Bivalent bispecific antibody and method for producing thereof

Abstract

The invention relates to the field of biotechnology, specifically to a bivalent bispecific antibody that comprises a first light chain and a first heavy chain of an antibody that specifically bind to a first antigen, and a second light chain and a heavy chain of an antibody that specifically bind to a second antigen, wherein the variable domains VL and VH are substituted for one another, and wherein the VL and VH have a modification at the C-terminus of FR4. Further provided is a method for producing said antibody. The invention makes it possible to produce bispecific antibodies in an efficient manner. 2 independent and 19 dependent claims, 9 tables, 9 figures, 10 examples.

Description

[0001] BIVALENT BISPECIFIC ANTIBODY AND METHOD FOR PRODUCING THEREOF

[0002] Field of the invention

[0003] The present invention relates to the field of biotechnology, specifically to a bivalent bispecific antibody that comprises a first light chain and a first heavy chain of an antibody that specifically bind to a first antigen, and a second light chain and a heavy chain of an antibody that specifically bind to a second antigen, wherein the variable domains VL and VH are substituted for one another, and wherein the VL and VH have a modification at the C-terminus of FR4, as well as to a method for producing thereof.

[0004] Background of the invention

[0005] Monoclonal antibodies in the form of chimeric, humanized or fully human molecules have proven to be useful as effective medicine for treating multiple disorders and diseases.

[0006] Naturally occurring human antibody molecules consist of two heavy chain homodimers, each of which forms a heterodimer in partnership with two identical light chain molecules. Conventional monoclonal antibodies in the form of whole molecules consist of bivalent ("twoarmed") heterodimers of heavy and light chains.

[0007] Diseases are often caused as a result of multiple pathologies and are accompanied by many concomitant diseases. Bispecific antibodies are capable of binding and thereby neutralizing two different antigens per antibody molecule. The potential for a significant improvement in the therapeutic properties (and value) of medicinal products as compared to monoclonal antibodies has made bispecific antibodies an active area of research. Over the past twenty years, the literature has described many solutions regarding engineered versions of bispecific antibodies, as described in Brinkmann, U and RE Kontermann, 2017, The Making of Bispecific Antibodies, MAbs; 209 Feb / Mar; 9(2):182-212, doi: 10.1080 / 19420862.2016.1268307.

[0008] There are many approaches to create molecules with combined antigen-binding domains, i.e. with antigen-binding domains that differ from one another. However each of these methods has its disadvantages.

[0009] Cross-linking by chemical methods is a time-consuming process, since the corresponding portions should be purified from homodimers and other undesirable by-products. In addition, the steps of chemical modification may alter the integrity of proteins, thus impairing stability thereof. Thus, the above method is typically ineffective and may lead to the loss of antibody activity. A method based on cell fusion (for example, production of hybridomas) is an arbitrary assembly of two heavy and two light chains, resulting in 10 combinations of antibodies. Target heteromultimeric antibodies are only a small part of the antibodies produced in this fashion. Isolation of target heteromultimeric proteins significantly reduces product yield and increases production costs.

[0010] Recombinant DNA techniques are employed to create various heteromultimeric antibodies, for example, single-chain Fv fragments, diabodies, etc. that are free of an Fc fragment. The main disadvantage of this type of an antibody molecule is an absent Fc domain, which results in antibody failure to trigger an effector function (such as, for example, complement activation, binding to an Fc receptor, etc.). Thus, there is a need for a bispecific antibody comprising a functional Fc domain.

[0011] Recombinant DNA techniques are further employed to design bispecific antibodies using the Knob-into-Holes technology. See international applications WO 9627011 and WO 9850431, as well as Merchant AM ET ALL., An efficient route to human bispecific IgG, Nat Biotechnol. 1998 Jul;16(7):677-81. One factor limiting the use of the above method is the fact that the light chains of the two initial antibodies should be identical to prevent mispairing and formation of undesirable and / or inactive molecules when expressed in a single cell.

[0012] The purity of bispecific antibody product depends on two factors as follows: a) heterodimeric assembly of two distinct heavy chains co-expressed in a cell, and b) correct pairing of two distinct light chains to the corresponding heavy chains.

[0013] The "Knob-into-Holes" technology to design bispecific antibodies solves the problem of correct heterodimeric assembly of two distinct heavy chains co-expressed in the cell. However, the use of the Knob-into-Holes technology to design bispecific antibodies makes it possible to achieve only about 25% yield of a properly assembled bispecific product, as the problem of correct pairing of two distinct light chains to the corresponding heavy chains is still unresolved.

[0014] The problem of correct pairing of two distinct light chains to the corresponding heavy chains is solved in various fashions:

[0015] 1. Use of an identical light chain in first and second antigen-binding portions of antibody (Van Blarcom T ET AL., Productive common light chain libraries yield diverse panels of high affinity bispecific antibodies, MAbs. 2018 Feb / Mar;10(2):256-268. doi:

[0016] 10.1080 / 19420862.2017.1406570).

[0017] The disadvantage of the above solution is non-universality thereof, since it may be problematic to select a light chain suitable for the both valencies. Furthermore, in case of amino acid substitutions in the light chain to optimize the properties of the antigen-binding fragment, the substitutions will affect the both valencies. Further, antibody binding to the second antigen may be disrupted. antigen-binding fragment specific for the first antigen are connected to one another via a linker of several amino acids.

[0018] This format has technological disadvantages, since it uses linkers either to fuse the antibody core (IgA, IgD, IgE, IgG or IgM) to a further binding protein (for example, scFv or scFab), or to fuse, for example, light and heavy variable domains (VH and VL) within scFv or a light chain (VL-CK(or CL)) to VH-CH1 within scFab. Linkers may cause problems in therapeutic settings. In fact, these foreign peptides can elicit an immune response against the linker itself or the junction region between the protein and the linker. Furthermore, the flexible nature of these peptides and the mobility thereof make them more prone to proteolytic cleavage, potentially leading to poor antibody stability, aggregation and increased immunogenicity.

[0019] 3. Modification of CHl-CK domains in a bispecific antibody to allow altering the interaction interface in bispecific antibody expression techniques so as to exclude the incorrect association of light chains. For example, the patent document WO 2017059551 provides various amino acid substitutions in CHI and / or CK that promote the preferred pairing between the desired heavy chain and the desired light chain.

[0020] Despite the above various bispecific antibody expression technologies, there is still a need in the art for improved purity of the bispecific antibody product, as well as for a scalable production solution for producing correctly assembled bispecific antibodies.

[0021] Disclosure of the essence of the invention

[0022] The new format of bivalent bispecific antibodies developed by the authors of the present invention surprisingly allows for producing a high yield of a product with correctly assembled two distinct heavy chains co-expressed in the cell and with correctly paired two distinct light chains and the respective heavy chains. This format of bivalent bispecific antibodies comprises a first light and a first heavy chain that specifically bind to a first antigen, and a second light and a second heavy chain that specifically bind to a second antigen, wherein the variable domains VL and VH are substituted for one another, and wherein the VL and VH have a modification at the C-terminus of FR4. Also, we developed a technology for producing these bispecific antibodies.

[0023] Consequently, the above results reduce production costs and lead to a scalable production solution for producing correctly assembled bispecific antibodies.

[0024] General definitions and general methods

[0025] Unless defined otherwise herein, all technical and scientific terms used in connection with the present invention will have the same meaning as is commonly understood by those skilled in the art. Furthermore, unless otherwise required by context, singular terms shall include plural terms, and the plural terms shall include the singular terms. Typically, the present classification and methods of cell culture, molecular biology, immunology, microbiology, genetics, analytical chemistry, organic synthesis chemistry, medical and pharmaceutical chemistry, as well as hybridization and chemistry of protein and nucleic acids described herein are well known by those skilled and widely used in the art. Enzyme reactions and purification methods are performed according to the manufacturer's guidelines, as is common in the art, or as described herein.

