US20170088611A1

US20170088611A1 - Single-chain multivalent binding protein compositions and methods

Info

Publication number: US20170088611A1
Application number: US15/252,115
Authority: US
Inventors: Lorenzo Benatuil; Chung-Ming Hsieh
Original assignee: AbbVie Inc
Current assignee: AbbVie Inc
Priority date: 2012-12-28
Filing date: 2016-08-30
Publication date: 2017-03-30
Also published as: US9458244B2; US20140221621A1

Abstract

Provided are protein, nucleic acid, and cellular libraries of single chain multivalent binding proteins (e.g., scDVD and scDVDFab molecules) and methods of using these of these libraries for the screening of single chain multivalent binding proteins using cell surface display technology (e.g., yeast display).

Description

RELATED APPLICATIONS

This application is a continuation of U.S. patent application Ser. No. 14/141,500, filed on Dec. 27, 2013, which claims priority from U.S. Provisional Patent Application Ser. No. 61/746,659, filed on Dec. 28, 2012, which are hereby incorporated by reference in their entirety.

BACKGROUND

I. Field
The present disclosure pertains to methods and compositions for producing single chain multivalent binding proteins that specifically bind to one or more desired target antigens. More specifically, the disclosure relates to protein, nucleic acid, and cellular libraries of single chain multivalent binding proteins (e.g., scDVD molecules) and methods of using these libraries for the screening of single chain multivalent binding proteins using cell surface display technology (e.g., yeast display).
II. Description of Related Art
A wide variety of multispecific antibody formats have been developed (see Kriangkum, J., et al., Biomol Eng, 2001. 18(2): p. 31-40). Amongst them tandem single-chain Fv molecules and diabodies, and various derivatives there of, are the most widely used formats for the construction of recombinant bispecific antibodies. More recently diabodies have been fused to Fc to generate more Ig-like molecules, named di-diabodies (see Lu, D., et al., J Biol Chem, 2004. 279(4): p. 2856-65). In addition, multivalent antibody construct comprising two Fab repeats in the heavy chain of an IgG and capable of binding four antigen molecules has been described (see WO 0177342A1, and Miller, K., et al., J Immunol, 2003. 170(9): p. 4854-61).
Despite the many bispecific antibody formats available to the skilled artisan, there is often a need for the skilled artisan to improve the affinity of the bispecific antibody through affinity maturation. However, conventional affinity maturation approaches rely upon screening for affinity matured variants of the component binding domains of the multispecific antibody followed by their reassembly into the original multispecific format. Such reassembly often results in a loss of the desired improvement in binding affinity or other desirable binding characteristics. Accordingly, there is a need in the art for improved constructs, formats, and screening methodologies for identifying affinity variants of multivalent binding proteins in their desired multivalent format.

SUMMARY

The present disclosure provides a novel compositions and methods useful for the generation of improved single-chain multivalent binding proteins (e.g., scDVD) capable of binding two or more antigens simultaneously with high affinity.
Accordingly, in one aspect, the disclosure provides a single chain multivalent binding protein.
In certain embodiments, the single chain multivalent binding protein has the general formula VH1-(X1)n-VH2-X2-VL1-(X3)n-VL2, wherein VH1 is a first antibody heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, X2 is a linker, VL1 is a first antibody light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites.
In certain embodiments, the single chain binding protein has the formula CH1-X0-VH1-(X1)n-VH2-X2-CL1-X4-VL1-(X3)n-VL2, wherein CH1 is a heavy chain constant domain, X0 is a linker with the proviso that it is not a constant domain, VH1 is a first antibody heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, X2 is a linker, CL1 is a light chain heavy domain, X4 is a linker with the proviso that it is not a constant domain, VL1 is a first antibody light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites. Optionally, the CL1 domain can be a kappa (hcκ or cκ) or a lambda (hλ or cλ) constant domain. In certain embodiments, CL1 is cκ.
In certain embodiments, X2 is a GS-rigid linker sequence. The GS rigid linker sequence can comprise an amino acid sequence selected from the group consisting of SEQ ID NOs:1-4.
In certain embodiments, the single chain multivalent binding protein has the general formula (VL1-(X1)n-VL2-X2-VH1-(X3)n-VH2, wherein VL1 is a first antibody light chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, X2 is a linker, VH1 is a first antibody heavy chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding site.
In certain embodiments, the single chain binding protein has the formula CL1-X0-VL1-(X1)n-VL2-X2-CH1-X4-VH1-(X3)n-VH2, wherein CL1 is a light chain constant domain, X0 is a linker with the proviso that it is not a constant domain, VL1 is a first antibody light chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, X2 is a linker, CH1 is a heavy chain constant domain, X4 is a linker with the proviso that it is not a constant domain, VH1 is a first antibody heavy chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding site. Optionally, the CL1 domain can be a kappa (hcκ or cκ) or a lambda (hλ or cλ) constant domain. In certain embodiments, CL1 is cκ.
In certain embodiments, X2 is a GS-rigid linker sequence. The GS rigid linker sequence can comprise an amino acid sequence selected from the group consisting of SEQ ID NOs:1-4.
In certain embodiments, the single chain multivalent binding protein is a single-chain dual variable domain immunoglobulin molecules (scDVD).
In certain embodiments, the single chain multivalent binding protein further comprising a cell surface anchoring moiety linked to the N and/or C terminus. In one embodiment, the anchoring moiety comprises the Aga2p polypeptide.
In another aspect, the disclosure provides a polynucleotide encoding a binding protein disclosed herein.
In another aspect, the disclosure provides a host cell expressing a binding protein disclosed herein.
In another aspect, the disclosure provides a diverse library of binding proteins.
In certain embodiments, the diverse library of binding proteins comprises a polypeptide chain having the general formula VH1-(X1)n-VH2-X2-VL1-(X3)n-VL2, wherein VH1 is a first heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second heavy chain variable domain, X2 is a linker, VL1 is a first light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second light chain variable domain, and n is 0 or 1, wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites, and wherein the amino acid sequences of VH1, X1, VH2, X2, VL1, X3, and/or VL2 independently vary within the library.
In certain embodiments, the diverse library of binding proteins comprises a polypeptide chain having the general formula CH1-X0-VH1-(X1)n-VH2-X2-CL1-X4-VL1-(X3)n-VL2, wherein CH1 is a heavy chain constant domain, X0 is a linker with the proviso that it is not a constant domain, VH1 is a first antibody heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, X2 is a linker, CL1 is a light chain constant domain, X4 is a linker with the proviso that it is not a constant domain, VL1 is a first antibody light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, the VH2 and VL2 respectively combine to form two functional antigen binding sites, and wherein the amino acid sequences of VH1, X1, VH2, X2, VL1, X3, and/or VL2 independently vary within the library. Optionally, the CL1 domain can be a kappa (hcκ or cκ) or a lambda (hcλ or cλ) constant domain. In certain embodiments, CL1 is cκ.
In certain embodiments, X2 is a GS-rigid linker sequence. The GS rigid linker sequence can comprise an amino acid sequence selected from the group consisting of SEQ ID NOs:1-4.
In certain embodiments, the diverse library of binding proteins comprises a polypeptide chain having the general formula (VL1-(X1)n-VL2-X2-VH1-(X3)n-VH2, wherein VL1 is a first antibody light chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, X2 is a linker, VH1 is a first antibody heavy chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, and n is 0 or 1, wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites, and wherein the amino acid sequences of VL1, X1, VL2, X2, VH1, X3, and/or VH2 independently vary within the library.
In certain embodiments, the diverse library of binding proteins comprises a polypeptide chain having the general formula CL1-X0-VL1-(X1)n-VL2-X2-CH1-X4-VH1-(X3)n-VH2, wherein CL1 is a light chain constant domain, X0 is a linker with the proviso that it is not a constant domain, VL1 is a first antibody light chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, X2 is a linker, CH1 is a heavy chain constant domain, X4 is a linker with the proviso that it is not a constant domain, VH1 is a first antibody heavy chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, the VH2 and VL2 respectively combine to form two functional antigen binding site, and wherein the amino acid sequences of VH1, X1, VH2, X2, VL1, X3, and/or VL2 independently vary within the library. In certain embodiments, the CL1 light chain. Optionally, the CL1 domain can be a kappa (hcκ or cκ) or a lambda (hcλ or cλ) constant domain. In certain embodiments, CL1 is cκ.
In certain embodiments, X2 is a GS-rigid linker sequence. The GS rigid linker sequence can comprise an amino acid sequence selected from the group consisting of SEQ ID NOs:1-4.
In certain embodiments, each binding proteins further comprises a cell surface anchoring moiety linked to the N or C terminus. In certain embodiments, the anchoring moiety is a cell surface protein. In one embodiment, the anchoring moiety is Aga2p.
In certain embodiments, the polypeptide chain is a scDVD or scDVDFab.
In certain embodiments, the amino acid sequence of at least one CDR of VH1, VH2, VL1 or VL2 independently varies within the library. In one embodiment, the amino acid sequence of HCDR3 of VH1, VH2 independently vary within the library. In one embodiment, the amino acid sequence of HCDR1 and HCDR2 of VH1 or VH2 independently vary within the library. In one embodiment, the amino acid sequence of HCDR1, HCDR2 and HCDR3 of VH1 or VH2 independently vary within the library. In one embodiment, the amino acid sequence of HCDR3 of VL1 or VL2 independently vary within the library. In one embodiment, the amino acid sequence of HCDR1 and HCDR2 of VL1 or VL2 independently vary within the library. In one embodiment, the amino acid sequence of HCDR1, HCDR2 and HCDR3 of VL1 or VL2 independently vary within the library.
In certain embodiments, X1 independently varies within the library and wherein X1 is selected from the amino acid sequences set forth in FIG. 2. In certain embodiments, X2 independently varies within the library and wherein X2 is (G₄S)n, where n=1-10 (SEQ ID NO: 53). In other embodiments, X2 is selected from the amino acid sequences set forth in FIG. 11B. In specific embodiments, X2 is selected from the amino acid sequences set forth in FIG. 11B when the polypeptide chain includes CH and CL domain. In certain embodiments, X3 independently varies within the library and X3 is selected from the amino acid sequences set forth in FIG. 2.
In certain embodiments, the library of binding proteins share at least 70, 75, 80, 85, 90, 95, 96, 97, 98, or 99 amino acid sequence identity with a reference binding protein. In certain embodiments, VH1 and VH2 of the reference binding protein specifically bind to different antigens.
In another aspect, the disclosure provides a diverse library of polynucleotides encoding a diverse library of binding proteins disclosed herein.
In another aspect, the disclosure provides a diverse library of expression vectors comprising a diverse library of polynucleotides disclosed herein.
In another aspect, the disclosure provides a library of transformed host cells, expressing the diverse library of binding proteins disclosed herein.
In certain embodiments, the binding proteins are anchored on the cell surface of a transformed host cell. In certain embodiments, the binding proteins are anchored on the cell surface through Aga1p.
In certain embodiments, the host cells are eukaryotic. In certain embodiments, the host cells are yeast, e.g., Saccharomyces cerevisiae, Saccharomyces carlsbergensis, Candida albicans, Candida kefyr, Candida tropicalis, Cryptococcus laurentii, Cryptococcus neoformans, Hansenula anomala, Hansenula polymorpha, Kluyveromyces fragilis, Kluyveromyces lactis, Kluyveromyces marxianus, Pichia pastoris, Rhodotorula rubra, Schizosaccharomyces pombe and Yarrowia lipolytica. In one embodiment, the yeast is Saccharomyces cerevisiae.
In another aspect, the disclosure provides a method of selecting a binding protein that specifically binds to a target antigen, the method comprising: providing a diverse library of transformed host cells expressing a diverse library of binding proteins disclosed herein; contacting the host cells with the target antigen; and selecting a host cell that bind to the target antigen, thereby identifying a binding protein that specifically binds to a target antigen.
In another aspect, the disclosure provides a method of selecting a binding protein that specifically binds to a first and a second target antigen simultaneously, the method comprising: providing a diverse library of transformed host cells expressing a diverse library of binding proteins disclosed herein; contacting the host cells with the first and second target antigen; and selecting a host cell that bind to the first and second target antigen, thereby identifying a binding protein that specifically binds to a first and a second target antigen simultaneously.
In certain embodiments of the methods disclosed herein, host cells that bind to the first and/or second antigen are selected by Magnetic Activated Cell Sorting using magnetically labeled antigen. In certain embodiments of the methods disclosed herein, host cells that bind to the first and/or second antigen are selected by Fluorescence Activated Cell Sorting using fluorescently labeled antigen.
In certain embodiments, the methods disclosed herein further comprise isolating the binding protein-encoding polynucleotide sequences from the selected host cells.
In another aspect, the disclosure provides a method of producing a binding protein comprising expressing in a host cell a binding protein that was selected using the methods disclosed herein.
In another aspect, the disclosure provides method of producing a diverse library of binding proteins that specifically binds to a target antigen, the method comprising: providing a first diverse library of scDVD or scDVDFab molecules, wherein the amino acid sequence of a first region of the scDVD or scDVDFab molecules is varied in the library, and wherein each member of the library binds to the target antigen; providing a second diverse library of scDVD or scDVDFab molecules, wherein the amino acid sequence of a second region of the scDVD or scDVDFab molecules is varied in the library, and wherein each member of the library binds to the target antigen; recombining the first and second libraries to produce a third diverse library of scDVD or scDVDFab molecules, wherein the third library comprises the first regions from the first library and the second region from the second library, thereby producing a diverse library of binding proteins that specifically binds to a target antigen.
In certain embodiments, the first and second libraries are recombined by yeast gap repair of polynucleotides encoding the libraries.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1A depicts an exemplary single chain dual variable domain (scDVD) molecules (FIG. 1A discloses “(G₄S)_n” as SEQ ID NO: 54), FIG. 1B depicts an exemplary full-length DVD-Ig molecule, and FIG. 1C depicts an exemplary a single chain Fv molecule.

