Cystine knot domain fusion protein dimer

Abstract

The present application relates to a protein dimer formed from a first fusion protein and a second fusion protein, wherein the first fusion protein comprises a cystine knot domain (CK) fused to a first pharmaceutically active protein and the second fusion protein comprises a CK domain fused to a second pharmaceutically active protein, wherein the two fusion proteins are covalently linked via their CK domains. The present application further relates to the individual fusion proteins, polynucleotides encoding the fusion proteins, vectors comprising the polynucleotides, and host cells comprising the vectors.

Description

Technical Field

[0001] This invention relates to the use of cystine knot (CK) domains, specifically the CK domain of the von Willebrand factor (CK). VWF The fusion of these components can improve the properties of pharmaceutically active proteins, specifically their physicochemical and pharmacokinetic properties. Background Technology

[0002] The purposeful arrangement of different IgG heavy and light chains provides bispecific antibodies that allow binding to two different antigens or two different epitopes on the same antigen. The combination of two different Fab arms in a single antibody molecule significantly expands the therapeutic options for monoclonal antibodies that historically bound to only one antigen or epitope. Today, bispecific antibodies are widely used in oncology therapy and other indications. Essentially, the combination of two arms with different binding epitopes allows for (i) inhibition of two cell surface receptors, (ii) simultaneous blocking of two ligands, (iii) cross-linking of two receptors, or (iv) recruitment of immune-active cells to tumor cells. This sophisticated strategy is impossible in the case of classic antibody formats.

[0003] The Fc domain and its associated effector functions are inherently part of a bispecific antibody in IgG format. The Fc effector functions may be beneficial or undesirable, depending on the specific target indication. The three-dimensional orientation and distance of the antigen-binding domain of this antibody are determined by the antibody format and IgG subclass used as the structural basis for the therapeutic agent.

[0004] Similar to these bispecific antibodies, a single Fc domain can serve as the structural basis for constructing bispecific therapeutic agents. Heterodimers formed from two Fc fusion protein monomers allow for the combination of different protein moieties fused to the C-terminus and / or N-terminus of the two Fc domains. For example, WO 2019 / 129053 A1 describes a fusion protein dimer using an antibody Fc region as its backbone. Both fusion proteins in this dimer contain one or more single-domain antibodies and optionally cytokines.

[0005] Similarly, the Fc domain has been fused with pharmaceutically active proteins such as coagulation factor VIII (FVIII) to prolong half-life or enhance various functions. This FVIII fusion protein is suitable for the treatment of hemophilia A.

[0006] Hemophilia is a group of inherited genetic disorders that impair the body's ability to control blood clotting. In its most common form, hemophilia A, there is a deficiency of clotting factor VIII (FVIII). Hemophilia A occurs in approximately 1 in 5,000 to 10,000 male births. FVIII protein is an essential cofactor with multifunctional properties in blood clotting. FVIII deficiency can be treated with plasma-derived FVIII concentrates or with recombinant FVIII. Treatment with FVIII concentrates can restore normal life for people with hemophilia.

[0007] Patients with hemophilia A may use FVIII for treatment or as a preventative therapy, administered several times a week. In preventative therapy, due to the continuous need for FVIII and its short half-life in the bloodstream (approximately 11 hours in humans), administration of 15–25 IU / kg FVIII three times weekly by body weight is necessary (Ewenstein et al., 2004). The short circulating half-life of FVIII and the associated need for frequent infusions of FVIII concentrate are major challenges in the treatment of hemophilia A.

[0008] In blood, under normal conditions, FVIII molecules associate with their cofactor von Willebrand factor (VWF), which stabilizes FVIII molecules and protects them from various forms of degradation. The non-covalent complex of FVIII and VWF has a high binding affinity of 0.2–0.3 nM (Vlot et al., 1996).

[0009] Historically, hemophilia A has been treated with FVIII derived from human plasma. Furthermore, since the 1990s, various recombinant FVIII (rFVIII) proteins have entered the market. However, neither plasma-derived nor recombinant FVIII proteins possess optimal pharmacokinetic properties. Like many other therapeutic proteins, they undergo peptidase metabolism and other clearance mechanisms, significantly limiting their in vivo half-life.

[0010] Attempts to extend the half-life of FVIII have included immunoglobulin Fc fusions (efmoroctocogalfa, Eloctate), the addition of polyethylene glycols (turoctocog alfa pegol, Esperoct; damoctocog alfa pegol, Jivi; rurioctocog alfa pegol, Adynovate), and single-chain constructs (lonoctocog alfa, Afstyla). All of these techniques have resulted in an approximately 1.5-fold increase in the half-life of FVIII (as reviewed in Tiede 2015). Sufficient literature has demonstrated that FVIII molecules form complexes with VWF in circulation, and that both molecules are simultaneously eliminated primarily through the VWF clearance pathway. Therefore, the half-life of FVIII is mainly determined by the half-life of VWF.

[0011] VWF and VWF fragments containing FVIII binding sites are known to stabilize FVIII, thereby preventing its rapid clearance, proteolytic digestion, uptake by antigen-presenting cells, and promoting higher bioavailability after subcutaneous administration. As shown by Yee et al. (2014), the human VWF D'D3 domain is sufficient to stabilize FVIII in plasma. However, the D'D3-Fc fusion protein only prolonged the half-life of FVIII in VWF- / - mice. In hemophilia A mice, the D'D3-Fc construct failed to prolong the half-life of FVIII because the competition between the VWF protein fragment and endogenous VWF for FVIII binding was ineffective.

[0012] WO 2014 / 011819 A2 describes the successful extension of the half-life of an FVIII construct containing a D'D3 domain of VWF, an Fc domain of IgG, and XTEN. Because this construct does not bind to endogenous VWF, the same half-life extension effect was observed in both VWF / FVIII double knockout mice and hemophilia A mice.

[0013] Other methods for increasing the half-life of therapeutic proteins include gene fusion with proteins that have longer natural half-lives, such as transferrin and albumin, or gene fusion with protein domains, such as the C-terminal peptide (CTP) of human chorionic gonadotropin. As reviewed by Strohl et al. (2015), various fusion proteins of therapeutic proteins and CTP have been developed and are currently in clinical trials. Therapeutic proteins include FSH (Elonva®), FVIIa, FIX, IFN-β, and gastrin.

[0014] WO 2017 / 198435 A1 describes a fusion protein comprising a major protein and one or more extended peptides, wherein the major protein is a mammalian protein (such as human VWF) or a fragment thereof. The extended peptide contains clusters of O-glycosylation sites having at least two O-glycosylated amino acids and may be derived from human VWF. Due to the presence of the extended peptide, the fusion protein has an increased half-life compared to the individual major protein, i.e., the mammalian protein or a fragment thereof. Fusion proteins can be used to increase the half-life of binding partner molecules, such as FVIII. The OCTA12 molecule falling under the definition of WO 2017 / 198435 A1 is a fusion of a VWF fragment capable of binding to FVIII with high affinity. Due to the absence of certain domains recognized by scavenging receptors (e.g., the A1 and D4 domains recognized by the SR-AI receptor), it has a significantly prolonged half-life compared to full-length VWF (a terminal half-life of up to approximately 200 hours after subcutaneous administration in humans, compared to approximately 18 hours for full-length VWF). In addition, it contains three extended peptide repeat sequences, which are 31-amino acid long sequences derived from VWF-derived quadruple O-glycosylation.

[0015] In addition to creating fusion proteins by combining two or more genes that originally encode a single protein and translating the resulting fusion gene, other methods for obtaining multifunctional proteins include combining proteins via dimerization. For example, WO 2013 / 156054A1 describes a fusion protein comprising a pharmaceutically active moiety and a heavy chain domain 2 (HD2) from IgM or IgE, which can be combined via dimerization of the HD2 domain. Summary of the Invention

[0016] This invention is particularly based on the discovery that, when isolated from the rest of the protein and fused with a pharmaceutically active protein, the cystine knot domain (CK) of human VWF... VWF It can be used through fused CK VWF Dimerization of the structural domain will link this pharmaceutically active protein with that of CK. VWF A fusion of second pharmaceutically active proteins. CK VWF The domains not only lead to strong intermolecular binding but also contribute to folding and increased stability of pharmaceutically active proteins. Dimerization is independent of CK. VWF Types of pharmaceutically active proteins with domain fusion. Furthermore, the inventors can demonstrate that not only CK... VWF It can be used for the dimerization of pharmaceutically active proteins, and such as CK. NDP Other cystine domains can also be used for the dimerization of pharmaceutically active proteins. This means that the domains can be generally used for protein dimerization, specifically for the dimerization of pharmaceutically active proteins.

[0017] Therefore, according to a first aspect, the present invention provides a protein dimer formed of a first fusion protein and a second fusion protein, wherein the first fusion protein comprises a CK fused to a first pharmaceutically active protein. VWF Furthermore, the second fusion protein comprises a CK fused to a second pharmaceutically active protein. VWF The two fusion proteins are described through their CK VWF Covalent connection of structural domains.

[0018] For example, through CK VWF Domain dimerization can be used to prepare compounds similar to the first CK. VWF The fused human VWF fragment (VWF-CK) VWF ) and with the second CK VWF FVIII (FVIII-CK) VWF A novel heterodimer of ).

[0019] According to one embodiment of the first aspect, the present invention provides a protein dimer formed of a first fusion protein and a second fusion protein, wherein the first fusion protein comprises an FVIII protein, preferably containing a first linker, the FVIII protein being coupled to a von Wöhlerbrand factor cystine knot domain (CK) via a second linker. VWF The second fusion protein contains a von Wöhlerbrand factor (VWF) fragment, which binds to CK via a third linker. VWF Domain fusion, wherein the third linker is an engineered peptide, and wherein the two fusion proteins are fused via their CK domains. VWF Covalent connection of structural domains.

[0020] The heterodimer contains a highly O-glycosylated extended peptide (EP), which is added to any fusion protein that forms the dimer, such as when linking FVIII or VWF to their corresponding CK. VWF It can be added to the linker of the domain, to the linker connecting the FVIII heavy chain and the light chain, or to the C-terminus or N-terminus of any fusion protein.

[0021] Compared to FVIII alone, the resulting FVIII-CK VWF + VWF-CK VWF (FVIII-CK) VWF -VWF) heterodimers exhibit improved properties, namely, increased expression levels, higher storage stability, and prolonged cycling half-life. Therefore, according to one embodiment of the first aspect, the first pharmaceutically active protein is an FVIII protein, and the second pharmaceutically active protein is a VWF fragment.

[0022] According to an alternative embodiment of the first aspect, the present invention provides a protein dimer formed of a first fusion protein and a second fusion protein, wherein the first fusion protein includes a cysteine ​​knot domain (CK) fused to a first pharmaceutically active protein, and the second fusion protein includes a CK domain fused to a second pharmaceutically active protein, wherein the two fusion proteins are covalently linked through their CK domains, provided that the first pharmaceutically active protein and the second pharmaceutically active protein are not selected from the group consisting of VWF, FVIII and fragments thereof.

[0023] For example, the CK fusion protein dimer was confirmed using a pair of completely different pharmaceutically active proteins, specifically CK... VWF Fusion protein dimer formation. In this example, the first fusion protein comprises a CK fused to a first pharmaceutically active single-domain antibody. VWF Or CK NDP (VHH1-CK) VWF The second fusion protein contains a CK fusion with a second pharmaceutically active single-domain antibody. VWF (VHH2-CK) VWF ).

[0024] The resulting VHH1-CK-VHH2 dimer exhibits improved properties, namely, increased shear stress resistance and thermal stability compared to dimers formed from other dimerizing domains (such as the immunoglobulin Fc domain). Therefore, according to one embodiment of the first aspect, the first pharmaceutically active protein is a first VHH protein, and the second pharmaceutically active protein is a second VHH protein. The single-chain antibody can bind to coagulation factors FIX and FX and can generate simulated FVIII activity.

[0025] Secondly, this invention relates to CK VWF The purpose of the domain pairs is to combine two pharmaceutically active proteins through dimer formation.

[0026] Thirdly, the present invention provides a fusion protein comprising CK VWF Domains, linkers, and pharmaceutically active proteins, wherein the CK VWF The structural domain can interact with the second CK. VWF Domains are covalently bonded.

[0027] Fourthly, the present invention relates to a polynucleotide that encodes a first fusion protein or a second fusion protein as described in the first aspect.

[0028] According to a fifth aspect, the present invention relates to a carrier containing the polynucleotides described in the fourth aspect.

[0029] In a sixth aspect, the present invention relates to a host cell containing the polynucleotide according to the fourth aspect or the carrier according to the fifth aspect, wherein the host cell is a mammalian cell.

[0030] Finally, in a seventh aspect, the present invention also relates to a pharmaceutical composition comprising the protein dimer according to the first aspect, said pharmaceutical composition for treating or preventing hemorrhagic conditions.

[0031] According to an eighth aspect, the present invention relates to the use of one or more EPs for reducing the aggregation tendency of a target protein, wherein the one or more EPs are fused to or inserted into the target protein. Attached Figure Description

[0032] Figure 1 FVIII-CK was shown. VWF -VWF(FVIII-CK) VWF + VWF-CK VWF A schematic diagram of the overall structure of a protein heterodimer.

[0033] Figure 2 FVIII-CK was shown. VWF Schematic diagram of -VWF heterodimers C9+C15, C10+C15, C11+C15, C12+C15, C13+C15, C9+C14, C10+C14 and C11+C14.

[0034] Connector 1 : Unlabeled - Contains furin cleavage site 1 (SEQ ID NO: 20); * - Contains furin cleavage site 2 (SEQ ID NO: 101); ** - Frin cleavage site is missing. C14 The discontinuous linker between the VWF precursor and the VWF fragment indicates post-translational processing via furin protease. The EP indicates a human VWF-derived EP with SEQ ID NO: 2.

[0035] Figure 3 FVIII-CK was shown. VWF Schematic diagram of -VWF heterodimers C12+C14, C13+C14, C11+C25, C11+C26, C13+C26, C27+C15, C27+C25 and C27+C26.

[0036] Connector 1 : Unlabeled - Contains furin cleavage site 1 (SEQ ID NO: 20); * - Contains furin cleavage site 2 (SEQ ID NO: 101); ** - Frin cleavage site is missing.C14 The discontinuous linker between the VWF precursor and the VWF fragment demonstrates post-translational processing via furin protease. ABV Indicates an albumin-binding VHH portion having SEQ ID NO:36.

[0037] Figure 4 FVIII-CK was shown. VWF Schematic diagram of -VWF heterodimers C28+C15, C28+C25, C28+C26, C29+C15, C29+C25 and C29+C26.

[0038] Connector 1 Contains furin cleavage site 1 (SEQ ID NO: 20). ABV The indicator has the albumin-binding VHH moiety of SEQ ID NO: 36. The EP indicates a human VWF-derived EP having SEQ ID NO: 2.

[0039] Figure 5 FVIII-CK was shown. VWF Schematic diagram of -VWF heterodimers C30+C15, C30+C25, C30+C26, C31+C15, C31+C25 and C31+C26.

[0040] Connector 1 Contains furin cleavage site 1 (SEQ ID NO: 20). ABV The EP indicates an albumin-bound VHH having SEQ ID NO: 36. The EP indicates a human VWF-derived EP having SEQ ID NO: 2.

[0041] Figure 6 (A) Shows the FVIII-CK detected in the culture supernatant after transient expression in Expi293F cells. VWF Bar chart of FVIII activity (FVIII:C) of the -VWF heterodimer. (B): FVIII-CK VWF Fusion proteins C11 and C27 through C31 (different in the number and location of EPs and the presence and location of ABVs) are associated with three different VWF-CKs. VWF Average FVIII:C activity of combinations of molecules (C15, C25, and C26, which also differ in the presence and location of ABV). Each bar represents the arithmetic mean and SD of the values ​​obtained from three replicate experiments.

[0042] Figure 7Western blot and non-reducing SDS-PAGE analysis of purified heterodimers C9+C15 and C11+C15 are shown. A, B, and C show FVIII-CK after Western blotting with FVIII detection (A), VWF detection (B), or after PAGE following Coomassie staining (C). VWF -VWF heterodimer. M, molecular weight marker. Molecular weight is expressed in kDa.

[0043] Figure 8 Western blot and non-reducing SDS-PAGE analyses of the purified heterodimeric fusion protein are shown. (A) through (C) show FVIII-CK analyses after Western blotting with FVIII detection (A), VWF detection (B), or after PAGE following Coomassie staining (C). VWF -VWF heterodimer. The numbers above each lane indicate the combination of fusion proteins: 1-C27+C25, 2-C27+C26, 3-C30+C25, 4-C31+C25, 5-C30+C26, 6-C31+C26, 7-C31+C15, 8-C30+C15, 9-C11+C25, 10-C11+C26, 11-C13+C26, 12-C27+C15, 13-C28+C26. M, molecular weight marker. Molecular weight is expressed in kDa.

[0044] Figure 9 A bar graph indicating the normalized binding level of the purified heterodimer relative to full-length human VWF (flVWF), as detected by surface plasmon resonance (SPR), is shown. FlVWF was coated onto a CM5 sensor chip, followed by injection of purified FVIII-CK. VWF -VWF heterodimer C9+C15, C11+C15, or control protein (OCTA12). The SPR signal detected 30 seconds after injection cessation was normalized relative to the binding of rFVIII (NUWIQ, set to 100%). OCTA12 is a negative control that does not bind to flVWF.

[0045] Figure 10 The following is illustrated by a single IV administration of FVIII-CK according to the invention. VWF A graph showing FVIII activity (FVIII:C) detected in HemA mouse plasma after -VWF heterodimer C9+C15 and C11+C15. rFVIII (NUWIQ) was used as a reference. Data are presented as mean + / - SD obtained from 5 animals at each time point.

[0046] Figure 11Results of pharmacokinetic (PK) experiments performed in WT mice are shown. The graph in (A) illustrates the effects of different FVIII-CK administrations at 200 IU / kg body weight (bw) IV. VWF / VWF-CK VWF Protein combination (FVIII-CK) VWF The level of FVIII activity (FVIII:C) 96 hours after the formation of the -VWF dimer. Each data point represents the FVIII:C level of one animal. The bars represent the geometric mean and SD of the values ​​obtained from five animals. (B) shows the formed FVIII-CK. VWF -The terminal half-life of the VWF heterodimer. Each data point represents the plasma half-life calculated for a mouse cohort. Each bar represents the geometric mean and SD for all five cohorts.

[0047] Figure 12 FVIII-CK was shown. VWF The effect of selective deletion of linkers and protease cleavage sites in fusion proteins C32, C35, C42, and C43 on heterodimer activity. (A) shows a schematic structure of the expressed heterodimer. catch Head 1 (B) shows the FVIII:C activity detected in the culture supernatant following transient expression of the indicated heterodimer. The FVIII:C cleavage site is missing.

[0048] Figure 13 FVIII-CK was shown. VWF The effect of selective deletion of linkers in fusion proteins C27, C44, and C45 on heterodimer activity. (A) shows a schematic structure of the expressed heterodimer. Connector 1 : Unlabeled - Contains furin cleavage site 1 (SEQ ID NO: 20). (B) Shows FVIII:C activity in culture supernatant after transient expression of the indicated heterodimer.

[0049] Figure 14 VWF-CK was shown VWF The effect of selective loss of the linker in the protein moiety on heterodimer activity. (A) shows a schematic structure of the expressed heterodimer. C14 and C41 The discontinuous linker between the VWF propeptide and the VWF fragment demonstrates post-translational processing via furin. Linker 1: Unlabeled - containing furin cleavage site 1 (SEQ ID NO: 20). (B) Shows FVIII:C activity in the culture supernatant following transient expression of the indicated heterodimer. Left inset: VWF-CK containing the VWF propeptide expressed with C27.VWF Comparison of fusion proteins C14 and C41. Right inset: VWF-CK with VWF propeptide deficiency expressed together with C27. VWF Comparison of fusion proteins C15 and C40.

[0050] Figure 15 (A) shows FVIII-CK VWF Schematic diagram of VWF heterodimers C32+C15 and C37+C15. EP indicates human VWF-derived EP with SEQ ID NO: 2. C32 and C37 differ only in the presence or absence of the three EP repeat sequences (C32 and C37) in linker 2. Linker 1: **-furin protease cleavage site deletion. (B) shows the results of thrombin generation assay (TGA) of purified C32+C15 and C37+C15 after transient expression in 293F cells.

[0051] Figure 16 Results of tail-clip hemorrhage assays performed in hemophilia A mice are shown. FVIII-CK was administered IV at 75 IU / kg body weight. VWF - Comparison of total blood loss (calculated based on photometric measurements of hemoglobin loss) after VWF heterodimers C32+C15 and C37+C15 relative to recombinant B-domain-deficient FVIII (BDD-FVIII). A formulation buffer (protein-free) was used as a negative control.

[0052] Figure 17 A schematic diagram shows the configuration of heterodimers H1, H2, H3, H5, H6, and H40 formed from the indicated fusion proteins, which comprise single-domain antibodies VHH1 and VHH2 and the CK domain (CK) of human VWF. VWF ) or Norrin's CK domain (CK NDP ). EP indicates a human VWF-derived EP with SEQ ID NO: 2.

[0053] Figure 18 This shows VHH1-CK expressed in HEK293 cells. VWF Non-reducing SDS-PAGE and Western blot analysis of the -VHH2 heterodimers H1, H2, H3, H5, and H6. (A) Coomassie staining; (B) Western blot analysis using His tagging; (C) Western blot analysis using Strep tagging. The numbers below each lane indicate the heterodimer fusion protein: 1-H1, 2-H2, 3-H3, 5-H5, 6-H6. Molecular weight is expressed in kDa.

[0054] Figure 19This shows VHH1-CK expressed in HEK293 cells. NDP Western blot analysis of the -VHH2 heterodimer H40. Western blot analysis was performed under non-reducing conditions (lanes 1 to 3) and reducing conditions (lanes 4 to 6) using (A) His tag detection and (B) Strep tag detection.

[0055] The positions of the H40 dimer in lane 1 and the H40 monomers (FP(FIX)20 and FP(FX)14) in lane 6 are indicated by arrows. Lanes 2 through 5 show unrelated proteins. Molecular weights are expressed in kDa.

[0056] Figure 20 The purified VHH1-CK is shown. VWF -Reducing and non-reducing SDS-PAGE analysis of the VHH2 heterodimer. (A) SDS-PAGE analysis of heterodimer fusion proteins H1, H4, and H5 under non-reducing conditions (lanes 1-3) and reducing conditions (lanes 5-7). The numbers below each lane indicate the dimer protein: H1 - lanes 1 and 5, H4 - lanes 2 and 6, H5 - lanes 3 and 7. (B) SDS-PAGE analysis of dimer fusion proteins H2 and H3 under non-reducing conditions (lanes 1 and 2) and reducing conditions (lanes 4 and 5). Lanes 1 and 4 show protein H2, and lanes 2 and 5 show protein H3. Molecular weight is expressed in kDa.

[0057] Figure 21 A schematic diagram of the configuration of heterodimer H4 and TPP349 is shown. H4 is composed of the VHH-Fc fusion protein VHH1-IgFc. k -Twin-Strep and VHH2-IgFc h -His formation. TPP349 is formed from the VHH-Fc fusion protein VHH1-IgFc. k and VHH2-IgFc h form.

[0058] Figure 22 VHH1-CK is shown VWF Dynamic light scattering (DLS) analysis of the -VHH2 heterodimers H1, H2, H3, H5, and H6. The superposition of the hydrodynamic radius distributions of the heterodimers is depicted. Figure labels are indicated next to the peaks. Each figure represents the average of three sets of measurements, with the standard deviation visualized as the shaded area around the graph line.

[0059] Figure 23 (A)VHH1-CK is shown by DLS analysis. VWFTemperature-induced unfolding and refolding of VHH2 dimers H1, H5, and H6, and (B) VHH1-IgFc-VHH2 heterodimers H4 and TPP349. Unfolding of molecules is illustrated in three subplots for each molecular class. The top subplot shows the 350 / 330 nm ratio (the figure represents the average of three sets of measurements, with the standard deviation visualized as the shaded area around the line), the middle subplot shows the first derivative of the 350 / 330 nm ratio, and the bottom subplot shows the cumulative radius. The bottom temperature scale is divided into two parts: the left part shows the rising temperature (up to 95 °C) used to test protein unfolding, and the right part shows the falling temperature at which refolding may be observed.

[0060] Figure 24 The images show (A) TPP349 (VHH1-IgFc-VHH2 heterodimer) and (B) H1 (VHH1-CK). VWF A comparison of the oscillatory stress stability of the -VHH2 heterodimer. The left side of each subplot shows a stack of SEC chromatograms of the complete product treated under different oscillatory stress conditions. The right side of each subplot depicts the percentage of the integrated area under the SEC peak relative to the unstressed sample.

[0061] Figure 25 VHH1-CK is shown VWF A comparison of the pH stability of -VHH2 and VHH1-IgFc-VHH2 heterodimers. (A) shows VHH1-CK VWF (a) Overlay of SEC chromatograms of unstressed (black) and stressed (treated, gray) samples of VHH2 dimer H5 (top) and VHH1-IgFc-VHH2 dimer TPP349 (bottom). (B) Depicts the relative areas of high molecular weight compounds (HMWC) in the unstressed (black) and stressed (gray) samples.

[0062] Figure 26 VHH1-CK is shown VWF Analysis of FVIII:C activity of VHH2 heterodimers H1, H2, H3, H5, and H6, as well as VHH1-IgFc-VHH2 dimer H4 and TPP349. FVIII activity is expressed as absorbance at 405 nm after baseline correction. The gray bars in the left section represent the signals obtained using human plasma FVIII standards (SHP). The black, white, and gray bars in the right section represent the signals obtained using heterodimers at the indicated protein concentrations.

[0063] Figure 27 The diagram shows the CK domain (CK) of single-domain antibodies VHH1 and VHH2, as well as human VWF. VWFA schematic diagram of the configuration of heterodimers H17, H18, H19, H20, H22, H28, and H30 formed by the specified fusion proteins of the half-life extension module. ABV2 indicates the anti-RSA single-domain antibody described in van Faassen et al., 2020. 3xEP, 6xEP, and 9xEP indicate three, six, and nine consecutive repeat sequences of human VWF-derived EP having SEQ ID NO: 2.

[0064] Figure 28 A schematic diagram illustrating the configuration of heterodimers H31, H32, H33, H34, H35, and H36 formed from the indicated fusion proteins is shown. Individual domains include the single-domain antibodies VHH1 and VHH2, and the CK domain (CK) of human VWF. VWF The half-life extension module ABV2 (anti-RSA single-domain antibody), 1xEP, and 3xEP. 1xEP and 3xEP respectively indicate one and three consecutive repeat sequences of human VWF-derived EP having SEQ ID NO: 2.

[0065] Figure 29 The purified CK is shown. VWF SDS-PAGE analyses of mediated heterodimers H3, H17, H18, H19, H20, H22, H28, H30, and H31 under reducing and non-reducing conditions (lanes 1-9). Analysis in (A) was performed under non-reducing conditions, and analysis in (B) was performed under reducing conditions. (C) SDS-PAGE analyses of H32, H33, H34, and H36 under reducing conditions (lanes 1-4) and non-reducing conditions (lanes 5-8). Molecular weight is expressed in kDa.

[0066] Figure 30 The results of pharmacokinetic (PK) studies of purified heterodimers H3, H5, H17, H18, H19, H20, H22, H28, H30, H31, H32, H33, H34, and H36 in Sprague-Dawley / CD rats are presented. This graph shows the change in the concentration (ng / mL) of the fusion protein dimer in rat plasma over time after intravenous administration of an equimolar dose. Data are presented as mean levels + / - SD obtained from 3 animals at each time point.