[0026] The term "KD" in this description refers to the affinity constant (or equilibrium constant) which is calculated from the ratio of Kd to Ka (i.e. Kd / Ka), and it is expressed as a molar concentration (M).

[0027] "Binding affinity" typically refers to the strength of the sum total of noncovalent interactions between a single binding site of a molecule (e.g. an antibody) and its binding partner (e.g. an antigen). Unless indicated otherwise, "binding affinity" refers to intrinsic (characteristic, true) binding affinity which reflects a 1 : 1 interaction between members of a binding pair (e.g. antibody and antigen). Affinity of a molecule X for its binding partner Y can typically be represented by the affinity constant (KD). The preferred Kd value is about 200 nM, 150 nM, 100 nM, 60 nM, 50 nM, 40 nM, 30 nM, 20 nM, 10 nM, 8 nM, 6 nM, 4 nM, 2 nM, 1 nM, or less. Affinity can be measured by common methods known in the art, including those described in the present description. Low-affinity antibodies typically bind an antigen slowly and tend to dissociate readily, whereas high-affinity antibodies typically bind an antigen faster and tend to remain bound longer. A variety of methods for measuring binding affinity are known in the art, any one of these methods may be used for the purposes of the present invention.

[0028] The term "Kd", "koff" or "kdis" refers to the off rate constant of a particular interaction between a binding molecule and antigen. The off rate constant koff can be measured using biolayer interferometry, for example, using the Octet™ system.

[0029] The term "Ka", "kon" or "on-rate" refers to the association rate constant.

[0030] As used in the present description and claims that follow, unless otherwise dictated by the context, the words "include" and "comprise", or variations thereof such as "includes", "including", "comprises", or "comprising", will be understood to imply the inclusion of a stated integer or group of integers but not the exclusion of any other integer or group of integers.

[0031] Bivalent bispecific antibody

[0032] The present invention relates to a bivalent bispecific antibody.

[0033] The antibody according to the invention is a monoclonal antibody.

[0034] The term "monoclonal antibody" or "mAb" refers to an antibody that is synthesized and isolated as an individual clonal population of cells. The antibody of the invention is a recombinant antibody.

[0035] The term "recombinant antibody" refers to an antibody that is expressed in a cell or cell line comprising nucleotide sequence(s) encoding an antibody, wherein said nucleotide sequence(s) is (are) not associated with the cell in nature.

[0036] The bivalent bispecific antibody according to the invention is an isolated antibody.

[0037] The term "isolated" used to describe various antibodies according to this description refers to an antibody which has been identified and isolated and / or regenerated from a cell or cell culture, in which the antibody is expressed. Impurities (contaminant components) from natural environment are materials which typically interfere with diagnostic or therapeutic uses of the polypeptide, and may include enzymes, hormones, and other proteinaceous or nonproteinaceous solutes. The isolated polypeptide is typically prepared by at least one purification step.

[0038] In one aspect, the present invention relates to a bivalent bispecific antibody that specifically binds to a first antigen and a second antigen, wherein said antibody comprises: a) a first light chain and a first heavy chain of the antibody, which specifically bind to a first antigen, wherein the first light chain comprises a light chain variable domain VL and a light chain constant domain, and wherein the first heavy chain comprises a heavy chain variable domain VH and heavy chain constant domains of antibody that include a first heavy chain constant domain CHI and an Fc fragment monomer comprising second CH2 and third CH3 heavy chain constant domains; and b) a second light chain and a second heavy chain of antibody specifically binding to a second antigen, wherein the second light chain comprises:

[0039] 1) a variable domain VH having a 4-amino acid deletion at the C-terminus of FR4 of the variable domain VH (VH-4),

[0040] 2) an amino acid sequence with EIK (Glu-Ile-Lys), and

[0041] 3) a light chain constant domain; and wherein the second heavy chain comprises:

[0042] 1) a variable domain VL having a deletion of 3 amino acids at the C-terminus of FR4 of the variable domain VL (VL-3),

[0043] 2) an amino acid sequence with SEQ ID NO: 1 (TVSS), and

[0044] 3) antibody heavy chain constant domains comprising a first heavy chain constant domain CHI and an Fc fragment monomer comprising second CH2 and third CH3 heavy chain constant domains; In some embodiments of the invention, the bivalent bispecific antibody comprises a second light chain having the following structure: VH'4- EIK - CL.

[0045] In some embodiments of the invention, the bivalent bispecific antibody comprises a second heavy chain having the following structure: VL'3- SEQ ID NO: 1 - CH1-CH2-CH3.

[0046] The term "antibody" or "immunoglobulin" (Ig) as used in the present description includes whole antibodies. The term "antibody" refers to a glycoprotein comprising at least two heavy (H) chains and two light (L) chains interconnected by disulfide bonds, or antigen-binding portion thereof. Each heavy chain comprises a heavy chain variable region (abbreviated referred to in the present description as VH) and a heavy chain constant region. Known are five types of mammalian antibody heavy chains denoted by Greek letters: a, 5, a, y and p. (Janeway C.A., Jr. et al, Immunobiology, 5th ed., publ. by Garland Publishing, 2001). The type of a heavy chain present defines the class of an antibody; these chains are found in IgA, IgD, IgE, IgG, and IgM antibodies, respectively. (Rhoades R.A., Pflanzer R.G., Human Physiology, 4th ed., publ. by Thomson Learning, 2002). Distinct heavy chains differ in size and composition; a and y contain approximately 450 amino acids, while p and a have approximately 550 amino acids. The constant region is identical in all antibodies of the same isotype, but differs in antibodies of different isotypes. Heavy chains y, a and 5 have a constant region composed of three constant domains CHI, CH2 and CH3 (in a line), and a hinge region for added flexibility (Woof J., Burton D., Nat Rev Immunol 4, 2004, cc.89-99); heavy chains p and a have a constant region composed of four constant domains CHI, CH2, CH3 and CH4 (Janeway C.A., Jr. et al., Immunobiology, 5th ed., Garland Publishing, 2001). In mammals, known are only two types of light chains denoted by lambda (X) and kappa (K). Each light chain consists of a light chain variable region (abbreviated referred to in the present description as VL) and light chain constant region. The approximate length of a light chain is 211 to 217 amino acids. Preferably the light chain is a kappa (K) light chain, and the constant domain CL is preferably C kappa (K).

[0047] VL and VH regions may be further subdivided into hyper-variability regions called complementarity determining regions (CDRs), located between regions that are more conserved, termed framework regions (FRs). Framework regions are responsible for acting as a scaffold for CDRs and aid in maintaining the overall structure of the four variable domains on the antibody. FRs are highly conserved regions of the variable portion of the antibody because they ensure proper folding of the antibody, which helps stabilize CDRs. Each VH and VL is composed of three CDRs and four FRs, arranged from amino-terminus to carboxy-terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. The variable regions of heavy and light chains contain a binding domain that interacts with an antigen. The constant regions of antibodies may mediate the binding of immunoglobulin to host tissues or factors, including various cells of the immune system (e.g. effector cells) and the first component (Clq) of the classical complement system.

[0048] The term "antigen-binding portion" of an antibody or "antigen-binding fragment" (or simply "antibody portion" or "antibody fragment"), as used in this description, refers to one or more fragments of an antibody that retain the capability of specific binding to an antigen. An example of a binding fragment included within the term "antigen-binding portion" of the antibody is a Fab fragment, i.e., a monovalent fragment consisting of VL, VH, CL, and CHI domains.