FIG. 2 is a schematic representation of an scDVD molecule and exemplary inter-variable domain linker amino acid sequences. The linkers between the VH1 and VH2 domains have amino acid sequences of SEQ ID NOs:9-30 listed from top to bottom. The linkers between the VL1 and VL2 domains have amino acid sequences of SEQ ID NOs:31-52 listed from top to bottom. FIG. 2 discloses “(G₄S)_n” as SEQ ID NO: 54.

FIG. 3 depicts the results of flow cytometry assays measuring the cell surface expression of scDVD or scFv on yeast cells.

FIG. 4A depicts the results of flow cytometry assays measuring the binding of DLL4 and/or VEGF to yeast cells expressing cell surface DLL4/VEGF-binding scDVD, and FIG. 4B depicts SOST and/or TNFa to yeast cells expressing cell surface SOST/TNFa-binding scDVD.

FIG. 5 depicts the results of flow cytometry assays measuring the binding of SOST and/or TNFa to yeast cells expressing cell surface SOST/TNFa-binding scDVD tagged with various epitope tags.

FIG. 6A depicts the amino acid sequence of an exemplary SOST/TNFa-binding scDVD molecule (SEQ ID NO:57) (FIG. 6A discloses “(G₄S)_n” as SEQ ID NO: 54). FIG. 6B depicts an exemplary SOST/TNFa-binding scDVD library design, with the VH3-9, SOST VH, V1-16 and MSL10VL sequences represented by SEQ ID NOs: 58-61, respectively (FIG. 6B discloses “(G₄S)_n” as SEQ ID NO: 54). FIG. 6C depicts the results of flow cytometry assays measuring the binding of SOST to yeast cells expressing parental or affinity matured cell surface SOST/TNFa-binding scDVD. FIG. 6D depicts the results of flow cytometry assays measuring the binding of SOST to yeast cells expressing parental or affinity matured cell surface SOST/TNFa-binding scDVD.

FIG. 7A depicts a schematic representation of an scDVD molecule and exemplary inter-VL domain linker amino acid sequences of SEQ ID NOs:62-73 listed from top to bottom (FIG. 7A discloses “(G₄S)_n” as SEQ ID NO: 54), and FIG. 7B depicts and results (as fold enrichment) of yeast display screens of SOST/TNFa-binding scDVD library comprising various inter-VL domain linker amino acid sequences.

FIG. 8 is a schematic representation of exemplary scDVD libraries disclosed herein and multiplexing methods of using these libraries.

FIG. 9 is a schematic representation of exemplary scDVD libraries disclosed herein.

FIG. 10A depicts an exemplary single chain dual variable domain Fab (scDVDFab) molecules, FIG. 10B depicts an exemplary full-length DVD-Ig molecule, and FIG. 10C depicts an exemplary a single chain DVD molecule (FIG. 10C discloses “(G₄S)_n” as SEQ ID NO: 54).

FIG. 11A depicts a schematic representation of an scDVDFab molecule, FIG. 11B depicts GS-rigid linker amino acid sequences (SEQ ID NOs:1-4), and FIG. 11C depicts a schematic of a scDVDFab with a GS-rigid linker (FIG. 11C discloses “G₃SG₃” as SEQ ID NO: 96 and “G₂SG₂” as SEQ ID NO: 97).

FIG. 12 depicts the results of flow cytometry assays measuring the expression of scDVDFab on the surface of yeast.

FIG. 13 depicts the results of flow cytometry assays showing that 1B/IL17 scDVDFab expressed on yeast retains its ability to bind both IL1B and/or IL17.

FIG. 14 depicts the results of flow cytometry assays showing that scDVDFab and DVD-Fab had similar binding profiles binding to both IL1B and IL17 on the surface of yeast.

DETAILED DESCRIPTION

The present disclosure provides a novel compositions and methods useful for the generation of improved single-chain multivalent binding proteins (e.g., scDVD) capable of binding two or more antigens simultaneously with high affinity.