[0067] Figure 31The results of pharmacokinetic (PK) studies of purified heterodimers H5, H17, H20, H31, H33, and H36 in Spra-Dowley / CD rats are presented. This graph shows the change in dimer concentration (ng / mL) in rat plasma over time after administration of an equimolar dose of SC. Data are presented as mean levels + / - SD obtained from 3 animals at each time point.

[0068] Figure 32 The abundance of initial HMWC present in the untreated sample is shown. The average values ​​of different subgroups of heterodimers grouped by the total amount of EP in the molecule are indicated by light gray bars. The HMWC content of dimers containing ABV2 is indicated by dark gray bars. Error bars depict the standard deviation of all samples within a group. Detailed Implementation

[0069] To provide a clear and consistent understanding of the specification and claims, as well as the scope of such terms, the following definitions are provided.

[0070] definition

[0071] As used herein, a "peptide" can consist of any number and type of amino acids, preferably naturally occurring amino acids linked by peptide bonds. Specifically, a peptide contains at least 3 amino acids, preferably at least 5, 7, 9, 12, or 15 amino acids. Furthermore, there is no upper limit to the length of the peptide. However, preferably, the peptide according to the invention has a length not exceeding 500 amino acids, more preferably not exceeding 300 amino acids, and even more preferably not exceeding 250 amino acids.

[0072] Therefore, the term "peptide" includes "oligopeptide" and "polypeptide," wherein oligopeptides are generally peptides with a length of 2 to 10 amino acids, and polypeptides are generally peptides with a length of more than 10 amino acids.

[0073] Due to post-translational modifications or in vitro chemical modifications, a protein, as used herein, may contain one or more polypeptide chains and optionally additional components. Proteins having more than one polypeptide chain are typically expressed as a single polypeptide chain by a single gene and are cleaved post-translationally. Therefore, the terms "polypeptide" and "protein" are used interchangeably. As used herein, polypeptides and proteins include chemically synthesized proteins as well as naturally synthesized proteins encoded by genes. Polypeptides or proteins can be obtained from natural sources such as human blood or produced as recombinant proteins in cell culture.

[0074] Post-translational modifications (PTMs) refer to the biochemical modifications made to proteins after they have been synthesized, i.e., after they have been translated from messenger RNA (mRNA) by ribosomes. These modifications typically occur in the endoplasmic reticulum or Golgi apparatus of eukaryotic cells and can alter the function, stability, position, or interactions of proteins with other molecules. Examples of PTMs include glycosylation, ubiquitination, sumoylation, prenylation, sulfation, myristylation, palmitoylation, attachment of glycosylphosphatidylinositol (GPI) anchors, and proteolytic cleavage.

[0075] As used herein, the term "fusion protein" refers to a protein created by linking two or more genes that originally encode individual proteins or protein fragments, wherein the components of the fusion protein are linked to each other by peptide bonds, either directly or via peptide linkers. As used herein, the term "fusion" refers to components linked by peptide bonds, either directly or via one or more peptide linkers.

[0076] As used herein, a "pharmaceutically active protein" is a protein that exhibits specific biological, pharmacological, or therapeutic effects on a target, tissue, or system in a living organism. Such proteins exert their activity by regulating, interacting with, or modulating one or more molecular, cellular, or physiological processes associated with a specific disease, symptom, or biological function. In the context of this invention, a pharmaceutically active protein can also achieve its function by modifying the physicochemical or pharmacokinetic properties of a second pharmaceutically active protein. Such pharmaceutically active proteins are also referred to as regulatory proteins.

[0077] According to the present invention, a "peptide linker" is a protein element that connects two fusion proteins, specifically connecting the FVIII heavy chain to the FVIII light chain or the FVIII light chain to the VWF moiety of a peptide. Peptide linkers are also simply referred to as "linkers." Peptide linkers contain structural amino acids, thereby allowing important domain interactions, enhancing stability, and reducing steric hindrance. In addition to structural amino acids, linkers may also contain functional motifs. An EP can be considered as a protein element or part of a linker. The length of a linker can range from 2 to 200 amino acids.

[0078] The "engineered peptide" according to the present invention is a peptide that is not derived from a living organism but contains an artificial amino acid sequence, specifically a peptide with a modular amino acid sequence obtained by combining the amino acid sequences of different proteins (such as flexible motifs or EP).

[0079] According to the present invention, a “flexible motif” is a motif that enhances the flexibility of peptide linkers.

[0080] As used herein, the term "therapeutic protein" refers to pharmaceutically active proteins, that is, proteins with therapeutic effects, i.e., a subgroup of proteins used as active pharmaceutical ingredients.

[0081] According to the present invention, the terms "protein precursor" and "preprotein" refer to inactive proteins (or peptides) that can be converted into active forms through post-translational modifications (e.g., by enzymatic cleavage of a portion of the amino acid sequence).

[0082] The correlation between two amino acid sequences or two nucleotide sequences is described by the parameter "sequence identity". For the purposes of this invention, the Needleman-Wunsch algorithm is used to determine the degree of sequence identity between two amino acid sequences (Needleman and Wunsch, 1970, J. Mol. Biol. 48: 443-453), which is implemented in the Needle program of the EMBOSS package (EMBOSS: The European Molecular Biology Open Software Suite, Rice et al., 2000, Trends Genet. 16: 276-277), preferably version 3.0.0 or later. Optional parameters used are: a gap opening penalty of 10, a gap extension penalty of 0.5, and an EBLOSUM62 substitution matrix (BLOSUM62 in the EMBOSS version). The Needle output marked "Longest Identity" (obtained using the no brief option) is used as the identity percentage, and calculated as follows:

[0083] (Number of identical residues x 100) / (Alignment length - Total number of gaps in alignment).

[0084] For the purposes of this invention, the Needleman-Wunsch algorithm (Needleman and Wunsch, 1970, ibid.) is used to determine the degree of sequence identity between two nucleotide sequences. This algorithm is implemented in the Needle program of the EMBOSS package (EMBOSS: European Molecular Biology Open Software Suite, Rice et al., 2000, ibid.), preferably version 3.0.0 or later. Optional parameters used are: a nick opening penalty of 10, a nick extension penalty of 0.5, and an EDNAFULL substitution matrix (EMBOSS version NCBI NUC4.4). The Needle output labeled "Longest Identity" (obtained using non-summary options) is used as the percentage identity, and is calculated as follows:

[0085] (Number of identical deoxyribonucleotides x 100) / (Alignment length - Total number of gaps in the alignment).

[0086] When the term "recombinant" is used to refer to cells, nucleic acids, proteins, or vectors, it means that the cells, nucleic acids, proteins, or vectors have been modified by introducing heterologous nucleic acids or proteins, or by altering native nucleic acids or proteins, or that the cells are derived from cells that have been so modified. Thus, for example, recombinant cells express genes that are not present in the natural (non-recombinant) form of the cell, or express naturally occurring genes at different levels or under different conditions.

[0087] As used herein, the term "half-life" is the time required for plasma / blood concentration to decrease by 50% after reaching quasi-equilibrium of distribution (as defined in Toutain et al., 2005). The term "half-life" is also referred to as "circulating half-life," "terminal half-life," or "elimination half-life."

[0088] As used herein, when the terms “transformation,” “stable transformation,” and “transgenic” are used to refer to a cell, it means that the cell contains a non-natural (e.g., heterologous) nucleic acid sequence that is integrated into its genome or carried as an episome and maintained across multiple generations.

[0089] As used herein, the term "fragment" refers to a polypeptide that, compared to a native or wild-type protein, lacks one or more amino acids at its N-terminus and / or C-terminus, but whose remaining amino acid sequence is identical to the corresponding position in the amino acid sequence deduced from full-length cDNA. Fragments are typically at least 50 amino acids in length.

[0090] As used herein, the term "glycosylation" refers to the linkage of a glycan to a molecule, such as a protein. Glycosylation can be an enzymatic reaction. Linkages can also be formed via covalent bonds. Therefore, as used herein, a glycosylated polypeptide is a polypeptide linked with one or more glycans. The phrase "highly glycosylated" means that molecules such as enzymes have undergone glycosylation at all or almost all available glycosylation sites (e.g., O-linked or N-linked glycosylation sites).

[0091] As used herein, the term "glycan" refers to a polysaccharide or oligosaccharide, or a carbohydrate segment of a glycoprotein or glycosylated polypeptide. Glycans can be homopolymers or heteropolymers of monosaccharide residues. Glycans typically contain at least three sugars and can be linear or branched. Glycans can include neutral sugar residues (e.g., glucose, N-acetylglucosamine (GlcNAc), N-acetylgalactosamine (GalNAc), galactose, mannose, fucose, arabinose, ribose, xylose, etc.), charged sugars (e.g., N-acetylneuraminic acid (sialic acid, NeuAc)), and / or modified sugars (e.g., 2'-fluororibose, 2'-deoxyribose, mannose phosphate, 6'-sulfonyl-N-acetyl-glucosamine, etc.).

[0092] As used herein, the term “O-glycan” refers to glycans that are typically covalently linked to serine and threonine residues of mammalian glycoproteins.

[0093] O-glycans can be linked to the -OH group of serine or threonine via an α-linked O-glycosidic bond on the GalNAc moiety. Other linkages include α-linked O-fucose, β-linked O-xylose, α-linked O-mannose, β-linked O-GlcNAc, α- or β-linked O-galactose, and α- or β-linked O-glucan.

[0094] According to the present invention, the terms “O-glycosylated cluster”, “O-glycan cluster” and “cluster of O-glycosylated amino acids” are used interchangeably and refer to two or more O-glycosylated amino acids.

[0095] As used herein, the term “sialylation” refers to the reaction of molecules, particularly polysaccharides, with sialic acid or its derivatives.

[0096] The transitional term "comprising," synonymous with "including," "containing," or "characterized by," is inclusive or open-ended and does not exclude additional unlisted elements or method steps. The transitional phrase "consisting of," except for impurities usually associated with it, excludes any element, step, or component not specified in the claim. When the phrase "consisting of" appears in the body clause of a claim rather than immediately following the preamble, it limits only to the elements set forth in that clause; other elements are not excluded as a whole from the claim. The transitional phrase "consisting essentially of" limits the scope of the claim to the specified materials or steps, and those materials or steps that do not materially affect the basis and novelty of the claimed invention. Claims "consisting essentially of" occupy an intermediate position between closed claims written in the "consisting of" format and fully open claims drafted in the "comprising" format.

[0097] In the context of this invention, the term "glycosylated protein," such as fusion protein, is used in the singular for practical reasons. Typically, in practice, proteins exist as compositions of protein molecules of the same type. However, in the case of glycosylated proteins, the glycosylation in each molecule of the composition is not uniform. For example, not all individual molecules in the composition achieve 100% glycosylation. Furthermore, differences may arise in the binding of glycans to specific O-glycosylation sites. Therefore, in this application, the reference to "fusion protein" also refers to compositions of fusion protein molecules having the same amino acid sequence but different O-glycan structures.

[0098] As used herein, the term “binding affinity” or “affinity” refers to the strength of binding between two molecules, particularly the strength of binding between a ligand and a protein target. Binding affinity is influenced by non-covalent intermolecular interactions between the two molecules, such as hydrogen bonds, electrostatic interactions, hydrophobic interactions, and van der Waals forces.

[0099] As used herein, “immune response” refers to either adaptive or innate immune responses. An innate immune response is a non-specific defense mechanism activated immediately or within hours of the presence of an antigen in the body. These mechanisms include physical barriers such as the skin, chemicals in the blood, and immune system cells that attack foreign cells within the body. Innate immune responses are activated by the chemical properties of the antigen. An adaptive immune response, on the other hand, is an antigen-specific immune response. For this to occur, the antigen must first be processed and recognized. Once the antigen is recognized, the adaptive immune system produces a large number of immune cells specifically designed to attack that antigen.

[0100] The term "VWF" refers to the blood protein von Wöhlerbrand factor.

[0101] As used herein, “extended peptide” or “EP” refers to a VWF-based O-glycosylation cluster 1 (SEQ ID NO: 2) glycosylated peptide.

[0102] The term "EP assembly" refers to two or more consecutive extended peptides.

[0103] As used in this article, "CK" VWF "Refers to the domain of the cystine knot structure based on VWF (SEQ ID NO:1).

[0104] As used in this article, "VHH domain" or "VHH fragment" or "VHH" refers to a single-domain antibody engineered from a heavy chain antibody found in camels.

[0105] As used herein, the terms “mammal” and “of mammals” refer to any vertebrate, including monotremes, marsupials, and placental mammals, that nurses their young and gives birth to live offspring (eumammals or placental mammals) or lays eggs (Metotherae or nonplacental mammals). Examples of mammal species include humans and other primates (e.g., monkeys, chimpanzees), rodents (e.g., rats, mice, guinea pigs), and ruminants (e.g., cows, pigs, horses).

[0106] As used herein, the expressions “C-terminus” and “N-terminus” define the portions of a protein closest to their C-terminus and N-terminus, or define the position of one fused portion relative to another. “N-terminus” means the first portion is closer to the N-terminus than the second portion. “C-terminus” means the first portion is closer to the C-terminus than the second portion.

[0107] Protein dimers and CK fusion proteins

[0108] According to a first aspect, the present invention provides a protein dimer formed of a first fusion protein and a second fusion protein, wherein the first fusion protein includes a cysteine ​​knot (CK) domain fused to a first pharmaceutically active protein, and the second fusion protein includes a CK domain fused to a pharmaceutically active protein, wherein the two fusion proteins are covalently linked through their CK domains.

[0109] A cysteine ​​knot (CK) is a protein structural motif containing three disulfide bridges formed by paired cysteine ​​residues. A portion between two segments of a polypeptide forms a loop, through which a third disulfide bond passes, thus forming a rotaxane substructure. CK motifs stabilize protein structures and are conserved across different species (Vitt et al. 2001; Sherbet 2011). There are three types of CK motifs with different disulfide bond topologies (Daly and Craik 2011): growth factor CK (GF-CK), inhibitory CK (I-CK) commonly found in spider and snail venoms, and cyclic CK (C-CK) or macrocyclic oligopeptides.

[0110] Therefore, the CK domain can be a growth factor CK, an inhibitor CK, or a cyclic CK. GF-CK is found in TGF-β, BMP, VEGF, NGF, and similar growth factors. It is characterized by a ring formed by two disulfide bonds and a peptide backbone, with a third disulfide group passing through the ring. I-CK is found in spider venom and snail venom, such as knottin, and in some protease inhibitors. C-CK is found in plant-derived peptides and is characterized by a cyclic backbone. Due to its cyclic nature, it forms extremely stable and knotted disulfide bonds.

[0111] VWF is a multimeric adhesion glycoprotein found in mammalian plasma with a variety of physiological functions. During primary hemostasis, VWF acts as an intermediary between specific receptors on the platelet surface and extracellular matrix components such as collagen. Furthermore, VWF acts as a carrier and stabilizer of procoagulant factor VIII. VWF is synthesized in endothelial cells and megakaryocytes as a 2813-amino acid precursor molecule. After being secreted into plasma, VWF circulates in various species with different molecular sizes. These VWF molecules consist of oligomers and polymers of mature subunits of 2050 amino acid residues. VWF can exist in plasma in polymeric form, ranging in size from approximately 500 kDa to 20,000 kDa (Furlan 1996).

[0112] The precursor polypeptide (pre-pro-VWF) consists of a 22-residue signal peptide, a 741-residue propeptide (domains D1-D2), and a 2050-residue polypeptide found in mature plasma von Wylerbrand factor (Fischer et al., 1994). The full-length VWF is identified by UniprotKB entry P04275 (version 224, April 12, 2017). The domain organization of the VWF is typically characterized as D3-TIL4-A1-A2-A3-D4-C1-C2-C3-CK. In the VWF, the cysteine ​​knot domain is located at the C-terminus of the protein.

[0113] Cystine knot domains in VWF (CK) VWF The primary function of the cystine knot is to provide structural stability to the protein. The cystine knot is formed by a unique arrangement of three disulfide bonds and a series of interconnected rings. This configuration creates a highly stable, compact, and rigid structure, thus contributing to the overall stability of the VWF protein. CK VWF It mediates the dimerization of proVWF in the endoplasmic reticulum and is essential for the long multimer required for hemostasis.

[0114] Norrin is a secreted signaling protein encoded by the NDP (Norrie Disease protein) gene. It plays a crucial role in vascular and neural development, particularly in the eye, ear, and central nervous system. Unlike many other growth factors, Norrin primarily functions by activating the Wnt / β-catenin signaling pathway. Norrin CK NDP The CK domain has the amino acid sequence of SEQ ID NO: 100.

[0115] The inventors were able to extract CK from the surrounding structure of VWF. VWF Structural domain, and use this CK VWF Domains create various fusion proteins, including those with a CK. VWF The fused VWF fragment with EP and another CK VWF The fused FVIII with missing B domain (BDD-FVIII). The fused CK. VWF The domains form stable dimers, which are covalently bonded to each other. Therefore, the VWF fragment and the FVIII fusion protein communicate via their CK... VWF The three disulfide bonds in the domain are stably dimerized. This provides a very strong connection between the two fusion proteins.

[0116] Disulfide bridges, also known as disulfide bonds, are covalent bonds formed between two cysteine ​​residues or between two separate protein chains within a protein. These bridges play a crucial role in stabilizing the tertiary and quaternary structures of proteins, thereby contributing to their overall function and stability.

[0117] Bond energy, or bond dissociation energy, is the energy required to break a specific covalent bond (in this case, a disulfide bond). For disulfide bonds, bond energies typically range from 50 kcal / mol to 70 kcal / mol (210 kJ / mol to 290 kJ / mol). In the case of three disulfide bonds connecting two proteins, the total bond energy can be estimated by adding the bond energies of each individual bridge. Assuming the average bond energy of each disulfide bond is 60 kcal / mol (251 kJ / mol), then CK... VWF The total bond energy of the three disulfide bridges in the dimer is approximately 180 kcal / mol (753 kJ / mol).

[0118] Therefore, the bond energies of fusion proteins can be in the ranges of 400 kJ / mol to 1150 kJ / mol, 450 kJ / mol to 1100 kJ / mol, 500 kJ / mol to 1050 kJ / mol, 550 kJ / mol to 1000 kJ / mol, 600 kJ / mol to 950 kJ / mol, 650 kJ / mol to 900 kJ / mol, and 700 kJ / mol to 850 kJ / mol.

[0119] As shown in the examples, the cysteine ​​knot domain of VWF forms a stable dimer in the context of the fusion protein. Example 10 below uses the CK domain of Norrin (CK... NDP This confirms the point.

[0120] Through the fusion of CK VWF Domain-mediated dimerization allows for the preparation of novel heterodimers of VHH-CK fusion proteins with various layouts. These include those in which CK... VWF The structural domain is placed in VHH-CK VWF The dimer configuration at the C-terminus or N-terminus of the fusion protein, or where CK... VWF The domains are positioned in a dimer configuration at the C-terminus of the first fusion protein and the N-terminus of the second fusion protein (or vice versa). Furthermore, the inventors also isolated CK. NDP It was also confirmed that a CK domain fusion protein can be formed, specifically VHH-CK. NDP Stable dimers of fusion proteins.

[0121] CK VWF and CK NDP Both belong to the GF-CK type. According to one embodiment, the CK domain is a GF-CK domain. According to a preferred embodiment, the CK domain is selected from CK... VWF and CK NDP According to another preferred embodiment, the CK domain is CK. VWF Structural domain.

[0122] According to one embodiment, CK VWF The domain has an amino acid sequence that is at least 90% identical to SEQ ID NO: 1. This identity can be 90%, 91%, 92%, 93%, 94%, 95%, 95.5%, 96%, 96.5%, 97%, 97.5%, 98%, 98.5%, 99%, 99.5%, or 100%. According to one embodiment, CK... VWF The domain has an amino acid sequence that is at least 95% identical to SEQ ID NO: 1. According to one embodiment, CK... VWF The domain has an amino acid sequence that is at least 98% identical to SEQ ID NO: 1. Example 6 shows a domain with CK. VWF A heterodimer of a domain variant (i.e., SEQ ID NO: 37) having the following amino acid substitutions: A2785I, V2794A, and L2799A.

[0123] According to one embodiment, CK NDP The domain has an amino acid sequence that is at least 90% identical to SEQ ID NO: 100. This identity can be 90%, 91%, 92%, 93%, 94%, 95%, 95.5%, 96%, 96.5%, 97%, 97.5%, 98%, 98.5%, 99%, 99.5%, or 100%. According to one embodiment, CK... NDP The domain has an amino acid sequence that is at least 95% identical to SEQ ID NO: 100. According to one embodiment, CK... NDP The domain has an amino acid sequence that is at least 98% identical to that of SEQ ID NO: 100.

[0124] In addition to strong binding, CK VWF The domains also provide several other functions that are beneficial to fusion couplers and therefore also beneficial to the protein dimers according to the invention. For example, compared to BDD-FVIII, the protein dimer (FVIII-CK) according to the invention... VWF + VWF-CK VWF (FVIII-CK) VWF-VWF) indicates an increased level of expression.

[0125] CK structural domain, specifically CK VWF The structural domain has the advantage of providing additive stabilizing effects to its fusion partner. (Compared to CK) VWF Comparison of the pharmacologically active protein with the domain-linked structure (VHH domain) and the same protein bound to the Fc domain revealed greater thermal stability and greater stability against oscillating stress during culture and against pH changes during purification (see Example 12). Additionally, the CK domain, specifically the CK... VWF The structural domains stabilize the fusion coupler during folding. When heated above the melting temperature, it interacts with CK. VWF The bound protein refolds (while the same VHH bound to the Fc domain does not refold), and shows almost no aggregation (see Example 12). Furthermore, its small size allows for efficient expression of the fusion protein.

[0126] Therefore, two pharmaceutically active proteins acting in combination (such as a single-domain antibody binding to coagulation factor IX (FIX) and a second single-domain antibody binding to coagulation factor X (FX) or FVIII and a VWF fragment binding to FVIII) pass through the CK domain, specifically CK VWF The advantage of domain dimerization is that pharmaceutically active proteins form stable covalently linked dimers in cells, which stabilize one or two couplers and allow for efficient expression and transport to the cell supernatant.

[0127] CK structural domain, specifically CK VWF This serves as the structural basis for constructing bispecific to tetraspecific therapeutic drugs. It consists of two CK fusion protein monomers, specifically CK... VWF The heterodimer formed by the fusion protein monomer allows for the interaction of two CK domains, specifically two CK domains. VWF A combination of up to four distinct protein motifs fused to the C-terminus and / or N-terminus of the CK domain. The protein may also incorporate either end (i.e., the CK domain). VWF The fusion of the N-terminus or C-terminus of the CK domain allows for different three-dimensional orientations and distances, which can be further modified by introducing protein linkers with individual properties. Fusiond proteins can possess different functional properties, such as providing binding to specific targets, enzymatic activity, inhibitory properties, prolonged half-life, receptor activation, or aggregation. Specifically, CK domain pairs... VWF The availability of four fusion sites within the domain pair allows for the construction of two fusion proteins with up to four functionally distinct protein motifs.

[0128] Therefore, CK VWFDomain fusion protein dimers, specifically CK VWF Domain fusion protein dimers are highly modular systems that allow for the combination of several pharmaceutically active proteins. This allows for the addition of various therapeutic functions, including regulatory proteins that modify the function of other pharmaceutically active proteins. However, the protein dimers according to the invention may include more than two or more pharmaceutically active proteins. Due to the modular design, other functional portions can also be introduced into each fusion protein, and thus into the protein dimers according to the invention. Specifically, as shown in the examples, EP or other half-life extension portions can be introduced.

[0129] For example, the insertion of EP increases storage stability and cycling half-life. Therefore, according to one embodiment, the first fusion protein and / or the second fusion protein comprises at least one copy of EP. As shown in the example, the EP having the sequence QEPGGLVVPPTDAPVSPTTLYVEDISEPPLH (SEQ ID NO: 2), namely the O-glycosylation cluster 1 of VWF (amino acids 1238-1268 of SEQ ID NO: 6), imparts increased expression levels, improved stability, and reduced aggregation tendency to the fusion protein according to the invention. Therefore, the EP according to the invention has an amino acid sequence with at least 90% identity to SEQ ID NO: 2. This identity can be 90%, 91%, 92%, 93%, 94%, 95%, 95.5%, 96%, 96.5%, 97%, 97.5%, 98%, 98.5%, 99%, 99.5%, or 100%. According to one embodiment, the sequence identity with SEQ ID NO: 2 is preferably at least 95%. According to one embodiment, the sequence identity of EP with SEQ ID NO: 2 is at least 98%. Example 6 shows a fusion protein having a sequence variant of SEQ ID NO: 2 (i.e., SEQ ID NO: 42) having the following amino acid substitution: G5A. According to one embodiment, two or more EPs have 100% sequence identity with SEQ ID NO: 2. According to one embodiment, the first fusion protein and / or the second fusion protein contains at least one copy of EP. For example, the first fusion protein and / or the second fusion protein contains 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 copies of EP. As shown in Example 14, by increasing the number of EPs, the tendency of fusion protein aggregation (i.e., the formation of high molecular weight components (HMWC)) is reduced. Specifically, the HMWC level decreases proportionally to the increase in the number of EPs in the fusion protein or fusion protein dimer. Fusion proteins with one copy of EP aggregate less than fusion proteins without EP. Fusion proteins with two copies of EP aggregate less than fusion proteins with one copy. According to one embodiment, the first and / or second fusion protein contains at least two copies of EP. Fusion proteins with four copies of EP aggregate less than fusion proteins with two copies. According to one embodiment, the first and / or second fusion protein contains at least four copies of EP. Fusion proteins with six copies of EP aggregate less than fusion proteins with four copies. According to one embodiment, the first and / or second fusion protein contains at least six copies of EP. Fusion proteins with nine copies of EP aggregate less than fusion proteins with six copies.According to one embodiment, the first fusion protein and / or the second fusion protein contain at least 9 copies of EP.

[0130] According to one embodiment, the first fusion protein and / or the second fusion protein comprises at least one half-life extension portion. Various half-life extension portions are known in the art. The half-life extension portion may be selected from immunoglobulin Fc domains, serum albumin, or portions thereof, or albumin-binding molecules. The Fc domain is the crystallizable fragment (Fc) region of the tail region of an IgG antibody, which interacts with a cell surface receptor called an Fc receptor. This interaction, along with the slower renal clearance of the larger molecule, increases the half-life of the protein to which it is linked. An exemplary Fc domain is the Fc domain of IgG1. Several fusion proteins containing albumin to increase the half-life of therapeutic proteins have been described, including fusions of factors VII, FVIII, and IX with albumin. According to one embodiment, full-length human serum albumin (HSA) is added to the fusion protein. The half-life of albumin is also regulated by an Fc receptor, namely the neonatal Fc receptor (FcRn). The HSA sequence added to the fusion protein preferably has the sequence according to Uniprot entry P02768.

[0131] Albumin-binding molecules can be selected from the group consisting of antibodies, antibody fragments, antibody mimics, and other albumin-binding proteins or portions thereof. Albumin-binding proteins and protein domains can include native or engineered albumin-binding domains of streptococcal protein G, albumin-binding fibronectin type III (Fn3) domains, albumin-binding single-domain antibodies, and engineered lipocalin.

