Cells for ex-vivo gene therapy that produce modified blood coagulation factor viii

Abstract

The present invention addresses the problem of providing novel cells for an ex-vivo gene therapy that produce a modified factor VIII. Cells for an ex-vivo gene therapy according to the present invention, as means for solving the problem, produce a modified human FVIII that comprises the 20-2351th amino acid sequence of the amino acid sequence shown in SEQ ID NO: 1, wherein the B domain (760-1667th amino acid sequence) is modified, the number of Asn residues subject to N-type glycosylation modification is 4-6 as counted from the N-terminal, and the function of a furin recognition site is deficient.

Description

Cells for ex vivo gene therapy producing modified blood coagulation factor VIII

[0001] The present invention relates to cells for ex vivo gene therapy that produce modified blood coagulation factor VIII.

[0002] Blood coagulation factor VIII (hereinafter "FVIII") is one of the factors responsible for hemostasis. Congenital abnormalities in the FVIII gene lead to deficiency or a decrease in blood FVIII concentration, resulting in a hemostatic disorder known as hemophilia A. Treatment for hemophilia A has traditionally been performed using plasma-derived FVIII preparations, but recombinant preparations are now also being used clinically. Human FVIII is a glycoprotein consisting of 2,332 amino acid residues. It is expressed primarily in liver cells as a precursor consisting of 2,351 amino acid residues (the amino acid sequence of which is shown in SEQ ID NO: 1), including a signal peptide consisting of 19 amino acid residues. It then undergoes cleavage of the signal peptide, glycosylation, and other processes to become a full-length form consisting of 2,332 amino acid residues. The full-length protein has a domain structure of A1-A2-B-A3-C1-C2, with linker regions called a1, a2, and a3 between the A1 and A2 domains, the A2 and B domains, and the B and A3 domains, respectively. The full-length form is cleaved by furin in the Golgi apparatus between Arg1648 at the C-terminus of the B domain and Glu1649 at the N-terminus of the a3 linker region (positions 1667 and 1668 in the amino acid sequence of SEQ ID NO: 1), converting it into a duplex heavy and light chain to become a mature form. After secretion from cells into the circulating blood, it is cleaved by thrombin between Arg372 at the C-terminus of the a1 linker region and Ser373 at the N-terminus of the A2 domain, between Arg740 at the C-terminus of the a2 linker region and Ser741 at the N-terminus of the B domain, and between Arg1689 at the C-terminus of the a3 linker region and Ser1690 at the N-terminus of the A3 domain, converting it into an active form and thereby exerting its specified function.

[0003] Hemophilia A is a hereditary disease caused by a single gene, and is therefore considered a suitable target disease for gene therapy, and therefore, in recent years, research and development of gene therapy for hemophilia A has been actively conducted (e.g., Non-Patent Document 1). Depending on the technique, gene therapy can be broadly divided into in vivo gene therapy, in which a vector for expressing a target gene in a living body is directly administered to a patient, and ex vivo gene therapy, in which a target gene is introduced into cells extracted from the patient and then returned to the patient, thereby expressing the target gene in the living body.

[0004] Research and development of ex vivo gene therapy for hemophilia A has been underway, for example, using a lentiviral vector, which is highly safe and enables long-term stable gene expression by efficiently incorporating a gene of interest into the host cell genome, to introduce a polynucleotide for producing a FVIII variant in vivo in which the B domain of human FVIII has been deleted into cells extracted from a patient and then return the cells to the patient (ID: NCT03818763, introduced in Non-Patent Document 1).However, it can be said that research and development of ex vivo gene therapy for hemophilia A is still in its infancy.

[0005] Yuji Kashiwakura et al., Advances in gene therapy for hemophilia, Journal of Thrombosis and Hemostasis 2021; 32(1): 17-25

[0006] An object of the present invention is to provide novel cells for ex vivo gene therapy that produce FVIII variants.

[0007] The ex vivo gene therapy cells of the present invention produce a modified human FVIII consisting of the amino acid sequence from amino acid 20 to amino acid 2351 in the amino acid sequence set forth in SEQ ID NO: 1, in which the B domain (amino acid sequence from amino acid 760 to amino acid 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted. Furthermore, in the ex vivo gene therapy cells of the present invention, the modified FVIII is any one of the following (a) to (c): (a) a protein consisting of the amino acid sequence of amino acids 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2; (b) a protein consisting of an amino acid sequence containing a deletion, substitution, addition or insertion of one or more amino acid residues in the amino acid sequence of amino acids 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2, and having biological activity equivalent to that of the protein of (a); (c) a protein consisting of an amino acid sequence having at least 90% sequence identity with the amino acid sequence of amino acids 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2, and having biological activity equivalent to that of the protein of (a). Furthermore, in the cells for ex vivo gene therapy of the present invention, the FVIII variant is any of the following (a) to (c): (a) a protein consisting of the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3; (b) a protein consisting of an amino acid sequence containing a deletion, substitution, addition or insertion of one or more amino acid residues in the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, and having biological activity equivalent to that of the protein of (a); (c) a protein consisting of an amino acid sequence having at least 90% sequence identity with the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, and having biological activity equivalent to that of the protein of (a). Furthermore, in the cells for ex vivo gene therapy of the present invention, the FVIII variant is any of the following (a) to (c):(a) a protein consisting of the amino acid sequence of amino acids 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4; (b) a protein consisting of an amino acid sequence comprising the amino acid sequence of amino acids 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4, with one or more amino acid residues deleted, substituted, added, or inserted, and having biological activity equivalent to that of the protein of (a); (c) a protein consisting of an amino acid sequence having at least 90% sequence identity with the amino acid sequence of amino acids 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4, and having biological activity equivalent to that of the protein of (a). Furthermore, in the cells for ex vivo gene therapy of the present invention, the FVIII variant is one in which the Pro residue at position 758 in the amino acid sequence shown in SEQ ID NO: 1 is substituted with another amino acid residue. Furthermore, in the cells for ex vivo gene therapy of the present invention, the other amino acid residue is a polar, uncharged amino acid residue selected from Thr, Ser, Asn, and Gln. Furthermore, in the cells for ex vivo gene therapy of the present invention, the FVIII variant is any of the following (a) to (c): (a) a protein consisting of the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5; (b) a protein consisting of an amino acid sequence containing a deletion, substitution, addition or insertion of one or more amino acid residues in the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, and having biological activity equivalent to that of the protein of (a); (c) a protein consisting of an amino acid sequence having at least 90% sequence identity with the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, and having biological activity equivalent to that of the protein of (a). Furthermore, in the cells for ex vivo gene therapy of the present invention, the FVIII variant is any of the following (a) to (c):(a) a protein consisting of the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14; (b) a protein consisting of an amino acid sequence comprising the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, with one or more amino acid residues deleted, substituted, added or inserted, and having biological activity equivalent to that of the protein of (a); (c) a protein consisting of an amino acid sequence having at least 90% sequence identity with the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, and having biological activity equivalent to that of the protein of (a). Furthermore, in the cells for ex vivo gene therapy of the present invention, the FVIII variant is any of the following (a) to (c): (a) a protein consisting of the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15; (b) a protein consisting of an amino acid sequence comprising the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, with one or more amino acid residues deleted, substituted, added or inserted, and having biological activity equivalent to that of the protein of (a); (c) a protein consisting of an amino acid sequence having at least 90% sequence identity with the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, and having biological activity equivalent to that of the protein of (a). Furthermore, in the cells for ex vivo gene therapy of the present invention, the FVIII variant is any of the following (a) to (c): (a) a protein consisting of the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16; (b) a protein consisting of an amino acid sequence containing a deletion, substitution, addition, or insertion of one or more amino acid residues in the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16, and having biological activity equivalent to that of the protein of (a); (c) a protein consisting of an amino acid sequence having at least 90% sequence identity with the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16, and having biological activity equivalent to that of the protein of (a). Furthermore, the cells for ex vivo gene therapy of the present invention are adipocytes. Furthermore, the transplant composition of the present invention comprises the cells for ex vivo gene therapy of the present invention and a pharmaceutically acceptable carrier.The transplant composition of the present invention further comprises an extracellular matrix component. The transplant composition of the present invention further comprises an angiogenic factor. The method for producing cells for ex vivo gene therapy of the present invention involves introducing into cells that can be transplanted into the body of a subject for ex vivo gene therapy an expression vector carrying a polynucleotide encoding at least the amino acid sequence of a human FVIII variant consisting of amino acids 20 to 2351 in the amino acid sequence set forth in SEQ ID NO: 1, in which the B domain (amino acid sequence 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted. The method for producing cells for ex vivo gene therapy of the present invention further comprises a viral vector as the expression vector. The method for producing cells for ex vivo gene therapy of the present invention further comprises a lentiviral vector as the viral vector. The expression vector for producing the cell for ex vivo gene therapy of the present invention comprises a polynucleotide encoding at least the amino acid sequence of a human FVIII variant consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence set forth in SEQ ID NO: 1, wherein the B domain (amino acid sequence of amino acids 760 to 1667) is modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site is deleted. The FVIII variant of the present invention is a human FVIII variant consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence set forth in SEQ ID NO: 1, wherein the B domain (amino acid sequence of amino acids 760 to 1667) is modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, the function of the furin recognition site is deleted, and the Pro residue at position 758 is substituted with another amino acid residue. In addition, in the FVIII variant of the present invention, the other amino acid residue is a polar, uncharged amino acid residue selected from Thr, Ser, Asn, and Gln. Furthermore, the polynucleotide of the present invention encodes at least the amino acid sequence of the FVIII variant of the present invention.The transplant composition of the present invention further comprises cells that produce human vWF and / or a fragment thereof in vivo. The cells of the present invention produce a human vWF fragment in vivo. The human vWF fragment of the present invention is a fragment that includes at least the D' and D3 domains, which are binding sites for FVIII. The cells of the present invention are a fragment of human vWF that has the SP, D', and D3 domains of the SP-D1-D2-D'-D3-A1-A2-A3-D4-B1-B2-B3-C1-C2-CK domain structure of the precursor, but does not have the A2, A3, D4, B1, B2, B3, C1, or C2 domains, and the A1 domain may have up to 50 amino acid residues on the N-terminal side, and may or may not have the D1, D2, or CK domains. Furthermore, in the cells of the present invention, the fragment of human vWF is any one of the following (a) to (c): (a) a protein consisting of the amino acid sequence shown in SEQ ID NO: 29 (b) a protein consisting of the amino acid sequence shown in SEQ ID NO: 29, which contains a deletion, substitution, addition, or insertion of one or more amino acid residues, and which has biological activity equivalent to that of the protein in (a) (c) a protein consisting of an amino acid sequence that has at least 90% sequence identity with the amino acid sequence shown in SEQ ID NO: 29, and which has biological activity equivalent to that of the protein in (a) Furthermore, in the cells of the present invention, the fragment of human vWF is any one of the following (a) to (c): (a) a protein consisting of the amino acid sequence shown in SEQ ID NO: 29, which contains a deletion, substitution, addition, or insertion of one or more amino acid residues, and which has biological activity equivalent to that of the protein in (a) (a) a protein consisting of the amino acid sequence shown in SEQ ID NO: 30; (b) a protein consisting of an amino acid sequence containing one or more deletions, substitutions, additions or insertions of amino acid residues in the amino acid sequence shown in SEQ ID NO: 30, and having biological activity equivalent to that of the protein of (a); (c) a protein consisting of an amino acid sequence having at least 90% sequence identity with the amino acid sequence shown in SEQ ID NO: 30, and having biological activity equivalent to that of the protein of (a). Furthermore, in the cells of the present invention, the fragment of human vWF is any one of the following (a) to (c):(a) a protein consisting of the amino acid sequence set forth in SEQ ID NO: 31; (b) a protein consisting of the amino acid sequence set forth in SEQ ID NO: 31, which contains a deletion, substitution, addition, or insertion of one or more amino acid residues, and which has biological activity equivalent to that of the protein of (a); (c) a protein consisting of an amino acid sequence having at least 90% sequence identity with the amino acid sequence set forth in SEQ ID NO: 31, and which has biological activity equivalent to that of the protein of (a). Furthermore, the pharmaceutical agent of the present invention comprises, as an active ingredient, cells that produce human vWF and / or a fragment thereof in vivo, for increasing the amount of FVIII variant secreted into the blood by transplanting the cells for ex vivo gene therapy of the present invention into the body of a subject of ex vivo gene therapy. Furthermore, the transplant composition of the present invention comprises cells that produce human vWF and / or a fragment thereof in vivo and a pharmaceutically acceptable carrier. In addition, the set of transplant compositions of the present invention includes a transplant composition comprising the ex vivo gene therapy cells of the present invention and a pharmaceutically acceptable carrier, and a transplant composition comprising the cells of the present invention that produce human vWF and / or a fragment thereof in vivo and a pharmaceutically acceptable carrier.