I. DEFINITIONS

Unless otherwise defined herein, scientific and technical terms used in connection with the present disclosure shall have the meanings that are commonly understood by those of ordinary skill in the art. The meaning and scope of the terms should be clear, however, in the event of any latent ambiguity, definitions provided herein take precedent over any dictionary or extrinsic definition. Further, unless otherwise required by context, singular terms shall include pluralities and plural terms shall include the singular. Generally, nomenclature used in connection with, and techniques of, cell and tissue culture, molecular biology, immunology, microbiology, genetics and protein and nucleic acid chemistry and hybridization described herein are those well known and commonly used in the art.
In order that the disclosure may be more readily understood, certain terms are first defined.
The term “multivalent binding protein” is used throughout this specification to denote a binding protein comprising two or more antigen binding sites, each of which can bind independently bind to an antigen.
The terms “dual variable domain immunoglobulin” or “DVD-Ig” refer to the multivalent binding proteins disclosed in, e.g., U.S. Pat. No. 8,258,268, which is herein incorporated by reference in its entirety.
The terms “single chain dual variable domain immunoglobulin” or “scDVD” refer to the antigen binding fragment of a DVD molecule that is analogous to an antibody single chain Fv fragment. scDVD are generally of the formula VH1-(X1)n-VH2-X2-VL1-(X3)n-VL2, where VH1 is a first antibody heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, X2 is a linker, VL1 is a first antibody light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, and n is 0 or 1, where the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites. An exemplary scDVD is depicted in FIGS. 1A-1C herein.
The terms “single chain dual variable domain immunoglobulin Fab” or “scDVDFab” refer to the antigen binding fragment of a DVD molecule that includes the variable heavy (VH) and light (VL) chains of a DVD-Ig. scDVD are generally of the formula CH1-X0-VH1-(X1)n-VH2-X2-CL1-X4-VL1-(X3)n-VL2, where CH1 is a heavy chain constant domain, X0 is a linker with the proviso that it is not a constant domain, VH1 is a first antibody heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, X2 is a linker, CL1 is a light chain constant domain, X4 is a linker with the proviso that it is not a constant domain, VL1 is a first antibody light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, and n is 0 or 1, where the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites. Optionally, the CL1 domain can be a kappa (hcκ or cκ) or a lambda (hcλ or cλ) constant domain. In certain embodiments, CL1 is cκ. An exemplary scDVDFab is depicted in FIG. 10A, herein.
The term “antibody”, as used herein, broadly refers to any immunoglobulin (Ig) molecule comprised of four polypeptide chains, two heavy (H) chains and two light (L) chains, or any functional fragment, mutant, variant, or derivation thereof, which retains the essential epitope binding features of an Ig molecule. Such mutant, variant, or derivative antibody formats are known in the art. Non-limiting embodiments of which are discussed below.
In a full-length antibody, each heavy chain is comprised of a heavy chain variable region (abbreviated herein as HCVR or VH) and a heavy chain constant region. The heavy chain constant region is comprised of three domains, CH1, CH2 and CH3. Each light chain is comprised of a light chain variable region (abbreviated herein as LCVR or VL) and a light chain constant region. The light chain constant region is comprised of one domain, CL. The VH and VL regions can be further subdivided into regions of hypervariability, termed complementarity determining regions (CDR), interspersed with regions that are more conserved, termed framework regions (FR). Each VH and VL is composed of three CDRs and four FRs, arranged from amino-terminus to carboxy-terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. Immunoglobulin molecules can be of any type (e.g., IgG, IgE, IgM, IgD, IgA and IgY), class (e.g., IgG 1, IgG2, IgG 3, IgG4, IgA1 and IgA2) or subclass.
The term “Fc region” is used to define the C-terminal region of an immunoglobulin heavy chain, which may be generated by papain digestion of an intact antibody. The Fc region may be a native sequence Fc region or a variant Fc region. The Fc region of an immunoglobulin generally comprises two constant domains, a CH2 domain and a CH3 domain, and optionally comprises a CH4 domain. Replacements of amino acid residues in the Fc portion to alter antibody effector function are known in the art (Winter, et al. U.S. Pat. Nos. 5,648,260; 5,624,821). The Fc portion of an antibody mediates several important effector functions e.g. cytokine induction, ADCC, phagocytosis, complement dependent cytotoxicity (CDC) and half-life/clearance rate of antibody and antigen-antibody complexes. In some cases these effector functions are desirable for therapeutic antibody but in other cases might be unnecessary or even deleterious, depending on the therapeutic objectives. Certain human IgG isotypes, particularly IgG1 and IgG3, mediate ADCC and CDC via binding to Fc.gamma.Rs and complement C1q, respectively. Neonatal Fc receptors (FcRn) are the critical components determining the circulating half-life of antibodies. In still another embodiment at least one amino acid residue is replaced in the constant region of the antibody, for example the Fc region of the antibody, such that effector functions of the antibody are altered. The dimerization of two identical heavy chains of an immunoglobulin is mediated by the dimerization of CH3 domains and is stabilized by the disulfide bonds within the hinge region (Huber et al. Nature; 264: 415-20; Thies et al 1999 J Mol Biol; 293: 67-79.). Mutation of cysteine residues within the hinge regions to prevent heavy chain-heavy chain disulfide bonds will destabilize dimeration of CH3 domains. Residues responsible for CH3 dimerization have been identified (Dall'Acqua 1998 Biochemistry 37: 9266-73.). Therefore, it is possible to generate a monovalent half-Ig. Interestingly, these monovalent half Ig molecules have been found in nature for both IgG and IgA subclasses (Seligman 1978 Ann Immunol 129: 855-70; Biewenga et al 1983 Clin Exp Immunol 51: 395-400). The stoichiometry of FcRn: Ig Fc region has been determined to be 2:1 (West et al 2000 Biochemistry 39: 9698-708), and half Fc is sufficient for mediating FcRn binding (Kim et al 1994 Eur J Immunol; 24: 542-548.). Mutations to disrupt the dimerization of CH3 domain may not have greater adverse effect on its FcRn binding as the residues important for CH3 dimerization are located on the inner interface of CH3 b sheet structure, whereas the region responsible for FcRn binding is located on the outside interface of CH2-CH3 domains. However the half Ig molecule may have certain advantage in tissue penetration due to its smaller size than that of a regular antibody. In one embodiment at least one amino acid residue is replaced in the constant region of the binding protein disclosed herein, for example the Fc region, such that the dimerization of the heavy chains is disrupted, resulting in half DVD Ig molecules.
The term “antigen-binding portion” of an antibody (or simply “antibody portion”), as used herein, refers to one or more fragments of an antibody that retain the ability to specifically bind to an antigen. It has been shown that the antigen-binding function of an antibody can be performed by fragments of a full-length antibody. Such antibody embodiments may also be bispecific, dual specific, or multi-specific formats; specifically binding to two or more different antigens. Examples of binding fragments encompassed within the term “antigen-binding portion” of an antibody include (i) a Fab fragment, a monovalent fragment consisting of the VL, VH, CL and CH1 domains; (ii) a F(ab′).sub.2 fragment, a bivalent fragment comprising two Fab fragments linked by a disulfide bridge at the hinge region; (iii) a Fd fragment consisting of the VH and CH1 domains; (iv) a Fv fragment consisting of the VL and VH domains of a single arm of an antibody, (v) a dAb fragment (Ward et al., (1989) Nature 341:544-546, Winter et al., PCT publication WO 90/05144 A1 herein incorporated by reference), which comprises a single variable domain; and (vi) an isolated complementarity determining region (CDR). Furthermore, although the two domains of the Fv fragment, VL and VH, are coded for by separate genes, they can be joined, using recombinant methods, by a synthetic linker that enables them to be made as a single protein chain in which the VL and VH regions pair to form monovalent molecules (known as single chain Fv (scFv); see e.g., Bird et al. (1988) Science 242:423-426; and Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883). Such single chain antibodies are also intended to be encompassed within the term “antigen-binding portion” of an antibody. Other forms of single chain antibodies, such as diabodies are also encompassed. Diabodies are bivalent, bispecific antibodies in which VH and VL domains are expressed on a single polypeptide chain, but using a linker that is too short to allow for pairing between the two domains on the same chain, thereby forcing the domains to pair with complementary domains of another chain and creating two antigen binding sites (see e.g., Holliger, P., et al. (1993) Proc. Natl. Acad. Sci. USA 90:6444-6448; Poljak, R. J., et al. (1994) Structure 2:1121-1123). Such antibody binding portions are known in the art (Kontermann and Dubel eds., Antibody Engineering (2001) Springer-Verlag. New York. 790 pp. (ISBN 3-540-41354-5). In addition single chain antibodies also include “linear antibodies” comprising a pair of tandem Fv segments (VH-CH1-VH-CH1) which, together with complementary light chain polypeptides, form a pair of antigen binding regions (Zapata et al. Protein Eng. 8(10):1057-1062 (1995); and U.S. Pat. No. 5,641,870).
As used herein, the terms “VH domain” and “VL domain” refer to single antibody variable heavy and light domains, respectively, comprising FR (Framework Regions) 1, 2, 3 and 4 and CDR (Complementary Determinant Regions) 1, 2 and 3 (see Kabat et al. (1991) Sequences of Proteins of Immunological Interest. (NIH Publication No. 91-3242, Bethesda).
As used herein, the terms “CH1 domain” and “CL1 domain” refer to single antibody heavy and light constant regions. A CL1 domain can be a Cκ or a Cλ, domain.
As used herein, the term “CDR” or “complementarity determining region” means the noncontiguous antigen combining sites found within the variable region of both heavy and light chain polypeptides. These particular regions have been described by Kabat et al., J. Biol. Chem. 252, 6609-6616 (1977) and Kabat et al., Sequences of protein of immunological interest. (1991), and by Chothia et al., J. Mol. Biol. 196:901-917 (1987) and by MacCallum et al., J. Mol. Biol. 262:732-745 (1996) where the definitions include overlapping or subsets of amino acid residues when compared against each other. The amino acid residues which encompass the CDRs as defined by each of the above cited references are set forth for comparison. Preferably, the term “CDR” is a CDR as defined by Kabat, based on sequence comparisons.
As used herein the term “framework (FR) amino acid residues” refers to those amino acids in the framework region of an immunogobulin chain. The term “framework region” or “FR region” as used herein, includes the amino acid residues that are part of the variable region, but are not part of the CDRs (e.g., using the Kabat definition of CDRs).
As used herein, the term “specifically binds to” refers to the ability of a binding polypeptide to bind to an antigen with an Kd of at least about 1×10⁻⁶M, 1×10⁻⁷M, 1×10⁻⁸M, 1×10⁻⁹M, 1×10⁻¹⁰M, 1×10⁻¹¹M, 1×10⁻¹²M, or more, and/or bind to an antigen with an affinity that is at least two-fold greater than its affinity for a nonspecific antigen. It shall be understood, however, that the binding polypeptide are capable of specifically binding to two or more antigens which are related in sequence. For example, the binding polypeptides disclosed herein can specifically bind to both human and a non-human (e.g., mouse or non-human primate) orthologs of an antigen.
The term “Polypeptide” as used herein, refers to any polymeric chain of amino acids. The terms “peptide” and “protein” are used interchangeably with the term polypeptide and also refer to a polymeric chain of amino acids. The term “polypeptide” encompasses native or artificial proteins, protein fragments and polypeptide analogs of a protein sequence. A polypeptide may be monomeric or polymeric.
The term “linker” is used to denote polypeptides comprising two or more amino acid residues joined by peptide bonds and are used to link one or more antigen binding portions. Such linker polypeptides are well known in the art (see e.g., Holliger, P., et al. (1993) Proc. Natl. Acad. Sci. USA 90:6444-6448; Poljak, R. J., et al. (1994) Structure 2:1121-1123). Preferred linkers include, but are not limited to, the amino acid linkers set forth in Table 7 herein.
The term “K_on”, as used herein, is intended to refer to the on rate constant for association of an antibody to the antigen to form the antibody/antigen complex as is known in the art.
The term “K_off”, as used herein, is intended to refer to the off rate constant for dissociation of an antibody from the antibody/antigen complex as is known in the art.
The term “Kd”, as used herein, is intended to refer to the dissociation constant of a particular antibody-antigen interaction as is known in the art.
The term “vector”, as used herein, is intended to refer to a nucleic acid molecule capable of transporting another nucleic acid to which it has been linked. One type of vector is a “plasmid”, which refers to a circular double stranded DNA loop into which additional DNA segments may be ligated. Another type of vector is a viral vector, wherein additional DNA segments may be ligated into the viral genome. Certain vectors are capable of autonomous replication in a host cell into which they are introduced (e.g., bacterial vectors having a bacterial origin of replication and episomal mammalian vectors). Other vectors (e.g., non-episomal mammalian vectors) can be integrated into the genome of a host cell upon introduction into the host cell, and thereby are replicated along with the host genome. Moreover, certain vectors are capable of directing the expression of genes to which they are operatively linked. Such vectors are referred to herein as “recombinant expression vectors” (or simply, “expression vectors”). In general, expression vectors of utility in recombinant DNA techniques are often in the form of plasmids. In the present specification, “plasmid” and “vector” may be used interchangeably as the plasmid is the most commonly used form of vector. However, the disclosure is intended to include such other forms of expression vectors, such as viral vectors (e.g., replication defective retroviruses, adenoviruses and adeno-associated viruses), which serve equivalent functions.
“Transformation”, as defined herein, refers to any process by which exogenous DNA enters a host cell. Transformation may occur under natural or artificial conditions using various methods well known in the art. Transformation may rely on any known method for the insertion of foreign nucleic acid sequences into a prokaryotic or eukaryotic host cell. The method is selected based on the host cell being transformed and may include, but is not limited to, viral infection, electroporation, lipofection, and particle bombardment. Such “transformed” cells include stably transformed cells in which the inserted DNA is capable of replication either as an autonomously replicating plasmid or as part of the host chromosome. They also include cells which transiently express the inserted DNA or RNA for limited periods of time.
The term “recombinant host cell” (or simply “host cell”), as used herein, is intended to refer to a cell into which exogenous DNA has been introduced. It should be understood that such terms are intended to refer not only to the particular subject cell, but, to the progeny of such a cell. Because certain modifications may occur in succeeding generations due to either mutation or environmental influences, such progeny may not, in fact, be identical to the parent cell, but are still included within the scope of the term “host cell” as used herein. Preferably host cells include prokaryotic and eukaryotic cells selected from any of the Kingdoms of life. Preferred eukaryotic cells include protist, fungal, plant and animal cells. Most preferably host cells include but are not limited to the prokaryotic cell line E. Coli; mammalian cell lines CHO, HEK 293 and COS; the insect cell line Sf9; and the fungal cell Saccharomyces cerevisiae.

