Polypeptide linkers for preparing multispecific antibodies

Abstract

The present invention relates to mutant polypeptide linkers for preparing multispecific antibodies, to said multispecific antibodies, and to methods for producing said multispecific antibodies.

Description

[Technical field]

[0001] The present invention relates to multispecific antibodies with improved properties that are useful in the medical field.

[0002] 2. Background of the Invention Therapeutic fusion proteins have become an important approach in drug development. Peptide linkers are frequently used to construct multi-domain proteins from different functional protein modules. The resulting multi-domain proteins are designed to bind targets cognate to the individual modules (or simultaneously exert the biological functions of the individual modules), either enhancing the biological effects associated with isolated single domains or providing novel biological activities not accessible by isolated single domains. Numerous examples of molecules using peptide linkers exist: single chain variable domains of antibodies (scFv), immunocytokines (cytokine-antibody fusions), bispecific antibodies (BsAb), etc. The choice of linker(s) for a particular fusion protein is dictated by considerations such as 1) whether the linker(s) need to be flexible in order for the various domains to fold into a particular tertiary structure (e.g., scFv-based antibodies), 2) whether the linker(s) need to be rigid in order to provide the necessary separation between the protein domains, or 3) whether the linker(s) must be cleavable to allow for the separation of the domains to produce the desired activity in vivo (Xiaoying Chen, et al, Adv Drug Deliv Rev. 2013. 65(10): 1357-1369). The choice of linker can be critical, as an inappropriate linker can reduce or eliminate the desired activity of the fusion protein. (Yumi Maeda, et al, Anal. Biochem. 1997. 249(2): 147-152).

[0003] A variety of linker sequences have been identified for use in constructing fusion proteins (Richard George and Jaap Heringa, Protein Engineering. 2003. 15(11): 871-879; Xiaoying Chen et al, Adv Drug Deliv Rev. 2013. 65(10): 1357-1369). There are also a variety of available databases compiling linker sequences for use in constructing fusion proteins: 1) SynLinker, compiled by the National University of Singapore (http: / / synlinker.syncti.org), and 2) International Congress of Synthetic Biology (http: / / parts.igem.org / Protein_domains / Linker); Centre for Integrative Bioinformatics, Vrije Universiteit Amsterdam (http: / / www.ibi.vu.nl / programs / linkerdbwww). International Patent Application WO 2013 / 005194 discloses multispecific antibodies constructed with a linker designed from an IgG1 hinge sequence followed by the N-terminus of the IgG1 CH2 domain sequence followed by an 8 amino acid semi-rigid linker sequence derived from the central portion of the IgA1 hinge. This semi-rigid linker, including the central portion of the IgA1 hinge, is important to separate both Fab domains of the antibody sufficiently to avoid steric hindrance from the C-terminus of the outer Fab1 affecting the antigen-binding paratope of Fab2 close to the inner Fc.

[0004] However, the inventors have demonstrated that the presence of multiple glycoforms does not facilitate the generation of consistent preparations of such multispecific antibodies, which are required for the development of therapeutics. Moreover, the characterization of such preparations would also be highly complex, which would make comparison of batches produced by different methods laborious.

[0005] Summary of the Invention The inventors then realised that this could be improved by redesigning the linker, in particular by eliminating the glycosylation site from the linker.

[0006] Thus, the present invention relates to the amino acid sequence EPKX1CDKX2HX3X4PPX5PAPELLGGPX6X7PPX8PX9PX 10 GG (SEQ ID NO: 1), wherein X1, X2, X3, X4, X5, X6, X7, X8, X9, X 10 are the same or different and are any amino acid as defined herein; with the proviso that the polypeptide does not comprise or consist of the sequence EPKSCDKTHTCPPCPAPELLGGPSTPPTPSPSGG (SEQ ID NO:5) or EPKSCDKTHTSPPSPAPELLGGPSTPPTPSPSGG (SEQ ID NO:6).

[0007] Such polypeptides are useful as linkers in fusion proteins, more particularly in multispecific antibodies, especially bispecific antibodies.

[0008] A subject of the present invention is therefore a multispecific antigen-binding fragment comprising at least two Fab fragments with different CH1 and CL domains, wherein each Fab fragment recognizes a different epitope of interest, said Fab fragments being arranged in tandem in any order, characterized in that the C-terminus of the CH1 domain of a first Fab fragment is linked to the N-terminus of the VH domain of a subsequent Fab fragment through a polypeptide linker, said polypeptide linker sequence comprising or consisting of the amino acid sequence EPKX1CDKX2HX3X4PPX5PAPELLGGPX6X7PPX8PX9PX10GG (SEQ ID NO: 1), wherein X1, X2, X3, X4, X5, X6, X7, X8, X9, X10, X11, X12, X13, X14, X15, X16, X17, X18, X19, X20, X21, X22, X23, X3, X4, X5, X6, X7, X8, X9, X10, X11, X12, X13, X14, X15, X16, X17, X18, X19 ...3, X24, X25, X26, X27, X28, X29, X30, X31, X32, X33, X34, X35, X36, X37, X38, X39, X40, X41, X42, X43, X44, X45, X46, X47, X48, X49, X50, X51, X52, X53, X54, X55, X 10are the same or different and are any amino acids; with the proviso that the linker sequence does not comprise or consist of the sequence EPKSCDKTHTCPPCPAPELLGGPSTPPTPSPSGG (SEQ ID NO:5) or EPKSCDKTHTSPPSPAPELLGGPSTPPTPSPSGG (SEQ ID NO:6).

[0009] Further provided is a multispecific antibody having two identical antigen-binding arms, each consisting of a multispecific antigen-binding fragment as defined herein.