[0132] According to one embodiment, the antibody fragment is selected from the group consisting of: Fab fragments, F(ab')2 fragments, Fab' fragments, and single-domain antibodies, specifically VHH domains. The Fab fragment is an antibody fragment consisting of variable regions of the heavy and light chains of the antibody, and a first constant region of the heavy chain. The Fab fragment can be generated by enzymatic digestion of a full-length antibody with papain. The F(ab')2 fragment is an antibody fragment consisting of two Fab fragments linked together by disulfide bonds. The F(ab')2 fragment can be generated by enzymatic digestion of a full-length antibody with pepsin. The Fab' fragment is an antibody fragment consisting of variable regions of the heavy and light chains of the antibody, and a portion of the constant region of the heavy chain. The Fab' fragment can be generated by enzymatic digestion of a full-length antibody with papain, followed by reduction of the disulfide bonds linking the heavy chains. These fragments are commonly used in research and diagnostic applications, and the methods used to generate the antibody fragments are not particularly limited and are known in the art.

[0133] Single-domain antibodies (sdAbs) are antibody fragments composed of a single monomeric variable antibody domain. Like whole antibodies, single-domain antibodies selectively bind to specific antigens. With a molecular weight of only 12-15 kDa, single-domain antibodies are much smaller than common antibodies (150-160 kDa). The first single-domain antibody was engineered from a heavy-chain antibody found in camels and was termed the VHH fragment or VHH domain (Hamers-Casterman et al., 1993). An alternative approach is to split the dimeric variable domain of common human or mouse immunoglobulin G (IgG) into monomers. Although most current research on single-domain antibodies is based on heavy-chain variable domains, single-domain antibodies derived from light chains have also been shown to bind specifically to target epitopes.

[0134] The antibody mimics according to the present invention can be selected from the group consisting of: single-chain variable fragments (scFv), affinity molecules, affilin, affimer, affitin, anticalin, DARPin, monofunctional antibodies, and peptide aptamers.

[0135] Single-chain variable fragments (scFvs) are a type of antibody fragment that consists of variable domains of the antibody's heavy and light chains linked together by short peptide linkers.

[0136] Affinities are small protein scaffolds engineered to bind to specific targets with high affinity and specificity. They are based on the binding domain of protein A, which acts as a natural ligand for the Fc region of an antibody. Affilins are a type of small protein scaffold engineered to bind to specific targets with high affinity and specificity. They are based on the ubiquitin family. Affimers are a type of protein scaffold engineered to bind to specific targets with high affinity and specificity. They are based on the cystatin family of cysteine ​​protease inhibitors, which act as cysteine ​​protease inhibitors in nature. Affitins are a type of protein scaffold engineered to bind to specific targets with high affinity and specificity. They are derived from the DNA-binding protein Sac7d. Anticalins are a type of protein scaffold engineered to bind to specific targets with high affinity and specificity. They are based on a protein called a lipid carrier protein, which is involved in the transport of small hydrophobic molecules. DARPin, or designed ankyrin repeat protein, is a type of protein scaffold engineered to bind to a specific target with high affinity and specificity. They are based on the ankyrin repeat protein family. Monofunctional antibodies are a type of antibody mimicry composed of a single protein domain engineered to bind to a specific target with high affinity and specificity. They are based on the fibronectin type III domain. Peptide aptamers are a type of protein scaffold composed of short peptide sequences engineered to bind to a specific target with high affinity and specificity. They are typically generated using phage display or other selective methods. There are no particular limitations on the corresponding antibody mimics and their generation methods, and they are known in the art.

[0137] According to one embodiment, the extended half-life portion is an albumin-binding VHH domain (ABV). Albumin-binding VHH domains are known in the art. An example of an albumin-binding VHH domain is MSA21 described in EP 2316852 B1. According to one embodiment, the albumin-binding VHH domain is the ABV shown in the example having SEQ ID NO: 36. The albumin-binding VHH domain according to the invention may have an amino acid sequence having at least 90% identity with SEQ ID NO: 36. This identity may be 90%, 91%, 92%, 93%, 94%, 95%, 95.5%, 96%, 96.5%, 97%, 97.5%, 98%, 98.5%, 99%, 99.5%, or 100%. According to one embodiment, the sequence identity with SEQ ID NO: 36 is preferably at least 95%. According to one embodiment, the albumin-binding VHH domain has at least 98% sequence identity with SEQ ID NO: 36.

[0138] The half-life extension portion may fuse with the N-terminus and / or C-terminus of the pharmaceutically active protein in the first fusion protein. The half-life extension portion may fuse with the N-terminus and / or C-terminus of the pharmaceutically active protein in the second fusion protein. According to one embodiment, the half-life extension portion fuses with the N-terminus and / or C-terminus of the first and / or second pharmaceutically active proteins. According to one embodiment, the half-life extension portion fuses with the CK in the first and / or second fusion protein. VWF The N-terminus and / or C-terminus of the domain are fused. Alternatively, the half-life extension portion forms part of the linker in the first and / or second fusion proteins.

[0139] The components of the fusion protein can be linked via peptide linkers. The linker length is preferably in the range of 2 to 200 amino acids. Linker lengths can be 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, and 95 amino acids. 100 amino acids, 105 amino acids, 110 amino acids, 115 amino acids, 120 amino acids, 125 amino acids, 130 amino acids, 135 amino acids, 140 amino acids, 145 amino acids, 150 amino acids, 155 amino acids, 160 amino acids, 165 amino acids, 170 amino acids, 175 amino acids, 180 amino acids, 185 amino acids, 190 amino acids, 195 amino acids, or 200 amino acids.

[0140] Peptide linkers can be rigid or flexible. Linkers can contain flexible motifs, which give the linker flexibility to allow for sufficient dexterity. The most flexible amino acids are the β-turn-forming amino acids glycine and serine (Huang and Nau, 2003). Therefore, a flexible motif can be a glycine-serine repeat sequence. A glycine-serine repeat sequence is a repeating unit of glycine and serine, such as (GGS). n Or (GGGGS) n They are common in flexible linkers. Glycine provides flexibility due to its small size, while serine increases the stability of the linker in aqueous solutions by forming hydrogen bonds with water molecules, thereby reducing adverse interactions with other protein motifs (Chen et al., 2013). Glycine-serine repeat sequences can be selected from (GS). n (GGS) n (GGGS) n and (GGGGS) n n can be an integer in the range of 1 to 10. n can be 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. Consistent with the single-letter amino acid codes, G represents glycine, and S represents serine. GGGS is SEQ ID NO: 75, and GGGGS is SEQ ID NO: 76.

[0141] Glycine-serine repeat sequences are examples of flexible repeat sequences. However, other suitable flexible motifs, such as (GGGX), exist. n (GXG) n elastin-like motifs and (SG)n Or (TG) n Repeated sequences are also suitable for the fusion proteins of the protein dimers of the present invention.

[0142] (GGGX) n The GGGGS motif is modified by altering the fourth position, where a different amino acid (X) occupies the position. This amino acid can be serine, threonine, or another small polar amino acid. This modification maintains flexibility while potentially altering the hydrophilicity, steric hindrance, or conformational preference of the linker.

[0143] (GXG) n This motif comprises glycine residues separated by different amino acids (X), which may be serine, alanine, or other small residues. The presence of glycine ensures flexibility, while the intermediate amino acids modulate the properties of the linker.

[0144] Elastin-like motifs: such as "(VPGXG)" n The sequence VPGXG is inspired by elastin and can impart elasticity and flexibility, where V is valine, P is proline, G is glycine, and X can be any amino acid (typically small amino acids like alanine). These motifs can also undergo reversible phase transitions in response to temperature changes, which can be used for protein purification and the design of responsive materials. VPGXG is SEQ ID NO: 77.

[0145] (Ser-Gly) n Or (Thr-Gly) n Repeating sequences: Due to the presence of serine or threonine, these motifs (repetitive units with serine or threonine followed by glycine) provide flexibility and a higher degree of hydrophilicity. They can be used to improve the solubility of protein constructs.

[0146] To provide rigidity and well-defined spatial orientation, other motifs, such as polyproline (PPII) helices, can be used.

[0147] Polyproline (PPII) helices: Proline-rich sequences can form a type of secondary structure called polyproline helices. While not as flexible as glycine-rich linkers, polyproline helices provide a rigid and extended structure that can separate functional domains. They are often used when a more defined distance or orientation between domains is desired.

[0148] In addition, the peptide linker may contain one or more copies of EP. The peptide linker may contain a half-life extension moiety, preferably albumin-bound VHH (ABV).

[0149] According to one embodiment, one of the pharmaceutically active proteins, the FVIII protein, contains a first linker ( Figure 1 (Connector 1 in the text). According to one embodiment, the CK of the first fusion protein... VWF The structural domain is connected through a second connector ( Figure 1 The second linker 2) is fused to the pharmaceutically active protein. The second linker can be a flexible or rigid linker. According to one embodiment, the linker length is in the range of 5 to 180 amino acids. According to one embodiment, the linker length is in the range of 10 to 160 amino acids. According to one embodiment, the linker length is in the range of 12 to 140 amino acids.

[0150] Preferably, the second connector is a flexible connector. According to one embodiment, the second connector includes a flexible sequence. The flexible sequence is selected from (GGS). n (GGGS) n and (GGGGS) n In this embodiment, n is an integer in the range of 1 to 10. n can be 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. Consistent with the single-letter amino acid codes, G represents glycine and S represents serine. These motifs give the second linker flexibility, thereby allowing sufficient interaction, specifically binding to the VWF and FVIII binding domains. According to one embodiment, the second linker contains (GGGGS)2 and / or (GGGGS)4 motifs.

[0151] According to one embodiment, two consecutive copies of the GGGGS motif are located at the N-terminus and / or C-terminus of the second connector. According to one embodiment, the (GGGGS)2 motif is located at the N-terminus. According to one embodiment, the (GGGGS)2 motif is located at the C-terminus. According to one embodiment, (GGGGS... n The motif (where n ≥ 2) is located at the C-terminus. According to one embodiment, (GGGGS)2 (SEQ ID NO: 78), (GGGGS)4 (SEQ ID NO: 79), (GGGGS)6 (SEQ ID NO: 80), or (GGGGS)8 (SEQ ID NO: 81) is located at the C-terminus of the second linker. The longer, more flexible linker facilitates interaction between the first pharmaceutically active protein and its binding partner or with the second pharmaceutically active protein. Therefore, (GGGGS) n A connector (where n ≥ 2) is preferred. According to one embodiment, the (GGGGS)2 motif is located at each of the N-terminus and the C-terminus.

[0152] According to one embodiment, the second connector is cleavable. According to one embodiment, the second connector includes a thrombin cleavage site. The advantage of having a protease cleavage site is that the first pharmaceutical protein FVIII can be separated from the second pharmaceutically active protein (e.g., a VWF fragment). The thrombin cleavage site may be defined by SEQ ID NO: 19. The reason for choosing the thrombin cleavage site is that it is part of the natural FVIII sequence.

[0153] According to one embodiment, the second connector contains at least one copy of the EP. As shown in Example 2, the second connector has an FVIII-CK EP. VWF -VWF fusion protein dimers, namely C27, C30, and C31, showed increased expression levels and prolonged half-life. Additionally, as shown in Example 9, this type of FVIII-CK contains EP in the second linker. VWF -VWF fusion protein dimer (i.e., C32) exhibits improved thrombin generation kinetics in vitro and in vivo. Figure 15 The second connector can, for example, contain one, two, three, four, five, six, seven, or eight copies of the EP. The example shown is FVIII-CK. VWF Proteins C27, C30, C31, and C32 have three EPs in the second linker. Therefore, according to one embodiment, the second linker contains at least two copies of the EPs. According to one embodiment, the second linker contains at least three copies of the EPs. The EPs may be distributed along the length of the linker, with structural amino acids or other elements of the linker between them. Alternatively, two or more EPs may be assembled adjacently (i.e., in a continuous sequence). According to one embodiment, all EPs in the second linker are assembled in a continuous sequence. According to one embodiment, the second linker is formed solely by the assembly of EPs. According to one embodiment, the second linker contains at least two copies of a flexible motif (specifically, a GGS, GGGS, or GGGGS motif at the C-terminus and / or N-terminus of the EP assembly). According to one embodiment, the second linker contains at least two copies of a C-terminal GGGGS motif of the EP assembly.

[0154] According to a preferred embodiment, the half-life extension portion is part of the second connector. According to another preferred embodiment, the half-life extension portion is ABV and is part of the second connector. According to one embodiment, the first connector contains at least two copies of a flexible motif (specifically, a GGS, GGGS, or GGGGS motif at the C-terminus and / or N-terminus of ABV). According to one embodiment, the first connector contains at least two copies of a GGGGS motif at both the C-terminus and N-terminus of ABV.

[0155] According to one embodiment, the CK of the second fusion protein VWFThe domain fuses with the pharmaceutically active protein via a third linker. The length of the third linker is preferably in the range of 2 to 200 amino acids. According to one embodiment, the linker length is in the range of 10 to 160 amino acids. According to another embodiment, the linker length is in the range of 12 to 140 amino acids.

[0156] The third connector can be a flexible connector or a rigid connector. Preferably, the third connector is a flexible connector. According to one embodiment, the third connector comprises components preferably selected from (GGS). n (GGGS) n and (GGGGS) n The flexible motifs. In this embodiment, n is an integer in the range of 1 to 10. n can be 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. Consistent with the single-letter amino acid codes, G represents glycine and S represents serine. These motifs give the third linker flexibility to allow sufficient interaction between the second pharmaceutical protein and its binding partner or with the first pharmaceutically active protein, and specifically to bind to the VWF and FVIII binding domains. According to one embodiment, the third linker contains (GGGGS)2, (GGGGS)4, and / or (GGGGS)6 motifs.

[0157] According to one embodiment, the third linker contains at least one copy of the EP. The construct used in the examples contains two EPs fused to the C-terminus of the VWF fragment, and thus can be considered part of the linker. The third linker may, for example, contain one, two, three, four, five, six, seven, or eight copies of the EP. According to one embodiment, the third linker contains at least two copies of the EP. According to one embodiment, the third linker contains at least three copies of the EP. The EPs may be distributed along the length of the linker, with structural amino acids or other elements of the linker between them. Alternatively, two or more EPs may be assembled adjacently (i.e., in a continuous sequence). According to one embodiment, all EPs in the third linker are assembled in a continuous sequence. According to one embodiment, the third linker is formed by assembling only EPs. According to one embodiment, the third linker contains at least two copies of a flexible motif (specifically, a GGS, GGGS, or GGGGS motif at the C-terminus and / or N-terminus of the EP assembly). According to one embodiment, the third linker contains at least two copies of a GGGGS motif at the C-terminus of the EP assembly.

[0158] According to one embodiment, the third connector does not contain any complete domains of VWF, and preferably does not contain any other portions of the D' and D3 domains besides EP. According to a preferred embodiment, the half-life extension portion is part of the third connector. According to a preferred embodiment, the half-life extension portion is ABV and is part of the third connector. According to one embodiment, the third connector contains at least two copies of the C-terminal and / or N-terminal GGS, GGGS, or GGGGS motif of ABV. According to one embodiment, the third connector contains at least two copies of the C-terminal and N-terminal GGGGS motif of ABV.

[0159] The first or second pharmaceutically active protein may be a synthetic protein, a naturally occurring protein, or a fragment of the latter. According to one embodiment, the first and second pharmaceutically active proteins are mammalian proteins or fragments thereof. According to one embodiment, the first or second pharmaceutically active protein is preferably a human protein or a fragment thereof.

[0160] According to one embodiment, the pharmaceutically active proteins of the first fusion protein and the second fusion protein may be the same or different. The pharmaceutically active protein may be, individually, a coagulation factor, a transport protein, a protease inhibitor, an immunoglobulin, a cell-associated plasma protein, an apolipoprotein, a complement factor, a growth factor, an anti-angiogenic protein, a highly glycosylated protein, a blood factor, or another blood protein.

[0161] Coagulation factors, specifically human coagulation factors, preferably selected from the group consisting of: fibrinogen (FI), prothrombin (FII), tissue factor (FIII), factor V (FV), factor VII (FVII), factor VIII (FVIII), factor IX (FIX), factor X (FX), factor XI (FXI), factor XII (FXII) and factor XIII (FXIII), VWF and ADAMTS13.

[0162] It should be understood that coagulation factors FI, FII, FV, FVII, FVIII, FIX, FX, FXI, FXII, and FXIII can be in inactive or activated forms. Therefore, in the context of this invention, references to FI, FII, FV, FVII, FVIII, FIX, FX, FXI, FXII, and FXIII respectively include the activated forms FIa (fibrin), FIIa (thrombin), FVa, FVIIa, FVIIIa, FIXa, FXa, FXIa, FXIIa, and FXIIIa, unless otherwise expressly stated or if the activated form is logically excluded from the context. Therefore, for example, in this case, FI, FII, FV, FVII, FVIII, FIX, FX, FXI, FXII, and FXIII can be interpreted as FI / FIa, FII / FIIa, FV / FVa, FVII / FVIIa, FVIII / FVIIIa, FIX / FIXa, FX / FXa, FXI / FXIa, FXII / FXIIa, and FXIII / FXIIIa.

[0163] Transport proteins, specifically human transport proteins, can be selected from albumin, transferrin, ceruloplasmin, haptoglobin, hemoglobin, and hemoglobin-binding protein.

[0164] According to one embodiment, the mammalian protein is a protease inhibitor, specifically a human protease inhibitor. Examples of such protease inhibitors are β-antithrombin, α-antithrombin, pre-latent-antithrombin, oxidized antithrombin, 2-macroglobulin, C1 inhibitor, tissue factor pathway inhibitor (TFPI), heparin cofactor II, protein C inhibitor (PAI-3), protein C, protein S, and protein Z.

[0165] Examples of immunoglobulins (such as polyclonal and monoclonal antibodies) include IgG1, IgG2, IgG3, IgG4, IgA, IgA1, IgA2, IgM, IgE, IgD, and Bence Jones protein.

[0166] Cell-associated plasma proteins can be, for example, fibronectin, thromboglobulin, or platelet factor 4. Examples of apolipoproteins are apolipoprotein AI, apolipoprotein A-II, and apolipoprotein E.

[0167] The complement factors according to the present invention are, for example, factor B, factor D, factor H, factor I, C3b-inactivator, properdin, C4-binding protein, etc.

[0168] Examples of growth factors include platelet-derived growth factor (PDGF), epidermal growth factor (EGF), transforming growth factor α (TGF-α), transforming growth factor β (TGF-β), fibroblast growth factor (FGF), and hepatocyte growth factor (HGF).

[0169] Antiangiogenic proteins include latent antithrombin, pre-latent antithrombin, oxidized antithrombin, and plasminogen.

[0170] Highly glycosylated proteins include α-1-acid glycoprotein, anti-chymotrypsin, α-trypsin inhibitor, α-2-HS glycoprotein, or C-reactive protein. Blood factors may include, for example, erythropoietin, interferon, tumor factors, tPA, or G-CSF.

[0171] Other human blood proteins include histidine-rich glycoproteins, mannan-binding lectins, C4-binding proteins, fibronectin, GC-globulins, plasminogen / plasmin, α-1 microglobulins, and C-reactive proteins.

[0172] The pharmaceutically active proteins are specifically selected from VWF, prothrombin, fibrinogen, FIII, FV, FVII, FVIII, FIX, FX, FXI, FXII, FXIII, ADAMTS13, antithrombin, α-1 antitrypsin, C1 inhibitor, antichymotrypsin, PAI-1, PAI-3, α2-macroglobulin, TFPI, heparin cofactor II, protein C, protein S, protein Z, and fragments of the aforementioned proteins.

[0173] According to one embodiment, the first pharmaceutical protein and / or the second pharmaceutical protein is an antigen-binding molecule. According to one embodiment, the antigen-binding molecule binds to any of the following: FIX, FIXa, FX, VWF, activated protein C, protease microtubule connexin-1, protein Z-dependent protease, antithrombin, protein S, ADAMTS13, platelet GpIIb / IIIa receptor, complement C5, complement C3, P-selectin, human serum albumin, or FcRn.

[0174] Antigen-binding molecules can be selected from the group consisting of antibodies, antibody fragments, and antibody mimics. According to one embodiment, the antibody fragment is selected from the group consisting of Fab fragments, F(ab')2 fragments, Fab' fragments, and single-domain antibodies, specifically VHH domains.

[0175] The antibody mimics according to the present invention can be selected from the group consisting of: single-chain variable fragments (scFv), affinity molecules, affilin, affimer, affitin, anticalin, DARPin, monofunctional antibodies, and peptide aptamers.

[0176] In this regard, according to one embodiment, the first and second pharmaceutically active proteins are specified to be selected from the group consisting of VWF, FVIII, and fragments thereof. Preferably, the CK domain is the CK domain as defined above. VWF Structural domain.

[0177] According to one embodiment, the first pharmaceutical protein and / or the second pharmaceutical protein is a VHH domain. According to one embodiment, the VHH domain binds to any of the following: FIX, FIXa, FX, VWF, activated protein C, protease microtubule connexin 1, protein Z-dependent protease, antithrombin, protein S, ADAMTS13, platelet GpIIb / IIIa receptor, complement C5, complement C3, P-selectin, human serum albumin, or FcRn.

[0178] According to one embodiment, the first pharmaceutically active protein is a FIX-binding VHH domain (VHH1). According to one embodiment, the FIX-binding VHH domain comprises an amino acid sequence having at least 90% identity. This identity can be 90%, 91%, 92%, 93%, 94%, 95%, 95.5%, 96%, 96.5%, 97%, 97.5%, 98%, 98.5%, 99%, 99.5%, or 100%. According to a preferred embodiment, the FIX-binding VHH domain comprises an amino acid sequence having at least 95% identity with SEQ ID NO: 72. According to a more preferred embodiment, the FIX-binding VHH domain comprises an amino acid sequence having at least 98% identity with SEQ ID NO: 72.

[0179] According to one embodiment of the protein dimer, the second pharmaceutically active protein is a VHH domain (VHH2) that binds to FX. According to one embodiment, the VHH domain that binds to FX comprises an amino acid sequence having at least 90% identity with SEQ ID NO: 73. This identity can be 90%, 91%, 92%, 93%, 94%, 95%, 95.5%, 96%, 96.5%, 97%, 97.5%, 98%, 98.5%, 99%, 99.5%, or 100%. According to a preferred embodiment, the VHH domain that binds to FX comprises an amino acid sequence having at least 95% identity with SEQ ID NO: 73. According to a more preferred embodiment, the VHH domain that binds to FX comprises an amino acid sequence having at least 98% identity with SEQ ID NO: 73.

[0180] Heterodimer VHH1-CK VWF-VHH2 forms a VIIIa-mimicking bispecific molecule, which is equivalent to the VIIIa-mimicking bispecific antibody that constitutes the API of Hemlibra.

[0181] Another example is a combination of an immunoglobulin targeting vascular endothelial growth factor A (VEGF-A) and an immunoglobulin targeting delta-like ligand 4 (DLL4). By simultaneously targeting VEGF-A and DLL4, heterodimers can be designed to inhibit tumor angiogenesis and enhance the efficacy of anti-VEGF-A therapy. A known example of a combination of these pharmaceutical activities is ABT-165, a dual variable domain immunoglobulin (DVD-Ig) fusion protein that targets both VEGF-A and DLL4. According to one embodiment, the first pharmaceutically active protein is a VHH domain that binds to VEGF-A. According to one embodiment, the second pharmaceutically active protein is a VHH domain that binds to DLL4.

[0182] Another example is the combination of a first single-chain variable fragment (scFv) that specifically binds to CD19 (a cell surface protein found on B-cell lymphoblasts) and a second scFv that binds to CD3 (a protein on the surface of T cells). By bridging these two immune cell populations, the dimer promotes the formation of immune synapses between T cells and B-cell lymphoblasts, ultimately leading to targeted killing of cancerous B cells. An example of a fusion protein with both activities is Blinatumomab (Blincyto), a bispecific T-cell conjugate (BiTE) antibody construct used to treat acute lymphoblastic leukemia (ALL).

[0183] Another example is ozoralizumab. Ozoralizumab is a 38 kDa humanized trivalent bispecific construct composed of two anti-TNFα single-domain antibodies and an anti-HSA single-domain antibody, used to treat inflammatory diseases.

[0184] An exemplary second connector (and its corresponding sequence ID), specifically VHH–CK VWFThe second linkers of the protein dimer are linker 10-1 (SEQ ID NO: 56), linker 10-2 (SEQ ID NO: 57), linker 10-3 (SEQ ID NO: 58), and linker 10-4 (SEQ ID NO: 59). The amino acid sequences of these linkers are shown in Table 19 (below). Variants of linker 2-1 have the sequence of SEQ ID NO: 46. According to one embodiment, the amino acid sequence of the first linker has at least 90%, at least 95%, or at least 98% identity with sequences selected from the following: SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 45, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, and SEQ ID NO: 59. According to one embodiment, the amino acid sequence of the first linker is selected from SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 45, SEQ ID NO: 46, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58 and SEQ ID NO: 59.

[0185] Exemplary third adapters (and their corresponding sequence IDs) are adapter 10-1 (SEQ ID NO: 56), adapter 10-2 (SEQ ID NO: 57), adapter 10-3 (SEQ ID NO: 58), and adapter 10-4 (SEQ ID NO: 59). The amino acid sequences of these adapters are shown in Tables 18 and 20 (hereinafter). According to one embodiment, the amino acid sequence of the third adapter has at least 90%, at least 95%, or at least 98% identity with sequences selected from the following: SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 55, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, and SEQ ID NO: 59. According to one embodiment, the amino acid sequence of the third adapter is selected from SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, and SEQ ID NO: 59.

[0186] Another example is the combination of FVIII and VWF segments, as shown in the example below.

[0187] According to one embodiment of the first aspect, the present invention provides a protein dimer formed of a first fusion protein and a second fusion protein, wherein the first fusion protein comprises an FVIII protein, preferably containing a first linker, the FVIII protein being coupled to a von Wöhlerbrand factor cystine knot domain (CK) via a second linker. VWF The second fusion protein contains a von Wöhlerbrand factor (VWF) fragment, which binds to CK via a third linker. VWF Domain fusion, wherein the third linker is an engineered peptide, and wherein the two fusion proteins are fused via their CK domains. VWF Covalent connection of structural domains.

[0188] In CK VWF In the case of domain-mediated dimerization, novel variants of VWF and FVIII can be prepared, thereby utilizing the stabilizing effect of VWF on FVIII upon binding. According to one embodiment, the first pharmaceutically active protein is the FVIII protein.

[0189] In humans, factor VIII is encoded by the F8 gene, which contains 187,000 base pairs across six exons. The transcribed mRNA is 9.029 base pairs long and is translated into a 2.351-amino acid protein, with 19 amino acids removed. In humans, the FVIII molecule is glycosylated at 31 amino acids by 25 N-glycosylation chains and 6 O-glycosylation chains (Kannicht et al., 2013).

[0190] Following translation, the amino acid chain is cleaved by a specific protease, resulting in a heavy chain of approximately 200 kDa and a light chain of approximately 80 kDa. The domain organization is typically characterized as A1-A2-B-A3-C1-C2. The light chain is a combination of domains A3-C1-C2. The heavy chain is composed of domains A1-A2-B. According to one embodiment, the FVIII protein comprises an FVIII heavy chain and an FVIII light chain. The FVIII heavy chain found in plasma exhibits a heterogeneous composition with molecular weights ranging from 90 kDa to 200 kDa. This size variation is attributed to heterogeneity in its glycosylation, splicing variants, and the presence of proteolytic products such as the B-domain-depleted heavy chain A1-A2. The full-length amino acid sequence of FVIII is identified by amino acids 20 to 2.351 in UniProtKB P00451 (Sequence Version 1, July 21, 1986).