II. SINGLE-CHAIN MULTIVALENT BINDING PROTEINS

In one aspect, the disclosure provides single-chain multivalent binding proteins that can bind to two antigen simultaneously. In certain embodiments, the single-chain multivalent binding proteins generally comprise a polypeptide of the formula VH1-(X1)n-VH2-X2-VL1-(X3)n-VL2, where VH1 is a first antibody heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, X2 is a linker, VL1 is a first antibody light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, and n is 0 or 1, where the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites.
In certain embodiments, the single chain binding protein has the formula CH1-X0-VH1-(X1)n-VH2-X2-CL1-X4-VL1-(X3)n-VL2, wherein CH1 is a heavy chain constant domain, X0 is a linker with the proviso that it is not a constant domain, VH1 is a first antibody heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, X2 is a linker, CL1 is a light chain heavy domain, X4 is a linker with the proviso that it is not a constant domain, VL1 is a first antibody light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites. Optionally, the CL1 domain can be a kappa (hcκ or cκ) or a lambda (hλ or cλ) constant domain. In certain embodiments, CL1 is cκ.
In certain embodiments, X2 is a GS-rigid linker sequence. The GS rigid linker sequence can comprise an amino acid sequence selected from the group consisting of SEQ ID NOs:1-4.
In certain embodiments, the single-chain multivalent binding proteins generally comprise a polypeptide of the formula VL1-(X1)n-VL2-X2-VH1-(X3)n-VH2, where VL1 is a first antibody light chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, X2 is a linker, VH1 is a first antibody heavy chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VHL2 is a second antibody heavy chain variable domain, and n is 0 or 1, where the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites
In certain embodiments, the single chain binding protein has the formula CL1-X0-VL1-(X1)n-VL2-X2-CH1-X4-VH1-(X3)n-VH2, wherein CL1 is a light chain constant domain, X0 is a linker with the proviso that it is not a constant domain, VL1 is a first antibody light chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, X2 is a linker, CH1 is a heavy chain constant domain, X4 is a linker with the proviso that it is not a constant domain, VH1 is a first antibody heavy chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding site. Optionally, the CL1 domain can be a kappa (hcκ or cκ) or a lambda (hλ or cλ) constant domain. In certain embodiments, CL1 is cκ.
In certain embodiments, X2 is a GS-rigid linker sequence. The GS rigid linker sequence can comprise an amino acid sequence selected from the group consisting of SEQ ID NOs:1-4.
In certain embodiments, the single-chain multivalent binding proteins are single-chain dual variable domain immunoglobulin molecules (scDVD). An exemplary scDVD is depicted in FIGS. 1A-1C herein. In other embodiments, the single-chain multivalent binding proteins are single-chain dual variable domain immunoglobulin Fab molecules (scDVDFab). An exemplary scDVDFab is depicted in FIG. 10A, herein.
In certain embodiments, the multivalent binding proteins comprise a cell surface anchoring moiety linked to the N and/or C terminus. Any molecule that can display the binding protein on the surface of a cell can be employed including, without limitation, cell surface protein and lipids. In certain embodiments, the anchoring moiety comprises the Aga2p polypeptide.
The antibody variable domains for the use in the single-chain multivalent binding proteins disclosed herein can be obtained using recombinant DNA techniques from a parent antibody (or DVD-Ig) generated by any method known in the art. In a certain embodiments, the variable domain is a murine heavy or light chain variable domain. In a certain embodiments, the variable domain is a CDR grafted or a humanized variable heavy or light chain domain. In a certain embodiments, the variable domain is a human heavy or light chain variable domain.
In certain embodiments, the first and second variable domains are linked directly to each other using recombinant DNA techniques. In certain embodiments, the variable domains are linked via a linker sequence. Preferably two variable domains are linked. Three or more variable domains may also be linked directly or via a linker sequence. The variable domains may bind the same antigen or may bind different antigens. Single-chain multivalent binding proteins molecules disclosed herein may include one immunoglobulin variable domain and one non-immunoglobulin variable domain such as ligand binding domain of a receptor, active domain of an enzyme. Single-chain multivalent binding proteins molecules may also comprise two or more non-Ig domains.
The linker sequence may be a single amino acid or a polypeptide sequence. Preferably the linker sequences are selected from the group consisting of consisting of the amino acid sequences set forth in FIG. 2 herein.
In certain embodiments, a heavy chain or light chain constant domain is linked to the single-chain multivalent binding proteins domains using recombinant DNA techniques. Additionally or alternatively, in certain embodiments, the DVD heavy chain is linked to an Fc region. The Fc region may be a native sequence Fc region, or a variant Fc region. In certain embodiments, the Fc region is a human Fc region. In one embodiment the Fc region includes an Fc region from IgG1, IgG2, IgG3, IgG4, IgA, IgM, IgE, or IgD.

III. LIBRARIES OF MULTIVALENT BINDING PROTEIN

In one aspect, the disclosure provides libraries of single-chain multivalent binding proteins (e.g., scDVD molecules). Such libraries are particularly useful for selecting multivalent binding proteins with improved properties relative to a reference binding molecule (e.g., improved binding kinetics or thermostability). Exemplary libraries and methods are set forth in FIGS. 8 and 9.
In certain embodiments, the library of binding proteins comprises a polypeptide chain having the general formula VH1-(X1)n-VH2-X2-VL1-(X3)n-VL2, wherein VH1 is a first heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second heavy chain variable domain, X2 is a linker, VL1 is a first light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second light chain variable domain, and n is 0 or 1, wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites, and wherein the amino acid sequences of VH1, X1, VH2, X2, VL1, X3, and/or VL2 independently vary within the library. In one embodiment, the polypeptide chain is a scDVD.
In certain embodiments, the diverse library of binding proteins comprises a polypeptide chain having the general formula CH1-X0-VH1-(X1)n-VH2-X2-CL1-X4-VL1-(X3)n-VL2, wherein CH1 is a heavy chain constant domain, X0 is a linker with the proviso that it is not a constant domain, VH1 is a first antibody heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, X2 is a linker, CL1 is a light chain constant domain, X4 is a linker with the proviso that it is not a constant domain, VL1 is a first antibody light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, the VH2 and VL2 respectively combine to form two functional antigen binding sites, and wherein the amino acid sequences of VH1, X1, VH2, X2, VL1, X3, and/or VL2 independently vary within the library. Optionally, the CL1 domain can be a kappa (hcκ or cκ) or a lambda (hcλ or cλ) constant domain. In certain embodiments, CL1 is cκ. In one embodiment, the polypeptide chain is a scDVDFab.
In certain embodiments, X2 is a GS-rigid linker sequence. The GS rigid linker sequence can comprise an amino acid sequence selected from the group consisting of SEQ ID NOs:1-4.
In certain embodiments, the binding proteins further comprise a polypeptide chain having the general formula (VL1-(X1)n-VL2-X2-VH1-(X3)n-VH2, wherein VL1 is a first heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second heavy chain variable domain, X2 is a linker, VH1 is a first light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VH2 is a second light chain variable domain, and n is 0 or 1, wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites, and wherein the amino acid sequences of VL1, X1, VL2, X2, VH1, X3, and/or VH2 independently vary within the library. In one embodiment, the polypeptide chain is a scDVD.
In certain embodiments, the diverse library of binding proteins comprises a polypeptide chain having the general formula CL1-X0-VL1-(X1)n-VL2-X2-CH1-X4-VH1-(X3)n-VH2, wherein CL1 is a light chain constant domain, X0 is a linker with the proviso that it is not a constant domain, VL1 is a first antibody light chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, X2 is a linker, CH1 is a heavy chain constant domain, X4 is a linker with the proviso that it is not a constant domain, VH1 is a first antibody heavy chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, the VH2 and VL2 respectively combine to form two functional antigen binding site, and wherein the amino acid sequences of VH1, X1, VH2, X2, VL1, X3, and/or VL2 independently vary within the library. In certain embodiments, the CL1 light chain. Optionally, the CL1 domain can be a kappa (hcκ or cκ) or a lambda (hcλ or cλ) constant domain. In certain embodiments, CL1 is cκ. In one embodiment, the polypeptide chain is a scDVDFab.
In certain embodiments, X2 is a GS-rigid linker sequence. The GS rigid linker sequence can comprise an amino acid sequence selected from the group consisting of SEQ ID NOs:1-4.
Any region of the polypeptide chains can be varied independently in the libraries disclosed herein. In certain embodiments, the amino acid sequences of at least one CDR of VH1, VH2, VL1 or VL2 independently varies within the library. In one embodiment, the amino acid sequences of HCDR3 of VH1, VH2 independently vary within the library. In one embodiment, the amino acid sequences of HCDR1 and HCDR2 of VH1 or VH2 independently vary within the library. In one embodiment, the amino acid sequences of HCDR1, HCDR2 and HCDR3 of VH1 or VH2 independently vary within the library. In one embodiment, the amino acid sequences of HCDR3 of VL1 or VL2 independently vary within the library. In one embodiment, the amino acid sequences of HCDR1 and HCDR2 of VL1 or VL2 independently vary within the library. In one embodiment, the amino acid sequences of HCDR1, HCDR2 and HCDR3 of VL1 or VL2 independently vary within the library.
The linker regions X1, X2 and/or X3 can be also be varied independently in the libraries disclosed herein. Any length and sequence of linkers can be employed. Suitable amino acid sequences for use in linker X1, X2 and/or X3 are set forth in FIG. 2 herein. In other embodiments, X2 is selected from the amino acid sequences set forth in FIG. 11B. In specific embodiments, X2 is selected from the amino acid sequences set forth in FIG. 11B when the polypeptide chain includes CH and CL domain.
In certain embodiments, the libraries disclosed herein are used in cell surface display techniques (e.g., yeast display as described in Wittrup, et al. U.S. Pat. No. 6,699,658, incorporated herein by reference). Accordingly, in certain embodiments, each binding protein in the library further comprises a cell surface anchoring moiety linked to the N and/or C terminus. Any molecule that can display the binding proteins on the surface of a cell can be employed including, without limitation, cell surface protein and lipids. In certain embodiments, the anchoring moiety comprise the Aga2p polypeptide.
In certain embodiments, each binding protein in the library further comprises an epitope tag that that can be recognized by binding protein (e.g., an antibody). Suitable tags include without limitation, include His, HA, c-myc, Flag, HSV, S, AcV5, E2, E, and StrepII tags.
In certain embodiments, the library of binding proteins are employed to affinity mature a reference binding protein (e.g., scDVD or scDVDFab). Accordingly, in certain embodiments, the library of binding proteins share at least 70, 75, 80, 85, 90, 95, 96, 97, 98, or 99 amino acid sequence identity with a reference binding protein (e.g., scDVD or scDVDFab). In certain embodiments, the VH1 and VH2 of the reference binding protein specifically bind to different antigens.
In another aspect, the disclosure provides libraries of polynucleotides encoding the diverse library of binding proteins. The libraries can be produced by any art recognized means. In certain embodiments, the libraries are produced by combining portions of other libraries by overlap PCR In certain embodiments, libraries are produced by combining portions of other libraries by gap repair transformation in yeast cells. In certain embodiments, the nucleic acids encoding the binding proteins are operably linked to one or more expression control elements (e.g., promoters or enhancer elements).
In another aspect, the disclosure provides libraries of expression vectors comprising the diverse library of polynucleotides disclosed herein. Any vectors suitable of expressing the binding proteins can be employed.
In another aspect, the disclosure provides a library of transformed host cells, expressing the diverse library of binding proteins disclosed herein. In certain embodiments, the individual transformed cells in the library of transformed host cells express only one species from the diverse library binding proteins.
Any cells, prokaryotic or eukaryotic, are suitable for use as host cells. In certain embodiments, the host cells are yeast including, without limitation, Saccharomyces cerevisiae, Saccharomyces carlsbergensis, Candida albicans, Candida kefyr, Candida tropicalis, Cryptococcus laurentii, Cryptococcus neoformans, Hansenula anomala, Hansenula polymorpha, Kluyveromyces fragilis, Kluyveromyces lactis, Kluyveromyces marxianus, Pichia pastoris, Rhodotorula rubra, Schizosaccharomyces pombe and Yarrowia lipolytica.
In certain embodiments, the expressed binding proteins are anchored on the surface of the host cell. Any means for anchoring can be employed. In certain embodiments, the binding proteins are anchored on the cell surface through Aga1p. This is usually achieved by the fusion of the Aga2p protein the N and/or C terminus of the binding protein.