[0049] "Kabat numbering scheme" or "numbering according to Kabat" as used in the present application refers to the system for numbering of amino acid residues that are more variable (i.e. hypervariable) than other amino acid residues in variable regions of heavy and light chains of antibody (Kabat et al. Arm. N.Y. Acad. Sci., 190:382-93 (1971); Kabat etal. Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242(1991)).

[0050] The antibody of the present invention "which binds" a target antigen refers to an antibody that binds the antigen with sufficient affinity such that the antibody can be used as a diagnostic and / or therapeutic agent targeting a protein or cell or tissue expressing the antigen, and slightly cross-reacts with other proteins. According to analytical methods: fluorescence-activated cell sorting (FACS), radioimmunoassay (RIA) or ELISA, in such embodiments, the degree of antibody binding to a non-target protein is less than 10 % of antibody binding to a specific target protein. With regard to the binding of antibody to a target molecule, the term “specific binding” or phrases “specifically binds to” or “is specific for” a particular polypeptide or an epitope on a particular target polypeptide means binding that is significantly (measurably) different from a non-specific interaction.

[0051] Specific binding may be measured, for example, by determining binding of a molecule as compared to binding of a control molecule. For example, specific binding may be determined by competition with another molecule that is similar to the target, for example, an excess of nonlabeled target. In this case, specific binding is indicated if the binding of the labeled target to a probe is competitively inhibited by the excess of unlabeled target. As used in the present description, the term "specific binding" or phrases "specifically binds to" or "is specific for" a particular polypeptide or an epitope on a particular target polypeptide may be described by example of a molecule having a Kd for the target of at least about 200 nM, or at least about 150 nM, or at least about 100 nM, or at least about 60 nM, or at least about 50 nM, or at least about 40 nM, or at least about 30 nM, or at least about 20 nM, or at least about 10 nM, or at least about 8 nM, or at least about 6 nM, or at least about 4 nM, or at least about 2 nM, or at least about 1 nM, or greater. In one embodiment, the term "specific binding" refers to binding where a molecule binds to a particular polypeptide or epitope on a particular polypeptide without substantially binding to any other polypeptide or epitope on a polypeptide.

[0052] The term "bispecific antibody" refers to an antibody having antigen-binding domains that are capable of specific binding to two distinct epitopes on a single biological molecule or capable of specific binding to epitopes on two distinct biological molecules. The bispecific antibody is also referred to herein as having "dual specificity" or as being a "dual specificity" antibody.

[0053] The fragment crystallizable region ("Fc region, Fc") of an immunoglobulin is the terminal region of an immunoglobulin molecule that interacts with cell surface Fc receptor as well as with some proteins of the complement system. This property allows antibodies to activate the immune system. In IgG, IgA and IgD isotypes, the Fc region is composed of two identical protein fragments from the second and third constant domains of the two heavy chains, respectively; in IgM and IgE isotypes, the Fc contains three heavy chain constant domains (CH2, CH3, and CH4 domains) in each polypeptide chain.

[0054] The "Fc fragment monomer" refers to the Fc region from the second and third constant domains of any one of the two heavy chains.

[0055] In some embodiments of the invention, the bivalent bispecific antibody comprises an antibody light chain constant domain selected from a kappa light chain constant domain (CK) or a lambda light chain constant domain (CL).

[0056] In mammals, known are only two types of light chains denoted by lambda (X) and kappa (K). The constant domain of the lambda light chain is designated CL, and that of the kappa light chain is designated CK.

[0057] In some embodiments of the invention, the bivalent bispecific antibody comprises a first antigen which is a cancer antigen.

[0058] In some embodiments of the invention, the bivalent bispecific antibody comprises a first cancer antigen selected from the group comprising BCMA or GD2.

[0059] The B cell maturation antigen (BCMA, TNFRSF17 and CD269) is a member of the tumor necrosis factor (TNF) receptor superfamily.

[0060] GD2 refers to ganglioside GD2.

[0061] In some embodiments of the invention, the bivalent bispecific antibody comprises a second antigen which is a T cell receptor.

[0062] In some embodiments of the invention, the bivalent bispecific antibody comprises a T cell receptor which is CD3.

[0063] CD3 (cluster of differentiation 3) is a molecule present on the surface of T lymphocytes.

[0064] In some embodiments of the invention, the bivalent bispecific antibody comprises an Fc fragment that belongs to human IgG. In some embodiments of the invention, the bivalent bispecific antibody comprises an Fc fragment isotype which is human IgGl.

[0065] In some embodiments of the invention, the bivalent bispecific antibody comprises mutations L234A and L235A according to the EU numbering scheme of amino acids of antibodies in the CH2 region.

[0066] In some embodiments of the invention, the bivalent bispecific antibody comprises the mutations M252Y, S254T, T256E according to the EU numbering scheme for amino acids of antibodies in the CH2 region.

[0067] In some embodiments of the invention, the bivalent bispecific antibody comprises deletions 446G and 447K, according to the EU numbering scheme of amino acids of antibodies, in the CH3 region.

[0068] In some embodiments of the invention, the bivalent bispecific antibody comprises a third constant domain CH3 of one heavy chain and a third constant domain of other heavy chain that contact one another via surfaces that are modified to form the bivalent bispecific antibody, wherein these modifications in the third constant domains of heavy chains are substitutions to provide for heterodimerization.

[0069] In some embodiments of the invention, the bivalent bispecific antibody comprises CH3 domains of antibody that are further modified by introduction of cysteine as an amino acid into the corresponding positions of each CH3 domain so that a disulfide bridge may form between the both CH3 domains.

[0070] In some embodiments of the invention, the bivalent bispecific antibody comprises a CH3 domain of one heavy chain, which is modified to form Knob, and the CH3 domain of the other heavy chain is modified to form Hole, or vice versa.

[0071] "Knobs-into-holes" (interactions of the "knobs -into-holes" type) is an approach that enables to circumvent the problem associated with mispaired byproducts. This approach aims at forcing the pairing of two different antibody heavy chains by introducing mutations into the CH3 domains to modify the contact interfaces. On one chain, bulky amino acids were replaced by amino acids with short side chains to create a "hole". Conversely, amino acids with larger side chains were introduced into the other CH3 domain to create a "knob". Co-expression of these two heavy chains produced a high yield of the heterodimer formation ("knob-hole") relative to the homodimer formation ("hole-hole" or "knob-knob") (WO9627011 and WO9850431, as well as Merchant AM ET ALL., An efficient route to human bispecific IgG, Nat Biotechnol. 1998 Jul;16(7):677-81).

[0072] In some embodiments of the invention, the bivalent bispecific antibody comprises a CH3 domain of one heavy chain, which has amino acid substitutions S354C / T366W, and the CH3 domain of the other heavy chain has amino acid substitutions Y349C / T366S / L368A / Y407V. In some embodiments of the invention, the bivalent bispecific antibody includes a CH3 domain of one heavy chain, which has amino acid substitutions Y349C / T366S / L368A / Y407, and the CH3 domain of another heavy chain has amino acid substitutions S354C / T366W.

[0073] In some embodiments of the invention, the bivalent bispecific antibody comprises a first antigen which is BCMA and a second antigen which is CD3.

[0074] In some embodiments of the invention, the bivalent bispecific antibody comprises a first antigen which is GD2 and a second antigen which is CD3.

[0075] The application materials provide the following antibodies: 12-001, 12-002, 12-003, 12- 004, 12-005, 12-006, 13-001, 13-002, 13-003, 13-004, 13-005, 13-006, 02-002, 03-003, which are described in Table 1.

