Modified soluble T cell receptor

Abstract

The present invention provides an engineered chimeric soluble T cell receptor (ETCR) comprising (i) all or part of a TCR α chain fused to all or part of an antibody constant domain, and (ii) all or part of a TCR β chain fused to all or part of an antibody constant domain. (i) and (ii) each contain an engineered linker, an engineered binding interface between the TCR and antibody domains, and one or more mutagen(s) in the TCR domain that stabilize the ETCR. The ETCR is characterized by recognizing a specific peptide-MHC (pMHC) complex and exhibiting biological function.

Description

[Technical field]

[0001] FIELD OF THEINVENTION The present invention relates generally to modified chimeric soluble T cell receptors and compositions thereof, and therapies in the treatment of disease. [Background technology]

[0002] 2. Background of the Invention T lymphocytes play a central role in adaptive immunity through their response to a wide variety of foreign antigens presented as peptides associated with major histocompatibility molecules (MHC). Specific recognition of peptide-MHC (pMHC) complexes is achieved by membrane-bound, multicomponent cell surface glycoproteins called T cell receptors (TCRs). Natural TCRs are heterodimeric cell surface proteins of the immunoglobulin superfamily that bind to the invariant protein of the CD3 complex, which is involved in mediating signal transduction. TCRs exist in αβ and γδ types, which are structurally similar but have quite different anatomical locations and, presumably, functions. In particular, αβ-TCRs appear on more than 95% of all T lymphocytes and assemble in an almost unlimited repertoire of diversity to provide humans with a central defense against both exogenous and endogenous diseases.

[0003] Antibodies and TCRs are just two types of molecules that specifically recognize antigens, with TCRs being the only receptors for certain peptide antigens presented by MHC, where foreign peptides are often the only sign of abnormality in cells. Similar to antibodies, interest has also been raised in the development of soluble antigen-specific TCRs and their derivatives as drug candidates to expand therapeutic targets of intracellular epitopes. Specific TCR:pMHC interactions can also be exploited as powerful diagnostic tools to detect infections, disease markers and specific cells expressing the corresponding pMHC complexes. However, unlike antibodies, TCRs are generally very unstable when expressed as soluble molecules and often face problems such as low expression yields, aggregation and misfolding. Potential explanations include extensive glycosylation, unstable constant domains and inefficient chain pairing.

[0004] A number of papers have described the generation of TCR heterodimers utilizing native disulfide bridges in the hinge regions connecting each subunit (Garboczi et al., (1996), Nature 384(6605):134-141; Garboczi et al., (1996), PNAS USA 91:11408-11412; Davodeau et al., (1993), J. Biol. Chem. 268(21):15455-15460; Golden et al., (1997), J. Imm. Meth. 206:163-169). However, although such TCRs could be recognized by TCR-specific antibodies, none of them were shown to recognize native ligands that display misfolded complementarity determining regions (CDRs). Recently, WO 2004 / 074322 described a soluble TCR that is capable of recognizing its natural ligand, correctly folded so as to be stable over a period of time, and can be produced in reasonable quantities. This TCR comprises a TCR α chain extracellular domain that dimerizes into a TCR β chain extracellular domain that is linked by an artificial disulfide bond between constant domain residues CαS48-CβT57. Based on such a soluble TCR format, an innovative bispecific TCR drug, Tebentafusp, was developed and showed benefits for patients with metastatic melanoma. Similarly, US 2018 / 021682 also described several other artificial disulfide bonds between constant domain residues (CαR53, P89, Y10 and CβS54, A19 and E20) and constant domain / variable domain residues (Vα46, 47 (IMGT numbering) and Cβ60, 61). Very recently, Karen et al. described the computational design of soluble TCRs. Using both computational methods and experimental screening with Rosetta, they identified seven mutations in Cα and Cβ that significantly improved full-length TCR association and expression (Karen et al., (2020), Nat. Comm., Vol. 11:2330).In particular, such soluble TCRs designed based on the modification of natural TCRs by either artificial disulfide bonds or mutagenesis, which are usually highly glycosylated, especially in the constant domain, may potentially cause unspecified potential as drug candidates.To avoid such drawbacks, in some cases, such soluble TCRs are produced in E. coli and assembled by protein refolding process, which results in a relatively complicated manufacturing procedure.

[0005] The high degree of sequence identity (30%-70%) between the variable (V) and constant (C) domains of TCRs and antibodies suggests that the TCR folds into a β-sheet sandwich structure that pairs similarly with the heavy (H) and light (L) chains of antibody Fab fragments. Given the similar overall structure and heterodimeric binding of TCRs and antibody Fabs, attempts have been made to generate TCR-antibody chimeric proteins as an alternative method to obtain soluble TCRs. Chimeric TCR formats described so far include a) direct injection of all or part of the TCR into a fragment crystallizable (Fc) domain to generate immunoglobulin-like aggregates, and b) injection of the TCR into an antibody fabC domain with or without additional stabilizing domains (e.g., Fc region, leucine zipper) to generate fab-like aggregates. It mainly involves the injection of V domains (Jack et al., (1994) Proc. Natl. Acad. Sci. USA 91:12654-12658; Mark et al., (1987) Proc. Natl. Acad. Sci. USA 84:2936-2940; Greg et al., (1988) J. Biol. Chem. 264(13):7310-7316; Be (1991) Proc. Natl. Acad. Sci. USA 88:8077-8081; Jonathan et al. (1997) J. Exp. Med. 186(8):1333-1345; Jonathan et al. (1999) Cell. Immunol. 192:175-184; AU729,406; US6,911,204). However, although the correct function of such chimeric proteins was observed in a few cases, the extremely low expression levels (30ng / ml to 1μg / ml) hindered further application of this protein as a pharmaceutical preparation. Indeed, many differences became evident upon careful investigation of the TCR and antibody structures, providing an explanation for the unsatisfactory results of simple injection of TCR-antibody chimeras to date. The TCR is more extensive across the middle than the Fab (about 56 Å vs. about 46 Å) due to protrusions from the loops of the Cβ domain in what appears to be a general feature of all β-chains.The TCR is also more asymmetric and recessed than the Fab, with the β-sheet crossing the Cα / Cβ interface at a more parallel angle, shifting approximately 5 Å off-center from the pseudo-two-fold symmetric position for Cα / Cβ. This asymmetry is accentuated by the smaller size of the Cα domain compared to Cβ. Thus, instead of simple injection of wild-type TCR and antibodies, comprehensive design based on structural features is required to enhance compatibility and thus generate stable and functional chimeras. Summary of the Invention [Problem to be solved by the invention]

[0006] Considering the importance of soluble TCRs, it is desirable to provide an alternative method for producing such molecules with native functions and great development potential. In the present invention, we have produced stable, soluble and functional TCRs (ETCRs) and TCR derivatives (bispecific ETCRs) in eukaryotic expression systems. Moreover, we have observed potent in vitro antitumor activity using the bispecific TCRs of the present invention. [Means for solving the problem]

[0007] BRIEF SUMMARY OF THE PRESENTINVENTION In one aspect, the disclosure provides a polypeptide complex comprising a first polypeptide comprising, from N-terminus to C-terminus, a first TCR alpha chain variable domain of a first TCR and a first antibody constant domain (C1) operably associated therewith, and a second polypeptide comprising, from N-terminus to C-terminus, a first TCR beta chain variable domain of the first TCR and a second antibody constant domain (C2) operably associated therewith, wherein C1 and C2 are capable of forming a dimer via their natural inter-chain bonds and interactions. In certain embodiments, the first TCR has a first antigen specificity.

[0008] In certain embodiments, C1 and C2 are selected from the group consisting of IgG1 (IMGT Accession Nos. J00228, Z17370, AL122127, MG920252, MG92025, MG920246, MG920247, MG920248, MG920249, MG920250, MG920251, MG920253), IgG2 (IMGT Accession Nos. J00230, AJ250170, AF449616, AF449618, AF928742, MH025828, MH025829, MH025830, MH025832, MH025833, MH025834), and / or IgG3 (IMGT Accession Nos. J00230, AJ250170, AF449616, AF449618, AF928742, MH025828, MH025829, MH025830, MH025832, MH025833, MH025834). , MH025835, MH025836), IgG3 (IMGT accession numbers: X03604, K01313, X16110, X99549, AJ390236, AJ390237, AJ390238, AJ390241, AJ390242, AL122127, AJ390247, AJ390252, AJ390254, AJ390260, AJ390262, AJ390272, AJ390276, MG920256, MG920255, MG920254, MH025837, MG920257, MG920258, MG920259, MG920 260, MG786813, MG920261), IgG4 (IMGT Accession Nos. K01316, AL928742), IgM (IMGT Accession Nos. X14940, K01307, X57331, AC254827), IgA1 (IMGT Accession Nos. J00220, IMGT000035), IgA2 (IMGT Accession Nos. J00221, M60192, S71043), IgD (IMGT Accession Nos. K02875, X57331) and IgE (IMGT Accession Nos. J00222, L00022, IMGT000025, AL928742). an antibody heavy chain (CH1 domain) selected from the group consisting of Cλ1 (IMGT Accession No. J00252, X51755), Cλ2 (IMGT Accession No. J00253, X06875, AJ491317), Cλ3 (IMGT Accession No. J00254, K01326, X06876, D87017), Cλ6 (IMGT Accession No. J03011), Cλ7 (IMGT Accession No. X51755, M61771, X51755, M61771, KM455557), Cκ1 (IMGT Accession No. J00241), Cκ2 (IMGT Accession No. M11736),The light chain constant domain (Cλ domain or Cκ domain) is selected from the group consisting of Cκ3 (IMGT Accession No.: M11737), Cκ4 (IMGT Accession No.: AF017732) and Cκ5 (IMGT Accession No.: AF113887).

[0009] In certain embodiments, C1 comprises a modified CH1 domain selected from the group consisting of IgG1, IgG2, IgG3, IgG4, IgM, IgA1, IgA2, IgD and IgE, and C2 comprises a modified λ or κ light chain constant domain (Cλ domain or Cκ domain) derived from a human immunoglobulin, the Cλ domain being selected from the group consisting of Cλ1, Cλ2, Cλ3, Cλ6 and Cλ7.

[0010] In certain embodiments, C1 comprises a modified lambda or kappa light chain constant domain (Clambda domain or Cκ domain) from a human immunoglobulin, where the Clambda domain is selected from the group consisting of Clambda1, Clambda2, Clambda3, Clambda6 and Clambda7, and C2 comprises a modified CH1 domain selected from the group consisting of IgG1, IgG2, IgG3, IgG4, IgM, IgA1, IgA2, IgD and IgE.

[0011] In certain embodiments, a) C1 comprises a modified CH1 domain derived from human immunoglobulin G1 (IgG1) and C2 comprises a modified Cλ1 domain derived from a human immunoglobulin; b) C1 comprises a modified CH1 domain derived from human immunoglobulin G2 (IgG2) and C2 comprises a modified Cλ1 domain derived from a human immunoglobulin; c) C1 comprises a modified CH1 domain derived from human immunoglobulin G3 (IgG3) and C2 comprises a modified Cλ1 domain derived from a human immunoglobulin; and d) C1 comprises a modified CH1 domain derived from human immunoglobulin G4 (IgG4). and C2 comprises a modified Cλ1 domain derived from a human immunoglobulin; e) C1 comprises a modified CH1 domain derived from a human immunoglobulin G1 (IgG1) and C2 comprises a modified Cλ2 domain derived from a human immunoglobulin; f) C1 comprises a modified CH1 domain derived from a human immunoglobulin G2 (IgG2) and C2 comprises a modified Cλ2 domain derived from a human immunoglobulin; g) C1 comprises a modified CH1 domain derived from a human immunoglobulin G3 (IgG3) and C2 comprises a modified Cλ2 domain derived from a human immunoglobulin; i) C1 comprises a modified CH1 domain derived from human immunoglobulin G1 (IgG1) and C2 comprises a modified Cλ1 domain derived from a human immunoglobulin; j) C1 comprises a modified CH1 domain derived from human immunoglobulin G2 (IgG2) and C2 comprises a modified Cλ3 domain derived from a human immunoglobulin; k) C1 comprises a modified CH1 domain derived from human immunoglobulin G3 (IgG3) and C2 comprises a modified Cλ3 domain derived from a human immunoglobulin. domains, l) C1 comprises a modified CH1 domain derived from human immunoglobulin G4 (IgG4) and C2 comprises a modified Cλ3 domain derived from a human immunoglobulin, m) C1 comprises a modified CH1 domain derived from human immunoglobulin G1 (IgG1) and C2 comprises a modified Cλ6 domain derived from a human immunoglobulin, n) C1 comprises a modified CH1 domain derived from human immunoglobulin G2 (IgG2) and C2 comprises a modified Cλ6 domain derived from a human immunoglobulin, o) C1 comprises a modified CH1 domain derived from human immunoglobulin G3 (IgG3);C2 comprises a modified Cλ6 domain from a human immunoglobulin, p) C1 comprises a modified CH1 domain from a human immunoglobulin G4 (IgG4) and C2 comprises a modified Cλ6 domain from a human immunoglobulin, q) C1 comprises a modified CH1 domain from a human immunoglobulin G1 (IgG1) and C2 comprises a modified Cλ7 domain from a human immunoglobulin, r) C1 comprises a modified CH1 domain from a human immunoglobulin G2 (IgG2) and C2 comprises a modified Cλ7 domain from a human immunoglobulin, s) C1 comprises a modified CH1 domain from a human immunoglobulin G3 (IgG3) and C2 comprises a modified Cλ7 domain from a human immunoglobulin, and t) C1 comprises a modified CH1 domain from a human immunoglobulin G4 (IgG4) and C2 comprises a modified Cλ7 domain from a human immunoglobulin.

