Multispecific binding proteins based on pseudo-fab

Abstract

Provided are binding proteins comprising a pseudo-Fab domain comprising a stabilized knock-out domain and a second VH / VL forming a first functional antigen binding domain. Also provided are multispecific binding proteins comprising at least one pseudo-Fab. Also provided are multispecific binding proteins, nucleic acids encoding binding proteins and multispecific binding proteins, expression vectors, host cells, pharmaceutical compositions, and methods of treatment comprising administering the binding proteins or multispecific binding proteins described herein.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to European application No. 18306840.2, filed on 24 December 2018 and European application No. 19305813.8, filed on 21 June 2019, the contents of which are incorporated herein by reference for all purposes. Background Technology

[0003] The asymmetry inherent in the structure of natural antibodies is a prerequisite for generating multispecific binding proteins with two (e.g., bispecific antibodies) or more binding specificities. For example, by isolating different asymmetric binding arms or one or more Fvs on a Fab, bispecific antibodies can have the flexibility to bind two different antigens or epitopes simultaneously. However, despite these advantages, many multispecific antibody technologies suffer from process and manufacturing problems due to mispairing of the asymmetric heavy and light chains. For example, many of these technologies have the so-called “light chain problem,” in which random pairing of two different light chains with the heavy chain produces multiple chain pairing combinations other than the desired combination. In some cases, the light chain problem can be avoided by using a common light chain that enables binding to two antigens or epitopes. However, this may be infeasible for many antibodies, as this form requires antibody regeneration in transgenic mice. Furthermore, rare antibodies such as broadly neutralizing anti-HIV antibodies derived from human patients are not suited to such a form. Therefore, alternative and creative solutions to the mispairing problem remain needed. Summary of the Invention

[0004] This disclosure is based on the discovery of a novel heterodimerizing domain known as a "stable knockout domain" that can be used to form "pseudo-Fab". As disclosed herein, pseudo-Fab can be incorporated into a wide variety of binding proteins and binding forms to impart multispecific binding properties. In some aspects, the pseudo-Fab moiety can facilitate the preferential production, synthesis, or purification of desired multispecific binding proteins while minimizing or eliminating undesirable chain mispairing that typically results with conventional multispecific binding protein forms. In one aspect, this disclosure provides a binding protein comprising:

[0005] The first pseudo-Fab portion includes (1) a first VL domain (VLa), which pairs with a first VH domain (VHa) to form a first functional antigen binding site for binding target antigen A; and (2) a first stable knockout VH domain (VHX), which pairs with a first stable knockout VL domain (VLX) to form a first stable knockout domain.

[0006] The stable knockout domain comprises (3) one or more inactivating mutations that eliminate binding to the target antigen; and (4) one or more engineered interstrand disulfide bonds.

[0007] In one aspect, this disclosure provides a binding protein comprising:

[0008] The first pseudo-Fab portion includes (1) a first VL domain (VLa), which pairs with a first VH domain (VHa) to form a first functional antigen binding site for binding target antigen A; and (2) a first stable knockout VH domain (VHX), which pairs with a first stable knockout VL domain (VLX) to form a first stable knockout domain.

[0009] The stable knockout domain comprises (3) one or more inactivating mutations relative to the wild-type domain that eliminate its binding to the target antigen; and (4) one or more engineered interchain disulfide bonds relative to the reference Fab molecule that impart enhanced thermal stability (Tm) to the pseudo-Fab molecule, wherein the reference Fab molecule is identical to the pseudo-Fab molecule except that, in the pseudo-Fab molecule, the CH1 and CL domains of the reference Fab molecule are replaced by VHX and VLX domains.

[0010] In some embodiments, the binding protein is a multispecific binding protein, and the multispecific binding protein further comprises at least:

[0011] The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen-binding site that binds to target antigen B.

[0012] In one aspect, this disclosure provides a multispecific binding protein comprising:

[0013] a) A first pseudo-Fab portion comprising (1) a first VL domain (VLa) that pairs with a first VH domain (VHa) to form a first functional antigen binding site for binding target antigen A; and (2) a first stable knockout VH domain (VHX) that pairs with a first stable knockout VL domain (VLX) to form a first stable knockout domain.

[0014] b) A first Fab portion comprising (3) a second VL domain (VLb) paired with a second VH domain (VHb) to form a second functional antigen-binding site for binding target antigen B; (4) a first CH1 domain paired with a first CL domain; and

[0015] The stable knockout domain comprises (5) one or more inactivating mutations that eliminate its binding to the target antigen; and (6) one or more engineered interchain disulfide bonds; or alternatively, the stable knockout domain comprises (5) one or more inactivating mutations that eliminate its binding to the target antigen relative to the wild-type domain; and (6) one or more engineered interchain disulfide bonds that impart enhanced thermal stability (Tm) to the pseudo-Fab relative to the reference Fab molecule, wherein the reference Fab molecule is identical to the pseudo-Fab molecule except that, in the pseudo-Fab molecule, the CH1 and CL domains of the reference Fab molecule are replaced by VHX and VLX domains.

[0016] On the other hand, this disclosure provides a multispecific binding protein comprising:

[0017] a) A first pseudo-Fab portion comprising (1) a first VL domain (VLa) paired with a first VH domain (VHa) to form a first functional antigen binding site for binding target antigen A; and (2) a first stable knockout VH domain (VHX) paired with a first stable knockout VL domain (VLX) to form a stable knockout domain, provided that the first pseudo-Fab portion does not contain a CH1 domain paired with a CL domain.

[0018] The stable knockout domain comprises (3) one or more inactivating mutations that eliminate its binding to the target antigen; and (4) one or more engineered interchain disulfide bonds; or alternatively, the stable knockout domain comprises (3) one or more inactivating mutations that eliminate its binding to the target antigen relative to the wild-type domain; and (4) one or more engineered interchain disulfide bonds that impart enhanced thermal stability (Tm) to the pseudo-Fab relative to the reference Fab molecule, wherein the reference Fab molecule is identical to the pseudo-Fab molecule except that, in the pseudo-Fab molecule, the CH1 and CL domains of the reference Fab molecule are replaced by VHX and VLX domains;

[0019] b) A first Fab portion comprising (5) a second VL domain (VLb) paired with a second VH domain (VHb) to form a second functional antigen-binding site for binding target antigen B; (6) a first CH1 domain paired with a first CL domain; and

[0020] c) A connector portion that operatively connects the first Fab portion and the first dummy Fab portion.

[0021] On the other hand, this disclosure provides a multispecific binding protein comprising:

[0022] a) A first pseudo-Fab portion comprising (1) a first VL domain (VLa) that pairs with a first VH domain (VHa) to form a first functional antigen binding site for binding target antigen A; and (2) a first stable knockout VH domain (VHX) that pairs with a first stable knockout VL domain (VLX) to form a first stable knockout domain.

[0023] b) A first Fab portion comprising (3) a second VL domain (VLb) paired with a second VH domain (VHb) to form a second functional antigen-binding site for binding target antigen B; (4) a first CH1 domain paired with a first CL domain; and

[0024] The stable knockout domain comprises (5) one or more inactivating mutations that eliminate its binding to the target antigen; and (6) one or more engineered interchain disulfide bonds; or alternatively, the stable knockout domain comprises (5) one or more inactivating mutations that eliminate its binding to the target antigen relative to the wild-type domain; and (6) one or more engineered interchain disulfide bonds that impart enhanced thermal stability (Tm) to the pseudo-Fab relative to the reference Fab molecule, wherein the reference Fab molecule is identical to the pseudo-Fab molecule except that, in the pseudo-Fab molecule, the CH1 and CL domains of the reference Fab molecule are replaced by VHX and VLX domains.

[0025] c) A connector portion that operatively connects the first Fab portion and the first dummy Fab portion.

[0026] In some implementations, the joint portion is on one or more heavy chains.

[0027] In some embodiments, the multispecific binding protein further includes a third VL domain (VLc) that pairs with a third VH domain (VHc) to form a third functional antigen-binding site for binding target antigen C.

[0028] In some embodiments, the multispecific binding protein independently comprises one or two pseudo-Fab moieties and one or two Fab moieties.

[0029] In some embodiments, the linker portion is a peptide linker. In some embodiments, the peptide linker is of formula (Gly4Ser). n The Gly-Ser connector, where n is 1-10 (SEQ ID NO:101).

[0030] In some embodiments, the heterodimerizing domain comprises a full-length IgG antibody. In some embodiments, the heterodimerizing domain comprises the Fc domain of a full-length IgG antibody or a functional fragment thereof.

[0031] In some embodiments, the binding protein comprises separate protein chains selected from one of the following groups:

[0032] (a) VHa-CH1-L1-VHb-L2-VHX and VLa-CL and VLb-L3-VLX;

[0033] (b) VHa-L2-VHX-L1-VHb-CH1 and VLa-L3-VLX and VLb-CL;

[0034] (c) VHa-CH1-L1-VHa-CH1 and VHb-L2-VHX-L3-VHb-L4-VHX and two chains VLb-L5-VLX and two chains VLa-CL;

[0035] The chains in (a) and (b) may exist once or twice, and L1, L2, L3, L4 and L5 may be the same or different joints independently.

[0036] This disclosure provides a multispecific antibody comprising:

[0037] a) The first pseudo-Fab portion, which includes:

[0038] (1) A first VL domain (VLa), which pairs with a first VH domain (VHa) to form a first antigen-binding site for binding target antigen A;

[0039] (2) A first stable knockout VL domain (VLX) is paired with a first stable knockout VH domain (VHX) to form a first disulfide bond stable knockout (dsKO) domain.

[0040] (3) First heterodimerization domain (HD1);

[0041] The first dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interchain disulfide bonds; or alternatively, comprises (i) one or more inactivating mutations that eliminate the binding to the target antigen relative to the wild-type domain; and (ii) one or more engineered interchain disulfide bonds that impart enhanced thermal stability (Tm) to the pseudo-Fab relative to the reference Fab molecule, wherein the reference Fab molecule is identical to the pseudo-Fab molecule except that, in the pseudo-Fab molecule, the CH1 and CL domains of the reference Fab molecule are replaced by VHX and VLX domains.

[0042] b) The first Fab portion, comprising:

[0043] (1) A second VL domain (VLb) that pairs with a second VH domain (VHb) to form a second functional antigen-binding site that binds to target antigen B;

[0044] (2) The first CH1 domain, which is paired with the first CL domain; and

[0045] (3) Second heterodimerization domain (HD2).

[0046] In some implementations, the first heterodimerization domain (HD1) is operatively connected to the C-terminus of the VHX domain of the pseudo-Fab portion.

[0047] In some implementations, the second heterodimerization domain (HD2) is operatively connected to the C-terminus of the first CH1 domain of the first Fab portion.

[0048] In some implementations, the first and second heterodimerization domains comprise first and second Fc domains.

[0049] In some implementations, the Fc domain comprises a general structure: hinge-CH2 domain-CH3 domain.

[0050] In some implementations, the Fc domain contains one or more knock-in-hole (KIH) mutations.

[0051] In some implementations, one Fc domain includes a first CH3 domain containing one or both of the S354C and T366W mutations, and another Fc domain includes a second CH3 domain containing one or both of the Y349C, T366S, L368A, and Y407V mutations.

[0052] In some implementations, the Fc domain contains H435R and / or Y436F mutations.

[0053] In some embodiments, the first pseudo-Fab portion comprises a first polypeptide chain and a second polypeptide chain, the first polypeptide chain having a structure represented by the following formula:

[0054] (Ia)N-VHa-L1-VHX-C

[0055] Furthermore, the second polypeptide chain has a structure represented by the following formula:

[0056] (IIa)N-VLa-L2-VLX-C

[0057] L1 and L2 are connectors that may exist or not exist independently, and N and C represent N-terminals and C-terminals, respectively.

[0058] In some embodiments, the first pseudo-Fab portion comprises a first polypeptide chain and a second polypeptide chain, the first polypeptide chain having a structure represented by the following formula:

[0059] (Ib)N-VHX-L1-VHa-C

[0060] Furthermore, the second polypeptide chain has a structure represented by the following formula:

[0061] (IIb)N-VLX-L2-VLa-C

[0062] L1 and L2 are connectors that may exist or not exist independently, and N and C represent N-terminals and C-terminals, respectively.

[0063] In some embodiments, the first pseudo-Fab portion comprises a first polypeptide chain and a second polypeptide chain, the first polypeptide chain having a structure represented by the following formula:

[0064] (Ic)N-VLa-L1-VHX-C

[0065] Furthermore, the second polypeptide chain has a structure represented by the following formula:

[0066] (IIc)N-VHa-L2-VLX-C

[0067] L1 and L2 are connectors that may exist or not exist independently, and N and C represent N-terminals and C-terminals, respectively.

[0068] In some embodiments, the first pseudo-Fab portion comprises a first polypeptide chain and a second polypeptide chain, the first polypeptide chain having a structure represented by the following formula:

[0069] (Id)N-VHX-L1-VLa-C

[0070] Furthermore, the second polypeptide chain has a structure represented by the following formula:

[0071] (IId)N-VLX-L2-VHa-C

[0072] L1 and L2 are connectors that may exist or not exist independently, and N and C represent N-terminals and C-terminals, respectively.

[0073] In some implementations, at least two of target antigen A, target antigen B, and target antigen C are different target antigens.

[0074] In some embodiments, at least one of the target antigens is a ligand of a cell surface receptor, and at least one of the target antigens is a cell surface receptor.

[0075] In some implementations, the antigen-binding site is derived from different antibodies.

[0076] In some implementations, target antigen A, target antigen B, and target antigen C are the same target antigen.

[0077] In some implementations, the antigen-binding site binds to different epitopes on the same target antigen.

[0078] In some implementations, the antigen binding site binds to the same epitope on the same target antigen.

[0079] In some implementations, the antigen-binding site is derived from the same antibody.

[0080] In some implementations, the melting temperature (T) of the dummy Fab portion m It is at least 4 degrees Celsius higher than the reference Fab molecule.

[0081] In some implementations, the engineered interchain disulfide bond is VH44C-VL100C.

[0082] In some implementations, the engineered interchain disulfide bond is VH105C-VL43C.

[0083] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in the VHX domain of the pseudoFab portion.

[0084] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRH3 of the VHX domain.

[0085] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRH2 of the VHX domain.

[0086] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRH1 of the VHX domain.

[0087] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in the VLX domain of the pseudoFab portion.

[0088] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRL3 of the VLX domain.

[0089] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRL2 of the VLX domain.

[0090] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRL1 of the VLX domain.

[0091] In some embodiments, the VHX domain of the pseudo-Fab moiety comprises an amino acid sequence selected from SEQ ID NO:77, SEQ ID NO:78, and SEQ ID NO:79.

[0092] In some implementations, the VLX / VHX pair is selected from:

[0093] (i) VLX containing the amino acid sequence of SEQ ID NO:76 and VHX containing the amino acid sequence of SEQ ID NO:77;

[0094] ii) VLX containing the amino acid sequence of SEQ ID NO:76, and VHX containing the amino acid sequence of SEQ ID NO:78; and

[0095] iii) VLX containing the amino acid sequence of SEQ ID NO:76 and VHX containing the amino acid sequence of SEQ ID NO:79.

[0096] In some embodiments, the binding protein further includes one or more additional binding domains operatively linked to the N-terminus or C-terminus of the binding protein.

[0097] In some implementations, the one or more additional binding structural domains are operatively connected to the N-terminus of the first or second dummy Fab portion.

[0098] In some implementations, the one or more additional bonding domains are operatively connected to the N-terminus of the first or second Fab portion.

[0099] On the other hand, this disclosure provides a multispecific binding protein comprising four polypeptide chains forming at least two antigen-binding sites, wherein

[0100] (a) The first polypeptide comprises a structure represented by the following formula:

[0101] VLa-L1-VLX[I]

[0102] (b) The second polypeptide comprises a structure represented by the following formula:

[0103] VHa-L2-VHX-FC1[II]

[0104] (c) The third polypeptide contains a structure represented by the following formula:

[0105] VLb-CL[III]

[0106] (d) The fourth polypeptide contains a structure represented by the following formula:

[0107] VHb-CH1-FC2[IV]

[0108] in:

[0109] VLa is the variable domain of the first immunoglobulin light chain;

[0110] VLb is the variable domain of the second immunoglobulin light chain;

[0111] VHa is the variable domain of the first immunoglobulin heavy chain;

[0112] VHb is the variable domain of the second immunoglobulin heavy chain;

[0113] VLX is a stable knockout of the variable structure domain of the light chain;

[0114] VHX is a stable knockout heavy chain variable structure domain;

[0115] CL is the constant domain of the immunoglobulin light chain;

[0116] CH1 is the constant domain of the heavy chain of immunoglobulin CH1;

[0117] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0118] L1 and L2 are amino acid linkers that can be the same or different independently;

[0119] in

[0120] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0121] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0122] (3) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0123] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0124] On the other hand, this disclosure provides an antigen-binding protein comprising six polypeptide chains forming four antigen-binding sites, wherein

[0125] (a) The first and second polypeptides comprise a structure represented by the following formula:

[0126] VLa-L1-VLX[I] and [II]

[0127] (b) The third and fourth polypeptides comprise structures represented by the following formula:

[0128] VLb-CL[III] and [IV]

[0129] (c) The fifth polypeptide comprises a structure represented by the following formula:

[0130] VHa-L2-VHX-L3-VHb-CH1-FC1[V]

[0131] (d) The sixth polypeptide comprises a structure represented by the following formula:

[0132] VHa-L2-VHX-L3-VHb-CH1-FC2[VI]

[0133] in:

[0134] VLa is the variable domain of the first immunoglobulin light chain;

[0135] VLb is the variable domain of the second immunoglobulin light chain;

[0136] VHa is the variable domain of the first immunoglobulin heavy chain;

[0137] VHb is the variable domain of the second immunoglobulin heavy chain;

[0138] VLX is a stable knockout of the variable structure domain of the light chain;

[0139] VHX is a stable knockout heavy chain variable structure domain;

[0140] CL is the constant domain of the immunoglobulin light chain;

[0141] CH1 is the constant structural domain of the immunoglobulin heavy chain;

[0142] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0143] L1, L2, and L3 are amino acid linkers.

[0144] in

[0145] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0146] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0147] (3) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0148] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0149] On the other hand, this disclosure provides an antigen-binding protein comprising six polypeptide chains forming four antigen-binding sites, wherein

[0150] (a) The first and second polypeptides comprise a structure represented by the following formula:

[0151] VLa-L1-VLX[I] and [II]

[0152] (b) The third and fourth polypeptides comprise structures represented by the following formula:

[0153] VLb-CL[III] and [IV]

[0154] (c) The fifth polypeptide comprises a structure represented by the following formula:

[0155] VHb-CH1-L3-VHa-L2-VHX-FC1[V]

[0156] (d) The sixth polypeptide comprises a structure represented by the following formula:

[0157] VHb-CH1-L3-VHa-L2-VHX-FC2[VI]

[0158] in:

[0159] VLa is the variable domain of the first immunoglobulin light chain;

[0160] VLb is the variable domain of the second immunoglobulin light chain;

[0161] VHa is the variable domain of the first immunoglobulin heavy chain;

[0162] VHb is the variable domain of the second immunoglobulin heavy chain;

[0163] VLX is a stable knockout of the variable structure domain of the light chain;

[0164] VHX is a stable knockout heavy chain variable structure domain;

[0165] CL is the constant domain of the immunoglobulin light chain;

[0166] CH1 is the constant structural domain of the immunoglobulin heavy chain;

[0167] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0168] L1, L2, and L3 are amino acid linkers.

[0169] in

[0170] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0171] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0172] (3) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0173] The dsKO domain contains (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0174] On the other hand, this disclosure provides an antigen-binding protein comprising six polypeptide chains forming four antigen-binding sites, wherein

[0175] (a) The first and second polypeptides comprise a structure represented by the following formula:

[0176] VLa-L1-VLX[I] and [II]

[0177] (b) The third and fourth polypeptides comprise structures represented by the following formula:

[0178] VLb-CL[III] and [IV]

[0179] (c) The fifth polypeptide comprises a structure represented by the following formula:

[0180] VHa-L2-VHX-L3-VHa-L4-VHX-FC1[V]

[0181] (d) The sixth polypeptide comprises a structure represented by the following formula:

[0182] VHb-CH1-L5-VHb-CH1-FC2[VI]

[0183] in:

[0184] VLa is the variable domain of the first immunoglobulin light chain;

[0185] VLb is the variable domain of the second immunoglobulin light chain;

[0186] VHa is the variable domain of the first immunoglobulin heavy chain;

[0187] VHb is the variable domain of the second immunoglobulin heavy chain;

[0188] VLX is a stable knockout of the variable structure domain of the light chain;

[0189] VHX is a stable knockout heavy chain variable structure domain;

[0190] CL is the constant domain of the immunoglobulin light chain;

[0191] CH1 is the constant structural domain of the immunoglobulin heavy chain;

[0192] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0193] L1, L2, L3, L4, and L5 are amino acid linkers.

[0194] in

[0195] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0196] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0197] (3) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0198] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0199] On the other hand, this disclosure provides an antigen-binding protein comprising four polypeptide chains forming three antigen-binding sites, wherein:

[0200] (a) The polypeptide comprises a structure represented by the following formula:

[0201] VLa-L1-VLX[I]

[0202] (b) The second polypeptide comprises a structure represented by the following formula:

[0203] VHa-L2-VHX-FC1[II]

[0204] (c) The third polypeptide comprises a structure represented by the following formula:

[0205] VLb-L3-VLc-L4-CL[III]

[0206] (d) The fourth polypeptide comprises a structure represented by the following formula:

[0207] VHc-L5-VHb-L6-CH1-FC2[IV]

[0208] in:

[0209] VLa is the variable domain of the first immunoglobulin light chain;

[0210] VLb is the variable domain of the second immunoglobulin light chain;

[0211] VLc is the variable domain of the third immunoglobulin light chain;

[0212] VHa is the variable domain of the first immunoglobulin heavy chain;

[0213] VHb is the variable domain of the second immunoglobulin heavy chain;

[0214] VHc is the variable domain of the third immunoglobulin heavy chain;

[0215] CL is the constant domain of the immunoglobulin light chain;

[0216] CH1 is the constant domain of the heavy chain of immunoglobulin CH1;

[0217] VLX is the first stable knockout variable structural domain of the light chain;

[0218] VHX is the first stable knockout heavy chain variable structure domain;

[0219] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0220] L1, L2, L3, L4, L5, and L6 are amino acid linkers.

[0221] in

[0222] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0223] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0224] (3) The third VL domain (VLc) pairs with the third VH domain (VHc) to form a first functional antigen-binding site for binding target antigen C;

[0225] (4) The polypeptide of Formula III and the polypeptide of Formula IV form a cross-linked light chain-heavy chain pair (CODV).

[0226] (5) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0227] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0228] On the other hand, this disclosure provides an antigen-binding protein comprising four polypeptide chains forming three antigen-binding sites, wherein:

[0229] (a) The polypeptide comprises a structure represented by the following formula:

[0230] VLa-L1-VLX [I]

[0231] (b) The second polypeptide comprises a structure represented by the following formula:

[0232] VHa-L2-VHX-FC1 [II]

[0233] (c) The third polypeptide comprises a structure represented by the following formula:

[0234] VLb-L3-VLc-L4-CL [III]

[0235] (d) The fourth polypeptide comprises a structure represented by the following formula:

[0236] VHc-L5-VHb-L6-CH1-FC2 [IV]

[0237] in:

[0238] VLa is the variable domain of the first immunoglobulin light chain;

[0239] VLb is the variable domain of the second immunoglobulin light chain;

[0240] VLc is the variable domain of the third immunoglobulin light chain;

[0241] VHa is the variable domain of the first immunoglobulin heavy chain;

[0242] VHb is the variable domain of the second immunoglobulin heavy chain;

[0243] VHc is the variable domain of the third immunoglobulin heavy chain;

[0244] CL is the constant domain of the immunoglobulin light chain;

[0245] CH1 is the constant domain of the heavy chain of immunoglobulin CH1;

[0246] VLX is the first stable knockout variable structural domain of the light chain;

[0247] VHX is the first stable knockout heavy chain variable structure domain;

[0248] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0249] L1, L2, L3, L4, L5, and L6 are amino acid linkers.

