Dcr3 variant

Abstract

To provide a DcR3 variant that has binding activity (preferably neutralizing activity) to a ligand of Decoy Receptor 3 (DcR3), and that exhibits a reduced level of aggregate formation compared to wild-type DcR3 when produced using a mammalian cell as a host, and / or that demonstrates improved in vivo pharmacokinetics; as well as DNA encoding the DcR3 variant, a vector comprising the DNA, a transformant obtained by introducing the vector, a method for producing the variant using the transformant, and a prophylactic or therapeutic agent for autoimmune diseases, inflammatory diseases, or allergic diseases comprising the variant as an active ingredient.SOLUTION: A DcR3 variant, which is a modified form of wild-type DcR3, exhibiting improved in vivo pharmacokinetics compared to the wild-type DcR3. The DcR3 variant may comprise a part of DcR3 and a part of a TNF superfamily molecule.SELECTED DRAWING: None

Description

[Technical Field]

[0001] The present invention relates to DcR3 variants, which are variants of wild-type DcR3. More specifically, the present invention relates to DcR3 variants that have binding activity (preferably neutralizing activity) to DcR3 ligands and that, when produced using mammalian-derived cells as hosts, produce fewer aggregates and / or exhibit improved pharmacokinetics compared to wild-type DcR3. [Background technology]

[0002] The tumor necrosis factor (TNF) superfamily (TNFSF) and the TNF receptor superfamily (TNFRSF) form a family of 18 structurally similar ligands and 29 structurally similar receptors, respectively. Antibodies and Fc fusion proteins against many molecules in this family have been developed and marketed, and have shown therapeutic effects in the treatment of various autoimmune diseases (Non-Patent Document 1).

[0003] Many TNFRSFs are expressed on the cell membrane and transmit signals downstream upon ligand binding, but some molecules are decoy receptors (DcRs) that are not involved in signal transduction. Four decoy receptors have been identified so far: DcR1, DcR2, DcR3, and OPG (Osteoprotegerin).

[0004] OPG is a soluble decoy receptor for RANKL and TRAIL, and inhibits signal transduction by competing with the RANKL receptor and TRAIL receptor for binding to their ligands. Meanwhile, DcR1 and DcR2 are decoy receptors for TRAIL, and DcR3 is a decoy receptor for the three molecules LIGHT, TL1A, and FasL, that neutralize their ligands by competitively inhibiting their binding to receptors that transmit signals (Non-Patent Document 2).

[0005] DcR3 is a soluble molecule consisting of 300 amino acid residues. It contains a signal peptide at the N-terminus followed by four cysteine-rich domains (CRDs) characteristic of TNFRSF (CRD1, CRD2, CRD3, and CRD4). The C-terminus contains a heparan sulfate-binding domain (HBD) rich in basic amino acids that contains a heparan sulfate-binding motif. LIGHT, TL1A, and FasL all bind to DcR3 via CRD2 and CRD3 (Non-Patent Documents 3, 4, and 5).

[0006] In addition to its function as a decoy receptor through ligand neutralization, DcR3 also functions as an immunoregulatory molecule based on the activity of its HBD. For example, it has been reported that DcR3 directly binds to glycosaminoglycans (GAGs), including heparan sulfate, on the cell membranes of monocytes, macrophages, and dendritic cells via its HBD, resulting in various immunosuppressive and immunostimulatory effects, such as Th2 induction through dendritic cell differentiation, M2 macrophage induction, enhanced monocyte adhesion, osteoclast differentiation, and reduced expression of MHC class II molecules (Non-Patent Documents 6 and 12).

[0007] DcR3 ligands have been reported to be involved in autoimmune diseases, inflammatory diseases, allergies, cancer, infectious diseases, and various other inflammatory responses. For example, LIGHT, TL1A, and FasL are all included in susceptibility loci for inflammatory bowel disease (IBD), and the existence of multiple genetic polymorphisms associated with the pathology of TL1A in particular has been reported. Furthermore, there have been reports of increased expression of DcR3 ligands in the blood or tissues of IBD patients, and of improvement of the pathology of mouse enteritis models by DcR3 ligand inhibition (Non-Patent Documents 6 to 9).

[0008] Although DcR3 expression in normal human tissues is extremely low, its expression is induced by infection or tissue damage. Furthermore, blood levels of DcR3 are known to be elevated in various autoimmune or inflammatory diseases, such as IBD, systemic lupus erythematosus (SLE), atopic dermatitis (AD), and rheumatoid arthritis (RA). Although no DcR3 homologue has been identified in mice, improvements in pathology have been confirmed in mouse pathological models, such as type I diabetes, multiple sclerosis, and nephritis, and administration of human DcR3 transgenic mice and plasmid or recombinant DcR3 have been shown to be effective (Non-Patent Documents 6 and 10).

[0009] Genentech cloned the human DcR3 gene and demonstrated that a fusion of DcR3 with the Fc region of human IgG1 binds to soluble human FasL and inhibits human FasL-dependent apoptosis in vitro (Patent Document 1).

[0010] Eli Lilly and Company has reported that FLINT, a protease-resistant DcR3 variant, was obtained by a single amino acid mutation (R218Q) in wild-type DcR3, and that its pharmacokinetics were improved compared to wild-type DcR3 in mice and monkeys. However, when wild-type DcR3 and FLINT were administered intravenously to cynomolgus monkeys at 0.5 mg / kg, their blood half-lives were extremely short, at 9 hours and 12.3 hours, respectively (Patent Documents 2 and 3, Non-Patent Document 11). [Prior art documents] [Patent documents]

[0011] [Patent Document 1] Patent No. 4303883 [Patent Document 2] US Patent US 6,835,814 B1 [Patent Document 3] US Patent US 6,965,012 B1 [Non-patent literature]

[0012] [Non-Patent Document 1] Nature Reviews Drug Discovery,2013,12:p.147-168 [Non-patent document 2] Nature Reviews Cancer,2002.2:p.420-430 [Non-patent document 3] Structure,2011,19:pp.162-171 [Non-patent document 4] Structure,2014,22:p.1252-1262 [Non-Patent Document 5] Structure,2016,24:p.2016-2023 [Non-patent document 6] Biochemical Pharmacology,2011,81:p.838-847 [Non-Patent Document 7] Immunology, 2009, 128: pp. 451-458 [Non-patent document 8] PNAS,2006,103:p.8441-8446 [Non-Patent Document 9] Am.J.Physiol.Gastrointest.Liver Physiol.,2003,285:p.G754-G760 [Non-Patent Document 10] Journal of Biomedical Science,2017,24:39 [Non-Patent Document 11] Drug Metabolism and Disposition,2003,31:p.502-507 [Non-Patent Document 12] J.Immunol.,2006,176:p.173-180 Summary of the Invention [Problem to be solved by the invention]

[0013] FLINT, an amino acid variant of wild-type DcR3, has extremely poor pharmacokinetics, and as a recombinant formulation whose mechanism of action is ligand neutralization, frequent administration is required, making it undesirable as a pharmaceutical. Therefore, DcR3 variants that can ensure a constant administration interval through improved pharmacokinetics are expected to be useful as pharmaceuticals.

[0014] Furthermore, as far as is known to date, no functional DcR3 variants have been found that produce reduced amounts of aggregates when expressed, isolated, and purified in mammalian host cells and that have neutralizing activity against DcR3 ligands.

[0015] The present invention aims to provide DcR3 variants that have binding activity (preferably neutralizing activity) against DcR3 ligands and that, when DcR3 protein is produced using mammalian-derived cells as a host, produce fewer aggregates and / or exhibit improved pharmacokinetics compared to wild-type DcR3; DNA encoding the DcR3 variants; vectors containing the DNA; transformants obtained by introducing the vectors; methods for producing the variants using the transformants; and pharmaceutical compositions containing the variants as active ingredients and agents for preventing or treating autoimmune diseases, inflammatory diseases, or allergies. [Means for solving the problem]

[0016] In order to solve the above problems, the present invention provides the following inventions. [1] A DcR3 variant that is a variant of wild-type Decoy Receptor 3 (hereinafter abbreviated as DcR3), which shows improved pharmacokinetics compared to the wild-type DcR3. [2] The DcR3 variant according to [1], which has one or more N-glycoside-linked complex-type sugar chains. [3] The DcR3 variant according to [1] or [2], which has neutralizing activity against at least one of LIGHT, TL1A, and FasL. [4] The DcR3 variant according to any one of [1] to [3], which has neutralizing activity against all of LIGHT, TL1A, and FasL. [5] The DcR3 variant according to any one of [1] to [3], which does not have neutralizing activity against FasL, but has neutralizing activity against one or more of LIGHT and TL1A. [6] The DcR3 variant according to any one of [1] to [3] and [5], which does not have neutralizing activity against FasL but has neutralizing activity against LIGHT and TL1A. [7] A DcR3 variant comprising: a first chimeric cysteine-rich region consisting of an amino acid sequence in which at least a portion of the cysteine-rich domain (hereinafter abbreviated as CRD) of wild-type DcR3 is replaced with at least a portion of the cysteine-rich domain of a TNF receptor superfamily molecule other than DcR3; or a second chimeric cysteine-rich region consisting of an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted or added in the amino acid sequence of the first chimeric cysteine-rich region. [8] The DcR3 variant according to [7], which has one or more N-glycoside-linked complex-type sugar chains. [9] The DcR3 variant according to [7] or [8], which has neutralizing activity against at least one of LIGHT, TL1A, and FasL.

[10] The DcR3 variant according to any one of [7] to [9], which has neutralizing activity against all of LIGHT, TL1A, and FasL.

[11] The DcR3 variant according to any one of [7] to [9], which does not have neutralizing activity against FasL, but has neutralizing activity against one or more of LIGHT and TL1A.

[12] The DcR3 variant according to any one of [7] to [9] and

[11] , which does not have neutralizing activity against FasL but has neutralizing activity against LIGHT and TL1A.

[13] The DcR3 variant according to any one of [7] to

[12] , wherein the TNF receptor superfamily molecule is OPG.

[14] At least a portion of the cysteine-rich domain of the wild-type DcR3 is selected from the group consisting of all or a portion of CRD1, all or a portion of CRD2, all or a portion of CRD3, and all or a portion of CRD4; A DcR3 variant described in any one of [7] to

[13] , wherein at least a portion of the cysteine-rich domain of the TNF receptor superfamily molecule is selected from all or a portion of CRD1, all or a portion of CRD2, all or a portion of CRD3, and all or a portion of CRD4.

[15] The first chimeric cysteine-rich region is replacement of a portion of the CRD1 of the wild-type DcR3 with a portion of the CRD1 of the TNF receptor superfamily molecule that corresponds to the portion of the CRD1 of the wild-type DcR3; replacement of the entire CRD1 of said wild-type DcR3 with the entire CRD1 of said TNF receptor superfamily molecule; replacement of a portion of the CRD2 of the wild-type DcR3 with a portion of the CRD2 of the TNF receptor superfamily molecule that corresponds to the portion of the CRD2 of the wild-type DcR3; replacement of the entire CRD2 of said wild-type DcR3 with the entire CRD2 of said TNF receptor superfamily molecule; replacement of a portion of the CRD3 of the wild-type DcR3 with a portion of the CRD3 of the TNF receptor superfamily molecule that corresponds to the portion of the CRD3 of the wild-type DcR3; replacement of the entire CRD3 of said wild-type DcR3 with the entire CRD3 of said TNF receptor superfamily molecule; Substitution of a portion of the CRD4 of the wild-type DcR3 with a portion of the CRD4 of the TNF receptor superfamily molecule that corresponds to the portion of the CRD4 of the wild-type DcR3; and Replacement of the entire CRD4 of the wild-type DcR3 with the entire CRD4 of the TNF receptor superfamily molecule. The DcR3 variant according to

[14] , having one or more substitutions selected from the following:

[16] The DcR3 variant according to

[15] , wherein the first chimeric cysteine-rich region retains a part or the entirety of the CRD2 of the wild-type DcR3.

[17] The DcR3 variant according to

[15] or

[16] , wherein the first chimeric cysteine-rich region retains a part or the entirety of the CRD3 of the wild-type DcR3.

[18] A DcR3 variant described in any one of

[14] to

[17] , wherein the first chimeric cysteine-rich region comprises the following amino acid sequence (a), (b), (c) or (d), and the second chimeric cysteine-rich region comprises the following amino acid sequence (e): (a) an amino acid sequence in which the CRD1 of the wild-type DcR3 in the cysteine-rich region is substituted with the CRD1 of OPG; (b) an amino acid sequence in which the CRD4 of the wild-type DcR3 in the cysteine-rich region is substituted with the CRD4 of OPG; (c) an amino acid sequence in the cysteine-rich region of the wild-type DcR3, in which the CRD1 of the wild-type DcR3 is replaced with the CRD1 of OPG and the CRD4 of the wild-type DcR3 is replaced with the CRD4 of OPG; (d) an amino acid sequence in which the 103rd to 123rd amino acids from the N-terminus of the amino acid sequence of (a), (b), or (c) are substituted with the corresponding amino acid sequence of the cysteine-rich domain of OPG. (e) an amino acid sequence in which 1 to 30 amino acids are deleted, substituted, inserted or added in the amino acid sequence of (a), (b), (c) or (d) above;

[19] The amino acid sequence of (a) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 26 or 50, the amino acid sequence of (b) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 28 or 52, the amino acid sequence of (c) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30 or 54, The DcR3 variant according to

[18] , wherein the amino acid sequence (d) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 32 or 56.

[20] A DcR3 variant described in

[18] or

[19] , in which the amino acid sequence (e) has one or more amino acids selected from the group consisting of Glu at the 57th position, Arg at the 58th position, and Arg at the 60th position from the N-terminus of the amino acid sequence of (a), (b), (c), or (d) substituted with other amino acids.

[21] A DcR3 variant described in any one of

[18] to

[20] , wherein the amino acid sequence (e) has a substitution of Glu at position 57 and Arg at position 58 from the N-terminus of the amino acid sequence (a), (b), (c) or (d) with other amino acids.

[22] A DcR3 variant described in any one of

[18] to

[21] , wherein the amino acid sequence (e) has one or more substitutions selected from the group consisting of a substitution of Glu at position 57 from the N-terminus of the amino acid sequence (a), (b), (c) or (d) with Lys, Leu, Arg, Val, Ala, Phe, His, Ile or Met, a substitution of Arg at position 58 with Asp, Glu or Thr, and a substitution of Arg at position 60 with Lys.

[23] A DcR3 variant described in any one of

[18] to

[22] , wherein the amino acid sequence (e) has a substitution of Glu at the 57th position from the N-terminus of the amino acid sequence (a), (b), (c) or (d) with Lys, Leu, Arg, Val, Ala, Phe, His, Ile or Met, and a substitution of Arg at the 58th position from the N-terminus with Asp, Glu or Thr.

[24] The DcR3 variant according to any one of

[18] to

[23] , wherein the amino acid sequence (e) has a substitution selected from the following (f) to (i): (f) Substitution of Asn at the 131st and 144th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) above with other amino acids (g) Substitution of Asn at positions 131, 144, and 157 from the N-terminus of the amino acid sequence of (b), (c), or (d) with other amino acids (h) Substitution of Thr at position 133 and Ser at position 146 from the N-terminus of the amino acid sequence of (b), (c), or (d) above with other amino acids. (i) Substitution of Thr at position 133, Ser at position 146, and Thr at position 159 from the N-terminus of the amino acid sequence of (b), (c), or (d) above with other amino acids

[25] The DcR3 variant according to any one of

[18] to

[24] , wherein the amino acid sequence (e) has a substitution selected from the following (f') to (i'): (f') Substitution of Asn at the 131st and 144th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) above with Ser (g') Substitution of Asn at the 131st, 144th, and 157th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) above with Ser (h') Substitution of Thr at position 133 and Ser at position 146 from the N-terminus of the amino acid sequence of (b), (c), or (d) with Ala (i') Substitution of Thr at position 133, Ser at position 146, and Thr at position 159 from the N-terminus of the amino acid sequence of (b), (c), or (d) with Ala

[26] A DcR3 variant described in any one of

[18] to

[25] , wherein the amino acid sequence (e) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 58, 60, 62, 64, 66, 68, 70, 180, 182, 184, 186, 188, 270, 272, 274, 276, 278, 280, 282, 284 or 286.

[27] The DcR3 variant described in any one of [7] to

[26] , wherein the DcR3 variant comprises the first or second chimeric cysteine-rich region and part or all of the heparan sulfate binding region of the wild-type DcR3 bound to the C-terminal side of the first or second chimeric cysteine-rich region, or the DcR3 variant comprises the first or second chimeric cysteine-rich region but does not comprise the heparan sulfate binding region of the wild-type DcR3.

[28] The DcR3 variant is (I) an amino acid sequence set forth in SEQ ID NO: 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, or 46, or an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted, or added in the amino acid sequence; and (II) An amino acid sequence set forth in SEQ ID NO: 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, 180, 182, 184, 186, 188, 270, 272, 274, 276, 278, 280, 282, 284, or 286, or an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted, or added. The DcR3 variant according to

[27] , comprising any one of the amino acid sequences selected from the following:

[29] The DcR3 variant described in [7] to

[28] , which comprises an Fc region derived from a human IgG1, IgG2, or IgG4 antibody, or a mutant Fc region consisting of an amino acid sequence in which one or several amino acids have been deleted, substituted, inserted, or added in the amino acid sequence of the Fc region.

[30] A DcR3 variant described in

[29] , in which the Fc region or the mutant Fc region is linked to the C-terminal side of the first or second chimeric cysteine-rich region via another region or a linker.

[31] The DcR3 variant described in

[29] or

[30] , wherein the mutant Fc region has a substitution of Cys at position 220 (EU index) in the amino acid sequence of the heavy chain of human IgG1 with Ser.

[32] The DcR3 variant described in

[31] , wherein the mutant Fc region has a substitution of Leu at position 234 with Ala, Leu at position 235 with Ala, and Gly at position 237 with Ala, as shown in the EU index, in the amino acid sequence of the heavy chain of human IgG1.

[33] A DcR3 variant according to

[31] or

[32] , wherein the mutant Fc region has a substitution of Asn at position 434 (EU index) in the amino acid sequence of the heavy chain of human IgG1 with Ala.

[34] The DcR3 variant described in

[31] , wherein the mutant Fc region has a substitution of Met at position 252 with Tyr, Ser at position 254 with Thr, and Thr at position 256 with Glu, as shown in the EU index of the amino acid sequence of the heavy chain of human IgG1.

[35] The DcR3 variant according to

[29] or

[30] , wherein the mutant Fc region has a substitution of Ser at position 228 with Pro, Leu at position 235 with Glu, and Arg at position 409 with Lys, as shown in the EU index, in the amino acid sequence of the heavy chain of human IgG4.

[36] The DcR3 variant described in

[29] or

[30] , which comprises a mutant Fc region consisting of the amino acid sequence set forth in SEQ ID NO: 72, 74, 156, 158, 160, 162, 164, 166, 311, 312 or 313.

[37] The DcR3 variants are selected from the group consisting of SEQ ID NOs: 76, 78, 80, 82, 84, 86, 88, 90, 92, 94, 96, 98, 150, 168, 170, 172, 174, 176, 178, 190, 192, 194, 196, 198, 200, 202, 204, 206, 208, 210, 212, 214, 216, 218, 220, 222, 224, 226, 228, 230, 232, 234, 236, 238, 240, 242, 244, 246, 248, 250, 252, 254, 256, 258, 260, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298, 300, 301, 302, 303, 304, 305, 306, 307 The DcR3 variant according to any one of

[29] to

[36] , comprising the amino acid sequence set forth in any one of [4], 266, 268, 288, 290, 292, 294, 296, 298, 300, 302, 304, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, or 337, or an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted, or added within the amino acid sequence.

[38] A DcR3 variant composition comprising the DcR3 variant according to any one of [1] to

[37] .

[39] The composition described in

[38] , which comprises a DcR3 variant having one or more N-glycoside-linked complex-type glycans and a DcR3 variant having no N-glycoside-linked complex-type glycans.

[40] A DNA encoding the DcR3 variant according to any one of [1] to

[37] .

[41] A recombinant vector containing the DNA described in

[40] .

[42] A transformant obtained by introducing the recombinant vector according to

[41] into a host cell.

[43] The transformant according to

[42] , wherein the host cell is a cell derived from a mammal.

[44] The transformant according to

[43] , wherein the mammalian-derived cells are CHO cells. A method for producing a DcR3 variant or a DcR3 variant composition, comprising culturing a transformant described in any one of

[45] ,

[42] to

[44] in a culture medium to produce and accumulate a DcR3 variant described in any one of [1] to

[37] , and purifying the DcR3 variant from the resulting culture medium.

[46] A DcR3 variant or a DcR3 variant composition produced by the production method described in

[45] .

[47] A pharmaceutical composition comprising, as an active ingredient, the DcR3 variant or DcR3 variant composition according to any one of [1] to

[39] and

[46] .

[48] The pharmaceutical composition according to

[47] , which is an agent for preventing or treating an autoimmune disease, an inflammatory disease, or an allergic disease.

[49] A method for preventing or treating an autoimmune disease, inflammatory disease, or allergic disease, comprising administering the pharmaceutical composition according to

[47] or

[48] to a patient in need of prevention or treatment of an autoimmune disease, inflammatory disease, or allergic disease. [Effects of the Invention]

[0017] The present invention provides DcR3 variants that have binding activity (preferably neutralizing activity) for DcR3 ligands and that, when DcR3 protein is produced using mammalian-derived cells as a host, produce fewer aggregates than wild-type DcR3 and / or exhibit improved pharmacokinetics; DNA encoding the DcR3 variants; vectors containing the DNA; transformants obtained by introducing the vectors; methods for producing the variants using the transformants; and pharmaceutical compositions containing the variants as active ingredients and agents for preventing or treating autoimmune diseases, inflammatory diseases, or allergies. [Brief explanation of the drawings]

[0018] [Figure 1]Figure 1A shows the results of SDS-PAGE of various wild-type DcR3 controls produced in mammalian cells. Lanes 1 and 3 show DcR3 FL-Fc, lanes 2 and 4 show DcR3 FL-FLAG, and lanes 5 and 6 show S195-Fc(g1S), which were electrophoresed under non-reducing and reducing conditions, respectively. Figure 1B shows electrophoresis of commercially available human DcR3-Fc (lane 7) produced in HEK293 cells under non-reducing conditions and detection by immunoblotting with an anti-human IgG antibody. Figure 1C shows commercially available insect cell-produced DcR3 FL-Fc (lanes 8 and 9), S195-Fc(g1S) (lanes 10 and 11), DcR3 FL-Fc(g1S) (lanes 14, 15, 18, and 19), and R218Q-Fc(g1S) (lanes 12, 13, 16, and 17) electrophoresed under non-reducing and reducing conditions, respectively. M indicates a molecular weight marker (Bio-Rad). [Figure 2] FIG. 2 shows an alignment of the immature amino acid sequences of human DcR3 and human OPG, with the cysteine-rich domains designated CRD1, CRD2, CRD3, and CRD4, respectively. [Figure 3] Figure 3 shows the domain structures of the various wild-type DcR3 controls, various DcR3 variants, and human OPG. A: DcR3 FL-Fc, B: S195-Fc, C: chimera A-Fc, D: 103-123OPG-Fc, E: N-glycosylation-2-substituted Fc, F: N-glycosylation-3-substituted Fc, G: chimera B-Fc, H: chimera C-Fc, and I: OPG. The vertical lines in CRD4 of E and F indicate two or three N-glycosylation residue substitutions, respectively, and DD in I indicates the Death domain. [Figure 4]Figure 4 shows the results of SDS-PAGE of various DcR3 variants produced in mammalian cells. Lanes 1 and 4 show chimera A-Fc (IEGRMD g1S) produced in Expi293 cells; lanes 2 and 5 show chimera A-Fc (IEGRMD g1S) produced in CHO-S cells; lanes 3 and 6 show chimera B-Fc (g1S) produced in Expi293 cells; and lanes 7 and 8 show chimera C-Fc (IEGRMD g1S) produced in Expi293 cells. These were electrophoresed under non-reducing or reducing conditions. M indicates a molecular weight marker (Bio-Rad). [Figure 5] 5 shows melting curves measured by the DSF method for R218Q-Fc, S195-Fc, and chimera A-Fc (IEGRMD g1S). The vertical axis represents fluorescence intensity (RFU) (103), and the horizontal axis represents temperature (°C). [Figure 6] Figure 6 shows the binding of various wild-type DcR3 controls and various DcR3 variants to human primary cells and CHO cells. Each cell was reacted with 10 μg / mL of each DcR3 variant, then stained with 0.1 μg / mL of PE-labeled anti-human antibody, and the PE fluorescence intensity was measured by flow cytometry. The vertical axis of the graph shows the geometric mean (Geo.Mean) of PE. The upper panel shows the staining results for HUVECs, the middle panel shows the staining results for hepatocytes, and the lower panel shows the staining results for CHO cells. [Figure 7] Figure 7 shows the time course of blood concentrations of S195-Fc and chimera A-Fc (IEGRMD g1S) after iv administration of 10 mg / kg to BALB / c mice. The vertical axis represents blood concentration (ng / mL), and the horizontal axis represents time (hr) after administration. [Figure 8]Figure 8A shows the binding of various wild-type DcR3 controls and various DcR3 variants to RANKL (OPG ligand), and Figure 8B shows the binding of various wild-type DcR3 controls and various DcR3 variants to TRAIL (OPG ligand). After capturing various wild-type DcR3 controls and various DcR3 variants on a plate immobilized with anti-human antibody, RANKL or TRAIL diluted to various concentrations was added and binding was assessed. Detection was performed using biotinylated anti-RANKL antibody or biotinylated anti-TRAIL antibody and streptavidin-HRP. The horizontal axis shows the concentration of RANKL or TRAIL (pg / mL), and the vertical axis shows absorbance (the value obtained by subtracting the absorbance at 570 nm from the absorbance at 450 nm). [Figure 9] Figure 9 shows the neutralizing activity of various wild-type DcR3 controls and various DcR3 variants against LIGHT. The graph shows IL-8 production from HT-29 cells when LIGHT was added at 100 ng / mL, and various wild-type DcR3 controls and various DcR3 variants at 0.1, 1, or 10 μg / mL. The vertical axis shows IL-8 concentration (pg / mL), and the horizontal axis shows various DcR3 variants added as inhibitors. [Figure 10] Figure 10 shows the neutralizing activity of various wild-type DcR3 controls and various DcR3 variants against TL1A. The graph shows IFN-γ production from human T cells when TL1A was added at 100 ng / mL, and various wild-type DcR3 controls and various DcR3 variants at 0.1, 1, or 10 μg / mL. The vertical axis shows IFN-γ concentration (pg / mL), and the horizontal axis shows various DcR3 variants added as inhibitors. [Figure 11] Figure 11 shows the neutralizing activity of various wild-type DcR3s and various DcR3 variant controls against FasL. The figure shows the amount of ATP produced by live Jurkat cells in RLU when FasL was added at 100 ng / mL, and various wild-type DcR3 controls and various DcR3 variants at 0.01, 0.1, or 1 μg / mL. The vertical axis represents cell viability (RLU × 106) as measured by ATP-dependent chemiluminescence, and the horizontal axis represents various DcR3 variants added as inhibitors. [Figure 12A]Figure 12A shows the binding of chimera A-Fc (g4PEK) and its variants with reduced FasL-binding ability (g4PEK) to each DcR3 ligand. Chimera A-Fc (g4PEK) or each variant with reduced FasL-binding ability was captured on a sensor chip with immobilized anti-human antibodies, and DcR3 ligands (human FasL, human LIGHT, and human TL1A) were applied as analytes. The vertical axis represents the amount of binding (RU), and the horizontal axis represents time (sec). [Figure 12B] Figure 12B shows the binding of the FasL-binding-reduced mutant (g4PEK) to each DcR3 ligand. Each FasL-binding-reduced mutant was captured on a sensor chip coated with an anti-human antibody, and the sensorgrams were obtained by passing DcR3 ligands (human FasL, human LIGHT, and human TL1A) through the sensor chip as an analyte. The vertical axis represents the amount of binding (RU), and the horizontal axis represents time (sec). [Figure 12C] Figure 12C shows the binding of the FasL-binding-reduced mutant (g4PEK) to each DcR3 ligand. Each FasL-binding-reduced mutant was captured on a sensor chip coated with an anti-human antibody, and the sensorgrams were obtained by passing DcR3 ligands (human FasL, human LIGHT, and human TL1A) through the sensor chip as an analyte. The vertical axis represents the amount of binding (RU), and the horizontal axis represents time (sec). [Figure 13] Figure 13 shows the results of SDS-PAGE of wild-type DcR3 controls produced in various mammalian cells. SDS-PAGE was performed by electrophoresis of commercially available human DcR3-Fc produced in various mammalian cells under reducing or non-reducing conditions. The mammalian cells used were HEK293 cells (Abcam) in lanes 1 and 4, CHO cells (AdipoGen) in lanes 2 and 5, and HEK293 cells (Enzo) in lanes 3 and 6. M indicates a molecular weight marker (Bio-Rad). [Figure 14A] Figure 14A shows the percentages of monomer, aggregate, and degradation product contents calculated from the peak areas of SEC-HPLC or SEC-UPLC of Protein A-purified mutants with reduced FasL binding activity, as well as the results for all mutants with reduced FasL binding activity (g4PEK). [Figure 14B] Figure 14B shows the percentages of monomer, aggregate, and degradation product contents calculated from the peak areas of SEC-HPLC or SEC-UPLC of Protein A-purified mutants with reduced FasL binding. Figure 14B shows the results for the selected mutant Fc fusion products with reduced FasL binding. [Figure 14C] FIG. 14C shows the percentages (%) of monomer, aggregate, and degradant contents of various chimera A-variant Fc fusion proteins purified with Protein A, calculated from the peak areas of SEC-HPLC or SEC-UPLC. [Figure 15A] Figure 15A shows the results of BIAcore analysis of the binding activity of various DcR3 variants to each DcR3 ligand. Various DcR3 variants were captured on a sensor chip immobilized with anti-human antibodies, and the kinetic constants (k, k, and K) were measured when a trimeric DcR3 ligand (human FasL, human LIGHT, or human TL1A) was injected as an analyte. Figure 15B shows the results for various chimera A-Fc antibodies with different Fc sequences. [Figure 15B] Figure 15B shows the results of BIAcore analysis of the binding activity of various DcR3 variants to each DcR3 ligand. Various DcR3 variants were captured on a sensor chip immobilized with anti-human antibodies, and the kinetic constants (k, k, and K) were measured when a trimeric DcR3 ligand (human FasL, human LIGHT, or human TL1A) was injected as an analyte. Figure 15B shows the results for variants with reduced FasL binding. [Figure 16A] Figure 16A shows the kinetic constants (KD values) for each DcR3 ligand calculated by BIAcore during selection of mutants with reduced FasL binding activity, compared with those of chimera A-Fc (g4PEK). Figure 16A shows the results for single amino acid substitution mutants. [Figure 16B] Figure 16B shows the kinetic constants (KD values) for each DcR3 ligand calculated by BIAcore during selection of mutants with reduced FasL binding activity, compared with those of chimera A-Fc (g4PEK). Figure 16B shows the results for mutants with two amino acid substitutions. [Figure 17A]Figure 17A shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble LIGHT. Figure 17A shows the inhibitory activity of various chimeric A-Fc with different Fc sequences against LIGHT-dependent CXCL10 production from IFN-γ-stimulated intestinal myofibroblasts. The vertical axis shows CXCL10 concentration (ng / mL), and the horizontal axis shows the concentration of added DcR3 variants (ng / mL). [Figure 17B] Figure 17B shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble LIGHT. Figure 17B shows the inhibitory activity of chimera A-Fc with different mutated Fc sequences against LIGHT-dependent CXCL10 production from IFN-γ-stimulated intestinal myofibroblasts. The vertical axis shows CXCL10 concentration (ng / mL), and the horizontal axis shows the concentration of added DcR3 variants (ng / mL). [Figure 17C] Figure 17C shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble LIGHT. Figure 17C shows the inhibitory activity of single-amino acid-substituted variants with reduced FasL binding activity against LIGHT-dependent IL-8 production from HT-29 cells. The vertical axis shows IL-8 concentration (ng / mL), and the horizontal axis shows the concentration of added DcR3 variants (ng / mL). [Figure 17D] Figure 17D shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble LIGHT. Figure 17D shows the inhibitory activity of two-amino acid substitution variants with reduced FasL binding activity against LIGHT-dependent CXCL10 production from IFN-γ-stimulated intestinal myofibroblasts. The vertical axis shows CXCL10 concentration (ng / mL), and the horizontal axis shows the concentration of added DcR3 variants (ng / mL). [Figure 18A] Figure 18A shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble TL1A. Figure 18A shows the inhibitory activity of various chimeric A-Fc with different Fc sequences against TL1A-dependent IFN-γ production from IL-12- and IL-18-stimulated human T cells. The vertical axis shows IFN-γ concentration (pg / mL), and the horizontal axis shows the concentration of the added DcR3 variant (ng / mL). [Figure 18B]Figure 18B shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble TL1A. Figure 18B shows the inhibitory activity of chimera A-Fc with different mutated Fc sequences against TL1A-dependent IFN-γ production from IL-12- and IL-18-stimulated human T cells. The vertical axis shows IFN-γ concentration (pg / mL), and the horizontal axis shows the concentration of the added DcR3 variant (ng / mL). [Figure 18C] Figure 18C shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble TL1A. Figure 18C shows the inhibitory activity of single-amino acid-substituted variants with reduced FasL binding activity against TL1A-dependent IFN-γ production from IL-12- and IL-18-stimulated human T cells. The vertical axis shows IFN-γ concentration (pg / mL), and the horizontal axis shows the concentration of the added DcR3 variant (ng / mL). [Figure 18D] Figure 18D shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble TL1A. Figure 18D shows the inhibitory activity of two-amino acid-substituted variants with reduced FasL binding against TL1A-dependent IFN-γ production from IL-12- and IL-18-stimulated human T cells. The vertical axis shows IFN-γ concentration (pg / mL), and the horizontal axis shows the concentration of the added DcR3 variant (ng / mL). [Figure 19A] Figure 19A shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble FasL. Figure 19A shows the inhibitory activity of various chimeric A-Fc with different Fc sequences against A3 cell death. The vertical axis shows cell survival (RLU × 106) as measured by ATP-dependent chemiluminescence, and the horizontal axis shows the concentration (ng / mL) of each DcR3 variant added. [Figure 19B] Figure 19B shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble FasL. Figure 19B shows the inhibitory activity of chimera A-Fc with different mutated Fc sequences against Jurkat cell death. The vertical axis shows cell survival (RLU × 106) as indexed by ATP production, and the horizontal axis shows the concentration (ng / mL) of each DcR3 variant added. [Figure 19C]Figure 19C shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble FasL. Figure 19C shows the inhibitory activity of single-amino acid-substituted variants with reduced FasL binding activity against Jurkat cell death. The vertical axis shows cell survival (RLU × 106) as measured by ATP production, and the horizontal axis shows the concentration (ng / mL) of each DcR3 variant added. [Figure 19D] Figure 19D shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble FasL. Figure 19D shows the inhibitory activity of two-amino acid-substituted variants with reduced FasL binding activity against Jurkat cell death. The vertical axis shows cell survival (RLU × 106) as measured by ATP production, and the horizontal axis shows the concentration (ng / mL) of each DcR3 variant added. [Figure 20A] Figure 20A shows the results of evaluating the binding activity of S195-Fc and various chimeric A-Fc with different Fc sequences to a cell line overexpressing membrane-type LIGHT. Each cell was reacted with various DcR3 variants, stained with a PE-labeled anti-human antibody, and the fluorescence intensity of PE was measured by flow cytometry. The vertical axis of the graph shows the geometric mean (Geo.Mean) of PE. [Figure 20B] Figure 20B shows the results of evaluating the binding activity of S195-Fc and various chimeric A-Fc with different Fc sequences to cells expressing membrane-type TL1A and membrane-type FasL. Each cell was reacted with various DcR3 variants, stained with PE-labeled anti-human antibody, and the fluorescence intensity of PE was measured by flow cytometry. The vertical axis of the graph shows the geometric mean (Geo.Mean) of PE. [Figure 21A] Figure 21A shows the results of evaluating the binding activity of S195-Fc, various chimeric A-Fc, and variants with reduced FasL binding to a cell line overexpressing membrane-type LIGHT. Each cell line was reacted with various DcR3 variants, stained with PE-labeled anti-human antibody, and the PE fluorescence intensity was measured by flow cytometry. The vertical axis of the graph shows the geometric mean (Geo.Mean) of PE. [Figure 21B]Figure 21B shows the results of evaluating the binding activity of S195-Fc, various chimeric A-Fc, and variants with reduced FasL binding to a cell line overexpressing membrane-type TL1A. Each cell was reacted with various DcR3 variants, stained with PE-labeled anti-human antibody, and the fluorescence intensity of PE was measured by flow cytometry. The vertical axis of the graph shows the geometric mean (Ge.Mean) of PE. [Figure 21C] Figure 21C shows the results of evaluating the binding activity of chimera A-Fc and a mutant with reduced FasL binding (g4PEK) to a cell line overexpressing membrane-type FasL. Each cell line was reacted with various DcR3 mutants, stained with PE-labeled anti-human antibody, and the fluorescence intensity of PE was measured by flow cytometry. The vertical axis of the graph shows the geometric mean (Geo.Mean) of PE. [Figure 22] Figure 22 shows the results of evaluating the binding activity of chimera A-Fc to membrane-type LIGHT on primary cells. Figure 22A shows the results of measuring the expression of membrane-type LIGHT on activated human T cells using a PE-labeled anti-LIGHT antibody. Figure 22B shows the geometric mean (Geo.Mean) calculated by flow cytometry of the binding of Alexa Fluor 488-labeled chimera A-Fc to membrane-type LIGHT on activated human T cells. In both figures, the vertical axis shows the geometric mean (Geo.Mean) of fluorescence intensity. [Figure 23] 23 shows the results of assessing the binding activity of chimera A-Fc to membrane-type TL1A on primary cells. The binding of Alexa Fluor 647-labeled chimera A-Fc to membrane-type TL1A on HUVEC cells was measured by flow cytometry in the presence or absence of a competitor protein, and the calculated geometric mean (Geo.Mean) is shown. [Figure 24] Figure 24 shows the results of sandwich ELISA for the binding activity of chimera A-Fc to FasL derived from primary cells, measured against soluble FasL in the culture supernatant of AICD-induced human T cells. Figure 24A shows the absorbance at 450 nm when the culture supernatant was added to a plate capturing chimera A-Fc, and Figure 24B shows the absorbance at 450 nm when the culture supernatant was detected with an anti-FasL antibody. [Figure 25]FIG. 25 shows the elution times (minutes) of various chimera A-Fc proteins with different Fc sequences and variants with reduced FasL binding ability in hydrophobic interaction chromatography (HIC). [Figure 26] FIG. 26 shows the Tm values (° C.) calculated by the Differential Scanning Fluorimetry (DSF) method for various chimera A-Fc antibodies with different Fc sequences and variants with reduced FasL binding ability. [Figure 27] Figure 27 shows the values of the blood half-life (h) during the elimination phase and the area under the blood concentration-time curve (AUC0-∞) (μg*h mL) of chimera A-Fc (Eg1S) with different Fc sequences and its variants with reduced FasL binding activity after intravenous administration of 10 mg / kg in BALB / c mice. [Figure 28A] Figure 28A shows the macroscopic pathology score index in a drug efficacy test of chimeric A-Fc using a mouse acute xenogeneic GVHD model. [Figure 28B] Figure 28B shows the pathology scores for each individual and the average for each group, with the pathology score on the vertical axis and each treatment group on the horizontal axis. [Figure 29] Figure 29 shows the number of human CD45-positive, human CD3- and CD4-positive, and human CD3- and CD8-positive cells per spleen for each individual in each group and the average number for each group in a drug efficacy test of chimeric A-Fc using a mouse acute xenogeneic GVHD model. Group 1 shows no human cell transfer, Group 2 shows a group in which human cells were transferred and a DNP antibody was administered, and Group 3 shows a group in which human cells were transferred and a chimeric A-Fc was administered. The vertical axis shows the number of cells positive for each surface marker, and the horizontal axis shows each treatment group. [Figure 30A] Figure 30A shows the results of BIAcore assay of the binding activity of various DcR3 variants to human DcR3 ligands, including the kinetic constants (k, k, and K) when trimers of human FasL, human LIGHT, or human TL1A were used as the human DcR3 ligand. [Figure 30B]Figure 30B shows the results of BIAcore measurement of the binding activity of various DcR3 variants to cynomolgus monkey DcR3 ligands, showing various kinetic constants (ka, kd, KD) when trimers of cynomolgus monkey FasL, cynomolgus monkey LIGHT, or cynomolgus monkey TL1A were used as cynomolgus monkey DcR3 ligands. [Figure 31] Figure 31 shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble LIGHT. It also shows the inhibitory activity of chimeric A-Fc with various mutant Fc and variants with reduced FasL binding ability against LIGHT-dependent CXCL10 production by intestinal myofibroblasts. The vertical axis shows CXCL10 concentration (ng / mL), and the horizontal axis shows the concentration of added DcR3 variants (ng / mL). [Figure 32] Figure 32 shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble TL1A. It also shows the inhibitory activity of chimeric A-Fc and variants with reduced FasL binding, each containing various mutant Fc, against TL1A-dependent IFN-γ production by human T cells. The vertical axis represents the IFN-γ concentration (pg / mL), and the horizontal axis represents the concentration of the added DcR3 variant (ng / mL). [Figure 33] Figure 33 shows the results of evaluating the neutralizing activity of various DcR3 variants against soluble FasL. Figure 33 also shows the inhibitory activity of chimera A-Fc and variants with reduced FasL binding ability containing various mutant Fc against Jurkat cell death. The vertical axis shows cell survival (RLU × 106) as measured by ATP-dependent chemiluminescence, and the horizontal axis shows the concentration (ng / mL) of each DcR3 variant added. [Figure 34A] 34A shows the results of flow cytometry evaluation of the binding activity of various DcR3 variants to a cell line overexpressing membrane-type LIGHT. The vertical axis of the graph shows the geometric mean (Geo.Mean) of PE. [Figure 34B] Figure 34B shows the results of flow cytometry evaluation of the binding activity of various DcR3 variants to a cell line overexpressing membrane-type TL1A. The vertical axis of the graph shows the geometric mean (Geo.Mean) of PE. [Figure 34C]34C shows the results of flow cytometry evaluation of the binding activity of various DcR3 variants to a cell line overexpressing membrane-type FasL. The vertical axis of the graph shows the geometric mean (Geo.Mean) of PE. [Figure 35] Figure 35 shows the results of evaluating the neutralizing activity of various DcR3 variants against membrane-type LIGHT. Figure 35 also shows the inhibitory activity of chimera A-Fc with various mutant Fc and variants with reduced FasL binding against membrane-type LIGHT-dependent CXCL10 production by intestinal myofibroblasts. The vertical axis shows CXCL10 concentration (ng / mL), and the horizontal axis shows the concentration of the added DcR3 variant (ng / mL). [Figure 36] Figure 36 shows the results of evaluating the neutralizing activity of various DcR3 variants against membrane-type TL1A. Figure 36 also shows the inhibitory activity of chimeric A-Fc and variants with reduced FasL binding, each containing various mutant Fc, against membrane-type TL1A-dependent IFN-γ production by human CD4+ T cells. The vertical axis shows IFN-γ concentration (pg / mL), and the horizontal axis shows the concentration of the added DcR3 variant (ng / mL). [Figure 37] Figure 37 shows the results of evaluating the neutralizing activity of various DcR3 variants against membrane-type FasL. It also shows the inhibitory activity of chimera A-Fc and variants with reduced FasL binding ability containing various mutant Fc against membrane-type FasL-dependent cell death in Jurkat cells. The vertical axis of the graph shows the percentage (%) of Annexin V-positive dead cells, and the horizontal axis shows the concentration (ng / mL) of each DcR3 variant added. [Figure 38A] Figure 38A shows the results of sandwich ELISA assay of the binding activity of various DcR3 variants to recombinant human LIGHT. The vertical axis shows the absorbance at 450 nm minus the absorbance at 570 nm, and the horizontal axis shows the concentration (ng / mL) of added recombinant human LIGHT. [Figure 38B]Figure 38B shows the results of sandwich ELISA of the binding activity of various DcR3 variants to LIGHT derived from human primary cells against soluble LIGHT in the culture supernatant of human T cells. The white bars represent the results of culture supernatant from human T cells cultured without stimulation, and the black bars represent the results of culture supernatant from human T cells stimulated with anti-CD3 and anti-CD28 antibodies. The vertical axis represents the absorbance at 450 nm minus the absorbance at 570 nm. [Figure 38C] Figure 38C shows the results of sandwich ELISA assay of the binding activity of various DcR3 variants to recombinant human TL1A. The vertical axis represents the absorbance at 450 nm minus the absorbance at 570 nm, and the horizontal axis represents the concentration of recombinant TL1A (ng / mL). [Figure 38D] Figure 38D shows the results of sandwich ELISA assay of the binding activity of various DcR3 variants to TL1A derived from human primary cells against soluble TL1A in the culture supernatant of human PBMCs. The white bars show the results for culture supernatants from unstimulated human PBMCs, while the black bars show the results for culture supernatants from human PBMCs stimulated with immune complexes. The vertical axis represents the absorbance at 450 nm minus the absorbance at 570 nm. [Figure 38E] Figure 38E shows the results of sandwich ELISA assay of the binding activity of various DcR3 variants to recombinant human FasL. The vertical axis shows the absorbance at 450 nm minus the absorbance at 570 nm, and the horizontal axis shows the human FasL concentration (ng / mL). [Figure 38F] Figure 38F shows the results of sandwich ELISA of the binding activity of various DcR3 variants to FasL derived from human primary cells against soluble FasL in the culture supernatant of AICD-induced human T cells. The white bars show the results using culture supernatant of human T cells without AICD induction, and the black bars show the results using culture supernatant of AICD-induced human T cells. The vertical axis shows the value obtained by subtracting the absorbance at 570 nm from the absorbance at 450 nm. [Figure 39]Figure 39 shows the values of the blood half-life (h) during the elimination phase and the area under the blood concentration-time curve up to infinity, AUC0-∞ (μg*h·mL), after various DcR3 variants were administered iv at 10 mg / kg to BALB / c mice. [Figure 40] Figure 40 shows the results of immunoblotting of various non-Fc-fused DcR3 variants transiently expressed in mammalian cells. Lanes 1 and 5 show a 30-fold concentrated culture supernatant from S195-His6 expression. Lanes 2 and 6 show a 6-fold diluted culture supernatant from chimera A-His6 expression. Lanes 3 and 7 show a 6-fold diluted culture supernatant from E57K-His6 expression. Lanes 4 and 8 show a 6-fold diluted culture supernatant from 45-18-His6 expression. These were electrophoresed under non-reducing or reducing conditions and detected by immunoblotting with an anti-6x-His tag antibody. DETAILED DESCRIPTION OF THE INVENTION

