JP2014218510A - Fusion proteins forming trimers - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide fusion proteins forming trimers which allow efficient recombinant manufacture combined with good trimerization properties and improved pharmaceutical properties.SOLUTION: The present invention relates to a fusion protein comprising (i) a collectin family trimerization domain comprising a. a collectin family Carbohydrate Recognition Domain; and b. a collectin family neck region; (ii) a linker element; and(iii) an effector polypeptide, wherein the effector polypeptide is located N-terminally in the collectin family neck region. The fusion proteins, and the nucleic acid encoding the same are suitable as pharmaceutical composition or for therapeutic, diagnostic and/or research applications as described herein.

Description

(Description)
(Field of Invention)
The present invention refers to a fusion protein comprising a neck region of a collectin trimerization domain and a carbohydrate recognition domain, a linker element, and an effector polypeptide. Furthermore, the present invention refers to a nucleic acid encoding said fusion protein. As described herein, the fusion proteins, nucleic acids and cells are suitable as pharmaceutical compositions or for therapeutic, diagnostic and / or research applications.

(background)
Collectin proteins are known to form stable oligomers. Collectin is one of 18 group members that make up the protein lectin superfamily containing a structural protein fold called C-type lectin domain (Zelensky et al., FEBS Journal 2005, Vol 272, p 6179-6217). . Lectins are proteins that bind to carbohydrates, and C-type lectins require calcium for binding. Since the C-type lectin domain is involved in carbohydrate binding, this domain is also called carbohydrate recognition domain (CRD). Collectins are involved in innate immunity and, among other functions, neutralize pathogens by binding to carbohydrates (eg present on viruses and bacteria). In addition, collectins regulate immune functions such as complement activation and inflammation effects. The basic structural features of collectins are the collagen-like domain and the lectin domain, which are constituents that represent the name of the collectin. Some members have been shown to contain additional structural features, thus they contain the following components: i) N-terminal collagen domain, ii) connected to this N-terminal collagen domain, Alpha helix segment, also called the neck region, and iii) C-terminal CRD (Figure 35). Collectins trimerize non-covalently through the “triple helix collagen” region, the “coiled coil neck” region and the CRD region. In humans, the collectin group includes serum mannose binding proteins, liver, kidney, placenta and lung collectins. Four lung collectins, including lung surfactant proteins -A and -D (SP-A and SP-D), containing an N-terminal cysteine involved in disulfide-mediated oligomerization of preformed trimers It is known. For example, SP-D forms a trimer tetramer and SP-A forms a trimer hexamer (Kisore et al., Mol. Immunol. 2006, Vol. 43, 1293-1315. ).

  In an attempt to provide a trimer complex of TNF superfamily cytokines, recombinant fusion proteins comprising TNF cytokines and multimerization components have been proposed as one possible approach (eg, US Pat. However, the disclosed construct exhibited a trimerization domain that is high molecular weight and has inefficient trimerization properties.

  Schneider et al. (J Exp Med 187 (1989), 1205-1213) describe that the trimer of TNF cytokines is stabilized by a stabilization motif located at the N-terminus. In CD95L, stabilization of the CD95L receptor binding domain trimer is probably caused by the N-terminal amino acid domain located in the vicinity of the cell membrane.

  Shiraishi et al. (Biochem Biophys Res Commun 322 (2004), 197-202) suggests that the receptor binding domain of CD95L can be stabilized by an artificial α-helix coiled coil (leucine zipper) motif located at the N-terminus. It is described that there is sex. However, the orientation of the polypeptide chains relative to each other (eg, parallel or antiparallel directions) has been found to be almost impossible to predict. In addition, it is difficult to determine the optimal number of heptad repeats in a coiled coil zipper motif. In addition, coiled coil structures tend to form polymer aggregates after changes in pH and / or ionic strength.

  McAlinden et al. (J of Biol Chem, 2002, 277 (43): 41274-41281) correspond to the human type IIA procollagen amino acid sequence and the first two heptad repeats of the neck domain of rat surfactant protein (SP-D). The preparation of a fusion protein between 14 amino acid sequences is disclosed.

  Patent Document 2 discloses the preparation of a fusion protein between surfactant protein-D and CD40L containing a signal sequence, a collagen domain and a neck domain. The disadvantage of these fusion proteins is that they lead to highly immunogenic multimeric aggregates and they do not produce biochemically defined trimer ligands.

  By using collectin trimerization domains as a means to form controlled trimers, avoidance of the stated problems existing in the art can be achieved. Attempts to do this have been made in the field. However, only the coiled-coil neck region of the collectin trimerization domain (CRD) has been used in such attempts. As described in U.S. Patent No. 6,047,032, it is possible to use the coiled coil-like neck region of SP-D itself as a trimerization domain fused to a protein domain at either the N or C terminus.

  However, when using a coiled coil neck region alone, it is difficult to determine the optimal number of heptad iterations (full length) to obtain a stable trimer. The presented portion of the SP-D neck region does not itself form a sufficiently stable trimer, but for its length or degree of repeat to produce a stabilized trimer fusion protein. Need to be optimized. In addition, coiled coil structures tend to form polymer aggregates after changes in pH and / or ionic strength. Thus, the collectin neck region α-helix bundle exists as a trimer molecule only under conditions that mimic or approximate physiological conditions. This implies that purification strategies using pH fluctuations and / or changes in ionic strength may have an adverse effect on the trimer state of the fusion protein based only on the neck.

  Attempts have also been made to oligomerize antibody fragments by using collectin trimerization domains. For example, fusion proteins containing anti-CD89-Fab or anti-CD64-Fab (neck + CRD domain) fused to recombinant human fragment SP-D have been studied, which target pathogens against neutrophils Has been found to be effective. The proposed fusion protein has been created by chemical cross-linking, which creates a mixture of protein products and requires a complex purification regime to obtain the desired protein species. It is.

  For human SP-D, a variant (SPD_F335A, Crouch et al., JBC 281: 18008-18014) in which the amino acid phenylalanine 335 (corresponding to amino acid 355 of SEQ ID NO: 21) is mutated to alanine is described. This mutant showed very weak carbohydrate binding.

International Publication No. 01/49866 International Publication No. 01/42298 International Publication No. 95/31540

  To enable an efficient manufacturing process for Fab-SP-D based fusion proteins, rather than requiring tedious purification techniques, controlled production of defined products instead of crude mixtures A process that enables it is desirable.

  The inventors have overcome the problems present in the art with the fusion proteins disclosed herein, and various effector polypeptides, such as cytokines of the TNF superfamily, or antibody fragments or single chain antibodies. It has been found that controlled production of the mer is possible.

  The object of the present invention was to provide a trimer-forming fusion protein that allows the efficient production of recombinants that combine good trimerization properties and improved pharmaceutical properties.

(Summary of Invention)
The present invention
(I) a. A collectin family carbohydrate recognition domain; and b. A collectin family trimerization domain comprising a collectin family neck region;
(Ii) a linker element; and (iii) an effector polypeptide, wherein the effector polypeptide is located at the N-terminus of the collectin family neck region
A fusion protein comprising

  The invention further relates to a nucleic acid molecule encoding a fusion protein as described herein, a cell or non-human organism transformed or transfected with a nucleic acid molecule as described herein.

  The invention also relates to a fusion protein, nucleic acid molecule or cell as described herein for use as a medicament.

  The invention further provides for the treatment and / or treatment of neoplasms, inflammatory, infectious, degenerative, hereditary, proliferative and vascular diseases, and premalignant and malignant cancerous conditions, cancer and congenital abnormalities. Or, for use in prevention, related to fusion proteins, nucleic acid molecules, or cells as described herein.

The present invention also relates to pharmaceutical or diagnostic compositions comprising a fusion protein, nucleic acid molecule or cell as described herein as an active agent. Fusion proteins, nucleic acid molecules or cells as described herein are neoplastic, inflammatory, infectious, degenerative, hereditary, proliferative and vascular diseases, as well as premalignant and malignant cancers. It can be used for the preparation of pharmaceutical compositions in the prevention and / or treatment of sexual conditions, cancer and birth defects.
Accordingly, the present invention provides the following items:
(Item 1)
(I) a. A collectin family carbohydrate recognition domain; and b. A collectin family trimerization domain comprising a collectin family neck region;
(Ii) a linker element; and (iii) a fusion protein comprising an effector polypeptide located at the N-terminus of the collectin family neck region.
(Item 2)
Item 1 wherein the carbohydrate recognition domain is a carbohydrate recognition domain of surfactant protein-D, surfactant protein-A1, surfactant protein-A2, mannan-binding protein-C, liver collectin 1, placental collectin 1 or collectin-11 The fusion protein described.
(Item 3)
Item 3. The fusion protein according to item 1 or item 2, wherein the carbohydrate recognition domain comprises one or more amino acid mutations or substitutions.
(Item 4)
The carbohydrate recognition domain is a carbohydrate recognition domain of surfactant protein D, and the substitution is a substitution of phenylalanine 355 of SEQ ID NO: 21, and is mutated to a polar amino acid selected from aspartic acid, glutamic acid, asparagine and glutamine 4. The fusion protein according to item 3.
(Item 5)
Item 1-4, wherein the collectin family neck region is a neck region of surfactant protein-D, surfactant protein-A1, surfactant protein-A2, mannan-binding protein-C, liver collectin 1, placental collectin 1 or collectin-11 The fusion protein according to any one of the above.
(Item 6)
6. The fusion protein according to any one of items 1 to 5, wherein the linker element has an amino acid length of 25 or less.
(Item 7)
Item 7. The fusion protein according to any one of items 1 to 6, wherein the linker element is constructed with an amino acid selected from alanine, threonine, glycine and serine.
(Item 8)
Item 8. The fusion protein according to any one of Items 1 to 7, wherein the effector polypeptide is a TNF superfamily cytokine, a receptor binding domain thereof, or a receptor for the cytokine.
(Item 9)
8. The fusion protein according to one of items 1 to 7, wherein the effector polypeptide is a TNF superfamily cytokine, a receptor binding domain thereof, or a mutant of a receptor for the cytokine.
(Item 10)
10. Any one of items 8 or 9, wherein the effector protein is a TNF superfamily cytokine, receptor binding domain or variant thereof, and the N-terminal stalk region of such cytokine is not included in the fusion protein. The fusion protein according to 1.
(Item 11)
Item 11. The item 1-10, wherein the TNF superfamily cytokine or its receptor binding domain is selected from TRAIL, CD95L, APRIL, LIGHT and RANK-L or their respective receptor binding domains. Fusion protein.
(Item 12)
11. The fusion protein according to any one of items 1 to 10, wherein the receptor for cytokine is selected from TRAIL-R1, TRAIL-R2, LTbetaR, TNFRSF13B, TNFRSF17 and IL-4R-alpha.
(Item 13)
Item 8. The fusion protein according to any one of Items 1 to 7, wherein the effector polypeptide is an antibody, a single chain antibody, or an antibody or a fragment of a single chain antibody.
(Item 14)
14. The fusion protein according to item 13, wherein the antibody, the single-chain antibody, or the fragment of the antibody or the single-chain antibody is prepared against a TNF superfamily cytokine, a receptor binding domain thereof, or a receptor for the cytokine.
(Item 15)
Item 15. The fusion protein according to item 14, wherein the receptor for cytokine is IL4R-alpha.
(Item 16)
The fusion protein according to any one of the above items, comprising an antibody, a single-chain antibody, or an antibody or a fragment of a single-chain antibody at the C-terminus of the collectin family carbohydrate recognition domain.
(Item 17)
The fusion protein according to any one of items 1 to 16, for use as a pharmaceutical product.
(Item 18)
Items 8-12 or 14 and 14 for the treatment of proliferative disorders associated with TNF cytokine dysfunction, including tumors, infections, inflammatory diseases, metabolic diseases, autoimmune disorders, degenerative diseases and transplant rejection 15. The fusion protein according to any one of 15.