[0191] The human FVIII heavy chain according to the present invention contains at least domains A1 and A2, and may further contain part or all of the B domain. The amino acid sequence of the human FVIII heavy chain without the B domain is identified by SEQ ID NO: 3. The amino acid sequence of the human FVIII heavy chain including the B domain is identified by SEQ ID NO: 4.

[0192] According to one embodiment, the FVIII heavy chain does not contain the FVIII B domain (BDD-FVIII). In this case, the FVIII heavy chain in the first fusion protein has an amino acid sequence similar to or identical to SEQ ID NO: 3. The heavy chain without the FVIII B domain may contain an amino acid sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identity with SEQ ID NO: 3. Example 6 illustrates a fusion protein having a variant of the FVIII heavy chain sequence of SEQ ID NO: 38, said variant having the following amino acid substitution: V592A. According to one embodiment, the heavy chain is at least 95% identical to SEQ ID NO: 3. According to one embodiment, the heavy chain is at least 98% identical to SEQ ID NO: 3.

[0193] According to one embodiment, the FVIII heavy chain contains an FVIII B domain. In this case, the FVIII heavy chain of the fusion protein has an amino acid sequence similar to or identical to SEQ ID NO: 4. The heavy chain having the FVIII B domain may contain an amino acid sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identity with SEQ ID NO: 4. According to one embodiment, the heavy chain is at least 95% identical to SEQ ID NO: 4. According to one embodiment, the heavy chain is at least 98% identical to SEQ ID NO: 4.

[0194] According to one embodiment, the FVIII light chain comprises the domain organization A3-C1-C2. The human FVIII light chain having the domain organization A3-C1-C2 has the sequence of SEQ ID NO: 5. The FVIII light chain of the fusion protein may have an amino acid sequence similar to or identical to SEQ ID NO: 5. According to one embodiment, the FVIII light chain comprises an amino acid sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identity with SEQ ID NO: 5. According to one embodiment, the light chain is at least 95% identical to SEQ ID NO: 5. According to one embodiment, the light chain is at least 98% identical to SEQ ID NO: 5. Example 6 illustrates a fusion protein having a sequence variant of the FVIII light chain of SEQ ID NO: 39, said variant having the following amino acid substitution: S1732T.

[0195] According to one embodiment, the C-end of the FVIII heavy chain is fused to the N-end of the FVIII light chain via a first connector. The connector connecting the heavy chain and the light chain is known in the art. One example is NUWIQ. ® The connector in the first connector is SFSQNSRHQAYRYRRG (SEQ ID NO: 21). This connector contains a sequence of B domains derived from FVIII. According to one embodiment, the first connector preferably contains a sequence of B domains derived from FVIII.

[0196] The first connector can be a flexible connector or a rigid connector. According to one embodiment, the connector length is in the range of 5 to 180 amino acids. According to one embodiment, the connector length is in the range of 10 to 160 amino acids. According to one embodiment, the connector length is in the range of 12 to 140 amino acids.

[0197] Preferably, the first connector is a flexible connector. According to one embodiment, the first connector includes a flexible sequence. The flexible sequence may be selected from (GGS). n (GGGS) n and (GGGGS) n In this embodiment, n is an integer in the range of 1 to 10. n can be 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. Consistent with the single-letter amino acid codes, G represents glycine, and S represents serine. These motifs give the first linker flexibility, thereby allowing sufficient interaction between the FVIII heavy chain and the light chain.

[0198] According to one embodiment, the first adapter is a cleavable adapter, i.e., it contains a protease cleavage site. The advantage of having a protease cleavage site is that FVIII can be processed intracellularly into its native double-stranded conformation. According to one embodiment, the first adapter comprises a furin protease cleavage site. The furin protease cleavage site is chosen because it is a naturally occurring cleavage site in wild-type FVIII. The furin protease cleavage site may have the amino acid sequence of SEQ ID NO: 20 or SEQ ID NO: 1.

[0199] According to one embodiment, FVIII-CK VWF The first linker of the -VWF heterodimer is an engineered peptide.

[0200] According to one embodiment, the first connector contains at least one copy of the EP. The first connector may, for example, contain one, two, three, four, five, six, seven, or eight copies of the EP. The example shown is FVIII-CK. VWF The -VWF heterodimer has three EPs in the first linker. Therefore, according to one embodiment, the first linker contains at least two copies of the EPs. According to one embodiment, the first linker contains at least three copies of the EPs. The EPs may be distributed along the length of the linker, with structural amino acids or other elements of the linker between them. Alternatively, two or more EPs may be assembled adjacently (i.e., in a continuous sequence). According to one embodiment, all EPs in the first linker are assembled in a continuous sequence. According to one embodiment, the first linker is formed by assembling only EPs. According to one embodiment, the third linker contains at least two copies of a flexible motif (specifically, a C-terminal and / or N-terminal GGS, GGGS, or GGGGS motif assembled by the EPs). According to one embodiment, the third linker contains at least two copies of a C-terminal GGGGS motif assembled by the EPs.

[0201] According to one embodiment of the protein dimer, the first, second, and / or third linkers contain at least two copies of the GGS, GGGS, or GGGGS motif on either side of the EP assembly and / or on either side of the extended half-life portion.

[0202] Exemplary first adapters (and their corresponding sequence IDs) are adapter 1-1 (SEQ ID NO: 8), adapter 1-2 (SEQ ID NO: 9), adapter 1-3 (SEQ ID NO: 10), adapter 1-4 (SEQ ID NO: 11), adapter 1-5 (SEQ ID NO: 12), adapter 1-6 (SEQ ID NO: 43), and adapter 1-7 (SEQ ID NO: 44). The amino acid sequences of these adapters are shown in Table 1 (below). A variant of adapter 1-3 has SEQ ID NO: 41. According to one embodiment, the amino acid sequence of the first adapter has at least 90%, at least 95%, or at least 98% identity with sequences selected from the following: SEQ ID NO: 8, SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 43, and SEQ ID NO: 44. According to one embodiment, the amino acid sequence of the first linker is identical to that selected from the following sequences: SEQ ID NO: 8, SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 41, SEQ ID NO: 43 and SEQ ID NO: 44.

[0203] According to one embodiment, FVIII-CK VWF The second linker of the -VWF heterodimer is an engineered peptide. An exemplary second linker (and its corresponding sequence ID) is specifically FVIII-CK. VWFThe second linkers of the -VWF protein dimer are linker 2-1 (SEQ ID NO: 13), linker 2-2 (SEQ ID NO: 14), linker 2-3 (SEQ ID NO: 15), linker 2-4 (SEQ ID NO: 16), and linker 2-5 (SEQ ID NO: 45). The amino acid sequences of these linkers are shown in Tables 2 and 14 (hereinafter). A variant of linker 2-1 has the sequence of SEQ ID NO: 46. According to one embodiment, the amino acid sequence of the first linker has at least 90%, at least 95%, or at least 98% identity with sequences selected from the following: SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, and SEQ ID NO: 45. According to one embodiment, the amino acid sequence of the first linker is selected from SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 45, SEQ ID NO: 46, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58 and SEQ ID NO: 59.

[0204] According to one embodiment, FVIII-CK VWF -The third connector of the VWF heterodimer, i.e., the linker between the VWF fragment and CK. VWF The connector of the structural domain does not contain the C3 structural domain of the VWF. Preferably, the third connector does not contain the C3 and C2 structural domains of the VWF. According to one embodiment, the third connector does not contain the C3, C2, and C1 structural domains. According to one embodiment, the third connector does not contain any complete structural domain of the VWF. According to one embodiment, the third connector does not contain any other part of the D' and D3 structural domains other than EP. An exemplary third connector (and its corresponding sequence ID), specifically FVIII-CK. VWFThe third linkers of the -VWF protein dimer are linker 3-1 (SEQ ID NO: 17), linker 3-2 (SEQ ID NO: 18), and linker 3-3 (SEQ ID NO: 55). The amino acid sequences of these linkers are shown in Tables 4 and 16 (hereinafter). A variant of linker 3-1 has the sequence of SEQ ID NO: 74. According to one embodiment, the amino acid sequence of the third linker has at least 90%, at least 95%, or at least 98% identity with sequences selected from: SEQ ID NO: 17, SEQ ID NO: 18, and SEQ ID NO: 55. According to one embodiment, the amino acid sequence of the third linker is selected from SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 55, and SEQ ID NO: 74.

[0205] The extended half-life portion can be linked to the remainder of the first or second fusion protein via a fourth linker, i.e., to the first or second pharmaceutically active protein or both CK proteins. VWF One of the structural domains is connected. The fourth connector can be a flexible connector or a rigid connector. According to one embodiment, the connector length is in the range of 5 to 180 amino acids. According to one embodiment, the connector length is in the range of 10 to 160 amino acids. According to one embodiment, the connector length is in the range of 12 to 140 amino acids.

[0206] According to one embodiment, the second pharmaceutically active protein is a VWF fragment. Therefore, according to one embodiment of the first aspect, the pharmaceutically active protein of the first fusion protein is an FVIII protein, and the pharmaceutically active protein of the second fusion protein is a VWF fragment.

[0207] Compared to FVIII alone, the resulting FVIII-CK VWF -VWF fusion protein heterodimers exhibit improved properties, namely, increased expression levels, greater storage stability, and prolonged cycling half-life.

[0208] The human VWF according to the invention has an amino acid sequence of any sequence in the UniprotKB P04275 sequence, particularly SEQ ID NO: 6 (isotype 1). The VWF contains two clusters of O-glycosylated amino acids. The first cluster of O-glycosylated amino acids is located between amino acids 1238 and 1268 of SEQ ID NO: 6. The second cluster includes amino acids 1468 to 1487 of SEQ ID NO: 6.

[0209] According to one embodiment, the VWF fragment comprises an amino acid sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identity with the segment of SEQ ID NO: 6. According to one embodiment, the VWF fragment is at least 95% identical to the segment of SEQ ID NO: 6. According to one embodiment, the light chain is at least 98% identical to the segment of SEQ ID NO: 6.

[0210] Fragments of human VWF do not contain either domain C3 or CK. Furthermore, relative to mature human VWF (TIL3-D3-TIL4-A1-A2-A3-D4-C1-C2-C3-CK), one or more of the domains A1, A2, A3, D4, C1, and C2 may be missing. For example, VWF fragments may have a domain organization selected from the following groups: TIL3-D3-TIL4-A1, TIL3-D3-TIL4-A1-A2, TIL3-D3-TIL4-A1-A2-A3, TIL3-D3-TIL4-A1-A2-A3, TIL3-D3-TIL4-A1-A2-A3-D4, TIL3-D3-TIL4-A1-A2-A3-D4-C1, and TIL3-D3-TIL4-A1-A2-A3-D4-C1-C2.

[0211] In this regard, the segment of SEQ ID NO: 6, specifically the segment beginning with amino acid 764 of SEQ ID NO: 6. Amino acids 764 to 1035 of SEQ ID NO: 6 contain the FVIII binding domain of VWF. The segment can be, for example, the segment defined in WO2015 / 185758 A2. As shown in WO 2015 / 185758 A2, the complex of the defined FVIII with the VWF fragment exhibits reduced binding to the phospholipid membrane compared to FVIII alone, and reduced binding to collagen III and heparin compared to the complex of FVIII with the full-length VWF. The VWF segment preferably begins with amino acid 764 of SEQ ID NO: 6 and preferably ends with amino acids in the range of 1905 to 2153 of SEQ ID NO: 6. According to one embodiment, the VWF fragment ends with amino acids in the range of 2030 to 2153 of SEQ ID NO: 6. According to another embodiment, the VWF fragment ends with amino acids in the range of 2100 to 2153 of SEQ ID NO: 6.

[0212] According to one embodiment, the VWF fragment comprises an amino acid sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identity with SEQ ID NO: 7. The VWF fragment having the amino acid sequence of SEQ ID NO: 7 is based on the segment of amino acids 764 to 1268 of SEQ ID NO: 6, with two amino acid substitutions, namely C1099A and C1142A. Replacing these two cysteine ​​residues with alanine eliminates the ability of the VWF fragment to form a multimer. This modified VWF fragment is used in the fusion protein of the example. According to one embodiment, the VWF fragment is at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identical to SEQ ID NO: 7. Example 6 illustrates a fusion protein having a sequence variant of SEQ ID NO: 7 (i.e., SEQ ID NO: 40) with the following additional amino acid substitution: A1164V.

[0213] According to one embodiment, at least one copy of the EP is directly fused to the C-terminus of the VWF fragment. One, two, three, four, five, or six copies of the EP may be fused to the C-terminus of the VWF fragment. The example shown is FVIII-CK. VWF / VWF-CK VWF The heterodimer has three EPs at the C-terminus of the VWF fragment, one of which is part of the fragment, and the other two EPs are fused to said part. The VWF protein having the amino acid sequence SEQ ID NO: 7 binds to two copies of the EPs, wherein the amino acid sequence added to the C-terminus SEQ ID NO: 2 is a modified derivative of the OCTA12 sequence described in WO 2017 / 198435 A1.

[0214] According to one embodiment, at least one copy of EP forms the N-terminus and / or C-terminus of the first fusion protein. For example, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 copies of EP can form the N-terminus and / or C-terminus of the first fusion protein. According to one embodiment, at least three copies of EP form the N-terminus and / or C-terminus of the first fusion protein.

[0215] According to one embodiment, at least one copy of EP forms the N-terminus and / or C-terminus of the second fusion protein. For example, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 copies of EP form the N-terminus and / or C-terminus of the second fusion protein. According to one embodiment, at least three copies of EP form the N-terminus and / or C-terminus of the first fusion protein. The example shown is FVIII-CK. VWF / VWF-CK VWF The heterodimer has three EPs at the C-terminus of the VWF fragment, which also form the C-terminus of the second fusion protein.

[0216] According to one embodiment, at least one of the N-terminus of the first fusion protein, the C-terminus of the first fusion protein, the N-terminus of the second fusion protein, and the C-terminus of the second fusion protein contains at least one EP.

[0217] According to one embodiment, at least two of the N-terminus of the first fusion protein, the C-terminus of the first fusion protein, the N-terminus of the second fusion protein, and the C-terminus of the second fusion protein contain at least one EP.

[0218] According to one embodiment, in the first fusion protein, the first pharmaceutically active protein is located at CK. VWF The N-terminus of the structural domain, and optionally, there exists a domain located at CK. VWF The third pharmaceutically active protein is located at the C-terminus of the CK domain. According to one embodiment, in the first fusion protein, the first pharmaceutically active protein is located at the CK domain. VWF The C-terminus of the domain, and optionally, there exists a domain located at CK. VWF A third pharmaceutically active protein at the N-terminus of the domain. Preferably, the third pharmaceutically active protein connects to CK via an additional linker. VWF Structural domain connections.

[0219] According to one embodiment, in the second fusion protein, the second pharmaceutically active protein is located at CK. VWF At the N-terminus of the structural domain, and optionally, there exists a domain located at CK. VWF The fourth pharmaceutically active protein is located at the C-terminus of the domain. According to one embodiment, in the second fusion protein, the second pharmaceutically active protein is located at the C-terminus. VWF At the C-terminus of the structural domain, and optionally, there exists a region located at CK. VWF A fourth pharmaceutically active protein located at the N-terminus of the domain. Preferably, the fourth pharmaceutically active protein binds to CK via a peptide linker. VWF Structural domain connections.

[0220] Of course, additional protein components, specifically additional pharmaceutically active proteins, can be added to each fusion protein. The first and second fusion proteins can each contain one, two, three, four, or more pharmaceutically active proteins. In cases where each fusion protein has more than two pharmaceutically active proteins, the additional pharmaceutically active proteins do not directly interact with the CK. VWF It is not a link, but specifically a link with other pharmaceutically active proteins through peptide linkers.

[0221] Furthermore, it is meaningful for the fusion protein forming the protein dimer according to the invention to have more than one copy of the pharmaceutically active protein. According to one embodiment, the first fusion protein contains more than one copy of the first pharmaceutically active protein. For example, the first fusion protein contains at least two, at least three, or at least four copies of the first pharmaceutically active protein. The copies of the first pharmaceutically active protein can be directly linked to each other or indirectly linked through peptide linkers. According to one embodiment, copies of the second pharmaceutically active protein are linked to each other through copies of a second linker.

[0222] According to one embodiment, the second fusion protein contains more than one copy of the second pharmaceutically active protein. For example, the second fusion protein contains at least two, at least three, or at least four copies of the second pharmaceutically active protein. The copies of the second pharmaceutically active protein may be directly linked to each other or indirectly linked through peptide linkers. According to one embodiment, the copies of the second pharmaceutically active protein are linked to each other through copies of a third linker.

[0223] Compared to FVIII alone, the protein dimer FVIII-CK according to the present invention VWF -VWF has an increased half-life. According to one embodiment, the half-life of the fusion protein is extended by at least 20%. According to one embodiment, the half-life of the fusion protein is extended by at least 30%. According to one embodiment, the half-life of the fusion protein is extended by at least 40%. According to one embodiment, the half-life of the fusion protein is extended by at least 50%. According to one embodiment, the half-life of the fusion protein is extended by at least 60%.

[0224] The protein dimer FVIII-CK according to the present invention VWF -VWF showed reduced binding to endogenous VWF after administration to patients. According to one embodiment, the binding to VWF was at most 11% of the binding level of FVIII alone. The binding was determined by surface plasmon resonance (SPR).

[0225] The half-life (t) can be calculated by performing linear regression analysis on the log-linear portion of a single plasma concentration-time curve or by using nonlinear regression of a single-phase exponential decay model. 1 / 2Exemplary software used for computation is GraphPad Prism version 6.07 (La Jolla, CA 92037, USA) and WinNonlin version 6.4 (Pharsight Corporation, Mountain View, CA, USA).

[0226] The calculation is based on the following equation:

[0227]

[0228]

[0229] K el =Eliminate rate constant

[0230] t ½ =Eliminate half-life

[0231] c = concentration

[0232] t = time

[0233] In FVIII-CK VWF In the case of -VWF heterodimer, according to one embodiment, the FVIII protein is located in the CK of the first fusion protein. VWF The N-terminus of the structural domain. According to a preferred embodiment, FVIII connects to CK via a second connector. VWF The N-terminus of the domain is connected. According to one embodiment, the VWF fragment is located in the CK region of the second fusion protein. VWF The N-terminus of the structural domain. According to a preferred embodiment, the VWF segment connects to CK via a third connector. VWF N-terminal connections of structural domains.

[0234] The first fusion protein of the protein dimer is selected, for example, from the proteins shown in the examples: C9 (SEQ ID NO: 22), C10 (SEQ ID NO: 23), C11 (SEQ ID NO: 24), C12 (SEQ ID NO: 25), C13 (SEQ ID NO: 26), C27 (SEQ ID NO: 27), C28 (SEQ ID NO: 28), C29 (SEQ ID NO: 29), C30 (SEQ ID NO: 30), C31 (SEQ ID NO: 31), C32 (SEQ ID NO: 47), C35 (SEQ ID NO: 48), C42 (SEQ ID NO: 51), C43 (SEQ ID NO: 52), C44 (SEQ ID NO: 53), and C45 (SEQ ID NO: 54). Preferred embodiments of the first fusion protein are C11, C13, and C27. Tables 3 and 15 below show the components (and their sequence IDs) that form these fusion proteins.

[0235] According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identity with a sequence selected from the following: SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31, SEQ ID NO: 47, SEQ ID NO: 48, SEQ ID NO: 51, SEQ ID NO: 52, SEQ ID NO: 53, and SEQ ID NO: 54.

[0236] According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO: 22. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO: 23. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO: 24. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO: 25. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO: 26. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO: 27. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 28. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 29. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 30. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 31. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 47. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 48. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 51.According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO: 52. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO: 53. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO: 54.

[0237] In FVIII-CK VWF In the case of -VWF heterodimers, the second fusion protein is selected, for example, from the proteins shown in the examples: C14 (SEQ ID NO: 32), C15 (SEQ ID NO: 33), C25 (SEQ ID NO: 34), C26 (SEQ ID NO: 35), C40 (SEQ ID NO: 49), and C41 (SEQ ID NO: 50). Tables 5 and 17 below show the components (and sequence IDs) that form these fusion proteins. Preferred embodiments of the second fusion protein are C15, C25, and C26.

[0238] According to one embodiment, the second fusion protein comprises an amino acid sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identity with a sequence selected from the following: SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, SEQ ID NO: 49, and SEQ ID NO: 50.

[0239] According to one embodiment of the protein dimer, the second fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 32. According to one embodiment of the protein dimer, the second fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 33. According to one embodiment of the protein dimer, the second fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 34. According to one embodiment of the protein dimer, the second fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 35. According to one embodiment of the protein dimer, the second fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 49. According to one embodiment of the protein dimer, the second fusion protein comprises an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98% identity with SEQ ID NO: 50.

[0240] According to one embodiment, the protein dimer is formed from a combination of a first fusion protein and a second fusion protein selected from the following: C11 plus C15, C11 plus C25, C11 plus C26, C13 plus C26, C27 plus C15, C27 plus C25, C27 plus C26, C28 plus C26, C30 plus C15, C30 plus C25, C30 plus C26, C31 plus C15, C31 plus C25, and C31 plus C26.

[0241] In VHH1-CK VWFIn the case of -VHH2 heterodimers, the first fusion protein of the protein dimer is selected, for example, from the proteins shown in the examples: FP(FIX)1 (SEQ ID NO: 60), FP(FIX)2 (SEQ ID NO: 62), FP(FIX)3 (SEQ ID NO: 64), FP(FIX)4 (SEQ ID NO: 68), FP(FIX)5 (SEQ ID NO: 82), FP(FIX)6 (SEQ ID NO: 83), FP(FIX)7 (SEQ ID NO: 84), FP(FIX)8 (SEQ ID NO: 86), FP(FIX)9 (SEQ ID NO: 88), FP(FIX)10 (SEQ ID NO: 90), FP(FIX)11 (SEQ ID NO: 93), FP(FIX)12 (SEQ ID NO: 95), and FP(FIX)20 (SEQ ID NO: 98). It should be noted that FP(FIX) binds to both FIX and FX. Table 19 below shows the components (and their sequence IDs) that form these fusion proteins. According to one embodiment of the protein dimer, the first fusion protein comprises an amino acid sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identity with a sequence selected from the following: SEQ ID NO: 60, SEQ ID NO: 62, SEQ ID NO: 64, SEQ ID NO: 68, SEQ ID NO: 82, SEQ ID NO: 83, SEQ ID NO: 84, SEQ ID NO: 86, SEQ ID NO: 88, SEQ ID NO: 90, SEQ ID NO: 93, SEQ ID NO: 95, and SEQ ID NO: 98.

[0242] In VHH1-CK VWFIn the case of -VHH2 heterodimers, the second fusion protein is selected, for example, from the proteins shown in the examples: FP(FX)1 (SEQ ID NO: 61), FP(FX)2 (SEQ ID NO: 63), FP(FX)3 (SEQ ID NO: 65), FP(FX)4 (SEQ ID NO: 69), FP(FX)5 (SEQ ID NO: 85), FP(FX)6 (SEQ ID NO: 87), FP(FX)7 (SEQ ID NO: 89), FP(FX)8 (SEQ ID NO: 91), FP(FX)9 (SEQ ID NO: 92), FP(FX)10 (SEQ ID NO: 94), FP(FX)11 (SEQ ID NO: 96), FP(FX)12 (SEQ ID NO: 97), and FP(FX)14 (SEQ ID NO: 99). Table 19 below shows the components (and their sequences) that form these fusion proteins. According to one embodiment of the protein dimer, the second fusion protein comprises an amino acid sequence having at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% identity with a sequence selected from the following: SEQ ID NO: 61, SEQ ID NO: 63, SEQ ID NO: 65, SEQ ID NO: 69, SEQ ID NO: 85, SEQ ID NO: 87, SEQ ID NO: 89, SEQ ID NO: 91, SEQ ID NO: 92, SEQ ID NO: 94, SEQ ID NO: 96, SEQ ID NO: 97, and SEQ ID NO: 99.

[0243] According to one embodiment, the protein dimer is formed from a combination of a first fusion protein and a second fusion protein selected from the following: H1 (FP(FIX)1 plus FP(FX)1), H2 (FP(FIX)2 plus FP(FX)2), H3 (FP(FIX)3 plus FP(FX)3), H5 (FP(FIX)4 plus FP(FX)4), H6 (FP(FIX)4 plus FP(FX)1), H17 (FP(FIX)5 plus FP(FX)4), H18 (FP(FIX)6 plus FP(FX)4), H19 (FP(FIX)7 plus FP(FX)4), H20 (FP(FIX)4 ...4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP(FX)2), H3 (FP(FIX)4 plus FP( H22 (FP(FIX)8 plus FP(FX)6), H28 (FP(FIX)5 plus FP(FX)3), H30 (FP(FIX)9 plus FP(FX)4), H31 (FP(FIX)4 plus FP(FX)7), H32 (FP(FIX)10 plus FP(FX)8), H33 (FP(FIX)4 plus FP(FX)4), H34 (FP(FIX)11 plus FP(FX)6), H35 (FP(FIX)11 plus FP(FX)10), and H36 (FP(FIX)12 plus FP(FX)11).

[0244] Secondly, the present invention relates to CK structural domain pairs, specifically CK VWF The purpose of the domain pair is to combine two pharmaceutically active proteins through dimer formation, one of which is the CK domain, specifically the CK... VWF The CK domain fuses with the pharmaceutically active protein to form the first fusion protein, and another CK domain, specifically the CK... VWF The domain fuses with a pharmaceutically active protein to form a second fusion protein.

[0245] In the application according to the second aspect, the CK domain, the pharmaceutically active protein, and any elements of the first and second fusion proteins are defined as described above with respect to the first aspect. Specifically, if the second fusion protein contains a fragment of VWF, then the fragment does not contain CK. VWF The C3 structural domain of VWF is included, and the third connector is an engineered connector that does not contain the C3 structural domain of VWF.

[0246] Thirdly, the present invention provides a fusion protein comprising VWF CK. VWF Domains, linkers, and pharmaceutically active proteins, wherein the CK VWF The structural domain can interact with the second CK. VWF Domains are covalently bonded.

[0247] Thirdly, the definition of a fusion protein is similar to that of the first or second fusion protein described in the first aspect. Therefore, the pharmaceutically active protein of the second aspect, as well as any elements of the first and second fusion proteins, are defined as described above with respect to the first aspect. Furthermore, the adapter can be defined according to the definitions of the second and third adapters above.