[0019] Preferred embodiments for carrying out the present invention will be described below. Note that the embodiment described below is an example of a typical embodiment of the present invention, and the scope of the present invention should not be construed as being narrow.

[0020] Two or more of the embodiments described below can be combined, and such combinations are also encompassed by the present invention.

[0021] The present invention relates to DcR3 variants, which are variants of wild-type DcR3. Specifically, the present invention relates to DcR3 variants that have DcR3 ligand binding activity (or neutralizing activity) and exhibit improved pharmacokinetics compared to wild-type DcR3 and / or reduced aggregation tendency compared to wild-type DcR3 when produced in mammalian cells, as well as DcR3 variants that contain a cysteine-rich domain in which one or more amino acid mutations have been introduced into the cysteine-rich domain of wild-type DcR3.

[0022] 1. Wild type DcR3 DcR3, also known as Decoy Receptor 3, DCR3, TNFRSF6B (Tumor necrosis factor receptor superfamily member 6B), TR6, or M68, is a soluble decoy receptor lacking a transmembrane domain and belonging to the TNF receptor superfamily. DcR3 binds to three ligands, LIGHT, TL1A, and FasL, competitively inhibiting their binding to the receptor and neutralizing the ligands. In addition to ligand neutralization, DcR3 also directly binds to glycosaminoglycans (GAGs), including heparan sulfate, on the cell membrane of monocytes, macrophages, and dendritic cells via its heparan sulfate-binding domain (HBD), resulting in various immunosuppressive and immunostimulatory effects [Biochemical. Pharmacology, 2011, 81:838-847; J. Immunol., 2006, 176:173-180].

[0023] Naturally occurring DcR3 has CRD1, CRD2, CRD3, CRD4, and HBD, in that order from the N-terminus. Naturally occurring DcR3 also has a region between CRD1 and CRD2, a region between CRD2 and CRD3, a region between CRD3 and CRD4, and a region between CRD4 and HBD. The cysteine-rich region of naturally occurring DcR3 is the region from the N-terminus of CRD1 to the C-terminus of CRD4, and includes CRD1, CRD2, CRD3, and CRD4, as well as the region between CRD1 and CRD2, the region between CRD2 and CRD3, and the region between CRD3 and CRD4. "Wild-type DcR3" refers to a molecule containing a cysteine-rich region and an HBD, where the cysteine-rich region and HBD are wild-type (i.e., the cysteine-rich region and HBD are identical to those of naturally occurring DcR3). Therefore, in addition to naturally occurring DcR3, "wild-type DcR3" also encompasses naturally occurring DcR3 variants, such as genetic polymorphisms and isoforms, as long as they contain the cysteine-rich region and HBD of naturally occurring DcR3. Furthermore, "wild-type DcR3" also encompasses immature DcR3 and mature DcR3. "Immature DcR3" refers to DcR3 with a signal peptide, and "mature DcR3" refers to DcR3 from which the signal peptide has been cleaved. The signal peptide may be a sequence derived from naturally occurring DcR3, an artificial sequence, a sequence derived from an expression vector, or a sequence derived from another protein suitable for the host cell in which naturally occurring DcR3 is expressed. Furthermore, when the cleavage site differs depending on the signal peptide used, sequences with different N-terminal amino acids in the mature form are also included in "wild-type DcR3." The CRD of naturally occurring DcR3 and the CRD of wild-type DcR3 are also referred to as "wild-type CRD," and the cysteine-rich region of naturally occurring DcR3 and the cysteine-rich region of wild-type DcR3 are also referred to as "wild-type cysteine-rich region."

[0024] The wild-type DcR3 of the present invention is not limited by its origin, and examples thereof include DcR3 derived from various eukaryotes. Examples include DcR3 derived from amphibians such as frogs, birds such as chickens, mammals such as primates including humans, and artiodactyls such as pigs and cows. When the DcR3 variants of the present invention are used in humans, it is preferable to use human-derived DcR3 as the wild-type DcR3.

[0025] The cDNA sequence of human DcR3 is shown in SEQ ID NO: 1, and the corresponding mRNA sequence has been registered with GenBank (NCBI, USA) under accession number NM_003823.3. The amino acid sequence of human DcR3 is shown in SEQ ID NO: 2 and has been registered with GenBank (NCBI, USA) under accession number NP_003814.1. Immature human DcR3 has a signal peptide at the N-terminus followed by four CRDs (CRD1, CRD2, CRD3, and CRD4) characteristic of the TNF receptor superfamily, and a basic amino acid-rich HBD at the C-terminus. Mature human DcR3 is the immature form from which the signal peptide has been truncated. The amino acid sequence of mature human DcR3 is shown, for example, in SEQ ID NO: 4, and the nucleotide sequence of DNA encoding the amino acid sequence of mature human DcR3 is shown, for example, in SEQ ID NO: 3. In the present invention, of the amino acid sequence of human DcR3 (SEQ ID NO: 2), the region from the N-terminus at positions 30 to 70 is defined as CRD1 (SEQ ID NO: 6), the region from positions 73 to 113 as CRD2 (SEQ ID NO: 8), the region from positions 115 to 150 as CRD3 (SEQ ID NO: 10), the region from positions 153 to 193 as CRD4 (SEQ ID NO: 12), and the region from positions 196 to 300 as HBD (SEQ ID NO: 48) (Figure 2). The nucleotide sequences of DNA encoding the amino acid sequences of CRD1, CRD2, CRD3, CRD4, and HBD of human DcR3 are represented by, for example, SEQ ID NOs: 5, 7, 9, 11, and 47, respectively. In addition, there are multiple definitions of the CRD amino acid sequence other than those described above, but any definition of the CRD amino acid sequence can be used for the DcR3 variants of the present invention using publicly known information [UniProt O95407, GenBank NP_003814.1, Structure, 2011, 19: p. 162-171].

[0026] As mentioned above, ligands for DcR3 include LIGHT, TL1A, and FasL, all of which belong to TNFSF.

[0027] LIGHT (lymphotoxin-like, exhibits inducible expression, and competes with herpes simplex virus (HSV) glycoprotein D (gD) for HVEM, a receptor expressed by T lymphocytes) is commonly referred to as TNFSF14 (Tumor necrosis factor superfamily member 14), LTg, HVEM-L, or CD258. The human LIGHT mRNA sequence and corresponding cDNA sequence are registered with GenBank (NCBI, USA) under accession number NM_003807.4, and the amino acid sequence is registered under accession number NP_003798.2. Soluble LIGHT is generated by shedding of the extracellular domain by protease after expression on the cell membrane as membrane-type LIGHT. The cleavage site in membrane-type LIGHT is between amino acids 82 and 83 of NP_003798.2. Both the soluble and membrane-type forms are functional.

[0028] Tumor necrosis factor (TNF)-like cytokine 1A (TL1A) is also commonly referred to as TNFSF15 (TNF superfamily member 15), TL1, VEGI, or VEGI-251. The mRNA sequence and corresponding cDNA sequence of human TL1A have been registered with GenBank (NCBI, USA) under accession number NM_005118.3, and the amino acid sequence under accession number NP_005109.2. Soluble TL1A is generated by shedding of the extracellular domain by protease after expression of membrane-type TL1A on the cell membrane. The cleavage site in membrane-type TL1A is between amino acids 71 and 72 of NP_005109.2, and both the soluble and membrane-type forms are functional.

[0029] FasL (Fas ligand) is also commonly referred to as FASLG, TNFSF6 (Tumor necrosis factor superfamily member 6), CD178, or APT1LG1. The mRNA sequence and corresponding cDNA sequence of human FasL have been registered with GenBank (NCBI, USA) under accession number NM_000639.2, and the amino acid sequence under accession number NP_000630.1. Soluble FasL is generated by shedding of the extracellular domain by proteases after expression of membrane-bound FasL on the cell membrane. The cleavage site of membrane-bound FasL is between amino acids 81 and 82 or between amino acids 129 and 130 of NP_000630.1. In vivo, membrane-bound FasL has been reported to be the predominant functional ligand.

[0030] Genes encoding proteins in eukaryotes often have genetic polymorphisms or isoforms. Genes used in the present invention that have mutations in their nucleotide or amino acid sequences due to such polymorphisms are also included in the genes encoding LIGHT, TL1A, or FasL in the present invention.

[0031] 2.DcR3 variants The DcR3 variants of the present invention contain a chimeric cysteine-rich region.

[0032] 2-1. Chimeric cysteine-rich region The chimeric cysteine-rich region of the present invention comprises an amino acid sequence in which one or more amino acid mutations have been introduced into the amino acid sequence of the cysteine-rich region of wild-type DcR3. The phrase "a mutation has been introduced into a certain amino acid sequence / base sequence" refers to the substitution, deletion, insertion, or addition of one or more amino acids / bases in the sequence. "Substitution, deletion, insertion, or addition" also encompasses a combination of two or more mutations selected from substitution, deletion, insertion, and addition. Mutations are introduced into at least one or more CRDs selected from CRD1, CRD2, CRD3, and CRD4 of the cysteine-rich region of wild-type DcR3. Mutations may or may not be introduced into the regions between CRD1 and CRD2, between CRD2 and CRD3, and between CRD3 and CRD4 of the cysteine-rich region of wild-type DcR3. When mutations are introduced into one or more regions selected from the region between CRD1 and CRD2, the region between CRD2 and CRD3, and the region between CRD3 and CRD4, the number of amino acids constituting each region after mutation introduction is typically 1 to 10, preferably 1 to 7, more preferably 1 to 5, even more preferably 1 to 3, and even more preferably 1 to 2. The amino acid sequence of each region after mutation introduction is not particularly limited. Mutations introduced into the cysteine-rich region of wild-type DcR3 include both naturally occurring mutations and artificial mutations. Examples of chimeric cysteine-rich regions of the present invention include chimeric cysteine-rich regions consisting of an amino acid sequence in which one or more amino acids have been substituted, deleted, inserted, or added to the amino acid sequence of the cysteine-rich region of wild-type DcR3. Examples of such chimeric cysteine-rich regions include the first and second chimeric cysteine-rich regions described below.

[0033] 2-1-1. First chimeric cysteine-rich region The first chimeric cysteine-rich region comprises an amino acid sequence of the cysteine-rich region of wild-type DcR3 in which at least a portion of the CRD of wild-type DcR3 is substituted with another peptide or protein, i.e., the first chimeric cysteine-rich region comprises an amino acid sequence derived from the cysteine-rich region of wild-type DcR3 and an amino acid sequence derived from another peptide or protein.

[0034] The portion of the cysteine-rich region of wild-type DcR3 that is substituted with another peptide or protein is preferably at least a portion of at least one CRD selected from CRD1, CRD2, CRD3, and CRD4. Thus, the portion of the cysteine-rich region of wild-type DcR3 that is substituted with another peptide or protein can be selected from all or a portion of CRD1, all or a portion of CRD2, all or a portion of CRD3, and all or a portion of CRD4.

[0035] The amino acid sequence of the first chimeric cysteine-rich region also includes an amino acid sequence in which, in addition to the amino acid sequence of the cysteine-rich region of wild-type DcR3, at least a portion of the CRD of wild-type DcR3 is substituted with another peptide or protein. The portion other than the CRD substituted with another peptide or protein can be selected from the region between CRD1 and CRD2, the region between CRD2 and CRD3, and the region between CRD3 and CRD4. The portion other than the CRD substituted with another peptide or protein may be a single portion or two or more portions. For example, the amino acid sequence of the first chimeric cysteine-rich region also includes an amino acid sequence in which the amino acid sequence of the region between CRD1 and CRD2, the region between CRD2 and CRD3, or the region between CRD3 and CRD4 is substituted with another peptide or protein.

[0036] The portion of the cysteine-rich region of wild-type DcR3 that is substituted with another peptide or protein may be a single portion or two or more portions. When at least a portion of a CRD in the cysteine-rich region of wild-type DcR3 is substituted, the substituted portion may be all or part of one CRD or all or part of multiple CRDs. However, it is more preferable that all or part of CRD2 and / or all or part of CRD3 of wild-type DcR3 involved in binding to LIGHT, TL1A, and FasL is retained, and all or part of the other CRDs are substituted with other peptides or proteins.

[0037] The other peptide or protein to be substituted may be either a natural or artificial peptide or protein, and examples include at least a portion of a CRD derived from a protein other than DcR3. In a preferred embodiment, the other peptide or protein to be substituted is at least a portion of a CRD of a TNF receptor superfamily (TNFRSF) molecule other than DcR3. That is, in a preferred embodiment, the first chimeric cysteine-rich region consists of an amino acid sequence in which at least a portion of the cysteine-rich domain of wild-type DcR3 is substituted with at least a portion of the cysteine-rich domain of a TNFRSF molecule other than DcR3. The at least a portion of the CRD of the TNFRSF molecule can be selected from all or a portion of CRD1, all or a portion of CRD2, all or a portion of CRD3, and all or a portion of CRD4. Therefore, the first chimeric cysteine-rich region may be a replacement of a portion of the CRD1 of wild-type DcR3 with a portion of the CRD1 of the TNFRSF molecule that corresponds to a portion of the CRD1 of wild-type DcR3, a replacement of the entire CRD1 of wild-type DcR3 with the entire CRD1 of the TNFRSF molecule, a replacement of a portion of the CRD2 of wild-type DcR3 with a portion of the CRD2 of the TNFRSF molecule that corresponds to a portion of the CRD2 of wild-type DcR3, a replacement of the entire CRD2 of wild-type DcR3 with the entire CRD2 of the TNFRSF molecule, It may have one or more substitutions selected from: a substitution of a portion of the CRD3 of wild-type DcR3 with a portion of the CRD3 of the TNFRSF molecule that corresponds to a portion of the CRD3 of wild-type DcR3; a substitution of the entire CRD3 of wild-type DcR3 with the entire CRD3 of the TNFRSF molecule; a substitution of a portion of the CRD4 of wild-type DcR3 with a portion of the CRD4 of the TNFRSF molecule that corresponds to a portion of the CRD4 of wild-type DcR3; and a substitution of the entire CRD4 of wild-type DcR3 with the entire CRD4 of the TNFRSF molecule.

[0038] Examples of the first chimeric cysteine-rich region include a chimeric cysteine-rich region in which the CRD1 of wild-type DcR3 is replaced with the CRD1 of the TNFRSF molecule (preferably, the other CRDs of wild-type DcR3 are retained in the chimeric cysteine-rich region), a chimeric cysteine-rich region in which the CRD4 of wild-type DcR3 is replaced with the CRD4 of the TNFRSF molecule (preferably, the other CRDs of wild-type DcR3 are retained in the chimeric cysteine-rich region), or a chimeric cysteine-rich region in which the CRD1 of wild-type DcR3 is replaced with the CRD1 of the TNFRSF molecule and the CRD4 of wild-type DcR3 is replaced with the CRD4 of the TNFRSF molecule (preferably, the other CRDs of wild-type DcR3 are retained in the chimeric cysteine-rich region). In addition to one or more of these substitutions, the first chimeric cysteine-rich region also includes a chimeric cysteine-rich region that further has a substitution of a portion of the CRD2 of wild-type DcR3 with a portion of the CRD2 of the TNFRSF molecule corresponding to that portion, and / or a substitution of a portion of the CRD3 of wild-type DcR3 with a portion of the CRD3 of the TNFRSF molecule corresponding to that portion.

[0039] There are 29 receptors in the human TNFRSF, and each receptor has a CRD in its N-terminal extracellular domain. Typically, each CRD consists of six Cys residues that form three disulfide bonds, and each receptor has one to four CRDs [Trends Biochem Sci, 2002.27:p-19-26.]

[0040] These TNFRSFs include DcR1(TNFRSF10C, TRAIL-R3, LIT, TRID, CD263), DcR2(TNFRSF10D, TRAIL-R4, TRUNDD,CD264), TNFR type I(TNFRSF1A, TNF-R, CD120a, TNFAR, TNF-R55, TNFR60), TNFR type II(TNFRSF1B, TNFBR, CD120b, TNFR80, p75, TNF-R75), LTBR(Lymphotoxin beta). receptor, TNFRSF3, TNFR III, TNFCR, TNFR-RP, TNFR2-RP, OX-40(TNFRSF4, ACT35, TXGP1L, CD134), CD40(TNFRSF5, Bp50, p50), Fas(Fas cell surface death). receptor, TNFRSF6, CD95, APO-1, APT1, FAS1, CD27(TNFRSF7, S152, Tp55), CD30(TNFRSF8, Ki-1), 4-1BB( TNFRSF9, CD137, ILA, DR4(TNFRSF10A, Apo2, TRAILR-1, CD261), DR5(TNFRSF10B, TRAIL-R2, KILLER, TRIC K2A, TRICKB, CD262, RANK(TNFRSF11A, CD265, FEO), FN14(TNFRSF12A, TweakR, CD266), TACI(TNFRSF13B). CD267、IGAD2、BAFFR(TNFRSF13C、CD268)、HVEM(TNFRSF14、ATAR、TR2、LIGHTR、HVEA、CD270)、NGFR(nerve growth factor receptor、TNFRSF16、p75NTR、CD271)、BCMA(TNFRSF17、BCM、CD269、TNFRSF13A)、GITR(TNFR SF18、AITR、CD357)、TROY(TNFRSF19、CROWN-ALPHA、CROWN、TRADE)、RELT(TNFRSF19L)、DR6(Death Receptor 6, TNFRSF21, CD358, DR3(Death Receptor 3, TNFRSF25, TRAMP, WSL-1, LARD, WSL-LR, DDR3, TR3, APO-3), EDAR(ectodysplasin A).receptor, ED3, DL, ED5, EDA3, Edar, ED1R, EDA1R), EDAR2R (ectodysplasin A2 receptor, XEDAR, EDAA2R, EDA-A2R, TNFRSF27), or osteoprotegerin (OPG, TNFRSF11B, TR1, OCIF).

[0041] The other peptide or protein that replaces at least a portion of the wild-type DcR3 in the first chimeric cysteine-rich region is not limited, but is particularly preferably OPG among TNFRSFs.

[0042] OPG is not limited to its origin, but examples thereof include OPG derived from various eukaryotes, such as OPG derived from amphibians such as frogs, birds such as chickens, mammals such as primates including humans, artiodactyls such as pigs and cows, and rodents including mice.

[0043] The cDNA sequence of human OPG is represented by SEQ ID NO: 13, and the corresponding mRNA sequence is registered with GenBank (NCBI, USA) under accession number NM_002546.3. The amino acid sequence of human OPG is represented by SEQ ID NO: 14, and the corresponding mRNA sequence is registered with GenBank (NCBI, USA) under accession number NP_002537.3.

[0044] Immature human OPG (SEQ ID NO: 14) has a signal peptide at the N-terminus. Mature human OPG is the immature form from which the signal peptide has been cleaved. The amino acid sequence of mature human OPG is represented, for example, by SEQ ID NO: 16, and the nucleotide sequence of DNA encoding the amino acid sequence of mature human OPG is represented, for example, by SEQ ID NO: 15. In the present invention, within the amino acid sequence of immature human OPG (SEQ ID NO: 14), the region from the N-terminus to the 22nd to 62nd positions is defined as CRD1 (SEQ ID NO: 18), the region from the 65th to 105th positions is defined as CRD2 (SEQ ID NO: 20), the region from the 107th to 142nd positions is defined as CRD3 (SEQ ID NO: 22), and the region from the 145th to 185th positions is defined as CRD4 (SEQ ID NO: 24). The nucleotide sequences of DNA encoding the amino acid sequences of CRD1, CRD2, CRD3, and CRD4 of human OPG are represented, for example, by SEQ ID NOs: 17, 19, 21, and 23, respectively. In addition, there are multiple definitions of the CRD amino acid sequence other than those described above, but any definition of the CRD amino acid sequence can be used for the DcR3 variants of the present invention using publicly known information [UniProt O00300, GenBank NP_002537.3].

[0045] Genes encoding proteins in eukaryotes often have genetic polymorphisms or isoforms. Genes used in the present invention that have mutations in their nucleotide or amino acid sequences due to such polymorphisms are also included in the genes encoding OPG of the present invention.

[0046] OPG binds to RANKL and neutralizes its activity, thereby suppressing osteoclast-mediated bone destruction [J. Immunol., 2012, 189: pp. 245-252]. OPG also binds to TRAIL and neutralizes its activity, thereby inhibiting TRAIL-mediated apoptosis [Am. J. Cancer. Res., 2012, 2: pp. 45-64]. Because neutralization of either ligand may cause undesired activity, it is desirable that the DcR3 variants of the present invention do not have neutralizing activity against either RANKL or TRAIL.

[0047] In a preferred embodiment, the first chimeric cysteine-rich region comprises or consists of the following amino acid sequence (a), (b), (c), or (d): (a) an amino acid sequence in which the CRD1 of wild-type DcR3 in the cysteine-rich region of wild-type DcR3 is substituted with the CRD1 of OPG (preferably, the other CRDs of wild-type DcR3 are retained in the amino acid sequence); (b) an amino acid sequence in which the CRD4 of wild-type DcR3 in the cysteine-rich region of wild-type DcR3 is replaced with the CRD4 of OPG (preferably, the other CRDs of wild-type DcR3 are retained in the amino acid sequence); (c) an amino acid sequence in the cysteine-rich region of wild-type DcR3 in which the CRD1 of wild-type DcR3 is replaced with the CRD1 of OPG and the CRD4 of wild-type DcR3 is replaced with the CRD4 of OPG (preferably, the other CRDs of wild-type DcR3 are retained in this amino acid sequence); (d) An amino acid sequence of (a), (b), or (c) above, in which a portion of the CRD2 of wild-type DcR3 is replaced with the corresponding portion of the CRD2 of OPG and / or a portion of the CRD3 of wild-type DcR3 is replaced with the corresponding portion of the CRD3 of OPG (preferably, the other CRDs of wild-type DcR3 are retained in the amino acid sequence).

[0048] An example of the amino acid sequence (d) above is an amino acid sequence in which the portion from positions 103 to 123 from the N-terminus of the amino acid sequence of (a), (b), or (c) is substituted with the corresponding portion of the amino acid sequence of the CRD of OPG. In this amino acid sequence, an amino acid sequence including the portion from positions 18 to 36 of the CRD3 of wild-type DcR3 and the two amino acid residues following the C-terminus of the amino acid sequence of CRD3 is substituted with the portion of OPG corresponding to said amino acid sequence.

[0049] A specific example of the amino acid sequence of (a) above is the amino acid sequence of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 26 or 50; a specific example of the amino acid sequence of (b) above is the amino acid sequence of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 28 or 52; a specific example of the amino acid sequence of (c) above is the amino acid sequence of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30 or 54; and a specific example of the amino acid sequence of (d) above is the amino acid sequence of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 32 or 56.

[0050] SEQ ID NO: 26 is the amino acid sequence of chimera B-HBD (a DcR3 variant in which CRD1 in wild-type DcR3 (SEQ ID NO: 4) was replaced with the CRD1 of OPG), SEQ ID NO: 28 is the amino acid sequence of chimera C-HBD (a DcR3 variant in which CRD4 in wild-type DcR3 (SEQ ID NO: 4) was replaced with the CRD4 of OPG), SEQ ID NO: 30 is the amino acid sequence of chimera A-HBD (a DcR3 variant in which CDR1 and CDR4 in wild-type DcR3 (SEQ ID NO: 4) were replaced with the CDR1 and CDR4 of OPG, respectively), SEQ ID NO: 32 is the amino acid sequence of 103-123OPG-HBD (a DcR3 variant in which the amino acid sequence including the 18th to 36th amino acids of CRD3 and the two amino acids at the C-terminus thereof was replaced with human OPG in chimera A-HBD), SEQ ID NO: 50 is the amino acid sequence of chimera B (wild-type DcR3), SEQ ID NO: 52 represents the amino acid sequence of chimera C (a DcR3 variant in which CRD4 in wild-type DcR3 (SEQ ID NO: 4) has been replaced with the CRD1 of OPG and the heparan sulfate binding region has been deleted); SEQ ID NO: 54 represents the amino acid sequence of chimera A (a DcR3 variant in which CDR1 and CDR4 in wild-type DcR3 (SEQ ID NO: 4) have been replaced with the CDR1 and CDR4 of OPG, respectively, and the heparan sulfate binding region has been deleted); and SEQ ID NO: 56 represents the amino acid sequence of 103-123OPG (a DcR3 variant in which the amino acid sequence containing positions 18 to 36 of CRD3 and two amino acids at the C-terminus thereof has been replaced with human OPG in chimera A).

[0051] In a preferred embodiment, DcR3 variants containing a first chimeric cysteine-rich region include DcR3 variants that have binding activity for at least one of LIGHT, TL1A, and FasL, DcR3 variants that have binding activity for all of LIGHT, TL1A, and FasL, DcR3 variants that have no binding activity for FasL and have binding activity for either LIGHT or TL1A, or DcR3 variants that have no binding activity for FasL and have binding activity for LIGHT and TL1A.

[0052] In the present invention, the term "having binding activity to a ligand" is used to mean that the binding activity of a DcR3 variant comprising a first chimeric cysteine-rich region to the ligand is equivalent to and not significantly reduced compared to the binding activity of wild-type DcR3 to the ligand, and also means that the binding activity is significantly enhanced compared to the binding activity of wild-type DcR3 to the ligand. For example, when measured by surface plasmon resonance (SPR), the dissociation constant (K D If the difference is less than 3 times, it can be determined that the antibody has binding activity to the ligand.

[0053] In the present invention, the expression "has no binding activity to a ligand" refers to a DcR3 variant comprising a first chimeric cysteine-rich region, and includes both undetectable and significantly reduced binding activity to the ligand compared with that of wild-type DcR3. For example, when measured by SPR, if the KD value of a DcR3 variant is more than three times that of wild-type DcR3, or if the Rmax of a DcR3 variant is less than 5, the binding activity to the ligand is determined to be significantly reduced, and the DcR3 variant can be defined as having no binding activity to the ligand.

[0054] In a particularly preferred embodiment, the DcR3 variant containing the first chimeric cysteine-rich region includes a DcR3 variant with reduced binding activity to FasL. The term "FasL-reduced binding variant" refers to a DcR variant containing a chimeric cysteine-rich region that has no binding activity to FasL but has binding activity to one or more of LIGHT and TL1A, or a DcR3 variant containing a chimeric cysteine-rich region that has no binding activity to FasL but has binding activity to LIGHT and TL1A.

