Hetero-dimeric multi-specific antibody format
11236150 · 2022-02-01
Assignee
Inventors
Cpc classification
A61P29/00
HUMAN NECESSITIES
C07K16/2866
CHEMISTRY; METALLURGY
A61P37/06
HUMAN NECESSITIES
C07K16/2809
CHEMISTRY; METALLURGY
C07K2317/60
CHEMISTRY; METALLURGY
C07K16/00
CHEMISTRY; METALLURGY
C07K2317/62
CHEMISTRY; METALLURGY
International classification
C07K16/00
CHEMISTRY; METALLURGY
C07K16/24
CHEMISTRY; METALLURGY
Abstract
The invention relates to a novel hetero-dimeric multi-specific format of multiple antibody variable domains comprising a core of two split variable domain pairs wherein both variable light domains and the two cognate variable heavy domains are positioned in tandem on two separate protein chains, respectively.
Claims
1. A hetero-dimeric protein comprising a first and a second single-chain protein, wherein said first single-chain protein comprises a first amino acid sequence consisting of (from the N- to the C-terminus): (ia) a first VL domain; (iia) a first polypeptide linker and (iiia) a second VL domain, and wherein said second single-chain protein comprises a second amino acid sequence consisting of (from the N- to the C-terminus): (ib) a first VH domain; (iib) a second polypeptide linker and (iiib) a second VH domain, and wherein said first VL domain forms a first cognate pair of variable domains with specificity to a first target antigen with either said first or said second VH domain, and said second VL domain forms a second cognate pair of variable domains with specificity to a second target antigen with the other of said VH domains, and wherein at least one of said first or said second single-chain protein further comprises: (iv) a third functional domain that is fused via a third polypeptide linker to said first or said second amino acid sequence; (v) at least one additional domain as a fourth functional domain that is fused via a fourth polypeptide linker to said first or said second amino acid sequence, so that said hetero-dimeric protein is at least tetraspecific; and wherein said hetero-dimeric protein does not comprise (i) a cognate pair of a first and a second immunoglobulin constant domain, wherein said first immunoglobulin constant domain is comprised in said first single-chain protein and wherein said second immunoglobulin constant domain is comprised in said second single-chain protein, and wherein said hetero-dimeric protein does not comprise (ii) any further pair of heteroassociation domains, in which one heteroassociation domain of said further pair of heteroassociation domains is located on the first single-chain protein, and the other heteroassociation domain is located on the second single-chain protein, other than said first and second cognate pairs of variable domains, wherein at least one of said VL and/or VH domains comprises human framework regions, wherein at least one of said VL domains comprises (i) human Vκ framework regions I to III; (ii) CDR domains CDR1, CDR2 and CDR3; and (iii) a framework region IV, which is a human Vλ germ line sequence for framework region IV.
2. The hetero-dimeric protein of claim 1, further comprising (vi) a fifth functional domain that is fused via a fifth polypeptide linker to said first and said second amino acid sequence; or (vi) a fifth and a sixth functional domain that are fused via a fifth and a sixth polypeptide linker, respectively, to said first and said second amino acid sequence.
3. The hetero-dimeric protein of claim 1, wherein said first polypeptide linker consists of from 5 to 20 amino acid residues.
4. The hetero-dimeric protein of claim 1, wherein (a) said first VL domain (ia) and said first VH domain (ib) form a first cognate pair of variable domains with specificity to a first target antigen, and said second VL domain (iia) and said second VH domain (iib) form a second cognate pair of variable domains with specificity to a second target antigen; or (b) said first VL domain (ia) and said second VH domain (iib) form a first cognate pair of variable domains with specificity to a first target antigen, and said second VL domain (iia) and said first VH domain (ib) form a second cognate pair of variable domains with specificity to a second target antigen.
5. The hetero-dimeric protein of claim 1, wherein said third and/or fourth functional domains are independently selected from the list of: binding domains, toxins, enzymes, hormones, signaling proteins, and albumins; particularly wherein said third and/or fourth functional domains are independently selected from binding domains; particularly wherein said binding domains are independently selected from the list of: antibody-based binding domains, particularly scFv fragments, Fab fragments and single antibody variable domains, and binding domains based on alternative scaffolds, particularly ankyrin-based domains, fynomers, avimers, anticalins and binding sites being built into constant regions of antibodies.
6. The hetero-dimeric protein of claim 1, wherein at least one of said VL and/or VH domains comprises CDR regions derived from a parental rabbit antibody.
7. The hetero-dimeric protein of claim 1, wherein the cognate pair of one of said first and said second VL and VH domains is specific for an antigen selected from the list of: a cancer target; and a target present on immune effector cells.
8. A nucleic acid sequence or two nucleic acid sequences encoding the first and the second single-chain proteins of the hetero-dimeric protein of claim 1.
9. A vector or two vectors comprising the nucleic acid sequence or the two nucleic acid sequences of claim 8.
10. A host cell or host cells comprising the vector or the two vectors of claim 9.
11. A method for producing the hetero-dimeric protein of claim 1, or the first and the second single-chain proteins of said hetero-dimeric protein, comprising (i) providing a nucleic acid sequence or two nucleic acid sequences encoding the first and the second single-chain proteins of the hetero-dimeric protein of claim 1, or a vector or two vectors comprising said nucleic acid sequence or nucleic acid sequences, expressing said nucleic acid sequence or nucleic acid sequences, or said vector or vectors, and collecting said hetero-dimeric protein, or (ii) providing a host cell or host cells comprising said vector or vectors, culturing said host cell or said host cells; and collecting said first and second single-chain proteins, or said hetero-dimeric protein, from the cell culture.
12. A pharmaceutical composition comprising the hetero-dimeric protein of claim 1 and a pharmaceutically acceptable carrier.
13. The hetero-dimeric protein of claim 1 for use in the treatment of a disease, wherein at least one of said cognate pairs of VL and VH domains, or of said third or fourth functional domain is able to specifically interact with a target of therapeutic relevance in the disease.