[0008] According to the present invention, novel cells for ex vivo gene therapy that produce FVIII variants can be provided.

[0009] 1 is a diagram showing the differences in the amino acid sequences of the B domain and those before and after it of five types of FVIII variants (ΔB, N6, N6Δfurin, N5Δfurin, N5Δfurin(P739T)) in Experiment 1 of Experiment A of the Example.

[0033] FIG. 1 is a diagram showing the structure of the lentiviral vector plasmid producing N5Δfurin(P739T).

[0034] FIG. 2 is a graph showing the average introduced copy number of FVIII variant genes in the preadipocytes introduced with the five types of FVIII variant genes prepared in Experiment 2 of Experiment A of the Example.

[0035] FIG. 3 is a graph showing the amount of FVIII antigen in the culture medium of the preadipocytes introduced with the five types of FVIII variant genes prepared.

[0036] FIG. 4 is a graph showing the FVIII activity of the culture medium of the preadipocytes introduced with the five types of FVIII variant genes prepared.

[0037] FIG. 5 is a graph showing the specific activity (activity per antigen amount) of the FVIII variants secreted into the culture medium from the preadipocytes introduced with the five types of FVIII variant genes prepared. 1 is a graph showing the amount of FVIII antigen in the blood of mice into which the prepared five types of FVIII variant gene-introduced preadipocytes were subcutaneously transplanted into the back in Experiment 4 of Experiment A of the Example. 2 is a graph showing the average introduced copy number of FVIII variant genes in the prepared seven types of FVIII variant gene-introduced preadipocytes in Experiment 2 of Experiment B of the Example. 3 is a graph showing the amount of FVIII antigen in the culture medium of the prepared seven types of FVIII variant gene-introduced preadipocytes. 4 is a graph showing the FVIII activity of the culture medium of the prepared seven types of FVIII variant gene-introduced preadipocytes. 5 is a graph showing the specific activity (activity per antigen amount) of the FVIII variants secreted into the culture medium from the prepared seven types of FVIII variant gene-introduced preadipocytes. 6 is the configuration of the vWF3-producing lentiviral vector plasmid in Experiment 1 of Experiment C of the Example. 1 is a graph showing the amount of FVIII antigen in the blood of mice that were mixed and transplanted subcutaneously into the back with the three types of human vWF fragment gene-introduced preadipocytes prepared in Experiment 2 of Experiment C of the Example, and N5Δfurin (P739T) gene-introduced preadipocytes in Experiment 3 of Experiment C of the Example, using atelocollagen as a scaffold material, and the amount of FVIII antigen in the blood of mice that were mixed and transplanted subcutaneously into the back with N5Δfurin (P739T) gene-introduced preadipocytes in Experiment 3 of Experiment C of the Example.1 is a graph showing the amount of FVIII antigen in the blood of mice that were mixed and transplanted subcutaneously into the dorsal region or inguinal region with vWF3 gene-introduced preadipocytes and N5Δfurin (P739T) gene-introduced preadipocytes in Experiment 4 of Example Experiment C. 2 is a graph showing the amount of FVIII antigen in the blood of mice that were mixed and transplanted into the inguinal region with vWF3 gene-introduced preadipocytes and N5Δfurin (P739T) gene-introduced preadipocytes at the ratio of the number of three types of transplanted cells in Experiment 5 of Example Experiment C.

[0010] The ex vivo gene therapy cells of the present invention produce a modified human FVIII consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence shown in SEQ ID NO: 1, in which the B domain (amino acid sequence of amino acids 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted.

[0011] The FVIII variant of the present invention is a variant of human FVIII (full-length FVIII consisting of 2332 amino acid residues) consisting of the amino acid sequence from 20 to 2351 in the amino acid sequence shown in SEQ ID NO: 1, in which the B domain (amino acid sequence from 760 to 1667) is modified and which has the following two structural features: (A) the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus; and (B) the function of the furin recognition site is lost.

[0012] A FVIII variant having the structural feature (A) is known from a document by Hongzhi Z. Miao et al. (Bioengineering of coagulation factor VIII for improved secretion. Blood. 2004; 103: 3412-3419). The B domain of human FVIII contains 19 Asn residues that are N-glycosylated. This document reports that when genes for FVIII variants in which the B domain has been modified to have an amino acid sequence in which the number of N-glycosylated Asn residues increases by one from the N-terminus of the B domain are introduced into COS-1 cells and the variants are produced in vitro, the secretion amount of the variants increases with the number of Asn residues up to a variant (N8) having an amino acid sequence with eight Asn residues (however, the secretion amount of a variant (N7) having an amino acid sequence with seven Asn residues is not reported).

[0013] FVIII variants having the structural feature (B) are known from a document by G. N. Nguyen et al. (Novel factor VIII variants with a modified furin cleavage site improve the efficacy of gene therapy for hemophilia A. Journal of Thrombosis and Haemostasis. 2017; 15: 110-121). In this paper, a FVIII variant (human FVIII-BDD) that retains only three amino acid residues counting from the N-terminus of the B domain of human FVIII and 11 amino acid residues counting from the C-terminus (neither of which has an Asn residue that is subject to N-glycosylation) was described. The furin recognition site (Arg1645-His1646-Gln1647-Arg1648 in the full-length form) consisting of four amino acid residues counting from the C-terminus of the B domain was mutated to eliminate its function, thereby preventing furin cleavage between Arg1648 at the C-terminus of the B domain and Glu1649 at the N-terminus of the a3 linker region (human FVIII-BDDΔfurin). The gene for this FVIII variant was introduced into hemophilia A / CD4 knockout mice using an adeno-associated virus vector, and the resulting FVIII was then purified in vivo. It has been reported that the secretion amount of human FVIII-BDDΔfurin produced in vivo is greater than that of human FVIII-BDD produced in vivo.

[0014] However, the document by Hongzhi Z. Miao et al. does not describe the use of a FVIII variant having structural feature (A) in ex vivo gene therapy, and the document by G. N. Nguyen et al. does not describe the use of a FVIII variant having structural feature (B) in ex vivo gene therapy.

[0015] Furthermore, a FVIII variant having both structural feature (A) and structural feature (B) is described in JP 2019-527541 A, which also mentions the use of a polynucleotide encoding the amino acid sequence in gene therapy, but does not provide any specific explanation or evidence.

[0016] In the present invention, the structural feature (A) "the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus" corresponds to variants N4, N5, and N6 described in the literature by Hongzhi Z. Miao et al.

[0017] In the present invention, the structural feature (B) "lack of function of the furin recognition site" is not particularly limited as long as it is an embodiment that prevents cleavage by furin between Arg1648 at the C-terminus of the B domain and Glu1649 at the N-terminus of the a3 linker region, and examples include an embodiment in which at least one of the amino acid residues that constitute the furin recognition site, which is made up of four amino acid residues counting from the C-terminus of the B domain, is deleted or substituted with one or more other amino acid residues.

[0018] A specific example of a FVIII variant of the present invention having both structural feature (A) and structural feature (B) is a protein consisting of the amino acid sequence from 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2, in which the number of Asn residues to be N-glycosylated, counting from the N-terminus, of structural feature (A) is four (abbreviation: N4Δfurin).

[0019] Another specific example is a protein (abbreviation: N5Δfurin) consisting of the amino acid sequence from amino acid 20 to amino acid 1652 in the amino acid sequence set forth in SEQ ID NO: 3, in which the number of Asn residues to be N-glycosylated, as counted from the N-terminus, is five, as shown in structural feature (A).

[0020] Another specific example is a protein (abbreviation: N6Δfurin) consisting of the amino acid sequence from amino acid 20 to amino acid 1670 in the amino acid sequence set forth in SEQ ID NO: 4, in which the number of Asn residues to be N-glycosylated, as counted from the N-terminus, is six, as shown in structural feature (A).

[0021] Furthermore, in the FVIII variant of the present invention, the Pro residue at position 758 in the amino acid sequence of SEQ ID NO: 1 may be substituted with another amino acid residue. The Pro residue at position 758 in the amino acid sequence of SEQ ID NO: 1 (Pro739 in the full-length form) is located at a thrombin recognition site for cleavage by thrombin between Arg740 at the C-terminus of the a2 linker region and Ser741 at the N-terminus of the B domain. Substitution of this Pro residue with a polar, uncharged amino acid residue such as Thr, Ser, Asn, or Gln can increase the amount secreted into the blood (increase the blood concentration) and the specific activity (activity per antigen amount).

[0022] A specific example of a FVIII variant having both structural feature (A) and structural feature (B) and in which the Pro residue at position 758 in the amino acid sequence shown in SEQ ID NO: 1 is substituted with a Thr residue is a protein consisting of the amino acid sequence from positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, in which the number of Asn residues to be N-glycosylated is five, counting from the N-terminus, according to structural feature (A) (abbreviation: N5Δfurin(P739T)). A specific example of a FVIII variant having both structural feature (A) and structural feature (B) and in which the Pro residue at position 758 in the amino acid sequence shown in SEQ ID NO: 1 is substituted with a Ser residue is a protein consisting of the amino acid sequence from positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, in which the number of Asn residues to be N-glycosylated is five, counting from the N-terminus, according to structural feature (A) (abbreviation: N5Δfurin(P739S)). A specific example of a FVIII variant having both structural feature (A) and structural feature (B) and in which the Pro residue at position 758 in the amino acid sequence shown in SEQ ID NO: 1 is substituted with an Asn residue is a protein consisting of the amino acid sequence from positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, in which the number of Asn residues to be N-glycosylated according to structural feature (A) is five, counting from the N-terminus (abbreviation: N5Δfurin(P739N)). A specific example of a FVIII variant having both structural feature (A) and structural feature (B) and in which the Pro residue at position 758 in the amino acid sequence shown in SEQ ID NO: 1 is substituted with a Gln residue is a protein consisting of the amino acid sequence from positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16, in which the number of Asn residues to be N-glycosylated according to structural feature (A) is five, counting from the N-terminus (abbreviation: N5Δfurin(P739Q)).

[0023] The FVIII variants of the present invention include the amino acid sequence of positions 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, the amino acid sequence of positions 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, and the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16, each of which has one or more amino acid residues deleted, substituted, or the like. These proteins include proteins that have additions or insertions and have biological activity equivalent to that of the proteins consisting of the amino acid sequence of amino acids 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, the amino acid sequence of amino acids 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, and the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16.and the amino acid sequence having at least 90% sequence identity with the amino acid sequence of amino acids 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, the amino acid sequence of amino acids 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, and the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16, The present invention also includes proteins having biological activity equivalent to that of a protein consisting of the amino acid sequence of amino acids 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, the amino acid sequence of amino acids 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, and the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16. Here, "plurality" refers to an integer of 50 or less, preferably an integer of 30 or less, more preferably an integer of 20 or less, and even more preferably an integer of 10 or less, for example, 2 to 9, 2 to 7, 2 to 5, or 2 to 3.The sequence identity with the amino acid sequence of positions 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, the amino acid sequence of positions 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, and the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16 is preferably at least 91%, more preferably at least 92%, more preferably at least 93%, more preferably at least 94%, more preferably at least 95%, more preferably at least 96%, more preferably at least 97%, more preferably at least 98%, and even more preferably at least 99%. The percent identity refers to a value calculated using software (e.g., FASTA, DANASYS, BLAST, etc.) that calculates identity between multiple (two) amino acid sequences with default settings. "Equivalent biological activity" refers to, for example, the expression of the function of FVIII as a blood coagulation factor, and biological activities such as activation and inactivation, which are of the same quality.