[0250] in

[0251] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0252] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0253] (3) The third VL domain (VLc) pairs with the third VH domain (VHc) to form a first functional antigen-binding site for binding target antigen C;

[0254] (4) The polypeptide of Formula III and the polypeptide of Formula IV form a cross-linked light chain-heavy chain pair (CODV).

[0255] (5) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0256] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen of the reference Fab molecule; and (ii) one or more engineered interchain disulfide bonds.

[0257] In other aspects, some embodiments relate to all binding proteins described herein, wherein the dsKO domain comprises (i) one or more inactivating mutations relative to the wild-type domain, eliminating its binding to the target antigen; and (ii) one or more engineered interchain disulfide bonds relative to the reference Fab molecule, conferring enhanced thermal stability (Tm) to the pseudo-Fab molecule, wherein the reference Fab molecule is identical to the pseudo-Fab molecule except that, in the pseudo-Fab molecule, the CH1 and CL domains of the reference Fab molecule are replaced by VHX and VLX domains. In those embodiments, binding to the target antigen is measured by methods known in the art (e.g., but not limited to surface plasmon resonance), and thermal stability is measured by methods known in the art (e.g., but not limited to differential scanning calorimetry).

[0258] In some embodiments, the FC1 and FC2 domains contain one or more kilometre structures (KIH) mutations, wherein the mutations promote heterodimerization of the Fc domain of the polypeptide.

[0259] In some implementations, one of FC1 or FC2 includes a first CH3 domain containing one or both of the S354C and T366W mutations, and the other of FC1 or FC2 includes a second CH3 domain containing one or both of the Y349C, T366S, L368A, and Y407V mutations, wherein the mutations promote heterodimerization of the Fc domain.

[0260] In some implementations, the FC1 or FC2 domain contains H435R and / or Y436F mutations.

[0261] In some implementations, at least two of target antigen A, target antigen B, and target antigen C are different target antigens.

[0262] In some embodiments, at least one of the target antigens is a ligand of a cell surface receptor, and at least one of the target antigens is a cell surface receptor.

[0263] In some implementations, the antigen-binding site is derived from different antibodies.

[0264] In some implementations, target antigen A, target antigen B, and target antigen C are the same target antigen.

[0265] In some implementations, the antigen-binding site binds to different epitopes on the same target antigen.

[0266] In some implementations, the antigen binding site binds to the same epitope on the same target antigen.

[0267] In some implementations, the antigen-binding site is derived from the same antibody.

[0268] In some implementations, the melting temperature (T) of the dummy Fab portion m It is at least 4 degrees Celsius higher than the reference Fab molecule.

[0269] In some implementations, the engineered interchain disulfide bond is VH44C-VL100C.

[0270] In some implementations, the engineered interchain disulfide bond is VH105C-VL43C.

[0271] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in the VHX domain of the pseudoFab portion.

[0272] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRH3 of the VHX domain.

[0273] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRH2 of the VHX domain.

[0274] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRH1 of the VHX domain.

[0275] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in the VLX domain of the pseudoFab portion.

[0276] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRL3 of the VLX domain.

[0277] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRL2 of the VLX domain.

[0278] In some embodiments, at least one of the one or more inactivating mutations that eliminate binding to the target antigen is present in CDRL1 of the VLX domain.

[0279] In some embodiments, the VHX domain of the pseudo-Fab moiety comprises an amino acid sequence selected from SEQ ID NO:77, SEQ ID NO:78, and SEQ ID NO:79.

[0280] In some implementations, the VLX / VHX pair is selected from:

[0281] (i) VLX containing the amino acid sequence of SEQ ID NO:76 and VHX containing the amino acid sequence of SEQ ID NO:77;

[0282] ii) VLX containing the amino acid sequence of SEQ ID NO:76, and VHX containing the amino acid sequence of SEQ ID NO:78; and

[0283] iii) VLX containing the amino acid sequence of SEQ ID NO:76 and VHX containing the amino acid sequence of SEQ ID NO:79.

[0284] On the other hand, this disclosure provides the use of a stable knockout domain for reducing heavy-light chain mispairing in multispecific binding proteins, wherein the stable knockout domain comprises VHX and VLX domains, the VHX and VLX domains comprising (3) one or more inactivating mutations that eliminate their binding to the target antigen; and (4) one or more engineered interchain disulfide bonds that confer enhanced thermal stability (Tm) to the pseudo-Fab relative to the reference Fab molecule, wherein the reference Fab molecule is identical to the pseudo-Fab molecule except that, in the pseudo-Fab molecule, the CH1 and CL domains of the reference Fab molecule are replaced by the VHX and VLX domains.

[0285] In some implementations, the engineered interchain disulfide bond is VH44C-VL100C.

[0286] In some implementations, the engineered interchain disulfide bond is VH105C-VL43C.

[0287] In some embodiments, the VHX domain of the pseudo-Fab moiety comprises an amino acid sequence selected from SEQ ID NO:77, SEQ ID NO:78, and SEQ ID NO:79.

[0288] In some implementations, the VLX / VHX pair is selected from:

[0289] (i) VLX containing the amino acid sequence of SEQ ID NO:76 and VHX containing the amino acid sequence of SEQ ID NO:77;

[0290] ii) VLX containing the amino acid sequence of SEQ ID NO:76, and VHX containing the amino acid sequence of SEQ ID NO:78; and

[0291] iii) VLX containing the amino acid sequence of SEQ ID NO:76 and VHX containing the amino acid sequence of SEQ ID NO:79.

[0292] In some implementations, the dummy Fab lacks the CH1 and CL structural domains.

[0293] In some embodiments, an isolated nucleic acid molecule is provided, the isolated nucleic acid molecule comprising a nucleotide sequence encoding one or more binding proteins. In some embodiments, a set of isolated nucleic acid molecules is provided, the set of isolated nucleic acid molecules comprising one or more nucleotide sequences encoding one or more binding proteins.

[0294] In some embodiments, an expression vector comprising the nucleic acid molecules is provided. In some embodiments, a set of expression vectors comprising the set of nucleic acid molecules is provided.

[0295] In some embodiments, an isolated host cell is provided, the isolated host cell comprising the nucleic acid molecule or the expression vector. In some embodiments, an isolated host cell is provided, the isolated host cell comprising the complete set of nucleic acid molecules or the complete set of expression vectors.

[0296] In some embodiments, a method for generating the binding protein is provided, the method comprising culturing the host cell under conditions that cause the binding protein to be expressed; and purifying the binding protein from the host cell.

[0297] In some embodiments, a pharmaceutical composition is provided comprising a pharmaceutically acceptable carrier and a therapeutically effective amount of the multispecific binding protein. In some embodiments, a multispecific binding protein is provided for use as a drug.

[0298] In some implementations, a method is provided for treating a disorder in which antigen activity is harmful, the method comprising administering an effective amount of a multispecific binding protein to a subject in need.

[0299] The above-described invention is non-limiting, and other features and advantages of the disclosed compositions and methods will become apparent from the following detailed description and claims of the invention.

[0300] Sequence Description

[0301] SEQ ID NO:1: Variable light chain sequence of trastuzumab.

[0302] SEQ ID NO:2: Variable heavy chain sequence of trastuzumab.

[0303] SEQ ID NO:3: Variable heavy chain sequence of trastuzumab KO variant 1.

[0304] SEQ ID NO:4: Variable heavy chain sequence of trastuzumab ko variant 2.

[0305] SEQ ID NO:5: Variable heavy chain sequence of trastuzumab ko variant 3.

[0306] SEQ ID NO:6: Anti-IL13--VL-G4S-anti-Her2-(trastuzumab-Q100C)-VL.

[0307] SEQ ID NO:7: Anti-IL13--VH-G4S-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0308] SEQ ID NO:8: Anti-IL13-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0309] SEQ ID NO:9: Anti-IL13-VH-(G4S)2-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0310] SEQ ID NO:10: Anti-IL13-VL3-IGK.

[0311] SEQ ID NO:11: Anti-IL13-VH2-IGHG1.

[0312] SEQ ID NO:12: Anti-TNFα-VL-G4S-anti-Her2-(trastuzumab-Q100C)-VL.

[0313] SEQ ID NO:13: Anti-TNFα-VH-G4S-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0314] SEQ ID NO:14: Anti-TNFα-VL-(G4S)2-anti-Her2-(trastuzumab-Q100C)-VL.

[0315] SEQ ID NO:15: Anti-TNFα-VH-(G4S)2-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0316] SEQ ID NO: 16: anti-TNFa-VL-huIGKC.

[0317] SEQ ID NO: 17: anti-TNFa-VH-huIgG1.

[0318] SEQ ID NO:18: Anti-IL6R-VL-G4S-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0319] SEQ ID NO:19: Anti-IL6R-VH-G4S-Anti-Her2-(Trastuzumab-G44C)-VH-Fc-huIgG1.

[0320] SEQ ID NO:20: Anti-IL6R-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL

[0321] SEQ ID NO:21: Anti-IL6R-VH-(G4S)2-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0322] SEQ ID NO:22: Anti-IL6R-VL-huIGKC.

[0323] SEQ ID NO:23: Anti-IL6R-VH-huIgG1.

[0324] SEQ ID NO:24: Anti-CTLA4-VL-G4S-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0325] SEQ ID NO:25: Anti-CTLA4-VH-G4S-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0326] SEQ ID NO:26: Anti-CTLA4-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0327] SEQ ID NO:27: Anti-CTLA4-VH-(G4S)2-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0328] SEQ ID NO:28: anti-CTLA4-VL-huIGKC.

[0329] SEQ ID NO:29: Anti-CTLA4-VH-huIgG1.

[0330] SEQ ID NO:30: Anti-PD1-VL-G4S-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0331] SEQ ID NO:31: Anti-PD1-VH-G4S-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0332] SEQ ID NO:32: Anti-PD1-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL

[0333] SEQ ID NO:33: Anti-PD1-VH-(G4S)2-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0334] SEQ ID NO:34: Anti-huPD-1-VL-huIGKC.

[0335] SEQ ID NO:35: Anti-huPD-1-VH-huIgG1.

[0336] SEQ ID NO:36: Anti-IL4-VL-G4S-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0337] SEQ ID NO:37: Anti-IL4-VH-G4S-Anti-Her2-(Trastuzumab-G44C)-VH-Fc-huIgG1.

[0338] SEQ ID NO:38: Anti-IL4-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0339] SEQ ID NO:39: Anti-IL4-VH-(G4S)2-anti-Her2-(trastuzumab-G44C)-VH-Fc-huIgG1.

[0340] SEQ ID NO:40: Anti-IL4-VL1-IGKC.

[0341] SEQ ID NO:41: Anti-IL4-VH1-IgG1.

[0342] SEQ ID NO:42: Anti-PD1-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0343] SEQ ID NO:43: Anti-PD1-VH-(G4S)2-anti-Her2-(trastuzumab-G44C-Var2)-VH-Fc-huIgG1.

[0344] SEQ ID NO:44: Anti-CTLA4-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0345] SEQ ID NO:45: Anti-CTLA4-VH-(G4S)2-anti-Her2-(trastuzumab-G44C-Var2)-VH-Fc-huIgG1.

[0346] SEQ ID NO:46: Anti-IL4-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0347] SEQ ID NO:47: Anti-IL4-VH-(G4S)2-anti-Her2-(trastuzumab-G44C-Var2)-VH-Fc-huIgG1.

[0348] SEQ ID NO:48: Anti-IL13-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0349] SEQ ID NO:49: Anti-IL13-VH-(G4S)2-Anti-Her2-(Trastuzumab-G44C-Var2)-VH-DKTHT-His6.

[0350] SEQ ID NO:50: Anti-IL13-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0351] SEQ ID NO:51: Anti-IL13-VH-(G4S)2-anti-Her2-(trastuzumab-VH_Var1-G44C)-VH-Fc-huIgG1.

[0352] SEQ ID NO:52: Anti-IL4-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0353] SEQ ID NO:53: Anti-IL4-VH-(G4S)2-anti-Her2-(trastuzumab-G44C-Var2)-VH-Fc-huIgG1 (Chu).

[0354] SEQ ID NO:54: Anti-PD1-huIGKC.

[0355] SEQ ID NO: 55: anti-PD1-VH-huIgG1 (ethyl-RF).

[0356] SEQ ID NO:56: Anti-IL13-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0357] SEQ ID NO:57: Anti-IL13-VH-(G4S)2-anti-Her2-(trastuzumab-G44C-Var2)-VH-Fc-huIgG1 (Chu).

[0358] SEQ ID NO:58: Anti-PD1-VL-huIGKC.

[0359] SEQ ID NO: 59: anti-PD1-VH-huIgG1 (ethyl-RF).

[0360] SEQ ID NO:60: Anti-PD1-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0361] SEQ ID NO:61: Anti-PD1-VH-(G4S)2-anti-Her2-(trastuzumab-G44C-Var2)-VH-Fc-huIgG1 (Puer).

[0362] SEQ ID NO:62: Anti-IL13-VL huIGKC.

[0363] SEQ ID NO: 63: anti-IL13-VH-huIgG1 (ethyl-RF).

[0364] SEQ ID NO:64: Anti-CTLA4-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0365] SEQ ID NO:65: Anti-CTLA4-VH-(G4S)2-anti-Her2-(trastuzumab-G44C-Var2)-VH-Fc-huIgG1 (Chu).

[0366] SEQ ID NO:66: Anti-PD1-VL-huIGKC.

[0367] SEQ ID NO: 67: anti-PD1-VH-huIgG1 (ethyl-RF).

[0368] SEQ ID NO:68: Anti-IL4-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0369] SEQ ID NO:69: Anti-IL4-VH-(G4S)2-anti-Her2-(trastuzumab-G44C-Var2)-VH-Fc-huIgG1 (Chu).

[0370] SEQ ID NO:70: Anti-IL13-VL huIGKC.

[0371] SEQ ID NO: 71: anti-IL13-VH-huIgG1 (ethyl-RF).

[0372] SEQ ID NO:72: Anti-PD1-VL-(G4S)2-Anti-Her2-(Trastuzumab-Q100C)-VL.

[0373] SEQ ID NO:73: Anti-PD1-VH-(G4S)2-anti-Her2-(trastuzumab-G44C-Var2)-VH-Fc-huIgG1 (Puer).

[0374] SEQ ID NO:74: Anti-PD1-VL-huIGKC.

[0375] SEQ ID NO: 75: anti-PD1-VH-huIgG1 (ethyl-RF).

[0376] SEQ ID NO:76: ds ko trastuzumab variable light chain sequence.

[0377] SEQ ID NO:77: ds ko trastuzumab variant 1 variable heavy chain sequence.

[0378] SEQ ID NO:78: ds ko trastuzumab variant 2 variable heavy chain sequence.

[0379] SEQ ID NO:79: ds ko trastuzumab variant 3 variable heavy chain sequence.

[0380] SEQ ID NO:80: Anti-TCRα / βx anti-CD123 wild type

[0381] SEQ ID NO:81: Anti-TCRα / βx anti-CD123-dsTrasKO2

[0382] SEQ ID NO:82: anti-TCRa / β-dsTrasKO2 x anti-CD123

[0383] SEQ ID NO:83: Anti-CD3εx and anti-CD123 wild type

[0384] SEQ ID NO:84: Anti-CD3εx anti-CD123-dsTrasKO2

[0385] SEQ ID NO:85: Anti-CD3ε-dsTrasKO2 x Anti-CD123

[0386] SEQ ID NO:86: Anti-CD3εx and anti-CD123 wild type

[0387] SEQ ID NO:87: Anti-CD3εx anti-CD123-dsTrasKO2

[0388] SEQ ID NO:88: Anti-CD3ε-dsTrasKO2 x Anti-CD123

[0389] SEQ ID NO:89: Anti-TCRα / βx anti-TNP negative control - wild type

[0390] SEQ ID NO:90: ​​Anti-TCRα / βx anti-TNP-dsTrasKO2 negative control

[0391] SEQ ID NO:91: Anti-TCRα / β-dsTrasKO2 x Anti-TNP negative control

[0392] SEQ ID NO:92: Anti-TNP x Anti-CD123 negative control - wild type

[0393] SEQ ID NO:93: Anti-TNP x Anti-CD123-dsTrasKO2 negative control

[0394] SEQ ID NO:94: Anti-TNP-dsTrasKO2 x Anti-CD123 negative control

[0395] SEQ ID NO:95: Anti-CD3εx anti-TNP negative control - wild type

[0396] SEQ ID NO:96: Anti-CD3εx anti-TNP-dsTrasKO2 negative control

[0397] SEQ ID NO:97: Anti-CD3ε-dsTrasKO2 x Anti-TNP negative control

[0398] SEQ ID NO:98: Anti-CD3εx anti-TNP negative control - wild type

[0399] SEQ ID NO:99: Anti-CD3εx anti-TNP-dsTrasKO2 negative control

[0400] SEQ ID NO:100: Anti-CD3ε-dsTrasKO2 x Anti-TNP negative control Attached Figure Description

[0401] The foregoing and other features and advantages of the invention will be more fully understood from the following detailed description of illustrative embodiments, taken in conjunction with the accompanying drawings. The patent or application document contains at least one color drawing. Upon request and payment of the necessary fees, the official authority will provide a copy of this patent or patent application publication with color drawings.

[0402] Figures 1A and 1B schematically depict dimer dual-specificity tandem molecules containing pseudoFab fragments in which the CH1 / CL pairs are replaced by disulfide-stabilized knockout domains (dsKO). Figure 1A depicts the tandem-(Fv-Fab x Fv-pseudoFab) molecule, and Figure 1B depicts the tandem-(Fv-pseudoFab x Fv-Fab) molecule.

[0403] Figure 2 schematically depicts an exemplary dimeric bispecific IgG molecule ((Fv-pseudoFab)x(Fv-Fab)-Fc)). A disulfide-stabilized knockout domain (dsKO) replaces the CH / CL domain of one Fab arm of the IgG molecule. A peptide linker (e.g., G4S (SEQ ID NO: 102) or (G4S)2 (SEQ ID NO: 103)) connects the first antigen-binding site (Fv) to the dsKO domain to form a pseudoFab moiety that binds antigen target A. An Fc heterodimerization domain with a club-and-mortar (KIH) mutation or an RF mutation connects the pseudoFab moiety to a second Fab-binding arm that binds antigen target B.

[0404] Figures 3A-3C schematically depict a comparison of the monomer fractions of the trastuzumab WT VH / VL-substituted antibody construct containing CH1 / CL (Figure 3A) and the monomer fractions of the disulfide-stable trastuzumab knockout (“dsTrastKO”) VH / VL-substituted antibody construct containing CH1 / CL (Figure 3B), as well as the thermal stability of the two constructs (Figure 3C).

[0405] Figure 4 schematically depicts the PDB structure 1N8Z of trastuzumab (green) binding to HER2 (brown). The four inactivating mutation sites (R50, R59, Y33, and Y103) that eliminate the binding to HER2 are annotated in yellow.

[0406] Figures 5A and 5B graphically depict the results of binding experiments demonstrating that dsTrastuKO variants 1-3 no longer bind to HER2.

[0407] Figures 6A-6B depict the pseudo-IgG and pseudo-Fab constructs used as some controls in the experiments.

[0408] Figures 7A-7D depict representative bispecific forms and purification results according to certain exemplary embodiments. Figure 7A shows the arrangement of individual domains within the IgG scaffold. Fv1 (anti-IL4) is fused to the VL / VH of dsTrasKO2 via a (G4S)2 linker (SEQ ID NO: 103). Fv2 (anti-IL13) retains the wild-type conformation. Figure 7B shows the reduced (one light chain and one heavy chain) and oxidized forms of the antibody using 4%-12% Bis / Tris MOPS sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The purity of the products is shown in Figure 7C using analytical size exclusion chromatography. Molecular integrity was validated by complete quality analysis using an Agilent 6540 Ultra HD (UHD) Q-TOF system equipped with a Jet Stream dual ESI interface and an Agilent 1290 / 1260 Infinity LC system (Figure 7D).

[0409] Figure 8 shows the analytical hydrophobic interaction chromatography (HIC) results, indicating that the dsTrasKO2 bispecific constructs are correctly paired and do not contain any unexpected species.

[0410] Figure 9 schematically depicts the crystal structure of the pseudo-Fab IL13-dsTrasKO2 construct with the ultrastructure shown in Figure 6A. The structure is... The solution is as follows. The structure shows the overlap of the IL13-VH / VL domains of the TrasKO2-IL13 pseudo-Fab (light gray) and Fab anti-IL13 (dark gray).

[0411] Figures 10A-10B schematically depict dimer bispecific tandem molecules containing pseudoFab fragments in which the CH1 / CL pairs are replaced by disulfide-stabilized knockout domains (dsKO). Figure 10A depicts the tandem-(Fv-Fab x Fv-pseudoFab)-IgG molecule, and Figure 10B depicts the tandem-(Fv-pseudoFab x Fv-Fab)-IgG molecule.

[0412] Figures 11A-11B schematically depict dimer dual-specificity tandem molecules containing pseudoFab fragments in which the CH1 / CL pairs are replaced by disulfide-stabilized knockout domains (dsKO). Figure 11A (Fc heterodimerization domains with both KIH and RF mutations) and Figure 11B (Fc heterodimerization domains with only RF mutations) depict the (((Fv-pseudoFab)[HC]-(Fv-pseudoFab))x((Fv-Fab)[HC]-(Fv-Fab)))-Fc molecule.

[0413] Figures 12A-12D depict a representative bispecific tandem IgG design in which a first pseudofabric of the Fv1 domain (with binding specificity against the first target antigen A (i.e., GITR)) and the dsTrasKO domain is attached to each Fv2 domain of a conventional IgG antibody (Pogalizumab) (with binding specificity against the second target antigen B (i.e., Ox40)). Figure 12A shows the arrangement of individual domains within the IgG scaffold. Fv1 (anti-GITR) is fused to the VL / VH of dsTrasKO2 via the (G4S)2 linker (SEQ ID NO:103). Fv2 (anti-Ox40) retains the wild-type conformation. Fv1-dsTrasKO2 is fused to Fv2-Ck / CH1 via the (G4S)2 linker (SEQ ID NO:103). Figure 12B shows the reduced (two light chains and one heavy chain) and oxidized forms of the antibody using 4%–12% Bis / TrisMOPS sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Product purity is shown in Figure 12C using analytical size exclusion chromatography. Molecular integrity was validated by complete quality analysis using an Agilent 6540 Ultra High Definition (UHD) Q-TOF system equipped with a JetStream dual ESI interface and an Agilent 1290 / 1260 Infinity LC system (Figure 12D).