[0010] In a preferred embodiment, - two identical antigen-binding arms, each consisting of a multispecific antigen-binding fragment as defined herein; - dimerized CH2 and CH3 domains of immunoglobulins; - an IgA, IgG, or IgD hinge region linking the C-terminus of the CH1 domain to the N-terminus of the CH2 domain of the antigen-binding arm; In one embodiment, a multispecific antibody having an immunoglobulin-like structure is provided, comprising:

[0011] Thus, the present invention more particularly provides a multispecific antibody, preferably a bispecific antibody, comprising two heavy chains and four light chains, wherein each heavy chain comprises: A. The Fc region of an immunoglobulin, including the hinge-CH2-CH3 domains Including, b. the Fc region is linked to the heavy chain CH1-VH of the Fab of Antibody 1 (Ab1) by the hinge domain; c. which is then linked to the heavy chain CH1-VH of the Fab of antibody 2 (Ab2) by a polypeptide linker sequence, which links the N-terminus of the heavy chain VH domain of the Fab of Ab1 to the C-terminus of the CH1 domain of Ab2; wherein the four light chains include an Ab1 light chain and an Ab2 light chain associated with their cognate heavy chain domains; The polypeptide linker sequence has the amino acid sequence EPKX1CDKX2HX3X4PPX5PAPELLGGPX6X7PPX8PX9PX 10 characterised in that it comprises or consists of GG (SEQ ID NO: 1), Here, X1, X2, X3, X4, X5, X6, X7, X8, X9, X 10 are the same or different and are any amino acids; with the proviso that the linker sequence does not comprise or consist of the sequence EPKSCDKTHTCPPCPAPELLGGPSTPPTPSPSGG (SEQ ID NO:5) or EPKSCDKTHTSPPSPAPELLGGPSTPPTPSPSGG (SEQ ID NO:6).

[0012] In certain embodiments, the polypeptide linker sequence is EPKSCDKTHTSPPAPAPELLGGPGGPPGPGPGGG (SEQ ID NO:2); EPKSCDKTHTSPPAPAPELLGGPAAPPAPAPAGG (SEQ ID NO: 3); and EPKSCDKTHTSPPAPAPELLGGPAAPPGPAPGGG (SEQ ID NO: 4) The nucleic acid comprises or consists of a sequence selected from the group consisting of:

[0013] A further subject of the present invention is a multispecific antigen-binding fragment or a polypeptide comprising, preferably consisting of, the heavy chain of a multispecific antibody, preferably a bispecific antibody, as defined herein.

[0014] The present invention further provides a polynucleotide comprising a sequence encoding such a polypeptide.

[0015] Host cells transfected with an expression vector containing said polynucleotide are also part of the invention.

[0016] A further subject of the present invention is a method for producing a multispecific antibody, preferably a bispecific antibody, as described herein, comprising the following steps: a) culturing, in suitable medium and culture conditions, a host cell expressing an antibody heavy chain as defined herein and an antibody light chain as defined herein; and b) recovering said produced antibody from the culture medium or from the cultured cell. [Brief description of the drawings]

[0017] [Figure 1] FIG. 1 is a diagram of a full-length BiXAb bispecific antibody of the invention. [Diagram 2] FIG. 2 is a diagram of a bispecific construct that does not contain an Fc domain and contains only two Fab domains connected by a linker (Fab-Fab). [Figure 3A] FIG. 3A shows size-exclusion chromatography analysis of BiXAb2b. [Figure 3B] FIG. 3B shows size-exclusion chromatography analysis of BiXAb3b. [Figure 3C] FIG. 3C shows size-exclusion chromatography analysis of Fab-Fab3b. [Figure 4A] FIG. 4A shows the MS spectrum of the LC-MS analysis of Fab-Fab3a. [Figure 4B] FIG. 4B shows the MS spectrum of the LC-MS analysis of Fab-Fab3b. [Figure 5A] FIG. 5A shows the MS spectrum of the LC-MS analysis of BiXAb3a. [Figure 5B] FIG. 5B shows the MS spectrum of the LC-MS analysis of BiXAb3b.

[0018] Detailed Description of the Invention definition The basic structure of a naturally occurring antibody molecule is a Y-shaped tetrameric quaternary structure consisting of two identical heavy chains and two identical light chains held together by non-covalent interactions and by interchain disulfide bonds.

[0019] In mammalian species, there are five types of heavy chains (α, δ, ε, γ, and μ): which determine the immunoglobulin class (isotype) (IgA, IgD, IgE, IgG, and IgM), respectively. In the γ, α, and δ heavy chains, the N-terminal variable domain (VH) of the heavy chain is followed by a constant region containing three domains (numbered from the N-terminus to the C-terminus as CH1, CH2, and CH3), whereas the constant region of the μ and ε heavy chains is composed of four domains (numbered from the N-terminus to the C-terminus as CH1, CH2, CH3, and CH4). The CH1 and CH2 domains of IgA, IgG, and IgD are separated by flexible hinges that vary in length between the different classes and, in the case of IgA and IgG, the different subtypes (IgG1, IgG2, IgG3, and IgG4), with hinges of 15, 12, 62 (or 77), and 12 amino acids, respectively, and IgA1 and IgA2 with hinges of 20 and 7 amino acids, respectively.

[0020] There are two types of light chains, lambda and kappa: they can associate with any of the heavy chain isotypes, but within a given antibody molecule, both are of the same kind. Both light chains appear to be functionally identical. Following their N-terminal variable domain (VL) is a constant region consisting of a single domain called CL.

[0021] The heavy and light chains are paired by protein / protein interactions between the CH1 and CL domains and between the VH and VL domains, and the two heavy chains are associated by protein / protein interactions between their CH3 domains.

[0022] The antigen-binding regions correspond to the arms of the Y-shaped structure, consisting of each intact light chain paired with the VH and CH1 domains of the heavy chain, and are called Fab fragments (for Fragment antigen binding). Fab fragments are first generated from native immunoglobulin molecules by digestion with papain, which cleaves the antibody molecule in the hinge region on the amino-terminal side of the interchain disulfide bond, thereby releasing two identical antigen-binding arms. Other proteases, such as pepsin, also cleave antibody molecules in the hinge region, but on the carboxy-terminal side of the interchain disulfide bond, thereby releasing a fragment consisting of two identical Fab fragments, which remain linked through disulfide bonds; reduction of the disulfide bonds in the F(ab')2 fragment generates the Fab' fragment.

[0023] The part of the antigen-binding region corresponding to the VH and VL domains is called the Fv fragment (for Fragment variable); it contains the CDRs (complementarity determining regions), which form the antigen-binding site (also called the paratope).

[0024] The effector region of an antibody, involved in its binding to effector molecules on immune cells, corresponds to the stem of the Y-shaped structure, which contains the CH2 and CH3 domains of the paired heavy chains (or the CH2, CH3 and CH4 domains, depending on the class of the antibody), and is called the Fc (for Fragment crystallisable) region.

[0025] Because the two heavy chains and the two light chains are identical, a naturally occurring antibody molecule has two identical antigen-binding sites and therefore binds two identical epitopes simultaneously.

[0026] In the context of the present invention, a "multispecific antigen-binding fragment" is defined herein as a molecule having two or more antigen-binding regions, each recognizing a different epitope. The different epitopes may be produced by the same antigenic molecule or by different antigenic molecules. The term "recognizing" or "recognizing" means that the fragment specifically binds to a target antigen.