[0055] In a preferred embodiment, the DcR3 variant of the present invention is a DcR3 variant that has neutralizing activity against at least one of LIGHT, TL1A, and FasL, a DcR3 variant that has neutralizing activity against all of LIGHT, TL1A, and FasL, a DcR3 variant that does not have neutralizing activity against FasL and has neutralizing activity against either LIGHT or TL1A, or a DcR3 variant that does not have neutralizing activity against FasL and has neutralizing activity against LIGHT and TL1A.

[0056] In the present invention, the expression "neutralizing activity against a certain ligand" includes the inhibition of the binding of the ligand to a receptor on the cell membrane surface by binding to a DcR3 variant, and the inhibition of the binding of the ligand to a receptor on the cell membrane surface by inhibiting the binding of the ligand to a receptor on the cell membrane surface, i.e., the inhibition of the biological activity of the ligand (e.g., biological activities such as cytokine production, proliferation promotion, and apoptosis induction in cells).

[0057] In the present invention, the expression "having neutralizing activity" of a DcR3 variant is used to mean that the neutralizing activity of the DcR3 variant against the ligand is not significantly different from the neutralizing activity of wild-type DcR3 against the ligand, and that the neutralizing activity is significantly enhanced compared to the neutralizing activity of wild-type DcR3 against the ligand.

[0058] In the present invention, the expression "having no neutralizing activity" of a DcR3 variant is used to mean that the neutralizing activity of the DcR3 variant against the ligand is significantly reduced compared to the neutralizing activity of wild-type DcR3 against the ligand.

[0059] In a particularly preferred embodiment, the DcR3 variant of the present invention is a DcR3 variant with reduced binding ability to FasL. The term "mutant with reduced FasL binding ability" refers to a DcR3 variant that does not have neutralizing activity against FasL but has neutralizing activity against one or more of LIGHT and TL1A, or a DcR3 variant that does not have neutralizing activity against FasL but has neutralizing activity against LIGHT and TL1A.

[0060] 2-1-2. Second chimeric cysteine-rich region The second chimeric cysteine-rich region has the amino acid sequence of the first chimeric cysteine-rich region, but with 1 to 30 amino acids deleted, substituted, inserted or added.

[0061] Methods for obtaining a polypeptide having an amino acid sequence in which one or more amino acids are deleted, substituted, inserted, or added in the amino acid sequence of the first chimeric cysteine-rich region include site-directed mutagenesis [Molecular Cloning, A Laboratory Manual, Second Edition, Cold Spring Harbor Laboratory Press (1989), Current Protocols in Molecular Biology, John Wiley & Sons (1987-1997), Nucleic Acids Research, 10, 6487 (1982), Proc. Natl. Acad. Sci. USA., 79, 6409, (1982), Gene, 34, 315 (1985), Proc. Natl. Acad. Sci. USA., 82, 488 (1985)].

[0062] Mutations (modifications) made to the first chimeric cysteine-rich region include both natural mutations and artificial amino acid substitutions, deletions, insertions, or additions. The amino acid sequence of the second chimeric cysteine-rich region may be an amino acid sequence obtained by substituting, deleting, inserting, or adding one or more, preferably 1 to 30, more preferably 1 to 10, even more preferably 1 to 5, and even more preferably 1 to 3 amino acids in the amino acid sequence of the first chimeric cysteine-rich region, or an amino acid sequence having 80% or more, preferably 85% or more, more preferably 90% or more, for example, 93% or more, 95% or more, 97% or more, 98% or more, or 99% or more identity to the amino acid sequence of the first chimeric cysteine-rich region. To describe amino acid substitutions, for example, a substitution of Ser for Asn at the 131st position from the N-terminus of the amino acid sequence to be substituted can be represented as N131S.

[0063] In a preferred embodiment, the second chimeric cysteine-rich region comprises or consists of the amino acid sequence of (e) below: (e) An amino acid sequence in which 1 to 30 amino acids are deleted, substituted, inserted or added in the amino acid sequence of (a), (b), (c) or (d) above.

[0064] Examples of mutations (modifications) that can be made to the first chimeric cysteine-rich region include the addition or deletion of a glycosylation site, which can control the biological activity or properties of the DcR3 variant of the present invention, its pharmacokinetics such as blood half-life, or its physical or chemical properties such as protein stability.

[0065] Glycosylation generally refers to the formation of an N-glycosidic linkage of a sugar chain to an asparagine residue of a peptide or protein, and / or an O-glycosidic linkage of a sugar chain to a serine or threonine residue of the peptide or protein. Examples of O-glycans added to DcR3 variants include core 1 and core 2, and examples of N-glycans include high-mannose, hybrid, or complex glycans, with complex glycans being preferred.

[0066] Examples of mutations (modifications) that can be made to the first chimeric cysteine-rich region include substituting at least one amino acid in the amino acid sequence of the first chimeric cysteine-rich region with an amino acid to which a sugar chain can be attached via an N-glycosidic or O-glycosidic bond, and adding a sugar chain; it is particularly preferable to add an N-glycosidic sugar chain.

[0067] Furthermore, the present invention also encompasses, for example, the removal of sugar chains by substituting amino acids involved in N-glycosidic bonds at at least one or more positions, preferably two or more positions, in the amino acid sequence of a chimeric cysteine-rich region with other amino acids. Generally, when a peptide or protein is expressed in yeast, insect cells, or mammalian cells, an Asn-X-Thr / Ser sequence (where X is any amino acid residue other than Pro) is recognized to form an N-glycosidic bond in the sugar chain. For example, N-glycosidic sugar chains can be removed by substituting Asn, Ser, or Thr in the Asn-X-Thr / Ser sequence present in a DcR3 variant with other amino acids.

[0068] As the second chimeric cysteine-rich region, in order to reduce aggregation of the DcR3 variant of the present invention, a chimeric cysteine-rich region having a glycan N-glycosidically linked to Asn at position 157 from the N-terminus of the amino acid sequence of the first chimeric cysteine-rich region (e.g., the amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of SEQ ID NO: 30) is particularly preferred.

[0069] In a preferred embodiment, the amino acid sequence of the second chimeric cysteine-rich region from which the glycosylation site has been removed (one embodiment of the amino acid sequence of (e) above) is (f) substitution of Asn at the 131st and 144th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) above (for example, an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence shown in SEQ ID NO: 28, 30, 32, 52, 54, or 56) with other amino acids; (g) substitution of Asn at the 131st, 144th, and 157th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) above (for example, an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence shown in SEQ ID NO: 28, 30, 32, 52, 54, or 56) with other amino acids; (h) substitution of Thr at position 133 and Ser at position 146 from the N-terminus of the amino acid sequence of (b), (c), or (d) above (for example, an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence shown in SEQ ID NO: 28, 30, 32, 52, 54, or 56) with other amino acids; and (i) Substitution of Thr at position 133, Ser at position 146, and Thr at position 159 from the N-terminus of the amino acid sequence of (b), (c), or (d) above (e.g., an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence shown in SEQ ID NO: 28, 30, 32, 52, 54, or 56) with other amino acids and having a substitution selected from:

[0070] In a more preferred embodiment, the amino acid sequence from which the glycosylation site has been removed (one embodiment of the amino acid sequence of (e) above) is (f') substitution of Asn at the 131st and 144th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) above (e.g., an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence shown in SEQ ID NO: 28, 30, 32, 52, 54, or 56) with Ser; (g') substitution of Asn at the 131st, 144th, and 157th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) above (e.g., an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence shown in SEQ ID NO: 28, 30, 32, 52, 54, or 56) with Ser; (h') substitution of Thr at the 133rd position and Ser at the 146th position from the N-terminus of the amino acid sequence of (b), (c), or (d) above (for example, an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence shown in SEQ ID NO: 28, 30, 32, 52, 54, or 56) with Ala, and (i') Substitution of Thr at position 133, Ser at position 146, and Thr at position 159 from the N-terminus of the amino acid sequence of (b), (c), or (d) above (for example, an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence shown in SEQ ID NO: 28, 30, 32, 52, 54, or 56) with Ala and having a substitution selected from:

[0071] Examples of amino acid sequences having the above-mentioned (f') substitution include an amino acid sequence in which Asn at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30 or SEQ ID NO: 54 has been substituted with Ser for Asn at positions 131 and 144 from the N-terminus (N131S / N144S) (an amino acid sequence in which Asn at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 34 or SEQ ID NO: 58 has been substituted with Ser).

[0072] SEQ ID NO: 34 represents the amino acid sequence of N131S / N144S-HBD (a DcR3 variant in which Asn at positions 131 and 144 from the N-terminus in chimera A-HBD has been replaced with Ser), and SEQ ID NO: 58 represents the amino acid sequence of N131S / N144S (a DcR3 variant in which Asn at positions 131 and 144 from the N-terminus in chimera A has been replaced with Ser).

[0073] Examples of amino acid sequences having the above-mentioned (h') substitution include an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus in the amino acid sequence set forth in SEQ ID NO: 30 or SEQ ID NO: 54, in which Thr at the 133rd position and Ser at the 146th position from the N-terminus are substituted with Ala (T133A / S146A) (an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus in the amino acid sequence set forth in SEQ ID NO: 36 or SEQ ID NO: 60).

[0074] SEQ ID NO: 36 represents the amino acid sequence of T133A / S146A-HBD (a DcR3 variant in which Thr at position 133 and Ser at position 146 from the N-terminus in chimera A-HBD are replaced with Ala), and SEQ ID NO: 60 represents the amino acid sequence of T133A / S146A (a DcR3 variant in which Thr at position 133 and Ser at position 146 from the N-terminus in chimera A are replaced with Ala).

[0075] Examples of amino acid sequences having the above-mentioned (g') substitution include an amino acid sequence in which Asn at positions 1 to 164 from the N-terminus in the amino acid sequence set forth in SEQ ID NO: 30 or SEQ ID NO: 54 has been substituted with Ser for Asn at positions 131, 144, and 157 from the N-terminus (N131S / N144S / N157S) (an amino acid sequence in which amino acids at positions 1 to 164 from the N-terminus in the amino acid sequence set forth in SEQ ID NO: 38 or SEQ ID NO: 62).

[0076] SEQ ID NO: 38 represents the amino acid sequence of N131S / N144S / N157S-HBD (a DcR3 variant in which Asn at positions 131, 144, and 157 from the N-terminus in chimera A-HBD has been replaced with Ser), and SEQ ID NO: 62 represents the amino acid sequence of N131S / N144S / N157S (a DcR3 variant in which Asn at positions 131, 144, and 157 from the N-terminus in chimera A has been replaced with Ser).

[0077] Examples of amino acid sequences having the substitution (i') above include amino acid sequences in which Thr at position 133, Ser at position 146, and Thr at position 159 from the N-terminus are substituted with Ala (T133A / S146A / T159A) in the amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30 or SEQ ID NO: 54 (amino acid sequences consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 40 or SEQ ID NO: 64).

[0078] Sequence number 40 represents the amino acid sequence of T133A / S146A / T159A-HBD (a DcR3 variant in which Thr at position 133 from the N-terminus, Ser at position 146, and Thr at position 159 from the N-terminus in chimera A-HBD are replaced with Ala), and sequence number 64 represents the amino acid sequence of T133A / S146A / T159A (a DcR3 variant in which Thr at position 133 from the N-terminus, Ser at position 146, and Thr at position 159 from the N-terminus in chimera A are replaced with Ala).

[0079] For example, mutations (modifications) made to the first chimeric cysteine-rich region can be made in CRD1 and / or CRD4 of the first chimeric cysteine-rich region without mutations in CRD2 and CRD3 of the first chimeric cysteine-rich region involved in the binding of LIGHT, TL1A, and FasL. This allows for the production of DcR3 variants containing a chimeric cysteine-rich region that maintains the binding activity of the chimeric cysteine-rich region to LIGHT, TL1A, and FasL. On the other hand, mutations in CRD2 and / or CRD3 of the first chimeric cysteine-rich region involved in the binding of LIGHT, TL1A, or FasL allow for the production of DcR3 variants containing a chimeric cysteine-rich region that alters the binding activity of the chimeric cysteine-rich region to LIGHT, TL1A, or FasL. That is, by introducing the above-mentioned mutations, it is possible to obtain DcR3 variants containing chimeric cysteine-rich regions that have the desired binding properties to LIGHT, TL1A, or FasL.

[0080] In a preferred embodiment, the DcR3 variant containing the second chimeric cysteine-rich region includes a DcR3 variant containing a chimeric cysteine-rich region that has binding activity for at least one of LIGHT, TL1A, and FasL, a DcR3 variant containing a chimeric cysteine-rich region that has binding activity for all of LIGHT, TL1A, and FasL, a DcR3 variant containing a chimeric cysteine-rich region that has no binding activity for FasL and has binding activity for either LIGHT or TL1A, or a DcR3 variant containing a chimeric cysteine-rich region that has no binding activity for FasL and has binding activity for LIGHT and TL1A, etc. The meanings of the expressions "having binding activity for a ligand" and "not having binding activity for a ligand" regarding the DcR3 variants of the present invention are as described above.

[0081] In a particularly preferred embodiment, the DcR3 variant containing the second chimeric cysteine-rich region includes a DcR3 variant with reduced binding activity to FasL. The term "variant with reduced FasL binding activity" refers to a DcR3 variant containing a chimeric cysteine-rich region that has no binding activity to FasL and has binding activity to one or more of LIGHT and TL1A, or a DcR3 variant containing a chimeric cysteine-rich region that has no binding activity to FasL and has binding activity to LIGHT and TL1A.

[0082] In a preferred embodiment, the DcR3 variant containing the second chimeric cysteine-rich region includes a DcR3 variant containing a chimeric cysteine-rich region that has neutralizing activity against at least one of LIGHT, TL1A, and FasL, a DcR3 variant containing a chimeric cysteine-rich region that has neutralizing activity against all of LIGHT, TL1A, and FasL, a DcR3 variant containing a chimeric cysteine-rich region that has no neutralizing activity against FasL but has neutralizing activity against either LIGHT or TL1A, or a DcR3 variant containing a chimeric cysteine-rich region that has no neutralizing activity against FasL but has neutralizing activity against LIGHT and TL1A, etc. The meanings of the expressions "having neutralizing activity against a ligand" and "not having neutralizing activity against a ligand" regarding the DcR3 variants of the present invention are as described above.

[0083] In a particularly preferred embodiment, the DcR3 variant containing the second chimeric cysteine-rich region includes a DcR3 variant with reduced neutralizing activity against FasL. The term "FasL-binding-reduced variant" refers to a DcR3 variant containing a chimeric cysteine-rich region that does not have neutralizing activity against FasL but has neutralizing activity against one or more of LIGHT and TL1A, or a DcR3 variant containing a chimeric cysteine-rich region that does not have neutralizing activity against FasL but has neutralizing activity against LIGHT and TL1A.

[0084] For example, DcR3 variants containing a chimeric cysteine-rich region that have reduced FasL-binding ability can be obtained by preparing modified DcR3 or DcR3 variants in which the ligand-binding site of each DcR3 or DcR3 variant, as predicted by crystal structure analysis, is substituted with Ala or another amino acid, and then measuring the binding activity and neutralizing activity for LIGHT, TL1A, or FasL ligand. Alternatively, DcR3 or DcR3 variants can be obtained by preparing a gene library in which the amino acids surrounding the ligand-binding site are randomly substituted with other amino acids, displaying the library on phage, yeast, mammalian cells, etc., and screening using the binding activity and neutralizing activity for LIGHT, TL1A, or FasL ligand as indicators.

[0085] The amino acid sequence of the chimeric cysteine-rich region contained in the FasL-binding-reducing variant (one form of the amino acid sequence (e) above) includes, for example, an amino acid sequence in which one or more amino acids selected from the group consisting of Glu at position 57, Arg at position 58, and Arg at position 60 from the N-terminus of the amino acid sequence of (a), (b), (c), or (d) above are substituted with other amino acids.

[0086] The amino acid to be substituted for Glu at position 57 is not particularly limited and can be appropriately selected from 19 types of amino acids excluding Glu out of 20 types of amino acids (Glu, Ala, Asp, Lys, Leu, Cys, Phe, Gly, His, Ile, Met, Asn, Pro, Gln, Arg, Ser, Thr, Val, Trp, Tyr). Preferably, the amino acid is selected from Lys, Leu, Arg, Val, Ala, Phe, His, Ile, and Met, more preferably from Lys, Leu, Arg, and Val, even more preferably from Lys, Arg, and Val, and even more preferably from Lys and Arg.

[0087] The amino acid to be substituted for Arg at position 58 is not particularly limited and can be appropriately selected from 19 types of amino acids excluding Arg out of the 20 types of amino acids, but is preferably selected from Asp, Glu, and Thr, and more preferably selected from Asp and Glu.

[0088] The amino acid to be substituted for Arg at position 60 is not particularly limited and can be appropriately selected from 19 types of amino acids excluding Arg among the 20 types of amino acids, but is preferably Lys.

[0089] In a preferred embodiment, the one or more amino acids selected from the group consisting of Glu at position 57, Arg at position 58, and Arg at position 60 are one amino acid consisting of Glu at position 57.

[0090] In yet another preferred embodiment, the one or more amino acids selected from the group consisting of Glu at position 57, Arg at position 58, and Arg at position 60 are the two amino acids Glu at position 57 and Arg at position 58. In this embodiment, it is preferable to combine substitution of Glu at position 57 with Lys, Leu, Arg, Val, Ala, Phe, His, Ile, or Met with substitution of Arg at position 58 with Asp, Glu, or Thr, it is more preferable to combine substitution of Glu at position 57 with Lys, Leu, Arg, or Val with substitution of Arg at position 58 with Asp or Glu, and it is even more preferable to combine substitution of Glu at position 57 with Lys or Arg with substitution of Arg at position 58 with Asp or Glu.

[0091] The second chimeric cysteine-rich region may have one or more amino acids substituted with other amino acids other than Glu at position 57, Arg at position 58, and Arg at position 60. Examples of amino acids other than Glu at position 57, Arg at position 58, and Arg at position 60 include Trp at position 53, Asn at position 54, Tyr at position 55, and Leu at position 56 from the N-terminus of the amino acid sequence of (a), (b), (c), or (d) above.

[0092] The amino acid to be substituted for Trp at position 53 is not particularly limited and can be appropriately selected from 19 amino acids excluding Trp out of the 20 amino acids, preferably Asp and Asn. Substitution of Trp at position 53 with another amino acid can be combined with, for example, substitution of Glu at position 57 with another amino acid.

[0093] The amino acid to be substituted for Asn at position 54 is not particularly limited and can be appropriately selected from 19 amino acids excluding Asn out of the 20 amino acids, but is preferably Asp. Substitution of Asn at position 54 with another amino acid can be combined with, for example, substitution of Glu at position 57 with another amino acid.

[0094] The amino acid to be substituted for Tyr at position 55 is not particularly limited and can be appropriately selected from 19 types of amino acids excluding Tyr out of the 20 types of amino acids, but is preferably selected from Thr, Asp, Gln, and Glu. Substitution of Tyr at position 55 with another amino acid can be combined with, for example, substitution of Glu at position 57 with another amino acid.

[0095] The amino acid to be substituted for Leu at position 56 is not particularly limited and can be appropriately selected from 19 types of amino acids excluding Leu out of the 20 types of amino acids, but is preferably selected from Asp, Gln, Thr, Glu, Gly, Asn, and Pro. Substitution of Leu at position 56 with another amino acid can be combined with, for example, substitution of Glu at position 57 with another amino acid.

[0096] Specific examples of the amino acid sequence having one or more amino acids selected from the group consisting of Glu at the 57th position, Arg at the 58th position, and Arg at the 60th position from the N-terminus of the amino acid sequence of (a), (b), (c), or (d) above substituted with other amino acids include: an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with Lys in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 42); an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with Lys in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 66); an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with Leu in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 44); an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with Leu in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 68); an amino acid sequence in which Arg at the 60th position from the N-terminus is substituted with Lys in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 46); an amino acid sequence in which Arg at the 60th position from the N-terminus is substituted with Lys in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 70); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Arg; an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with Arg in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 180); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Val; an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54, in which Glu at the 57th position from the N-terminus is substituted with Val (an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 182); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Ala; an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with Ala in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 270); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Phe; an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with Phe in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 272); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with His; an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with His in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 274); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Ile; an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54, in which Glu at the 57th position from the N-terminus is substituted with Ile (an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 276); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Met; an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54, in which Glu at the 57th position from the N-terminus is substituted with Met (an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 278); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Lys and Arg at the 58th position is substituted with Asp; an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with Lys and Arg at the 58th position from the N-terminus is substituted with Asp in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 184); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Lys and Arg at the 58th position is substituted with Glu; an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54, in which Glu at the 57th position from the N-terminus is substituted with Lys and Arg at the 58th position is substituted with Glu (an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 186); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Arg and Arg at the 58th position is substituted with Asp; an amino acid sequence in which Glu at the 57th position from the N-terminus is substituted with Arg and Arg at the 58th position from the N-terminus is substituted with Asp in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 188); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Lys and Arg at the 58th position is substituted with Thr; an amino acid sequence in which Glu at position 57 from the N-terminus is substituted with Lys and Arg at position 58 from the N-terminus is substituted with Thr in the amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54 (an amino acid sequence consisting of amino acids at positions 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 280); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Leu and Arg at the 58th position is substituted with Glu; an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54, in which Glu at the 57th position from the N-terminus is substituted with Leu and Arg at the 58th position is substituted with Glu (an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 282); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Val and Arg at the 58th position is substituted with Thr; an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54, in which Glu at the 57th position from the N-terminus is substituted with Val and Arg at the 58th position is substituted with Thr (an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 284); an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30, in which Glu at the 57th position from the N-terminus is substituted with Val and Arg at the 58th position is substituted with TGlu; An amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 54, in which Glu at the 57th position from the N-terminus is substituted with Val and Arg at the 58th position is substituted with Glu (an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 286). etc.

[0097] SEQ ID NO: 42 is the amino acid sequence of chimera A-E57K-HBD (a DcR3 variant in which Glu at the 57th position from the N-terminus in chimera A-HBD has been replaced with Lys), SEQ ID NO: 44 is the amino acid sequence of chimera A-E57L-HBD (a DcR3 variant in which Glu at the 57th position from the N-terminus in chimera A-HBD has been replaced with Leu), SEQ ID NO: 46 is the amino acid sequence of chimera A-R60K-HBD (a DcR3 variant in which Arg at the 60th position from the N-terminus in chimera A-HBD has been replaced with Lys), and SEQ ID NO: 66 is the amino acid sequence of chimera A-E57K (a DcR3 variant in which N SEQ ID NO: 68 is the amino acid sequence of chimera A-E57L (a DcR3 variant in which Glu at the 57th position from the N-terminus is substituted with Leu in chimera A); SEQ ID NO: 70 is the amino acid sequence of chimera A-R60K (a DcR3 variant in which Arg at the 60th position from the N-terminus is substituted with Lys in chimera A); SEQ ID NO: 180 is the amino acid sequence of chimera A-E57R (a DcR3 variant in which Glu at the 57th position from the N-terminus is substituted with Arg in chimera A); SEQ ID NO: 182 is the amino acid sequence of chimera A-E57V SEQ ID NO: 184 is the amino acid sequence of chimera A-E57K_R58D (a DcR3 variant in chimera A in which Glu at the 57th position from the N-terminus is substituted with Lys and Arg at the 58th position from the N-terminus is substituted with Asp). SEQ ID NO: 186 is the amino acid sequence of chimera A-E57K_R58E (a DcR3 variant in chimera A in which Glu at the 57th position from the N-terminus is substituted with Lys and Arg at the 58th position from the N-terminus is substituted with Glu). SEQ ID NO: 188 is the amino acid sequence of chimera A-E57R_R58D (a DcR3 variant in chimera A in which Glu at the 57th position from the N-terminus is substituted with Val). SEQ ID NO: 270 is the amino acid sequence of chimera A-E57A (a DcR3 variant in which Glu at the 57th position from the N-terminus is substituted with Arg and Arg at the 58th position from the N-terminus is substituted with Asp); SEQ ID NO: 272 is the amino acid sequence of chimera A-E57F (a DcR3 variant in which Glu at the 57th position from the N-terminus is substituted with Phe in chimera A); SEQ ID NO: 274 is the amino acid sequence of chimera A-E57H (a DcR3 variant in which Glu at the 57th position from the N-terminus is substituted with His in chimera A);SEQ ID NO: 276 is the amino acid sequence of chimera A-E57I (a DcR3 variant in which Glu at the 57th position from the N-terminus in chimera A is substituted with Ile), SEQ ID NO: 278 is the amino acid sequence of chimera A-E57M (a DcR3 variant in which Glu at the 57th position from the N-terminus in chimera A is substituted with Met), SEQ ID NO: 280 is the amino acid sequence of chimera A-E57K_R58T (a DcR3 variant in which Glu at the 57th position from the N-terminus in chimera A is substituted with Lys and Arg at the 58th position from the N-terminus in chimera A is substituted with Thr), SEQ ID NO: 282 is the amino acid sequence of chimera A-E57K_R SEQ ID NO: 284 represents the amino acid sequence of chimera A-E57V_R58T (a DcR3 variant in which Glu at position 57 from the N-terminus is substituted with Lys and Arg at position 58 from the N-terminus is substituted with Thr). SEQ ID NO: 286 represents the amino acid sequence of chimera A-E57V_R58E (a DcR3 variant in which Glu at position 57 from the N-terminus is substituted with Val and Arg at position 58 from the N-terminus is substituted with Glu).

[0098] 2-2. Other areas The DcR3 variants of the present invention may or may not contain one or more additional regions attached to the C-terminus of the first or second chimeric cysteine-rich region. Examples of additional regions include part or all of the region between the CRD4 and HBD in wild-type DcR3, part or all of the region following the C-terminus of the CRD4 amino acid sequence in TNF receptor superfamily (TNFRSF) molecules other than DcR3, and part or all of the HBD in wild-type DcR3. The expression "another region attached to the C-terminus of the first or second chimeric cysteine-rich region" is used to encompass cases where the additional region is attached directly to the C-terminus of the first or second chimeric cysteine-rich region, as well as cases where the additional region is attached to the C-terminus of the first or second chimeric cysteine-rich region via another additional region.

[0099] In one embodiment, the DcR3 variant of the present invention comprises a part or all of the region located between CRD4 and HBD in wild-type DcR3 as the additional region linked to the C-terminus of the first or second chimeric cysteine-rich region, and in this embodiment, the additional region is preferably directly linked to the C-terminus of the first or second chimeric cysteine-rich region.

[0100] In another embodiment, the DcR3 variant of the present invention comprises, as the additional region linked to the C-terminus of the first or second chimeric cysteine-rich region, part or all of the region following the C-terminus of the amino acid sequence of CRD4 in a TNF receptor superfamily (TNFRSF) molecule other than DcR3. In this embodiment, the additional region is preferably linked directly to the C-terminus of the first or second chimeric cysteine-rich region. The description of TNFRSF is the same as above. TNFRSF is preferably OPG.

[0101] When the CRD4 of wild-type DcR3 is replaced with the CRD4 of OPG in the first or second chimeric cysteine-rich region (e.g., a first chimeric cysteine-rich region consisting of the amino acid sequences (b) to (d) above), the DcR3 variant of the present invention preferably comprises, as another region linked to the C-terminus of the chimeric cysteine-rich region, multiple amino acid residues subsequent to the C-terminus of the amino acid sequence of OPG CRD4. That is, when the CR4 of wild-type DcR3 is replaced with the CRD4 of OPG, the multiple amino acid residues subsequent to the C-terminus of the amino acid sequence of wild-type DcR3 are also preferably replaced with multiple amino acid residues subsequent to the C-terminus of the amino acid sequence of OPG CRD4. Preferred examples of the multiple amino acid residues subsequent to the C-terminus of the amino acid sequence of OPG CRD4 include, for example, amino acid residues 186 to 194 of the amino acid sequence of OPG (SEQ ID NO: 14), although the number of amino acid residues substituted with the multiple amino acid residues subsequent to the C-terminus of the CRD4 of wild-type DcR3 can be adjusted as appropriate. The number of amino acid residues following the C-terminus of the amino acid sequence of OPG's CRD4 is typically 1 to 12, preferably 1 to 10, and more preferably 1 to 9, 1 to 6, or 1 to 3. Furthermore, when the CRD4 in the first or second chimeric cysteine-rich region (e.g., a first chimeric cysteine-rich region consisting of the amino acid sequences (a) and (d) above) is derived from wild-type DcR3, the C-terminus of the amino acid sequence of CRD4 in the first chimeric cysteine-rich region is preferably linked to an amino acid sequence following the C-terminus of the amino acid sequence of CRD4 of wild-type DcR3. An example of such an amino acid sequence is amino acids 194 to 195 of SEQ ID NO: 2, although the number of amino acid residues constituting the amino acid sequence following the C-terminus of the amino acid sequence of CRD4 of wild-type DcR3 can be adjusted as appropriate. The number of amino acid residues constituting the amino acid sequence following the C-terminus of CRD4 of wild-type DcR3 is usually 1 to 10, preferably 1 to 5, more preferably 1 to 3, and even more preferably 1 to 2.

[0102] In yet another embodiment, the DcR3 variant of the present invention comprises part or all of the HBD of wild-type DcR3 as the other region linked to the C-terminus of the first or second chimeric cysteine-rich region. In this embodiment, the other region may be linked directly to the C-terminus of the first or second chimeric cysteine-rich region, or may be linked to the C-terminus of the first or second chimeric cysteine-rich region via part or all of the region present between the CRD4 and HBD in wild-type DcR3, or via part or all of the region following the C-terminus of the amino acid sequence of CRD4 in a TNFRSF molecule other than DcR3.

[0103] It is preferable that the DcR3 variant of the present invention does not include HBD as the other region bound to the C-terminal side of the first or second chimeric cysteine-rich region, and instead includes amino acid residues 186 to 194 of the amino acid sequence of OPG (sequence number 14).

[0104] 3. Modified DcR3 containing glycans The DcR3 variants of the present invention also include DcR3 variants containing at least one sugar chain. As long as at least one sugar chain is bound to the cysteine-rich region or any other amino acid residue contained in the above-mentioned DcR3 variants, any DcR3 variant containing any sugar chain is included in the present invention.

[0105] A glycoprotein has one or more sugar chains. When a glycoprotein has two or more sugar chains, the glycoprotein may have one type of sugar chain or two or more types of sugar chains. Examples of sugar chains that glycoproteins have include sugar chains that are N-glycoside-linked to amino acid residues of peptides or proteins (e.g., asparagine residues, etc.), and sugar chains that are O-glycoside-linked to amino acid residues of peptides or proteins (e.g., serine residues, threonine residues, etc.). Examples of O-glycan include core 1 and core 2, and examples of N-glycan include high-mannose, hybrid, and complex sugar chains, with complex sugar chains being preferred.

[0106] 4. DcR3 variants containing Fc regions The DcR3 variants of the present invention also include proteins in which a homologous or heterologous peptide, polypeptide, or protein is bound or fused to the N- or C-terminus of a chimeric cysteine-rich region (or a chimeric cysteine-rich region containing other regions), either directly or, if necessary, via an appropriate peptide linker. The number of amino acids constituting the peptide linker is not particularly limited, and examples include 4, 5, 6, or 15 amino acids.

[0107] Examples of polypeptides or proteins to be bound to or fused to a chimeric cysteine-rich region include polypeptides or proteins such as immunoglobulin constant regions or Fc regions, peptides that bind to FcRn (neonatal Fc receptor), albumin, protein A, protein G, β-galactosidase, glutathione S-transferase (GST), maltose-binding protein, polyhistidine, and FLAG peptide. Preferably, they are immunoglobulin Fc regions or mutants thereof (mutant Fc regions), and more preferably, mammalian-derived immunoglobulin Fc regions or mutants thereof (mutant Fc regions).

[0108] When used in humans, the Fc region of an immunoglobulin (also referred to as an antibody) is preferably a human immunoglobulin Fc region. Examples of immunoglobulin classes and subclasses include, but are not limited to, IgG, IgD, IgE, IgM, IgA, IgG1, IgG2, IgG2a, IgG2b, IgG2c, IgG3, IgG4, and IgA1. For use in humans, it is preferable to use a human immunoglobulin class and subclass. Furthermore, when using an immunoglobulin Fc region as a polypeptide or protein to be bound or fused to a chimeric cysteine-rich region, the immunoglobulin Fc region is preferably bound or fused to the C-terminus of the chimeric cysteine-rich region.

[0109] Immunoglobulins are composed of heavy and light chain polypeptides, and the heavy chain constant region of human IgG is composed of, in order from the N-terminus, a CH1 domain, a hinge domain, a CH2 domain, and a CH3 domain. The IgG Fc region of the present invention also includes a region comprising the CH2 domain and the CH3 domain, as well as a region comprising a part or all of the hinge domain and the CH2 domain and the CH3 domain. Each domain contained in the IgG Fc region of the present invention can be identified by a number in the EU index. Specifically, the hinge domain is identified by positions 216 to 230 of the EU index, the CH2 domain is identified by positions 231 to 340, and the CH3 domain is identified by positions 341 to 447 of the EU index.

[0110] Polypeptides or proteins bound to or fused to a chimeric cysteine-rich region also include modified polypeptides or proteins in which one or more amino acids have been substituted, deleted, inserted, or added to alter the biological activity or properties of the DcR3 variant, its pharmacokinetics such as blood half-life, or its physical or chemical properties such as protein stability. Modifications of polypeptides or proteins bound to or fused to a chimeric cysteine-rich region include both natural mutations and artificial amino acid substitutions, deletions, insertions, or additions. Modified polypeptides or proteins include, for example, mutant Fc regions consisting of an amino acid sequence in which one or several amino acids have been deleted, substituted, inserted, or added to the amino acid sequence of the Fc region of an immunoglobulin. Examples of the amino acid sequence of a mutant Fc region include an amino acid sequence in which one or more, preferably 2 to 30, more preferably 2 to 10, and particularly preferably 2 to 5 amino acids have been substituted, deleted, inserted, or added in the amino acid sequence of an immunoglobulin Fc region, or an amino acid sequence that has 80% or more, preferably 85% or more, and more preferably 90% or more identity to the amino acid sequence of an immunoglobulin Fc region.

[0111] The addition or deletion of the aforementioned glycosylation site is also included in the modification of the polypeptide or protein to be bound or fused to the chimeric cysteine-rich region. For example, a glycosylation site can be added by adding or inserting a polypeptide containing an N-glycosylation site. Specific polypeptide sequences include GGNGT and YGNGT, each consisting of five amino acids (WO 2014 / 153111).

[0112] Substitutions in human IgG1 that reduce or eliminate complement-dependent cytotoxicity (CDC) activity include substitutions of one or more amino acids selected from Leu at position 234 (L234), Leu at position 235 (L235), Asp at position 265 (D265), Asp at position 270 (D265), Lys at position 322 (K322), Pro at position 329 (P329), and Pro at position 331 (P331) as shown in the EU index with other amino acids, and specific examples include substitutions of one or more amino acids selected from L234, L235, D270, K322, P329, and P331 as shown in the EU index with Ala, and substitutions of P331 with Ser or Gly [J. Immunol., 2000, 164: pp. 4178-4184, Cell. Immunol.,2000,200:p.16-26].

[0113] Substitutions in human IgG1 that reduce or eliminate effector activities such as antibody-dependent cellular cytotoxicity (ADCC) activity and antibody-dependent cellular phagocytosis (ADCP) include Asn at position 297 (N297), Leu at position 234 (L234), Leu at position 235 (L235), Gly at position 237 (G237), Cys at position 226 (C226), Cys at position 229 (C229), Pro at position 238 (P238), Glu at position 233 (E233), and Ser at position 267 (S267), as shown in the EU index. , Leu at position 328 (L328), Pro at position 331 (P331), etc., with other amino acids. Specific examples include amino acid substitutions such as substitution of N297 with Ala (N297A), N297 with Gln (N297Q), N297 with Gly (N297G), or substitution of L234 with Ala (L234A), L235 with Ala (L235A), or G237 with Ala (G237A), as shown in the EU index. Note that " / " means "and" (the same applies hereinafter).