[0012] In certain embodiments, C1 comprises a modified CH1 according to any one of SEQ ID NOs: 11, 13, 15 and 17, and / or C2 comprises a modified Cλ according to any one of SEQ ID NOs: 1, 3, 5, 7 and 9.

[0013] In certain embodiments, the first Vα is operably linked to C1 via a first junction domain, and the first Vβ is operably linked to C2 via a second junction domain.

[0014] In certain embodiments, C1 comprises a modified CH1 and C2 comprises a modified Cλ, wherein the first junction domain comprises any one of SEQ ID NOs: 19, 21, and 23, and / or the second junction domain comprises any one of SEQ ID NOs: 25, 27, 29, 31, 33, and 35, preferably wherein the second junction domain comprises EDLXNVXP, where X is any amino acid.

[0015] In certain embodiments, the TCR Vβ comprises a mutagen at one or more positions selected from 10, 13, 19, 24, 48, 54, 77, 90, 91, 123 and 125 (IMGT numbering) of the framework region, preferably, the TCR Vβ comprises at least one mutation at position 13 or at least two mutations at positions 90 and 91.

[0016] In certain embodiments, Cλ or CH1 comprises a mutagen at one or more positions selected from positions 30, 31 and 33.

[0017] In another aspect, a recent disclosure provides a multispecific antigen-binding complex comprising a first antigen-binding moiety comprising the aforementioned polypeptide complex, and a second antigen-binding moiety, wherein the first antigen-binding moiety has a first antigen specificity.

[0018] In certain embodiments, the second antigen-binding moiety binds to a different epitope on the first antigen or preferably has a second antigen specificity that is different from the first antigen specificity and is conjugated at the N-terminus or C-terminus of the first polypeptide of the first antigen-binding moiety or the second polypeptide of the first antigen-binding moiety.

[0019] In certain embodiments, the first antigen specificity and the second antigen specificity are directed to two different antigens or to two different epitopes on one antigen.

[0020] In certain embodiments, the multispecific antigen-binding complex comprises a first antigen-binding moiety and a second antigen-binding moiety, wherein the first antigen-binding moiety comprises a first polypeptide comprising, from N-terminus to C-terminus, a first TCR alpha chain variable domain of a first TCR and a first antibody constant domain (C1) operably associated therewith, and a second polypeptide comprising, from N-terminus to C-terminus, a first TCR beta chain variable domain of the first TCR and a second antibody constant domain (C2) operably associated therewith, wherein C1 and C2 are capable of forming a dimer via their natural inter-chain binding and interaction. The first TCR has a first antigen specificity.

[0021] In certain embodiments, the second antigen-binding portion has specificity for a different epitope on the first antigen.

[0022] In certain embodiments, the second antigen-binding moiety has a second antigen specificity that is different from the first antigen specificity and is conjugated at the N-terminus or C-terminus of the first polypeptide of the first antigen-binding moiety or the second polypeptide of the first antigen-binding moiety.

[0023] In certain embodiments, one of the first and second antigen specificities is directed against a T cell-specific receptor molecule and / or a natural killer cell (NK cell)-specific receptor molecule, and the other is directed against a tumor-associated antigen and / or a tumor neoantigen.

[0024] In a particular embodiment, the first antigen-binding portion comprises a TCR Vα and a TCR Vβ, wherein Vα comprises an amino acid sequence selected from SEQ ID NOs: 37, 41 and 45, and Vβ comprises an amino acid sequence selected from SEQ ID NOs: 39, 43 and 47, and preferably the second antigen-binding portion comprises an scFv selected from SEQ ID NO: 49.

[0025] In certain embodiments, the first antigen-binding portion binds to HLA*A*02:01-NY-ESO-1 peptide (SLLMWITQC) (SEQ ID NOs: 37-40, 45-48) and the second antigen-binding portion binds to Cluster of Differentiation 3 (CD3) (SEQ ID NOs: 49-50).

[0026] In certain embodiments, the first antigen-binding portion binds to HLA*A*02:01-GP100 peptide (YLEPGPVTV) (SEQ ID NOs: 41-44) and the second antigen-binding portion binds to CD3 (SEQ ID NOs: 49-50).

[0027] In certain embodiments, the second antigen-binding portion comprises a single-chain variable fragment (scFv) comprising both a heavy chain variable domain and a light chain variable domain covalently conjugated via a flexible linker.

[0028] In another aspect, the disclosure provides an isolated polynucleotide that encodes a polypeptide complex provided herein, or a multispecific antigen-binding complex provided herein.

[0029] In one aspect, the disclosure provides an isolated vector comprising a polynucleotide provided herein.

[0030] In one aspect, the disclosure provides a host cell comprising an isolated polynucleotide provided herein, or an isolated vector provided herein.

[0031] In one aspect, provided herein is a conjugate comprising a polypeptide complex or a multispecific antigen-binding complex as provided herein.

[0032] In one aspect, the disclosure provides a method of expressing a polypeptide complex provided herein or a multispecific antigen-binding complex provided herein, comprising culturing a host cell provided herein under conditions in which the polypeptide complex or multispecific antigen-binding complex is expressed.

[0033] In one aspect, the disclosure provides a method of producing a polypeptide complex as provided herein, comprising: a) introducing into a host cell a first polynucleotide encoding a first polypeptide comprising, from N-terminus to C-terminus, a first TCR alpha chain variable domain of the first TCR and a first antibody constant domain (C1) operably associated therewith, and a second polypeptide comprising, from N-terminus to C-terminus, a first TCR beta chain variable domain of the first TCR and a second antibody constant domain (C2) operably associated therewith, wherein C1 and C2 are capable of forming a dimer via their natural inter-chain bonds and interactions, and wherein the first TCR has a first antigen specificity; and b) allowing the host cell to express the polypeptide complex.

[0034] In one aspect, provided herein is a method of producing a multispecific antigen binding complex as provided herein comprising the steps of: a) introducing into a host cell a first polynucleotide encoding a first polypeptide comprising, from N-terminus to C-terminus, a first TCR alpha chain variable domain of the first TCR and a first antibody constant domain (C1) operably associated therewith, and a second polypeptide comprising, from N-terminus to C-terminus, a first TCR beta chain variable domain of the first TCR and a second antibody constant domain (C2) operably associated therewith, wherein C1 and C2 are capable of forming a dimer through their natural interchain bonds and interactions, wherein the first TCR has a first antigen specificity and the second antigen binding moiety has a second antigen specificity different from the first antigen specificity and is conjugated at the N-terminus or C-terminus of the first polypeptide of the first antigen binding moiety or the second polypeptide of the first antigen binding moiety; and b) allowing the host cell to express the multispecific antigen binding complex.

[0035] In certain embodiments, the methods of producing a multispecific antigen-binding complex provided herein further comprise the step of isolating the polypeptide complex.

[0036] In one aspect, provided herein is a composition comprising a polypeptide complex provided herein or a multispecific antigen-binding complex provided herein.

[0037] In one aspect, provided herein is a pharmaceutical composition comprising a polypeptide complex provided herein or a multispecific antigen-binding complex provided herein, and a pharma- ceutically acceptable carrier.

[0038] In one aspect, the disclosure provides a method for treating a condition or disease, such as cancer, in a subject in need thereof, comprising administering to the subject a therapeutically effective amount of a polypeptide complex provided herein or a multispecific antigen-binding complex provided herein. In certain embodiments, when both the first antigen and the second antigen are modulated, the condition can be alleviated, eliminated, treated or prevented.

[0039] In another aspect, the disclosure provides kits comprising the polypeptide complexes provided herein for the detection, diagnosis, prognosis or treatment of a disease or condition. [Brief description of the drawings]

[0040] [Figure 1] FIG. 1 is a schematic diagram showing the truncated and mutated amino acid positions in the antibody constant region in step 1 of Example 1. [Diagram 2] FIG. 1 shows the display and binding activity assay results of phage-displayable chimeric TCR intermediates obtained by screening in step 2 of Example 1. [Diagram 3] FIG. 1 shows the results of comparing the affinities of chimeric TCR intermediates obtained by screening in step 2 of Example 1 using dsTCR and scTCR. [Figure 4] FIG. 1 is a schematic diagram showing the positions of incorporation between VTCRC IgG1, IgG4, IgA1, and VTCRC κ and λ linkers in step 3 of Example 1. [Diagram 5]FIG. 1 shows the results of comparing chimeric TCR phage linkers in step 4 of Example 1. [Figure 6] FIG. 1 shows the results of phage display and binding activity assay of cloned phages in 1G4 affinity maturation in Example 1, step 5. [Figure 7A] Figure 7A shows the modeled structure of a representative CTCR, with artificial disulfide bonds depicted as spheres. [Figure 7B] Figure 7B shows the modeled structure of a representative ETCR. The long FG loop that helps stabilize Vβ in CTCR is absent in ETCR, resulting in a less stable β-strand structure. [Figure 8] Figure 8a shows the junction domain structure of CTCR. The black arrows indicate the ends of the junction domain analyzed from the structure, and the red dashed arrows indicate potential polar contacts formed between the junction domain and the FG loop. Figure 8b shows the junction domain structure of ETCR(SSAS). The black arrows indicate the ends of the junction domain analyzed from the overlap of CTCR and ETCR. [Figure 9] Figure 1 shows the modelled structure of a representative CTCR, with residues involved in the variable and constant domain binding interfaces shown as sticks covered by grey mesh. [Figure 10A] Figure 10A shows the detailed structure of the Vβ-Cβ binding interface of a representative CTCR. Residues involved in the binding are shown as sticks, red arrows and yellow dashed lines indicate polar contacts, and orange circles indicate non-polar contacts. [Figure 10B] Figure 10B shows the detailed structure of the Vβ-Cλ binding interface of a representative ETCR. Residues involved in binding are shown as sticks, and red arrows indicate the absence of polar contacts. [Figure 11A] FIG. 11A shows the results of SDS-PAGE of representative ETCRs. [Figure 11B] FIG. 11B shows KD ELISA results of representative ETCRs. [Figure 12A] Figure 12A shows the overlap results of ETCR1 (cyan) and ETCR2 (magenta), with residues in FR1 shown as sticks. [Figure 12B] FIG. 12B shows the results of SDS-PAGE of representative ETCR2 mutants. [Figure 13] 13A to 13E show representative sensorgrams of SPR analysis of ETCR and CTCR. [Figure 14] FIG. 1 shows representative ETCR1 and CTCR1 FACS results. [Figure 15] Schematic diagram showing the bispecific ETCRs tested. The gene product of anti-CD3scFv was amplified and inserted into the N-terminus of the TCR Vβ domain, the C-terminus of the antibody Cλ domain, the N-terminus of the TCR Vα domain, and the C-terminus of the antibody CH1 domain, respectively, to generate bispecific ETCR1-E1.1, ETCR1-E1.2, ETCR1-E1.3, and ETCR1-E1.4 (Figure 15A-D). For the CTCR, the gene product of anti-CD3scFv was amplified and inserted into the N-terminus of the TCR Vβ domain to generate CTCR1-E1.1 (Figure 15E). [Figure 16] Figure 1 shows SDS-PAGE results of bispecific ETCR1, lanes 1-4: supernatant of bispecific ETCR1, lanes 5-8: corresponding purified bispecific ETCR1. [Figure 17A] FIG. 17A shows the dose-dependent results of redirecting T cell killing towards T2 cells at 18 h. [Figure 17B] FIG. 17B shows the dose-dependent results of redirecting T cell killing towards T2 cells at 24 h. [Figure 18] FIG. 1 shows dose-dependent results of redirecting T cell killing against A375 cells at 72 h. [Figure 19] FIG. 1 is a diagram showing the types of ETCR. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0041] Detailed Description The present invention will be described in detail below, but it should be understood that the present invention is not limited to the specific methodology, protocol and reagent described herein, because they may vary.It should also be understood that the terminology used herein is only for the purpose of describing specific embodiments, and is not intended to limit the scope of the present invention, which is limited only by the scope of the claims that follow.Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art.