[0027] An antibody "specifically binds" to a target antigen if it binds with greater affinity, avidity, more readily, and / or for a longer period than it binds to other substances. "Specifically binding" or "preferentially binding" does not necessarily require (although it can include) exclusive binding. Generally, but not necessarily, reference to binding refers to preferential binding.

[0028] The terms "subject," "individual," and "patient" are used interchangeably herein and refer to a mammal being evaluated for and / or treated for treatment. The subject may be a human, but also includes other mammals, particularly mammals useful as experimental models for human disease, such as mice, rats, rabbits, dogs, and the like.

[0029] The term "treatment" or "treating" refers to a measure, application, or therapy, where a subject, including a human, receives medical care for the purpose of directly or indirectly improving the subject's condition. In particular, the term refers in some embodiments to reducing incidence or alleviating symptoms, eliminating recurrence, preventing recurrence, preventing onset, improving symptoms, improving prognosis, or a combination thereof. Those skilled in the art will understand that treatment does not necessarily result in complete abolition or elimination of symptoms. For example, with respect to cancer, "treatment" or "treating" can refer to slowing the growth, proliferation, or metastasis of neoplastic or malignant cells, preventing or delaying the onset of neoplastic or malignant cell growth, proliferation, or metastasis, or some combination thereof.

[0030] Multispecific antibody design Provided herein are multispecific antigen-binding fragment(s), and multispecific antibody constructs comprising said fragments, each multispecific antigen-binding fragment consisting essentially of tandemly arranged Fab fragments separated by a linker of the invention.

[0031] Such fragments and constructs preferably comprise chains derived from a human immunoglobulin, preferably IgG, more preferably IgG1.

[0032] In the case of multispecific antigen-binding fragments comprising more than two different Fab fragments, the polypeptide linkers separating the Fab fragments may be the same or different.

[0033] According to a preferred embodiment of the multispecific antibody of the invention, it has two identical antigen-binding arms, each consisting of a multispecific antigen-binding fragment as defined above. The antigen-binding arms can be linked to each other in various ways, depending on the intended use of the antibody.

[0034] If it is desired to obtain an antibody that has no Fc-mediated effects, the antibody will not contain an Fc region at all. In this case, the two antigen-binding arms can be, for example: - by homodimerization of the antigen-binding arms through interchain disulfide bonds provided by the polypeptide linker(s) separating the Fab fragments; and / or -by adding a polypeptide extension containing a cysteine ​​residue at the C-terminus of each antigen-binding arm, which allows for the formation of an interchain disulfide bond, and resulting in a hinge-like structure upon homodimerization of the polypeptide extensions; as a non-limiting example, the polypeptide extensions can be, for example, the hinge sequence of IgG1, IgG2, or IgG3; through a semi-rigid linker connecting the C-terminus of the heavy chains of the two antigen-binding arms to form a single polypeptide chain and maintaining the antigen-binding arms at a sufficient distance from each other, They can be linked together.

[0035] Alternatively, if effector functions such as CDC inhibition, ADCC inhibition, or ADP are desired, the multispecific antibodies of the invention may further comprise an Fc domain that confers these effector functions. The choice of Fc domain will depend on the type of effector function desired.

[0036] In this case, the multispecific antibody of the invention has an immunoglobulin-like structure comprising: - two identical multispecific antigen-binding arms as defined above; - dimerized immunoglobulin CH2 and CH3 domains; - the hinge region of either IgA, IgG or IgD, linking the C-terminus of the CH1 domain of the antigen-binding arm to the N-terminus of the CH2 domain, or alternatively, if the CH4 domain following the CH3 domain is derived from IgM or IgE, the C-terminus of the CH1 domain of the antigen-binding arm may in this case be linked directly to the N-terminus of the CH2 domain.

[0037] Preferably, the CH2 and CH3 domains, the hinge region and / or the CH4 domain are derived from the same immunoglobulin or from an immunoglobulin of the same isotype and subclass as the CH1 domain of the antigen-binding arm.

[0038] The CH2 domain, CH3 domain, and optionally the CH4 domain, as well as the hinge region, derived from a natural immunoglobulin may be used. They may also be mutated, if desired, for example, to modulate the effector functions of the antibody. In some cases, all or part of the CH2 or CH3 domain may be omitted.

[0039] The present invention more particularly provides bispecific, tetravalent antibodies comprising two binding sites for their respective targets and a functional Fc domain enabling activation of effector functions such as antibody-dependent cellular cytotoxicity (ADCC) and phagocytosis.

[0040] Such preferred antibodies of the invention are full-length antibodies. They preferably comprise heavy and light chains derived from a human immunoglobulin, preferably IgG, more preferably IgG1.

[0041] The light chains may be lambda or kappa light chains; they are preferably kappa light chains.

[0042] In a preferred embodiment, the linker of the invention links IgG Fab domains in a tetravalent Fab bispecific antibody format, the amino acid sequence of which comprises at least two Fab heavy chain sequences connected by the polypeptide linker, co-expressed with an appropriate IgG light chain sequence, followed by a native hinge sequence, followed by an IgG Fc sequence.

[0043] An example of an antibody of the invention, designated BiXAb antibody, having an IgG-like structure is illustrated in Figure 1. The antibodies designated BiXAb2a, BiXAb2b, BiXAb2c, and BiXAb3b described below are specific examples.

[0044] Bispecific antibodies of the invention typically include - the assembled continuous heavy chain of Fc (hinge-CH2-CH3), followed by the heavy chain of the Fab of antibody 1 (CH1-VH) and consecutively the heavy chain of the Fab of antibody 2 (CH1-VH), the latter being connected by a polypeptide linker sequence of the invention; Including, During protein expression, the resulting heavy chains associate into dimers, whereas the co-expressed light chains (VL-CL) of Antibody 1 and Antibody 2 associate with their cognate heavy chains to form the final tandem F(ab)'2-Fc molecule; Antibody 1 (Ab1) and Antibody 2 (Ab2) are different.

[0045] In a preferred embodiment, a) a Fab fragment having the CH1 and C-κ domains derived from human IgG1 / κ and the VH and VL domains of Ab1; b) a Fab fragment having the CH1 and C-κ domains from human IgG1 / κ and the VH and VL domains of Ab2; c) a mutated light chain CL constant domain derived from the human kappa constant domain; d) Mutated heavy chain CH1 constant domain Two Fab fragments having different CH1 and CL domains, wherein the Fab fragments are tandemly arranged in the following order: the C-terminus of the CH1 domain of the Fab fragment of Ab1 linked to the N-terminus of the VH domain of the Fab fragment of Ab2 through a polypeptide linker; a hinge region of human IgG1 linking the C-terminus of the CH1 domain of a fragment of Ab2 to the N-terminus of the CH2 domain, -Dimerized CH2 and CH3 domains of human IgG1.

