therapeutic effects while minimizing adverse effects, and therefore can be effectively used in treatments requiring T cell-mediated killing. A bispecific antibody that specifically binds to CD3 and GPNMB according to the present invention can be named an "anti-GPNMB / anti-CD3 bispecific antibody."

[0016] In an embodiment of the anti-GPNMB / anti-CD3 bispecific antibody of the present invention, an antibody is provided that comprises a first domain that specifically binds to GPNMB, forming one of the variable regions of the antibody, and a second domain that specifically binds to CD3, forming the other variable region. The first domain that specifically binds to GPNMB may be cross-reactive with human and monkey GPNMB. Additionally, the second domain that specifically binds to CD3 may be cross-reactive with human and monkey CD3.

[0017] In the first and second domains, some amino acids may be substituted, inserted, and / or deleted as long as the properties consistent with the objectives of the present invention, such as affinity and specificity for GPNMB and CD3, respectively, are maintained. For example, conservative amino acid substitutions may be present. Conservative substitutions mean that an original amino acid residue is replaced with another amino acid residue having similar properties.

[0018] For example, lysine, arginine, and histidine have similar properties in that they have basic side chains, aspartic acid and glutamic acid have similar properties in that they have acidic side chains, glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine, and tryptophan have similar properties in that they have uncharged polar side chains, alanine, valine, leucine, threonine, isoleucine, proline, phenylalanine, and methionine have similar properties in that they have nonpolar side chains, and tyrosine, phenylalanine, tryptophan, and histidine have similar properties in that they have aromatic side chains.

[0019] In an embodiment of the present invention, the first domain may comprise a heavy chain variable region (VH) comprising an H-CDR1 represented by any one of the amino acid sequences selected from the group consisting of SEQ ID NOs: 1, 7, 8, 9, 10, and 11; an H-CDR2 represented by the amino acid sequence of SEQ ID NO: 2; and an H-CDR3 represented by the amino acid sequence of SEQ ID NO: 3; and a light chain variable region (VL) comprising an L-CDR1 represented by the amino acid sequence of SEQ ID NO: 4; an L-CDR2 represented by the amino acid sequence of SEQ ID NO: 5 or 12; and an L-CDR3 represented by the amino acid sequence of SEQ ID NO: 6.

[0020] In an embodiment of the present invention, the first domain may comprise a heavy chain variable region (VH) comprising an H-CDR1 represented by any one of the amino acid sequences selected from the group consisting of SEQ ID NOs: 1, 7, 8, 9, 10, and 11; an H-CDR2 represented by the amino acid sequence of SEQ ID NO: 2; and an H-CDR3 represented by the amino acid sequence of SEQ ID NO: 3; and a light chain variable region (VL) comprising an L-CDR1 represented by the amino acid sequence of SEQ ID NO: 4; an L-CDR2 represented by the amino acid sequence of SEQ ID NO: 5; and an L-CDR3 represented by the amino acid sequence of SEQ ID NO: 6.

[0021] In another embodiment of the present invention, the first domain may comprise a heavy chain variable region (VH) comprising an H-CDR1 represented by the amino acid sequence of SEQ ID NO: 1; an H-CDR2 represented by the amino acid sequence of SEQ ID NO: 2; and an H-CDR3 represented by the amino acid sequence of SEQ ID NO: 3; and a light chain variable region (VL) comprising an L-CDR1 represented by the amino acid sequence of SEQ ID NO: 4; an L-CDR2 represented by the amino acid sequence of SEQ ID NO: 12; and an L-CDR3 represented by the amino acid sequence of SEQ ID NO: 6.

[0022] In addition, the heavy chain variable region (VH) of the first domain can be represented by the amino acid sequence of SEQ ID NO: 26, 27, 28, 29, 30, or 31. In addition, the light chain variable region (VL) of the first domain can be represented by the amino acid sequence of SEQ ID NO: 32 or 33.

[0023] Additionally, the first domain may comprise an scFv type in which the heavy chain variable region and the light chain variable region are linked to each other via a linker. Any amino acid linker may be used as the linker as long as it can link the light chain variable region and the heavy chain variable region to each other. For example, such an scFv may have the amino acid sequence represented by any one of SEQ ID NOs: 19, 20, 21, 22, 23, 24, and 25.

[0024] In an embodiment of the present invention, the second domain may comprise a heavy chain variable region (VH) comprising an H-CDR1 represented by the amino acid sequence of SEQ ID NO: 13; an H-CDR2 represented by the amino acid sequence of SEQ ID NO: 14; and an H-CDR3 represented by the amino acid sequence of SEQ ID NO: 15; and a light chain variable region (VL) comprising an L-CDR1 represented by the amino acid sequence of SEQ ID NO: 16; an L-CDR2 represented by the amino acid sequence of SEQ ID NO: 17; and an L-CDR3 represented by the amino acid sequence of SEQ ID NO: 18.