FIG. 1 is an SEC of affinity purified CD95L-ASPD. FIG. 2 is a silver gel of SEC fractions A1 to A11 derived from affinity-purified CD95L-ASPD. FIG. 3 shows caspase activity in Jurkat cells induced by affinity purified CD95L-ASPD derived SEC fractions A1-A15. FIG. 4 is the cytotoxicity of CD95L-ASPD against WM35, HT1080 and HeLa cells. FIG. 5 is an SEC of affinity-purified LIGHT-ASPD. FIG. 6 is the binding of HVEM-Fc to immobilized LIGHT-ASPD. FIG. 7 is a Western blot from HEK cells transiently transfected with the TRAIL construct. FIG. 8 shows caspase activity in Jurkat T cells. FIG. 9 is a size exclusion chromatography of TRAIL-ASPD. FIG. 10 shows the cytotoxic activity of TRAIL-ASPD against human cancer cells. FIG. 11 shows TRAIL-ASPD-induced caspase activity in Jurkat. FIG. 12 is a cytotoxicity assay with TRAIL-ASPD or TRAIL-DSPD against HT1080 cells. FIG. 13 is a western blot from transiently transfected HEK cells transiently transfected with a TRAIL-SPD construct or a TRAIL receptor selective SPD construct. FIG. 14 shows that TRAIL receptor selective ligands (TRAILR1mut and TRAILR2mut) immobilized on Streptactin plates are detected differently by TRAIL receptor 1-Fc or TRAIL receptor 2-Fc. FIG. 15 is the binding of the TRAIL receptor to a receptor selective “mutein” ligand. FIG. 16 is a size exclusion chromatography of affinity purified TRAILR1mut-ASPD. FIG. 17 is silver-stained SDS-PAGE of SEC fractions A1 to A14 derived from affinity-purified TRAILR1mut-ASPD. FIG. 18 shows caspase activity in Jurkat cells of SEC fractions A1 to A14 derived from affinity-purified TRAILR1mut-ASPD. FIG. 19 is a size exclusion chromatography of affinity purified TRAILR2mut-ASPD. FIG. 20 is a silver-stained SDS-PAGE of SEC fractions A1 to A14 derived from affinity-purified TRAILR2mut-ASPD. FIG. 21 is a Jurkat killing assay Jurkat of SEC fractions A1-A14 derived from affinity purified TRAILR2mut-ASPD. FIG. 22 is the cytotoxic activity of TRAIL-ASPD, TRAILR1mut-ASPD and TRAILR2mut-ASPD against human cancer cells. The receptor selective TRAIL-SPD protein is highly soluble. FIG. 24 is an SEC of affinity purified TRAIL-ASPD_F335A. FIG. 25 is a silver-stained SDS-PAGE of SEC fractions A1 to A13. FIG. 26 shows the cytotoxic effect of TRAIL-ASPD_F335A on human cancer cells. FIG. 27 is an SEC of affinity purified TRAIL-ASPD_F335D. FIG. 28 is a silver-stained SDS-PAGE of SEC derived from affinity purified TRAIL-ASPD_F335D. FIG. 29 shows the cytotoxic effect of TRAIL-SPD_F335D on human cancer cells. FIG. 30 shows the binding of TRAIL-ASPD fusion protein to carbohydrate. FIG. 31 is the pharmacokinetics of TRAIL-ASPD (A) or TRAIL-ASPD_F335D (B) fusion protein. FIG. 32 shows caspase activity in primary human hepatocytes. FIG. 33 is a Western blot of supernatant from HEK293 cells transiently transfected with a trimeric APRIL construct. TACI-Fc binds to APRIL-ASPD. FIG. 35 is a schematic diagram of the domain structure of collectin SP-D. The collagen region and neck region trimerize collectin, and the N-terminus further oligomerizes the trimer into a trimer tetramer or hexamer. CRD mediates binding to carbohydrates and is also involved in trimer formation. FIG. 36 is a schematic diagram showing the general structure of TNF-SF protein.

Represents the cell membrane; the N-terminus is located within the cell; 1. antiparallel β-fold of receptor binding domain (RBD); 2. RBD and cell membrane interface; Protease cleavage site.
FIG. 37 is a schematic diagram showing the structure of a natural TNF-SF trimer. The cylindrical structure (1) represents RBD, the N-terminus (2) forms a stalk and connects the RBD with the cell membrane. FIG. 38 is a schematic diagram illustrating the modifications introduced to minimize TNF-SF-RBD. N-terminal stalk has been deleted. 1. 1. antiparallel β-fold of receptor binding domain (RBD); Interface between RBD and cell membrane. FIG. 39 is an affinity purified Sp-sc006-ASPD-St silver gel. FIG. 40 is a size exclusion chromatography of affinity purified Sp-sc006-ASPD-St.

(Detailed description of the invention)
A fusion protein as disclosed herein can be a monomeric protein or a multimeric protein. Preferably, the fusion protein exists as a trimeric complex consisting of three monomer units which may be the same or different. Preferably, the trimeric complex consists of three identical fusion proteins. In a further preferred embodiment, the complex comprises a covalent bond between the three fusion proteins described herein (eg, disulfides between cysteines of collectin trimerization domains as described herein. Formed by cross-linking covalent bonds).

  Such a trimer complex exhibits biological activity. However, oligomers of the trimer complex (eg, certain complexes in which the basic trimer structure is present 2, 3 or 4 times) have also been found to have biological activity. Accordingly, oligomers of the trimer complex are also preferred.

The fusion protein includes the following elements:
(I) a. A collectin family carbohydrate recognition domain; and b. Collectin family neck region;]
A collectin family trimerization domain comprising (ii) a linker element; and (iii) an effector polypeptide, wherein the effector polypeptide is located at the N-terminus of the collectin family neck region.

  A collectin trimerization domain, as used herein, is generally derived from the C-terminal portion of a collectin polypeptide. A trimerization domain as used herein comprises a coiled-coil region (referred to as a neck region in certain embodiments) and a carbohydrate recognition domain (also referred to herein as a CRD).

  The collectin trimerization domain may comprise any collectin family member. Such members and their structures have been summarized, for example, in Hkansson et al. (Protein Science, 2000, 9: 1607-1617), and include surfactant protein-D (accession number: P35247), surfactant protein-A1 (active protein). Session Number: Q8IWL2), Surfactant Protein-A2 (Accession Number: Q8IWL1), Mannan Binding Protein-C (Accession Number: P11226), Liver Collectin 1 (Accession Number: Q9Y6Z7), Placenta Collectin 1 (Accession Number: Q5KU26), or collectin 11 (accession number: Q9BWP8). Similar to the CRD, the coiled coil region (neck region) may be selected from the collectins. However, it should be understood that the coiled-coil (neck region) and the CRD need not be from the same collectin.

  The collectin trimerization domain as described herein may be from a different species than the TNF superfamily cytokine or its receptor binding domain as described herein. Alternatively, the collectin trimerization domain as described herein may be derived from the same species as the TNF superfamily cytokine or receptor binding domain described herein. In a preferred embodiment, the collectin domain as described herein is human and the TNF superfamily cytokine or receptor binding domain thereof as described herein is human.

  The CRD may comprise a variant, for example a variant of surfactant protein-D or collectin-11, which does not bind to mannose. Such variants can be identified by methods known to those skilled in the art, such as the method disclosed in Crouch et al. (J Biol Chem, 2006, 281 (26): 18008-18014). The collectin trimerization domain (ii) may further comprise a variant comprising at least one amino acid substitution as described herein and can be made as described herein. Such amino acid substitutions may modify the collectin trimerization domain binding to its ligand mannose and are described herein when used in therapy and / or as a pharmaceutical composition. May change the clearance rate of such fusion proteins. The modification can result in reduced or no binding to mannose, resulting in a low clearance rate. Such a modification is achieved, for example, by an amino acid substitution acting on the amino acid at position F355 of human surfactant protein-D of SEQ ID NO: 21, in particular by the amino acid substitutions F355A, F355S, F355T, F355E, F355D, F355K, or F355R. obtain. Particularly preferred is F355D substitution. Alternatively, the modification can result in enhanced binding to mannose and high clearance rates. Such a modification can be achieved, for example, by an amino acid substitution acting on the amino acid at position F355 of human surfactant protein-D of SEQ ID NO: 21, in particular the amino acid substitution F355L, F355Y or F355W.

  The neck region as used herein may include a coiled coil structure. This neck may be located near the CRD of the collectin, or in certain cases, away from the CRD.

  In a preferred embodiment, the collectin trimerization domain comprises surfactant protein-D neck and carbohydrate binding domain (CRD) domains, particularly amino acids 217-375, 218-derived from human surfactant protein-D of SEQ ID NO: 21. 375, 219-375, 220-375, 221-375, 222-375, 223-375, 224-375, 225-375. In another preferred embodiment, the collectin trimerization domain is a surfactant protein-D neck domain, particularly amino acids 217-257, 218-257, 219-257, 220, derived from human surfactant protein-D of SEQ ID NO: 21. -257, 221-257, 222-257, 223-257, 224-257 or 225-257. In another preferred embodiment, the collectin trimerization domain is a neck and carbohydrate binding domain (CRD) domain of collectin-11, particularly amino acids 110-271, 116-271 of human collectin-11 of SEQ ID NO: 22 or 121-271. In another preferred embodiment, the collectin trimerization domain is a neck domain of collectin-11, in particular the amino acid of human collectin-11 of SEQ ID NO: 22

including.

A flexible linker element is disposed between the collectin trimerization domain and the effector polypeptide polypeptide as described herein. This flexible linker element preferably has a length of 25 amino acids or less. In certain embodiments, the linker element is 3-30 amino acids long, in particular 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 , 19, 20, 21, 22, 23, 24, 25, 26, 27, 28 or 30 amino acids in length. In one embodiment, the length of the linker element is 5-25 amino acids, 8-20 amino acids or 10-20 amino acids. More preferably, the linker length is 9-15 amino acids. In general, the linker element used herein can be composed of any known amino acid or artificial amino acid derivative. In certain embodiments, the linker element is constructed of small, hydrophobic, uncharged amino acids. In general, a linker element according to the present invention may comprise an amino acid selected from G, S, A and T. The linker element is preferably a glycine / serine linker, ie a peptide linker consisting essentially of the amino acids glycine and serine. In a particularly preferred embodiment, the linker is the amino acid sequence (GSS) _a (SSG) _b (GSG) _c , wherein a, b, c are 0, 1, 2, 3, 4, 5 or 6, respectively. ). When a TNF superfamily cytokine or its receptor binding domain is pre-terminated by G (eg, human TRAIL (SEQ ID NO: 10)), such G is a linker sequence (GSS) _a (SSG) _b (GSG) It will be apparent to one skilled in the art that the first G of the linker in _c can be formed. In principle, the building block of the linker element may be composed of 1, 2, 3, 4 or more amino acids, and the linkers useful herein are 3 It should be understood that the invention is not limited to linkers made of the basic unit of elements. In general, a linker element as used herein may be composed of a basic unit or a series of amino acids.

  An effector polypeptide as used herein is an extracellular polypeptide or a fragment of a protein that includes the extracellular portion of such a protein. In general, a protein or fragment thereof can be used as an effector polypeptide. The effector polypeptide may be a polypeptide of any origin and may be selected from, for example, a mammalian polypeptide, a vertebrate polypeptide, an insect polypeptide, a bacterial polypeptide, a plant polypeptide, a yeast polypeptide. Mammalian polypeptides can include mouse, rat, human, horse, goat, dog, rabbit, cat, sheep, hamster, donkey, monkey and other polypeptides. In certain embodiments, the polypeptide is a human polypeptide or a humanized polypeptide. Such an extracellular protein or extracellular portion of the protein can be selected from, for example, a cell surface protein, a secreted extracellular protein, an extracellular region of a transmembrane protein, and the like. In a particular embodiment of the invention, the effector polypeptide is selected from the group consisting of a TNF superfamily cytokine, its receptor binding domain, a receptor for the cytokine and / or an antibody or fragment thereof.

  In one embodiment, the effector polypeptide is a TNF superfamily cytokine or a receptor binding domain thereof. Preferably, the cytokine is a mammalian, particularly human cytokine, or receptor binding domain thereof (including allelic variants and / or derivatives thereof). In addition, TNF cytokines can bind to the corresponding cytokine receptor and preferably can activate the receptor (which can cause apoptotic or proliferative activity) that receptor. A binding domain is preferred. Cytokines may be selected, for example, from members of the TNF superfamily, such as human TNFSF-1 to -18 as shown in Table 1, preferably

Or they can be selected from their receptor binding domains. Preferred receptor binding domains for each protein are shown in Table 1 (NH ₂ -aa to COOH-aa), eg, amino acids 59-205 or 60-205 of LTA (SEQ ID NO: 1), TNFα (SEQ ID NO: 2) Amino acids 86-233, LTB (SEQ ID NO: 3) amino acids 82-244 or 86-244, OX40L (SEQ ID NO: 4) amino acids 52-183 or 55-183, CD40L (SEQ ID NO: 5) amino acids 112-261 or 117-261, amino acids 51-193 or 56-193 of CD27L (SEQ ID NO: 7), amino acids 97-234, 98-234 or 102-234 of CD30L (SEQ ID NO: 8), amino acids 86- of CD137L (SEQ ID NO: 9) 254, amino acids 161-317 of RANKL (SEQ ID NO: 11), TW Amino acids 103-249, 104-249 or 105-249 of EAK (SEQ ID NO: 12), amino acids 112-247 or 113-247 of APRIL1 (SEQ ID NO: 13), amino acids 112-250 or 113- of APRIL2 (SEQ ID NO: 14) 250, amino acids 140-285 of BAFF (SEQ ID NO: 15), amino acids 91-240 of LIGHT (SEQ ID NO: 16), amino acids 91-251 or 93-251 of TL1A (SEQ ID NO: 17), amino acids of GITRL (SEQ ID NO: 18) 52 to 177, amino acids 245 to 391 of EDA-A1 (SEQ ID NO: 19), and amino acids 245 to 389 of EDA-A2 (SEQ ID NO: 20).

  In one embodiment, the effector polypeptide may be IL4R-alpha (accession number P24394).

