Binders for inhibiting formation of multimeric proteins

Abstract

To provide variants of OB-fold proteins, in particular of the Sac7d family, that are able to bind a subunit of a multimeric protein and inhibit the formation of the multimer.SOLUTION: The present invention provides a polypeptide comprising a variant of a protein of the Sac7d family, the variant comprising 4 to 22 mutated amino acids in the binding site of the protein of the Sac7d family, that binds to a subunit of a multimeric protein and that does not bind to the subunit of the protein when such subunit protein is involved in the fully formed multimeric protein in its native state.SELECTED DRAWING: None

Description

Technical Field

[0001] The present invention relates to the field of molecular biology, and more particularly to the development of novel molecules and conjugates having the ability to inhibit the formation of multimeric protein complexes.

Background Art

[0002] Introduction Some proteins with biological properties can be active as multimeric proteins, i.e., polypeptides formed by multiple polypeptide chains called subunits. Multimeric proteins or polypeptides can be formed from the same subunits (homomultimers) or from different subunits (heteromultimers). Thus, a heteromultimeric polypeptide represents a protein containing two or more different polypeptide chains, and a homomultimeric polypeptide represents a protein containing two or more polypeptide chains that are all identical.

[0003] Generally, in vivo, there is an equilibrium between monomeric subunits and soluble forms of multimeric proteins, where monomeric subunits assemble to form multimeric proteins and multimeric proteins disassemble and release their multimeric forms. The ratio between these different species is governed by a concentration-based equilibrium that shifts the multimeric form to a relatively high concentration. Such an equilibrium allows the concentration of the multimeric and active soluble form to be regulated to maintain homeostasis.

[0004] One strategy for reducing the biological activity of the soluble form of multimeric proteins, particularly their membrane-bound forms inserted and stabilized in the cell membrane, is to avoid the formation of biologically active multimeric proteins or to shift the equilibrium towards the monomeric (or multimeric-inactive) subunit side. This can be done by blocking the assembly of subunits, for example, by sterically blocking the site of multimer assembly.

[0005] Accordingly, there is interest in developing molecules that bind to the monomeric portion of a multimeric protein, inhibit multimer formation, or shift the monomer-multimer species equilibrium.

[0006] Applicants propose to produce such molecules by creating novel artificial proteins that are based on variants of the OB fold domain, which can be obtained by particular screening methods, and that bind to the subunits of multimeric proteins and can inhibit the formation of multimeric complexes.

[0007] EP1930342 (Patent Document 1) discloses a library containing variants of Sac7d and a method for screening such a library. Sac7d variants that bind to PulD are disclosed. As shown in the description of FIG. 12, such variants appear to bind to the protein isolated from the membrane and thus to the protein in its native multimeric state. EP19300342 thus discloses variants that bind to the subunits of multimeric proteins but do not bind to the subunits of the protein (i.e., to the multimeric protein) when the subunit protein is involved in a fully formed multimeric protein in its native state.

[0008] WO2017161096 (Patent Document 2) discloses molecules that are conjugates of a targeting portion, which is an antibody mimetic, and an activator. Such antibody mimetics are referred to as variants of Sac7d, and this document also shows that the target can be a member of the Sac7d superfamily. However, this document does not disclose such variants, nor does it disclose the fact that it is possible to specifically discover variants that bind to the subunits of a protein and do not bind to the native multimerized protein. Furthermore, in view of the fact that the antibody mimetic is intended to direct the activator to the target, it is clear that this document assumes variants that bind to the fully formed target (i.e., the multimeric molecule).

[0009] Dias and Roqie (Biotechnol Bioeng. 2017 Mar; 114(3); 481-491 (Non-Patent Document 1)) disclose protein scaffolds as affinity reagents for purification, but do not mention the possibility of using variants of the Sac7d family of proteins that bind to subunit proteins with epitopes involved in the formation of such multimers or sub-multimers, thereby inhibiting the formation of biologically active multimers by preventing the formation of such multimers or sub-multimers in the context of OB-fold proteins, particularly biologically active multimeric proteins.

[0010] Dienhelt et al. (PLoS One. 2010 May 19;5(5):e10728 (Non-Patent Document 2)) disclose a high-affinity synbody that specifically binds to AKT1 as an example of a method for determining high-affinity ligands for target proteins in a single discovery step. This document makes no mention of the ability of such ligands to bind only to subunits of multimers and prevent protein multimerization.

[0011] Behar et al. (J Chromatogr A. 2016 Apr 8;1441:44-51 (Non-Patent Document 3)) disclose the ability of Sac7d variants in affinity columns. This document does not teach or suggest that such variants can be designed to prevent protein multimerization and that they can bind to subunits but not to fully multimerized proteins.

[0012] In summary, none of these documents disclose or suggest the possibility of obtaining scaffolds that can bind to subunits of multimeric proteins when not involved in the multimer, but cannot bind to such subunits when involved in the multimeric protein.

Prior Art Documents

Patent Documents

[0013]

Patent Document 1

Patent Document 2

Non-Patent Document

[0014]

Non-Patent Document 1

Non-Patent Document 2

Non-Patent Document 3

Summary of the Invention

[0015] The present invention relates to OB-fold proteins, particularly variants of proteins of the Sac7d family, which inhibit the formation of biologically active multimeric proteins by binding to subunit proteins with epitopes involved in the formation of such multimers or sub-multimers, thereby preventing the formation of biologically active multimers.

[0016] The present invention also encompasses polypeptides comprising variants of such proteins of the Sac7d family.

[0017] It should be noted that this variant contains 4 to 22 mutant amino acids in the binding site of the protein of the Sac7d family and binds to the subunits of the multimeric protein, but does not bind to the subunits of the protein when the subunit protein is involved in a fully formed multimeric protein in its native state.

[0018] This specification discloses the Sac7d protein and family below, but it should be noted that this teaching is also applicable to other OB fold domains as disclosed in WO2007139397. The present invention is also applicable to the SH3 domain, a small protein domain of about 60 amino acid residues that was first reported as a conserved sequence in the viral adapter protein v-Crk and is reported under PF00018 in the PFAM database. The SH3 domain has a characteristic β-barrel fold consisting of five or six β-strands arranged as two closely packed antiparallel β-sheets. Its linker region may contain short helices. The OB fold and the SH3 domain share homology, and in light of the knowledge of the sequences and structures of these domains, it should be noted that in either the OB fold or the SH3 domain, it is possible to determine which amino acids the amino acids disclosed below for Sac7d correspond to.

[0019] In certain embodiments, the multimeric protein is a member of the TNF-α superfamily, more particularly TNF-α, RANKL or TRAIL (TNFSF10).

[0020] In certain embodiments, the variant is present within the polypeptide and is thus covalently attached by an amine bond to another protein of biological interest.

[0021] In certain embodiments, the variant is conjugated to an organic molecule that exhibits some function.

[0022] The present invention also relates to these variants for therapeutic, diagnostic or purification uses. The present invention also relates to compositions containing this variant, particularly oral or topical (dermal) compositions.

[0023] The present invention also relates to a method capable of identifying such variants that bind to the surface of a subunit involved in the formation of a multimeric protein (the surface involved in protein-protein interaction in the multimer).

[0024] According to the present invention, it is possible to obtain a polypeptide comprising a variant of a protein of the Sac7d family, wherein the variant contains 4 to 22 mutant amino acids at the binding site of the protein of the Sac7d family and binds to the subunit of the multimeric protein, but does not bind to the subunit of the protein when the subunit protein is involved in the fully formed multimeric protein in its native state. The sequence of Sac7d is: Note that it is TIFF2025108528000001.tif11143.

[0025] It is possible to obtain the above polypeptide, wherein the mutant amino acid of the Sac7d variant is selected from the group consisting of V2, K3, K5, K7, Y8, K9, G10, E11, K13, E14, T17, K21, K22, W24, V26, G27, K28, M29, S31, T33, Y34, D35, D36, N37, G38, K39, T40, G41, R42, A44, S46, E47, K48, D49, A50 and P51 of Sac7d.

[0026] It is possible to obtain the above polypeptide, wherein the Sac7d variant contains 4 to 17 mutant amino acids selected from the group consisting of K7, Y8, K9, E11, K21, K22, W24, V26, M29, S31, T33, D35, T40, G41, R42, A44 and S46 of Sac7d.

[0027] In a preferred embodiment, the multimeric protein is a member of the tumor necrosis factor superfamily, preferably TNF-α, RANKL or TRAIL.

[0028] In a preferred embodiment, the polypeptide has the sequence It includes TIFF2025108528000002.tif44143.

[0029] The polypeptide may also include SEQ ID NO:16 in which 1 to 13, preferably 1 to 11, more preferably 1 to 10, still more preferably 1 to 8, still more preferably 1 to 6, still more preferably 1 to 5, still more preferably 1 to 4, still more preferably 1 to 3, still more preferably 2, and most preferably 1 amino acid selected from the group consisting of V7, M8, F9, K11, V21, H22, M24, Q26, L29, E35, D41, F44 and P46 is substituted with another amino acid.

[0030] In another aspect, the polypeptide has the sequence It includes TIFF2025108528000003.tif31142.

[0031] In some aspects, the polypeptide is linked or fused to a variant of a protein of the Sac7d family that binds to a subunit of a multimeric protein, to another protein or polypeptide, particularly to another variant of a protein of the Sac7d family or to an antibody, particularly an antibody that binds to IL17, TNF-α or Her2 / neu.

[0032] In another aspect, the polypeptide is conjugated to an organic molecule.

[0033] It also enables the obtaining of a gene construct containing a DNA sequence encoding the above polypeptide, a vector containing such a gene construct, and a host cell containing such a gene construct within its genome.

[0034] It is also possible to carry out a method for producing a polypeptide, which includes culturing a cell culture, wherein the cell is transformed with a gene construct, and recovering the polypeptide.

[0035] Methods for obtaining and identifying proteins that bind to monomers of multimeric proteins are also available.

[0036] One of ordinary skill in the art can also use the disclosed polypeptides for use in medicaments for inhibiting multimerization of multimeric proteins or for treating the diseases shown below, as the disclosed polypeptides.

[0037] The present invention also relates to a polypeptide disclosed herein that binds to a monomer of TNF-α but does not bind to trimerized TNF-α for use in the treatment of rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, Crohn's disease, ulcerative colitis, chronic psoriasis, hidradenitis suppurativa, juvenile idiopathic arthritis, Behçet's disease or psoriasis vulgaris.

[0038] The present invention also relates to a polypeptide disclosed herein that binds to a monomer of TNF-α but does not bind to trimerized TNF-α for use in the treatment of cancer in combination with chemotherapy or treatment with CAR-T cells.

[0039] The present invention also relates to a polypeptide disclosed herein that binds to a monomer of RANKL but does not bind to trimerized RANKL for use in the prevention of bone loss (e.g., in postmenopausal patients suffering from postmenopausal osteoporosis (PMO) or in patients with cancer (especially breast cancer and prostate cancer)).

[0040] The present invention also relates to a method of treating a patient in need thereof, comprising administering a therapeutically effective amount (an amount that provides therapeutic activity) of a polypeptide disclosed herein, the polypeptide binding to a monomer of TNF-α but not binding to trimerized TNF-α, and the patient having rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, Crohn's disease, ulcerative colitis, chronic psoriasis, hidradenitis suppurativa, juvenile idiopathic arthritis, Behçet's disease or psoriasis vulgaris.