[0025] In addition, the heavy chain variable region (VH) of the second domain can be represented by the amino acid sequence of SEQ ID NO: 44, 45, or 46. In addition, the light chain variable region (VL) of the second domain can be represented by the amino acid sequence of SEQ ID NO: 42 or 43.

[0026] Additionally, the second domain may comprise an scFv type in which the heavy chain variable region and the light chain variable region are linked to each other via a linker. Any amino acid linker may be used as the linker, as long as it can link the light chain variable region and the heavy chain variable region to each other. For example, such an scFv may have the amino acid sequence represented by SEQ ID NO: 49 or SEQ ID NO: 51. Additionally, the nucleic acid encoding the amino acid sequence may have the nucleotide sequence represented by SEQ ID NO: 50 or SEQ ID NO: 52, respectively.

[0027] In an embodiment of the present invention, each of the first domain and the second domain may further comprise an Fc region, and the Fc region may be derived from the heavy chain constant region (CH) of IgG1, IgG2, IgG3, or IgG4.

[0028] As used herein, the term "Fc region" refers to the C-terminal region of an immunoglobulin heavy chain containing a portion of the constant region. The Fc region may typically contain the CH2 and CH3 regions of the heavy chain constant region of an antibody, and may include wild-type and variant Fc regions.

[0029] In an embodiment of the present invention, one of the Fc regions in the first domain and the second domain may have a knob structure, and the other may have a hole structure. For example, if the Fc region of the first domain has a knob structure, the Fc region of the second domain may have a hole structure, and if the Fc region of the first domain has a hole structure, the Fc region of the second domain may have a knob structure.

[0030] As used herein, the term "knob-into-hole structure" refers to a structure obtained by introducing mutations into the CH3 regions of two different Ig heavy chains, such that a knob structure is introduced into one Ig heavy chain CH3 region and a hole structure is introduced into the other Ig heavy chain CH3 region, allowing the two regions to form a heterodimer.

[0031] Typically, in the amino acid residues forming the knob structure, hydrophobic amino acid residues having large side chains are replaced with hydrophobic amino acid residues having small side chains, and in the amino acid residues forming the hole structure, hydrophobic amino acid residues having small side chains are replaced with hydrophobic amino acid residues having large side chains, although the present invention is not limited thereto.

[0032] Specifically, substitutions can be made for several amino acids (Q347R, S354C, D399V, and F405T) in the CH3 region of the Fc region of the first domain, and for several amino acids (Y349C, K360E, and K409W) in the CH3 region of the Fc region of the second domain. Thus, the first and second domains can be linked to each other via a disulfide bond or a knob-into-hole structure (preferably a knob-into-hole structure) to form a bispecific antibody. However, the present invention is not limited thereto. Here, amino acid residues are numbered according to EU numbering.

[0033] In one embodiment, an Fc region having a hole structure may have the amino acid sequence of SEQ ID NO: 47, and an Fc region having a knob structure may have the amino acid sequence of SEQ ID NO: 48. Thus, the first domain may comprise an Fc region having the amino acid sequence of SEQ ID NO: 47 or SEQ ID NO: 48, and in that case, the second domain may comprise an Fc region having the amino acid sequence of SEQ ID NO: 48 or SEQ ID NO: 47.

[0034] In addition, LALA mutations (L243A, L245A) may be present in the Fc region of each of the first and second domains. When LALA mutations are present in the Fc region, the antibody does not exhibit antibody-dependent cellular cytotoxicity (ADCC) efficacy. Therefore, the antibody may exhibit selective toxicity only against cancer cells, including GPNMB-expressing cells, but not against other normal cells. Here, amino acid residues are numbered according to EU numbering.

[0035] In an embodiment of the invention, the first domain may be represented by the amino acid sequence of SEQ ID NO: 34 or 35, and the second domain may be represented by the amino acid sequence of SEQ ID NO: 36, 40, or 41.

[0036] Additionally, aspects of the present invention provide a polynucleotide encoding the amino acid sequence of the first domain and a polynucleotide encoding the amino acid sequence of the second domain. In one embodiment, the polynucleotide encoding the amino acid sequence of the first domain may be the nucleic acid sequence of SEQ ID NO:53 or SEQ ID NO:54.

[0037] A polynucleotide can be easily derived by one skilled in the art from the amino acid sequence of the bispecific antibody.

[0038] Additionally, in another aspect of the present invention, there is provided an expression vector comprising a polynucleotide encoding the first domain and a polynucleotide encoding the second domain, respectively.

[0039] As used herein, the term "expression vector" refers to a recombinant vector capable of expressing a target protein in a host cell, and means a gene construct containing operably linked essential control elements so that an inserted gene is expressed. The polynucleotides encoding the amino acid sequences of the first domain and the second domain may be inserted into separate vectors or into a single vector.

[0040] As used herein, the term "operably linked" means that a nucleic acid expression control sequence and a nucleic acid sequence encoding a desired protein are functionally linked so as to perform a desired function. Operable linking of recombinant vectors can be achieved using genetic engineering techniques well known in the art, and site-specific DNA cleavage and ligation can be easily achieved using enzymes commonly known in the art.

