WO2025259749A1 - Stable single-chain variable fragments - Google Patents

Abstract

The present invention relates to the field of single-chain variable fragments (scFvs) having improved biophysical properties and uses thereof.

Description

STABLE SINGLE-CHAIN VARIABLE FRAGMENTS

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 63/658,923, filed June 12, 2024.

DESCRIPTION OF THE TEXT FILE SUBMITTED ELECTRONICALLY

[0002] The present application contains a Sequence Listing, which has been submitted electronically in XML format and is hereby incorporated by reference in its entirety. The computer readable format copy of the Sequence Listing, which was created on June 3, 2025, is named 10322-W001-SEC_ST26.xml and is 92,464 bytes in size.

FIELD OF THE INVENTION

[0003] The present invention relates to the field of single-chain variable fragments (scFvs) having improved biophysical properties and uses thereof. Such improved biophysical properties include yield, aggregation, thermal stability, and/or hydrophobicity.

BACKGROUND OF THE INVENTION

[0004] Single-chain variable fragments (scFvs) are valuable components of antibodybased therapeutics due to their small size. However, scFv-containing molecules usually have lower stability, higher likelihood to aggregate, and a tendency to undergo domain swapping. One solution to improve stability and reduce aggregation of scFvs is to introduce two cysteines that form a disulfide between the variable heavy (VH) and variable light (VL) domains as a cys-clamp. Despite the improved stability imparted by a cys-clamp, molecules containing cys- clamped scFvs typically have lower yields and can exhibit a reduced pharmacokinetic (PK) half-life compared to their non-cys-clamped counterparts. In addition, introducing cys-clamps between the VH and VL domains allows the possibility for the scFvs to undergo domain swapping and remain locked in that configuration upon disulfide formation between the swapped VH and VL.

[0005] There is therefore a need for improved methods of making scFvs that are more stable. Such scFvs are useful in treating subjects in need thereof. SUMMARY OF THE INVENTION

[0006] The present invention provides a single chain variable fragment (scFv) comprising a heavy chain variable region (VH), a light chain variable region (VL), and a linker connecting the VH and VL, wherein the linker comprises a cysteine (linker cysteine), and the VH or VL comprises a cysteine (VH cysteine or VL cysteine). In an embodiment, the VH comprises a cysteine. In an embodiment, the VL comprises a cysteine. In an embodiment, the scFv comprises a disulfide bond between the linker cysteine and the VH cysteine or the VL cysteine. In an embodiment, the scFv is more stable than an scFv without the disulfide bond. [0007] The present invention provides a single chain variable fragment (scFv) comprising a heavy chain variable region (VH), a light chain variable region (VL), a linker comprising a cysteine, and wherein the linker connects the VH and VL, and: the scFv is in a VH-VL (HL) orientation from N-terminus to C-terminus and comprises a VH cysteine at position 13, 61, 62, 83, 84, or 85 according to Kabat numbering; or the scFv is in a VL-VH (LH) orientation from N-terminus to C-terminus and comprises a VL cysteine at position 42, 45, 57, 81, or 83, or a VH cysteine at position 101 according to Kabat numbering.

[0008] In an embodiment, the scFv is in a VH-VL (HL) orientation from N-terminus to C-terminus and comprises a VH cysteine at position 13. In an embodiment, the scFv is in a VH-VL (HL) orientation from N-terminus to C-terminus and comprises a VH cysteine at position 61. In an embodiment, the scFv is in a VH-VL (HL) orientation from N-terminus to C- terminus and comprises a VH cysteine at position 62. In an embodiment, the scFv is in a VH- VL (HL) orientation from N-terminus to C-terminus and comprises a VH cysteine at position 83. In an embodiment, the scFv is in a VH-VL (HL) orientation from N-terminus to C-terminus and comprises a VH cysteine at position 84. In an embodiment, the scFv is in a VH-VL (HL) orientation from N-terminus to C-terminus and comprises a VH cysteine at position 85.

[0009] In an embodiment, the scFv is in a VL-VH (LH) orientation from N-terminus to C-terminus and comprises a VL cysteine at position 42. In an embodiment, the scFv is in a VL- VH (LH) orientation from N-terminus to C-terminus and comprises a VL cysteine at position 45. In an embodiment, the scFv is in a VL-VH (LH) orientation from N-terminus to C-terminus and comprises a VL cysteine at position 57. In an embodiment, the scFv is in a VL-VH (LH) orientation from N-terminus to C-terminus and comprises a VL cysteine at position 81. In an embodiment, the scFv is in a VL-VH (LH) orientation from N-terminus to C-terminus and comprises a VL cysteine at position 83. In an embodiment, the scFv is in a VL-VH (LH) orientation from N-terminus to C-terminus and comprises a VH cysteine at position 101. [0010] In an embodiment, the linker comprises a cysteine at a position between 1-15.

[0011] In an embodiment, the linker comprises a cysteine at position 3, 4, 8, 9,10,14 or

13 according to linear numbering. In an embodiment, the linker comprises a cysteine at position 3. In an embodiment, the linker comprises a cysteine at position 4. In an embodiment, the linker comprises a cysteine at position 8. In an embodiment, the linker comprises a cysteine at position 9. In an embodiment, the linker comprises a cysteine at position 10. In an embodiment, the linker comprises a cysteine at position 14. In an embodiment, the linker comprises a cysteine at position 13.

[0012] In an embodiment, the scFv is in an HL orientation, and the VH cysteine and linker cysteine are at positions 84 (VH) / 9 (linker); 61 (VH) / 14 (linker); 85 (VH) / 10 (linker); 62 (VH) / 14 (linker); 13 (VH) / 3 (linker); or 83 (VH) / 8 (linker). In an embodiment, the VH cysteine and linker cysteine are at positions 84 (VH) / 9 (linker). In an embodiment, the VH cysteine and linker cysteine are at positions 61 (VH) / 14 (linker). In an embodiment, the

VH cysteine and linker cysteine are at positions 85 (VH) / 10 (linker). In an embodiment, the

VH cysteine and linker cysteine are at positions 62 (VH) / 14 (linker). In an embodiment, the

VH cysteine and linker cysteine are at positions 13 (VH) / 3 (linker). In an embodiment, the

VH cysteine and linker cysteine are at positions 83 (VH) / 8 (linker).

[0013] In an embodiment, the scFv is in an LH orientation, and the VL cysteine and linker cysteine are at positions 81 (VL) / 8C (linker); 42 (VL) / 9 (linker); 45 (VL) / 10 (linker); 83 (VL) / 4 (linker); or 57 (VL) / 13 (linker). In an embodiment, the VL cysteine and linker cysteine are at positions 81 (VL) / 8C (linker). In an embodiment, the VL cysteine and linker cysteine are at positions 42 (VL) / 9 (linker). In an embodiment, the VL cysteine and linker cysteine are at positions 45 (VL) / 10 (linker). In an embodiment, the VL cysteine and linker cysteine are at positions 83 (VL) / 4 (linker). In an embodiment, the VL cysteine and linker cysteine are at positions 57 (VL) / 13 (linker).

[0014] In an embodiment, the scFv is in an orientation from N-terminus to C-terminus of VL-VH (LH), and the VH cysteine is at position 101 and the linker cysteine is at position 13. [0015] In an embodiment, the scFv comprises a disulfide bond between the VH cysteine and a linker cysteine, or between the VL cysteine and the linker cysteine.

[0016] The present invention provides an scFv comprising a VH, a VL, a linker connecting the VH and VL, and an N-terminal linker comprising a cysteine (N-terminal linker cysteine). In an embodiment, the VH comprises a cysteine (VH cysteine). In an embodiment, the VL comprises a cysteine (VL cysteine). In an embodiment, the scFv comprises a disulfide bond between the N-terminal linker cysteine and the VH cysteine. In an embodiment, the scFv comprises a disulfide bond between the N-terminal linker cysteine and the VL cysteine. In an embodiment, the scFv is in an HL orientation and comprises a VL cysteine at position 42, 45, 57, or 81 according to Kabat numbering. In an embodiment, the scFv is in an LH orientation and comprises a VH cysteine at position 61, 62, 83, 84, or 85 according to Kabat numbering. In an embodiment, the scFv comprises an N-terminal linker cysteine at position -1, -2, -3, -4, - 5, -6, -7, -8, -9, -10, -11 according to N-terminal linker linear numbering.

[0017] In an embodiment, the scFv comprises an amino acid sequence given by one of SEQ ID NO: 31-37 and 39-87. In an embodiment, the scFv comprises an amino acid sequence given by one of SEQ ID NO: 1-4 and 38. In an embodiment, the scFv comprises an amino acid sequence comprising one of SEQ ID NO: 31-37 and 39-87. In an embodiment, the scFv comprises an amino acid sequence comprising one of SEQ ID NO: 1-4 and 38.

[0018] The present invention provides a construct comprising the scFv of the present invention. In an embodiment, the construct is a multispecific construct.

[0019] The present invention provides an scFv of the present invention for the manufacture of a medicament for the treatment of a disease or disorder. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0020] The present invention provides a construct of the present invention for the manufacture of a medicament for the treatment of a disease or disorder. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0021] The present invention provides an scFv of the present invention for use in treating a disease or disorder. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0022] The present invention provides a construct of the present invention for use in treating a disease or disorder. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease. [0023] The present invention provides an scFv of the present invention for use in therapy. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0024] The present invention provides a construct of the present invention for use in therapy. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0025] The present invention provides a use of an scFv of the present invention for the manufacture of a medicament for the treatment of disease or disorder. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0026] The present invention provides a use of a construct of the present invention for the manufacture of a medicament for the treatment of disease or disorder. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0027] The present invention provides an scFv of the present invention for use as a medicament.

[0028] The present invention provides a construct of the present invention for use as a medicament.

[0029] The present invention provides an scFv of the present invention for use in the treatment of a disease or disorder. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0030] The present invention provides a construct of the present invention for use in the treatment of a disease or disorder. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease. [0031] The present invention provides a method of treating a disease or disorder in a patient, said method comprising administering to the patient an scFv of the present invention. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0032] The present invention provides a method of treating a disease or disorder in a patient, said method comprising administering to the patient a construct of the present invention. In an embodiment, the disease is cancer. In an embodiment, the disease is an autoimmune disease. In an embodiment, the disease is a cardiac disease. In an embodiment, the disease is an inflammatory disease.

[0033] In an embodiment, the disease is cancer, an immune disease, an allergic disease, an infectious disease, or a non-infectious disease.

[0034] The present invention provides a pharmaceutical composition comprising an scFv of the present invention and one or more pharmaceutically acceptable carriers, diluents, or excipients.

[0035] The present invention provides a pharmaceutical composition comprising a construct of the present invention and one or more pharmaceutically acceptable carriers, diluents, or excipients.

[0036] In an embodiment, an scFv of the present invention is more stable than an scFv without the disulfide bond.

BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1A, IB, and 1C. Prior scFv cys-clamps exhibit poor behavior. A) ProA yields and percentages of aggregate species (HMW%) are shown for T cell engagers (TCEs) with (black) and without (gray) a cys-clamp (VH-103C/VL-43C) in two different CD3- binding scFvs and the same tumor associated antigen (TAA). B) Pharmacokinetic profiles of TAA-targeting TCEs across four bispecific antibody formats where CD3-binding scFvs are either cys-clamped (CC+) or non-cys-clamped (CC-). C) Potential disulfide formation in native (top row) and domain-swapped (bottom row) isomers containing a VH-to-VL cys-clamp (left column) or a VH/VL-to-scFv linker cys-clamp (right column). [0038] Figure 2A and 2B. In silico design of cys-clamps between the scFv linker and VL/VH domains. Models of the CD3 binder 1 scFv with the scFv linker added are shown on the right for both HL (Fig. 2A) and LH (Fig. 2B) orientations of the scFv. Six residues utilized in HL scFv linker cys-clamps and six residues utilized in LH scFv linker cys-clamps are labeled and shown as sticks in the zoomed out scFv structures on the right. Each of the twelve designed cys-clamps are shown as zoomed in, overlaid images of multiple structural models that utilize the same VH or VL position in the respective scFv linker cys-clamp. The disulfide-bonded residues between the VH or VL domains and the scFv linker are shown as sticks. The VH or VL position utilized for each designed cys-clamp is labeled in Kabat numbering, and linker cys-clamp positions are numbered linearly with the first position of the (G4Q)3 linker designated as position 1. The number of models the position was found during the in silico disulfide scan out of the 1000 HL or 1000 LH models is noted as n.

[0039] Figure 3A, 3B, and 3C. Yields, aggregation levels, melting temperatures, and HIC retention times of multispecific antibodies with scFv cys-clamp variants. Figure

3 A) TCEs in format 1 and format 2 containing CD3 binder 1 or CD3 binder 2 scFvs and a TAA target 1 binder with no cys-clamp (CC-), a VH-to-VL cys-clamp ( 103 C/43 C), or scFv linker cys-clamps 61C, 62C, 83C, and 84C with HL scFv orientation. Linker cys-clamps 57C and 101C with LH scFv orientation are also shown. Initial aggregation levels (HMW%) and the change in aggregation levels after 1 week incubation at 50°C (AHMW%) were both measured at 70 mg/mL protein concentration. Figure 3B) TCEs in format 1 format containing CD3 binder 1 or CD3 binder 2 scFvs and a TAA target 1 binder with no cys-clamp (CC-), a VH-to-VL cys- clamp (103C/43C), or scFv linker cys-clamps 13C and 85C with HL scFv orientation. No cys- clamp (CC-), VH-to-VL cys-clamp (103C/43C), and scFv linker cys-clamps 42C, 45C, 81C, and 83C are also shown with LH scFv orientation. Initial aggregation levels (HMW%) and the change in aggregation levels after 1 week incubation at 50°C (AHMW%) were both measured at 5 mg/mL protein concentration. Figure 3C) TCEs in format 1 containing CD3 binder 1 or CD3 binder 2 scFvs and a TAA target 5 binder, and multispecific antibodies in format 3 with Target 6 binder scFvs and Target 7 binder with no cys-clamp (CC-), a VH-to-VL cys-clamp (103C/43C for TCEs or 44C/100C for format 3 multispecific antibodies), or scFv linker cys- clamps 13C, 61C, 62C, 83C, 84C, and 85C with HL scFv orientation. No cys-clamp (CC-), VH-to-VL cys-clamp (103C/43C for TCEs or 44C/100C for format 3 multispecific antibodies), and scFv linker cys-clamps 42C, 45C, 57C, 81C, 83C, and 101C are also shown with LH scFv orientation. Initial aggregation levels (HMW%) and the change in aggregation levels after 2 week incubation at 40°C (AHM\V%) were measured at 70 mg/mL protein concentration. denotes variants were tested at 5 mg/mL concentration due to insufficient material to produce and test samples at 70 mg/mL concentration. Light gray panels indicate variants that were not made. “N.D ” denotes variants where materials were insufficient to produce experimental data. [0040] Figure 4A and 4B. Binding affinities and TDCC activities of TCE molecules with scFv cys-clamp variants. Figure 4A) Binding affinities (KD) toward CD3de antigen of format 1 TCEs containing CD3 binder 1 scFvs with no cys-clamp, VH-to-VL cys-clamp (103C/43C), or scFv linker cys-clamps. Figure 4B) EC50 values of mini-mAb and format 1 TCEs containing CD3 binder 1 scFvs in a target cell-based TDCC activity assay.

[0041] Figure 5A and 5B. Pharmacokinetic profiles of TCEs with CD3 binder 1 and CD3 binder 2 scFvs. Serum concentration levels of TCE molecules following 1 mg/kg intravenous administration in huFcRn Tg32 mice are shown for CD3 binder 1 scFv-containing mini-mAbs (Figure 5A) and CD3 binder 2 scFv-containing format 1 and mini-mAbs (Figure 5B).

[0042] Figure 6. Crystal structures of scFvs with cys-clamp variants. CD3 binder 2 scFv crystal structures with various cys-clamp variants are shown. VH domains are shown in white, VL domains are shown in gray, and the scFv linker is shown in black. Disulfide residues are shown as sticks and their positions are labeled in Kabat numbering for VH/VL residues, and linearly for (G4Q)3 linker residues.

[0043] Figure 7A and 7B. Yields and aggregation levels of 20 diverse Fvs to Target 2 converted to scFv-Fcs with no cys-clamp, VH-to-VL cys-clamp, and linker cys-clamp variants. Columns represent 10 mAbs against Target 2 with kappa light chains and 10 mAbs against Target 2 with lambda light chains, where the Fvs from the mAbs have been converted to scFvs and produced in scFv-Fc format in both HL and LH scFv orientations with cys-clamp variants, which are represented in the rows. Protein yields (Figure 7A) and aggregation levels (Figure 7B) are shown in each cell representing a converted scFv in HL or LH orientation with a cys-clamp variant. Cells are colored to scale by yields or aggregation levels. Black cells represent samples that were not produced or that had insufficient yields for analysis.

[0044] Figure 8A and 8B. HIC retention times and binding affinities of 20 diverse Fvs to Target 2 converted to scFv-Fcs with no cys-clamp, VH-to-VL cys-clamp, and linker cys-clamp variants. Columns represent 10 mAbs against Target 2 with kappa light chains and 10 mAbs against Target 2 with lambda light chains, where the Fvs from the mAbs have been converted to scFvs and produced in scFv-Fc format in both HL and LH scFv orientations with cys-clamp variants, which are represented in the rows. HIC retention times (Figure 8A) and binding affinities (Figure 8B) are shown in each cell representing a converted scFv in HL or LH orientation with a cys-clamp variant. Cells are colored to scale by HIC retention times or KD values. Black cells represent samples that were not produced or that had insufficient yields for analysis.

[0045] Figure 9. Schematic illustration of converting scFv linker cys-clamps to scFv N-terminal linker cys-clamps. Cys-clamps between a cysteine in the scFv linker between VH and VL and a cysteine in either VH or VL (A) can be converted to structurally similar cys-clamps using an N-terminal linker (B) by swapping the ordering of the VH and VL domains, translating the scFv linker to the N-terminus of the scFv, and placing the cys-clamp between a cysteine in the N-terminal linker and a cysteine in either VH or VL.

[0046] Figure 10. Yields, HIC retention times, and aggregation levels of N-terminal linker scFv cys-clamp variants. Molecule formats on the right were produced with scFv cys- clamp variants: no cys-clamp (CC-), VH-to-VL cys-clamp (103C/43C), or N-terminal linker cys-clamps in HL and LH scFv orientations. For each N-terminal linker cys-clamp, three adjacent cysteine positions in the N-terminal linker were screened.

[0047] Figure 11. Characterization of N-terminal linker scFv cys-clamp variants.

DETAILED DESCRIPTION

[0048] Single-chain variable fragments (scFvs) are valuable components of antibodybased therapeutics due to their small size. However, scFv-containing molecules usually have lower stability, higher likelihood to aggregate, and a tendency to undergo domain swapping. One solution to improve stability and reduce aggregation of scFvs is to introduce two cysteines that form a disulfide between the variable heavy (VH) and variable light (VL) domains as a cys-clamp. Despite the improved stability imparted by a cys-clamp, molecules containing cys- clamped scFvs typically have lower yields (Figure 1 A) and can exhibit a reduced pharmacokinetic (PK) half-life (Figure IB) compared to their non-cys-clamped counterparts. In addition, introducing cys-clamps between the VH and VL domains allows the possibility for the scFvs to undergo domain swapping and remain locked in that configuration upon disulfide formation between the swapped VH and VL (Figure 1C, left column).

[0049] In order to improve the manufacturability and pharmacokinetic behavior of scFv-containing molecules, scFv cys-clamps that utilize the scFv linker that connect the VH and VL domains were assessed. Placing the scFv cys-clamp between the scFv linker and the VH or VL domains decreased the flexibility of the scFv linker, reduced disassociation of the VH and VL domains in the scFv, and shielded the cys-clamp from the protein surface. In addition, a cys-clamp between the scFv linker and the VH or VL domain disfavors the cys- clamp from being formed if scFvs are domain swapped, and therefore domain swapping will be more energetically disfavored (Figure 1C).

[0050] scFvs containing disulfide bonds are disclosed in PCT Publication Number WO 2021/030657.

[0051] Described herein are scFvs with a single disulfide bond between a residue in either the VH or VL, and a residue in the linker. A single disulfide bond is sufficient to improve biophysical properties of scFvs.

[0052] A fragment antigen-binding (“Fab”) refers to an antigen binding domain having a heavy chain variable region (HCVR) and a heavy chain constant domain (CHI), and a light chain variable region (LCVR) and a light chain constant domain.

[0053] An Fv refers to an antigen binding domain having a VH domain and a VL.

[0054] A “single chain variable fragment” or “scFv” refers to a single polypeptide protein that has a VH and VL connected by a linker. An scFv can be in a VH-VL orientation, which means the VH is N-terminal to the VL, with a linker connecting the VH and VL. An scFv can be in a VL-VH orientation, which means the VL is N-terminal to the VH, with a linker connecting the VL and VH.

[0055] The linker that connects the VH and VL, or the N-terminal linker, can be a linker that is known in the art, such as GQ or GS linkers. Preferably, the linker that connects the VH and VL is a peptide linker that is between 15 and 20 amino acids or between 12 and 20 amino acids. The N-terminal linker can be a linker known in the art, as described herein, or as shown in Table 2 herein.

[0056] Exemplary linkers include G4Q based linkers (e.g. (G4Q)3 and (G4Q)4) and G4S based linkers (e.g. (G4S)3 and (G4S)4), and linkers described in PCT Publication Number WO 2022/096704. Linkers are also described in U.S. Patents 4,751 ,180 and 4,935,233 or WO 88/09344. An scFv or construct of the present invention may have one, two, three, four, five, or six repeats of the linker. For example two SG4S repeats would be binding domain (e.g. scFv)-SGGGGSSGGGGS-binding domain (e g. scFv) (SGGGGSSGGGGS given by SEQ ID NO: 30). In an embodiment, the linker is a polypeptide linker. In an embodiment, the linker comprises a sequence selected from the group consisting of (Gly3Ser)3 (SEQ ID NO: 9), (Gly4Ser)3 (SEQ ID NO: 10), (Gly3Ser)4 (SEQ ID NO: 11), (Gly4Ser)4 (SEQ ID NO: 12), (Gly3Ser)5 (SEQ ID NO: 13), (Gly4Ser)5 (SEQ ID NO: 88), (Gly3Ser)6 (SEQ ID NO: 14), (Gly4Ser)6 (SEQ ID NO: 15), GSADDAKKDAAKKDAAKKDDAKKDDAGS (SEQ ID NO: 16), GSADDAKKDAAKKDAAKKDDAKKDDAKKDAGS (SEQ ID NO: 17), (Gly3Gln)2 (SEQ ID NO: 18), (Gly4Gln)2 (SEQ ID NO: 19), (Gly3Gln)3 (SEQ ID NO: 20), (Gly4Gln)3

(SEQ ID NO: 21), (Gly3Gln)4 (SEQ ID NO: 22), (Gly4Gln)4 (SEQ ID NO: 23), (Gly3Gln)5

(SEQ ID NO: 24), (Gly4Gln)5 (SEQ ID NO: 25), (Gly3Gln)6 (SEQ ID NO: 26), (Gly4Gln)6

(SEQ ID NO: 27), (Gly3Ser)2 (SEQ ID NO: 28), and (Gly4Ser)2 (SEQ ID NO: 29).