  More preferably, the TNF superfamily cytokine or its receptor binding domain is selected from CD95L or TRAIL, or their receptor binding domains. In a particularly preferred embodiment, the TNF superfamily cytokine or its receptor binding domain comprises the extracellular portion of a TNF cytokine comprising a receptor binding domain without a membrane located domain.

  In a preferred embodiment, the TNF superfamily cytokine or its receptor binding domain of the fusion protein is selected from human CD95L (SEQ ID NO: 6), in particular amino acids 142-281 or 144-281 of human CD95L.

  In a further preferred embodiment, the TNF superfamily cytokine of the fusion protein or its receptor binding domain is human TRAIL (SEQ ID NO: 10), in particular amino acids 95-281, 116-281, 117-281, 118- of human TRAIL. 281, 119-281 or 120-281. In another preferred embodiment, human TRAIL comprises any amino acid from amino acid 95 to amino acid 281 of SEQ ID NO: 10 as the starting amino acid.

  In a further preferred embodiment of the invention, the cytokine of the TNF superfamily of fusion proteins as described herein or a receptor binding domain thereof is TRAIL-receptor 1 (TRAILR1) and / or TRAIL-receptor 2 It includes a variant of the TNF superfamily cytokine or its receptor binding domain that binds to (TRAILR2) and / or activates them. The binding and / or activity of this variant can be determined, for example, by assays such as those disclosed herein (eg, in the Examples or van der Slot et al. (PNAS, 2006, 103: 8634-8639), Kelley et al. (By the assay disclosed in J. Biol. Chem., 2005, 280: 2205-2215) or MacFarlane et al. (Cancer Res., 2005, 65: 1125-111270).

  The mutants may be made by any technique, which is known to those skilled in the art (eg, and der Sloot et al. (PNAS, 2006, 103: 8634-8639), Kelley et al. (J. Biol. Chem. , 2005, 280: 2205-2215) or MacFarlane et al. (Technology disclosed in Cancer Res., 2005, 65: 112125-11270), any type of structural mutation (eg, amino acid substitution, deletion, duplication). And / or insertion). A preferred embodiment is the generation of substitution. The substitution may affect at least one amino acid of a TNF superfamily cytokine or its receptor binding domain as described herein. In preferred embodiments, the substitution may affect at least one amino acid of TRAIL (eg, human TRAIL (eg, SEQ ID NO: 10)). Preferred substitutions in this regard are the following of human TRAIL of SEQ ID NO: 10:

Affects at least one of the amino acids. Preferred amino acid substitutions of human TRAIL of SEQ ID NO: 10 are:

At least one of the substitutions.

  Amino acid substitutions can affect the binding and / or activity of TRAIL (eg, human TRAIL) to either TRAILR1 or TRAILR2. Alternatively, amino acid substitutions can affect TRAIL (eg, human TRAIL) binding to and / or activity on both TRAILR1 and TRAILR2. The binding and / or activity of TRAILR1 and / or TRAILR2 can be positively influenced (ie, stronger, more selective binding or more specific binding and / or greater activation of the receptor). Alternatively, the binding and / or activity of TRAILR1 and / or TRAILR2 can be negatively affected (ie, weaker, less selective binding or less specific binding and / or lower activation or receptor of the receptor) No activation).

  Examples of variants of TRAIL having amino acid substitutions that affect both the binding and / or activity of both TRAILR1 and TRAILR2 can be found in, for example, Table 1 of MacFarlane et al. (See above) and the following 2 amino acids of SEQ ID NO: 10 Human TRAIL variants with substitutions, Y213W and S215D, or the following single amino acid substitution Y189A may be included.

  Examples of variants of TRAIL having amino acid substitutions that affect the binding and / or activity of TRAILR1 can be found in, for example, Table 1 of MacFarlane et al. (See above), the following four amino acid substitutions of SEQ ID NO: 10, K201R, Y213W and S215D, or human TRAIL variants with the following five amino acid substitutions, Q193S, N199V, K201R, Y213W and S215D, can be included, or can be found in Table 2 of Kelley et al. (See above): Human TRAIL variants with 6 amino acid substitutions, Y213W, S215D, Y189A, Q193S, N199V and K201R or Y213W, S215D, Y189A, Q193S, N199R and K201R may be included.

  Examples of variants of TRAIL having amino acid substitutions that affect TRAILR2 binding and / or activity are found, for example, in Table 1 of MacFarlane et al. (See above) or in Table 2 of Kelley et al. (See above). Human TRAIL variants having the following six amino acid substitutions of number 14, Y189Q, R191K, Q193R, H264R, I266L and D267Q, can be included, or can be found in Table 2 of van der Slot et al. Human TRAIL variants with a single amino acid substitution D269H, the following two amino acid substitutions, D269H and E195R or D269H and T214R may be included.

  In a further preferred embodiment, the cytokine part of the fusion protein is derived from human LIGHT (SEQ ID NO: 16), in particular from amino acids 91 to 240 of SEQ ID NO: 16. LIGHT is a member of the TNF superfamily and its receptors have been identified as lymphotoxin-beta receptor (LTbetaR) and herpesvirus entry mediator (HVEM) / ATAR / TR2 Both of these lack the cytoplasmic sequence termed the “death domain”. Since LIGHT associates with LTbetaR and HVEM as cell receptors, it is expected to have a biological function similar to that of LTalpha and LTbeta. As previous studies have shown, LIGHT induces cell death of HT-29 cells, as does LT alpha (Harrop, J. Aetal. Et al. (1998) J. Biol. Chem. 273, 27548-27556. Zhai, Y. et al. (1998) J. Clin. Invest. 102, 1142-1151 and Rooney, IA et al. (2000) J. Biol. Chem. 275, 14307-14315), development into smooth muscle cells. It causes growth arrest in RD cells after a genetic change and stimulates secretion of IL-8 and RANTES from the cells (Hikichi, Y. et al. (2001) Biochem. Biophys. Res. Commun. 289, 670-677). It has also been reported that LIGHT is one of the CD28-independent costimulatory molecules in T cells (Tamada, K. et al. (2000) J. Immunol. 164, 4105-4110).

  In one embodiment, the TNF superfamily cytokine is RANK-L. Human RANK ligand (RANK-L) is a member of the tumor necrosis factor superfamily of proteins known to be important regulators of the immune system, bone development and homeostasis (Anderson et al., Nature 390: 175-179, 1997). . In addition, this ligand includes tumor necrosis factor-related activation-inducing cytokine (TRANCE) (Wong et al., J. Exp. Med. 186: 2075, 1997), osteoclast differentiation inhibitory factor ligand (OPGL) (Lacey et al., Cell). 93: 165, 1998), and osteoclast differentiation factor (ODF) (Yasuda et al., Proc. Natl. Acad. Sci. 95: 3597, 1998). Tumor necrosis family members mediate various and sometimes opposite biological responses such as proliferation, apoptosis, cell survival and differentiation.

  In a further embodiment, the TNF superfamily cytokine is TWEAK. Recently, it was observed that picomolar concentrations of TWEAK promote normal endothelial cell proliferation and that TWEAK induces angiogenesis in an in vivo rat corneal model (Lynch et al., 1999, The Journal of Biological Chemistry). , 274 (13) pp. 8455-8459), TWEAK (TNFLSF member 12) was found to be a direct and potent inducer of angiogenesis (vascular formation and growth).

  In an even more preferred embodiment, the cytokine portion of the fusion protein is human APRIL (SEQ ID NO: 13 or 14), in particular amino acids 112-247 or 113-247 of SEQ ID NO: 13, or 112-250 or 113 of SEQ ID NO: 14. Derived from ~ 250. APRIL (TNFLS member 13) is a trimeric cytokine that binds to the receptors TNFRSF13B and TNFRSF17. Recently, it has been shown that the addition of recombinant APRIL to various tumor cells stimulates their proliferation and therefore APRIL may be involved in the regulation of tumor cell growth (Hahne M Et al., 1998; J. Exp. Med. 188: 1185-1190). It has also been suggested that APRIL may be involved in monocyte / macrophage mediated immune processes.

  The TNF superfamily cytokine used herein as an effector polypeptide may be cleaved N-terminally in certain embodiments. Cleavage as used in this respect shall refer to the deletion of all amino acids in the stalk region or the deletion of some amino acids in the stalk region.

  TNF superfamily proteins are tethered to the membrane via the 15-30 amino acid N-terminal portion (so-called stalk region). The stalk region contributes to trimerization and provides a certain distance to the cell membrane. However, the stalk region does not form part of the receptor binding domain (RBD).

  Importantly, RBD is characterized by a unique localization of its N- and C-terminal amino acids. The amino acids are located immediately adjacent to the trimer axis and are centered on the trimer axis. The first N-terminal amino acid of RBD forms an antiparallel β chain with the C-terminal amino acid of RBD (FIGS. 36 and 38).

  Therefore, the antiparallel beta chain of RBD forms an interface with the cell membrane, which is connected to the cell membrane via amino acids in the stalk region and is anchored in the cell membrane. In certain embodiments of the invention, the stalk region of the TNF superfamily cytokine used as an effector polypeptide is deleted. This has the unique advantage that steric hindrance between the stalk region and the linker collectin can be avoided. Otherwise, the linker connecting the C-terminus of one of the TNF-SF-RBD domains to the N-terminus of the collectin domain is sterically hindered by stalk, thereby causing instability and / or aggregate formation. It can happen.

  In certain embodiments, cleaving the TNF superfamily cytokine by deletion of the N-terminal stalk region is a prerequisite for C-terminal positioning of the collectin trimerization domain. In certain embodiments, it is particularly preferred that the collectin-based trimer fusion protein comprises a receptor binding domain of a TNF-SF cytokine lacking any amino acids from the stalk region.

  At least one functional binding to each cytokine receptor is a collectin-based trimeric TNF-SF fusion polypeptide, where the effector TNF superfamily cytokine is located at the N-terminus of the collectin trimerization domain The ability to form ordered trimeric structures including sites is critically associated with the deletion of all or part of the stalk region of TNF superfamily cytokines in certain embodiments of the invention.

  In the fusion proteins of the invention as described herein, the collectin trimerization domain is located at the C-terminus of the effector polypeptide. Accordingly, the fusion protein comprises an effector polypeptide as described herein, and a collectin family neck domain and a collectin family CRD domain (eg, the neck domain of surfactant protein D and the CRD and / or neck domain, or collectin- 11 neck domains and a collectin trimerization domain comprising CRD and / or neck domains (both as described herein), wherein the domains of the effector polypeptide Located at the C-terminus. In this embodiment, it is preferred that the collectin trimerization domain comprises a neck domain and a CRD.

The fusion proteins disclosed herein comprise an effector polypeptide at the N-terminus of the collectin trimerization domain. The inventors have found that controlled formation of trimers can only occur if this orientation is met. Only constructs having an effector polypeptide at the N-terminus of the collectin trimerization domain lead to a controlled and stable trimer and have a minimal tendency to form larger aggregates during handling and purification. From the published crystal structure of the neck-CRD portion of recombinant human SP-D (Shrive, AK et al., 2003, J. Mol. Biol. 331: 509-523), each monomer (monomera) The relative positions of the N and C terminal amino acids can be obtained. The amino terminal amino acid (A205 [A225] ^* ) of the trimer is solvent exposed and placed directly close to the top of the parallel coiled-coil structure, while the trimer (trimere) ) C-terminal (F355 [F375] ^* ) amino acid side chains are buried in the CRD domain and are placed in symmetrically spaced triangular positions. The COOH group of F355 [F375] ^* forms an isosceles triangle having a side length of approximately 23 angstroms. Importantly, the protein core of the neck-CRD trimer is within this triangle, and therefore only radial space relative to the neck-CRD core is accessible to the fusion partner.

  From the relative position to each other, the N- and C-terminal amino acids in the neck-CRD part of the SP-D trimer can be concluded not to be equivalent for use as a seam for the construction of the fusion protein.

As a disadvantage in the case of TNF-SF protein, a C-terminal fusion comprising an extracellular region with a stalk region (which means fusing the TNF-SF fragment to the C-terminal amino acid of the neck-CRD; F355 [F375] ^* ) can give rise to macromolecular aggregates due to the trimerizing force of the TNF-SF fragment itself.
^{* The} numbers presented in [] refer to SEQ ID NO: 21.

  Thus, fusion of a TNF-SF-RBD domain with its endogenous stalk C-terminus to the neck-CRD construct will not yield a defined trimeric assembly. Thus, in a particular embodiment of the invention, a truncated polypeptide of TNF-SF-RBD (TNF superfamily receptor binding domain) lacking a stalk region is used as an effector polypeptide.