14. The hetero-dimeric protein of claim 7, wherein said antigen is CD3.
15. The hetero-dimeric protein of claim 1 for use in the treatment of a human disease, wherein at least one of said cognate pairs of VL and VH domains, or of said third or fourth functional domain is able to specifically interact with a target of therapeutic relevance in the disease.
16. The hetero-dimeric protein of claim 1 for use in the treatment of a human disease selected from cancer, an inflammatory and an autoimmune disease, wherein at least one of said cognate pairs of VL and VH domains, or of said third or fourth functional domain is able to specifically interact with a target of therapeutic relevance in the disease.
17. The hetero-dimeric protein of claim 1, wherein said first polypeptide linker consists of from 6 to 15 amino acid residues.
Description
FIGURES
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
DETAILED DESCRIPTION OF THE INVENTION
(13) Here we present a novel format exhibiting quantitative hetero-dimeric assembly of two protein chains containing multiple antibody variable domains. This format consists of a core of two split variable domain pairs (two Fv fragments) wherein both variable light domains and both variable heavy domains each are positioned on a separate protein chain, thereby driving hetero-dimerization of the two protein chains. Up to two additional variable domains in the scFv format with high intra- and inter-domain stability are fused to the amino- and/or the carboxyl-terminus of either peptide chain, resulting in an up to hexa-specific hetero-dimeric protein.
(14) Thus, in a first aspect the present invention relates to a hetero-dimeric protein comprising a first and a second single-chain protein, wherein said first single-chain protein comprises a first amino acid sequence consisting of (from the N- to the C-terminus): (ia) a first VL domain; (iia) a first polypeptide linker and (iiia) a second VL domain, and
wherein said second single-chain protein comprises a second amino acid sequence consisting of (from the N- to the C-terminus): (ib) a first VH domain; (iib) a second polypeptide linker and (iiib) a second VH domain, and
wherein said first VL domain forms a first cognate pair of variable domains with specificity to a first target antigen with either said first or said second VH domain, and said second VL domain forms a second cognate pair of variable domains with specificity to a second target antigen with the other of said VH domains, and wherein at least one of said first or said second single-chain protein further comprises (iv) at least one additional domain as third functional domain that is fused via a third polypeptide linker to said first or said second amino acid sequence.
(15) Throughout this specification and the claims which follow, unless the context requires otherwise, the word “comprise”, and variations such as “comprises” and “comprising”, will be understood to imply the inclusion of a stated integer, composition or step or group of integers or steps, while any additional integer, composition or step or group of integers, compositions or steps may optionally be present as well, including embodiments, where no additional integer, composition or step or group of integers, compositions or steps are present. With respect to such latter embodiments, the term “comprising” thus includes the narrower term “consisting of”.
(16) Several documents are cited throughout the text of this specification. Each of the documents cited herein (including all patents, patent applications, scientific publications, manufacturer's specifications, instructions, GenBank Accession Number sequence submissions etc.), whether supra or infra, is hereby incorporated by reference in its entirety to the extent possible under the respective patent law. Nothing herein is to be construed as an admission that the invention is not entitled to antedate such disclosure by virtue of prior invention.
(17) In the context of the present invention, the terms “VL domain” and “VH domain” refer to the variable light chain domain, and the variable heavy chain domain, respectively, of antibodies. In the context of the present invention, the term “antibody” refers to immunoglobulin molecules and immunologically active portions of immunoglobulin molecules, i.e. molecules that contain an antigen binding site that specifically binds to an antigen, i.e. including antibody portions comprising at least an antigen-binding fragment of an antibody.
(18) In the context of the present invention, an antibody, or any binding molecule in general, is considered to “specifically bind” to an antigen (in the case of an antibody), or to a cognate binding partner (in the case of a binding molecule in general), if it has a dissociation constant K.sub.D to said antigen/cognate binding partner as target of 100 μM or less, preferably 50 μM or less, preferably 30 μM or less, preferably 20 μM or less, preferably 10 μM or less, preferably 5 μM or less, more preferably 1 μM or less, more preferably 900 nM or less, more preferably 800 nM or less, more preferably 700 nM or less, more preferably 600 nM or less, more preferably 500 nM or less, more preferably 400 nM or less, more preferably 300 nM or less, more preferably 200 nM or less, even more preferably 100 nM or less, even more preferably 90 nM or less, even more preferably 80 nM or less, even more preferably 70 nM or less, even more preferably 60 nM or less, even more preferably 50 nM or less, even more preferably 40 nM or less, even more preferably 30 nM or less, even more preferably 20 nM or less, and even more preferably 10 nM or less.
(19) In the context of the present invention, the term “functional domains” refers to a proteinaceous domain having a predefined function, such as enzymatic activity or specific binding to a cognate ligand, wherein said proteinaceous domain is a structured domain having at least a secondary structure element. Methods for the determining the presence of secondary structure in polypeptides or proteins, such as X-ray crystallography, circular dichroism (CD), vibrational circular dichroism (VCD), NMR, or FT-IR, or for predicting the presence of secondary structure in polypeptides, such as PEP-FOLD (Shen et al., J. Chem. Theor. Comput. 10 (2014) 4745-4758) are well known to the practitioner in the art. In particular embodiments, said proteinaceous domain is a structured domain having a tertiary structure. In particular embodiments, said proteinaceous domain comprises at least about 20 amino acid residues (see Heitz et al., Biochemistry 38 (1999) 10615-25), particularly at least about 50 amino acid residues, more particularly at least about 100 amino acid residues.
(20) In the context of the present invention, the term “polypeptide linker” refers to a linker consisting of a chain of amino acid residues linked by peptide bonds that is connecting two domains, each being attached to one end of the linker. In particular embodiments, the polypeptide linker has a continuous chain of between 2 and 30 amino acid residues (e.g. 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 amino acid residues). In particular embodiments, the polypeptide linker is non-structured polypeptide. As mentioned above, methods for the determining the presence of secondary structure in polypeptides, such as X-ray crystallography, circular dichroism (CD), vibrational circular dichroism (VCD), NMR, or FT-IR, or for predicting the presence of secondary structure in polypeptides, such as PEP-FOLD (Shen et al., J. Chem. Theor. Comput. 10 (2014) 4745-4758) are well known to the practitioner in the art.