[0024] The cells used to produce the ex vivo gene therapy cells of the present invention are not particularly limited, as long as they are capable of producing the FVIII variant of the present invention in vivo when transplanted into the body of a subject of ex vivo gene therapy (expression is carried out as a protein bound to a signal peptide consisting of the amino acid sequence of amino acids 1 to 19 in the amino acid sequence set forth in SEQ ID NO: 1), and a specific example thereof is adipocytes. Adipocytes include mature adipocytes, which are spherical fat-storing cells containing oil droplets, as well as preadipocytes, which have the ability to differentiate into mature adipocytes. Adipocytes may be autologous or allogeneic. Cells for ex vivo gene therapy using adipocytes are described in detail in Japanese Patent No. 4879867, and the contents of Japanese Patent No. 4879867 can be referenced as appropriate in the present invention.

[0025] The cells for ex vivo gene therapy of the present invention can be produced by introducing an expression vector carrying a polynucleotide encoding at least the amino acid sequence of a human FVIII variant consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence shown in SEQ ID NO: 1, in which the B domain (amino acid sequence of amino acids 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted, into, for example, cells extracted from a subject of ex vivo gene therapy or a healthy individual, for example, adipocytes collected from adipose tissue by a method known per se.

[0026] Any known method may be used to introduce into adipocytes an expression vector carrying a polynucleotide encoding at least the amino acid sequence of a human FVIII variant consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence set forth in SEQ ID NO: 1, in which the B domain (amino acid sequence of amino acids 760 to 1667) is modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site is deleted. However, the expression vector is preferably a viral vector, which is excellent in gene transfer efficiency into cells, maintenance of intracellular replication, stability, expression efficiency, etc. As the viral vector, known vectors such as retroviral vectors, lentiviral vectors, adenoviral vectors, and adeno-associated viral vectors can be used. However, it is preferable to use a lentiviral vector, which is excellent in safety and enables long-term stable gene expression by efficiently incorporating a gene of interest into the host cell genome. Gene transfer into target cells using a lentiviral vector can be carried out by a method known per se, for example, by transfecting a lentiviral vector plasmid derived from human immunodeficiency virus type 1 (HIV-1) incorporating a gene of interest together with a packaging plasmid into packaging cells such as 293T cells or COS cells, and then infecting the target cells with the lentiviral vector carrying the gene of interest, obtained as a cell culture supernatant (protamine sulfate, polybrene, or the like may be added to increase infection efficiency). Tools for producing lentiviral vectors carrying a gene of interest have also been commercialized (for example, those available from Takara Bio, Funakoshi, and Cosmo Bio).

[0027] The lentiviral vector carrying a polynucleotide encoding at least the amino acid sequence of a human FVIII variant consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence set forth in SEQ ID NO: 1, in which the B domain (amino acid sequence of amino acids 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted can be obtained as a cell culture supernatant by transfecting, into packaging cells such as 293T cells or COS cells, a lentiviral vector plasmid incorporating, as a target gene, a polynucleotide encoding at least the amino acid sequence of a human FVIII variant consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence set forth in SEQ ID NO: 1, the B domain (amino acid sequence of amino acids 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted, together with a packaging plasmid.

[0028] The polynucleotide encoding at least the amino acid sequence of a human FVIII variant consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence shown in SEQ ID NO: 1, in which the B domain (amino acid sequence of amino acids 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted is, for example, a polynucleotide encoding the amino acid sequence of a protein in which the human FVIII variant consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence shown in SEQ ID NO: 1 has been modified, the B domain (amino acid sequence of amino acids 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted, is linked to a signal peptide consisting of the amino acid sequence of amino acids 1 to 19 in the amino acid sequence shown in SEQ ID NO: 1. A polynucleotide encoding the amino acid sequence of a protein in which a human FVIII variant consisting of the amino acid sequence of 20 to 2351 in the amino acid sequence shown in SEQ ID NO: 1, in which the B domain (amino acid sequence of 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted, is linked to a signal peptide consisting of the amino acid sequence of 1 to 19 in the amino acid sequence shown in SEQ ID NO: 1, for example. A modified human FVIII consisting of the amino acid sequence of amino acids 20 to 2351 in SEQ ID NO: 1, in which the B domain (amino acid sequence of amino acids 760 to 1667) is modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site is deleted, can be prepared by chemical synthesis based on the amino acid sequence of a protein bound to a signal peptide consisting of the amino acid sequence of amino acids 1 to 19 in the amino acid sequence of SEQ ID NO: 1 (the base sequence may be codon optimized as necessary).

[0029] Specific examples of polynucleotides encoding the amino acid sequence of a protein in which a variant of human FVIII consisting of the amino acid sequence at positions 20 to 2351 in the amino acid sequence shown in SEQ ID NO: 1 is linked to a signal peptide consisting of the amino acid sequence at positions 1 to 19 in the amino acid sequence shown in SEQ ID NO: 1, in which the B domain (amino acid sequence at positions 760 to 1667) is modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site is deleted. These polynucleotides include a polynucleotide consisting of the base sequence shown in SEQ ID NO: 6 (a polynucleotide encoding the amino acid sequence of a protein in which N4Δfurin consisting of the amino acid sequence at positions 1 to 1607 in the amino acid sequence shown in SEQ ID NO: 2 is linked to a signal peptide), a polynucleotide consisting of the base sequence shown in SEQ ID NO: 7 (a polynucleotide encoding the amino acid sequence of a protein in which N5Δfurin consisting of the amino acid sequence at positions 1 to 1652 in the amino acid sequence shown in SEQ ID NO: 3 is linked to a signal peptide), and a polynucleotide consisting of the base sequence shown in SEQ ID NO: 8 (a polynucleotide encoding the amino acid sequence of a protein in which N5Δfurin consisting of the amino acid sequence at positions 1 to 1652 in the amino acid sequence shown in SEQ ID NO: 9). a polynucleotide consisting of the base sequence (a polynucleotide encoding the amino acid sequence of a protein in which N6Δfurin consisting of the amino acid sequence of positions 1 to 1670 in the amino acid sequence shown in SEQ ID NO: 4 is bound to a signal peptide); a polynucleotide consisting of the base sequence shown in SEQ ID NO: 9 (a polynucleotide encoding the amino acid sequence of a protein in which N5Δfurin (P739T) consisting of the amino acid sequence of positions 1 to 1652 in the amino acid sequence shown in SEQ ID NO: 5 is bound to a signal peptide); a polynucleotide consisting of the base sequence shown in SEQ ID NO: 17 (a polynucleotide encoding the amino acid sequence of a protein in which N5Δfurin (P739S) consisting of the amino acid sequence of positions 1 to 1652 in the amino acid sequence shown in SEQ ID NO: 14 is bound to a signal peptide); a polynucleotide consisting of the base sequence shown in SEQ ID NO: 18 (a polynucleotide encoding the amino acid sequence of a protein in which N5Δfurin (P739N) consisting of the amino acid sequence of positions 1 to 1652 in the amino acid sequence shown in SEQ ID NO: 15 is bound to a signal peptide);Examples of such polynucleotides include a polynucleotide consisting of the base sequence shown in SEQ ID NO: 19 (a polynucleotide encoding the amino acid sequence of a protein in which N5Δfurin (P739Q), consisting of the amino acid sequence of amino acids 1 to 1652 in the amino acid sequence shown in SEQ ID NO: 16, is linked to a signal peptide). However, polynucleotides encoding the amino acid sequence of a protein in which a variant of human FVIII consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence shown in SEQ ID NO: 1, in which the B domain (amino acid sequence of amino acids 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted, is not limited to these polynucleotides, and include polynucleotides that hybridize to the complementary strands of these polynucleotides under stringent conditions and have the sequence The polynucleotides include those encoding at least a protein having biological activity equivalent to that of a protein consisting of the amino acid sequence of positions 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, the amino acid sequence of positions 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, and the amino acid sequence of positions 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16. Furthermore, a polynucleotide having at least 90%, preferably at least 91%, more preferably at least 92%, more preferably at least 93%, more preferably at least 94%, more preferably at least 95%, more preferably at least 96%, more preferably at least 97%, more preferably at least 98%, and even more preferably at least 99% sequence identity with each of the base sequences shown in SEQ ID NOs: 6 to 9 and 17 to 19, and further comprising the amino acid sequence of positions 20 to 1607 in the amino acid sequence shown in SEQ ID NO: 2:The present invention includes polynucleotides encoding at least a protein having biological activity equivalent to that of a protein consisting of the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, the amino acid sequence of amino acids 20 to 1670 in the amino acid sequence shown in SEQ ID NO: 4, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 5, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14, the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 15, and the amino acid sequence of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16. Here, "hybridization under stringent conditions" includes, for example, hybridization at 30 to 50°C in 3 to 4×SSC (150 mM sodium chloride, 15 mM sodium citrate, pH 7.2), 0.1 to 0.5% SDS for 1 to 24 hours, preferably hybridization at 40 to 45°C in 3.4×SSC, 0.3% SDS for 1 to 24 hours, followed by washing. Examples of washing conditions include successive washing at room temperature with a solution containing 2xSSC and 0.1% SDS, and with a 1xSSC solution and a 0.2xSSC solution. However, the above combination of conditions is merely an example, and a person skilled in the art would be able to achieve stringency similar to that described above by appropriately combining the above factors that determine hybridization stringency and other factors (e.g., concentration, length, and GC content of the hybridization probe, hybridization reaction time, etc.).

[0030] The cells for ex vivo gene therapy of the present invention, when transplanted into the body of a subject for ex vivo gene therapy for hemophilia A, produce in vivo a modified human FVIII consisting of the amino acid sequence of amino acids 20 to 2351 in the amino acid sequence set forth in SEQ ID NO: 1, in which the B domain (amino acid sequence of amino acids 760 to 1667) has been modified, the number of Asn residues to be N-glycosylated is 4 to 6 counting from the N-terminus, and the function of the furin recognition site has been deleted, and therefore are useful for ex vivo gene therapy for hemophilia A. The number of cells for ex vivo gene therapy of the present invention to be transplanted is determined taking into consideration the age, weight, sex, condition and severity of the disease of the subject for ex vivo gene therapy, etc.

[0031] When the cells for ex vivo gene therapy of the present invention are produced using adipocytes, they can be transplanted by injection into subcutaneous tissue or adipose tissue using a method known per se. In this case, the cells for ex vivo gene therapy of the present invention are preferably transplanted as a transplant composition together with a pharmaceutically acceptable carrier such as physiological saline, Ringer's solution, phosphate buffer, culture medium, serum, or body fluid. The transplant composition may further contain various extracellular matrix components (e.g., collagen, laminin, heparan sulfate, gelatin) as a scaffold material. The transplant composition may also contain angiogenic factors (whether naturally occurring or produced by genetic engineering techniques) such as vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), acidic fibroblast growth factor (aFGF), platelet-derived growth factor (PDGF), or hepatocyte growth factor (HGF) to promote the formation of blood vessels around the transplanted cells and the efficient secretion of the produced FVIII variants into the bloodstream after transplantation.