[0046] Ab1 and Ab2 can be any antibody of interest, in particular any antibody of therapeutic interest. In a particular embodiment, Ab1 and Ab2 are different and are independently selected from the group consisting of anti-EGFR antibodies and anti-HER2 / neu antibodies. In a preferred embodiment, Ab1 and Ab2 are different and are independently selected from the group consisting of cetuximab or a mutant derivative thereof on the one hand and trastuzumab or a mutant derivative thereof on the other hand.

[0047] In another specific embodiment, Ab1 and Ab2 are different and are independently selected from the group consisting of an anti-CD38 antibody and an anti-PD-L1 antibody.

[0048] Such antibodies are useful as pharmaceuticals, more particularly in the treatment of cancer.

[0049] Throughout this specification, amino acid sequences are defined according to Kabat et al, Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md. (1991).

[0050] Another example of a construct of the invention, which is a multispecific antigen-binding fragment Fab-Fab, not containing an Fc domain, is illustrated in Figure 2. A specific example, designated Fab-Fab3a, is described below.

[0051] Such Fab-Fab constructs typically contain two different Fab domains. Such antibodies have only one Fab domain that binds antigen 1 and antigen 2, respectively. They have the same light chain as the corresponding BiXAb antibodies, however the heavy chain of the Fab-Fab is truncated such that its most C-terminal residue is cysteine-220 (EU numbering).

[0052] Linker Design The polypeptide linker sequence according to the invention is the amino acid sequence EPKX1CDKX2HX3X4PPX5PAPELLGGPX6X7PPX8PX9PX 10 GG (SEQ ID NO: 1), wherein X1, X2, X3, X4, X5, X6, X7, X8, X9, X 10 are the same or different and are any amino acids; with the proviso that the linker sequence does not comprise or consist of the sequence EPKSCDKTHTCPPCPAPELLGGPSTPPTPSPSGG (SEQ ID NO:5) or EPKSCDKTHTSPPSPAPELLGGPSTPPTPSPSGG (SEQ ID NO:6).

[0053] The polypeptide linker sequence consists of less than 80 amino acids, preferably less than 60 amino acids, more preferably less than 40 amino acids.

[0054] In certain embodiments, X1, X2, and X3 are the same or different and are threonine (T) or serine (S).

[0055] In another particular embodiment, X1, X2, and X3 are the same or different and are any amino acid other than threonine (T) or serine (S), preferably, X1, X2, and X3 are the same or different and are selected from the group consisting of Ala (A), Gly (G), Val (V), Asn (N), Asp (D), and Ile (I), more preferably, X1, X2, and X3 are the same or different and may be Ala (A) or Gly (G).

[0056] Alternatively, X1, X2, and X3 may be the same or different and may be Leu (L), Glu (E), Gln (Q), Met (M), Lys (K), Arg (R), Phe (F), Tyr (T), His (H), Trp (W), preferably Leu (L), Glu (E), or Gln (Q).

[0057] In certain embodiments, X4 and X5 are the same or different and are any amino acid selected from the group consisting of serine (S), cysteine ​​(C), alanine (A), and glycine (G).

[0058] In a preferred embodiment, X4 is serine (S) or cysteine ​​(C).

[0059] In a preferred embodiment, X5 is alanine (A) or cysteine ​​(C).

[0060] In certain embodiments, X6, X7, X8, X9, X 10 are the same or different and are any amino acid other than threonine (T) or serine (S). 10 are the same or different and are selected from the group consisting of Ala (A), Gly (G), Val (V), Asn (N), Asp (D), and Ile (I).

[0061] Or X6, X7, X8, X9, X 10are the same or different and can be Leu (L), Glu (E), Gln (Q), Met (M), Lys (K), Arg (R), Phe (F), Tyr (T), His (H), Trp (W), preferably Leu (L), Glu (E), or Gln (Q).

[0062] In a preferred embodiment, X6, X7, X8, X9, X 10 are the same or different and are selected from the group consisting of Ala (A) and Gly (G).

[0063] In a further preferred embodiment, X6 and X7 are identical and are preferably selected from the group consisting of Ala (A) and Gly (G).

[0064] In a preferred embodiment, the polypeptide linker sequence comprises or consists of the sequence of SEQ ID NO:1, X1, X2, and X3 are the same or different and are threonine (T), serine (S); X4 is serine (S) or cysteine ​​(C); X5 is alanine (A) or cysteine ​​(C); X6, X7, X8, X9, X 10 are the same or different and are selected from the group consisting of Ala (A) and Gly (G).

[0065] In certain embodiments, the polypeptide linker sequence is EPKSCDKTHTSPPAPAPELLGGPGGPPGPGPGGG (SEQ ID NO:2); EPKSCDKTHTSPPAPAPELLGGPAAPPAPAPAGG (SEQ ID NO: 3); and EPKSCDKTHTSPPAPAPELLGGPAAPPGPAPGGG (SEQ ID NO: 4) The nucleic acid comprises or consists of a sequence selected from the group consisting of:

[0066] In another preferred embodiment, the polypeptide linker sequence comprises or consists of the sequence of SEQ ID NO:1, X1, X2, and X3 are the same or different and are Ala (A) or Gly (G): X4 is serine (S) or cysteine ​​(C); X5 is alanine (A) or cysteine ​​(C); X6, X7, X8, X9, X 10 are the same or different and are selected from the group consisting of Ala (A) and Gly (G).

[0067] Generation of antibodies The nucleic acids encoding the heavy and light chains of the antibody of the invention are inserted into an expression vector. The light and heavy chains can be cloned into the same or different expression vectors. The DNA segments encoding the immunoglobulin chains are operably linked to control sequences in the expression vector(s) that ensure the expression of immunoglobulin polypeptides. Such control sequences include signal sequences, promoters, enhancers, and transcription termination sequences. Expression vectors are typically replicable in the host organisms either as episomes or as an integrated part of the host chromosomal DNA. Generally, expression vectors will contain a selection marker, such as tetracycline or neomycin, to allow detection of cells transformed with the desired DNA sequence.