[0041] A variety of expression host / vector combinations can be used to express bispecific antibodies. Expression vectors suitable for eukaryotic hosts include, but are not limited to, expression control sequences derived from SV40, bovine papilloma virus, adenovirus, adeno-associated virus, cytomegalovirus, and retroviruses. Expression vectors that can be used in bacterial hosts include bacterial plasmids derived from Escherichia coli, such as pET, pRSET, pBluescript, pGEX2T, pUC vectors, colE1, pCR1, pBR322, pMB9, and their derivatives; broad-host-range plasmids such as RP4; and the like.

[0042] Additionally, in an embodiment of the present invention, a host cell transformed with an expression vector is provided. Expression vectors containing a polynucleotide encoding the first domain and a polynucleotide encoding the second domain, respectively, can be inserted into a host cell to form a transformant. Suitable host cells for the vectors include prokaryotic cells such as Escherichia coli, Bacillus subtilis, Streptomyces sp., and Pseudomonas sp. Host cells can include eukaryotic cells, including yeast such as Saccharomyces cerevisiae, and higher eukaryotic cells such as insect cells.

[0043] In addition, the host cell may also be derived from a plant or a mammal.Preferably, the host cell may include, but is not limited to, monkey kidney cells (COS7 cells), NSO cells (myeloma cells of mouse origin), SP2 / 0 cells (myeloma cells of mouse origin), other myeloma cell lines, Chinese hamster ovary (CHO) cells, MDCK, HuT 78 cells, and HEK293 cells, with CHO cells being preferred.

[0044] Meanwhile, in another aspect of the present invention, there is provided a method for producing an anti-GPNMB / anti-CD3 bispecific antibody, comprising the steps of culturing host cells and purifying the anti-GPNMB / anti-CD3 antibody.

[0045] In detail, a method for producing a bispecific antibody may comprise the steps of inserting a polynucleotide encoding a first domain and a polynucleotide encoding a second domain into a vector to construct a recombinant vector, transforming the recombinant vector into a host cell and culturing it, and isolating and purifying the bispecific antibody from the cultured transformant.

[0046] Bispecific antibodies can be produced in large quantities by culturing transformants expressing recombinant vectors in a nutrient medium. The medium and culture conditions can be appropriately selected from those known in the art depending on the type of host cell. During culture, conditions such as temperature, medium pH, and culture time can be appropriately adjusted to suit cell growth and mass production of proteins.

[0047] Additionally, in an aspect of the present invention, there is provided a pharmaceutical composition for preventing or treating cancer, comprising the bispecific antibody or a fragment thereof.

[0048] The bispecific antibody can specifically bind to CD3-expressing T cells and GPNMB-expressing cancer cells. The cancer may be one or more selected from the group consisting of colorectal cancer, lung cancer, brain cancer, pancreatic cancer, ovarian cancer, breast cancer, prostate cancer, liver cancer, thyroid cancer, head and neck cancer, gastric cancer, bladder cancer, non-Hodgkin's lymphoma, skin cancer, melanoma, leukemia, neuroblastoma, and glioblastoma. However, the cancer is not limited thereto and may include any cancer in which GPNMB is expressed. The bispecific antibody can induce T cells through specific binding to CD3, thereby inducing the death of GPNMB-expressing cancer cells or inhibiting their proliferation.

[0049] The pharmaceutical composition may further comprise a pharmaceutically acceptable carrier, such as a binder, a glidant, a disintegrant, an excipient, a solubilizer, a dispersant, a stabilizer, a suspending agent, a dye, a flavoring, etc., which may be used for oral administration, a buffer, a preservative, an analgesic, a solubilizer, an isotonic agent, a stabilizer, etc., which may be used for an injection mixture, or a base, an excipient, a lubricant, a preservative, etc., which may be used for topical administration.

[0050] The pharmaceutical composition can be prepared in various ways by mixing with the pharmaceutically acceptable carrier as described above. For example, for oral administration, the pharmaceutical composition can be formulated in the form of tablets, troches, capsules, elixirs, suspensions, syrups, wafers, or the like. For injection, the pharmaceutical composition can be formulated in the form of unit dose ampoules or multiple dose forms.

[0051] The pharmaceutical composition can be administered in a pharmaceutically effective amount for treating cancer cells or their metastasis, or inhibiting cancer growth. The effective amount may vary depending on various factors such as the type of cancer, the patient's age, weight, nature, and severity of symptoms, the type of current treatment, the number of treatments, dosage form, and administration route, and can be easily determined by experts in the corresponding fields.

[0052] The pharmaceutical composition may be administered together or sequentially with the pharmacological or physiological components described above, and may also be administered in combination with additional conventional therapeutic agents, in which case the pharmaceutical composition may be administered sequentially or simultaneously with the conventional therapeutic agent. Such administration may be single or multiple administrations. Taking all of the above factors into consideration, it is important to administer the minimum amount that allows for maximum effect without adverse effects, and such an amount can be easily determined by those skilled in the art.