[0032] Furthermore, the cells for ex vivo gene therapy of the present invention may be transplanted into the body of a subject of ex vivo gene therapy together with cells that in vivo produce human von Willebrand factor (hereinafter referred to as "vWF") or a fragment thereof (e.g., a fragment containing at least the D' domain and D3 domain, which are binding sites for FVIII), which contribute to the stabilization and transport of FVIII in circulating blood. Interestingly, by transplanting the cells for ex vivo gene therapy of the present invention together with cells that in vivo produce human vWF or a fragment thereof into the body of a subject of ex vivo gene therapy, the cells that in vivo produce human vWF or a fragment thereof have the effect of increasing the amount of FVIII variants secreted into the blood (increasing their blood concentration) produced by the cells for ex vivo gene therapy of the present invention.

[0033] Human vWF is a glycoprotein expressed in vascular endothelial cells and bone marrow megakaryocytes as a precursor (the amino acid sequence of which is shown in SEQ ID NO:28) consisting of 2,813 amino acid residues, including a signal peptide consisting of 22 amino acid residues. The precursor has a domain structure of SP-D1-D2-D'-D3-A1-A2-A3-D4-B1-B2-B3-C1-C2-CK, and upon cleavage of the signal peptide (SP), pro-vWF is generated, which consists of a propeptide (D1-D2) consisting of 741 amino acid residues and a mature subunit (D'-D3-A1-A2-A3-D4-B1-B2-B3-C1-C2-CK) consisting of 2,050 amino acid residues. Pro-vWF forms a dimer via disulfide bonds at the C-terminus, and then undergoes multimer formation by disulfide bonds between dimers at the N-terminus, glycosylation, and cleavage of the propeptide, thereby exerting its desired function.

[0034] Examples of cells that produce in vivo a fragment of human vWF containing at least the D' and D3 domains, which are binding sites for FVIII, and that may be transplanted into the body of a subject of ex vivo gene therapy together with the cells for ex vivo gene therapy of the present invention include cells that produce in vivo a fragment that has the SP, D', and D3 domains but does not have the A2, A3, D4, B1, B2, B3, C1, or C2 domains of the domain structure SP-D1-D2-D'-D3-A1-A2-A3-D4-B1-B2-B3-C1-C2-CK that the precursor has, and that may have up to 50 amino acid residues on the N-terminal side of the A1 domain, and that may or may not have the D1, D2, or CK domains. Here, the amino acid sequences of the SP, D1, D2, D', D3, A1, and CK domains are the amino acid sequence of amino acids 1 to 22, the amino acid sequence of amino acids 23 to 386, the amino acid sequence of amino acids 387 to 763, the amino acid sequence of amino acids 764 to 864, the amino acid sequence of amino acids 865 to 1242, the amino acid sequence of amino acids 1243 to 1479, and the amino acid sequence of amino acids 2664 to 2813, respectively, in the amino acid sequence of the precursor shown in SEQ ID NO: 28.

[0035] A specific example of a fragment having the SP, D', and D3 domains but not the A2, A3, D4, B1, B2, B3, C1, or C2 domains of the SP-D1-D2-D'-D3-A1-A2-A3-D4-B1-B2-B3-C1-C2-CK domain structure of the precursor, wherein the A1 domain may have up to 50 amino acid residues on the N-terminal side, and the D1, D2, and CK domains may or may not be present, is a fragment consisting of the SP, D1, D2, D', and D3 domains, 26 amino acid residues counting from the N-terminal side of the A1 domain, a further 3 amino acid residues, and the CK domain (a protein consisting of the amino acid sequence set forth in SEQ ID NO: 29, abbreviated as vWF1). The "26 amino acid residues counting from the N-terminus of the A1 domain" correspond to the amino acid residues constituting the amino acid sequence of positions 1243 to 1268 in the amino acid sequence of the precursor shown in SEQ ID NO: 28, and include multiple amino acid residues (serine and threonine) to which O-glycans are bound. The "further 3 amino acid residues" correspond to the amino acid residues of positions 1269, 1270, and 2663 in the amino acid sequence of the precursor shown in SEQ ID NO: 28. In other words, the amino acid sequence of vWF1 is a combination of the amino acid sequence of positions 1 to 1270 and the amino acid sequence of positions 2663 to 2813 in the amino acid sequence of the precursor shown in SEQ ID NO: 28. Furthermore, of the domain structure of SP-D1-D2-D'-D3-A1-A2-A3-D4-B1-B2-B3-C1-C2-CK possessed by the precursor, a fragment having the SP, D', and D3 domains but not the A2, A3, D4, B1, B2, B3, C1, and C2 domains, wherein the A1 domain may have up to 50 amino acid residues on the N-terminal side, and which may or may not have the D1, D2, and CK domains, may be a fragment consisting of the SP, D1, D2, D', and D3 domains and 26 amino acid residues counting from the N-terminal side of the A1 domain (a protein consisting of the amino acid sequence shown in SEQ ID NO: 30, abbreviated as vWF2), or a fragment consisting of the SP, D', and D3 domains and 26 amino acid residues counting from the N-terminal side of the A1 domain (a protein consisting of the amino acid sequence shown in SEQ ID NO: 31, abbreviated as vWF3).

[0036] Fragments of human vWF include, for example, proteins having one or more deletions, substitutions, additions, or insertions of amino acid residues in the amino acid sequence shown in SEQ ID NO: 29, the amino acid sequence shown in SEQ ID NO: 30, or the amino acid sequence shown in SEQ ID NO: 31, and having biological activity equivalent to that of proteins consisting of the amino acid sequences shown in SEQ ID NO: 29, the amino acid sequence shown in SEQ ID NO: 30, or the amino acid sequence shown in SEQ ID NO: 31. Fragments of human vWF also include proteins having at least 90% sequence identity with the amino acid sequence shown in SEQ ID NO: 29, the amino acid sequence shown in SEQ ID NO: 30, or the amino acid sequence shown in SEQ ID NO: 31, and having biological activity equivalent to that of proteins consisting of the amino acid sequence shown in SEQ ID NO: 29, the amino acid sequence shown in SEQ ID NO: 30, or the amino acid sequence shown in SEQ ID NO: 31. Here, "multiple" refers to an integer of 50 or less, preferably an integer of 30 or less, more preferably an integer of 20 or less, and even more preferably an integer of 10 or less, for example, 2 to 9, 2 to 7, 2 to 5, or 2 to 3. The sequence identity with the amino acid sequence set forth in SEQ ID NO:29, SEQ ID NO:30, and SEQ ID NO:31 is preferably at least 91%, more preferably at least 92%, more preferably at least 93%, more preferably at least 94%, more preferably at least 95%, more preferably at least 96%, more preferably at least 97%, more preferably at least 98%, and even more preferably at least 99%. The percent identity refers to a value calculated using software (e.g., FASTA, DANASYS, BLAST, etc.) that calculates identity between multiple (two) amino acid sequences with default settings. "Equivalent biological activity" refers to biological activity, such as the function of vWF (e.g., stabilization and transport of FVIII in circulating blood), being of the same quality.

[0037] Cells that produce human vWF or a fragment thereof in vivo, which may be transplanted into the body of a subject of ex vivo gene therapy together with the cells for ex vivo gene therapy of the present invention, can be produced, for example, by introducing an expression vector carrying a polynucleotide that encodes at least the amino acid sequence of human vWF or a fragment thereof into cells extracted from a subject of ex vivo gene therapy or a healthy individual, for example, adipocytes collected from adipose tissue by a method known per se, in accordance with the method for producing the cells for ex vivo gene therapy of the present invention described above.

[0038] Specific examples of polynucleotides encoding the amino acid sequence of a human vWF fragment include a polynucleotide consisting of the nucleotide sequence shown in SEQ ID NO: 32 (a polynucleotide encoding the amino acid sequence shown in SEQ ID NO: 29), a polynucleotide consisting of the nucleotide sequence shown in SEQ ID NO: 33 (a polynucleotide encoding the amino acid sequence shown in SEQ ID NO: 30), and a polynucleotide consisting of the nucleotide sequence shown in SEQ ID NO: 34 (a polynucleotide encoding the amino acid sequence shown in SEQ ID NO: 31). However, polynucleotides encoding the amino acid sequence of a human vWF fragment are not limited to these polynucleotides, and include polynucleotides that hybridize to the complementary strands of these polynucleotides under stringent conditions and encode at least proteins having biological activity equivalent to that of the amino acid sequence shown in SEQ ID NO: 29, the amino acid sequence shown in SEQ ID NO: 30, or the amino acid sequence shown in SEQ ID NO: 31, or to proteins having the respective amino acid sequences. Furthermore, the present invention also includes polynucleotides having at least 90%, preferably at least 91%, more preferably at least 92%, more preferably at least 93%, more preferably at least 94%, more preferably at least 95%, more preferably at least 96%, more preferably at least 97%, more preferably at least 98%, and even more preferably at least 99% sequence identity to the nucleotide sequences shown in SEQ ID NOs: 32 to 34, respectively, and encoding at least a protein having biological activity equivalent to the amino acid sequence shown in SEQ ID NO: 29, the amino acid sequence shown in SEQ ID NO: 30, or the amino acid sequence shown in SEQ ID NO: 31. Here, "hybridization under stringent conditions" includes, for example, hybridization at 30 to 50°C in 3 to 4 x SSC (150 mM sodium chloride, 15 mM sodium citrate, pH 7.2), 0.1 to 0.5% SDS for 1 to 24 hours, or preferably hybridization at 40 to 45°C in 3.4 x SSC, 0.3% SDS for 1 to 24 hours, followed by washing.Examples of washing conditions include successive washing at room temperature with a solution containing 2xSSC and 0.1% SDS, and with a 1xSSC solution and a 0.2xSSC solution, etc. However, the above combination of conditions is merely an example, and a person skilled in the art would be able to achieve a stringency similar to that described above by appropriately combining the above factors that determine hybridization stringency and other factors (e.g., concentration, length, and GC content of the hybridization probe, hybridization reaction time, etc.).

[0039] Examples of methods for transplanting cells that in vivo produce human vWF or a fragment thereof together with the ex vivo gene therapy cells of the present invention into the body of a subject for ex vivo gene therapy include a method in which a transplant composition containing the ex vivo gene therapy cells of the present invention and a pharmaceutically acceptable carrier is further supplemented with cells that in vivo produce human vWF or a fragment thereof and transplanted. However, a transplant composition containing cells that in vivo produce human vWF or a fragment thereof and a pharmaceutically acceptable carrier may be prepared separately from the transplant composition containing the ex vivo gene therapy cells of the present invention and a pharmaceutically acceptable carrier, and the two transplant compositions may be transplanted as a set simultaneously or at intervals. In this case, the pharmaceutically acceptable carrier in the transplant composition containing cells that in vivo produce human vWF or a fragment thereof and a pharmaceutically acceptable carrier may be the same as the pharmaceutically acceptable carrier in the transplant composition containing the ex vivo gene therapy cells of the present invention and a pharmaceutically acceptable carrier. Furthermore, a transplant composition comprising cells that produce human vWF or a fragment thereof in vivo and a pharmaceutically acceptable carrier can also contain various extracellular matrix components, angiogenic factors, etc., similar to the transplant composition comprising cells for ex vivo gene therapy of the present invention and a pharmaceutically acceptable carrier.

[0040] When cells that produce human vWF or a fragment thereof in vivo are transplanted into the body of a subject for ex vivo gene therapy together with the cells for ex vivo gene therapy of the present invention, the ratio of the number of transplanted cells (cells for ex vivo gene therapy of the present invention and cells that produce human vWF or a fragment thereof in vivo) is determined, for example, from the range of 1:100 to 100:1, taking into consideration the number of copies of the introduced gene in each cell.

[0041] The present invention will be described in detail below with reference to examples, but the present invention should not be construed as being limited to the following description.