(21) This invention is characterized by the following: The use of antibody variable domains to create a hetero-dimeric format, where both VL are located on one protein chain while the corresponding VH are located on a second protein chain. The hetero-dimeric core domain allows appending of additional functional domains, such as binding domains, to create tri-, tetra-, penta- or hexaspecific entities. Multiple examples for highly efficient pairing of the hetero-dimeric core assembly. Simple solution to combinatorial screening of multiple binding-domain pools that share a common hetero-dimeric core domain.
(22) In a particular embodiment, the invention relates to a hetero-dimeric protein wherein said first or said second single-chain protein further comprises (v) a fourth functional domain that is fused via a fourth polypeptide linker to said first or said second amino acid sequence.
(23) In a particular embodiment, the invention relates to a hetero-dimeric protein wherein said first or said second single-chain protein further comprises (vi) a fifth functional domain that is fused via a fifth polypeptide linker to said first or said second amino acid sequence.
(24) In a particular embodiment, the invention relates to a hetero-dimeric protein wherein said first or said second single-chain protein further comprises (vii) a sixth functional domain that is fused via a sixth polypeptide linker to said first or said second amino acid sequence.
(25) In particular embodiments, said hetero-dimeric protein comprises said third and said fourth functional domain. In such embodiments, said hetero-dimeric protein is tetravalent, in particular embodiments, said hetero-dimeric protein is tetraspecific.
(26) In particular embodiments, said hetero-dimeric protein comprises said third, said fourth, said fifth and said sixth functional domain. In such embodiments, said hetero-dimeric protein is hexavalent, in particular embodiments, said hetero-dimeric protein is hexaspecific.
(27) In particular embodiments, said hetero-dimeric protein does not comprise a cognate pair of a first and a second immunoglobulin constant domain, wherein said first immunoglobulin constant domain is comprised in said first single-chain protein and wherein said second immunoglobulin constant domain is comprised in said second single-chain protein. In particular embodiments, at least one of said first and said second single-chain proteins does not comprise an immunoglobulin constant domain. In a particular embodiment, both said first and said second single-chain proteins do not comprise an immunoglobulin constant domain.
(28) In particular embodiments, said hetero-dimeric protein does not comprise a cognate pair of a first proteinaceous interaction domain comprised in said first single-chain protein and a second proteinaceous interaction domain comprised in said second single-chain protein other than the cognate pairs of (i) said first VL domain and said first VH domain and (ii) said second VL domain and said second VH domain.
(29) In particular embodiments, said first polypeptide linker consists of from 5 to 20 amino acid residues, particularly from 6 to 15 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4; and n being selected from 1, 2, 3, 4, and 5.
(30) In particular other embodiments, said first polypeptide linker consists of from 11 to 20 amino acid residues, particularly from 11 to 15 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4; and n being selected from 3, 4, and 5.
(31) In particular embodiments, said second polypeptide linker consists of from 5 to 20 amino acid residues, particularly from 6 to 15 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4; and n being selected from 1, 2, 3, 4, and 5.
(32) In particular other embodiments, said second polypeptide linker consists of from 11 to 20 amino acid residues, particularly from 11 to 15 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4; and n being selected from 3, 4, and 5.
(33) In particular embodiments, said third, fourth, fifth and/or sixth polypeptide linkers independently consist of from 8 to 20 amino acid residues, particularly from 10 to 15 amino acid residues. In particular embodiments, said polypeptide linkers independently have the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4, particularly 4; and n being selected from 1, 2, 3, 4, and 5, particularly from 2 and 3.
(34) In particular embodiments, said first VL domain (ia) and said first VH domain (ib) form a first cognate pair of variable domains with specificity to a first target antigen, and said second VL domain (iia) and said second VH domain (iib) form a second cognate pair of variable domains with specificity to a second target antigen. In such embodiment, said first and said second single-chain protein form said hetero-dimeric protein in a parallel arrangement of said single-chain proteins.
(35) In particular such embodiments, said first polypeptide linker consists of from 10 to 20 amino acid residues, particularly from 12 to 17 amino acid residues, particularly 15 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4, particularly 4; and n being selected from 1, 2, 3, 4, and 5, particularly 3.
(36) In particular such embodiments, said second polypeptide linker consists of from 10 to 20 amino acid residues, particularly from 12 to 17 amino acid residues, particularly 15 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4, particularly 4; and n being selected from 1, 2, 3, 4, and 5, particularly 3.
(37) In particular such embodiments, said third, fourth, fifth and/or sixth polypeptide linkers independently consist of from 10 to 20 amino acid residues, particularly from 12 to 17 amino acid residues, particularly 15 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4, particularly 4; and n being selected from 1, 2, 3, 4, and 5, particularly 3.
(38) In particular other embodiments, said first VL domain (ia) and said second VH domain (iib) form a first cognate pair of variable domains with specificity to a first target antigen, and said second VL domain (iia) and said first VH domain (ib) form a second cognate pair of variable domains with specificity to a second target antigen. In such embodiment, said first and said second single-chain protein form said hetero-dimeric protein in an anti-parallel arrangement of said single-chain proteins.
(39) In particular such embodiments, said first polypeptide linker consists of from 5 to 12 amino acid residues, particularly from 5 to 10 amino acid residues, particularly 6 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4, particularly 2; and n being selected from 1, 2, 3, 4, and 5, particularly 2.
(40) In particular such embodiments, said second polypeptide linker consists of from 5 to 12 amino acid residues, particularly from 6 to 10 amino acid residues, particularly 8 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4, particularly 3; and n being selected from 1, 2, 3, 4, and 5, particularly 2.
(41) In particular such embodiments, said third, fourth, fifth and/or sixth polypeptide linkers independently consist of from 10 to 20 amino acid residues, particularly from 8 to 12 amino acid residues, particularly 10 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (G.sub.mS).sub.n; with m being independently selected from 2, 3, and 4, particularly 4; and n being selected from 1, 2, 3, 4, and 5, particularly 2.