[0053] The present invention provides a method for preventing or treating cancer, which comprises the step of administering a pharmaceutical composition to a subject.

[0054] As used herein, the term "subject" refers to a mammal, preferably a human, suffering from or at risk of a condition or disease that can be alleviated, inhibited, or treated by administration of a pharmaceutical composition.

[0055] As used herein, the term "administration" means introducing a given substance into a subject in any suitable manner, and pharmaceutical compositions can be administered via any route as long as the route allows the pharmaceutical composition to reach the target tissue. Such administration methods may include, but are not limited to, intraperitoneal, intravenous, intramuscular, subcutaneous, intradermal, oral, topical, intranasal, pulmonary, or rectal administration. In the case of oral administration, it may be desirable to formulate the oral composition in such a way that the active substance is coated or the composition is protected from digestion in the stomach, in view of the digestion of proteins. In addition, pharmaceutical compositions may be administered by any device that allows the active ingredient to migrate to its target cells.

[0056] Additionally, the present invention provides use of the pharmaceutical composition for the manufacture of a medicament for preventing or treating cancer. [Example]

[0057] The present invention will now be described in more detail through the following examples, which are for illustrative purposes only and do not limit the scope of the present invention.

[0058] Example 1. Generation of anti-GPNMB / anti-CD3 bispecific antibodies Example 1.1. Selection of anti-GPNMB antibodies To select GPNMB-specific antibodies, the researchers used recombinant DNA technology to insert the DNA sequence to be expressed into the genome of a bacteriophage that parasitizes E. coli, and then used phage display technology to express the inserted gene on the surface of the phage in a fused form with one of the phage coat proteins.

[0059] Single colonies were collected from the final amplified population of the synthetic phage display scFv library. Subsequently, the colonies were cultured in 1.5 mL of SB / carbenicillin at 37°C and 220 rpm until the OD600 reached approximately 0.8-1.0. They were then cultured for 12 hours or more under conditions of 1 mM IPTG, 30°C, and 200 rpm. The reaction products were centrifuged at 5,500 rpm for 5 minutes, and each supernatant was then added to an ELISA plate coated with GPNMB antigen. The plate was then incubated at room temperature for 2 hours and washed four times with PBST (1x PBS, 0.05% Tween 20). A 1:5000 dilution of HRP / anti-hFab-HRP conjugate in 1% BSA / 1x PBS was then added to the plate, and the reaction was allowed to proceed at room temperature for 1 hour. Subsequently, the plate was washed again four times with PBST (1x PBS, 0.05% Tween 20), after which TMB solution was added to the plate and the reaction was allowed to proceed for 5-10 minutes, followed by the addition of TMB stop solution.

[0060] Subsequently, the OD values ​​were read at a measurement wavelength of 450 nm using a TECAN Sunrise, and clones with high OD values ​​were identified as individual clones. As a result, 23 clones that specifically bind to human GPNMB were selected, and their amino acid sequences were identified. Using the selected clones, their binding ability to GPNMB-expressing cancer cell lines was checked. As a result, only one clone was identified that bound to the cell line. Based on this, six additional clones with enhanced protein-binding and cell-binding abilities were obtained by affinity maturation.

[0061] As a result, a total of seven clones that bind to GPNMB protein and GPNMB-expressing cancer cell lines were obtained, and the selected clones were named clone GPNMB1 (sequence number 55), clone GPNMB2 (sequence number 56), clone GPNMB3 (sequence number 57), clone GPNMB4 (sequence number 58), clone GPNMB5 (sequence number 59), clone GPNMB6 (sequence number 60), and clone GPNMB7 (sequence number 61), respectively.

[0062] The variable region sequences of the clones are shown in SEQ ID NOs: 26 to 33, respectively, and the CDR amino acid sequences in each variable region, identified according to Kabat numbering, are shown in Table 1 below.

[0063] [Table 1]

[0064] In addition, clone GPNMB6 (SEQ ID NO: 18), which exhibits the highest affinity among anti-GPNMB antibodies, was engineered to have a knob-into-hole structure.

[0065] Example 1.2. Selection of anti-CD3 antibodies To select antibodies specific to human and monkey CD3, the mouse SP34 antibody was humanized and antibodies that bound to CD3 with various affinities were selected. Among these, clone A15, consisting of the amino acid sequence of SEQ ID NO: 36, and clone E15, consisting of the amino acid sequence of SEQ ID NO: 41, were obtained. In addition, an antibody having the amino acid sequence of SEQ ID NO: 40 (Hu38E4.v1, manufactured by Genentech) was produced and used. Additionally, for bispecific antibodies, anti-CD3 antibodies (Hu38, A15, or E15) were constructed to have knob-into-hole structures.