[0114] Other modifications that increase the binding activity to the inhibitory receptor FcγRIIb include substitution of Gly at position 236 with Asp (G236D), substitution of Leu at position 328 with Phe (L328F), substitution of Ser at position 239 with Asp (S239D), and substitution of Ser at position 267 with Glu (S267E), as shown in the EU index [Curr. Opin. Cell. Biol., 2009, 20: pp. 685-691]. Furthermore, in the low pH environment of the endosome, FcRn (neonatal Fc Modifications that enhance binding to the antibody receptor (EU index), prevent antibody elimination, and extend blood half-life include substitution of Thr at position 250 with Gln (T250Q), substitution of Met at position 428 with Leu (M428L), substitution of Met at position 252 with Tyr (M252Y), substitution of Ser at position 254 with Thr (S254T), substitution of Thr at position 256 with Glu (T256E), and Examples of such mutations include substitution of Met at position 252 with Tyr (M252Y), substitution of Ser at position 254 with Thr (S254T), substitution of Thr at position 256 with Glu (T256E), substitution of Met at position 428 with Leu (M428L), substitution of Asn at position 434 with Ser (N434S), substitution of Asn at position 434 with Ala (N434A), and substitution of Asn at position 434 with His (N434H) [J. Immunol., 2009, 182: pp. 7663-7671, MAbs, 2017, 9: pp. 844-853].

[0115] In a preferred embodiment, the variant Fc region has a substitution of Cys to Ser at position 220 according to the EU index in the amino acid sequence of the heavy chain of an antibody belonging to the human IgG1 subclass. Examples of mutant Fc regions in this form include an immunoglobulin Fc region (hereinafter referred to as g1S, SEQ ID NO: 72) having an amino acid sequence in which Cys at position 220 (EU index), which is involved in binding to the light chain, has been substituted with Ser (C220S) in an Fc region (hereinafter referred to as g1S, SEQ ID NO: 72) obtained by removing Glu at position 216 (EU index) contained in the CH1 domain and hinge domain from the human IgG1 heavy chain constant region (SEQ ID NO: 153); and an immunoglobulin Fc region (hereinafter referred to as Eg1S, SEQ ID NO: 156) having an amino acid sequence in which Cys at position 220 (EU index), which is involved in binding to the light chain, has been substituted with Ser (C220S) in an Fc region (hereinafter referred to as Eg1S, SEQ ID NO: 153) obtained by removing the CH1 domain from the human IgG1 heavy chain constant region (SEQ ID NO: 153) (positions 216 to 447 in EU index). When the DcR3 variants of the present invention comprise a mutant Fc region of this type, the first chimeric cysteine-rich region contained in the DcR3 variants of the present invention is preferably a first chimeric cysteine-rich region in which the CRD1 of wild-type DcR3 is replaced with the CRD1 of OPG and the CRD4 of wild-type DcR3 is replaced with the CRD4 of OPG. Furthermore, when the DcR3 variants of the present invention comprise a mutant Fc region of this type, the second chimeric cysteine-rich region contained in the DcR3 variants of the present invention preferably comprises a substitution of Glu at position 57 with another amino acid, or a substitution of Glu at position 57 with another amino acid and Arg at position 58 with another amino acid. The other amino acid substituted for Glu at position 57 is preferably selected from Lys, Leu, Arg, Val, Ala, Phe, His, Ile, and Met, and more preferably selected from Lys, Leu, Arg, and Val. The other amino acid substituted for Arg at position 58 is preferably selected from Asp, Glu and Thr, more preferably Asp and Glu.When substituting another amino acid for Glu at position 57 with another amino acid for Arg at position 58, it is preferable to combine substituting Lys, Leu, Arg, Val, Ala, Phe, His, Ile, or Met for Glu at position 57 with Asp, Glu, or Thr for Arg at position 58, it is more preferable to combine Lys, Leu, Arg, or Val for Glu at position 57 with Asp or Glu for Arg at position 58, and it is even more preferable to combine Lys or Arg for Glu at position 57 with Asp or Glu for Arg at position 58.

[0116] In another preferred embodiment, the variant Fc region has a substitution of Ser at position 228 (EU index) with Pro, Leu at position 235 (EU index) with Glu, and Arg at position 409 (EU index) with Lys in the amino acid sequence of the heavy chain of an antibody belonging to the human IgG4 subclass. An example of a variant Fc region in this embodiment is the Fc of an immunoglobulin having an amino acid sequence obtained by removing the CH1 domain from the heavy chain constant region of human IgG4 (SEQ ID NO: 154) and substituting Ser at position 228 (EU index) with Pro, Leu at position 235 with Glu, and Arg at position 409 (EU index) with Lys (hereinafter referred to as g4PEK, SEQ ID NO: 74) as disclosed in WO 2006 / 33386. When the DcR3 variants of the present invention comprise a mutant Fc region of this type, the first chimeric cysteine-rich region contained in the DcR3 variants of the present invention is preferably a first chimeric cysteine-rich region in which the CRD1 of wild-type DcR3 is replaced with the CRD1 of OPG and the CRD4 of wild-type DcR3 is replaced with the CRD4 of OPG. Furthermore, when the DcR3 variants of the present invention comprise a mutant Fc region of this type, the second chimeric cysteine-rich region contained in the DcR3 variants of the present invention preferably comprises a substitution of Glu at position 57 with another amino acid, or a substitution of Glu at position 57 with another amino acid and Arg at position 58 with another amino acid. The other amino acid substituted for Glu at position 57 is preferably selected from Lys, Leu, Arg, Val, Ala, Phe, His, Ile, and Met, and more preferably selected from Lys, Leu, Arg, and Val. The other amino acid substituted for Arg at position 58 is preferably selected from Asp, Glu and Thr, more preferably Asp and Glu.When substituting another amino acid for Glu at position 57 with another amino acid for Arg at position 58, it is preferable to combine substituting Lys, Leu, Arg, Val, Ala, Phe, His, Ile, or Met for Glu at position 57 with Asp, Glu, or Thr for Arg at position 58, it is more preferable to combine Lys, Leu, Arg, or Val for Glu at position 57 with Asp or Glu for Arg at position 58, and it is even more preferable to combine Lys or Arg for Glu at position 57 with Asp or Glu for Arg at position 58.

[0117] In yet another preferred embodiment, the mutant Fc region has a substitution of Leu at position 234 (EU index) with Ala, Leu at position 235 (EU index) with Ala, and Gly at position 237 (EU index) with Ala in the amino acid sequence of the heavy chain of an antibody belonging to the human IgG1 subclass. Examples of mutant Fc regions in this embodiment include immunoglobulin Fc regions having the amino acid sequence of the Fc region of g1S shown in SEQ ID NO: 72 or Eg1S shown in SEQ ID NO: 156, in which Leu at position 234 is substituted with Ala, Leu at position 235 is substituted with Ala, and Gly at position 237 is substituted with Ala (hereinafter referred to as g1S LALAGA or Eg1S LALAGA; SEQ ID NOs: 162 and 164). When the DcR3 variants of the present invention comprise a mutant Fc region of this type, the first chimeric cysteine-rich region contained in the DcR3 variants of the present invention is preferably a first chimeric cysteine-rich region in which the CRD1 of wild-type DcR3 is replaced with the CRD1 of OPG and the CRD4 of wild-type DcR3 is replaced with the CRD4 of OPG. Furthermore, when the DcR3 variants of the present invention comprise a mutant Fc region of this type, the second chimeric cysteine-rich region contained in the DcR3 variants of the present invention preferably comprises a substitution of Glu at position 57 with another amino acid, or a substitution of Glu at position 57 with another amino acid and Arg at position 58 with another amino acid. The other amino acid substituted for Glu at position 57 is preferably selected from Lys, Leu, Arg, Val, Ala, Phe, His, Ile, and Met, and more preferably selected from Lys, Leu, Arg, and Val. The other amino acid substituted for Arg at position 58 is preferably selected from Asp, Glu and Thr, more preferably Asp and Glu.When substituting another amino acid for Glu at position 57 with another amino acid for Arg at position 58, it is preferable to combine substituting Lys, Leu, Arg, Val, Ala, Phe, His, Ile, or Met for Glu at position 57 with Asp, Glu, or Thr for Arg at position 58, it is more preferable to combine Lys, Leu, Arg, or Val for Glu at position 57 with Asp or Glu for Arg at position 58, and it is even more preferable to combine Lys or Arg for Glu at position 57 with Asp or Glu for Arg at position 58.

[0118] In yet another preferred embodiment, the mutant Fc region has a substitution of Asn at position 434 (as shown in the EU index) with Ala in the amino acid sequence of the heavy chain of an antibody belonging to the human IgG1 subclass. Examples of mutant Fc regions in this embodiment include immunoglobulin Fc regions having the amino acid sequence of the g1S Fc region shown in SEQ ID NO: 72 or the Eg1S Fc region shown in SEQ ID NO: 156, in which Asn at position 434 is substituted with Ala (hereinafter referred to as g1S N434A or Eg1S N434A; SEQ ID NOs: 312, 160), and immunoglobulin Fc regions having the amino acid sequence of the g1S LALAGA Fc region shown in SEQ ID NO: 162 or the Eg1S LALAGA Fc region shown in SEQ ID NO: 164, in which Asn at position 434 is substituted with Ala (hereinafter referred to as g1S LALAGANA or Eg1S LALAGANA; SEQ ID NOs: 313, 166). When the DcR3 variants of the present invention comprise a mutant Fc region of this type, the first chimeric cysteine-rich region contained in the DcR3 variants of the present invention is preferably a first chimeric cysteine-rich region in which the CRD1 of wild-type DcR3 is replaced with the CRD1 of OPG and the CRD4 of wild-type DcR3 is replaced with the CRD4 of OPG. Furthermore, when the DcR3 variants of the present invention comprise a mutant Fc region of this type, the second chimeric cysteine-rich region contained in the DcR3 variants of the present invention preferably comprises a substitution of Glu at position 57 with another amino acid, or a substitution of Glu at position 57 with another amino acid and Arg at position 58 with another amino acid. The other amino acid substituted for Glu at position 57 is preferably selected from Lys, Leu, Arg, Val, Ala, Phe, His, Ile, and Met, and more preferably selected from Lys, Leu, Arg, and Val. The other amino acid substituted for Arg at position 58 is preferably selected from Asp, Glu and Thr, more preferably Asp and Glu.When substituting another amino acid for Glu at position 57 with another amino acid for Arg at position 58, it is preferable to combine substituting Lys, Leu, Arg, Val, Ala, Phe, His, Ile, or Met for Glu at position 57 with Asp, Glu, or Thr for Arg at position 58, it is more preferable to combine Lys, Leu, Arg, or Val for Glu at position 57 with Asp or Glu for Arg at position 58, and it is even more preferable to combine Lys or Arg for Glu at position 57 with Asp or Glu for Arg at position 58.

[0119] In yet another preferred embodiment, the mutant Fc region has a substitution of Met at position 252 (EU index) with Tyr, Ser at position 254 (EU index) with Thr, and Thr at position 256 (EU index) with Glu in the amino acid sequence of the heavy chain of an antibody belonging to the human IgG1 subclass. Examples of mutant Fc regions in this embodiment include immunoglobulin Fc regions having the amino acid sequence of the Fc region of g1S shown in SEQ ID NO: 72 or Eg1S shown in SEQ ID NO: 156, in which Met at position 252 is substituted with Tyr, Ser at position 254 with Thr, and Thr at position 256 with Glu (hereinafter referred to as g1S YTE or Eg1S YTE; SEQ ID NOs: 311 and 158). When the DcR3 variants of the present invention comprise a mutant Fc region of this type, the first chimeric cysteine-rich region contained in the DcR3 variants of the present invention is preferably a first chimeric cysteine-rich region in which the CRD1 of wild-type DcR3 is replaced with the CRD1 of OPG and the CRD4 of wild-type DcR3 is replaced with the CRD4 of OPG. Furthermore, when the DcR3 variants of the present invention comprise a mutant Fc region of this type, the second chimeric cysteine-rich region contained in the DcR3 variants of the present invention preferably comprises a substitution of Glu at position 57 with another amino acid, or a substitution of Glu at position 57 with another amino acid and Arg at position 58 with another amino acid. The other amino acid substituted for Glu at position 57 is preferably selected from Lys, Leu, Arg, Val, Ala, Phe, His, Ile, and Met, and more preferably selected from Lys, Leu, Arg, and Val. The other amino acid substituted for Arg at position 58 is preferably selected from Asp, Glu and Thr, more preferably Asp and Glu.When substituting another amino acid for Glu at position 57 with another amino acid for Arg at position 58, it is preferable to combine substituting Lys, Leu, Arg, Val, Ala, Phe, His, Ile, or Met for Glu at position 57 with Asp, Glu, or Thr for Arg at position 58, it is more preferable to combine Lys, Leu, Arg, or Val for Glu at position 57 with Asp or Glu for Arg at position 58, and it is even more preferable to combine Lys or Arg for Glu at position 57 with Asp or Glu for Arg at position 58.

[0120] Examples of mutant Fc regions include, but are not limited to, mutant Fc regions that comprise or consist of the amino acid sequence set forth in SEQ ID NO: 72, 74, 156, 158, 160, 162, 164, 166, 311, 312, or 313.

[0121] Examples of peptide linkers that can be added to the N-terminus or C-terminus of a chimeric cysteine-rich region to further link the same or a different peptide, polypeptide, or protein include, but are not limited to, peptide linkers such as an IEG RMD linker or a GS linker, and chemical linkers.

[0122] The most preferred DcR3 variants of the present invention include DcR3 variants comprising a first or second chimeric cysteine-rich region and an Fc region or a mutant Fc region.

[0123] In one embodiment, the amino acid sequence contained in the DcR3 variant of the present invention includes, for example, the amino acid sequence set forth in SEQ ID NO: 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, or 46, or an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted, or added to the amino acid sequence. In this embodiment, the DcR3 variant of the present invention may consist of or contain the above amino acid sequence, but preferably consists of the above amino acid sequence. The above amino acid sequence includes the HBD.

[0124] In another embodiment, the amino acid sequence contained in the DcR3 variant of the present invention includes, for example, the amino acid sequence set forth in SEQ ID NO: 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, 180, 182, 184, 186, 188, 270, 272, 274, 276, 278, 280, 282, 284, or 286, or an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted, or added within the amino acid sequence. In this embodiment, the DcR3 variant of the present invention may consist of or contain the above amino acid sequence, but preferably consists of the above amino acid sequence. Note that the above amino acid sequence does not include the HBD.

[0125] In another embodiment, examples of amino acid sequences contained in DcR3 variants of the present invention include amino acid sequences set forth in SEQ ID NO: 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, 180, 182, 184, 186, 188, 270, 272, 274, 276, 278, 280, 282, 284, or 286, or amino acid sequences in which 1 to 30 amino acids have been deleted, substituted, inserted, or added within the amino acid sequence, and an amino acid sequence set forth in SEQ ID NO: 72, 74, 156, 158, 160, 162, 164, 166, 311, 312, or 313 as a mutant Fc region.

[0126] In yet another embodiment, the amino acid sequences contained in the DcR3 variants of the present invention include, for example, those represented by SEQ ID NOs: 76, 78, 80, 82, 84, 86, 88, 90, 92, 94, 96, 98, 150, 168, 170, 172, 174, 176, 178, 190, 192, 194, 196, 198, 200, 202, 204, 206, 208, 210, 212, 214, 216, 218, 220, 222, 224, 226, 228, 230, 232, 234, 236, 238, 240, 242, 244, 246, 248, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 300, 301, 302, 303, 304, 305, 306, 3 252, 254, 256, 258, 260, 262, 264, 266, 268, 288, 290, 292, 294, 296, 298, 300, 302, 304, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, or 337, or amino acid sequences in which 1 to 30 amino acids have been deleted, substituted, inserted, or added within the amino acid sequences. In this embodiment, the DcR3 variants of the present invention may consist of or contain the above amino acid sequences, but preferably consist of the above amino acid sequences. The above amino acid sequence includes an Fc region or a mutant thereof (mutant Fc region).

[0127] SEQ ID NO: 76 is the amino acid sequence of chimera B-Fc(IEGRMD g1S) (a DcR3 variant in which chimera B, an IEGRMD linker, and Fc(g1S) are fused), SEQ ID NO: 78 is the amino acid sequence of chimera C-Fc(IEGRMD g1S) (a DcR3 variant in which chimera C, an IEGRMD linker, and Fc(g1S) are fused), SEQ ID NO: 80 is the amino acid sequence of chimera A-Fc(IEGRMD g1S) (a DcR3 variant in which chimera B, an IEGRMD linker, and Fc(g1S) are fused), SEQ ID NO: 82 is the amino acid sequence of chimera A-Fc(g4PEK) (a DcR3 variant in which chimera A and Fc(g4PEK) are fused), SEQ ID NO: 84 is the amino acid sequence of 103-123OPG-Fc(g4PEK) (a DcR3 variant in which 103-123OPG and Fc(g4PEK) are fused), SEQ ID NO: 86 is the amino acid sequence of N131S / N SEQ ID NO: 88 is the amino acid sequence of T133A / S146A-Fc(g4PEK) (a DcR3 variant in which N131S / N144S and Fc(g4PEK) are fused), SEQ ID NO: 90 is the amino acid sequence of N131S / N144S / N157S-Fc(g4PEK) (a DcR3 variant in which T133A / S146A and Fc(g4PEK) are fused), SEQ ID NO: 92 is the amino acid sequence of T133A / S146A / T159A-Fc(g4PEK) (a DcR3 variant in which T133A / S146A / T159A and Fc(g4PEK) are fused); SEQ ID NO: 94 is the amino acid sequence of chimera A-E57K-Fc(g4PEK) (a DcR3 variant in which chimera A-E57K and Fc(g4PEK) are fused); SEQ ID NO: 96 is the amino acid sequence of chimera A-E57L-Fc(g4PEK) ) (a DcR3 variant in which chimera A-E57L and Fc(g4PEK) are fused), SEQ ID NO: 98 is the amino acid sequence of chimera A-R60K-Fc(g4PEK) (a DcR3 variant in which chimera A-R60K and Fc(g4PEK) are fused), SEQ ID NO: 150 is the amino acid sequence of chimera A-Fc(g1S) (a DcR3 variant in which chimera A and Fc(g1S) are fused), SEQ ID NO: 168 is the amino acid sequence of chimera A-Fc(Eg1S) (a DcR3 variant in which chimera A andSEQ ID NO: 170 is the amino acid sequence of chimera A-Fc(Eg1S-YTE) (a DcR3 variant formed by fusing chimera A with Fc(Eg1S-YTE)); SEQ ID NO: 172 is the amino acid sequence of chimera A-Fc(Eg1S-N434A) (a DcR3 variant formed by fusing chimera A with Fc(Eg1S-N434A)); SEQ ID NO: 174 is the amino acid sequence of chimera A-Fc(g1S-LALAGA) (a DcR3 variant formed by fusing chimera A with Fc(g1S-LALAGA)); SEQ ID NO: 176 is the amino acid sequence of chimera A-Fc(Eg1S-LALAGA) (a DcR3 variant formed by fusing chimera A with Fc(Eg1S-LALAGA)). SEQ ID NO: 178 is the amino acid sequence of chimera A-Fc(Eg1S-LALAGANA) (a DcR3 variant formed by fusing chimera A with Fc(Eg1S-LALAGANA)); SEQ ID NO: 190 is the amino acid sequence of chimera A-E57R-Fc(g4PEK) (a DcR3 variant formed by fusing chimera A with Fc(g4PEK)); SEQ ID NO: 192 is the amino acid sequence of chimera A-E57V-Fc(g4PEK) (a DcR3 variant formed by fusing chimera A with Fc(g4PEK)). SEQ ID NO: 194 is the amino acid sequence of chimera A-E57K_R58D-Fc(g4PEK) (a DcR3 variant in which chimera A-E57K_R58D and Fc(g4PEK) are fused), SEQ ID NO: 196 is the amino acid sequence of chimera A-E57K_R58E-Fc(g4PEK) (a DcR3 variant in which chimera A-E57K_R58E and Fc(g4PEK) are fused), SEQ ID NO: 198 is the amino acid sequence of chimera A-E57R_R58D-Fc(g4PEK) (a DcR3 variant in which chimera A-E57R_R58D and Fc(g4PEK) are fused). SEQ ID NO: 200 is the amino acid sequence of chimera A-E57K-Fc(Eg1S) (a DcR3 variant formed by fusing chimera A-E57K with Fc(Eg1S)); SEQ ID NO: 202 is the amino acid sequence of chimera A-E57L-Fc(Eg1S) (a DcR3 variant formed by fusing chimera A-E57L with Fc(Eg1S)); SEQ ID NO: 204 is the amino acid sequence of chimera A-E57R-Fc(Eg1S) (a DcR3 variant formed by fusing chimera A-E57R with Fc(Eg1S));SEQ ID NO: 206 is the amino acid sequence of chimera A-E57V-Fc(Eg1S) (a DcR3 variant in which chimera A-E57V and Fc(Eg1S) are fused), SEQ ID NO: 208 is the amino acid sequence of chimera A-E57K_R58D-Fc(Eg1S) (a DcR3 variant in which chimera A-E57K_R58D and Fc(Eg1S) are fused), SEQ ID NO: 210 is the amino acid sequence of chimera A-E57K SEQ ID NO: 212 is the amino acid sequence of chimera A-E57R_R58D-Fc(Eg1S) (a DcR3 variant in which chimera A-E57R_R58D and Fc(Eg1S) are fused), SEQ ID NO: 214 is the amino acid sequence of chimera A-E57K-Fc(Eg1S) SEQ ID NO: 216 is the amino acid sequence of chimera A-E57L-Fc(Eg1S YTE) (a DcR3 variant formed by fusing chimera A-E57L with Fc(Eg1S YTE)); SEQ ID NO: 218 is the amino acid sequence of chimera A-E57R-Fc(Eg1S YTE) (a DcR3 variant formed by fusing chimera A-E57R with Fc(Eg1S YTE)); SEQ ID NO: 220 is the amino acid sequence of chimera A-E57V-Fc(Eg1S YTE) (a DcR3 variant formed by fusing chimera A-E57V with Fc(Eg1S YTE)); SEQ ID NO: 222 is the amino acid sequence of chimera A-E57K_R58D-Fc(Eg1S YTE) (a DcR3 variant formed by fusing chimera A-E57K_R58D with Fc(Eg1S YTE)); SEQ ID NO: 224 is the amino acid sequence of chimera A-E57K_R58E-Fc(Eg1S YTE) (a DcR3 variant formed by fusing chimera A-E57K_R58E with Fc(Eg1S YTE)); SEQ ID NO: 226 is the amino acid sequence of chimera A-E57R_R58D-Fc(Eg1S YTE) (a DcR3 variant formed by fusing chimera A-E57R_R58D with Fc(Eg1S YTE)); SEQ ID NO: 228 is the amino acid sequence of chimera A-E57K-Fc(Eg1S N434A) (a DcR3 variant in which chimera A-E57K and Fc(Eg1S N434A) are fused), SEQ ID NO: 230 is the amino acid sequence ofSEQ ID NO: 232 is the amino acid sequence of chimera A-E57R-Fc(Eg1S N434A) (a DcR3 variant formed by fusing chimera A-E57R with Fc(Eg1S N434A)). SEQ ID NO: 234 is the amino acid sequence of chimera A-E57V-Fc(Eg1S N434A) (a DcR3 variant formed by fusing chimera A-E57V with Fc(Eg1S N434A)). SEQ ID NO: 236 is the amino acid sequence of chimera A-E57K_R58D-Fc(Eg1S SEQ ID NO: 238 is the amino acid sequence of chimera A-E57K_R58E-Fc(Eg1S N434A) (a DcR3 variant formed by fusing chimera A-E57K_R58E with Fc(Eg1S N434A)); SEQ ID NO: 240 is the amino acid sequence of chimera A-E57R_R58D-Fc(Eg1S N434A) (a DcR3 variant formed by fusing chimera A-E57R_R58D with Fc(Eg1S N434A)); SEQ ID NO: 242 is the amino acid sequence of chimera A-E57K-Fc(Eg1S SEQ ID NO: 244 is the amino acid sequence of chimera A-E57L-Fc(Eg1S LALAGA) (a DcR3 variant formed by fusing chimera A-E57L with Fc(Eg1S LALAGA)); SEQ ID NO: 246 is the amino acid sequence of chimera A-E57R-Fc(Eg1S LALAGA) (a DcR3 variant formed by fusing chimera A-E57R with Fc(Eg1S LALAGA)); SEQ ID NO: 248 is the amino acid sequence of chimera A-E57V-Fc(Eg1S LALAGA) (a DcR3 variant formed by fusing chimera A-E57V with Fc(Eg1S SEQ ID NO: 250 is the amino acid sequence of chimera A-E57K_R58D-Fc(Eg1S LALAGA) (a DcR3 variant formed by fusing chimera A-E57K_R58D and Fc(Eg1S LALAGA)); SEQ ID NO: 252 is the amino acid sequence of chimera A-E57K_R58E-Fc(Eg1S LALAGA) (a DcR3 variant formed by fusing chimera A-E57K_R58E andSEQ ID NO: 254 is the amino acid sequence of chimera A-E57R_R58D-Fc(Eg1S LALAGA) (a DcR3 variant formed by fusing chimera A-E57R_R58D with Fc(Eg1S LALAGA)); SEQ ID NO: 256 is the amino acid sequence of chimera A-E57K-Fc(Eg1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57K with Fc(Eg1S LALAGANA)); SEQ ID NO: 258 is the amino acid sequence of chimera A-E57L-Fc(Eg1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57L with Fc(Eg1S SEQ ID NO: 260 is the amino acid sequence of chimera A-E57R-Fc(Eg1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57R with Fc(Eg1S LALAGANA)); SEQ ID NO: 262 is the amino acid sequence of chimera A-E57V-Fc(Eg1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57V with Fc(Eg1S LALAGANA)); SEQ ID NO: 264 is the amino acid sequence of chimera A-E57K_R58D-Fc(Eg1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57K_R58D with Fc(Eg1S SEQ ID NO: 266 is the amino acid sequence of chimera A-E57K_R58E-Fc(Eg1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57K_R58E and Fc(Eg1S LALAGANA)), SEQ ID NO: 268 is the amino acid sequence of chimera A-E57R_R58D-Fc(Eg1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57R_R58D and Fc(Eg1S LALAGANA)). SEQ ID NO: 288 is the amino acid sequence of chimera A-E57A-Fc(g4PEK) (a DcR3 variant formed by fusing chimera A-E57A with Fc(g4PEK)); SEQ ID NO: 290 is the amino acid sequence of chimera A-E57F-Fc(g4PEK) (a DcR3 variant formed by fusing chimera A-E57F with Fc(g4PEK)); SEQ ID NO: 292 is the amino acid sequence of chimera A-E57H-Fc(g4PEK) (a DcR3 variant formed by fusing chimera A-E57H with Fc(g4PEK));SEQ ID NO: 294 is the amino acid sequence of chimera A-E57I-Fc(g4PEK) (a DcR3 variant in which chimera A-E57I and Fc(g4PEK) are fused), SEQ ID NO: 296 is the amino acid sequence of chimera A-E57M-Fc(g4PEK) (a DcR3 variant in which chimera A-E57M and Fc(g4PEK) are fused), SEQ ID NO: 298 is the amino acid sequence of chimera A-E57K_R58, SEQ ID NO: 300 is the amino acid sequence of chimera A-E57L_R58E-Fc(g4PEK) (a DcR3 variant in which chimera A-E57K_R58T and Fc(g4PEK) are fused), SEQ ID NO: 302 is the amino acid sequence of chimera A-E57L_R58E-Fc(g4PEK) (a DcR3 variant in which chimera A-E57L_R58E and Fc(g4PEK) are fused), SEQ ID NO: 303 is the amino acid sequence of chimera A-E57V_R58E-Fc(g4PEK) (a DcR3 variant in which chimera A-E57V_R58E and Fc(g4PEK) are fused). SEQ ID NO: 304 is the amino acid sequence of chimera A-E57V_R58E-Fc(g4PEK) (a DcR3 variant in which chimera A-E57V_R58E and Fc(g4PEK) are fused). SEQ ID NO: 314 is the amino acid sequence of chimera A-Fc(g1S SEQ ID NO: 315 is the amino acid sequence of chimera A-Fc(g1S N434A) (a DcR3 variant formed by fusing chimera A with Fc(g1S N434A)); SEQ ID NO: 316 is the amino acid sequence of chimera A-Fc(g1S LALAGANA) (a DcR3 variant formed by fusing chimera A with Fc(g1S LALAGANA)); SEQ ID NO: 317 is the amino acid sequence of chimera A-E57K-Fc(g1S YTE) (a DcR3 variant formed by fusing chimera A with Fc(g1S YTE)); SEQ ID NO: 318 is the amino acid sequence of chimera A-E57L-Fc(g1S YTE) (a DcR3 variant formed by fusing chimera A with Fc(g1S SEQ ID NO: 319 is the amino acid sequence of chimera A-E57R-Fc(g1S YTE) (a DcR3 variant formed by fusing chimera A-E57R with Fc(g1S YTE)); SEQ ID NO: 320 is the amino acid sequence of chimera A-E57V-Fc(g1S YTE) (a DcR3 variant formed by fusing chimera A-E57V with Fc(g1S YTE)); SEQ ID NO: 321 is the amino acid sequence of chimera A-E57K_R58D-Fc(g1S YTE) (a DcR3 variant formed by fusing chimera A-E57K_R58D with Fc(g1S YTE)); SEQ ID NO: 322 is the amino acid sequence of chimera A-E57K_R58E-Fc(g1S YTE) (chimera A-E57K_R58E and Fc(g1SSEQ ID NO: 323 is the amino acid sequence of chimera A-E57R_R58D-Fc(g1S YTE) (a DcR3 variant formed by fusing chimera A-E57R_R58D with Fc(g1S YTE)); SEQ ID NO: 324 is the amino acid sequence of chimera A-E57K-Fc(g1S N434A) (a DcR3 variant formed by fusing chimera A-E57K with Fc(g1S N434A)); SEQ ID NO: 325 is the amino acid sequence of chimera A-E57L-Fc(g1S N434A) (a DcR3 variant formed by fusing chimera A-E57L with Fc(g1S N434A)); SEQ ID NO: 326 is the amino acid sequence of chimera A-E57R-Fc(g1S SEQ ID NO: 327 is the amino acid sequence of chimera A-E57V-Fc(g1S N434A) (a DcR3 variant formed by fusing chimera A-E57V with Fc(g1S N434A)); SEQ ID NO: 328 is the amino acid sequence of chimera A-E57K_R58D-Fc(g1S N434A) (a DcR3 variant formed by fusing chimera A-E57K_R58D with Fc(g1S N434A)); SEQ ID NO: 329 is the amino acid sequence of chimera A-E57K_R58E-Fc(g1S N434A) (a DcR3 variant formed by fusing chimera A-E57K_R58E with Fc(g1S SEQ ID NO: 330 is the amino acid sequence of chimera A-E57R_R58D-Fc(g1S N434A) (a DcR3 variant formed by fusing chimera A-E57R_R58D with Fc(g1S N434A)); SEQ ID NO: 331 is the amino acid sequence of chimera A-E57K-Fc(g1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57K with Fc(g1S LALAGANA)); SEQ ID NO: 332 is the amino acid sequence of chimera A-E57L-Fc(g1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57L with Fc(g1S LALAGANA)); SEQ ID NO: 333 is the amino acid sequence of chimera A-E57R-Fc(g1S The amino acid sequence of chimera A-E57R-Fc(g1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57R and Fc(g1S LALAGANA)), SEQ ID NO: 334, isSEQ ID NO: 335 is the amino acid sequence of chimera A-E57K_R58D-Fc(g1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57K_R58D with Fc(g1S LALAGANA)); SEQ ID NO: 336 is the amino acid sequence of chimera A-E57K_R58E-Fc(g1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57K_R58E with Fc(g1S LALAGANA)); SEQ ID NO: 337 is the amino acid sequence of chimera A-E57R_R58D-Fc(g1S LALAGANA) (a DcR3 variant formed by fusing chimera A-E57R_R58D with Fc(g1S LALAGANA)); 1 shows the amino acid sequence of a DcR3 variant fused with DcR3-LALAGANA.

[0128] In all of the above aspects, the mutations (modifications) made to the amino acid sequence include both natural mutations and artificial amino acid substitutions, deletions, insertions, or additions. Furthermore, in all of the above aspects, the amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted, or added includes an amino acid sequence in which 1 or 2 or more, preferably 2 to 30, more preferably 2 to 10, and particularly preferably 2 to 5 amino acids have been substituted, deleted, inserted, or added, or an amino acid sequence having 80% or more, preferably 85% or more, more preferably 90% or more, for example, 93% or more, 95% or more, 97% or more, 98% or more, or 99% or more identity to the amino acid sequence.

[0129] The DcR3 variants of the present invention may be further chemically modified to alter their biological activity or properties, pharmacokinetics such as blood half-life, or physical or chemical properties such as protein stability.

[0130] Examples of chemical modifications include polyethylene glycol (PEG) modification, acetylation, amidation, and phosphorylation, with PEGylation being particularly preferred. PEGylation involves binding one or more PEG molecules to the N-terminal amino group of a protein, or to an amino acid residue having a functional group in its side chain, such as the ε-amino group of Lys, a carboxyl group, a thiol group, or a hydroxyl group.

[0131] The average molecular weight of the PEG molecule is not limited to the following, but can be in the range of about 3,000 to about 50,000.

[0132] Methods for attaching PEG molecules to modified DcR3 include introducing an active group, such as a carboxyl group, formyl (aldehyde) group, N-hydroxysuccinimide ester group, amino group, thiol group, or maleimide group, into the terminal portion of PEG and reacting it with a functional group, such as an amino group, carboxyl group, thiol group, or hydroxyl group, on the side chain of the modified DcR3.

[0133] The signal peptide is not limited to, but includes, when the N-terminal CRD domain of the DcR3 variant is derived from human DcR3, the signal peptide of human DcR3 consisting of the amino acid sequence of positions 1 to 29 of SEQ ID NO: 2. When the N-terminal CRD domain of the DcR3 variant is derived from human OPG, the signal peptide of human OPG consisting of the amino acid sequence of positions 1 to 21 of SEQ ID NO: 14 is also included. The signal peptide of the present invention also includes artificial sequences, sequences derived from expression vectors, and sequences derived from other proteins suitable for the host cell in which the DcR3 variant is expressed. DcR3 variants containing a signal peptide are immature polypeptides, and in one embodiment, the signal peptide is cleaved during maturation. The DcR3 variants of the present invention also include those with a different N-terminus, in which the signal peptide is cleaved at a site other than the predicted site.

[0134] 5. Method for producing DcR3 variants The DcR3 variants of the present invention can be produced by expressing DNA encoding the DcR3 variants in host cells using methods such as those described in Molecular Cloning: A Laboratory Manual, 3rd edition, Cold Spring Harbor Laboratory Press (2001), for example, by the following method.

[0135] The signal peptide is not limited to, but includes, when the N-terminal CRD domain of the DcR3 variant is derived from human DcR3, the signal peptide of human DcR3 consisting of the amino acid sequence of positions 1 to 29 of SEQ ID NO: 2. When the N-terminal CRD domain of the DcR3 variant is derived from human OPG, the signal peptide includes the amino acid sequence of positions 1 to 21 of SEQ ID NO: 14, derived from human OPG. Furthermore, artificial sequences, sequences derived from expression vectors, and sequences derived from other proteins suitable for host cells expressing the DcR3 variants can all be used to produce the DcR3 variants of the present invention. The DcR3 variants of the present invention also include those with a different N-terminus, where the signal peptide is cleaved at a position different from the predicted cleavage site.

[0136] The DcR3 variants of the present invention can be obtained by artificial design based on the amino acid sequence of the cysteine-rich region of wild-type DcR3 before amino acid substitution, the amino acid sequence of the cysteine-rich region of a DcR3 variant, or the amino acid sequence of a DcR3 variant, or by analyzing mutants. Site-directed mutagenesis using PCR with primers is preferred as a method for introducing mutations (Kunkel et al., Proc. Natl. Acad. Sci. USA, 1985, 82:488-492). Other methods include total synthesis of the mutated gene, or PCR using primers containing the mutation to separate the gene into a segment before and a segment after the mutation site, ligating the two fragments via the overlapping region containing the mutation, and inserting them into a vector using in-fusion cloning (Clontech, Inc.), for example.

[0137] Methods for identifying mutants with the desired binding mode include preparing a library with randomly introduced mutations, displaying it on phage or yeast, and screening for binding activity to obtain mutants with the desired binding mode. Another method involves expressing a vector containing DNA mutations that replace specific amino acids with different amino acids in host cells, thereby obtaining mutants with the desired binding mode. DNA encoding the DcR3 variants of the present invention can be synthesized using a DNA synthesizer by designing a nucleotide sequence encoding the amino acid sequence of the DcR3 variant of the present invention from the amino acid sequence of the DcR3 variant. Alternatively, DNA can be isolated by PCR using human or other cDNA as a template.

[0138] The DNA encoding the DcR3 variant of the present invention obtained above is inserted downstream of the promoter of an appropriate expression vector to prepare a recombinant vector, which is then introduced into a host cell compatible with the expression vector.