IV. SINGLE-CHAIN MULTIVALENT BINDING PROTEIN SCREENING METHODS

In another aspect, the disclosure provides a method of selecting a binding protein (e.g., scDVD or scDVDFab) that specifically binds to a target antigen. The method generally comprises: a) providing a diverse library of transformed host cells expressing a diverse library of binding proteins disclosed herein; b) contacting the host cells with the target antigen; and c) selecting a host cell that bind to the target antigen, thereby identifying a binding protein that specifically binds to a target antigen.
In another aspect, the disclosure provides a method of selecting a binding protein that specifically binds to a first and a second target antigen simultaneously. The method generally comprises: a) providing a diverse library of transformed host cells expressing a diverse library of binding proteins disclosed herein; b) contacting the host cells with the first and second target antigen; and c) selecting a host cell that bind to the first and second target antigen, thereby identifying a binding protein that specifically binds to a first and a second target antigen simultaneously.
In certain embodiments of the foregoing methods, host cells that bind to the first and/or second antigen are selected by Magnetic Activated Cell Sorting using magnetically labeled antigen. In certain embodiments of the foregoing methods, host cells that bind to the first and/or second antigen are selected by Fluorescence Activated Cell Sorting using fluorescently labeled antigen.
Any host cells, prokaryotic or eukaryotic, are suitable for use in the foregoing methods. In certain embodiments, the host cells are yeast including, without limitation, Saccharomyces cerevisiae, Saccharomyces carlsbergensis, Candida albicans, Candida kefyr, Candida tropicalis, Cryptococcus laurentii, Cryptococcus neoformans, Hansenula anomala, Hansenula polymorpha, Kluyveromyces fragilis, Kluyveromyces lactis, Kluyveromyces marxianus, Pichia pastoris, Rhodotorula rubra, Schizosaccharomyces pombe and Yarrowia lipolytica.
In certain embodiments, the expressed binding proteins are anchored on the surface of the host cell. Any means for anchoring can be employed. In certain embodiments, the binding proteins are anchored on the cell surface through Aga1p. This is usually achieved by the fusion of the Aga2p protein to one or more chain of the binding protein.
After selection of antigen-binding host cells, the polynucleotides encoding the binding proteins expressed by those cells can be isolated using any standard molecular biological means. These polynucleotides can be isolated and re-expressed in another cellular or acellular system as desired. Alternatively, these polynucleotides can be further modified and screened using the methods disclosed herein. In certain embodiments, the isolated polynucleotides are recombined with other polynucleotides (including libraries disclosed herein) to produce new, hybrid polynucleotides encoding novel binding proteins.
In certain embodiments, multiplex methods of screening libraries are employed. In such methods, each individual library is barcoded by one or more epitope tags that allows for differentiating one library or a subgroup or libraries from another library or a subgroup of libraries. Unique tag or tags are peptide sequences attached at the N-, C-, or both termini, or in the linker between VH and VL domains. The libraries are differentiated by binders (e.g., antibodies) to the epitope tags using flow cytometry or fluorescence activated cell sorting. The method of differentiation of libraries can be additive (a library having one or more tags distinct from the others) or subtractive (a library missing one ore more tags from the others). The libraries can be kept separately or combined (i.e. multiplexed) for analysis or cell sorting.
In the multiplex methods, the libraries are generally introduced to organisms that are amenable to magnetic and fluorescent activated cell sorting including, but not limited to, bacteria, yeast, and mammalian cells.
The libraries separated and distinguished by one or more tags can differ according to one or more of the following attributes: 1) antibody germline subgroups or sequences, light chain isotypes (kappa vs. lambda), or combinations thereof (e.g. specific VH/VL pairs); 2) natural or synthetic (or a combination thereof) antibody or TCR sequences; 3) cell type (B, T, plasma cells, etc); 4) tissues (peripheral blood, spleen, lymph node, bone marrow, tonsil, cord blood, etc); 5) species (human, mouse, rat, llama, rabbit, chicken, hamster, shark, etc); 6) protein scaffolds (antibodies, T cell receptors, etc); ormats (antibody and its fragments scFv, Fab, dAb, DVD-Ig, DVD-Fab, scDVD, scDVDFab, etc); 7) diversity and locations (framework vs. CDR diversity, HCDR3 size and diversity, HC vs. LC diversity, DVD-Ig linkers, domain orientation, etc; and/or 8) operation logistics (operators, lab locations, cell sorters, etc)
In certain embodiments, multiple diverse libraries are created, where each library contains clones that vary at a different discreet region of a reference binding protein. Each library is then screened separately for binding to the desired antigen(s) and the selected clones from each library are recombined to from a new library for screening. For example, to facilitate the affinity maturation of a reference binding protein, two distinct, diverse libraries can be created: a first diverse library in which only the HCDR1 and HCDR2 regions of a reference antibody are varied; and a second diverse library in which only the HCDR3 region of a reference antibody are varied. The first and the second library can be screened using the methods disclosed herein (e.g., using yeast display) to identify binding molecules with improved antigen binding characteristics. The polynucleotides encoding the selected binding proteins can then be recombined (e.g., by overlap PCR or yeast GAP repair) to form a third library comprising the HCDR1 and HCDR2 regions from the first library and the HCDR3 regions form second library. This third library can then be screened using the methods disclosed herein to identify binding proteins with further improved antigen binding characteristics. Exemplary libraries and methods are set forth in FIGS. 8 and 9.
Binding proteins selected using the methods disclosed herein can be isolated and re-expressed in another cellular or acellular system as desired.

V. ENGINEERED MULTIVALENT BINDING PROTEINS

In certain preferred embodiments, the single-chain multivalent binding proteins produced using the methods and compositions disclosed herein exhibit improved properties (e.g., affinity or stability) with respect to a corresponding parental reference binding protein. For example, the engineered binding protein may dissociate from its target antigen with a k_offrate constant of about 0.1 s⁻¹or less, as determined by surface plasmon resonance, or inhibit the activity of the target antigen with an IC₅₀of about 1×10⁻⁶M or less. Alternatively, the binding protein may dissociate from the target antigen with a k_offrate constant of about 1×10⁻²s⁻¹or less, as determined by surface plasmon resonance, or may inhibit activity of the target antigen with an IC₅₀of about 1×10⁻⁷M or less. Alternatively, the binding protein may dissociate from the target with a k_offrate constant of about 1×10⁻³s⁻¹or less, as determined by surface plasmon resonance, or may inhibit the target with an IC₅₀of about 1×10⁻⁸M or less. Alternatively, binding protein may dissociate from the target with a k_offrate constant of about 1×10⁻⁴s⁻¹or less, as determined by surface plasmon resonance, or may inhibit its activity with an IC₅₀of about 1×10⁻⁹M or less. Alternatively, binding protein may dissociate from the target with a k_offrate constant of about 1×10⁻⁵s⁻¹or less, as determined by surface plasmon resonance, or inhibit its activity with an IC₅₀of about 1×10⁻¹° M or less. Alternatively, binding protein may dissociate from the target with a k_offrate constant of about 1×10⁻⁵s⁻¹or less, as determined by surface plasmon resonance, or may inhibit its activity with an IC₅₀of about 1×10⁻¹¹M or less.
In certain embodiments, the engineered binding protein comprises a heavy chain constant region, such as an IgG1, IgG2, IgG3, IgG4, IgA, IgE, IgM or IgD constant region. Preferably, the heavy chain constant region is an IgG1 heavy chain constant region or an IgG4 heavy chain constant region. Furthermore, the binding protein can comprise a light chain constant region, either a kappa light chain constant region or a lambda light chain constant region. The binding protein comprises a kappa light chain constant region. In certain embodiments, the scDVD is reformatted into a DVD-Ig or a DVD-Fab molecule (scDVDFab).
In certain embodiments, the engineered binding protein comprises an engineered effector function known in the art (see, e.g., Winter, et al. U.S. Pat. Nos. 5,648,260; 5,624,821). The Fc portion of a binding protein mediates several important effector functions e.g. cytokine induction, ADCC, phagocytosis, complement dependent cytotoxicity (CDC) and half-life/clearance rate of binding protein and antigen-binding protein complexes. In some cases these effector functions are desirable for therapeutic binding protein but in other cases might be unnecessary or even deleterious, depending on the therapeutic objectives. Certain human IgG isotypes, particularly IgG1 and IgG3, mediate ADCC and CDC via binding to FcγRs and complement C1q, respectively. Neonatal Fc receptors (FcRn) are the critical components determining the circulating half-life of binding proteins. In still another embodiment at least one amino acid residue is replaced in the constant region of the binding protein, for example the Fc region of the binding protein, such that effector functions of the binding protein are altered.
In certain embodiments, the engineered binding protein is derivatized or linked to another functional molecule (e.g., another peptide or protein). For example, a labeled binding protein disclosed herein can be derived by functionally linking a binding protein or binding protein portion disclosed herein (by chemical coupling, genetic fusion, noncovalent association or otherwise) to one or more other molecular entities, such as another binding protein (e.g., a bispecific binding protein or a diabody), a detectable agent, a cytotoxic agent, a pharmaceutical agent, and/or a protein or peptide that can mediate associate of the binding protein with another molecule (such as a streptavidin core region or a polyhistidine tag).
Useful detectable agents with which a binding protein or binding protein portion disclosed herein may be derivatized include fluorescent compounds. Exemplary fluorescent detectable agents include fluorescein, fluorescein isothiocyanate, rhodamine, 5-dimethylamine-1-napthalenesulfonyl chloride, phycoerythrin and the like. A binding protein may also be derivatized with detectable enzymes, such as alkaline phosphatase, horseradish peroxidase, glucose oxidase and the like. When a binding protein is derivatized with a detectable enzyme, it is detected by adding additional reagents that the enzyme uses to produce a detectable reaction product. For example, when the detectable agent horseradish peroxidase is present, the addition of hydrogen peroxide and diaminobenzidine leads to a colored reaction product, which is detectable. A binding protein may also be derivatized with biotin, and detected through indirect measurement of avidin or streptavidin binding.
In other embodiment, the engineered binding protein is further modified to generate glycosylation site mutants in which the 0- or N-linked glycosylation site of the binding protein has been mutated. One skilled in the art can generate such mutants using standard well-known technologies. Glycosylation site mutants that retain the biological activity, but have increased or decreased binding activity, are another object of the present invention.
In still another embodiment, the glycosylation of the engineered binding protein or antigen-binding portion disclosed herein is modified. For example, an aglycoslated binding protein can be made (i.e., the binding protein lacks glycosylation). Glycosylation can be altered to, for example, increase the affinity of the binding protein for antigen. Such carbohydrate modifications can be accomplished by, for example, altering one or more sites of glycosylation within the binding protein sequence. For example, one or more amino acid substitutions can be made that result in elimination of one or more variable region glycosylation sites to thereby eliminate glycosylation at that site. Such aglycosylation may increase the affinity of the binding protein for antigen. Such an approach is described in further detail in PCT Publication WO2003016466A2, and U.S. Pat. Nos. 5,714,350 and 6,350,861, each of which is incorporated herein by reference in its entirety.
Additionally or alternatively, an engineered binding protein disclosed herein can be further modified with an altered type of glycosylation, such as a hypofucosylated binding protein having reduced amounts of fucosyl residues or a binding protein having increased bisecting GlcNAc structures. Such altered glycosylation patterns have been demonstrated to increase the ADCC ability of binding proteins. Such carbohydrate modifications can be accomplished by, for example, expressing the binding protein in a host cell with altered glycosylation machinery. Cells with altered glycosylation machinery have been described in the art and can be used as host cells in which to express recombinant binding proteins disclosed herein to thereby produce a binding protein with altered glycosylation. See, for example, Shields, R. L. et al. (2002) J. Biol. Chem. 277:26733-26740; Umana et al. (1999) Nat. Biotech. 17:176-1, as well as, European Patent No: EP 1,176,195; PCT Publications WO 03/035835; WO 99/54342 80, each of which is incorporated herein by reference in its entirety. Using techniques known in the art a practitioner may generate binding proteins exhibiting human protein glycosylation. For example, yeast strains have been genetically modified to express non-naturally occurring glycosylation enzymes such that glycosylated proteins (glycoproteins) produced in these yeast strains exhibit protein glycosylation identical to that of animal cells, especially human cells (U.S. patent Publication Nos. 20040018590 and 20020137134 and PCT publication WO2005100584 A2).