[0076] These antibodies are given for illustrative purposes to confirm the operability of the bivalent bispecific antibody format according to the invention, as well as to confirm surprising properties thereof. These antibodies should not be construed as somehow limiting the bivalent bispecific antibody according to the invention.

[0077] The yield parameters of the product with the correct assembly of two distinct heavy chains and the correct pairing between two distinct light chains and the corresponding heavy chains do not depend on the heavy and light chain variable fragments of the bispecific antibody and specificity thereof for antigens.

[0078] The bivalent bispecific antibodies of the invention may be used to treat a variety of diseases, in particular oncological diseases.

[0079] Table 1 - Characteristics of antibodies

[0080]

[0081]

[0082] Method for producing antibody

[0083] In one aspect, the present invention relates to a method for producing any of the above bivalent bispecific antibodies that includes the steps of: a) transforming a host cell

[0084] - with expression vectors comprising nucleic acid molecules encoding the first light chain and the first heavy chain of the bispecific antibody,

[0085] - with expression vectors comprising nucleic acid molecules encoding the second light chain and the second heavy chain of the bispecific antibody, b) culturing the host cell under conditions suitable for synthesis of said bivalent bispecific antibody; and c) isolating said bivalent bispecific antibody from cell culture.

[0086] The term "vector" as used herein means a nucleic acid molecule capable of transporting another nucleic acid to which it has been linked. In some embodiments of the invention, the vector is a plasmid, i.e. a circular double stranded piece of DNA into which additional DNA segments may be ligated. In some embodiments of the invention, the vector is a viral vector, wherein additional DNA segments may be ligated into the viral genome. In some embodiments of the invention, vectors are capable of autonomous replication in a host cell into which they are introduced (e.g. bacterial vectors having a bacterial site of replication origin and episomal mammalian vectors). In further embodiments of the invention, vectors (e.g. non-episomal mammalian vectors) may be integrated into the genome of a host cell upon introduction into a host cell, and thereby are replicated along with the host gene. Moreover, certain vectors are capable of directing the expression of genes to which they are operably linked. Such vectors are referred to herein as "recombinant expression vectors" (or simply, "expression vectors").

[0087] The term "recombinant host cell" (or simply "host cell") as used herein refers to a cell into which a recombinant expression vector has been introduced. It should be understood that "recombinant host cell" and "host cell" refer not only to a particular subject cell but to the progeny of such cell as well. Since modifications may occur in succeeding generations due to either mutation or environmental influences, such progeny may not, in fact, be identical to a parental cell; however, such cells are still included within the scope of the term "host cell" as used herein.

[0088] Transformation may be carried out by any known technique of introducing polynucleotides into a host cell. Methods for introducing heterologous polynucleotides into mammalian cells are well known in the art and include dextran-mediated transfection, cationic polymer-nucleic acid complex transfection, calcium phosphate precipitation, polybrene-mediated transfection, protoplast fusion, encapsulation of the polynucleotides in liposomes, and direct microinjection of DNA into nuclei. In addition, the nucleic acid molecules may be introduced into mammalian cells by viral vectors.

[0089] Mammalian cell lines used as hosts for transformation are well known in the art and include a plurality of immortalized cell lines available. These include, e.g., Chinese hamster ovary (CHO) cells, NSO cells, SP2 cells, HEK-293T cells, FreeStyle 293 cells (Invitrogen), NIH-3T3 cells, HeLa cells, baby hamster kidney (BHK) cells, African green monkey kidney cells (COS), human hepatocellular carcinoma cells (e.g., Hep G2), A549 cells, and a number of other cell lines. Cell lines are selected by way of determining which cell lines have high expression levels and provide for necessary characteristics of the protein being produced. Other cell lines that may be used are insect cell lines, such as Sf9 or Sf21 cells. When the recombinant expression vectors encoding the above bispecific antibody or a portion thereof are introduced into mammalian host cells, the above bispecific antibody is produced by culturing the host cells for a period of time sufficient to express the above bispecific antibody or a portion thereof according to the invention in the host cells, or, more preferably, secrete the above bispecific antibody into the culture medium in which the host cells are cultured. The above bispecific antibody may be isolated from culture medium using standard protein purification techniques. Plant host cells include e.g. Nicotiana, Arabidopsis, duckweed, com, wheat, potato, etc. Bacterial host cells include Escherichia and Streptomyces species. Yeast host cells include Schizosaccharomyces pombe, Saccharomyces cerevisiae and Pichia pastoris.

[0090] Furthermore, level of production of the bispecific antibody of the invention from a producing cell line may be enhanced using a number of known techniques. For example, the glutamine synthetase gene expression system (the GS system) is a common approach for enhancing expression under certain conditions. The GS system is discussed in whole or part in connection with EP Nos. 0216846, 0256055, 0323997 and 0338841.

[0091] It is likely that the bispecific antibody of the invention in different cell lines or host cells will have different glycosylation patterns from each other. However, the bispecific antibody disclosed herein is part of this invention, regardless of the state of glycosylation of the binding molecules and, in general, regardless of the presence or absence of post-translational modifications.

[0092] The above host cell does not relate to a host cell produced using human embryos.

[0093] The above host cell does not relate to a host cell produced by modifying the genetic integrity of human germline cells.

[0094] The present invention relates to methods for producing the bivalent bispecific antibodies according to the present invention. One embodiment of the invention relates to a method for producing bivalent bispecific antibodies as defined herein, comprising producing a recombinant host cell capable of expressing the bivalent bispecific antibody, culturing said host cells under conditions suitable for expression of the bivalent bispecific antibodies, and isolating the resulting bivalent bispecific antibodies. The bivalent bispecific antibody produced by such expression in such recombinant host cells is referred to herein as "bivalent bispecific antibody".

[0095] Brief description of drawings

[0096] Figure l is a schematic representation of the structure of the molecule of the bispecific anti- GD2 / anti-CD3 antibody in the classic knob-into-hole antibody format, with the following designations:

[0097] VH - heavy chain variable domain;

[0098] VL - light chain variable domain;

[0099] CHI - first heavy chain constant domain;

[0100] CH2 - second heavy chain constant domain;

[0101] CH3 - third heavy chain constant domain;

[0102] 1 - position of knob-into-hole mutations S354C and T366W, as well as Y349C, T366S, L368A and Y407V - in the third constant domain of the heavy chain.

[0103] Figure 2 is a schematic representation of the structure of the molecule of the bispecific anti- BCMA / anti-CD3 antibody, in the anti-CD3 portion of which the variable domains VL and VH are substituted for one another, wherein the VL and VH have a modification at the C-terminus of FR4, with the following designations:

[0104] VH'4- heavy chain variable domain with a 4-amino acid deletion at the C-terminus of FR4 of the variable domain;

[0105] VL'3- light chain variable domain with a 3 -amino acid deletion at the C-terminus of FR4 of the variable domain;

[0106] CHI - first heavy chain constant domain;

[0107] CH2 - second heavy chain constant domain;

[0108] CH3 - third heavy chain constant domain;

[0109] 1 - position of knob-into-hole mutations S354C and T366W, as well as Y349C, T366S, L368A and Y407V - in the third constant domain of the heavy chain;

[0110] 2 - additional 4 terminal amino acids (TVSS) at the C-terminus of FR4 of the VL'3variable domain;

[0111] 3 - additional 3 terminal amino acids (EIK) at the C-terminus of FR4 of the VH'4variable domain.