[0068] In one example, both the heavy and light chain coding sequences (e.g., sequences encoding VH and VL, VH-CH1, or VL-CL) are contained in one expression vector. In another example, each of the heavy and light chains of an antibody is cloned into an individual vector. In the latter case, expression vectors encoding the heavy and light chains may be co-transfected into one host cell for expression of both chains, which can associate to form an intact antibody, either in vivo or in vitro.

[0069] In certain embodiments, a host cell is co-transfected with three independent expression vectors, such as plasmids, which result in the simultaneous production of all three chains (i.e., the heavy chain HC and the two light chains LC1 and LC2, respectively) and secretion of a multispecific antibody.

[0070] More specifically, the three vectors may be advantageously used in the following molecular ratio: 3:2:2 (HC:LC1:LC2).

[0071] The recombinant vector for expressing the antibody described herein typically contains a nucleic acid encoding the amino acid sequence of the antibody operably linked to a promoter, either constitutive or inducible. The vector may be suitable for replication and integration in prokaryotic cells, eukaryotic cells, or both. Typical vectors contain transcription and translation terminators, initiation sequences, and promoters useful for regulating the expression of the nucleic acid encoding the antibody. The vector optionally contains a gene expression cassette that contains at least one independent termination sequence, a sequence that allows the cassette to replicate in both eukaryotes and prokaryotes, i.e., shuttle vectors, and a selection marker for both prokaryotic and eukaryotic systems.

[0072] Multispecific antibodies as described herein can be produced in prokaryotic or eukaryotic expression systems, such as bacteria, yeast, filamentous fungi, insect, and mammalian cells. It is not necessary for the recombinant antibodies of the invention to be glycosylated or expressed in eukaryotic cells, however, expression in mammalian cells is generally preferred. Examples of useful mammalian host cell lines are human embryonic kidney (293 cells), baby hamster kidney cells (BHK cells), Chinese hamster ovary cells / - or +DHFR (CHO, CHO-S, CHO-DG44, Flp-in CHO cells), African green monkey kidney cells (VERO cells), and human liver cells (Hep G2 cells).

[0073] Mammalian tissue cell culture is preferred for expressing and producing the polypeptides, since many suitable host cell lines capable of secreting intact immunoglobulins have been developed in the art, including CHO cell lines, various Cos cell lines, HeLa cells, preferably myeloma cell lines, or transformed B cells or hybridomas.

[0074] In a most preferred embodiment, the multispecific, preferably bispecific, antibodies of the invention are produced by using a CHO cell line, most advantageously a CHO-S cell line.

[0075] Expression vectors for these cells may contain expression control sequences, such as an origin of replication, a promoter, and an enhancer, as well as necessary processing information sites, such as ribosome binding sites, RNA splice sites, polyadenylation sites, and transcription termination sequences. Preferred expression control sequences are promoters derived from immunoglobulin genes, SV40, adenovirus, bovine papilloma virus, cytomegalovirus, and the like.

[0076] Vectors containing the polynucleotide sequences of interest (e.g., sequences encoding heavy and light chains, and expression control sequences) can be introduced into host cells by well-known methods, which vary depending on the type of cellular host. For example, calcium phosphate treatment or electroporation may be used for other cellular hosts (see generally Sambrook et al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Press, 2nd ed., 1989)). When the heavy and light chains are cloned on separate expression vectors, co-transfection of the vectors results in the expression and assembly of intact immunoglobulins.

[0077] Host cells are transformed or transfected with the vector (e.g., by chemical transfection or electroporation) and cultured in a conventional nutrient medium (or modified as appropriate) to induce promoters, select transformants, or amplify genes encoding the desired sequences.

[0078] Once expressed, the complete antibody, its dimer, individual light and heavy chains, or other immunoglobulin forms of the invention can be further isolated or purified to obtain preparations that are substantially homogeneous for further assays and applications. Standard protein purification methods known in the art can be used. For example, suitable purification procedures may include fractionation on immunoaffinity or ion exchange columns, precipitation with ethanol, high performance liquid chromatography (HPLC), sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), precipitation with ammonium sulfate, and gel filtration (see generally, Scopes, Protein Purification (Springer-Verlag, NY, 1982)). Substantially pure immunoglobulins of at least about 90-95% homogeneity are preferred, with 98-99% or more homogeneity being most preferred.

[0079] In vitro production allows for scale-up to obtain large quantities of the desired multispecific, preferably bispecific, antibodies of the invention. Such methods may use homogenous suspension cultures, for example in airlift reactors or continuous stirred reactors, or immobilized or encapsulated cell cultures, for example in hollow fibers, microcapsules, agarose microbeads or ceramic cartridges.

[0080] therapeutic application A further aspect of the invention is a pharmaceutical composition comprising an antibody according to the invention. Another aspect of the invention is the use of an antibody according to the invention for the manufacture of a pharmaceutical composition. A further aspect of the invention is a method for the manufacture of a pharmaceutical composition comprising an antibody according to the invention.

[0081] In another aspect, the invention provides a composition, e.g. a pharmaceutical composition, comprising an antibody as defined herein formulated together with a pharmaceutical carrier.

[0082] A composition of the present invention can be administered by a variety of methods known in the art.

[0083] The invention, having thus been generally described above, will be more readily understood by reference to the following examples, which are provided by way of illustration and not by way of limitation of the invention.

[0084] Working Example design The first bispecific antibody of the invention is an antibody designated BiXAb2a, which has the following structure: i) a contiguous heavy chain comprising: the heavy chain variable region (VH) of trastuzumab corresponding to SEQ ID NO: 7, The wild-type CH1 constant domain from human IgG1 corresponding to SEQ ID NO:8 (residue at Kabat position 192 is a threonine) a polypeptide linker connecting the two Fab heavy chains consisting of EPKSCDKTHTSPPAPAPELLGGPGGPPGPGPGGG (SEQ ID NO:2); the heavy chain variable region (VH) of cetuximab corresponding to SEQ ID NO: 9, a mutated CH1 constant domain derived from human IgG1 corresponding to SEQ ID NO: 10 (residue 192 according to Kabat is mutated from threonine to glutamic acid), a wild-type hinge region derived from human IgG1 corresponding to SEQ ID NO: 11, the wild type CH2 domain of human IgG1 corresponding to SEQ ID NO: 12, The wild type CH3 domain of human IgG1 corresponding to SEQ ID NO: 13. Thus, the bispecific antibody of the invention has a continuous heavy chain of SEQ ID NO:14 (701 residues). ii) a wild-type trastuzumab light chain consisting of SEQ ID NO: 15; iii) A cetuximab light chain having a mutated constant domain derived from human kappa corresponding to SEQ ID NO: 16 (Kabat residues Ser114 and Asn137 are mutated to Ala and Lys, respectively).