[0057] As used herein, “VH cysteine,” “linker cysteine,” and “VL cysteine” refers to a cysteine residue in either the VH, linker, or VL, respectively. Exemplary cysteine positions in the linker are shown in Figure 2 (9C,14C,10C,3C,8C,4C,13C). It is understood that these positions are exemplary and that the cysteine in the linker can be moved in either direction while still retaining the desired effect (e.g. improved stability). For example, the linker cysteine can be moved one or two residues in either direction, and it is envisaged that improved stability is retained. An “N-terminal linker cysteine” refers to a cysteine in the N-terminal linker. An scFv comprising an N-terminal linker refers to an scFv that has a polypeptide sequence N- terminal to the scFv.

[0058] Exemplary cysteine positions include: VH:83C (cysteine at position 83 in the VH according to Kabat numbering)/linker:8C (cysteine at position 8 in the scFv linker according to linear numbering) for heavy chain variable region (HCVR)-light chain variable region (LCVR) orientation (N-terminus to C-terminus), VH:85C/linker: 10C for HCVR-LCVR orientation (N-terminus to C-terminus), VL:57C/linker: 13C for LCVR- HCVR orientation (N- terminus to C-terminus), and VL:81C/linker:8C for LCVR- HCVR orientation (N-terminus to C-terminus).

[0059] Sequences of anti-CD3 scFvs (without the cysteines used to form the disulfides) in the HL orientation are SEQ ID NO: 1-4, and an CD3 -binder- l_scFv_(G4Q)3 in the LH orientation is SEQ ID NO: 38. The four cysteines in each of these sequences form native disulfides, one in VH and one in VL. The designed VH-to-VL cys-clamps or linker cys-clamps as disclosed herein are additional to these cysteines. For example, amino acid sequences of scFvs in the Examples are given by SEQ ID NO: 3 or SEQ ID NO: 4, with the addition of a cysteine in the linker and in the VH or VL. SEQ ID NO: 31-37 and 39-45 provide amino acid sequences for the CD3-binder-l_scFv_(G4Q)3_HL of SEQ ID NO: 3 with additional cysteines of the present invention. SEQ ID NOs 46-60 provide amino acid sequences for the CD3 -binder- 2_scFv_(G4Q)3 with additional cysteines of the present invention. SEQ ID NOs 61-87 provide amino acid sequences of CD3 binders with N-terminal linkers. scFvs of the present invention and constructs comprising said scFvs are envisioned to comprise one or more of the aforementioned SEQ ID NOs.

[0060] The numbering of amino acids herein refers to Kabat numbering (Kabat, et al., Ann. NY Acad. Sci. 190:382-93 (1971); Kabat et al., Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242 (1991), with the exception of (i) the linker (connecting the linker and the VH or VL) cysteine positions which are numbered linearly with the first position of the linker located at the most N-terminal part of the linker, and designated as position 1; (ii) and the N-terminal linker cysteine positions, which are numbered linearly (N-terminal linker linear numbering) with the first position of the linker located at the most C-terminus of the linker (and designated as position -1). For example, in the context of an N-terminal linker, “-6” refers to a position that is N-terminal six amino acids from the C-terminus (e.g. GCGGGGQ-VH).

[0061] A “multispecific construct” is a construct (e.g. polypeptide or protein) that is able to specifically bind more than one target. For example, a multispecific construct may comprise a Fab that specifically binds one target, and an attached scFv of the present invention that specifically binds another target.

[0062] “Specifically binds” refers to binding to an antigen (target) with a dissociation constant (KD) is <10-7 M as measured via a surface plasma resonance technique (e.g., BIACore, GE-Healthcare Uppsala, Sweden) or Kinetic Exclusion Assay (KinExA, Sapidyne, Boise, Idaho).

[0063] Molecule stability can be determined according to procedures known in the art and as described herein. For example, Differential Scanning Fluorimetry (DSF) is a method to determine melting temperature (Tm). One may also determine aggregation levels after heat stress via analytical size exclusion chromatography.

[0064] The scFvs or constructs of the present invention can readily be produced in mammalian cells, non-limiting examples of which includes CHO, NSO, HEK293 or COS cells. The host cells are cultured using techniques well known in the art.

[0065] In certain embodiments, the present invention provides vectors comprising a nucleic acid encoding an scFv or construct of the invention or a portion thereof. Examples of vectors include, but are not limited to, plasmids, viral vectors, non-episomal mammalian vectors and expression vectors, for example, recombinant expression vectors. Vectors containing the polynucleotide sequences of interest (e.g., the polynucleotides encoding the polypeptides of the scFv or construct and expression control sequences) can be transferred into the host cell by well-known methods, which vary depending on the type of cellular host. Examples of vectors include, but are not limited to, plasmids, viral vectors, non-episomal mammalian vectors and expression vectors, for example, recombinant expression vectors. [0066] In certain embodiments, the present invention provides host cells into which a recombinant expression vector of the invention has been introduced. A host cell can be any prokaryotic cell or eukaryotic cell. Prokaryotic host cells include gram negative or gram positive organisms, for example E. coli or bacilli. Higher eukaryotic cells include insect cells, yeast cells, and established cell lines of mammalian origin. Examples of suitable mammalian host cell lines include Chinese hamster ovary (CHO) cells or their derivatives such as CHO and related cell lines which grow in serum -free media (see Rasmussen et al., 1998, Cytotechnology 28:31) or CHO strain DXB-11, which is deficient in DHFR (see Urlaub et al., 1980, Proc. Natl. Acad. Sci. USA 77:4216-20). Additional CHO cell lines include CHO-K1 (ATCC#CCL-61), EM9 (ATCC# CRL-1861), and UV20 (ATCC# CRL-1862). Additional host cells include the COS-7 line of monkey kidney cells (ATCC CRL 1651) (see Gluzman et al., 1981, Cell 23: 175), L cells, C127 cells, 3T3 cells (ATCC CCL 163), AM-l/D cells (described in U.S. Patent No. 6,210,924), HeLa cells, BHK (ATCC CRL 10) cell lines, the CV1/EBNA cell line derived from the African green monkey kidney cell line CV1 (ATCC CCL 70) (see McMahan et al., 1991, EMBO J.10:2821), human embryonic kidney cells such as 293, 293 EBNA or MSR 293, human epidermal A431 cells, human Colo205 cells, other transformed primate cell lines, normal diploid cells, cell strains derived from in vitro culture of primary tissue, primary explants, HL-60, U937, HaK or Jurkat cells. Appropriate cloning and expression vectors for use with bacterial, fungal, yeast, and mammalian cellular hosts are described by Pouwels et al. (Cloning Vectors: A Laboratory Manual, Elsevier, New York, 1985).

[0067] Typically, expression vectors used in any of the host cells will contain sequences for plasmid maintenance and for cloning and expression of exogenous nucleotide sequences. Such sequences, collectively referred to as "flanking sequences" in certain embodiments will typically include one or more of the following nucleotide sequences: a promoter, one or more enhancer sequences, an origin of replication, a transcriptional termination sequence, a complete intron sequence containing a donor and acceptor splice site, a sequence encoding a leader sequence for polypeptide secretion, a ribosome binding site, a polyadenylation sequence, a polylinker region for inserting the nucleic acid encoding the polypeptide to be expressed, and a selectable marker element. The leader sequence may comprise an amino acid sequence of SEQ ID NO: 5 (MDMRVPAQLLGLLLLWLRGARC) which is encoded by SEQ ID NO: 6 (atggacatgagagtgcctgcacagctgctgggcctgctgctgctgtggctgagaggcgccagatgc). The leader sequence may comprise an amino acid sequence of SEQ ID NO: 7. (MAWALLLLTLLTQGTGSWA) which is encoded by SEQ ID NO: 8 (atggcctggg ctctgctgct cctcacccte ctcactcagg gcacagggtc ctgggcc). The present invention contemplates molecule protein sequences without a leader sequence.

[0068] Various methods of protein purification may be employed to purify proteins, including, but not limited to, scFvs or constructs, and such methods are known in the art. [0069] The scFvs or constructs of the invention can be biosynthesized, purified, and formulated for administration by well-known methods. For example, an appropriate host cell, such as HEK 293 or CHO, is either transiently or stably transfected with an expression system. Vectors suitable for expression and secretion are well-known. Following expression and secretion, the medium is clarified to remove cells and the clarified medium is purified using any of many commonly-used techniques. For example, the medium may be applied to a Protein A or G column that has been equilibrated with a buffer, such as phosphate buffered saline (pH 7.4). The column is washed to remove nonspecific binding components. The bound protein is eluted, for example, by a pH gradient (such as 0.1 M sodium phosphate buffer pH 6.8 to 0.1 M sodium citrate buffer pH 2.5). Proteins are detected, such as by SDS-PAGE, and then are pooled. Further purification is optional, depending on the intended use. The protein may be concentrated and/or sterile filtered using common techniques. Other materials than the protein, such as host cell and growth medium components, and soluble aggregates and multimers of the scFv or construct, may be effectively reduced or removed by common techniques, including size exclusion, hydrophobic interaction, cation exchange, anion exchange, affinity, or hydroxyapatite chromatography. The purity of the protein after these chromatography steps is typically greater than 95%. The product may be frozen at -70 °C or may be lyophilized.

[0070] Vector DNA can be introduced into prokaryotic or eukaryotic cells via conventional transformation or transfection techniques. For stable transfection of mammalian cells, it is known that, depending upon the expression vector and transfection technique used, only a small fraction of cells may integrate the foreign DNA into their genome. In order to identify and select these integrants, a gene that encodes a selectable marker (e.g., for resistance to antibiotics) is generally introduced into the host cells along with the gene of interest. Additional selectable markers include those which confer resistance to drugs, such as G418, hygromycin and methotrexate. Cells stably transfected with the introduced nucleic acid can be identified by drug selection (e.g., cells that have incorporated the selectable marker gene will survive, while the other cells die), among other methods.

[0071] A polynucleotide encoding an amino acid sequence of a scFv or construct of the present invention can be any length as appropriate for the desired use or function, and can comprise one or more additional sequences, for example, regulatory sequences, and/or be part of a larger nucleic acid, for example, a vector. The skilled artisan will appreciate that, due to the degeneracy of the genetic code, each of the polypeptide sequences disclosed herein is encoded by a large number of other nucleic acid sequences. Mutations can also be introduced into a nucleic acid without significantly altering the biological activity of a polypeptide that it encodes. For example, one can make nucleotide substitutions leading to amino acid substitutions at nonessential amino acid residues.