[0414] Figures 13A-13D depict a representative trispecific CODV IgG design, in which a first pseudoFab with an Fv3 domain (having binding specificity against the first target antigen A (i.e., CD137)) and a dsTrasKO domain is paired with a CODV arm (having binding specificity against the second target antigen B (i.e., Ox40) (Fv1) and binding specificity against the third target antigen C (i.e., PD1) (Fv2)). Figure 13A shows the arrangement of individual domains within the CODV IgG scaffold. Fv1 (anti-Ox40) and Fv2 (anti-PD1) on the CODV arm are fused with the wild-type λ domain and CH1 domain. Fv3 (anti-CD137) on the Fab arm is fused with the VL / VH of dsTrasKO2 via a (G4S)2 linker (SEQ ID NO: 103). Figure 13B shows the reduced (two light chains and two heavy chains) and oxidized forms of the CODV antibody using 4%–12% Bis / Tris MOPS sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Product purity is shown in Figure 13C using analytical size exclusion chromatography. Molecular integrity was validated by complete quality analysis using an Agilent 6540 Ultra High Definition (UHD) Q-TOF system equipped with a JetStream dual ESI interface and an Agilent 1290 / 1260 Infinity LC system (Figure 13D).

[0415] Figure 14 illustrates a two-step purification process to ensure proper light chain pairing with the dsTrasKO2 knockout bispecific molecule.

[0416] Figures 15A-15B depict the cytotoxicity assay of human panT cells against THP-1 target cells co-incubated with bispecific antibodies anti-CD3εx and anti-CD123. Figure 15A corresponds to bispecific antibodies IDs 33, 34, and 35, and negative controls 45 and 46. Figure 15B corresponds to bispecific antibodies IDs 36, 37, and 38, and negative controls 47 and 48. T effector cells and CFSE-labeled THP-1 target cells were seeded at an effector cell to target cell ratio of 10:1 and co-incubated at 37°C for 20 h with serial dilutions (10 nM-0 nM) of the corresponding bispecific molecules. Dead cells were stained with 7-AAD and measured by flow cytometry. Cytotoxic activity was calculated based on the percentage of dead THP-1 target cells (7-AAD / CFSE double positive). The data show the relationship between the percentage of dead target cells [%] and the concentration of bispecific molecules [pM] as an average of two representative healthy donors. Detailed Implementation

[0417] I. definition

[0418] To make this disclosure easier to understand, the selected terms are defined as follows.

[0419] As used herein, the twenty common amino acids and their abbreviations follow conventional usage. Stereoisomers of the twenty common amino acids (e.g., D-amino acids); non-natural amino acids (such as α-disubstituted amino acids, N-alkyl amino acids, lactic acid, and other unconventional amino acids) may also be suitable components of the polypeptide chain of the binding protein of this invention. Examples of unconventional amino acids include: 4-hydroxyproline, γ-carboxyglutamic acid, cN,N,N-trimethyllysine, cN-acetyllysine, O-phosphoserine, N-acetylserine, N-formylmethionine, 3-methylhistidine, 5-hydroxylysine, uN-methylarginine, and other similar amino acids and imino acids (e.g., 4-hydroxyproline). In the polypeptide representation used herein, the left-hand direction is the amino-terminal direction and the right-hand direction is the carboxyl-terminal direction, according to standard usage and convention. Naturally occurring residues can be categorized into several classes based on common side-chain characteristics (see Table 1).

[0420]

[0421] Table 1.

[0422] Conservative amino acid substitutions can involve the exchange of a member from one category with a member from another category. Conservative amino acid substitutions can encompass amino acid residues that are not naturally occurring, typically incorporated through chemical peptide synthesis rather than through biological systems. These include peptide-like compounds and other reversed or inverted forms of amino acid residues. Non-conservative substitutions can involve the exchange of a member from one category with a member from another category.

[0423] As used herein, the term "mutation" or "mutated" refers to an alteration of the amino acid sequence by deletion, insertion, and / or substitution of one or more amino acids. A mutation is introduced with respect to a given sequence (e.g., the amino acid sequence that specifically recognizes the VL1 and / or VH1 pair of epitope 1). The term "non-mutated" refers to any amino acid sequence that exhibits functional properties, such as any sequence that still exhibits binding properties. This is clarified as follows: mutating VH1 / VL1 causes it to bind to the epitope non-specifically. The non-mutated form of this VH1 / VL1 still binds specifically to epitope 1. Therefore, each VH / VL domain of an antibody that binds to any epitope is suitable for mutation to be used as a scaffold protein of the present invention.

[0424] As used herein, the term "variant" refers to an amino acid sequence that is at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% identical to the amino acid sequence from which it is derived (e.g., SEQ ID NO:1 or SEQ ID NO:2). The percentage of identity between two sequences is determined using a mathematical algorithm described by Karlin and Altschul, Proc. Natl. Acad. Sci. USA 90, 5873-5877, 1993. Such an algorithm is incorporated into the BLASTN and BLASTP procedures of Altschul et al. (1990) J. Mol. Biol. 215, 403-410. For vacancy-based alignments used for comparative purposes, vacancy-based BLAST is employed, as described in Altschul et al. (1997) Nucleic Acids Res. 25, 3389-3402. When using the BLAST and vacancy-based BLAST procedures, the default parameters of the respective procedures are used. Alternatively, variants may be defined as having up to 20, 15, 10, 5, 4, 3, 2, or 1 amino acid substitutions, particularly conserved amino acid substitutions. Conservative substitutions are well known in the art (see, for example, Creighton (1984) Proteins. WH Freeman and Company). An overview of the physical and chemical properties of amino acids is given in Table 1 above. In a particular embodiment, a conserved substitution is a substitution made with amino acids that commonly have at least one property according to Table 1 (i.e., column 1 and / or column 2). The term "variant" also includes fragments. Fragments have an N-terminal and / or C-terminal deletion of up to 20, 15, 10, 5, 4, 3, 2, or 1 amino acid in total. Additionally or alternatively, variants may be modified, for example, by adding N-terminal and / or C-terminal amino acids of up to 50, 40, 30, 20, 10, 5, 4, 3, 2, or 1 amino acid in total.

[0425] As used herein, the terms "antigen," "target antigen," or "antigen target" refer to a molecule or part of a molecule that can be specifically bound by the binding protein of the present invention and can also be used in animals to generate antibodies that can specifically bind to an epitope of that antigen. A target antigen may have one or more epitopes. For each target antigen recognized by the binding protein, the binding protein can compete with an intact antibody that recognizes said target antigen.

[0426] As used herein, the terms "binding protein" or "binding polypeptide" refer to a polypeptide (e.g., an antibody or fragment thereof) containing at least one binding site responsible for selectively binding to a target antigen of interest (e.g., a human antigen). Exemplary binding sites include, but are not limited to, variable domains of antibodies, ligand-binding sites of receptors, or receptor-binding sites of ligands. In some aspects, a binding polypeptide contains multiple (e.g., two, three, four, or more) binding sites. In some aspects, a binding protein is not a therapeutic enzyme.

[0427] As used in this article, the term "Her2" or "HER2" refers to human epidermal growth factor receptor 2, which is a member of the epidermal growth factor receptor family.

[0428] As used herein, the term "binding protein" refers to a non-naturally occurring, recombinant, or engineered molecule capable of specifically binding to at least one antigen. In a particular embodiment, the binding protein comprises at least one VH / VL pair that specifically binds to the antigen.

[0429] Generating bispecific binding proteins by co-expressing two light chains and two heavy chains in a single host cell can be highly challenging due to the low yield of the desired bispecific binding protein and the difficulty in removing closely associated mispaired binding protein contaminants (Suresh et al., Proc. Natl. Acad. Sci. USA 83, 7989-7993, 1986). This is because the heavy chain forms a homodimer as well as the desired heterodimer, referred to in this paper as the "heavy chain pairing problem." Additionally, the light chain may mispair with non-homologous heavy chains, referred to in this paper as the "light chain pairing problem." Therefore, in addition to the desired bispecific binding protein, co-expression of the two antibodies may generate up to nine unwanted types.

[0430] As used herein, a “heterodimerization domain” refers to a subunit of a bispecific or multispecific binding protein that promotes, guides, or compels the correct assembly of light chains and their homologous heavy chains to produce the desired protein while preventing mispairing of the corresponding light or heavy chains.

[0431] As used herein, the term "heterodimerized Fc" or "functional fragment of heterodimerized Fc" refers to a mutant form of a constant structural domain (e.g., CH2-CH3 or CH2-CH3-CH4) that is mutated with respect to the naturally occurring Fc moiety because it no longer forms a homodimer but instead forms a heterodimer with the corresponding mutated Fc moiety. Therefore, the term refers to a portion of the two chains forming the heterodimer. Several such pairs are known in the art and include, for example, kilometre hummus (KIH) variants or EV-RWT variants.

[0432] Rigdeway and colleagues generated CH3 interfaces favorable for heterodimer assembly by replacing a small side chain on one CH3 interface with a larger side chain to produce a pestle, and replacing a large side chain on another CH3 domain with a smaller side chain to produce a mortar. Testing of these variants showed preferential heterodimerization. Phage display further extended this original pestle-mortar structure mutation to identify other suitable combinations for generating bispecific IgG antibodies, thereby testing for additional substitutions that allow disulfide bond formation. The pestle-mortar structure variants are further described in U.S. Patent Nos. 5,732,168 and 8,216,805, which are incorporated herein by reference. Thus, in one embodiment, the CH3 domain of one FC domain or heterodimerizing domain contains the mutants Y349C, T366S, L368A, and Y407V, and the CH3 domain of the other FC domain or heterodimerizing domain contains the mutants S354C and T366W (amino acid positions indicated by reference to the IgG1 sequence).

[0433] As used herein, the term "homodimerization domain" refers to a domain that mediates the homodimerization of two similar domains (e.g., two heavy chains). Heavy chain pairing is mediated by the last domain of the constant region (i.e., CH3 in the IgG molecule), which forms a high-affinity homodimer complex (KD of approximately 10 pM). Additional interactions exist in the hinge region responsible for the covalent connection of the two heavy chains formed after heavy chain assembly. Interactions in the CH3 homodimer involve approximately 16 residues at the CH3-CH3 interface, as shown for human γ1CH3, where a patch formed by six residues at the center of the interface (T366, L368, F405, Y407, and K409) strongly promotes stability. Homodimerization domains include, but are not limited to, variants of the Fc region and its effector modifications, fragments of any of these variants, the CH2 domain or a fragment thereof, the CH3 domain or a fragment thereof, the CH4 domain or a fragment thereof, etc.

[0434] Naturally occurring antibodies typically comprise tetramers. Each such tetramer usually consists of two identical pairs of polypeptide chains, each pair having a full-length "light" chain (typically with a molecular weight of about 25 kDa) and a full-length "heavy" chain (typically with a molecular weight of about 50–70 kDa). As used herein, the terms "heavy chain" and "light chain" refer to any immunoglobulin polypeptide having a variable domain sequence sufficient to confer specificity against a target antigen. The amino-terminal portion of each light and heavy chain typically includes a variable domain of about 100 to 110 or more amino acids, which is typically responsible for antigen recognition. The carboxyl-terminal portion of each chain typically defines a constant domain responsible for effector function. Therefore, in naturally occurring antibodies, full-length heavy chain IgG immunoglobulin peptides include a variable domain (VH) and three constant domains (CH1, CH2, and CH3), wherein the VH domain is located at the amino terminus of the peptide and the CH3 domain is located at the carboxyl terminus, and full-length light chain immunoglobulin peptides include a variable domain (VL) and a constant domain (CL), wherein the VL domain is located at the amino terminus of the peptide and the CL domain is located at the carboxyl terminus.

[0435] Human light chains are typically classified as κ and λ light chains, and human heavy chains are typically classified as μ, δ, γ, α, or ε, with antibody isotypes defined as IgM, IgD, IgG, IgA, and IgE, respectively. IgG has several subclasses, including but not limited to IgG1, IgG2, IgG3, and IgG4. IgM has multiple subclasses, including but not limited to IgM1 and IgM2. IgA is similarly subdivided into multiple subclasses, including but not limited to IgA1 and IgA2. Within both the full-length light and heavy chains, variable and constant domains are typically linked by a “J” region of about 12 or more amino acids, and the heavy chain also includes a “D” region of about 10 or more amino acids. See, for example, Fundamental Immunology (edited by Paul, W., Raven Press, 2nd ed., 1989), which is incorporated herein by reference in its entirety for all purposes. The variable region of each light / heavy chain pair typically forms an antigen-binding site. The variable domains of naturally occurring antibodies typically exhibit the same overall structure as relatively conserved framework regions (FRs) linked by three hypervariable regions (also known as complementarity-determining regions or CDRs). The CDRs from the two chains in each pair are usually aligned along the framework regions, which allows for binding to specific epitopes. Both the light and heavy chain variable domains typically contain domains FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4, from the amino terminus to the carboxyl terminus.

[0436] As used herein, the term “CDR set” refers to a group of three CDRs present in a single variable region capable of binding an antigen. The exact boundaries of these CDRs have been defined in different ways according to different systems. The system described by Kabat (Kabat et al., SEQUENCES OF PROTEINS OF IMMUNOLOGICAL INTEREST (National Institutes of Health, Bethesda, Maryland) (1987) and (1991)) not only provides a well-defined residue numbering system applicable to any variable region of an antibody, but also provides precise residue boundaries defining three CDRs. These CDRs may be referred to as Kabat CDRs. Chothia and colleagues (Chothia and Lesk, 1987, J. Affol. Biol. 196:901-17; Chothia et al., 1989, Nature 342:877-83) found that, despite significant diversity at the amino acid sequence level, certain sub-regions within the Kabat CDR adopt nearly identical peptide skeleton architectures. These sub-regions are named L1, L2, and L3 or H1, H2, and H3, where “L” and “H” designate the light chain region and the heavy chain region, respectively. These regions may be referred to as Chothia CDRs, and they have boundaries that overlap with Kabat CDRs. Other boundaries of CDRs that overlap with Kabat CDRs have been described in Padlan, 1995, FASEB J.9:13339; MacCallum, 1996, J.Mol.Biol.262(5):732-45; and Lefranc, 2003, Dev.Comp.Immunol.27:55-77. There are still other CDR boundary definitions that may not strictly follow one of the systems in this paper, but will still overlap with Kabat CDRs, although these other CDR boundaries may be shortened or lengthened given that specific residues or groups of residues or even the entire CDR do not significantly affect the prediction or experimental findings of antigen binding. The methods used in this paper can utilize the CDR defined according to any of these systems, but some implementations use the CDR defined by Kabat or Chothia. Identification of the predicted CDR using amino acid sequences is well known in the art, as in Martin, AC, “Protein sequence and structure analysis of antibody variable domains”, in Antibody Engineering, Vol. 2, edited by Kontermann R., Dikel S., Springer-Verlag, Berlin, pp. 33-51 (2010).The amino acid sequence of the heavy chain and / or light chain variable domains can also be examined to identify the CDR sequence using other conventional methods, such as comparing it with known amino acid sequences of other heavy and light chain variable regions to identify regions with sequence hypervariability. Numbered sequences can be aligned visually or using an alignment program (such as one from the CLUSTAL program suite), as described in Thompson, 1994, Nucleic Acids Res. 22:4673-80. Molecular models are routinely used to correctly depict the framework and CDR regions, thereby correcting sequence-based alignments.

[0437] In some implementations, the CDR / FR definition in the immunoglobulin light or heavy chain should be determined based on the IMGT definition (Lefranc et al., Dev. Comp. Immunol., 2003, 27(1): 55-77; www.imgt.org).

[0438] As used herein, the term "Fc" refers to a molecule (in monomeric or multimeric form) containing a sequence of a non-antigen-binding fragment produced by antibody digestion or by other means, and may contain a hinge region. While IgG1 and IgG2 are used in exemplary embodiments, the original immunoglobulin source of native Fc is typically human and can be any immunoglobulin. Fc molecules consist of monomeric polypeptides that can be covalently (i.e., disulfide bonds) and non-covalently associated to form dimers or multimers. The number of intermolecular disulfide bonds between the monomeric subunits of a native Fc molecule ranges from 1 to 4, depending on the class (e.g., IgG, IgA, and IgE) or subclass (e.g., IgG1, IgG2, IgG3, IgA1, and IgGA2). An example of Fc is a disulfide-bonded dimer produced by papain digestion of IgG. As used herein, the term "native Fc" is general for monomeric, dimer, and multimeric forms.

[0439] F(ab) fragments typically comprise a light chain and a heavy chain with VH and CH1 domains, wherein the VH-CH1 heavy chain portion of the F(ab) fragment cannot form a disulfide bond with another heavy chain polypeptide. As used herein, F(ab) fragments may also comprise a light chain containing two variable domains separated by amino acid linkers and a heavy chain containing two variable domains and a CH1 domain separated by amino acid linkers.

[0440] The F(ab') fragment typically includes a light chain and a portion of a heavy chain containing more constant regions (between the CH1 and CH2 domains), allowing interchain disulfide bonds to form between the two heavy chains to form the F(ab')2 molecule.

[0441] As used in this article, the term "T" m "T" refers to the melting temperature of binding proteins, antigen-binding proteins, and antibodies, and is a key parameter for the thermal stability of antigen-binding proteins. m This generally involves the thermal stability of the Fv segment (i.e., the variable region heavy chain and light chain (VH / VL)). T can be measured by differential scanning calorimetry (DSC). m .

[0442] One embodiment of this disclosure provides a binding protein with biological and immune specificity to one to three target antigens. Another embodiment of this disclosure provides a nucleic acid molecule comprising a nucleotide sequence encoding a polypeptide chain forming such a binding protein. Yet another embodiment of this disclosure provides an expression vector comprising a nucleic acid molecule containing a nucleotide sequence encoding a polypeptide chain forming such a binding protein. Still another embodiment of this disclosure provides a host cell that expresses such a binding protein (i.e., a nucleic acid molecule or vector comprising a polypeptide chain encoding such a binding protein).

[0443] As used herein, the terms “antigen” or “target antigen” or “antigen target” refer to a molecule or part of a molecule that can be bound by a binding protein and, additionally, can be used in animals to generate antibodies that can bind to an epitope of that antigen. A target antigen may have one or more epitopes. For each target antigen recognized by a binding protein, the binding protein can compete with an intact antibody that recognizes that target antigen.

[0444] As used herein, the terms “epitope” or “target epitope” or “epitope target” refer to any determinant, such as a polypeptide determinant, capable of specifically binding to an immunoglobulin or T-cell receptor. For example, but by no means limited, target epitope A may be a first epitope on a first antigen, and target epitope B may be a second epitope on a first antigen. Alternatively, target epitope B may be a second epitope on a second antigen. In some embodiments, the epitope determinant comprises chemically active surface groups of a molecule, such as amino acids, sugar side chains, phosphoryl groups, or sulfonyl groups, and in some embodiments may have specific three-dimensional structural features and / or specific charge features. An epitope is an antigenic region bound by an antibody or by an antigen-binding fragment of an antibody or by a binding protein. In some embodiments, when a binding protein preferentially recognizes its target antigen in a complex mixture of proteins and / or macromolecules, it is said to bind antigen specifically. In some embodiments, when the equilibrium dissociation constant is <10 -8 When M, the equilibrium dissociation constant is <1 -9 When M, or when the dissociation constant is <10 -10 When M occurs, it is said that the binding protein specifically binds to the antigen.

[0445] As used herein, the term "linker" refers to 0-100 consecutive amino acid residues. Linkers may be present or absent, and may be identical or different. Linkers may all have the same amino acid sequence or may all have different amino acid sequences.

[0446] In some embodiments, the term "linker" refers to 1-15 consecutive amino acid residues. Typically, linkers provide flexibility and spatial separation between two amino acids or between two polypeptide domains. Linkers can be inserted between VH, VL, CH, and / or CL domains, depending on the molecular form, to provide sufficient flexibility and mobility for the light and heavy chain domains, for example, to fold into a cross-variable region immunoglobulin. At the amino sequence level, linkers are typically inserted at transitions between variable domains, between variable and knockout domains, or between variable and constant domains, respectively. Since the approximate size of immunoglobulin domains is well understood, transitions between domains can be identified. The precise location of a domain transition can be determined by locating peptide extensions where secondary structural elements (such as β-sheets or α-helices) have not formed (as demonstrated by experimental data or as can be determined by techniques such as modeling or secondary structure prediction). In some exemplary embodiments, linkers can be inserted between Fab domains to generate tandem Fab antibodies. In a particular embodiment, a linker can be inserted between the N-terminus of the VH domain of a first Fab and the C-terminus of the CH1 domain of a second Fab.

[0447] The identity and sequence of amino acid residues in a linker can vary depending on the type of one or more secondary structural elements to be achieved in the linker. For example, glycine, serine, and alanine are suitable for linkers with the greatest flexibility. If a more rigid and elongated linker is required, certain combinations of glycine, proline, threonine, and serine are useful. Depending on the desired properties, any combination of amino acid residues with other amino acid residues can be considered a linker to construct larger peptide linkers.

[0448] In some embodiments, the linker comprises: a single glycine (Gly) residue; a diglycine peptide (Gly-Gly); a tripeptide (Gly-Gly-Gly); a peptide having four glycine residues (Gly-Gly-Gly-Gly; SEQ ID NO: 104); a peptide having five glycine residues (Gly-Gly-Gly-Gly-Gly; SEQ ID NO: 105); a peptide having six glycine residues (Gly-Gly-Gly-Gly-Gly-Gly; SEQ ID NO: 106); a peptide having seven glycine residues (Gly-Gly-Gly-Gly-Gly-Gly; SEQ ID NO: 107); and a peptide having eight glycine residues (Gly-Gly-Gly-Gly-Gly-Gly-Gly; SEQ ID NO: 108).

[0449] In some implementations, the linker contains small amino acids such as Gly, Ala, or Ser.

[0450] In some embodiments, the connector comprises Gly and Ser, or GS, GGS, GGGS (SEQ ID NO: 109) or GGGGS (SEQ ID NO: 102). In some embodiments, the connector comprises (Gly-Gly-Gly-Gly-Ser)2 (i.e., (GGGGS)2 (SEQ ID NO: 103)). In some embodiments, the connector comprises (Gly-Gly-Gly-Gly-Ser)3 (i.e., (GGGGS)3 (SEQ ID NO: 110)).

[0451] In some embodiments, the linker comprises Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 102), the peptide Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO:103), the peptide Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO:110) and the peptide Gly-Gly-Ser-Gly-Ser-Ser-Gly-Ser-Gly-Gly (SEQ ID NO:111).

[0452] In some embodiments, the linker comprises a single Ser residue; a single Val residue; a dipeptide selected from Arg-Thr, Gln-Pro, Ser-Ser, Thr-Lys, and Ser-Leu; Thr-Lys-Gly-Pro-Ser (SEQ ID NO:112), Thr-Val-Ala-Ala-Pro (SEQ ID NO:113), Gln-Pro-Lys-Ala-Ala (SEQ ID NO:114), Gln-Arg-Ile-Glu-Gly (SEQ ID NO:115); Ala-Ser-Thr-Lys-Gly-Pro-Ser (SEQ ID NO:116), Arg-Thr-Val-Ala-Ala-Pro-Ser (SEQ ID NO:117), and Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO:118). NO:118), His-Ile-Asp-Ser-Pro-Asn-Lys (SEQ ID NO:119) and Asp-Lys-Thr-His-Thr (SEQ ID NO:120).

[0453] In some embodiments, the two tandem Fabs are connected via a (Gly-Gly-Gly-Gly-Ser)2 connector (SEQ ID NO:103). In some embodiments, the connector between the stable knockout domain and the VH / VL pair is a (Gly-Gly-Gly-Gly-Ser)2 connector (SEQ ID NO:103).