[0042] Experiment A: Study on preadipocytes transfected with various FVIII variant genes (Part 1) Experiment 1: Preparation of preadipocytes transfected with FVIII variant genes (1) Preparation of lentiviral vector plasmids producing FVIII variants Polynucleotides were synthesized to which SbfI and MluI restriction enzyme sites were added at the 5' end and the 3' end, respectively, of the nucleotide sequences corresponding to the amino acid sequences of proteins in which a signal peptide was bound to each of the five FVIII variants listed below. These polynucleotides were then inserted into the SbfI / MluI site of the SIN-type lentiviral vector pLVSIN-EF1α Neo Vector (Takara Bio Inc.; the same applies hereinafter) to prepare lentiviral vector plasmids producing FVIII variants. Each nucleotide sequence was codon-optimized based on the corresponding amino acid sequence using GeneArt GeneOptimizer software (Thermo Fisher Scientific).(a) A nucleotide sequence (SEQ ID NO: 11) corresponding to the amino acid sequence (SEQ ID NO: 10) of a protein in which a signal peptide is bound to NovoEight (Novo Nordisk Pharma K.K., Eur J Haematol. 2014 Nov; 93(5): 369-376) (ΔB), a recombinant B-domain deleted human FVIII preparation; (b) A nucleotide sequence (SEQ ID NO: 13) corresponding to the amino acid sequence (SEQ ID NO: 12) of a protein in which a signal peptide is bound to a FVIII variant (N6) in which the number of Asn residues to be N-glycosylated in the B-domain of human FVIII is set to 6 from the N-terminus and the function of the furin recognition site is maintained. (c) A nucleotide sequence (SEQ ID NO: 8) corresponding to the amino acid sequence (SEQ ID NO: 4) of a protein in which a signal peptide is bound to a FVIII variant (N6Δfurin) in which the number of Asn residues to be N-glycosylated in the B domain of human FVIII is 6, counting from the N-terminus, and the function of the furin recognition site is deleted. (d) A nucleotide sequence (SEQ ID NO: 7) corresponding to the amino acid sequence (SEQ ID NO: 3) of a protein in which a signal peptide is bound to a FVIII variant (N5Δfurin) in which the number of Asn residues to be N-glycosylated in the B domain of human FVIII is 5, counting from the N-terminus, and the function of the furin recognition site is deleted. (e) A nucleotide sequence (SEQ ID NO: 9) corresponding to the amino acid sequence (SEQ ID NO: 5) of a protein in which a signal peptide is bound to a FVIII variant (N5Δfurin(P739T)) in which the number of Asn residues N-glycosylated in the B domain of human FVIII, counting from the N-terminus, is set to 5, the function of the furin recognition site is deleted, and the 758th Pro residue is replaced with a Thr residue. The differences in the amino acid sequences of the B domain and those before and after it (positions relative to the full-length FVIII) of the above five FVIII variants are shown in Figure 1. The structure of the lentiviral vector plasmid producing N5Δfurin(P739T) is shown in Figure 2.

[0043] (2) Preparation of lentiviral vector producing FVIII variant The lentiviral vector plasmid (5.5 μg) producing the FVIII variant prepared in (1), serum-free DMEM (1.5 mL), Lentiviral High Titer Packaging Mix (Takara Bio Inc., 7 μL), and PEIpro transfection reagent (Avantor, 45 μL) were mixed by gentle pipetting and allowed to stand at room temperature for 15 minutes to prepare a transfection solution. Approximately 24 hours before transfection, 1 × 10 ７ The transfection solution was added dropwise to Lenti-X293T cells (Takara Bio Inc.) seeded and cultured in 10 mL of 10% FBS-containing DMEM in a 10 cm dish. The cells were then cultured. Approximately 24 hours after transfection, the medium was replaced with 10 mL of fresh 10% FBS-containing DMEM, and the cells were further cultured. The cells were cultured at 37°C and 5% CO. ２ Approximately 48 hours after transfection, the virus-containing culture medium was collected and centrifuged and filtered to obtain the FVIII variant-producing lentiviral vector as the supernatant.

[0044] (3) Obtaining human preadipocytes transfected with FVIII variant genes Adipose tissue obtained from healthy volunteers was dispensed into 50 mL centrifuge tubes in 1 g aliquots, and 3 mL of 2 mg / mL collagenase-containing Hank's balanced salt solution (Sigma) was added, followed by shaking for 1 hour at 37 ° C. After 1 hour, 10 mL of 20% FBS and 40 μg / mL gentamicin-containing Dulbecco's Modified Eagle's Medium / Nutrient Mixture F-12 Ham (DMEM-HAM / 20% FBS, Sigma; the same applies below) was added and stirred, followed by centrifugation at 400 g for 1 minute, and the solution containing the precipitate fraction was removed. Further, 10 mL of DMEM-HAM / 20% FBS was added and stirred, followed by centrifugation at 400 g for 1 minute. This series of operations to remove the solution containing the precipitate fraction was repeated twice to obtain a supernatant fraction containing preadipocytes. The supernatant fraction containing preadipocytes thus obtained was filtered through a 500 μm mesh into a new 50 mL centrifuge tube, and then 10 mL of DMEM-HAM / 20% FBS was added to the original 50 mL centrifuge tube for washing, and the washing solution was also filtered. The filtrate thus obtained was added to a T150 flask filled with DMEM-HAM / 20% FBS that had been warmed to 37°C in advance, and further DMEM-HAM / 20% FBS was added so as to minimize the amount of air bubbles remaining in the flask. The flask was then placed so that the bottom of the flask was the culture surface (ceiling surface), and the flask was incubated at 37°C and 5% CO ２ The cells were cultured in an incubator for 7 days. After 7 days, the culture medium was removed from the flask, the culture surface was washed with PBS, and the preadipocytes were recovered by trypsinization. The preadipocytes were centrifuged and suspended in a cell cryopreservation medium (CELLBANKER1: Takara Bio Inc., the same applies hereinafter) and frozen.

[0045] (4) Preparation of FVIII mutant gene-introduced preadipocytes The preadipocytes frozen in (3) were thawed, suspended in MSF-BM / MSF-Supplement A (Shimadzu Diagnostics, the same applies hereinafter), and centrifuged to remove the cell cryopreservation solution. The cells were then resuspended in MSF-BM / MSF-Supplement A, placed in a T225 flask, and incubated at 37°C, 5% CO using MSF-BM / MSF-Supplement A. ２The cells were cultured in an incubator for 4 days. The next day, the culture medium in the flask was removed, the culture surface was washed with PBS, and the preadipocytes were recovered by trypsinization and centrifuged. The centrifuged preadipocytes were suspended in MSF-BM / MSF-Supplement A, and the cell concentration was measured to be 9.5 × 10 ５ The cells were separated and centrifuged, and then suspended in a solution prepared by mixing the supernatant (FVIII variant-producing lentiviral vector) obtained in (2) with protamine sulfate to a concentration of 100 μg / mL, to give a final concentration of 4.75 × 10 ４ The transfected cell suspension was seeded at 2 mL / well on a 6-well plate, and from the next day, the cells were incubated at 37°C, 5% CO using MSF-BM / MSF-Supplement A. ２ The cells were expanded in an incubator with passage every 3 or 4 days to prepare preadipocytes transfected with the FVIII variant gene.

[0046] Experiment 2: Evaluation of the prepared FVIII variant gene-introduced preadipocytes For each of the five types of FVIII variant gene-introduced preadipocytes prepared in Experiment 1, at the time of passage on days 10 to 14 after seeding the cell suspension onto a 6-well plate in Experiment 1(4), the culture medium in the flask was removed, the culture surface was washed with PBS, and a portion of the cells recovered by trypsinization was centrifuged, suspended in a cell cryopreservation solution, and frozen. The frozen cells were thawed, suspended in MSF-BM / MSF-Supplement A, and centrifuged to remove the cell cryopreservation solution. The cells were then resuspended in MSF-BM / MSF-Supplement A, placed in a T75 flask, and incubated at 37°C, 5% CO using MSF-BM / MSF-Supplement A. ２ The cells were cultured in an incubator, with passage every 3 or 4 days. Passage was performed on days 8, 11, and 14 after the start of culture. At the time of passage on days 11 and 14, the culture medium in the flask was removed, the culture surface was washed with PBS, and the cells were recovered by trypsinization. The cells were then centrifuged, suspended in MSF-BM / MSF-Supplement A, and the cell concentration was measured. The cells were then collected at 2 × 10 ５The cells were collected, centrifuged, suspended in PBS, frozen, and stored until use in evaluation. A portion of the culture medium removed from the flask was also frozen and stored until use in evaluation.

[0047] (1) Calculation of the Average Introduced Copy Number of FVIII Variant Genes (Calculation Method) Frozen cells were thawed, and DNA was extracted using the DNeasy Blood & Tissue Kit (QIAGEN) according to the protocol. Specifically, 20 μL of Proteinase K and 4 μL of RNase A (QIAGEN) were added to the thawed cells, mixed, and left at room temperature for 5 minutes. Then, 0.2 mL of Buffer AL was added, mixed, and heated at 56°C for 10 minutes. After 10 minutes, 0.2 mL of ethanol was added, mixed, placed in a spin column, placed in a 2 mL collection tube, and centrifuged at 6,000 g for 1 minute. The centrifuged spin column was placed in a new 2 mL collection tube, and 0.5 mL of Buffer AW1 was added. The column was then centrifuged at 6,000 g for 1 minute. The centrifuged spin column was then placed in a new 2 mL collection tube, 0.5 mL of Buffer AW2 was added, and the column was centrifuged at 20,000 g for 3 minutes. The centrifuged spin column was then placed in a new 1.5 mL collection tube, 0.2 mL of Buffer AE was added, and the column was left at room temperature for 1 minute, followed by centrifugation at 6,000 g for 1 minute to elute DNA, whose concentration was measured using an absorbance meter. The DNA extracted from the cells was then used to calculate the average introduced copy number of the FVIII variant gene using a Bio-Rad QX200 Droplet Digital PCR system, a C1000 Touch thermal cycler, a PX1 PCR Plate Sealer, and QuantaSoft software according to the manufacturer's instructions.

[0048] (Calculation Results) The average introduced copy numbers of FVIII variant genes calculated from DNA extracted from cells recovered at each passage on days 11 and 14 after the start of culture in a T75 flask were added together and divided by 2, and the resulting value is shown in Figure 3. As is clear from Figure 3, the average introduced copy numbers were similar and not significantly different between gene-introduced preadipocytes producing ΔB, N6, N6Δfurin, N5Δfurin, or N5Δfurin (P739T).

[0049] (2) Measurement of FVIII antigen amount in culture medium The frozen culture medium was thawed, and the FVIII antigen amount in the culture medium was measured using an ELISA kit (Asserachrom FVIII, Roche; the same applies hereinafter) according to the protocol. The FVIII antigen amount was quantified as a relative value (% value when the FVIII antigen amount in healthy subject plasma was set as 100%) based on the absorbance at 450 nm (subwavelength 570 nm) of a plate reader, using a calibration curve prepared for the FVIII antigen amount in healthy subject plasma. The FVIII antigen amount in the culture medium collected at each passage on days 11 and 14 after the start of culture in a T75 flask was divided by 3 (the number of days of passage), and the sum was divided by 2. The values ​​are shown in Figure 4. As is clear from FIG. 4, the amount of FVIII antigen in the culture medium showed a very high value for ΔB, but the values ​​for the other FVIII variants were half or less than the value for ΔB, with no significant difference.

[0050] (3) Measurement of FVIII activity of culture medium The frozen culture medium was thawed, and the FVIII activity of the culture medium was measured using a Coatest SP FVIII kit (Chromogenix) according to the protocol. The FVIII activity was quantified as a relative value (% value when the FVIII activity of healthy human plasma was set as 100%) based on the absorbance at 450 nm (subwavelength 570 nm) using a plate reader, based on a calibration curve prepared for the FVIII activity of healthy human plasma. The FVIII activity of the culture medium collected at each passage on days 11 and 14 after the start of culture in a T75 flask was divided by 3 (the number of days of passage), and the sum was divided by 2. The resulting values ​​are shown in Figure 5. As is clear from Figure 5, the FVIII activity of the culture medium was highest for ΔB, followed by N5Δfurin and then N5Δfurin(P739T). The values ​​for N6 and N6Δfurin were less than half of that of ΔB.