[0248] Exemplary fusion proteins are fusion proteins (and corresponding sequence IDs) as follows: C9 (SEQ ID NO: 22), C10 (SEQ ID NO: 23), C11 (SEQ ID NO: 24), C12 (SEQ ID NO: 25), C13 (SEQ ID NO: 26), C27 (SEQ ID NO: 27), C28 (SEQ ID NO: 28), C29 (SEQ ID NO: 29), C30 (SEQ ID NO: 30), C31 (SEQ ID NO: 31), C14 (SEQ ID NO: 32), C15 (SEQ ID NO: 33), C25 (SEQ ID NO: 34), and C26 (SEQ ID NO: 35), C32 (SEQ ID NO: 47), C35 (SEQ ID NO: 48), C40 (SEQ ID NO: 49), C41 (SEQ ID NO: 50), C42 (SEQ ID NO: 51), C43 (SEQ ID NO: 52), C44 (SEQ ID NO: 53), C45 (SEQ ID NO: 54), FP(FIX)1 (SEQ ID NO: 60), FP(FIX)2 (SEQ ID NO: 62), FP(FIX)3 (SEQ ID NO: 64), FP(FIX)4 (SEQ ID NO: 68), FP(FX)1 (SEQ ID NO: 61), FP(FX)2 (SEQ IDNO: 63), FP(FX)3 (SEQ ID NO: 65), FP(FX)4 (SEQ ID NO: 69), FP(FIX)5 (SEQ ID NO:82), FP(FIX)6 (SEQ ID NO: 83), FP(FIX)7 (SEQ ID NO: 84), FP(FX)5 (SEQ ID NO: 85), FP(FIX)8 (SEQ ID NO: 86), FP(FX)6 (SEQ ID NO: 87), FP(FIX)9 (SEQ ID NO: 88), FP(FX)7 (SEQ ID NO: 89), FP(FIX)10 (SEQ ID NO: 90), FP(FX)8 (SEQ ID NO: 91), FP(FX)9 (SEQID NO: 92), FP(FIX)11 (SEQ ID NO: 93), FP(FX)10 (SEQ ID NO: 94), FP(FIX)12 (SEQ IDNO: 95), FP(FX)11 (SEQ ID NO: 96), FP(FX)12 (SEQ IDNO: 97), FP(FIX)20 (SEQ ID NO: 98), and FP(FX)14 (SEQ ID NO: 99). Tables 3, 5, 15, 17, and 19 below show the components (and their sequence IDs) that form these fusion proteins.

[0249] According to one embodiment, the amino acid sequence of the fusion protein has at least 90%, at least 95%, or at least 98% identity with sequences selected from the following: SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31, SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, SEQ ID NO: 47, SEQ ID NO: 48, SEQ ID NO: 49, SEQ ID NO: 50, SEQ ID NO: 51, SEQ ID NO: 52, SEQ ID NO: 53, and SEQ ID NO: 54.

[0250] According to one embodiment, the amino acid sequence of the fusion protein is identical to that selected from the following sequences: SEQ ID NO:22, SEQ ID NO:23, SEQ ID NO:24, SEQ ID NO:25, SEQ ID NO:26, SEQ ID NO:27, SEQ ID NO:28, SEQ ID NO:29, SEQ ID NO:30, SEQ ID NO:31, SEQ ID NO:32, SEQ ID NO:33, SEQ ID NO:34, SEQ ID NO:35, SEQ ID NO:47, SEQ ID NO:48, SEQ ID NO:49 and SEQ ID NO:50, SEQ ID NO:51, SEQ ID NO:52, SEQ ID NO:53 and SEQ ID NO:54.

[0251] According to one embodiment, the amino acid sequence of the fusion protein has at least 90%, at least 95%, or at least 98% identity with a sequence selected from the following: SEQ ID NO: 60, SEQ ID NO: 62, SEQ ID NO: 64, SEQ ID NO: 68, SEQ ID NO: 61, SEQ ID NO: 63, SEQ ID NO: 65, SEQ ID NO: 69, SEQ ID NO: 82, SEQ ID NO: 83, SEQ ID NO: 84, SEQ ID NO: 85, SEQ ID NO: 86, SEQ ID NO: 87, SEQ ID NO: 88, SEQ ID NO: 89, SEQ ID NO: 90, SEQ ID NO: 91, SEQ ID NO: 92, SEQ ID NO: 93, SEQ ID NO: 94, SEQ ID NO: 95, SEQ ID NO: 96, SEQ ID NO: 97, SEQ ID NO: 98, and SEQ ID NO: 99.

[0252] According to one embodiment, the amino acid sequence of the fusion protein is identical to that selected from the following sequences: SEQ ID NO:60, SEQ ID NO:62, SEQ ID NO:64, SEQ ID NO:68, SEQ ID NO:61, SEQ ID NO:63, SEQ ID NO:65, SEQ ID NO:69, SEQ ID NO:82, SEQ ID NO:83, SEQ ID NO:84, SEQ ID NO:85, SEQ ID NO:86, SEQ ID NO:87, SEQ ID NO:88, SEQ ID NO:89, SEQ ID NO:90, SEQ ID NO:91, SEQ ID NO:92, SEQ ID NO:93, SEQ ID NO:94, SEQ ID NO:95, SEQ ID NO:96, SEQ ID NO:97, SEQ ID NO:98 and SEQ ID NO:99.

[0253] Polynucleotides

[0254] According to a fourth aspect, the present invention provides an isolated polynucleotide comprising a nucleic acid sequence encoding a fusion protein according to a third aspect of the present invention.

[0255] The isolated polynucleotide can be a DNA molecule or an RNA molecule. The isolated polynucleotide is preferably a DNA molecule, specifically a cDNA molecule. Techniques for isolating or cloning polynucleotides encoding peptides are known in the art and include isolation from genomic DNA, preparation from cDNA, or combinations thereof. Polynucleotides can be cloned from such genomic DNA, for example, by using well-known polymerase chain reaction (PCR) or by screening expression libraries with antibodies to detect cloned DNA fragments with common structural features (see, for example, Innis et al., 1990). Other nucleic acid amplification procedures can be used, such as ligase chain reaction (LCR), ligation-activated transcription (LAT), and polynucleotide-based amplification (NASBA).

[0256] The isolated polynucleotide sequence may contain a first part encoding the FVIII heavy chain, a second part encoding the first linker, a third part encoding the FVIII light chain, a fourth part encoding the second linker, and a part encoding CK. VWF Part 5.

[0257] According to one embodiment, the first part encodes the FVIII heavy chain, which has at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with SEQ ID NO: 3.

[0258] According to one embodiment, the second part encodes a first connector, the first connector having at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with a sequence selected from the group consisting of: SEQ ID NO: 8, SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 41, SEQ ID NO: 43, and SEQ ID NO: 44.

[0259] According to one embodiment, the third part encodes the FVIII light chain, which has at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with SEQ ID NO: 5.

[0260] According to one embodiment, the fourth part encodes a second connector, the second connector having at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with a sequence selected from the group consisting of: SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 45, SEQ ID NO: 46, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, and SEQ ID NO: 59.

[0261] According to one embodiment, the fifth part is encoded as CK. VWF The CK VWF It has at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with SEQ ID NO: 1.

[0262] According to one embodiment, the sixth part encodes three consecutive EPs, which have at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with SEQ ID NO:2.

[0263] According to one embodiment of the polynucleotide, the sequence of the isolated polynucleotide may include a first portion encoding the VWF fragment, a second portion encoding the third linker, and a portion encoding the CK. VWF The third part.

[0264] According to an alternative embodiment of the polynucleotide, the isolated polynucleotide sequence encodes VHH, the second part of the third linker, and CK. VWF The third part.

[0265] According to another alternative embodiment of the polynucleotide, the isolated polynucleotide sequence encodes VHH, the second part of the third linker, and CK. NDP The third part.

[0266] According to one embodiment, the first part encodes a VWF fragment having at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with SEQ ID NO: 6.

[0267] According to one embodiment, the first part encodes VHH, which has at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with SEQ ID NO: 72.

[0268] According to one embodiment, the first part encodes VHH, which has at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with SEQ ID NO: 72.

[0269] According to one embodiment, the second part encodes a third connector having at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with a sequence selected from the group consisting of: SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 55, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, SEQ ID NO: 59, and SEQ ID NO: 74.

[0270] According to one embodiment, the third part is encoded as CK. VWF The CK VWF It has at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with SEQ ID NO: 1.

[0271] According to one embodiment, the third part is encoded as CK. NDP The CK VWF It has at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with SEQ ID NO: 100.

[0272] According to one embodiment, the polynucleotide encodes a fusion protein selected from the following: C9 (SEQ ID NO: 22), C10 (SEQ ID NO: 23), C11 (SEQ ID NO: 24), C12 (SEQ ID NO: 25), C13 (SEQ ID NO: 26), C27 (SEQ ID NO: 27), C28 (SEQ ID NO: 28), C29 (SEQ ID NO: 29), C30 (SEQ ID NO: 30), C31 (SEQ ID NO: 31), C14 (SEQ ID NO: 32), C15 (SEQ ID NO: 33), C25 (SEQ ID NO: 34), C26 (SEQ ID NO: 35), C32 (SEQ ID NO: 47), C35 (SEQ ID NO: 48), C40 (SEQ ID NO: 49), C41 (SEQ ID NO: 50), C42 (SEQ ID NO: 51), C43 ...50), C43 (SEQ ID NO: 51), C42 (SEQ ID NO: 51), NO: 52), C44 (SEQ ID NO: 53) and C45 (SEQ ID NO: 54).

[0273] According to one embodiment, a polynucleotide encodes a fusion protein having at least 90%, at least 95%, or at least 98% sequence identity with a sequence selected from the following: SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31, SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, SEQ ID NO: 47, SEQ ID NO: 48, SEQ ID NO: 49, SEQ ID NO: 50, SEQ ID NO: 51, SEQ ID NO: 52, SEQ ID NO: 53, and SEQ ID NO: 54.

[0274] According to one embodiment, the polynucleotide encodes a fusion protein selected from the following: FP(FIX)1 (SEQ ID NO: 60), FP(FIX)2 (SEQ ID NO: 62), FP(FIX)3 (SEQ ID NO: 64), FP(FIX)4 (SEQ ID NO: 68), FP(FIX)1 (SEQ ID NO: 61), FP(FIX)2 (SEQ ID NO: 63), FP(FIX)3 (SEQ ID NO: 65), and FP(FIX)4 (SEQ ID NO: 69), FP(FIX)5 (SEQ ID NO: 82), FP(FIX)6 (SEQ ID NO: 83), FP(FIX)7 (SEQ ID NO: 84), FP(FX)5 (SEQ ID NO: 85), FP(FIX)8 (SEQ ID NO: 86), FP(FX)6 (SEQ ID NO: 87), FP(FIX)9 (SEQ ID NO: 88), FP(FX)7 ...89), FP(FIX)5 (SEQ ID NO: 82), FP(FIX)6 (SEQ ID NO: 83), FP(FIX)6 (SEQ ID NO: 87), FP(FIX)9 (SEQ ID NO: 88), FP(FX)7 (SEQ ID NO: 89), FP(FIX)6 (SEQ ID NO: 89), FP(FIX)6 (SEQ ID NO: 89), 89), FP(FIX)10 (SEQ ID NO: 90), FP(FX)8 (SEQ ID NO: 91), FP(FX)9 (SEQ ID NO: 92), FP(FIX)11 (SEQ ID NO: 93), FP(FX)10 (SEQ ID NO: 94), FP(FIX)12 (SEQ ID NO: 95), FP(FX)11 (SEQ ID NO: 96), FP(FX)12 (SEQ ID NO: 97), FP(FIX)20 (SEQ ID NO: 98) and FP(FX)14 (SEQ ID NO: 99).

[0275] According to one embodiment, a polynucleotide encodes a fusion protein having at least 90%, at least 95%, or at least 98% sequence identity with a sequence selected from the following: SEQ ID NO: 60, SEQ ID NO: 62, SEQ ID NO: 64, SEQ ID NO: 68, SEQ ID NO: 61, SEQ ID NO: 63, SEQ ID NO: 65, SEQ ID NO: 69, SEQ ID NO: 82, SEQ ID NO: 83, SEQ ID NO: 84, SEQ ID NO: 85, SEQ ID NO: 86, SEQ ID NO: 87, SEQ ID NO: 88, SEQ ID NO: 89, SEQ ID NO: 90, SEQ ID NO: 91, SEQ ID NO: 92, SEQ ID NO: 93, SEQ ID NO: 94, SEQ ID NO: 95, SEQ ID NO: 96, SEQ ID NO: 97, SEQ ID NO: 98, and SEQ ID NO: 99.

[0276] expression carrier

[0277] Fifthly, the present invention also relates to expression vectors comprising polynucleotides according to the fourth aspect of the present invention.

[0278] The expression vector further preferably includes control elements, such as promoters, transcription and translation termination signals. According to the second aspect, the polynucleotide and control elements can be coupled together to produce a recombinant expression vector, which may include one or more restriction sites, thereby allowing the insertion or substitution of a polynucleotide encoding a polypeptide at such sites. The polynucleotide can be inserted into a suitable expression vector for expression. In creating the expression vector, the coding sequence is located within the expression vector such that the coding sequence is operatively linked to a suitable control sequence for expression.

[0279] The recombinant expression vector can be any vector (e.g., plasmid or virus) that allows for convenient recombinant DNA procedures and enables the expression of polynucleotides according to the fourth aspect of this invention. The choice of expression vector typically depends on its compatibility with the host cell to which it will be introduced. The expression vector can be a linear or closed circular plasmid.

[0280] Expression vectors are preferably suitable for expression in mammalian cells. Expression vectors can be autonomously replicating vectors, i.e., vectors existing as extrachromosomal entities whose replication is independent of chromosome replication, such as plasmids, extrachromosomal elements, small chromosomes, or artificial chromosomes. To achieve autonomous replication, the vector may further include an origin of replication, thereby enabling the vector to replicate autonomously in the appropriate host cell. The origin of replication can be any plasmid replicon that functions in the cell and mediates autonomous replication. The terms "origin of replication" or "plasmid replicon" refer to the polynucleotide that enables the plasmid or vector to replicate in vivo.

[0281] The vector is preferably one that integrates into the genome upon introduction into the host cell and replicates along with the chromosome into which it is integrated. For integration into the host cell genome, the expression vector can rely on any other elements of the expression vector to achieve genome integration via homologous or non-homologous recombination. Alternatively, the vector may contain additional polynucleotides to guide integration into the precise location of the host cell genome within the chromosome via homologous recombination.

[0282] The vector of the present invention preferably contains one or more (e.g., several) selectable markers, thereby allowing convenient selection of transformed, transfected, transduced, etc., cells. A selectable marker is a gene whose product provides resistance to microbial agents or viruses, resistance to heavy metals, or protrophic nutrition for auxotrophs.

[0283] The procedures for connecting the above-described elements to construct the recombinant expression vector of the present invention are well known to those skilled in the art (see Green and Sambrook 2012; Chapter 3).

[0284] According to one embodiment, the backbone of the vector according to the fifth aspect is selected from pCDNA3, pCDNA3.1, pCDNA3.2, pCDNA3.3, pCDNA3.4, pCDNA4, pCDNA5, pCDNA6, pCEP4, pCEP-puro, pCET1019, pCMV, pEF1, pEF4, pEF5, pEF6, pExchange, pEXPR, pIRES, and pSCAS.

[0285] host cells

[0286] According to a sixth aspect, the present invention provides a host cell comprising a polynucleotide according to a fourth aspect of the invention or an expression vector according to a fifth aspect of the invention. The expression vector according to the fifth aspect is introduced into the host cell such that the expression vector is maintained as a chromosomal integrator or as a self-replicating extrachromosomal vector as described above. The choice of host cell depends largely on the gene encoding the polypeptide and its origin.

[0287] According to one embodiment, a fusion protein is generated by expression in a mammalian host cell line. The fusion protein is preferably generated in a human host cell line. Generally, any human host cell line is suitable for the expression of the fusion protein. The host cell is preferably of human origin to ensure that the fusion protein is properly processed during folding and receives appropriate post-translational modifications (e.g., glycosylation, hydroxylation, phosphorylation, and sulfation). Specifically, a human kidney cell line is used to obtain a favorable glycosylation profile of the fusion protein. A preferred human kidney cell line is the HEK cell line, specifically the HEK 293 cell line.

[0288] Examples of HEK cell lines used for generating glycosylated peptides include HEK 293F, Expi293F (Thermo Scientific A14527), Flp-In™-293 (Invitrogen R75007), 293 (ATCC® CRL-1573), 293 EBNA, 293H (Thermo Scientific 11631017), 293S, 293T (ATCC® CRL-3216™), 293T / 17 (ATCC® CRL11268™), 293T / 17 SF (ATCC® ACS4500™), HEK 293 STF (ATCC® CRL 3249™), and HEK-293.2sus (ATCC® CRL-1573™). The preferred cell line for peptide generation is the HEK 293F cell line.

[0289] Other suitable human cell lines as expression host cells include, but are not limited to, cell lines derived from myeloid leukemia cells. Specific examples of host cells are K562, NM-F9, NM-D4, NM-H9D8, NM-H9D8-E6, NM H9D8-E6Q12, GT-2X, GT-5s, and cells derived from any of these host cells. K562 is a human myeloid leukemia cell line residing at the American Type Culture Collection (ATCC CCL-243). The other cell lines are derived from K562 cells and were selected due to specific glycosylation characteristics.

[0290] Other mammalian host cell lines suitable for producing the fusion protein according to the invention include cell lines of hamster, mouse, and monkey origin. Suitable host cells include Chinese hamster ovary cells (CHO cells, such as DG44, DXB11, and K1 [ATCC CCL-61, including its glutamine auxotroph derivatives CHOZn, SAFCCHOGS]) and neonatal hamster kidney (BHK) cells.

[0291] Pharmaceutical Compositions and Medical Uses

[0292] Therefore, the fusion protein according to the first aspect is particularly effective as an active ingredient for medical treatment. Preferably, it is particularly effective for treating or preventing hemorrhagic conditions. The fusion protein according to the first aspect described herein can be administered alone or in the form of a pharmaceutical composition.

[0293] Therefore, according to the seventh aspect, the present invention provides a fusion protein according to the first aspect, said fusion protein for treating hemorrhagic conditions.

[0294] According to one embodiment, the fusion protein can be formulated with at least one pharmaceutically acceptable carrier. The fusion protein-based pharmaceutical composition can be prepared and administered to a subject by any method known in the pharmaceutical field. See, for example, Goodman and Gilman, *The Pharmacological Basis of Therapeutics*, edited by Hardman et al., McGraw-Hill Professional (10th ed., 2001); *Remington: The Science and Practice of Pharmacy*, edited by Gennaro, Lippincott Williams & Wilkins (20th ed., 2003); and *Pharmaceutical Dosage Forms and Drug Delivery Systems*, edited by Ansel et al., Lippincott Williams & Wilkins (7th ed., 1999). Additionally, the pharmaceutical compositions according to the embodiments can also be formulated to include other medically useful pharmaceutical or biological agents. Pharmaceutical compositions typically comprise a combination of a therapeutically effective amount of a fusion protein and a pharmaceutically acceptable carrier. A pharmaceutically acceptable carrier is any carrier known or established in the art. Exemplary pharmaceutically acceptable carriers include pyrogen-free sterile water and pyrogen-free sterile saline solutions. Other forms of pharmaceutically acceptable carriers that can be used in embodiments of the invention include binders, disintegrants, surfactants, absorption enhancers, humectants, adsorbents, lubricants, fillers, spreaders, moisturizers, preservatives, stabilizers, emulsifiers, solubilizers, salts for controlling osmotic pressure, diluents such as buffers, and excipients typically used depending on the intended use of the formulation. These pharmaceutically acceptable carriers are optionally selected and used based on the unit dose of the final formulation.

[0295] Therefore, the present invention also relates to a method for treating or preventing a patient’s hemorrhagic condition, the method comprising administering to the patient a pharmaceutical composition according to the seventh aspect.

[0296] As used in this article, "hemorrhagic disorders" refer to diseases or conditions that impair normal hemostasis. Examples of hemorrhagic disorders include, for instance, hemophilia A, hemophilia B, factor VIII deficiency, factor XI deficiency, von Willebrand disease, Glanzmann's thromboasthenia, Bernard-Soulier syndrome, idiopathic thrombocytopenic purpura, and intracerebral hemorrhage.

[0297] As used in this article, "hemophilia" refers to a group of bleeding disorders associated with an increased clot formation time compared to that of healthy individuals without hemophilia. Hemophilia includes hemophilia A, a condition that causes the production of defective factor VIII; hemophilia B, a condition that causes the production of defective factor IX; and acquired hemophilia A, a rare bleeding disorder caused by autoantibodies against FVIII.

[0298] The bleeding disorder is preferably hemophilia A or hemophilia B. Treatment may be, for example, treatment of hemophilia in previously untreated patients (PUPS), or immune tolerance induction (ITI) therapy and / or other treatments related to hemophilia.

[0299] For internal use, the pharmaceutical composition may be administered to the patient via any conventional route of administration, such as oral, parenteral, or inhalation. Parenteral administration includes intravenous injection, subcutaneous injection, intraperitoneal injection, intramuscular injection, liquid formulations, suspensions, emulsions, and drops. For parenteral administration, the pharmaceutical composition should be an injectable formulation, such as a liquid formulation or suspension.

[0300] In other embodiments, the pharmaceutical composition is administered orally to a patient. In these embodiments, the form of the drug includes solid formulations (such as tablets, coated tablets, powders, granules, capsules, and pills), liquid formulations (such as liquids, e.g., eye drops, nasal drops), suspensions, emulsions, and syrups, inhalers (such as aerosols, nebulizers, and nebulizers), and liposome encapsulations. In still some other embodiments, glycosylated peptides, protein complexes, or pharmaceutical compositions are administered by inhalation to a patient's airway, thereby targeting the patient's trachea and / or lungs.

[0301] According to one embodiment of the seventh aspect, the use includes intravenous injection or non-intravenous injection. Non-intravenous injection is preferably subcutaneous injection.

[0302] Uses of EP

[0303] According to an eighth aspect, the present invention relates to the use of one or more EPs for reducing the aggregation tendency of a target protein, wherein the one or more EPs are fused to or inserted into the target protein. As shown in Example 14, in the presence of this number of EPs, the tendency of the fusion protein to aggregate (i.e., form high molecular weight components (HMWCs)) is reduced by the EPs. Specifically, the number of HMWCs decreases proportionally to the increase in the number of EPs fused to the target protein. A fusion protein having one copy of an EP aggregates less than a fusion protein without an EP.

[0304] According to one embodiment, the target protein is a pharmaceutically active protein as defined above. According to one embodiment, the target protein is a CK fusion protein as defined above. According to one embodiment, the target protein is defined according to a first fusion protein and a second fusion protein according to the first aspect. According to one embodiment, the target protein is a protein dimer according to the first aspect.

[0305] According to one embodiment, the use includes fusing at least two copies of the EP with a target protein. For example, adding 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 copies of the EP to the target protein. A fusion protein with two copies of the EP aggregates less than a fusion protein with one copy. According to one embodiment, the use includes fusing at least two copies of the EP with a target protein. According to one embodiment, the use includes inserting at least two copies of the EP into the target protein. A fusion protein with four copies of the EP aggregates less than a fusion protein with two copies. According to one embodiment, the use includes fusing at least four copies of the EP with a target protein. According to one embodiment, the use includes inserting at least four copies of the EP into the target protein. A fusion protein with six copies of the EP aggregates less than a fusion protein with four copies. According to one embodiment, the use includes fusing at least six copies of EP with a target protein. According to one embodiment, the use includes inserting at least six copies of EP into a target protein. A fusion protein having nine copies of EP aggregates less than a fusion protein having six copies. According to one embodiment, the use includes fusing at least nine copies of EP with a target protein. According to one embodiment, the use includes inserting at least nine copies of EP into a target protein.

[0306] Example

[0307] Example 1-FVIII-CK VWF and VWF-CK VWF Design, cloning, and expression of fusion proteins

[0308] Experimental Objective

[0309] Design and generate codes for various FVIII-CK VWF and VWF-CK VWF The cDNA of the fusion protein was cloned and then inserted into an expression vector. FVIII-CK VWF and VWF-CK VWF Fusion proteins and their combinations are recombinantly expressed in human cells.

[0310] method

[0311] cDNA design, codon optimization, gene synthesis and cloning

[0312] In order to generate the code FVIII-CK VWF and VWF-CK VWF The expression vector for the protein variant was constructed using Golden Gate cloning technology (Engler et al., 2008). For this purpose, a vector encoding CK was synthesized at Twist Biosciences. VWF The fusion variant's cDNA fragment is cloned into a Golden Gate-compatible donor vector. The donor vector containing the desired cDNA construct, along with a proprietary receptor backbone and discrete donor vectors containing regulatory elements, is used in the Golden Gate assembly reaction. This strategy is used to generate vectors for transient mammalian expression containing a gpCMV promoter at the 5' end of the desired cDNA construct and a marmot hepatitis virus post-transcriptional regulatory element (WPRE) at the 3' end of the cDNA construct.

[0313] The vector construct was transformed into E. coli (NEB5-α) cells, and single clones were selected after incubation overnight at 37°C on LB agar plates containing ampicillin.

[0314] Plasmid DNA preparation should be performed using either the QIAprep DNA Mini Kit (Qiagen) or the NucleoBond® Xtra Maxi Plus EF Kit (Macherey-Nagel), as recommended by the manufacturer. Verify the integrity of the construct by sequencing, paying particular attention to the correct orientation and integrity of the gene encoding the desired fusion protein variant.

[0315] Protein expression

[0316] As recommended by the manufacturer, FVIII-CK was transiently expressed in Expi293F cells (Thermo Fisher Scientific) at a scale of 500-1000 mL. VWF and VWF-CK VWF Fusion construct pairs. 4-5 days post-transfection, cell culture supernatant containing the product was collected by centrifugation at 2000 xg for 20 minutes.

[0317] result

[0318] Tables 1 to 3 show the FVIII-CK encoded by the cloned cDNA constructs. VWF The fusion protein and the adaptor used. Regarding the first adaptor, FVIII-CK... VWF The molecules were produced in “NUWIQ-like” and “ReFacto-like” versions (fusion proteins C9 and C10, respectively). C9 and C10 contain linker sequences 1-1 and 1-2, respectively, that are side-connected to the deleted FVIII B domain, as found in commercially available rFVIII substitute products NUWIQ (Octapharma AG) and ReFacto (Pfizer AG). Based on “NUWIQ-like” and “ReFacto-like” sequences, first adapter variants 1-3, 1-4, and 1-5 were designed, which contain a combination of GS adapters and EPs inserted into the remaining sequence of the FVIII B domain: adapter 1-3: three EPs and flanking G4S adapters inserted into the NUWIQ-like B domain sequence; adapter 1-4: a ReFacto-like B domain sequence with the inserted three EPs and G4S adapters; and adapter 1-5: similar to adapter 1-4, but with the furin cleavage site removed (deleted amino acid RHQR, Table 1).

[0319] Table 1: For FVIII-CK VWF The first connector of the fusion protein.

[0320] The first joint is the joint that connects the FVIII heavy chain to the FVIII light chain. Figure 1 ). EP This indicates the extended peptide with SEQ ID NO:2. Lowercase numbers indicate the number of repetitions of the sequence element in parentheses.

[0321]

[0322]

[0323] The above variants are used in combination with four variants containing a second linker containing the thrombin-cleaving sequence IEPRSFS. The sequence variants differ in the presence and location of the EP, as well as in the placement of the half-life extension portion, i.e., the albumin-binding VHH domain (ABV, Table 2).

[0324] Table 2: For FVIII-CK VWF The second linker of the fusion protein.

[0325] The second linker connects the FVIII light chain to the CK of the first fusion protein. VWF Cuttable joints for structural domain connections ( Figure 1 ). EP This indicates an extended peptide having SEQ ID NO: 2. ABV Indicates albumin-binding VHH with SEQ ID NO: 36. Lowercase numbers indicate the number of repetitions of the sequence element in parentheses.

[0326]

[0327] Table 3: FVIII-CK VWF An overview of fusion proteins and their components.

[0328] ABV Indicates an albumin-binding VHH moiety having SEQ ID NO: 36.