VI. PRODUCTION OF MULTIVALENT BINDING PROTEINS

Engineered binding proteins of the present disclosure may be produced by any of a number of techniques known in the art. For example, expression from host cells, wherein expression vector(s) encoding the heavy and light chains is (are) transfected into a host cell by standard techniques. The various forms of the term “transfection” are intended to encompass a wide variety of techniques commonly used for the introduction of exogenous DNA into a prokaryotic or eukaryotic host cell, e.g., electroporation, calcium-phosphate precipitation, DEAE-dextran transfection and the like. Although it is possible to express the binding proteins disclosed herein in either prokaryotic or eukaryotic host cells, expression of binding proteins in eukaryotic cells is preferable, and most preferable in mammalian host cells, because such eukaryotic cells (and in particular mammalian cells) are more likely than prokaryotic cells to assemble and secrete a properly folded and immunologically active binding protein.
Preferred mammalian host cells for expressing the recombinant binding proteins disclosed herein include Chinese Hamster Ovary (CHO cells) (including dhfr-CHO cells, described in Urlaub and Chasin, (1980) Proc. Natl. Acad. Sci. USA 77:4216-4220, used with a DHFR selectable marker, e.g., as described in R. J. Kaufman and P. A. Sharp (1982) Mol. Biol. 159:601-621), NS0 myeloma cells, COS cells and SP2 cells. When recombinant expression vectors encoding binding protein genes are introduced into mammalian host cells, the binding proteins are produced by culturing the host cells for a period of time sufficient to allow for expression of the binding protein in the host cells or, more preferably, secretion of the binding protein into the culture medium in which the host cells are grown. Binding proteins can be recovered from the culture medium using standard protein purification methods.
Host cells can also be used to produce functional binding protein fragments, such as Fab fragments or scFv molecules. It will be understood that variations on the above procedure are within the scope of the present disclosure. For example, it may be desirable to transfect a host cell with DNA encoding functional fragments of either the light chain and/or the heavy chain of a binding protein of this disclosure. Recombinant DNA technology may also be used to remove some, or all, of the DNA encoding either or both of the light and heavy chains that is not necessary for binding to the antigens of interest. The molecules expressed from such truncated DNA molecules are also encompassed by the binding proteins disclosed herein. In addition, bifunctional binding proteins may be produced in which one heavy and one light chain are a binding protein disclosed herein and the other heavy and light chain are specific for an antigen other than the antigens of interest by crosslinking a binding protein disclosed herein to a second binding protein by standard chemical crosslinking methods.
In a preferred system for recombinant expression of a binding protein, or antigen-binding portion thereof, disclosed herein, a recombinant expression vector encoding both the binding protein heavy chain and the binding protein light chain is introduced into dhfr-CHO cells by calcium phosphate-mediated transfection. Within the recombinant expression vector, the binding protein heavy and light chain genes are each operatively linked to CMV enhancer/AdMLP promoter regulatory elements to drive high levels of transcription of the genes. The recombinant expression vector also carries a DHFR gene, which allows for selection of CHO cells that have been transfected with the vector using methotrexate selection/amplification. The selected transformant host cells are cultured to allow for expression of the binding protein heavy and light chains and intact binding protein is recovered from the culture medium. Standard molecular biology techniques are used to prepare the recombinant expression vector, transfect the host cells, select for transformants, culture the host cells and recover the binding protein from the culture medium. Still further the disclosure provides a method of synthesizing a recombinant binding protein disclosed herein by culturing a host cell disclosed herein in a suitable culture medium until a recombinant binding protein disclosed herein is synthesized. The method can further comprise isolating the recombinant binding protein from the culture medium.

II. EXEMPLIFICATION

The present disclosure is further illustrated by the following examples which should not be construed as further limiting. The contents of Sequence Listing, figures and all references, patents and published patent applications cited throughout this application are expressly incorporated herein by reference.

Example 1. Generation of Single Chain Dual Variable Domain Molecules

The design of a scDVD molecule derived from a DVD-Ig is shown schematically in FIGS. 1A-1C. For comparison, the schematic diagrams of a DVD-Ig (FIG. 1B) and a scFv (FIG. 1C) have also been presented. The scDVD protein includes both the variable heavy and light chains of a DVD-Ig in their entirety with the carboxyl terminus of the VH domains tethered to the amino terminus of the VL domains through a Gly₄Ser peptide linker (SEQ ID NO: 54) of 30, 35, 40 or 45 amino acids. VH1 and VH2 are paired connected with a specific linker sequence of 6 to 14 amino acids. VL1 and VL2 are paired connected with a specific linker sequence (SL) of 6 amino acids. Sequences encoding the variable regions were PCR amplified from DVD-Ig expression vectors. Primers were designed in such a way that amplified DNAs have the necessary overlap sequence to perform additional overlapping PCRs. The final fragment contains the VH domains, the long Gly₄Ser linker (SEQ ID NO: 54), the VL domains and a peptide tag used to monitor expression of the scDVD on the surface of yeast. The construct is cloned by homologous recombination into a pYD yeast expression vector using DH5α chemically competent bacteria. Clones from the transformation were screened by bacteria colony PCR for the presence of the correct construct.
Several different linker sequences were evaluated for linking the VH domains or VL domains (see FIG. 2). The SL linkers correspond to the first 6 to 14 amino acids amino acids of the IgG1 constant region (ASTKGPSVFPLAPS (SEQ ID NO: 55)), or corresponding to the first 6 to 14 amino acids of the IgK constant region (RTVAAPSVFIFPPS (SEQ ID NO: 56)). The GS linkers correspond to 6 to 14 amino acids with repeats of Gly₄Ser (SEQ ID NO: 54). The RL linkers correspond to sequences of 6 to 14 amino acids rich in Proline.

Example 2. scDVD Expression on the Surface of Yeast

The expression of scDVD on the surface of yeast and the suitability of the selected epitope tags for monitoring expression were evaluated. scDVD expression on the surface of yeast was monitored by flow cytometry analysis using antibodies against scDVD epitope tags. The expression of scDVD on the surface of yeast was found to be comparable to that observed for scFv molecules, with about 50% of the yeast cells expressing the scDVD construct (FIG. 3A). However, scDVD expression shows a lower mean fluorescence intensity compared to scFv, suggesting a lower number of scDVD molecules were expressed by single cell. FIG. 3A (right dot-plot) shows this difference when two different yeast cultures (one expressing scDVD and another expressing scFv) are labeled together in the same tube. Both constructs are expressed in about 50% of the cells (data not shown) but scFv clones have a higher mean fluorescence.
The length of the long Gly₄Ser linker (SEQ ID NO: 54) did not greatly impact the ability of the cells to express the scDVD. A Gly₄Ser linker (SEQ ID NO: 54) of 30 amino acids seemed to have a negative impact on the expression while there was no difference in expression when using Gly₄Ser (SEQ ID NO: 54) of 35, 40 or 45 amino acids (FIG. 3B).

Example 3. scDVD Retains the Ability of DVD-Ig to Bind Both Targets

Two different DVD-Igs were expressed as scDVD on the surface of yeast using pYD vectors with three different tags (AcV5, E or StrepII peptide tags). Each construct was incubated with biotinylated antigens under the same conditions and concentrations. scDVD expression was monitored using epitope tags specific antibodies made in mouse, goat and rabbit, respectively. Fluorochrome labeled donkey anti-mouse, goat or rabbit antibodies were used as detection reagents. Mean fluorescence is shown in each individual dot-plot. DLL4/VEGF scDVD retains its ability to bind both DLL4 and/or VEGF (FIG. 4A). There is no difference in binding (mean fluorescence intensity) when the scDVD is incubated with DLL4, VEGF, or a mixture of the two antigens. The same findings were observed for TNF/SOST scDVD. This scDVD retains its ability to bind both TNF and/or Sclerostin (FIG. 4B). There is no difference in binding (mean fluorescence intensity) when the scDVD is incubated with TNF, SOST, or a mixture of the two antigens. Yeast cells express many copies of scDVD on the cell surface, accordingly, the simultaneous binding to both antigens could theoretically be due to some scDVD molecules on a cell binding to one antigen and other scDVD molecules on the same cell binding independently to the second antigen. However, the mean fluorescence do not change when the scDVD is incubated with one antigen, the other antigen or a mix of both antigens, suggesting that the scDVD molecules are binding both antigens simultaneously.

Example 4. scDVD Binds Both Antigens Regardless the Tag Used to Monitor its Expression on the Surface of Yeast

In yeast display, expression tags are used to monitor the antibody expression and to normalize the antigen-binding signal for expression, thus eliminating artifacts due to host expression bias. This allows for fine discrimination between mutants with different affinities towards their target. Experiments were performed to determine if any given functional DVD-Ig, when expressed as a scDVD, maintains its binding capabilities towards its two cognate targets regardless of the tag used to monitor its expression on the surface of yeast. Specifically, TNF/SOST DVD-Ig was expressed as scDVD on the surface of yeast using three different tags (AcV5, E or StrepII peptide tags). The three constructs were exposed to the same biotinylated antigens (TNF and Sclerostin) under the same conditions and concentrations. scDVD expression was monitored using tag-specific antibodies made in mouse (anti-AcV5; Abcam), goat (anti-E; Abcam) and rabbit (anti-StrepII; GeneScript). Fluorochrome labeled donkey anti-mouse (PerCP), goat (PE) or rabbit (DyLight488) antibodies were used as detection reagents (see Tables 1-3 herein). Antigen binding was monitored by APC conjugated streptavidin or Dylight633 conjugated neutravidin. All samples were analyzed by flow cytometry. FIG. 5 shows that it is feasible to use different peptide tags to monitor scDVD expression and binding on the surface of yeast.