[0454] In some embodiments having a CODV-Fab portion, L1 and L2 are located on the light chain and L3 and L4 are located on the heavy chain, wherein L1 is 3 to 12 amino acid residues long, L2 is 3 to 14 amino acid residues long, L3 is 1 to 8 amino acid residues long, and L4 is 1 to 3 amino acid residues long. In some embodiments, L1 is 5 to 10 amino acid residues long, L2 is 5 to 8 amino acid residues long, L3 is 1 to 5 amino acid residues long, and L4 is 1 to 2 amino acid residues long. In some embodiments, L1 is 7 amino acid residues long, L2 is 5 amino acid residues long, L3 is 1 amino acid residue long, and L4 is 2 amino acid residues long. In some embodiments, L1 is 10 amino acid residues long, L2 is 10 amino acid residues long, L3 is 0 amino acid residues long, and L4 is 0 amino acid residues long. In some embodiments, L1, L2, L3, and L4 each have an independently selected length of 0 to 15 amino acids (e.g., 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 amino acids), wherein at least two linkers have a length of 1 to 15 amino acids (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 amino acids). In some embodiments, L1, L2, L3, and L4 are Asp-Lys-Thr-His-Thr (SEQ ID NO: 120). In some embodiments, one or more linkers comprise the sequence Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO: 118). In some embodiments, L1 comprises the sequence Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO: 118). In some embodiments, L1 contains the sequence Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO: 118), L2 contains the sequence Thr-Lys-Gly-Pro-Ser-Arg (SEQ ID NO: 121), L3 contains the sequence Ser, and L4 contains the sequence Arg-Thr. In some embodiments, L3 contains the sequence Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO: 118). In some embodiments, L1 contains the sequence Ser, L2 contains the sequence Arg-Thr, L3 contains the sequence Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO: 118), and L4 contains the sequence Thr-Lys-Gly-Pro-Ser-Arg (SEQ ID NO: 121).

[0455] In some implementations, L1, L2, L3, and L4 each independently contain a sequence selected from the following: (Gly-Gly-Gly-Gly-Ser) n(where n is an integer between 0 and 5; SEQ ID NO: 122), Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 103), Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO:110), Ser, Arg-Thr, Thr-Lys-Gly-Pro-Ser (SEQ ID NO:112), Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO:118) and Gly-Gly-Ser-Gly-Ser-Ser-Gly-Ser-Gly-Gly (SEQ ID NO:111). In some embodiments, L1 contains the sequence Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO: 118), L2 contains the sequence Thr-Lys-Gly-Pro-Ser (SEQ ID NO: 112), L3 contains the sequence Ser, and L4 contains the sequence Arg-Thr. In some embodiments, L1 contains the sequence Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 103), L2 contains the sequence Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 103), L3 is 0 amino acids long, and L4 is 0 amino acids long. In some embodiments, L1 comprises the sequence Gly-Gly-Ser-Gly-Ser-Ser-Ser-Gly-Gly (SEQ ID NO: 111), L2 comprises the sequence Gly-Gly-Ser-Gly-Ser-Ser-Gly-Gly (SEQ ID NO: 111), L3 is 0 amino acids long, and L4 is 0 amino acids long. In some embodiments, L1 comprises the sequence Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 110), L2 is 0 amino acids long, L3 comprises the sequence Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 110), and L4 is 0 amino acids long.In some embodiments, L1 and L2 are zero amino acids in length, and L3 and L4 each contain sequences independently selected from (Gly-Gly-Gly-Gly-Ser). n (where n is an integer between 0 and 5; SEQ ID NO: 122), Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 103), Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO:110), Ser, Arg-Thr, Thr-Lys-Gly-Pro-Ser (SEQ ID NO:112), Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO:118) and Gly-Gly-Ser-Gly-Ser-Ser-Gly-Ser-Gly-Gly (SEQ ID NO:111). In some implementations, L3 and L4 are zero amino acids in length, and L1 and L2 each contain sequences independently selected from the following: (Gly-Gly-Gly-Gly-Ser). n (where n is an integer between 0 and 5; SEQ ID NO: 122), Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 103), Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser-Gly-Gly-Gly-Gly-Ser (SEQ ID NO:110), Ser, Arg-Thr, Thr-Lys-Gly-Pro-Ser (SEQ ID NO:112), Gly-Gln-Pro-Lys-Ala-Ala-Pro (SEQ ID NO:118) and Gly-Gly-Ser-Gly-Ser-Ser-Gly-Ser-Gly-Gly (SEQ ID NO:111).

[0456] In some embodiments, one or more adapters comprise sequences derived from naturally occurring sequences located at the junction between antibody variable domains and antibody constant domains (e.g., as described in WO 2012 / 135345). For example, in some embodiments, the adapter comprises a sequence found between an endogenous VH domain and a CH1 domain, or at the transition between an endogenous VL domain and a CL domain (e.g., κ or λ). In some embodiments, the adapter comprises a sequence found between an endogenous human VH domain and a CH1 domain, or at the transition between an endogenous human VL domain and a CL domain (e.g., human κ or λ).

[0457] The examples listed above are not intended to limit the scope of this disclosure in any way, and linkers containing randomly selected amino acids are suitable for use in the binding proteins described herein, wherein the randomly selected amino acids are selected from valine, leucine, isoleucine, serine, threonine, lysine, arginine, histidine, aspartic acid, glutamic acid, asparagine, glutamine, glycine, and proline. Further descriptions of linker sequences can be found, for example, in WO2012135345 and WO 2017 / 180913, which are incorporated by reference.

[0458] As used herein, the term "valence" refers to the number of binding sites of a binding protein, epitope, antigen-binding protein, or antibody. For example, the term "monovalent binding protein" refers to a binding protein having one antigen-binding site. The term "bivalent binding protein" refers to a binding protein having two binding sites. The term "trivalent binding protein" refers to a binding protein having three binding sites. The term "tetravalent binding protein" refers to a binding protein having four binding sites. In a particular embodiment, a bivalent binding protein may bind to one antigen target. In other embodiments, a bivalent binding protein may bind to two different antigen targets. In a particular embodiment, a trivalent binding protein may bind to one antigen target, i.e., it is monospecific. In other embodiments, a trivalent binding protein may bind to two different antigen targets, i.e., it is bispecific. In other embodiments, a trivalent binding protein may bind to three different antigen targets, i.e., it is trispecific. In a particular embodiment, a tetravalent binding protein may bind to one antigen target, i.e., it is monospecific. In other embodiments, a tetravalent binding protein may bind to two different antigen targets, i.e., it is bispecific. In other embodiments, the tetravalent binding protein can bind to three different antigen targets, i.e., it is trispecific. In other embodiments, the tetravalent binding protein can bind to four different antigen targets, i.e., it is quadrispecific.

[0459] As used herein, the term "specificity" refers to the number of binding specificities of a binding protein, epitope, antigen-binding protein, or antibody. For example, the term "monospecific binding protein" refers to a binding protein that specifically binds to one antigen target. The term "bispecific binding protein" refers to a binding protein that specifically binds to two different antigen targets. The term "trispecific binding protein" refers to a binding protein that specifically binds to three different antigen targets. The term "tetraspecific binding protein" refers to a binding protein that specifically binds to four different antigen targets, and so on.

[0460] As used herein, the term "selective recognition site" refers to a modification in a binding protein that allows selective recognition by an affinity reagent that binds to the selective recognition site. Examples of selective recognition sites include binding sites for protein A in the Fc portion of immunoglobulins.

[0461] As used herein, the term "affinity reagent" refers to a reagent containing a ligand immobilized on a matrix and specifically binding to surface groups of a molecule, such as amino acid or sugar side chains, and typically possessing specific three-dimensional structural features and specific charge characteristics. Affinity reagents are tools in affinity chromatography, enabling purification through specific interactions between the ligand and the product. An example of an affinity reagent is "protein L," which refers to recombinant protein L immobilized on a matrix to form a ligand with affinity for a subset of the variable domains of the immunoglobulin κ light chain. Such a matrix can be a resin. Another example of an affinity reagent is "KappaSelect," which refers to a recombinant camel-derived 13kDa single-chain antibody immobilized on a matrix to form a ligand with affinity for the constant domains of the human immunoglobulin κ light chain. Yet another example of an affinity reagent is protein A. Protein A is a 42kDa surface protein originally discovered on the cell wall of the bacterium Staphylococcus aureus. Crystallographic refinement has revealed that the major binding site of protein A is located in the Fc region, between the CH2 and CH3 domains. Furthermore, protein A has been shown to bind to human IgG molecules containing the IgG F(ab')2 fragment from the human VH3 gene family. Protein A can bind with strong affinity to the Fc portion of immunoglobulins from certain species.

[0462] The dissociation constant (K) of the bound protein D) can be determined, for example, by surface plasmon resonance. Typically, surface plasmon resonance analysis uses a BIAcore system (Pharmacia Biosensor; Piscatavir, NJ) to measure the real-time binding interaction between the ligand (target antigen on the biosensor matrix) and the analyte (binding protein in solution) via surface plasmon resonance (SPR). Surface plasmon resonance analysis can also be performed by immobilizing the analyte (binding protein on the biosensor matrix) and presenting the ligand (target antigen). As used herein, the term "K" is used... D "" refers to the dissociation constant of the interaction between a specific binding protein and a target antigen.

[0463] As used herein, the term "specifically binding" refers to binding a protein or its antigen-binding fragment at a rate of at least about 1 x 10-1. -6 M, 1x10 -7 M, 1x10 -8 M, 1x10 -9 M, 1x10 -10 M, 1x10 -11 M, 1x10 -12 The ability of M or larger Kd molecules to bind antigens containing epitopes, and / or the ability to bind epitopes with an affinity at least twice that of the binding protein or its antigen-binding fragment for nonspecific antigens. The binding affinity of antigens to binding proteins or antibodies can be determined using surface plasmon resonance (SPR) with a BIAcore instrument.

[0464] As used herein, the term "reference Fab molecule" refers to a molecule identical to a pseudo-Fab molecule except that, in pseudo-Fab molecules, the CH1 and CL domains of the reference Fab molecule are replaced by VHX and VLX domains. A "reference Fab molecule" is a molecule in which the variable domains are identical to those of the pseudo-Fab molecule and which also possess CH1 and CL domains. In pseudo-Fab molecules, the CH1 and CL domains of the reference Fab molecule are replaced by VHX and VLX domains.

[0465] As used herein, the term "nucleic acid" refers to a polymeric or oligomeric macromolecule, or biomolecule, that is essential for all known forms of life. Nucleic acids, including DNA (deoxyribonucleic acid) and RNA (ribonucleic acid), are derived from monomers called nucleotides. Most naturally occurring DNA molecules consist of two complementary biopolymer chains that coil around each other to form a double helix. DNA chains are also referred to as polynucleotides, which are composed of nucleotides. Each nucleotide consists of a nitrogenous nucleotide base and a monosaccharide called deoxyribose or ribose, and a phosphate group. Naturally occurring nucleotides include guanine (G), adenine (A), thymine (T), uracil (U), or cytosine (C). Nucleotides are linked together in the chain by covalent bonds between the sugar of one nucleotide and the phosphate of the next, creating an alternating sugar-phosphate backbone. If the sugar is deoxyribose, the polymer is DNA. If the sugar is ribose, the polymer is RNA. Typically, polynucleotides are formed by phosphodiester bonds between individual nucleotide monomers.

[0466] As used herein, the term "polynucleotide" refers to a single-stranded or double-stranded nucleic acid polymer with a length of at least 10 nucleotides. It should be understood that nucleotides containing polynucleotides can be ribonucleotides or deoxyribonucleotides or modified forms of either type of nucleotide. Such modifications include base modifications (such as bromouridine), ribose modifications (such as cytarabine and 2',3'-dideoxyribose), and internucleotide linkage modifications (such as thiophosphates, dithiophosphates, phosphoroselenoate, diselenophosphate, phosphoroanilothioate, phoshoraniladate, and phosphoroamidate). The term "polynucleotide" specifically includes both single-stranded and double-stranded forms of DNA.

[0467] "Separated polynucleotide" is a polynucleotide of genomic, cDNA or synthetic origin or some combination thereof, wherein the separated polynucleotide: (1) is not associated with all or part of the polynucleotide in which the separated polynucleotide is found in nature, (2) is associated with a polynucleotide in which it is not connected, or (3) is not present in nature as part of a larger sequence.

[0468] An “isolated polypeptide” is a polypeptide that: (1) is free from at least some other polypeptides that are normally found with it; (2) is substantially free from other polypeptides from the same source (e.g., from the same species); (3) is expressed by cells from a different species; (4) has been isolated from at least about 50% of the polynucleotides, lipids, carbohydrates, or other materials associated with it in nature; (5) is not associated with portions of polypeptides associated with the “isolated polypeptide” in nature (through covalent or non-covalent interactions); (6) is operatively associated with polypeptides that are not associated with it in nature (through covalent or non-covalent interactions); or (7) is not present in nature. Such an isolated polypeptide may be encoded by genomic DNA, cDNA, mRNA, or other RNA of synthetic origin, or any combination thereof. In an exemplary embodiment, the isolated polypeptide is substantially free from polypeptides or other contaminants found in its natural environment that would interfere with its application (therapeutic, diagnostic, preventative, research, or other applications).

[0469] As used herein, the term "expression vector" is also referred to as an expression construct, and generally refers to a plasmid or virus designed for protein expression in a cell. The term "vector" refers to a protein or polynucleotide or a mixture thereof that can be introduced into a cell or that can introduce the protein and / or nucleic acid contained therein into a cell. Examples of vectors include, but are not limited to, plasmids, phages, viruses, or artificial chromosomes. In particular, vectors are used to transport gene products of interest (e.g., foreign or heterologous DNA) into a suitable host cell. Vectors may contain "replicon" polynucleotide sequences that facilitate autonomous replication of the vector within the host cell. Foreign DNA is defined as heterologous DNA, which is DNA not naturally found in the host cell, such as replication vector molecules, encoding selectable or screenable markers, or encoding transgenes. Once inside the host cell, the vector can replicate independently of or simultaneously with the host chromosomal DNA, and may produce several copies of the vector and its inserted DNA. Additionally, the vector may contain essential elements that allow the inserted DNA to be transcribed into mRNA molecules or otherwise cause the inserted DNA to be replicated into multiple copies of RNA. Vectors can also contain “expression control sequences” that regulate the expression of genes of interest. Typically, expression control sequences are polypeptides or polynucleotides, such as, but not limited to, promoters, enhancers, silencers, insulators, or repressors. In vectors containing more than one polynucleotide encoding one or more gene products of interest, expression can be controlled together or individually by one or more expression control sequences. More specifically, each polynucleotide contained in the vector can be controlled by a separate expression control sequence, or all polynucleotides contained in the vector can be controlled by a single expression control sequence. Polynucleotides contained in a single vector controlled by a single expression control sequence can form an open reading frame. Some expression vectors additionally contain sequence elements adjacent to the inserted DNA that increase the half-life of the expressed mRNA and / or allow the translation of the mRNA into protein molecules. Therefore, numerous mRNA and polypeptide molecules encoded by the inserted DNA can be rapidly synthesized.

[0470] As used herein, the term "host cell" refers to a cell in which a recombinant expression vector has been introduced. The term "recombinant host cell" or "host cell" is intended not only to refer to a specific test cell but also to its progeny. Because progeny cells may undergo modifications due to mutations or environmental influences, they may actually differ from parent cells, but such cells are still included within the scope of the term "host cell" as used herein. Numerous host cell expression systems can be used to express binding proteins, including bacterial, yeast, baculovirus, and mammalian expression systems (as well as phage display expression systems). A suitable example of a bacterial expression vector is pUC19. For recombinant expression of the binding protein, a host cell is transformed or transfected with one or more recombinant expression vectors carrying a DNA fragment encoding a polypeptide chain of the binding protein, such that the polypeptide chain is expressed in the host cell and secreted into the culture medium in which the host cell is cultured in an exemplary embodiment, thereby allowing the binding protein to be recovered from the culture medium.

[0471] As used herein, the term "pharmaceutical composition" refers to a compound or composition that, when appropriately administered to a patient, can induce a desired therapeutic effect.

[0472] As used herein, the terms “pharmaceutically acceptable carrier” or “physiologically acceptable carrier” refer to one or more formulation materials suitable for achieving or enhancing the delivery of binding proteins.

[0473] The terms "effective amount" and "therapeutic effective amount" when used with respect to a pharmaceutical composition comprising one or more binding proteins (e.g., antibodies or antigen-binding fragments thereof) refer to an amount or dose sufficient to produce the desired therapeutic outcome. More specifically, a therapeutic effective amount is an amount of binding protein (e.g., an antibody or antigen-binding fragment thereof) sufficient to inhibit one or more clinically defined pathological processes associated with the treated condition for a period of time. Effective amounts can vary depending on the specific antibody-like binding protein used and also depend on a variety of factors and conditions related to the patient being treated and the severity of the disorder. For example, if a binding protein or multispecific binding protein is to be administered in vivo, factors such as the patient's age, weight, and health status, as well as dose-response curves and toxicity data obtained in preclinical animal work, would be among those considered. The determination of an effective amount or therapeutic effective amount of a given pharmaceutical composition is entirely within the competence of those skilled in the art.

[0474] As used herein, the term "method for producing binding proteins" refers to a recombinant method of protein expression using techniques well known in the art.

[0475] II. fake Fab section

[0476] In some embodiments, the binding molecule described herein includes at least one pseudo-Fab moiety. As used herein, the "pseudo-Fab" moiety is similar to the Fab moiety of a conventional antibody because it includes a functional antigen-binding portion formed by pairing a variable light chain (VL) domain with a variable heavy chain (VH) domain. However, while the VL and VH domains of a conventional Fab are directly fused or linked to a constant light chain (CL) domain and a constant heavy chain 1 (CH1) domain, respectively, the pseudo-Fab moiety lacks the CH1 and CL domains. Instead, the VL and VH domains of the pseudo-Fab are operatively linked to a second pair of stable knockout VL and VH domains (referred to herein as VLX and VHX), which form an inactive or non-functional binding portion (referred herein as a "stable knockout" portion or domain) that cannot specifically bind to the target antigen. The stable knockout domains described herein cannot specifically bind to any target antigen. The pseudo-Fab moiety lacks the CH and CL domains.

[0477] Although it cannot selectively bind to target antigens, the VLX and VHX domains of pseudoFab still preferentially associate with each other to form stable chain pairs. Therefore, by attaching pseudoFab to one or more additional binding domains with different specificities, the inherent stability of the VLX / VHX chain pairing of pseudoFab can drive the heterodimerization of the chains of the desired multispecific binding molecules.

[0478] Therefore, the pseudo-Fab of this disclosure comprises or consists of a first polypeptide chain and a second polypeptide chain, the first polypeptide chain having a structure represented by the following formula:

[0479] (I)VHa-L1-VHX; or

[0480] (II)VHb-L1-VHX

[0481] Furthermore, the second polypeptide chain has a structure represented by the following formula:

[0482] (III) VLa-L2-VLX; or

[0483] (IV)VLb-L2-VLX

[0484] in

[0485] VHX associates with VLX to form a knockout domain.

[0486] VH associates with VL to form the first functional antigen-binding domain, and

[0487] L1 and L2 are connectors that may or may not be present.

[0488] In some embodiments, the first polypeptide chain of the pseudo-Fab has the structure VH-L1-VHX, and the second polypeptide chain of the pseudo-Fab has the structure VL-L2-VLX.

[0489] In other embodiments, the first polypeptide of the pseudo-Fab has the structure VHX-L1-VH, and the second polypeptide chain has the structure VLX-L2-VL.

[0490] In some implementations, the binding protein comprises separate protein chains selected from one of the following groups:

[0491] (a) VHa-CH1-L1-VHb-L2-VHX and VLa-CL and VLb-L3-VLX;

[0492] (b) VHa-L2-VHX-L1-VHb-CH1 and VLa-L3-VLX and VLb-CL;

[0493] (c) VHa-CH1-L1-VHa-CH1 and VHb-L2-VHX-L3-VHb-L4-VHX and two chains VLb-L5-VLX and two chains VLa-CL;

[0494] The chains in (a) and (b) may exist once or twice, and L1, L2, L3, L4 and L5 may be the same or different joints independently.

[0495] (a) Knockout of structural domains

[0496] The "knockout" domain of pseudoFab can be generated by any means that results in the elimination or reduction of binding affinity or specificity of the normally functional antigen-binding site. In some embodiments, the knockout domain of pseudoFab has been rendered inactive or nonfunctional by mutations in one or more of the VLX and VHX domains of the knockout domain. In one embodiment, the knockout modification is an amino acid substitution. In other embodiments, the knockout modification is an amino acid insertion or deletion. In yet another embodiment, the knockout modification is a combination of one or more amino acid substitutions, insertions, and deletions. In still other embodiments, the knockout domain is rendered nonfunctional by covalent modification with, for example, a portion that interferes with the ability of the variable domain to bind to the target antigen.

[0497] In some embodiments, knockout modification eliminates binding by generating repulsion or disruption of the stable antigen-binding protein complex. In some embodiments, knockout modification may include replacing amino acids that normally form contact with the target antigen with amino acids that generate charge-charge repulsion with the target antigen. Alternatively or additionally, mutations that destabilize the π-π interacting complex may be introduced.

[0498] In some embodiments, knockout modification involves replacing a charged amino acid with an uncharged amino acid. In other embodiments, knockout modification involves replacing an uncharged amino acid with a charged amino acid. In other embodiments, knockout modification involves replacing a polar amino acid with a nonpolar amino acid. In other embodiments, knockout modification involves replacing a charged amino acid with a polar and uncharged amino acid, and replacing a polar and uncharged amino acid with a nonpolar hydrophobic amino acid.

[0499] In some embodiments, the knockout modification is introduced at the amino acid position that interacts with the antigen formation binding. For example, the knockout modification may be located on the protein surface where antigen-antibody interactions typically occur. In some exemplary embodiments, the modification may be introduced into the complementarity-determining region (CDR) of one or both of the VLX or VHX domains.

[0500] In some embodiments, the knockout modification is the substitution of residues in CDRH1 of the VHX domain. In another embodiment, the knockout modification is the substitution of residues in CDRH2 of the VHX domain. In yet another embodiment, the knockout modification is the substitution of residues in CDRH3 of the VHX domain.

[0501] In some embodiments, the knockout modification is the substitution of residues in CDRL1 of the VLX domain. In another embodiment, the knockout modification is the substitution of residues in CDRL2 of the VLX domain. In yet another embodiment, the knockout modification is the substitution of residues in CDRL3 of the VLX domain.

[0502] In some exemplary embodiments, the arginine in the CDR of the VHX or VLX domain is mutated to glutamic acid. In other embodiments, one or more tyrosines in the CDR of the VHX or VLX domain are mutated to alanine.

[0503] In some embodiments, the modification produces a knockout domain that has absolutely no target antigen binding activity to which its functional non-knockout (i.e., wild-type) counterpart is derived. Alternatively, the binding functionality of the knockout domain may be substantially reduced compared to its functional non-knockout counterpart, while still retaining a certain level of detectable binding.

[0504] Scaffolds for generating knockout domains can be obtained, for example, from protein databases (PDBs). PDBs are crystallographic databases of three-dimensional structural data for large biomolecules such as proteins and nucleic acids. Knockout domains can be generated by specifically mutating amino acids, which are predicted to be involved in antigen binding through computer-based modeling of the antigen-binding domain and its homologous target antigen. Subsequent binding studies can be performed using methods known in the art (e.g., surface plasmon resonance) and reveal with high precision whether binding function has been eliminated and whether the interaction between the binding domain and its target has been disrupted.

[0505] (b) Stable knockout domain

[0506] In some embodiments, the knockout domain is a stable knockout domain that exhibits increased thermal stability compared to its wild-type counterpart. For example, in some exemplary embodiments, the melting temperature (T0) of the knockout domain is... m Within a temperature range of at least 0.25°C, at least 0.5°C, at least 0.75°C, at least 1°C, at least 2°C, at least 3°C, at least 4°C, at least 5°C, or at least 10°C compared to the wild-type counterpart from which it is derived. In other exemplary embodiments, the T domain is knocked out. m This represents an increase compared to its wild-type counterparts. For example, compared to the wild-type counterparts from which it was derived, T... m The temperature can be increased by at least 0.25°C, at least 0.5°C, at least 0.75°C, at least 1°C, at least 2°C, at least 3°C, at least 4°C, at least 5°C, or at least 10°C. Thermal stability can be measured by differential scanning fluorometry (DSF) or other bioanalytical methods routinely used by those skilled in the art.