[0041] The present invention also relates to a method of treating a patient in need thereof, comprising administering a therapeutically effective amount (an amount providing therapeutic activity) of the polypeptide disclosed herein, which binds to the monomer of TNF-α but not to trimerized TNF-α, and wherein the patient is being treated for cancer by chemotherapy or using CAR-T cells.

[0042] The present invention also relates to a method of preventing or treating bone loss in a patient in need thereof, comprising administering a therapeutically effective amount (an amount providing therapeutic activity) of the polypeptide disclosed herein, which binds to the monomer of RANKL but not to trimerized RANKL, and wherein the patient is a postmenopausal patient suffering from postmenopausal osteoporosis (PMO) or a patient with cancer (particularly breast cancer and prostate cancer), and the bone loss is a side effect of such cancer.

[0043] A method of detecting the presence of or quantifying a subunit of a multimeric protein in a sample, comprising: a. exposing the sample to a variant that binds to the subunit under conditions that allow such binding; b. recovering the variant and / or detecting or measuring the amount of the subunit. is also contemplated.

[0044] A method of purifying a subunit of a multimeric protein in a sample, comprising: a. exposing the sample to a variant disclosed herein that binds to the subunit under conditions that allow such binding; b. recovering the variant bound to the subunit; c. eluting the subunit from the variant. is also contemplated. DETAILED DESCRIPTION OF THE INVENTION

[0045] Detailed Description of the Invention Sac7d protein family The Sac7d family is defined by its association with the Sac7d protein and corresponds to a family of 7 kDa DNA-binding proteins isolated from extremophilic bacteria. It is disclosed herein as a representative species of the OB fold domain that is preferably used in the context of the present invention. Since the SH3 domain shares homology with the OB fold domain, the teachings regarding Sac7d are also applicable to the SH3 scaffold.

[0046] These proteins and this family are described in particular in WO2008 / 068637. Thus, in the context of the present invention, a protein belongs to the Sac7d family when it has one of the sequences SEQ ID NO: 1 to SEQ ID NO: 14, or when it has a sequence corresponding to the consensus sequence SEQ ID NO: 15 (derived from SEQ ID NO: 1 to SEQ ID NO: 9 and SEQ ID NO: 12 to SEQ ID NO: 14, in this consensus sequence, the dash - indicates the absence of an amino acid and not all proteins have the same size). This Sac7d family is also known in particular as Sac7d or Sac7e protein from Sulfolobus acidocaldarius, Sso7d protein from Sulfolobus solfataricus, DBP 7 also called Sto7 from Sulfolobus tokodaii, Ssh7b protein from Sulfolobus shibatae, Ssh7a protein from Sulfolobus shibatae, Mse7 from Metallosphaera sedula, Mcu7 from Metallosphaera cuprina, Aho7a or Aho7b or Aho7c from Acidianus hospitalis, Sis7a or Sis7b from Sulfolobus islandicus, and p7ss protein from Sulfolobus solfataricus. In view of the extensive similarity of the sequences of the proteins of the Sac7d family, it is directly possible and easy to identify the amino acids of another protein other than Sac7d corresponding to specific amino acids of Sac7d. In particular, the person skilled in the art can also use the teaching of WO2008 / 068637 which shows (in the drawings) and describes (in the specification) that such proteins are superimposable.

[0047] Note that the number of mutant residues in the variant (relative to the wild-type protein) is preferably from 4 to 25, or more specifically from 4 to 22. The present invention preferably has at least 4, more preferably at least 5, more preferably at least 6, more preferably at least 7 or 8, even more preferably at least 10, but usually less than 25, more preferably less than 22, even more preferably less than 20 or less than 15 or 14 substituted amino acids compared to the wild-type OB-fold protein (or domain). Note that all and any ranges (e.g., 5 to 20 or 7 to 25, etc.) are contemplated in the present invention. Particularly preferred ranges are from 4 to 17 and from 6 to 17, from 4 to 14 and from 6 to 14.

[0048] It is preferred that 7, 8, 9, 10, 11, 12, 13 or 14 amino acids are mutated within the binding site of the OB-fold domain relative to the wild-type OB-fold domain. These mutations are introduced into the amino acids corresponding to V2, K3, K5, K7, Y8, K9, G10, E11, K13, E14, T17, K21, K22, W24, V26, G27, K28, M29, S31, T33, Y34, D36, N37, G38, K39, T40, R42, A44, S46, E47, K48, D49, A50 and P51 of Sac7d (SEQ ID NO:1).

[0049] In certain embodiments, the number of mutant amino acids is from 7 to 14 (including the upper and lower limits).

[0050] In one particular embodiment, these variants may also include amino acid insertions as described above.

[0051] As shown, the proteins of the Sac7d family are DBP 7, also called Sac7d or Sac7e from Sulfolobus acidocaldarius, Sso7d from Sulfolobus solfataricus, Sto7 from Sulfolobus tokodaii, Ssh7b from Sulfolobus shibatae, Ssh7a from Sulfolobus shibatae, Mse7 from Metallosphaera sedula, Mcu7 from Metallosphaera cuprina, Aho7a or Aho7b or Aho7c from Acidianus hospitalis, Sis7a or Sis7b from Sulfolobus islandicus, and p7ss from Sulfolobus solfataricus. The Sac7d, Sso7d, Sac7e, Ssh7b, Ssh7a, DBP 7, Sis7a (three alleles), Mse7, Mcu7, Aho7a, Aho7b, and Aho7c proteins are represented by SEQ ID NOs: 1 to 14, respectively.

[0052] Variants of the proteins of the Sac7d family may be called nanofitin. Thus, the present invention may preferably be carried out by variants of the proteins represented by SEQ ID NOs: 1 to 14, or proteins having the sequence SEQ ID NO: 15, particularly variants of Sac7d.

[0053] In a preferred embodiment, the mutant amino acids are selected from the group consisting of V2, K3, K5, K7, Y8, K9, G10, E11, K13, E14, T17, K21, K22, W24, V26, G27, K28, M29, S31, T33, Y34, D35, D36, N37, G38, K39, T40, G41, R42, A44, S46, E47, K48, D49, A50 and P51 (with reference to the alignment of SEQ ID NO:1 and Figure 1). It should be noted that the variant should contain mutant amino acids (preferably 7 to 12 as described above) selected from among these amino acids, and may also contain other mutant amino acids (preferably 0 to 5) in other regions (i.e., selected from among other residues of Sac7d). As described above, A59, R60, A61, E62, R63, E64 and / or K66 may be deleted.

[0054] In a preferred embodiment, the mutant amino acids are selected from the group consisting of K7, Y8, K9, E11, K21, K22, W24, V26, M29, S31, T33, D35, T40, G41, R42, A44 and S46.

[0055] The variants disclosed herein can be obtained by the methods described in WO2008 / 068637 (in particular, enrichment of various rounds of combinatorial libraries using ribosome display).

[0056] Thus, the variant of the OB fold domain is preferably a variant of a protein of the Sac7d family (Sac7d variant) that contains 5 to 20 (preferably 7 to 14) mutant amino acids at the binding site of the protein. Preferably, the mutant amino acids of the Sac7d variant are selected from the group consisting of V2, K3, K5, K7, Y8, K9, G10, E11, K13, E14, T17, K21, K22, W24, V26, G27, K28, M29, S31, T33, Y34, D35, D36, N37, G38, K39, T40, G41, R42, A44, S46, E47, K48, D49, A50 and P51 (refer to the alignment of SEQ ID NO:1 and Figure 1). Preferably, the Sac7d variant contains 4 to 17 mutant amino acids selected from the group consisting of K7, Y8, K9, E11, K21, K22, W24, V26, M29, S31, T33, D35, T40, G41, R42, A44 and S46.

[0057] The advantage of the method described in WO 2008 / 068637 is that it is possible to obtain variants of the OB fold protein by screening a combinatorial library that contains or expresses a plurality of variants in which a specific number of amino acids have been "randomized", i.e., substituted with random amino acids. Screening of these libraries makes it possible to identify variants of these proteins that specifically bind, usually with high affinity (the application WO 2008 / 068637 actually describes affinities on the order of 1 nanomolar), to a target of interest that is different from the natural ligand of the wild-type protein from which the combinatorial library was generated. A specific method for identifying variants that bind to subunits of multimeric proteins so as to prevent the formation of their multimers is described below.

[0058] Binding site of Sac7d protein OB-fold proteins are known in the art. They are described, inter alia, in the above-mentioned documents and also in Arcus (Curr Opin Struct Biol. 2002 Dec; 12(6): 794-801). The OB-fold is in a cylindrical form with five β-sheets. Most OB-fold proteins use the same binding interface for their natural ligands, which can be oligosaccharides, oligonucleotides, proteins, metal ions or catalytic substrates. This binding interface contains residues located mainly within the β-sheets. Certain residues located within loops can also be involved in the binding of an OB-fold protein to its natural ligand. Thus, applications WO 2007 / 139397 and WO 2008 / 068637 and the Arcus document (2002, supra) describe OB-fold protein domains for binding to their natural ligands.

[0059] In particular, document WO 2008 / 068637 describes in detail a method for identifying the binding domain of an OB-fold protein. By aligning various sequences and 3D structures of proteins having an OB-fold domain using the websites WU-Blast2 (http: / / www.ebi.ac.uk / blast2 / index.html) (Lopez et al., 2003, Nucleic Acids Res 31, 3795-3798), T-COFFEE (http: / / www.ch.embnet.org / software / TCoffee.html) (Notredame et al., 2000, J Mol Biol 302, 205-217) and DALI lite (http: / / www.ebi.ac.uk / DaliLite / ) (Holm and Park, 2000, Bioinformatics 16, 566-567), it is possible to identify its binding domain, in particular the positions of the amino acids that can be modified. Using the sequence of Sac7d (SEQ ID NO:1) as a reference, these are residues V2, K3, K5, K7, Y8, K9, G10, E11, K13, E14, T17, K21, K22, W24, V26, G27, K28, M29, S31, T33, Y34, D35, D36, N37, G38, K39, T40, G41, R42, A44, S46, E47, K48, D49, A50 and P51.

[0060] The binding domains of other OB-fold proteins can be identified as described in WO 2008 / 068637. This application shows that it is possible to perform superposition of the 3D structures of OB-fold proteins or domains (in this application, 10 domains including Sac7d were used) using the DALI website (http: / / www.ebi.ac.uk / dali / interactive.html) (Holm and Sander, 1998, Nucleic Acids Res 26, 316-319). Thus, for any OB-fold protein (or any OB-fold domain), it is easy to identify the amino acids involved in its binding site and corresponding to the above Sac7d amino acids. Therefore, providing amino acids that can be mutated in one of these proteins enables the identification of corresponding amino acids in any other OB-fold domain.

[0061] It is also possible to delete some amino acids from the OB-fold scaffold. Again, using the Sac7d sequence as a reference, the residues that can be deleted are A59, R60, A61, E62, R63, E64 and / or K66.

[0062] The teachings of WO 2008 / 068637 also disclose that amino acids can optionally be inserted into the loops of OB fold proteins, particularly proteins of the Sac7d family; in particular, insertions of 1 to 15 amino acid residues can be made in loop 3 (as defined in FIGS. 1b and 2 of WO 2008 / 068637), for example in the region of residues 25 - 30 of Sac7d, preferably between residues 27 - 28, insertions of 1 to 15 amino acid residues can be made in loop 4 (as defined in FIGS. 1b and 2 of WO 2008 / 068637), for example in the region of residues 35 - 40 of Sac7d, preferably between residues 37 - 38, and insertions of 1 to 20 residues can be made in loop 1 (as defined in FIGS. 1b and 2 of WO 2008 / 068637), for example in the region of residues 7 - 12 of Sac7d, preferably between residues 9 - 10.