(42) In particular embodiments, said third, fourth, fifth and/or sixth functional domains are independently selected from the list of: binding domains, toxins, enzymes, hormones, signaling proteins, and albumins.
(43) In particular embodiments, said third, fourth, fifth and/or sixth functional domains are independently selected from binding domains.
(44) In particular such embodiments, binding domains are independently selected from the list of: antibody-based binding domains including but not limited to scFv, Fab and single antibody variable domains, single domain antibodies based on the VNAR structure from shark, and binding domains based on alternative scaffolds including but limited to ankyrin-based domains, fynomers, avimers, anticalins, fibronectins, and binding sites being built into constant regions of antibodies (e.g. f-star technology)
(45) In particular such embodiments, said binding domains are antibody-based binding domains selected from: single-chain Fv fragments and single antibody variable domains.
(46) In certain such embodiments, the order of variable domain in such a single chain Fv fragment is selected from (from N-terminus to C-terminus) VL-(linker)-VH and VH-(linker)-VL. In certain embodiments, the order of variable domains is the same for all single-chain Fv fragments comprised in the hetero-dimeric protein. In certain embodiments, three VL domains are linked to each other by said first polypeptide linker and one of said third, fourth and fifth polypeptide linkers, respectively, for example where a single-chain Fv fragment in the order VL-(linker)-VH is C-terminal from said first amino acid sequence. In certain embodiments, three VH domains are linked to each other by said second polypeptide linker and one of said third, fourth and fifth polypeptide linkers, respectively, for example where a single-chain Fv fragment in the order VL-(linker)-VH is N-terminal from said second amino acid sequence (see
(47) In certain other embodiments, the variable domain of any such antibody-based binding domain that is directly linked via the corresponding linker to the N- and/or the C-terminus of said first or second amino acid sequence is (a) a VH domain in case that it is fused to said first amino acid sequence, and (b) a VL domain in case that it is fused to said second amino acid sequence. Thus, a VH domain is fused to the N- and/or the C-terminus of a VL-linker-VL core region, and a VL domain is fused to the N- and/or the C-terminus of a VH-linker-VH core region (see, for example,
(48) In particular embodiments, said third, fourth, fifth and/or sixth binding domains are single-chain Fv fragments.
(49) In particular such embodiments, the polypeptide linker connecting the variable domains of said single-chain Fv fragments consists of between 15 and 25 amino acid residues, particularly 20 amino acid residues. In particular embodiments, said polypeptide linker has the sequence (GGGGS).sub.n, with n being selected from 3, 4, and 5, particularly 4.
(50) In particular embodiments, the at least one of said antibody variable domains comprises CDR regions derived from a parental rabbit antibody.
(51) In particular embodiments, at least one of said antibody variable domains comprises human framework regions.
(52) In particular such embodiments, at least one of said VL domains comprises (i) human Vκ framework regions I to III; (ii) CDR domains CDR1, CDR2 and CDR3; and (iii) a framework region IV, which is selected from
(53) a. a human Vλ germ line sequence for framework region IV, particularly a Vλ germ line sequence selected from the list of: SEQ ID NO. 16 to 22 according to WO 2014/206561;
(54) b. a Vλ-based sequence, which is (bi) a consensus Vλ sequence from human Vλ germ line sequences for framework region IV, particularly SEQ ID NO. 17 according to WO 2014/206561; or (bii) a consensus Vλ sequence from rearranged human Vλ sequences for framework region IV, particularly a Vλ consensus sequence selected from the list of: SEQ ID NO. 16 and 17 according to WO 2014/206561; and
c. a Vλ-based sequence, which has one or two mutations, particularly one mutation, compared to the closest human Vλ germ line sequence for framework region IV.
(55) In certain embodiments, the cognate pair of one of said first and said second VL and VH domains is specific for an antigen selected from the list of: a cancer target; and a target present on immune effector cells, such as CD3.
(56) In particular such embodiments, said third, fourth, fifth and/or sixth binding domains are single-chain Fv fragments with specificity for a target selected from the list of: a cancer target, and a target present on immune effector cells, such as CD3.
(57) In the context of the present application the term “target” refers to a cognate binding partner of a binding domain, such as an antigen of an antibody that is specifically bound by such binding domain.
(58) In particular embodiments, said target is a cancer target, in particular an antigen or an epitope that is present on the surface of one or more tumour cell types or tumour-associated cells in an increased concentration and/or in a different steric configuration as compared to the surface of non-tumour cells. Particularly, said cancer target is present on the surface of one or more tumour or tumour stroma cell types, but not on the surface of non-tumour cells.
(59) In other particular embodiments, said target is an antigen or epitope that is preferentially expressed on cells involved in autoimmune diseases. In other embodiments, said antigen or epitope is preferentially expressed on cells involved in an inflammatory disease.
(60) In particular embodiments, said target is a target present on immune effector cells. In particular embodiments, said target is CD3.