[0042] The term "isolated" as used herein refers to a state obtained from the natural state by artificial means. A particular "isolated" substance or component may exist in nature, possibly because its natural environment is altered, or the substance is isolated from its natural environment, or both. For example, a particular non-isolated polynucleotide or polypeptide naturally exists in a particular living animal body, and the same polynucleotide or polypeptide of high purity isolated from such a natural state is referred to as an isolated polynucleotide or polypeptide. The term "isolated" does not exclude mixed artificial or synthetic substances, or other impurities that do not affect the activity of the isolated substance.

[0043] The term "vector" as used herein refers to a nucleic acid vehicle into which a polynucleotide can be inserted. If the vector allows expression of a protein encoded by the inserted polynucleotide, the vector is called an expression vector. A vector can carry genetic material elements that are expressed in a host cell by transformation, transduction or transfection into the host cell. Vectors are well known to those skilled in the art and include, but are not limited to, plasmids, phages, cosmids, artificial chromosomes, such as yeast artificial chromosomes (YACs), bacterial artificial chromosomes (BACs) or P1-derived artificial chromosomes (PACs), phages, such as lambda phages or M13 phages, and animal viruses. Animal viruses that can be used as vectors include, but are not limited to, retroviruses (including lentiviruses), adenoviruses, adeno-associated viruses, herpes viruses (e.g., herpes simplex viruses), poxviruses, baculoviruses, papilloma viruses, papova viruses (e.g., SV40). A vector may contain multiple elements for regulating expression, including, but not limited to, a promoter sequence, a transcription initiation sequence, an enhancer sequence, a selection element, and a reporter gene. In addition, a vector may contain an origin of replication.

[0044] The term "host cell" as used herein refers to a cell line that can be modified to produce a protein, a protein fragment, or a peptide of interest. Host cells include, but are not limited to, cultured cells, such as mammalian cultured cells derived from rodents (rat, mouse, guinea pig, or hamster), such as CHO, BHK, NSO, SP2 / 0, YB2 / 0, or mammalian cultured cells derived from human tissue, or hybridoma cells, yeast cells, and insect cells, as well as cells contained in transgenic animals or cultured tissue. This term encompasses not only the specific subject cell, but also the progeny of such cells. Since certain modifications may occur in succession due to either mutation or environmental influences, such progeny may not be identical to the parent cell, but still fall within the scope of the term "host cell".

[0045] The term "SPR" or "surface plasmon resonance" as used herein refers to and includes the optical phenomenon that allows the analysis of biospecific interactions in real time by detecting changes in protein concentration in a biosensor matrix, for example, using the BIAcore system (PharmaciaBiosensorAB, Uppsala, Sweden and Piscataway, NJ).For further description, see Example 5 and Jonsson, U. et al. (1993) Ann.Biol.Clin. 51:19-26; Jonsson, U. et al. (1991) Biotechniques 11:620-627; Johnsson, B. et al. (1995) J.Mol.Recognit. 8:125-131; and Johnnson, B. et al. (1991) Anal.Biochem. 198:268-277.

[0046] The term "cancer" as used herein refers to any one of solid and non-solid tumors, e.g., leukemia, mediated by the growth, proliferation or metastasis of tumor or malignant cells, causing a medical condition.

[0047] The term "treatment, treating or treated" as used herein in the context of treating a condition, whether in a human or animal, generally refers to treatments and therapies in which some desired therapeutic effect is achieved, such as inhibition of progression of the condition, including reduction in the rate of progression, halting the rate of progression, regression of the condition, amelioration of the condition, and cure of the condition. It also includes treatment as a preventative measure (i.e., prophylaxis, prevention). In cancer, "treating" can refer to attenuation or delay of the growth, proliferation or metastasis of tumors or malignant cells, or some combination thereof. In tumors, "treatment" includes removal of all or part of the tumor, inhibition or delay of tumor growth and metastasis, prevention or delay of tumor onset, or some combination thereof.

[0048] The term "effective amount" or "therapeutically effective amount" as used herein refers to an amount of an active compound or substance, a composition, or a dosage amount for containing an active compound, which, when administered according to a desired treatment regimen, is effective to produce some desired therapeutic effect, commensurate with a reasonable benefit / risk ratio. For example, when used in connection with the treatment of a target antigen-related disease or condition, an "effective amount" refers to an amount or concentration of an antibody or antigen-binding portion thereof that is effective to treat the disease or condition.

[0049] The term "pharmaceutical acceptable" as used herein means that the vehicle, diluent, excipient and / or salt thereof is chemically and / or physically compatible with the other ingredients in the formulation and physiologically compatible with the recipient.

[0050] As used herein, the term "pharmaceutically acceptable carrier and / or excipient" refers to a carrier and / or excipient that is pharmacologically and / or physiologically compatible with the subject and the active agent, which are well known in the art (see, for example, Remington's Pharmaceutical Sciences. Gennaro AR, ed., 19th ed. Pennsylvania: Mack Publishing Company, 1995), including, but not limited to, pH adjusters, surfactants, adjuvants and ionic strength enhancers. For example, pH adjusters include, but are not limited to, phosphate buffers, surfactants include, but are not limited to, cationic, anionic or nonionic surfactants, such as Tween-80, and ionic strength enhancers include, but are not limited to, sodium chloride.

[0051] As used herein, the term "subject" includes any human or non-human animal. The term "non-human animal" includes all vertebrates, e.g., mammals, and non-mammals, e.g., non-human primates, sheep, dogs, cats, horses, cows, chickens, amphibians, reptiles, etc. Except where otherwise stated, the terms "patient" or "subject" are used interchangeably.

[0052] The experimental methods in the following examples are conventional unless otherwise specified. EXAMPLES

[0053] Example 1. TCR variable domain genes are ligated with antibody constant region genes and further associated with the gIII gene of a phage to screen for TCR heterodimer formats that can be displayed on the surface of the phage.

[0054] Step 1. Two different TCR V domain genes: CTCR2 and CTCR2 are ligated with antibody constant region genes The antibody heavy chain constant region includes the CH1 domain or full length IgG1, IgG2, IgG4, IgM, IgA1, IgA2, IgD, IgE and the antibody light chain constant region Cκ, Cλ. In particular, two additional mutations in the antibody constant domain were designed, tested and compared to wild type: 1) interchain disulfide bond mutation: cysteine ​​to serine (if the C at the interchain disulfide bond position in the antibody constant region is mutated to S, the constant region CH1 is designated as κG1s, κG1s-reverse), 2) N-glycosylation mutation: asparagine to glutamine. The truncations and mutation positions are shown in Figure 1. Ligation results in V TCR C IgG1、IgG2、IgG4、IgA1、IgA2、IgM、IgD、IgE and V TCR C κ、λ Gene product (V TCR These gene products were inserted into the phagemid vector pcom3XX-DT, where V TCR C IgG1、IgG2、IgG4、IgA1、IgA2、IgM、IgD、IgE V TCR C κ、λ was conjugated to the gIII gene expressing the c-Myc.6His tag, and V TCR C κ、λexpresses a Flag tag. Because the phage is capable of self-association, the two expressed polypeptides spontaneously complex and are displayed on the phage as a functional heterodimer. The corresponding CTCR is also inserted into the same phagemid vector, where VβCβ is conjugated with a gIII gene expressing a c-Myc.6His tag and VαCα expresses a Flag tag. The corresponding single-chain TCR is assembled as Vα-(G4S)-Vβ and inserted into the plasmid vector pFL249 fused with a gIII gene expressing a 6His.c-Myc tag.

[0055] Step 2. Optimization of phage culture conditions To screen for phages displaying various TCR heterodimer formats, clones were inoculated into 600 μl of 2YT medium (containing 10 g / L yeast extract, 16 g / L trypsin, 5 g / ml, 0.1 mg / ml ampicillin and 2% glucose, pH 7.0). Strains were incubated at OD 600 =0.3-0.5, 7.5E9pfu of helper phage M13KO7 (Invitrogen) was added, and further incubated for 45 minutes in a 37°C incubator. The cells were pelleted by centrifugation at 4,000g for 10 minutes, resuspended in 600μl of 2YT medium containing 0.1mg / ml ampicillin, 0.05 kanamycin, and 5mM MgSO4, and cultured for 36 hours in a shaker at 25°C. Phage supernatant displaying TCR heterodimers can be obtained by centrifugation.

[0056] Step 3. Detecting phage display and binding activity of TCR heterodimers by phage ELISA The display level of TCR heterodimer on phage was detected by sandwich ELISA method. ELISA plates were coated with anti-Flag (2 μg / ml) to capture ETCR or CTCR heterodimer phages, and with anti-c-myc (1 μg / ml) to capture scTCR. The plates were blocked with blocking solution (3% BSA) for 1 h. Phage supernatant diluted 1:1 was added and incubated at room temperature (20-25 °C) for 2 h. After washing six times with 1 × PBST, anti-phage M13 alkaline phosphatase conjugated antibody was used to detect the display level of TCR on the phage surface. The binding ability of TCR heterodimer on phage was detected by ELISA method. ELISA plates were coated with 4 μg / ml SA overnight and blocked with blocking solution (3% BSA) for 1 h. 2 μg / ml biotinylated pMHCI monomer was added and incubated at RT for 1 h. After washing six times with 1×PBST, phage supernatant diluted 1:1 was added and incubated at RT for 1 h, and then anti-phage M13 alkaline phosphatase conjugated antibody was used to detect the binding activity of TCR-displaying phages. Finally, 64 ETCR formats were screened, and 6 optimal formats were obtained: λG1, λG1s, κG1s, λG1s-reverse, λG4s-reverse, λA1s-reverse, which revealed better display level and binding ability, as shown in Figure 2.

[0057] Figure 2A and Figure 2B show the display levels of TCR1 and TCR2 in different TCR types, respectively. Four clones were randomly selected to test each displayed TCR. The results suggested that all ETCRs, CTCRs and scTCRs could be well displayed by phage.

[0058] Figures 2C and 2D show the binding ability of TCR1 and TCR2, respectively, in different TCR formats. Differential binding to specific pMHCI was observed with different TCR formats, where non-specific binding to pMHCI was not observed.

[0059] The presentation level and binding activity of the optimal ETCR format were further determined and compared with CTCR and scTCR by phage relative quantification ELISA. The number of phage in the initial well was 5E10 pfu at 1:3 dilution. The results show that the presentation level of the optimal ETCR format is better than scTCR and dsTCR. Also, as shown in Figure 3, the binding affinity of our TCR λG4s-reverse is 2-6 times better than that of dsTCR. To further optimize the ETCR format, a linker domain was then designed.

[0060] Step 4. Linker optimization of chimeric TCR structures To improve the expression level and binding activity of ETCR, the FR4 portion of TCR Vα or Vβ was truncated and then directly connected to the constant domain of the antibody. The results showed that the stability of TCR was affected and the binding activity was reduced.

[0061] Meanwhile, linkers of different lengths, including SS, SSA, SSAS, SSASS, SSASSS, were inserted between the C-terminus of ETCR variable domain and the N-terminus of antibody constant domain, and the specific positions are shown in Figure 4. Four clones were randomly selected to detect the phage display level and binding activity. Among all the linkers, the SSAS linker showed better ability than the other linkers (Figure 5). Therefore, λG4s-reverse-SSAS was selected as the final ETCR format for TCR affinity maturation.

[0062] Step 5. Construction and screening of TCR affinity maturation libraries The native TCR 1G4 was selected for affinity maturation as a proof-of-concept test. The Vα and Vβ genes of 1G4 were synthesized and cloned into our phagemid vector containing λG4s-reverse-SSAS ETCR backbone, resulting in 1G4 ETCR (template for affinity maturation, wild type, WT). The affinity of the native TCR is so low that the binding signal is undetectable using ELISA. Then, we made site-directed mutations on CDR2 and CDR3 of 1G4 ETCR according to the references, and used ETCR heterodimer format to detect the binding activity, as shown in Figure 6, proving the feasibility of TCR affinity maturation.