[0063] In particular, obtaining variants of the OB fold of Sac7d protein WO 2007 / 139397 describes the use of a library of OB fold proteins in which the OB domain is modified by introducing mutations into this domain for the binding of the protein to its natural ligand. In particular and as described hereinabove, the modified OB fold domain comprises a) at least one modified amino acid residue in a β - strand of the OB fold domain binding surface, compared to the naturally occurring OB fold domain, or b) at least one modified amino acid residue in a β - strand of the OB fold domain binding surface and at least one modified amino acid residue in a strand of the OB fold domain loop region, or c) at least one modified amino acid residue in a strand of the OB fold domain loop region. Generally, the modified OB fold domain exhibits altered binding properties compared to the naturally occurring OB fold domain.

[0064] WO 2008 / 068637 describes the use of a library based on the Sac7d protein to obtain ligands having affinity for a target of interest. The method described in WO 2008 / 068637 involves generating a combinatorial library containing a plurality of DNA molecules that all have the same sequence except for the presence of specific random mutations that introduce mutations at specific amino acids in the binding site of this wild-type OB-fold protein. In particular, in WO 2008 / 068637, the wild-type OB-fold protein is the Sac7d protein, and mutations are introduced therein, in particular, at amino acids selected from K7, Y8, K9, E11, K21, K22, W24, V26, M29, S31, T33, D35, T40, G41, R42, A44 and S46 or other amino acids, such as V26, G27, K28, M29, S31, R42, A44, S46, E47 and K48. These amino acids are based on the Sac7d sequence represented by SEQ ID NO:1.

[0065] WO 2012 / 150314 shows that mutations of one protein of the Sac7d family can be transferred to another protein of the same family. This mobility allows mutants of another protein of the Sac7d family to be generated starting from mutants of one protein of that family. The first mutant can in particular be obtained by carrying out the process of WO 2008 / 068637. Thus, in particular using the teachings of WO 2012 / 150314 and FIG. 1, it is possible to obtain mutants of any protein of the Sac7d family starting from mutants of any other protein of such a family. As an example, in the case of mutants of Sac7d, a person skilled in the art only needs to use the sequence alignment of FIG. 1 to introduce the mutant amino acids of the Sac7d mutant into the scaffold of another protein.

[0066] It has already been shown that it is possible to obtain such variants for a particular target (especially when the target is a protein or a peptide). A person skilled in the art can refer to variants that bind to immunoglobulins (Behar et al., Protein Engineering, Design & Selection vol. 26 no. 4 pp. 267-275, 2013) or other proteins (WO WO 2008 / 068637). Gera et al. (J Mol Biol. 2011 Jun 17; 409(4):601-16) also showed the possibility of obtaining such proteins starting from Sso7d. Gocha et al. (Scientific Reports 7, Article number: 12021 (2017)) also reported the ability to obtain variants derived from Sso7d.

[0067] Examples of OB fold or SH3 domain Non-limiting examples of OB-fold proteins that can be used in accordance with the present invention include Sac7d, Sso7d, the N-terminal domain of SEB (Papageorgiou et al., 1998), the A chain of Shiga-like toxin IIe (PDB 2bosa), human neutrophil activating peptide 2 (NAP-2, PDB 1tvxA), the molybdenum-binding protein (modg) of Azotobacter vinelandii (PDB 1h9j), the N-terminal domain of SPE-C (Roussel et al., 1997), the B5 subunit of Escherichia coli (E. coli) Shiga-like toxin (Kitov et al., 2000), Cdc13 (Mitton-Fry et al., 2002), the cold shock DNA-binding domain of human Y-box protein YB-1 (Kloks et al., 2002), E. coli inorganic pyrophosphatase EPPase (Samygina et al., 2001), or any protein listed in Table 3 of the literature by (Arcus, 2002), for example, 1krs (lysyl-tRNA synthetase LysS, E. coli), 1c0aA (Asp-tRNA synthetase, E. coli), 1b8aA (Asp-tRNA synthetase, P. kodakaraensis), 1lylA (lysyl-tRNA synthetase LysU, E. coli), 1quqA (replication protein A, 32 kDa subunit, human), 1quqB (replication protein A, 14 kDa subunit, human), 1jmcA (replication protein A, 70 kDa subunit (RPA70) fragment, human), 1otc (telomere end-binding protein, O. nova), 3ullA (mitochondrial ssDNA-binding protein, human), 1prtF (pertussis toxin S5 subunit, B. pertussis), 1bcpD (pertussis toxin S5 subunit (ATP-binding), B. pertussis), 3chbD (cholera toxin, V. cholerae), 1tiiD (heat-labile toxin, E. coli), 2bosA (verotoxin-1 / shiga toxin, B-pentamer, E. coli), 1br9 (TIMP-2, human), 1an8 (superantigen SPE-C, S. pyogenes), 3seb (superantigen SPE, S.Aureus (S. aureus)), 1aw7A (Toxic shock syndrome toxin, S. aureus), 1jmc (Major cold shock protein, Escherichia coli), 1bkb (Translation initiation factor 5a, P. aerophylum), 1sro (S1 RNA binding domain of PNPase, Escherichia coli), 1d7qA (Translation initiation factor 1, elF1a, Human), 1ah9 (Translation initiation factor 1, IF1, Escherichia coli), 1b9mA (Mo-dependent transcriptional regulator ModE, Escherichia coli), 1ckmA (RNA guanylyltransferase, Chorella virus, PBCV-1), 1a0i (ATP-dependent DNA ligase, Bacteriophage T7), 1snc (Staphylococcal nuclease, S. aureus), 1hjp (DNA helicase RuvA subunit, N-terminal domain, Escherichia coli), 1pfsA (Gene V protein, Pseudomonas bacteriophage pf3), 1gvp (Gene V protein, Filamentous bacteriophage (f1, M13)), 1gpc (Gene 32 protein (gp32) core, Bacteriophage T4), 1wgjA (Inorganic pyrophosphatase, S. cerevisiae) and 2prd (Inorganic pyrophosphatase, T. thermophilus).

[0068] Non-exhaustive examples of proteins having an SH3 domain are signal transduction adapter proteins, CDC24, Cdc25, PI3 kinase, Phospholipase, Ras GTPase activating protein, Vav proto-oncogene, GRB2, p54 S6 kinase 2 (S6K2), SH3D21, C10orf76 (Potentially), STAC3, some myosins, SHANK1, 2, 3, ARHGAP12, C8orf46, TANGO1, Integrase, Focal adhesion kinase (FAK, PTK2), Proline-rich tyrosine kinase (Pyk2, CADTK, PTK2beta) or TRIP10 (cip4).

[0069] Multimeric protein Note that a subunit is a single protein molecule that aggregates (or "coassembles") with other protein molecules (other subunits) to form a protein complex called a multimer or multimeric protein.

[0070] Among the preferred multimeric proteins, those skilled in the art can refer to the proteins of the TNF superfamily. The tumor necrosis factor (TNF) superfamily is a protein superfamily of type II transmembrane proteins that contain a TNF homology domain and form trimers.

[0071] Members of this superfamily are released from the cell membrane by extracellular proteolytic cleavage and can function as cytokines (e.g., TNF-α involved in inflammation).

[0072] These proteins are mainly expressed by immune cells and regulate not only immune responses and inflammation but also various cellular functions including proliferation, differentiation, apoptosis, and embryogenesis.

[0073] This superfamily includes 19 members (lymphotoxin α, tumor necrosis factor, lymphotoxin β, OX40 ligand, CD40 ligand, Fas ligand, CD27 ligand, CD30 ligand, CD137 ligand, TNF-related apoptosis-inducing ligand, receptor activator of nuclear factor kappa B ligand (RANKL), TNF-related apoptosis weak inducer, proliferation-inducing ligand, B cell-activating factor, LIGHT, vascular endothelial growth inhibitor, TNF superfamily member 18, and ectodysplasin A).

[0074] These proteins bind to receptors that are members of the TNF receptor superfamily, which includes 29 members.

[0075] Among the proteins of the TNF superfamily, those specifically mentioned are - TNF-α that binds to tumor necrosis factor receptor 1 (1A), tumor necrosis factor receptor 2 (1B), or lymphotoxin β receptor - RANKL that binds to RANK, osteoprotegerin, or TWEAK receptor - TRAIL (TNFSF10) that interacts with TNFRSF10B It is.

[0076] Tumor necrosis factor (TNF, tumor necrosis factor α, TNFα, cachectin, or cachexin) is a cell signaling protein (cytokine) involved in systemic inflammation and is one of the cytokines that make up the acute-phase response. It is mainly produced by activated macrophages, but can also be produced by many other cell types, such as CD4+ lymphocytes, NK cells, neutrophils, mast cells, eosinophils, and nerve cells. Dysregulation of TNF production has been associated with various human diseases, including Alzheimer's disease [6], cancer [7], major depression [8], psoriasis [9], and inflammatory bowel disease (IBD).

[0077] Several antibodies against TNF have been developed to treat various diseases. - Adalimumab is used to treat rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, Crohn's disease, ulcerative colitis, chronic psoriasis, hidradenitis suppurativa, and juvenile idiopathic arthritis. - Certolizumab pegol (CDP870) is used in the treatment of Crohn's disease, rheumatoid arthritis, psoriatic arthritis, and ankylosing spondylitis. - Infliximab is used in the treatment of Crohn's disease, ulcerative colitis, psoriasis, psoriatic arthritis, ankylosing spondylitis, and rheumatoid arthritis. It is also used in Behcet's disease and other conditions. Note that infliximab is administered by intravenous infusion because the digestive system breaks down the drug and cannot be administered orally. - Golimumab is used in the treatment of rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, and ulcerative colitis and is used by subcutaneous or intravenous injection. - Etanercept is used to treat rheumatoid arthritis, juvenile rheumatoid arthritis and psoriatic arthritis, psoriasis vulgaris and ankylosing spondylitis.

[0078] TNF acts, in particular, by promoting the release of other cytokines, including chemokines, IL-1, IL-6, GM-CSF and IL-10, by increasing adhesion molecules, by allowing angiogenesis, and by releasing tissue-destructive enzymes. It also activates T cells.

[0079] TNF can also play a role in cancer, by promoting an inflammatory microenvironment near the tumor, thereby limiting the effectiveness of treatment (chemotherapy, CAR-T cells...).

[0080] Therefore, variants of OB-fold proteins (especially from the Sac7d family) that have the ability to inhibit the formation of functional TNF trimers can be used for the same applications as the above antibodies and to treat the same diseases. Furthermore, such variants can be administered orally or topically (applied to the skin) to the patient, which are administration modes not available for antibodies, so it is advantageous to use them.

[0081] Another advantage is an enhanced long-term safety profile due to the ability to selectively neutralize the activity of soluble TNFα rather than its membrane-bound form. Blocking membrane-bound TNF can actually (i) trigger an immune response against the blocking factor (Deora et al., MABS. 2017; 9(4):680-695) and (ii) induce unwanted side effects, such as the risk of inducing multiple sclerosis.