[0066] Example 1.3. Introduction of vectors for antibody expression 2.0 × 10 6Expi293F cells at a density of 1000 cells / ml were passaged in Expi293 medium at 125±10 rpm in a shaking incubator at 37°C and 8% CO2. At the time of transfection, cell number and cell viability were determined to determine if the cells exhibited 95% or greater cell viability. Cells were cultured at 5×10 in a 500 mL culture flask. 8 The cells were dispensed, and then Expi293 medium was added to adjust the final volume to 170 mL (based on 200 mL). 200 μg of antibody expression vector was mixed with Opti-MEM I medium to a total of 1,500 μL, and incubation was carried out at room temperature for 5 minutes.

[0067] 540 μl of transfection reagent was mixed with Opti-MEM I medium to a total volume of 1,500 μl, and the mixture was incubated at room temperature for 5 minutes. The Opti-MEM I medium containing the vector and transfection reagent was gently mixed and incubated at room temperature for 20 minutes. The reaction mixture was then added to a flask containing Expi293F cells. The cells were incubated in a shaking incubator at 125±10 rpm, 37°C, and 8% CO2 for 16 to 20 hours. Then, 1 ml of transfection enhancer I and 10 ml of transfection enhancer II were added, and the mixture was incubated for 6 days to obtain candidate antibodies.

[0068] Example 1.4. Generation of anti-GPNMB / anti-CD3 bispecific antibodies The culture was centrifuged at 4,000 rpm for 30 minutes and filtered through a 0.22 μm filter to remove cell debris and obtain the supernatant. 1 ml of Mabselect Xtra resin was added to the column, and equilibration was achieved with a volume of Protein A binding buffer corresponding to 10 times the resin volume.

[0069] The supernatant was then loaded onto the column by gravity. After loading was complete, the column was washed with a Protein A binding buffer at a volume equivalent to 10 times the resin volume. Subsequently, IgG elution buffer was added to the column to perform elution. The eluate was neutralized by adding 25 μl of 1.5 M Tris-Cl per ml of eluate, and then the concentration was measured at OD 280 nm. The eluate whose concentration had been measured was subjected to buffer exchange with PBS by dialysis.

[0070] The sample was then concentrated or diluted to approximately 1.0-2.0 g / L and loaded onto a HiLoad 16 / 600 Superdex 200 pg column (GE Healthcare, 28989335) in an AKTA Purifier 900. The mobile phase was 20 mM sodium phosphate (pH 7.0) containing 200 mM sodium chloride, and the flow rate was 1.0 ml / min. The eluted fractions were then collected approximately 50-70 minutes after sample loading. The eluted fractions were subjected to Coomassie Blue staining using 4%-12% Bis-Tris PAGE (Invitrogen, 0321BOX), and the fractions containing the 150 kDa antibody were subsequently collected. The fractions were subsequently concentrated using 30 kDa Amicon centrifugal filter units (Merck, UFC803024).

[0071] FIG. 1 illustrates the results of co-expression and purification of a bispecific antibody comprising an anti-GPNMB antibody fragment (SEQ ID NO: 22) and an anti-CD3 antibody fragment (SEQ ID NO: 1), showing that a protein sample corresponding to 150 kDa was obtained.

[0072] Example 1.5. Determining the Purity of Bispecific Antibodies by HPLC Analysis For HPLC analysis, 50 mM sodium phosphate (pH 6.0) was used as the equilibration buffer. The solution obtained by adding sodium chloride to 500 mM sodium phosphate (pH 6.0) and then filtering through a 0.45 μm bottle-top filter (Nalgene, 597-4520) was used as the elution buffer. A cation column (Thermofisher, 054993) was connected to an HPLC system (Waters, 2695 / 2489) and then equilibrated with the equilibration buffer. The sample to be analyzed was diluted 10-fold or more in the equilibration buffer to prepare the loading sample. The mobile phase was analyzed in a flow-through manner with the equilibration buffer at 0.5 mL / min for 10 minutes, followed by a linear gradient of 0% to 100% elution buffer over 40 minutes. After analysis was completed, the purity was determined by calculating the area in the chromatogram measured at UV 280 nm.

[0073] HPLC analysis was performed on the purified protein sample, with the results shown in Figure 1. As a result, it was confirmed that anti-GPNMB antibody fragments and anti-CD3 antibody fragments were almost absent in the GPNMB / CD3 bispecific antibody sample (Figure 2).

[0074] Example 2. Analysis of the affinity of bispecific antibodies to GPNMB protein The quantitative binding affinity of the purified bispecific antibody to recombinant human GPNMB was measured using a Biacore T-200 biosensor (GE Healthcare, USA). GPNMB purified from HEK293 cells was immobilized on a CM5 chip (GE Healthcare) using an amine-carboxyl chemistry until an Rmax of 200 was achieved. Next, serially diluted GPNMB / CD3 bispecific antibody in HBS-EP buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA, 0.005% surfactant P20) was bound to the chip for 120 seconds at concentrations ranging from 0.078 nM to 5 nM. The antibody was then allowed to flow at a flow rate of 30 μL / min for 1800 seconds to reach dissociation. Dissociation of antibody bound to GPNMB was induced by flowing 10 mM glycine-HCl (pH 1.5) at a flow rate of 30 μL / min for 30 s (Table 2). The affinity was determined by the kinetic rate constant (K on and K. off ) and the equilibrium dissociation constant (K D ) were obtained using the Biacore T-200 evaluation software (Table 3).