[0063] Example 2: Recombining TCR variable domains with antibody constant domains to generate TCR antibody chimeric proteins (ETCRs) 1. TCR sequence HLA*A*02:01NY-ESO-1 (SLLMWITQC)-specific TCR with a non-native disulfide bond between CαS48-CβT57, named CTCR1 (SEQ ID NOs: 37-40 and 63-66, the amino acid sequence of Vα is shown as SEQ ID NO: 37, the amino acid sequence of Vβ is shown as SEQ ID NO: 39, the amino acid sequence of Cα is shown as SEQ ID NO: 63, and the amino acid sequence of Vβ is shown as SEQ ID NO: 65), and CTCR2 An HLA*A*02:01GP100(YLEPGPVTV)-specific TCR with a non-native disulfide bond between CαS48-CβT57, named YLEPGPVTV (SEQ ID NOs: 41-44 and 63-66, the amino acid sequence of Vα is shown as SEQ ID NO: 41, the amino acid sequence of Vβ is shown as SEQ ID NO: 43, the amino acid sequence of Cα is shown as SEQ ID NO: 63, and the amino acid sequence of Vβ is shown as SEQ ID NO: 65) was selected for proof-of-concept studies. The IMGT numbering convention was used for all TCR variable domains.

[0064] Another HLA*A*02:01NY-ESO-1 (SLLMWITQC) specific TCR with a non-native disulfide bond between CαS48-CβT57, designated CTCR3 (SEQ ID NOs: 45-48 and 63-66, the amino acid sequence of Vα is shown as SEQ ID NO: 45, the amino acid sequence of Vβ is shown as SEQ ID NO: 47, the amino acid sequence of Cα is shown as SEQ ID NO: 63, and the amino acid sequence of Vβ is shown as SEQ ID NO: 65) was also selected for further testing. The IMGT numbering convention was used for all TCR variable domains.

[0065] SEQ ID NO: 37, CAb1-NY-ESO-1_VαAA: AQSVAQPEDQVNVAEGNPLTVKCTYSVSGNPYLFWYVQYPNRGLQFLLKYLGDSALVKGSYGFEAEFNKSQTSFHLKKPSALVSDSALYFCAVRDIRSGAGSYQLTFGKGTKLSVIP SEQ ID NO: 38, CAb1-NY-ESO-1_Vα DNA: gcccagtccgtggctcagcccgaggaccaagtgaacgtggccgagggcaaccctctgaccgtgaagtgcacctattccgtgagcggcaacccctatctgttttggtacgtgcagtaccccaacagaggactgcagttttctgctgaagtatctgggagacagcgctctggtgaagg gaagctacggcttcgaagccgagttcaacaagagccagacctccttccatctgaagaagcctagcgctctggtgagcgactccgctctgtacttctgcgccgtcagagacatcagaagcggcgccggaagctaccagctgaccttcggcaagggcaccaagctgagcgtgatccct SEQ ID NO: 39, CAb1-NY-ESO-1_VβAA: SAVISQKPSRDIKQRGTSLTIQCQVDKRLALMFWYRQQPGQSPTLIATAWTGGEATYESGFVIDKFPISRPNLTFSTLTTVSNMSPEDSSIYLCSVGGSGAADTQYFGPGTRLTVL SEQ ID NO: 40, CAb1-NY-ESO-1_VβDNA: agcgccgtgatcagccagaagcctagcagagacatcaaacagaggggcacatctctgaccatccagtgccaagtggacaagagactcgctctgatgttctggtatagacagcagcccggacagtccccccacactgatcgccaccgcttggaccggcggagaagccacctacg agtccggcttcgtgatcgacaagttccccatctctagacccaatctgaccttttccacactgaccgtgtccaacatgagccccgaggactccagcatttatctgtgtagcgtgggaggcagcggagctgccgatacccagtacttcggccccggaaccagactgaccgtgctg SEQ ID NO: 41, CAb2-GP100_VαAA: AQQGEEDPQALSIQEGENATMNCSYKTSINNLQWYRQNSGRGLVHLILIRSNEREKHSGRLRVTLDTSKKSSSLLITASRAADTASYFCATDGSTPMQFGKGTRLSVIA SEQ ID NO: 42, CAb2-GP100_Vα DNA: gctcagcaaggcgaagaggatccccaagctctgagcattcaagagggcgagaacgccaccatgaactgctcctacaagaccagcatcaacaacctccagtggtatagacagaacagcggcagaggactggtgcatctgattctgattagaagcaacgagagagagaagcactccggaaggctgagggtgacactggatacaagcaagaagagcagctctctgctgatcaccgcttccagagccgctgacaccgccagctacttctgcgccaccgacggcagcacccctatgcagttcggcaagggcacaagactcagcgtgatcgcc Sequence number 43, CAb2-GP100_VβAA: DGGITQSPKYLFRKEGQNVTLSCEQNLNHDAMYWYRQDPGQGLRLIYYSWAQGDFQKGDIAEGYSVSREKKESFPLTVTSAQKNPTAFYLCASSWGAPYEQYFGPGTRLTVT Sequence number 44, CAb2-GP100_VβDNA: gacggcggcatcacccagtcccccaagtatctgtttagaaaggagggccagaatgtgacactgagctgcgagcagaatctgaaccacgacgccatgtactggtacagacaagaccccggccaaggactgaggctgatctattacagctgggcacaaggagacttccagaagggcgacatcgccgagggatacagcgtgtctagagagaagaaggagagctttcctctgaccgtgaccagcgcccagaagaatcccaccgccttctatctgtgtgccagcagctggggagctccctacgagcagtatttcggacccggcacaagactgaccgtgaca Sequence number 45, CAb3-NY-ESO-1_VαAA: QEVTQIPAALSVPEGENLVLNCSFTDSAIYNLQWFRQDPGKGLTSLLLITPWQREQTSGRLNASLDKSSGRSTLYIAASQPGDSATYLCAVRPLLDGTYIPTFGRGTSLIVHP Sequence number 46, CAb3-NY-ESO-1_Vα DNA: caagaagtgacacagatccctgccgctctgtctgtgcctgagggcgaaaacctggtgctgaactgcagcttcaccgacagcgccatctacaacctgcagtggttcagacaggaccccggcaagggactgacaagcctgctgctgattaccccttggcagagagagcagaccagcggcagactgaatgccagcctggataagtcctccggcagaagcaccctgtatatcgccgcttctcagcctggcgatagcgccacatatctgtgtgccgtcagacccctgctggacggcacatatatccccacctttggcagaggcaccagcctgatcgtgcaccct Sequence number 47, CAb3-NY-ESO-1_Vβ AA: GVTQTPKFQVLKTGQSMTLQCAQDMNHEYMSWYRQDPGMGLRLIHYSVAIQTTDRGEVPNGYNVSRSTIEDFPLRLLSAAPSQTSVYFCASSYLGNTGELFFGEGSRLTVL Sequence number 48, CAb3-NY-ESO-1_Vβ DNA: ggagttacacagacccctaagttccaggtgctgaaaaccggccagagcatgaccctgcagtgcgcccaggatatgaaccacgagtacatgagctggtacaggcaggatccaggcatgggcctgagactgatccactactctgtggccatccagaccaccgacagaggcgaagtgcccaacggctacaacgtgtccagatccaccatcgaggacttcccactgagactgctgtctgctgcccctagccagacctccgtgtacttttgtgccagcagctacctgggcaacaccggcgagctgttttttggcgagggctccagactgaccgtgctg SEQ ID NO: 49, anti-CD3-scFv AA: AIQMTQSPSSLSASVGDRVTITCRASQDIRNYLNWYQQKPGKAPKLLIYYTSRLESGVPSRFSGSGSGTDYTLTISSLQPEDFATYYCQQGNTLPWTFGQGTKVEIKGGGGSGGGGSGGGGSGGGGSGGGGSGGGSEVQLVESGGGLVQPGGSLRLSCAASGYSFTGYTMNWVRQAPGKGLEWVALINPYKGVSTYNQKFKDRFTISVDKSKNTAYLQMNSLRAEDTAVYYCARSGYYGDSDWYFDVWGQGTLVTVSS SEQ ID NO: 50, anti-CD3-scFv DNA: gccatccagatgacgcaaagtccatcaagtctgagcgccagcgtgggcgacagagtgaccatcacctgcagagccagccaggacatcagaaattacctgaattggtaccagcagaagcctggcaaggctccaaagctcctcatatattatacatcgagattagaatctggtgttccaagcagattcagcggcagcggcagcggcaccgactacaccctgaccatcagcagcctgcagcctgaggacttcgccacctactactgccagcagggcaataccctgccttggacatttggacagggtaccaaggtggaaattaaaggcggcggcggaagcggaggcggagggtcgggtggcggaggttcaggtggaggagggtctggtggaggctcagaggtacaacttgtggagtcaggcggtggactagtccaaccaggaggatctttacgcttatcttgtgccgccagcggctacagcttcaccggctacaccatgaattgggtgagacaggctcccggtaagggcctggagtgggtggccctgatcaatccttacaagggcgtgagcacctacaatcagaagttcaaggacagattcaccatcagcgtggacaagagcaagaataccgcctacctgcagatgaatagcctgagagccgaggacaccgccgtgtactactgcgccagaagcggctactacggcgacagcgactggtactttgatgtttgggggcaaggtacacttgtcactgtaagctcc Sequence number 51, CAb1-NY-ESO-1_αFL AA: AQSVAQPEDQVNVAEGNPLTVKCTYSVSGNPYLFWYVQYPNRGLQFLLKYLGDSALVKGSYGFEAEFNKSQTSFHLKKPSALVSDSALYFCAVRDIRSGAGSYQLTFGKGTKLSVIPNIQNPDPAVYQLRDSKSSDKSVCLFTDFDSQTNVSQSKDSDVYITDKCVLDMRSMDFKSNSAVAWSNKSDFACANAFNNSIIPEDT SEQ ID NO: 52, CAb1-NY-ESO-1_αFL DNA: gcccagtccgtggctcagcccgaggaccaagtgaacgtggccgagggcaaccctctgaccgtgaagtgcacctattccgtgagcggcaacccctatctgttttggtacgtgcagtaccccaacagaggactgcagtttctgctgaagtatctgggagacagcgctctggtgaagggaagctacggcttcgaagccgagttcaacaagagccagacctccttccatctgaagaagcctagcgctctggtgagcgactccgctctgtacttctgcgccgtcagagacatcagaagcggcgccggaagctaccagctgaccttcggcaagggcaccaagctgagcgtgatccctaacatccagaaccccgatcccgccgtgtaccagctgagggacagcaagtccagcgacaagtccgtgtgtctgttcaccgacttcgactcccagaccaacgtgtcccagagcaaggatagcgacgtgtacatcaccgacaagtgcgtcctcgacatgaggtccatggacttcaagagcaacagcgccgtggcttggagcaacaagagcgacttcgcttgcgccaacgccttcaacaacagcatcatccccgaggacacc SEQ ID NO: 53, CAb1-NY-ESO-1_βFL AA: SAVISQKPSRDIKQRGTSLTIQCQVDKRLALMFWYRQQPGQSPTLIATAWTGGEATYESGFVIDKFPISRPNLTFSTLTVSNMSPEDSSIYLCSVGGSGAADTQYFGPGTRLTVLEDLKNVFPPEVAVFEPSEAEISHTQKATLVCLATGFYPDHVELSWWVNGKEVHSGVCTDPQPLKEQPALDNSRYALSSRLRVSATFWQDPRNHFRCQVQFYGLSENDEQDWTRAKPVTQIVSAEAWGRAD sequence number 54、CAb1-NY-ESO-1_βFL DNA: agcgccgtgatcagccagaagcctagcagagacatcaaacagaggggcacatctctgaccatccagtgccaagtggacaagagactcgctctgatgttctggtatagacagcagcccggacagtcccccacactgatcgccaccgcttggaccggcggagaagccacctacgagtccggcttcgtgatcgacaagttccccatctctagacccaatctgaccttttccacactgaccgtgtccaacatgagccccgaggactccagcatttatctgtgtagcgtgggaggcagcggagctgccgatacccagtacttcggccccggaaccagactgaccgtgctggaggatctgaagaacgtgtttccccccgaggtggccgtgtttgagcccagcgaggccgagattagccacacccagaaggccacactggtgtgtctggccaccggcttttaccccgaccacgtggaactgagctggtgggtgaacggcaaggaggtgcactccggcgtgtgtaccgatccccagcctctgaaggagcagcccgccctcaacgatagcagatacgctctgtcctccagactgagagtgagcgccacattctggcaagaccccagaaaccactttagatgccaagtgcagttctacggactgagcgaaaacgacgagtggacacaagatagagccaagcccgtgacccagatcgtgagcgccgaggcttggggcagagccgat Accession No. 55, CAb2-GP100_αFL AA: AQQGEEDPQALSIQEGENATMNCSYKTSINNLQWYRQNSGRGLVHLILIRSNEREKHSGRLRVTLDTSKKSSSLLITASRAADTASYFCATDGSTPMQFGKGTRLSVIANIQKPDPAVYQLRDSKSSDKSVCLFTDFDSQTNVSQSKDSDVYITDKCVLDMRSMDFKSNSAVAWSNKSDFACANAFNNSIIPEDT Accession No. 56, CAb2-GP100_αFL DNA: gctcagcaaggcgaagaggatccccaagctctgagcattcaagagggcgagaacgccaccatgaactgctcctacaagaccagcatcaacaacctccagtggtatagacagaacagcggcagaggactggtgcatctgattctgattagaagcaacgagagagagaagcactccggaaggctgagggtgacactggatacaagcaagaagagcagctctctgctgatcaccgcttccagagccgctgacaccgccagctacttctgcgccaccgacggcagcacccctatgcagttcggcaagggcacaagactcagcgtgatcgccaacatccagaagcccgaccccgccgtgtaccagctgagagactccaagagcagcgacaagagcgtgtgtctgttcaccgacttcgactcccagaccaacgtgagccagtccaaggacagcgacgtgtacatcaccgacaagtgcgtgctggacatgaggagcatggacttcaagtccaacagcgccgtggcttggtccaacaaatccgatttcgcttgcgccaatgccttcaacaactccatcatccccgaggacaca Accession No. 57, CAb2-GP100_βFL AA: DGGITQSPKYLFRKEGQNVTLSCEQNLNHDAMYWYRQDPGQGLRLIYYSWAQGDFQKGDIAEGYSVSREKKESFPLTVTSAQKNPTAFYLCASSWGAPYEQYFGPGTRLTVTEDLKNVFPPEVAVFEPSEAEISHTQKATLVCLATGFYPDHVELSWWVNGKEVHSGVCTDPQPLKEQPALNDSRYALSSRLRVSATFWQDPRNHFRCQVQFYGLSENDEWTQDRAKPVTQIVSAEAWGRAD Accession No. 58, CAb2-GP100_βFL DNA: gacggcggcatcacccagtcccccaagtatctgtttagaaaggagggccagaatgtgacactgagctgcgagcagaatctgaaccacgacgccatgtactggtacagacaagaccccggccaaggactgaggctgatctattacagctgggcAcaagggacttccagaagggcgacatcg ccgagggatacagcgtgtctagagagaagaaggagagctttcctctgaccgtgaccagcgcccagaagaatcccaccgccctttatctgtgtgccagcagctggggagctccctacgagcagtatttcggacccggcacaagactgaccgtgacagaggatctgaagaacgtcttccctccc gaggtggctgtgttcgagccctccgaggccgagatctcccacacccagaaggccaccctcgtgtgtctggctaccggcttctaccccgaccacgtggagctgagctggtgggtgaacggcaaagaggtgcatagcggcgtgtgtaccgacccccagcctctgaaagagcaacccgctctga acgactccagatacgctctgtcctccagactgagggtctccgccacattttggcaagaccctagaaaccactttagatgtcaagtgcagttctacggactgagcgagaatgatgagtggacacaagacagagccaagcccgtgacacagattgtcagcgccgaggcttggggaagagctgat sequence number 59、CAb3-NY-ESO-1_VαAA: QEVTQIPAALSVPEGENLVLNCSFTTDSAIYNLQWFRQDPGKGLTSLLLITPWQREQTSGRLNASLDKSSGRSTLYIAASQPGDSATYLCAVRPLLDGTYIPTFGRGTSLIVHP sequence number 60、CAb3-NY-ESO-1_VαDNA: caagaagtgacacagatccctgccgctctgtctgtgcctgagggcgaaaacctggtgctgaactgcagcttcaccgacagcgccatctacaacctgcagtggttcagacaggaccccggcaagggactgacaagcctgctgctgattaccccttggcagagagagcagaccagcggcagactgaatgccagcctggataagtcctccggcagaagcaccctgtatatcgccgcttctcagcctggcgatagcgccacatatctgtgtgccgtcagacccctgctggacggcacatatatccccacctttggcagaggcaccagcctgatcgtgcaccct Accession No. 61, CAb3-NY-ESO-1_VβAA: GVTQTPKFQVLKTGQSMTLQCAQDMNHEYMSWYRQDPGMGLRLIHYSVAIQTTDRGEVPNGYNVSRSTIEDFPLRLLSAAPSQTSVYFCASSYLGNTGELFFGEGSRLTVL Accession No. 62, CAb3-NY-ESO-1_VβDNA: ggagttacacagacccctaagttccaggtgctgaaaaccggccagagcatgaccctgcagtgcgcccaggatatgaaccacgagtacatgagctggtacaggcaggatccaggcatgggcctgagactgatccactactctgtggccatccagaccaccgacagaggcgaagtgcccaacggctacaacgtgtccagatccaccatcgaggacttcccactgagactgctgtctgctgcccctagccagacctccgtgtacttttgtgccagcagctacctgggcaacaccggcgagctgttttttggcgagggctccagactgaccgtgctg Accession No. 63, CαAA: NIQKPDPAVYQLRDSKSSDKSVCLFTDFDSQTNVSQSKDSDVYITDKCVLDMRSMDFKSNSAVAWSNKSDFACANAFNNSIIPEDT Accession No. 64, Cα cDNA: aacatccagaagcccgaccccgccgtgtaccagctgagagactccaagagcagcgacaagagcgtgtgtctgttcaccgacttcgactcccagaccaacgtgagccagtccaaggacagcgacgtgtacatcaccgacaagtgcgtgctggacatgaggagcatggacttcaagtccaacagcgccgtggcttggtccaacaaatccgatttcgcttgcgccaatgccttcaacaactccatcatccccgaggacaca Accession No. 65, Cβ AA: EDLKNVFPPEVAVFEPSEAEISHTQKATLVCLATGFYPDHVELSWWVNGKEVHSGVCTDPQPLKEQPALNDSRYALSSRLRVSATFWQDPRNHFRCQVQFYGLSENDEWTQDRAKPVTQIVSAEAWGRAD Accession No. 66, Cβ cDNA: gaggatctgaagaacgtcttccctcccgaggtggctgtgttcgagccctccgaggccgagatctcccacacccagaaggccaccctcgtgtgtctggctaccggcttctaccccgaccacgtggagctgagctggtgggtgaacggcaaagaggtgcatagcggcgtgtgtaccgacccccagcctctgaaagagcaacccgctctgaacgactccagatacgctctgtcctccagactgagggtctccgccacattttggcaagaccctagaaaccactttagatgtcaagtgcagttctacggactgagcgagaatgatgagtggacacaagacagagccaagcccgtgacacagattgtcagcgccgaggcttggggaagagctgat 2. Generation of TCR-Antibody Chimeric Proteins (ETCR) The constant domains Cα and Cβ of CTCR1 and CTCR2 were replaced by the constant domains CH1 and Cλ / Cκ of IgA, IgD, IgE, IgG and IgM antibodies, either fused or not to an Fc domain, to generate multiple ETCRs for further analysis.