[0139] Bases in the DNA sequence encoding the DcR3 variant can be substituted to optimize codons for expression in the host, thereby improving the yield of the desired DcR3 variant. When constructing the DNA encoding the DcR3 variant described above, DNA encoding a signal peptide of a secretory protein can be added to the 5' end. Using this DNA, a recombinant vector can be constructed in the same manner as described above, and introduced into host cells, allowing the peptide to be secreted into the culture medium and produced. Examples of signal peptides include sequences derived from DcR3 or OPG, artificial sequences, sequences derived from expression vectors, and sequences derived from other proteins suitable for the host cell.

[0140] In one embodiment, examples of the base sequence of DNA encoding the amino acid sequence of a DcR3 variant include the base sequences set forth in SEQ ID NOs: 25, 27, 29, 31, 33, 35, 37, 39, 41, 43, and 45.

[0141] In another aspect, examples of the base sequence of DNA encoding the amino acid sequence of a DcR3 variant include the base sequences set forth in SEQ ID NOs: 49, 51, 53, 55, 57, 59, 61, 63, 65, 67, 69, 179, 181, 183, 185, 187, 269, 271, 273, 275, 277, 279, 281, 283, and 285.

[0142] In yet another embodiment, the base sequence of DNA encoding the amino acid sequence of a DcR3 variant is, for example, SEQ ID NOs: 75, 77, 79, 81, 83, 85, 87, 89, 91, 93, 95, 97, 149, 167, 169, 171, 173, 175, 177, 189, 191, 193, 195, 197, 199, 201, 203, 205, 207, 209, 211 , 213, 215, 217, 219, 221, 223, 225, 227, 229, 231, 233, 235, 237, 239, 241, 243, 245, 247, 249, 251, 253, 255, 257, 259, 261, 263, 265, 267, 287, 289, 291, 293, 295, 297, 299, 301 and 303.

[0143] Any expression vector can be used as long as it is capable of autonomous replication in the host cell to be used or of being integrated into a chromosome and contains a suitable promoter in a position where the DNA encoding the polypeptide can be transcribed.

[0144] Any host cell can be used as long as it is capable of expressing a gene encoding a DcR3 variant, such as yeast, insect cells, or animal cells. Preferred are yeast, animal cells, or insect cells that are capable of glycosylation of proteins.

[0145] Examples of yeast include Saccharomyces ( Saccharomyces ) genus, Schizosaccharomyces ( Schizosaccharomyces ) genus, Kluyveromyces ( Kluyveromyces ) genus, Trichosporon ( Trichosporon ) genus, Schwanniomyces ( Schwanniomyces ) genus, Pichia ( Pichia ) genus, Candida ( Candida ) Microorganisms belonging to the genus, for example, Saccharomyces cerevisiae , Schizosaccharomyces pombe , Kluyveromyces lactis , Trichosporon pullulans , Schwanniomyces alluvius , or Candida utilis etc.

[0146] Examples of insect cells include Sf9 and Sf21 ovarian cells of Spodoptera frugiperda [Baculovirus Expression Vectors, A Laboratory Manual, W.H. Freeman and Company, New York (1992)], High 5 ovarian cells of Trichoplusia ni (Invitrogen), and S2 (Schneider 2) cells derived from late embryos of Drosophila melanogaster (Thermo Scientific).

[0147] Examples of animal cells include human Namalwa cells, monkey COS cells, and Chinese hamster CHO cells [Journal of Experimental Medicine, 108, 945 (1958); Proc. Natl. Acad. Sci. USA, 60, 1275 (1968); Genetics, 55, 513 (1968); Chromosoma, 41, 129 (1973); Methods in Cell Science, 18, 115 (1996); Radiation Research, 148, 260 (1997); Proc. Natl. Acad. Sci. USA, 77, 4216 (1980); Proc. Natl. Acad. Sci. USA, 60, 1275 (1968); Cell, 6, 121 (1975); Molecular Cellgenetics, Appendix I, II (pp. 883-900)), CHO / DG44, CHO-K1 (ATCC number: CCL-61), Freestyle CHO-S cells, CHO cells lacking the dihydrofolate reductase gene [Proc. Natl. Acad. Sci. USA, 77, 4216 (1980)], CHO cells lacking the 6-fucosyltransferase gene (WO 2005 / 035586, WO 02 / 31140), human 293 cells (ATCC number: CRL-1573), Freestyle 293F cells, Expi293 cells (Thermo Scientific), DUkXB11 (ATCC number: CCL-9096), Pro-5 (ATCC number: CCL-1781), CHO-S (Life Technologies, Cat#11619), Pro-3, rat myeloma cells YB2 / 3HL.P2.G11.16Ag.20 (also referred to as YB2 / 0), mouse myeloma cells NSO, mouse myeloma cells SP2 / 0-Ag14, or Syrian hamster cells BHK and HBT5637 (Japanese Patent Laid-Open No. 63-299).

[0148] When yeast is used as the host cell, examples of the expression vector include YEP13 (ATCC No. 37115), YEp24 (ATCC 37051), YCp50 (ATCC 37419), pHS19, and pHS15.

[0149] Any promoter may be used as long as it can be expressed in a yeast strain, and examples include promoters of glycolytic genes such as hexose kinase, the PHO5 promoter, the PGK promoter, the GAP promoter, the ADH promoter, the gal 1 promoter, the gal 10 promoter, the heat shock polypeptide promoter, the MFα1 promoter, and the CUP 1 promoter.

[0150] Any method for introducing a recombinant vector can be used as long as it is a method for introducing DNA into yeast, and examples thereof include the electroporation method [Methods Enzymol., 194, 182 (1990)], the spheroplast method [Proc. Natl. Acad. Sci. USA, 75, 1929 (1978)], the lithium acetate method [J. Bacteriology, 153, 163 (1983)], or the method described in Proc. Natl. Acad. Sci. USA, 75, 1929 (1978).

[0151] When insect cells are used as hosts, peptides can be expressed by methods described in, for example, Current Protocols in Molecular Biology, Baculovirus Expression Vectors, A Laboratory Manual, W.H. Freeman and Company, New York (1992), Bio / Technology, 6, 47 (1988), etc.

[0152] That is, a recombinant gene transfer vector and a defective baculovirus genome are co-transfected into insect cells to obtain a recombinant virus in the insect cell culture supernatant, and then the recombinant virus can be used to infect insect cells to express the peptide.

[0153] Examples of gene transfer vectors used in this method include pVL1392, pVL1393 (Becton Dickinson), and pBlueBac4.5 (Invitrogen).

[0154] As the baculovirus, for example, Autographa californica nuclear polyhedrosis virus, which is a virus that infects insects of the Noctuidae family, can be used.

[0155] Methods for co-introducing the above-mentioned recombinant gene transfer vector and the above-mentioned baculovirus into insect cells to prepare a recombinant virus include, for example, the calcium phosphate method (JP 2-227075 A) or the lipofection method [Proc. Natl. Acad. Sci. USA, 84, 7413 (1987)].

[0156] When S2 (Schneider 2) cells (Thermo Scientific) derived from late embryos of the insect cell Drosophila melanogaster are used as hosts, peptides can be expressed by introducing a gene transfer vector such as pMTBiPV5-HisA (Thermo Scientific) into the host cells using the calcium phosphate method, for example, as described in Mol. Biotechnol., 2015, 10: pp. 914-922.

[0157] When animal cells are used as hosts, examples of expression vectors include pCI mammalian expression vector (Promega), pcDNA3.1(+) (Invitrogen), pcDNA I / Amp, pcDNA I, pcDM8 (Funakoshi), pAGE107 (Japanese Patent Laid-Open No. 3-22979, Cytotechnology, 3, 133 (1990)), pAS3-3 (Japanese Patent Laid-Open No. 2-227075), pCDM8 (Nature, 329, 840 (1987)), pREP4 (Invitrogen), pAGE103 (J. Biochem., 101, 1307 (1987)), pAGE210, pME18SFL3, or pKANTEX93 (International Publication No. 97 / 10354).

[0158] Any promoter that functions in animal cells can be used, including, for example, the promoter of the cytomegalovirus (CMV) IE (immediate early) gene, the SV40 early promoter, a retrovirus promoter, a metallothionein promoter, a heat shock promoter, or an SRα promoter. The enhancer of the human CMV IE gene may also be used together with the promoter.

[0159] Any method for introducing a recombinant vector into an animal cell can be used as long as it is a method for introducing DNA into an animal cell, such as electroporation [Cytotechnology, 3, 133 (1990)], calcium phosphate method (JP Patent Publication No. 2-227075), lipofection [Proc. Natl. Acad. Sci. USA, 84, 7413 (1987)], or the method described in Virology, 52, 456 (1973).

[0160] The DcR3 variants of the present invention can be transiently expressed using expression vectors for the DcR3 variants obtained by the above-mentioned methods, or expression vectors obtained by modifying these vectors.

[0161] Any host cells capable of expressing DcR3 variants can be used for introducing the expression vector, but for example, COS-7 cells (ATCC No. CRL1651) can be used [Methods in Nucleic Acids Res., CRC Press, 283 (1991)]. Expression vectors can be introduced into COS-7 cells using the DEAE-dextran method [Methods in Nucleic Acids Res., CRC Press, (1991)] or the lipofection method [Proc. Natl. Acad. Sci. USA, 84, 7413 (1987)].

[0162] When CHO-S cells or Expi293 cells (Thermo Scientific) are used, the expression vector is introduced by lipofection [Proc. Natl. Acad. Sci. USA, 84, 7413 (1987)] or the like.

[0163] Using the expression vectors for the DcR3 variants of the present invention obtained by the above-mentioned methods, or modified expression vectors thereof, transformants that stably express the DcR3 variants can be obtained. After introducing the expression vector, transformants that stably express the recombinant antibody are selected by culturing in an animal cell culture medium containing an agent such as G418 sulfate (Japanese Patent Application Laid-Open No. 2-257891).

[0164] When the transformant is a transformant obtained using a eukaryotic host such as yeast, either a natural medium or a synthetic medium may be used as a medium for culturing the transformant, as long as it contains a carbon source, a nitrogen source, and / or inorganic salts that can be assimilated by the transformant and allows the transformant to be cultured efficiently.

[0165] The carbon source may be any that can be utilized by the transformant, and examples thereof include carbohydrates such as glucose, fructose, sucrose, molasses containing these, starch or starch hydrolysates, organic acids such as acetic acid and propionic acid, and alcohols such as ethanol or propanol.

[0166] Examples of nitrogen sources that can be used include ammonia, ammonium salts of inorganic or organic acids such as ammonium chloride, ammonium sulfate, ammonium acetate, and / or ammonium phosphate, other nitrogen-containing compounds, as well as peptone, meat extract, yeast extract, corn steep liquor, casein hydrolysate, soybean meal and soybean meal hydrolysate, various fermentation bacteria and digests thereof, and the like.

[0167] Examples of inorganic salts that can be used include monopotassium phosphate, dipotassium phosphate, magnesium phosphate, magnesium sulfate, sodium chloride, ferrous sulfate, manganese sulfate, copper sulfate, and / or calcium carbonate.

[0168] Cultivation is preferably carried out under aerobic conditions such as shaking culture or submerged aeration and stirring culture. The culture temperature is preferably 15 to 40°C, and the culture time is usually preferably 16 hours to 7 days. The pH during cultivation is preferably maintained at 3.0 to 9.0. The pH can be adjusted using inorganic or organic acids, alkaline solutions, urea, calcium carbonate, ammonia, or the like.

[0169] If necessary, antibiotics such as ampicillin or tetracycline may be added to the medium during the culture.

[0170] Media for culturing transformants obtained using insect cells as hosts include commonly used TNM-FH medium (Becton Dickinson), Sf-900 II SFM medium (Invitrogen), ExCell400, ExCell405 (all manufactured by JRH Biosciences), Grace's insect medium [Nature, 195, 788 (1962)], Schneider's Medium (Thermo Fisher), etc.

[0171] The transformant obtained using insect cells as a host is preferably cultured for 1 to 5 days at a pH of 6 to 7 and a temperature of 25 to 30° C. Furthermore, an antibiotic such as gentamicin may be added to the medium during culture, if necessary.

[0172] Media for animal cell culture include RPMI1640 medium (Invitrogen), GIT medium (Nihon Pharmaceutical Co., Ltd.), EX-CELL301 medium (JRH), EX-CELL325 PF CHO serum-free medium (Sigma-Aldlich), IMDM medium (Invitrogen), Hybridoma-SFM medium (Invitrogen), Eagle's minimal essential medium (MEM) [Science, 122, 501 (1952)], Dulbecco's modified Eagle's medium [Virology, 8, 396 (1959)], 199 medium [Proceedings of the Society for Biological Medicine, 73, 1 (1950)], and media containing various additives such as FBS.

[0173] Animal cells are preferably cultured in the presence of 5% CO2, at a pH of 6 to 8, at a temperature of 30 to 40°C, for 1 to 7 days. If necessary, antibiotics such as kanamycin or penicillin may be added to the medium during culture.

[0174] The resulting transformant is cultured in a medium to express and accumulate the DcR3 variant or DcR3 variant in the culture supernatant. The expression level of the DcR3 variant or DcR3 variant can be increased by using a DHFR amplification system (Japanese Patent Laid-Open Publication No. 2-257891) or the like.

[0175] As described above, the DcR3 variant of the present invention can be produced by producing and accumulating it in the culture and then collecting it from the culture.

[0176] To isolate and purify the peptide produced by the transformant, conventional methods for isolating and purifying proteins can be used.

[0177] For example, when a DcR3 variant of the present invention is secreted extracellularly, the DcR3 variant can be recovered from the culture supernatant by centrifuging the culture to obtain a culture supernatant, and then purified products can be obtained from the culture supernatant by using, alone or in combination, conventional protein isolation and purification methods, such as solvent extraction, salting out with ammonium sulfate or the like, desalting, precipitation with organic solvents, anion exchange chromatography using resins such as diethylaminoethyl (DEAE)-Sepharose or DIAION HPA-75 (Mitsubishi Chemical Corporation), cation exchange chromatography using resins such as S-Sepharose FF (GE Healthcare), hydrophobic chromatography using resins such as butyl Sepharose or phenyl Sepharose, gel filtration using molecular sieves, affinity chromatography, chromatofocusing, or electrophoresis such as isoelectric focusing.

[0178] For example, when the DcR3 variant of the present invention has an immunoglobulin Fc capable of binding to Protein G or Protein A, Protein G chromatography or Protein A chromatography can be used as the affinity chromatography method, in which Protein G or Protein A is bound to a carrier as an affinity ligand [Monoclonal Antibodies—Principles and practice, Third edition, Academic Press (1996); Antibodies—A Laboratory Manual, Cold Spring Harbor Laboratory (1988)]. Furthermore, methods used in protein purification, such as gel filtration, ion exchange chromatography, and ultrafiltration, can also be combined.

[0179] Furthermore, DcR3 variants obtained by the above-mentioned methods can be further chemically modified with peptides, sugar chains, PEG, or the like using conventional chemical modification methods.

[0180] 6. Methods for evaluating the biological activity, physical properties, and kinetics of DcR3 variants The DcR3 variants of the present invention preferably include DcR3 variants that have neutralizing activity against at least one of LIGHT, TL1A, and FasL, DcR3 variants that have neutralizing activity against all of LIGHT, TL1A, and FasL, DcR3 variants that do not have neutralizing activity against FasL but have neutralizing activity against either LIGHT or TL1A, and DcR3 variants that do not have neutralizing activity against FasL but have neutralizing activity against LIGHT and TL1A.

[0181] The biological activities such as neutralizing activity, physical properties, and pharmacokinetics of the DcR3 variants of the present invention can be measured using the following methods.

[0182] (1) Preparation of the ligand The origin of LIGHT, TL1A, and FasL used in the present invention is not limited, and examples thereof include eukaryotically derived LIGHT, TL1A, and FasL. Examples of eukaryotically derived LIGHT, TL1A, and FasL include yeast, insect, and mammalian LIGHT, TL1A, and FasL. Preferred examples include LIGHT, TL1A, and FasL derived from primates, including humans, or rodents, including mice.

[0183] Cells expressing LIGHT, TL1A, or FasL or their ligands can be obtained by introducing an expression vector containing a cDNA encoding full-length or partial LIGHT, TL1A, or FasL into appropriate host cells such as E. coli, yeast, insect cells, or animal cells. Alternatively, the ligands can be purified from various human cultured cells or human tissues expressing large amounts of LIGHT, TL1A, or FasL. These cultured cells or tissues can also be used directly as ligands. Furthermore, synthetic peptides containing partial sequences of LIGHT, TL1A, or FasL can be prepared by chemical synthesis methods such as the Fmoc or tBoc method.

[0184] Furthermore, LIGHT, TL1A, and FasL can be produced by introducing DNA encoding LIGHT, TL1A, or FasL into host cells and expressing it using methods described in Molecular Cloning, A Laboratory Manual, Second Edition, Cold Spring Harbor Laboratory Press (1989), for example, by the following method.

[0185] Soluble LIGHT, TL1A, and FasL are produced by shedding the extracellular domain of membrane-bound LIGHT, TL1A, and FasL, respectively, after they are expressed on the cell membrane. Although the cleavage site is identified for each ligand, any sequence near the cleavage site can be included when preparing a recombinant soluble ligand. For example, recombinant soluble LIGHT can be prepared from the region 66-240, 74-240, or 83-240 from the N-terminus of the amino acid sequence of membrane-bound LIGHT. Recombinant soluble TL1A can be prepared from the region 72-251 from the N-terminus of the amino acid sequence of membrane-bound TL1A. Soluble FasL can be prepared from the region 130-281, 134-281, or the like from the N-terminus of the amino acid sequence of membrane-bound FasL.

[0186] Recombinant soluble ligands can be prepared by adding a His tag, FLAG tag, or the like to the N-terminus of the above region and isolated by affinity purification. Examples of the amino acid sequences of soluble recombinant LIGHT, TL1A, and FasL include those set forth in SEQ ID NOs: 114, 116, 118, 120, 122, 124, 126, 128, 130, and 306. Examples of the nucleotide sequences of DNA encoding the amino acid sequences of soluble recombinant LIGHT, TL1A, and FasL include those set forth in SEQ ID NOs: 113, 115, 117, 119, 121, 123, 125, 127, 129, and 305.

[0187] Functional LIGHT, TL1A, or FasL can form homotrimers. Soluble LIGHT, TL1A, or FasL prepared by any of the above methods can be analyzed by SEC-MALS to determine their molecular weight. SEC-MALS is an analytical method in which the molecular weight is calculated using the maximum scattering intensity detected by a multi-angle light scattering (MALS) detector for each peak separated by SEC (size exclusion chromatography)-HPLC at a wavelength of 215 nm. Examples of HPLC systems include Shimadzu's Prominence, SEC columns include Tosoh's TSKgel, and MALS detectors include Wyatt Technology's miniDAWN TREOS. Methods for isolating homotrimers from crude products that form non-trimers include SEC purification. Examples of SEC purification methods include fractionating the crude product by molecular weight using an AKTApurifier HPLC system manufactured by GE Healthcare and a Superdex 200 Increase 10 / 300 GL SEC column manufactured by GE Healthcare, and recovering only the homotrimer molecular weight fraction.

[0188] (2) Evaluation of the binding activity of DcR3 variants containing the human Fc region Methods for measuring the binding activity of DcR3 variants to soluble LIGHT, soluble TL1A, or soluble FasL include, for example, enzyme-linked immunosorbent assay (ELISA) binding assays and Biacore kinetic analysis. Examples of ligands used include transfected cells or recombinant proteins obtained by introducing an expression vector containing a DNA sequence encoding the extracellular domain of each ligand of LIGHT, TL1A, or FasL into Escherichia coli, yeast, insect cells, animal cells, or the like, as well as serum, plasma, or culture supernatant containing the ligand obtained from human tissues or human cells.

[0189] In ELISA, for example, anti-human IgG antibodies are immobilized on a 96-well plate, reacted with DcR3 variants, and then each ligand is dispensed and allowed to bind. After washing, the plate is reacted with unlabeled anti-ligand antibodies or receptor-Fc fusions of each ligand, followed by anti-Fc antibodies labeled with biotin, enzymes, or chemiluminescent substances. Alternatively, labeled anti-ligand antibodies or receptor-Fc fusions of each ligand are reacted, and detection is performed according to the label to measure binding of DcR3 variants to the ligands. Furthermore, in the case of Biacore, for example, the Biacore T100 or Biacore T200 is used to measure the binding kinetics of each ligand to DcR3 variants, and the results are analyzed using the analysis software provided with the instrument. Specifically, anti-human IgG antibodies are immobilized on a CM5 sensor chip by amine coupling, and then the DcR3 variants are injected to allow binding to the sensor chip in appropriate amounts. Then, multiple ligands at known concentrations are injected to measure binding and dissociation. The obtained data are subjected to kinetic analysis using the software provided with the instrument based on a 1:1 binding model to obtain various parameters. Alternatively, each ligand is immobilized on a sensor chip, for example, by amine coupling, and then multiple DcR3 variants at known concentrations are injected to measure binding and dissociation. The obtained data are subjected to kinetic analysis using the software provided with the instrument based on a bivalent binding model to obtain various parameters. Alternatively, a sensor chip Protein A is prepared by immobilizing MabSelectSure ligand, a protein with an IgG-binding domain variant of Protein A, on the sensor chip, and then injecting DcR3 variants to allow appropriate amounts to bind to the sensor chip. Then, multiple ligands at known concentrations are injected to measure binding and dissociation. The obtained data are subjected to kinetic analysis using the software provided with the instrument based on a 1:1 binding model to obtain various parameters.

[0190] The binding activity of DcR3 variants to membrane-type LIGHT, membrane-type TL1A, or membrane-type FasL can be measured by flow cytometry using, for example, transfected cells obtained by introducing expression vectors containing DNA sequences encoding the full-length ligands of LIGHT, TL1A, or FasL into animal cells, or human cells such as PBMCs or HUVECs in which expression of the membrane-type ligands has been induced by stimulation with mitogens such as PHA-L, PMA, or ionomycin, or by stimulation with anti-CD3 and anti-CD28 antibodies, cytokines, IgG, or immune complexes. Specifically, the binding activity can be measured by reacting DcR3 variants with cells expressing the membrane-type ligands and then reacting with fluorescently labeled anti-Fc antibodies, or by reacting DcR3 variants labeled with biotin or a fluorescent dye, followed by washing and detecting the fluorescence intensity corresponding to the labeled substance using a flow cytometer.

[0191] DcR3 variants of the present invention that can be measured by the above-mentioned method include, for example, DcR3 variants that have binding activity to at least one of LIGHT, TL1A, and FasL, DcR3 variants that have binding activity to all of LIGHT, TL1A, and FasL, DcR3 variants that have no binding activity to FasL and have binding activity to either LIGHT or TL1A, or DcR3 variants that have no binding activity to FasL and have binding activity to LIGHT and TL1A.

[0192] (3) Evaluation of the neutralizing activity of DcR3 variants Methods for measuring the neutralizing activity of DcR3 variants include, for example, measuring the binding of LIGHT, TL1A, or FasL to the corresponding receptor in a solution containing the DcR3 variant, or adding the corresponding ligand to cells expressing the LIGHT, TL1A, or FasL receptor in a medium containing the DcR3 variant, and measuring cellular functions such as cytokine production and cell proliferation.

[0193] The inhibitory activity against the binding of human LIGHT to its corresponding receptor can be measured by the following method, for example, using the method described in U.S. Patent No. 8,974,787 B2. In a reaction solution containing a DcR3 variant, HVEM or LTβR labeled with biotin, a fluorescent dye, or the like is reacted with cells expressing membrane-type LIGHT, and after washing, the fluorescence intensity corresponding to the label is detected using a flow cytometer. Alternatively, in a reaction solution containing a DcR3 variant, LIGHT labeled with biotin, a fluorescent dye, or the like is reacted with cells expressing HVEM or LTβR, and after washing, the fluorescence intensity corresponding to the label is detected using a flow cytometer. The inhibitory activity of a DcR3 variant against the binding of LIGHT to its corresponding receptor is confirmed by a decrease in the fluorescence intensity compared to when the DcR3 variant is not added.

[0194] Using the same method as above, the inhibitory activity against the binding of TL1A or FasL to the corresponding receptor can also be measured.

[0195] Preferred DcR3 variants of the present invention include those that have inhibitory activity against the binding of at least one of LIGHT, TL1A, and FasL to their corresponding receptors, those that have inhibitory activity against all of LIGHT, TL1A, and FasL to their corresponding receptors, those that have no inhibitory activity against the binding of FasL to its corresponding receptor but have inhibitory activity against either LIGHT or TL1A to their corresponding receptors, and those that have no inhibitory activity against the binding of FasL to its corresponding receptor but have inhibitory activity against the binding of LIGHT and TL1A to their corresponding receptors. In the present invention, the phrase "having no inhibitory activity against the binding of a certain ligand to its corresponding receptor" is used to encompass a meaning in which the inhibitory activity is significantly reduced compared to wild-type DcR3.

[0196] The inhibitory activity against cellular functions induced by the addition of LIGHT can be measured, for example, using the method described in U.S. Patent No. 8,974,787 B2. In a medium containing a DcR3 variant, cells such as HT-29 (ATCC No. HTB-38), which have been confirmed to express the LIGHT receptors HVEM and LTβR, are used to measure LIGHT-dependent production of chemokines such as IL-8, CCL5, and CCL20 using culture supernatants with ELISA, alphaLISA (Perkin Elmer), or CBA assay (BD Biosciences). Alternatively, similar methods are used to measure soluble or membrane-type LIGHT-dependent production of chemokines such as CXCL-10 using cells such as intestinal myofibroblasts (Lonza) stimulated with IFN-γ. Alternatively, instead of soluble or membrane-type LIGHT, similar assays can be performed using LIGHT induced to express in PBMCs or T cells stimulated with anti-CD3 and anti-CD28 antibodies, or PMA and ionomycin. The inhibitory activity of DcR3 variants on cellular functions induced by the addition of soluble or membrane-type LIGHT is confirmed by a decrease in the production of the chemokine compared to when the DcR3 variant is not added. Furthermore, in vivo, improvements in survival, body weight, pathology, pathology, and human cell engraftment following administration of DcR3 variants can be evaluated in an acute graft-versus-host disease (GVHD) model in which human PBMCs are allogeneically transplanted into immunodeficient mice, using methods such as those described in U.S. Patent No. 8,974,787 B2.

[0197] The inhibitory activity against cell function induced by the addition of TL1A can be measured, for example, using the method described in Mucosal Immunology, 2015, 8:545-558, as follows: Blood, PBMCs, pan T cells, CD4+ T cells, or memory CD4+ T cells derived from primates such as humans or rodents such as mice are used in a medium containing a modified DcR3. The cells are stimulated with a cytokine cocktail of IL-12, IL-18, and TL1A, or IL-12, IL-18, IL-15, and TL1A, and TL1A-dependent cytokine production, such as IFN-γ, IL-6, GM-CSF, TNF-α, IL-5, IL-13, IL-17, or IL-22, is measured by ELISA, alphaLISA, CBA assay, or the like using the culture supernatant. Alternatively, using the method described in Immunity, 2002, 16:479-492, for example, pan T cells, CD4+ T cells, or memory CD4+ T cells derived from primates such as humans or rodents such as mice are stimulated with anti-CD3 and anti-CD28 antibodies in a medium containing a DcR3 variant, and TL1A-dependent IFN-γ and IL-2 production is measured using the same method as described above. Alternatively, measurements can be performed using a cell line expressing membrane-type TL1A instead of soluble TL1A. Alternatively, measurements can be performed using TL1A induced from PBMCs or monocytes stimulated with IgG or immune complexes. The inhibitory activity of DcR3 variants on cellular functions induced by the addition of TL1A is confirmed by a decrease in the cytokines or inhibition of cell proliferation compared to when the DcR3 variant is not added. Furthermore, in vivo, improvements in survival, body weight, pathology, etc. due to administration of DcR3 variants can be evaluated, for example, in the TNBS (2,4,6-trinitrobenzenesulfonic acid)-induced colitis model described in Mucosal Immunology, 2011, 4: pp. 172-185, or the DSS (dextran sodium sulfate)-induced colitis model described in Mucosal Immunology, 2014, 7: pp. 1492-1503.In addition, the effect of improving pathology can be evaluated in a TL1A-dependent inflammation, allergic disease, or autoimmune disease model in rodents such as mice.

[0198] The inhibitory activity of FasL-induced cell functions can be measured, for example, using the method described in J. Rheumatol., 2013, 40:1316-1326, as follows. Using Jurkat cells (DSMZ No.: ACC 282) in a medium containing a DcR3 variant, apoptosis dependent on soluble FasL or antibody-crosslinked soluble FasL is measured by Annexin V / Propium Iodide staining, BrdU incorporation into viable cells, or ATP levels. Alternatively, a similar method can be used to measure apoptosis using a cell line expressing membrane-type FasL instead of soluble FasL. Alternatively, a similar method can be used to measure FasL induced from stimulated PBMCs or T cells. The inhibitory activity of DcR3 variants on FasL-induced cell functions is confirmed by a decrease in apoptosis compared to when no DcR3 variant is added.

[0199] DcR3 variants of the present invention that can be measured by the above-mentioned method include, for example, DcR3 variants that have inhibitory activity against at least one or more of the cellular functions induced by the addition of LIGHT, TL1A, and FasL, DcR3 variants that have inhibitory activity against all of the cellular functions induced by the addition of LIGHT, TL1A, and FasL, DcR3 variants that do not have inhibitory activity against cellular functions induced by the addition of FasL but have inhibitory activity against one or more of the cellular functions induced by the addition of LIGHT and TL1A, and DcR3 variants that do not have inhibitory activity against cellular functions induced by the addition of FasL but have inhibitory activity against cellular functions induced by the addition of LIGHT and TL1A. In the present invention, the expression "does not have inhibitory activity against cellular functions induced by the addition of a certain ligand" is used to mean that the inhibitory activity is significantly reduced compared to wild-type DcR3.

[0200] (4) Evaluation of OPG ligand reactivity of DcR3 variants One of the features of the DcR3 variants of the present invention is that any or all of the parts of CRD1, CRD4, and CRD3 that are not involved in ligand binding have sequences derived from OPG.

[0201] The lack of binding activity of DcR3 variants to the OPG ligands RANKL and TRAIL can be evaluated, for example, by methods similar to those used to measure binding activity to LIGHT, TL1A, and FasL, as described above.

[0202] Another feature of the DcR3 variant of the present invention is that it does not have neutralizing activity against RANKL and TRAIL.

[0203] The neutralizing activity of DcR3 variants against RANKL can be evaluated, for example, by measuring TRAP (tartrate-resistant acid phosphatase) activity, which is an assay for differentiation of osteoclast precursor cells stimulated by RANKL, as described in J. Immunol., 2012, 189: pp. 245-252.

[0204] Furthermore, the neutralizing activity of DcR3 variants against TRAIL can be evaluated, for example, by measuring apoptosis induction by soluble TRAIL or cross-linked soluble TRAIL in human cancer cell lines expressing DR4 or DR5 using a method similar to the method for measuring the neutralizing activity against FasL described above.

[0205] (5) Dynamics of DcR3 mutants The DcR3 variants of the present invention are preferably DcR3 variants that have reduced or no binding to heparan sulfate contained in heparan sulfate proteoglycans (HSPGs) on the cell membrane, and more preferably those that do not have a heparan sulfate binding domain (HBD).

[0206] The presence or absence of binding to heparan sulfate on the cell membrane via the HBD can be measured, for example, using the method described in J. Immunol., 2006, 176:173-180, as follows: Any cells, such as CHO cells, human cell lines, vascular endothelial cells, hepatocytes, or blood cells, are reacted with wild-type DcR3 or DcR3 variants, followed by a fluorescently labeled detection antibody or wild-type DcR3 or DcR3 variants labeled with biotin, a fluorescent dye, or the like. After washing, the cells can be measured by detecting the fluorescence intensity corresponding to the label using a flow cytometer (FCM). The reduction or loss of cell membrane binding due to the deletion of the HBD of DcR3 can be determined by the reduction in fluorescence intensity measured by FCM.

[0207] Methods for assessing whether the above-mentioned binding to the cell membrane is mediated by the HBD of DcR3 include, for example, using the method described in J. Immunol., 2006, 176: pp. 173-180, adding GAGs such as heparin or heparan sulfate as inhibitors during the reaction of wild-type DcR3 or DcR3 variants, or treating the cells in advance with an enzyme such as heparinase or trypsin, and then reacting with wild-type DcR3 or DcR3 variants.

[0208] One of the features of the DcR3 variants of the present invention is that they exhibit improved pharmacokinetics compared to wild-type DcR3 due to reduced heparan sulfate-mediated elimination in vivo. In the present invention, "improved pharmacokinetics" means that they have a longer half-life in the blood or a higher area under the blood concentration-time curve (AUC) up to infinity compared to wild-type DcR3.

[0209] Instead of wild-type DcR3, molecules containing the CRD of wild-type DcR3 can also be used as controls. Specific examples include DcR3 FL-Fc (amino acid sequence: SEQ ID NO: 100, DNA sequence: SEQ ID NO: 99), DcR3 FL-FLAG (amino acid sequence: SEQ ID NO: 104, DNA sequence: SEQ ID NO: 103), S195-Fc (amino acid sequence: SEQ ID NO: 102, DNA sequence: SEQ ID NO: 101), which fuses the CRD of wild-type DcR3 with Fc, and R128Q-Fc (U.S. Patents US6,835,814 B1, US6,965,01 B1), which fuses Fc to the C-terminus of full-length DcR3 (amino acid sequence: SEQ ID NO: 112, DNA sequence: SEQ ID NO: 111) with a single amino acid mutation in the HBD. When these molecules containing the CRD of wild-type DcR3 are used as controls, they are also referred to as wild-type DcR3 controls.

[0210] The blood concentration of a DcR3 variant in rodents (e.g., mice) or non-human primates (e.g., cynomolgus monkeys) can be measured by intravenously or subcutaneously administering a desired dose, drawing blood at desired times, and then using a DcR3 or human Fc detection system. Kinetic parameters such as blood half-life and AUC can be calculated from the blood concentration profile using techniques such as moment analysis, for example, as described in Pharmacokinetics, 1999, 14:286-293.

[0211] Furthermore, because it is known that the more sialic acids added to the terminus of an N-glycan, the better its pharmacokinetics (J. Pharm. Sci., 2015, 104: pp. 1866-1884), it is preferable that the DcR3 variant of the present invention having an N-glycan also has a large number of sialic acids added. The number of sialic acids added per protein molecule can be calculated, for example, by separating sialic acids labeled with a sialic acid fluorescent labeling reagent kit (Takara) or the like by reverse-phase HPLC and comparing the results with a standard curve for sialic acids.

[0212] (6) Evaluation of the physical properties of DcR3 variants One of the characteristics of the DcR3 variants of the present invention is that when expressed in mammalian cells, isolated, or purified, they have a lower aggregate content than wild-type DcR3.

[0213] Whether or not aggregates are formed during protein expression and secretion can be determined, for example, by collecting host cells and / or culture supernatants into which a recombinant vector carrying a DcR3 variant has been introduced, and examining the approximate molecular weight under non-reducing conditions by immunoblotting with an anti-DcR3 antibody or, in the case of Fc fusions or tagged forms such as His or FLAG, antibodies against those antibodies. If the protein aggregates during expression and / or secretion, one or more bands larger than the predicted molecular weight will be detected.

[0214] The molecular weight of a protein can be calculated from the amino acid sequence. To obtain a more accurate molecular weight, an analytical method using SEC-MALS can be used. For each peak detected at a wavelength of 215 nm separated by SEC (Size Exclusion Chromatography)-HPLC, the molecular weight can be calculated using the maximum scattering intensity detected by a multi-angle light scattering (MALS) detector. Examples of HPLC systems include Shimadzu's Prominence, SEC columns include Tosoh's TSKgel, and MALS detectors include Wyatt Technology's miniDAWN TREOS.

[0215] Whether an isolated or purified protein is aggregated can be determined by subjecting the protein to SDS-PAGE under non-reducing conditions and detecting the protein by CBB staining or by immunoblotting using the same method as described above. The aggregate content can also be determined by calculating the percentage of each peak from the area of each peak detected at a wavelength of 215 nm by gel filtration chromatography (SEC) using HPLC. Examples of SEC columns include Tosoh's TSKgel G3000SW and Waters' ACQUITY UPLC Protein BEH SEC. When analyzed using the above method, the aggregate content is preferably 0 to 60%, more preferably 0 to 40%, even more preferably 0 to 30%, even more preferably 0 to 20%, and most preferably 0 to 10%.