[0085] A second bispecific antibody of the invention is an antibody designated BiXAb2b, which is composed of the same sequence except for the linker, which is EPKSCDKTHTSPPAPAPELLGGPAAPPAPAPAGG (SEQ ID NO: 3).

[0086] A third bispecific antibody of the invention is an antibody designated BiXAb2c, which is composed of the same sequences except for the linker, which is EPKSCDKTHTSPPAPAPELLGGPAAPPGPAPGGG (SEQ ID NO: 4).

[0087] A fourth bispecific antibody of the invention is an antibody designated BiXAb3b, which has the following structure: i) a continuous heavy chain comprising: the heavy chain variable region (VH) of atezolizumab corresponding to SEQ ID NO: 23 (EVQLVESGGGLVQPGGSLRLSCAASGFTFSDSWIHWVRQAPGKGLEWVAWISPYGGSTYYADSVKGRFTISADTSKNTAYLQMNSLRAEDTAVYYCARRHWPGGFDYWGQGTLVTVSS); a mutated CH1 constant domain derived from human IgG1 corresponding to SEQ ID NO: 10 (residue 192 according to Kabat is mutated from threonine to glutamic acid), a polypeptide linker connecting the two Fab heavy chains consisting of SEQ ID NO:3; The heavy chain variable region (VH) of daratumumab corresponds to SEQ ID NO: 17 (EVQLLESGGGLVQPGGSLRLSCAVSGFTFNSFAMSWVRQAPGKGLEWVSAISGSGGGTYYADSVKGRFTISRDNSKNTLYLQMNSLRAEDTAVYFCAKDKILWFGEPVFDYWGQGTLVTVSS); the wild-type CH1 constant domain from human IgG1 corresponding to SEQ ID NO: 8 (residue 192 according to Kabat is a threonine), a wild-type hinge region derived from human IgG1 corresponding to SEQ ID NO: 11, the wild type CH2 domain of human IgG1 corresponding to SEQ ID NO: 12, the wild type CH3 domain of human IgG1 corresponding to SEQ ID NO: 13, [Thus, the heavy chain of the present invention is SEQ ID NO: 18 [ka] having a continuous heavy chain of ii) SEQ ID NO: 19 [ka] atezolizumab light chain with a mutated constant domain derived from human kappa corresponding to (Kabat residues Ser114 and Asn137 are mutated to Ala and Lys, respectively); iii) SEQ ID NO: 20 [ka] A wild-type daratumumab light chain consisting of

[0088] For comparison, an antibody designated BiXAb3a was also generated, which differs from the BiXAb3b antibody by a linker consisting of EPKSCDKTHTSPPSPAPELLGGPSTPPTPSPSGG (SEQ ID NO: 6).

[0089] Another construct of the present invention is designated Fab-Fab3b; it contains the same sequence as BiXAb3b, except that the hinge, CH2 domain and CH3 domain are deleted in the heavy chain. Thus, Fab-Fab3b is SEQ ID NO: 21 [ka] The heavy chains are sequential.

[0090] For comparison, a construct designated Fab-Fab3a was also generated, which is composed of the same sequence as BiXAb3a, except for the hinge, and lacks the CH2 and CH3 domains in the heavy chain.

[0091] SEQ ID NOs:7-16 are shown below. Sequence number 7 [ka] Sequence number 8 [ka] Sequence number 9 [ka] Sequence number 10 [ka] Sequence number 11 [ka] Sequence number 12 [ka] Sequence number 13 [ka] Sequence number 14 [ka] Sequence number 15 [ka] Sequence number 16 [ka]

[0092] Gene synthesis The amino acid sequences of an anti-HER2 antibody (trastuzumab, clone humAb4D5-8) and an anti-EGFR antibody (cetuximab) were used to design DNA sequences after codon optimization for mammalian expression using the GeneScript program. For the heavy chain, DNA encoding the signal peptide, variable region, and CH1 constant domain of Fab1 followed by a pseudo hinge linker, and the variable and CH1 constant domains of Fab2 with flanking sequences for digestion with restriction enzymes was synthesized by GeneScript. For the light chain, DNA encoding the signal peptide and the variable and constant kappa regions was synthesized by GeneScript.

[0093] A PCR reaction using Pfu Turbo Hot Start was performed to amplify the inserts, which were then digested with NotI+ApaI and NotI+HindIII for the heavy and light chains, respectively. The double-digested heavy chain fragments were ligated with NotI+ApaI treated pcDNA3.1 expression vector (Invitrogen), into which the human IgG1 CH1+hinge+CH2+CH3 domains had already been inserted. The double-digested light chain fragments were ligated with NotI+HindIII treated pcDNA3.1 expression vector (Invitrogen). Plasmid DNA was verified by double-stranded DNA sequencing.

[0094] Expression and purification The bispecific antibodies of the invention were generated using transient gene expression by co-transfection of three genes encoded on separate vectors in a molar ratio of 2:3:3=HC:LC1:LC2 (one contiguous heavy chain (HC) and two light chains (LC)) in serum-free suspension medium (CHO SFM-II medium™ from Life Technologies)-adapted CHO-S cells. Typically, for a 50 mL medium scale expression study, a total of 50 μg of plasmid DNA (25 μg heavy chain 1, 12.5 μg trastuzumab light chain, and 12.5 μg cetuximab light chain) was mixed in a 1.5 mL Eppendorf tube and 1 mL of CHO SFM medium containing 25 μL of 3 mg / mL PEI transfection reagent (Polyplus) pH 7.0 was added and incubated at room temperature for 20 min. The DNA-PEI mixture was added to 49 mL of Invitrogen FreeStyle CHO-S cells in a 125 mL shake flask at 1–2 × 10 6 The cells were loaded at 1000μg / mL. The cells were shaken for another 6 days. The supernatant was collected by centrifuging the cells at 3,000 rpm for 15 min. The expression titer of BiXAb in the supernatant was determined using a ForteBio Protein A biosensor (Octet® system). The bispecific monoclonal antibodies (BiXAb) were then purified using MabSelect SuRe (GE Healthcare Life Sciences) in Protein A affinity medium. The antibodies were eluted from Protein A using 0.1 M glycine pH 3.5. The purified antibodies in Dulbecco's PBS (Lonza BE17-512Q) were sterile filtered (0.2 μM sterile filter from Techno Plastic Products AG) and the final concentration was determined by reading the absorbance at 280 nm using an Eppendorf Biospectrometer®.