[0082] Receptor Activator of Nuclear Factor-kappa B Ligand (RANKL) is an apoptosis regulatory gene that is a binding partner of osteoprotegerin (OPG), a ligand for the receptor RANK, and controls cell proliferation by regulating the protein levels of Id4, Id2, and Cyclin D1. It is thought to be involved in the formation of bone metastases in breast or prostate cancer. Overproduction of RANKL has also been associated with various degenerative bone diseases, such as rheumatoid arthritis and psoriatic arthritis. Denosumab is an antibody against RANKL that is used to treat postmenopausal patients suffering from postmenopausal osteoporosis (PMO) and is also used to reduce cancer treatment-induced bone loss in breast and prostate cancer. It may also be used in multiple myeloma to prevent the progression of bone complications.

[0083] Therefore, variants of OB-fold proteins (especially those derived from the Sac7d family) that have the ability to inhibit the formation of functional RANKL trimers can be used for the same applications and to treat or prevent the same diseases or complications.

[0084] TNF-related apoptosis-inducing ligand (TRAIL) is a protein that functions as a ligand that induces the process of cell death called apoptosis. TRAIL and its receptors have been used as targets for various anticancer therapies. TRAIL has also been shown to be a pathogen or protective factor in various lung diseases, particularly pulmonary hypertension. Recently, it was shown that neutralization of TNFSF10 improves the functional outcome in a mouse model of Alzheimer's disease (Cantarella et al., Brain. 2015 Jan; 138(1): 203-216). Therefore, variants of OB-fold proteins (especially those derived from the Sac7d family) that have the ability to inhibit the formation of functional TRAIL trimers can be used for the treatment of Alzheimer's disease, to treat the disease or prevent its further progression.

[0085] Method for screening Sac7d variants The present invention also provides a method for obtaining a protein that binds to a monomer of a multimeric protein, comprising: a. providing a combinatorial library of variants of a Sac7 family protein, wherein 5 to 22 residues of the binding interface of the OB-fold protein to its natural ligand are randomized, and the OB-fold protein is Sac7d or Sac7e derived from Sulfolobus acidocaldarius, or Sso7d derived from Sulfolobus solfataricus, or DBP7 derived from Sulfolobus tokodaii, or Ssh7b derived from Sulfolobus shibatae, or Ssh7a derived from Sulfolobus shibatae, or p7ss derived from Sulfolobus solfataricus, and the residues randomized in the variant are selected from among the residues corresponding to V2, K3, K5, K7, Y8, K9, G10, E14, T17, K21, K22, W24, V26, G27, K28, M29, S31, T33, D35, D36, N37, G38, K39, T40, G41, R42, A44, S46, E47, K48, D49, A50 and P51 of Sac7d; b. expressing the variant (in particular, by ribosome, yeast display); c. presenting the variant to a multimeric protein conjugated to a solid support such that multimerization is inhibited, presenting the surface involved in multimerization of the multimeric protein and utilized for binding; d. selecting the variant that binds to the surface; e. obtaining a sublibrary of the selected variant; f. optionally, repeating steps (b) to (e) 1 to 3 times; g. selecting a variant of the sublibrary, the variant comprising 5 to 25 residues mutated at the binding interface of the starting protein of the Sac7d protein to its natural ligand and binding to the surface of the multimeric protein involved in multimerization to inhibit multimerization of the multimeric protein. A method comprising the above steps is provided.

[0086] Thus, this method allows for the identification from a library of variants of a variant that binds to its subunit and inhibits or delays the formation of a biologically functional multimer.

[0087] The variant inhibits the multimerization of a multimer (whether homomultimer or heteromultimer) protein. In fact, it should be noted that protein-protein binding is a dynamic process, and two polypeptides are bound at one point and separated at another. The major challenge in obtaining a stable multimer is the affinity of the two proteins (for one another), which will in particular control the duration of binding of these proteins. In the present application, when shifting the equilibrium between free monomers and monomers bound to multimeric proteins, the variant is said to inhibit multimer formation. This is done by competition between the variant and other monomers for the unbound monomeric form. When the free unbound monomer binds to the variant, the multimer formed decreases, thus shifting the equilibrium. The action of the variant disclosed herein is thus understood to be a competing partner for the unbound free monomer portion of the multimeric protein that binds to such a monomer at the epitope such that binding to other native partners is hindered, thereby limiting the formation of a functional multimeric protein.

[0088] Step a) consists of providing a library of variants of the OB fold or SH3 protein, in particular variants of proteins of the Sac7d family. These variants within the library are randomly mutated at a specific number of amino acids located at the binding site. The prior art (in particular, WO2007139397 and WO 2008 / 068637) details how the binding sites of these proteins can be determined. In particular, one skilled in the art can overlay the OB fold or SH3 fold onto the Sac7d scaffold and use the amino acids corresponding to those shown above.

[0089] Step b) consists of the expression of variants. Such expression is carried out by any method in the art, for example, by bacteria, yeast or ribosome display. Ribosome display is particularly suitable. Two patent application documents provide all the guidance necessary to carry out this step.

[0090] In step c), the expressed variant is presented on a subunit of a multimeric protein conjugated to a solid support. Such conjugation is carried out under conditions where multimerization of the protein (protein-protein binding within the multimer) is inhibited such that the face of the subunit involved in multimerization of the multimeric protein is presented and thus available for binding by the variant (i.e., the protein is not in multimeric form).

[0091] In accordance with steps d) and e), variants that bind to the protein presented in step c) are selected, thereby providing a sublibrary containing the selected variants. Such selection steps are known in the art and are classical in the process of identifying binders within a library.

[0092] Steps b)-e) can typically be repeated 1-3 times in order to improve the enrichment of the sublibrary (i.e., the limitation of the number of variants in a series of sublibraries), in particular by increasing the stringency of binding (especially in order to obtain binders with improved affinity). Thereby, ultimately, the quality of the binder is improved and ultimately it becomes possible to select very specific and affinity binders (step g).

[0093] In order to achieve the goal of obtaining variants that bind to the face involved in protein multimerization, it is important to ensure that step c), i.e., such a face is presented to the variants of the library.

[0094] This is the following step: a. A step of grafting / conjugating a ligand to a multimeric protein, b. A step of immobilizing the grafted / conjugated protein on a solid surface via the ligand, c. A step of washing the surface to remove unbound multimeric protein, can be achieved by conjugating the multimeric protein to a solid ligand according to the following.

[0095] The ligand grafted in such a can be biotin, which can be easily used for protein conjugation. The fact that the ligand is conjugated to the multimeric form of the protein can ensure that it is not bound to the surface of the subunit involved in multimerization and stabilization of the multimer.

[0096] The immobilization of the grafted / conjugated protein on a solid surface (especially beads) is carried out via the ligand. (Especially using streptavidin or neutravidin on the solid surface) The biotin / avidin binding is particularly suitable for such step b.

[0097] Finally, the solid surface is washed to remove unbound multimeric protein and to "dismultimerize" the protein, i.e., to remove the subunits that are involved in the multimer but not bound to the surface. As a result, only the "lonely" subunits bound via the ligand will remain on the surface. Such subunits "present" the surface involved in multimerization as a target for variants of the library.

[0098] Note that the surface involved in multimerization of the multimeric protein can be determined by crystallization of the multimer. Alternatively, or in addition, site-directed mutagenesis followed by cross-linking experiments (e.g., SDS-PAGE or size exclusion) enables the determination of the amino acids involved in protein-protein interactions and formation of the multimer.

[0099] When using the above method, the multimer can first be bound to a solid surface. By performing different washings, ultimately, subunits that are involved in the multimer and not bound to the solid surface are released, thereby presenting the surface involved in protein-protein interactions in the multimer.

[0100] To confirm that multimerization is inhibited, one of ordinary skill in the art can use methods known in the art: after co-incubation of the isolated variant and monomer, one of ordinary skill in the art can confirm that inhibition has been inhibited by size exclusion or SDS-PAGE. These methods are known in the art.

[0101] Furthermore, it is considered possible to obtain variants for any monomer involved in the multimeric protein. In fact, the surface involved in protein-protein interactions in the multimer evolves to exhibit a certain tropism with respect to protein-protein interactions, and thus could interact with variants of the OB fold domain. Therefore, it is considered that there is a competitive advantage in that this surface binds to the variant presented on the display.

[0102] Generation of identified variants The sequence of the identified variant can be cloned into any suitable vector by any molecular genetic method known in the art.

[0103] These recombinant DNA constructs containing nucleotide sequences encoding polypeptides containing the above variants are used in combination with vectors such as plasmids, phagemids, phages or viral vectors.

[0104] These recombinant acid molecules can be generated by the techniques described in Sambrook et al., 1989 (Sambrook J, Fritschi EF and Maniatis T (1989) Molecular cloning: a laboratory manual, Cold Spring Harbor Press, New York). Alternatively, the DNA sequence can be chemically synthesized, for example, using a synthesizer.

[0105] The recombinant constructs of the present invention include an expression vector capable of expressing RNA and thus generating a protein from the above gene sequences. The vector can thus further include regulatory sequences including an appropriate promoter operably linked to the open reading frame (ORF) of the gene sequences disclosed herein. The vector can further include a selectable marker sequence, such as an antibiotic resistance gene. When bacteria are used as the expression host, specific initiation and bacterial secretion signals may also be required for efficient translation of the coding sequence.

[0106] Generation of Molecules Cells are transfected or transformed with a vector containing a sequence encoding a polypeptide comprising the above variant.

[0107] Cells are cultured under conditions in which the protein is expressed and preferably secreted. The cell culture conditions are those commonly used in the production of recombinant antibodies and are known in the art. Such conditions known in the art can also be optimized by those skilled in the art as needed. Kunert and Reinhart (Appl Microbiol Biotechnol. 2016; 100: 3451-3461) have summarized such methods and provide sufficient reference materials thereon.

[0108] One skilled in the art can use a production system of bacteria, phage (Shukra et al., Eur J Microbiol Immunol (Bp). 2014; 4(2): 91-98) or eukaryotes.

[0109] One skilled in the art may prefer to use eukaryotic cells to obtain appropriate post-translational modifications, such as glycosylation.

[0110] In particular, one skilled in the art can use CHO (Chinese hamster ovary) cells, PER.C6 cells (human cell line, Pau et al., Vaccine. 2001 21;19(17-19):2716-21), HEK293b cells (human embryonic kidney cells 293), NS0 cells (non-secretory mouse myeloma-derived cell line) or EB66 cells (duck cell line Valneva, Lyons, France).

[0111] Also provided by the present disclosure are host cells comprising at least one DNA construct encoding a polypeptide comprising the above variants. The host cell can be any cell in which an expression vector is available. As described above, it can be a higher eukaryotic host cell, such as a mammalian cell, a lower eukaryotic host cell, such as a yeast cell, or a prokaryotic cell, such as a bacterial cell.

[0112] Introduction of the recombinant construct into the host cell is carried out by any method known in the art (e.g., calcium phosphate transfection, lipofection, DEAE, dextran-mediated transfection, electroporation or phage infection). The vector can be inserted into the genome of the host cell or maintained as an episomal vector (e.g., bacterial artificial chromosome or yeast artificial chromosome). When introduced into the genome of the cell, such introduction can be randomized or targeted using methods known in the art (such as homologous recombination, etc.).

[0113] Bacterial host and expression Useful expression vectors for use in bacteria are constructed by inserting recombinant DNA sequences into a functional reading phase having a functional promoter, along with appropriate translation start and termination signals. The vector may contain one or more phenotypic selection markers and an origin of replication to ensure the management of the vector and, if desired, to provide amplification within the host.