[0112] Figure 3 is a schematic representation of the structure of the molecule of the bispecific anti- GD2 / anti-CD3 antibody, in the anti-CD3 portion of which the variable domains VL and VH are substituted for one another, wherein the substitution of VL and VH takes place incompletely, with the following designations:

[0113] VH - heavy chain variable domain;

[0114] VL - light chain variable domain;

[0115] CHI - first heavy chain constant domain;

[0116] CH2 - second heavy chain constant domain;

[0117] CH3 - third heavy chain constant domain;

[0118] 1 - position of knob-into-hole mutations S354C and T366W, as well as Y349C, T366S, L368A and Y407V - in the third constant domain of the heavy chain;

[0119] 2 - additional 4 terminal amino acids (TVSS) at the C-terminus of FR4 of the VL'3variable domain;

[0120] 3 - additional 3 terminal amino acids (EIK) at the C-terminus of FR4 of the VH'4variable domain.

[0121] Figure 4 is a schematic representation of the structure of the molecule of a bispecific anti- BCMA / anti-CD3 antibody, in the anti-BCMA portion of which the variable domains VL and VH are substituted by one another, with the following designations:

[0122] VH - heavy chain variable domain;

[0123] VL - light chain variable domain;

[0124] CHI - first heavy chain constant domain;

[0125] CH2 - second heavy chain constant domain;

[0126] CH3 - third heavy chain constant domain;

[0127] 1 - position of knob-into-hole mutations S354C and T366W, as well as Y349C, T366S, L368A and Y407V - in the third constant domain of the heavy chain.

[0128] Figure 5 is a schematic representation of the structure of the molecule of the bispecific anti- BCMA / anti-CD3 antibody, in the anti-CD3 portion of which the Fab domains VL-CK and VH- CH1 are substituted by one another, with the following designations:

[0129] VH - heavy chain variable domain;

[0130] VL - light chain variable domain;

[0131] CHI - first heavy chain constant domain;

[0132] CH2 - second heavy chain constant domain;

[0133] CH3 - third heavy chain constant domain;

[0134] 1 - position of knob-into-hole mutations S354C and T366W, as well as Y349C, T366S, L368A and Y407V - in the third constant domain of the heavy chain. Figure 6 is a schematic representation of the structure of the molecule of the bispecific anti- BCMA / anti-CD3 antibody, in the anti-BCMA portion of which the Fab domains VL-CK and VH- CH1 are substituted by one another, with the following designations:

[0135] VH - heavy chain variable domain;

[0136] VL - light chain variable domain;

[0137] CHI - first heavy chain constant domain;

[0138] CH2 - second heavy chain constant domain;

[0139] CH3 - third heavy chain constant domain;

[0140] 1 - position of knob-into-hole mutations S354C and T366W, as well as Y349C, T366S, L368A and Y407V - in the third constant domain of the heavy chain.

[0141] Figure 7 is a schematic representation of the structure of the molecule of the bispecific anti- BCMA / anti-CD3 antibody, in the anti-CD3 portion of which the constant domains CK and CHI are substituted by one another, with the following designations:

[0142] VH - heavy chain variable domain;

[0143] VL - light chain variable domain;

[0144] CHI - first heavy chain constant domain;

[0145] CH2 - second heavy chain constant domain;

[0146] CH3 - third heavy chain constant domain;

[0147] 1 - position of knob-into-hole mutations S354C and T366W, as well as Y349C, T366S, L368A and Y407V - in the third constant domain of the heavy chain.

[0148] Figure 8 is a schematic representation of the structure of the molecule of the bispecific anti- BCMA / anti-CD3 antibody, in the anti-BCMA portion of which the constant domains CK and CHI are substituted by one another, with the following designations:

[0149] VH - heavy chain variable domain;

[0150] VL - light chain variable domain;

[0151] CHI - first heavy chain constant domain;

[0152] CH2 - second heavy chain constant domain;

[0153] CH3 - third heavy chain constant domain;

[0154] 1 - position of knob-into-hole mutations S354C and T366W, as well as Y349C, T366S, L368A and Y407V - in the third constant domain of the heavy chain.

[0155] Figure 9 is a gradient polyacrylamide gel electrophoregram of the bispecific antibody that specifically binds to CD3 and GD2, where:

[0156] 1) Protein molecular weight marker;

[0157] 2) 02-002 2 pg under reducing conditions;

[0158] 3) 02-002 2 pg under non-reducing conditions. Examples

[0159] The following examples are provided for better understanding of the invention. These examples are for purposes of illustration only and are not to be construed as limiting the scope of the invention in any manner.

[0160] All publications, patents, and patent applications cited in this specification are incorporated herein by reference. Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended embodiments.

[0161] Materials and general methods

[0162] General information regarding the nucleotide sequences of human immunoglobulin light and heavy chains is given in: Kabat, E.A., et al., Sequences of Proteins of Immunological Interest, 5th ed., Public Health Service, National Institutes of Health, Bethesda, MD (1991). Amino acids of antibody chains are numbered according to the EU numbering scheme (Edelman, G.M., et al., Proc. Natl. Acad. Sci. USA 63 (1969) 78-85; Kabat, E.A., et al., Sequences of Proteins of Immunological Interest, 5th ed., Public Health Service, National Institutes of Health, Bethesda, MD, (1991).

[0163] Recombinant DNA techniques

[0164] Standard methods were used to manipulate DNA as described in Sambrook, J. et al, Molecular cloning: A laboratory manual; Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1989. The molecular biological reagents were used according to the manufacturer protocols.

[0165] Gene synthesis

[0166] Desired gene segments were prepared from oligonucleotides made by chemical synthesis. The gene segments of 300-1400 bp long, which were flanked by singular restriction sites, were assembled by annealing and ligation of oligonucleotides including PCR amplification and subsequently cloned via the restriction sites. The DNA sequences of the subcloned gene fragments were confirmed by DNA sequencing.

[0167] DNA sequence determination

[0168] DNA sequences were determined by Sanger sequencing.

[0169] DNA and protein sequence analysis and sequence data management

[0170] The Unipro's UGENE suite version 1.29 and SnapGene Viewer were used for sequence creation, mapping, analysis, annotation and illustration.

[0171] Expression vectors For the expression of the antibodies described in the application materials, variants of expression plasmids intended for expression of antibodies in prokaryotic cells (E.coli), transient expression in eukaryotic cells (e.g., in CHO cells) were applied. Beside the antibody expression cassette the vectors contained: an origin of replication which allows replication of said plasmid in E. coli, genes which confer resistance in E. coli to various antibiotics (e.g. to ampicillin, kanamycin).

[0172] The fusion genes comprising the described antibody chains as described below were generated by PCR and / or gene synthesis and assembled with known recombinant methods and techniques by connection of the according nucleic acid segments, e.g. using unique restriction sites in the corresponding vectors. The subcloned nucleic acid sequences were verified by DNA sequencing. For transient transfections, larger quantities of the plasmids were prepared by plasmid preparation from transformed E. coli cultures.

[0173] Example 1. Design and preparation of genetic constructs of bispecific antibodies that specifically bind to CD3 and tumor BCMA

[0174] To produce the bispecific antibodies, developed was a variety of construct variants where the regions were substituted.

[0175] Variants: 12-001, 12-002, 12-003, 12-004, 12-005, 12-006, 13-001, 13-002, 13-003, 13- 004, 13-005, 13-006.

[0176] The candidates 12-001, 12-003, 13-001, 13-004 use VH-CHl-to-VL-CK Fab-chain substitution format, whereas in the candidates 12-001 and 13-001, the substitution takes place for anti-CD3 chains (Figure 5), and in the candidates 12-003 and 13-004, the substitution takes place for anti-BCMA chains (Figure 6).