[0066] SEQ ID NO:1, modified Cλ1AA: PTVTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKADGSPVKAGVETTKPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEKTVAPTECS SEQ ID NO:2, modified Cλ1 DNA: cccacggtcactctgttcccgccctcctctgaggagctccaagccaacaaggccacactagtgtgtctgatcagtgacttctacccgggagctgtgacagtggcttggaaggcagatggcagccccgtcaaggcgggagtggagacgacc aaaccctccaaacagagcaacaacaagtacgcggccagcagctacctgagcctgacgcccgagcagtggaagtcccacagaagctacagctgccaggtcacgcatgaagggagcaccgtggagaagacagtggcccctacagaatgttca SEQ ID NO:3, modified Cλ2AA: P S VTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKAD S SPVKAGVETT T PSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEKTVAPTECS SEQ ID NO:4, modified Cλ2 DNA: ccctcggtcactctgttcccgccctcctctgaggagcttcaagccaacaaggccacactggtgtgtctcataagtgacttctacccgggagccgtgacagtggcttggaaagcagatagcagccccgtcaaggcgggagtggagaccaccacaccctccaaacaaagcaacaacaagtacgcggccagcagctatctgagcctgacgcctgagcagtggaagtcccacagaagctacagctgccaggtcacgcatgaagggagcaccgtggagaagacagtggcccctacagaatgttca Sequence number 5, modified Cλ3 AA: PSVTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKADSSPVKAGVETTTPSKQSNNKYAASSYLSLTPEQWKSHKSYSCQVTHEGSTVEKTVAPTECS Sequence number 6, modified Cλ3 DNA: ccctcggtcactctgttcccaccctcctctgaggagcttcaagccaacaaggccacactggtgtgtctcataagtgacttctacccgggagccgtgacagttgcctggaaggcagatagcagccccgtcaaggcgggggtggagaccaccacaccctccaaacaaagcaacaacaagtacgcggccagcagctacctgagcctgacgcctgagcagtggaagtcccacaaaagctacagctgccaggtcacgcatgaagggagcaccgtggagaagacagttgcccctacggaatgttca Sequence number 7, modified Cλ6 AA: PSVTLFPPSSEELQANKATLVCLISDFYPGAVKVAWKADGSPVNTGVETTTPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEKTVAPAECS Sequence number 8, modified Cλ6 DNA: ccatcggtcactctgttcccgccctcctctgaggagcttcaagccaacaaggccacactggtgtgcctgatcagtgacttctacccgggagctgtgaaagtggcctggaaggcagatggcagccccgtcaacacgggagtggagaccaccacaccctccaaacagagcaacaacaagtacgcggccagcagctacctgagcctgacgcctgagcagtggaagtcccacagaagctacagctgccaggtcacgcatgaagggagcaccgtggagaagacagtggcccctgcagaatgttca SEQ ID NO: 9, Modified Cλ7 AA: PSVTLFPPSSEELQANKATLVCLVSDFYPGAVTVAWKADGSPVKVGVETTKPSKQSNNKYAASSYLSLTPEQWKSHRSYSCRVTHEGSTVEKTVAPAECS SEQ ID NO: 10, Modified Cλ7 DNA: ccctcggtcactctgttcccaccctcctctgaggagcttcaagccaacaaggccacactggtgtgtctcgtaagtgacttctacccgggagccgtgacagtggcctggaaggcagatggcagccccgtcaaggtgggagtggagaccaccaaaccctccaaacaaagcaacaacaagtatgcggccagcagctacctgagcctgacgcccgagcagtggaagtcccacagaagctacagctgccgggtcacgcatgaagggagcaccgtggagaagacagtggcccctgcagaatgctct SEQ ID NO: 11, Modified IgG1 CH1 AA: TKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKV SEQ ID NO: 12, Modified IgG1 CH1 DNA: accaagggcccatcggtcttccccctggcaccctcctccaagagcacctctgggggcacagcggccctgggctgcctggtcaaggactacttccccgaaccggtgacggtgtcgtggaactcaggcgccctgaccagcggcgtgcacaccttcccggctgtcctacagtcctcaggactctactccctcagcagcgtggtgaccgtgccctccagcagcttgggcacccagacctacatctgcaacgtgaatcacaagcccagcaacaccaaggtggacaagaaagttg Accession No. 13, Modified IgG2 CH1 AA: TKGPSVFPLAPCSRSTSESTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSNFGTQTYTCNVDHKPSNTKVDKTV Accession No. 14, Modified IgG2 CH1: accaagggcccatcggtcttccccctggcgccctgctccaggagcacctccgagagcacagccgccctgggctgcctggtcaaggactacttccccgaaccggtgacggtgtcgtggaactcaggcgctctgaccagcggcgtgcacaccttcccagctgtcctacagtcctcaggactctactccctcagcagcgtggtgaccgtgccctccagcaacttcggcacccagacctacacctgcaacgtagatcacaagcccagcaacaccaaggtggacaagacagtt Accession No. 15, Modified IgG3 CH1 AA: TKGPSVFPLAPCSRSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYTCNVNHKPSNTKVDKRV Accession No. 16, Modified IgG3 CH1 DNA: accaagggcccatcggtcttccccctggcgccctgctccaggagcacctctgggggcacagcggccctgggctgcctggtcaaggactacttccccgaaccggtgacggtgtcgtggaactcaggcgccctgaccagcggcgtgcacaccttcccggctgtcctacagtcctcaggactctactccctcagcagcgtggtgaccgtgccctccagcagcttgggcacccagacctacacctgcaacgtgaatcacaagcccagcaacaccaaggtggacaagagagtt Accession No. 17, Modified IgG4 CH1 AA: TKGPSVFPLAPCSRSTSESTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTKTYTCNVDHKPSNTKVDKRV Accession No. 18, Modified IgG4 CH1 DNA: accaagggcccatccgtcttccccctggcgccctgctccaggagcacctccgagagcacagccgccctgggctgcctggtcaaggactacttccccgaaccggtgacggtgtcgtggaactcaggcgccctgaccagcggcgtgcacaccttcccggctgtcctacagtcctcaggactctactccctcagcagcgtggtgaccgtgccctccagcagcttgggcacgaagacctacacctgcaacgtagatcacaagcccagcaacaccaaggtggacaagagagtt 3. Materials and Methods 3.1 Modeling of Antibody and TCR Homology Antibody and TCR structural models were constructed based on the amino acid sequences using MODELLER. Then, all modeled segments were assembled to construct α- and β-chimeric chain structural models. The relative orientation between the two modeled chains was predicted by taking the angle of the TCR structure with the most similar overall sequence. All molecular visualization and analysis work was performed using PyMOL software (Schrodinger).