(61) In certain embodiments, said first and said second single-chain protein are selected from the following list, wherein VLA, VLB, VHA, and VHB correspond to said first and second VL and VH domains, and VLC, VLD, VLE, VLF, VHC, VHD, VHE, and VHF are part of single-chain fragments with a linker corresponding to said third, fourth, fifth and/or sixth functional domain, respectively, linked via third, fourth, fifth and/or sixth linkers LINKER3, LINKER4, LINKER5 and LINKER6) to the core domain (in bold letters); all constructs are written in the direction N- to C-terminus:
(62) A (parallel; 6Fvs):
(63) TABLE-US-00001 chain 1: VLC-(linker)-VHC-(LINKER3)-VLA-(LINKER1)-VLB- (LINKER4)-VLD-(linker)-VHD chain 2: VLE-(linker)-VHE-(LINKER5)-VHA-(LINKER2)-VHB- (LINKER6)-VLF-(linker)-VHF
B (anti-parallel 6Fvs):
(64) TABLE-US-00002 chain 1: VLC-(linker)-VHC-(LINKER3)-VLA-(LINKER1)-VLB- (LINKER4)-VLD-(linker)-VHD chain 2: VLE-(linker)-VHE-(LINKER52)-VHB-(LINKER2)-VHA- (LINKER6)-VLF-(linker)-VHF
C1 (anti-parallel 4 Fvs) (see
(65) TABLE-US-00003 chain 1: VLC-(linker)-VHC-(LINKER3)-VLA-(LINKER1)-VLB chain 2: VLD-(linker)-VHD-(LINKER4)-VHB-(LINKER2)-VHA
C2 (anti-parallel 4 Fvs) (see
(66) TABLE-US-00004 chain 1: VLC-(linker)-VHC-(LINKER3)-VLA-(LINKER1)-VLB chain 2: VHB-(LINKER2)-VHA-(LINKER4)-VLD-(linker)-VHD
C3 (anti-parallel 4 Fvs):
(67) TABLE-US-00005 chain 1: VLA-(LINKER1)-VLB-(LINKER3)-VLC-(linker)-VHC chain 2: VLD-(linker)-VHD-(LINKER4)-VHB-(LINKER2)-VHA
C4 (anti-parallel 4 Fvs):
(68) TABLE-US-00006 chain 1: VLA-(LINKER1)-VLB-(LINKER3)-VLC-(linker)-VHC chain 2: VHB-(LINKER2)-VHA-(LINKER4)-VLD-(linker)-VHD
D1 (parallel 4 Fvs) (see
(69) TABLE-US-00007 chain 1: VLC-(linker)-VHC-(LINKER3)-VLA-(LINKER1)-VLB chain 2: VLD-(linker)-VHD-(LINKER4)-VHA-(LINKER2)-VHB
D2 (parallel 4 Fvs):
(70) TABLE-US-00008 chain 1: VLC-(linker)-VHC-(LINKER3)-VLA-(LINKER1)-VLB chain 2: VHA-(LINKER2)-VHB-(LINKER4)-VLD-(linker)-VHD
D3 (parallel 4 Fvs):
(71) TABLE-US-00009 chain 1: VLA-(LINKER1)-VLB-(LINKER3)-VLC-(linker)-VHC chain 2: VLD-(linker)-VHD-(LINKER4)-VHA-(LINKER2)-VHB
D4 (parallel 4 Fvs):
(72) TABLE-US-00010 chain 1: VLA-(LINKER1)-VLB-(LINKER3)-VLC-(linker)-VHC chain 2: VHA-(LINKER2)-VHB-(LINKER4)-VLD-(linker)-VHD
(73) In this format the localization of two split variable heavy domains VHB and VHC on one protein chain and the two corresponding variable light domains VLB and VLC on the other protein chain (VH-VH/VL-VL) prevents the formation of intra-chain domain pairings resulting in inactive single-chain diabody (scDb)-like structures as it would be the case if the VH-VL/VH-VL orientation of the conventional diabody—similar to the design suggested by Kipriyanov et al—had been used to drive hetero-dimerization. In contrast, the VH-VH/VL-VL-orientation forces the formation of exclusively hetero-dimeric bi- to hexa-specific proteins.
(74) There is the theoretical possibility that the VH/VL domain pairing of the target A and B binding VHA-VHB/VLA-VLB core domain would result in an inactive core domain due to the inappropriate pairing of VHA with VLB and VHB with VLA resulting in VHA-VLB and VHB-VLA pairs. Unexpectedly and surprisingly, such inactive variants have not been observed so far. Without wishing to be bound by theory, dimerization could be driven towards cognate pairing due to the more efficient packing of the CDRs of cognate pairs as opposed to potential packing interferences occurring in non-matching pairings.
(75) In order to further drive the hetero-dimerization towards active pairing in the VH-VH/VL-VL core domain, the knob-into-hole or similar technologies could be applied in one or—if reciprocally applied—both VL/VH pairs of the VH-VH/VL-VL core domain. Thus, in certain embodiments, the active pairing in the VH-VH/VL-VL core domain of said hetero-dimeric protein is further supported by a technology selected from: knob-into-hole, and inter-chain cysteine bridges.
(76) In a second aspect, the present invention relates to one or two nucleic acid sequences encoding said first and a second single-chain proteins.
(77) In a third aspect, the present invention relates to one or two vectors comprising said one or two nucleic acid sequences.
(78) In a fourth aspect, the present invention relates to a host cell or host cells comprising one or two vectors.
(79) In a fourth aspect, the present invention relates to a method for producing the first and second single-chain proteins, or the hetero-dimeric protein, of the present invention, comprising (i) providing a nucleic acid or nucleic acids according to the present invention, or a vector or vectors according to the present invention, expressing said nucleic acid or nucleic acids or said vector or vectors and collecting said first and second single-chain proteins, or said hetero-dimeric protein, from the expression system, or (ii) providing a host cell or host cells of the present invention, culturing said host cell or host cells, and collecting said first and second single-chain proteins, or said hetero-dimeric protein, from the cell culture.
(80) In a fifth aspect, the present invention relates to a pharmaceutical composition comprising the hetero-dimeric protein of the present invention and a pharmaceutically acceptable carrier.
(81) In a sixth aspect, the present invention relates to the hetero-dimeric protein of the present invention for use in the treatment of a disease selected from cancer, an inflammatory and an autoimmune disease, wherein at least one of said cognate pairs of VL and VH domains, or of said third, fourth, fifth, or sixth functional domain is able to specifically interact with a target of therapeutic relevance in the corresponding disease.
(82) In a seventh aspect the present invention relates to a method for treating a patient suffering from a disease selected from cancer, an inflammatory and an autoimmune disease, comprising administering to a subject an effective amount of the hetero-dimeric protein of the present invention, wherein at least one of said cognate pairs of VL and VH domains, or of said third, fourth, fifth, or sixth functional domain is able to specifically interact with a target of therapeutic relevance in the corresponding disease.