[0216] Furthermore, the DcR3 variants of the present invention are characterized by being less hydrophobic than wild-type DcR3. The hydrophobicity of a protein can be measured using a hydrophobic interaction chromatography (HIC) column, which interacts with the hydrophobic regions present on the surface of the protein. Examples of HIC columns include TSKgel Butyl-NPR (Tosoh Corporation).

[0217] Furthermore, the DcR3 variants of the present invention are characterized by improved thermal stability compared to wild-type DcR3. Methods for measuring protein thermal stability include calorimetry such as DSC (Differential Scanning Calorimetry), spectroscopy that obtains autofluorescence or circular dichroism (CD) spectra during thermal or chemical denaturation, and DSF (Differential Scanning Fluorimetry), which uses a fluorescent dye (e.g., Sypro Orange) to detect the exposure of hydrophobic regions present inside the protein as the temperature increases [J Am Chem Soc, 2009, 131: 3794-3795].

[0218] DSF for assessing protein thermal stability can be performed using, for example, the method described in J. Pharm. Sci., 2013, 102: pp. 2471-2483, where the fluorescence intensity at each temperature can be measured. Alternatively, a melting curve can be plotted using software such as BioRad's CFX Manager, and the Tm (thermal unfolding transition midpoints) value can be calculated. Similarly, DSC for assessing protein thermal stability can be performed using, for example, the method described in J. Pharm. Sci., 2012, 101: pp. 955-964, where the heat capacity and Tm value at each temperature can be calculated.

[0219] 7. DcR3 variant composition DcR3 variant compositions of the present invention include compositions consisting of DcR3 variant molecules. Examples of DcR3 variant compositions of the present invention include compositions containing multiple DcR3 variant molecules that have the same primary amino acid sequence and that may be generated by post-translational modifications, such as oxidation / reduction, glycosylation, and sulfation, within the amino acid sequence. DcR3 variants of the present invention may include, for example, DcR3 variants bearing one or more N-glycoside-linked complex-type glycans and DcR3 variants lacking N-glycoside-linked complex-type glycans. The proportion of DcR3 variants bearing one or more N-glycoside-linked complex-type glycans relative to the total number of DcR3 variants of the present invention is, for example, preferably 70-100%, more preferably 90-99%, and particularly preferably 95-98%.

[0220] 8. Pharmaceutical Compositions Containing DcR3 Variants One embodiment of the present invention is a composition comprising an effective amount of a DcR3 variant of the present invention. Compositions containing the DcR3 variants of the present invention can be used as active ingredients in drugs for the prevention or treatment of autoimmune diseases, inflammatory diseases, or allergic diseases, including mucosal diseases. Specifically, autoimmune diseases, inflammatory diseases, or allergic diseases can be prevented or treated by administering a pharmaceutical composition containing the DcR3 variant of the present invention to a patient in need of prevention or treatment of an autoimmune disease, inflammatory disease, or allergic disease.

[0221] Pathological conditions or diseases for which the composition of the present invention can be used include, but are not limited to, inflammatory bowel disease (IBD), systemic lupus erythematosus, psoriasis, chronic graft-versus-host disease, acute graft-versus-host disease, Crohn's disease, ulcerative colitis, inflammatory bowel disease, multiple sclerosis, celiac disease, idiopathic thrombotic thrombocytopenic purpura, myasthenia gravis, Sjogren's syndrome, scleroderma, asthma, uveitis, epidermal hyperplasia, alopecia areata, Behcet's disease, Takayasu's arteritis, cartilage inflammation, bone degradation, arthritis, juvenile arthritis, juvenile rheumatoid arthritis, oligoarticular juvenile rheumatoid arthritis, polyarticular juvenile rheumatoid arthritis, systemic-onset juvenile rheumatoid arthritis, juvenile ankylosing spondylitis, juvenile enteropathic arthritis, juvenile reactive arthritis, juvenile Reiter's syndrome, SEA syndrome (seronegative, enthesopathy, arthropathy syndrome), juvenile skin disease, Myositis, juvenile psoriatic arthritis, juvenile scleroderma, juvenile systemic lupus erythematosus, juvenile vasculitis, oligoarticular rheumatoid arthritis, polyarticular rheumatoid arthritis, systemic-onset rheumatoid arthritis, ankylosing spondylitis, enteropathic arthritis, reactive arthritis, Reiter's syndrome, dermatomyositis, psoriatic arthritis, vasculitis, myositis, polymyositis, dermatomyositis, osteoarthritis, polyarteritis nodosa, Wegener's granulomatosis, arteritis, rheumatoid arthritis Examples of such conditions include inflammatory, autoimmune, or allergic diseases such as polymyalgia, sarcoidosis, sclerosis, primary biliary cirrhosis, sclerosing cholangitis, dermatitis, atopic dermatitis, atherosclerosis, Still's disease, chronic obstructive pulmonary disease, Guillain-Barré syndrome, type 1 diabetes, Graves' disease, Addison's disease, Raynaud's phenomenon, autoimmune hepatitis, or Wiskott-Aldrich syndrome.

[0222] Pharmaceutical compositions containing the DcR3 variants of the present invention may contain the DcR3 variants as an active ingredient or a mixture with any other active ingredient for treatment. These pharmaceutical preparations are prepared by mixing the active ingredient with one or more pharmacologically acceptable carriers and by any method well known in the technical field of pharmaceuticals.

[0223] The content of the DcR3 variant of the present invention in the pharmaceutical composition varies depending on the dosage form, the pharmacologically acceptable dose of the DcR3 variant of the present invention, etc., but is, for example, about 0.01 to 100% by weight. The content of the pharmacologically acceptable carrier in the pharmaceutical preparation varies depending on the dosage form, the pharmacologically acceptable dose of the DcR3 variant of the present invention, etc., but is, for example, 0 to 99.9% by weight.

[0224] The route of administration is preferably the most effective for the treatment, and examples thereof include oral administration and parenteral administration such as intravenous, subcutaneous, oral, intratracheal, rectal, intramuscular, or intraperitoneal administration.

[0225] The dosage forms include tablets, powders, granules, syrups, and injections.

[0226] For example, liquid preparations such as syrups can be prepared using water, sugars such as sucrose, sorbitol, or fructose, glycols such as polyethylene glycol or propylene glycol, oils such as sesame oil, olive oil, or soybean oil, preservatives such as p-hydroxybenzoic acid esters, flavors such as strawberry flavor or peppermint, etc. Furthermore, tablets, powders, granules, etc. can be prepared using excipients such as lactose, glucose, sucrose, or mannitol, disintegrating agents such as starch or sodium alginate, lubricants such as magnesium stearate or talc, binders such as polyvinyl alcohol, hydroxypropyl cellulose, or gelatin, surfactants such as fatty acid esters, plasticizers such as glycerin, etc.

[0227] Preparations suitable for parenteral administration preferably consist of a sterile aqueous solution containing an active compound that is isotonic with the recipient's blood. For example, in the case of injections, solutions for injection are prepared using a carrier such as a saline solution, a glucose solution, or a mixture of saline and a glucose solution.

[0228] Furthermore, these parenteral preparations may also contain one or more auxiliary ingredients selected from the diluents, preservatives, flavors, excipients, disintegrants, lubricants, binders, surfactants, plasticizers, and the like exemplified for oral preparations.

[0229] Pharmaceuticals containing the DcR3 variants of the present invention can be safely administered to mammals (eg, humans, mice, rats, rabbits, dogs, cats, cows, horses, pigs, monkeys, etc.).

[0230] The dosage and frequency of administration of the DcR3 variants of the present invention vary depending on the dosage form, the patient's age, weight, disease, and the nature or severity of the symptoms to be treated. Oral administration typically involves administering 0.01 mg to 1 g, preferably 0.05 to 50 mg, per adult, once or several times per day. Parenteral administration, such as intravenous administration, typically involves administering 0.001 to 100 mg, preferably 0.01 to 10 mg, per adult, once or several times per day. However, these dosages and frequency of administration vary depending on the various conditions mentioned above. [Example]

[0231] Examples of the present invention will be described below, but the present invention is not limited to these examples.

[0232] [Example 1] Evaluation of aggregation properties of wild-type DcR3 in mammalian cells A fusion protein (DcR3 FL-Fc) (Fig. 3A, SEQ ID NO: 100) of wild-type DcR3 (also referred to as full-length DcR3) (SEQ ID NO: 4), the linker sequence IEGRM (SEQ ID NO: 106), and the human IgG1 Fc region (g1S) (SEQ ID NO: 72), a fusion protein (S195-Fc) (Fig. 3B, SEQ ID NO: 102) of human DcR3 lacking the HBD region (SEQ ID NO: 108), the linker sequence IEGRM (SEQ ID NO: 106), and the human IgG1 Fc region (SEQ ID NO: 72), and a protein (DcR3 FL-FLAG) (SEQ ID NO: 104) in which a FLAG tag (SEQ ID NO: 110) was added to full-length DcR3 (SEQ ID NO: 4) were prepared as follows.

[0233] For DcR3 FL-Fc, a DNA fragment of the signal peptide sequence, a DNA fragment of human DcR3 (sequence number 3), a DNA fragment of the linker sequence IEGRM (sequence number 105), and a DNA fragment of Fc(g1S) (sequence number 71) were artificially synthesized (by Genewiz or Sigma) and inserted into a pCIpuro vector (a partially modified version of Promega's pCI) that had been digested with the restriction enzymes NheI and SalI (New England Biolabs) using the In-Fusion HD Cloning Kit (Clontech). Escherichia coli DH5α competent cells (Toyobo) were then transformed to obtain transformants containing the DcR3 FL-Fc DNA fragment (sequence number 99). Similarly, for S195-Fc, a DNA fragment of the signal peptide sequence, a DNA fragment of human DcR3 minus the HBD (SEQ ID NO: 107), a DNA fragment of the linker sequence IEGRM (SEQ ID NO: 105), and a DNA fragment of Fc(g1S) (SEQ ID NO: 71) were artificially synthesized, and a transformant into which the S195-Fc DNA fragment (SEQ ID NO: 101) was inserted was obtained. For DcR3 FL-FLAG, a DNA fragment of the signal peptide sequence, a DNA fragment of human DcR3 (SEQ ID NO: 3), and a DNA fragment of the FLAG tag (SEQ ID NO: 109) were artificially synthesized, and a transformant into which the DcR3 FL-FLAG DNA fragment (SEQ ID NO: 103) was inserted was obtained.

[0234] The plasmids obtained from each transformant were transfected into Freestyle CHO-S cells, Freestyle 293F cells, or Expi293 cells (all from Thermo Scientific) to transiently express the proteins. Plasmid transfection was performed using Freestyle MAX Reagent, 293Fectin Transfection Reagent, or ExpiFectamine 293 (all from Thermo Scientific).

[0235] After culturing the transfected cells for several days, the culture supernatants were collected and affinity-purified using MabSelect SuRe (GE Healthcare) to extract DcR3 FL-Fc and S195-Fc. Anti-FLAG M2 affinity gel (Sigma) was used to purify DcR3 FL-FLAG. The culture supernatants containing DcR3 FL-Fc and S195-Fc were passed through a resin-packed column, washed with PBS (Nacalai Tesque), and eluted with elution buffer (20 mM citric acid, 50 mM NaCl, pH 3.4). The column was immediately neutralized with neutralization buffer (1 M Na phosphate, pH 7.0). The culture supernatant of DcR3 FL-FLAG was similarly loaded onto the resin, washed with PBS, and then eluted with elution buffer (100 mM glycine-HCl, pH 3.5) and immediately neutralized with neutralization buffer (1 M Tris-HCl, pH 8.0). The absorbance at 280 nm (A 280) was measured, and consecutive fractions with high values were collected. The buffer solution of the collected fractions was replaced with PBS using a NAP column (GE Healthcare), and the purified protein was passed through a 0.22 μm filter. The concentrations were calculated based on the extinction coefficients at 280 nm of DcR3 FL-Fc, S195-Fc, and DcR3 FL-FLAG of 1.03, 1.17, and 0.77, respectively. After SDS-PAGE under non-reducing and reducing conditions with 100 mM DTT, the gel was stained with Coomassie (Nacalai Tesque) to confirm the molecular weight. The predicted molecular weights of the monomers of DcR3 FL-Fc, S195-Fc, and DcR3 FL-FLAG, based on their amino acid sequences, are approximately 56.4 kDa, 44.7 kDa, and 31.0 kDa, respectively. Under non-reducing conditions, the Fc fusion products DcR3 FL-Fc and S195-Fc exist as dimers, with estimated molecular weights of approximately 112.8 kDa and 89.4 kDa, respectively.

[0236] SDS-PAGE under non-reducing conditions revealed that most of the DcR3 FL-Fc, S195-Fc, and DcR3 FL-FLAG transiently expressed in mammalian cells remained in the sample wells, and the electrophoresed samples formed a smear or ladder-like pattern, indicating that all recombinants were highly aggregated (Fig. 1A).

[0237] Furthermore, when immunoblotting was performed on commercially available full-length DcR3-Fc (Abcam), which was produced using HEK293 cells as host cells, under non-reducing conditions using rabbit anti-human IgG Fc polyclonal antibody as the primary antibody and goat anti-rabbit IgG antibody (Dako) as the secondary antibody, most of the DcR3-Fc was found to exist as aggregates, similar to the purified product described above (Figure 1B).

[0238] On the other hand, commercially available DcR3 FL-Fc (R&D) expressed in insect Sf21 cells showed little aggregation, indicating differences in aggregation ability depending on the type of host cell used to produce wild-type DcR3 (Figure 1C). Furthermore, using insect S2 cells as the host, we produced DcR3 FL-Fc (SEQ ID NO: 100), R218Q-Fc (U.S. Patents US6835814B1, US6965012B1, SEQ ID NO: 340), which contains the R218Q mutation in human DcR3 (SEQ ID NO: 2) by substituting Arg at position 218 with Gln, and S195-Fc (SEQ ID NO: 102), which lacks the HBD, and evaluated their aggregation ability. We artificially synthesized a DNA fragment of human DcR3 (SEQ ID NO: 3), a DNA fragment of human DcR3 with the R218Q mutation (SEQ ID NO: 111), or a DNA fragment of human DcR3 lacking the HBD (SEQ ID NO: 107), along with a DNA fragment of the linker sequence IEGRM (SEQ ID NO: 105) and a DNA fragment of Fc(g1S) (SEQ ID NO: 71). Using pMTBiPV5-HisA (Thermo Scientific) and the Drosophila Expression System (Thermo Scientific), we isolated S2 cell lines stably expressing DcR3 FL-Fc, R218Q-Fc, and S195-Fc. Culture supernatants from these stable cell lines were affinity purified using MabSelect SuRe as described above.

[0239] As a result, DcR3 FL-Fc, R218Q-Fc, and S195-Fc produced in S2 cells showed little aggregation (Figure 1C). Quantitative evaluation of aggregate content was performed by gel filtration chromatography (SEC) (TSKgel G3000 SWXL 7.8 mm x 300 mm) (Tosoh Corporation) using HPLC (Shimadzu Corporation). Expi293-derived and S2-derived S195-Fc were analyzed, and the percentages (%) of monomer, aggregates, and degradation products calculated from the peak areas are shown in Table 1.

[0240] [Table 1]

[0241] These results demonstrate that wild-type DcR3 increases the amount of aggregates produced when expressed in mammalian cells.

[0242] [Example 2] Preparation of DcR3 variants that do not aggregate in a mammalian cell expression system No human DcR3 variants that do not aggregate in mammalian cell expression systems have been known. Therefore, we attempted to generate DcR3 variants that maintain DcR3 activity while reducing the amount of aggregate formation. DcR3 is a soluble molecule consisting of 300 residues. It contains an N-terminal signal peptide followed by four cysteine-rich domains (CRD1, CRD2, CRD3, and CRD4) characteristic of the TNF receptor superfamily (TNFRSF). The C-terminal domain contains a basic amino acid-rich heparan sulfate binding domain (HBD) containing a heparan sulfate-binding motif (Fig. 2 and 3A). The DcR3 ligands LIGHT, TL1A, and FasL all bind via CRD2 and CRD3 of DcR3. Therefore, we prepared a DcR3 variant by substituting CRD1 and / or CRD4 of DcR3 with a soluble decoy receptor, osteoprotegerin (OPG), a related TNFRSF molecule, while retaining the region containing CRD2 and CRD3 of DcR3.

[0243] Chimera A-Fc (FIG. 3C, SEQ ID NO: 80 or 82) was obtained by fusing chimera A (SEQ ID NO: 54) in which CRD1 and CRD4 of human DcR3 were replaced with the amino acid sequence of human OPG, CRD2 and CRD3 were replaced with the amino acid sequence of human DcR3, and the amino acid sequence of the HBD was deleted with an Fc sequence (IEGRMD g1S (SEQ ID NO: 339) or g4PEK (SEQ ID NO: 74) in which Ser at position 228 of the heavy chain of human IgG4 as shown in the EU index was replaced with Pro, Leu at position 235 with Glu, and Arg at position 409 with Lys). Chimera B (SEQ ID NO: 50) in which CRD1 of human DcR3 was replaced with the amino acid sequence of human OPG, and CRD2, CRD3, and CRD4 were replaced with the amino acid sequence of human DcR3, and the amino acid sequence of the HBD was deleted with an Fc sequence (IEGRMD g1S (SEQ ID NO: 339) or g4PEK (SEQ ID NO: 74) in which Ser at position 228 of the heavy chain of human IgG4 as shown in the EU index was replaced with Pro, Leu at position 235 with Glu, and Arg at position 409 with Lys). Various DcR3 variants with different structures were prepared using the following methods: chimera B-Fc (Fig. 3G, SEQ ID NO: 76) in which CRD1, CRD2, and CRD3 of human DcR3 are fused with the Fc sequence (IEGRMD g1S); chimera C-Fc (Fig. 3H, SEQ ID NO: 78) in which chimera C (SEQ ID NO: 52) was prepared by fusing the Fc sequence (IEGRMD g1S) with human DcR3; and 103-123OPG-Fc(g4PEK) (Fig. 3D, SEQ ID NO: 84) in which the amino acid sequence of positions 18 to 36 of CRD3 of chimera A and two amino acids at its C-terminus were substituted with human OPG. In addition, the DcR3 variant in which the CRD4 is derived from DcR3 has the TS sequence, which is the amino acid sequence from positions 194 to 195 of DcR3 (sequence number 2), added to the C-terminus of CRD4, and the DcR3 variant in which the CRD4 is derived from OPG has the SGNSESTQK sequence, which is the amino acid sequence from positions 186 to 194 of OPG (sequence number 14), added to the C-terminus of CRD4.

[0244] DNA fragments encoding the signal peptide sequence, chimera A, chimera B, chimera C, or 103-123OPG, with the HBD sequence removed (chimera A: sequence number 53, chimera B: sequence number 49, chimera C: sequence number 51, 103-123OPG: sequence number 55) were artificially synthesized and linked to DNA fragments encoding the Fc sequence (IEGRMD g1S or g4PEK) (sequence numbers 338 and 73), and inserted into the pCIpuro vector in the same manner as in Example 1 to obtain plasmids into which DNA fragments encoding various DcR3 variants (chimera A-Fc: sequence number 79 or 81, chimera B-Fc: sequence number 75, chimera C-Fc: sequence number 77, 103-123OPG-Fc: sequence number 83) were inserted. The obtained plasmid was introduced into any of host cells, Freestyle CHO-S cells, Freestyle 293F cells, or Expi293 cells, as in Example 1, to transiently express the protein, and affinity purification was performed from the culture supernatant using MabSelectSuRe.

[0245] The percentages (%) of monomer, aggregates, and degradation products of the various DcR3 variants prepared were calculated by SEC-HPLC using the same method as in Example 1, or by SEC (Waters) using an ACQUITY UPLC Protein BEH SEC 4.6 mm x 150 mm (Table 2).

[0246] [Table 2]

[0247] As a result, all of the DcR3 variants, chimera A-Fc, chimera B-Fc, chimera C-Fc, and 103-123OPG-Fc, showed reduced aggregation compared to wild-type DcR3. Chimera A-Fc was most effective in reducing aggregation, followed by chimera B-Fc and chimera C-Fc. Therefore, it was demonstrated that substitution with OPG-derived CRD1 and CRD4 each reduced aggregation, and that combined substitution with both CRD1 and CRD4 further reduced aggregation.

[0248] When chimera A-Fc, chimera B-Fc, and chimera C-Fc (all IEGRM D g1S) were subjected to SDS-PAGE under non-reducing and reducing conditions with 100 mM DTT, almost no smear or ladder was observed in chimera A-Fc and chimera B-Fc, even when produced using mammalian cells as hosts (Figure 4).

[0249] To calculate absolute molecular weights, chimera A-Fc (IEGRMD g1S or g4PEK) and 103-123OPG-Fc (g4PEK) were analyzed by SEC (TSKgel G3000 SWXL 7.8 mm x 300 mm) (Tosoh Corporation) using HPLC (Prominence, Shimadzu Corporation). After separation at a flow rate of 0.75 mL / min using 50 mmol / L phosphate buffer (pH 7.0, 500 mmol / L NaCl) as the mobile phase, absolute molecular weights (% uncertainty) were calculated for each peak detected at 215 nm using the scattered light intensity detected by a multi-angle light scattering (MALS) detector (miniDAWN TREOS, Wyatt Technology). The results are shown in Table 3.

[0250] [Table 3]

[0251] The molecular weights of the dimers predicted from the amino acid sequences of chimera A-Fc (IEGRMD g1S or g4PEK) and 103-123OPG-Fc were 92.0 kDa, 90.1 kDa, and 90.5 kDa, respectively, confirming that all DcR3 variants exist as dimers.

[0252] [Example 3] Analysis of N-glycosylation rate of chimeric A-Fc(g4PEK) and evaluation of its effect on aggregation Based on the amino acid sequence, chimera A-Fc(g4PEK) (SEQ ID NO: 82) contains three Asn positions predicted for N-glycosylation in the OPG-derived CRD4 (N131, N144, and N157) and one Asn position in the Fc region (N260). Therefore, the presence or absence of N-glycosylation in chimera A-Fc(g4PEK) was evaluated as follows. The N-glycans of reduced and alkylated chimera A-Fc(g4PEK) were cleaved by PNGaseF treatment, followed by protein digestion with trypsin, endoproteinase Asp-N, and chymotrypsin. The resulting peptide mixture was dissolved in 5% (v / v) acetonitrile / 0.1% formic acid and analyzed by liquid chromatography-electrospray ionization-mass spectrometry (LC-ESI-MS). Analysis was performed using a MAGIC2000 HPLC (Michrom Bioresources) equipped with a C18 reverse-phase column (0.2 mm x 150 mm) (GL Science) and an LTQ Orbitrap XL ion trap-Orbitrap hybrid mass spectrometer (Thermo Scientific) with a gradient elution of 5-65% (v / v) acetonitrile / 0.1% formic acid. The resulting peptide fragments were identified by MASCOT analysis (Matrix Science) of the amino acid sequence of chimera A-Fc(g4PEK). N-glycosylated Asn was converted to Asp by PNGaseF treatment, resulting in a mass increase of 0.984 Da. This information was used to identify peptides with shifted peaks in the MS spectrum and the glycosylation sites within these peptides. Results indicated that N-glycosylation was observed at all Asn residues: N131, N144, N157, and N260. Of these, N157 and N260 had glycosylated peptide fragments in almost all peptide fragments, but for N131 and N144, both glycosylated and non-glycosylated fragments were detected.

[0253] Next, to evaluate the effect of N-glycosylation on aggregation of the three Asn residues (N131, N144, and N157) in OPG-derived CRD4, we constructed mutants with amino acid substitutions to remove either two N131 and N144 or three N131, N144, and N157 glycosylation sites. Two mutants with two glycosylation sites were constructed: N131S / N144S-Fc(g4PEK) (SEQ ID NO: 86), in which N131 and N144 were substituted with Ser; and T133A / S146A-Fc(g4PEK) (SEQ ID NO: 88), in which T133 and S146 were substituted with Ala. As triple-deglycosylated Fc, N131S / N144S / N157S-Fc(g4PEK) (SEQ ID NO: 90) was prepared by substituting N131, N144, and N157 with Ser, and T133A / S146A / T159A-Fc(g4PEK) (SEQ ID NO: 92) was prepared by substituting T133, S146, and T159 with Ala. DNA fragments encoding the signal peptide sequence and the respective deglycosylated Fc fragments (SEQ ID NOs: 85, 87, 89, and 91) were artificially synthesized and inserted into the pCIpuro vector as in Example 1 and then transfected into Expi293 cells. Each deglycosylated Fc was transiently expressed, and affinity purified from the culture supernatant using MabSelectSuRe. Each of the prepared deglycosylated products was analyzed by SEC-UPLC (apparatus: ACQUITY UPLC, column: ACQUITY UPLC Protein BEH SEC 200 Å, 1.7 μm, 4.6 × 150 mm) (Waters). The percentages (%) of monomer, aggregate, and degraded product calculated from the peak areas are shown in Table 4.

[0254] [Table 4]

[0255] The aggregate content of the double-deglycosylated DcR3 was slightly increased for all amino acid substitutions compared to chimera A-Fc (Table 2), and further increased for the triple-deglycosylated DcR3 for all amino acid substitutions. This suggests that, of the three N-glycosylations (N131, N144, and N157) on the OPG-derived CRD4, the N157 glycan in particular contributes to the reduction of chimera A aggregates. On the other hand, even when all three N-glycosylations were removed, the aggregate fraction was reduced compared to DcR3 FL-Fc and S195-Fc (Table 1). These results suggest that the N-glycosylation and OPG sequences each have the effect of reducing wild-type DcR3 aggregates, and that combining the two further reduces aggregates.

[0256] [Example 4] Evaluation of the physical properties of various DcR3 variants To analyze the aggregation-reducing effects of the various DcR3 variants prepared in Examples 2 and 3, the following physicochemical analyses were performed. Hydrophobic interaction chromatography (HIC) is an analytical method in which the elution time from the column increases with increasing hydrophobicity of the protein surface. Using a hydrophobic chromatography column (TSKgel Butyl-NPR 4.6 mm × 35 mm) (Tosoh Corporation), 8 μg of sample was separated at a flow rate of 0.5 mL / min using a gradient of Buffer A (2 mmol / L ammonium sulfate, 20 mmol / L Tris buffer, pH 7.0) and Buffer B (20 mmol / L Tris buffer, pH 7.0) as the mobile phase. The elution times (minutes) detected at a wavelength of 215 nm are shown in Table 5.

[0257] [Table 5]

[0258] Compared with the single amino acid mutant R218Q-Fc of full-length DcR3 produced in insect S2 cells, S2-derived S195-Fc, mammalian cell-derived chimera A-Fc (IEGRMD g1S, g4PEK), 103-123OPG-Fc (g4PEK), and the double-deglycosylated form all showed shorter elution times, suggesting that the reduced hydrophobicity improved their physical properties and reduced their aggregation. Furthermore, chimera C and the triple-deglycosylated form, which showed a slightly weaker aggregation-reducing effect, were equally or more hydrophobic than R218Q-Fc, indicating a correlation between the rate of aggregation and the hydrophobicity of the protein.

[0259] Next, the thermal stability of the proteins was evaluated by differential scanning fluorimetry (DSF). 9.5 μg of each DcR3 variant was mixed with 1 μL of SYPRO Orange Protein Gel Stain (Invitrogen) diluted 50-fold with water in a 20 μL system in a 96-well white microplate (BioRad). The temperature was increased from 20°C to 95°C in 0.5°C increments for 10 seconds using a C1000 thermal cycler (BioRad). Fluorescence at each temperature was detected in the FRET channel, and melting curves (Figure 5) and melting temperatures (Tm values) (°C) (Table 6) were calculated using CFX Manager software (BioRad).

[0260] [Table 6]

[0261] Compared with the single amino acid mutant R218Q-Fc of full-length DcR3 produced in insect S2 cells, the S2-derived S195-Fc showed a significantly higher Tm value and improved thermal stability, suggesting that the HBD region contributes to thermostability. Furthermore, the mammalian cell-derived chimera A-Fc (IEGRMD g1S) showed an even higher Tm value than S195-Fc, demonstrating that the DcR3 variant, in which part of the CRD of DcR3 was replaced with part of the CRD of OPG, not only reduced aggregation but also improved the thermal stability of the protein (Table 6, Figure 5).

[0262] [Example 5] Evaluation of the reactivity of DcR3 variants lacking the heparan sulfate binding domain (HBD) to normal human cells and CHO cells It has been reported that the pharmacokinetics of wild-type DcR3 and its mutant FLINT (R218Q mutation) are extremely poor in mice and cynomolgus monkeys (Drug Metabolism and Disposition, 2003, 31: pp. 502-507). One of the reasons for this is thought to be that the heparan sulfate-binding domain (HBD) present in wild-type DcR3 directly binds to heparan sulfate proteoglycans on the cell membrane (J. Immunol., 2006, 176: 173-180).

[0263] Therefore, we analyzed the reactivity of the HBD-deleted DcR3 variants S195-Fc, chimera A-Fc (IEGRMD g1S, g4PEK), and 103-123OPG-Fc (g4PEK) with human primary cells and CHO cells (producer cells) by flow cytometry (FCM). DcR3 FL-Fc, which contains the HBD, was used as a positive control, and K194-Fc and anti-DNP antibody (g4PEK) were used as negative controls. K194-Fc (amino acid sequence: SEQ ID NO: 152; DNA base sequence: SEQ ID NO: 151) is a protein comprising the amino acid sequence from Met1 to Lys194 of OPG (SEQ ID NO: 14) fused with an IEG RMD linker (SEQ ID NO: 106) and the amino acid sequence of Fc(g1S) (SEQ ID NO: 72). This protein was transiently expressed in Expi293 cells as described in Example 1 and purified from the culture supernatant using Mabselect SuRe (GE Healthcare). The anti-DNP antibody (g4PEK) was prepared by inserting the variable regions of the anti-2,4-dinitrophenol (DNP) antibody described in Clin. Cancer Res., 2005, 11(8), pp. 3126-3135, into a vector encoding the Fc sequence (g4PEK), expressing the antibody in CHO cells, and then purifying it from the culture supernatant using Protein A. Human primary cells used were HUVECs (Lonza) and Male Human Hepatocytes (Bioreclamation IVT). Human primary cells were cultured in collagen-coated adhesive plates (IWAKI) using the medium specified for each cell type according to the package insert. CHO cells were cultured in suspension using EX-CELL 325 PF CHO Serum-Free Medium (Sigma-Aldrich).

[0264] HUVECs and hepatocytes were detached using a 0.02% EDTA solution and a cell scraper, and then passed through a cell strainer (40 μm). The detached HUVECs, hepatocytes, and CHO cells separated from the suspension culture were washed with FCM buffer (PBS containing 1% BSA, 1 mmol / L EDTA, and 0.05% NaN3) and then suspended in FCM buffer.

[0265] Next, 1×10 5 Cells were seeded at 1000 cells / well in a 96-well U-bottom plate (Falcon), and each prepared Fc fusion protein was added at 10 μg / mL and incubated on ice for 1 hour. After washing with FCM buffer, the cells were suspended in a LIVE / DEAD Fixable Aqua Dead Cell Stain Kit (Molecular Probes) and 0.1 μg / mL Goat F(ab')2 Anti-Human IgG R-phycoerythrin Conjugate (Southern Biotech) and stained on ice for 1 hour. Human FcR Blocking Reagent (Miltenyi Biotech) was added for staining HUVECs and hepatocytes. After washing with FCM buffer, the fluorescence intensity was analyzed using a FACS Fortessa flow cytometer (BD Biosciences).

[0266] Analysis of PE staining intensity for live cell fractions negative for the LIVE / DEAD Fixable Aqua Dead Cell Stain Kit revealed that DcR3 FL-Fc(IgG1) (R&D) exhibited significant binding to HUVECs, hepatocytes, and CHO cells, whereas the HBD-deleted DcR3 mutant did not bind to any of these cells (Figure 6).

[0267] [Example 6] Evaluation of pharmacokinetics of DcR3 variants in mice

[0268] [Table 7]

[0269] Mouse kinetic studies were performed on S195-Fc and each DcR3 variant listed in Table 7. For chimera A-Fc (g4PEK) and 103-123OPG-Fc (g4PEK), each DcR3 variant was generated by stable expression in CHO cells using the method described below. DNA fragments encoding the signal peptide sequence and each DcR3 variant (SEQ ID NOs: 81 and 83) were artificially synthesized and inserted into recombinant vectors prepared by the method described in WO 2012 / 081628 in the same manner as in Example 1. The vectors were then electroporated into CHO cells. Culture and drug selection were performed using standard methods, and stable expression strains were identified when the viable cell ratio recovered to approximately 98%. These stable expression strains were cultured for a certain period in a medium such as EX-CELL 325 PF CHO Serum-Free Medium (Sigma-Aldrich). The culture supernatant was then collected, and each DcR3 variant was purified using the method described in Example 1.

[0270] S195-Fc and each DcR3 variant were administered intravenously to 5-6-week-old BALB / c mice (female) at a single dose of 10 mg / kg (n = 2 or 3). Blood samples were collected via the tail vein at random times post-administration, and the concentrations of S195-Fc and each DcR3 variant in the blood were measured as follows. Streptavidin-immobilized beads were reacted with biotinylated monkey anti-human IgG polyclonal antibodies, and the bound S195-Fc and each DcR3 variant in the serum were detected with Alexa Fluor 647-labeled monkey anti-human IgG polyclonal antibodies. Measurements were performed using a Gyrolab xP workstation (Gyros AB), and kinetic parameters were calculated by moment analysis. The same standard substance as the test substance administered to the animals was used to construct the calibration curve. Figure 7 shows the time courses of blood concentrations of S195-Fc produced in S2 cells and chimera A-Fc (IEGRMD g1S) produced in CHO-S cells. Table 8 shows the blood half-life (h) of the elimination phase after a single administration and AUC0-∞ (μg*h / mL), which is the area under the blood concentration-time curve up to infinity, for S195-Fc and each DcR3 variant shown in Table 7.

[0271] [Table 8]

[0272] Compared with the wild-type DcR3 control, S195-Fc, the blood concentration profiles of the DcR3 variants were significantly improved (Figure 7). The half-lives were similar, but the AUC0-∞ was more than 10-fold improved (Table 8).

[0273] The blood half-lives of wild-type DcR3 and FLINT (R218Q mutant) following a single intravenous dose of 0.5 mg / kg in CD-1 mice were reported to be 1.2 and 3.1 hours, respectively, and the AUC0-∞ (μg*h / mL) was 0.48 and 0.36, respectively (Drug Metabolism and Disposition, 2003.31:pp.502-507.). When comparing the concentration profiles multiplied by the dose ratio assuming a linear relationship, the DcR3 variant showed higher exposure than wild-type DcR3 and FLINT.

[0274] [Example 7] Evaluation of binding activity to DcR3 ligand (1) Preparation of DcR3 ligand Soluble recombinant forms of human, cynomolgus monkey, or mouse DcR3 ligands (LIGHT, TL1A, FasL) were produced (amino acid sequences: SEQ ID NOs: 114, 116, 118, 120, 122, 124, 126, 128, 130; DNA base sequences: SEQ ID NOs: 113, 115, 117, 119, 121, 123, 125, 127, 129).

[0275] Soluble recombinant LIGHT from human, cynomolgus monkey, or mouse had a His tag (His10) and a GS linker (GGGSGGGSGGGSIEGR) attached to the N-terminus, and the extracellular domain of LIGHT (human LIGHT: Asp74-Val240 (sequence number 132), cynomolgus monkey LIGHT: Asp74-Val240 (sequence number 134), mouse LIGHT: Asp72-Val239 (sequence number 136)) linked downstream.

[0276] Soluble recombinant human, cynomolgus monkey, or mouse TL1A had a His tag (His6) and a GS linker (GGGSGGGSGGGS) attached to the N-terminus, with the extracellular domain of TL1A (human TL1A: Leu72-Leu251 (sequence number 138), cynomolgus monkey TL1A: Leu72-Leu251 (sequence number 140), mouse TL1A: Ile94-Leu270 (sequence number 142)) linked downstream.

[0277] The soluble recombinant FasL of human, cynomolgus monkey, or mouse had a His tag (His6) attached to the N-terminus and the extracellular domain of FasL (human FasL: Pro134-Leu281 (SEQ ID NO: 144), cynomolgus monkey FasL: Pro133-Leu280 (SEQ ID NO: 146), mouse FasL: Pro132-Leu279 (SEQ ID NO: 148)) linked downstream.

[0278] A DNA fragment of the signal peptide sequence and a DNA sequence of a tagged soluble DcR3 ligand were artificially synthesized (Genewiz) and inserted downstream of the CMV promoter of the pCI-Hygro2.01 vector (a partially modified version of Promega's pCI) using the In-Fusion HD Cloning Kit (Clontech), and Escherichia coli DH5α competent cells (Toyobo) were transformed with the vector.