[0072] Transformed cells can be cultured under conditions that promote expression of the polypeptide, and the polypeptide recovered by conventional protein purification procedures. [0073] Polypeptides contemplated for use herein include substantially homogeneous recombinant mammalian polypeptides substantially free of contaminating endogenous materials. Cells containing the nucleic acid encoding the molecules of the present invention also include hybridomas.

[0074] In certain embodiments, a vector comprising a nucleic acid molecule as described herein is provided. In certain embodiments, the invention comprises a host cell comprising a nucleic acid molecule as described herein. In certain embodiments, a nucleic acid molecule encoding a scFv or construct as described herein is provided. In certain embodiments, a pharmaceutical composition comprising at least one scFv or construct described herein is provided.

[0075] Glutaminyl and asparaginyl residues are frequently deamidated to the corresponding glutamyl and aspartyl residues, respectively. Alternatively, these residues are deamidated under mildly acidic conditions. Either form of these residues falls within the scope of this invention.

[0076] Other modifications include hydroxylation of proline and lysine, phosphorylation of hydroxyl groups of seryl or threonyl residues, methylation of the a-amino groups of lysine, arginine, and histidine side chains (T. E. Creighton, Proteins: Structure and Molecular Properties, W. H. Freeman & Co., San Francisco, 1983, pp. 79-86), acetylation of the N-terminal amine, and amidation of any C-terminal carboxyl group.

[0077] Another type of covalent modification of the molecules included within the scope of this invention comprises altering the glycosylation pattern of the protein. As is known in the art, glycosylation patterns can depend on both the sequence of the protein (e.g., the presence or absence of particular glycosylation amino acid residues, discussed below), or the host cell or organism in which the protein is produced.

[0078] Glycosylation of polypeptides is typically either N-linked or O-linked. N-linked refers to the attachment of the carbohydrate moiety to the side chain of an asparagine residue. The tri-peptide sequences asparagine-X-serine and asparagine-X-threonine, where X is any amino acid except proline, are the recognition sequences for enzymatic attachment of the carbohydrate moiety to the asparagine side chain. Thus, the presence of either of these tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars N-acetylgalactosamine, galactose, or xylose, to a hydroxyamino acid, most commonly serine or threonine, although 5-hydroxyproline or 5- hydroxylysine may also be used.

[0079] The scFvs and constructs of the present invention may be used to treat a disease or disorder in a patient. It is envisioned that said scFvs and constructs can be useful in the treatment of a disease or disorder in which binding at least one target and eliciting a biological effect is beneficial.

[0080] As used interchangeably herein, "treatment" and/or "treating" and/or "treat" are intended to refer to all processes wherein there may be a slowing, interrupting, arresting, controlling, stopping, or reversing of the progression of the disorders described herein, but does not necessarily indicate a total elimination of all disorder symptoms. Treatment includes administration of an scFv or construct of the present invention for treatment of a disease or condition in a human that would benefit from activity of an scFv or construct of the present invention, and includes: (a) inhibiting further progression of the disease; and (b) relieving the disease, i.e., causing regression of the disease or disorder or alleviating symptoms or complications thereof.

[0081] Therapeutically effective amounts (or dose) of an scFv or construct of the present invention can be administered. The amount of an scFv or construct that constitutes a therapeutically dose may vary with the indication treated, the weight of the patient, the calculated skin surface area of the patient. Dosing of an scFv or construct can be adjusted to achieve the desired effects. In many cases, repeated dosing may be required. Dosages and the frequency of administration may vary according to such factors as the route of administration, the particular an scFv or construct employed, the nature and severity of the disease to be treated, whether the condition is acute or chronic, and the size and general condition of the subject.

[0082] An scFv or construct, or a pharmaceutical composition containing such a molecule, can be administered by any feasible method. Protein therapeutics will ordinarily be administered by a parenteral route, for example by injection, since oral administration, in the absence of some special formulation or circumstance, would lead to hydrolysis of the protein in the acid environment of the stomach. Subcutaneous, intramuscular, intravenous, intraarterial, intralesional, or peritoneal bolus injection are possible routes of administration. An scFv or construct be administered via infusion, for example intravenous or subcutaneous infusion. [0083] scFv and constructs can be administered in the form of a composition comprising one or more additional components such as a physiologically acceptable carrier, excipient or diluent. Optionally, the composition additionally comprises one or more physiologically active agents. In various particular embodiments, the composition comprises one, two, three, four, five, or six physiologically active agents in addition to one or more scFv or construct.

[0084] As used herein, an “effective amount” means the amount of an scFv or construct of the present invention or pharmaceutical composition comprising such an scFv or construct that will elicit the biological or medical response of or desired therapeutic effect on a tissue, system, animal, mammal, or human that is being sought by the researcher, medical doctor, or other clinician. An effective amount of the scFv or construct may vary according to factors such as the disease state, age, sex, and weight of the individual, and the ability of the scFv or construct to elicit a desired response in the individual. An effective amount is also one in which any toxic or detrimental effect of the scFv or construct is outweighed by the therapeutically beneficial effects. Such benefit includes improving signs or symptoms of inflammatory disease(s). An effective amount can be readily determined by one skilled in the art, by the use of known techniques, and by observing results obtained under analogous circumstances. An effective amount of an scFv or construct of the present invention may be administered in a single dose or in multiple doses. In determining the effective amount for a patient, a number of factors are considered by the attending medical practitioner, including, but not limited to: the patient's size (e.g., weight or mass), body surface area, age, and general health; the specific disease or disorder involved; the degree of, or involvement, or the severity of the disease or disorder; the response of the individual patient; the particular compound administered; the mode of administration; the bioavailability characteristics of the preparation administered; the dose regimen selected; the use of concomitant medication; and other relevant circumstances known to medical practitioners.

EXAMPLES

EXAMPLE 1: DESIGN OF CYS-CLAMPS

[0085] To design cys-clamps between the scFv linker and the VH or VL domains, a crystal structure of CD3 binder 1 was used as the input structure for in silico design. The scFv linker, which is missing in the crystal structure and has the sequence (G4Q)3, was first modeled in using Rosetta Remodel (Rosetta version 3.12). One thousand models of the CD3 binder 1 scFv including the (G4Q)3 linker for each the VH-linker-VL (HL) and the VL-linker-VH (LH) scFv orientations were produced. From these 2000 models, which had diverse scFv linker conformations, the Rosetta Disulfidize mover was used to search for residue pairs between the scFv linker and the VH or VL domains that could satisfy the distance and geometry constraints of a disulfide bond when mutated to cysteine (Figure 2).

[0086] Overall, 12 designed scFv cys-clamps utilizing the scFv linker were selected, including 6 in the HL orientation and 6 scFv in the LH orientation. Cys-clamps involving the most frequently found VH or VL positions in our in silico disulfide scan across the two sets of 1000 models were prioritized, and structurally diverse positions across the VH or VL domains were selected. Cys-clamps that involved mutating a proline residue were deprioritized. Sequences of anti-CD3 scFvs (without the cysteines used to form the disulfides) in the HL orientation are SEQ ID NO: 1-4, and an CD3-binder-l_scFv_(G4Q)3 in the LH orientation is SEQ ID NO: 38. The four cysteines in each of these sequences form native disulfides, one in VH and one in VL. The designed VH-to-VL cys-clamps or linker cys-clamps as disclosed herein are additional to these cysteines. For example, amino acid sequences of scFvs in the Examples are given by SEQ ID NO: 3 or SEQ ID NO: 4, with the addition of a cysteine in the linker and in the VH or VL. SEQ ID NO: 31-37 and 39-45 provide amino acid sequences for the CD3-binder-l_scFv_(G4Q)3_HL of SEQ ID NO: 3 with additional cysteines of the present invention. SEQ ID NOs 46-60 provide amino acid sequences for the CD3-binder- 2_scFv_(G4Q)3 with additional cysteines of the present invention. SEQ ID NOs 61-87 provide amino acid sequences of CD3 binders with N-terminal linkers.

EXAMPLE 2: YIELD, AGGREGATION, AND MELTING TEMPERATURES

[0087] The 12 scFv linker cys-clamps were tested in T cell engager (TCE) molecules with either CD3 binder 1 or CD3 binder 2 as cys-clamped scFvs in two different formats, format 1 or format 2, and targeting the either Target 1 or Target 5 as the TAA (Figure 3). We also tested the cys-clamps in multispecific antibodies with a Target 6 scFv and a Target 7 binder.

[0088] In addition, controls were included with no cys-clamp as well as controls with a VH-to-VL cys-clamp between residues 103C in VH and 43 C in VL for CD3 binder scFvs or 44C in VH and 100C in VL for Target 6 scFvs. The molecules were cloned into pBMVl. l or pBMV2.1 mammalian expression vectors. HCs for TCEs were constructed in IgGl with SEFL2.2 (R309C/N314G/V321C) and YTE (M265Y/S267T/T269E) mutations in the CH2 domains. The HCs in the format 1 molecules additionally contained charge pair mutations in the CH3 domains (E377K, D427K in the scFv-containing HC, and K420D, K440D, K470D in the Target 1 Fab-containing HC). After sequence confirmation, DNA was prepared using Maxi plasmid purification kits, and plasmids were mixed at mass-based ratios according to the chain counts for each molecule. Format 3 multispecific antibodies were constructed in IgGl with SEFL2.2 (R309C/N314G/V321C) and LALA (L247A/L248A) mutations in the CH2 domains/ [0089] All molecules were expressed in a CHO cell expression hosts at IL (Figure 3 A, format 1) or 40 mL (Figure 3B, format 2; Figure 3C, format 1 and 3) scale after transfection via Lipofectamine LTX (Gibco #15338-100). After selection and scaling of cultures, cultured media was harvested from the cell cultures after 7-10 days of production. Protein A affinity purification was used to capture the proteins (Mab Select SuRe) followed by cation exchange chromatography (CEX)(Capto SP ImpRes). The samples were then dialyzed into sodium acetate buffer (10 mM NaOAc, 9% sucrose, pH 5.2) and concentrated to 70 mg/mL (Figure 3A, 3C) or 5 mg/mL (Figure 3B). The purified samples were analyzed via size exclusion chromatography (SEC)(Waters, ACQUITY UPLC Protein BEH 200 A, 1.7 um, 4.6x300 mm (Figure 3A) or Sepax Zenix-C 300 A, 3 um, 4.6x300mm (Figure 3B, 3C)) to assess sample purity and the ratios of aggregate species and desired species. The samples were also analyzed via hydrophobic interaction chromatography (HIC)(YMC BioPro HIC HT, 2.3 um, 4.6x3.5 cm).

[0090] As shown in Figure 3, the VH 103C/VL 43C cys-clamp variants showed varied effects on yield compared to non-cys-clamped controls. The 103C/43C cys-clamp variants showed similar or higher yields than the non-cys-clamped controls for CD3 binder 1 and reduced yield for CD3 binder 2 for format 1 molecules with HL scFvs, while the 103 C/43 C cys-clamp variants of either CD3 binder had more significantly reduced yields in format 2 molecules. Yields were also reduced for format 1 molecules with Target 5 binders and LH CD3 scFvs containing the 103C/43C cys-clamp (Figure 3C). The VH-to-VL cys-clamp for non-CD3 binders, VH 44C/VL 100C, reduced yield for the format 3 molecule with an LH scFv but had minimal impact on yield for the format 3 molecule with an HL scFv (Figure 3C).