[0329]

[0330]

[0331] Four different VWF-CK designs were created, each with two different third connector sequences. VWF Molecule. Connector 3-1 consists of two consecutive EP connectors, followed by a long flexible G4S connector. Connector 3-2 is used in CK. VWF Albumin-binding VHH (ABV) was added before the domain (Table 4). As a result, ABV became a sequence with a GS linker attached to its side (in fusion protein C26). Alternatively, ABV was associated with VWF-CK. VWF C-terminal fusion of molecules (in C25, Table 5).

[0332] Table 4: Used for VWF fragments and CK VWF The third connector for the structural domain.

[0333] The third linker connects the VWF fragment of the second fusion protein to CK. VWF Connectors for structural domains ( Figure 1 ). EP This indicates an extended peptide having SEQ ID NO: 2. ABV Indicates albumin-bound VHH with SEQ ID NO: 36. Lowercase numbers indicate the number of repetitions of the sequence element in parentheses.

[0334]

[0335] VWF-CK VWF The layout of the fusion protein is shown in Table 5.

[0336] Table 5: VWF-CK VWF An overview of fusion proteins and their components.

[0337] ABV This indicates that albumin binds to VHH (SEQ ID NO: 36).

[0338]

[0339] The results of the protein expression experiment are presented in Example 2.

[0340] Example 2-FVIII-CK VWF + VWF-CK VWF Dimer formation and factor VIII activity (FVIII:C) of fusion proteins

[0341] Experimental Objective

[0342] Testing FVIII-CK VWF and VWF-CK VWF The ability of the fusion protein to dimerize was assessed. The dimers formed were characterized by chromogenic factor VIII activity (FVIII:C) analysis. The effects of the residual FVIII B domain sequence (“NUWIQ-like” or “ReFacto-like”) and the presence of EP and flanking GS linkers in the B domain sequence on FVIII activity in the expression supernatant were evaluated.

[0343] method

[0344] FVIII-CK was transiently expressed in Expi293F cells (Thermo Fisher Scientific) at a scale of 3 mL. VWF and VWF-CK VWF Fusion constructs were developed. Experiments were performed in triplicate. Four days post-transfection, cell culture supernatant was collected by centrifugation at 4800 xg for 30 minutes. FVIII:C activity was assessed on a BCSXP system (Siemens) using the FVIII chromogenic assay kit (Siemens). Expression experiments were performed in two separate rounds. Schematic structures of all heterodimers expressed in both expression rounds are depicted on [the diagram / image ... Figures 2 to 5 middle.

[0345] result

[0346] The first round of expression experiments included five different FVIII-CKs. VWF Fusion proteins (C9 to C13 in Table 3), each corresponding to one of the two different VWF-CK proteins in Table 5. VWF The fusion proteins (i.e., C14 and C15) were expressed together. Analysis of the FVIII activity of the 10 heterodimers formed showed that the FVIII:C levels in the expression supernatant ranged from 1.05 IU / ml to 5.36 IU / ml. Figure 6 A). The FVIII activity levels of the dimer formed with the VWF fragment variant containing the propeptide (C14) ranged from 1.1 IU / ml (for C12+C14) to 2.09 IU / ml (for C11+C14). In contrast, significantly higher activity was measured for the dimer formed with C15 (in the range of 2.87 IU / ml (for C10+C15) to 5.36 IU / ml (for C11+C15)). FVIII-CK co-expressed with C14 or C15... VWF The highest expression levels were achieved in the C11 construct, which contained three EPs and flanking GS connectors in the remaining sequence of the inserted “NUWIQ-like” B domain (2.09 IU / ml for C11+C14 and 5.36 IU / ml for C11+C15).

[0347] FVIII-CK VWF The fusion proteins C9 (“NUWIQ-like”) and C10 (“ReFacto-like”) differ in the amino acid short segments in linker 1 (Table 3).

[0348] To further evaluate whether EP insertion affects expression levels, fusion proteins containing the same furin cleavage site in adapter 1 and differing only in the presence of EP in the same adapter were compared: C9+C14 / C15 vs. C11+C14 / C15 and C10+C14 / C15 vs. C12+C14 / C15. The higher FVIII activity level of C11, compared to the values ​​of the C9 and C10 heterodimers (C9 and C10 represent “NUWIQ-like” and “ReFacto-like” sequences, respectively, without an additional module in the first adapter), confirms that the increased expression level can be mediated by FVIII-CK. VWF Add EP to the part to mediate ( Figure 6A). Specifically, the data confirm that EP has a positive effect on the expression levels of molecules containing "NUWIQ-like" adapter 1, but no positive effect on molecules containing "ReFacto-like" adapter 1.

[0349] Two different FVIII-CK variants containing the remaining sequence of the “ReFacto-like” B domain. VWF Proteins (C12 and C13) and VWF-CK VWF The fusion proteins C14 and C15 were combined to form four different heterodimers: C12+C14 / C15 and C13+C14 / C15. Among them, the heterodimer of the C13 fusion protein, containing a GS adaptor group with a replaced FVIII B domain and EP but lacking a furin cleavage site, achieved the highest FVIII:C activity level (4.36 IU / mL for C13+C15). Figure 6 A).

[0350] In summary, insert FVIII-CK VWF The flexible GS linker and EP in the first linker of the protein with VWF-CK without propeptide VWF The combination of variants achieved the highest FVIII activity level, indicating the highest protein stability and correct folding.

[0351] In the second part of the experiment, six additional FVIII-CKs that differ in the number and location of EPs, as well as in the presence and placement of the extended half-life portion (ABV, Table 3), will be compared. VWF The fusion protein differs from three different VWF-CK antibodies (ABV, Table 5) in the presence and location of the same albumin-binding single-domain antibody. VWF Molecular assembly. Therefore, FVIII-CK... VWF With VWF-CK VWF Eighteen combinations of fusion proteins were co-expressed in the human embryonic kidney 293 cell line, and their FVIII activity was compared with that of the monomeric rFVIII (BDD-FVIII; NUWIQ). All 18 fusion protein combinations exhibited FVIII:C activity. Their activity levels (3.4 IU / ml to 12.8 IU / ml) were significantly higher than the expression level of the rFVIII control (0.22 IU / ml). Figure 6 B).

[0352] Data indicates that FVIII-CK VWFThe heterodimer formed by the construct and EP inserted into both linkers (linker 1 and linker 2 in C27, C30, and C31) showed higher expression levels than the fusion proteins (C11, C28, and C29) with EP only in linker 1. In most cases, FVIII-CK... VWF VWF-CK contains VHH moieties (ABV in C25 and C26) that bind to albumin. VWF Co-expression of the chain mediates a higher expression level than its counterpart lacking this modification (C15, Figure 5 B) The level of expression mediated is higher.

[0353] Example 3-FVIII-CK VWF -VWF(FVIII-CK) VWF + VWF-CK VWF Purification and structural characterization of protein dimers

[0354] Experimental Objective

[0355] To assess the size, integrity, and heterodimerization of the expressed fusion protein, the selected constructs were purified. SDS-PAGE analysis was performed to control for the homogeneity and purity of the expressed constructs and the dimers formed by their combination. Western blot analysis was performed to confirm the identity and integrity of the formed heterodimers.

[0356] method

[0357] FVIII-CK was achieved by applying two consecutive affinity chromatography steps. VWF Purification of VWF heterodimer: First, VWF was subjected to affinity chromatography, followed by FVIII, and intermediate and final rebuffering steps were performed.

[0358] In summary, the collected cell culture supernatant was filtered through a 0.2 µm PES filter and directly loaded onto VWF affinity resin (VOLTselect affinity resin, Thermo Fisher Scientific) at a loading volume of approximately 100 CV and a contact time of at least 3 minutes. The column was equilibrated with 0.05 M Tris, 0.1 M NaCl, and 0.02% polysorbate 80 at pH 7.0, and eluted with 0.05 M Tris, 0.1 M NaCl, and 1 M MgCl2 at pH 7.0. The eluent was re-buffered into a formulation matrix (171.1 mM NaCl, 7.1 mM L-arginine, 26.3 mM sucrose, 3.4 mM trisodium citrate, 1.7 mM CaCl2, 0.1 mM poloxamer 188, pH 7.0) using a Cytiva Sephadex G-25 desalting column for matrix exchange, and then loaded onto FVIII Select affinity resin. The FVIII Select column was equilibrated with 0.3 M NaCl, 0.02 M CaCl2, 0.02 M L-histidine, and 0.02% polysorbate 80 at pH 6.5, and eluted with 1.5 M NaCl, 0.02 M CaCl2, 0.02 M L-histidine, 50% ethylene glycol, and 0.02% polysorbate 80 at pH 6.5.

[0359] Finally, the resulting eluent was re-buffered into the formulation buffer as described above using a Sephadex G-25 desalting column (Stopfan).

[0360] SDS-PAGE

[0361] Samples from different purification steps were analyzed by non-reducing SDS-PAGE. Samples were denatured by incubation with LDS sample buffer. Samples were run on 4%–12% BisTris gels (NuPage) at 175 V for 70 minutes.

[0362] Coomassie staining was performed using ready-to-use Coomassie staining solution (Thermo Scientific, Page Blue protein staining solution) at room temperature (RT) for 3 hours, followed by washing and destaining in MilliQ water until the background was clear.

[0363] Protein blot

[0364] According to the manufacturer's (Thermo Fisher Scientific) protocol, proteins are transferred to PVDF membranes in the iBlot 2 gel dry transfer device using the iBlot 2 transfer stack (Ingenium).

[0365] Following blotting, the membrane was blocked with TBS Superblock (Thermo Fisher Scientific) and proteins were detected using either antibody SAF8C-AP (for human FVIII, Coachrom) or 19818 / ABIN190120 (for human VWF, QED Bioscience). The blot was further washed and incubated with the corresponding HRP-conjugated secondary antibodies (anti-sheep, Ingenium, for FVIII, and anti-mouse, BioRad, for VWF). Signals were detected using the Supersignal West Pico kit (Thermo Fisher Scientific).

[0366] result

[0367] Based on SDS-PAGE analysis, the selected FVIII-CK from Example 2 WF -VWF heterodimers were successfully purified to a purity of >90%. Impurities associated with the product, such as FVIII or VWF homodimers, were effectively removed in two consecutive affinity purification steps.

[0368] The results of the analysis are summarized in Figure 7 and 8 middle. Figure 7 The results show the purified C9+C15 and C11+C15 heterodimers in Western blotting with FVIII detection (A), VWF detection (B), and SDS-PAGE with Coomassie staining (C). Heterodimers were formed, as indicated by a single band in the Coomassie gel. Figure 7 C). The band directly above the 250 kDa marker band corresponds to FVIII-CK. VWF -VWF heterodimers C9+C15 and C11+C15, because in the use of FVIII ( Figure 7 A) and VWF staining ( Figure 7 A similar signal was detected in the Western blot analysis performed in B). The small band below 250 kDa represents furin cleavage products. Interestingly, the furin cleavage site, although present in both FVIII-CK... VWF The process is the same in both C9 and C11 molecules, but the processing efficiency in C11 is lower than that in C9. This is in Figure 7 At approximately 75 kDa on the anti-FVIII blot in A and Figure 7A stronger furin protease cleavage product band of the C9+C15 heterodimer at >150 kDa was visible on the anti-VWF blot in B. The reduced processing of C11 (complexed with C15) can be attributed to the presence of three EPs in the first linker of the fusion protein and explained by the strong negative charge on the EPs, which in turn prevents the interaction of furin protease with its cleavage site.

[0369] Figure 8 The purified heterodimers from the second expression round of Example 2 are shown. These molecules contain an additional half-life extension in the form of an albumin-binding single-domain antibody. After purification, the molecules were analyzed by Western blotting with FVIII detection (A), VWF detection (B), and SDS-PAGE with Coomassie staining (C). The following protein dimers were obtained from... Figure 8 The lane numbers in the diagram are described as follows: 1 - a dimer formed by C27 and C25, 2 - C27 + C26, 3 - C30 + C25, 4 - C31 + C25, 5 - C30 + C26, 6 - C31 + C26, 7 - C31 + C15, 8 - C30 + C15, 9 - C11 + C25, 10 - C11 + C26, 11 - C13 + C26, 12 - C27 + C15, 13 - C28 + C26.

[0370] All proteins exhibited characteristic bands of heterodimers at >250 kDa, with some small bands indicating furin cleavage products. These data confirm that the CK domain enables monomer dimerization regardless of its location within the fusion protein: at the C-terminus of albumin-bound VHH (in constructs C15, C26, C11, C27, C29, and C31) and at the N-terminus (in constructs C25, C28, and C30).

[0371] Example 4-CK VWF Mediated binding of protein dimers to full-length VWF

[0372] Experimental Objective

[0373] Evaluation of FVIII-CK VWF - The ability of VWF heterodimers C9+C15 and C11+C15 to bind to human full-length VWF.

[0374] method

[0375] Binding to full-length VWF (flVWF) was measured using surface plasmon resonance (SPR) on a Biacore T200 instrument (Stopvan). Purified flVWF (Sekisui) was coated onto a CM5 chip via amine coupling using an amine coupling kit (Stopvan) according to the manufacturer's instructions. FlVWF was immobilized at approximately 1000 response units (RU) in three different flow cells. The run buffer consisted of 20 mM HEPES, 150 mM NaCl, 5 mM CaCl2, and 0.05% Tween 20. The surface was regenerated after each analyte injection using regeneration buffer (20 mM HEPES, 600 mM NaCl, 350 mM CaCl2, and 0.05% Tween 20). Purified FVIII-CK was injected in triplicate at a fixed concentration of 8.5 IU / ml FVIII:C into three different flow cells in randomized order. VWF -VWF fusion protein dimer. The binding level, measured 30 seconds after analyte injection, was normalized by dividing RU by the molecular weight of the corresponding protein and expressed as rFVIII as 100% binding percentage.

[0376] result

[0377] Further analysis of purified FVIII-CK was performed using SPR. VWF -VWF heterodimers C9+C15 and C11+C15 bind to flVWF. Both heterodimers showed very low flVWF binding levels, 1.8% and 1.2%, respectively. These values ​​are compared with the negative control -dimer VWF fragment OCTA12 (0.6% binding). Figure 9 )quite.

[0378] Example 5-FVIII-CK VWF Pharmacokinetics of -VWF protein dimers C9+C15 and C11+C15

[0379] Experimental Objective

[0380] To explore the effects of dimerization of the FVIII and VWF fusion proteins and the presence of EP in linker 1, two purified heterodimeric proteins, C9+C15 and C11+C15, were tested in a pharmacokinetic study of BDD-FVIII in hemophilia A (HemA) mice.

[0381] method

[0382] Male B6;129S-F8tm1Kaz / J (F8- / -) mice aged 5 to 8 weeks were obtained from Jackson Laboratory (Bar Harbor, Maine, USA). Based on FVIII:C activity, purified FVIII-CK was administered via tail vein injection at a dose of 200 IU / kg body weight. VWF Animals were treated with either the -VWF protein dimer formulation or rFVIII. A summary of the study is presented in Table 6. Blood samples were collected at the indicated time points. Blood samples were taken from five animals in each group at each time point. Each mouse was used for two sampling points. Blood was collected in tubes containing 3.8% sodium citrate solution. Immediately after collection, blood samples were placed on crushed ice and plasma was separated within one hour of sampling by centrifugation at 3350 xg at 4°C for 15 minutes. Plasma samples were stored at -80°C until analysis was performed using the FVIII:C assay (Coamatic Factor VIII Assay Kit; Chromogenix, Bedford, MA, USA).

[0383] Table 6: Research Overview

[0384]

[0385] result

[0386] The research results are presented in Figure 10 And in Table 7. As expected, rFVIII shows a half-life (T... 1 / 2 The duration is 7.61 hours. Two types of CK... VWF The mediated heterodimers C9+C15 and C11+C15 exhibited significantly different PK spectra. At a value of 9.01 h, the T0 of C9+C15 was significantly different. 1 / 2 Only slightly longer than that of monomer rFVIII. C11+C15 shows the optimal PK spectrum, where T 1 / 2 It lasted 28.21 hours (3.7 times that of rFVIII), with a maximum C max It was 363.6% (1.2 times that of rFVIII), and the highest AUC was 7947.72 hours*% (2.3 times that of rFVIII).

[0387] Table 7: PK analysis of FVIII:C data measured in HemA mouse plasma

[0388]

[0389] These results confirm that the FVIII fusion protein communicates with its CK. VWF Covalent linkage of the domain to the VWF fragment fusion protein and the addition of EP via gene fusion improved the PK parameters of heterodimers in hemophilia A mice. The most significant improvement came from heterodimer formation and FVIII-CK. VWF The combination of specific EP in the linker 1 of the fusion protein portion.

[0390] Confirmation of results for Example 6 - FVIII and VWF sequence variants

[0391] Experimental Objective

[0392] It was confirmed that the CK domain, FVIII heavy chain, FVIII light chain, EP linker, and minor amino acid sequence variations in the VWF fragment did not affect the biological characteristics of the protein heterodimer.

[0393] result

[0394] As shown in Table 8, additional FVIII-CK based on C11 was generated. VWF Fusion protein, including FVIII heavy chain, EP (in linker 1), FVIII light chain, linker 2, and CK. VWF One or all of them exhibit sequence variations. Furthermore, as shown in Table 9, a C15-based VWF-CK was generated. VWF Fusion protein, including VWF fragment and CK VWF One or all of them have sequence variations.

[0395] Table 8: Additional FVIII-CK with C11 sequence variants VWF Fusion protein.

[0396]

[0397] Table 9: Additional VWF-CK with C15 sequence variants VWF Fusion protein.

[0398]

[0399] Using the FVIII-CK mentioned above VWF and VWF-CK VWF Protein sequence variants produce the following heterodimers:

[0400] Table 10: FVIII-CK VWF - Composition of VWF heterodimer sequence variants.

[0401]

[0402]

[0403] In addition, experiments according to Examples 2 to 4 were repeated using combinations of sequence variants C11a to C11f and C15a to C15e.

[0404] Example 7 - Sequence modification, albumin-bound VHH and additional extended peptides against FVIII-CK VWF Effects of VWF dimer on pharmacokinetics

[0405] Experimental Objective

[0406] To investigate a) the deletion of the furin cleavage site in the FVIII sequence, and b) the insertion of additional EP into the FVIII-CK sequence. VWF The fusion pair formed by the protein portion and c) albumin-bound VHH portion as complementary parts of the fusion protein is FVIII-CK. VWF - Effects of VWF heterodimer on pharmacokinetics.

[0407] method

[0408] Male C57BL / 6J mice aged 6 to 8 weeks were obtained from Janvier Labs (France). Based on FVIII:C activity, purified FVIII-CK was administered via tail vein injection at a dose of 200 IU / kg body weight. VWF -VWF heterodimer was used to treat the animals. A summary of the study is presented in Table 11. Blood samples were collected at the indicated time points. Blood samples were taken from five animals in each group at each time point. Each mouse was used for two sampling points. Blood was collected in tubes containing 3.8% sodium citrate solution. Immediately after collection, the blood samples were placed on crushed ice and plasma was separated within 1 hour of sampling by centrifugation at 3350 xg at 4°C for 15 minutes. Plasma samples were stored at -80°C until analysis.

[0409] The FVIII activity (FVIII:C) of aliquots of mouse plasma was analyzed. FVIII:C was measured using a capture assay. In short, FVIII-CK was captured from mouse plasma using a custom-made anti-VWF single-domain antibody (specific to the human VWF-D'D3 domain) from mouse plasma. VWF -VWF protein dimer. FVIII activity was assessed by Biophen FVIII:C assay (Hyphen Biomed) after blocking and washing steps.

[0410] Table 11: Overview of Pharmacokinetic Studies

[0411]

[0412] result

[0413] In pharmacokinetic studies conducted in WT mice, the FVIII-CK mice listed in Table 11 will be used. VWF + VWF-CK VWF Thirteen heterodimers formed by the fusion protein combinations were tested compared to the C11+C15 dimer (see Example 5). The protein modules and structures of the various fusion proteins are listed in Tables 3 and 5 (Example 1) and illustrated in [the table / image / image]. Figures 2 to 5 The research results are presented in... Figure 11 And in Table 12.

[0414] The half-life (15.42 hours) of the heterodimer C11+C15 in WT mice was 1 / 1.8 of the half-life (28.21 hours – Example 5, Table 7) detected in hemophilia A mice. These data are consistent with the results published by Dumont et al. in 2012, who observed similar differences in the half-life of the FVIII-Fc fusion protein in hemophilia A mice compared to the WT mouse strain.

[0415] VWF-CK VWF Fusion proteins C25 and C26 contain an albumin-binding VHH moiety (C25) at the C-terminus of the protein or contain a CK region inserted into the fusion protein sequence. VWF The albumin-binding VHH portion preceding the domain (C26). In the heterodimer with the FVIII fusion protein C11, C25 and C26 exhibit improved terminal half-lives compared to their unmodified counterparts C15 (18.3 h for C11+C25 and 17.35 h for C11+C26, compared to 15.42 h for C11+C15). Figure 11 B). Additionally, C11+C25 and C11+C26 showed higher protein recoveries during cycling (for C11+C25 and C11+C26, C...). max The levels were 3.15 U / ml and 3.47 U / ml, respectively, compared to 2.79 U / ml for C11+C15. These improvements also enabled the detection of higher FVIII:C plasma levels of C11+C25 and C11+C26 at 96 hours post-protein administration. Figure 11 A). This confirms the introduction of albumin-binding VHH into VWF-CK. VWF The sequence located in CK VWFThe position of either the C-terminus (in C25) or the N-terminus (in C26) of the domain makes FVIII-CK VWF -VWF heterodimers exhibit significantly improved pharmacokinetic properties.

[0416] For those with FVIII-CK VWF Further improvements in pharmacokinetic properties were observed in the heterodimer of construct C13, a modified version of the C11 molecule lacking the furin cleavage site within the WT amino acid sequence of FVIII. This was observed in the VWF-CK... VWF In the heterodimer of protein C26, the C13 chain mediates the terminal half-life of 19.43 hours and the highest plasma FVIII:C level (0.15 U / mL for C13+C26) detected in WT mice. Figure 11 ).

[0417] Compared to C11, the FVIII fusion protein C27 falls between FVIII-LC and CK. VWF The linker between the domains (linker 2) contains three additional EPs. In combination with C15, this modification improves the terminal half-life of C27+C15 by 1.15-fold compared to C11+C15. At 96 hours post-protein administration, the insertion of additional EPs resulted in nearly twice the FVIII:C level of the C27+C15 heterodimer in WT mouse plasma. Figure 11 ).

[0418] Add the albumin-binding VHH moiety to C15 (i.e., in CK). VWF Then it fuses with the C-terminus of the VWF fusion protein, or with CK. VWF Inserting VHH before the structural domain generates C25 and C26 respectively (see above). Figures 2 to 5 In the case of combination with the FVIII fusion construct C27, this modification did not further improve the half-life beyond the heterodimer C27+C15, but it did have a positive effect on the recovery of the molecule (for C27+C25 and C27+C26, C...). max The concentrations were 3.45 U / ml and 3.33 U / ml, respectively, compared to 2.58 U / ml for C27+C15.

[0419] FVIII-CK with ABV added to the C-terminus VWF Fusion protein (i.e., placed in CK) VWF The C-terminus of the structural domain (C28 and C30) mediates the improved half-life of the heterodimer. Compared with C27+C15 (excluding ABV) (T... 1 / 2Compared to 17.86 hours, the half-life of C30+C15 is significantly prolonged (T = 17.86 hours). 1 / 2 = 25.81 hours). However, inserting VWF-CK VWF The second ABV, which combines with or is fused to the couple (C26 and C25, respectively), did not promote a further improvement in the half-life of the dimer, but improved the recovery level (C30+C15 versus C30+C25 and C30+C26). max Differences (Table 12).

[0420] FVIII-CK with placed heterodimer VWF A partial CK domain N-terminal ABV fusion protein (C31) mediates a lower molecular weight than ABV molecules with C-terminal fusion (C30). Figure 11 And Table 12) T 1 / 2 Values ​​and FVIII activity levels.

[0421] In summary, these data demonstrate that all modifications, deletion of the FVIII cleavage site of furin, addition of EP, and insertion of albumin-binding single-domain antibody sequences into the fusion protein can induce heterodimer FVIII-CK. VWF - Improved pharmacokinetic properties of the VWF molecule. Importantly, the positions of EP and ABV within the heterodimer play a crucial role. Data indicate that placing ABV within the heterodimer in FVIII-CK... VWF The C-terminus of the ABV moiety has the greatest potential for improving the half-life of the molecule. However, the ABV moiety and the heterodimer VWF-CK... VWF Partial fusion improved the recovery of FVIII activity in mouse plasma.

[0422]

[0423] Example 8-FVIII-CK VWF and VWF-CK VWF Flexible linkers in fusion proteins for FVIII-CK VWF Effect of VWF protein dimer activity

[0424] Experimental Objective

[0425] Evaluation in FVIII-CK VWF The flexible G4S linker is used at positions 1 and 2 of the fusion protein and as a VWF-CK linker. VWF Linker 3 in the fusion protein is responsible for the formation of FVIII-CK. VWF The effect of VWF protein dimer activity.

[0426] method

[0427] According to the manufacturer's (Thermo Fisher Scientific) protocol, as described in Example 1, the FVIII-CK clone... VWF and VWF-CK VWF The fusion construct was transiently expressed in Expi293F cells in a 3 mL format. Four days post-transfection, cell culture supernatant was collected by centrifugation at 4800 x g for 30 min. FVIII:C activity was assessed using an FVIII colorimetric assay kit (Siemens) on a BCS XP system (Siemens). All expression experiments were performed in two independent replicates.

[0428] result

[0429] The replacement of FVIII-CK was studied. VWF The effects of an additional combination (linker 1) of the G4S linker of the FVIII B domain and the VWF extended peptide (EP) in the fusion protein were investigated. The effects of the remaining FVIII B domain amino acids (“NUWIQ-like” or “ReFacto-like” amino acid sequences, see Example 1 for details) located between the FVIII heavy and light chains on the function of the fusion protein were further evaluated. VWF Functional effects of the linker and EP (linker 2) between the FVIII light chain of the fusion protein and the CK module.

[0430] The initial scaffold molecule used for the selective deletion of linker 1 was the fusion protein C27. To ensure these findings are universally applicable to the different B-domain remnants, C27 and FVIII-CK, which differs from C27 in terms of linker 1 sequence, were compared. VWF The fusion protein C32 (including “ReFacto-like” linkers 1–5 but not “NUWIQ-like” linkers 1–3) underwent similar G4S linker deletion. All FVIII-CK VWF All versions are the same as VWF-CK VWF Co-expression of the fusion protein C15.

[0431] Three variants of the C32 fusion protein were created: C43, which lacks the G4S linker of the EP repeat sequence sidelink in linker 1 (connecting the LC and HC domains of FVIII); C42, which lacks the linker of the EP sidelink in linker 2 (connecting the LC and CK domains of FVIII); and C35, which lacks all G4S linkers from both linker 1 and linker 2. The fusion protein is schematically shown in... Figure 12 In A.

[0432] Two variants of the C27 molecule were created: C44, which lacks the G4S linker of the EP repeat sequence in linker 1 (connecting the LC and HC domains of FVIII), and C45, which lacks the linker of the EP repeat sequence in linker 2 (connecting the LC and CK domains of FVIII). The fusion protein is schematically shown in... Figure 13 In A.

[0433] FVIII-CK VWF The fusion proteins were designed and generated as shown in Tables 13 to 15 (see also Tables 1 and 2 for adapter nomenclature).