[0279] The resulting plasmid was introduced into Expi293 cells (Thermo Scientific) to transiently express the protein. The plasmid was introduced using ExpiFectamine 293 (Thermo Scientific), and after culturing for 3 days, the culture supernatant was collected.

[0280] Human FasL and cynomolgus monkey FasL were introduced into Freestyle CHO-S cells (Thermo Scientific) for transient protein expression. The plasmids were introduced using Freestyle MAX Reagent (Thermo Scientific), and the cells were cultured for 3 days, after which the culture supernatants were collected.

[0281] Protein purification from the culture supernatant of Expi293 cells was performed using Ni Sepharose Fast Flow Resin and His Buffer Kit (both from GE Healthcare). The culture supernatant was passed through a column packed with the resin, washed with washing buffer (60 mmol / L imidazole, 20 mmol / L sodium phosphate, 0.5 mol / L NaCl, pH 7.4), and then eluted with elution buffer (250 mmol / L imidazole, 20 mmol / L sodium phosphate, 0.5 mol / L NaCl, pH 7.4).

[0282] Protein purification from the culture supernatant of Freestyle CHO-S cells was performed using Complete His-Tag Purification Resin (Roche) and His Buffer Kit (GE Healthcare). The culture supernatant was passed through a resin-packed column, washed with washing buffer (2 mmol / L imidazole, 20 mmol / L sodium phosphate, 0.5 mol / L NaCl, pH 7.4), and then eluted with elution buffer (250 mmol / L imidazole, 20 mmol / L sodium phosphate, 0.5 mol / L NaCl, pH 7.4).

[0283] Each elution fraction was eluted with PBS using a NAP column (GE Healthcare) and sterilized by passing through a 0.22 μm filter. The purity of the resulting purified protein was confirmed by SDS-PAGE. Analysis of multimerization by SEC-UPLC (instrument: ACQUITY UPLC, column: ACQUITY UPLC Protein BEH SEC 200 Å, 1.7 μm, 4.6 × 150 mm) (Waters) revealed a peak corresponding to the molecular weight of the trimer for almost all recombinant soluble DcR3 ligands. However, mouse LIGHT did not exhibit a trimer peak and was therefore a monomer.

[0284] (2) Binding activity measurement using BIAcore The binding activity of the wild-type DcR3 control and DcR3 variants shown in Table 9 to DcR3 ligands (LIGHT, TL1A, and FasL) was analyzed by SPR using a BIAcore T-100 (GE Healthcare). HBS-EP+ Buffer was used as the buffer.

[0285] Anti-human antibodies (10,000 RU) were immobilized on a Series S Sensor Chip CM5 using a Human Antibody Capture Kit (GE Healthcare). Various wild-type DcR3 and DcR3 variants were then captured by flowing at 10 μL / min for 30 seconds. Protein-free buffer was then passed through the reference flow cell. Binding of analyte (human, cynomolgus monkey, or mouse DcR3 ligands diluted to 0.08-80 nmol / L) was monitored by flowing at 10 μL / min for 2 minutes, followed by buffer flow for 3 minutes to monitor dissociation. Next, 3 mol / L magnesium chloride was flowed at 20 μL / min for 1 minute to perform the regeneration reaction. Using BIAcore T-100 evaluation software and a 1:1 binding model, each DcR3 ligand was evaluated as a monomer (human LIGHT monomer: 20.8 kDa, human TL1A monomer: 22.1 kDa, human FasL monomer: 17.7 kDa, cynomolgus monkey LIGHT monomer: 20.8 kDa, cynomolgus monkey TL1A monomer: 22.0 kDa, cynomolgus monkey FasL monomer: 17.7 kDa, mouse LIGHT monomer: 20.9 kDa, mouse TL1A monomer: 21.5 kDa, mouse FasL monomer: 17.7 kDa). a , k d , K. D ) were calculated (Tables 10-12). Measurements confirmed that, except for mouse LIGHT, which failed to form trimers, various wild-type DcR3 and DcR3 variants bound to the respective DcR3 ligands of human, cynomolgus monkey, and mouse.

[0286] [Table 9]

[0287] [Table 10]

[0288] [Table 11]

[0289] [Table 12]

[0290] [Example 8] Evaluation of binding activity to OPG ligands (RANKL, TRAIL) RANKL binds to RANK and is known to be involved in bone resorption, while TRAIL binds to TRAIL receptors and is involved in cell death. Using the various human DcR3 recombinants prepared, the binding activity to RANKL and TRAIL was analyzed by ELISA. As positive controls for binding to OPG ligands, the C-terminal deletion of OPG K194-Fc prepared in Example 5, RANK-Fc (Enzo Life Sciences), and TRAIL R1-Fc (R&D Systems) were used. As a negative control, an anti-DNP antibody (IgG1) [the variable region of the anti-2,4-dinitrophenol (DNP) antibody described in Clin. Cancer Res., 2005, 11(8), pp. 3126-3135 was inserted into a vector encoding the Fc of IgG1, expressed in CHO cells, and purified with Protein A] was used.

[0291] Anti-human IgG1 (American Qualex) diluted to 10 μg / mL in PBS was dispensed at 50 μL / well into a 96-well plate (MAXISORP NUNC-IMMUNO PLATE, Thermo Scientific) and allowed to stand overnight at 4°C for adsorption. After removing the immobilization solution, 100 μL / well of 1% Block Ace, prepared by dissolving 1 g of Block Ace powder (DS Pharma Biomedical) in 100 mL of water, was dispensed. The plate was allowed to stand at room temperature for 1 hour for blocking, and then washed three times with PBS containing 0.1% Tween (hereafter referred to as PBST). Next, 50 μL / well of various human DcR3 recombinants or various DcR3 variants diluted to 1 μg / mL with 1% BSA-PBS and RANK-Fc were dispensed onto plate No. 1, and 50 μL / well of various wild-type DcR3 or various DcR3 variants diluted to 1 μg / mL with 1% BSA-PBS and TRAIL R1-Fc were dispensed onto plate No. 2, and the plates were left to stand at room temperature for 1 hour.

[0292] After washing each plate three times with PBST, 50 μL / well of RANKL (Peprotech) diluted to 0.64-50,000 pg / mL in 1% BSA-PBS was dispensed into plate No. 1, and 50 μL / well of TRAIL (Peprotech) diluted to 0.64-50,000 pg / mL in 1% BSA-PBS was dispensed into plate No. 2, and the plates were left to stand at room temperature for 1 hour.

[0293] After washing the plates three times with PBST, 50 μL / well of biotinylated anti-RANKL antibody (Peprotech) diluted to 0.4 μg / mL in 1% BSA-PBS was dispensed into plate No. 1, and 50 μL / well of biotinylated anti-TRAIL antibody (R&D) diluted to 0.4 μg / mL in 1% BSA-PBS was dispensed into plate No. 2, and the plates were left to stand at room temperature for 1 hour.

[0294] After each plate was washed three times with PBST, streptavidin-HRP (PIERCE) diluted 10,000 times with 0.1% Block Ace was dispensed at 50 μL / well and allowed to stand at room temperature for 1 hour.

[0295] After washing each plate three times with PBST, 50 μL of TMB+Substrate Chromogen (Dako) was dispensed into each well and allowed to stand at room temperature for 1 minute. 50 μL of 0.5 mol / L sulfuric acid solution was dispensed into each well to stop the color reaction. The absorbance was measured using a plate reader at a sample wavelength of 450 nm and a reference wavelength of 570 nm.

[0296] As a result, K194-Fc bound to the OPG ligands RANKL and TRAIL, and RANK-Fc and TRAIL R1-Fc bound to RANKL and TRAIL, respectively, whereas DcR3 FL-Fc (R&D), S195-Fc, chimera A-Fc (IEGRMD g1S), and chimera B-Fc (IEGRMD g1S) did not bind to any of the ligands (Figure 8). Together with the results of Example 7, it was confirmed that the DcR3 variants prepared by substituting a portion of the CRD of DcR3 with a portion of the CRD of OPG exhibit binding activity to DcR3 ligands equivalent to that of wild-type DcR3 but do not bind to OPG ligands.

[0297] [Example 9] Measurement of DcR3 ligand neutralizing activity The neutralizing activities of various wild-type DcR3 control and DcR3 mutants against LIGHT, TL1A, and FasL were measured by the methods described below. DNP antibodies were prepared by the methods described in Examples 5 and 8.

[0298] (1) LIGHT neutralization activity measurement The neutralizing activity of wild-type DcR3 and various DcR3 variants was measured using the human colon cancer cell line HT-29 (ATCC number: HTB-38) using LIGHT-induced IL-8 production as an indicator. Cell culture and neutralizing activity evaluation were performed in McCoy's 5A medium (Gibco) supplemented with 10% FBS (Gibco) and penicillin / streptomycin (Nacalai Tesque).

[0299] HT-29 cells were cultured at 2 x 10 in a 96-well adherent culture plate (Sumitomo Bakelite Co., Ltd.). 4After seeding at 1000 sb / well, various wild-type DcR3 or various DcR3 variants were added to a final concentration of 0.1, 1, or 10 μg / mL. Human LIGHT (Gly66-Val240) (described in JP2013153749) tagged with a FLAG tag (DYKDDDDK) at the N-terminus was then added to a final concentration of 0.1 μg / mL, and the total volume of the culture medium was adjusted to 200 μL / well. After culturing for 3 days at 37°C in a 5% CO2 incubator, the culture supernatant was collected, and the IL-8 concentration in the culture supernatant was measured using the AlphaLISA IL-8 Immunoassay Research Kit (Perkin Elmer).

[0300] The measurement results confirmed that various human DcR3 variants reduced the amount of IL-8 produced in a concentration-dependent manner, and exhibited LIGHT-neutralizing activity (FIG. 9).

[0301] (2) TL1A neutralizing activity measurement The neutralizing activity of wild-type DcR3 and various DcR3 variants was measured using human T cells with TL1A as an indicator of IFN-γ production. X-VIVO15 medium (Lonza) was used for cell culture and neutralizing activity evaluation.

[0302] Frozen PBMCs from healthy donors (AllCells) were thawed in a 37°C water bath and suspended in medium containing DNase I (STEMCELL) preheated to 37°C. After shaking at low speed for 2 hours at 37°C, T cells were isolated using the EasySep Human T cell Enrichment Kit (STEMCELL). The isolated T cells were cultured in a 96-well suspension culture plate (Sumitomo Bakelite) at a concentration of 1 x 10 5After seeding at 1000 cells / well, various wild-type DcR3 controls or various DcR3 variants were added to final concentrations of 0.1, 1, or 10 μg / mL. Recombinant His10 human TL1A prepared in Example 6 was then added to a final concentration of 0.1 μg / mL. Human IL-12 (Miltenyi Biotech) was added to a final concentration of 2 ng / mL, and recombinant human IL-18 (MBL) was added to a final concentration of 50 ng / mL, and the total culture volume was adjusted to 200 μL / well. After 3 days of culture at 37°C in a 5% CO2 incubator, the culture supernatant was collected, and the IFN-γ concentration in the culture supernatant was measured using an AlphaLISA IFN-γ Immunoassay Research Kit (Perkin Elmer).

[0303] The measurement results confirmed that various DcR3 variants reduced the amount of IFN-γ produced in a concentration-dependent manner, and exhibited TL1A neutralizing activity (FIG. 10).

[0304] (3) FasL neutralization activity measurement The neutralizing activity of wild-type DcR3 control and various DcR3 variants was measured using the Jurkat T-cell leukemia cell line (DSMZ number: ACC 282) using FasL-induced apoptosis as an indicator. Cell culture and neutralizing activity evaluation were performed using RPMI 1640 medium (Nacalai Tesque) supplemented with 10% FBS and penicillin / streptomycin.

[0305] Jurkat cells were cultured in a 96-well suspension culture plate at 5x10 4After seeding at 1000 cells / well, various wild-type DcR3 or DcR3 variants were added to final concentrations of 0.01, 0.1, or 1 μg / mL. Recombinant human His6 Fas Ligand (CST Japan) was then added to a final concentration of 0.1 μg / mL, and the culture medium was adjusted to a total volume of 100 μL / well. After overnight incubation at 37°C in a 5% CO2 incubator, 100 μL / well of CellTiter-Glo (Promega) was added to the Jurkat culture plate, and the number of viable cells was measured using a luminometer (Veritas, Promega) to measure ATP production.

[0306] As a result of the measurement, it was confirmed that various DcR3 variants increased the amount of ATP produced from living cells in a concentration-dependent manner, and exhibited FasL-neutralizing activity (FIG. 11).

[0307] [Example 10] Ligand neutralizing activity and binding activity of mutants with reduced FasL binding Using chimera A-Fc(g4PEK), we prepared and evaluated single-amino acid substitution variants that retained binding and neutralizing activity against the DcR3 ligands TL1A and LIGHT but reduced binding and neutralizing activity against FasL. Variants were prepared by substituting any amino acid in CRD2 or CRD3 shown in Table 13 with Ala or an amino acid other than Ala, and their neutralizing activity against DcR3 ligands was measured using the same method as in Example 9. As a result, the single amino acid substitutions of chimera A-E57K-Fc(g4PEK) (amino acid sequence: SEQ ID NO: 94, DNA base sequence: SEQ ID NO: 93), chimera A-E57L-Fc(g4PEK) (amino acid sequence: SEQ ID NO: 96, DNA base sequence: SEQ ID NO: 95), in which Glu at position 57 from the N-terminus of chimera A-Fc(g4PEK) was replaced with Lys or Leu, and chimera A-R60K-Fc(g4PEK) (amino acid sequence: SEQ ID NO: 98, DNA base sequence: SEQ ID NO: 97), in which Arg at position 60 was replaced with Lys, showed reduced FasL-selective neutralizing activity (Table 13).

[0308] The DcR3 ligand-binding activities of the chimera A-E57K-Fc(g4PEK) (SEQ ID NO: 94), chimera A-E57L-Fc(g4PEK) (SEQ ID NO: 96), and chimera A-R60K-Fc(g4PEK) (SEQ ID NO: 98), which have reduced FasL-selective neutralizing activity, were measured using the same method as in Example 7. The results showed that the binding of only FasL was significantly reduced for the chimera A-E57K-Fc(g4PEK), chimera A-E57L-Fc(g4PEK), and chimera A-R60K-Fc(g4PEK) variants, while the binding of TL1A and LIGHT was maintained (Fig. 12A, B, C).

[0309] [Table 13]

[0310] [Example 11] Evaluation of the aggregation properties of commercially available wild-type DcR3 in mammalian cells Commercially available full-length DcR3-Fc (Abcam) produced using HEK293 cells as host cells, commercially available full-length DcR3-Fc (AdipoGen) produced using CHO cells as host cells, and a commercially available Fc fusion of DcR3 molecule (Enzo) lacking approximately half of the C-terminus of the HBD produced using HEK293 cells as host cells were electrophoresed under reducing and non-reducing conditions in the same manner as in Example 1. The molecular weight of the monomer predicted from the electrophoretic mobility under reducing conditions was approximately 50 to 60 kDa, but the mobility under non-reducing conditions was significantly higher than the predicted dimer molecular weight, and most of the commercially available products existed as aggregates (Figure 13).

[0311] [Example 12] Preparation of mutants with reduced FasL binding To select a DcR3 ligand variant that binds to TL1A and LIGHT but has reduced binding affinity to FasL, we created a single amino acid substitution variant (E57X; X is any amino acid other than Glu) in which the 57th Glu residue from the N-terminus of chimera A-Fc(g4PEK) (sequence number 82) (E57) was replaced with an amino acid other than Glu. Furthermore, for E57K, E57L, E57R, and E57V, in which E57 was substituted with Lys, Leu, Arg, or Val, two-amino acid substitutions (E57X_W53Z, E57X_N54Z, E57X_Y55Z, E57X_L56Z, and E57X_R58Z; each two-amino acid substitution was assigned a variant number shown in Figure 14A) were used to prepare fusions of Fc(g4PEK) with the adjacent amino acid, Trp (W53), Asn (N54), Tyr (Y55), Leu (L56), and Arg (R58), which are located at the 53rd, 54th, and 58th positions from the N-terminus of chimera A-Fc(g4PEK) (SEQ ID NO: 82), respectively, by a specific amino acid Z (Z is Asp, Glu, Asn, Gln, Pro, Thr, or Gly).

[0312] Of these, chimera A-E57K (amino acid sequence: SEQ ID NO: 66, DNA base sequence: SEQ ID NO: 65), chimera A-E57R (amino acid sequence: SEQ ID NO: 180, DNA base sequence: SEQ ID NO: 179), chimera A-E57V (amino acid sequence: SEQ ID NO: 182, DNA base sequence: SEQ ID NO: 181), chimera A-E57K_R58D (variant number: 45-10, amino acid sequence: SEQ ID NO: 184, DNA base sequence: SEQ ID NO: 183), chimera A-E57K_R58E (variant number: 45-18, amino acid sequence: SEQ ID NO: 186, DNA base sequence: SEQ ID NO: 185), and chimera A were subjected to various mutations shown in Table 14. Fusions with a portion of the Fc sequence of the recombinant human Fc were prepared (nucleotide sequences: SEQ ID NOs: 213, 217, 219, 221, 223, 227, 231, 233, 235, 237, 255, 259, 261, 263, 265, 149, 167, 169, 171, 173, 175, 177; amino acid sequences: SEQ ID NOs: 214, 218, 220). , 222, 224, 228, 232, 234, 236, 238, 256, 260, 262, 264, 266, 317, 319, 320, 321, 322, 324, 326, 327, 328, 329, 331, 333, 334, 335, 336, 150, 168, 170, 172, 174, 176, 178). In Table 14, E216 represents Glu at position 216 of the human IgG1 heavy chain as indicated by the EU index. The mutation sites C220S, M252Y, S254T, T256E, N434A, L234A, L235A, and G237A indicate substitutions of Cys at position 220 with Ser, Met at position 252 with Tyr, Ser at position 254 with Thr, Thr at position 256 with Glu, Asn at position 434 with Ala, Leu at position 234 with Ala, Leu at position 235 with Ala, and Gly at position 237 with Ala, respectively, in the human IgG1 heavy chain shown by the EU index.

[0313] [Table 14]

[0314] Chimera A-Fc(g1S) (nucleotide sequence: SEQ ID NO: 149, amino acid sequence: SEQ ID NO: 150) was prepared by deleting the IEGRMD sequence from chimera A-Fc(IEGRMD g1S) (SEQ ID NO: 80) containing the linker sequence IEGRMD (SEQ ID NO: 106) prepared in Example 2, and stably expressing it in CHO cells by the method described in Example 6. Chimera A-Fc(Eg1S) (nucleotide sequence: SEQ ID NO: 167, amino acid sequence: SEQ ID NO: 168) was prepared by stably expressing a sequence in which chimera A (SEQ ID NO: 54) and Eg1S were linked together in CHO cells by the method described in Example 6.

[0315] Chimera A-Fc (Eg1S YTE, Eg1S N434A, Eg1S LALAGA, Eg1S LALAGANA) (nucleotide sequence: SEQ ID NOs: 169, 171, 175, 177; amino acid sequence: SEQ ID NOs: 170, 172, 176, 178) was obtained by transient expression in CHO-S cells of sequences in which each Fc was fused to the C-terminus of Chimera A (SEQ ID NO: 54).

[0316] Chimera A-Fc (g1S YTE, g1S N434A, g1S LALAGA, g1S LALAGANA) (amino acid sequences: SEQ ID NOs: 314, 315, 174, 316) was obtained by transiently expressing each Fc at the C-terminus of Chimera A (SEQ ID NO: 54) in Expi293 cells.

[0317] Plasmids expressing fusions of each amino acid substitution with Fc(g4PEK) were prepared by PCR amplification of two regions, one from the NheI site to the site where the amino acid substitution was introduced and the other from the site where the amino acid substitution was introduced to the SalI site, using either the DNA sequence of chimera A-Fc(g4PEK) or E57X-Fc (g4PEK; X is K, R, or V) as a template and PCR primers designed to contain the mutation sites. The regions were then inserted under the CMV promoter of the pCIpuro vector in the same manner as in Example 1.

[0318] Fusions of each amino acid substitution with the Fc of Eg1S YTE, Eg1S N434A, or Eg1S LALAGANA were prepared by PCR amplification of two regions, one from the EcoRI site to the site where the amino acid substitution was introduced, and the other from the site where the amino acid substitution was introduced to the Bsu36I site, using the DNA sequence of one of the chimera A-Fc sequences containing the respective mutant Fc sequences as a template and PCR primers designed to contain the mutation sites. The amplified regions were then inserted into the EcoRI and Bsu36I sites of the chimera A-various mutant Fc vector used as a template to create plasmids.

[0319] Fusions of each amino acid substitution with the Fc of g1S YTE, g1S N434A, or g1S LALAGANA were prepared by PCR amplification of two regions, one from the EcoRI site to the Glu site to be deleted and the other from the Glu site to the Bsu36I site, using the DNA sequence of one of the above-mentioned amino acid substitution-Fc vectors as a template and PCR primers designed to remove E216. Plasmids were then prepared by inserting the amplified regions into the EcoRI and Bsu36I sites of each amino acid substitution-Fc vector used as a template.

[0320] Each plasmid was introduced into Expi293 cells for transient expression, and the culture supernatant was affinity purified using MabSelect SuRe. The content of monomer, aggregate, and degradation product of the prepared chimera A-Fc and each of the single- and double-amino acid substituted-Fc was calculated from the peak area analyzed by SEC-UPLC (ACQUITY UPLC Protein BEH SEC 4.6 mm × 150 mm) (Waters) or SEC-HPLC (TSKgel SuperSW3000 4.0 μm, 4.6 mm × 300 mm) (Tosoh) in the same manner as in Example 2.

[0321] As a result, many of the chimera A-Fc containing various mutant Fc fragments and the one- and two-amino acid substitution-Fc fragments maintained a lower aggregate content than S195-Fc (Fig. 14A, B, C).

[0322] [Example 13] Evaluation of binding activity to DcR3 soluble trimeric ligand The binding activities of DcR3-Fc, S195-Fc, various chimera A-Fc with different Fc sequences, and various mutants with reduced FasL binding activity prepared in Example 12 to soluble human LIGHT trimer, soluble human TL1A trimer, and soluble human FasL trimer were evaluated using the method described in Example 7 with some modifications.

[0323] (1) Preparation of a soluble trimeric DcR3 ligand For human soluble recombinant LIGHT, a FLAG tag (DYKDDDDK) was added to the N-terminus, and the extracellular domain of LIGHT (Asp74-Val240) (SEQ ID NO: 132) was linked downstream (FLAG-LIGHT) (nucleotide sequence: SEQ ID NO: 305, amino acid sequence: SEQ ID NO: 306). For human soluble recombinant TL1A, a His tag (His6) and a GS linker (GGGSGGGSGGGS) were added to the N-terminus, and the extracellular domain of TL1A (Leu72-Leu251) (SEQ ID NO: 138) was linked downstream (His6-TL1A) (nucleotide sequence: SEQ ID NO: 115, amino acid sequence: SEQ ID NO: 116). Each plasmid was prepared in the same manner as in Example 7, and transient expression was performed in Expi293 cells.

[0324] FLAG-LIGHT was purified using ANTI-FLAG M2 Affinity Gel (Sigma). The culture supernatant was passed through a resin-packed column, washed with washing buffer (50 mM Tris HCl, 150 mM NaCl, pH 7.4), and then eluted with elution buffer (0.1 M glycine hydrochloride, pH 3.5).

[0325] His6-TL1A was purified as follows. The culture supernatant was applied to a column packed with Complete His-Tag Purification Resin (Roche), washed with washing buffer (50 mM NaH2PO4 pH 8.0, 300 mM NaCl), and then eluted with elution buffer (50 mM NaH2PO4 pH 8.0, 300 mM NaCl, 250 mM imidazole). The eluate was replaced with PBS using a NAP column (GE Healthcare). The column was then applied to a column packed with Ni Sepharose Fast Flow Resin (GE Healthcare). The column was washed with washing buffer (60 mM imidazole, 20 mM sodium phosphate, 0.5 M NaCl, pH 7.4) prepared using a His Buffer Kit (GE Healthcare), and then eluted with elution buffer (250 mM imidazole, 20 mM sodium phosphate, 0.5 M NaCl, pH 7.4).

[0326] The eluted fractions of FLAG-LIGHT and His6-TL1A were eluted with PBS using a NAP column (GE Healthcare) and sterilized through a 0.22 μm filter. The purified proteins were subjected to gel filtration chromatography (SEC) (TSKgel G3000 SWXL 7.8 mm x 300 mm) (Tosoh Corporation) using HPLC (Shimadzu Corporation), and the trimer fraction was isolated.

[0327] The soluble human recombinant FasL used was Human His6 Fas Ligand / TNFSF6 (Cell Signaling Technology), which had a His tag (His6) attached to the N-terminus and the extracellular domain of FasL (Pro134-Leu281) (SEQ ID NO: 144) linked downstream. SEC-MALS was performed as in Example 2, and it was confirmed to be a trimer.

[0328] (2) Binding activity measurement using BIAcore DcR3-Fc (R&D) and S195-Fc were evaluated as described in Example 7. Chimera A-g4PEK (SEQ ID NO: 82) was prepared in Example 6. Chimera A-Fc (g1S, Eg1S) was prepared in Example 12. As various variants with reduced FasL binding, the Fc(g4PEK) fusion products prepared in Example 9 or Example 12 were evaluated.

[0329] The binding activity of the human DcR3 trimeric ligand was analyzed by SPR. The binding activity to human LIGHT and human TL1A was measured using a BIAcore T-100 (GE Healthcare), and the binding activity to human FasL was measured using a BIAcore T-100 (GE Healthcare) or a BIAcore T-200 (GE Healthcare). HBS-EP+ Buffer was used as the buffer.

[0330] Anti-human antibodies (10,000 RU) were immobilized on a Series S Sensor Chip CM5 using a Human Antibody Capture Kit (both from GE Healthcare). Wild-type DcR3 and DcR3 variants were then captured by flowing at 10 μL / min for 30 seconds. Protein-free buffer was then passed through the reference flow cell. Binding of each human DcR3 trimeric ligand diluted to 0.02-80 nmol / L was monitored by flowing at 30 μL / min for 2 minutes, followed by dissociation by flowing buffer for 3 minutes. Next, 3 mol / L magnesium chloride was flowed at 30 μL / min for 1 minute to perform the regeneration reaction. The binding activity to human LIGHT and human TL1A was measured using BIAcore T-100 evaluation software with a 1:1 binding model. The ligands were treated as trimers (human LIGHT trimer: 62.4 kDa, human TL1A trimer: 66.2 kDa). The kinetic constants (k a , k d , K. DThe binding activity to human FasL was measured using BIAcore T-100 evaluation software with a 1:1 binding model or BIAcore T-200 evaluation software with a 1:1 binding model. The ligand was a monomer (human FasL monomer: 19.8 kDa), and each kinetic constant (k a , k d , K. D ) was calculated.

[0331] As a result, it was confirmed that chimera A-Fc(g1S), chimera A-Fc(Eg1S), and chimera A-Fc(g4PEK) all bind to each DcR3 trimeric ligand (Fig. 15A).

[0332] Among the variants (g4PEK) with reduced FasL binding activity prepared in Example 12, the following variants were prepared: chimera A-E57K-Fc (g4PEK) (amino acid sequence: SEQ ID NO: 94, DNA base sequence: SEQ ID NO: 93) (hereinafter, sometimes referred to as "E57K-Fc"), chimera A-E57L-Fc (g4PEK) (amino acid sequence: SEQ ID NO: 96, DNA base sequence: SEQ ID NO: 95) (hereinafter, sometimes referred to as "E57L-Fc"), and chimera A-E57R-Fc (g4PEK) (amino acid sequence: SEQ ID NO: 190, DNA base sequence: SEQ ID NO: 189) (hereinafter, sometimes referred to as "E57R-Fc"). -Fc"), chimera A-E57V-Fc(g4PEK) (amino acid sequence: SEQ ID NO: 192, DNA base sequence: SEQ ID NO: 191) (hereinafter sometimes referred to as "E57V-Fc"), chimera A-E57A-Fc(g4PEK) (amino acid sequence: SEQ ID NO: 288, DNA base sequence: SEQ ID NO: 287) (hereinafter sometimes referred to as "E57A-Fc"), chimera A-E57F-Fc(g4PEK) (amino acid sequence: SEQ ID NO: 290, DNA base sequence: SEQ ID NO: 289) (hereinafter sometimes referred to as "E57F-Fc"), chimera A-E57H-Fc (g4PEK) (amino acid sequence: SEQ ID NO: 292, DNA base sequence: SEQ ID NO: 291) (hereinafter, sometimes referred to as "E57H-Fc"), chimera A-E57I-Fc(g4PEK) (amino acid sequence: SEQ ID NO: 294, DNA base sequence: SEQ ID NO: 293) (hereinafter, sometimes referred to as "E57I-Fc"), chimera A-E57M-Fc(g4PEK) (amino acid sequence: SEQ ID NO: 296, DNA base sequence: SEQ ID NO: 295) (hereinafter, sometimes referred to as "E57M-Fc"), and two amino acid substituted chimera A-E57K_R58D-Fc(g4PEK) (modified No.: 45-10, amino acid sequence: SEQ ID NO: 194, DNA base sequence: SEQ ID NO: 193) (hereinafter may be referred to as "45-10-Fc"), chimera A-E57K_R58T-Fc(g4PEK) (variant No.: 45-11, amino acid sequence: SEQ ID NO: 298, DNA base sequence: SEQ ID NO: 297) (hereinafter may be referred to as "45-11-Fc"), chimera A-E57K_R58E-Fc(g4PEK) (variant No.: 45-18, amino acid sequence: SEQ ID NO: 196, DNA base sequence: SEQ ID NO: 195) (hereinafter may be referred to as "45-18-Fc"),Chimera A-E57L_R58E-Fc(g4PEK) (variant number: 46-4, amino acid sequence: SEQ ID NO: 300, DNA base sequence: SEQ ID NO: 299) (hereinafter, sometimes referred to as "46-4-Fc"), chimera A-E57R_R58D-Fc(g4PEK) (variant number: 82-5, amino acid sequence: SEQ ID NO: 198, DNA base sequence: SEQ ID NO: 197) (hereinafter, sometimes referred to as "82-5-Fc"), chimera A-E57V_R58T-Fc(g4PEK) (variant number: 85-6, amino acid sequence: SEQ ID NO: 302, DNA base sequence: SEQ ID NO: 301) (hereinafter sometimes referred to as "85-6-Fc"), chimera A-E57V_R58E-Fc(g4PEK) (variant number: 85-8, amino acid sequence: SEQ ID NO: 304, DNA base sequence: SEQ ID NO: 303) (hereinafter sometimes referred to as "85-8-Fc"), both had lower K for LIGHT trimer and TL1A trimer than chimera A-Fc(g4PEK), D The value is less than 3-fold and the K D The values were reduced to more than three times or the Rmax value was reduced to less than 5, confirming that binding to FasL was selectively reduced (Fig. 16A, B).

[0333] Of these, E57K-Fc, E57L-Fc, E57R-Fc, E57V-Fc, 45-10-Fc, 45-18-Fc, and 82-5-Fc (all with g4PEK Fc) were purified to 95% or more monomers by SEC-HPLC (column: TSKgel G3000 SWXL 7.8 mm x 300 mm, Tosoh Corporation; HPLC: Shimadzu Corporation), and the kinetic constants of each purified product measured by BIAcore are shown in Figure 15B.

[0334] [Example 14] Evaluation of neutralizing activity of DcR3 soluble ligands The neutralizing activities of various DcR3 variants against soluble human LIGHT, soluble human TL1A, and soluble human FasL were evaluated using the method described in Example 9 with some modifications.

[0335] (1) Construction of various chimeric A-Fc antibodies with different Fc sequences The chimera A-Fc (g1S, Eg1S) used was that prepared in Example 12. Each chimera A-Fc prepared in Example 12, in which the Fc of Eg1S YTE, Eg1S N434A, Eg1S LALAGA, or Eg1S LALAGANA shown in Table 14 was fused to the C-terminus of chimera A, was produced by transient expression in CHO-S cells and purified to 95% or more of the monomer by SEC (Superdex 200 Increase 10 / 300 GL) (GE Healthcare) using an AKTA purifier (GE Healthcare).

[0336] (2) Soluble human LIGHT neutralizing activity Chimeric A-Fc (g1S, Eg1S, Eg1S YTE, Eg1S N434A, Eg1S LALAGA, Eg1S LALAGANA, g4PEK) was evaluated for its inhibitory activity against LIGHT-dependent CXCL10 production from IFN-γ-stimulated intestinal myofibroblasts (Lonza). Intestinal myofibroblasts were cultured in collagen I-coated flasks (BD) using SmGM-2 Bullet Kit (Lonza) medium. 1x10 cells were plated on a collagen I-coated 96-well plate (BD). 4 After seeding cells at 1000 cells / well, IFN-γ (final concentration: 10 ng / mL), trimeric FLAG-LIGHT (prepared in Example 13) (final concentration: 20 ng / mL), and various DcR3 variants (final concentrations: 19.5, 78.1, 313, 1250, 5000, 20000 ng / mL, or 4.88, 19.5, 78.1, 313, 1250, 5000 ng / mL) were added to the cells and cultured for 3 days. Culture supernatants were collected, and CXCL10 concentrations in the culture supernatants were measured using a CXCL10 / IP-10 (human) AlphaLisa Detection Kit (Perkin Elmer). Chimera A-Fc containing all Fc sequences inhibited CXCL10 production in a concentration-dependent manner and neutralized soluble LIGHT (Figures 17A and 17B).

[0337] The FasL-binding-reducing variants E57K-Fc, E57L-Fc, E57R-Fc, and E57V-Fc (all of which have g4PEK Fc) were evaluated for their inhibitory activity against LIGHT-dependent IL-8 production from HT-29 cells by the same method as in Example 9. As a result, it was confirmed that all variants inhibited IL-8 production in a concentration-dependent manner and had neutralizing activity against soluble LIGHT ( Figure 17C ).

[0338] The FasL-binding-reducing variants 45-10-Fc, 45-18-Fc, and 82-5-Fc (all with g4PEK Fc) were tested for their inhibitory activity against LIGHT-dependent CXCL10 production from IFN-γ-stimulated intestinal myofibroblasts in a well containing 2 x 10 intestinal myofibroblasts, similar to chimera A-Fc. 4 The test compounds were evaluated at final concentrations of 19.5, 78.1, 313, 1250, 5000, and 20000 ng / mL. As a result, it was confirmed that all of the variants inhibited CXCL10 production in a concentration-dependent manner and had neutralizing activity against soluble LIGHT (Figure 17D).

[0339] (3) Soluble human TL1A neutralizing activity The inhibitory activity of various DcR3 variants against TL1A-dependent IFN-γ production from IL-12- and IL-18-stimulated human T cells was evaluated using the same method as in Example 9. Chimera A-Fc containing any of the Fc sequences inhibited IFN-γ production in a concentration-dependent manner, demonstrating its neutralizing activity against soluble TL1A (Fig. 18A and B). Furthermore, the FasL-binding-reduced variants E57K-Fc, E57L-Fc, E57R-Fc, E57V-Fc, 45-10-Fc, 45-18-Fc, and 82-5-Fc (all of which have g4PEK Fc) also showed neutralizing activity against soluble TL1A (Fig. 18C and D).

[0340] (4) Soluble human FasL neutralizing activity The inhibitory activity of various DcR3 variants against FasL-dependent cell death in Jurkat cells or Jurkat subclone A3 cells was evaluated using the same method as in Example 9. Chimeric A-Fc containing any of the Fc sequences inhibited A3 cell death in a concentration-dependent manner, demonstrating neutralizing activity against soluble FasL (Fig. 19A and B). Meanwhile, the FasL-binding-reduced variants E57K-Fc, E57L-Fc, E57R-Fc, E57V-Fc, 45-10-Fc, 45-18-Fc, and 82-5-Fc (all of which have g4PEK Fc) significantly reduced their inhibitory activity against Jurkat cell death, demonstrating selective reduction in neutralizing activity against soluble FasL (Fig. 19C and D).

[0341] [Example 15] Evaluation of binding activity to DcR3 membrane-type ligands The binding activity of various DcR3 variants to membrane-type human LIGHT, membrane-type human TL1A, and membrane-type human FasL was assessed by flow cytometry using strains overexpressing the membrane-type ligands. HEK293 cells overexpressing membrane-type human LIGHT were used as described in US8974787. Membrane-type TL1A and membrane-type FasL were amplified by PCR using commercially available ORF clones (Origene) as templates, with a Met residue and a FLAG tag (DYKDDDDK) added to the N-terminus (nucleotide sequence: SEQ ID NOs: 307 and 309; amino acid sequence: SEQ ID NOs: 308 and 310). These were inserted downstream of the CMV promoter in the pCIpuro vector using the In-Fusion HD Cloning Kit (Clontech), and used to transform E. coli DH5α competent cells (Toyobo).