[0091] Format 2 TCE variants containing HL scFvs with linker cys-clamps 62C, 83C, and 84C, or LH scFvs with linker cys-clamps 57C and 101C all showed similar or improved yields compared to the non-cys-clamped controls, as did format 1 variants containing HL scFvs with the 62C, 83C, and 85C linker cys-clamp or LH scFvs with 45C, 81C, and 83C linker cys- clamps. Format 3 molecules with an HL scFv and 85C cys-clamp or an LH scFv and 83C cys- clamp showed similar yields compared to the non-cys-clamped control and improved yields compared to the VH-to-VL cys-clamped molecules. The VH-to-VL ( 103 C/43 C or 44C/100C) cys-clamp reduced initial aggregation levels compared to the non-cys-clamped controls for all format 1, format 2, and format 3 molecules with the exception of format 1 Target 1 TCEs with LH scFvs where the 103C/43C increased aggregation levels by 1%. Most linker cys-clamps reduced aggregation levels compared to non-cysclamped controls, including 61C, 62C, 83C, 84C, and 85C in molecules with HL scFvs, and 42C, 45C, 57C, 81C, and 83C in molecules with LH scFvs.

[0092] For each of these cys-clamped molecule variants, aggregation levels showed no to little change (0-3%) after incubation for 1 week at 50°C (Figure 3 A, 3B) or incubation for 2 weeks at 40°C (Figure 3C), with the exception of the Target 1 format 1 TCE containing an HL scFv with linker cys-clamp 13C, which showed large amounts of precipitation after heated incubation (Figure 3B).

[0093] The 103C/43C cys-clamp decreased melting temperatures (Tm) for most TCEs compared to the non-cys-clamped controls except for format 1 TCEs with HL scFvs using CD3 binder 1, where Tms were slightly increased by about 1°C. All format 2 linker cys-clamp variants exhibited increased Tms by 1-5°C compared to the non-cysclamped controls with the exception of the LH 101C cys-clamp. Format 1 TCEs and format 3 antibodies containing HL scFvs with 62C, 62C, or 85C linker cys-clamps had increased Tms of about 5°C compared to the non-cysclamped controls, and those containing LH scFvs with 42C, 45C, 81C, or 83C cys- clamps each exhibited increased Tms of 1-5°C compared to non-cys-clamped controls. The 103 C/43 C cys-clamp increased HIC retention time across all TCE molecules compared to the non-cys-clamped controls except for in the format 2 TCE with CD3 binder 2 where retention time was slightly decreased. Format 2 TCEs containing HL scFvs with linker cys-clamps 61C, 62C, 83C, and 84C or LH scFv with the 57C linker cys-clamp, and format 1 TCEs containing an HL scFv with linker cys-clamp 61C, 62C, 83C, 84C, or 85C, or LH scFvs with linker cys- clamps 42C, 45C, 57C, 81C, or 83C all had HIC retention times that were unchanged or slightly reduced compared to the non-cys-clamped controls. HIC retention times format 3 molecules were similar across all non-cys-clamped controls and cys-clamp variants.

[0094] Overall, the linker cys-clamps that showed the most significant improvements across all tested biophysical properties (yield, aggregation, thermal stability, hydrophobicity) were 83C and 85C in HL CD3 scFvs and 81C and 83C in LH scFvs.

EXAMPLE 3: BINDING AFFINITY AND ACTIVITY

[0095] A subset of molecules with various cys-clamps were tested to determine the effects of different cys-clamps on antigen binding affinity and on T-cell dependent cellular cytotoxicity (TDCC) activity. For TCE molecules in format 1 and containing CD3 binder 1 scFvs, the binding affinities (KD equilibrium dissociation constants) were measured using a Sartorius Octet HTX instrument with streptavidin SAX biosensor tips with a biotinylated (1 biotin/molecule), polyclonal capture antibody (Jackson ImmunoResearch, catalog# 109-005- 098). The TCE molecules were captured on the biosensor tips and then incubated with a 6-point dilution series of soluble human CD3de antigen (ACROBiosystems, catalog# 50-201-9702). Experiments were run using the 96-tip mode with 0.3 Hz data acquisition rate at 27°C and 1000 RPM flow rate. Data was processed with Genedata Screener 19 software and globally fit to a 1 : 1 binding model to determine the association rate constant (ka) and the dissociation rate constant (kd). The equilibrium dissociation constant (KD) was then calculated as a ratio of kd/ka.

[0096] TDCC activity was also measured for TCE molecules in format 1 and format 2 containing CD3 binder 1 scFvs. Primary human T-cells were thawed and washed 3X in sterile 10% ImmunoCult T-Cell Expansion Medium (ICM)(RPME10%FBS/lx GlutaMAX/HEPES/NEAA/NaPyr/2-Mercaptoethanol) and then re-suspended in ICM at 30,000 cells/20 mL. Target cells were also washed 3X and resuspended in ICM at 3000 cells/20 mL. 20 mL T-cells were aliquoted into 384-well tissue culture plates, then 20 mL of target cells were added. 10 mL of TCE molecules were added to a final concentration range of 2.4E-04 nM-1000 nM, then the cells were incubated for 24 hours at 37°C and 5% carbon dioxide. Following 24 hour incubation, %ly sis of target cells was measured using BioGio Luciferase assay (Promega, catalog# G7940).

[0097] As shown in Figure 4A, the VH-to-VL cys-clamp (103C/43C) reduced binding affinity of the format 1 TCE molecule containing a CD3 binder 1 scFv in HL orientation by nearly two-fold compared to the no cys-clamp control, whereas the same cys-clamp increased binding affinity of the format 1 TCE molecule containing a CD3 binder 1 scFv in LH orientation by two-fold. All linker cys-clamp variants (HL 13C, HL 85C, LH 42C, LH 45C, LH 81C, and LH 85C) maintained binding affinity within a 0.2-0.8 nM range of the respective no cys-clamp controls. HL 13C, HL 85C, LH 42C, LH 45C, and LH 85C cys-clamp variants showed slightly increased affinities compared to the respective no cys-clamp controls, while the LH 81C variant exhibited reduced binding affinity by 0.5 nM compared to the no cys-clamp control. Similarly, the VH-to-VL cys-clamp (103C/43C) reduced TDCC activity in format 2 TCE with CD3 binder 1 scFv in HL orientation, while HL linker cys-clamps 61C, 62C, and 83C exhibited EC50 values within 0.1-0.25 nM of the no cys-clamp control (Figure 4B). In format 1 TCEs with CD3 binder 1 scFv in HL orientation, the VH-to-VL cys-clamp also significantly increased the EC50 for TDCC activity compared to the no cys-clamp control, but had little effect on TDCC activity in the same molecule format with the scFv in LH orientation. Linker cys-clamps in the HL orientation (13C and 85C) increased TDCC activity and reduced EC50 compared to the no cys-clamp control, and linker cys-clamps in the LH orientation (42C, 45C, 81C, 83C) had insignificant effect on TDCC activity compared to the no cys-clamp control. Overall, all scFv linker cys-clamps showed no large detrimental effects on antigen binding or on TDCC activity in the molecule formats and CD3 binder tested.

EXAMPLE 4: PHARMACOKINETICS

[0098] The effects of scFv cys-clamp variants on the pharmacokinetic (PK) behavior of a subset of TCE molecules was determined. Endotoxin levels were measured for samples of each TCE molecule on a Charles River Endosafe MCS instrument to ensure that endotoxin levels were less than 1 Eu/mg. A single-dose PK study was conducted using male huFcRn transgenic line 32 homozygous (huFcRn Tg32) mice, which were dosed via intravenous bolus injection at 1 mg/kg. Serum samples were collected from each animal at 6, 24, 48, 72, 120, 168, 240, 336, 504, 672, 840, and 1008 hours, and levels of TCE molecules in serum samples were determined via capture with a CD3 anti-idiotype antibody followed by detection with an antiHuman Fc antibody.

[0099] As shown in Figure 5, TCE molecules containing CD3 binder 1 with no cys- clamp or the VH-to-VL cys-clamp (Figure 5A) showed faster clearance in huFcRn Tg32 mice than those containing CD3 binder 2 with the same cys-clamp variants (Figure 5B). For TCE molecules containing either CD3 binder, and in either format 1 or format 2, addition of the VH- to-VL cys-clamp (103C/43C) significantly increased rate of clearance compared to the no cys- clamp controls. In contrast, addition of any scFv linker cys-clamps (HL 83C, HL 61C, HL 62C, HL 84C, LH 57C) had no effect on rate of clearance compared to the no cys-clamp controls. Therefore TCE molecules containing scFvs with linker-based cys-clamps exhibited superior pharmacokinetic behavior compared to those containing scFvs with the VH-to-VL 103 C/43 C cys-clamp.

EXAMPLE 5: CRYSTAL STRUCTURES

[00100] To confirm the formation of disulfide bonds for designed cys-clamps, crystal structures were determined of CD3 binder 2 with a VH-to-VL cys-clamp ( 103 C/43 C) or scFv linker cys-clamps (HL 72C, HL 73C, LH 57C). Cys-clamp variants of the CD3 binder 2 scFv were cloned with a C-terminal 6xHis tag into pBMV2.1 mammalian expression vectors via Golden Gate assembly of DNA fragments synthesized by Twist Bioscience. After sequence confirmation, DNA was prepared using Maxi plasmid purification kits. The scFv molecules were expressed in a CHO cell expression hosts at 250 mL scale after transfection using Lipofectamine LTX (Gibco #15338-100). After selection and scaling of cultures, cultured media was harvested from the cell cultures after 7-10 days of production. Nickel affinity purification was used to capture the proteins (HisTrap) followed by size exclusion chromatography (SEC)(HiLoad Superdex 75 26/60). The samples were concentrated to 15 mg/mL and dialyzed into 20 mM HEPES, 0.15M NaCl, pH 7.5. Protein crystals were produced using hanging drop vapor diffusion methods, and all X-ray diffraction data were collected using a CLS Synchotron BM as the X-ray source and Pilatus3 as the detector. Refinement was performed using Phenix-phaser software. Resolution ranged from 1.49-1.95 A for the determined structures. As shown in Figure 6, the VH-to-VL cys-clamp disulfide, as well as linker cys-clamp disulfides in both HL and LH scFvs were ordered, indicating that the designed cys-clamps were properly formed. In addition, presence of linker cys-clamps allowed 3-6 residues of the (G4Q)3 scFv linker residues to be resolved in structures with linker cys-clamped scFvs, which were missing density in the VH-to-VL cys-clamp structure.