LITERATURE
(83) R1. Skerra, A., and Plückthun, A. (1988). Assembly of a functional immunoglobulin Fv fragment in Escherichia coli. Science 240, 1038-1041. R2. Röthlisberger et al., (2005). Domain interactions in the Fab fragment: A comparative evaluation of the single-chain Fv and Fab format engineered with variable domains of different stability. J Mol Biol 347, 773-789. R3. Ridgway et al., 1996. ‘Knobs-into-holes’ engineering of antibody CH3 domains for heavy chain heterodimerization. Protein Eng. 9, 617-621. R4. Zhu (1997) Remodeling domain interfaces to enhance heterodimer formation. Protein Sci. 6, 781-788 R5. Schaefer, W., et al., 2011b. Immunoglobulin domain crossover as a generic approach for the production of bispecific IgG antibodies. Proc. Natl. Acad. Sci. U.S.A. 108, 11187-11192. R6. Holliger et al., “Diabodies”: small bivalent and bispecific antibody fragments. Proc. Natl. Acad. Sci. U.S.A. 90, 6444-6448. R7. Arndt et al., 1999. Abispecific diabody that mediates natural killer cell cytotoxicity against xeno-transplantated human Hodgkin's tumors. Blood 94, 2562-2568. R8. Kipriyanov et al., 1999. Bispecific tandem diabody for tumor therapy with improved antigen binding and pharmacokinetics. J. Mol. Biol. 293, 41-56. R9. Alt et al., 1999. Novel tetravalent and bispecific IgG-like antibody molecules combining single-chain diabodies with the immunoglobulin gamma1 Fc or CH3 region. FEBS Lett. 454, 90-94. R10. Johnson et al., 2010. Effector cell recruitment with novel Fv-based dual-affinity retargeting protein leads to potent tumor cytolysis and in vivo B-cell depletion. J. Mol. Biol. 399, 436-449. R11. De Jonge et al., (1995) Production and characterization of bispecific single-chain antibody fragments. Mol. Immunol. 32, 1405-1412. R12. Reiter et al., (1994) Engineering interchain disulfide bonds into conserved framework regions of Fv fragments: improved biochemical characteristics of recombinant immunotoxins containing disulfide-stabilized Fv. Protein Eng. 7, 697-704. R13. Pack, P., and Plückthun, A. (1992). Miniantibodies: Use of amphipathic helices to produce functional, flexibly linked dimeric Fv fragments with high avidity in Escherichia coli. Biochemistry 31, 1579-1584. R14. Schoonjans et al., Fab chains as an efficient heterodimerization scaffold for the production of recombinant bispecific and trispecific antibody derivatives. J Immunol. 2000 Dec. 15; 165(12):7050-7. R15. Orcutt et al., 2009. A modular IgG-scFv bispecific antibody topology. Pro-tein Eng. Des. Sel. 23, 221-228. R16. Wu, C. et al., 2007. Simultaneous targeting of multiple disease mediators by a dual-variable-domain immunoglobulin. Nat. Biotechnol. 25, 1290-1297. R17. “mAbs”; Köhler & Milstein, Nature. 256 (1975) 495-7 R18. Umaña et al., 1999. Engineered glycoforms of an antineuroblastoma IgG1 with optimized antibody-dependent cellular cytotoxic activity. Nat. Biotechnol. 17, 176-180 R19. Yu, Y. J. et al. Sci. Trans. Med. 3, 84ra44 (2011). R20. Hinton P R. et al., 2004. Engineered human IgG antibodies with longer serum half-lives in primates. J Biol Chem. 279(8):6213-6. R21. Spiess et al., 2015. Alternative molecular formats and therapeutic applications for bispecific antibodies. Mol Immunol. 2015 Jan. 27. R22. Davis et al., 2013. Readily isolated bispecific antibodies with native immunoglobulin format. U.S. Pat. No. 8,586,713. Regeneron Pharmaceuticals, Inc. R23. Shahied L S, et al., Bispecific minibodies targeting HER2/neu and CD16 exhibit improved tumor lysis when placed in a divalent tumor antigen binding format. J Biol Chem. 2004 Dec. 24; 279(52):53907-14. Epub 2004 Oct. 7. R24. Milstein. C and Cuello. A. C. (1983) Nature, 305, 537-54 R25. Chang et al., The dock and lock method: a novel platform technology for building multivalent, multifunctional structures of defined composition with retained bioactivity. Clin Cancer Res. 2007 Sep. 15; 13(18 Pt 2):5586s-5591s. R26. Deyev et al., (2003). Design of multivalent complexes using the barnase⋅barstar module. Nature biotechnology, 21(12), 1486-1492. R27. Pack, P., and Plückthun, A. (1992). Miniantibodies: Use of amphipathic helices to produce functional, flexibly linked dimeric Fv fragments with high avidity in Escherichia coli. Biochemistry 31, 1579-1584. R28. Halin et al. (2003). Synergistic therapeutic effects of a tumor targeting antibody fragment, fused to interleukin 12 and to tumor necrosis factor α. Cancer research, 63(12), 3202-3210. R29. D. Müller et al., Improved pharmacokinetics of recombinant bispecific antibody molecules by fusion to human serum albumin J. Biol. Chem., 282 (2007), pp. 12650-12660 R30. EP1293514 R31. Milstein C, and Cuello A C (1983) Hybrid hybridomas and their use in immunohistochemistry. Nature 305:537-540
EXAMPLES
Example 1: Construction of Multispecific Formats
(84) For the construction of the hetero-dimeric multi-specific formats that were named multispecific antibody-based therapeutics by cognate hetero-dimerization (MATCH), four well characterized variable domains were chosen that are directed against human tumor necrosis factor alpha (TNF), human interleukin-5 receptor (IL5R), human CD3 epsilon (CD3) and interleukin-23 receptor (IL23R), respectively. Based on the known binding characteristics of the respective variable domains in the scFv format, the activity and thereby correct association of cognate VL/VH pairs was assessed in the context of the multi-specific molecules. The respective variable domains in the periphery of the molecule were either located at the amino (N)-terminus or the carboxyl (C)-terminus of each protein chain as single-chain Fv (scFv) fragments, or located in the hetero-dimerization core domain. In contrast to the peripheral scFv fragments for which the VL and VH were positioned on the same protein chain, the cognate variable domains VL and VH of the core domain were located on the two different protein chains. In the examples presented below the target-binding domains located in the two core domains are directed against CD3 or TNF, respectively. The variable domains binding to IL23R or IL5R have been used for the peripheral scFv modules that were fused either to the N- or C-terminus of the core domain using a flexible amino acid linker of 10 or 15 amino acids.