[0075] [Table 2]

[0076] [Table 3]

[0077] Example 3. Analysis of the affinity of bispecific antibodies to GPNMB-expressing cells and CD3-expressing cells Example 3.1. Analysis of the affinity of bispecific antibodies to human GPNMB-expressing cancer cell lines The affinity of the bispecific antibody for GPNMB-expressing cells was determined using flow cytometry. Specifically, 5 ml tubes containing 100 μl of FACS buffer (2% FBS / PBS) were prepared, and 3 × 10 cells of each of three cancer cell lines (SK MEL2, U87MG, and T98G) were added to each tube.5 The cells were added at a concentration of 100 μg per tube, and each tube was treated with 0.5 μg of primary antibody. The cells were then incubated for 30 minutes at 4°C in the dark. Subsequently, 1 ml of FACS buffer was added. The cells were centrifuged at 1,500 rpm at 4°C for 3 minutes, and the supernatant was then removed.

[0078] Next, each tube was treated with 0.2 μg of a fluorescent dye-labeled secondary antibody capable of specifically binding to the primary antibody in 100 μl of FACS buffer. The mixture was then incubated for 30 minutes at 4°C in the dark. Subsequently, 1 ml of FACS buffer was added, and the mixture was centrifuged at 1,500 rpm and 4°C for 3 minutes. The supernatant was then removed to obtain a sample. Then, 200 μl of a buffer prepared at a ratio of FACS buffer to BD Cytofix™ = 1:4 was added to the sample to suspend the cells, and analysis was performed using a BD FACSCalibur. The results are shown in Figure 3.

[0079] As shown in Figure 3, GPNMB (SEQ ID NO: 35 ) / A15(SEQ ID NO: 36 ) The bispecific antibody was found to bind to three types of cancer cell lines. A negative control antibody, used as an irrelevant control, did not bind to any of the three types of cancer cell lines.

[0080] Example 3.2. Analysis of the affinity of bispecific antibodies to human CD3-expressing cancer cell lines Flow cytometry was used to determine whether the bispecific antibody prepared in Example 1 also exhibits affinity for CD3-expressing cells. Specifically, 5 ml tubes containing 100 μl of FACS buffer (2% FBS / PBS) were prepared, and 3 × 10 Jurkat cell line cells were added to each tube. 5The cells were added at a concentration of 100 μg per tube, and each tube was treated with 0.5 μg of primary antibody. After incubation for 30 minutes at 4°C in the dark, 3 ml of FACS buffer was added, followed by centrifugation at 1,500 rpm and 4°C for 3 minutes, after which the supernatant was removed.

[0081] Next, each tube was treated with 0.2 μg of a fluorescent dye-labeled secondary antibody capable of specifically binding to the primary antibody in 100 μl of FACS buffer. The mixture was then incubated for 30 minutes at 4°C in the dark. Subsequently, 1 ml of FACS buffer was added, and the mixture was centrifuged at 1,500 rpm and 4°C for 3 minutes. The supernatant was then removed to obtain a sample. Then, 200 μl of a buffer prepared at a ratio of FACS buffer to BD Cytofix™ = 1:4 was added to the sample to suspend the cells, and analysis was performed using a BD FACSCalibur. The results are shown in Figure 3.

[0082] As shown in Figure 3, the GPNMB (SEQ ID NO: 35) / A15 (SEQ ID NO: 36) bispecific antibody was found to bind to the cell line Jurkat. A negative control antibody used as an irrelevant control was Jurkat cell line Did not bond.

[0083] Example 4. Evaluation of cell killing efficacy of bispecific antibodies against tumor cell lines Using human peripheral blood mononuclear cells (PBMCs) and the bispecific antibody prepared in Example 1, the GPNMB-specific tumor cell-killing efficacy of the bispecific antibody against three types of GPNMB+ tumor cells (SK-MEL-2, U87MG, and T98G) was determined according to the following method.

[0084] Example 4.1. Construction of target cell lines Three types of GPNMB+ tumor cells (SK-MEL-2, U87MG, and T98G) were harvested using 1x trypsin-EDTA solution and centrifuged at 1,200 rpm at 4°C for 5 minutes. The supernatant was then removed and resuspended in cRPMI (RPMI, A10491-01 + 10% FBS + 55 μM β-ME). The cell number was then quantified. Each cell line suspension was diluted to 1.0 x 10 5 The cells were prepared at a concentration of 1000 cells / ml, added to a 6-well plate at 1 ml / well, and incubated in a CO2 incubator at 37°C for 1 day to prepare cell lines. Next, transduction was performed at a multiplicity of infection of 3 (MOI 3) using IncuCyte® NucLight Red lentiviral reagent (EF-1 alpha promoter, puromycin selection).