[0114] Prokaryotic host organisms suitable for transformation include Escherichia coli, Bacillus subtilis, Salmonella typhimurium, and various species of Pseudomonas, Streptomyces, and Staphylococcus.

[0115] Eukaryotic hosts and expression Examples of eukaryotic host cells include vertebrate cells, insect cells, and yeast cells. In particular, those skilled in the art can use the above cells.

[0116] Transformed or transfected cells are cultured according to methods known in the art, and the polypeptide is recovered from the intracellular or extracellular fraction (depending on whether it is secreted).

[0117] Isolation of molecules The recombinant multispecific protein produced can be separated and purified from the intracellular or extracellular fraction by any of a variety of known separation methods that utilize the physical or chemical properties of the protein.

[0118] In particular, those skilled in the art can use methods such as sedimentation, ultrafiltration, various types of liquid chromatography, such as molecular sieve chromatography (gel filtration), adsorption chromatography, ion exchange chromatography, and affinity chromatography, dialysis, and combinations thereof.

[0119] In general, any method known in the art and used to purify recombinant polypeptides is suitable for purifying the molecules disclosed herein.

[0120] If a tag (e.g., a polyhistidine tag) is introduced into the recombinant sequence, one of ordinary skill in the art can use this tag to purify the molecule. However, it is preferred to use affinity to purify the molecule.

[0121] In particular, the fact that the molecules generated herein can bind to a specific target can be used, and any affinity method (affinity column, FACS, beads) can be used to isolate such molecules.

[0122] One particular advantage of the molecules disclosed herein is that they do not require glycosylation to be active and can therefore be produced in any type of cell and do not need to be eukaryotic cells. They are particularly well produced in bacterial cells.

[0123] Modification of variants Variants having the ability to alter the formation of the multimeric form of the target, particularly isolated by the above methods, can be modified by any method known in the art.

[0124] Preparation of polypeptides containing variants It is possible to prepare a DNA sequence containing two coding sequences, one a variant disclosed herein and the other a protein or peptide of interest. The resulting expressed protein will therefore be a polypeptide containing both proteins. The vector can be constructed to include a sequence encoding a linker located between the two proteins in the expressed polypeptide.

[0125] Accordingly, the invention also encompasses a polypeptide comprising a variant of an OB-fold protein (preferably of the Sac7d family) that binds to a subunit of a multimeric protein and that is linked (preferably through an amine bond as described above) to another protein or polypeptide.

[0126] In certain embodiments, the other protein or polypeptide comprises another variant of a protein of the Sac7d family.

[0127] In another embodiment, the other protein or polypeptide is an antibody. In this embodiment, the variant of the OB-fold domain is fused to at least one of the heavy or light chains of an immunoglobulin monomer, preferably at the N-terminus or C-terminus of its light or heavy chain. In another embodiment, the variant can be fused to both the heavy and light chains.

[0128] To obtain such compounds, one of ordinary skill in the art can use a gene construct comprising a DNA sequence selected from the group consisting of: a. a sequence encoding the heavy chain of an antibody fused at its 3'-end to a sequence encoding a variant of an OB-fold protein (optionally together with a sequence encoding a linker); b. a sequence encoding the heavy chain of an antibody fused at its 5'-end to a sequence encoding a variant of an OB-fold protein (optionally together with a sequence encoding a linker); c. a sequence encoding the light chain of an antibody fused at its 3'-end to a sequence encoding a variant of an OB-fold protein (optionally together with a sequence encoding a linker); d. a sequence encoding the light chain of an antibody fused at its 5'-end to a sequence encoding a variant of an OB-fold protein (optionally together with a sequence encoding a linker).

[0129] ​This fusion can be carried out at the N-terminus and / or C-terminus of the antibody chain (heavy chain and / or light chain). Notably, when using particularly small OB fold domains (about 70 amino acids), such as proteins from the Sac7d family, it has been noted that it is possible to obtain a molecule having a structure of an antibody with an additional binding region consisting of an antibody region and a modified OB fold domain (two heavy chains paired with two light chains, and such dimers paired together).

[0130] In certain embodiments, the antibody portion of the proteins disclosed herein is an IgG molecule.

[0131] In another embodiment, the antibody portion of the proteins disclosed herein is an IgA molecule.

[0132] In another embodiment, the antibody portion of the proteins disclosed herein is an IgM molecule.

[0133] In another embodiment, the antibody portion of the proteins disclosed herein is an IgD molecule.

[0134] In another embodiment, the antibody portion of the proteins disclosed herein is IgE molecule.

[0135] The antibody can be a human antibody, a rodent antibody (e.g., a mouse or rat antibody), a cat antibody, a dog antibody, a chicken antibody, a goat antibody, a camelid antibody (e.g., a camel, llama, alpaca or nanobody), a shark antibody or an antibody from any other species. It can be a chimeric or humanized antibody. As described in Wikipedia, a humanized antibody is an antibody from a non-human species whose protein sequence has been modified to increase their similarity to antibody variants produced naturally in humans. A chimeric antibody contains sequences from different species.

[0136] Preferably, the antibody, which is part of the molecule disclosed herein, is an antibody comprising two identical heavy chains and two identical light chains (about 400 to 500 amino acids, usually approximately 450 amino acids). Therefore, the antibody comprises identical Fab variable regions. This antibody is thus a monospecific antibody in which both parts of the antibody (the combination of the light and heavy chains) bind to the same epitope of the antigen.

[0137] However, the antibody may exhibit different heavy and / or light chains. In particular, in some embodiments, the antibody is a bispecific antibody. The term "antibody" thus encompasses not only "classical antibodies" having the same heavy and light chains as described above, but also modified antibodies having two or more specificities.

[0138] In certain embodiments, the antibody exhibits one heavy chain and one light chain from one antibody and another heavy chain and another light chain from another antibody.

[0139] The antibody is: Cell surface receptors: insulin receptor, low density lipoprotein receptor-related protein 1, transferrin receptor, epidermal growth factor receptor, epidermal growth factor receptor variant III, vascular endothelial growth factor receptor 1, vascular endothelial growth factor receptor 2, Her2, Her3, Her4, PMSA, IGF-1R, GITR, RAGE, CD28, Cell surface proteins: mesothelin, EpCam, CD19, CD20, CD38, CD3, TIM-3, CEA, cMet, ICAM1, ICAM3, MadCam, a4b7, CD7, CD4, CD138, Angiogenic factors and growth factors: VEGF, angiopoietin 2, HGF, PDGF, EGF, GM-CSF, HB-EGF, TGF, Immune checkpoint inhibitors or activators: PD-1, PD-L1, CTLA4, CD28, B7-1, B7-2, ICOS, ICOS-L, B7-H3, B7-H4, LAG3, KIR, 4-1BB, OX40, CD27, CD40L, TIM3, A2aR, Circulating proteins: TNFa, IL23, IL12, IL33, IL4, IL13, IL5, IL6, IL4, IFNg, IL17, RANKL, Bace1, α-synuclein, Tau, amyloid, It can bind to a target selected from the group consisting of.

[0140] In particular, it is envisaged that the antibody targets IL17 or a cell surface receptor particularly involved in cancer.

[0141] Therefore, - A polypeptide comprising the above variant that binds to TNF-α and an antibody that binds to IL17 (for example, the reference antibody secukinumab), - A polypeptide comprising the above variant that binds to TNF-α and an antibody that binds to Her2 / neu (the reference antibodies that bind to Her2 / neu are trastuzumab and pertuzumab), - A polypeptide comprising the above variant that binds to RANKL and an antibody that binds to TNF-α (the reference antibodies that bind to TNF-α are infliximab, adalimumab, certolizumab pegol, ozoralizumab and golimumab), are particularly envisaged.

[0142] Alternatively, the polypeptide may comprise the above variant and a biologically active molecule, such as erythropoietin, interferon or etanercept.

[0143] In another aspect, variants of the OB-fold domain (particularly variants of proteins of the Sac7d family) are conjugated to organic molecules. This can be done by any method known in the art. In particular, one of ordinary skill in the art can chemically link a molecule to a protein. As molecules, one of ordinary skill in the art can refer to anti-proliferative agents (cytotoxic and cytostatic agents) including cytotoxic compounds (e.g., broad-spectrum), angiogenesis inhibitors, cell cycle progression inhibitors, PBK / m-TOR / AKT pathway inhibitors, MAPK signaling pathway inhibitors, kinase inhibitors, protein chaperone inhibitors, HDAC inhibitors, PARP inhibitors, Wnt / Hedgehog signaling pathway inhibitors, RNA polymerase inhibitors, and proteasome inhibitors. One of ordinary skill in the art can also use anti-inflammatory molecules.

[0144] In particular, DNA-binding or alkylating agents such as anthracyclines (doxorubicin, epirubicin, idarubicin, daunorubicin) and their analogs, alkylating agents such as calicheamicin, dactinomycin, mitomycin, pyrrolobenzodiazepine, etc. can be referred to. One of ordinary skill in the art can also refer to cell cycle progression inhibitors such as CDK inhibitors, Rho-kinase inhibitors, checkpoint kinase inhibitors, aurora kinase inhibitors, PLK inhibitors, and KSP inhibitors. One of ordinary skill in the art can also refer to thalidomide and its derivatives lenalidomide and pomalidomide. For treating inflammatory disorders, one of ordinary skill in the art can also use cyclooxygenase-2 inhibitors, 5-lipoxygenase inhibitors, quercetin, and / or resveratrol as molecules conjugated to polypeptides containing the variant.

[0145] Use of variants The variant can be used particularly in a therapeutic method.

[0146] In particular, variants targeting TNFα target the monomeric or dimeric form of the protein and can be used, in particular, for the treatment of inflammatory diseases (such as inflammatory bowel disease (IBD), rheumatoid arthritis (RA), psoriasis, etc.) or for the treatment of said diseases.

[0147] Variants targeting RANKL target its monomeric or dimeric form and can be used, in particular, for the treatment of inflammatory diseases (such as the above-mentioned diseases) or bone diseases (such as osteoporosis) or for the treatment of said diseases.

[0148] Variants targeting the TRAIL protein target its monomeric or dimeric form and can be used for the treatment of Alzheimer's disease or for the treatment of said diseases.

[0149] Accordingly, the present invention relates to a method for treating the above-mentioned diseases, which comprises administering a therapeutically effective amount of the OB-fold variant (in particular, a variant of a protein of the Sac7d family) to a patient in need thereof.

[0150] The terms "therapeutically effective amount" or "effective amount", as used herein, are amounts sufficient to produce beneficial or desired results, such as clinical results, and the "effective amount" depends on the circumstances to which it is applied. An effective amount is an amount that provides a therapeutic benefit while minimizing side effects or adverse effects.

[0151] One skilled in the art can administer the variant by any method in the art.

[0152] In particular, one skilled in the art can inject the variant. In another aspect, one skilled in the art can apply the variant topically (either to the patient's skin or to the eye) as disclosed in WO 2014 / 173899. In another aspect, one skilled in the art can administer the variant orally as disclosed in WO 2016 / 062874.

[0153] Variants or polypeptides containing variants can also be used in diagnostic methods.