[0434] Table 13: For FVIII-CK VWF The first connector of the fusion protein.

[0435] The first joint is the joint that connects the FVIII heavy chain to the FVIII light chain. Figure 1 ). EP Indicates the VWF EP with SEQ ID NO:2. Lowercase numbers indicate the number of repetitions of the sequence element in parentheses. See Table 1 for characteristics of connectors 1-1 to 1-5.

[0436]

[0437] Table 14: For FVIII-CK VWF The second linker of the fusion protein.

[0438] The second linker connects the FVIII light chain to the CK of the first fusion protein. VWF Connectors for structural domains ( Figure 1 ). EP Indicates a VWF EP with SEQ ID NO: 2. Lowercase numbers indicate the number of repetitions of the sequence elements in parentheses. See Table 2 for characteristics of connectors 2-1 to 2-4.

[0439]

[0440] Table 15: FVIII-CK VWF An overview of fusion proteins and their components.

[0441]

[0442] To explore whether a flexible G4S linker connecting the VWF fragment and the CK domain is necessary for mediating proper folding and heterodimer activity, novel molecules were designed. C40 is a version of C15 lacking linker 3. C41 is a version of C14 lacking linker 3. VWF-CK VWF Fusion proteins C40 and C41 were designed and generated as shown in Tables 16 and 17. All co-expressions were associated with FVIII-CK.VWF The fusion protein C27 was used together. The resulting heterodimer was designed to be presented in... Figure 14 In A.

[0443] Table 16: Used for VWF fragments and CK VWF The third connector for the structural domain.

[0444] The third linker is the CK of the second fusion protein. VWF The connector between the structural domain and the VWF segment ( Figure 1 ). EP This indicates the extended peptide having SEQ ID NO: 2. Lowercase numbers indicate the number of repetitions of the sequence element in parentheses.

[0445]

[0446] VWF-CK VWF The layout of the fusion protein is shown in Table 17.

[0447] Table 17: VWF-CK VWF An overview of fusion proteins and their components.

[0448]

[0449] In the first round of expression, four different versions of FVIII-CK were generated, each carrying a GS connector with different combinations of EP and the remaining sequence of the inserted “ReFacto-like” B domain. VWF Molecules (i.e., C32, C35, C42, and C43) Figure 12 A). Different FVIII-CK VWF Protein and the same VWF-CK VWF The fusion protein C15 was combined to form four different heterodimers: C32+C15, C35+C15, C42+C15, and C43+C15. The C32+C15 heterodimer achieved an FVIII:C activity level of 8.01 IU / mL in the expression supernatant. For C35, which contains two EP modules but lacks the GS adapter that lateralizes them, or C42, which only carries the GS adapter group that replaces the FVIII B domain and the EP, the FVIII activity of the heterodimer with C15 was significantly reduced to 2.97 IU / mL and 2.9 IU / mL, respectively. In contrast, the GS adapter group lacking the EP lateralized in the B domain but containing the CK... VWF The C43 and C15 combination of the C-terminal GS linker group of the EP domain ahead of the domain achieves an FVIII:C level of 7.22 IU / mL, which is only slightly lower than the FVIII activity of the heterodimer C32+C15, which contains both the entire EP and GS linker group. Figure 12B).

[0450] In the second round of expression, the FVIII-CK of the “NUWIQ-like” residual sequence with the B domain was investigated. VWF Three versions of the molecule combined with C15. These combinations form three different FVIII-CK molecules. VWF -VWF heterodimers C27+C15, C44+C15 and C45+C15 ( Figure 13 A). The heterodimer formed from C27 (containing both the entire EP and the flanking GS adapter group) and C15 achieved an FVIII:C level of 3.46 IU / mL in the expression supernatant. For CK-deficient samples... VWF The heterodimers C15 and C45 of the C-terminal GS linker group, which are EP-side connected to the front of the domain, showed a significant decrease in FVIII activity to 0.66 IU / mL. The GS linker group lacking EP-side connection was present in the B domain, but not in the CK domain. VWF The heterodimer of C15 and C44, with EP-side linkers upstream of the domain, achieved an FVIII:C activity of 3.26 IU / mL. This activity is only slightly lower than that of FVIII-CK, which carries two GS linkers. VWF -VWF heterodimer C27+C15 activity (3.46 IU / mL) Figure 13 B).

[0451] In summary, for FVIII-CK independent of the B-domain residual sequence VWF The G4S linker group with EP side-attached in the B domain of the molecule is for FVIII-CK VWF -VWF-CK VWF Proper folding, stability, and FVIII activity of the heterodimer are unnecessary. In contrast, the CK placed at the C-terminus... VWF The G4S connector of the EP side-connected section in front of the structural domain (i.e., between the FVIII light chain and the CK structural domain) is for FVIII-CK. VWF -VWF-CK VWF Proper folding of the heterodimer and FVIII activity are crucial.

[0452] For heterodimer VWF-CK VWF Partially, four molecules were compared, all of which were associated with the same FVIII-CK. VWF C27 co-expression in the mate body (schematically shown) Figure 14 (In Institute A). The FVIII:C activity measurement results of the formed heterodimer are presented in... Figure 14 In B, the results show that for both VWF-CK... VWFFor molecular variants (C14 and C15), the flexible linker between the VWF and CK domains is unnecessary and has no significant effect on heterodimer folding and activity.

[0453] Example 9 - Effects of the extended peptide in the second linker on thrombin generation kinetics and FVIII-CK VWF Effects of VWF heterodimer on in vivo potency

[0454] Experimental Objective

[0455] The effects of the extended peptide repeat sequence in the second linker (specified in Table 2) were evaluated.

[0456] method

[0457] Two FVIII-CK mice were compared in thrombin generation assay (TGA) and tail-cut hemorrhage assay in hemophilia A (HemA) mice. VWF -VWF dimer, C32+C15 and C37+C15, differ only in FVIII-CK VWF The linker 2 contains three extended peptide repeat sequences (C32 contains linker 2-2 with EP, and C37 contains linker 2-1 without EP).

[0458] The Thrombin Generation Assay (TGA), developed by Thrombinoscope BV (Netherlands), is used for real-time monitoring of thrombin generation. The method is based on the fluorescence detection of a peptide substrate, which undergoes thrombin-mediated conversion into a fluorescent compound that can be quantified using a photometer. This assay is a general physiological function test of the thrombotic hemostatic system. It measures the kinetics of thrombin generation in FVIII-deficient plasma after the addition of FVIII or an FVIII-containing sample. Measurements were performed in 96-well plates at 37°C using a Fluoroskan Ascent FL fluorometer. Thrombin generation was monitored by the activation of the thrombin-mediated fluorescent substrate.

[0459] Four key parameters indicative of the activity of the FVIII construct were monitored by TGA: hysteresis time (time from activation of the coagulation cascade to the first detection of thrombin), ETP (intrinsic thrombin potential - area under the thrombin signal curve), peak thrombin (maximum thrombin concentration), and ttp (time to peak - time to reach maximum thrombin concentration). All parameters are presented as a percentage of the values ​​obtained for BDD-FVIII, which were used as controls in each experiment.

[0460] The in vivo performance of the two constructs compared to BDD-FVIII was evaluated through a tail-cutting hemorrhage study in HemA mice. Male B6;129S-F8tm1Kaz / J (F8- / -) mice were treated with a single intravenous (iv) injection of BDD-FVIII, C32+C15, or C37+C15 at a dose of 75 IU / kg body weight after an adaptation period of at least 10 mice per group. Each protein dimer was administered at an application volume of 5 ml / kg body weight. One minute after administration, the animals were anesthetized (80 mg / kg, 10 ml / kg sodium pentobarbital, intraperitoneally (ip)) to ensure that stage II tolerance had been reached. Six minutes later, the animals underwent tail cutting, in which the portion 3 mm from the tip of the tail was severed with a sharp scalpel. The animal was then immediately placed on a temperature-controlled bleeding rack with its tail suspended in a sample tube filled with 0.9% saline solution and maintained at 37°C. Blood was collected over a 30-minute period, and total blood loss was determined by spectrophotometric hemoglobin measurement at 405 nm. Blood loss was calculated using hemoglobin standards prepared according to defined blood volumes and expressed as blood loss volume (µl).

[0461] result

[0462] The functional comparison of the two constructs differs only in the presence or absence of the EP repeat sequence in connector 2 of part FVIII (C32 versus C27, schematically depicted in...). Figure 15 In A), functional differences in construct performance were revealed. In TGA assays, there were no differences between the two constructs in terms of ETP and peak thrombin parameters. However, and surprisingly, two parameters related to thrombin generation kinetics differed between the two molecules tested. Specifically, C32+C15, containing three EP repeat sequences in linker 2, showed a shorter hysteresis time and a shorter time to peak compared to C37+C15 without these EPs. Figure 15 B).

[0463] Compared with mice treated with the same dose of C32+C15, HemA mice treated with C37+C15 experienced greater blood loss, further supporting the significant effect of EP in vivo. Figure 16 ).

[0464] Overall, these findings confirm the functional importance of the EP repeat sequence in linker 2. Since linker 2 contains a thrombin cleavage site, the adjacent EP repeat sequence appears to have a positive impact on FVIII activation, and thus a positive impact on construct-mediated thrombin generation kinetics.

[0465] Example 10 - Design, cloning, expression, and CK of single-domain antibody fusion proteins VWF and CK NDP Mediated dimerization

[0466] Experimental Objective

[0467] Single-domain antibodies were identified and cloned into human coagulation factors FIX and FX. Antibodies encoding different single-domain antibodies were generated and cloned into human Norrin (VHH-CK). NDP Fusion protein) or human VWF (VHH-CK) VWF The cDNA of the fusion protein containing the CK domain was cloned into an expression vector for recombinant expression of the selected VHH-CK. VWF and VHH-CK NDP The fusion protein was purified and its dimerization ability was tested.

[0468] method

[0469] Sequences of VHH1 and VHH2 were obtained from a camel immunotherapy library after immunization of llamas and alpacas with natural human plasma-derived clotting factors FIXa and FX. Immunization, peripheral blood mononuclear cell (PBMC) isolation, and phage display were performed according to a standard protocol used by the service provider, Abcore Inc. (USA). Briefly, multiple productive hemorrhages were collected throughout the animal immunization period. Serum titers of anti-FIXa and anti-FX were monitored by ELISA. At the end of the immunization protocol, PBMCs with the highest target-specific antibody titers were isolated from productive hemorrhages, and RNA was subsequently extracted. The extracted RNA was reverse transcribed, and the resulting cDNA was used to construct a DNA library encoding VHH. Both libraries were displayed on the surface of phages and underwent multiple rounds of affinity-based selection against the relevant target proteins (FIXa and FX). Individual clones were tested by ELISA based on the resulting pools of phages rich in target-specific binding agents. The genes of positive clones were analyzed using Sanger sequencing, and their target binding characteristics were confirmed using Octet (BLI) technology.

[0470] Gene synthesis, cloning, and protein expression

[0471] Codon-optimized cDNA encoding the fusion protein was ordered from IDT (Integrated DNA Technologies Inc.). As described in Example 1, a plasmid was constructed via one-step cloning, allowing the corresponding cDNA cassette to be inserted into the expression vector. The plasmid identity was verified by sequencing. The protein was transiently produced in Expi293F cells. Protein expression and secretion were monitored by Western blotting using anti-His tag (GenScript A00186-100) and anti-Strep tag (Abcam ab97023) antibodies after SDS-PAGE under non-reducing conditions. Western blotting analysis was performed on the cell supernatant 48 hours post-transfection. On day 7, cell culture supernatant was collected and immediately purified or frozen at -70°C. The protein-containing supernatant was analyzed by SDS-PAGE and Coomassie staining under both reducing and non-reducing conditions.

[0472] VHH-CK VWF Purification and analysis of heterodimers of fusion proteins

[0473] VHH1-CK was achieved by applying two consecutive affinity chromatography steps. VWF Purification of the -VHH2 dimer: affinity chromatography was performed on the Twin-Strep tag, followed by affinity chromatography on the His tag, and finally a rebuffering step was performed.

[0474] In short, the collected cell culture supernatant was loaded onto Strep-Tactin XT 4Flow resin (IBA Lifesciences). The column was equilibrated with 100 mM Tris / HCl pH 8, 150 mM NaCl, and 1 mM EDTA, and eluted with 100 mM Tris / HCl, 150 mM NaCl, 1 mM EDTA, and 50 mM biotin at pH 8.0. For the second affinity step, the HisTrap Excel column (Stenofan) was equilibrated with PBS and 350 mM NaCl, and eluted with PBS, 350 mM NaCl, and 0.5 M imidazole at pH 7.5. Finally, the resulting eluent was re-buffered with PBS (for heterodimers H2 and H3) or 150 mM sodium phosphate buffer, 150 mM NaCl at pH 6.5 (for H1, H5, and H6) using a HiTrap desalting column (Stenofan).

[0475] result

[0476] The composition of the fusion protein is indicated in Table 19. A schematic overview of its structure is shown in... Figure 17 middle.

[0477] Table 18: Linkers for VHH-CKVWF and VHH-CKNDP fusion proteins

[0478] The location of the linker in the fusion protein indicates Figure 17 middle. EP Indicates a VWF-derived extended peptide having SEQ ID NO: 2. Lowercase numbers indicate the number of repetitions of the sequence element in parentheses.

[0479]

[0480] Table 19: Overview of VHH-CK fusion protein and its components

[0481]

[0482]

[0483] For all fusion protein combinations analyzed by Western blot, signals from both tags were detected at the expected molecular weight of the dimeric protein. Therefore, VHH-CK VWF and VHH-CK NDP All combinations of tests for the fusion protein formed a dimer. Figure 18 and 19 All CKs listed in Table 19 VWF The mediated heterodimers (i.e., H1, H2, H3, H5, and H6) can all be purified using the two-step affinity chromatography purification protocol described above. The purified heterodimers were analyzed by SDS-PAGE, revealing the expected molecular weights under both reducing and non-reducing conditions. Figure 20 ).

[0484] The results confirmed that human Norrin (CK) NDP ) and VWF (CK) VWF The CK domain of VHH1-CK has the ability to mediate dimerization, regardless of its position in the fusion protein. Various layouts of VHH1-CK were successfully formed. VWF -VHH2 heterodimer, in which CK VWF The domain is placed at the C-terminus or N-terminus of the two fusion proteins, or in which CK is located. VWF The domain is placed at the C-terminus of the first fusion protein and the N-terminus of the second fusion protein. These configurations provide multiple options for the assembly of the CK domain fusion partner, such as... Figure 17As described in the text. Furthermore, fusion proteins can not only dimerize, but also trimerize and tetramerize. Therefore, the CK domain of human VWF and Norrin (NDP) can provide a universal tool for the dimerization of any combination of soluble proteins and potentially for their trimerization and tetramerization.

[0485] Furthermore, the successful inclusion of EP in the fusion protein suggests a possible mechanism for extending the half-life of dimeric single-domain antibody molecules.

[0486] Example 11 - by VHH-CK VWF Stability and function of heterodimers formed by fusion proteins

[0487] Experimental Objective

[0488] Research on VHH-CK VWF The stability of the protein dimer was compared with that of the heterodimer obtained by dimerization through the fusion of the same VHH domain with the Fc domain of IgG1. The biological function of the heterodimer was analyzed, namely, its simultaneous binding to activated coagulation factor IX (FIXa) and coagulation factor X (FX), and its FVIII-like activity was tested.

[0489] method

[0490] Further regarding VHH-CK of Example 10 VWF The fusion protein generated four additional cDNAs encoding fusions of single-domain antibodies and the Fc domain of IgG1, which were cloned into corresponding expression vectors and expressed in HEK293 cells. The single-domain antibody-IgFc fusion was used as described above for CK... VWF The fusion proteins describe the same two VHH sequences (i.e., VHH1 for human coagulation factor FIX and VHH2 for human FX). Using a knock-in-hole technique for efficient Fc chain pairing (Ridgway et al., 1996), two distinct heterodimers, H4 and TPP349, were successfully formed from four different VHH-IgFc fusion proteins. H4 is composed of the fusion protein VHH1-IgFc. k -Twin-Strep (SEQ ID NO: 66) and VHH2-IgFc h -His (SEQ ID NO: 67) is formed, and TPP349 is formed from the fusion protein VHH1-IgFc. k (SEQ ID NO: 70) and VHH2-IgFc h(SEQ ID NO: 71) Formation. H4 and TPP349 differ in the presence or absence of the (H4) or (TPP349) IgG1 hinge region, the presence or absence of the (H4) or (TPP349) C-terminal His tag and Strep tag, and the sequence of the linker connecting the Fc scaffold to VHH1 and VHH2. The layout of the Fc dimerized protein is schematically shown in Figure 21 This is described in Tables 20 and 21.

[0491] The expression and purification of Fc dimer H4 were performed as described in Example 10 for the CK dimer. The second Fc-mediated dimer, TPP349, was formed from a fusion protein expressed without His and Strep tags and purified by protein A (Amsphere A3) affinity chromatography. The protein was captured and subsequently eluted with 100 mM glycine at pH 2.7, then immediately neutralized and the buffer was replaced with PBS. Mass spectrometric analysis of the final protein sample indicated that the content of mismatched and monomeric species was less than 10% (data not shown).

[0492] Dynamic light scattering and melting temperature analysis

[0493] Dynamic light scattering (DLS) and melting temperature (Tm) analysis were performed using a Prometheus Panta instrument from Nanotemper. Samples were diluted to 0.2 mg / mL in buffer (1x PBS pH 7.4 for H2, H3, TPP349, and H4; 150 mM sodium phosphate / 150 mM NaCl pH 6.5 for H1, H5, and H6) and transferred to standard capillaries. Each sample was measured in triplicate in two separate experiments. DLS measurements were performed in high-sensitivity mode, with 10 acquisitions per capillary at 100% laser intensity and 25°C. Viscosities and refractive indices at 20°C and 405 nm laser wavelength using the BufferBuilder tool in PR.Panta control software v1.6.3 were calculated using 1x PBS pH 7.4 and 150 mM sodium phosphate / 150 mM NaCl pH 6.5. For 1x PBS at pH 7.4, the calculated viscosity was approximately 1.020 mPa*s, and the refractive index was 1.332. The values ​​for 150 mM sodium phosphate / 150 mM NaCl at pH 6.5 were 1.082 mPa*s and 1.338, respectively. Size distribution analysis was performed using PR.Panta analysis software v1.6.3.

[0494] Simultaneous acquisition using nano-differential scanning fluorescence (nanoDSF) and DLS was performed, with heating and cooling rates of 1 °C / min and 100% laser intensity, along a thermal ramp from 25 °C to 95 °C and back to 25 °C. Tm was determined by nanoDSF using the first derivative of the emission ratio at 350 nm and 330 nm. Accumulation was monitored by the increase in the hydrodynamic radius (cumulative radius) measured by DLS. All evaluations were performed using PR.Panta analysis software v1.6.3.

[0495] Oscillating stress

[0496] Oscillatory stress was achieved using a ThermoMixer® C from Eppendorf. 1.5 mL Eppendorf tubes were filled with 60 µL of sample (0.2 mg / mL; 150 mM sodium phosphate, 150 mM NaCl pH 6.5) and oscillated at 1850 rpm and 2000 rpm for 3 h, or at 1850 rpm for 3 h, 5 h, or 7 h. The temperature was adjusted to 20 °C during oscillation stress. Stressed and unstressed samples were analyzed by size exclusion chromatography on a Thermo Fisher Scientific HPLC system equipped with a Superdex 200 Increase 10 / 300 GL column (Stopvan). Analyses were performed at a flow rate of 0.56 mL / min, with 1x PBS pH 7.4 as the mobile phase and an injection volume of 45 µL. Chromatograms were collected at 280 nm, and the absorption traces were integrated using a Chromeleon 7.2 (Thermo Fisher Scientific). The peak area recorded at 280 nm was used to calculate the protein recovery after oscillation stress.

[0497] pH stability

[0498] A pH change to 3.5 was induced by adding 100 µL of sample (0.2 mg / mL) to defined volumes of HCl (0.6 µL of 1.5 M HCl for 1X PBS, pH 7.4 (H2, H3, and TPP349) and 2.34 µL of 3 M HCl for 150 mM sodium phosphate, 150 mM NaCl, pH 6.5 (H1, H5, and H6)) followed by rapid mixing to ensure a uniform pH decrease throughout the sample volume. After incubation for 100 min, the pH was neutralized by transferring the sample to a new tube containing an equimolar volume of NaOH equal to the previously added HCl. Unstressed and stressed samples were analyzed by SE-HPLC (see Oscillating Stress). Due to the introduction of inconsistent volumetric dilution, the relative areas of high molecular weight compounds (HMWCs) were compared before and after the pH change by adding HCl and NaOH.

[0499] Biological Layer Interference (BLI)

[0500] The Octet BLI Discovery V13.0 instrument was used to assess the combination of VHH1-CK with FIXa and FX via biolayer interferometry (BLI), and the data were analyzed using Octet Analysis Studio V13.0. VWF -VHH2 and VHH1-Fc-VHH2 heterodimers were loaded at a concentration of 2 µg / mL onto an AR2G (second-generation amine-reactive) sensor in 10 mM acetate buffer at pH 5.5. A reference sensor loaded with 2 µg / mL eigentimod (ArgenX, an Fc receptor antagonist that does not bind to either coagulation factors FIXa or FX) was used.

[0501] The association of activated human coagulation factors FIXa and FX (Haematologic Technologies Inc.) was performed in CAB-T buffer at concentrations of 60 nM and 100 nM (10 mM HEPES, 100 mM NaCl, 2.5 mM CaCl2, and 0.05% (w / v) Tween-20 pH 7.4). All association steps were recorded at 30°C and 100 rpm. Assays were performed in the following order: A) FIXa association first, then FX association, or B) FX association first, then FIXa association. In the second step of the association assay, the concentration of the first coagulation factor was kept constant so that the recorded signal depended only on the newly added coagulation factor. Coupling of amines to the BLI sensor was ineffective for heterodimers H2 and H3. Therefore, following the manufacturer's instructions (Thermo Fisher Scientific), these constructs and the heterodimer H4 were biotinylated using EZ-LinkNHS-PEG4-biotin at a molar coupling ratio of 3:1. In loading reconnaissance, the ideal loading concentrations for H2 and H3 were determined to be 2 µg / mL, and the ideal loading concentration for heterodimer H4 was determined to be 1 µg / mL. These concentrations were loaded onto the streptavidin sensor (SA sensor), and the same association steps were applied to FIXa and FX as described above. Analysis included subtracting the nonspecific signal recorded by the reference sensor loaded with igamod.

[0502] The simulated FVIII activity of the heterodimer was evaluated using the BIOPHEN FVIII:C kit (Hyphen Biomed) according to the protocol provided by the manufacturer.

[0503] result

[0504] The molecular structure of the VHH1-IgFc-VHH2 heterodimer is shown in Figure 21 The details are provided in Tables 20 and 21.

[0505] Table 20: Linkers for VHH-Fc fusion proteins.

[0506] The location of the linker in the fusion protein indicates Figure 21 In the middle. Lowercase numbers indicate the number of repetitions of the sequence element within the parentheses.

[0507]

[0508] Table 21: Overview of the VHH-Fc fusion protein and its components.

[0509]

[0510]

[0511] In the following experiments, the VHH1-Fc-VHH2 dimer was compared with the VHH1-CK described in Example 10. VWF -VHH2 dimer comparison.

[0512] Size distribution and thermal stability

[0513] VHH1-CK determined experimentally by DLS VWF The hydrodynamic radius (R) of the -VHH2 and VHH1-Fc-VHH2 heterodimers h The results are presented in Table 22. Among the analyzed VHH1-CK-VHH2 dimers, only H2 and H3 (molecules carrying EP) showed single particle swarms, reflected only in a single peak detected in DLS analysis at a hydrodynamic radius of approximately 6 nm. All other VHH-CK dimers (H1, H5, and H6) showed the presence of two particle swarms; one in the range of 4 to 5 nm, and the second particle swarm having a hydrodynamic radius of approximately 30 nm. Figure 22 This indicates a lower tendency for H2 and H3 to form high molecular weight compounds, likely due to the multiple negative charges provided by the sialylated O-glycans of EP. Thermal stability was investigated by recording changes in the intrinsic fluorescence of the protein during pyrolysis via nanoDSF. Melting temperature (Tm) analysis revealed that all VHH-CK molecules exhibited higher thermal stability (Tm range of 71.2 °C to 73.1 °C) compared to VHH-Fc molecules (Tm values ​​of 64.7 °C and 68.1 °C, Table 22).

[0514] Table 22. DLS analysis of the size (intensity distribution of radius) and melting temperature (Tm) of the VHH1-CK-VHH2 and VHH1-Fc-VHH2 dimers.

[0515]

[0516] After heating to 95°C and then cooling again to 25°C, refolding processes of H1, H5, and H6 were observed, as evidenced by the inflection point observed in the 350 nm / 330 nm plot during capillary cooling. Figure 23 A, the top small plot shows the 350 nm / 330 nm ratio of the parameter, and the middle small plot shows the first derivative of the parameter ratio (on the right side of the chart). Additionally, for VHH1-CK... VWF -VHH2 dimer did not record protein aggregation ( Figure 23A, lower small plot, parameter cumulative radius). In contrast, no protein refolding was observed for IgFc-based dimers H4 and TPP349, where no inflection point was detected in the 350 nm / 330 nm plot during sample cooling (after initial protein unfolding). Figure 23 B, top and middle small images). Additionally, protein aggregation was observed during the unfolding analysis of the TPP349 molecule, which was visible as an increase in the cumulative radius at higher temperatures. Figure 23 B, lower small image).

[0517] Oscillation Stress Stability

[0518] The oscillatory stress stability of the molecules was investigated in two separate experiments. In the first experiment, the stability was investigated based on IgFc and CK. VWF A representative of the heterodimer group was subjected to oscillatory stress in solution at 1850 rpm and 2000 rpm for 3 hours. TPP349 (based on IgFc) and H1 (based on CK) VWF The results of the SEC analysis after oscillating stress are presented in Figure 24 SEC analysis revealed how much intact product remained after applying oscillatory stress. It was assumed that the protein moieties under study, invisible in the SEC analysis, aggregated and / or adsorbed onto the surface via exposed hydrophobic plaques. For TPP349, less than 40% of intact heterodimers were detected after 3 hours of oscillation at 1850 rpm, and almost no intact product was observed after 3 hours of oscillation at 2000 rpm. In contrast, over 60% of H1 remained intact under both oscillatory stress conditions. To understand the fluid dynamics and identify the threshold oscillation velocity crucial for potential protein unfolding, the fluid motion as a function of oscillation velocity was evaluated via video recording. Oscillations up to approximately 1750 rpm caused the fluid to rotate along the tube wall (data not shown). Significant changes in fluid behavior were observed above 1750 rpm, causing continuous inversion of the gas-liquid interface and thus increasing interfacial stress. Based on these results, a second experiment was conducted using an oscillation velocity of 1850 rpm and durations of 3, 5, and 7 hours, where the entire CK-based system was used. VWF The heterodimers were compared with those of TPP349. For TPP349, less than 47% of intact heterodimers were detected after 3 hours of oscillation at 1850 rpm. In contrast, after 3 hours of oscillation, more than 75% of the VHH-CK heterodimers (H1, H2, H3, H5, and H6) remained intact, and this did not change significantly after extending the oscillation period to 5 or 7 hours (Table 23).