  In a preferred embodiment, the fusion protein is TRAIL, in particular human TRAIL or its receptor binding domain or a variant of TRAIL as described herein, preferably 95-281 of human TRAIL (SEQ ID NO: 10). 116-281, 117-281, 118-281, 119-281 or 120-281, and collectin trimerization domains or variants thereof as described herein, in particular, CRD of surfactant protein-D And the neck domain, preferably amino acids 217-375, 218-375, 219-375, 220-375, 221-375, 222-375, 223-375, 224-375 of human surfactant protein-D of SEQ ID NO: 21, 225-375 (where Collectin trimerization domain comprises TRAIL or are located at the C-terminus of the variant TRAIL) as described herein. A preferred fusion protein in this regard is SEQ ID NO: 26 or 27. Alternatively, the fusion protein can be a linker as described herein, eg, amino acid sequence (GSS) a (SSG) b (GSG) c, where a, b, c are 0, 1, A linker having (2, 3, 4, 5 or 6). Preferably, the linker has a length of 9-15 amino acids.

  In a preferred embodiment, the fusion protein is TRAIL, in particular human TRAIL or its receptor binding domain or a variant of TRAIL as described herein, preferably 95-281 of human TRAIL (SEQ ID NO: 10). , 116-281, 117-281, 118-281, 119-281 or 120-281, and collectin trimerization domains or variants thereof as described herein, in particular, the neck of surfactant protein-D Domain, preferably amino acids 217-257, 218-257, 219-257, 220-257, 221-257, 222-257, 223-257, 224-257 or 225 of human surfactant protein-D of SEQ ID NO: 21 257 (where the collection Emissions trimerization domain comprises TRAIL or are located at the C-terminus of the variant TRAIL) as described herein. A preferred fusion protein in this regard is SEQ ID NO: 28. Alternatively, the fusion protein can be a linker as described herein, eg, amino acid sequence (GSS) a (SSG) b (GSG) c, where a, b, c are 0, 1, A linker having (2, 3, 4, 5 or 6).

  In another preferred embodiment, the fusion protein is TRAIL, in particular human TRAIL or its receptor binding domain or a variant of TRAIL as described herein, preferably 95 of human TRAIL (SEQ ID NO: 10). ~ 281, 116-281, 117-281, 118-281, 119-281 or 120-281, and a collectin trimerization domain or variant thereof as described herein, in particular of collectin-11 CRD and neck domain, preferably amino acids 110-271, 116-271 or 121-271 of human collectin-11 of SEQ ID NO: 22, wherein the collectin trimerization domain is as described herein Located at the C-terminus of TRAIL or mutant TRAIL) . Preferred fusion proteins in this regard are SEQ ID NOs: 29 and 30. Alternatively, the fusion protein can be a linker as described herein, eg, amino acid sequence (GSS) a (SSG) b (GSG) c, where a, b, c are 0, 1, A linker having (2, 3, 4, 5 or 6). Preferably, the linker has a length of 9-15 amino acids.

  In another preferred embodiment, the fusion protein is of TRAIL, in particular human TRAIL or its receptor binding domain, or a variant of TRAIL as described herein, preferably human TRAIL (SEQ ID NO: 10). 95-281, 116-281, 117-281, 118-281, 119-281 or 120-281, and collectin trimerization domains or variants thereof as described herein, in particular, collectin-11 Neck domain, preferably the amino acid of human collectin-11 of SEQ ID NO: 22

Wherein the collectin trimerization domain is located at the C-terminus of TRAIL or mutant TRAIL as described herein. A preferred fusion protein in this regard is SEQ ID NO: 31. Alternatively, the fusion protein can be a linker as described herein, eg, amino acid sequence (GSS) a (SSG) b (GSG) c, where a, b, c are 0, 1, A linker having (2, 3, 4, 5 or 6).

  In another preferred embodiment, the fusion protein is CD95L, in particular human CD95L or its receptor binding domain as described herein, eg, amino acids 21-160 of SEQ ID NO: 40, and human SP-D. It includes a collectin trimerization domain comprising a neck domain and a CRD (eg, amino acids 172 to 209 and 210 to 327 of SEQ ID NO: 40, respectively) or variants thereof as described herein. The fusion protein may comprise a linker (eg, a flexible linker), more preferably a glycine / serine linker as described herein having a length of preferably 9-15 amino acids. A preferred fusion protein in this regard comprises SEQ ID NO: 40, particularly amino acids 21-327 of SEQ ID NO: 40.

  In another preferred embodiment, the fusion protein is LIGHT, in particular human LIGHT or its receptor binding domain as described herein, preferably amino acids 21-170 of SEQ ID NO: 41, and human SP-D. A collectin trimerization domain comprising a neck domain and a CRD (eg, amino acids 182-219 and 220-337 of SEQ ID NO: 41, respectively) or a variant thereof as described herein. The cytokine and collectin domains are connected by a linker (eg, a glycine / serine linker as described herein, preferably having a length of 9-15 amino acids). A preferred fusion protein in this regard comprises SEQ ID NO: 41, particularly amino acids 21-327 of SEQ ID NO: 41.

  In another preferred embodiment, the fusion protein is TRAIL, in particular human TRAIL or its receptor binding domain or a variant of TRAIL as described herein (eg, amino acids of SEQ ID NO: 43 (wild type TRAIL)). 21-181, amino acids 21-181 of SEQ ID NO: 47 (TRAILR1mut) or amino acids 21-181 of SEQ ID NO: 48 (TRAILR2mut)). Further, the fusion protein can be a collectin trimerization domain selected from a neck domain of human SP-D and optionally a CRD (eg, amino acids 193 to 230 and 231 to 384 of SEQ ID NO: 43, respectively); And variants thereof as described in (eg, variants as shown in SEQ ID NO: 49 or 50). The fusion polypeptide includes both the human SP-D neck region and the CRD. The cytokine and collectin domains are connected by a linker (eg, a glycine / serine linker as described herein). Preferably, the linker has a length of 9-15 amino acids.

  Preferred fusion proteins in this regard are (i) SEQ ID NO: 43, particularly amino acids 21-348 of SEQ ID NO: 43, (ii) SEQ ID NO: 44, particularly amino acids 21-230 of SEQ ID NO: 44, (iii) SEQ ID NO: 47 In particular, amino acids 21 to 348 of SEQ ID NO: 47, (iv) SEQ ID NO: 48, in particular amino acids 21 to 348 of SEQ ID NO: 48, (v) SEQ ID NO: 49, in particular amino acids 21 to 348 of SEQ ID NO: 49, or ( vi) comprises amino acids 21 to 348 of SEQ ID NO: 50, in particular SEQ ID NO: 50;

  In another preferred embodiment, the fusion protein is TRAIL, in particular human TRAIL or its receptor binding domain or a variant of TRAIL as described herein above, and the neck domain of human collectin 11 and human collectin It contains 11 CRDs (eg, amino acids 193-224 and 225-347 of SEQ ID NO: 45, respectively), a collectin trimerization domain. The cytokine and collectin domains are connected by a linker (eg, a glycine / serine linker as described herein above, preferably having a length of 9-15 amino acids). Preferred fusion proteins in this regard include SEQ ID NO: 45 and SEQ ID NO: 46, particularly amino acids 21-347 of SEQ ID NO: 45 or amino acids 21-229 of SEQ ID NO: 46.

  In another preferred embodiment, the fusion protein is APRIL, particularly human APRIL or a receptor binding domain thereof as described herein (eg, amino acids 21-158 of SEQ ID NO: 51) and as described herein. Collectin trimerization domains such as those described above, particularly the neck domain and CRD of human SP-D or variants thereof as described herein (eg, amino acids 170-207 and 208 of SEQ ID NO: 51, respectively). ~ 325). The cytokine and collectin domains are connected by a linker (eg, a glycine / serine linker as described herein, preferably having a length of 9-15 amino acids). A preferred fusion protein in this regard comprises SEQ ID NO: 51, particularly amino acids 21-325 of SEQ ID NO: 51.

  In certain embodiments, the effector polypeptide is an antibody. Antibodies as used herein can include conventional wild type antibodies as well as modified and altered antibodies. Such modifications can include the production of single chain antibodies, antibody fragments, humanized antibodies and the like. In general, an antibody or antigen-binding amino acid sequence (complementarity determining region, CDR) can be an antibody or CDR from any species, as well as a recombinantly produced antibody or CDR.

  The antibody can be a complete antibody from human class A, E, D, M or G, preferably IgG, or an antigen-binding fragment thereof. Preferably, the antibody or antigen-binding fragment thereof is a chimeric antibody or humanized antibody having a human constant region, preferably an IgG1, IgG2, IgG3 or IgG4 subclass or a combination thereof, or a gene having altered effector function An engineered variant. Effector functions can be antibody-dependent cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC) (may by). A genetically engineered Fc portion may also have altered pharmacokinetic properties. Furthermore, fully human antibodies or antigen binding fragments thereof are preferred. More preferably, the antibody is a humanized or human antibody or fragment thereof additionally comprising a human or substantially human framework region. Also preferred are antibody fragments (eg, bivalent or monovalent antibody fragments such as F (ab) 2 or Fab fragments). On the other hand, antibodies are recombinant antibodies (eg, single chain antibodies or fragments thereof (eg, scFv fragments (Fv fragments are connected by any amino acid sequence, preferably small polar amino acids, more preferably glycine and And / or composed of serine and / or threonine))). Furthermore, the recombinant antibody can be a multimeric scFv assembly (eg, two or more scFv-fragments linked together on a single polypeptide).

  In certain embodiments, antibodies, single chain antibodies, or fragments of antibodies or single chain antibodies are generated against a TNF superfamily cytokine, its receptor binding domain, or a receptor for a cytokine. In a particular embodiment, the antibody is raised against IL4R-alpha.

  In certain embodiments, the antibody is an antibody or antibody fragment, eg, a chimeric or humanized antibody derived from mouse antibody X2 / 45 (Tony et al., 1994) produced by hybridoma cell line DSM ACC2882. The hybridoma cell line DSM ACC2882 was deposited in Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSZM), on the date of the microbiology of Mascheroder Weg 1b, 38124 Braunschweig, Germany. In certain special embodiments, the antibody or antibody fragment recognizes the human IL4Ra receptor extracellular domain. In a further embodiment, the heavy and light chain complementarity determining regions of mouse X2 / 45 have been used to generate human or humanized antibodies or recombinant humanized antibody fragments that recognize human IL4Ra. In a preferred embodiment, a humanized scFv antibody fragment as shown at positions 21-264 of SEQ ID NO: 52 is used as an effector polypeptide.

  A fusion protein as described herein may additionally comprise an N-terminal signal peptide domain, thereby allowing processing (eg, extracellular secretion) in a suitable host cell. Preferably, the N-terminal signal peptide domain contains a cleavage site for a protease, such as a signal peptidase, and thus can be removed after or during expression to yield a mature protein. In a preferred embodiment, the N-terminal signal peptide domain comprises the sequence of SEQ ID NO: 23, SEQ ID NO: 24 or SEQ ID NO: 25.

  In addition, the fusion protein may include a recognition / purification domain, such as a Strep-tag domain and / or a poly-His domain (which may be located at the N-terminus or C-terminus) (may comprises).

  The fusion protein is, for example, 1-50, preferably 10-30 amino acids long, can include and / or be connected to a recognition / purification domain as described herein, C Additional end flexible elements may be included.

  In certain embodiments, the fusion protein may additionally comprise an antibody, a single chain antibody, or a fragment of an antibody or single chain antibody fused to the C-terminus of the CRD. Such antibodies can be useful in certain embodiments, for example, to target a fusion protein to a particular molecule or site. Such constructs may also provide an opportunity to make fusion polypeptides having at least two different antigen specificities.

A further aspect of the invention relates to a nucleic acid molecule encoding a fusion protein as described herein. The nucleic acid molecule can be a DNA molecule (eg, a double-stranded or single-stranded DNA molecule) or an RNA molecule. This nucleic acid molecule can be a signal sequence as described herein, or other heterologous amino acid moiety for secretion or purification (preferably the N and / or C of the fusion protein as described herein. May encode a fusion protein or precursor thereof (eg, a proform or preproform of the fusion protein). The nucleic acid molecule encodes a fusion protein in which a heterologous amino acid moiety can be linked to the first and / or second domains via a protease cleavage site (eg, _a factor cleavage site for factor Xa, thrombin or IgA). May be.

  The nucleic acid molecule may be operably linked to an expression control sequence (eg, an expression control sequence that allows the nucleic acid molecule to be expressed in a desired host cell). The nucleic acid molecule may be placed in a vector (eg, a plasmid, bacteriophage, viral vector, chromosomal integration vector, etc.). Examples of suitable expression control sequences and vectors include, for example,

Or even listed in their latest version.

  A variety of expression vector / host cell systems may be used to express the nucleic acid sequence encoding the fusion protein of the invention. Suitable host cells include prokaryotic cells such as bacteria (eg E. coli), eukaryotic host cells such as yeast cells, insect cells, plant cells or animal cells, preferably mammalian cells, more preferably human cells. For example, but not limited to. Nucleic acid molecules encoding fusion proteins as described herein are suitable host cells (eg, E. coli, yeast cells, plant cells, insect cells, animal cells (eg, mammalian cells or humans). The expression in the cell)) may be optimized taking into account its codon usage.

  The present invention further relates to non-human organisms (eg, mice or rats) that have been transformed or transfected with nucleic acid molecules as described herein. Such organisms can include knockout organisms created by known genetic transfer methods including homologous recombination. Alternatively, such organisms can include transgenic organisms that contain several copies of a nucleic acid molecule as described herein. The production of transgenic organisms is known in the art.