[0067] 3.2 DNA manipulation CTCR1 and CTCR2 genes were synthesized by Genewiz Inc. CH1 and Cλ / Cκ genes were amplified by PCR from existing in-house DNA templates. For ETCR fused with Fc domain chimera (IgG-like ETCR), the gene product of light chain recombinant was inserted into a linearized vector containing CMV promoter, kappa signal peptide and WPRE regulator, while the gene product of heavy chain recombinant was inserted into a linearized vector containing human corresponding constant region CH2-CH3, CMV promoter and human antibody heavy chain signal peptide. For ETCR not fused with Fc domain chimera (Fab-like ETCR), each of light chain recombinant and heavy chain recombinant was inserted into a linearized vector containing CMV promoter, kappa signal peptide and WPRE regulator, respectively. Plasmid ligation, transformation and DNA preparation were performed using standard molecular biology protocols.

[0068] 3.3 Protein expression The constructed vectors of heavy and light chains were co-transfected into Expi293 cells (Thermofisher Scientific). The ratio of different vectors for co-transfection was optimized according to the predicted ETCR structure and the initial expression results shown in SDS-PAGE and Western blot. The transfection procedure followed the manual provided by the supplier. Briefly, 2.5 μg each of plasmid and 2.94 × 10 β-lactamase were transfected into Expi293 cells (Thermofisher Scientific). 6100 cells were transfected with 100% EDTA. Enhancer 1 and Enhancer 2 were added 20 hours post-transfection. Transfected cells were cultured at 37°C with 8% CO2 and 85% humidity on an orbital shaker, rotating at either 120 rpm (flasks) or 200 rpm (50 ml tubes). Five days post-transfection, supernatants were harvested by centrifugation and cell debris was removed by 0.22 μm filtration. Treated supernatants were concentrated as necessary before further testing.

[0069] 3.4 Measurement of ETCR concentration by ELISA For IgG-like ETCR, ELISA plates were coated with 1 μg / ml anti-human FC antibody in coating buffer (200 mM Na2CO3 / NaHCO3, pH 9.2). After overnight incubation at 4°C, plates were washed once with PBST washing buffer using a deep-well washer (Biotek ELx405). Plates were then blocked with 1% casein and incubated for 1 h at room temperature. Plates were washed 3 times with washing buffer and 100 μl of positive control (if present), negative control (if present) and diluted samples were added and incubated for 1 h at room temperature. Plates were washed 3 times with washing buffer and 100 μl of HRP-conjugated anti-lambda or anti-kappa antibody was added and incubated for 1 h at room temperature. The plate was washed six times with wash buffer, 100 μl of TMB was added and incubated for 10 minutes, then 100 μl of stop solution (2 M HCl, 100 μl / well) was added and the absorbance was read at 450 nm using a plate reader (Molecular Devices SpectraMaxMV5e).

[0070] For Fab-like ETCRs, ELISA plates were coated with 0.5 μg / ml anti-His antibody in coating buffer (200 mM Na2CO3 / NaHCO3, pH 9.2). After overnight incubation at 4°C, plates were washed once with PBST washing buffer using a deep-well washer (Biotek ELx405). Plates were then blocked with 1% casein and incubated for 1 h at room temperature. Plates were washed 3 times with washing buffer and 100 μl of positive control (if present), negative control (if present) and diluted samples were added and incubated for 1 h at room temperature. Plates were washed 3 times with washing buffer and 100 μl of HRP-conjugated anti-lambda or anti-kappa antibody was added and incubated for 1 h at room temperature. The plate was washed six times with wash buffer, 100 μl of TMB was added and incubated for 10 minutes, then 100 μl of stop solution (2 M HCl, 100 μl / well) was added and the absorbance was read at 450 nm using a plate reader (Molecular Devices SpectraMaxMV5e).

[0071] 3.5 Measuring target binding by ELISA ELISA plates were coated with 2 μg / ml streptavidin (SA) in coating buffer (200 mM Na2CO3 / NaHCO3, pH 9.2). After overnight incubation at 4°C, the plates were washed once with PBST washing buffer using a deep well washer (Biotek ELx405). The plates were then blocked with 1% casein and incubated for 1 h at room temperature. The plates were washed 3 times with washing buffer and 2.5 μg / ml of antigens HLA*A*02:01NY-ESO-1 (SLLMWITQC), HLA*A*02:01GP100 (YLEPGPVTV) (pHLA, provided by Kactus bio) were added and incubated for 1 h at room temperature. The plates were washed 3 times with washing buffer and positive control (if present), negative control (if present) and diluted samples were added and incubated for 1 h at room temperature. The plate was washed 3 times with wash buffer, 100 μl of HRP-conjugated anti-cmyc antibody was added and incubated for 1 hour at room temperature, the plate was washed 6 times with wash buffer, 100 μl of TMB was added and incubated for 10 minutes, then 100 μl of stop solution (2M HCl, 100 μl / well) was added and the absorbance was read at 450 nm using a plate reader (Molecular Devices SpectraMaxMV5e).

[0072] 4.Results As TCRs are naturally membrane proteins, their transfer to soluble forms does not always result in favorable drug-like properties. However, as reported decades ago, the introduction of an artificial disulfide bond CαS48-CβT57 into the native TCR non-covalent Cα-Cβ resulted in a stable enhancement of the soluble TCR form (Figure 7A, US Patent No. 7,666,604 B2). In contrast to the native TCR, the native disulfide bond was present in the antibody constant domain (Figure 7B). This may contribute to the stability of the chimera.

[0073] The inventors first evaluated the ability of CH1 and Cλ / Cκ to replace Cα and Cβ. Either CTCR1 or CTCR2 variable domain was recombined with the constant domain from IgG antibody and fused to Fc domain to generate IgG-like ETCR. The expression and binding of such ETCR were further determined by ELISA. Tables 1 and 2 list the construction, expression and binding results of IgG-like ETCR, and the collected supernatant was concentrated 10-fold for ELISA analysis. In general, most of the constructs expressed well, but the binding signal was relatively low, indicating that the ETCR part of the chimeric IgG-like ETCR may not be folded or associated correctly. It seems impossible to stabilize the ETCR part by the Fc domain, which is expected to further enhance the ability of the chimeric TCR. Nevertheless, by analyzing the results, the inventors found that the "reverse" fusion pattern was better than the "forward" fusion pattern, and in almost all the tested samples, Vα fused to CH1 and Vβ fused to CL produced better binding ability than others.

[0074] [Table 1]

[0075] [Table 2]

[0076] In addition, additional constant domains from IgA, IgD, IgE and IgM were also recombined with CTCR1 and CTCR2 variable domains to generate Fab-like ETCRs for large-scale screening (based on previous results, in further screening and modification, Fc domains are not fused with ETCRs). The expression and binding of such ETCRs were further determined by ELISA. Tables 3 and 4 list the construction, expression and binding results of Fab-like ETCRs, and the collected supernatants were concentrated 10-fold for ELISA analysis. In general, low expression levels and binding abilities were observed when TCR variable domains were fused with constant domains from IgA, IgD, IgE and IgM.

[0077] Based on these results, further modifications will focus on a Fab-like ETCR with constant domains containing IgG CH1 and Cλ / Cκ fused in a "reverse" pattern.

[0078] [Table 3]

[0079] [Table 4]

[0080] Example 3: Design and modification of ETCR junction domain In general, the junction domain between the variable domain and the constant domain in both TCR and antibody is important for their stabilization and function.However, the linkers of IgG and TCR are somewhat different.Therefore, linkers with different types and lengths are inserted between the variable domain and the constant domain of each chain.Various ETCRs are produced and their expression levels and binding abilities are examined.

[0081] result First, conventional flexible linkers containing various lengths of serine and alanine were inserted as linkers between the TCR variable domain and the antibody constant domain (TCR Vα-CH1 fusions, SEQ ID NOs: 19-24 and TCR Vβ-Cλ / Cκ fusions, SEQ ID NOs: 25-30).

[0082] Table 5 lists exemplary construction, expression and binding results of Fab-like ETCRs with flexible linkers between the variable and constant domains, and harvested supernatants were concentrated 10-fold for ELISA analysis. Junction domains containing SSAS as a linker (α chain, SEQ ID NO: 19 and β chain, SEQ ID NO: 25) showed the highest expression levels and binding signals in all chimeric assemblies tested (Table 5), indicating that although a small degree of flexibility was introduced between the domains, the steric hindrance of the variable and constant domains was not eliminated and the original TCR function was not fully restored.

[0083] SEQ ID NO: 19, Junction domain 1AA: SSAS SEQ ID NO: 20, Junction domain 1 DNA: tcgtcggcttca SEQ ID NO: 21, Junction domain 2AA: SSASS SEQ ID NO: 22, Junction domain 2 DNA: tcgtcggcttcatcg SEQ ID NO: 23, Junction domain 3AA: SSASSS SEQ ID NO: 24, Junction domain 3 DNA: tcgtcggcttcatcgtca SEQ ID NO: 25, Junction domain 4AA: SSASKAA SEQ ID NO: 26, Junction domain 4 DNA: agttcggcctcaaaggctgcc SEQ ID NO: 27, Junction domain 5AA: SSASSKAA SEQ ID NO: 28, Junction domain 5 DNA: tcgtcggcttcatcgaaggctgcc SEQ ID NO: 29, Junction domain 6AA: SSASSSKAA SEQ ID NO: 30, Junction domain 6 DNA: tcgtcggcttcatcgtcaaaggctgcc

[0084] [Table 5]

[0085] Then, based on the structural alignment, the inventors carefully aligned the antibody and TCR sequences and found that the junction defined in the germline sequence does not always match the domain. The inventors confirmed how the antibody and TCR junction overlap on the overlapping structure, and predicted the potential substitutions using the N-terminus of the antibody constant domain from the TCR junction (Figure 8, indicated as black arrow). In particular, the alignment of the structure of the TCR constant domain with that of the antibody revealed that the FG and DE loops of the TCR beta chain are significantly longer than the corresponding regions of the antibody constant domain, forming a strong interaction with the TCR junction domain (Figure 8A, indicated as red arrow). Due to the absence of long FG and DE loops in the chimeric ETCR of the present invention, the key positions in the original TCR junction domain that are likely to have unsaturated charged amino acids need to be rationally mutated to enhance stability.

[0086] Based on this concept, two junction domains (L1 and L2) of the β chain (between the TCR Vβ domain and Cλ) were first designed and tested using the λG4-reverse constant domain as an exemplary scaffold (SEQ ID NOs: 33-36, the amino acid sequence of L1 is shown as SEQ ID NO: 33, the amino acid sequence of L2 is shown as SEQ ID NO: 35, the junction domain of the α chain is also a flexible linker, the amino acid sequence of the flexible linker is shown in SEQ ID NO: 19). Table 6 lists exemplary construction, expression and binding results of Fab-like ETCRs with designed linkers inserted between the variable and constant domains, and the collected supernatants were concentrated 10-fold for ELISA analysis. The clones of the β-chain with the designed linker inserted showed comparable expression but better binding signals compared to the λG4-reverse backbone using both CTCR1 and CTCR2 variable domains, indicating that the use of the designed linkers L1 and L2 in the β-chain as junction domains allows for better structural compatibility and results in more native TCR-like association. Therefore, the λG4-reverse with the flexible linker and the designed linker inserted in the α-chain and β-chain, respectively, was used as an example backbone for further engineering.