[0177] The candidates 12-002, 12-004, 13-002, 13-005 use CHl-to-CK substitution format, whereas, in the candidates 12-002 and 13-002, the substitution takes place for anti-CD3 chains (Figure 7), and, in the candidates 12-004 and 13-005, the substitution takes place for anti-BCMA chains (Figure 8).

[0178] The candidates 12-005 and 13-003 (Figure 2) use VL-to-VH substitution format for anti- CD3 chains, in particular the variable domain VL with a 3 -amino acid deletion at the C-terminus of FR4 of the variable domain VL (VL'3) and additional amino acids of SEQ ID NO: 1 (TVSS) at the C-terminus is substituted by the variable domain VH with a 4-amino acid deletion at the C- terminus of FR4 of the variable domain VH (VH'4) and additional amino acids EIK (Glu-Ile-Lys) at the C-terminus; wherein the VL-VH substitution takes place to form the following heavy and light chains:

[0179] VL'3- SEQ ID NO: 1 - CH1-CH2-CH3, where VL'3is a variable domain VL with a 3- amino acid deletion at the C-terminus of FR4 of the variable domain VL, and VH'4- EIK - CL, where VH'4is a variable domain VH with a 4-amino acid deletion at the C-terminus of FR4 of the variable domain VH.

[0180] The candidates 12-006 and 13-006 use the VL-VH substitution format for the anti-BCMA chains (Figure 4).

[0181] For all candidates, the heavy chain constant domains are represented by human IgGl HC, where the CH3 domain of one heavy chain has the amino acid substitutions S354C / T366W and the CH3 domain of the other heavy chain has the amino acid substitutions Y349C / T366S / L368A / Y407V, for proper heterodimerization of the heavy chains of the bispecific antibody.

[0182] The light chain variable domain VL is fused to the light chain constant domain CK of human IgGl .

[0183] The genes for the heavy and light chains of antibodies to BCMA and CD3, and the genes for the dimerization domains were synthesized de novo. The genes for the heavy chain variable domain and the Fc fragment of human IgGl, for the light chain variable domain and CK, or for the variable domains and dimerization domains were fused using PCR and / or gene synthesis and assembly using known restriction-ligation methods.

[0184] The genes of the heavy and light chains of the antibody were cloned into pEE-Hc and pEE-Lc plasmids to generate protein in the IgGl format in mammalian cells.

[0185] Cloned nucleic acid sequences were confirmed by DNA sequencing. The resulting plasmids were produced in desired quantities in E.coli cells and purified using a plasmid DNA isolation kit.

[0186] The resulting gene constructs were transferred for transient production of proteins in CHO cell line.

[0187] Example 2. Production of full-length bispecific antibodies that specifically bind to CD3 and tumor antigen BCMA

[0188] The full-length antibodies 12-005, 12-006, 13-001, 13-002, 13-003, 13-004, 13-005, 13- 006 were produced in cells of an established cell line derived from Chinese hamster ovary cells (CHO-T line) using transient transfection in two replicates. Suspension culturing was conducted in orbital shake bioreactors using serum-free media. For transient expression, cells at a concentration of 2-2.2* 106cells / ml were transfected using linear polyethyleneimine. DNA / PEI ratio was 1 :7. On day 10 of culturing, the cell suspension was centrifuged under 2000 g for 15 min and filtered through 0.22 pm filter.

[0189] Antibodies were purified by affinity chromatography columns using a robotic station. The column was equilibrated with a buffer containing 50 mM NaPB (sodium phosphate buffer), 150 mM NaCl (pH 7.5), the filtered culture liquid with antibodies was applied, the column was then washed with 8 volumes of a buffer containing 50 mM NaPB, 150 mM NaCl and 4 volumes of 50 mM NaPB (pH 7.5). Protein was eluted with 6 column volumes of a solution of 50 mM NaPB, 100 mM NaCl (pH 3). The resulting samples were neutralized with 25 pl of IM phosphate buffer (pH 8).

[0190] Polyacrylamide gel electrophoresis was employed to control the purity of the bispecific antibodies. Electrophoresis was performed in 7.5% polyacrylamide gel under denaturing nonreducing conditions. The protein purity was determined by the intensity of band staining at a protein load of 10 pg per lane.

[0191] Table 2 - Productivity, yield post purification and % content of aggregates, monomers and antibody fragments when analyzed by polyacrylamide gel electrophoresis

[0192] The bispecific antibodies in the novel format according to the invention (12-005 and 13- 003) surprisingly showed the best productivity parameters (>248), yield post purification (>3.4) and high monomer yield (>64.3%) among bispecific antibodies in other test formats. Therefore, these anti-CD3 chains were used to produce other antibodies.

[0193] Example 3. Design and preparation of genetic constructs of bispecific antibodies that specifically bind to CD3 and tumor antigen GD2.

[0194] The chains of the second antigen-binding fragment, which specifically binds to GD2, of any bispecific antibody selected from 02-002 or 03-003 comprise heavy and light chain variable domains fused to heavy and light chain constant domains, respectively.

[0195] The heavy chain constant domains are represented by human IgGl HC, where the CH3 domain of one heavy chain has the amino acid substitutions S354C / T366W and the CH3 domain of the other heavy chain has the amino acid substitutions Y349C / T366S / L368A / Y407V, for proper heterodimerization of the heavy chains of the bispecific antibody. The light chain variable domain VL is fused to the light chain constant domain CK of human IgGl.

[0196] The chains of the first antigen -binding fragment, which specifically binds to CD3, of any bispecific antibody selected from 02-002 or 03-003 comprise heavy and light chain variable domains, whereas the candidate 03-003 also comprises deletions 446G, 447K at the C-terminus of the Fc to reduce antibody heterogeneity.

[0197] Also, the candidates 02-002, 03-003 (Figure 3) use VL-to-VH substitution format for anti- CD3 chains, in particular the variable domain VL with a 3 -amino acid deletion at the C-terminus of FR4 of the variable domain VL (VL-3) and additional amino acids of SEQ ID NO: 1 (TVSS) at the C-terminus is substituted by the variable domain VH with a 4-amino acid deletion at the C- terminus of FR4 of the variable domain VH (VH'4) and additional amino acids EIK (Glu-Ile-Lys) at the C-terminus; wherein the VL-VH substitution takes place to form the following heavy and light chains:

[0198] VL'3- SEQ ID NO: 1 - CH1-CH2-CH3, where VL'3is a variable domain VL with a 3- amino acid deletion at the C-terminus of FR4 of the variable domain VL, and

[0199] VH'4- EIK - CL, where VH'4is a variable domain VH with a 4-amino acid deletion at the C-terminus of FR4 of the variable domain VH.

[0200] The genes for the heavy and light chains of antibodies to GD2 and CD3, and the genes for the dimerization domains were synthesized de novo. The genes for the heavy chain variable domain and the Fc fragment of human IgGl, for the light chain variable domain and CK, or for the variable domains and dimerization domains were fused using PCR and / or gene synthesis and assembly using known restriction-ligation methods.

[0201] The genes of the heavy and light chains of the antibody were cloned into pEE-Hc and pEE- Lc plasmids to generate protein in the IgGl format in mammalian cells.

[0202] Cloned nucleic acid sequences were confirmed by DNA sequencing. The resulting plasmids were produced in desired quantities in E.coli cells and purified using a plasmid DNA isolation kit.

[0203] The resulting genetic constructs, pEE-anti-GD2-Hc, pEE-anti-GD2-Lc, pEE-anti-CD3-Hc and pEE-anti-CD3-Lc, were transferred to transient production of proteins in the CHO cell line.