[0085] Example 4.2. Preparation of target cell lines Specifically, cells were harvested using 1x trypsin-EDTA solution and centrifuged at 1,200 rpm at 4°C for 5 minutes. Subsequently, the supernatant was removed and resuspended in cRPMI (RPMI, A10491-01 + 10% FBS + 55 μM β-ME). The cell number was then quantified. Each cell line suspension was diluted to 1x10 5 The target cell line was prepared by preparing a concentration of 100 μl / ml of cells and adding it to a 96-well plate at 100 μl / well, and incubating the plate at 37° C. in a CO 2 incubator for 1 day.

[0086] Example 4.3. Preparation of peripheral blood mononuclear cells Cryopreserved peripheral blood mononuclear cells (PBMCs) were rapidly thawed in a water bath at 37°C and then transferred to a 50 ml conical tube. Thawing medium (RPMI, 11875-093 + 10% FBS + 55 μM β-ME) was added dropwise to the tube and mixed by shaking. The supernatant was then removed by centrifugation at 1,200 rpm and 4°C for 10 minutes, and the cells were resuspended in 30 ml of thawing medium. The number of cells was then quantified, and the concentration was determined to be 1.0 × 10 6Cells were suspended in cRPMI for each donor, adjusted to cells / ml.

[0087] Example 4.4. Plating of Peripheral Blood Mononuclear Cells and Antibodies Each antibody was diluted in cRPMI, then diluted 1 / 5 starting at 20 nM. Antibodies were applied to wells seeded with target cells one day earlier. 100 μl / well of pre-prepared PBMCs were then added to the wells to obtain a target:PBMC ratio of 1:10 (SK-MEL-2) or 1:20 (U87MG, T98G).

[0088] Example 4.5. Real-time cell imaging analysis using IncuCyte S3 Brightfield and red fluorescence were measured at 10x magnification at 2-hour intervals using an IncuCyte S3 microscope during incubation at 37°C in a CO2 incubator for 2 days. The results confirmed that the bispecific antibody exhibited dose-dependent cell-killing efficacy against three types of GPNMB-expressing cancer cells (Figure 4). In contrast, a bispecific antibody (irrelevant / A15) obtained by linking an irrelevant Ab to the A15 antibody did not induce cell death. On the other hand, a bispecific antibody was generated using A15 (SEQ ID NO: 36) and a third-party antibody that exhibited similar cell-binding ability to GPNMB (SEQ ID NO: 35) / A15 (SEQ ID NO: 36), and its cell-killing efficacy was determined. As a result, this bispecific antibody was found to exhibit lower efficacy than the GPNMB clone. Based on this, it is demonstrated that the GPNMB antibody (SEQ ID NO: 35) recognizes an effective epitope as a bispecific T cell-inducing antibody.

[0089] Example 5. Measurement of T cell activity induced by bispecific antibodies To analyze the degree of T cell activation caused by the bispecific antibody prepared according to Example 1, the T cell activation ability of the bispecific antibody was analyzed using the cell line IL2-luc2P Jurkat (Promega) against three cancer cell lines overexpressing GPNMB (SK MEL2, U87MG, and T98G).

[0090] Specifically, GPNMB-expressing SK MEL2, U87MG, and T98G cells were each plated in 100 μl of medium at 3 × 10 cells per well in a 96-well plate. 4 The target cell line was prepared by seeding the cells and incubating them in a humidified incubator at 37°C and 5% CO2 for 18 hours. GloResponse™ Frozen Thaw and Use (FTU) IL-2-luc2P Jurkat effector cells were thawed in a water bath at 37°C for 2 minutes. 4 mL of prewarmed assay medium (RPMI medium containing 10% FBS) was added to a 15 mL conical centrifuge tube, followed by 1 mL of thawed effector cells. The effector cell line was prepared by gently mixing. Next, 75 μl of medium was removed from the 96-well plate in which the target cells had been seeded, and 25 μl of FTU IL2-Luc2P Jurkat cells were added to the plate at 1 x 10 per well. 5 cells were added.

[0091] Antibodies were prepared by diluting 1 / 3 starting from 10 nM to obtain 10 points at a 3x dose. Then, 25 μl of the prepared antibodies at 10 concentrations was added to a pre-prepared 96-well plate containing FTU IL2-Luc2P Jurkat cells to achieve a 1x dose. Incubation was carried out for 5 hours at 37°C and 5% CO2 in a humidified incubator. The plate was then removed from the incubator and allowed to stand at room temperature for 10-15 minutes. Subsequently, 75 μl of Bio-Glo™ reagent was added per well, and the plate was allowed to stand for 5 minutes. Measurements were then performed using a GloMax™ Multi+ multiwell plate reader. The results showed that the bispecific antibodies induced T cell activation in all three cancer cell types, and the EC2 activity increased as the antibody's affinity for CD3 decreased. 50 Increased values ​​were observed. Bispecific antibodies generated using an irrelevant Ab and Hu38 did not induce T cell activation (Figure 5).