[0095] Analysis by SDS-PAGE Electrophoresis was performed under reducing and non-reducing conditions using Bio-Rad stain-free 4-15% gels and corresponding running buffers. Samples were prepared by combining purified BiXAb or Fab-Fab antibodies with 2x SDS sample buffer and heating at 95°C for 5 min. Preparation of reduced samples included the addition of NuPAGE reducing agent prior to heating. Apparent molecular weights were determined using Ladder Precision Plus Protein Unstained Standards (Bio-Rad).

[0096] Size exclusion chromatography analysis Protein aggregation is frequently observed in engineered protein molecules. We performed analytical size exclusion chromatography (SEC) to assay the content of high molecular weight species of our antibodies. We used a SEC-s3000 (300×7.8 mm) column (Biosep) and an Aktapurifier 10 system (GE Healthcare); the assay was performed at a flow rate of 1 mL / min using PBS buffer at pH 7.4.

[0097] SEC chromatograms of BiXAb2b (Figure 3A), BiXAb3a, BiXAb3b (Figure 3B), Fab-Fab3a, and Fab-Fab3b (Figure 3C) demonstrated that the main peaks corresponded to the expected sizes of monomeric BiXAb and Fab-Fab antibodies; these peaks accounted for 99.9-100.0% of all samples. Thus, we concluded that the antibodies containing the novel linker were completely devoid of high molecular weight species. The narrow and symmetric shape of the monomer peaks suggested that all BiXAb and Fab-Fab were correctly associated and represented by a single species.

[0098] Characterization of BiXAb by differential scanning calorimetry Differential scanning calorimetry (DSC) was used to test the thermal stability of BiXAb2b. Differential scanning calorimetry experiments were carried out using a Microcal™ VP-Capillary DSC system (Malvern Instruments).

[0099] Samples were centrifuged (20,000 × g, 5 min, 4°C) and their protein content was quantified using the IgG analysis program on a Nanodrop ND-1000 spectrophotometer (Thermo Scientific) prior to differential scanning calorimetry analysis. For the assay, samples were diluted in PBS to a final concentration of 1 mg / mL.

[0100] The pre-equilibration time was 3 min and the resulting thermograms were acquired from 20 to 110 °C at a scan rate of 60 °C / h, with a filtering period of 25 s and medium feedback. Before analyzing the samples, five buffer / buffer scans were measured to stabilize the instrument and a buffer / buffer scan was performed between each protein / buffer scan. The data were pre- and post-transition adjusted by fitting to a non-two-state unfolding model and subtracting the baseline.

[0101] DSC results demonstrated that the DSC profile of BiXAb2b exhibited two transitions: a smaller peak with a Cpmax of 96 Kcal / mol / °C and a Tm1 of 71.5°C, corresponding to the unfolding of both the CH2 and Fab domains, and a larger peak with a Cpmax of 190 Kcal / mol / °C and a Tm2 of 80.5°C, corresponding to the unfolding of the CH3 domain.

[0102] Analysis by liquid chromatography / mass spectrometry (LC-MS) LC-MS / MS data were acquired using a Dionex Ultimate 3000 system coupled to a Q-Exactive mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) and a Proswift RP-4H reversed-phase column (250 mm x 1 mm; Thermo Fisher). The column oven temperature was set at 65°C. 10 μl was injected for separation by LC. A mobile phase gradient consisting of LC-MS grade water with 0.1% formic acid (mobile phase A) and acetonitrile with 0.1% formic acid (mobile phase B) was delivered at a flow rate of 0.2 mL / min (total run time 20 min). Eluted antibody species were introduced into the Q-Exactive instrument by electrospray ionization (ESI), which was operated in positive ion mode using a full scan 15,000 resolution. Xcalibur 2.2 software (Thermo Fisher Scientific, Bremen, Germany) was used for instrument control and data file processing.

[0103] 4A and 4B show LC-MS analysis of Fab-Fab3a (containing a linker with SEQ ID NO: 6) and Fab-Fab3a (containing a linker with SEQ ID NO: 3), respectively. FIG. 4A demonstrates that the LC-MS spectrum of Fab-Fab3a with a linker corresponding to SEQ ID NO: 6 is significantly more heterogeneous than the related Fab-Fab3b antibody, which differs only in the composition of the linker (SEQ ID NO: 3). The linker sequence in the Fab-Fab3a antibody contains the PSTPPTPSPS (SEQ ID NO: 22) sequence, which is found in the human IgA1 hinge and is known to undergo O-linked glycosylation at two threonine and two serine residues. Since the glycosylation of these sites is heterogeneous, the product usually has multiple glycoforms, the population of which is strongly influenced by the expression conditions of the recombinant protein. The total number of N-linked and O-linked glycosylation sites in Fab-Fab3a is at least four, which explains the complex MS spectrum observed in FIG. 4A. In order to reduce the heterogeneity due to non-uniform O-linked glycosylation, a linker sequence corresponding to SEQ ID NO:3 was designed. Thus, some serine and threonine residues in the portion of the linker (whose sequence corresponds to the hinge sequence of human IgA1, which is known to undergo O-linked glycosylation) were replaced with glycine residues; furthermore, some other serine and threonine residues in the linker were also replaced with glycine. Figure 4B shows that the MS spectrum of Fab-Fab3b containing the linker of SEQ ID NO:3 is substantially simplified; this is because the Fab-Fab3b analyte is less complex than the Fab-Fab3a analyte, whose spectrum is shown in Figure 4A. Thus, the inventors concluded that a substantially more homogeneous BiXAb preparation was obtained by eliminating four O-linked glycosylation sites from the Fab-Fab3b antibody by replacing the linker of SEQ ID NO:6 with the linker of SEQ ID NO:3.

[0104] The difference in homogeneity of bispecific antibodies containing linkers of SEQ ID NO:6 and SEQ ID NO:3 was particularly evident from the analysis of the full-length BiXAb antibody, a symmetric molecule with two linker sequences and therefore an increased number of glycoforms in the BiXAb preparation. Furthermore, BiXAb has two N-linked glycosylation sites, one on each heavy chain of the Fc domain. The LC-MS spectra of BiXAb3a (constructed with linker of SEQ ID NO:6) and BiXAb3b (constructed with linker of SEQ ID NO:3) are presented in Figures 5A and 5B, respectively. BiXAb3a is expected to contain eight additional O-linked glycosylation sites compared to BiXAb3b based on the differences in the linker sequences. The large number of O-linked glycosylation sites in BiXAb3a explains the highly complex MS spectrum in Figure 5A. Both BiXAb antibodies have two N-glycosylation sites in the Fc domain, resulting in additional glycoforms that contribute to the heterogeneity of both BiXAbs, explaining the residual amount of heterogeneity of BiXAb3b observed in Figure 5B.