[0342] The resulting plasmids were transfected into CHO-K1 cells (ECACC) using Nucleofector and Cell Line Nucleofector Kit T (both from Lonza) and subjected to drug selection with 10 μg / mL Puromycin (Thermo Fisher Scientific). The resulting drug-resistant cells were stained with DyLight488-conjugated anti-human TL1A antibody (Novus Biologics) or APC-conjugated anti-human FasL antibody (BD Pharmingen), and high-expressing fractions were sorted using a cell sorter (Sony). After expansion, the cells were stained with PE-conjugated anti-human TL1A antibody (Novus Biologics) or PE-conjugated anti-human FasL antibody (BioLegend), and then sorted again to obtain cell lines highly expressing membrane-bound human TL1A or membrane-bound human FasL.

[0343] The binding activity of various DcR3 variants to membrane-type ligand-expressing strains and host cells was evaluated using the same method as in Example 5, with the following modifications: HEK293 and membrane-type LIGHT-expressing strains were reacted with 1 μg / mL of each protein and 10 ng / mL of the secondary antibody Goat F(ab')2 Anti-Human IgG R-phycoerythrin Conjugate (Southern Biotech). Membrane-type TL1A and membrane-type FasL-expressing strains were reacted with 1 or 10 μg / mL of each protein and 0.1 or 1 μg / mL of the secondary antibody.

[0344] As a result, chimera A-Fc (g1S, Eg1S, g4PEK) did not react with the host cells, 293 or CHO-K1 cells, but reacted specifically with the cells expressing each membrane-bound ligand. Therefore, it was confirmed that each chimera A with a different Fc sequence had binding activity to the membrane-bound DcR3 ligand (Fig. 20A, B). Furthermore, the binding activity of the FasL-binding-reduced variants E57K-Fc, E57L-Fc, E57R-Fc, E57V-Fc, 45-10-Fc, 45-18-Fc, and 82-5-Fc (all with g4PEK Fc) was similarly assessed. It was confirmed that they retained binding activity to membrane-bound human LIGHT and membrane-bound human TL1A, whereas their binding activity to membrane-bound human FasL was significantly reduced, except for E57L (Fig. 21A, B, C).

[0345] [Example 16] Evaluation of binding activity to DcR3 primary ligand The binding activity of various DcR3 variants to each DcR3 ligand derived from primary cells was evaluated by the following method.

[0346] (1) Primary LIGHT binding activity It is known that expression of membrane-type LIGHT is induced in activated human T cells (The Journal of Immunology, 2004, 173: pp. 502-507). Frozen PBMCs from healthy donors (AllCells) were thawed and stimulated overnight with PMA (Sigma) at a final concentration of 50 ng / mL and ionomycin (Sigma) at 1 μg / mL. The binding of chimeric A-Fc(Eg1S) to membrane-type LIGHT, whose expression was induced in CD3+ T cells, was evaluated as follows. The anti-DNP antibody (IgG1) described in Example 8 was used as a negative control. Stimulated PBMCs were collected and incubated with Human FcR Blocking Reagent (Miltenyi Biotech), followed by the addition of BV421-labeled CD3 antibody (BD Pharmingen), chimeric A-Fc antibody (7-AAD Staining Solution (BD Pharmingen)), or anti-DNP antibody (final concentration 0.4 μg / mL) labeled with Alexa Fluor 488 Antibody Labeling Kit (Thermo Scientific). After incubation, the cells were washed, and the Alexa Fluor 488 fluorescence intensity in CD3+ T cells in the live cell fraction was analyzed using a flow cytometer. To confirm the expression of membrane-type LIGHT in stimulated PBMCs, stimulated PBMCs were reacted with Human FcR Blocking Reagent (Miltenyi Biotech), followed by the addition of BV510-labeled CD3 antibody (BioLegend), 7-AAD Staining Solution (BD Pharmingen), and PE-labeled LIGHT antibody (LSBio) or PE-labeled mouse IgG1κ isotype control antibody (BioLegend). After the reaction, the cells were washed, and the PE fluorescence intensity in CD3+ T cells in the live cell fraction was analyzed using a flow cytometer.

[0347] As a result, it was confirmed that expression of LIGHT was induced on CD3-positive human T cells in stimulated PBMCs, and that Alexa Fluor 488-labeled chimeric A-Fc bound to membrane-type LIGHT on activated CD3-positive human T cells (Figures 22A and 22B).

[0348] (2) Primary TL1A binding activity HUVEC cells (Lonza) cultured as described in Example 5 were treated with recombinant human IL-1 alpha (R&D) at a final concentration of 10 ng / mL and the TACE inhibitor TAPI-1 (Calbiochem) at a final concentration of 20 μM for 24 hours. The binding of chimeric A-Fc (IEGRMD g1S) to the membrane-type TL1A induced by expression was evaluated in a competition experiment with the TL1A antibody 1D1 1.31 (US2015 / 0132311). The TL1A antibody 1D1 1.31 was constructed by linking the amino acid sequences of the VL and VH described in US2015 / 0132311 to the constant region of human IgG1. The antibody was transiently expressed in Expi293 cells as described in Example 1, and purified from the culture supernatant using Mabselect SuRe (GE Healthcare). The anti-DNP antibody (IgG1) described in Example 8 was used as a negative control. The collected cells were reacted with Human FcR Blocking Reagent (Miltenyi Biotech), and then either anti-DNP antibody, chimeric A-Fc, or TL1A antibody labeled with Zenon Alexa Fluor 647 Human IgG Labeling Kit (Molecular Probes) was added to a final concentration of 8.3 μg / mL. After the reaction, the cells were washed, and the Alexa Fluor 647 fluorescence intensity was analyzed using a flow cytometer. Under competitive conditions, the cells were reacted with unlabeled DNP antibody, TL1A antibody, or chimeric A-Fc at a final concentration of 50 μg / mL.

[0349] As a result, the binding of labeled chimera A-Fc competed with unlabeled TL1A antibody, and the binding of labeled TL1A antibody competed with unlabeled chimera A-Fc, confirming that chimera A bound to membrane-type TL1A on stimulated HUVEC cells (Figure 23).

[0350] (3) Primary FasL binding activity The binding of chimeric A-Fc (IEGRMD g1S) to primary soluble FasL produced by human T cells induced with activation-induced cell death (AICD) was evaluated as follows. Primary soluble FasL was prepared as follows. Specifically, human T cells isolated from frozen PBMCs of healthy donors in the same manner as in Example 9 were plated at 2 x 10 in a 96-well U-bottom plate (BD). 4 After 24 hours of incubation in 1 μg / mL PHA-L (eBioscience), IL-2 (Peptrotech) was added at a final concentration of 1 μg / mL and the cells were cultured for 5 days. After 5 days, the cells were harvested and seeded onto a 96-well U-bottom plate coated with 5 μg / mL of the anti-CD3 antibody OKT3 (BioLegend). After overnight incubation, AICD was induced and the culture supernatant was collected and concentrated to 1 / 10th the volume using an Amicon Ultra-15 (Millipore) column with a molecular weight cutoff of 10 kDa, pre-rinsed with sterile water.

[0351] A 96-well immunoplate (Thermo Scientific) coated with 10 μg / mL anti-human IgG antibody (American Qualex) was blocked with 1% Block Ace (DS Pharma Biomedical) and then 20 μg / mL chimeric A-Fc (IEGRMD g1S) or Fas-Fc (R&D) was captured. After washing, the plate was incubated with recombinant FasL (Abcam) or 10-fold concentrated AICD culture supernatant, which was used as a standard. After washing, the plate was incubated with biotinylated anti-FasL antibody (Abcam). After washing, the plate was incubated with streptavidin-HRP (Pierce). After washing again, TMB solution (Abcam) was added to develop color. The color reaction was stopped with 2N sulfuric acid solution, and the absorbance at 450 nm was measured.

[0352] The results for the plate capturing chimera A-Fc (IEGRMD g1S) are shown in Figure 24A, and the results for the plate capturing Fas-Fc are shown in Figure 24B. "10x AICDsup. CD3" indicates that culture supernatant from T cells in which AICD was induced under anti-CD3 antibody OKT3 stimulation conditions was used, while "none" indicates that culture supernatant from T cells in which AICD was not induced was used. For plates capturing either chimera A-Fc or Fas-Fc, FasL was detected only in the culture supernatant from AICD-induced T cells. This confirmed that chimera A-Fc binds to soluble FasL produced by AICD-induced T cells (Figures 24A and B).

[0353] [Example 17] Evaluation of physical properties of DcR3 variants For each chimera A-Fc (g1S, Eg1S, Eg1S YTE, Eg1S N434A, Eg1S LALAGANA, Eg1S LALAGA, g4PEK) and the FasL-binding-reduced variants E57K-Fc, E57L-Fc, E57R-Fc, E57V-Fc, 45-10-Fc, 45-18-Fc, and 82-5-Fc (all g4PEK), the elution time (min) was calculated by hydrophobic interaction chromatography (HIC) and the Tm value (°C) was calculated by differential scanning fluorimetry (DSF) in the same manner as in Example 4.

[0354] The results confirmed that the introduction of multiple amino acid mutations into the IgG1 Fc sequence did not significantly affect hydrophobicity. The elution times of all variants with reduced FasL binding were not significantly different from those of chimeric A-g4PEK, confirming that the substitution of one or two amino acids did not affect hydrophobicity (Figure 25).

[0355] The Tm values measured by DSF were influenced by the introduction of the YTE mutation, among the amino acid substitutions or insertions in the IgG1 Fc sequence. None of the Tm values of the FasL-binding-reducing variants were significantly different from those of chimeric A-g4PEK, confirming that the introduction of one or two amino acid substitutions in the CRD region of DcR3 did not affect thermal stability (Figure 26).

[0356] [Example 18] Evaluation of pharmacokinetics of DcR3 variants in mice Chimera A-Fc (Eg1S) and its FasL-binding-reduced variants E57K-Fc, E57L-Fc, E57R-Fc, E57V-Fc, 45-10-Fc, 45-18-Fc, and 82-5-Fc (all with g4PEK Fc) were evaluated for their pharmacokinetics in mice by the same method as described in Example 6. Chimera A-Fc (Eg1S) was stably expressed in CHO-K1 cells, and each of the FasL-binding-reduced variants was produced by transient expression in CHO-S cells. The resulting product was purified to 95% or more of the monomer by SEC (Superdex 200 Increase 10 / 300 GL) (GE Healthcare) using an AKTA purifier (GE Healthcare).

[0357] A single 10 mg / kg dose of each DcR3 variant was administered intravenously to 5- to 6-week-old BALB / c mice (female) (n = 2 or 3). The blood half-life (h) of the elimination phase after a single dose and the area under the blood concentration-time curve (AUC0-∞) (μg*h / mL) were calculated (Figure 27). DcR3 variants produced by transient expression in Expi293 cells were used as standards to measure serum DcR3 variant concentrations in mice administered E57K-Fc, E57R-Fc, or E57V-Fc.

[0358] As a result, compared to the wild-type DcR3 S195-Fc (IEGRMD g1S) described in Example 6, all of the DcR3 variants had significantly improved AUC.

[0359] [Example 19] In vivo efficacy evaluation of DcR3 variants The in vivo efficacy of chimeric A-Fc(g4PEK) was evaluated using a mouse acute xenogeneic graft-versus-host disease (GVHD) model.

[0360] (1) Establishment of a mouse acute xenogeneic GVHD model A mouse acute xenogeneic GVHD model was created using a method similar to that described in JP5209625. Six-week-old severe combined immunodeficient (SCID) female mice were intraperitoneally injected with 100 μg of rat anti-mouse IL2 receptor-β (IL2Rβ) chain antibody TMβ1 (Bio X Cell) on days -2 and 5 to deplete endogenous mouse natural killer cells. On day -1, the mice were irradiated sublethally with 1.7 Gy using a CellRad X-ray irradiator (Faxitron). On day 0, 3 x 10 6 Human PBMCs (AllCells) were transferred intraperitoneally, followed by intraperitoneal injection of 300 μg of chimeric A-Fc or DNP antibody (both g4PEK) adjusted with 100 μL of PBS. The chimeric A-Fc-treated group received additional 300 μg of chimeric A-Fc on days 4 and 8. After 12 days, the mice were scored for gross pathology due to the GVHD reaction, and the spleens were harvested from sacrificed mice to assess the number of human cells in the spleens.

[0361] (2) Evaluation of GVHD pathology score The macroscopic pathology observed on day 12 was scored based on four indices: coat condition, intestinal redness, activity status, and weight loss (Figure 28A). Each indice was scored as 0, 1, or 2 for absent, mild, or severe, respectively, and the GVHD pathology score was calculated by adding up all the indices.

[0362] As a result, an increase in the GVHD pathology score due to the transfer of human PBMCs and a decrease in the pathology score due to the administration of chimera A-Fc compared to the DNP antibody-administered group were confirmed, demonstrating the efficacy of chimera A-Fc (Figure 28B).

[0363] (3) Measurement of human cell count in the spleen Spleens were harvested from mice and disrupted using gentleMACS Dissociators (Miltenyi) to prepare splenocyte suspensions. After hemolysis with Lysing buffer (BD Biosciences), the cells were stained with PE / Cy7-conjugated anti-human CD45, FITC-conjugated anti-human CD3, APC-conjugated anti-human CD4, and PE-conjugated anti-human CD8 (all BioLegend). Each human cell subset was analyzed by flow cytometry. CountBright Absolute Counting Beads for flow cytometry (Thermo Fisher Scientific) were used to measure cell numbers. The fluorescent bead counts detected by flow cytometry and the abundance ratios of each cell subset were corrected for the known amount of beads added to calculate the total number of cells in the spleen.

[0364] As a result, human cells were detected in the spleens of mice after the transfer of human PBMCs, and a decrease in the number of human cells in the spleens of mice after administration of chimeric A-Fc was confirmed compared to the DNP antibody-administered group (FIG. 29).

[0365] [Example 20] Evaluation of binding activity to soluble DcR3 ligand trimer The binding activity of the chimera A-Fc having various mutant Fc and the variants with reduced FasL binding activity prepared in Example 12 to soluble human LIGHT trimer, soluble cynomolgus monkey LIGHT trimer, soluble human TL1A trimer, soluble cynomolgus monkey TL1A trimer, soluble human FasL trimer, or soluble cynomolgus monkey FasL trimer was evaluated as follows, with some modifications to the method described in Examples 7 and 13.

[0366] (1) Preparation of soluble DcR3 ligand trimers The soluble human DcR3 ligand trimer used was that prepared in Example 13. The soluble recombinant cynomolgus monkey LIGHT had a FLAG tag (DYKDDDDK) attached to the N-terminus and the extracellular domain of cynomolgus monkey LIGHT (Asp74-Val240) (SEQ ID NO: 134) linked downstream (FLAG-cynoLIGHT) (nucleotide sequence: SEQ ID NO: 341, amino acid sequence: SEQ ID NO: 342). The soluble recombinant cynomolgus monkey TL1A (His6-cynoTL1A) and soluble cynomolgus monkey FasL (His6-cynoFasL) (amino acid sequences: SEQ ID NOs: 122 and 124, respectively) were the same sequences as in Example 7. LIGHT and TL1A were transiently expressed in Expi293, and FasL was transiently expressed in CHO-S.

[0367] FLAG-cynoLIGHT and His6-cynoTL1A were purified as described in Example 13. His6-cynoFasL was purified by passing the culture supernatant through a column packed with Ni Sepharose Fast Flow resin (GE Healthcare), washing with a washing buffer (60 mM imidazole, 20 mM sodium phosphate, 0.5 M NaCl, pH 7.4) prepared using a His Buffer Kit (GE Healthcare), and then eluting with an elution buffer (250 mM imidazole, 20 mM sodium phosphate, 0.5 M NaCl, pH 7.4) and replacing the buffer with PBS.

[0368] The purified FLAG-cynoLIGHT and His6-cynoTL1A were subjected to gel filtration chromatography (SEC) using an AKTA purifier (GE Healthcare) on a Superdex 200 Increase 10 / 300 GL column (GE Healthcare). The purified His6-cynoFasL was analyzed by SEC-UHPLC (Shimadzu Nexera X2 instrument, ACQUITY UPLC Protein BEH SEC 200Å, 1.7 μm, 4.6 × 150 mm column (Waters)), and it was confirmed that more than 85% of the purified His6-cynoFasL was trimer.

[0369] (2) Binding activity measurement using BIAcore The chimeric A-Fc containing various mutant Fc and the variants with reduced FasL binding were the variants described in Example 12, which were purified to 95% or more of the monomer by SEC (Superdex 200 Increase 10 / 300 GL) (GE Healthcare) using AKTA purifier (GE Healthcare) or SEC (HiLoad 26 / 600 Superdex 200 pg) (GE Healthcare) using AKTA pure 25 (GE Healthcare).

[0370] The binding activity to human or cynomolgus monkey DcR3 ligand trimers was analyzed by SPR. BIAcore T-100 (GE Healthcare) was used to measure binding activity to human LIGHT, cynomolgus monkey LIGHT, human TL1A, and cynomolgus monkey TL1A, and BIAcore T-200 (GE Healthcare) was used to measure binding activity to human FasL and cynomolgus monkey FasL. HBS-EP+ Buffer was used as the buffer.

[0371] Various DcR3 variants were flowed through a Series S Sensor Chip Protein A (GE Healthcare) at 10 μL / min for 30 seconds for capture. Protein-free buffer was flowed through the reference flow cell. Subsequently, each human DcR3 ligand trimer or each cynomolgus monkey DcR3 ligand trimer diluted to 0.08-20 nmol / L was flowed as the analyte. Binding of human LIGHT trimer, human TL1A trimer, or each cynomolgus monkey DcR3 ligand trimer was monitored by flowing at 30 μL / min for 2 minutes, followed by buffer flow for 5 minutes to monitor dissociation. Binding of human FasL trimer was monitored by flowing at 30 μL / min for 1 minute, followed by buffer flow for 3 minutes to monitor dissociation. Next, 3 mol / L magnesium chloride was flowed at 30 μL / min for 1 minute to perform the regeneration reaction. The binding activity for human LIGHT, cynomolgus monkey LIGHT, human TL1A, and cynomolgus monkey TL1A was measured using BIAcore T-100 evaluation software with a 1:1 binding model. The binding activity for human FasL and cynomolgus monkey FasL was measured using BIAcore T-200 evaluation software with a 1:1 binding model. Each DcR3 ligand was measured as a trimer (human LIGHT trimer: 62.4 kDa, cynomolgus monkey LIGHT trimer: 57.9 kDa, human TL1A trimer: 66.2 kDa, cynomolgus monkey TL1A trimer: 66.1 kDa, human FasL trimer: 53.1 kDa, cynomolgus monkey FasL trimer: 53.1 kDa), and each kinetic constant (k a , k d , K. D ) was calculated.

[0372] As a result, it was confirmed that all chimera A-Fc containing various mutant Fc (Eg1S, Eg1S YTE, Eg1S N434A, Eg1S LALAGA, Eg1S LALAGANA) (respective amino acid sequences: SEQ ID NOs: 168, 170, 172, 176, 178) bound to the respective DcR3 trimer ligands of human and cynomolgus monkey (Figures 30A and B).

[0373] The FasL-binding-reducing variants E57K-Fc, E57R-Fc, E57V-Fc, 45-10-Fc, and 45-18-Fc (each of which has an Fc of Eg1S YTE, Eg1S N434A, or Eg1S LALAGANA) all exhibited reduced K binding to human LIGHT trimer and human TL1A trimer compared to chimera A-Eg1S. D However, there was no significant difference in the K values for the human FasL trimer in these mutants with reduced FasL binding activity. D The value was significantly increased compared to chimera A-Eg1S, confirming that the binding activity was selectively reduced in human FasL (Fig. 30A).

[0374] All of the mutants with reduced FasL binding having various mutant Fc showed a K D The K values were similar to those of chimera A-Eg1S, but E57R-Fc (Eg1S YTE, Eg1S N434A) tended to have a reduced binding activity to cynomolgus monkey TL1A trimer. On the other hand, all of the FasL-binding-reducing mutants showed a K value of 1000 for cynomolgus monkey FasL trimer. D The value was significantly increased compared to chimera A-Eg1S, confirming that binding to FasL was selectively reduced (FIG. 30B).

[0375] [Example 21] Evaluation of the neutralizing activity of DcR3 soluble ligands The neutralizing activities of FasL-binding-reduced variants with various mutant Fc against soluble human LIGHT, soluble human TL1A, and soluble human FasL were evaluated by the same method as in Example 9 or Example 14. The various DcR3 variants used were chimera A-Fc prepared in Example 12, or FasL-binding-reduced variants E57K-Fc and 45-18-Fc (Eg1S YTE or Eg1S LALAGANA Fc in both cases) prepared by transient expression in CHO-S cells, which were purified to 95% or more of the monomer by the method described in Example 14(1) or by SEC (HiLoad 16 / 600 Superdex 200 pg) (GE Healthcare) using an AKTA purifier (GE Healthcare).

[0376] (1) Soluble human LIGHT neutralizing activity The inhibitory activity of various DcR3 variants against LIGHT-dependent CXCL10 production by IFN-γ-stimulated intestinal myofibroblasts was evaluated using the method described in Example 14. Chimera A-Fc, E57K-Fc, and 45-18-Fc (each containing Eg1S YTE or Eg1S LALAGANA Fc) were found to inhibit CXCL10 production in a concentration-dependent manner and to have neutralizing activity against soluble LIGHT. The results are shown in Figure 31.

[0377] (2) Soluble human TL1A neutralizing activity The inhibitory activity of various DcR3 variants against TL1A-dependent IFN-γ production by IL-12- and IL-18-stimulated human T cells was evaluated using the method described in Example 9. Chimera A-Fc, E57K-Fc, and 45-18-Fc (each containing Eg1S YTE or Eg1S LALAGANA Fc) were found to inhibit IFN-γ production in a concentration-dependent manner and to have neutralizing activity against soluble TL1A. The results are shown in Figure 32.

[0378] (3) Soluble human FasL neutralizing activity The inhibitory activity of various DcR3 variants against FasL-dependent cell death in Jurkat cells was evaluated using the method described in Example 9. Chimera A-Fc (Eg1S YTE, Eg1S LALAGANA) concentration-dependently inhibited Jurkat cell death and demonstrated neutralizing activity against soluble FasL. On the other hand, E57K-Fc and 45-18-Fc (both with Eg1S YTE or Eg1S LALAGANA Fc) significantly reduced their inhibitory activity against Jurkat cell death, demonstrating selective reduction in neutralizing activity against soluble FasL. The results are shown in Figure 33.

[0379] In Figures 31 to 33, solid lines and X (-×-) represent the negative control anti-DNP antibody, solid lines and black circles (-●-) represent chimera A-Fc (Eg1S YTE), solid lines and white triangles (-△-) represent E57K-Fc (Eg1S YTE), solid lines and white circles (-○-) represent 45-18-Fc (Eg1S YTE), dotted lines and black circles (--●--) represent chimera A-Fc (Eg1S LALAGANA), dotted lines and white triangles (--△--) represent E57K-Fc (Eg1S LALAGANA), and dotted lines and white circles (--○--) represent 45-18-Fc (Eg1S LALAGANA).

[0380] [Example 22] Evaluation of binding activity to DcR3 membrane-type ligands The binding activities of chimera A and its variants with reduced FasL binding activity having various mutant Fc to membrane-type human LIGHT, membrane-type human TL1A, and membrane-type human FasL were evaluated in the same manner as in Example 15.

[0381] Chimera A-Fc (Eg1S) was produced from a stable CHO cell line and purified to >95% monomer by SEC. Chimera A-Fc (Eg1S YTE, Eg1S N434A, Eg1S LALAGA, Eg1S LALAGANA) was produced by tra...

Claims

1. A variant of wild-type Decay Receptor 3 (hereinafter abbreviated as DcR3), which is a variant of wild-type DcR3, and which exhibits improved pharmacokinetics compared to the wild-type DcR3.

2. The DcR3 variant of claim 1, which has one or more N-glycoside-linked complex-type sugar chains.

3. The DcR3 variant of claim 1, which has neutralizing activity against at least one of LIGHT, TL1A, and FasL.

4. The DcR3 variant according to claim 1, which has neutralizing activity against all of LIGHT, TL1A and FasL.

5. The DcR3 variant of claim 1, which has no neutralizing activity against FasL and has neutralizing activity against any one or more of LIGHT and TL1A.

6. The DcR3 variant of claim 1, which has no neutralizing activity against FasL and has neutralizing activity against LIGHT and TL1A.

7. A DcR3 variant comprising: a first chimeric cysteine-rich region consisting of an amino acid sequence in which at least a portion of the cysteine-rich domain (hereinafter abbreviated as CRD) of wild-type DcR3 is substituted with at least a portion of the cysteine-rich domain of a TNF receptor superfamily molecule other than DcR3; or a second chimeric cysteine-rich region consisting of an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted or added in the amino acid sequence of the first chimeric cysteine-rich region.

8. The DcR3 variant of claim 7, which has one or more complex-type N-glycoside-linked sugar chains.

9. The DcR3 variant according to claim 7, which has neutralizing activity against at least one of LIGHT, TL1A, and FasL.

10. The DcR3 variant according to claim 7, which has neutralizing activity against all of LIGHT, TL1A, and FasL.

11. The DcR3 variant of claim 7, which has no neutralizing activity against FasL and has neutralizing activity against any one or more of LIGHT and TL1A.

12. The DcR3 variant of claim 7, which has no neutralizing activity against FasL and has neutralizing activity against LIGHT and TL1A.

13. The DcR3 variant of claim 7, wherein the TNF receptor superfamily molecule is OPG.

14. at least a portion of the cysteine-rich domain of wild-type DcR3 is selected from all or a portion of CRD1, all or a portion of CRD2, all or a portion of CRD3, and all or a portion of CRD4; The DcR3 variant of claim 7, wherein at least a portion of the cysteine-rich domain of the TNF receptor superfamily molecule is selected from all or a portion of CRD1, all or a portion of CRD2, all or a portion of CRD3, and all or a portion of CRD4.

15. the first chimeric cysteine-rich region comprises: replacement of a portion of the CRD1 of the wild-type DcR3 with a portion of the CRD1 of the TNF receptor superfamily molecule corresponding to the portion of the CRD1 of the wild-type DcR3; replacement of the entire CRD1 of said wild-type DcR3 with the entire CRD1 of said TNF receptor superfamily molecule; replacement of a portion of the CRD2 of the wild-type DcR3 with a portion of the CRD2 of the TNF receptor superfamily molecule corresponding to the portion of the CRD2 of the wild-type DcR3; replacement of the entire CRD2 of said wild-type DcR3 with the entire CRD2 of said TNF receptor superfamily molecule; replacement of a portion of the CRD3 of the wild-type DcR3 with a portion of the CRD3 of the TNF receptor superfamily molecule corresponding to the portion of the CRD3 of the wild-type DcR3; replacement of the entire CRD3 of said wild-type DcR3 with the entire CRD3 of said TNF receptor superfamily molecule; Substitution of a portion of the CRD4 of the wild-type DcR3 with a portion of the CRD4 of the TNF receptor superfamily molecule that corresponds to the portion of the CRD4 of the wild-type DcR3; and Replacing the entire CRD4 of said wild-type DcR3 with the entire CRD4 of said TNF receptor superfamily molecule.

15. The DcR3 variant of claim 14, having one or more substitutions selected from:

16. 16. The DcR3 variant of claim 15, wherein the first chimeric cysteine-rich region retains a portion or the entirety of the CRD2 of the wild-type DcR3.

17. The DcR3 variant of claim 15, wherein the first chimeric cysteine-rich region retains a part or the entirety of the CRD3 of the wild-type DcR3.

18. The DcR3 variant of claim 14, wherein the first chimeric cysteine-rich region comprises the amino acid sequence (a), (b), (c) or (d) below, and the second chimeric cysteine-rich region comprises the amino acid sequence (e) below. (a) an amino acid sequence in which the CRD1 of wild-type DcR3 in the cysteine-rich region is substituted with the CRD1 of OPG; (b) an amino acid sequence in which the CRD4 of wild-type DcR3 in the cysteine-rich region is substituted with the CRD4 of OPG; (c) an amino acid sequence in the cysteine-rich region of the wild-type DcR3, in which the CRD1 of the wild-type DcR3 is replaced with the CRD1 of OPG and the CRD4 of the wild-type DcR3 is replaced with the CRD4 of OPG; (d) an amino acid sequence in which the 103rd to 123rd positions from the N-terminus in the amino acid sequence of (a), (b), or (c) are substituted with the corresponding positions in the amino acid sequence of the cysteine-rich domain of OPG; (e) An amino acid sequence in which 1 to 30 amino acids are deleted, substituted, inserted or added in the amino acid sequence of (a), (b), (c) or (d).

19. the amino acid sequence of (a) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 26 or 50, the amino acid sequence (b) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 28 or 52, the amino acid sequence of (c) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 30 or 54, The DcR3 variant according to claim 18, wherein the amino acid sequence (d) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 32 or 56.

20. The DcR3 variant described in claim 18, wherein the amino acid sequence (e) has one or more amino acids selected from the group consisting of Glu at the 57th position, Arg at the 58th position, and Arg at the 60th position from the N-terminus of the amino acid sequence (a), (b), (c), or (d) substituted with other amino acids.

21. The DcR3 variant described in claim 18, wherein the amino acid sequence (e) has a substitution of Glu at position 57 and Arg at position 58 from the N-terminus of the amino acid sequence (a), (b), (c), or (d) with other amino acids.

22. The DcR3 variant described in claim 18, wherein the amino acid sequence (e) has one or more substitutions selected from the group consisting of a substitution of Glu at the 57th position from the N-terminus of the amino acid sequence (a), (b), (c) or (d) with Lys, Leu, Arg, Val, Ala, Phe, His, Ile or Met, a substitution of Arg at the 58th position with Asp, Glu or Thr, and a substitution of Arg at the 60th position with Lys.

23. The DcR3 variant described in claim 18, wherein the amino acid sequence (e) has a substitution of Glu at the 57th position from the N-terminus of the amino acid sequence (a), (b), (c) or (d) with Lys, Leu, Arg, Val, Ala, Phe, His, Ile or Met, and a substitution of Arg at the 58th position from the N-terminus with Asp, Glu or Thr.

24. The DcR3 variant of claim 18, wherein the amino acid sequence of (e) has a substitution selected from the following (f) to (i): (f) Substitution of Asn at the 131st and 144th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) with other amino acids. (g) Substitution of Asn at the 131st, 144th, and 157th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) with other amino acids. (h) Substitution of Thr at the 133rd position and Ser at the 146th position from the N-terminus of the amino acid sequence of (b), (c), or (d) with other amino acids. (i) Substitution of Thr at position 133, Ser at position 146, and Thr at position 159 from the N-terminus of the amino acid sequence of (b), (c), or (d) with other amino acids.

25. The DcR3 variant of claim 18, wherein the amino acid sequence of (e) has a substitution selected from the following (f') to (i'): (f') Substitution of Asn at the 131st and 144th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) above with Ser (g') Substitution of Asn at the 131st, 144th, and 157th positions from the N-terminus of the amino acid sequence of (b), (c), or (d) above with Ser (h') Substitution of Thr at the 133rd position and Ser at the 146th position from the N-terminus of the amino acid sequence of (b), (c), or (d) with Ala. (i') Substitution of Thr at position 133, Ser at position 146, and Thr at position 159 from the N-terminus of the amino acid sequence of (b), (c), or (d) with Ala

26. The DcR3 variant of claim 18, wherein the amino acid sequence (e) is an amino acid sequence consisting of amino acids 1 to 164 from the N-terminus of the amino acid sequence set forth in SEQ ID NO: 58, 60, 62, 64, 66, 68, 70, 180, 182, 184, 186, 188, 270, 272, 274, 276, 278, 280, 282, 284 or 286.

27. The DcR3 variant described in claim 7, wherein the DcR3 variant comprises the first or second chimeric cysteine-rich region and part or all of the heparan sulfate binding region of the wild-type DcR3 linked to the C-terminal side of the first or second chimeric cysteine-rich region, or the DcR3 variant comprises the first or second chimeric cysteine-rich region but does not comprise the heparan sulfate binding region of the wild-type DcR3.

28. The DcR3 variant is (I) an amino acid sequence set forth in SEQ ID NO: 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, or 46, or an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted, or added in the amino acid sequence; and (II) An amino acid sequence set forth in SEQ ID NO: 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, 180, 182, 184, 186, 188, 270, 272, 274, 276, 278, 280, 282, 284, or 286, or an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted, or added in the amino acid sequence.

28. The DcR3 variant of claim 27, comprising any one of the amino acid sequences selected from:

29. The DcR3 variant described in claim 7, comprising an Fc region derived from a human IgG1, IgG2, or IgG4 antibody, or a mutant Fc region consisting of an amino acid sequence in which one or more amino acids have been deleted, substituted, inserted, or added in the amino acid sequence of the Fc region.

30. The DcR3 variant of claim 29, wherein the Fc region or the variant Fc region is linked to the C-terminal side of the first or second chimeric cysteine-rich region via another region or a linker.

31. The DcR3 variant of claim 29, wherein the mutant Fc region has a substitution of Cys at position 220 of the amino acid sequence of the heavy chain of human IgG1, as shown in the EU index, with Ser.

32. The DcR3 variant of claim 31, wherein the mutant Fc region has a substitution of Leu at position 234 with Ala, Leu at position 235 with Ala, and Gly at position 237 with Ala in the amino acid sequence of the heavy chain of human IgG1, as shown in the EU index.

33. The DcR3 variant of claim 31, wherein the mutant Fc region has a substitution of Asn at position 434 of the amino acid sequence of the heavy chain of human IgG1, as shown in the EU index, with Ala.

34. The DcR3 variant of claim 31, wherein the mutant Fc region has a substitution of Met at position 252 with Tyr, Ser at position 254 with Thr, and Thr at position 256 with Glu in the amino acid sequence of the heavy chain of human IgG1, as shown in the EU index.

35. The DcR3 variant described in claim 29, wherein the mutant Fc region has a substitution of Ser at position 228 with Pro, Leu at position 235 with Glu, and Arg at position 409 with Lys, as shown in the EU index, in the amino acid sequence of the heavy chain of human IgG4.

36. 30. The DcR3 variant of claim 29, wherein the DcR3 variant comprises a variant Fc region consisting of the amino acid sequence set forth in SEQ ID NO: 72, 74, 156, 158, 160, 162, 164, 166, 311, 312, or 313.

37. The DcR3 variants are selected from the group consisting of SEQ ID NOs: 76, 78, 80, 82, 84, 86, 88, 90, 92, 94, 96, 98, 150, 168, 170, 172, 174, 176, 178, 190, 192, 194, 196, 198, 200, 202, 204, 206, 208, 210, 212, 214, 216, 218, 220, 222, 224, 226, 228, 230, 232, 234, 236, 238, 240, 242, 244, 246, 248, 250, 252, 254, 256, 258, 260, and 262. 30, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, or 337, or an amino acid sequence in which 1 to 30 amino acids have been deleted, substituted, inserted, or added.

38. A DcR3 variant composition comprising the DcR3 variant of claim 1 or 7.

39. The composition of claim 38, comprising a DcR3 variant having one or more complex-type N-glycoside-linked sugar chains and a DcR3 variant having no complex-type N-glycoside-linked sugar chains.

40. A DNA encoding the DcR3 variant of claim 1 or 7.

41. A recombinant vector containing the DNA of claim 40.

42. A transformant obtained by introducing the recombinant vector according to claim 41 into a host cell.

43. The transformant according to claim 42, wherein the host cell is a cell derived from a mammal.

44. The transformant of claim 43, wherein the mammalian cell is a CHO cell.

45. A method for producing a DcR3 variant or a DcR3 variant composition, comprising culturing the transformant described in claim 42 in a culture medium to produce and accumulate the DcR3 variant described in claim 1 or 7, and purifying the DcR3 variant from the resulting culture medium.

46. 46. A DcR3 variant or DcR3 variant composition produced using the production method of claim 45.

47. A pharmaceutical composition comprising the DcR3 variant or DcR3 variant composition according to claim 1, 7 or 46 as an active ingredient.

48. The pharmaceutical composition according to claim 47, which is an agent for preventing or treating an autoimmune disease, an inflammatory disease, or an allergic disease.

49. A method for preventing or treating an autoimmune disease, inflammatory disease, or allergic disease, comprising administering the pharmaceutical composition of claim 47 to a patient in need of such prevention or treatment.

Images