(85) In order to explore different variations of the hetero-dimeric core assembly presented herein, the parallel as well as the anti-parallel orientation of the cognate variable domain pairs have been generated, each with either one or two additional scFv modules appended to the N or C-terminus of the core domain.
(86) In the antiparallel arrangement, the core domain has been constructed in the orientation VHA-VHB/VLB-VLA, from N-terminus to C-terminus of each protein chain (protein chains 1 through 9). In one embodiment a tetra-specific format is formed by an N-terminal fusion of one scFv module to each of the two protein chains (constructs consisting of protein chains 1+2). The corresponding tri-specific format contains a scFv module fused to only one of the two protein chains (Constructs 1+5). To investigate possible stabilization effects of the core domain assembly by engineered disulfide bridges, the two formats above have been generated also with a C-terminal cysteine that results in a crosslink of the cognate Fvs in the core domain of each protein chain. The respective hetero-dimeric formats consist of protein chains 3+4, for the tetra-specific format and protein chains 4+6 for the tri-specific format. In a variation of the antiparallel arrangement the scFv module located on the chain containing the tandem VH in the core domain was fused to the C-terminus instead of the N-terminus and was combined with a protein chain containing the assembled scFv module at the N-terminus resulting in a tetra-specific format (protein chains 1+7) or with protein chain containing only a core domain resulting in a tri-specific format (protein chains 5+7).
(87) In the parallel arrangement, the core domain has been constructed in the orientation VHA-VHB/VLA-VLB, from N-terminus to C-terminus of each protein chain arrangement. A tetra-specific format, with both scFv modules fused to the N-terminal side of the core domains, was generated by co-expression of the protein chains 9+10. The corresponding trispecific assembly, with a scFv module solely on the tandem VH containing chain, was generated by co-expression of protein chains 10+11.
(88) To generate the constructs outlined in Table 1 the amino acid sequences for the Fv domains and linkers were back-translated into corresponding nucleic sequences, which were de novo synthesized. The coding sequences were assembled and cloned by standard molecular biology techniques (e.g. Sambrook, J., et al., Molecular Cloning: A Laboratory Manual) into a suitable expression vector (e.g. pcDNA3.1, Invitrogen) for recombinant protein secretion.
Example 2: Expression and Purification
(89) The expression of the multispecific format assemblies was performed by co-transfection of the constructs into a suspension cell line (e.g. CHO-S Freestyle™, Invitrogen) by using a transient gene expression protocol (FreeStyle™ MAX system). The combination of the co-expressed expression vectors for the generation of the multispecific format assemblies is outlined in Table 2. After cultivation for several days the supernatant of the antibody fragment secreting cells was recovered for purification. The protein was captured on a suitable affinity resin (e.g. Capto L, GE Healthcare), washed extensively and eluted by a pH shift. The eluted protein was neutralized and buffer exchanged to yield the purified pools. The proteins were analyzed by size-exclusion high-performance liquid chromatography (SE-HPLC) (Table 3 and
(90) Using a single step affinity chromatography procedure all constructs could be eluted in a highly pure and monomeric fraction (
Example 3: Storage Stability Assessment
(91) Efficient MATCH chain dimerization was further demonstrated by the remarkable homogeneity of the protein content in protein L-purified samples. The protein was analyzed over the course of four weeks and storage at 4° C. and 37° C. with respect to oligomerization by SE-HPLC and degradation by SDS-PAGE (see
Example 4: Thermal Unfolding
(92) The midpoint of transition for the thermal unfolding of the tested constructs was determined by Differential Scanning Fluorimetry (DSF), essentially as described by Niesen (Niesen et al., Nat Protoc. 2 (2007) 2212-21). The DSF assay is performed in a qPCR machine (e.g. MX3005p, Agilent Technologies). The samples were diluted in buffer (citrate-phosphate pH 6.4, 0.25 M NaCl) containing a final concentration of 5×SYPRO orange in a total volume of 25 μL. Samples were measured in triplicates and a temperature ramp from 25-96° C. programmed. The fluorescence signal was acquired and the raw data was analyzed with the GraphPad Prism (GraphPad Software Inc.).
Example 5: Affinity Determination
(93) Binding affinities of individual target binding domains in the single-chain Fv (scFv) format as well as of the purified hetero-dimeric tetra-specific constructs to recombinant target proteins human IL-5 receptor (IL5R), human IL-23 receptor ECD (IL23R), human CD3 gamma-epsilon single-chain (CD3) were measured by surface plasmon resonance (SPR) using a MASS-1 SPR instrument (Sierra Sensors). For affinity measurements (done in HEPES running buffer: 0.01 M HEPES, 0.15 M NaCl, 0.05% Tween) human hetero-dimeric single-chain CD3γδ extracellular domain (produced in-house), human IL5R (R&D Systems), human IL23R (Trenzyme) and human TNF (Peprotech), target proteins were immobilized at 100-250 RUs using buffer systems optimized for each individual target, on a sensor chip (SPR-2 Affinity Sensor High Capacity Amine, Sierra Sensors) using a standard amine-coupling procedure. For human TNF-alpha (TNF) a standard amine sensor was used. Two-fold serial dilutions of purified hetero-dimeric tetra-specific constructs ranging from 90 to 0.703 nM were injected into the flow cells for 3 min (20 μl/min) and dissociation was allowed to proceed for 720 sec. After each injection cycle, surfaces were regenerated with a 45 second injection of 10 mM Glycine-HCl pH 1.5. Affinities were calculated by fitting sensograms of at least six concentrations, such that the average Chi.sup.2 is below 10% or R.sub.max. For TNF, no serial dilutions but only single concentration measurements at 90 nM were performed. Data is double-subtracted (reference channel and control cycle was subtracted).