[0051] (4) Calculation of Specific Activity (Activity per Antigen Amount) of FVIII Variants Secreted into Culture Medium The specific activity of the FVIII variants secreted into the culture medium was calculated from (2) the results of measuring the amount of FVIII antigen in the culture medium and (3) the results of measuring the FVIII activity of the culture medium. The FVIII activity of the culture medium collected at passage on day 11 after the start of culture in a T75 flask was divided by 3, then divided by the amount of FVIII antigen in the culture medium divided by 3. The sum of the two values ​​obtained by dividing the FVIII activity of the culture medium collected at passage on day 14 by 3 and then divided by the amount of FVIII antigen in the culture medium divided by 3 is shown in Figure 6. As is clear from Figure 6, the specific activity of the FVIII variants secreted into the culture medium was highest for N5Δfurin (P739T), which was significantly higher than those for ΔB and N6 (significant difference). The next highest value was N5Δfurin, followed by N6Δfurin, all of which were higher than the values ​​for ΔB and N6, although not significantly different from each other. Therefore, it was found that these FVIII variants secreted into the culture medium from gene-transfected preadipocytes producing N6Δfurin, N5Δfurin, and N5Δfurin (P739T) have high specific activities.

[0052] Experiment 3: Transplantation of FVIII variant gene-introduced preadipocytes into NOD / SCID mice Each of the five types of FVIII variant gene-introduced preadipocytes prepared in Experiment 1 was suspended in MSF-BM / MSF-Supplement A and randomly placed in a flask at 4 × 10 ４ cells / cm ２ The cells were seeded with MSF-BM / MSF-Supplement A and incubated at 37°C, 5% CO ２ The cells were cultured in an incubator. The next day, the medium was replaced with PGM-2 (Lonza, the same applies hereinafter), and the cells were cultured for an additional 3 days. The next day, the culture medium in the flask was removed, the culture surface was washed with PBS, and the cells were then recovered by trypsinization and spun down. The spun down cells were suspended in Ringer's solution containing albumin at a final concentration of 0.5% and spun down four times. This series of steps was then repeated four times to wash the cells, and the washed cells were then suspended in Fiblast Spray (Kaken Pharmaceutical, the same applies hereinafter). This prepared 200 μg / mL trafermin-containing cell suspension was mixed with an equal amount of Matrigel (Corning, the same applies hereinafter) to obtain a final concentration of 2 × 10 ７ A cell suspension for transplantation of 0.5 mL / mL (final concentration of trafermin: 100 μg / mL) was prepared. The prepared cell suspension for transplantation was administered subcutaneously to the left and right dorsal skin of 7- or 8-week-old NOD / SCID mice (Jackson Laboratory Japan) under isoflurane anesthesia using a 20G blunt needle cannula at a rate of 0.5 mL / site (2 × 10 ７ cells / animal).

[0053] Experiment 4: Evaluation of FVIII variant gene-transfected preadipocytes as cells for ex vivo gene therapy producing FVIII variants (measurement of FVIII antigen levels in the blood of transplanted mice) (Measurement method) Blood was collected from the cheek of each mouse under isoflurane anesthesia using an animal lancet (MEDIpoint, Inc.; the same applies hereinafter) on days 3, 7, and 14 after transplantation of each FVIII variant gene-transfected preadipocyte into NOD / SCID mice in Experiment 3. The collected blood was placed in a 1.5 mL centrifuge tube, and EDTA was added to a final concentration of 0.1-0.25% (v / v). Plasma samples were obtained by centrifugation at 5,000 rpm at 4°C for 10 minutes and frozen until use in evaluation. The frozen plasma samples were thawed, and the FVIII antigen levels in the blood of the transplanted mice were measured using an ELISA kit according to the protocol. The amount of FVIII antigen was quantified as a relative value (a percentage value when the amount of FVIII antigen in healthy subject plasma was set to 100%) based on a calibration curve prepared for the amount of FVIII antigen in healthy subject plasma, based on the absorbance at 450 nm (secondary wavelength 570 nm) on a plate reader.

[0054] (Measurement results) are shown in Figure 7. As is clear from Figure 7, the amount of FVIII antigen in the blood of the transplanted mice was highest for N5Δfurin (P739T), which was much higher than the amounts for ΔB and N6 (significant difference). The next highest value was for N5Δfurin, followed by the value for N6Δfurin, all of which were higher than the values ​​for ΔB and N6 (the former was significantly different from the value for ΔB). Therefore, it was found that gene-transfected preadipocytes producing N6Δfurin, N5Δfurin, and N5Δfurin (P739T) secreted greater amounts of FVIII variants into the blood than gene-transfected preadipocytes producing ΔB and N6.

[0055] Experiment B: Study on preadipocytes transfected with various FVIII variant genes (part 2) Experiment 1: Preparation of preadipocytes transfected with FVIII variant genes (1) Preparation of lentiviral vector plasmid producing FVIII variants (f) A polynucleotide consisting of a nucleotide sequence (SEQ ID NO: 17) corresponding to the amino acid sequence (SEQ ID NO: 14) of a protein in which a signal peptide is bound to a FVIII variant (N5Δfurin(P739S)) in which the number of N-glycosylated Asn residues in the B domain of human FVIII, counting from the N-terminus, is set to 5, the function of the furin recognition site is deleted, and the 758th Pro residue is replaced with a Ser residue, was inserted into the SbfI / MluI site of pLVSIN-EF1α Neo Vector to prepare a lentiviral vector plasmid producing N5Δfurin(P739S) by the following method. The N5Δfurin(P739T)-producing lentiviral vector plasmid prepared in Experiment A, Experiment 1 (1) (e) was subjected to two PCR reactions using the following two primer sets (Fw1&Re1, Fw2&Re2). Using a polynucleotide consisting of a base sequence corresponding to the amino acid sequence of the protein in which a signal peptide is bound to N5Δfurin(P739T) inserted into the plasmid as a template, two polynucleotide fragments were obtained: one with an XbaI restriction enzyme site at the 5' end and a P739S mutation at the 3' end, and the other with a P739S mutation at the 5' end and an MluI restriction enzyme site at the 3' end. The two polynucleotide fragments thus obtained and a polynucleotide linearized by treating the N5Δfurin (P739T)-producing lentiviral vector plasmid with restriction enzymes XbaI and MluI were ligated by in-fusion reaction to obtain the desired N5Δfurin (P739S)-producing lentiviral vector plasmid. Fw1: GTGCGGCCCCCTGTATTCTAG (SEQ ID NO: 20) Re1: CTTCTGCTCTCGATGGCATTG (SEQ ID NO: 21) Fw2: CATCGAGAGCAGAAGCTTCAGC (SEQ ID NO: 22) Re2: GTTGATTGTTCCAGACGCGTTC (SEQ ID NO: 23)

[0056] (g) A polynucleotide consisting of a nucleotide sequence (SEQ ID NO: 18) corresponding to the amino acid sequence (SEQ ID NO: 15) of a protein in which a signal peptide is linked to a FVIII variant (N5Δfurin(P739N)) in which the number of Asn residues to be N-glycosylated in the B domain of human FVIII, counting from the N-terminus, is set to 5, the function of the furin recognition site is deleted, and the 758th Pro residue is substituted with an Asn residue, was inserted into the SbfI / MluI site of pLVSIN-EF1α Neo Vector to prepare an N5Δfurin(P739N)-producing lentiviral vector plasmid in the same manner as the method for preparing the N5Δfurin(P739S)-producing lentiviral vector plasmid. The two primer sets (Fw1 & Re1, Fw2 & Re2) used to subject the N5Δfurin (P739T)-producing lentiviral vector plasmid to two PCR reactions were as follows: Fw1: GTGCGGCCCCCTGTATTCTAG (SEQ ID NO: 20) Re1: CTTCTGTTCTCGATGGCATTG (SEQ ID NO: 24) Fw2: CATCGAGAACAGAAGCTTCAGC (SEQ ID NO: 25) Re2: GTTGATTGTTCCAGACGCGTTC (SEQ ID NO: 23)

[0057] (h) A polynucleotide consisting of a nucleotide sequence (SEQ ID NO: 19) corresponding to the amino acid sequence (SEQ ID NO: 16) of a protein in which a signal peptide is linked to a FVIII variant (N5Δfurin(P739Q)) in which the number of N-glycosylated Asn residues in the B domain of human FVIII, counting from the N-terminus, is set to 5, the function of the furin recognition site is deleted, and the 758th Pro residue is substituted with a Gln residue, was inserted into the SbfI / MluI site of pLVSIN-EF1α Neo Vector to prepare an N5Δfurin(P739Q)-producing lentiviral vector plasmid in the same manner as the method for preparing the N5Δfurin(P739S)-producing lentiviral vector plasmid. The two primer sets (Fw1 & Re1, Fw2 & Re2) used to subject the N5Δfurin (P739T)-producing lentiviral vector plasmid to two PCR reactions were as follows: Fw1: GTGCGGCCCCCTGTATTCTAG (SEQ ID NO: 20) Re1: CTTCTCTGCTCGATGGCATTG (SEQ ID NO: 26) Fw2: CATCGAGCAGAGAAGCTTCAGC (SEQ ID NO: 27) Re2: GTTGATTGTTCCAGACGCGTTC (SEQ ID NO: 23)

[0058] (2) Preparation of lentiviral vector producing FVIII variants A lentiviral vector producing FVIII variants was prepared using the lentiviral vector plasmid producing FVIII variants prepared in (1) in the same manner as in (2) of Experiment 1 of Experiment A.

[0059] (3) Obtaining human preadipocytes transfected with FVIII variant genes Preadipocytes were obtained from adipose tissue obtained from healthy volunteers using a method similar to that used for obtaining human preadipocytes transfected with FVIII variant genes in Experiment 1 (3) of Experiment A, and were suspended in a cell freezing medium and frozen.

[0060] (4) Preparation of FVIII mutant gene-introduced preadipocytes The preadipocytes frozen in (3) were thawed, suspended in MSF-BM / MSF-Supplement A, and centrifuged to remove the cell cryopreservation solution. The cells were then resuspended in MSF-BM / MSF-Supplement A and placed in a T150 flask. The cells were then cultured in MSF-BM / MSF-Supplement A at 37°C and 5% CO ２ The cells were cultured overnight in an incubator. The next day, the culture medium in the flask was removed, the culture surface was washed with PBS, and the preadipocytes were recovered by trypsinization and centrifuged. The centrifuged preadipocytes were suspended in MSF-BM / MSF-Supplement A, and the cell concentration was measured to be 1.25 × 10 ５ The cells were separated and centrifuged, and then suspended in a solution prepared by mixing the supernatant (FVIII variant-producing lentiviral vector) obtained in (2) with protamine sulfate to a concentration of 100 μg / mL, to give a final concentration of 2.5 × 10 ４ The transfected cell suspension was seeded into T25 flasks at 5 mL / flask, and from the next day, the cells were cultured at 37°C, 5% CO using MSF-BM / MSF-Supplement A. ２ The cells were successively expanded in an incubator with passage every four days. On the eighth day after the gene transfer procedure, a second gene transfer procedure was performed in the same manner as above, followed by successive expansion with passage every three days. On the sixth day after the second gene transfer procedure (the fourteenth day after the first gene transfer procedure), the cells were again successively expanded with passage every three or four days in the same manner as above, thereby producing FVIII mutant gene-introduced preadipocytes (a total of three gene transfer procedures were performed).