[0154] Accordingly, the present invention is a method for detecting or quantifying the presence of subunits of a multimeric protein in a sample, comprising: a. exposing the sample to the disclosed variant that binds to the subunit but not to the fully formed multimeric protein under conditions such that such binding is possible; b. recovering the variant and / or detecting the subunit, measuring the amount of the subunit, or quantifying [ELISA, fluorescence, column] the subunit. The present invention relates to a method comprising the steps of.

[0155] The recovery in step b) can be carried out by various washing or methods common in the art. Detection or quantification can be carried out by any method, for example, ELISA or other methods in the art.

[0156] The present invention also provides a method for purifying the subunits of a multimeric protein in a sample, comprising: a. exposing the sample to the disclosed variant that binds to the subunit but not to the fully formed multimeric protein under conditions such that such binding is possible; b. recovering the variant bound to the subunit; c. eluting the subunit from the variant. The present invention encompasses a method comprising the steps of.

[0157] Step c can be carried out, in particular, by changing the composition of the buffer (increase / decrease in salt concentration, surfactant, chaotropic agent) or the pH (increase / decrease).

[0158] In these methods, it may be desirable to use a polypeptide comprising a variant fused to another protein that can be used to recover the variant bound to the subunit (e.g., in a column having an affinity for this other protein).

Brief Description of the Drawings

[0159]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Example

[0160] Example 1. Generation of NF against TNF or RANK ligand Screening was performed using a library containing random mutations at position 17 of the Sac7d sequence (SEQ ID NO: 1).

[0161] The mutation positions were K7, Y8, K9, E11, K21, K22, W24, V26, M29, S31, T33, D35, T40, G41, R42, A44 and S46.

[0162] The targets were TNF-α and RANKL.

[0163] The trimeric soluble form of the target was biotinylated and attached to solid beads through biotin / avidin binding. Then they were washed thoroughly to dissociate the trimers.

[0164] Screening essentially disclosed in WO 2008 / 068637 was performed by ribosome display expression of library variants and exposure to the attached target.

[0165] For the selection and identification of clones, biotinylated hTNFα (R&D Systems) and mTNFα (R&D Systems) were used. Biotinylation was performed by incubating a 50 μM solution of the target protein with a 5-fold molar excess of sulfo-succinimidyl-6-(biotinamido)hexanoate (sulfo-NHS-LC-LC-biotin, Pierce) in PBS (Sigma-Aldrich) for 1 hour on ice. The biotinylated protein was buffer-exchanged using a Pierce protein desalting spin column equilibrated in 20 mM Tris-HCl, 150 mM NaCl pH 7.4 (TBS). The degree of biotinylation was determined to be approximately 0.5 - 5 molecules of biotin per protein molecule using the HABA assay (Sigma). Generation of the library corresponding to random mutagenesis at the above positions in the Sac7d protein has been described previously (in particular, Mouratou et al., Methods Mol Biol. 2012;805:315-31; Mouratou et al., Proc Natl Acad Sci U S A. 2007 Nov 13;104(46):17983-8; WO 2008 / 068637).

[0166] The PCR-amplified library was transcribed and selection was performed at 4°C essentially as previously described (Mouratou et al., 2007 and 2012, supra, Binz et al., Nat Biotechnol 2004, 22:575-582). For the first two rounds of selection, streptavidin (66 nM, 100 μl / well; Sigma-Aldrich) or neutravidin (66 nM, 100 μl / well; Thermo Scientific) in TBS150 (50 mM Tris HCl, pH 7.4, 150 mM NaCl) was immobilized onto Maxisorp plates (Nunc) by alternating overnight incubations at 4°C. The wells were then blocked with 300 μl of 0.5% BSA (Sigma-Aldrich) in TBS150 for 1 h at 4°C. Biotinylated hTNFα (100 μl, 0.1 μM) in TBS150 containing 0.5% BSA was allowed to bind for 1 h at 4°C. Prior to each round of ribosome display, the wells were washed thoroughly with wash buffer WBT (50 mM Tris acetate, pH 7.5, 150 mM NaCl, 50 mM Mg(CH3COO-)2, 0.05% Tween 20). Rounds 3 and 4 were performed similarly except that the target was alternately presented to streptavidin- or neutravidin-coated magnetic beads.

[0167] Each round of ribosome display consisted of a 1-h pre-panning step on streptavidin (or neutravidin)-coated objects and a 1-h binding step to the target protein. After washing, RNA purification and reverse transcription, the DNA pool was amplified by RT-PCR.

[0168] Multiple variants were obtained for each target and further characterized to show that they bind to the target in other assays.

[0169] One assay to verify binding to the target is performed using variants generated to have a His tag.

[0170] Except for performing the blocking step at room temperature for 1 hour, the biotinylated antigen is immobilized on plates coated with streptavidin as described above. For the identification of positive clones, 100 μl of crude E. coli extracts are applied to wells with or without the immobilized antigen for 1 hour at room temperature. The binding is visualized using an anti-RGSHis antibody HRP conjugate (Qiagen) that detects the RGSHis6 tag of the variant and a solution of 1 mg / ml o-phenylenediamine substrate (OPD) in detection buffer (0.05 M citric acid, 0.05% hydrogen peroxide). The absorbance at 450 nm is measured using an ELISA plate reader. Quantitative ELISA is performed in the same manner except using the purified protein.

[0171] The characterization of a specific variant that binds to TNF-α called N11 is disclosed in more detail in the following examples.

[0172] The characterization of a variant that binds to RANKL called F10 is also described below.

[0173] Example 2. Further characterization of a specific variant (N11) Example 2.1. Competition with TNF receptor and known binders The ability of other binders to TNF-α that compete with the binding of N11 was evaluated.

[0174] Briefly, other binders (TNF receptor I (TNFR I), TNF receptor II (TNFR II) or two monoclonal antibodies adalimumab (ADA) and infliximab (IFX)) were co-incubated with N11 and soluble TNF.

[0175] As shown in Figure 2, the binding of N11 is neutralized by TNFR I, TNFR II and adalimumab, but the binding of N11 is not neutralized by infliximab.

[0176] Example 2.2. Cross-binding between human and mouse TNF The EC100 of N11-human TNFa (hTNFa) interaction was determined by ELISA using plates functionalized with hTNFa and a range of concentrations of N11.

[0177] Also, the cross-reactivity of N11 to hTNFa and mTNFa (mouse TNFa) was monitored by ELISA at a single concentration (EC100). Plates were functionalized with hTNFa and mTNFa. Wells not functionalized with TNFa were used as negative controls to highlight the specificity of the observed signals.

[0178] N11 was shown to cross-react between human and mouse TNF (Figure 3).

[0179] Example 2.3. Scanning of important residues To determine which residues were responsible for the observed binding, each residue of N11 (SEQ ID NO: 16) that differed from the Sac7d (SEQ ID NO: 1) sequence was replaced with alanine (alanine scan).

[0180] Subsequently, the resulting proteins were exposed to TNFa and binding was measured by ELISA (measurement of EC100 using plates functionalized with hTNFa and a range of concentrations of N11 variants).

[0181] If there was a loss of binding, this indicated that the residue was very important for binding to the subunit.

[0182] As shown in Figure 4, residues 31, 33, 40 and 42 were shown to be very important.

[0183] Residues 22 and 24 are also important, although their mutations essentially reduce their affinity.

[0184] Modification of other mutant residues essentially has no effect on binding. Therefore, the sequences SEQ ID NO:17 (including the very important amino acids 31, 33, 40, 42 of N11) and SEQ ID NO:18 (including the very important amino acids 31, 33, 40, 42 of N11 as well as residues 22 and 24) appear to be the minimum sequences that maintain binding to TNFa.

[0185] Conclusion The above information - Binds to human and mouse TNFa - Competes with ADA, TNFRI, TNFRII but not with IFX - Properties of important amino acids were used for in-silico modeling, and it was shown that N11 is likely to bind to the trimerization core of TNF (two models that both bind to the core were obtained). This was further experimentally demonstrated by modifying mutants of TNFα by site-directed mutagenesis (Example 3).

[0186] Furthermore, it was shown that N11 does not bind to other members of the TNF superfamily.

[0187] Example 3. Design and further characterization of TNF variants Four point mutants of TNFα: - TNF variant R32A - TNF variant E146A - TNF variant Y115A - TNF variant Y151A were prepared (Figure 5).

[0188] The binding of N11 to plates coated with TNFα variants was examined (Figure 6). For the variant Y151A, it was shown that binding was essentially inhibited.

[0189] However, the Y151 variant was still shown to be functional (Figure 7). This was done by the L929 assay. This indicates that the variant Y151A is likely to be folded sufficiently and that its residue is involved in the epitope recognized by N11.

[0190] Conclusion The N11 binding is significantly affected by the Y151A mutation. The Y151A mutation was shown to be fully functional. Residue Y151 is involved in the epitope of N11. Residue Y151 appears to be buried within the core of the TNFα trimer. This is a further discussion showing the interaction pattern based on the block of TNFα trimerization.

[0191] Example 4. Further binding analysis - inhibition of trimer formation Cross-linking experiments were performed.

[0192] Figure 8 shows that the bands could be due to the complex of N11 and the monomer, and that the dimer of TNFα was observed on the SDS-PAGE gel after cross-linking. On the other hand, no evidence of the interaction between N11 and the trimer of TNFα was confirmed. When TNFα was incubated with N11 at room temperature for 24 hours, lower molecular weight complexes that could be due to the monomer of TNFα complexed with one nanobody and the dimer of TNFα complexed with one nanobody were further observed.

[0193] In a similar cross-linking experiment where TNF was co-incubated with N11 over time, it was observed that the ratio of TNFα trimer and dimer decreased with the increase in incubation time, while the ratio of the monomer and dimer of TNFα complexed with N11 increased with the increase in incubation time.

[0194] Similar to what was observed in this cross-linking experiment, lower molecular weight species appeared after incubating TNFα with N11.

[0195] Example 5. Inhibition of trimer formation The above data strongly supports the hypothesis that N11 binds to both the monomer and dimer of TNFa, but does not bind to its trimer.

[0196] To reinforce this finding, mutants of TNFa were generated (Figure 9). a) Mutants that generate stabilized trimers of TNFa Stabilization of the trimer is achieved by the formation of disulfide bridges through the double mutation S95C / G148C. b) Mutants that prevent trimerization Position 119 is important for the trimerization of TNFα. Mutations at this 119th position prevent trimer formation.

[0197] In the Y119 mutant, a lower ratio of trimers was observed compared to WT TNFα (Figure 10).

[0198] The functions of the TNFa mutants were verified (Figure 11).

[0199] The stabilized trimer maintained functionality in the L929 assay (mutant S95C / G148C).

[0200] The mutant Y119 was not functional in the L929 assay. This observation was predicted from the fact that mutations at this position alter trimerization and that the active form of TNFα is a trimer. Since the L929 assay is not relevant to confirming that the mutation at Y119 does not impair the correct folding of each TNFα mutant, other tests were performed to verify the functionality of the Y119 mutant. A binding assay using infliximab showed that similar infliximab (IFX) binding was observed for both WT and Y119 TNFα variants. Since the Y119 mutant is still functional with respect to binding to IFX, this indicates that the protein is properly folded.

[0201] N11 was shown to be unable to bind to the trimer but was still able to bind sufficiently to TNFα in its low oligomeric state (Figure 12). The lack of binding to the stabilized TNFα trimer is in good correlation with the observations made in cross-linking experiments.

[0202] Binding of N11 to TNFα could also be monitored in Octet RED96 for the Y119 variant, but not for WT TNFα (not shown).