EXAMPLE 6: FURTHER CHARACTERIZATION

[00101] Whether the cys-clamp variants could improve biophysical properties across a diverse set of Fvs was determined. A set of 20 diverse monoclonal antibodies (mAbs) against Target 2 was used, including 10 with kappa (k) light chains and 10 with lambda (1) light chains, and converted the variable fragment (Fv) regions of these mAbs into scFvs with no cys-clamp, with a VH-to-VL cys-clamp, or with scFv linker cys-clamps. These scFv cys-clamp variants in a symmetric scFv-Fc format using an IgGl Fc backbone IgGl with SEFL2.2 (R309C/N314G/V321C) and YTE (M265Y/S267T/T269E) mutations in the CH2 domains were produced. All molecules were cloned into a pBMV2.1 mammalian expression vector via Golden Gate assembly of DNA fragments synthesized by Twist Bioscience. After sequence confirmation, DNA was prepared using Maxi plasmid purification kits. The molecules were then expressed using a stable CHO cell expression system at 4 mL scale using a fed-batch process. Cells were transfected via Lipofectamine LTX (Gibco #15338-100), and after selection and scaling of cultures, cultured media was harvested from the cell cultures after 7-10 days of production. Proteins were captured via Protein A affinity purification (GenScript Protein A MagBeads), normalized to 1-2 mg/mL, and then dialyzed into sodium acetate buffer (10 mM NaOAc, 9% sucrose, pH 5.2). The purified samples were analyzed via size exclusion chromatography (SEC)(Sepax Zenix-C 300 A, 3 um, 4.6x300mm) to assess sample purity and the ratios of aggregate species and desired species. The samples were also analyzed via hydrophobic interaction chromatography (HIC)(YMC BioPro HIC HT, 2.3 um, 4.6x3.5 cm). Binding affinities for select molecules were measured using a SAHC30M sensor chip (Carterra), which was loaded with an anti-human Fc capture antibody for 10 minutes. Test scFv-Fc molecules were then loaded for 10 minutes at a concentration range of 0.4-0.6 mg/mL. The target analyte was prepared in a six point 1 :3 dilution series. All data were fitted using the Kinetics software suite (Carterra) with a 1 : 1 Langmuir binding model.

[00102] As shown in Figure 7 A, addition of a VH-to-VL cys-clamp (VH 44C/ VL 100C) reduced yields for 17 out of 20 converted scFvs in the HL orientation and for 17 out of 20 converted scFvs in the LH orientation compared to the no cys-clamp control. In addition, while the VH-to-VL cys-clamp reduced aggregation levels for 5 scFv-Fcs in the HL orientation and 6 scFv-Fcs in the LH orientation, addition of the VH-to-VL cys-clamp increased levels of aggregation for most converted scFvs (Figure 7B). Overall, linker cys-clamps performed better than the VH-to-VL cys-clamp at successfully converting Fvs into scFvs by reducing aggregation without negatively impacting yield compared to the no cys-clamp controls. The best performing linker cys-clamps in each scFv orientation across the 20 mAb panel were the 85C linker cys-clamp in the HL scFv orientation and the 83C linker cys-clamp in the LH scFv orientation. 18 of 20 scFv-Fcs with the HL 85C linker cys-clamp produced with yields within 80% of the no cys-clamp control, and 14 of 19 HL scFv-Fcs where the no cys-clamp control produced enough yield for aggregation analysis showed reduced aggregation upon addition of the 85C linker cys-clamp compared to the no cys-clamp control. 9 of those 14 scFvs showed improvement in aggregation levels by more than 5%, compared to only 5 scFvs that showed 5% or greater improvement using the VH-to-VL cys-clamp (Table 1). In addition, yields of scFv-Fcs in the LH orientation were also largely unimpacted by addition of the 83C linker cys- clamp, where 15 of 20 molecules were produced at yields within 80% of the no cys-clamp control, compared to only 5 of 20 molecules for LH scFv-Fcs with the VH-to-VL cys-clamp. The LH 83C linker cys-clamp improved aggregation levels for 12 of 18 scFv-Fcs in the LH orientation compared to the no cys-clamp control, with 8 of those being improved by more than 5% (Table 1).

Table 1. Numbers of scFv-Fcs with unimpacted yields and improved aggregation levels across

20 Target 2 scFv-Fcs. [00103] The proportion of scFv-Fcs that showed yield within 80% of the no cys-clamp controls are shown for the VH-to-VL cys-clamp and the best performing linker cys-clamps in both HL and LH scFv orientations. The proportion of scFv-Fcs that showed reduced aggregation levels by greater than 0% or by greater than 5% compared to the no cys-clamp controls are shown for the same cys-clamp variants. The total number of scFv-Fcs tallied for aggregation level analysis (denominator) represent the total number of scFv-Fcs where the no cys-clamp control had sufficient yield for the aggregation assay.

[00104] As shown in Figure 8, addition of VH-to-VL cys-clamps or linker cys-clamps had no effect on or slightly reduced HIC retention times for scFv-Fcs in both HL and LH orientations compared to the no cys-clamp controls, with the exception of few converted scFvs where the HIC retention times were increased upon addition of cys-clamps. Binding affinities were also generally unaffected by the addition cys-clamps compared to the no cys-clamp control, however, two converted scFvs showed significantly increased KD’s by an order of magnitude after addition of the VH-to-VL or select linker cys-clamps in the HL orientation (Figure 8B). Overall, linker cys-clamps improved hydrophobicity for most Target 2 binders as shown by reduced HIC retention times of the scFv-Fc linker cys-clamp variants while having negligible effects on binding.

EXAMPLE 7: N-TERMINAL LINKER

[00105] The crystal structures of CD3 binder 2 scFvs with linker cys-clamps (Figure 6) revealed that structurally similar cys-clamps could be produced by translating the scFv linker and C-terminal VL or VH domain to the N-terminal portion of the scFv and reconnecting the VH and VL domains with another (G4Q)3 linker as illustrated in Figure 9. In this way, the cys- clamp disulfide and VH and VL conformations would be identical in three-dimensional space as the original molecule despite the relocation of the cysteine-containing linker to the N- terminal of the scFv and the swapping of the VH and VL domains in sequence. To test these N- terminal linker scFv cys-clamp designs, the same VH or VL cysteine positions characterized previously for scFv linker cys-clamps and designed N-terminal sequences with cysteines that were spatially similar to previous linker cys-clamps were used. Additional adjacent cysteines in the N-terminal linker were assessed as shown in Table 2.

Table 2, Summary of N-terminal linker scFv cys-clamp designs.

[00106] These N-terminal linker scFv cys-clamp designs were implemented in two molecules with two different formats: 1) a TCE in format 1 containing CD3 binder 1 scFv and Target 1 binder, and 2) a symmetric Fab-scFv-Fc format, where a Target 3 scFv is buried between the C-terminus of a Target 4 Fab and the Fc. These molecules were produced with no cys-clamp, the VH-to-VL cys-clamp, and N-terminal linker cys-clamps using an IgGl Fc backbone IgGl with SEFL2.2 (R309C/N314G/V321C) mutations in the CH2 domains. All molecules were cloned into a pBMV2.1 mammalian expression vector via Golden Gate assembly of DNA fragments synthesized by Twist Bioscience. After sequence confirmation, DNA was prepared using Maxi plasmid purification kits. The molecules were then expressed using a stable CHO cell expression system at 4 mL scale using a fed-batch process. Cells were transfected via Lipofectamine LTX (Gibco #15338-100), and after selection and scaling of cultures, cultured media was harvested from the cell cultures after 7-10 days of production. Proteins were captured via Protein A affinity purification (GenScript Protein A MagBeads), followed by cation exchange (CEX)(Capto SP ImpRes), then dialyzed into sodium acetate buffer (10 mM NaOAc, 9% sucrose, pH 5.2). The purified samples were analyzed via size exclusion chromatography (SEC)(Sepax Zenix-C 300 A, 3 um, 4.6x300mm) to assess sample purity and the ratios of aggregate species and desired species. The samples were also analyzed via hydrophobic interaction chromatography (HIC)(YMC BioPro HIC HT, 2.3 um, 4.6x3.5 cm).

[00107] As shown in Figure 10, N-terminal linker cys-clamp variants HL 42C/-8C, LH 62C/-4C, and LH 85C/-4C were the best performing N-terminal linker cys-clamps. While HL 42C/-8C slightly decreased yields compared to the no cys-clamp controls, aggregation was significantly reduced for the format 1 molecule and slightly reduced for the symmetric Fab- scFv-Fc molecule compared to the no cys-clamp control, which had relatively low levels of aggregation to start. LH 62C/-4C and LH 85C/-4C showed yields similar to the no cys-clamp controls but higher yields than the VH-to-VL cys-clamp variant for both format 1 and Fab- scFv-Fc molecules. Both LH 62C/-4C and LH 85C/-4C s N-terminal linker cys-clamps reduced aggregation levels for the Fab-scFv-Fc, whereas only LH 62C/-4C showed improved aggregation levels for the Format 1 as well compared to the no cys-clamp control. The LH 85C/-4C variant showed significantly reduced aggregation levels for the Fab-scFv-Fc molecule, but slightly increased aggregation levels in the Format 1 molecule compared to the no cys- clamp control. Overall, N-terminal linker cys-clamp variants HL 42C/-8C, LH 62C/-4C, and LH 85C/-4C showed improved biophysical properties compared to the VH-to-VL cys-clamp in both Format 1 and Fab-scFv-Fc molecules in both HL and LH scFv orientations.

[00108] Several N-terminal linker cys-clamp variants were characterized in format 1 TCEs with Target 5 binders as well as format 3 multispecific antibodies with Target 6 scFv binders and Target 7 binders. Molecules were produced as described in Example 2. As shown in Figure 11, the HL 42C/-8C and LH 61C/-2C N-terminal linker cys-clamps improved Tms for the TCE molecules by about 5°C compared to the non-cys-clamped controls, improved initial aggregation levels, and maintained aggregation levels after 2 week incubation at 40°C at below 2%. These N-terminal linker cys-clamps also maintained or reduced HIC retention times for TCE molecules compared to the non-cys-clamped controls, whereas the VH/VL cys-clamp increased HIC retention times. The HL 42C/-8C cys-clamp improved TCE yield compared to the non-cys-clamped control, while the LH 61C/-2C significantly reduced yield for the CD3 binder 1 -containing TCE. Biophysical properties for format 3 multispecific antibodies were improved by addition of the LH 62C/-4C N-terminal linker cys-clamp, including increased Tm by 5°C, and low aggregation levels both immediately after purification and after 2 week incubation at 40°C. However the yield was modestly impacted, and the HIC retention time was increased compared to the non-cys-clamped control.