[0061] Experiment 2: Evaluation of the prepared FVIII variant gene-introduced preadipocytes Each of the three types of FVIII variant gene-introduced preadipocytes prepared in Experiment 1 was passaged on the 25th and 29th days from the day of the first gene transfection procedure in (4) of Experiment 1. At the time of passage on the 29th day, the culture medium in the flask was removed, the culture surface was washed with PBS, and the cells were recovered by trypsinization, centrifuged, suspended in MSF-BM / MSF-Supplement A, and the cell concentration was measured. ５ The cells were collected, centrifuged, suspended in PBS, frozen, and stored until use in evaluation. A portion of the culture medium removed from the flask was also frozen and stored until use in evaluation.

[0062] (1) Calculation of the average introduced copy number of FVIII variant genes The average introduced copy number of FVIII variant genes was calculated from DNA extracted from cells recovered at passage 29 days after the first gene transfer procedure using the same method as in (1) of Experiment 2 of Experiment A. The results are shown in Figure 8 (N5Δfurin(P739S) is abbreviated as P739S, N5Δfurin(P739N) is abbreviated as P739N, and N5Δfurin(P739Q) is abbreviated as P739Q, respectively; the same applies hereinafter). 8 also shows the results for four types of FVIII variant gene-introduced preadipocytes prepared in Experiment 1 (1) of Experiment A using the ΔB-producing lentiviral vector plasmid, N5Δfurin-producing lentiviral vector plasmid, N5Δfurin (P739T)-producing lentiviral vector plasmid, and N6Δfurin-producing lentiviral vector plasmid, using the same method as described in Experiment 1 (2) to (4) (N5Δfurin (P739T) is abbreviated as P739T. The same applies below.). As is clear from FIG. 8, in the gene-introduced preadipocytes producing ΔB, N5Δfurin, N5Δfurin (P739T), N5Δfurin (P739S), N5Δfurin (P739N), N5Δfurin (P739Q), and N6Δfurin, the average introduced copy number was similar and not significantly different.

[0063] (2) Measurement of FVIII antigen amount in culture medium The amount of FVIII antigen in the culture medium collected at the time of passage on day 29 from the day of the first gene transfer procedure was measured using a method similar to that used for measuring the amount of FVIII antigen in the culture medium in Experiment 2(2) of Experiment A. The results are shown in Figure 9. As is clear from Figure 9, the amount of FVIII antigen in the culture medium showed a very high value for ΔB, but the values ​​for the other FVIII variants were half or less of the value for ΔB, and there was not much difference.

[0064] (3) Measurement of FVIII activity of culture medium The frozen culture medium was thawed, and the FVIII activity of the culture medium was measured using a Revohem FVIII synthetic substrate kit (Sysmex Corporation) according to the kit's protocol. The FVIII activity was quantified as a relative value (% value when the FVIII activity of healthy human plasma was set as 100%) based on the absorbance of a fully automated blood coagulation analyzer (CS-1600, Sysmex Corporation) using a calibration curve prepared for the FVIII activity of healthy human plasma. The FVIII activity of the culture medium recovered at the 29th day of passage from the first gene transfer procedure is shown in Figure 10. As is clear from Figure 10, the FVIII activity of the culture medium for N5Δfurin (P739N) and N5Δfurin (P739Q) was as high as that of ΔB.

[0065] (4) Calculation of specific activity (activity per antigen amount) of FVIII variants secreted into culture medium. The specific activity of the FVIII variants secreted into culture medium was calculated from (2) the results of measuring the amount of FVIII antigen in the culture medium and (3) the results of measuring the FVIII activity of the culture medium. The results are shown in Figure 11. As is clear from Figure 11, the specific activities of the FVIII variants secreted into culture medium were high for N5Δfurin(P739T), N5Δfurin(P739S), N5Δfurin(P739N), and N5Δfurin(P739Q), and were significantly higher than the value for ΔB (significant difference). They were also higher than the value for N5Δfurin (significant difference except for N5Δfurin(P739S)). Therefore, it was found that a FVIII variant in which the Pro residue at position 758 in the amino acid sequence of N5Δfurin, which consists of amino acids 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 3, was substituted with a polar, uncharged amino acid residue such as Thr, Ser, Asn, or Gln, has a high specific activity.

[0066] Experiment C: Investigation of mixed transplantation of FVIII variant gene-transfected preadipocytes and human vWF fragment gene-transfected preadipocytes Experiment 1: Preparation of human vWF fragment gene-transfected preadipocytes (1) Preparation of human vWF fragment-producing lentiviral vector plasmid A polynucleotide was synthesized by adding SbfI restriction enzyme sites to the 5' end and MluI restriction enzyme sites to the 3' end of the nucleotide sequence (SEQ ID NO: 32) corresponding to the amino acid sequence (SEQ ID NO: 29) of a fragment (vWF1) consisting of the SP, D1, D2, D', and D3 domains and 26 amino acid residues counting from the N-terminus of the A1 domain, plus 3 additional amino acid residues and a CK domain in the SP-D1-D2-D'-D3-A1-A2-A3-D4-B1-B2-B3-C1-C2-CK domain structure of the human vWF precursor. A vWF1-producing lentiviral vector plasmid was prepared by inserting the vector into the SbfI / MluI site. Similarly, a vWF2-producing lentiviral vector plasmid was prepared by inserting a polynucleotide consisting of a nucleotide sequence (SEQ ID NO: 33) corresponding to the amino acid sequence (SEQ ID NO: 30) of a fragment (vWF2) consisting of 26 amino acid residues counting from the N-terminus of the SP, D1, D2, D', and D3 domains and the A1 domain into the SbfI / MluI site of pLVSIN-EF1α Neo Vector, and a vWF3-producing lentiviral vector plasmid was prepared by inserting a polynucleotide consisting of a nucleotide sequence (SEQ ID NO: 34) corresponding to the amino acid sequence (SEQ ID NO: 31) of a fragment (vWF3) consisting of 26 amino acid residues counting from the N-terminus of the SP, D', and D3 domains and the A1 domain into the SbfI / MluI site of pLVSIN-EF1α Neo Vector. The nucleotide sequence of vWF1 was the corresponding nucleotide sequence in the nucleotide sequence of human vWF (SEQ ID NO: 35). The nucleotide sequences of vWF2 and vWF3 were codon-optimized. The structure of the vWF3-producing lentiviral vector plasmid is shown in Figure 12.

[0067] (2) Preparation of human vWF fragment-producing lentiviral vector A human vWF fragment-producing lentiviral vector was prepared using the human vWF fragment-producing lentiviral vector plasmid prepared in (1) in the same manner as in the preparation of the FVIII variant-producing lentiviral vector in Experiment 1 (2) of Experiment A.

[0068] (3) Obtaining human preadipocytes transfected with human vWF fragment genes Preadipocytes were obtained from adipose tissue obtained from healthy volunteers using a method similar to that used for obtaining human preadipocytes transfected with FVIII variant genes in (3) of Experiment 1 of Experiment A, and were then suspended in a cell freezing medium and frozen.

[0069] (4) Preparation of human vWF fragment gene-introduced preadipocytes The preadipocytes frozen in (3) were thawed, suspended in MSF-BM / MSF-Supplement A, and centrifuged to remove the cell cryopreservation solution. The cells were then resuspended in MSF-BM / MSF-Supplement A and placed in a T225 flask. The cells were then cultured in MSF-BM / MSF-Supplement A at 37°C and 5% CO ２ The cells were cultured in an incubator for 4 days. The next day, the culture medium in the flask was removed, the culture surface was washed with PBS, and the preadipocytes were recovered by trypsinization and centrifuged. The centrifuged preadipocytes were suspended in MSF-BM / MSF-Supplement A, and the cell concentration was measured to obtain a cell density of 2.5 × 10 ５ The cells were separated and centrifuged, and then suspended in a solution prepared by mixing the supernatant (human vWF fragment-producing lentiviral vector) obtained in (2) with protamine sulfate to a concentration of 100 μg / mL, to give a final concentration of 5.0 × 10 ４ The transfected cell suspension was seeded into T25 flasks at 5 mL / flask, and from the next day, the cells were cultured at 37°C, 5% CO using MSF-BM / MSF-Supplement A. ２The cells were cultured in an incubator. On the second day after the gene transfer operation, a second gene transfer operation was performed in the same manner as above, followed by sequential expansion with passage every 3 days. On the sixth day after the second gene transfer operation (the eighth day after the first gene transfer operation), the cells were again cultured in the same manner as above, followed by sequential expansion with passage every 3 or 4 days, to produce vWF1 gene-transfected preadipocytes. On the second day after the gene transfer operation, a second gene transfer operation was performed in the same manner as above, followed by culture. On the third day after the second gene transfer operation (the fifth day after the first gene transfer operation), the cells were again cultured in the same manner as above, followed by sequential expansion with passage every 3 or 4 days, to produce vWF2 gene-transfected preadipocytes and vWF3 gene-transfected preadipocytes (each gene-transfected preadipocyte was subjected to gene transfer three times in total).

[0070] Experiment 2: Evaluation of mixed transplantation of human vWF fragment gene-introduced preadipocytes with FVIII variant gene-introduced preadipocytes (measurement of FVIII antigen amount in the blood of transplanted mice) (Measurement method) N5Δfurin (P739T)-producing lentiviral vector prepared in Experiment 1 (1) of Experiment A by a method similar to that described in Experiment 1 (2) of Experiment A using the N5Δfurin (P739T)-producing lentiviral vector prepared in Experiment 1 (1) of Experiment A, and N5Δfurin (P739T) gene-introduced preadipocytes prepared in Experiment 1 (3) by a method similar to that described in Experiment 1 (4) (however, the gene introduc- tion operation was performed three times in total, on the second and seventh days after the first gene introduc- tion operation (five days after the second gene introduc- tion operation, with no intervening passage)), and each of the three types of human vWF fragment gene-introduced preadipocytes prepared in Experiment 1 were cultured in MSF-BM / MSF-Supplement. A was suspended in 4 × 10 ４ cells / cm ２ The cells were seeded with MSF-BM / MSF-Supplement A and incubated at 37°C, 5% CO ２The cells were cultured in an incubator. The next day, the medium was replaced with PGM-2, and the cells were cultured for an additional 3 days. The next day, the culture medium in the flask was removed, the culture surface was washed with PBS, and the cells were then recovered by trypsinization and spun down. The spun down cells were suspended in Ringer's solution containing albumin at a final concentration of 0.5% and spun down. This series of steps was repeated four times to wash the cells, and the washed cells were then suspended in a mixed solution of Fiblast Spray and Medgel (Nitta Gelatin Co., Ltd.) until the concentration of each cell was 8 × 10 ７ A cell suspension containing 200 μg / mL trafermin and 8 mg / mL Medgel was prepared. Next, the prepared N5Δfurin (P739T) gene-transfected preadipocyte suspension and the human vWF fragment gene-transfected preadipocyte suspension were mixed in equal amounts to prepare a mixed suspension of N5Δfurin (P739T) gene-transfected preadipocytes and human vWF fragment gene-transfected preadipocytes. This was further mixed with an equal amount of Matrigel to give a final cell concentration of 2 × 10 ７ A cell suspension for transplantation (final concentration of trafermin was 100 μg / mL and final concentration of Medgel was 4 mg / mL) was prepared. The prepared cell suspension for transplantation was administered subcutaneously into the left and right dorsal skin of 6-week-old NOD / SCID mice (CLEA Japan, the same applies hereinafter) under isoflurane anesthesia using a 20G blunt needle cannula at a rate of 0.5 mL / site (2 × 10 cells each of the two types of cells). ７ On days 3, 7, 14, 21, and 28 after the transplantation of the cell suspension for transplantation containing the FVIII variant gene-introduced preadipocytes and the human vWF fragment gene-introduced preadipocytes into NOD / SCID mice, blood was collected from the cheek of the mice using an animal lancet under isoflurane anesthesia, and the amount of FVIII antigen in the blood of the transplanted mice was measured in the same manner as in Experiment 4 of Experiment A.