[0203] Conclusion Cross-linking experiments showed that N11 binds to both the monomer and dimer of TNFα but not to the trimer. The epitope of N11 includes residues buried within the core of the TNFα trimer. N11 cannot bind to the stabilized trimer of TNFα. Binding of N11 is promoted when TNFα exists in its low oligomeric state.

[0204] Example 6. In vivo experiments Neutralization of TNFα by N11 was demonstrated in an in vivo setting.

[0205] A mixture of TNFα and N11 pre-incubated for 24 hours was administered i.v. to mice, and measurement of IL6 (a cytokine produced in response to TNFα) in the blood was performed 6 hours later.

[0206] Additional controls were used: N9, another (less effective) variant isolated in Example 1, N9mut42A (an N9 variant with a mutation that inhibits binding to TNF), and a variant of irrelevant Sac7d (Irr NF).

[0207] Figure 13 shows that both N11 and N9 are able to inhibit the production of IL6 in vivo and thereby inhibit the activity of TNFα in vivo.

[0208] Other in vivo experiments demonstrated the efficacy of various variants when used orally in a mouse model of acute colitis.

[0209] Example 7. Characterization of a specific variant (F10) Example 7.1. Binding of Sac7d variant A variant that binds to RANKL, called F10, was also characterized. The sequence of such a variant is shown by SEQ ID NO:19. Furthermore, another variant called F3 (sequence not shown) was also characterized.

[0210] Both variants F3 and F10 were shown to bind to RANKL and other members of the TNF superfamily (TNF-α, CD40 ligand and TRAIL).

[0211] Considering the fact that these Sac7d variants bind to various members of the TNF superfamily, the epitope must be located in a region homologous in all proteins, namely the trimerization interface. Therefore, if this variant is not involved in trimerization, it can bind to the monomer when such an interface is available, and once trimerization is achieved, it cannot bind to the protein.

[0212] It was shown using biolayer interferometry that the epitopes of F3 and F10 are different.

[0213] Example 7.2. Cross-binding between human and mouse TNF Cross-binding between human and mouse RANKL was determined using a protocol similar to that disclosed in Example 2.2.

[0214] Both F3 and F10 were shown to cross-react between human and mouse RANKL.

[0215] Example 7.3. In vitro experiments CD14+ monocytes were cultured in complete medium containing MCSF (macrophage colony-stimulating factor) in the presence of RANKL.

[0216] Such conditions induce the differentiation of these cells into osteoclasts.

[0217] The presence of the monoclonal antibody IK 22.5 (1 μg / ml) against RANKL inhibited the differentiation of these cells.

[0218] Various concentrations of Sac7d variants F3 and F10 were tested, and it was shown that inhibition of differentiation was obtained at a low concentration of 25 nM (about 50% inhibition).

[0219] Example 7.4. In vivo experiments Ovariectomized mice (female c57bl6 mice, 8 weeks old) were used as a model of bone loss. After recovery from surgery and 7 days of adaptation, the mice were randomly divided into a treatment group and a control group (n = 8 per group). A polypeptide containing the F10 variant together with a peptide that prolongs the half-life in vivo was administered daily by intraperitoneal injection at 10 mg / kg. Non-ovariectomized mice were included as healthy controls. During the experimental period, the body weights of the animals were monitored. After 6 weeks of treatment, the mice were anesthetized with isoflurane and euthanized by cervical injury. The hind legs and vertebrae were collected and stored at 4°C in 4% formalin until further analysis. For toxicity screening, the internal organs were collected, fixed in 4% formalin, and paraffin-embedded.

[0220] Analysis of the tissues showed that administration of such F10 variants protects the trabecular bone (bone volume / total volume ratio and bone surface) compared to untreated mice.

[0221] The anti-RANKL variant thus shows activity to protect against bone loss in vivo.

[0222] The data disclosed herein shows that by the screening method provided herein, it is possible to obtain variants of the Sac7d protein that can bind to the subunits of proteins of the TNF superfamily at their trimerization sites and prevent multimerization. These variants do not bind to the fully multimerized native protein.

[0223] In vivo experiments have shown that the variants can inhibit the action of multimerized proteins (perhaps, but not bound by this theory) by displacing the equilibrium between subunits in free or multimerized proteins. The above examples are only illustrative of specific proteins, but other variants can also be obtained using different libraries or further screening steps. The above examples are only illustrative of specific clinical conditions, but the reported results show that the variants interact with proteins involved in various clinical conditions, and thus these variants are useful in such diseases in which such proteins are involved. The examples are only illustrative of proteins of the TNF superfamily, but this screening test can be carried out using subunits involved in other multimeric proteins, and the results reported herein for TNF proteins (which bind to free-form subunits and not to multimerized forms) can be achieved by repeating the teaching in these other multimeric proteins.

[0224] Sequence Information SEQUENCE LISTING <110> AFFILOGIC <120> BINDERS FOR INHIBITING FORMATION OF MULTIMERIC PROTEINS <150> EP 18306324.7 <151> 2018-10-07 <160> 20 <170> PatentIn version 3.5 <210> 1 <211> 66 <212> PRT <213> Sulfolobus acidocaldarius <400> 1 Met Val Lys Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Glu Val Asp 1 5 10 15 Thr Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Val Ser Phe 20 25 30 Thr Tyr Asp Asp Asn Gly Lys Thr Gly Arg Gly Ala Val Ser Glu Lys 35 40 45 Asp Ala Pro Lys Glu Leu Leu Asp Met Leu Ala Arg Ala Glu Arg Glu 50 55 60 Lys Lys 65 <210> 2 <211> 64 <212> PRT <213> Sulfolobus solfataricus <400> 2 Met Ala Thr Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Glu Val Asp 1 5 10 15 Ile Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser Phe 20 25 30 Thr Tyr Asp Glu Gly Gly Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Gln Met Leu Glu Lys Gln Lys Lys 50 55 60 <210> 3 <211> 65 <212> PRT <213> Sulfolobus acidocaldarius <400> 3 Met Ala Lys Val Arg Phe Lys Tyr Lys Gly Glu Glu Lys Glu Val Asp 1 5 10 15 Thr Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Val Ser Phe 20 25 30 Thr Tyr Asp Asp Asn Gly Lys Thr Gly Arg Gly Ala Val Ser Glu Lys 35 40 45 Asp Ala Pro Lys Glu Leu Met Asp Met Leu Ala Arg Ala Glu Lys Lys 50 55 60 Lys 65 <210> 4 <211> 64 <212> PRT <213> Sulfolobus shibatae <400> 4 Met Val Thr Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Glu Val Asp 1 5 10 15 Thr Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser Phe 20 25 30 Thr Tyr Asp Glu Gly Gly Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Gln Met Leu Glu Lys Gln Lys Lys 50 55 60 <210> 5 <211> 64 <212> PRT <213> Sulfolobus shibatae <400> 5 Met Ala Thr Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Gln Val Asp 1 5 10 15 Ile Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser Phe 20 25 30 Thr Tyr Asp Glu Gly Gly Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Gln Met Leu Glu Lys Gln Lys Lys 50 55 60 <210> 6 <211> 64 <212> PRT <213> Sulfolobus tokodaii <400> 6 Met Val Thr Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Glu Val Asp 1 5 10 15 Ile Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser Phe 20 25 30 Thr Tyr Asp Asp Asn Gly Lys Thr Gly Arg Gly Ala Val Ser Glu Lys 35 40 45 Asp Ala Pro Lys Glu Leu Leu Gln Met Leu Glu Lys Ser Gly Lys Lys 50 55 60 <210> 7 <211> 64 <212> PRT <213> Sulfolobus islandicus <400> 7 Met Val Thr Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Gln Val Asp 1 5 10 15 Thr Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser Phe 20 25 30 Thr Tyr Asp Glu Gly Gly Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Gln Met Leu Glu Lys Gln Lys Lys 50 55 60 <210> 8 <211> 64 <212> PRT <213> Sulfolobus islandicus <400> 8 Met Thr Thr Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Gln Val Asp 1 5 10 15 Thr Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser Phe 20 25 30 Thr Tyr Asp Glu Gly Gly Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Gln Met Leu Glu Lys Gln Lys Lys 50 55 60 <210> 9 <211> 64 <212> PRT <213> Sulfolobus islandicus <400> 9 Met Thr Thr Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Gln Val Asp 1 5 10 15 Ile Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser Phe 20 25 30 Thr Tyr Asp Glu Gly Gly Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Gln Met Leu Glu Lys Gln Lys Lys 50 55 60 <210> 10 <211> 62 <212> PRT <213> Metallosphaera sedula. <400> 10 Met Ala Thr Lys Ile Lys Phe Lys Tyr Lys Gly Gln Asp Leu Glu Val 1 5 10 15 Asp Ile Ser Lys Val Lys Lys Val Trp Lys Val Gly Lys Met Val Ser 20 25 30 Phe Thr Tyr Asp Asp Asn Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Asn Met Ile Gly Lys Lys 50 55 60 <210> 11 <211> 62 <212> PRT <213> Metallosphaera cuprina <400> 11 Met Ala Thr Lys Ile Lys Phe Lys Tyr Lys Gly Gln Asp Leu Glu Val 1 5 10 15 Asp Ile Ser Lys Val Lys Lys Val Trp Lys Val Gly Lys Met Val Ser 20 25 30 Phe Thr Tyr Asp Asp Asn Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Ser Met Ile Gly Lys Lys 50 55 60 <210> 12 <211> 61 <212> PRT <213> Acidianus hospitalis <400> 12 Met Thr Thr Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Glu Val Asp 1 5 10 15 Ile Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser Phe 20 25 30 Thr Tyr Asp Asp Asn Gly Lys Thr Gly Arg Gly Ala Val Ser Glu Lys 35 40 45 Asp Ala Pro Lys Glu Leu Leu Glu Lys Leu Glu Lys Lys 50 55 60 <210> 13 <211> 62 <212> PRT <213> Acidianus hospitalis <400> 13 Met Ala Thr Lys Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Glu Val 1 5 10 15 Asp Ile Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser 20 25 30 Phe Thr Tyr Asp Asp Asn Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Asp Lys Leu Glu Lys Lys 50 55 60 <210> 14 <211> 60 <212> PRT <213> Acidianus hospitalis <400> 14 Met Ala Thr Lys Val Lys Phe Lys Tyr Lys Gly Glu Glu Lys Glu Val 1 5 10 15 Asp Ile Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Ile Ser 20 25 30 Phe Thr Tyr Asp Asp Asn Gly Lys Thr Gly Arg Gly Ala Val Ser Glu 35 40 45 Lys Asp Ala Pro Lys Glu Leu Leu Glu Lys Leu Lys 50 55 60 <210> 15 <211> 68 <212> PRT <213> Artificial Sequence <220> <223> Consensus sequence <220> <221> VARIANT <222> (2)..(2) <223> Xaa is V, A or T <220> <221> VARIANT <222> (3)..(3) <223> Xaa is T or K <220> <221> VARIANT <222> (4)..(4) <223> Xaa is - or K <220> <221> VARIANT <222> (6)..(6) <223> Xaa is R or K <220> <221> VARIANT <222> (15)..(15) <223> Xaa is E or Q <220> <221> VARIANT <222> (18)..(18) <223> Xaa is T or I <220> <221> VARIANT <222> (31)..(31) <223> Xaa is V or I <220> <221> VARIANT <222> (37)..(39) <223> Xaa Xaa Xaa is EGG or DN- <220> <221> VARIANT <222> (57)..(57) <223> Xaa is M or L <220> <221> VARIANT <222> (58)..(58) <223> Xaa is Q, D or E <220> <221> VARIANT <222> (59)..(59) <223> Xaa is M or K <220> <221> VARIANT <222> (61)..(67) <223> Xaa is -------, EKS--GK, EK---QK, ARAEREK, ARA-EKK, E-----K, EKS--G, ACAEREK or ACA-EKK <400> 15 Met Xaa Xaa Xaa Val Xaa Phe Lys Tyr Lys Gly Glu Glu Lys Xaa Val 1 5 10 15 Asp Xaa Ser Lys Ile Lys Lys Val Trp Arg Val Gly Lys Met Xaa Ser 20 25 30 Phe Thr Tyr Asp Xaa Xaa Xaa Gly Lys Thr Gly Arg Gly Ala Val Ser 35 40 45 Glu Lys Asp Ala Pro Lys Glu Leu Xaa Xaa Xaa Leu Xaa Xaa Xaa Xaa 50 55 60 Xaa Xaa Xaa Lys 65 <210> 16 <211> 65 <212> PRT <213> Artificial Sequence <220> <223> Sequence N11 derived from Sac7d <400> 16 Met Val Lys Val Lys Phe Val Met Phe Gly Lys Glu Lys Glu Val Asp 1 5 10 15 Thr Ser Lys Ile Val His Val Met Arg Gln Gly Lys Leu Val His Phe 20 25 30 Trp Tyr Glu Asp Asn Gly Lys Ile Asp Lys Gly Phe Val Pro Glu Lys 35 40 45 Asp Ala Pro Lys Glu Leu Leu Asp Met Leu Ala Arg Ala Glu Arg Glu 50 55 60 Lys 65 <210> 17 <211> 65 <212> PRT <213> Artificial Sequence <220> <223> Sequence bindig to TNFa derived from N11 <220> <221> misc_feature <222> (7)..(9) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (11)..(11) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (21)..(22) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (24)..(24) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (26)..(26) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (29)..(29) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (35)..(35) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (41)..(41) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (44)..(44) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (46)..(46) <223> Xaa can be any naturally occurring amino acid <400> 17 Met Val Lys Val Lys Phe Xaa Xaa Xaa Gly Xaa Glu Lys Glu Val Asp 1 5 10 15 Thr Ser Lys Ile Xaa Xaa Val Xaa Arg Xaa Gly Lys Xaa Val His Phe 20 25 30 Trp Tyr Xaa Asp Asn Gly Lys Ile Xaa Lys Gly Xaa Val Xaa Glu Lys 35 40 45 Asp Ala Pro Lys Glu Leu Leu Asp Met Leu Ala Arg Ala Glu Arg Glu 50 55 60 Lys 65 <210> 18 <211> 65 <212> PRT <213> Artificial Sequence <220> <223> Sequence bindig to TNFa derived from N11 <220> <221> misc_feature <222> (7)..(9) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (11)..(11) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (21)..(21) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (26)..(26) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (29)..(29) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (35)..(35) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (41)..(41) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (44)..(44) <223> Xaa can be any naturally occurring amino acid <220> <221> misc_feature <222> (46)..(46) <223> Xaa can be any naturally occurring amino acid <400> 18 Met Val Lys Val Lys Phe Xaa Xaa Xaa Gly Xaa Glu Lys Glu Val Asp 1 5 10 15 Thr Ser Lys Ile Xaa His Val Met Arg Xaa Gly Lys Xaa Val His Phe 20 25 30 Trp Tyr Xaa Asp Asn Gly Lys Ile Xaa Lys Gly Xaa Val Xaa Glu Lys 35 40 45 Asp Ala Pro Lys Glu Leu Leu Asp Met Leu Ala Arg Ala Glu Arg Glu 50 55 60 Lys 65 <210> 19 <211> 64 <212> PRT <213> Artificial Sequence <220> <223> Sequence F10 derived from Sac7d <400> 19 Val Lys Val Lys Phe Arg Ser Asp Gly Glu Glu Lys Glu Val Asp Thr 1 5 10 15 Ser Lys Ile Leu Leu Val Ala Arg Gln Gly Lys Ile Val Met Phe Gly 20 25 30 Tyr Asp Asp Asn Gly Lys Pro Gly Ile Gly Leu Val Trp Glu Lys Asp 35 40 45 Asp Pro Lys Glu Leu Leu Asp Met Leu Ala Arg Ala Glu Arg Gly Lys 50 55 60 <210> 20 <211> 64 <212> PRT <213> Artificial Sequence <220> <223> Sequence derived from F10 <400> 20 Val Lys Val Lys Phe Arg Ser Asp Gly Glu Glu Lys Glu Val Asp Thr 1 5 10 15 Ser Lys Ile Leu Leu Val Ala Arg Gln Gly Lys Ile Val Met Phe Gly 20 25 30 Tyr Asp Asp Asn Gly Lys Pro Gly Ile Gly Leu Val Trp Glu Lys Asp 35 40 45 Asp Pro Lys Glu Leu Leu Asp Met Leu Ala Arg Ala Glu Arg Glu Lys 50 55 60