  Fusion proteins, nucleic acids encoding them, transformed or transfected cells and trimer complexes or oligomers of trimer complexes (all as described herein) can be used as pharmaceuticals, diagnostics and And / or may be used for research applications. For these applications, both the effector polypeptide as described herein and the collectin trimerization domain as described herein may be of the same species to minimize immunological effects. For example, when such a protein is applied to a human, it is preferable to use a fusion protein derived from a human. In addition, an effector polypeptide as described herein to a neck-collectin trimerization domain as described herein (eg, a neck domain from surfactant protein-D or collectin-11). Fusion can lead to rapid clearance. Alternatively, a neck domain and CRD-collectin trimerization domain (eg, surfactant protein-D) as described herein of a TNF superfamily cytokine or receptor binding domain thereof as described herein. Or fusion to the collectin-11-derived neck and CRD domains) may lead to low clearance. Use of a variant of collectin trimerization domain as described herein may alter the clearance rate of the fusion protein in a manner as described herein.

  Further aspects of the invention include at least one fusion protein, nucleic acid encoding it, transformed or transfected cells and trimer complexes or oligomers of trimer complexes (all described herein). As a pharmaceutical agent or a diagnostic composition.

  At least one fusion protein, nucleic acid encoding it, transformed or transfected cells and trimer complexes or oligomers of trimer complexes (all as described herein) Sexual disorders, especially disorders associated with and / or associated with TNF cytokine dysfunction, such as tumors (eg solid tumors or lymphomas), infectious diseases, inflammatory diseases, metabolic diseases, self Treatment (eg, prevention and / or treatment) of a disorder selected from immune disorders (eg, rheumatic and / or arthritic diseases), degenerative diseases (eg, neurodegenerative diseases such as multiple sclerosis), apoptosis-related diseases and transplant rejection. ).

  Fusion proteins, nucleic acid molecules or cells as described herein can be used in the treatment of diseases, for example, neoplastic, inflammatory, infectious, degenerative, hereditary, proliferative and vascular diseases, as well as previous It can be used as a medicament in the prevention and / or treatment of malignant and malignant cancerous conditions, cancer and birth defects, or for the preparation of pharmaceutical compositions.

  In certain embodiments of the invention, the degenerative disease can include various tissue or organ degenerative diseases, such as neurodegenerative diseases, bone and skeletal degenerative diseases, skin, mucosal epithelial degenerative diseases, eg, osteoporosis. ), Osteopenia, Alzheimer's disease and the like.

  In one embodiment, the proliferative disease is a tumor. Tumors include head and neck tumors, tracheal tumors, anogenital tract tumors, gastrointestinal tumors, urological tumors, genital tumors, endocrine tumors, central and peripheral nervous system tumors Tumors, tumors of the skin and its appendages, soft tissue and bone tumors, lymphoid and hematopoietic tumors, and the like. Tumors can include, for example, neoplasms such as benign and malignant tumors, carcinomas, sarcomas, leukemias, lymphomas or dysplasia. In particular embodiments, the tumor is, for example, cancer of the head and neck, cancer of the trachea, cancer of the anogenital tract, cancer of the gastrointestinal tract, cancer of the skin and its appendages, central and peripheral nerves. Cancers of the system, urological cancer, genital cancer, endocrine cancer, soft tissue and bone cancer, hematopoietic and lymphoid cancers.

  In certain embodiments where the effector polypeptide is selected from the group consisting of a TNSF polypeptide or a cytokine receptor polypeptide, the present invention is for use in therapy, for example by proliferative disorders, in particular by dysfunction of TNF cytokines. Disorders caused, associated with and / or associated with them, such as tumors (eg solid tumors or lymphomas), infections, inflammatory diseases, metabolic diseases, autoimmune disorders (eg rheumatism and / or arthritis) Disease), degenerative diseases (eg, neurodegenerative diseases such as multiple sclerosis), apoptosis-related diseases and transplant rejection, as described herein for preparing a pharmaceutical composition Described herein for use of a fusion protein, nucleic acid molecule or cell. Also it relates to fusion proteins, nucleic acid molecules or cells as.

  In certain embodiments where the effector polypeptide is an antibody, a single chain antibody, or a fragment of an antibody or single chain antibody, the present invention is further used for therapy (eg, proliferative disorder, immunological disorder, Autoimmune disorders, inflammatory diseases, lymphomas, infectious diseases, metabolic diseases, autoimmune disorders (eg, rheumatic and / or arthritic diseases), degenerative diseases (eg, neurodegenerative diseases such as multiple sclerosis), apoptosis related diseases And as described herein for using a fusion protein, nucleic acid molecule or cell as described herein for preparing a pharmaceutical composition in the prevention and / or treatment of transplant rejection. Pertains to fusion proteins, nucleic acid molecules or cells.

  The composition may be administered as a monotherapy or as a combination therapy with additional pharmaceutical agents (eg, cytostatic or chemotherapeutic agents, corticosteroids and / or antibiotics). Preferably, the composition is administered with a tumor-selective apoptosis sensitizer and / or a tumor-selective apoptosis-inducing agent as described, for example, in Example 2.8.

  The fusion protein is administered to a subject in need thereof (especially a human patient) in a sufficient dose to treat the specific condition by suitable means. For example, the fusion protein may be formulated as a pharmaceutical composition with a pharmaceutically acceptable carrier, diluent and / or adjuvant. The effect and toxicity of treatment can be measured according to standard protocols. The pharmaceutical composition may be administered systemically (eg, intraperitoneally, intramuscularly or intravenously) or locally (eg, intranasally, subcutaneously or intrathecally). Intravenous administration is preferred.

  The dose of fusion protein administered will, of course, depend on the subject being treated, the subject's weight, the type and severity of the disease, the mode of administration and the judgment of the prescribing physician. For administration of the fusion protein, a daily dose of 0.001 to 100 mg / kg is preferred.

  (References)

(Basic structure of fusion protein)
In the following, the basic structure of the recombinant protein of the present invention is illustrated and shown for a TNF-superfamily cytokine as described herein. This illustration is not intended to limit the general scope of the invention, but gives an overall impression of the components present in the fusion protein.

As a basic structure, the fusion protein contains the following elements:
(I) a. A collectin family carbohydrate recognition domain; and b. Collectin family neck region;]
A collectin family trimerization domain comprising (ii) a linker element; and (iii) an effector polypeptide, wherein the effector polypeptide is located at the N-terminus of the collectin family neck region.

Various fragments of human collectin surfactant protein-D and collectin-11 are contemplated as trimerization domains as described herein.

1.4 Flexible linker element (GSS) a (SSG) b (GSG) c (wherein a, b and c are 0, 1, 2, 3, 4, 5 or 6, respectively)

As described herein, various fragments (eg, receptor binding domains) of TNF superfamily cytokines are contemplated.

(Example)
(1.1 Construction of TNF-SF protein stabilized by collectin-derived trimerization domain located at C-terminal)
Trimerization motifs obtained from human collectin-11 (Col11), collectin-11 “coiled coil” (CC11), human lung surfactant protein-D (SP-D), SP-D “coiled coil” (CCSPD) (Tables 2 and 3), the human receptor binding domain (RBD) of CD95L (“CD95L-RBD”; Glu142-Leu281), human TRAIL-RBD (Gln120-Gly281), human LIGHT-RBD (Glu91-Val240) and It fused to the C-terminus of human APRIL-RBD (Lys113 to Leu250), respectively.

Flexible linker element The linker element was placed in various lengths between TNSF-RBD and the trimerization domain (Table 4).

(1.2 Preparation of expression construct)
A nucleic acid molecule encoding a fusion protein as described herein can be cloned into a suitable vector for expressing the fusion protein. The molecular tools necessary to make such vectors are known to those skilled in the art and include tools, restriction enzymes, vectors, and suitable hosts for growing vectors.

  Step-tag II (amino acid sequence WSHPQFEK) was added to the C-terminus for purification and analysis strategies. This affinity tag was linked to the trimerization domain by a flexible linker element (amino acid sequence PSSSSSSA). To enable expression based on secretion, a signal peptide derived from human Igκ was fused to the N-terminus of the protein. The amino acid sequence of this fusion protein was back translated and their codon usage was optimized for mammalian cell based expression. Gene synthesis was performed by ENTELECHON GmbH (Regensburg, Germany). The final expression cassette was subcloned into the pcDNA4-HisMax-backbone using the unique Hind-III and Not-I sites of the plasmid. All expression cassettes were verified as usual by DNA sequencing.

  Data for the following constructs are presented herein (Tables 5a and 5b).

1.3 Expression and purification of genetically engineered ligands of the TNF superfamily
Hek293T cells grown in DMEM + GlutaMax (GibCo) supplemented with 10% FBS, 100 units / ml penicillin and 100 μg / ml streptomycin were transientized with a plasmid encoding a fusion protein as described herein. Transfected. Cell culture supernatant containing the recombinant protein was collected 3 days after transfection, removed impurities by centrifugation at 300 × g, and then filtered through a 0.22 μm sterile filter. For affinity purification, 4 ml of 50% Streptactin Sepharose (IBA GmbH, Goettingen, Germany) is packed into a 2 ml column and 30 ml phosphate buffered saline, pH 7.4 (PBS; Invitrogen Cat. 10010) or buffer. Equilibrated with W (100 mM Tris-HCl, 150 mM NaCl pH 8.0). The cell culture supernatant was poured onto the column at a flow rate of 2 ml / min at 4 ° C. The column is then washed with PBS or buffer W and the specifically bound protein is stepped by addition of 5 × 2 ml buffer E (PBS or buffer W containing 2.5 mM desthiobiotin, pH 7.4). Eluted. The protein content of the eluted fractions was analyzed by absorption spectroscopy and silver staining SDS-PAGE. The positive fraction was then concentrated by ultrafiltration (Sartorius, Vivaspin, 10,000 Da cutoff) and further analyzed by size exclusion chromatography (SEC).

  SEC was performed on a Superdex 200 column using an Aekta chromatography system (GE-Healthcare). The column was equilibrated with PBS (Invitrogen Cat. 10010) and the concentrated streptactin purified protein was loaded onto the SEC column at a flow rate of 0.5 ml / min. The eluate was monitored by absorbance at 280 nm. The apparent molecular weight of the purified protein was determined based on calibration of a Superdex 200 column using a gel filtration standard protein (Bio-Rad GmbH, Muenchen, Germany).

(1.4 Cell death assay)
To analyze caspase activation, a cell assay with Jurkat A3 permanent human T cell line (cat.no.CRL2570, ATCC) was used. Jurkat cells were grown in flasks using RPMI 1640 medium + GlutaMAX (GibCo) supplemented with 10% FBS (Biochrom), 100 units / ml penicillin and 100 μg / ml streptomycin (GibCo). Prior to the assay, 100,000 cells were seeded in 96 well microtiter plates for each well. Various solutions containing the above proteins with or without cross-linked antibodies were added to the wells (final volume: 200 μl), followed by incubation at 37 ° C. for 3 hours. Cells are lysed by adding 20 μl lysis buffer (250 mM HEPES, 50 mM MgCl ₂ , 10 mM EGTA, 5% Triton-X-100, 100 mM DTT, 10 mM AEBSF, pH 7.5) and the plates are on ice for 30 minutes. Incubated for ~ 2 hours. Apoptosis occurs in parallel with increased caspase activity. Therefore, the extent of apoptosis was measured by using cleavage of the specific caspase substrate Ac-DEVD-AFC (Biomol). For the caspase activity assay, 20 μl of cell lysate was transferred to a black 96 well microtiter plate. After the addition of 80 μl buffer containing 50 mM HEPES, 1% sucrose, 0.1% CHAPS, 50 μM Ac-DEVD-AFC and 25 mM DTT, pH 7.5, the plate is transferred to a Tecan Infinite F500 microtiter plate reader. The increase in fluorescence intensity was monitored (excitation wavelength 400 nm, emission wavelength 505 nm).

  To measure cell death in HT1080 fibrosarcoma cells, HeLa cervical cancer cells and WM35 melanoma cells, 15,000 cells were cultured in RPMI 1640 medium supplemented with 10% FBS (Biochrom) + GlutaMAX (GibCo). Placed in a 96-well plate in the evening. For Colo205 cells, 50,000 cells were left overnight. The next day, cells were stimulated with the indicated ligand and incubated for an additional 18 hours. For HeLa and HT1080 cells, cycloheximide (Sigma) was used at a final concentration of 2.5 μg / ml during stimulation with ligand. Cell death of HT1080, HeLa and WM35 was quantified by staining with buffer KV (0.5% crystal violet, 20% methanol). After staining, the wells were washed with water and air dried. The dye was eluted with methanol and the optical density at 595 nm was measured using an ELISA reader. Viability of Colo205 cells was quantified by MTS assay (Promega).