(94) Affinities of hetero-dimeric tetra-specific constructs to each of the four targets were generally very similar to the affinities of the individual binding domains (scFvs) used in the tetra-specific format, including those CDRs whose immune reactivity is putatively dependent upon proper dimerization (i.e., those displayed by the dimer-forming Fvs targeting TNFα and CD3ε, respectively). This demonstrates full functionality of each variable domain in the tetra-specific constructs and confirms correct assembly of the cognate variable domain pairs.
(95) Additionally, each of the three multispecifics was capable of binding all four target antigens simultaneously, seemingly irrespective of the order of antigen-encounter, as demonstrated by SPR analysis of immobilized MATCH protein (
(96) It is important to acknowledge that while these data suggest proper inter-MATCH chain assembly, they do not necessarily indicate the absence of non-cognate variable domain associations, specifically the “inverted” pairing of MATCH chains that would produce chimeric CDR sets. It has been suggested that CDR sets influence the efficiency of VL-VH pairing, and our SE-HPLC, SDS-PAGE and SPR data would appear to suggest that cognate pairing of MATCH chains is highly favored. However, in an attempt to assess the degree of MATCH chain inverted pairing, we performed a SE-HPLC analysis of antibody and antibody-antigen complexes after incubation of the MATCH proteins with the molar equivalent of trimeric TNFα (i.e., 3-fold excess TNFα epitope). When applying this method of analysis to the parental anti-TNFα scFv (data not shown), SE-HPLC traces showed discrete peaks consistent with three distinct antibody-antigen complex populations, reflecting the disparate size of 1-, 2- and 3-times scFv:TNFα complexes. Additionally, a peak that was consistent with the presence of residual, non-complexed TNFα in solution was observed, whereas non-complexed scFv was completely absent from solution, thus validating the application of this method to identify “inactive” anti-TNFα antibody.
(97) Separation of MATCH protein and MATCH-antigen complexes was less efficient due to the larger molecular weight of the multispecific molecules. However, our results (
(98) TABLE-US-00011 TABLE 1 Constructs Protein Core Core chains Linker 1 domain Fv 1 Linker 2 domain Fv 2 Linker 3 1 scFv (αlL23R) GGGGSGGGGS VL (αTNFa) GGSGGS VL (αCD3) (SEQ ID (SEQ ID NO: 1) NO: 3) 2 scFv (αlL5R) GGGGSGGGGS VH (αCD3) GGSGGS VH (αTNFa) (SEQ ID (SEQ ID NO: 1) NO: 3) 3 scFv (αlL23R) GGGGSGGGGS VL (αTNFa) GGSGGS VL (αCD3) GSC (SEQ ID (SEQ ID NO: 1) NO: 3) 4 scFv (αlL5R) GGGGSGGGGS VH (αCD3) GGSGGS VH (αTNFa) GSC (SEQ ID (SEQ ID NO: 1) NO: 3) 5 VL (αTNFa) GGSGGS VL (αCD3) (SEQ ID NO: 3) 6 VL (αTNFa) GGSGGS VL (αCD3) GSC (SEQ ID NO: 3) 7 VH (αCD3) GGGSGGGS VH (αTNFa) GGGGSGGGGS scFv (αlL5R) (SEQ ID (SEQ ID NO: 4) NO: 1) 8 VL (αTNFa) GGSGGS VL (αCD3) (SEQ ID NO: 3) 9 scFv (αlL23R) GGGGSGGGGS VL (αTNFa) GGGGSGGG VL (αCD3) GGGGS GSGGGGS (SEQ ID (SEQ ID NO: 2) NO: 2) 10 scFv (αlL5R) GGGGSGGGGS VH (αCD3) GGGGSGGG VH (αTNFa) GGGGS GSGGGGS (SEQ ID (SEQ ID NO: 2) NO: 2) 11 VL (αTNFa) GGGGSGGG VL (αCD3) GSGGGGS (SEQ ID NO: 2)
(99) TABLE-US-00012 TABLE 2 Multispecific format assemblies Protein ID (Numab) Assembly Protein chain 1 Protein chain 2 PRO356 1 (see FIG. 1) 1 2 PRO469 2 1 5 PRO357 3 (see FIG. 2) 3 4 PRO470 4 4 6 PRO358 5 (see FIG. 3) 1 7 PRO471 6 5 7 PRO355 7 (see FIG. 4) 9 10 PRO468 8 10 11
(100) TABLE-US-00013 TABLE 3 Size exclusion chromatograms after 1-step purification Protein ID Monomer (internal) Assembly ID content FIG. PRO356 Assembly 1 93.9 5A PRO357 Assembly 3 94.4 5B PRO358 Assembly 5 93.9 5C PRO355 Assembly 7 90.4 5D
(101) TABLE-US-00014 TABLE 4 Midpoint of unfolding for the proteins determined by differential scanning fluorimetry Protein ID Assembly (internal) ID Tm [° C.] PRO356 1 67.99 PRO469 2 67.24 PRO357 3 71.27 PRO470 4 70.34 PRO358 5 68.51 PRO471 6 67.98 PRO355 7 67.33 PRO468 8 66.67
(102) TABLE-US-00015 TABLE 5 Affinity of hetero-dimeric tetra-specific constructs Affinity to Affinity to Affinity to Affinity to IL5R CD3 IL23R TNF Protein ID [M] M] [M] [M] scFvs 2.32E−10 8.57E−09 1.50E−10 2.02E−10 PRO355 1.03E−10 2.01E−08 6.54E−10 3.30E−10 PRO356 1.26E−10 7.14E−09 3.41E−10 2.01E−10 PRO357 1.28E−10 6.69E−09 3.58E−10 1.81E−10 PRO358 2.12E−10 5.60E−09 4.14E−10 2.11E−10