[0087] SEQ ID NO:31, Junction domain 7AA: EDLNKVFP SEQ ID NO:32, Junction domain 7 DNA: gaggacctgaacaaggtgttccca SEQ ID NO: 33, Junction domain 9AA: EDLSNVSP SEQ ID NO: 34, Junction domain 9 DNA: gaggacctgtccaatgtcagtccc SEQ ID NO: 35, Junction domain 8AA: EDLKNVFP SEQ ID NO:36, Junction domain 8 DNA: gaggacctgaaaaacgtgttccca

[0088] [Table 6]

[0089] Example 4: Design and modification of the Vβ-CL binding interface of ETCR By carefully analyzing the native TCR structure, we found that the binding of the variable and constant domains of native TCR is usually provided by three separate regions Vα-Cα, Vβ-Cβ and Vα-Cβ (Figure 9). Among them, the largest binding region in Vβ-Cβ was found to be highly organized and composed of several H-bonds and salt bridges as well as a hydrophobic core, and showed very strong binding affinity (Figure 10A, polar contacts are indicated by red arrows and yellow dashed lines, and non-polar contacts are indicated by orange circles). Then, we superimposed the chimeric ETCR structure onto the native TCR and further observed that the highly organized interactions of native TCR were completely disrupted by replacing Cβ with antibody constant domain Cλ (Figure 10B). By analyzing the overlapping model, we determined the key positions in Cλ that may contribute to the binding between Vβ and Cλ, and further designed and tested mutagen. Exemplary designs and mutations are listed below. The IMGT numbering convention was used for all TCR variable domains.

[0090] [Table 7]

[0091] material and method Protein purification 6xHis tagged proteins were purified by AKTApureM25 with NiSepharose™ Excel chromatography resin (GE Healthcare) on the column. Wash Buffer A: 50 mM sodium phosphate, 150 mM NaCl, pH 7.2. Wash Buffer B: 50 mM sodium phosphate, 150 mM NaCl, 500 mM imidazole, pH 7.2. The purification process is generally described as follows: The column is equilibrated with Wash Buffer A at 1 ml / min. The sample is applied at 1 ml / min using the sample inlet. The column is washed with Wash Buffer A at 1 ml / min. The column is washed with 2%, 4%, 10%, 100% Wash Buffer B. Fractions are collected at 1.0 ml / vial during the washing process.

[0092] The pre-purified protein can be further purified by AKTApure M25 with Superdex™ 75 / 200 increase chromatography resin (GE Healthcare) on the column. Washing buffer: 137 mM sodium phosphate, 2.68 mM NaCl, 1.76 mM KCl, 10 mM KH2PO4, 10 mM Na2HPO4, pH 7.4. The purification process is generally described as follows: Concentrate the protein by ultrafiltration to an appropriate loading volume. Wash the column with distilled water. Equilibrate the column with washing buffer. Apply the sample onto the column. Elute the column with washing buffer at 0.5 ml / min until no material appears in the flow-through. Store the purified protein at -80°C for future use.

[0093] Quantification of purified proteins The purified protein was preliminarily characterized using A280. The absorbance value of the protein solution at 280 nm was measured by Nanodrop2000 using 50 mM sodium phosphate, 150 mM NaCl, pH 7.2 as blank buffer. Protein concentration (mg / ml)=A280 / Extinction coefficient.

[0094] SDS-PAGE was also used for characterization. The running voltage was constant at 200 V for 35 min, and Coomassie blue was used for staining after electrophoresis.

[0095] The purity of the samples was finally determined using size-exclusion high performance liquid chromatography. An Agilent 1200 HPLC system with a TSK GEL G3000SWXL column was used. Wash buffer: 50 mM sodium phosphate. The brief process is generally described as follows: The column is equilibrated with 50 mM sodium phosphate, 150 mM NaCl, pH 7.0. The sample was applied onto the column and the UV absorbance at 280 nm was monitored. The purity was estimated by integrating the chromatogram.

[0096] result The λG4-reverse with flexible linker SSAS (SEQ ID NO: 19) and designed linker L1 (SEQ ID NO: 33) inserted into the α-chain and β-chain, respectively, was used as an example scaffold for further modification. Based on the structural analysis, key positions including G30, A31 and T33 in antibody Cλ were identified and mutated to G30D, A31H and T33E, respectively, to generate TCR-like interactions.

[0097] Single mutations as well as combinatorial mutations were generated, expressed, purified and characterized. Figure 11A shows the electrophoresis results of the supernatant of the exemplary mutagens, and clear bands were visible by SDS-PAGE, suggesting that such mutations significantly enhanced the expression of chimeric ETCR1. Further binding ELISA tests revealed binding abilities comparable to the native TCR CTCR1 (Figure 11B, slight differences may occur due to different detection tags), suggesting that the rational mutagen-mediated reconstructed interactions between G30D-Vβ123R, A31H-Vβ125T and T33E-Vβ10R strongly stabilized the whole ETCR1 structure. ETCR1 was well expressed and functioned with such mutations, whereas variability was observed for ETCR2 with the same mutagens.

[0098] To further improve the compatibility of the β chain, both the binding interfaces of ETCR1 and ETCR2 were carefully analyzed by overlapping again. The structural analysis revealed that the framework 1 region (FR1) of the TCR variable domain located at the VC binding interface is significantly different between ETCR1 and ETCR2, providing an explanation for the above-mentioned variability (Figure 12A, this difference is indicated by red arrows). Based on the λG4-reverse and bM1 designs in which the flexible linker SSAS (SEQ ID NO: 19) and the designed linker L1 (SEQ ID NO: 33) were inserted into the α chain and β chain formats, the positions selected according to the structural analysis of CTCR2FR1 were individually replaced with the corresponding amino acids of ETCR1FR1 to further improve the compatibility of ETCR2, and as a result, one position of R13 of the ETCR2 variable β domain (Vβ) was identified. Figure 12B shows an example electrophoresis result of the supernatant of the R13 mutagen, and a clear band was visible by SDS-PAGE, suggesting that the FR1 mutant significantly enhanced the expression of chimeric ETCR2 compared to bM1.Further Q-ELISA test also revealed the enhanced expression level of ETCR2-bM1 with R13K / T mutation compared to parental ETCR2-bM1 (957 / 755 nM vs. 208 nM), suggesting that the R13 mutation in Vβ together with the mutation in Cλ strongly stabilizes the whole ETCR2 structure.

[0099] Example 5: SPR analysis of ETCR The detailed binding ability of ETCR was then determined using SPR technology.

[0100] method The binding affinity of ETCR and MHC-peptide (pMHC) antigen was detected using BiacoreT200 (or Biocore8K). The general process is described as follows: pMHC antigen was immobilized on a CM5 sensor chip (GE). Step concentrations of analyte and running buffer (50 mM sodium phosphate, 150 mM NaCl, 0.05% Tween20, pH 7.4) were systematically injected into the chip at a flow rate of 30 μL / min for a 120 s association phase and a 2400 s dissociation phase. After each cycle, the sensor chip surface was fully regenerated with 10 mM glycine (pH 1.5). Surface channel Fc1 without capture ligand was used as a control surface for reference subtraction. The final data of each interaction was subtracted from the reference Fc1 and buffer channel data. The molar concentration of the analyte was calculated using a molecular weight of 50.5 kDa, and the experimental data was fitted by evaluation with Biacore8K.

[0101] result FIG. 13 shows the sensorgrams of ETCR and CTCR, and in general, very similar binding characteristics were observed from the sensorgrams. In detail, Table 13 and Tables 8-9 list the SPR results of exemplary ETCR1 and ETCR2. Chimeric ETCR and CTCR had qualitatively similar binding ability. In particular, ETCR had a better K than CTCR. on , which may benefit from a stable and compatible antibody constant domain compared to the CTCR.

[0102] [Table 8]

[0103] [Table 9]

[0104] Example 6: FACS analysis of ETCR The binding ability of ETCR was then assessed against tumor cell lines using FACS technology.

[0105] method The binding ability of the designed ETCR was evaluated using A375 tumor cell line (A375 is an HLA*A*02:01 and NY-ESO-1 double positive cell line). The cell line was obtained from the American Type Culture Collection (ATCC) and maintained in DMEM medium supplemented with 10% fetal bovine serum (FBS).

[0106] 10 per well 5 An aliquot of cells was taken and washed with 1% bovine serum albumin (BSA) before being incubated with serially diluted ETCR in a 96-well round-bottom plate for 1 h at 4°C. After washing three times with 1% BSA, the plate was further incubated with PE-conjugated goat anti-human c-myc antibody for 30 min at 4°C. After washing the plate again three times, the cells were analyzed by flow cytometry using a FACSCantoII cytometer (BD Biosciences) and the associated fluorescence intensity was quantified using FlowJo software. EC50 values ​​were obtained by Prism software (GraphPad Software, Inc) using a four-parameter nonlinear regression analysis.

[0107] result In Figure 14, we show the FACS results of an example ETCR1. The chimeras ETCR1-bM1 and CTCR1 had qualitatively similar binding properties. Nevertheless, ETCR1-bM2 had significantly better binding properties than CTCR1 and ETCR1-bM1, which was not shown by ELISA or SPR. We hypothesize that subtle structural differences in the global interactions of G30D-Vβ123R, A31H-Vβ125T and T33E-Vβ10R may be highlighted under conditions of low antigen density (10-50 copies of antigen per A375 cell) compared to the single interaction of T33E-Vβ10R.

[0108] Example 7: Design and modification of the Vβ framework domain of ETCR To further test the compatibility of the chimeric format of the present invention, the inventors fused the mutations introduced in FR1 of the modified antibody constant domain of the present invention and the TCR variable domain to different TCR germlines, but could not generate stable ETCR for one germline pair reported so far as 1G4.The inventors hypothesize that, apart from the constant domain and binding interface, the variable domain of TCR may also have an effect on the stability of ETCR.Therefore, the FR region of Vβ was then comprehensively designed for better stability.

[0109] TCR sequence Another HLA*A*02:01NY-ESO-1 (SLLMWITQC)-specific TCR (SEQ ID NOs: 45-48) with a non-native disulfide bond between CαS48-CβT57, designated CTCR3, was selected for testing. The IMGT numbering convention was used for all TCR variable domains.

[0110] material and method Antibody and TCR homology modelling Antibody and TCR structural models were constructed based on the amino acid sequences using MODELLER. All modeled segments were then combined to construct α- and β-chimeric chain structural models. The relative orientation between the two modeled chains was predicted by taking the angle of the TCR structure with the most similar overall sequence. All molecular visualization and analysis work was performed using PyMOL software (Schrodinger).

[0111] result First, the variable domain of CTCR3 was amplified and fused to a modified antibody constant domain containing the bM2 design (λG4-reverse with flexible linker SSAS (SEQ ID NO: 19) and designed linker L1 (SEQ ID NO: 35) inserted into the α and β chains). However, the resulting ETCR-bM2 expressed only approximately 50 nM and was not purifiable. To investigate whether further stabilization of the Vβ domain of the TCR would benefit TCR expression, stability and association when expressing soluble TCR in mammalian cells, the inventors used molecular modeling simulations to identify mutations that stabilize the Vβ domain. The inventors scanned across all residue positions in the FR region of the Vβ domain and used FoldX to model all potential point mutations (except cysteine) and calculate the energies of such mutations. By analyzing and ranking such energy data, 180 mutations from the theoretical 1500 mutations were selected for further experimental confirmation. Finally, the following mutations were identified that could significantly improve the expression level: M19Y (bM41), A24K (bM42), A24R (bM43), M48F (bM39), H54Y (bM44), H54W (bM45), H54A (bM40), N77E (bM46), R90T (bM37), R90V (bM47), and L91I (bM38). The stabilizing mutations were then combined into various variants containing 2 to 4 mutations. Among all the combinations, R90T-L91I (bM37-bM38) showed the highest expression level, reaching 1612 nM. Table 10 lists the SPR results of an example ETCR3. The chimeric ETCR and CTCR had qualitatively similar binding abilities, indicating that mutations in the TCR variable domain also contribute to ETCR stabilization.

[0112] [Table 10]

[0113] Example 8: Design and modification of bispecific ETCRs After successfully generating stable expressing ETCR and confirming that the chimeric format can bind to the natural ligand, the inventors went on to construct bispecific formats and test their functions in vitro. The λG4-reverse and bM1 designs, in which the flexible linker SSAS (SEQ ID NO: 19) and the designed linker L1 (SEQ ID NO: 33) were inserted into the α-chain and β-chain, respectively, were used as ETCR1 formats.

[0114] method DNA manipulations and plasmid construction Anti-CD3scFv antibody (SEQ ID NO: 49-50) genes were synthesized by Genewiz Inc. The gene products of anti-CD3scFv were amplified and inserted into the N-terminus of the TCR Vβ domain, the C-terminus of the antibody Cλ domain, the N-terminus of the TCR Vα domain, and the C-terminus of the antibody CH1 domain, respectively, to generate bispecific ETCR1-E1.1, ETCR1-E1.2, ETCR1-E1.3, and ETCR1-E1.4 (Figure 15A-D). For CTCR, the gene product of anti-CD3scFv was amplified and inserted into the N-terminus of the TCR Vβ domain to generate CTCR1-E1.1 (Figure 15E, reported format). Plasmid ligation, transformation, and DNA preparation were performed using standard molecular biology protocols.