[0204] Example 4. Production, isolation and purification of bispecific aCD3 / aGD2 antibodies from suspension culture of mammalian cells.

[0205] Full-length bispecific antibodies selected from 02-002 or 03-003 were produced in CHO cell growth medium. Following transfection of cells with expression vectors, orbital feed-batch cultivation was performed in serum-free medium for 7 days. Secretion of the antibodies in question was monitored using the Pall ForteBio's Octet RED96 system for molecular interactions analysis on protein A biosensors.

[0206] Target proteins were isolated from culture liquid by affine HPLC on a chromatography system. The culture liquid was loaded onto a Protein A column, thereafter the column was washed with PBS and the protein was eluted with a solution of 0.1 M glycine buffer pH 3, thereafter the protein solutions were neutralized. The proteins were then transferred to PBS pH 7.4 by dialysis; thereafter, the resulting solutions was filtered off (0.22 pm). The products were stored at -70 °C. The purity of the resulting protein solutions was assessed using SDS gel electrophoresis in a 4- 15% gradient polyacrylamide gel. Figure 9 shows data for the bispecific antibody 02-002 for illustrative purposes only. The bispecific antibody 03-003 showed analogous results.

[0207] Example 5. Interaction affinity between bispecific antibodies and human CD3 antigen (epsilon and delta subunits)

[0208] The interaction between bispecific antibodies and the human CD3 antigen (epsilon and delta subunits) was tested on the ForteBio Octet RED96 instrument using AR2G biosensors (ForteBio). The test used recombinant human CD3 antigen (epsilon and delta subunits). The experiment consisted of the following steps: activating sensors, loading protein onto sensors, quenching unreacted activated groups, recording baselines, recording analyte association, recording dissociation. Measurements were carried out at 30 °C. The sensors were activated in an aqueous solution comprising 20 mM EDC and 10 mM sNHS for 300 s. The antibody 02-002 was loaded onto the surface of biosensors in a 10 mM sodium -acetate buffer solution with pH 5.0 for 1000 s. The concentration of loading protein was 20 pg / ml. Unreacted active centers on the sensor surface were quenched in IM aqueous solution of ethanolamine with pH 8.5 for 300 s (pH value was adjusted by adding hydrochloric acid). The baseline and all subsequent steps of the experiment were carried out in a kinetic buffer solution. At the association step (step duration was 300 s), sensors with loaded protein were immersed into wells containing solutions of antigen at a concentration of 6.25 pg / ml, prepared in a kinetic buffer. At the dissociation step (600 s), the sensors were dipped into wells with a kinetic buffer, where the baseline was recorded.

[0209] Binding curves were processed using the Octet Data Analysis software (Version 8.2) using the 1 : 1 interaction model. Table 3 shows processing results. The bispecific antibodies 02-002 and 13-003 were selected for illustrative purposes only. The bispecific antibody 02-002 has the same complete set of 6 CDRs (LCDR1+LCDR2+LCDR3+HCDR1+HCDR2+HCDR3) of the antigenbinding fragment to CD3 as those ofbispecific antibodies 03-003, 13-003 and 12-005.

[0210] Table 3 - Kinetic constants for interactions between bispecific antibodies 02-002 / 13-003 and human CD3 antigen

[0211] Thus, it can be concluded that all bispecific antibodies specifically bind to the human CD3 antigen.

[0212] Example 6. Determination of affinity of bispecific antibody that specifically binds to CD3 and tumor antigen BCMA

[0213] The antibody 13-003 at a concentration of 10 pg / ml was immobilized on the surface of Protein A biosensors (ForteBio) for 300 s. The analysis was performed at 30 °C using a kinetic buffer solution. After setting the baseline in a kinetic buffer solution (120 s), the sensors with immobilized antibodies were dipped into wells containing the solution of analyte (BCMA), where the association of the complex took place for 300 seconds. For each test antibody, a sensorgram for the solution of antigen (Avi-His-hBCMA-HSA) at concentrations of 5 pg / ml (65.2 nM), 2.5 pg / ml (32.6 nM), 1.25 pg / ml (16.3 nM) and a reference signal (reference sensorgram) of an antigen-free kinetic buffer solution were recorded. The complex dissociation in a buffer solution was then detected for 600 seconds. To check for a nonspecific interaction between the analyte and the sensors (negative control), antibody-free sensors were used. At the loading step, the negative control sensors were dipped into antibody-free sodium acetate buffer, all other steps were analogous to antibody-loaded sensors.

[0214] Binding curves, after subtracting a reference signal, were analyzed using the Octet Data Analysis (Version 9.0) software using 1 : 1 interaction model based on 3 sensorgrams, sensorgrams are shown in Table 4.

[0215] Table 4 - Kinetic constants for antibodies / human BCMA interaction

[0216] Thus, it can be concluded that the tested bispecific antibody specifically binds to the human antigen BCMA.

[0217] Example 7. Determination of thermal stability of bispecific antibodies

[0218] Antibodies were heated in PBS pH 7.4 using an amplifier in plastic test tubes at 50°C for 48 hours, followed by a shift to +4°C. After the end of the program, the samples were analyzed before and following heating using analytical gel chromatography on a TSK Gel G3000 SWxl column. The areas of the target peaks of the samples before and following heating were compared, and the delta change in the monomer peak areas was calculated. The results are shown in Table 5.

[0219] Table 5 - Ratios of peak areas on chromatograms of antibody products before and after heating.

[0220] Thus, all test bispecific antibodies that specifically bind to CD3 and the GD2 antigen have acceptable aggregation stability.

[0221] Example 8. Aggregation stability of bispecific antibody that specifically binds to CD3 and tumor antigen BCMA

[0222] Determination of aggregation stability of monoclonal bispecific antibodies by sizeexclusion (SE) HPLC and dynamic light scattering (DLS).

[0223] The test consisted of the following steps: determination of aggregation temperature (DLS), subjecting of samples to thermal stress at 40 °C for 144 hours, analysis of stressed samples and of control intact samples by (SE) HPLC, particle size analysis of stressed samples and control samples by DLS.

[0224] Table 6 shows the results of determination of mean particle size, of hydrodynamic radius of the target fraction, of poly dispersity of the target fraction and of scattering intensity of the target fraction by DLS for the intact sample (not subjected to stress) and a stressed sample. Table 7 shows the results of SE HPLC determination on an intact (non-stressed) sample and a stressed sample.

[0225] Table 6 - Results of testing antibodies by DLS

[0226] Table 7 - SE HPLC data, peak areas per fragment, target fraction (monomer) anc aggregates

[0227] The tested bispecific antibody that specifically binds to CD3 and the tumor antigen BCMA have acceptable aggregation stability.

[0228] Example 9. Test to determine specific activity of aCD3 / aGD2 antibodies on Jurkat- NFAT-Luc reporter cell line

[0229] This test is carried out using the Jurkat-NFAT-luc reporter cell line, which is created based on the Jurkat cell line comprising the gene encoding firefly luciferase under the control of the NFAT promoter; SK-N-BE(2) cells were used as target cells. Jurkat-NFAT-Luc cells were cultured at 37°C under 5% CO2 on RPML1640 medium (10% FBS, 10 pg / ml gentamicin, 2mM L-glutamine, and 200 pg / ml hygromycin); SK-N-BE(2) cell line was cultured in DMEM / F12 medium (10% FBS, 10 mcg / ml gentamicin and 2mM L-glutamine).