(1.5 Hepatocyte cytotoxicity assay)
To measure the effect of TRAIL fusion protein, primary human hepatocytes were prepared from healthy donors and cultured in Williams E medium using 25,000 cells per well in a 96 well plate. On the second day, the medium was changed to DMEM-F12 supplemented with 10% FCS, human insulin, Pen / Strep, minimal essential medium (MEM), sodium pyruvate and 10 mM Hepes, and cultured for another day. On day 3, cells were stimulated with various concentrations of the indicated protein in the presence or absence of cross-linking antibody (StrepMabImo, IBA GmbH). To assess the potential hepatotoxic effects of ligand and chemotherapeutic treatment, various concentrations of TRAIL-ASPD_F335D were co-incubated with 5 mM doxorubicin or 5 mM gemcitabine. Cells were incubated for 5 or 24 hours at 37 ° C. and 5% CO ₂ and then lysed for measurement of caspase activity as described in the “Cell Death Assay” column.

(1.6 Streptactin-ELISA)
A 96 well microplate coated with streptactin was used to measure receptor binding to the assembled ligand. Therefore, supernatant, mouse serum or purified protein from transiently transfected HEK293 cells was immobilized on streptactin plates (IBA GmbH) for 1-3 hours in PBS. Samples were diluted in ELISA binding / blocking buffer (PBS, 0.1% Tween-20, 20% SuperBlock T20-PBS (Pierce)). Plates were washed with PBS + 0.1% Tween-20, mouse anti-TRAIL antibody (Pharmingen, clone RIK-2), TRAIL receptor 1-Fc (R & D Systems), TRAIL receptor 2-Fc (R & D Systems), TACI- Fc (R & D
Systems) or HVEM-Fc (R & D Systems) for 1 hour at room temperature. Plates were washed again and Fc protein was detected with anti-human- or anti-mouse-Fc specific peroxidase conjugated antibody (Sigma). The color reaction was performed by the addition of 100 μl TMB substrate (Kem-En-Tec Diagnostics) per well, and after adding 25 μl 25% H ₂ SO ₄ as stop solution, the absorbance at 450 nm and 630 nm was determined by ELISA. Measurement was performed using a reader. Values were calculated as 450 nm-630 nm using MS Excel.

(1.7 Mannan binding assay)
ELISA plates (Nunc Maxisorp) were placed in sterile coating buffer (15 mM Na ₂ CO ₃ , 35 mM NaHCO ₃ , 0.025% NaN ₃ , pH 9.6) at 4 ° C. with 10 μg / well yeast mannan (Sigma). Incubate overnight. Plates first buffer _{BB (20mM Tris, 140mM NaCl,} 5mM CaCl 2, 0.1% BSA and 20% SuperBlock T20-PBS (Pierce )) for 1 h at room temperature, then shown of various concentrations For 90 minutes in buffer BB. The plate was washed with buffer WB (20 mM Tris, 140 mM NaCl, 5 mM CaCl ₂ , 0.05% Tween-20) and detection was performed in buffer BB using streptactin-HRP (IBA GmbH). Plates were washed and developed with TMB substrate (Kem-En-Tec Diagnostics). After adding 25 μl of 25% H ₂ SO ₄ as a stop solution, the absorbance at 450 nm and 630 nm was measured using an ELISA reader. Values were calculated as 450 nm-630 nm using MS Excel.

(1.8 Pharmacokinetics of TRAIL-SPD fusion protein)
Male CD1 mice (Charles River) were intravenously injected with 10 μg of protein dissolved in 300 μl of PBS (Invitrogen). Blood was collected at 0 minutes (before administration), 5 minutes, 30 minutes, 2 hours, 6 hours and 24 hours. Two samples were taken for each time point. Blood samples were processed to obtain serum and stored at -15 ° C. The concentration of TRAIL fusion protein was measured using ELISA as described below (Chapter 1.9) and the half-life was calculated (GraphPad Prism v4.0).

(1.9 ELISA for quantifying TRAIL constructs in mouse serum)
To quantify the concentration of TRAIL protein (from pharmacokinetic studies) in mouse serum, an ELISA method using 96 well microplates was used.

ELISA plates were coated with 2 μg / ml mouse anti-TRAIL (clone RIK-2; Pharmingen) for 1 hour at 37 ° C. Wash with PBS + 0.1% Tween-20 and block the plate with StartingBlock ^™ (Pierce) for 30 minutes at 37 ° C, then add serum, calibration and control samples at 0.2% and 5% concentrations And incubated at 37 ° C. for 1 hour. Calibration samples and control samples are prepared from each TRAIL batch (TRAIL-ASPD or TRAIL-ASPD-F335A or TRAIL-ASPD-F335D) and 0.2% or 5 to them to account for potential matrix effects. % Pooled untreated CD1 mouse serum was supplemented. Control samples (high, medium and low concentrations of TRAIL constructs) were added in a defined assay window as a quality control to ensure the accuracy and accuracy of TRAIL quantification. Plates were washed again and TRAIL constructs containing StrepTag were detected by Strep Tactin-POD (IBA) diluted 1: 1000. All samples and proteins were diluted with ELISA buffer (PBS, 0.1% Tween-20, 5% StartingBlock (Pierce)). The color reaction was initiated after adding 100 μl TMB substrate (Kem-En-Tec Diagnostics) per well. After adding 25 μl of 25% H 2 SO 4 as a stop solution, the absorbance at 450 nm and 630 nm was measured using an ELISA reader. Values were calculated as 450 nm-630 nm using MS Excel.

(2. Results)
(2.1 Characterization of CD95L fusion protein (CD95L-ASPD))
0.5 ml (0.86 mg protein) from CD95L-ASPD affinity purified with Streptactin was loaded onto a Superdex 200 column at a flow rate of 0.5 ml / min using PBS as a running buffer. 0.5 ml fractions were collected (A1-A11 are shown). The main peak retention capacity at 11.92 ml corresponded to 170 kDa as determined from the size exclusion standard. This indicated that the protein was a trimer composed of glycosylated monomers. The calculated molecular weight of the monomeric polypeptide is 38 kDa. Aliquots of fractions A1-A11 were used for SDS-PAGE and caspase activity. Only certain trimer peaks (fractions A7-A10) were used in the final analysis. The results are shown in FIG.

  An aliquot from the size exclusion chromatography of affinity purified CD95L-ASPD was used for silver staining following reducing SDS-PAGE. A band (indicated by the arrow) detected at approximately 40-45 kDa corresponded to CD95L-ASPD. Trimeric species were present in fractions A7-A10. The results are shown in FIG.

  Jurkat cells were incubated with aliquots in final 8-fold dilutions from SEC-derived fractions A1-A15 using affinity purified CD95L-ASPD. After 3 hours of incubation, cells were lysed and caspase activity was measured by a fluorogenic assay. Since Jurkat cells are known to require highly cross-linked ligands, the fraction corresponding to the trimeric peak (fractions A7-A10) induced clear but weak caspase activity in Jurkat. Thus, the aggregated and undefined species in fractions A1-A6 are strong inducers of caspase activation (not used further). Importantly, only certain trimer species (A7-A10) were collected and used for final analysis. The results are shown in FIG.

  Human cancer cell lines HT1080 (A), HeLa (B) or WM35 (C) at the indicated concentrations in the presence or absence of cross-linked antibody (2.5 microgram / ml anti-Step-tag II) Of the purified trimeric CD95L-ASPD. Cells were incubated for 18 hours and cytotoxicity was analyzed by crystal violet staining. As a result, CD95L-ASPD induced cell death in HeLa cervical cancer and HT1080 fibrosarcoma, but not in WM35 melanoma cells. The results are shown in FIG.

  The amino acid sequence of CD95L-ASPD is shown below.

1-20: Secretory signal peptide (Sp; underlined)
21-160: CD95L receptor binding domain 161-171: flexible linker element (A-linker; italic)
172 to 209: Coiled coil “neck” region of human SP-D 210 to 327: C-type lectin domain 328 to 338 of human SP-D: Linker element (GGSPSSSSSSSA)
339-346: Strep-tag II (WSHPQFEK).

(2.2 Characterization of LIGHT-fusion protein (LIGHT-ASPD))
From the affinity-purified LIGHT-ASPD, 0.5 ml (1.56 mg) was loaded onto a Superdex 200 column and separated at 0.5 ml / min using PBS as a running buffer. The major peak detected at 11.96 ml corresponded to a size of 170-180 kDa, indicating that LIGHT-ASPD is a trimer composed of three glycosylated monomers. . Trimeric peaks (fractions A7-A10) were collected and used for final analysis. The inset shows silver stained SDS-PAGE of two independent purified trimer LIGHT-ASPD batches (named 0917 and 0918). The results are shown in FIG.

  Various concentrations (0-10 micrograms) of affinity purified and SEC purified trimer LIGHT-ASPD were used for immobilization via Strep-tagII on Streptactin-coated microplates. LIGHT-ASPD was then detected in an ELISA setup using 100 ng / ml of receptor HVEM and TRAIL receptor 1 Fc fusion protein, respectively. The ELISA signal for HVEM-Fc increased with increasing amount of immobilized ligand, while no signal was detected for TRAIL receptor 1-Fc over the entire range analyzed. There wasn't. This indicated that LIGHT-ASPD is a functional molecule capable of binding to its receptor HVEM. The results are shown in FIG.

  The amino acid sequence of the LIGHT-ASPD fusion protein is shown below.

1-20: Secretory signal peptide (Sp; underlined)
21-170: LIGHT receptor binding domains 171-181: flexible linker element (A-linker; italic)
182-219: Coiled coil “neck” region of human SP-D 220-337: C-type lectin domain 338-348 of human SP-D: linker element (GGSPSSSSSSSA)
349-356: Strep-tag II (WSHPQFEK).

(2.3 Characterization of TRAIL fusion protein)
HEK293 cells were transiently transfected with 24 different expression vectors encoding TRAIL fusion proteins (Table 6).

The supernatant was used for SDS-PAGE and the TRAIL construct was detected by Western blot analysis using an antibody specific for Strep-tag II.

  The specific band detected is indicated by the arrow. The expression intensity was dependent on the type of trimerization motif used in the construction (SPD> 69 / T4 / collectin 11 / CCSPD / CC11), as well as the length of the linker element (A> B> C> D). The results are shown in FIG.

  Jurkat cells, along with supernatant from transiently transfected HEK cells, in the presence of cross-linked antibody (2.5 microgram / ml anti-Strep-tag II) (black bar, anti-Strep-tag II) or non- Incubated in presence (white bar) for 3 hours. The supernatant contains TRAIL fusion proteins with various trimerization motifs (T4, 69, SPD, CCSPD, Col11, CC11) fused via various linker elements (A, B, C and D linkers). Contained. As a negative control, cell supernatant from untransfected cells was used. Jurkat cells were lysed and analyzed for caspase activity by a fluorogenic assay.

  As a result, caspase activity was reduced by the type of linker element used (A> B> C> D) and the used Fold-On. While TRAIL constructs containing collectin-11 or a coiled coil of collectin-11 (CCCol11) were expressed (as shown by Western blot analysis) but were not functional, the SPD-derived fold-on motif was A functional TRAIL ligand was generated. The results are shown in FIG.

  Affinity purified TRAIL-ASPD was subjected to SEC by loading 0.5 ml (0.4 mg protein) onto a Superdex 200 column at 0.5 ml / min using PBS as the running buffer. Protein elution was monitored by absorbance at 280 nm and 0.5 ml fractions were collected. A holding volume of 12.28 ml corresponds to 135-140 kDa determined from the size exclusion standard. This indicated that TRAIL-ASPD is a homotrimer because the calculated molecular weight of the monomeric polypeptide is 40 kDa. Importantly, for all fusion proteins analyzed by SEC consisting of the wild-type TRAIL-RBD sequence, an additional peak at approximately 8 ml was observed corresponding to the inactive and aggregated TRAIL fusion protein. From the collected fractions A1-A14, only the trimer peak (A8-A10) was used for further analysis. The results are shown in FIG.

  Human cancer cell lines HeLa, HT1080, Colo205 or WM35 with the indicated concentration of purified trimeric TRAIL-ASPD in the presence or absence of cross-linked antibody (2.5 microgram / ml anti-Strep-tag II) Incubated for 18 hours in the presence. Cell death was quantified by crystal violet staining (HeLa, WM35 and HT1080) or MTS assay (Colo205). The increase in viability of Colo205 cells at high ligand concentrations appears to be due to the limitations of cross-linked antibodies. The results are shown in FIG.

  Various concentrations (A) or constant concentrations (B) of affinity purified and SEC purified trimeric TRAIL-ASPD were used for immobilization on Streptactin coated 96 well plates. Plates were then incubated with 100,000 Jurkat cells per well for 5 hours at 37 ° C., 5% CO 2 and caspase activity was measured by fluorogenic assay. To analyze the specificity, cells were added after incubating plates (B) for 30 minutes with various concentrations of antagonist anti-TRAIL antibodies (clone RIK-2, Pharmingen) as indicated. The results are shown in FIG.