[0507] In some embodiments, the pseudo-Fab comprises an engineered intrachain disulfide bond between the VHX or VLX domains of the knockout domain of the pseudo-Fab, which imparts enhanced stability. Typically, this modification involves replacing at least one amino acid in the VHX domain with a cysteine ​​(Cys) residue and at least one amino acid in the VLX domain with a Cys residue. Cys residues can be introduced at positions in the VHX and VLX domains that allow disulfide bond formation after dimerization. In some embodiments, mutations can be made at the VH and VL interface to improve stability at the VH / VL interface. Specific amino acid mutants in the VH and VL domains can improve stability by introducing non-natural cysteine ​​residues that form disulfide bridges.

[0508] A first set of disulfide bond stabilizing mutations can be made on amino acid residues in the VH and VL domains. In an exemplary embodiment, the disulfide bond is formed by Cys at position 44 of the VHX domain and Cys at position 100 of the VLX domain. Alternatively, this set of disulfide bond stabilizing mutations may be referred to as the “VH44C / VL100C” mutation set. The first set of disulfide bond stabilizing mutations is described in further detail in Reiter et al., Nature Biotechnology, Vol. 14, pp. 1239-1245, 1996, which is incorporated herein by reference for all purposes.

[0509] A second set of disulfide bond stabilizing mutations can be made to amino acid residues in the VH and VL domains. In an exemplary embodiment, the disulfide bond is formed by Cys at position 105 of the VHX domain and Cys at position 43 of the VLX domain. Alternatively, the second set of disulfide bond stabilizing mutations may be referred to as the “VH105C / VL43C” mutation set. Other disulfide bond stabilizing mutations are further described in detail in U.S. Patent No. 9,527,927, which is incorporated herein by reference for all purposes.

[0510] In some embodiments, the VLX domain of the pseudo-Fab contains a variant of SEQ ID NO:1, said variant having at least one knockout modification. For example, the VLX domain may contain at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% or 100% of the amino acid sequence of SEQ ID NO:1, if no knockout modification is used.

[0511] In some embodiments, the VHX domain of the pseudo-Fab contains a variant of SEQ ID NO:2, said variant having at least one knockout modification. For example, the VHX domain may contain at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% or 100% of the amino acid sequence of SEQ ID NO:2, if no knockout modification is used.

[0512] In one embodiment, the VLX domain of the pseudo-Fab contains the amino acid sequence of SEQ ID NO:76, and the VHX domain of the pseudo-Fab contains an amino acid sequence selected from SEQ ID NO:77, SEQ ID NO:78 and SEQ ID NO:79.

[0513] II. Multispecific binding peptides containing pseudofabe

[0514] In other respects, multispecific binding proteins incorporating the pseudoFab moiety described herein are provided. The highly modular nature of the pseudoFab moiety allows for the formation of numerous multispecific structures.

[0515] In some embodiments, the multispecific binding protein of this disclosure includes an N-terminus and / or a C-terminus attached to one or both strands of the pseudoFab moiety to form additional binding specificity of the pseudoFab-containing multivalent binding protein.

[0516] In some implementations, the multispecific binding protein includes:

[0517] a) The first pseudo-Fab portion, which includes:

[0518] (1) A first VL domain (VLa), which pairs with a first VH domain (VHa) to form a first antigen-binding site for binding target antigen A;

[0519] (2) A first stable knockout VL domain (VLX) is paired with a first stable knockout VH domain (VHX) to form a first disulfide bond stable knockout (dsKO) domain.

[0520] (3) First heterodimerization domain (HD1);

[0521] The first dsKO domain includes (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds;

[0522] b) The first Fab portion, comprising:

[0523] (1) A second VL domain (VLb) that pairs with a second VH domain (VHb) to form a second functional antigen-binding site that binds to target antigen B;

[0524] (2) The first CH1 domain is paired with the first CL domain;

[0525] and

[0526] (3) Second heterodimerization domain (HD2).

[0527] In some embodiments, the first and second heterodimerizing domains of the binding protein comprise first and second Fc domains. In some embodiments, the Fc domain comprises a universal structure: a hinge-CH2 domain-CH3 domain.

[0528] (a) Fc heterodimerization domain

[0529] In some embodiments, the multispecific binding protein of this disclosure may further comprise an Fc heterodimerization C1 or C2 domain. In some embodiments, the Fc heterodimerization domain is selected from heterodimerized Fc or fragments thereof, particularly the kilometre humeral (KIH) variant of the Fc moiety and its effector-modified variants; or heterodimerized Fc or fragments thereof, particularly the EV-RWT variant of Fc and its effector-modified variants. In some embodiments, Fc contains one or more amino acid mutations. One possibility is, for example, removing the selective recognition site of the first affinity reagent and introducing the selective recognition site of the second affinity reagent by mutations selected from H435R and Y436F. Alternatively, only the selective recognition site of the second affinity reagent may be introduced.

[0530] (b) Homodimerization domain

[0531] In some embodiments, the multispecific binding protein of this disclosure may also include a homodimerization domain. In some embodiments, the homodimerization domain is selected from variants of the Fc region and its effector modifications; one or more CH2 domains (e.g., of IgG, IgE, or IgM); one or more CH3 domains (e.g., of IgG, IgA, or IgD); and one or more CH4 domains (e.g., of IgE or IgM).

[0532] III. Multimer-binding proteins containing pseudo-Fab

[0533] (a) Symmetrical tetravalent construct

[0534] In another embodiment, the binding peptide comprises a binding protein containing pseudofabrication, said pseudofabrication binding protein comprising another polypeptide chain that associates with the pseudofabrication polypeptide chain to form an additional binding domain. These pseudofabrication binding peptides are further fused to an Fc heterodimerization domain to form half of a conventional Y-shaped antibody.

[0535] In some implementations, the antigen-binding protein comprises six polypeptide chains forming four antigen-binding sites, wherein

[0536] (a) The first and second polypeptides comprise a structure represented by the following formula:

[0537] VLa-L1-VLX[I] and [II]

[0538] (b) The third and fourth polypeptides comprise structures represented by the following formula:

[0539] VLb-CL[III] and [IV]

[0540] (c) The fifth polypeptide comprises a structure represented by the following formula:

[0541] VHa-L2-VHX-L3-VHb-CH1-FC1[V]

[0542] (d) The sixth polypeptide comprises a structure represented by the following formula:

[0543] VHa-L2-VHX-L3-VHb-CH1-FC2[VI]

[0544] in:

[0545] VLa is the variable domain of the first immunoglobulin light chain;

[0546] VLb is the variable domain of the second immunoglobulin light chain;

[0547] VHa is the variable domain of the first immunoglobulin heavy chain;

[0548] VHb is the variable domain of the second immunoglobulin heavy chain;

[0549] VLX is a stable knockout of the variable structure domain of the light chain;

[0550] VHX is a stable knockout heavy chain variable structure domain;

[0551] CL is the constant domain of the immunoglobulin light chain;

[0552] CH1 is the constant structural domain of the immunoglobulin heavy chain;

[0553] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0554] L1, L2, and L3 are amino acid linkers.

[0555] in

[0556] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0557] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0558] (3) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0559] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0560] These molecules can also be referred to as “tandem-(Fv-pseudoFab x Fv-Fab)” (see Figure 10B).

[0561] In some implementations, the antigen-binding protein comprises six polypeptide chains forming four antigen-binding sites, wherein

[0562] (a) The first and second polypeptides comprise a structure represented by the following formula:

[0563] VLa-L1-VLX[I] and [II]

[0564] (b) The third and fourth polypeptides comprise structures represented by the following formula:

[0565] VLb-CL[III] and [IV]

[0566] (c) The fifth polypeptide comprises a structure represented by the following formula:

[0567] VHb-CH1-L3-VHa-L2-VHX-FC1[V]

[0568] (d) The sixth polypeptide comprises a structure represented by the following formula:

[0569] VHb-CH1-L3-VHa-L2-VHX-FC2[VI]

[0570] in:

[0571] VLa is the variable domain of the first immunoglobulin light chain;

[0572] VLb is the variable domain of the second immunoglobulin light chain;

[0573] VHa is the variable domain of the first immunoglobulin heavy chain;

[0574] VHb is the variable domain of the second immunoglobulin heavy chain;

[0575] VLX is a stable knockout of the variable structure domain of the light chain;

[0576] VHX is a stable knockout heavy chain variable structure domain;

[0577] CL is the constant domain of the immunoglobulin light chain;

[0578] CH1 is the constant structural domain of the immunoglobulin heavy chain;

[0579] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0580] L1, L2, and L3 are amino acid linkers.

[0581] in

[0582] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0583] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0584] (3) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0585] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0586] These molecules can also be referred to as “tandem-(Fv-Fab x Fv-pseudoFab-IgG)” (see Figure 10A).

[0587] (b) Asymmetric four-specific molecules

[0588] Dimerization of binding peptides (only one of which contains pseudofabrication) yields asymmetric constructs with additional specificity. These binding peptides containing pseudofabrication are further fused into the Fc heterodimerization domain to form half of a conventional full-length Y-shaped IgG antibody.

[0589] In some implementations, the multispecific binding protein comprises four polypeptide chains forming at least two antigen-binding sites, wherein

[0590] (a) The first polypeptide comprises a structure represented by the following formula:

[0591] VLa-L1-VLX[I]

[0592] (b) The second polypeptide comprises a structure represented by the following formula:

[0593] VHa-L2-VHX-FC1[II]

[0594] (c) The third polypeptide contains a structure represented by the following formula:

[0595] VLb-CL[III]

[0596] (d) The fourth polypeptide contains a structure represented by the following formula:

[0597] VHb-CH1-FC2[IV]

[0598] in:

[0599] VLa is the variable domain of the first immunoglobulin light chain;

[0600] VLb is the variable domain of the second immunoglobulin light chain;

[0601] VHa is the variable domain of the first immunoglobulin heavy chain;

[0602] VHb is the variable domain of the second immunoglobulin heavy chain;

[0603] VLX is a stable knockout of the variable structure domain of the light chain;

[0604] VHX is a stable knockout heavy chain variable structure domain;

[0605] CL is the constant domain of the immunoglobulin light chain;

[0606] CH1 is the constant domain of the heavy chain of immunoglobulin CH1;

[0607] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0608] L1 and L2 are amino acid linkers that can be the same or different independently.

[0609] in

[0610] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0611] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0612] (3) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0613] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0614] These molecules can also be referred to as dimer bispecific IgG molecules (Fv-pseudoFab)x(Fv-Fab)-Fc (see Figures 2 and 7).

[0615] In another embodiment, the binding peptide comprises a binding protein containing pseudofabrication, said pseudofabrication binding protein comprising another polypeptide chain that associates with the pseudofabrication polypeptide chain to form an additional binding domain. These pseudofabrication binding peptides are further fused to an Fc heterodimerization domain to form half of a conventional Y-shaped antibody. Dimerization of such molecules produces constructs with additional specificity.

[0616] In some implementations, the antigen-binding protein comprises six polypeptide chains forming four antigen-binding sites, wherein

[0617] (a) The first and second polypeptides comprise a structure represented by the following formula:

[0618] VLa-L1-VLX[I] and [II]

[0619] (b) The third and fourth polypeptides comprise structures represented by the following formula:

[0620] VLb-CL[III] and [IV]

[0621] (c) The fifth polypeptide comprises a structure represented by the following formula:

[0622] VHa-L2-VHX-L3-VHa-L4-VHX-FC1[V]

[0623] (d) The sixth polypeptide comprises a structure represented by the following formula:

[0624] VHb-CH1-L5-VHb-CH1-FC2[VI]

[0625] in:

[0626] VLa is the variable domain of the first immunoglobulin light chain;

[0627] VLb is the variable domain of the second immunoglobulin light chain;

[0628] VHa is the variable domain of the first immunoglobulin heavy chain;

[0629] VHb is the variable domain of the second immunoglobulin heavy chain;

[0630] VLX is a stable knockout of the variable structure domain of the light chain;

[0631] VHX is a stable knockout heavy chain variable structure domain;

[0632] CL is the constant domain of the immunoglobulin light chain;

[0633] CH1 is the constant structural domain of the immunoglobulin heavy chain;

[0634] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0635] L1, L2, L3, L4, and L5 are amino acid linkers.

[0636] in

[0637] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0638] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0639] (3) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO) domain.

[0640] The dsKO domain contains (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0641] The C-terminal heterodimerization domain can be a conventional Fc-mortar heterodimerization domain, a conventional Fc-RF heterodimerization domain, or a combination thereof. These molecules can also be referred to as “Fv-pseudoFab([HC]-(Fv-pseudoFab))x((Fv-Fab)[HC]-(Fv-Fab)))-Fc” (see Figure 11).

[0642] In some implementations, the antigen-binding protein comprises four polypeptide chains forming three antigen-binding sites, wherein:

[0643] (a) The first polypeptide comprises a structure represented by the following formula:

[0644] VLa-L1-VLX[I]

[0645] (b) The second polypeptide comprises a structure represented by the following formula:

[0646] VHa-L2-VHX-FC1[II]

[0647] (c) The third polypeptide comprises a structure represented by the following formula:

[0648] VLb-L3-VLc-L4-CL[III]

[0649] (d) The fourth polypeptide comprises a structure represented by the following formula:

[0650] VHc-L5-VHb-L6-CH1-FC2[IV]

[0651] in:

[0652] VLa is the variable domain of the first immunoglobulin light chain;

[0653] VLb is the variable domain of the second immunoglobulin light chain;

[0654] VLc is the variable domain of the third immunoglobulin light chain;

[0655] VHa is the variable domain of the first immunoglobulin heavy chain;

[0656] VHb is the variable domain of the second immunoglobulin heavy chain;

[0657] VHc is the variable domain of the third immunoglobulin heavy chain;

[0658] CL is the constant domain of the immunoglobulin light chain;

[0659] CH1 is the constant domain of the heavy chain of immunoglobulin CH1;

[0660] VLX is a stable knockout of the variable structure domain of the light chain;

[0661] VHX is a stable knockout heavy chain variable structure domain;

[0662] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0663] L1, L2, L3, L4, L5, and L6 are amino acid linkers.

[0664] in

[0665] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0666] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0667] (3) The third VL domain (VLc) pairs with the third VH domain (VHc) to form a third functional antigen binding site for binding target antigen C;

[0668] (4) The polypeptide of Formula III and the polypeptide of Formula IV form a cross-linked light chain-heavy chain pair (CODV).

[0669] (5) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO2) domain.

[0670] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen; and (ii) one or more engineered interstrand disulfide bonds.

[0671] In some implementations, the antigen-binding protein comprises four polypeptide chains forming three antigen-binding sites, wherein:

[0672] (a) The first polypeptide comprises a structure represented by the following formula:

[0673] VLa-L1-VLX[I]

[0674] (b) The second polypeptide comprises a structure represented by the following formula:

[0675] VHa-L2-VHX-FC1[II]

[0676] (c) The third polypeptide comprises a structure represented by the following formula:

[0677] VLb-L3-VLc-L4-CL[III]

[0678] (d) The fourth polypeptide comprises a structure represented by the following formula:

[0679] VHc-L5-VHb-L6-CH1-FC2[IV]

[0680] in:

[0681] VLa is the variable domain of the first immunoglobulin light chain;

[0682] VLb is the variable domain of the second immunoglobulin light chain;

[0683] VLc is the variable domain of the third immunoglobulin light chain;

[0684] VHa is the variable domain of the first immunoglobulin heavy chain;

[0685] VHb is the variable domain of the second immunoglobulin heavy chain;

[0686] VHc is the variable domain of the third immunoglobulin heavy chain;

[0687] CL is the constant domain of the immunoglobulin light chain;

[0688] CH1 is the constant domain of the heavy chain of immunoglobulin CH1;

[0689] VLX is a stable knockout of the variable structure domain of the light chain;

[0690] VHX is a stable knockout heavy chain variable structure domain;

[0691] FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and

[0692] L1, L2, L3, L4, L5, and L6 are amino acid linkers.

[0693] in

[0694] (1) The first VL domain (VLa) pairs with the first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A;

[0695] (2) The second VL domain (VLb) pairs with the second VH domain (VHb) to form a second functional antigen binding site that binds to target antigen B;

[0696] (3) The third VL domain (VLc) pairs with the third VH domain (VHc) to form a third functional antigen binding site for binding target antigen C;

[0697] (4) The polypeptide of Formula III and the polypeptide of Formula IV form a cross-linked light chain-heavy chain pair (CODV).

[0698] (5) The stable knockout VL domain (VLX) is paired with the stable knockout VH domain (VHX) to form a disulfide bond-stable knockout (dsKO2) domain.

[0699] The dsKO domain comprises (i) one or more inactivating mutations that eliminate its binding to the target antigen of the reference Fab molecule; and (ii) one or more engineered interchain disulfide bonds.

[0700] These molecules can also be referred to as “(CODV-Fab)x(pseudoFab)-Fc” (see Figure 13).

[0701] In specific embodiments of the first and second aspects of the invention, the binding protein comprises a light chain and heavy chain pair selected from the following: SEQ ID NO: 6 and 7; SEQ ID NO: 8 and 9; SEQ ID NO: 10 and 11; SEQ ID NO: 12 and 13; SEQ ID NO: 14 and 15; SEQ ID NO: 16 and 17; SEQ ID NO: 18 and 19; SEQ ID NO: 20 and 21; SEQ ID NO: 22 and 23; SEQ ID NO: 24 and 25; SEQ ID NO: 26 and 27; SEQ ID NO: 28 and 29; SEQ ID NO: 30 and 31; SEQ ID NO: 32 and 33; SEQ ID NO: 34 and 35; SEQ ID NO: 36 and 37; SEQ ID NO: 38 and 39; SEQ ID NO: 40 and 41; SEQ ID NO: 42 and 43; SEQ ID NO: 44 and 45; SEQ ID NO: 46 and 47 and SEQ ID NO: 48 and 49; SEQ ID NO: 6 and 7; SEQ ID NO: 8 and 9; SEQ ID NO: 10 and 11; SEQ ID NO: 12 and 13; SEQ ID NO: 14 and 15; SEQ ID NO: 16 and 17; SEQ ID NO: 18 and 19; SEQ ID NO: 20 and 21; SEQ ID NO: 22 and 23; SEQ ID NO: 24 and 25; SEQ ID NO: 26 and 27; SEQ ID NO: 28 and 29; SEQ ID NO: 30 and 31; SEQ ID NO: 32 and 33; SEQ ID NO: 34 and 35; SEQ ID NO: 36 and SEQ ID NO:50 and 51; SEQ ID NO:52 and 53 and SEQ ID NO:54 and 55; SEQ ID NO:56 and 57 and SEQ ID NO:58 and 59; SEQ ID NO:60 and 61 and SEQ ID NO:62 and 63; SEQ ID NO:64 and 65 and SEQ ID NO:66 and 67; SEQ ID NO:68 and 69 and SEQ ID NO:70 and 71; SEQ ID NO:72 and 73; and SEQ ID NO:74 and 75.

[0702] In some implementations, the first and second CLs are independently selected from the constant region light chain κ (CLκ) and the constant region light chain λ (CLλ).

[0703] In some embodiments, HD1 and HD2 each comprise an Fc region and a variant modified with its effector; a heterodimerized Fc region, particularly a kilometre hummock (KIH) variant of the Fc region and a variant modified with its effector; one or more CH2 domains (e.g., of IgG, IgE, or IgM); one or more CH3 domains (e.g., of IgG, IgA, or IgD); or one or more CH4 domains (e.g., of IgE or IgM).

[0704] In some implementations, one Fc domain includes a first CH3 domain containing one or both of the S354C and T366W mutations, and another Fc domain includes a second CH3 domain containing one or both of the Y349C, T366S, L368A, and Y407V mutations.

[0705] In one embodiment, the Fc region of HD1 or HD2 contains one or more amino acid mutations (e.g., selected from H435R or Y436F) that result in the removal of a selective recognition site for a second affinity reagent, or one or more amino acid mutations that result in the introduction of a selective recognition site for a third affinity reagent.

[0706] Table 2: Sequences of the selected binding proteins

[0707]

[0708]

[0709]

[0710]

[0711]

[0712]

[0713]

[0714]

[0715]

[0716]

[0717]

[0718]

[0719]

[0720]

[0721]

[0722]

[0723]

[0724]

[0725]

[0726]

[0727] Table 3: Sequences of Bispecific T cell engager antibodies

[0728]

[0729]

[0730]

[0731]

[0732]

[0733]

[0734]

[0735]

[0736]

[0737]

[0738]

[0739]

[0740]

[0741]

[0742]

[0743]

[0744]

[0745]

[0746]

[0747]

[0748]

[0749]

[0750]

[0751]

[0752]

[0753] (e) Nucleic acid

[0754] In a fifth aspect, the present invention relates to isolated nucleic acid molecules comprising a nucleotide sequence encoding a pseudoFab, the pseudoFab containing one or both of the binding proteins and multispecific binding proteins of the first, second or third aspects of the present invention.

[0755] One aspect of the present invention relates to a polynucleotide encoding a binding protein of any of the first to third aspects of the present invention.

[0756] Polynucleotides encoding polypeptides that form binding proteins are constructed using standard recombinant DNA methods, these polynucleotides are incorporated into recombinant expression vectors, and such vectors are introduced into host cells. See, for example, Sambrook et al., 2001, *MOLECULAR CLONING: A LABORATORY MANUAL* (Cold Spring Harbor Laboratory Press, 3rd edition). Enzymatic reactions and purification techniques can be performed according to the manufacturer's instructions, as commonly practiced in the art or as described herein. Unless specifically defined, the terminology used in analytical chemistry, synthetic organic chemistry, and medical and medicinal chemistry as described herein, as well as the laboratory procedures and techniques for analytical chemistry, synthetic organic chemistry, and medical and medicinal chemistry described herein, are those well known and commonly used in the art. Similarly, conventional techniques can be used for chemical synthesis, chemical analysis, drug preparation, formulation, delivery, and patient treatment.

[0757] Other aspects of this disclosure relate to isolated nucleic acid molecules containing nucleotide sequences encoding any of the binding proteins described herein. In some embodiments of the fifth aspect of the invention, the isolated nucleic acid molecule contains at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identical sequences to the nucleic acid encoding any of the binding proteins described herein.

[0758] Certain aspects of this disclosure relate to suites of polynucleotides. In some embodiments, one or more polynucleotides are vectors (e.g., expression vectors). These suites can be used, in particular, to generate one or more binding proteins described herein, such as the heterodimerizing domains, divalent, trivalent, tetravalent, or multivalent binding proteins of this disclosure.

[0759] In some implementations, the isolated nucleic acid is operatively linked to a heterologous promoter to direct the transcription of a nucleic acid sequence encoding a binding protein. The promoter can refer to a nucleic acid control sequence that directs the transcription of a nucleic acid. The first nucleic acid sequence is operatively linked to the second nucleic acid sequence when the first nucleic acid sequence is positioned to have a functional relationship with the second nucleic acid sequence. For example, if the promoter affects the transcription or expression of a coding sequence, then the promoter is operatively linked to the coding sequence of the binding protein. Examples of promoters may include, but are not limited to, promoters derived from the genomes of viruses (such as polyomavirus, fowlpox virus, adenovirus (such as adenovirus 2), bovine papillomavirus, avian sarcoma virus, cytomegalovirus, retrovirus, hepatitis B virus, simian virus 40 (SV40), etc.), heterologous eukaryotic promoters (such as actin promoter, immunoglobulin promoter, heat shock promoter, etc.), CAG promoter (Niwa et al., Gene 108(2):193-9, 1991), phosphoglycerate kinase (PGK) promoter, tetracycline-inducible promoter (Masui et al., Nucleic Acids Res. 33:e43, 2005), lac system, trp system, tac system, trc system, major operon and promoter regions of bacteriophage λ, promoter of 3-phosphoglycerate kinase, promoter of yeast acid phosphatase, and promoter of yeast α-mating factor. The polynucleotide encoding the binding protein of this disclosure may be under the control of a constitutive promoter, an inducible promoter, or any other suitable promoter described herein, or other suitable promoters that will be readily recognized by those skilled in the art.