Claims

1. A multispecific antigen-binding fragment comprising at least two Fab fragments, each Fab fragment recognizing a different epitope of interest, the Fab fragments being arranged in tandem in any order, the C-terminus of the CH1 domain of a first Fab fragment being linked to the N-terminus of the VH domain of a subsequent Fab fragment through a polypeptide linker, the polypeptide linker sequence being the amino acid sequence EPKX 1 CDKX 2 HX 3 X 4 PPX 5 PAPELLGGPX 6 X 7 PPX 8 PX 9 PX 10 GG (SEQ ID NO: 1), １ , X ２ and X ３ are the same or different and are threonine (T) or serine (S); X ４ and X ５ are the same or different and are selected from the group consisting of serine (S), cysteine ​​(C), alanine (A) and glycine (G); and X ６ , X ７ , X ８ , X ９ , X １０ and are the same or different and are selected from the group consisting of alanine (A), glycine (G), valine (V), asparagine (N), aspartic acid (D), and isoleucine (I).

2. The multispecific antigen-binding fragment of claim 1, wherein the multispecific antigen-binding fragment comprises at least two Fab fragments having different CH1 and CL domains.

3. A multispecific antibody having two identical antigen-binding arms, each consisting of a multispecific antigen-binding fragment of claim 1 or 2.

4. - two identical antigen-binding arms, each consisting of a multispecific antigen-binding fragment as defined in claim 1 or 2; - dimerized CH2 and CH3 domains of immunoglobulins; - an IgA, IgG, or IgD hinge region linking the C-terminus of the CH1 domain to the N-terminus of the CH2 domain of the antigen-binding arm; The multispecific antibody of claim 3 having an immunoglobulin-like structure comprising:

5. A multispecific antibody comprising two heavy chains and four light chains, wherein each heavy chain is a. The Fc region of an immunoglobulin, including the hinge-CH2-CH3 domains Including, b. the Fc region is linked to the heavy chain CH1-VH of Fab of Antibody 1 (Ab1) by the hinge domain; c. which is then linked to the heavy chain CH1-VH of the Fab of antibody 2 (Ab2) by a polypeptide linker sequence, which links the N-terminus of the heavy chain VH domain of the Fab of Ab1 to the C-terminus of the CH1 domain of Ab2; wherein the four light chains include an Ab1 light chain and an Ab2 light chain associated with their cognate heavy chain domains; The polypeptide linker sequence has the amino acid sequence EPKX 1 CDKX 2 HX 3 X 4 PPX 5 PAPELLGGPX 6 X 7 PPX 8 PX 9 PX 10 comprising or consisting of GG (SEQ ID NO:1), Here, X １ , X ２ and X ３ are the same or different and are threonine (T) or serine (S); X ４ and X ５ are the same or different and are selected from the group consisting of serine (S), cysteine ​​(C), alanine (A) and glycine (G); and X ６ , X ７ , X ８ , X ９ , X １０ and are the same or different and are selected from the group consisting of alanine (A), glycine (G), valine (V), asparagine (N), aspartic acid (D) and isoleucine (I).

6. The multispecific antibody of claim 5 which is a bispecific antibody.

7. X ４ The multispecific antigen-binding fragment or multispecific antibody of any of claims 1 to 6, wherein is serine (S) or cysteine ​​(C).

8. X ５ The multispecific antigen-binding fragment or multispecific antibody of any of claims 1 to 7, wherein is alanine (A) or cysteine ​​(C).

9. X ６ , X ７ , X ８ , X ９ , X １０ The multispecific antigen-binding fragment or multispecific antibody of any of claims 1 to 8, wherein are the same or different and are selected from the group consisting of Ala (A) and Gly (G).

10. X ６ and X ７ are identical and selected from the group consisting of Ala (A) and Gly (G).

11. The polypeptide linker sequence is EPKSCDKTHTSPPAPAPELLGGPAAPPAPAPAGG (SEQ ID NO:3); EPKSCDKTHTSPPAPAPELLGGPGGPPGPGPGGG (SEQ ID NO:2); and 11. The multispecific antigen-binding fragment or the multispecific antibody of any of claims 1 to 10, comprising or consisting of a sequence selected from the group consisting of: EPKSCDKTHTSPPAPAPELLGGPAAPPGPAPGGG (SEQ ID NO: 4).

12. The multispecific antigen-binding fragment or multispecific antibody of any one of claims 1 to 11, which recognizes both EGFR and HER2 / neu.

13. 6. The multispecific antibody of claim 5, wherein Ab1 and Ab2 are different and are independently selected from the group consisting of cetuximab or a mutant derivative thereof on the one hand and trastuzumab or a mutant derivative thereof on the other hand.

14. The multispecific antigen-binding fragment or multispecific antibody of any one of claims 1 to 11, which recognizes both CD38 and PD-L1.

15. A polypeptide comprising the heavy chain of an antigen-binding fragment or the heavy chain of a multispecific antibody as defined in any one of claims 1 to 14.

16. A polynucleotide comprising a sequence encoding the polypeptide of claim 15.

17. A host cell transfected with an expression vector comprising the polynucleotide of claim 16.

18. 20. The host cell of claim 17, further transformed with at least two polynucleotides encoding two different light chains, a first light chain that is specifically paired with a first VH / CH1 region of the heavy chain; and a second light chain that is specifically paired with a second VH / CH1 region of the heavy chain.

19. The host cell of claim 18, further transformed with a third polynucleotide encoding a third light chain that is distinct from the first light chain and the second light chain and that specifically pairs with a third VH / CH1 region of the heavy chain.

20. A method for preparing an antigen-binding fragment or a multispecific antibody as defined in any one of claims 1 to 14, comprising the step of culturing a host cell of any one of claims 17 to 19 and recovering the antigen-binding fragment or antibody from the culture medium.

21. A multispecific antigen-binding fragment or a multispecific antibody as defined in any one of claims 1 to 14 for use as a medicament.

22. A multispecific antigen-binding fragment or a multispecific antibody as defined in any of claims 12 to 14 for use in the treatment of cancer.

Images