  HT1080 cells were incubated on the same 96-well plate with the indicated concentrations of purified trimeric TRAIL-ASPD or TRAIL-DSPD. The next day, cell death was quantified by crystal violet staining. Use of the D linker reduced biological activity approximately 4.5-fold, as indicated by EC50 values of 27 ng / ml and 6 ng / ml for TRAIL-DSPD and TRAIL-ASPD, respectively. The results are shown in FIG.

  The nucleic acid and amino acid sequences of the TRAIL fusion polypeptide are shown below.

SEQ ID NO: 42: Sp-TRAIL-ASPD expression cassette endonuclease restriction sites are underlined (HindIII, AAGCTT; BamHI, GGATCC; NotI, GCGGCCGC). The translation initiation codon is bold.

1-20: Secretory signal peptide (Sp; underlined)
21-181: TRAIL receptor binding domains 182-192: flexible linker element (A-linker; italic)
193-230: Coiled coil “neck” region of human SP-D 231-348: C-type lectin domain 349-359 of human SP-D: Linker element (GGSPSSSSSSSA)
360-367: Strep-tag II (WSHPQFEK)

1-20: Secretory signal peptide (Sp; underlined)
21-181: TRAIL receptor binding domains 182-192: flexible linker element (A-linker; italic)
193-230: Coiled coil “neck” region 231-238 of human SP-D: Linker element (PSSSSSSA)
239-246: Strep-tag II (WSHPQFEK)

1-20: Secretory signal peptide (Sp; underlined)
21-181: TRAIL receptor binding domains 182-192: flexible linker element (A-linker; italic)
193-224: Coiled coil “neck” region of human collectin-11 225-347: C-type lectin domain 348-357 of human collectin-11: linker element (GSPSSSSSSA)
358-365: Strep-tag II (WSHPQFEK)

1-20: Secretory signal peptide (underlined)
21-181: TRAIL receptor binding domains 182-193: flexible linker element (A-linker; GSS GSS GSS GSG italic)
194 to 229: Coiled coil “neck” region 230-238 of human collectin 11: Linker element (GPSSSSSSA)
239-246: Strep-tag II (WSHPQFEK).

(2.4 Characterization of receptor selective TRAIL ("mutein") fusion proteins)
HEK293 cells were transiently transfected with expression plasmids encoding various TRAIL receptor selective SPD constructs.

Supernatants were collected 3 days after transfection and aliquots were used for Western blotting using antibodies specific for SDS-PAGE and Strep-tag II. Specific bands were detected at approximately 38 kDa (SPD fusion protein) and 28 kDa (coiled coil SPD fusion protein). The amount of protein expressed depends on the ligand itself (TRAILR1 mutein> TRAILR2 mutein> TRAIL), secondly depends on the ligand length used (A> D), and thirdly, the three used Dependent on the merging motif (SPD> CCSPD). The apparent molecular weight was as expected from the calculated size (40 kDa and 27 kDa for SPD and CCSPD fusion proteins, respectively). The results are shown in FIG.

  The selectivity of TRAIL receptor 1 or TRAIL receptor 2 for SPD / ccSPD and TRAIL, TRAILR1mut and TRAILR2mut fusion proteins was demonstrated by Streptactin-ELISA. Therefore, TRAIL-SPD fusion protein in the supernatant from transiently transfected HEK293 cells was immobilized on a microplate coated with streptactin. Cell supernatant from untransfected cells served as a negative control. The results are shown in FIG. Specifically bound protein was detected by either a constant concentration (A, B) or various concentrations (C, D) of either TRAIL receptor 1-Fc or TRAIL receptor 2-Fc. As shown in (A), the ligand TRAILR1mut variant fused to SPD is detected by TRAIL receptor 1, whereas the ligand TRAILR2mut is not detected. As shown in (B), the ligand TRAILR2mut is preferentially detected by TRAIL receptor 2, while the TRAILR1mut construct and the TRAIL wild type construct are similarly well detected. As shown in C, TRAIL receptor 1-Fc bound to TRAIL-R1mut-ASPD and TRAIL-ASPD as well throughout the receptor titration range, while TRAIL-R2mut-ASPD Not detected. As shown in D, TRAIL receptor 2-Fc bound similarly well to TRAIL-R2mut-ASPD and TRAIL-ASPD over the analyzed receptor measurement range, while the signal for TRAIL-R1mut-ASPD is , Decreased rapidly with decreasing receptor concentration.

  1 microgram / ml affinity purified trimer TRAIL-ASPD, TRAILR1mut-ASPD or TRAILR2mut-ASPD in 100 microliters of PBS via Strep-tagII onto streptactin-coated microplates Used for immobilization. Bound ligand was detected in an ELISA setting using TRAIL receptor 1 (A) and TRAIL receptor 2 (B) Fc fusion proteins. As shown in (A), TRAIL receptor 1 preferentially bound to receptor-selective TRAILR1mut-ASPD compared to TRAILR2mut-ASPD. As shown in (B), TRAIL receptor 2 preferentially bound to TRAILR2mut-ASPD compared to TRAILR1mut-ASPD. In conclusion, constructed TRAIL variants fused to SPD are receptor selective. The results are shown in FIG.

  Affinity purified TRAILR1mut-ASPD was subjected to SEC by filling a Superdex 200 column with 0.5 ml (0.95 mg protein). The results are shown in FIG. The protein was separated at 0.5 ml / min using PBS as a running buffer, and 0.5 ml fractions were collected (fractions A1-A14 are shown). A retention volume of 12.46 ml corresponded to 140-145 kDa as determined by the size exclusion standard. The small peak at 10.83 ml showed some aggregating species, but importantly, no peaks were detected at the beginning of running (8 ml), indicating that one of the wild type TRAIL amino acid sequences This molecule is shown to be much more soluble than the protein containing the part.

  Aliquots from size-exclusion chromatography of affinity purified TRAILR1mut-ASPD were used for silver staining as shown in FIG. 17 after non-reducing (A) or reducing (B) SDS-PAGE. Two bands were detected at 35 and 70 kDa under non-reducing conditions, whereas a single band at 40 kDa (indicated by arrows) was detected under reducing conditions. This indicated the formation of disulfide bridged molecules. The trimer species was present in fractions A8-A11 and was used for later analysis.

  Jurkat cells in the absence of 2.5 micrograms / ml cross-linked antibody (white bars) with aliquots in final 80-fold dilutions from SEC-derived fractions A1-A14 of affinity purified TRAILR1mut-ASPD (white bars) or Incubated in the presence (black bar). The results are shown in FIG. As a negative control, Jurkat cells were incubated with medium alone. After 3 hours of incubation, Jurkat cells were lysed and caspase activity was measured by a fluorogenic assay. Jurkat cells have been shown to primarily express TRAIL receptor 2, so that all fractions induce significant caspase activity even when TRAILR1mit-ASPD is cross-linked by a Strep-tag II specific antibody I did not. This indicated that TRAILR1mut-ASPD did not bind to TRAIL receptor 2.

  Affinity purified TRAILR2mut-ASPD was subjected to size exclusion chromatography by loading 0.5 ml (0.5 mg protein) on a Superdex 200 column as shown in FIG. Protein was separated at 0.5 ml / min using PBS as a running buffer, and 0.5 ml fractions were collected (fractions A1-A14 are shown). A retention capacity of 12.60 ml corresponds to 130-135 kDa determined from the size exclusion standard. This indicated that TRAILR2mut-ASPD was a homotrimer, as calculated from the expected monomer weight of 40 kDa. Importantly, more than 95% was present in the trimer peak fraction and no aggregates were detected. The trimer peak was used for later analysis.

  An aliquot from the size exclusion chromatography of affinity purified TRAILR2mut-ASPD was used for silver staining as shown in FIG. 20 following non-reducing (A) or reducing (B) SDS-PAGE. Under non-reducing conditions, two bands were detected at 35 and 70 kDa, while under reducing conditions, a single band of approximately 40 kDa (indicated by arrows) was detected. This indicated the formation of disulfide bridged molecules. The trimeric species was present in fractions A9-A11 and was used for later analysis.

  The result of the Jurkat cell killing assay using TRAILR2mut-ASPD is shown in FIG. Jurkat cells with aliquots from SEC-derived fractions A1-A14 of affinity purified TRAILR2mut-ASPD in the absence of cross-linked antibody (2.5 microgram / ml anti-Strep-tagII) (white bar) or Incubated in the presence (black bar). Samples were used at a final 640-fold dilution. Cells were lysed after 3 hours of incubation and caspase activity was measured by fluorogenic assay. Jurkat cells have been shown to predominantly express TRAIL receptor 2, which requires a multimerizing ligand type for efficient signal transduction, so that TRAILR2mut-ASPD has caspase activity when cross-linked. Induced. This indicated that TAILR2mut-ASPD is a functional molecule.

  The cytotoxic activity of TRAIL-ASPD, TRAILR1mut-ASPD and TRAILR2mut-ASPD against various human cancer cells is shown in FIG. The indicated cell lines HT1080 (A and B), Hela (C and D) or Colo205 (E and F) can be obtained in the absence (A, C and E) or presence (B , D and F) were treated with various concentrations of purified trimer TRAIL-ASPD, TRAILR1mut-ASPD or TRAILR2mut-ASPD. Cells were incubated with the indicated concentrations of ligand for 18 hours and cell death was quantified by crystal violet staining (HT1080 and HeLa) or MTS assay (Colo205). As a result, the ligand TRAIL-ASPD induced cell death against the three cell lines tested, and TRAILR2mut-ASPD showed excellent cell killing activity. In contrast, TRAILR1mut-ASPD selective for TRAIL receptor 1 was not active against any cell line tested.

  Affinity purified TRAILR2mut-ASPD was concentrated 20-fold into PBS by centrifugation through a 10 kDa membrane to give a 2.5 mg / ml solution. From the concentrate, 0.1 ml was subjected to size exclusion chromatography. As a result, only the trimer peak was detected and no aggregates were detected, indicating that this composition has improved production capacity (FIG. 23). Similar results were achieved for TRAILR1mut-ASPD, where even 5.4 mg / ml concentrated solution showed no signs of aggregation (not shown). In contrast, all tested fusion proteins containing a receptor binding domain composed of a wild type TRAIL sequence showed aggregation with 40% aggregates at concentrations as low as 0.4 mg / ml.

  The amino acid sequence of the receptor selective TRAIL mutein fusion polypeptide is shown below.

1-20: Secretory signal peptide (Sp; underlined)
21-181: TRAILR1mut receptor binding domain 182-192: flexible linker element (A-linker; italic)
193-230: Coiled coil “neck” region of human SP-D 231-348: C-type lectin domain 349-359 of human SP-D: Linker element (GGSPSSSSSSSA)
360-367: Strep-tag II (WSHPQFEK)

1-20: Secretory signal peptide (Sp; underlined)
21-181: TRAILR2mut receptor binding domain 182-192: flexible linker element (A-linker; italic)
193-230: Coiled coil “neck” region of human SP-D 231-348: C-type lectin domain 349-359 of human SP-D: Linker element (GGSPSSSSSSSA)
360-367: Strep-tag II (WSHPQFEK).

(2.5 SPD Carbohydrate-Characterization of Variant)
Affinity purified TRAIL-ASPD_F335A was subjected to size exclusion chromatography by packing a Superdex 200 column with 0.5 ml PBS solution (0.4 mg protein) as shown in FIG. Protein was separated at 0.5 ml / min using PBS as running buffer and 0.5 ml fractions were collected (A1-A13 are shown). A retention volume of 12.27 ml corresponds to 135-145 kDa determined from the size exclusion standard. This indicated that TRAIL-ASPD_F335A is a homotrimer, as calculated from the expected monomer weight of 40 kDa. Two additional peaks at 8.32 and 10.68 ml indicated the formation of TRAIL-ASPD_F335A aggregates. Only the trimer peak was used for later analysis.

  From size exclusion chromatography, aliquots taken from fractions A1-A13 were separated by reducing SDS-PAGE and the gel was silver stained (FIG. 25). The band detected at approximately 40 kDa corresponded to a calculated molecular weight of 40 kDa for TRAIL-ASPD_F335A. The positive fractions (A8, A9, A10) corresponding to the SEC run trimer molecule were pooled and used for further analysis.

  The amino acid sequence of the TRAIL-SPD carbohydrate variant fusion protein is shown below.

1-20: Secretory signal peptide (Sp; underlined)
21-181: TRAIL receptor binding domains 182-192: flexible linker element (A-linker; italic)
193-230: Coiled coil “neck” region of human SP-D 231-348: C-type lectin domain of human SP-D (Phe mutation (bold))
349-359: Linker element (GGSPSSSSSSSA)
360-367: Strep-tag II (WSHPQFEK)

1-20: Secretory signal peptide (Sp; underlined)
21-181: TRAIL receptor binding domains 182-192: flexible linker element (A-linker; italic)
193-230: Coiled coil “neck” region of human SP-D 231-348: C-type lectin domain of human SP-D (Asp mutation (bold))
349-359: Linker element (GGSPSSSSSSSA)
360-367: Strep-tag II (WSHPQFEK).