[0760] In some embodiments, isolated nucleic acids are incorporated into a vector. In some embodiments, the vector is an expression vector. An expression vector may include one or more regulatory sequences operatively linked to the polynucleotide to be expressed. As used herein, the term "regulatory sequence" includes promoters, enhancers, and other expression control elements (e.g., polyadenylation signals). Examples of suitable enhancers may include, but are not limited to, enhancer sequences from mammalian genes (e.g., globulins, elastases, albumins, alpha-fetoproteins, insulin, etc.) and enhancer sequences from eukaryotic viruses (e.g., the SV40 enhancer (bp 100-270) post-OMI, the cytomegalovirus early promoter enhancer, the polyoma enhancer post-OMI, the adenovirus enhancer, etc.). Examples of suitable vectors may include, for example, plasmids, granulocytes, episomes, transposons, and viral vectors (e.g., adenovirus, vaccinia virus, Sindbisvirus, measles, herpesvirus, lentivirus, retrovirus, adeno-associated virus vector, etc.). Expression vectors can be used to transfect host cells, such as bacterial cells, yeast cells, insect cells, and mammalian cells. Biologically functional viral and plasmid DNA vectors capable of expression and replication within a host are known in the art and can be used to transfect any cell of interest.

[0761] In a sixth aspect, the present invention relates to an expression vector comprising the nucleic acid molecule of the fifth aspect of the present invention.

[0762] Other aspects of this disclosure relate to vector systems comprising one or more vectors encoding first, second, third, and fourth polypeptide chains of any of the binding proteins described herein. In some embodiments of the sixth aspect of the invention, the vector system comprises a first vector encoding a first polypeptide chain of a binding protein, a second vector encoding a second polypeptide chain of a binding protein, a third vector encoding a third polypeptide chain of a binding protein, and a fourth vector encoding a fourth polypeptide chain of a binding protein. In some embodiments, the vector system comprises a first vector encoding first and second polypeptide chains of a binding protein and a second vector encoding third and fourth polypeptide chains of a binding protein. In some embodiments, the vector system comprises a first vector encoding first and third polypeptide chains of a binding protein and a second vector encoding second and fourth polypeptide chains of a binding protein. In some embodiments, the vector system comprises a first vector encoding first and fourth polypeptide chains of a binding protein and a second vector encoding second and third polypeptide chains of a binding protein. In some embodiments, the vector system comprises a first vector encoding first, second, third, and fourth polypeptide chains of a binding protein. The one or more vectors of the vector system can be any of the vectors described herein. In some embodiments, the one or more vectors are expression vectors.

[0763] (f) Isolated host cells

[0764] In a seventh aspect, the present invention relates to isolated host cells comprising the nucleic acid molecule of the fifth aspect of the invention or the expression vector of the sixth aspect of the invention.

[0765] Other aspects of this disclosure relate to isolated host cells comprising one or more isolated polynucleotides, polynucleotide suites, vectors, and / or vector systems described herein. In some embodiments of the seventh aspect of the invention, the host cell is a bacterial cell (e.g., *E. coli* cells). In some embodiments, the host cell is *E. coli* DH5α cells. In some embodiments, the host cell is a yeast cell (e.g., *Saccharomyces cerevisiae* cells). In some embodiments, the host cell is an insect cell. Examples of insect host cells may include, for example, *Drosophila* cells (e.g., S2 cells), *Trichoplusia ni* cells (e.g., High Five™ cells), and *Spodoptera frugiperda* cells (e.g., Sf21 or Sf9 cells). In some embodiments, the host cell is a mammalian cell. Examples of mammalian host cells may include, for example, human embryonic kidney cells (e.g., 293 cells or subclones used for growth in suspension cultures), Expi293TM cells, CHO cells, juvenile hamster kidney cells (e.g., BHK, ATCC CCL 10), mouse supporting cells (e.g., TM4 cells), monkey kidney cells (e.g., CV1 ATCC CCL 70), African green monkey kidney cells (e.g., VERO-76, ATCC CRL-1587), human cervical cancer cells (e.g., HELA, ATCC CCL 2), canine kidney cells (e.g., MDCK, ATCC CCL 34), Buffalo rat hepatocytes (e.g., BRL 3A, ATCC CRL 1442), human lung cells (e.g., W138, ATCC CCL75), human hepatocytes (e.g., Hep G2, HB 8065), and mouse mammary tumor cells (e.g., MMT). 060562, ATCCCCL51), TRI cells, MRC 5 cells, FS4 cells, human hepatocellular carcinoma lineages (e.g., Hep G2), myeloma cells (e.g., NS0 and Sp2 / 0 cells), etc.

[0766] IV. Preparation method

[0767] 1. Expression methods

[0768] Other aspects of this disclosure relate to methods for generating any of the binding proteins described herein. In some embodiments, the methods include:

[0769] a) Culture host cells containing isolated nucleic acids, vectors, and / or vector systems (e.g., any isolated nucleic acids, vectors, and / or vector systems described herein) under conditions that enable host cells to express the binding protein; and

[0770] b) Isolation of the binding protein from the host cell. Methods of culturing host cells under conditions that express the protein are well known to those skilled in the art. Methods of isolating the protein from the cultured host cell are well known to those skilled in the art, including, for example, affinity chromatography (e.g., two-step affinity chromatography, including protein A affinity chromatography followed by size exclusion chromatography).

[0771] 2. Purification methods

[0772] In a fourth aspect, the present invention relates to a method for purifying the multispecific binding protein of the third aspect of the present invention, the method comprising the following steps:

[0773] (i) Applying a solution containing a multispecific binding protein to a first, second, or third affinity reagent, wherein the first, second, or third affinity reagent specifically binds to a recognition site of the multispecific binding protein against the first, second, or third affinity reagent; and

[0774] (ii) Recover multispecific binding proteins that do not bind to the first, second, or third affinity reagent, or multispecific binding proteins that bind to the first or third affinity reagent.

[0775] Each of the first, second, and third affinity reagents binds to a different recognition site of the multispecific binding protein.

[0776] In some embodiments, the method includes the additional steps of: applying a solution containing the multispecific binding protein recovered in step (ii) to a first, second, or third affinity reagent, wherein the affinity reagent is different from the affinity reagent used in step (i); and recovering the multispecific binding protein that does not bind to the first, second, or third affinity reagent or that binds to the first or third affinity reagent.

[0777] In some implementations, the first affinity reagent binds to Cκ; the second affinity reagent is protein A; and / or the third affinity reagent is protein G.

[0778] In some embodiments, the binding protein of this disclosure is purified by protein A affinity chromatography, κ light chain affinity chromatography (e.g., using Kappa Select resin according to the manufacturer's instructions; GE Healthcare), and optionally λ light chain affinity chromatography (e.g., using Lambda Fab Select resin according to the manufacturer's instructions; GE Healthcare). In some embodiments, the binding protein comprises two Fc regions, each Fc region containing a CH3 domain, and only one CH3 domain contains an amino acid substitution at positions 435 and 436 of human IgG1 or IgG4 according to the EU index, wherein said amino acid substitution is H435R and Y436F. In some embodiments, the binding protein of this disclosure is purified sequentially by protein A affinity chromatography, followed by κ light chain affinity chromatography (e.g., using Kappa Select resin according to the manufacturer's instructions; GE Healthcare), and optionally λ light chain affinity chromatography (e.g., using Lambda Fab Select resin according to the manufacturer's instructions; GE Healthcare). For example, in some embodiments, the binding protein is contacted with protein A, eluted from protein A under conditions suitable for separating the binding protein from binding proteins containing 0 or 2 CH3 domains (which contain amino acid substitutions of H435R and Y436F), contacted with a κ light chain affinity medium (e.g., as used in KappaSelect resin; GE Healthcare), and eluted from the κ light chain affinity medium under conditions suitable for separating the binding protein from binding proteins containing only the λCL domain (e.g., according to the manufacturer's instructions).

[0779] Suitable elution conditions for protein A are known in the art, including but not limited to stepwise elution gradients from pH 4.5 to 2.8. In some embodiments, protein A or a variant of protein A suitable for protein purification is employed. In some embodiments, protein A is attached to the substrate or resin, for example, as part of the chromatographic medium. In some embodiments, after elution from the κ light chain affinity medium, the binding protein is contacted with a λ light chain affinity medium (e.g., as used in LambdaFabSelect resin; GE Healthcare) and eluted from the λ light chain affinity medium under conditions suitable for separating the binding protein from binding proteins containing only the κCL domain (e.g., according to the manufacturer's instructions). In some embodiments, the binding protein of this disclosure is detected using HIC chromatography. In some embodiments, the binding protein comprises: a first polypeptide chain containing a λCL domain; a CH3 domain of a second polypeptide chain containing amino acid substitutions at positions 354 and 366 of human IgG1 or IgG4 according to the EU index, wherein the amino acid substitutions are S354C and T366W; a CH3 domain of a third polypeptide chain containing amino acid substitutions at positions 349, 366, 368, 407, 435, and 436 of human IgG1 or IgG4 according to the EU index, wherein the amino acid substitutions are Y349C, T366S, L368A, Y407V, H435R, and Y436F; and a fourth polypeptide chain containing a κCL domain. In some embodiments, the binding protein is produced by a host cell. In some embodiments, the binding protein is purified from cell culture medium or host cell extract. In some embodiments, the binding protein is secreted by a host cell or produced and extracted from a host cell (e.g., before contact with protein A). In some embodiments, the binding protein is in cell culture medium or host cell extract when contacted with protein A. In some embodiments, the binding protein is purified from other binding proteins, peptides, and / or other cellular components.

[0780] In some embodiments, a stabilizing knockout domain is used to facilitate the preferential synthesis or purification of a desired multispecific binding protein, wherein the stabilizing knockout domain comprises (1) one or more inactivating mutations that eliminate binding to the target antigen; and (2) one or more engineered interchain disulfide bonds that impart enhanced thermal stability (Tm) relative to a reference Fab molecule, wherein the reference Fab molecule is identical to the pseudo Fab molecule except that, in the pseudo Fab molecule, the CH1 and CL domains of the reference Fab molecule are replaced by VHX and VLX domains.

[0781] V. Formulations / Pharmaceutical Compositions

[0782] In an eighth aspect, the present invention relates to pharmaceutical compositions comprising a pharmaceutically acceptable carrier and a therapeutically effective amount of any of the proteins of the first to third aspects of the present invention.

[0783] Therapeutic or pharmaceutical compositions comprising binding proteins are within the scope of this disclosure. Some embodiments of the eighth aspect of the invention comprise pharmaceutical compositions comprising a therapeutically effective amount of any of the binding proteins or binding protein-drug conjugates as described herein, mixed with a pharmaceutically or physiologically acceptable formulation suitably chosen for the manner of administration.

[0784] Acceptable formulations are generally non-toxic to recipients at the dosage and concentration used.

[0785] In some embodiments, the pharmaceutical composition may contain formulation materials for modifying, maintaining, or preserving, for example, the composition's pH, molar osmotic pressure concentration, viscosity, clarity, color, isotonicity, odor, sterility, stability, dissolution or release rate, adsorption, or permeation. Suitable formulation materials include, but are not limited to, amino acids (such as glycine, glutamine, asparagine, arginine, or lysine), antimicrobial agents, antioxidants (such as ascorbic acid, sodium sulfite, or sodium bisulfite), buffers (such as borates, bicarbonates, Tris-HCl, citrates, phosphates, or other organic acids), swelling agents (such as mannitol or glycine), chelating agents (such as ethylenediaminetetraacetic acid (EDTA)), complexing agents (such as caffeine, polyvinylpyrrolidone, β-cyclodextrin, or hydroxypropyl-β-cyclodextrin), fillers, monosaccharides, disaccharides, and other carbohydrates (such as glucose, mannose, or dextrin), proteins (such as serum albumin, gelatin, or immunoglobulins), colorants, flavoring agents, and diluents, emulsifiers, hydrophilic polymers (such as polyvinylpyrrolidone), low molecular weight peptides, salt-forming antiions (such as sodium), and preservatives (such as benzalkonium chloride, benzoic acid, and water). Salicylic acid, thimerosal, phenethyl alcohol, methylparaben, propylparaben, chlorhexidine, sorbic acid, or hydrogen peroxide; solvents (such as glycerol, propylene glycol, or polyethylene glycol); sugar alcohols (such as mannitol or sorbitol); suspending agents; surfactants or wetting agents (such as pluronics; PEG; dehydrated sorbitol esters; polysorbates (such as polysorbate 20 or polysorbate 80); triton; tromethamine; lecithin; cholesterol or tyloxapal); stability enhancers (such as sucrose or sorbitol); tension enhancers (such as alkali metal halides (e.g., sodium chloride or potassium chloride) or mannitol sorbitol); delivery media, diluents, excipients, and / or adjuvants (see, for example, Remington's Pharmaceutical, incorporated herein by reference for any purpose). Sciences (18th edition, edited by ARGennaro, Mack Publishing Company 1990 and subsequent editions).

[0786] In some implementations, the optimal pharmaceutical composition will be determined by a technician based on factors such as the intended route of administration, delivery method, and required dosage. Such compositions may affect the physical state, stability, in vivo release rate, and in vivo clearance rate of the binding protein.

[0787] In some embodiments, the primary medium or carrier in the pharmaceutical composition may be aqueous or non-aqueous. For example, a medium or carrier suitable for injection may be water, physiological saline solution, or artificial cerebrospinal fluid, possibly supplemented with other materials commonly used in compositions for parenteral administration. Neutral buffered saline or saline mixed with serum albumin are other exemplary mediums. Other exemplary pharmaceutical compositions comprise a Tris buffer of about pH 7.0-8.5 or an acetate buffer of about pH 4.0-5.5, and may also include sorbitol or suitable substitutes. In one embodiment of this disclosure, the binding protein composition can be prepared for storage by mixing the selected composition having the desired purity with an optional formulation in the form of a lyophilized cake or aqueous solution. Furthermore, the binding protein can be formulated as a lyophilized product using a suitable excipient such as sucrose.

[0788] In some embodiments, the pharmaceutical compositions of this disclosure may be selected for parenteral or subcutaneous delivery. Alternatively, the compositions may be selected for inhalation or for delivery via the digestive tract, such as oral administration. The preparation of such pharmaceutically acceptable compositions is within the scope of the art.

[0789] In some embodiments, the components are formulated to be present at concentrations acceptable to the treatment site. For example, a buffer solution is used to maintain the composition at a physiological pH or slightly lower, typically in the pH range of about 5 to about 8.

[0790] When considering parenteral administration, the therapeutic composition used can be in a pyrogen-free, parenterally acceptable aqueous solution containing the desired binding protein in a pharmaceutically acceptable medium. A particularly suitable medium for parenteral injection is sterile distilled water in which the binding protein is formulated into a properly preserved sterile isotonic solution. Another preparation may involve formulating the desired molecule with an agent that provides controlled or sustained release of the product (such as injectable microspheres, biodegradable particles, polymeric compounds such as polylactic acid or polyglycolic acid), beads, or liposomes), which can then be delivered via accumulation injection. Hyaluronic acid can also be used, and this can have the effect of promoting a prolonged duration of circulation. Other suitable means for introducing the desired molecule include implantable drug delivery devices.

[0791] In one embodiment, the pharmaceutical composition can be formulated for inhalation. For example, the binding protein can be formulated as a dry powder for inhalation. The binding protein inhalation solution can also be formulated using a propellant for aerosol delivery. In yet another embodiment, the solution can be nebulized.

[0792] It is also considered that certain formulations can be administered orally. In one embodiment of this disclosure, multispecific binding proteins administered in this manner may be formulated with or without carriers typically used in mixtures of solid dosage forms (such as tablets and capsules). For example, capsules may be designed to release the active portion of the formulation in the gastrointestinal tract at the point of maximizing bioavailability and minimizing pre-systemic degradation. Additional agents may be included to facilitate the absorption of the binding proteins. Diluents, flavoring agents, low-melting-point waxes, vegetable oils, lubricants, suspending agents, tablet disintegrants, and binders may also be employed.

[0793] Another pharmaceutical composition may involve an effective amount of a multispecific binding protein in a mixture of non-toxic excipients suitable for manufacturing tablets. Solutions can be prepared in unit dose form by dissolving the tablets in sterile water or another suitable medium. Suitable excipients include, but are not limited to, inert diluents such as calcium carbonate, sodium carbonate, or sodium bicarbonate, lactose, or calcium phosphate; or binders such as starch, gelatin, or gum arabic; or lubricants such as magnesium stearate, stearic acid, or talc.

[0794] Further pharmaceutical compositions disclosed herein will be apparent to those skilled in the art, including formulations comprising binding proteins in sustained-release or controlled-release formulations. Techniques for formulating a variety of other sustained-release or controlled-release methods, such as liposome carriers, biodegradable microparticles or porous beads, and accumulation injections, are also known to those skilled in the art. Further examples of sustained-release formulations include semi-permeable polymer matrices in the form of molded articles (e.g., films or microcapsules). Sustained-release matrices may include polyesters, hydrogels, polylactide, copolymers of L-glutamic acid and γ-ethyl-L-glutamic acid esters, poly(2-hydroxyethyl-methacrylate), ethylene vinyl acetate, or poly(-)-3-hydroxybutyric acid. Sustained-release compositions may also include liposomes, which can be prepared by any of several methods known in the art.

[0795] In some embodiments, pharmaceutical compositions intended for in vivo administration must generally be sterile. This can be achieved by filtration through a sterile filter membrane. In the case of lyophilization of the composition, sterilization can be performed using this method before or after lyophilization and reconstitution. Compositions intended for parenteral administration can be stored in lyophilized form or in solution. Additionally, parenteral compositions are typically placed in containers with sterile access ports, such as intravenous infusion bags or vials with stoppers that can be punctured by a hypodermic needle.

[0796] Once a pharmaceutical composition has been formulated, it can be stored in sterile vials as a solution, suspension, gel, emulsion, solid, or as a dehydrated or lyophilized powder. Such formulations can be stored in ready-to-use form or in a form that needs to be reconstituted before administration (e.g., lyophilized).

[0797] This disclosure also covers kits for producing a single dose unit. The kit may each contain a first container with a dried multispecific binding protein and a second container with an aqueous formulation. The scope of this disclosure also includes kits containing single-compartment and multi-compartment pre-filled syringes (e.g., liquid syringes and lyosyringes).

[0798] The effective amount of a therapeutically applied protein-binding drug composition will depend, for example, on the treatment context and objectives. Those skilled in the art will recognize that the appropriate therapeutic dose level will therefore vary in part depending on: the molecule delivered, the indication for which the protein-binding drug is used, the route of administration, and the patient's size (weight, body surface or organ size) and condition (age and general health status). Therefore, clinicians can titrate the dose and modify the route of administration to achieve optimal therapeutic effects.

[0799] The frequency of administration will depend on the pharmacokinetic parameters of the binding protein in the formulation used. Typically, clinicians will administer the composition until a dose is reached to achieve the desired effect. Therefore, the composition can be administered as a single dose, as two or more doses over time (which may or may not contain equal amounts of the desired molecule), or as a continuous infusion via an implanted device or catheter. Further refinement of the appropriate dose is performed by those skilled in the art in a conventional manner and within the scope of their usual duties. An appropriate dose can be determined by using appropriate dose-response data.

[0800] The drug composition can be administered via routes consistent with known methods, such as oral administration; intravenous, intraperitoneal, intracerebral (brain parenchyma), intraventricular, intramuscular, intraocular, intraarterial, portal vein, or intralesional injection; via a continuous release system; or via an implantable device. Where necessary, the composition can be administered by bolus injection, continuous infusion, or via an implantable device.

[0801] In some embodiments, the composition can also be administered via an implantation site using a membrane, sponge, or other suitable material to which the desired molecules have been adsorbed or encapsulated. When using an implantable device, the device can be implanted into any suitable tissue or organ, and the delivery of the desired molecules can be performed via diffusion, timed release bolus, or continuous administration.

[0802] VI. Treatment / Usage

[0803] In a ninth aspect, the present invention relates to a method for treating a disorder in which antigenic activity is harmful, said method comprising administering to a subject in need an effective amount of a binding protein of any of the first to third aspects of the invention. In some embodiments, a multispecific binding protein is provided for use as a drug.

[0804] The binding protein can be used in any known assay method, such as competitive binding assays, direct and indirect sandwich assays, and immunoprecipitation assays, for the detection and quantification of one or more target antigens. The binding protein will bind to the one or more target antigens with an affinity suitable for the assay method used.

[0805] For diagnostic applications, in some implementations, the binding protein can be labeled with a detectable portion. The detectable portion can be any part capable of generating a detectable signal directly or indirectly. For example, the detectable portion can be a radioactive isotope, such as... 3 H, 14 C 32 P, 35 S, 125 I, 99 Tc, 111 In or 67 Ga; fluorescent or chemiluminescent compounds, such as fluorescein isothiocyanate, rhodamine, or luciferin; or enzymes, such as alkaline phosphatase, β-galactosidase, or horseradish peroxidase.

[0806] Binding proteins can also be used for in vivo imaging. Animals can be given a detectably labeled binding protein, for example, into the bloodstream, and the presence and location of the labeled antibody in the host can be determined. Any partially labeled binding protein detectable in animals by MRI, radiology, or other detection methods known in the art can be used.

[0807] For clinical or research applications, in some implementations, the binding protein may be conjugated to a cytotoxic agent. Various antibodies conjugated to cytotoxic agents (i.e., antibody-drug conjugates) have been used to target specific tumor cells with cytotoxic payloads. Cytotoxic agents and connectors for conjugating said agents to antibodies are known in the art; see, for example, Parslow, AC et al. (2016) Biomedicines 4:14; and Kalim, M. et al. (2017) Drug Des. Devel. Ther. 11:2265-2276.

[0808] This disclosure also relates to kits comprising binding proteins and other reagents that can be used to detect levels of target antigens in biological samples. Such reagents may include detectable markers, blocked sera, positive and negative control samples, and detection reagents. In some embodiments, the kit comprises a composition containing any binding protein, polynucleotide, carrier, carrier system, and / or host cell described herein. In some embodiments, the kit comprises a container and a label or packaging instructions on or attached to said container. Suitable containers include, for example, bottles, vials, syringes, IV solution bags, etc. Containers can be made of a variety of materials such as glass or plastic. The container contains a composition (either alone or in combination with another composition) that is effective in treating, preventing, and / or diagnosing a condition, and the container may have a sterile inlet (e.g., the container may be an intravenous infusion bag or a vial with a stopper that can be punctured by a hypodermic needle). In some embodiments, the label or packaging instructions indicate that the composition is used for the prevention, diagnosis, and / or treatment of selected conditions. Alternatively or additionally, the product or kit may also include a second (or third) container containing pharmaceutically acceptable buffers such as bacteriostatic water for injection (BWFI), phosphate-buffered saline, Ringer's solution, and glucose solution. The product or kit may also include other materials required from a commercial and user perspective, including additional buffers, diluents, filters, needles, and syringes.