[0519] Table 23: Oscillation stress stability of VHH-Fc (TPP349) and VHH-CK constructs is shown as the recovery of the main peak area after oscillation at 1850 rpm for 3, 5 and 7 hours.

[0520]

[0521] Data confirms that compared to IgFc-based dimers, CK... VWF The mediated heterodimer exhibits superior stability against oscillating stress. This improved stability can represent a major advantage in biotechnology production, where proteins are subjected to continuous agitation during mixing, ultrafiltration / percolation, pumping, filling, and final product delivery.

[0522] pH stability

[0523] The production of biopharmaceuticals requires virus inactivation, which typically involves prolonged incubation at low pH followed by pH adjustment of the formulation.

[0524] The ability of dimer proteins to resist pH changes occurring during manufacturing process steps (e.g., elution at low pH) was evaluated by assessing the formation of high molecular weight compounds (HMWCs). HMWC formation before and after pH change treatment was analyzed by size exclusion chromatography (SEC). CK-based analysis was conducted. VWF The heterodimers H1, H2, H3, H5 and H6 were compared with the IgFc-based heterodimer TPP349 carrying the same VHH.

[0525] pH change treatment induces HMWC formation of IgFc-based dimer TPP349, but does not induce CK. VWF HMWC formation in dimer sample H5. This difference in stability was described by comparing the SE chromatograms of H5 and TPP349 before and after pH changes. The Fc fusion construct exhibited multiple different HMWC species at different retention times, indicating the presence of more than one aggregation mode. Figure 25 A). Furthermore, TPP349 was the only construct to exhibit an additional small fraction of low molecular weight species after treatment, indicating protein fragmentation. In contrast, based on CK... VWF pH changes in the heterodimers did not significantly alter a small fraction of the HMWC, and the differences between the constructs were minimal. Figure 25 B). The slight decrease in HMWC can be attributed to protein loss during processing due to adsorption onto surfaces (e.g., adsorption onto container walls, pipette tips, HPLC tubes, and column molten metal).

[0526] The results of this experiment highlight the importance of CK under manufacturing process conditions. VWFThe mediated dimer exhibits superior stability compared to the IgFc fusion protein dimer.

[0527] Binding to the target

[0528] All single-domain antibody fusion proteins were analyzed using the biolayer interferometry method (where heterodimer formation was caused by CK). VWF The VHH heterodimer binds to both FIX and FX in two sequences (either mediated by IgFc fusion). Based on the data presented in Table 24, all VHH heterodimers were able to bind to both coagulation factors FIX and FX simultaneously. The data confirm that regardless of the position of the VHH moiety relative to the CK domain (i.e., fused to the N-terminus or C-terminus), both single-domain antibodies retain their function, i.e., their ability to bind to their targets FIX and FX.

[0529] Table 24. FIX and FX vs. VHH1-CK VWF The binding of -VHH2 and VHH1-Fc-VHH2 heterodimers

[0530]

[0531]

[0532] Simulated FVIII activity

[0533] VHH1-CK was studied. VWF Whether the -VHH2 dimer can not only bind to FIX and FX, but also generate (mimicking) FVIII activity. To this end, purified heterodimers H1, H2, H3, H5, and H6 were analyzed in a colorimetric FVIII activity (FVIII:C) assay. The results of the measurements are plotted on... Figure 26 The FVIII activity of the heterodimer and the FVIII activity of standard human plasma (SHP) were expressed as absorbance at 405 nm (baseline-corrected average), representing the amount of active FX generated in the assay. Heterodimers were measured with protein concentration indicated on the X-axis. Heterodimers H5 and H6 exhibited significantly simulated FVIII activity compared to H1, H2, H3, H4, and TPP349, whose activity is not shown. Interestingly, and compared to H1 through H3, in H5 and H6, the heterodimer (fusion protein CK) showed significantly higher FVIII activity. VWF The FIXa of -VHH1) binds to the VHH arm and fuses with the C-terminus of the CK domain. Therefore, CK VWF The unique properties of the C-terminus of the domain dimer allow for the placement of VHH-bound coagulation factors FIXa and FX in a favorable orientation, and induce mimicking FVIII activity in the heterodimers H5 and H6. This confirms the CK-based... VWFDimerization of structural domains provides for a wide range of possible relative arrangements of molecules, which is something that other state-of-the-art dimerization techniques cannot achieve.

[0534] Example 12-VHH-CK VWF Cloning, expression, and purification of trivalent and tetravalent heterodimers of fusion proteins

[0535] Experimental Objective

[0536] Design and generate coded monovalent and divalent VHH-CK VWF cDNA of the fusion protein monomer. Simultaneous expression of cDNA and testing of the ability of the secreted fusion protein to dimerize into trivalent and tetravalent heterodimers. Purification of the secreted heterodimers from cell culture supernatant.

[0537] method

[0538] Gene synthesis, cloning, and protein expression

[0539] Codon-optimized cDNA was ordered from IDT (Integrated DNA Technologies). A protein expression plasmid encoding the secreted protein was constructed via one-step cloning. The coding region was verified by sequencing. The expression vector was chemically transfected, resulting in transient protein production in HEK 293 ALL cells. Cell counts were performed, and cell viability was measured at transfection and at the end of production. Plasmid identity was verified by sequencing 48 hours post-transfection. Protein expression and secretion were monitored by Western blotting following SDS-PAGE with either an anti-His tag (GenScript, A00186-100; 1:5000) or an anti-Strep tag (Abogen, ab97023; 1:8000). Western blot analysis was performed 48 hours post-transfection. At the end of production, the supernatant was analyzed by SDS-PAGE and Coomassie staining under both reducing and non-reducing conditions. On day 7, cell culture supernatant was collected and purified immediately after collection or frozen at -70°C.

[0540] Heterodimeric nanobody-CK VWF Purification and analysis of fusion proteins

[0541] Nanobody-CK was achieved by applying two consecutive affinity chromatography steps. VWFPurification of the fusion protein: Affinity chromatography was performed against the Twin-Strep tag, followed by affinity chromatography against the His tag, and finally a rebuffering step. In short, the collected cell culture supernatant was loaded onto a Strep-TactinXT 4Flow resin (IBA Life Sciences). The column was equilibrated with 100 mM Tris / HCl pH 8, 150 mM NaCl, and 1 mM EDTA, and eluted with 100 mM Tri / HCl, 150 mM NaCl, 1 mM EDTA, and 50 mM biotin at pH 8.0. For the second affinity step, the HisTrap Excel column (Stenofan) was equilibrated with PBS and 350 mM NaCl, and eluted with PBS, 350 mM NaCl, and 0.5 M imidazole at pH 7.5. Finally, the resulting eluent was rebuffered into PBS using a HiTrap desalting column (Stenofan).

[0542] result

[0543] The components and structures of fusion proteins and heterodimers are shown in Table 25 and Figure 27 and 28 middle.

[0544] Table 25: Nanobody CK VWF An overview of fusion proteins and their components.

[0545]

[0546]

[0547] The purpose of expressing the above molecule is to improve stability, prolong half-life, and link the functional group to the end of the simple protein dimer described in Example 10. An anti-albumin single-domain antibody was chosen as the functional group so that the dimer complex can bind to albumin in circulation. Since rats were used for in vivo testing of the new molecule, anti-rat serum albumin VHH R28 was chosen as ABV2 (van Faassen et al., 2020).

[0548] All Nanobodies-CK VWF The dimers were expressed as protein dimers and secreted into the cell supernatant because signals from both tags were detected at the expected molecular weight of the respective dimers (data not shown). The proteins were purified using the two-step affinity chromatography purification protocol described in Example 10. The purified heterodimers were analyzed by SDS-PAGE under reducing and non-reducing conditions. Figure 29 The results confirmed that the versatility of the fusion protein dimerization platform provides up to four ends for the fusion of functional protein groups.

[0549] Example 13-CK VWF Pharmacokinetics of VHH heterodimer mediated in rats

[0550] Experimental objective:

[0551] Successful implementation of CK-mediated anti-FIX / anti-FX VHH heterodimers in therapeutic molecules that mimic FVIII function requires appropriate pharmacokinetic (PK) properties. This study investigated whether a half-life extension module was needed to improve these properties to achieve sufficiently long dosing intervals.

[0552] To conduct experiments

[0553] - The pharmacokinetics (PK) of the fusion protein dimer were discovered in rats after intravenous (IV) administration.

[0554] - Investigate the effect of IV administration on the fusion of anti-rat serum albumin ABV2 with the fusion protein dimer on its PK.

[0555] - Investigate the effect of different amounts of EP fusion at different sites of the simulated FVIII protein after IV administration on PK.

[0556] - Investigate the PK characteristics of different heterodimers containing or without ABV2 and / or added EP after subcutaneous (SC) administration.

[0557] method

[0558] Research Design

[0559] This study was conducted in Spra-Dowley / CD rats. Animals were randomly assigned to 20 groups (n=3 per group). Fourteen groups received intravenous administration of the tested molecules, and six groups received SC administration, as described in Table 26. The purified heterodimeric fusion proteins (H3, H5, H17, H18, H19, H20, H22, H28, H30, H31, H32, H33, H34, and H36 – see Examples 10 and 11) were quantified by µBCA assay (Thermo Fisher Scientific) for dose adjustment. The smallest protein (H5) was administered intravenously at 1 mg / kg body weight (bw) and SC at 2 mg / kg body weight. All other FVIII mimic proteins were administered at the same molar dose as H5, adjusted based on their corresponding molecular weight (Table 26). At each time point, a 200 μl blood sample was collected in a sodium citrate tube via tail vein bleeding and stored at -80°C until analysis. IV samples were collected at the following time points after administration: 0 minutes (before administration), 5 minutes, 1 hour, 6 hours, 12 hours, 24 hours, 48 ​​hours, 60 hours, 72 hours, 96 hours, 120 hours, and 192 hours. SC samples were collected at the following time points after administration: 0 minutes (before administration), 1 hour, 12 hours, 24 hours, 48 ​​hours, 60 hours, 72 hours, 96 hours, 120 hours, and 192 hours.

[0560] Quantitative analysis of fusion protein dimers in rat plasma

[0561] Plasma concentrations of protein dimers were determined by immunoassay using a Gyrolab xPlore instrument (GyrosProtein Technologies, Uppsala, Sweden). The assay setup was as follows: Polyclonal anti-VWF antibody (Dako A / S) was biotinylated using EZ-Link sulfonyl-NHS-LC-biotin reagent (Thermo Fisher Scientific) according to the manufacturer's instructions, then diluted to 100 µg / ml in PBS-T and used as the capture reagent. Standards and rat plasma samples were diluted 1:10 in RexxipA buffer (GyrosProtein Technologies). An anti-His antibody conjugate (anti-His-tagged Alexa Fluor® 647 antibody, Bio-techne) was diluted to 25 nM in RexxipF and used for detection. Quantification was performed on a Gyrolab Bioaffy1000 CD instrument (GyrosProtein Technologies).

[0562] result

[0563] The tested fusion protein dimers exhibited broad PK properties after IV administration, depending on the individual design of the molecules:

[0564] Unmodified VHH-CK dimer H5 showed the shortest half-life (3.19 hours). The half-life of the protein dimer gradually increased with increasing amount of added EP (e.g., H33 with 1 EP showed a T... 1 / 2 The T-type H17 with 3 EP has a lifespan of 10.9 hours. 1 / 2 The T-type H18 has a lifespan of 19.2 hours and 6 EP. 1 / 2 (25.15 hours). The most significant effect was achieved by adding EP at the C-terminal position of the molecule (Table 26, Figure 27 , 28 and 30).

[0565] The added EP has different effects on the half-life of different variants. Interestingly, H30, with 9 consecutive EPs at the N-terminus, shows better results than H34 (T) with the same total number of EPs. 1 / 2 (40.3 hours) shorter T 1 / 2 (22.74 hours), but the EP is distributed at the three ends of the heterodimer (i.e., there are 3 EP at each end). Figure 27 , 28 And 30). Furthermore, H36, equipped with one EP at each of the four ends of the dimer (a total of 4 EPs), has a significantly longer T compared to H30. 1 / 2 (32.1 hours vs. 22.7 hours; Table 26). Two fusion protein dimers, H19 and H31, containing single-domain antibodies ABV2 at different positions, show different PK properties depending on their position in the molecule. Figure 27 , 28 (and 30 and Table 26). This indicates that the position of ABV2 in the VHH portion has a significant impact on the half-life of the dimer.

[0566] The effects of EP and ABV2 on the PK of VHH-CK dimer after SC administration were similar to those after IV administration. Figure 31 Furthermore, subcutaneous administration of VWF-derived EP or anti-RSA single-domain antibody ABV2 improved the bioavailability of the protein dimer. This was evident from the increased Cmax and AUC of all tested constructs compared to H5 (Table 26).

[0567] In summary, by altering the quantity and position of EP and / or ABV2, fusion protein dimers with a wide range of different PK properties were achieved, which means that a comprehensive selection of half-life and bioavailability of custom therapeutic molecules can ultimately be used to tune them.

[0568]

[0569] Benefiting from the teachings presented in the foregoing description and the associated drawings, those skilled in the art will conceive of many modifications and other embodiments of the invention set forth herein. Therefore, it will be understood that the invention should not be limited to the specific embodiments disclosed, and that modifications and other embodiments are intended to be included within the scope of the appended claims. While specific terminology is used herein, such terminology is used only in a general and descriptive sense and is not intended for limiting purposes.

[0570] References

[0571] Chen X, Zaro J, Shen WC. Fusion Protein Linkers: Property, Design and Functionality. Advanced Drug Delivery Review. 2013, 65(10): 1357-1369.

[0572] Daly NL, Craik DJ. Bioactive cystine knotproteins. Current Opinion in Chemical Biology, 2011, 15(3): 362-368.

[0573] Dumont JA, Liu T, Low SC, Zhang X, Kamphaus G, et al. Prolonged activity of a recombinantfactor VIII-Fc fusion protein in hemophilia A mice and dogs. Blood, 2012, 119(13): 3024-3030.

[0574] Engler C, Kandzia R, Marillonnet S. A one-pot, one-step precision cloning method with high throughput capability. PLoS ONE 2008, 3(11): e3647.

[0575] Ewenstein BM, Collins P, Tarantino MD, Negrier C, Blanchette V, Shapiro AD, Baker D, Spotts G, Sensel M, Yi SE, Gomperts ED. Hemophilia therapy innovation development of an advanced category recombinant factor VIII by a plasma / albumin-free method. Proceedings of a Special Symposium at the XIXth Congress of the International Society on Thrombosis and Haemostasis. 2004, Vol. 41: 1-16.

[0576] Fischer B, Mitterer A, Schlokat U, Den Bouwmeester R, Dorner F. Structural analysis of recombinant von Willebrand factor: identification of hetero- and homo-dimers. FEBS Letters, 1994, 351(3): 345-348. (Corrected in FEBS Letters, 1994, 353(3): 337.)

[0577] Furlan M. Von Willebrand factor: molecular size and functional activity. Ann Hematol. 1996, 72(6): 341-348.

[0578] Innis M, Gelfand D, Sninsky J, White T. *PCR: A Guide to Methods and Application*. Academic Press, New York, 1990.

[0579] Green MR and Sambrook J. Molecular Cloning: A Laboratory Manual. 4th ed. Cold Spring Harbor Laboratory Press, 2012, Cold Spring Harbor, New York.

[0580] Hamers-Casterman C, Atarhouch T, Muyldermans S, Robinson G, Hamers C, Songa EB, Bendahman N, Hamers R. Naturally occurring antibodies devoid of light chains. Nature 1993, 363(6428): 446-448.

[0581] Huang F and Nau WM. A Conformational Flexibility Scale for Amino Acids in Peptides. Angewandte Chemie International Edition. 2003, 42: 2269-2272.

[0582] Kannicht C, Ramstrøm M, Kohla G, Tiemeyer M, Casademunt E, Walter O, Sandberg H. Characterisation of the post-translational modifications of a novel, human cell line-derived recombinant human factor VIII. Thromb Research. 2013, 131(1): 78-88.

[0583] Krishnamurthy A and Jimeno A. Bispecific antibodies for cancer therapy: A review. Pharmacol Ther. 2018, 185: 122-134.

[0584] Needleman SB, Wunsch CD. A general method applicable to the search for similarities in the aminoacid sequence of two proteins. Journal of Molecular Biology. 1970, 48(3):443-453.

[0585] Ridgway JB, Presta LG, Carter P. 'Knobs-into-holes' engineering of antibody CH3 domains for heavychain heterodimerization. Protein Engineering. 1996, 9(7): 617-621.

[0586] Runcie K, Budman DR, John V, Seetharamu N. Bi-specific and tri-specific antibodies - the next big thing in solid tumor therapeutics. Molecular Medicine 2018, 24(1): 50.

[0587] Sherbet, GV. Growth Factor Families, Growth Factors and Their Receptors in Cell Differentiation. Cancer and Cancer Therapy, 2011, 3-5.

[0588] Strohl WR. Fusion Proteins for Half-Life Extension of Biologics as a Strategy to Make Biobetters. BioDrugs. 2015, 29(4): 215-239.

[0589] Tiede A. Half-life extended factor VIII for the treatment of hemophilia A. Journal of Thrombosis and Haemost. 2015, Vol. 13, Supplement 1: S176-179.

[0590] van Faassen H, Ryan S, Henry KA, Raphael S, Yang Q, et al. Serum albumin-binding VHHs with variable pH sensitivities enable tailored half-life extension of biologics. Federation of American Societies for Experimental Biology Journal (FASEB J). 2020, 34(6):8155-8171.

[0591] Vlot AJ, Koppelman SJ, Meijers JC, Dama C, van den Berg HM, Bouma BN, Sixma JJ, Willems GM. Kinetics of factor VIII-von Willebrand factor association. Blood. 1996, 87(5): 1809-1816.

[0592] Vitt, UA, Hsu, Y, Hsueh, AJW. Evolution and Classification of Cystine Knot-Containing Hormones and Related Extracellular Signaling Molecules. Molecular Endocrinology. 2001, 15 (5): 681-694.

[0593] Vollack-Hesse N, Oleshko O, Werwitzke S, Solecka-Witulska B, Kannicht C, Tiede A. Recombinant VWF fragments improve bioavailability of subcutaneous factor VIII in hemophilia A mice. Blood. 2021, 137(8): 1072-1081.

[0594] Yee A, Gildersleeve RD, Gu S, Kretz CA, McGee BM, Carr KM, Pipe SW, Ginsburg D. A von Willebrand factor fragment containing the D'D3 domains is sufficient to stabilize coagulation factor VIII in mice. Blood. 2014, 124(3): 445-452.

[0595] Sequence Overview

[0596]

[0597]

[0598]

Claims

1. A protein dimer formed from a first fusion protein and a second fusion protein, wherein the first fusion protein comprises an FVIII protein, preferably containing a first linker, the FVIII protein being coupled via a second linker to a von Willebrand factor cystine knot domain (CK). VWF The second fusion protein contains a von Wöhlerbrand factor (VWF) fragment, which binds to CK via a third linker. VWF Domain fusion, wherein the third linker is an engineered peptide, and wherein the two fusion proteins are fused via their CK domains. VWF Covalent connection of structural domains.

2. The protein dimer according to claim 1, wherein the CK VWF The domain has an amino acid sequence that is at least 90%, more preferably at least 95%, and most preferably at least 98% identical to SEQ ID NO:

1.

3. The protein dimer according to claim 1 or 2, wherein the first fusion protein and / or the second fusion protein comprises at least one copy of an extended peptide (EP), wherein the EP has an amino acid sequence having at least 90% identity with SEQ ID NO: 2 and contains a cluster of O-glycosylation sites, wherein the cluster contains at least two O-glycosylated amino acids.

4. The protein dimer according to claims 1 to 3, wherein the first fusion protein and / or the second fusion protein comprises at least one half-life extension portion, wherein the at least one half-life extension portion is preferably selected from the immunoglobulin Fc domain, serum albumin or a portion thereof, albumin-binding domain, and most preferably the albumin-binding V domain. H H-structure domain.

5. The protein dimer of claim 4, wherein the first pharmaceutically active protein is an FVIII protein, wherein the FVIII protein preferably comprises an FVIII heavy chain, the FVIII heavy chain being fused to an FVIII light chain via a first linker, wherein more preferably, the FVIII heavy chain lacks a B domain, and wherein most preferably, the first linker preferably comprises a sequence derived from a B domain of FVIII.

6. The protein dimer of claim 5, wherein the FVIII heavy chain comprises an amino acid sequence having at least 90%, more preferably at least 95%, and most preferably at least 98% identity with SEQ ID NO: 3, and / or the FVIII light chain comprises an amino acid sequence having at least 90%, preferably at least 95%, and more preferably at least 98% identity with SEQ ID NO:

5.

7. The protein dimer according to any one of claims 4 to 6, wherein the second pharmaceutically active protein is a fragment of VWF, wherein the VWF fragment preferably has an amino acid sequence having at least 90%, more preferably at least 95%, and most preferably at least 98% identity with SEQ ID NO:

7.

8. The protein dimer according to any one of the preceding claims, wherein the first linker and the second linker are engineered peptides, and / or wherein the third linker does not contain the C3 domain of VWF.

9. The protein dimer according to any one of the preceding claims, wherein the first linker, the second linker and / or the third linker comprises at least one copy of the EP, preferably at least two copies, more preferably at least three copies, wherein the EP is preferably assembled in a sequential order.

10. The protein dimer according to any one of the preceding claims, wherein the extended half-life portion a) is fused to the C-terminus of the first fusion protein and / or the second fusion protein via a fourth linker, or b) forms part of the third linker.

11. The protein dimer according to any one of the preceding claims, wherein the first linker, the second linker, and / or the third linker comprises a flexible motif, specifically selected from (GGS). n (GGGS) n Or (GGGGS) n , where n is an integer in the range of 1 to 10, and where G represents glycine and S represents serine.

12. The protein dimer according to any one of the preceding claims, wherein the amino acid sequence of the first linker has at least 90%, at least 95%, or at least 98% identity with a sequence selected from the group consisting of: SEQ ID NO: 8, SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 43, and SEQ ID NO:

44.

13. The protein dimer according to any one of the preceding claims, wherein the amino acid sequence of the second linker has at least 90%, at least 95%, or at least 98% identity with a sequence selected from the group consisting of: SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, and SEQ ID NO:

45.

14. The protein dimer according to any one of the preceding claims, wherein the amino acid sequence of the third linker has at least 90%, at least 95%, or at least 98% identity with a sequence selected from: SEQ ID NO: 17, SEQ ID NO: 18, and SEQ ID NO:

55.

15. The protein dimer according to any one of the preceding claims, wherein the first fusion protein is an FVIII fusion protein, the FVIII fusion protein comprising an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98%, and most preferably 100% identity with a sequence selected from the following: SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31, SEQ ID NO: 47, SEQ ID NO: 48, SEQ ID NO: 51, SEQ ID NO: 52, SEQ ID NO: 53, and SEQ ID NO:

54.

16. The protein dimer according to any one of the preceding claims, wherein the second fusion protein is a VWF fusion protein, the VWF fusion protein comprising an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98%, and most preferably 100% identity with a sequence selected from the following: SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, SEQ ID NO: 49, and SEQ ID NO:

50.

17. A protein dimer formed of a first fusion protein and a second fusion protein, wherein the first fusion protein comprises a cystine knot domain (CK) fused to a first pharmaceutically active protein, and the second fusion protein comprises a CK domain fused to a second pharmaceutically active protein, wherein the two fusion proteins are covalently linked through their CK domains, provided that the first pharmaceutically active protein and the second pharmaceutically active protein are not selected from the group consisting of VWF, FVIII and fragments thereof.

18. The protein dimer according to claim 17, wherein the CK domain is selected from CK. VWF domain and CK NDP Structural domain, preferably the CK structural domain is CK VWF Structural domain.

19. The protein dimer according to claim 17 or 18, wherein the CK VWF The domain has an amino acid sequence that is at least 90%, more preferably at least 95%, and most preferably at least 98% identical to SEQ ID NO:

1.

20. The protein dimer according to any one of claims 17 to 19, wherein the first fusion protein and / or the second fusion protein comprises at least one copy of an extended peptide (EP), wherein the EP has an amino acid sequence having at least 90% identity with SEQ ID NO: 2 and contains a cluster of O-glycosylation sites, wherein the cluster contains at least two O-glycosylated amino acids.

21. The protein dimer according to any one of claims 17 to 20, wherein the first fusion protein and / or the second fusion protein comprises at least one half-life extension portion, wherein the at least one half-life extension portion is preferably selected from immunoglobulin Fc domains, serum albumin or a portion thereof, albumin-binding domains, and most preferably, the half-life extension portion is albumin-binding V H H-structure domain.

22. The protein dimer according to any one of claims 17 to 21, wherein the first pharmaceutically active protein and the second pharmaceutically active protein are the same or different and preferably selected from the group consisting of: prothrombin, fibrinogen, FIII, FV, FVII, FIX, FX, FXI, FXII, FXIII, ADAMTS13, antithrombin, α-1 antitrypsin, C1 inhibitor, antichymotrypsin, PAI-1, PAI-3, α2-macroglobulin, TFPI, heparin cofactor II, protein C, protein S, protein Z, fragments of the aforementioned proteins, and VHH domain, wherein the VHH domain is preferably bound to: FIX, FIXa, FX, activated protein C, protease microtubule connexin-1, protein Z-dependent protease, antithrombin, protein S, ADAMTS13, platelet GpIIb / IIIa receptor, complement C5, complement C3, P-selectin, human serum albumin, or FcRn.

23. The protein dimer according to any one of claims 17 to 22, wherein the first pharmaceutically active protein is a VHH domain, the VHH domain specifically binding to FIX, wherein the VHH domain bound to FIX preferably comprises an amino acid sequence having at least 90%, more preferably at least 95%, and most preferably at least 98% identity with SEQ ID NO:

72.

24. The protein dimer of claim 23, wherein the second pharmaceutically active protein is a VHH domain, the VHH domain specifically binding to FX, wherein the VHH domain binding to FX preferably comprises an amino acid sequence having at least 90%, more preferably at least 95%, and most preferably at least 98% identity with SEQ ID NO:

73.

25. The protein dimer according to any one of claims 17 to 24, wherein the amino acid sequence of the first linker and / or the second linker has at least 90%, at least 95%, or at least 98% identity with a sequence selected from the group consisting of: SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, and SEQ ID NO:

59.

26. The protein dimer according to any one of claims 17 to 25, wherein the first fusion protein is a VHH fusion protein, the VHH fusion protein comprising an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98%, and most preferably 100% identity with a sequence selected from the following: SEQ ID NO: 60, SEQ ID NO: 62, SEQ ID NO: 64, SEQ ID NO: 68, SEQ ID NO: 82, SEQ ID NO: 83, SEQ ID NO: 84, SEQ ID NO: 86, SEQ ID NO: 88, SEQ ID NO: 90, SEQ ID NO: 93, SEQ ID NO: 95, and SEQ ID NO:

98.

27. The protein dimer of claim 26, wherein the second fusion protein is a VHH fusion protein, the VHH fusion protein comprising an amino acid sequence having at least 90%, preferably at least 95%, more preferably at least 98%, and most preferably 100% identity with a sequence selected from the following: SEQ ID NO: 61, SEQ ID NO: 63, SEQ ID NO: 65, SEQ ID NO: 69, SEQ ID NO: 85, SEQ ID NO: 87, SEQ ID NO: 89, SEQ ID NO: 91, SEQ ID NO: 92, SEQ ID NO: 94, SEQ ID NO: 96, SEQ ID NO: 97, and SEQ ID NO: 99.

Images