Example 5. Binding Selection of a TNF/SOST scDVD Derived Library Demonstrate Expression and Binding Improvement Compare with the Parental scDVD

In order to test the ability of scDVD format expressed on the surface of yeast to enhance and affinity mature DVD-Ig, an affinity maturation of a TNF/SOST DVD-Ig was performed using different libraries. These libraries were constructed to contain limited mutations in different CDRs of SOST variable domains. The TNF/SOST scDVD protein sequence is set forth in FIG. 6A. To design these libraries hypermutated CDR residues were identified from other human antibody sequences. The corresponding SOST CDR residues were then subjected to limited mutagenesis by PCR with primers having low degeneracy (79% parental nucleotide and 21% all other three nucleotides) at these positions to create three antibody libraries in the scDVD format suitable for yeast surface display. The first library (H1+H2) contained mutations in HCDR1 and HCDR2 of SOST VH domain. The second library (H3) contained mutations in HCDR3 of SOST VH domain and the third library (LC) contained mutations in all CDRs of SOST VL domain. To further increase the identity of SOST variable domains to the human germline framework sequence, a binary degeneracy (50% parental 50% germline) at certain positions were introduced into the libraries and certain residues were germline (see FIG. 6B). The introduced changes were as follows:

H1+H2 Library:

Limited mutagenesis of residues: D30, D31, S52, H53, G54, D55, F56 and D58
Germlining 7 residues: G16R, T23A, S74A, T77S, G82bS, M87T, I89L

H3 Library:

Limited mutagenesis of residues: N95, N96, R97, G98, Y99, G100, G100a, L100b
Germlining 7 residues: G16R, T23A, S74A, T77S, G82bS, M87T, I89L
Binary degeneracy between SOST VH and germline at G94K

LC Library:

Limited mutagenesis of residues: S27, S30, T32, S40, S94
NNK randomization at residues N95a, G95b and S95c
Binary degeneracy between SOST VL and germline at G3V
These libraries (see FIG. 6B) were separately transformed and displayed on yeast cells and selected against low concentration of biotinylated Sclerostin and TNF by magnetic then fluorescence activated cell sorting. Each library was differently tagged by one of StrepII, FLAG or E peptide tags. scDVD expression and antigen binding were monitored by flow cytometry as described above using the antibodies described on Tables 2 and 3 herein.
After 2 and 4 rounds of selection, the binding towards Sclerostin was notably improved compared to the binding of the parental molecule. Parental TNF/SOST scDVD binds to 300 nM of Sclerostin after an incubation for 1 hour at 37° C. No binding was observed when the parental molecule was incubated with 30 nM of Sclerostin. In contrast, after 2 rounds of selection the H3 library shows binding to 30 nM of Sclerostin, and after 4 round of selection the binding to 30 nM of Sclerostin is observed when the library output was incubated only for 20 minutes at room temperature (see FIG. 6C). Similar improvements were observed for the H1+H2 and LC libraries.
Once the diversity of each library is reduced to about 10³the plasmid DNA from each output was isolated and the libraries are recombined by PCR into a new library (rHC+LC). This library was transformed into yeast cells and displayed on cell surfaces to be selected against biotinylated Sclerostin. After selection the improvement in affinity is very notorious. As pointed out the parental construct is able to bind Sclerostin at 300 nM when incubated for 1 hour at 37° C. rHC+LC library output after 6 round of selection is able to bind 0.1 nM of Sclerostin when incubated only for 20 seconds at 4° C. (FIG. 6D). Although, no formal quantification of the affinity is done, an improvement of more than 100 folds is expected based on this results. It is clear that scDVD based libraries could be selected and enriched for better binders.

Example 6. Binding Selection of TNF/SOST scDVD Libraries Shows Enrichment of SL Linkers Between VL Domains

As discussed above, there is a clear need for linker engineering during the construction and optimization of DVD-Ig antibodies. Steric hindrance due to the proximity of the outer variable domain to the ligand binding site of the inner VD could, at least partially, be responsible for a reduced affinity of a domain when engineered as the inner variable domain. Accordingly, experiments were performed to determine if the scDVD approach could be used to engineer linkers to pair VHs or VLs in a DVD-Ig. To this end, a TNF/SOST scDVD library was made by introducing 12 different linkers: four SL linkers corresponding to the first 6, 8, 10 and 12 amino acids amino acids of the IgK constant region; four GS linkers with repeats of Gly₄Ser (SEQ ID NO: 54) of 6, 8, 10 and 12 amino acids; and four proline-rich RL linkers corresponding to 6, 8, 10 and 12 amino acids (see FIG. 7A). Additionally, residues S94, N95a, G95b and S95c of the LCDR3 of SOST VL were mutated by NNK randomization. After four rounds of selection using different concentrations of Sclerostin under different conditions, the library output showed enrichment in RL linkers especially of the longest size (12 and 10 amino acids; between 3 to 7 folds). Also, the GS linkers were significantly reduced (between 6 to 8 fold) (see FIG. 7B). This data clearly demonstrates that scDVD-based yeast surface display allows for the optimization and engineering of linkers to pair VHs or VLs.

TABLE 1

Peptide tags used on a panel
of yeast expression vectors

		SEQ		SEQ
Peptide	DNA	ID	Protein	ID	pYDsTEV
Tag	sequence	NO:	sequence	NO:	vectors

HIS*	CATCATCA	74	HHHHHH	85
	CCATCACC
	AT

V5	GGTAAGCC	75	GKPIPNPL	86	13767_pYDs_
	TATCCCTA		LGLDST		TEV_total
	ACCCTCTC
	CTCGGTCT
	CGATTCTA
	CG

c-MYC	GAACAAAA	76	EQKLISEE	87	pYDsTEV_c-MYC
	ACTTATTT		DL
	CTGAAGAA
	GATCTG

HA	TACCCATA	77	YPYDVPDY	88	pYDsTEV_HA
	CGATGTTC		A
	CGGATTAC
	GCT

HSV	AGCCAGCC	78	SQPELAPE	89	pYDsTEV_HSV
	AGAACTCG		DPED
	CTCCTGAA
	GACCCAGA
	GGAC

FLAG	GACTACAA	79	DYKDDDDK	90	pYDsTEV_FLAG
	GGACGACG
	ACGACAAG

StrepII	TGGAGCCA	80	WSHPQFEK	91	pYDsTEV_
	TCCGCAGT				StrepII
	TTGAGAAG

E2	TCCAGCAC
	81	SSTSSDFR	92	pYDsTEV_E2
	CTCGAGTG		DR
	ATTTTCGA
	GATCGC

S	AAGGAAAC	82	KETAAAKF	93	pYDsTEV_S
	CGCGGCTG		ERQHMDS
	CCAAGTTT
	GAACGCCA
	GCATATGG
	ATAGC

E	GGAGCGCC	83	GAPVPYPD	94	pYDsTEV_E
	TGTACCAT		PLEPR
	ATCCGGAT
	CCGCTGGA
	ACCGCGC

AcV5	AGCTGGAA	84	SWKDASGW	95	pYDsTEV_AcV5
	GGATGCGA		S
	GCGGCTGG
	AGC

*HIS tag is present in all pYDsTEV vectors downstream of all others tags.

TABLE 2

Commercially available anti-peptide tags antibodies
used to monitor ScDVD antibody expression on yeast.

Tag	Ab Source	Clone	Source	Catalog #

S	Mouse	SBSTAGa	Abcam	ab24838
S	Rabbit	Polyclonal		ab18588
AcV5	Mouse	AcV5	Abcam. Rabbit S tag	ab49581
			antibody
E2	Mouse	5E11	Abcam. AcV5 tag	ab977
			antibody
E	Rabbit	Polyclonal	Abcam T7 tag ®	ab3397
E	Goat	Polyclonal	Abcam	ab95868
E	Chicken	Polyclonal		ab18695
StrepII	Mouse	Strep-tag	Abcam. E tag antibody	MCA2489
StrepII	Rabbit	Polyclonal	Abcam. E tag antibody	A00626
HA	Mouse	HA-7	Sigma	H9658
HA	Goat	Polyclonal	Abcam	ab9134
HA	Rat (IgG1)	3F10	Roche	11-867-423
c-myc	Mouse	9E10	Sigma	M4439
c-myc	Rabbit	Polyclonal	Sigma	C3956
Flag	Mouse	M2	Sigma	F3165
Flag	Rabbit	Polyclonal	Sigma	F7425
HSV	Rabbit	Polyclonal	Sigma	H6030

TABLE 3

Commercially available secondary reagents used to monitor scFv antibody
expression and binding on the surface of yeast

Secondary reagent	Fluorocrome	Source	Catalog #

F(ab′)2 Frag. Donkey Anti-Rat IgG	PerCp	Jackson	712-126-150
		ImmunoResearch
F(ab′)2 Frag. Donkey Anti-Goat IgG	R-PE	Jackson
		ImmunoResearch
F(ab′)2 Frag. Donkey Anti-Rabbit IgG	DyLight-488	Jackson	705-116-147
		ImmunoResearch
F(ab′)2 Frag. Goat Anti-Rabbit IgG	R-PE	Jackson
		ImmunoResearch
F(ab′)2 Frag. Goat Anti-Rabbit IgG	Alexafluor	488	Invitrogen	711-486-152
Chicken anti mouse IgG (H + L)	PerCP	Jackson	111-116-144
		ImmunoResearch
F(ab′)2 Frag Donkey Anti-Mouse IgG	Alexafluor 633	ThermoScientific	715-126-151

Example 7. Generation of a Single Chain Dual Variable Domain Fab (scDVDFab) Including Constant Regions

Another design of a scDVDFab antibody derived from a DVD-Ig is shown schematically in FIGS. 10A-10C. For comparison, the schematic diagrams of a DVD-Ig (FIG. 10B) and a scDVD (FIG. 10C) have also been presented. In this example, the scDVDFab protein includes the variable heavy (VH) and light (VL) chains of a DVD-Ig in their entirety with the CH1 region of the heavy chain and the kappa constant region (Cκ) of the light chain. As shown in FIG. 10A, The VL domains fused to the Cκ are tethered to the VH domains fused to the CH1 through a GS-rigid peptide linker of 41, 49, 57 or 65 amino acids from the carboxyl terminus of the Ck region to the amino terminus of the VH domains. These linkers are shown in greater detail below. VL1 and VL2 are paired connected with specific linkers already described and used in DVD-Igs and scDVD. The same is for VH1 and VH2 pair. FIG. 11A contains a schematic representation of a scDVDFab linear sequence.
Sequences encoding the variable regions were PCR amplified from the DVD-Ig expression vectors. Primers were designed in such a way that amplified DNAs had the necessary overlap sequence to perform additional overlapping PCRs. The final fragment contained the linear sequence represented in FIG. 11A plus a peptide tag used to monitor expression of the scDVDFab on the surface of yeast. The construct was cloned by homologous recombination into a pYD yeast expression vector using DH5a chemically competent bacteria. Clones from the transformation were screened by bacteria colony PCR for the presence of the right construct.
GS-Rigid Linkers
The GS-rigid linkers were made by combinations of different Gly/Ser segments and proline rich rigid segments. The sequences of the linkers are below and a GS-rigid linker scheme could be found in FIG. 11B. More specifically the GS-rigid linkers are composed as follows:
N-terminus-G₃SG₃-left rigid segment-G₂SG₂-right rigid segment-G₃SG₃-C-terminus (“G₃SG₃” disclosed as SEQ ID NO: 96 and “G₂SG₂” disclosed as SEQ ID NO: 97)
where the rigid segments vary in length and amino acid composition. The following rigid segments have been tested:

	Right rigid segment in the linkers:
	(SEQ ID NO: 98)
	TPAPLPAPLPT 11 AA

	(SEQ ID NO: 99)
	TPAPTPAPLPAPLPT 15 AA

	(SEQ ID NO: 100)
	TPAPLPAPTPAPLPAPLPT 19 AA

	(SEQ ID NO: 101)
	TPAPLPAPLPAPTPAPLPAPLPT 23 AA

	Left rigid segments in the linkers:
	(SEQ ID NO: 5)
	TPLPAPLPAPT 11 AA

	(SEQ ID NO: 6)
	TPLPTPLPAPLPAPT 15 AA

	(SEQ ID NO: 7)
	TPLPAPLPTPLPAPLPAPT 19 AA

	(SEQ ID NO: 8)
	TPLPAPLPAPLPTPLPAPLPAPT 23 AA

	41 aminoacids GS-rigid linker:
	(SEQ ID NO: 1)
	GGGSGGGTPLPAPLPAPTGGSGGTPAPLPAPLPTGGGSGGG

	49 aminoacids GS-rigid linker:
	(SEQ ID NO: 2)
	GGGSGGGTPLPTPLPAPLPAPTGGSGGTPAPTPAPLPAPLP
	TGGGSGGG


	57 aminoacids GS-rigid linker:
	(SEQ ID NO: 3)
	GGGSGGGTPLPAPLPTPLPAPLPAPTGGSGGTPAPTPAPTP
	APLPAPLPTGGGSGGG


	65 aminoacids GS-rigid linker:
	(SEQ ID NO: 4)
	GGGSGGGTPLPAPLPAPLPTPLPAPLPAPTGGSGGTPAPTP
	APTPAPTPAPLPAPLPTGGGSGGG

Example 8. scDVDFab Expression on the Surface of Yeast

scDVDFab were expressed on the surface of yeast and the selected peptide tags were suitable for monitoring its expression. ScDVDFab expression on the surface of yeast was monitored by flow cytometry analysis and antibodies were used to detect peptide tags. A DVD-Ig was expressed as scDVDFab on the surface of yeast using pYD vectors and 4 different GS-rigid linkers. The expression of scDVDFab on the surface of yeast was comparable to that observed for scFv molecules reaching more than 50% of the yeast cells expressing the construct (FIG. 12). The length of the GS-rigid linker did not impact the ability of the cells to express the scDVDFab.

Example 9. ScDVDFab Retained the Ability of DVD-Ig to Bind Both Targets

Functional DVD-Ig expressed as scDVDFab maintained its binding capabilities towards its two targets on the surface of yeast. A DVD-Igs was expressed as scDVDFab on the surface of yeast using pYD vectors. Aliquots of the yeast culture were incubated with biotinylated antigens. scDVDFab expression was monitored by purified tag-specific antibodies. Fluorochrome labeled secondary antibodies were used as detection reagents. IL-1B/IL17 scDVDFab retains its ability to bind both IL1B and/or IL17 (FIG. 13).

Example 10. Binding to Both Targets is Comparable Between scDVDFab and DVD-Fab Formats Expressed on the Surface of Yeast

scDVDFab constructs bound both antigens in a similar way as the DVD-Fab bind them. A DVD-Ig was expressed as scDVDFab and DVD-Fab on the surface of yeast using pYD vectors. Aliquots of the yeast culture were incubated with biotinylated antigens. scDVDFab and DVD-Fab expression was monitored by purified tag-specific antibodies. Fluorochrome labeled secondary antibodies were used as detection reagents. The scDVDFab and DVD-Fab had similar binding profiles binding to both IL1B and IL17 on the surface of yeast. There is a small increase in the mean fluorescence of scDVDFab compared to DVD-Fab (FIG. 14).

Claims

We claim:

1. A single chain multivalent binding protein having the general formula VH1-(X1)n-VH2-X2-VL1-(X3)n-VL2, wherein VH1 is a first antibody heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, X2 is a linker, VL1 is a first antibody light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites.

2. A single chain multivalent binding protein having the general formula (VL1-(X1)n-VL2-X2-VH1-(X3)n-VH2, wherein VL1 is a first antibody light chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, X2 is a linker, VH1 is a first antibody heavy chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, and n is 0 or 1, and wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding site.

3. The binding protein of claim 1 or 2 which is a single-chain dual variable domain immunoglobulin molecules (scDVD).

4. The binding protein of any one of the preceding claims, further comprising a cell surface anchoring moiety linked to the N and/or C terminus.

5. The binding protein of claim 4, wherein the anchoring moiety comprises the Aga2p polypeptide.

6. A polynucleotide encoding a binding protein of any one of the preceding claims.

7. A host cell expressing a binding protein of any one of the preceding claims.

8. A diverse library of binding proteins comprising a polypeptide chain having the general formula VH1-(X1)n-VH2-X2-VL1-(X3)n-VL2, wherein VH1 is a first heavy chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VH2 is a second heavy chain variable domain, X2 is a linker, VL1 is a first light chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VL2 is a second light chain variable domain, and n is 0 or 1, wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites, and wherein the amino acid sequences of VH1, X1, VH2, X2, VL1, X3, and/or VL2 independently vary within the library.

9. A diverse library of binding proteins comprising a polypeptide chain having the general formula (VL1-(X1)n-VL2-X2-VH1-(X3)n-VH2, wherein VL1 is a first antibody light chain variable domain, X1 is a linker with the proviso that it is not a constant domain, VL2 is a second antibody light chain variable domain, X2 is a linker, VH1 is a first antibody heavy chain variable domain, X3 is a linker with the proviso that it is not a constant domain, VH2 is a second antibody heavy chain variable domain, and n is 0 or 1, wherein the VH1 and VL1, and the VH2 and VL2 respectively combine to form two functional antigen binding sites, and wherein the amino acid sequences of VL1, X1, VL2, X2, VH1, X3, and/or VH2 independently vary within the library.

10. The diverse library of claim 8 or 9, wherein each binding proteins further comprises a cell surface anchoring moiety linked to the N or C terminus.

11. The diverse library of claim 10, wherein the anchoring moiety is a cell surface protein.

12. The diverse library of claim 10, wherein the anchoring moiety is Aga2p.

13. The diverse library of any one of the preceding claims, wherein the polypeptide chain is a scDVD.

14. The library of any one of the preceding claims, wherein the amino acid sequence of at least one CDR of VH1, VH2, VL1 or VL2 independently varies within the library.

15. The library of any one of the preceding claims, wherein the amino acid sequence of HCDR3 of VH1, VH2 independently vary within the library.

16. The library of any one of the preceding claims, wherein the amino acid sequence of HCDR1 and HCDR2 of VH1 or VH2 independently vary within the library.

17. The library of any one of the preceding claims, wherein the amino acid sequence of HCDR1, HCDR2 and HCDR3 of VH1 or VH2 independently vary within the library.

18. The library of any one of the preceding claims, wherein the amino acid sequence of HCDR3 of VL1 or VL2 independently vary within the library.

19. The library of any one of the preceding claims, wherein the amino acid sequence of HCDR1 and HCDR2 of VL1 or VL2 independently vary within the library.

20. The library of any one of the preceding claims, wherein the amino acid sequence of HCDR1, HCDR2 and HCDR3 of VL1 or VL2 independently vary within the library.

21. The library of any one of the preceding claims, wherein X1 independently varies within the library and wherein X1 is selected from the amino acid sequences set forth in FIG. 2.

22. The library of any one of the preceding claims, wherein X2 independently varies within the library and wherein X2 is (G₄S)n, where n=1-10.

23. The library of any one of the preceding claims, wherein X3 independently varies within the library and wherein X3 is selected from the amino acid sequences set forth in FIG. 2.

24. The library of any one of the preceding claims, wherein the library of binding proteins share at least 70, 75, 80, 85, 90, 95, 96, 97, 98, or 99 amino acid sequence identity with a reference binding protein.

25. The library of any one of the preceding claims, wherein VH1 and VH2 of the reference binding protein specifically bind to different antigens.

26. A diverse library of polynucleotides encoding the diverse library of binding proteins of any one of the preceding claims.

27. A diverse library of expression vectors comprising the diverse library of polynucleotides of claim 26.

28. A library of transformed host cells, expressing the diverse library of binding proteins of any one of the preceding claims.

29. The library of transformed host cells of claim 28, wherein the binding proteins are anchored on the cell surface.

30. The library of transformed host cells of claim 28, wherein the binding proteins are anchored on the cell surface through Aga1p.

31. The library of transformed host cells of claim 28, wherein the host cells are eukaryotic.

32. The library of transformed host cells of claim 31, wherein the host cells are yeast.

33. The library of transformed host cells of claim 31, wherein the yeast is selected from the group consisting of Saccharomyces cerevisiae, Saccharomyces carlsbergensis, Candida albicans, Candida kefyr, Candida tropicalis, Cryptococcus laurentii, Cryptococcus neoformans, Hansenula anomala, Hansenula polymorpha, Kluyveromyces fragilis, Kluyveromyces lactis, Kluyveromyces marxianus, Pichia pastoris, Rhodotorula rubra, Schizosaccharomyces pombe and Yarrowia lipolytica.

34. The library of transformed host cells of claim 31, wherein the yeast is Saccharomyces cerevisiae.

35. A method of selecting a binding protein that specifically binds to a target antigen, the method comprising:

a) providing a diverse library of transformed host cells expressing the diverse library of binding proteins of any one of claims 8-25;

b) contacting the host cells with the target antigen; and

c) selecting a host cell that bind to the target antigen, thereby identifying a binding protein that specifically binds to a target antigen.

36. A method of selecting a binding protein that specifically binds to a first and a second target antigen simultaneously, the method comprising:

b) contacting the host cells with the first and second target antigen; and

c) selecting a host cell that bind to the first and second target antigen, thereby identifying a binding protein that specifically binds to a first and a second target antigen simultaneously.

37. The method of claim 35 or 36, wherein host cells that bind to the first and/or second antigen are selected by Magnetic Activated Cell Sorting using magnetically labeled antigen.

38. The method of claim 35, 36, or 37, wherein host cells that bind to the first and/or second antigen are selected by Fluorescence Activated Cell Sorting using fluorescently labeled antigen.

39. The method of any one of claims 35-38, further comprising isolating the binding protein-encoding polynucleotide sequences from the host cells selected in step (c).

40. A method of producing a binding protein, comprising expressing in a host cell a binding protein that was selected using the methods of any of claims 8-25.

41. A method of producing a diverse library of binding proteins that specifically binds to a target antigen, the method comprising:

a) providing a first diverse library of scDVD molecules, wherein the amino acid sequence of a first region of the scDVD molecules is varied in the library, and wherein each member of the library binds to the target antigen;

b) providing a second diverse library of scDVD molecules, wherein the amino acid sequence of a second region of the scDVD molecules is varied in the library, and wherein each member of the library binds to the target antigen;

c) recombining the first and second libraries to produce a third diverse library of scDVD molecules, wherein the third library comprises the first regions from the first library and the second region from the second library, thereby producing a diverse library of binding proteins that specifically binds to a target antigen.

42. The method of claim 41, wherein the first and second libraries are recombined by yeast gap repair of polynucleotides encoding the libraries.