[0809] In some embodiments, the binding protein of this disclosure is administered to a patient in need for the treatment or prevention of cancer. In some embodiments, this disclosure relates to methods for the prevention and / or treatment of proliferative diseases or disorders (e.g., cancer). In some embodiments, the methods include administering to a patient a therapeutically effective amount of at least one binding protein described herein or a pharmaceutical composition associated therewith. In some embodiments, the patient is a human being.

[0810] In some embodiments, the at least one binding protein is administered in combination with one or more anticancer therapies (e.g., any anticancer therapy known in the art, such as chemotherapy agents or therapies). In some embodiments, the at least one binding protein is administered prior to the one or more anticancer therapies. In some embodiments, the at least one binding protein is administered concurrently with the one or more anticancer therapies. In some embodiments, the at least one binding protein is administered after the one or more antiretroviral therapies.

[0811] Example

[0812] Before describing the invention in detail below, it should be understood that the invention is not limited to the specific methods, schemes, and reagents described herein, as these can be modified. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the invention, which will be limited only by the appended claims. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. In particular embodiments, the terminology used herein is defined as described in “A multilingual glossary of biotechnological terms: (IUPAC Recommendations),” edited by Leuenberger, H.G., Nagel, B., and Kolb, H. (1995), Helvetica Chimica Acta, CH-4010 Basel, Switzerland. Unless otherwise defined herein, scientific and technical terms used herein have the meaning commonly understood by one of ordinary skill in the art. In the event of any potential ambiguity, the definitions provided herein take precedence over any dictionary or non-inherent definition. Unless the context otherwise requires, singular terms shall include plural terms, and plural terms shall include singular terms. Unless otherwise stated, the use of “or” means “and / or”. The use of the term “including” and other forms such as “includes” and “included” is non-limiting. As used herein, unless otherwise stated, the singular forms “a,” “an,” and “the” include multiple references. Thus, for example, a reference to “a protein” includes multiple protein molecules.

[0813] Furthermore, unless otherwise indicated, the experiments described herein utilize routine molecular and cell biological and immunological techniques within the scope of the art. Such techniques are well known to skilled practitioners and are well explained in the literature. See, for example, Ausubel et al. (eds.), Current Protocols in Molecular Biology, John Wiley & Sons, Inc., NY, NY (1987–2008), including all supplements; Molecular Cloning: A Laboratory Manual (4th edition), MR Green and J. Sambrook; and Harlow et al., Antibodies: A Laboratory Manual, Chapter 14, Cold Spring Harbor Laboratory, Cold Spring Harbor (2013, 2nd edition).

[0814] Numerous references are cited throughout the text of this specification. Each reference cited herein (including all patents, patent applications, scientific publications, manufacturer's specifications, instructions, etc.) is incorporated herein by reference in its entirety. Nothing herein should be construed as an admission that the invention is not entitled to any work prior to such disclosure.

[0815] The elements of the invention will be described below. These elements are listed using specific embodiments; however, it should be understood that these specific embodiments can be combined in any manner and in any number to produce other embodiments. The differentiated descriptions of embodiments and preferred / specific embodiments should not be construed as limiting the invention to only the explicitly described embodiments. This specification should be understood to support and cover embodiments combining the explicitly described embodiments with any number of the disclosed and / or preferred elements. Furthermore, unless the context otherwise indicates, any permutation and combination of all the elements disclosed in this application should be considered.

[0816] Generally, the terminology used herein in conjunction with cell and tissue culture, molecular biology, immunology, microbiology, genetics, and protein and nucleic acid chemistry and hybridization is that which is well known and commonly used in the art. Unless otherwise indicated, the methods and techniques provided herein are generally performed according to conventional methods well known in the art and described in the various general and more specific references cited and discussed throughout this specification. Enzymatic reactions and purification techniques are performed according to the manufacturer's instructions, as commonly practiced in the art or as described herein. The terminology used herein in analytical chemistry, synthetic organic chemistry, and medical and medicinal chemistry, as well as the laboratory procedures and techniques described herein in analytical chemistry, synthetic organic chemistry, and medical and medicinal chemistry, are those well known and commonly used in the art. Conventional techniques are used in chemical synthesis, chemical analysis, drug preparation, formulation and delivery, and patient treatment.

[0817] Example 1: Identifying a suitable alternative stent

[0818] The binding proteins disclosed in this paper include knockout VH / VL domain substitutions for a CH1 / CL pair in various antibody forms (see Figures 1, 2, 7, 10-13). To identify suitable alternative scaffolds to replace the CH1 / CL dimer, the introduction of disulfide bridges into certain VH / VL pairs was investigated. Furthermore, the expressibility and fusion with another VH / VL pair were investigated.

[0819] To address the issue of proper light chain pairing, trastuzumab was identified as a suitable alternative scaffold to the CH1 / CL dimer.

[0820] Example 2: Using trastuzumab variable domains to replace CH1 / CL

[0821] The cysteine ​​residues at positions VH44 and VL100 were determined to be compatible to form a disulfide bond between VH and VL of trastuzumab (referred to herein as ds-trastuzumab), thereby stabilizing the heterodimer. To determine its thermal stability, the ds-trastuzumab domains were incubated at 1 mg / mL in D-PBS buffer (GIBCO) at 40 °C for 14 days. Control samples at the same concentrations were maintained at -80 °C and 4 °C. After stress testing, the total aggregate content of the samples was analyzed by analytical size exclusion chromatography (SEC). Analytical SEC was performed at 25 °C using a BioSECcurity instrument (PSS Polymer) with a TSKgel SuperSW3000 column (4.6 mm x 300 mm) and a TSKgel SuperSW HPLC guard column (Tosoh Bioscience). The analysis was performed at a flow rate of 0.25 mL / min using 250 mM NaCl and 100 mM sodium phosphate (pH 6.7), with detections at 280 nm and 260 nm. Static light scattering was detected at 436 nm. 5 μL of protein sample (1 mg / mL) was applied to the column. Data were evaluated using WinGPC software version 8.1 (PSS Polymer). For molecular weight estimation, the SEC column was calibrated using protein standards ranging from 6.5 to 670 kDa.

[0822] The ds-trastuzumab scaffold (with VH44 / VL100 cysteine ​​modification) was found to increase thermal stability by 4 °C (see Figures 3A-C).

[0823] Example 3: Studying the VH / VL linkage in a pseudo-IgG design

[0824] A pseudo-IgG1 construct was generated, containing a disulfide-stabilized trastuzumab framework as a replacement scaffold for CH1 / CL. The ds-trastuzumab scaffold was fused to the VH / VL pairs of interest using a G4S (SEQ ID NO:102) or (G4S)2 (SEQ ID NO:103) linker (see Figure 6).

[0825] Determine the melting point (T) m )

[0826] The melting point (T) was determined using differential scanning fluorometry (DSF). mSamples were diluted to a final concentration of 0.2 μg / μl in D-PBS buffer (Invitrogen) in white 96-well half-skirt plates (BIORAD), including a 4x concentrated solution of SYPRO-Orange dye (Invitrogen, 5000x stock solution in DMSO) in D-PBS. All measurements were performed in duplicate using a MyiQ2 real-time PCR instrument (BIORAD). The negative first derivative of the melting curve (-d(RFU) / dT) was generated in iQ5 software version 2.1 (BIORAD). The data were then exported to Excel for Tm determination and graphical display of the data.

[0827] Six antibody sequences were expressed in fusion with ds-trastuzumab, and the percentage of purified monomer and melting temperature of each fusion protein, as well as the pseudoIgG antibody without the ds-trastuzumab domain, are described in Table 4 below. Expression levels were slightly lower than WT-IgG, and the monomer content was also slightly lower than WT-IgG.

[0828] Table 4: Monomer content and melting temperature of pseudoIgG1 antibody-ds-trastuzumab fusion protein

[0829]

[0830]

[0831]

[0832] Example 4: Generating the ds-Tras knockout (dsTrasKO) variant

[0833] Trastuzumab binding activity was knocked out by introducing mutations in four key amino acid residues in the complementary region of the antibody, thus affecting the antibody's binding ability to the receptor tyrosine protein kinase erbB-2 (HER2). Trastuzumab was modeled based on the PDB structure 1N8Z (available in the PDB public database). Four positions on the heavy chain were identified that may disrupt the interaction between ds-trastuzumab and its antigen. These positions include both arginine residues R59 and R50, which bind glutamate and aspartate on HER2. Restoring the charge by mutating arginine to glutamate residues should result in HER2 rejection. Without intending to be bound by scientific theory, two other positions (tyrosine residues Y33 and Y105) appear to stabilize the complex through a π-π interaction with phenylalanine. Without intending to be bound by scientific theory, mutation of alanine should disrupt this interaction. Crystallographic data from the 1N8Z file were analyzed using the BIOVIA Discovery Studio suite. Special attention should be paid to the existing interface region between the antibody (trastuzumab fab) and its target (HER2 protein domain) (see Figure 4).

[0834] Analysis of the 1N8Z eutectic interface region

[0835] During antibody-target interface region analysis, the BIOVIA Discovery Studio non-bonded interaction monitor was used to verify the validity of all non-covalent interactions. Table 5 presents an excerpt of the results. For each interaction, Table 5 presents the distance between interacting atoms, the nature of the interaction type, and the characteristics of the interacting atoms.

[0836] Table 5.

[0837] Extract the 1N8Z PDB file from the nonbonded interaction monitor. Chains A, B, and C correspond to trastuzumab, respectively. Fab light chain, trastuzumab Fab heavy chain, and HER2.

[0838]

[0839]

[0840]

[0841]

[0842] Based on a table of non-covalent interactions between trastuzumab fab and HER2 protein, certain interactions are selected to eliminate them in order to disrupt the binding interaction.

[0843] Proposed mutational combinations that disrupt binding interactions

[0844] Based on non-bonded interaction data, the following four amino acid residues were selected for mutation in order to maintain the original trastuzumab sequence.

[0845] The composition of trastuzumab heavy chain arginine 50 residues:

[0846] This arginine residue in the antibody causes a salt-bridge interaction with HER2 residues glutamate 558 and aspartic acid 560. Mutating arginine 50 to a glutamate residue induces rejection of HER2 residues glutamate 558 and aspartic acid 560.

[0847] The details of arginine residue 59 in the heavy chain of trastuzumab:

[0848] This arginine residue in the antibody causes a salt-bridge interaction with HER2 residues glutamate 558 and aspartic acid 560. Mutating arginine 59 to a glutamate residue induces rejection of HER2 residues glutamate 558 and aspartic acid 560.

[0849] The situation regarding arginine residue 33 in the tyrosine heavy chain:

[0850] The antibody's tyrosine 33 residue enables a π-interaction with the HER2 residue phenylalanine 573. Mutating tyrosine 33 to an alanine residue disrupts these π-interactions between the antibody and its target.

[0851] The situation regarding arginine residue 105 in the tyrosine heavy chain:

[0852] The antibody's tyrosine 105 residue enables a π-interaction with the HER2 residue phenylalanine 573. Mutating tyrosine 105 to an alanine residue disrupts these π-interactions between the antibody and its target.

[0853] Table 6 presents specific combinations of amino acid residues to be mutated. These combinations produce two variants of the trastuzumab antibody.

[0854] Table 6. Trastuzumab knockout variants (TrasKO)

[0855]

[0856] Variant 1Fab contains all four point mutations. Variant 2Fab, with two positively charged residues reversed to negatively charged residues, possesses a repulsion region that rejects the HER2 protein into the original trastuzumab HER2 binding region. Variant 3Fab, with disruption of the π-interaction with HER2 phenylalanine 573, induces destabilization of the trastuzumab-HER2 complex. All three dsTrasKO variants were found to eliminate binding to HER2 (see Figures 5A-5B). Variant 2 (dsTrasKO2) was identified as the optimal producer.

[0857] Example 5: Evaluation of dstrasko2 as an alternative stent for CH1 / CL

[0858] The dsTrasKO variants were evaluated in pseudo-IgG (monospecific) form (see Figure 6A) and pseudo-Fab form (see Figure 6B). The tested pseudo-IgG and pseudo-Fab forms exhibited expression and purification characteristics similar to the parental antibody (see Table 7).

[0859] Table 7: Expression and purification results of pseudoIgG antibodies with CH1 / CL dsTrasKO2 domain substitution

[0860]

[0861]

[0862] Example 6: Evaluation of bispecific anti-inflammatory drugs with natural structure using dstrasko2 as an alternative scaffold for CH1 / CL body

[0863] The dsTrasKO2 domain was evaluated as a substitute scaffold in bispecific antibodies with a native IgG structure. Specifically, an anti-IL4-dsTrasKO(Var2)x anti-IL13-huIgG1 bispecific design with an RF mutation in the Fc was synthesized (see Figure 7).

[0864] i. Expression of dsTrasKO2 bispecific molecules

[0865] Expression plasmids encoding the two heavy chains (dsTrasKO2-palm and WT-mortar) and two light chains (dsTrasKO and WT-κ / λ) of the corresponding constructs were propagated in *E. coli* DH5α. Plasmids for transfection were prepared from *E. coli* using the Qiagen EndoFree Plasmid Mega Kit.

[0866] HEK 293-FS cells grown in serum-free suspension culture (Invitrogen) were transfected with the indicated light chain (LC) and heavy chain (HC) plasmids using a polyethyleneimine transfection reagent. After culturing at 37°C for 7 days, the cells were removed by centrifugation, and the supernatant was passed through a 0.22 μm filter to remove particles.

[0867] For purification, the antibody was captured on a MabSelect SuRe column (catalog number: 11-0034-93, GE Healthcare) and eluted with 0.1M citrate buffer (pH 3.0), followed by direct desalting using a HiPrep 26 / 10 desalting column (catalog number: 17-05087-02, GE Healthcare). Possible homodimeric dsTrasKO2 single-specific molecules were sorted using an additional KappaSelect capture step (0.1M glycine pH 2.7 elution buffer) (see Figure 14). The protein was used for further characterization after purification by size exclusion chromatography (SEC, as described in Example 2 above) using a Superdex 200 26 / 60 (GE) and a final ultrafiltration concentration step. See Table 8 for purification results of the dsTrasKO2 bispecific construct. See Figures 7A-7D for representative bispecific designs and purification results.

[0868] Table 8. Purification yield of dsTrasKO2 bispecific construct

[0869]

[0870]

[0871] Evaluate antigen binding of bispecific antibodies containing dsTrasKO2 substitutions for CH1 / CL pairs.

[0872] Surface plasmon resonance (SPR) was used to assess antigen-antibody binding. The binding of antigens to antibody constructs was measured using a BIAcore 3000 instrument (GE Healthcare) and HBS-EP buffer (GE Healthcare) with SPR. Human IL4 (IL004, Millipore), human IL13 (IL012, Millipore), human HER2 (1129-ER, R&D Systems), human TNFα (H8916, SIGMA Aldrich), human CTLA4 (CT4-H5229, ACROBiosystems), human PD-1 (8986-PD, R&D Systems), and human IL6Ra (227-SR / CF, R&D Systems) were used as antigens. Anti-human Fc capture antibodies (Human Antibody Capture Kit, GE Life Sciences) were immobilized via primary amine groups (11000 RU) on a research-grade CM5 chip (GE Life Sciences) using a standard procedure. The ligand was captured at a flow rate of 10 μl / min to achieve a regulated RU value resulting in maximum analyte binding of 30 RU. Binding to HER2 was tested by capturing the antigen with an anti-human Fc antibody. The antibody construct being tested was used as the analyte and injected at a concentration of 100 nM for 240 seconds, with a dissociation time of 300 seconds, at a flow rate of 30 μL / min. Binding kinetics were measured using the captured antibody by injecting serially diluted analytes from 3 nM to 100 nM. For human IL4 and IL13, dilution series of 0.1 nM to 3 nM and 0.8 nM to 25 nM were used, respectively. The chip surface was regenerated by injecting the regeneration buffer provided in the capture kit for 2 min. The sensor map was double-referenced using a blank chip surface and an HBS-EP buffer blank. Data analysis was performed using BIAevaluation software version 4.1. Binding characteristics of mAb, pseudoIgG, and bispecific antibodies are shown in Table 9 below.

[0873] Table 9. Binding kinetics measured by SPR: Comparison of mAb, pseudoIgG, and bispecific antibodies

[0874]

[0875]

[0876]

[0877] SPR results showed that various VH / VL fusions with dsTrasKO2 retained parental antigen binding affinity, including bispecific antibodies.

[0878] Analyzing protein integrity

[0879] Protein integrity and potential mispairing in the heterodimer constructs were analyzed by LC-MS. Protein samples were deglycosylated at 37°C for 15 h using 12.5 μg of protein (diluted to 0.5 mg / mL in D-PBS buffer treated with 0.5 μl PNGaseF (glycerol-free, New England Biolabs)). LC-MS analysis was performed using an Agilent 6540UHD Accurate-MassQ-TOF LC / MS instrument. Reversed-phase (RP) chromatography was performed at 180 μL / min using a Poroshell 300SB-C8 5 μm (75 x 0.5 mm) (Agilent) column and a Poroshell 300SB-C8 5 μm (2.1 x 12.5 mm) (Agilent) guard column. The eluents were LC water, 0.1% formic acid (A) and 90% acetonitrile, 10% LC water, 0.1% formic acid (B). 2 μg of protein was injected onto the column and eluted using a linear gradient from 0% to 100% B over 13 minutes. Data analysis was performed using MassHunter software B.06 (Agilent). Molecular mass was calculated based on the protein's amino acid sequence using GPMAW software version 9.13a2 (LighthouseData). The results showed that the dsTrasKO2-antibody fusion protein was largely intact and exhibited correct heterodimer pairing (see Table 10 below).

[0880] Table 10. LC-MS results of dsTrasKO2-antibody fusion protein

[0881]

[0882]

[0883] In addition to LC-MS analysis, bispecific samples were analyzed by hydrophobic interaction chromatography (HIC) to detect potential mispairing or unexpected species. Analytical HIC was performed using an LC10 HPLC instrument (Shimadzu) with a TSKgelEther-5PW 10μm (2x75mm) (Tosoh Bioscience) at 25°C. The analysis was run at a flow rate of 0.1 ml / min and detection was performed at 280 nm. 5 μg of undiluted protein sample was applied to the column. Gradient elution was performed from 0 to 30 min (0% to 100% B), followed by 10 min of 100% B and 15 min of reequilibration. Buffer A consisted of 1.5 M ammonium sulfate and 25 mM sodium phosphate (pH 7.0). Buffer B consisted of 25 mM sodium phosphate (pH 7.0). Data were evaluated using LabSolutions software version 5.85 (Shimadzu). Data shows that the dsTrasKO2 bispecific samples have uniform HIC spectra, indicating that there are no unexpected species and that the bispecific samples have correct pairings (see Figure 8).

[0884] Thermal stability: Comparison of pseudoIgG and bispecific substances

[0885] The thermal stability of dsTrasKO2 pseudoIgG and bispecific samples was evaluated as described in Example 2 above. The melting point (Tm) was determined using differential scanning fluorometry (DSF) as described in Example 3. Compared to the parental monoclonal IgG, the dsTrasKO2-based constructs showed a reduced melting temperature, and no impaired stability was detected in thermal stability assays over 2 weeks (40°C, 4°C, and -80°C) (see Table 11 below).

[0886] Table 11. Thermal stability of dsTrasKO2 antibody constructs.

[0887]

[0888]

[0889] Example 7: Crystal structure of pseudofab IL-13

[0890] Crystallization of dsTrasKO2-IL13-pseudoFab was performed for structural studies. The crystallization experiment was set up as a sitting-drop experiment using a standard two-drop 96-well MRC plate with a 1:1 protein:stock ratio and incubation at 20°C. Crystallization hits for both TrasKO2-CH1 / CL and anti-IL13-TrasKO2 were screened against various commercially available sparse matrix sieves. Screening drops of TrasKO2-CH1 / CL (stock solution 15 mg / ml) and anti-IL13-TrasKO2 (stock solution 15.6 mg / ml) were prepared by mixing 100 nmol of protein solution with 100 nmol of stock solution in 20 mM HEPES pH 7.5, 0.1 M sodium chloride, and equilibrated in a sitting-drop vapor diffusion experiment against 80 μl of stock solution. TrasKO2-CH1 / CL final crystals suitable for data collection and structural resolution were grown at 20 °C using a stock solution consisting of 0.1 M sodium citrate phosphate (pH 4.2), 20% (w / v) polyethylene glycol 8000, and 0.2 M sodium chloride. Diffraction-grade IL13-TrasKO2-resistant crystals were grown using a stock solution consisting of 0.1 M CHES (pH 9.5) and 20% (w / v) polyethylene glycol 8000. All crystals were cryoprotected by adding 25% (w / v) ethylene glycol (final concentration) before rapid cooling in liquid nitrogen.

[0891] Data collection and structure determination

[0892] Diffraction data were collected on the beamline PSII of the Swiss Light Source (SLS) in Villingen, Switzerland. The diffraction data were processed using a combination of XDS (Kabsch, 2010) and AIMLESS (Evans & Mushudov, 2013) from the CCP4 program suite (Winn et al., 2011). The crystals were TrasKO2-CH1 / CL, space group I23, with cell parameters of [missing information]. 90.00°, 90.00°, 90.00°, data expansion The resolution is [not specified]. The crystal resisting IL13-TrasKO2 belongs to space group P3221 with cell parameters 145.37 145.37 52.06 90.00 90.00 120.00, and diffracts to [not specified].

[0893] The CCP4 implementation scheme of Phaser was used to solve the structure via molecular substitution (McCoy et al., 2007). For TrasKO2-CH1 / CL, the modified form of the trastuzumab-VH-VL domain and the CH1 / CL domain of the pdb entry 1n8z were used as search models. Molecular substitutions for anti-IL13-TrasKO2 were performed using the modified forms of the anti-IL13 VH / VL domains of the Sanofi internal structure of Fab anti-IL13 and the VH / VL domains of 1n8z as phasing models. Atomic models were constructed through iterative rounds of manual model building and refinement using Coot (Emsley et al., 2010) and Refine (Bricogne, 2017). For TrasKO2-CH1 / CL, the R- and R-free factors of the final model are 19.2 / 23.2, and for anti-IL13-TrasKO2, they are 21.0 / 31.2 (see Figure 9 for crystal structure). Example 8: Using dstrasko2 Evaluation of bispecific antibodies with tandem structures as alternative scaffolds for CH1 / CL

[0894] The dsTrasKO2 domain was evaluated as an alternative scaffold in the design of bispecific tandem IgG. In particular, an anti-GITR-dsTrasKO2 x anti-OX40-κ]-huIgG1 tandem IgG design was synthesized and tested (see Figures 10-12).

[0895] Expression of dsTrasKO2 bispecific tandem molecule

[0896] Expression plasmids encoding the heavy chain and two light chains (dsTrasKO and WT-κ / λ) of the corresponding construct were propagated in *E. coli* DH5α. Plasmids for transfection were prepared from *E. coli* using the Qiagen EndoFree Plasmid Mega Kit.

[0897] HEK 293-FS cells grown in serum-free suspension culture (Invitrogen) were transfected with the indicated LC and HC plasmids using the polyethyleneimine transfection reagent. After culturing at 37°C for 7 days, the cells were removed by centrifugation, and the supernatant was passed through a 0.22 μm filter to remove particles.

[0898] For purification, the antibody was captured on a MabSelect SuRe column (catalog number: 11-0034-93, GE Healthcare) and eluted with 0.1M citrate buffer (pH 3.0), followed by direct desalting using a HiPrep 26 / 10 desalting column (catalog number: 17-05087-02, GE Healthcare). Possible homodimer dsTrasKO2 single-specific molecules were sorted using an additional KappaSelect capture step (0.1M glycine pH 2.7 elution buffer) (see Figure 14). Following size exclusion chromatography (SEC) using a Superdex 200 26 / 60 (GE) and a final ultrafiltration concentration step, the protein was used for further characterization.