SEQUENCES

SEQ ID NO: 1 CD3-binder-l_scFv_(G4S)3_HL

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYISYWAYW

GQGTLVTVSSGGGGSGGGGSGGGGSQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNY

PNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 2 CD3-binder-2_scFv_(G4S)3_HL

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYISYFAYWG

QGTLVTVSSGGGGSGGGGSGGGGSQTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNYP

NWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLLGGKAALTLSGVQPEDEAEYYCVLW

YSNRWVFGSGTKLTVL

SEQ ID NO: 3 CD3-binder-l_scFv_(G4Q)3_HL

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYISYWAYW

GQGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGN

YPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 4 CD3-binder-2_scFv_(G4Q)3_HL

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYISYFAYWG

QGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNY

PNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLLGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 5 Leader sequence

MDMRVPAQLLGLLLLWLRGARC

SEQ ID NO: 6 Nucleic acid encoding SEQ ID NO: 5 atggacatgagagtgcctgcacagctgctgggcctgctgctgctgtggctgagaggcgccagatgc

SEQ ID NO: 7 Leader sequence

MAWALLLLTLLTQGTGSWA

SEQ ID NO: 8 Nucleic acid encoding SEQ ID NO: 7 atggcctggg ctctgctgct cctcaccctc ctcactcagg gcacagggtc ctgggcc

(Gly₃Ser)₃ (SEQ ID NO: 9)

GGGSGGGSGGGS

(Gly₄Ser)₃ (SEQ ID NO: 10)

GGGGSGGGGSGGGGS

(Gly₃Ser)₄ (SEQ ID NO: 11)

GGGSGGGSGGGSGGGS (Gly₄Ser)₄ (SEQ ID NO: 12)

GGGGSGGGGSGGGGSGGGGS

(Gly₃Ser)₅ (SEQ ID NO: 13)

GGGSGGGSGGGSGGGSGGGS

(Gly₃Ser)₆ (SEQ ID NO: 14)

GGGSGGGSGGGSGGGSGGGSGGGS

(Gly₄Ser)₆ (SEQ ID NO: 15)

GGGGS GGGGSGGGGSGGGGSGGGGSGGGGS

GSADDAKKDAAKKDAAKKDDAKKDDAGS (SEQ ID NO: 16)

GSADDAKKDAAKKDAAKKDDAKKDDAKKDAGS (SEQ ID NO: 17)

(Gly₃Gln)₂ (SEQ ID NO: 18)

GGGQGGGQ

(Gly₄Gln)₂ (SEQ ID NO: 19)

GGGGQGGGGQ

(Gly₃Gln)₃ (SEQ ID NO: 20)

GGGQGGGQGGGQ

(Gly₄Gln)₃ (SEQ ID NO: 21)

GGGGQGGGGQGGGGQ

(Gly₃Gln)₄ (SEQ ID NO: 22)

GGGQGGGQGGGQGGGQ

(Gly₄Gln)₄ (SEQ ID NO: 23)

GGGGQGGGGQGGGGQGGGGQ

(Gly₃Gln)₅ (SEQ ID NO: 24)

GGGQGGGQGGGQGGGQGGGQ

(Gly₄Gln)₅ (SEQ ID NO: 25)

GGGGQGGGGQGGGGQGGGGQGGGGQ

(Gly₃Gln)₆ (SEQ ID NO: 26)

GGGQGGGQGGGQGGGQGGGQGGGQ

(Gly₄Gln)₆ (SEQ ID NO: 27)

GGGGQGGGGQGGGGQGGGGQGGGGQGGGGQ

(Gly₃Ser)₂ (SEQ ID NO: 28)

GGGSGGGS

(Gly₄Ser)₂ (SEQ ID NO: 29) GGGGSGGGGS

SGGGGSSGGGGS (SEQ ID NO: 30)

SEQ ID NO: 31 CD3-binder-l_scFv_(G4Q)3_HL_VH-103C_VL-43C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYISYWAYC

GQGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGN

YPNWVQKKPGQCPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 32 CD3-binder-l_scFv_(G4Q)3_HL_VH-13C_linker-3C

EVQLVESGGGLVCPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYISYWAYW

GQGTLVTVSSGGCGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGN

YPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 33 CD3-binder-l_scFv_(G4Q)3_HL_VH-61C_linker-14C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRSKYNNY

ATYYACAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYISYWAYW

GQGTLVTVSSGGGGQGGGGQGGGCQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGN

YPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 34 CD3-binder-l_scFv_(G4Q)3_HL_VH-62C_linker-14C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRSKYNNY

ATYYADCVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYISYWAYW

GQGTLVTVSSGGGGQGGGGQGGGCQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGN

YPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 35 CD3-binder-l_scFv_(G4Q)3_HL_VH-83C_linker-8C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTVYLQMNNLCTEDTAVYYCARAGNFGSSYISYWAYW

GQGTLVTVSSGGGGQGGCGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGN

YPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 36 CD3-binder-l_scFv_(G4Q)3_HL_VH-84C_linker-9C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTVYLQMNNLKCEDTAVYYCARAGNFGSSYISYWAYW

GQGTLVTVSSGGGGQGGGCQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGN

YPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 37 CD3-binder-l_scFv_(G4Q)3_HL_VH-85C_linker-10C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTVYLQMNNLKTCDTAVYYCARAGNFGSSYISYWAYW

GQGTLVTVSSGGGGQGGGGCGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGN YPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 38 CD3-binder-l_scFv_(G4Q)3_LH

QTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPGT

PARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG

GQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI S YW AYWGQGTL VT VS S

SEQ ID NO: 39 CD3-binder-l_scFv_(G4Q)3_LH_VH-103C_VL-43C

QTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQCPRGLIGGTKFLAPGT

PARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG

GQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI SYWAYCGQGTLVTVSS

SEQ ID NO: 40 CD3-binder-l_scFv_(G4Q)3_LH_VL-42C_linker-9C

QTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGCAPRGLIGGTKFLAPGT

PARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGGC

QGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIR

SKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYIS YW AYWGQGTL VTVSS

SEQ ID NO: 41 CD3-binder-l_scFv_(G4Q)3_LH_VL-45C_linker-10C

QTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPCGLIGGTKFLAPGT

PARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG

GCGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI SYW AYWGQGTL VT VS S

SEQ ID NO: 42 CD3-binder-l_scFv_(G4Q)3_LH_VL-57C_linker-13C

QTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPCT

PARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG

GQGGCGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI SYW AYWGQGTL VT VS S

SEQ ID NO: 43 CD3-binder-l_scFv_(G4Q)3_LH_VL-81C_linker-8C

QTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPGT

PARFSGSLSGGKAALTLSGVQPCDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGC

GQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI SYW AYWGQGTL VT VS S

SEQ ID NO: 44 CD3-binder-l_scFv_(G4Q)3_LH_VL-83C_linker-4C

QTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPGT

PARFSGSLSGGKAALTLSGVQPEDCAEYYCVLWYSNRWVFGSGTKLTVLGGGCQGGG

GQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI SYW AYWGQGTL VT VS S SEQ ID NO: 45 CD3-binder-l_scFv_(G4Q)3_LH_VH-101C_linker-13C

QTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPGT

PARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG

GQGGCGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI

SYWCYWGQGTLVTVSS

SEQ ID NO: 46 CD3-binder-2_scFv_(G4Q)3_HL_VH-103C_VL-43C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYISYFAYCG

QGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNY

PNWVQKKPGQCPRGLIGGTKFLAPGTPARFSGSLLGGKAALTLSGVQPEDEAEYYCVL

WYSNRWVFGSGTKLTVL

SEQ ID NO: 47 CD3-binder-2_scFv_(G4Q)3_HL_VH-13C_linker-3C

EVQLVESGGGLVCPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYISYFAYWG

QGTLVTVSSGGCGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNY

PNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLLGGKAALTLSGVQPEDEAEYYCVL WYSNRWVFGSGTKLTVL

SEQ ID NO: 48 CD3-binder-2_scFv_(G4Q)3_HL_VH-61C_linker-14C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARIRSKYNNY

ATYYACAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYISYFAYWG

QGTLVTVSSGGGGQGGGGQGGGCQQTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNY

PNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLLGGKAALTLSGVQPEDEAEYYCVL WYSNRWVFGSGTKLTVL

SEQ ID NO: 49 CD3-binder-2_scFv_(G4Q)3_HL_VH-62C_linker-14C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARIRSKYNNY

ATYYADCVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYISYFAYWG

QGTLVTVSSGGGGQGGGGQGGGCQQTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNY

PNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLLGGKAALTLSGVQPEDEAEYYCVL WYSNRWVFGSGTKLTVL

SEQ ID NO: 50 CD3-binder-2_scFv_(G4Q)3_HL_VH-83C_linker-8C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTAYLQMNNLCTEDTAVYYCVRNANFGTSYISYFAYWG

QGTLVTVSSGGGGQGGCGQGGGGQQTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNY

PNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLLGGKAALTLSGVQPEDEAEYYCVL WYSNRWVFGSGTKLTVL

SEQ ID NO: 51 CD3-binder-2_scFv_(G4Q)3_HL_VH-84C_linker-9C

EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARIRSKYNNY

ATYYADAVKDRFTISRDDSKNTAYLQMNNLKCEDTAVYYCVRNANFGTSYISYFAYW

GQGTLVTVSSGGGGQGGGCQGGGGQQTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGN

YPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLLGGKAALTLSGVQPEDEAEYYCVL WYSNRWVFGSGTKLTVL

SEQ ID NO: 52 CD3-binder-2_scFv_(G4Q)3_HL_VH-85C_linker-10C EVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARIRSKYNNY ATYYADAVKDRFTISRDDSKNTAYLQMNNLKTCDTAVYYCVRNANFGTSYISYFAYW GQGTLVTVSSGGGGQGGGGCGGGGQQTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGN

YPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLLGGKAALTLSGVQPEDEAEYYCVL WYSNRWVFGSGTKLTVL

SEQ ID NO: 53 CD3-binder-2_scFv_(G4Q)3_LH

QTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPG

TPARFSGSLLGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG GQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARI RSKYNNYATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYI S YF AYWGQGTL VT VS S

SEQ ID NO: 54 CD3-binder-2_scFv_(G4Q)3_LH_VH-103C_VL-43C

QTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNYPNWVQKKPGQCPRGLIGGTKFLAPG

TPARFSGSLLGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG GQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARI RSKYNNYATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYI S YF AYCGQGTL VT VS S

SEQ ID NO: 55 CD3-binder-2_scFv_(G4Q)3_LH_VL-42C_linker-9C

QTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNYPNWVQKKPGCAPRGLIGGTKFLAPG

TPARFSGSLLGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG CQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARI RSKYNNYATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYI SYF AYWGQGTL VT VS S

SEQ ID NO: 56 CD3-binder-2_scFv_(G4Q)3_LH_VL-45C_linker-10C

QTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNYPNWVQKKPGQAPCGLIGGTKFLAPG

TPARFSGSLLGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG GCGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARI RSKYNNYATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYI SYF AYWGQGTL VT VS S

SEQ ID NO: 57 CD3-binder-2_scFv_(G4Q)3_LH_VL-57C_linker-13C

QTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPC

TPARFSGSLLGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG GQGGCGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARI RSKYNNYATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYI SYF AYWGQGTL VT VS S

SEQ ID NO: 58 CD3-binder-2_scFv_(G4Q)3_LH_VL-81C_linker-8C

QTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPG

TPARFSGSLLGGKAALTLSGVQPCDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGC GQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARI RSKYNNYATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYI SYF AYWGQGTL VT VS S

SEQ ID NO: 59 CD3-binder-2_scFv_(G4Q)3_LH_VL-83C_linker-4C

QTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPG TPARFSGSLLGGKAALTLSGVQPEDCAEYYCVLWYSNRWVFGSGTKLTVLGGGCQGGG GQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYI S YF AYWGQGTL VT VS S

SEQ ID NO: 60 CD3-binder-2_scFv_(G4Q)3_LH_VH-101C_linker-13C

QTVVTQEPSLTVSPGGTVTLTCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPG

TPARFSGSLLGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQGGG

GQGGCGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVREAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTAYLQMNNLKTEDTAVYYCVRNANFGTSYI SYFCYWGQGTLVTVSS

SEQ ID NO: 61 CD3-binder-l_scFv_(G4Q)3_HL_VL-42C_N-term-linker-6C

GCGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI SYW AYWGQGTL VTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTG AVTSGNYPNWVQKKPGCAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEA EYYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 62 CD3-binder-l_scFv_(G4Q)3_HL_VL-42C_N-term-linker-7C

GCQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVA RIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSS YISYW AYWGQGTL VTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSS TGAVTSGNYPNWVQKKPGCAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPED EAEYYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 63 CD3-binder-l_scFv_(G4Q)3_HL_VL-42C_N-term-linker-8C

GCGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWV ARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGS SYISYW AYWGQGTL VTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGS STGAVTSGNYPNWVQKKPGCAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPE DEAEYYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 64 CD3-binder-l_scFv_(G4Q)3_HL_VL-45C_N-term-linker-5C

GCGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIR SKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYIS YW AYWGQGTL VTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTG AVTSGNYPNWVQKKPGQAPCGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEA EYYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 65 CD3-binder-l_scFv_(G4Q)3_HL_VL-45C_N-term-linker-6C

GCGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI SYW AYWGQGTL VTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTG AVTSGNYPNWVQKKPGQAPCGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEA EYYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 66 CD3-binder-l_scFv_(G4Q)3_HL_VL-45C_N-term-linker-7C

GCQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVA RIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSS YISYW AYWGQGTL VTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSS TGAVTSGNYPNWVQKKPGQAPCGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPED

EAE YYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 67 CD3-binder-l_scFv_(G4Q)3_HL_VL-57C_N-term-linker-4C

GCGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIRS

KYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYISY

WAYWGQGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGA VTSGNYPNWVQKKPGQAPRGLIGGTKFLAPCTPARFSGSLSGGKAALTLSGVQPEDEAE YYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 68 CD3-binder-l_scFv_(G4Q)3_HL_VL-57C_N-term-linker-5C

QCGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARIR

SKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYIS

YWAYWGQGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTG AVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPCTPARFSGSLSGGKAALTLSGVQPEDEA EYYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 69 CD3-binder-l_scFv_(G4Q)3_HL_VL-57C_N-term-linker-6C

GCGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEWVARI

RSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNFGSSYI

SYWAYWGQGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTG

AVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPCTPARFSGSLSGGKAALTLSGVQPEDEA EYYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 70 CD3-binder-l_scFv_(G4Q)3_HL_VL-81C_N-term-linker-9C

GCGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEW

VARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNF

GSSYISYWAYWGQGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTITC GSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQ PEDEAE YYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 71 CD3-binder-l_scFv_(G4Q)3_HL_VL-81C_N-term-linker-10C

QCGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLE

WVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGN

FGSSYISYWAYWGQGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTIT CGS STGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPGTPARF SGSLSGGKAALTLSGV QPCDEAEYYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 72 CD3-binder-l_scFv_(G4Q)3_HL_VL-81C_N-term-linker-l 1C

GCGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGL

EWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAG

NFGSSYISYWAYWGQGTLVTVSSGGGGQGGGGQGGGGQQTVVTQEPSLTVSPGGTVTI TCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFLAPGTPARFSGSLSGGKAALTLSG VQPCDEAE YYCVLWYSNRWVFGSGTKLTVL

SEQ ID NO: 73 CD3-binder-l_scFv_(G4Q)3_LH_VH-61C_N-term-linker-lC

GGCQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKFL

APGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGGQ

GGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLEW

VARIRSKYNNYATYYACAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGNF GS S YIS YW AYWGQGTL VT VS S SEQ ID NO: 74 CD3-binder-l_scFv_(G4Q)3_LH_VH-61C_N-term-linker-2C

GGCQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKF

LAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGG

QGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLE

WVARIRSKYNNYATYYACAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGN

FGSSYISYWAYWGQGTLVTVSS

SEQ ID NO: 75 CD3-binder-l_scFv_(G4Q)3_LH_VH-61C_N-term-linker-3C

GCGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKF

LAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGG

QGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLE

WVARIRSKYNNYATYYACAVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGN

FGSSYISYWAYWGQGTLVTVSS

SEQ ID NO: 76 CD3-binder-l_scFv_(G4Q)3_LH_VH-62C_N-term-linker-2C

GGCQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKF

LAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGG

QGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLE

WVARIRSKYNNYATYYADCVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGN

FGSSYISYWAYWGQGTLVTVSS

SEQ ID NO: 77 CD3-binder-l_scFv_(G4Q)3_LH_VH-62C_N-term-linker-3C

GCGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTKF

LAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGGG

QGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGLE

WVARIRSKYNNYATYYADCVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAGN

FGSSYISYWAYWGQGTLVTVSS

SEQ ID NO: 78 CD3-binder-l_scFv_(G4Q)3_LH_VH-62C_N-term-linker-4C

GCGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTK

FLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGG

GQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGL

EWVARIRSKYNNYATYYADCVKDRFTISRDDSKNTVYLQMNNLKTEDTAVYYCARAG

NFGS S YIS YW AYWGQGTL VT VS S

SEQ ID NO: 79 CD3-binder-l_scFv_(G4Q)3_LH_VH-83C_N-term-linker-8C

GCGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLI

GGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTV

LGGGGQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAP

GKGLEWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLCTEDTAVYYC

ARAGNFGSSYISYW AYWGQGTL VTVSS

SEQ ID NO: 80 CD3-binder-l_scFv_(G4Q)3_LH_VH-83C_N-term-linker-9C

GCGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRG

LIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLT

VLGGGGQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQA

PGKGLEWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLCTEDTAVYY

C ARAGNFGS S YIS YW AYWGQGTL VT VS S

SEQ ID NO: 81 CD3-binder-l_scFv_(G4Q)3_LH_VH-83C_N-term-linker-10C QCGGGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPR GLIGGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKL TVLGGGGQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQ APGKGLEWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLCTEDTAVY YC ARAGNFGS S YIS YW AYWGQGTL VT VS S

SEQ ID NO: 82 CD3-binder-l_scFv_(G4Q)3_LH_VH-84C_N-term-linker-6C

GCGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGG TKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLG GGGQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGK GLEWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKCEDTAVYYCAR AGNFGS S YIS YW AYWGQGTL VT VS S

SEQ ID NO: 83 CD3-binder-l_scFv_(G4Q)3_LH_VH-84C_N-term-linker-7C

GCQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIG GTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVL GGGGQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPG KGLEWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKCEDTAVYYCA RAGNFGS S YIS YW AYWGQGTL VT VS S

SEQ ID NO: 84 CD3-binder-l_scFv_(G4Q)3_LH_VH-84C_N-term-linker-8C

GCGQGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLI GGTKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTV LGGGGQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAP GKGLEWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKCEDTAVYYC ARAGNFGSSYISYW AYWGQGTL VTVSS

SEQ ID NO: 85 CD3-binder-l_scFv_(G4Q)3_LH_VH-85C_N-term-linker-4C

GCGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGTK FLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGGG GQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKGL EWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTCDTAVYYCARAG NFGS S YIS YW AYWGQGTL VT VS S

SEQ ID NO: 86 CD3-binder-l_scFv_(G4Q)3_LH_VH-85C_N-term-linker-5C

GCGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGGT KFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLGG GGQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGKG LEWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTCDTAVYYCARA GNFGS S YIS YW AYWGQGTL VT VS S

SEQ ID NO: 87 CD3-binder-l_scFv_(G4Q)3_LH_VH-85C_N-term-linker-6C

GCGGGGQQTVVTQEPSLTVSPGGTVTITCGSSTGAVTSGNYPNWVQKKPGQAPRGLIGG TKFLAPGTPARFSGSLSGGKAALTLSGVQPEDEAEYYCVLWYSNRWVFGSGTKLTVLG GGGQGGGGQGGGGQEVQLVESGGGLVQPGGSLKLSCAASGFTFNKYAINWVRQAPGK GLEWVARIRSKYNNYATYYADAVKDRFTISRDDSKNTVYLQMNNLKTCDTAVYYCAR AGNFGS S YIS YW AYWGQGTL VT VS S

SEQ ID NO: 88 (Gly₄Ser)₅

GGGGSGGGGSGGGGSGGGGSGGGGS

Claims

CLAIMS What is claimed:

1. A single chain variable fragment (scFv), comprising a heavy chain variable region (VH), a light chain variable region (VL), a linker comprising a cysteine (linker cysteine), and wherein the linker connects the VH and VL, and: a. the scFv is in a VH-VL (HL) orientation from N-terminus to C-terminus and comprises a VH cysteine at position 13, 61, 62, 83, 84, or 85 according to Kabat numbering; or b. the scFv is in a VL-VH (LH) orientation from N-terminus to C-terminus and comprises a VL cysteine at position 42, 45, 57, 81, or 83, or a VH cysteine at position 101 according to Kabat numbering.

2. The scFv of Claim 1, wherein the linker cysteine is at a position between 1-15 according to linear numbering.

3. The scFv of Claim 1 or Claim 2, wherein the linker cysteine is at position 3, 4, 8, 9, 10, 13, or 14 according to linear numbering.

4. The scFv of any one of Claims 1-3, wherein the scFv is in an HL orientation, and the VH cysteine and linker cysteine are at positions 84 (VH) / 9 (linker); 61 (VH) / 14 (linker); 85 (VH) / 10 (linker); 62 (VH) / 14 (linker); 13 (VH) / 3 (linker); or 83 (VH) / 8 (linker).

5. The scFv of any one of Claims 1-3, wherein the scFv is in an LH orientation, and the VL cysteine and linker cysteine are at positions 81 (VL) / 8C (linker); 42 (VL) / 9 (linker); 45 (VL) / 10 (linker); 83 (VL) / 4 (linker); or 57 (VL) / 13 (linker).

6. The scFv of any one of Claims 1-3, wherein the scFv is in an orientation from N- terminus to C-terminus of VL-VH (LH), and the VH cysteine is at position 101 and the linker cysteine is at position 13.

7. The scFv of any one of claims 1-6, comprising a disulfide bond between the VH cysteine and a linker cysteine, or between the VL cysteine and a linker cysteine.

8. An scFv comprising a VH, a VL, a linker connecting the VH and VL, an N-terminal linker comprising a cysteine, and a disulfide bond between the N-terminal linker cysteine and a VH cysteine, or between the N-terminal linker cysteine and a VL cysteine.

9. The scFv of Claim 8, wherein the scFv is in an HL orientation and comprises a VL cysteine at position 42, 45, 57, or 81 according to Kabat numbering.

10. The scFv of Claim 8, wherein the scFv is in an LH orientation and comprises a VH cysteine at position 61, 62, 83, 84, or 85 according to Kabat numbering.

11. The scFv of any one of Claims 8-10, comprising an N-terminal linker cysteine at position -1, -2, -3, -4, -5, -6, -7, -8, -9, -10, -11 according to N-terminal linker linear numbering.

12. A construct comprising the scFv of any one of Claims 1-11.

13. The construct of Claim 12, which is a multispecific construct.

14. Use of the scFv of any one of Claims 1-11 or the construct of Claim 12 or 13 for the manufacture of a medicament for the treatment of disease or disorder.

15. The scFv of any one of Claims 1-11 or the construct of Claim 12 or 13 for use as a medicament.

16. The scFv of any one of Claims 1-11 or the construct of Claim 12 or 13 for use in the treatment of a disease or disorder.

17. A method of treating a disease or disorder in a patient, said method comprising administering to the patient the scFv of any one of Claims 1-11 or the construct of Claim 12 or 13.

18. The use of Claim 14, the scFv of Claim 15 or 16, or the method of Claim 17, wherein the disease or disorder is cancer or an immune disease.

19. A pharmaceutical composition comprising the scFv of any one of Claims 1-11 or the construct of Claim 12 or 13 and one or more pharmaceutically acceptable carriers, diluents, or excipients.