Claims

1. A polypeptide comprising a variant of a Sac7d family protein, wherein the variant comprises 4 to 22 mutant amino acids at the binding site of the Sac7d family protein, binds to a subunit of a multimeric protein, and does not bind to the subunit of the protein when the subunit protein is involved in the multimeric protein that is fully formed in its native state.

2. The polypeptide according to claim 1, wherein the mutant amino acids of the Sac7d variant are selected from the group consisting of V2, K3, K5, K7, Y8, K9, G10, E11, K13, E14, T17, K21, K22, W24, V26, G27, K28, M29, S31, T33, Y34, D35, D36, N37, G38, K39, T40, G41, R42, A44, S46, E47, K48, D49, A50 and P51 of Sac7d.

3. The polypeptide according to claim 1 or 2, wherein the Sac7d variant comprises 4 to 17 mutant amino acids selected from the group consisting of K7, Y8, K9, E11, K21, K22, W24, V26, M29, S31, T33, D35, T40, G41, R42, A44 and S46 of Sac7d.

4. The polypeptide according to any one of claims 1 to 3, wherein the multimeric protein is a member of the tumor necrosis factor superfamily, preferably selected from the group consisting of TNF-α, RANKL and TRAIL.

5. The polypeptide according to any one of claims 1 to 4, comprising SEQ ID NO:17 or SEQ ID NO:

18.

6. The polypeptide according to any one of claims 1 to 5, comprising SEQ ID NO:

16.

7. The polypeptide according to any one of claims 1 to 5, comprising SEQ ID NO:16 in which 1 to 8 amino acids selected from the group consisting of V7, M8, F9, K11, V21, H22, M24, Q26, L29, E35, D41, F44 and P46 are replaced with another amino acid.

8. The polypeptide according to any one of claims 1 to 4, comprising SEQ ID NO:19 or SEQ ID NO:

20.

9. The polypeptide according to any one of claims 1 to 8, wherein a variant of the protein of the Sac7d family that binds to the subunit of the multimeric protein is linked or fused to another protein or polypeptide.

10. The polypeptide according to claim 9, wherein the another protein or polypeptide comprises another variant of the protein of the Sac7d family.

11. The polypeptide according to claim 9, wherein the another protein or polypeptide is an antibody, preferably binding to IL17, TNF-α or Her2 / neu.

12. The polypeptide according to any one of claims 1 to 11, which is conjugated to an organic molecule.

13. A gene construct comprising a DNA sequence encoding the polypeptide according to any one of claims 1 to 11.

14. A vector comprising the gene construct according to claim 13.

15. A host cell comprising the gene construct according to claim 13 within its genome.

16. A method for producing the polypeptide according to any one of claims 1 to 11, comprising the following steps: a. culturing a cell culture, wherein the cell is transformed with the gene construct according to claim 13, and b. recovering the polypeptide.

17. A method for obtaining a protein that binds to a monomer of a multimeric protein, comprising the following steps: a. A step of providing a combinatorial library of variants of a protein of the Sac7d family, wherein 4 to 22 residues of the binding interface of the OB-fold protein to its natural ligand are randomized, and the OB-fold protein is Sac7d or Sac7e derived from Sulfolobus acidocaldarius, or Sso7d derived from Sulfolobus solfataricus, or DBP 7 derived from Sulfolobus tokodaii, or Ssh7b derived from Sulfolobus shibatae, or Ssh7a derived from Sulfolobus shibatae, or p7ss derived from Sulfolobus solfataricus, and the residues randomized in the variant are selected from the residues corresponding to V2, K3, K5, K7, Y8, K9, G10, E14, T17, K21, K22, W24, V26, G27, K28, M29, S31, T33, D35, D36, N37, G38, K39, T40, G41, R42, A44, S46, E47, K48, D49, A50 and P51 of Sac7d, b. A step of expressing the variant, c. A step of presenting the variant to the multimeric protein conjugated to a solid support so that multimerization is inhibited, d. A step of selecting the variant that binds to the surface, e. A step of obtaining a sub-library of the selected variant, f. Optionally, a step of repeating steps (b) to (e) 1 to 3 times, g. A step of selecting a variant of the sub-library, wherein the variant contains 5 to 20 residues mutated at the binding interface of the starting protein of the Sac7d protein to its natural ligand and binds to the surface of the multimeric protein involved in multimerization to inhibit the multimerization of the multimeric protein.

18. The conjugation of the multimeric protein to the solid ligand is a. A step of grafting / conjugating a ligand to the multimeric protein, b. A step of immobilizing the grafted / conjugated protein on a solid surface via the ligand, c. A step of washing the surface to remove unbound multimeric protein, The method according to claim 17, comprising **Claim 19** A polypeptide according to any one of claims 1 to 12, for use as a medicament. **Claim 20** A polypeptide according to any one of claims 1 to 12 for use in the treatment of rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, Crohn's disease, ulcerative colitis, chronic psoriasis, hidradenitis suppurativa, juvenile idiopathic arthritis, Behçet's disease or plaque psoriasis, which binds to the monomer of TNF-α but not to trimerized TNF-α. **Claim 21** A polypeptide according to any one of claims 1 to 12 for use in the treatment of cancer in combination with chemotherapy or treatment with CAR-T cells, which binds to the monomer of TNF-α but not to trimerized TNF-α. **Claim 22** A polypeptide according to any one of claims 1 to 4 and 8 to 12 for use in the prevention of bone loss in postmenopausal patients suffering from bone loss, such as postmenopausal osteoporosis (PMO), or in patients with cancer, particularly breast cancer and prostate cancer, which binds to the monomer of RANKL but not to trimerized RANKL. **Claim 23** A polypeptide according to any one of claims 1 to 12 for use in inhibiting the multimerization of multimeric proteins. **Claim 24** A method for detecting or quantifying the presence of subunits of a multimeric protein in a sample, comprising: a. exposing the sample to a variant according to any one of claims 1 to 12 that binds to the subunit, under conditions such that such binding is possible; b. recovering the variant and / or detecting the subunit or measuring the amount of the subunit. The method comprising. **Claim 25** A method for purifying a subunit of a multimeric protein in a sample, comprising: a. exposing the sample to a variant according to any one of claims 1 to 12 that binds to the subunit, under conditions such that such binding is possible; b. recovering the variant bound to the subunit; c. eluting the subunit from the variant. The method comprising.

Images