[0899] For representative bispecific design and purification results, see Figures 12A-12D.

[0900] Example 9: Evaluation of trispecific antibodies with CODV structure using dstrasko2 as an alternative scaffold for CH1 / CL body

[0901] The dsTrasKO2 domain was evaluated as an alternative scaffold in the design of a trispecific cross-variable domain (CODV). Specifically, the CODV-anti-Ox40 x anti-PD1]x anti-CD137-dsTrasKO2-huIgG1-LALA-KIH-RF construct was synthesized and tested (see Figure 13).

[0902] Expression of dsTrasKO2 trispecific CODV molecule

[0903] Expression plasmids encoding the heavy chain and two light chains (dsTrasKO and WT) of the corresponding construct, as well as expression plasmids encoding the two heavy chains (CODV-codone and dsTrasKO2-codone) and two light chains (CODV and dsTrasKO) of the corresponding construct, were propagated in *E. coli* DH5α. Plasmids for transfection were prepared from *E. coli* using the Qiagen EndoFree Plasmid Mega Kit.

[0904] HEK 293-FS cells grown in serum-free suspension culture (Invitrogen) were transfected with the indicated LC and HC plasmids using the polyethyleneimine transfection reagent. After culturing at 37°C for 7 days, the cells were removed by centrifugation, and the supernatant was passed through a 0.22 μm filter to remove particles.

[0905] For purification, antibodies were captured on a MabSelect SuRe column (catalog number: 11-0034-93, GE Healthcare) and eluted with 0.1M citrate buffer (pH 3.0), followed by direct desalting using a HiPrep 26 / 10 desalting column (catalog number: 17-05087-02, GE Healthcare). Samples were further purified on a MonoS cation exchange column (catalog number: 17-5169-01, GE Healthcare, with a 0–1M NaCl salt gradient in 0.01M L-histidine pH 6.0 buffer). Following an ultrafiltration concentration step, the protein was used for further characterization.

[0906] For representative trispecific design and purification results, see Figures 13A-13D.

[0907] Example 10: Evaluation of bispecific T cell adaptor antibodies using dstrasko2 as an alternative scaffold for CH1 / CL

[0908] General Method

[0909] Analytical size exclusion chromatography (SEC)

[0910] Analytical SEC was performed at 25°C using an AdvanceBio 300 column (4.6 mm x 300 mm) and an AdvanceBio 300 guard column (Agilent Technologies) on a BioSECcurity instrument (PSS Polymer). The analysis was performed at a flow rate of 0.5 mL / min using 2x concentrated D-PBS buffer (Thermo Fisher Scientific) and detection was performed at 280 nm. 10 μl of protein sample (1 mg / mL) was applied to the column. Data were evaluated using WinGPC software version 8.1 (PSSPolymer). For molecular weight estimation, the SEC column was calibrated using a protein calibration standard mixture (Agilent Technologies).

[0911] Analytical hydrophobic interaction chromatography (HIC)

[0912] Analytical HIC was performed at 25°C using an LC10 HPLC instrument (Shimadzu) or a Vanquish HPLC instrument (Thermo Fisher Scientific) equipped with a TSKgel Butyl-NPR column (2.5 μm, 4.6 x 35 mm) (Tosoh Bioscience). The analysis was run at a flow rate of 1 mL / min and detected at 280 nm. 5 μg of undiluted protein sample was applied to the column. Gradient elution was performed from 15% B to 85% B over 7 min, followed by 1 min to 100% B, then 1 min to 15% B, and finally equilibration at 15% B for 3 min. Buffer A consisted of 1.5 M ammonium sulfate and 25 mM sodium phosphate (pH 7.0). Buffer B consisted of 25 mM sodium phosphate (pH 7.0). Data were evaluated using LabSolutions software version 5.85 (Shimadzu) or Chromeleon 7 software (Thermo Fisher Scientific).

[0913] nanoDSF

[0914] The onset temperature (Tinitial) and melting point (Tm) of protein denaturation were determined using nano-differential scanning fluorimetry (nanoDSF). Samples were diluted to a final concentration of 0.5 μg / μl in preparation buffer and loaded in duplicate into nanoDSF capillaries (Nanotemper Technologies). All measurements were performed using a Prometheus NT.plex nanoDSF device (Nanotemper Technologies). Heating was performed at a rate of 1 °C / min from 20 °C to 95 °C. Data were recorded using PR. Thermal Control Software version 2.3.1 (Nanotemper Technologies) and analyzed using PR. Stability Analysis Software version 1.0.3 (Nanotemper Technologies).

[0915] Surface plasmon resonance (SPR)

[0916] The binding of the antigen to the antibody construct was measured using surface plasmon resonance (SPR) with a BIAcore 8K instrument (GE Healthcare) and HBS-EP+ buffer (GE Healthcare). For binding kinetics and affinity determination, human CD3εδ-Fc-His and human CD123-Fc-His fusion proteins (both from internal sources) were used as antigens. Anti-His capture antibodies (His Capture Kit, GE Life Sciences) were immobilized via primary amine groups (11000 RU) on a research-grade CM5 chip (GE Life Sciences) using a standard procedure. The antigen was captured at a flow rate of 10 μL / min through the surface of the anti-His capture chip for 90 seconds to achieve antibody binding levels between 10 RU and 30 RU. The antibody was serially injected at 30 μL / min in two-fold dilutions from 100 nM to 3.1 nM (for determining CD123 affinity) and from 400 nM to 3.1 nM or 100 nM to 3.1 nM (for determining CD3 binding affinity) for 240 seconds. Dissociation was measured by injecting HBS-EP+ buffer for 1200 seconds at a rate of 30 μL / min. The chip surface was regenerated by injecting regeneration buffer (His Capture Kit, GE Life Sciences). The sensor plots were double-referenced using a blank chip surface and an HBS-EP+ buffer blank. The data were fitted with a 1:1 Langmuir binding model using Biacore 8K evaluation software version 1.11.7442 (GE Healthcare) to determine the kinetic and affinity constants ka, kd, and KD.

[0917] For the assessment of the relative binding levels (%Rmax) of antibodies to CD3 and CD123, antibodies were captured onto a sensor chip via anti-Fc affinity capture. In this assay, human CD3εδ-FLAG-His (#CT038-H2508H, SinoBiological) and human CD123 (#301-R3 / CF, R&D Systems) proteins were used. Anti-human Fc capture antibodies (Human Antibody Capture Kit, GE Life Sciences) were immobilized via primary amine groups (11000 RU) on a research-grade CM5 chip (GE Life Sciences) using a standard procedure. Antibodies were captured at a flow rate of 10 μl / min to achieve a regulated RU value resulting in maximum analyte binding of 10 to 30 RU. The antigen was used as the analyte and injected at concentrations of 400 nM and 100 nM (for CD3εδ-FLAG-His) or 100 nM (for CD123). The antigen was injected for 240 seconds, with a dissociation time of 300 seconds, at a flow rate of 30 μL / min. The chip surface was regenerated by 2 min injections with the regeneration buffer provided by the capture kit. The sensor map was double-referenced using a blank chip surface and an HBS-EP buffer blank. Data analysis and binding level determination were performed using Biacore 8K evaluation software version 1.11.7442 (GE Healthcare). The %Rmax value was calculated by dividing the maximum binding level by the theoretical Rmax value. Rmax = Rcapture, binding stoichiometry N, antibody molecular weight Mw (Ab), and antigen molecular weight Mw (Ag) were used to determine the binding level. * N * The theoretical Rmax value is calculated using (Mw(Ag) / Mw(Ab)).

[0918] mass spectrometry

[0919] Protein integrity and potential mispairing in the heterodimer constructs were analyzed by LC-MS. Protein samples were deglycosylated at 37°C for 16 h using 12.5 μg of protein (diluted to 0.17 mg / mL in LC-MS grade water (Thermo Scientific) treated with 0.5 μL PNGaseF (glycerol-free, New England Biolabs). LC-MS analysis was performed using an Orbitrap Fusion Lumos Tribrid mass spectrometer. Reversed-phase (RP) chromatography was performed at 300 μL / min using a MabPac RP HPLC column (analytical 4 μm particle size, 2.1 x 100 mm) (Thermo Scientific). The eluents were LC water, 0.1% formic acid (A), and 90% acetonitrile, 10% LC water, 0.1% formic acid (B). 2 μg of protein solution was injected onto the column, and elution was performed using a linear gradient from 0% to 95% B over 12 min. Data analysis was performed using Expressionist software version 13.0.3 (Genedata). Molecular mass was calculated based on the amino acid sequence of the protein using GPMAW software version 10.32b1 (Lighthouse data).

[0920] Cytotoxicity assay using bispecific TrasKO2 molecules

[0921] Bispecific TrasKO2 molecules were analyzed using primary human T cells in a cytotoxicity assay. Human peripheral blood mononuclear cells (PBMCs) from healthy donor blood were isolated in Leucosep tubes (Greiner Bio-One, #227290) using 15 mL Histopaque (Sigma-Aldrich, #10771) and centrifuged at 1000 x g for 10 min. The isolated PBMCs were washed twice in autoMACS wash buffer (Miltenyi Biotec, #130-091-222) supplemented with 5% MACSBSA stock solution (Miltenyi Biotec, #130-091-370). Primary human T cells were isolated from human PBMCs using the manufacturer's protocol with a MACSpro separator (Miltenyi Biotec) and a Pan T cell isolation kit (Miltenyi Biotec, #130-096-535). The isolated human T cells were separated into 5 x 10⁻⁶ cells. 6THP-1 target cells / mL were resuspended in RPMI GlutaMAX I medium (Gibco, #72400) supplemented with 10% FCS HI (Gibco, #10082-147). Prior to cytotoxicity assays, THP-1 target cells (ADCC TIB-202) were stained with 1 μM CFSE (Invitrogen, #C1157) at 37°C for 15 min. Cells were washed twice in RPMI + GlutaMAX I medium and centrifuged at 400 x g for 5 min. The cells were then resuspended at 5 x 10⁻⁶ cells / mL. 5 Cells / mL were resuspended in RPMI medium supplemented with 10% FCS HI. CFSE-labeled THP-1 cells were mixed with human pan T cells at an effector-to-target cell ratio of 10:1 and seeded in 96-well assay plates (Greiner BioOne, #650185) at a total volume of 100 μL / well. Bispecific TrasKO2 molecules were added to the cells in 11 dilution series from 10 nM to 0 nM (1:6 dilution) at a volume of 5 μL / well, and the cells were incubated at 37 °C and 5% CO2 for 20 h. After incubation, the cells were stained with 5 μg / mL 7-AAD (Invitrogen, #A1310) at 4 °C for 30 min. To determine cytotoxicity, dead target cells were measured by gating CFSE / 7-AAD double-positive THP-1 cells on an LSRII flow cytometer (BD), and EC50 values ​​were determined using Xlfit software.

[0922] The dsTrasKO2 domain was evaluated as an alternative scaffold in the design of bispecific T-cell connector antibodies. Bispecific T-cell connectors were generated by using anti-TCRα / β or one of two different anti-CD3ε as the effector arm and anti-CD123 as the target arm. Negative controls for both arms were generated using a TNP antibody sequence (tinitrophenol antibody). The bispecific protein was purified using MabSelect Sure, followed by KappaSelect and SEC. The bispecific antibody was expressed as wild-type bispecific IgG (detecting naturally occurring mispairing) or with a TrasKO2 substitution on one of the Fab arms (generating two possible reengineered Fab arms). The biophysical characterization of the bispecific antibody is described in Table 12 below.

[0923] Table 12. Biophysical characterization of the dsTrasKO2-T cell connector antibody construct.

[0924]

[0925]

[0926]

[0927] In addition to the biophysical characterization described above, the bispecific TrasKO2 molecule was also analyzed in a cell-based cytotoxicity assay. As shown in Figure 15A, all three anti-CD3εx anti-CD123 bispecific antibodies exhibited considerable and robust activity. The presence of the dsTrasKO2 domain in antibody IDs 34 and 35 did not negatively impact activity, but reduced chain mispairing. Negative control antibodies IDs 45 and 46 containing TNP antibody sequences were used. As shown in Figure 15B, all three anti-CD3εx anti-CD123 bispecific antibodies with alternative anti-CD3ε binding domains also exhibited considerable and robust activity. The presence of the dsTrasKO2 domain in antibody IDs 37 and 38 did not negatively impact activity, but reduced chain mispairing. As negative controls, antibody IDs 47 and 48 containing TNP antibody sequences were used.

Claims

1. A binding protein comprising: At least one pseudo-Fab portion, the at least one pseudo-Fab portion comprising (1) a first VL domain (VLa), the first VL domain pairing with a first VH domain (VHa) to form a first functional antigen-binding site for binding target antigen A; (2) a first stable knockout VH domain (VHX), the first stable knockout VH domain pairing with a first stable knockout VL domain (VLX) to form a first stable knockout domain, wherein the VLX consists of the amino acid sequence of SEQ ID NO: 76, and The VHX consists of the amino acid sequence of SEQ ID NO:

78.

2. The binding protein according to claim 1, wherein the binding protein is a multispecific binding protein, and the multispecific binding protein further comprises at least a second VL domain (VLb), the second VL domain pairing with a second VH domain (VHb) to form a second functional antigen binding site for binding target antigen B.

3. The binding protein according to claim 1 or 2, wherein the binding protein is a multispecific binding protein, further comprising: The first Fab portion includes (1) a second VL domain (VLb) that pairs with a second VH domain (VHb) to form a second functional antigen binding site for binding target antigen B; and (2) a first CH1 domain that pairs with a first CL domain.

4. The binding protein of claim 3, wherein the binding protein further comprises a linker portion operatively connecting the first Fab portion and the at least one dummy Fab portion. The stable knockout domain comprises (5) one or more inactivating mutations that eliminate its binding to the target antigen; and (6) one or more engineered interstrand disulfide bonds.

5. The binding protein of claim 4, wherein the linker moiety is a Gly-Ser linker of the formula (Gly4Ser)n, wherein n is 1-10. n 1. A method of treating a patient with a disease or disorder comprising administering to the patient a therapeutically effective amount of a compound of Formula (I): ###0001### (I) or a pharmaceutically acceptable salt thereof, 6. The binding protein according to claim 4, wherein the linker portion is a heterodimerization domain.

7. The binding protein according to claim 1 or 2, wherein it independently comprises one or two first pseudo-Fab moieties and one or two first Fab moieties.

8. The binding protein of claim 2, comprising a separate protein chain selected from the group consisting of: (a) VHa-CH1-L1-VHb-L2-VHX and Vla-CL and VLb-L3-VLX; (b) Vha-L2-VHX-L1-VHb-CH1 and Vla-L3-VLX and VLb-CL; (c)Vha-CH1-L1-Vha-CH1 and VHb-L2-VHX-L3-VHb-L4-VHX and two chains VLb-L5-VLX and two chains Vla-CL; The chains in (a) and (b) exist once or twice, and L1, L2, L3, L4 and L5 are independently identical or different joints.

9. The binding protein according to claim 1 or 2, wherein the at least one pseudo-Fab moiety comprises a first polypeptide chain and a second polypeptide chain, the first polypeptide chain having a structure represented by the following formula: (Ia)N-VHa-L1-VHX-C Furthermore, the second polypeptide chain has a structure represented by the following formula: (Iia)N-Vla-L2-VLX-C L1 and L2 are connectors that may or may not exist independently, and N and C represent N-terminals and C-terminals, respectively.

10. The binding protein according to claim 1 or 2, wherein the at least one pseudo-Fab moiety comprises a first polypeptide chain and a second polypeptide chain, the first polypeptide chain having a structure represented by the following formula: (Ib)N-VHX-L1-VHa-C Furthermore, the second polypeptide chain has a structure represented by the following formula: (IIb)N-VLX-L2-VLa-C L1 and L2 are connectors that may or may not exist independently, and N and C represent N-terminals and C-terminals, respectively.

11. The binding protein according to claim 1 or 2, wherein the at least one pseudo-Fab moiety comprises a first polypeptide chain and a second polypeptide chain, the first polypeptide chain having a structure represented by the following formula: (Ic)N-VLa-L1-VHX-C Furthermore, the second polypeptide chain has a structure represented by the following formula: (IIc)N-VHa-L2-VLX-C L1 and L2 are connectors that may or may not exist independently, and N and C represent N-terminals and C-terminals, respectively.

12. The binding protein according to claim 1 or 2, wherein the at least one pseudo-Fab moiety comprises a first polypeptide chain and a second polypeptide chain, the first polypeptide chain having a structure represented by the following formula: (Id)N-VHX-L1-VLa-C Furthermore, the second polypeptide chain has a structure represented by the following formula: (Iid)N-VLX-L2-Vha-C L1 and L2 are connectors that may or may not exist independently, and N and C represent N-terminals and C-terminals, respectively.

13. The binding protein of claim 1 or 2, further comprising one or more additional binding domains operatively linked to the N-terminus or C-terminus of the binding protein.

14. The binding protein of claim 13, wherein the one or more additional binding domains are operatively linked to the N-terminus of the at least one pseudo-Fab moiety.

15. The binding protein of claim 3, wherein the binding protein is a multispecific binding protein, wherein: a) The at least one fake Fab portion further comprises: First heterodimerization domain (HD1). b) The first Fab portion further includes: The second heterodimerization domain (HD2).

16. The binding protein of claim 15, wherein the first and second heterodimerization domains comprise first and second Fc domains.

17. The binding protein of claim 16, wherein the first and second Fc domains comprise a universal structure: a hinge-CH2 domain-CH3 domain.

18. The binding protein of claim 2, wherein the binding protein is a multispecific binding protein comprising four polypeptide chains forming at least two antigen-binding sites, wherein (a) The first polypeptide comprises a structure represented by the following formula: VLa-L1-VLX [I] (b) The second polypeptide comprises a structure represented by the following formula: Vha-L2-VHX-FC1 [II] (c) The third polypeptide comprises a structure represented by the following formula: VLb-CL [III] (d) The fourth polypeptide contains a structure represented by the following formula: VHb-CH1-FC2 [IV] in: CL is the constant domain of the immunoglobulin light chain; CH1 is the constant domain of the heavy chain of immunoglobulin CH1; FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and L1 and L2 are independent amino acid linkers that are the same or different.

19. The binding protein of claim 2, further comprising six polypeptide chains forming four antigen-binding sites, wherein... (a) The first and second polypeptides contain structures represented by the following formula: VLa-L1-VLX [I] and [II] (b) The third and fourth polypeptides contain structures represented by the following formula: VLb-CL [III] and [IV] (c) The fifth polypeptide comprises a structure represented by the following formula: VHa-L2-VHX-L3-VHb-CH1-FC1 [V] (d) The sixth polypeptide contains a structure represented by the following formula: VHa-L2-VHX-L3-VHb-CH1-FC2 [VI] in: CL is the constant domain of the immunoglobulin light chain; CH1 is the constant structural domain of the immunoglobulin heavy chain; FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and L1, L2, and L3 are amino acid linkers.

20. The binding protein of claim 2, further comprising six polypeptide chains forming four antigen-binding sites, wherein... (a) The first and second polypeptides contain structures represented by the following formula: VLa-L1-VLX [I] and [II] (b) The third and fourth polypeptides contain structures represented by the following formula: VLb-CL [III] and [IV] (c) The fifth polypeptide comprises a structure represented by the following formula: VHb-CH1-L3-VHa-L2-VHX-FC1 [V] (d) The sixth polypeptide contains a structure represented by the following formula: VHb-CH1-L3-VHa-L2-VHX-FC2 [VI] in: CL is the constant domain of the immunoglobulin light chain; CH1 is the constant structural domain of the immunoglobulin heavy chain; FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and L1, L2, and L3 are amino acid linkers.

21. The binding protein according to claim 2, further comprising six polypeptide chains forming four antigen-binding sites, wherein... (a) The first and second polypeptides contain structures represented by the following formula: VLa-L1-VLX [I] and [II] (b) The third and fourth polypeptides contain structures represented by the following formula: VLb-CL [III] and [IV] (c) The fifth polypeptide comprises a structure represented by the following formula: VHa-L2-VHX-L3-VHa-L4-VHX-FC1 [V] (d) The sixth polypeptide contains a structure represented by the following formula: VHb-CH1-L5-VHb-CH1-FC2 [VI] in: CL is the constant domain of the immunoglobulin light chain; CH1 is the constant structural domain of the immunoglobulin heavy chain; FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and L1, L2, L3, L4, and L5 are amino acid linkers.

22. The binding protein according to claim 2, further comprising four polypeptide chains forming three antigen-binding sites, wherein: (a) The first polypeptide comprises a structure represented by the following formula: VLa-L1-VLX [I] (b) The second polypeptide comprises a structure represented by the following formula: VHa-L2-VHX-FC1 [II] (c) The third polypeptide comprises a structure represented by the following formula: VLb-L3-VLc-L4-CL [III] (d) The fourth polypeptide contains a structure represented by the following formula: VHc-L5-VHb-L6-CH1-FC2 [IV] in: VLc is the variable domain of the third immunoglobulin light chain; VHc is the variable domain of the third immunoglobulin heavy chain; CL is the constant domain of the immunoglobulin light chain; CH1 is the constant domain of the heavy chain of immunoglobulin CH1; FC1 and FC2 are Fc domains containing the immunoglobulin hinge region and the CH2 and CH3 immunoglobulin heavy chain constant domains; and L1, L2, L3, L4, L5, and L6 are amino acid linkers. in (1) The third VL domain (VLc) pairs with the third VH domain (VHc) to form a third functional antigen-binding site for binding target antigen C; and (2) The polypeptide of Formula III and the polypeptide of Formula IV form a cross-linked light chain-heavy chain pair (CODV).

23. The binding protein of claim 16, wherein the Fc domain comprises one or more kilopod (KIH) mutations.

24. The binding protein according to claim 1 or 2, wherein the melting temperature (T) of the pseudo-Fab moiety is... m It is at least 4 degrees Celsius higher than the reference Fab molecule.

25. The binding protein according to claim 2, wherein target antigen A and target antigen B are different epitopes of the same antigen.

26. Use of a pseudo-Fab having a stable knockout domain for reducing heavy-light chain mispairing in a multispecific binding protein, wherein the stable knockout domain comprises VHK and VLX domains, the VHK and VLX domains comprising (1) one or more inactivating mutations relative to the wild-type domains, eliminating their binding to the target antigen; and (2) one or more engineered interchain disulfide bonds relative to a reference Fab molecule, conferring enhanced thermal stability (Tm) to the pseudo-Fab, wherein the reference Fab molecule is identical to the pseudo-Fab molecule except that, in the pseudo-Fab reference molecule, the CH1 and CL domains of the reference Fab molecule are replaced by the VHX and VLX domains. VLX is a stable knockout of the variable domain of the immunoglobulin light chain; VHX is a stable knockout of the variable domain of the immunoglobulin heavy chain; The VLX consists of the amino acid sequence of SEQ ID NO: 76, and the VHX consists of the amino acid sequence of SEQ ID NO:

78.

27. An isolated nucleic acid molecule comprising a nucleotide sequence encoding a binding protein according to any one of the preceding claims.

28. An expression vector comprising the nucleic acid molecule according to claim 27.

29. An isolated host cell comprising the nucleic acid molecule of claim 27 or the expression vector of claim 28.

30. A method for producing a binding protein according to any one of claims 1-25, comprising culturing a host cell according to claim 29 under conditions that cause the binding protein to be expressed; and purifying the binding protein from the host cell.

31. A pharmaceutical composition comprising a pharmaceutically acceptable carrier and a therapeutically effective amount of the binding protein according to any one of claims 1-25.

Images