Claims

1. A first domain that specifically binds to GPNMB, and A second domain that specifically binds to CD3 1. An anti-GPNMB / anti-CD3 bispecific antibody comprising: the first domain comprises a heavy chain variable region (VH) comprising H-CDR1 represented by the amino acid sequence of SEQ ID NO: 11; H-CDR2 represented by the amino acid sequence of SEQ ID NO: 2; and H-CDR3 represented by the amino acid sequence of SEQ ID NO: 3; and a light chain variable region (VL) comprising L-CDR1 represented by the amino acid sequence of SEQ ID NO: 4; L-CDR2 represented by the amino acid sequence of SEQ ID NO: 5; and L-CDR3 represented by the amino acid sequence of SEQ ID NO: 6; and the second domain comprises a heavy chain variable region (VH) comprising H-CDR1 represented by the amino acid sequence of SEQ ID NO: 13; H-CDR2 represented by the amino acid sequence of SEQ ID NO: 14; and H-CDR3 represented by the amino acid sequence of SEQ ID NO: 15; and a light chain variable region (VL) comprising L-CDR1 represented by the amino acid sequence of SEQ ID NO: 16; L-CDR2 represented by the amino acid sequence of SEQ ID NO: 17; and L-CDR3 represented by the amino acid sequence of SEQ ID NO: 18; Anti-GPNMB / anti-CD3 bispecific antibody.

2. 2. The anti-GPNMB / anti-CD3 bispecific antibody of claim 1, wherein the first domain has a variable region in which a heavy chain variable region and a light chain variable region are linked to each other via a linker.

3. 3. The anti-GPNMB / anti-CD3 bispecific antibody of claim 2, wherein the variable region of the first domain has the amino acid sequence of SEQ ID NO:

24.

4. The anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 3, wherein the second domain has a variable region in which a heavy chain variable region and a light chain variable region are linked to each other via a linker.

5. 5. The anti-GPNMB / anti-CD3 bispecific antibody of claim 4, wherein the variable region of the second domain has the amino acid sequence of SEQ ID NO:

49.

6. The anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 5, wherein the first domain further comprises an Fc region.

7. The anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 6, wherein the second domain further comprises an Fc region.

8. 8. The anti-GPNMB / anti-CD3 bispecific antibody of claim 6 or 7, wherein the Fc region is derived from the heavy chain constant region (CH) of IgG1, IgG2, IgG3, or IgG4.

9. The anti-GPNMB / anti-CD3 bispecific antibody of claim 6 or 7, wherein one of the Fc regions in the first domain and the second domain has a knob structure and the other has a hole structure.

10. 7. The anti-GPNMB / anti-CD3 bispecific antibody of claim 6, wherein the first domain comprises an Fc region represented by the amino acid sequence of SEQ ID NO: 47 or 48.

11. 8. The anti-GPNMB / anti-CD3 bispecific antibody of claim 7, wherein the second domain comprises an Fc region represented by the amino acid sequence of SEQ ID NO: 48 or 47.

12. 12. The anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 11, wherein the first domain is represented by the amino acid sequence of SEQ ID NO:

35.

13. 13. The anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 12, wherein the second domain is represented by the amino acid sequence of SEQ ID NO:

36.

14. The anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 13, which specifically binds to T cells and GPNMB-expressing cancer cells.

15. A polynucleotide encoding the amino acid sequence of the first domain of the anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 14.

16. An expression vector containing the polynucleotide of claim 15.

17. A host cell transformed with the expression vector of claim 16.

18. 1. A method for producing an anti-GPNMB / anti-CD3 bispecific antibody, comprising: A host cell transformed with an expression vector containing a polynucleotide encoding the amino acid sequence of the first domain and a polynucleotide encoding the amino acid sequence of the second domain of the anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 14, or Culturing a host cell transformed with an expression vector containing a polynucleotide encoding the amino acid sequence of the first domain of the anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 14 and an expression vector containing a polynucleotide encoding the amino acid sequence of the second domain of the anti-GPNMB / anti-CD3 bispecific antibody of any one of claims 1 to 14; and Purifying anti-GPNMB / anti-CD3 antibodies A method comprising:

19. A pharmaceutical composition for preventing or treating cancer, comprising: comprising an anti-GPNMB / anti-CD3 bispecific antibody, A pharmaceutical composition, wherein the bispecific antibody is an anti-GPNMB / anti-CD3 bispecific antibody according to any one of claims 1 to 14.

20. 20. The pharmaceutical composition of claim 19, wherein the cancer is one or more selected from the group consisting of colorectal cancer, lung cancer, brain cancer, pancreatic cancer, ovarian cancer, breast cancer, prostate cancer, liver cancer, thyroid cancer, head and neck cancer, gastric cancer, bladder cancer, non-Hodgkin's lymphoma, skin cancer, melanoma, leukemia, neuroblastoma, and glioblastoma.

21. 21. Use of the pharmaceutical composition according to claim 19 or 20 for the manufacture of a medicament for preventing or treating cancer.

Images