  FIG. 26 shows the cytotoxic effect of TRAIL-ASPD_F335A on human cancer cells. The indicated human cancer cell lines were combined with various concentrations of affinity purified and SEC purified trimer TRAIL-ASPD_F335A in the presence of cross-linked antibody (2.5 microgram / ml anti-Strep-tag II) or non- Incubated overnight in the presence. Cell viability was quantified by crystal violet staining (HT1080, HeLa and WM35) or MTS (Colo205). The increased viability of Colo205 cells at high concentrations of ligand appears to be due to the limitations of cross-linked antibodies.

  Affinity purified TRAIL-ASPD_F335D was subjected to size exclusion chromatography by packing 0.5 ml (0.2 mg protein) into a Superdex 200 column as shown in FIG. Protein was separated at 0.5 ml / min using PBS as a running buffer, and 0.5 ml fractions were collected (fractions A1-A13 are shown). A retention volume of 12.29 ml corresponds to 135-145 kDa determined from the size exclusion standard. This indicated that TRAIL-ASPD_F335D is a homotrimer, as calculated from the expected monomer weight of 40 kDa. The peak at 8.35 corresponded to the inactive TRAIL-ASPD_F335D aggregate typically found for all fusion proteins containing part of the wild type TRAIL amino acid sequence.

  From size exclusion chromatography, aliquots of affinity purified TRAIL-ASPD_F335D from the collected fractions A1-A13 were separated by reducing SDS-PAGE and the gel was silver stained (FIG. 28). The band detected at approximately 40 kDa (indicated by the arrow) corresponded to a calculated molecular weight of 40 kDa for TRAIL-ASPD_F335D. Fractions containing trimer protein (fractions A8-A10) were pooled and used for further analysis.

  Human cancer cell lines HT1080 (A), HeLa (B), WM35 (C) or Colo205 (D) were purified at various concentrations in the presence or absence of cross-linking antibody (anti-Strep-tag II). Incubated overnight with the trimer TRAIL-ASPD_F335D. Cell viability was quantified by crystal violet staining (HT1080, HeLa and WM35) or MTS (Colo205). The data shows that TRAIL-ASPD_F335D can induce cell death in the exemplified cancer cell lines (FIG. 29). The increased viability of Colo205 cells at high concentrations of ligand appears to be due to the limitations of cross-linked antibodies.

(2.6 Analysis of carbohydrate binding characteristics of SPD trimerization motif variants)
Wild-type full-length oligomeric SP-D proteins from several species as well as human SP-D trimer neck + CRD have been shown to bind to several different carbohydrates. In addition, human SP-D neck + CRD has also been shown to exert immunomodulatory effects by acting as a chemotactic factor for immune cells such as neutrophils (Cai et al., 1999, Am J). Physiol Lung Cell Mol Physiol 276: 131-136). Other cells can also be recruited by SP-D. Since the addition of maltose inhibited chemotaxis function, it was shown that the chemotaxis effect of human SP-D neck + CRD depends on the sugar-binding function. Therefore, a ligand for TNFSF having SP-D mediated chemotaxis function may have a superior activity compared to a ligand having a natural amino acid sequence or a construct thereof. For example, in scenarios where cellular effects are desirable, such as cancer treatment, such described ligands may be desirable.

  In addition, ligands where SP-D does not have a carbohydrate function may be desirable in other settings. For human SP-D, a variant in which the amino acid phenylalanine 335 (corresponding to amino acid 355 of SEQ ID NO: 21) has been mutated to alanine has been described (SPD_F335A, Crouch et al., JBC 281: 18008-18014). This mutant showed very weak carbohydrate binding. However, when carbohydrate linkage is not desired, the introduction of charged amino acids (eg, acidic amino acids) may be better than F335A. Therefore, mutant SPD_F335D may be superior to F335A mutant.

  To analyze the binding of TRAIL fusion protein to carbohydrate, mannan derived from yeast was immobilized on a microplate, and the binding of TRAIL-SPD, TRAIL-SPD_F335A or TRAIL-SPD_F335D was detected by ELISA. The results are shown in FIG. As expected, the ELISA signal increased with increasing TRAIL-ASPD concentration. In contrast, the carbohydrate variant form of TRAIL-ASPD_F335A showed a very low ELISA signal. In addition, the newly constructed variant TRAIL-ASPD_F335D represented the lowest ELISA signal (see inset and arrow). This suggested that mutant F335D had a lower mannan binding affinity compared to the SP-D mutant form of F335A described above.

(2.7 Pharmacokinetics of TRAIL-SPD fusion protein)
To measure the half-life of the TRAIL-SPD fusion protein, 10 micrograms of TRAIL-ASPD (A) or TRAIL-ASPD_F335D (B) was injected intravenously into male CD1 mice and after several time points (pre-dose, Serum samples were taken after 5 minutes, 30 minutes, 2 hours, 6 hours and 24 hours). TRAIL protein in mouse serum was quantified by ELISA and the data was used to calculate half-life. The results are shown in FIG. For the two proteins analyzed, a half-life of 7-14 hours was calculated for TRAIL-ASPD (A) and TRAIL-ASPD_F335D (B). During the observation period, the animals did not die and showed no signs of intolerance. This data shows at least an 80-fold improvement in serum half-life compared to wild-type TRAIL (Kelley et al. 2001) reported to have a half-life in the range of 3-5 minutes in rodents.

(2.8 Cytotoxicity of TRAIL-ASPD fusion protein)
To analyze the potential hepatotoxic effects of TRAIL-ASPD, TRAIL-ASPD_F335A, or TRAIL-ASPD_F335D, primary human hepatocytes (PHH) were used at various concentrations with or without cross-linking antibody (anti-Strep-tag II). Incubated with the indicated TRAIL-SPD fusion protein. As a control, CD95L-T4 (described in WO2008 / 025516), a stabilized variant of CD95L, was used. The results are shown in FIG.

  In addition, the effect of co-incubation of PHH with 5 mM chemotherapeutic drug was analyzed for TRAIL-ASPD_F335D. After 5 hours (A, B and E) or 24 hours (C, D and F) incubation, cells were lysed and caspase activity was assessed by fluorogenic assay.

  As a result, all TRAIL-SPD fusion proteins analyzed did not induce hepatotoxic effects even when the ligand was secondarily cross-linked by the antibody. In contrast, CD95L-T4 is hepatotoxic as shown by an increase in active caspase (AD). Co-incubation of primary human hepatocytes with trimeric TRAIL-ASPD_F335D with chemotherapeutic drugs for 5 hours did not induce caspase activity (E). However, after 24 hours incubation with doxorubicin, soluble TRAIL-ASPD_F335D induced a strong caspase activity signal (F).

  This suggests that the TRAIL fusion proteins of the present invention may not exhibit undesirable hepatotoxicity in medical applications. Accordingly, TRAIL fusion proteins are preferably apoptotic sensitizers and / or apoptosis inducers (eg, chemotherapeutic drugs such as oxaliplatin, cisplatin, 5-fluorouracil, etoposide, gemcitabine, irinotecan), or the Bcl2 family, in particular, It is administered in combination with a drug that is a Bcl2 binding molecule (eg, a small molecule or peptide compound) that binds to Bcl2 or Bclxl polypeptide.

(2.9 Characterization of APRIL fusion protein)
HEK293 cells were transiently transfected with an expression vector encoding APRIL-A69 (WO20080255516), APRIL-ASPD, APRIL-ACCSPD or APRIL-ACol11. After 3 days, the supernatant was analyzed for secreted protein by Western blotting. The results are shown in FIG. For detection of APRIL fusion protein, an antibody specific for Strep-tag II was used. Arrows indicate specific bands detected around 40 kDa (representing APRIL-ASPD and APRIL-ACol11, respectively) and specific bands detected around 25 kDa (respectively APRIL-A69 and APRIL-ACCSPD). . Thus, the APRIL expression cassette was functional and protein secretion indicated that the protein was properly folded. For the other TNFSF proteins analyzed, the highest secreted protein level was found for APRIL fused to a trimerization motif composed of a coiled coil “neck” + CRD of human SP-D ( APRIL-ASPD, lane number 2). APRIL-ASPD was used to analyze binding to the receptor TACI.

  To demonstrate that the constructed APRIL-ASPD fusion protein is functional, the binding of APRIL to a known receptor, TACI, was evaluated (FIG. 34). Therefore, APRIL-ASPD in the supernatant from transiently transfected HEK293 cells was immobilized on a microplate coated with streptactin. Cell supernatant from untransfected HEK293 cells served as a negative control. The specifically bound protein was detected by various concentrations of TACI-Fc and then incubated with an anti-human Fc specific antibody conjugated to peroxidase. As a result, the ELISA signal was enhanced as the concentration of TACI-Fc was increased, indicating that APRIL-ASPD is a functional molecule.

  The amino acid sequence of the APRIL fusion protein is shown below.

1-20: signal secretion peptide (underlined)
21-158: APRIL-RBD
159-169: flexible linker element (A-linker; GSS GSS GSS GS italic)
170-207: Coiled coil “neck” region of human SP-D 208-325: C-type lectin domain 326-336 of human SP-D: Linker element (GGSPSSSSSSSA)
337-344: Strep-tag II (WSHPQFEK).

(Example 3 Production of fusion protein with single chain antibody as effector polypeptide)
An exemplary amino acid sequence for a single chain (sc) Fv-SPD fusion protein is shown below (SEQ ID NOs: 52, 53).

1-20: signal secretion peptide (underlined)
21-140: heavy chain variable domain 141-155: linker element 156-264: light chain variable domain 265-276: A-linker 277-431: SPD-motif (motiv) (neck + CRD)
432-441: Linker element 442-450: Strep-tag II (WSHPQFEK)

1-20: signal secretion peptide (underlined)
21-140: heavy chain variable domain 141-155: linker element 156-264: light chain variable domain 265-276: A-linker 277-431: SPD-motif (motiv) (neck + CRD)
432-441: Linker element 442-450: Strep-tag II (WSHPQFEK).

  The protein was expressed and subjected to affinity chromatography as described in column 1.3. Aliquots of the eluate were separated by SDS-PAGE under reducing or non-reducing conditions. A single band at 40-50 kDa, indicated by the arrow (see FIG. 39), corresponding to sc006-ASPD-St with an expected molecular weight of 45.8 kDa can be detected.

  The protein was expressed, affinity purified and subjected to size exclusion chromatography as described in column 1.3. 50 micrograms of affinity purified protein is loaded onto a Superdex 200 column and a chromatogram is shown (see FIG. 40). The main peak at 12.63 ml corresponds to a molecular weight of 160 ± 15 kDa, similar to the predicted species of three scFv molecules built into trimers via SPD. The molecule can be functionally tested in an ELISA setting by immobilizing the antigen and detecting sc006-SPD-St.

  The fusion protein produced in this experiment includes a single chain antibody made against IL4R-alpha.

Claims (12)

(I) a. A collectin family carbohydrate recognition domain; and b. A collectin family trimerization domain comprising a collectin family neck region;
(Ii) a linker element; is positioned at the N-terminus of and (iii) the collectin family neck region, it Oh effector polypeptide, the effector polypeptide is an antibody, single chain antibody or an antibody or single chain antibody A fusion protein comprising an effector polypeptide that is a fragment of The carbohydrate recognition domain is a carbohydrate recognition domain of surfactant protein-D, surfactant protein-A1, surfactant protein-A2, mannan-binding protein-C, liver collectin 1, placental collectin 1 or collectin-11. The fusion protein according to 1. The fusion protein according to claim 1 or 2 , wherein the carbohydrate recognition domain comprises a single amino acid mutation or substitution. The carbohydrate recognition domain is a carbohydrate recognition domain of surfactant protein D, and the substitution is a substitution of phenylalanine 355 of SEQ ID NO: 21, and is mutated to a polar amino acid selected from aspartic acid, glutamic acid, asparagine and glutamine The fusion protein according to claim 3. The collectin family neck region, surfactant protein -D, surfactant protein -A1, surfactant protein -A2, mannan-binding protein -C, liver collectin 1, a neck region of the placenta collectin 1 or collectin -11 Claim 1-4 The fusion protein according to any one of the above. The fusion protein according to any one of claims 1 to 5 , wherein the linker element has an amino acid length of 25 or less. The fusion protein according to any one of claims 1 to 6, wherein the linker element is constructed with an amino acid selected from alanine, threonine, glycine and serine. 2. The fusion protein of claim 1, wherein the antibody, single chain antibody or antibody or fragment of a single chain antibody is made against a TNF superfamily cytokine, its receptor binding domain or a receptor for a cytokine. . 9. The fusion protein of claim 8, wherein the receptor for cytokine is IL4R-alpha. The fusion protein according to any one of claims 1 to 9, comprising an antibody, a single chain antibody, or an antibody or a fragment of a single chain antibody at the C-terminus of the collectin family carbohydrate recognition domain. A pharmaceutical comprising the fusion protein according to any one of claims 1 to 10. 10. The prophylactic disorder according to claim 1 or 9 for the treatment of proliferative disorders associated with TNF cytokine dysfunction, including tumors, infections, inflammatory diseases, metabolic diseases, autoimmune disorders, degenerative diseases and transplant rejection. Medicinal products containing fusion proteins.