[0230] The test was performed in a white 96-well culture plate designed for luminescence assays. The test antibodies were diluted from 20 pg / ml. 25,000 Jurkat-NFAT-Luc effector cells, 25,000 SK-N-BE(2) target cells, and the test antibodies were added to the wells of the plate. The final volume of cell suspension and antibodies in a well was 100 pl; all suspension components were prepared in RPMI-1640 medium (10% FBS, 10 pg / ml gentamicin and 2 mM L-glutamine).

[0231] The plate was incubated for 4 h at 37°C, 5% CO2, luciferase substrate (BIOKAD) was then added to all wells, and the luminescence intensity in the wells was measured. Luminescence was measured on the Spark plate reader (Tecan), data processing and graphing were performed using Magellan software. The resulting experimental data for luminescence vs the concentration of the test products were approximated using a 4-parameter logistic function. Key evaluation parameters were maximum value of luminescence / activation of the reporter and EC50.

[0232] Thus, it has been shown that antibodies 02-002 and 03-003 induce, with comparable efficiency, reporter activation in the Jurkat-NFAT-Luc cell line in the presence of GD2-expressing cells, Sk-N-BE(2) (neuroblastoma). EC50 values are shown in Table 8.

[0233] Table 8 - ECso

[0234] It has also been shown that when the test used a non-GD2-expressing line, A431 (human lung carcinoma), no reporter activation was observed, which confirms the specific action of these antibodies.

[0235] Example 10. Analysis of ability of the bispecific antibody that specifically binds to CD3 and tumor antigen BCMA to activate reporter T lymphocytic Jurkat NFAT-luc Cl.l line in the presence of RPMI8226 multiple myeloma cells.

[0236] Ability of anti-BCMA / anti-CD3 antibodies to activate T cell signaling by means of binding to the CD3 subunit of the T cell receptor (TCR) in the presence of a target multiple myeloma cell line RPMI8226 was measured by way of detecting a luminescent signal following incubating the antibodies with reporter T lymphocytic human T cells, Jurkat NFAT-luc Cl.l, in the presence of RPMI8226 multiple myeloma cells.

[0237] 20,000 RPMI8226 cells and 20,000 Jurkat NFAT-luc Cl. l cells were incubated in a growth medium with serial dilutions of test antibodies for 24 hours. After the specified period of time, to the cells was added a luciferase detection reagent, incubation was carried out for 8 minutes in the dark, followed by recording of the luminescent signal using a plate reader. Half-effective concentration EC50 was calculated using a four-parameter logistic regression model, the data is shown in Table 9.

[0238] Table 9 - EC50 of anti-BCMA / anti-CD3 antibodies upon activation of Jurkat NFAT-luc

[0239] Cl.l in the presence of multiple myeloma cells RPMI8226

[0240] Thus, the test bispecific antibody that specifically bind to CD3 and the tumor antigen BCMA is capable of activating the Jurkat NFAT-luc Cl. l reporter cell line in the presence of multiple myeloma.

Claims

Claims1. A bivalent bispecific antibody that specifically binds to a first antigen and a second antigen, wherein said antibody comprises: a) a first light chain and a first heavy chain of the antibody, which specifically bind to a first antigen, wherein the first light chain comprises a light chain variable domain VL and a light chain constant domain, and wherein the first heavy chain comprises a heavy chain variable domain VH and heavy chain constant domains of antibody that include a first heavy chain constant domain CHI and an Fc fragment monomer comprising second CH2 and third CH3 heavy chain constant domains; and b) a second light chain and a second heavy chain of antibody specifically binding to a second antigen, wherein the second light chain comprises:1) a variable domain VH having a 4-amino acid deletion at the C-terminus of FR4 of the variable domain VH (VH-4),2) an amino acid sequence with EIK (Glu-Ile-Lys), and3) a light chain constant domain; and wherein the second heavy chain comprises:1 ) a variable domain VL having a deletion of 3 amino acids at the C-terminus of FR4 of the variable domain VL (VL-3),2) an amino acid sequence with SEQ ID NO: 1 (TVSS), and3) antibody heavy chain constant domains comprising a first heavy chain constant domain CHI and an Fc fragment monomer comprising second CH2 and third CH3 heavy chain constant domains;2. The bivalent bispecific antibody according to claim 1, wherein the second light chain has the following structure: VH’4- EIK - CL.

3. The bivalent bispecific antibody according to claim 1, wherein the second heavy chain has the following structure: VL-3- SEQ ID NO: 1 - CH1-CH2-CH3.

4. The bivalent bispecific antibody according to claim 1, wherein the light chain constant domain of the antibody is selected from a kappa light chain constant domain (CK) or lambda light chain constant domain (CL).

5. The bivalent bispecific antibody according to claim 1, wherein the first antigen is a cancer antigen.

6. The bivalent bispecific antibody according to claim 5, wherein the first cancer antigen is selected from the group comprising BCMA or GD2.

7. The bivalent bispecific antibody according to claim 1, wherein the second antigen is a T cell receptor.

8. The bivalent bispecific antibody according to claim 7, wherein the T cell receptor is CD3.

9. The bivalent bispecific antibody according to claim 1, wherein the Fc fragment belongs to human IgG.

10. The bivalent bispecific antibody according to claim 9, wherein the Fc fragment isotype is human IgGl.

11. The bivalent bispecific antibody according to claim 10, wherein the antibody comprises the mutations L234A and L235A according to the EU numbering scheme for amino acids of antibodies in the CH2 region.

12. The bivalent bispecific antibody according to claim 10, wherein the antibody comprises the mutations M252Y, S254T, T256E according to the EU numbering scheme for amino acids of antibodies in the CH2 region.

13. The bivalent bispecific antibody according to claim 10, wherein the antibody comprises the deletion 446G and 447K according to the EU numbering scheme for amino acids of antibodies in the CH3 region.

14. The bivalent bispecific antibody according to claim 1, wherein the third constant domain CH3 of one heavy chain and a third constant domain of other heavy chain contact one another via surfaces that are modified to form the bivalent bispecific antibody, wherein these modifications in the third constant domains of heavy chains are substitutions to provide for heterodimerization.

15. The bivalent bispecific antibody according to claim 14, wherein the CH3 domains of antibody are further modified by introduction of cysteine as an amino acid into the corresponding positions of each CH3 domain so that a disulfide bridge may form between the both CH3 domains.

16. The bivalent bispecific antibody according to claim 15, wherein the CH3 domain of one heavy chain is modified to form Knob, and the CH3 domain of another heavy chain is modified to form Hole, or vice versa.

17. The bivalent bispecific antibody according to claim 16, wherein the CH3 domain of one heavy chain has amino acid substitutions S354C / T366W, and the CH3 domain of another heavy chain has amino acid substitutions Y349C / T366S / L368A / Y407V.

18. The bivalent bispecific antibody according to claim 17, wherein the CH3 domain of one heavy chain has amino acid substitutions Y349C / T366S / L368A / Y407, and the CH3 domain of another heavy chain has amino acid substitutions S354C / T366W.

19. The bivalent bispecific antibody according to claim 1, wherein the first antigen is BCMA and the second antigen is CD3.

20. The bivalent bispecific antibody according to claim 1, wherein the first antigen is GD2 and the second antigen is CD3.

21. A method for producing the bivalent bispecific antibody according to claims 1 to 20, wherein the method comprises the steps of: a) transforming a host cell- with expression vectors comprising nucleic acid molecules encoding the first light chain and the first heavy chain of the bispecific antibody,- with expression vectors comprising nucleic acid molecules encoding the second light chain and the second heavy chain of the bispecific antibody, b) culturing the host cell under conditions suitable for synthesis of said bivalent bispecific antibody; and c) isolating said bivalent bispecific antibody from cell culture.

Images