[0115] Protein expression, purification and other characterization methods were as described above. result All anti-CD3scFv antibody (SEQ ID NO:50) ETCR fusions were successfully expressed and purified in Expi293 cells. Figure 16 shows the SDS-PAGE data of the produced bispecific ETCR proteins in the supernatant and after purification. All bands of the correct molecular weight, i.e., approximately 78 Kd, in the non-reducing gel were clearly observed. The purity achieved was over 99% when the purified samples were further investigated by SEC-HPLC. The data showed that the ETCR bispecific proteins were successfully expressed and assembled.

[0116] The binding properties of all bispecific ETCRs were then examined by SPR. Table 11 shows the SPR results of an exemplary bispecific ETCR1. Fusion of anti-CD3 scFv at different positions did not significantly affect the binding affinity of bispecific ETCR1, which showed a better K on Similar binding properties, but slightly better, were again observed by E1.1 and E1.3. Nevertheless, variable results were observed with respect to the binding properties of anti-CD3scFv. The binding affinity of ETCR1-E1.1 and ETCR-E1.3, in which anti-CD3scFv is conjugated to the N-terminus of the TCR Vα and Vβ domains, was almost 10-fold higher than that of ETCR-E1.2 and ETCR-E1.4, in which anti-CD3scFv is conjugated to the C-terminus of the antibody CH1 and Cλ domains, indicating that the steric hindrance of the antibody constant domain appears to be stronger than that of the TCR variable domain.

[0117] [Table 11]

[0118] Example 9: In vitro T2 cell killing assay of bispecific ETCR In vitro functional assays were performed to confirm the activity of the designed bispecific ETCR in T cell-engaged killing of T2 antigen-presenting cells loaded with specific peptides. T2 cells are HLA*A*02:01 positive and, in particular, are defective in peptide transporters involved in antigen processing (TAP), and thus cannot properly translocate endogenous (processed) peptides to the site of MHC loading in the endoplasmic reticulum, the Golgi apparatus. Thus, peptide-pulsed T2 cells can be used to monitor cytotoxic T cell responses against exogenous antigens of interest in a non-competitive environment. Compared to tumor cell lines, T2 cells loaded with specific peptides successfully delivered high antigen density, resulting in better killing properties.

[0119] method Peripheral blood mononuclear cells (PBMCs) from healthy donors were freshly isolated from heparinized venous blood by density centrifugation with Ficoll-Paque PLUS (GE Healthcare-17-1440-03). After 6 days of culture in RPMI1640 medium supplemented with 10% FBS, 1% penicillin / streptomycin solution, 50 units of human IL-2 ligand protein per ml, and 10 ng / ml OKT3 antibody, PBMCs were passed through an EasySep column to enrich for CD8+ T cells. CD8+ T cells from the negative selection column were used as effector cells.

[0120] T2 cells (174 9CEM.T2, ATCC CRL-1992™) were obtained from ATCC and maintained at 37° C., 5% CO2 in IMDM medium supplemented with 20% FBS and penicillin / streptomycin. Prior to use, they were counted to 1×10 6 The cells were resuspended in medium to 100 μg / ml and pulsed with peptides at a peptide concentration of 20 μg / ml for 90 minutes in a 37° C., 5% CO2 incubator. The pulsed T2 cells were then labeled with 20 nM Far-Red in DPBS for 30 minutes, then washed twice and resuspended to the designated cell density.

[0121] The cells and bispecific ETCR were then mixed and incubated for the designated time. For analysis, 100 μl of PI (1:500 dilution in PBS) was added to each well and FACS was performed.

[0122] result In vitro functional assays were performed to confirm the activity of the designed bispecific ETCR in T cell-engaged killing of antigen-presenting cells T2 loaded with specific peptides. T2 cells loaded with irrelevant peptides as well as irrelevant ETCR2 were used as negative controls. Figure 17 shows the dose-dependent cell killing function of the exemplary bispecific ETCR at 18h and 24h. No non-specific killing was observed using any of the negative controls. Moreover, ETCR-E1.1 (2.3pM) was approximately 20-fold more potent than ETCR-E1.3 (39pM) (Table 12), indicating that the best killing function redirection was produced by anti-CD3scFv conjugated to the N-terminus of the Vβ domain instead of the Vα of the TCR (no killing effect was observed under the same conditions using bispecific ETCR with anti-CD3scFv conjugated to the C-terminus of any antibody constant domain, data not shown). Notably, using the same anti-CD3 scFv and conjugation site, ETCR-E1.1 (2.3 pM) exhibited significantly more potent killing than CTCR1-E1.1 (25 pM), 10-fold, suggesting that the altered antibody constant domains as well as the TCR variable domains provided better overall stability of the chimeric ETCR than CTCR.

[0123] [Table 12]

[0124] Example 10: In vitro tumor cell line killing assay of bispecific ETCR In vitro functional assays were also performed to confirm the activity of the designed bispecific ETCR in T cell-engaged killing of the tumor cell line A375. A375 tumor cells are HLA*A*02:01 and NY-ESO-1 double positive with an antigen density of 10-50 copies per cell, making them suitable for testing the killing of NY-ESO-1-specific bispecific TCRs.

[0125] method A method for isolation of CD8+ T cells is described in Example 9.

[0126] A375 cells were obtained from ATCC (ATCC CRL1619™) and maintained in DMEM supplemented with 10% FBS. For the killing assay, 50 μl / well of diluted bispecific ETCR was added to a black 96-well flat-bottom plate. 50 μl / well of isolated CD8+ T cells were added to 10 4 The indicated ratios were added to A375 cells at 100 / well and incubated for the designated time. For analysis, the plates were washed once with DPBS, and dissection variability was detected by CellTiter-Glo (CTG) assay kit (Promega, Cat. No. G755B).

[0127] result In vitro functional assays were performed to confirm the activity of the designed bispecific ETCRs in T cell-mediated killing of tumor cell line A375. Unrelated ETCR2 was used as a negative control. Figure 18 shows the dose-dependent cell killing function of the exemplary ETCRs. No non-specific killing was observed using any of the negative controls. In general, the sharp decrease in antigen density of A375 tumor cell line compared with T2 cell line reduced the killing ability of all bispecific ETCRs, and in particular, the killing function of ETCR-E1.3 was almost completely lost. Furthermore, ETCR-E1.1 (3.6nM) showed a remarkably 70-fold stronger killing than CTCR-E1.1 (264nM) (Table 13). These data again suggest that the overall stability of chimeric ETCR is better than CTCR due to the modified antibody constant domain and TCR variable domain, and such advantage is expanded under low antigen density conditions that frequently appear in real tumor microenvironment.

[0128] [Table 13]

Claims

1. 1. A polypeptide complex comprising: a first polypeptide comprising, from N-terminus to C-terminus, a first TCR alpha chain variable domain of a first TCR (TCR Vα) and a first antibody constant domain (C1) operably linked thereto, the first antibody constant domain comprising a CH1 domain of an IgG4 or IgG1 isotype; and a second polypeptide comprising, from N-terminus to C-terminus, a first TCR beta chain variable domain of a first TCR (TCR Vβ) and a second antibody constant domain (C2) operably linked thereto, the second antibody constant domain comprising a lambda light chain constant domain (Cλ domain), wherein the Cλ domain is selected from the group consisting of Cλ1, Cλ2, Cλ3, Cλ6 and Cλ7, and wherein the Cλ domain comprises one or more substitutions compared to any one of SEQ ID NOs: 1, 3, 5, 7 and 9, wherein the one or more substitutions are selected from a substitution of an amino acid D at position 30, a substitution of an amino acid H at position 31, and a substitution of an amino acid E at position 33; A polypeptide complex, in which a first polypeptide and a second polypeptide form a dimer.

2. The polypeptide complex of claim 1 , wherein the CH1 domain comprises an amino acid sequence of any one of SEQ ID NOs: 11 and 17.

3. 3. The polypeptide complex of claim 1 or 2, wherein a first TCR Vα operably binds to C1 via a first junction domain and a first TCR Vβ operably binds to C2 via a second junction domain.

4. The polypeptide complex of claim 3, wherein the first junction domain comprises any one of the amino acid sequences of SEQ ID NOs: 19, 21 and 23, and the second junction domain comprises any one of the amino acid sequences of SEQ ID NOs: 25, 27, 29, 31, 33 and 35.

5. The TCR Vβ has the following within the framework regions: According to IMGT numbering (a) R13K or R13T, (b) R90T or R90V, (c) L91I, (d) M48F, (e) H54A, H54Y or H54W, (f) M19Y, (g) A24K or A24R, (h) N77E 2. The polypeptide complex of claim 1, comprising the amino acid sequence of any one of SEQ ID NOs: 39, 43 and 47 together with one or more of the following substitutions:

6. 6. A multispecific antigen-binding complex comprising a first antigen-binding portion comprising the polypeptide complex of any one of claims 1 to 5, and a second antigen-binding portion, wherein the first antigen-binding portion binds to a first antigen and has a first antigen specificity, and the second antigen-binding portion binds to a second antigen and has a second antigen specificity different from the first antigen specificity, wherein the second antigen is a different epitope from the first antigen or a different antigen from the first antigen.

7. 7. The multispecific antigen-binding complex of claim 6, wherein the second antigen-binding moiety is conjugated at the N-terminus or C-terminus of the first polypeptide of the first antigen-binding moiety or the second polypeptide of the first antigen-binding moiety.

8. 8. The multispecific antigen-binding complex of claim 7, wherein the second antigen-binding moiety is conjugated at the N-terminus of the first polypeptide of the first antigen-binding moiety or the second polypeptide of the first antigen-binding moiety.

9. 7. The multispecific antigen-binding complex of claim 6, wherein the first antigen specificity is directed against a T-cell specific receptor molecule or a natural killer cell (NK cell) specific receptor molecule and the second antigen specificity is directed against a tumor-associated antigen, a tumor neoantigen or a T-cell specific receptor molecule.

10. 7. The multispecific antigen-binding complex of claim 6, wherein the first antigen-binding moiety binds to HLA*A*02:01-NY-ESO-1 peptide (SLLMWITQC) or HLA*A*02:01-GP100 peptide (YLEPGPVTV) and the second antigen-binding moiety binds to cluster of differentiation 3 (CD3).

11. 11. The multispecific antigen-binding complex of claim 10, wherein the first antigen-binding portion comprises a TCR Vα and a TCR Vβ, wherein Vα comprises an amino acid sequence selected from SEQ ID NOs: 37, 41 and 45, and Vβ comprises an amino acid sequence selected from SEQ ID NOs: 39, 43 and 47.

12. 7. The multispecific antigen-binding complex of claim 6, wherein the second antigen-binding moiety comprises a single-chain variable fragment (scFv) comprising both a heavy chain variable domain and a light chain variable domain covalently conjugated via a flexible linker.

13. 13. The multispecific antigen-binding complex of claim 12, wherein the second antigen-binding portion comprises an scFv as set forth in SEQ ID NO:

49.

14. An isolated polynucleotide encoding a polypeptide complex according to any one of claims 1 to 5, or a multispecific antigen-binding complex according to any one of claims 6 to 13.

15. 15. An isolated vector comprising the polynucleotide of claim 14.

16. 16. A host cell comprising the isolated polynucleotide of claim 14 or the isolated vector of claim 15.

17. A conjugate comprising a polypeptide complex according to any one of claims 1 to 5 or a multispecific antigen-binding complex according to any one of claims 6 to 13.

18. A method for expressing a polypeptide complex of any one of claims 1 to 5 or a multispecific antigen-binding complex of any one of claims 6 to 13, comprising culturing a host cell of claim 16 under conditions in which said polypeptide complex is expressed.

19. A pharmaceutical composition comprising a polypeptide complex according to any one of claims 1 to 5 or a multispecific antigen-binding complex according to any one of claims 6 to 13, and a pharma- ceutically acceptable carrier.

20. A polypeptide complex according to any one of claims 1 to 5 for the treatment of a condition in a subject in need thereof, said condition being able to be alleviated, eliminated, treated or prevented when the antigen targeted by the TCR is modulated.

21. 14. A multispecific antigen-binding complex according to any one of claims 6 to 13 for the treatment of a condition in a subject in need thereof, wherein when said first antigen and said second antigen are both modulated, said condition can be alleviated, eliminated, treated or prevented.

22. 22. The polypeptide complex or multispecific antigen-binding complex of claim 20 or 21, wherein the condition is cancer.

23. 14. A kit for the treatment of a condition in a subject in need thereof comprising a polypeptide complex according to any one of claims 1 to 5 or a multispecific antigen-binding complex according to any one of claims 6 to 13.

Images