[0071] (Measurement results) are shown in Figure 13. In addition, Figure 13 also shows the results of measurements using the same method as above, with a final concentration of 2 x 10 ７ A cell suspension for transplantation containing only N5Δfurin (P739T) gene-transduced preadipocytes was prepared and administered subcutaneously to the left and right dorsal regions at 0.5 mL / site (2 × 10７ 13 shows the results of a cotransplantation of N5Δfurin(P739T) gene-introduced preadipocytes and vWF3 gene-introduced preadipocytes, the amount of FVIII antigen in the blood of the transplanted mice was higher when N5Δfurin(P739T) gene-introduced preadipocytes were mixed and transplanted than when N5Δfurin(P739T) gene-introduced preadipocytes alone were transplanted, when N5Δfurin(P739T) gene-introduced preadipocytes were mixed and transplanted with vWF1 gene-introduced preadipocytes, or when N5Δfurin(P739T) gene-introduced preadipocytes were mixed and transplanted with vWF2 gene-introduced preadipocytes (significant differences were observed in all cases). Thus, it was found that the amount of FVIII variant secreted into the blood was improved by cotransplanting human vWF fragment gene-introduced preadipocytes and FVIII variant gene-introduced preadipocytes.

[0072] Experiment 3: Evaluation of scaffold materials (extracellular matrix components) when human vWF fragment gene-introduced preadipocytes were mixed and transplanted with FVIII variant gene-introduced preadipocytes (measurement of FVIII antigen levels in the blood of transplanted mice) (Measurement method) The N5Δfurin (P739T) gene-introduced preadipocytes described in Experiment 2 and the vWF3 gene-introduced preadipocytes prepared in Experiment 1 were cultured, collected, centrifuged, washed, and then suspended in Ringer's solution containing albumin at a final concentration of 0.5% to a cell concentration of 2 × 10 ８ Next, the prepared N5Δfurin (P739T) gene-transfected preadipocyte suspension and the vWF3 gene-transfected preadipocyte suspension were mixed in equal amounts to prepare a mixed suspension of N5Δfurin (P739T) gene-transfected preadipocytes and vWF3 gene-transfected preadipocytes, which was then further mixed with an equal amount of 1.2% or 2.0% atelocollagen (Kokensha, the same applies hereinafter) to give a final cell concentration of 5 × 10 ７ A cell suspension for transplantation (final concentration of atelocollagen: 0.6% or 1.0%) was prepared at 0.2 mL / site using a 20G blunt needle cannula into the left and right dorsal skin of 6-week-old NOD / SCID mice under isoflurane anesthesia. The prepared cell suspension for transplantation was administered subcutaneously at 0.2 mL / site (2 × 10 cells each of the two types of cells). ７On days 3, 7, 14, 21, 28, 35, 42, 49, and 56 after the transplantation of the cell suspension for transplantation containing the FVIII variant gene-introduced preadipocytes and the vWF3 gene-introduced preadipocytes into NOD / SCID mice, blood was collected from the cheek of the mice using an animal lancet under isoflurane anesthesia, and the amount of FVIII antigen in the blood of the transplanted mice was measured by the same method as the method for measuring the amount of FVIII antigen in the blood of the transplanted mice in Experiment 4 of Experiment A.

[0073] (Measurement results) are shown in Figure 14. In addition, Figure 14 also shows the results of measurements using the same method as above, with a final concentration of 5 x 10 ７ A cell suspension for transplantation containing only N5Δfurin (P739T) gene-transduced preadipocytes (final concentration of atelocollagen: 0.6%) was prepared and administered subcutaneously to the left and right dorsal regions at 0.2 mL / site (2 × 10 ７ As is clear from Figure 14, the amount of FVIII antigen in the blood of the transplanted mice was higher when atelocollagen was added to the transplant cell suspension containing N5Δfurin(P739T) gene-introduced preadipocytes and vWF3 gene-introduced preadipocytes than when atelocollagen was added to the transplant cell suspension containing only N5Δfurin(P739T) gene-introduced preadipocytes. The values ​​obtained when a cell suspension containing atelocollagen at a final concentration of 1.0% were higher for a long period of time than the values ​​obtained when a cell suspension for transplantation containing only N5Δfurin(P739T) gene-transfected preadipocytes was added with a final concentration of 0.6% atelocollagen, and the values ​​obtained when a cell suspension for transplantation containing N5Δfurin(P739T) gene-transfected preadipocytes and vWF3 gene-transfected preadipocytes was added with a final concentration of 0.6% atelocollagen (there were significant differences in the values ​​obtained in both cases). Thus, it was found that the use of atelocollagen as a scaffold material when human vWF fragment gene-transfected preadipocytes are mixed and transplanted with FVIII variant gene-transfected preadipocytes improves the amount of FVIII variant secreted into the blood.

[0074] Experiment 4: Evaluation of transplantation site when human vWF fragment gene-introduced preadipocytes and FVIII variant gene-introduced preadipocytes are mixed and transplanted (measurement of FVIII antigen amount in the blood of transplanted mice) (Measurement method) The N5Δfurin (P739T) gene-introduced preadipocytes described in Experiment 2 and the vWF3 gene-introduced preadipocytes prepared in Experiment 1 were cultured, collected, centrifuged, washed, and then suspended in Ringer's solution containing albumin at a final concentration of 0.5% to a cell concentration of 2 × 10 ８ Next, the prepared N5Δfurin(P739T) gene-transfected preadipocyte suspension and the prepared vWF3 gene-transfected preadipocyte suspension were mixed in equal amounts to prepare a mixed suspension of N5Δfurin(P739T) gene-transfected preadipocytes and vWF3 gene-transfected preadipocytes, which was then further mixed with an equal amount of 2.0% atelocollagen to give a final cell concentration of 5 × 10 ７ A cell suspension for transplantation (final concentration of atelocollagen: 1.0%) was prepared at 0.2 mL / site (2 × 10 cells / mL for each of the two types of cells) was administered subcutaneously to the left and right dorsal regions of 6-week-old NOD / SCID mice under isoflurane anesthesia using a 20 G blunt needle cannula or to the left and right groin regions using a 25 G injection needle. ７ On days 3, 7, 14, 21, 28, 35, 42, 49, and 56 after the transplantation of the cell suspension for transplantation containing the FVIII variant gene-introduced preadipocytes and the vWF3 gene-introduced preadipocytes into NOD / SCID mice, blood was collected from the cheek of the mice using an animal lancet under isoflurane anesthesia, and the amount of FVIII antigen in the blood of the transplanted mice was measured by the same method as the method for measuring the amount of FVIII antigen in the blood of the transplanted mice in Experiment 4 of Experiment A.

[0075] (Measurement results) are shown in Figure 15. In addition, Figure 15 also shows the results of measurements using the same method as above, with a final concentration of 5 x 10 ７ A cell suspension for transplantation containing only N5Δfurin (P739T) gene-transduced preadipocytes (final concentration of atelocollagen: 1.0%) was prepared and administered at 0.2 mL / site (2 × 10 ７15 shows the results of a co-transplantation of N5Δfurin(P739T) gene-introduced preadipocytes and vWF3 gene-introduced preadipocytes into the inguinal region, the amount of FVIII antigen in the blood of the transplanted mice was higher than when only N5Δfurin(P739T) gene-introduced preadipocytes were transplanted into the inguinal region, or when N5Δfurin(P739T) gene-introduced preadipocytes and vWF3 gene-introduced preadipocytes were co-transplanted subcutaneously on the back (significant differences were observed for both cases). Thus, it was found that when human vWF fragment gene-introduced preadipocytes and FVIII variant gene-introduced preadipocytes were co-transplanted into the inguinal region, the amount of FVIII variant secreted into the blood was improved.

[0076] Experiment 5: Evaluation of the ratio of transplanted cell numbers of human vWF fragment gene-introduced preadipocytes when mixed and transplanted with FVIII variant gene-introduced preadipocytes (measurement of FVIII antigen amount in the blood of transplanted mice) (Measurement method) The N5Δfurin (P739T) gene-introduced preadipocytes described in Experiment 2 and the vWF3 gene-introduced preadipocytes prepared in Experiment 1 were each cultured, collected, centrifuged, washed, and then suspended in Ringer's solution containing albumin at a final concentration of 0.5% in the same manner as in Experiment 2 to prepare cell suspensions with the following cell concentrations for each cell type: (N5Δfurin (P739T) gene-introduced preadipocyte suspension): 5 x 10 ７ cells / mL 3.3 x 10 ７ cells / mL (vWF3 gene-transfected preadipocyte suspension) 1 x 10 ８ cells / mL ・ 5 x 10 ７ cells / mL ・ 2.5 x 10 ７Next, either of the prepared N5Δfurin (P739T) gene-introduced preadipocyte suspensions was mixed with either of the prepared vWF3 gene-introduced preadipocyte suspensions in equal amounts to prepare a mixed suspension of N5Δfurin (P739T) gene-introduced preadipocytes and vWF3 gene-introduced preadipocytes, and an equal amount of 1.6% atelocollagen was further mixed to prepare three types of cell suspensions for transplantation (final concentration of atelocollagen: 0.8%), each with a final cell concentration as follows: (1) N5Δfurin (P739T) gene-introduced preadipocytes: 1.25 × 10 ７ cells / mL, vWF3 gene-transfected preadipocytes were 6.25 × 10 ６ cells / mL (2) N5Δfurin (P739T) gene-transduced preadipocytes were 1.25 × 10 ７ cells / mL, vWF3 gene-transfected preadipocytes were 1.25 × 10 ７ cells / mL (3) N5Δfurin (P739T) gene-transduced preadipocytes were 8.25 × 10 ６ cells / mL, vWF3 gene-transfected preadipocytes were 2.5 × 10 ７ Each of the three types of cell suspensions for transplantation prepared was administered to the left and right inguinal regions of 6-week-old NOD / SCID mice under isoflurane anesthesia using a 25G injection needle at a dose of 0.4 mL / site for cell suspension for transplantation (1) (1 × 10 cells / mL for N5Δfurin (P739T) gene-transduced preadipocytes). ７ cells / animal, and vWF3 gene-transduced preadipocytes were 5 x 10 ６ cells / animal), and the cell suspension for transplantation (2) was also administered at 0.4 mL / site (1 × 10 cells each of the two types). ７ cells / animal), and the cell suspension for transplantation (3) was administered at 0.6 mL / site (N5Δfurin (P739T) gene-transduced preadipocytes were administered at 1 × 10 ７ cells / animal, and vWF3 gene-transduced preadipocytes were 3 x 10 ７On days 3, 7, 14, 21, 28, 35, 42, 49, and 56 after the transplantation of the cell suspension for transplantation containing the FVIII variant gene-introduced preadipocytes and the vWF3 gene-introduced preadipocytes into NOD / SCID mice, blood was collected from the cheek of the mice using an animal lancet under isoflurane anesthesia, and the amount of FVIII antigen in the blood of the transplanted mice was measured by the same method as the method for measuring the amount of FVIII antigen in the blood of the transplanted mice in Experiment 4 of Experiment A.

[0077] (Measurement results) are shown in Figure 16. As is clear from Figure 16, the amount of FVIII antigen in the blood of transplanted mice increased as the ratio of the number of transplanted vWF3 gene-introduced preadipocytes to the number of transplanted N5Δfurin (P739T) gene-introduced preadipocytes increased, and the value when transplant cell suspension (3) was administered was higher than the values ​​when transplant cell suspension (1) and transplant cell suspension (2) were administered (significant differences were observed in both cases). Thus, when human vWF fragment gene-introduced preadipocytes were mixed and transplanted with FVIII variant gene-introduced preadipocytes, it was found that increasing the ratio of the number of transplanted human vWF fragment cells to the number of transplanted FVIII variant gene-introduced preadipocytes improved the amount of FVIII variant secreted into the blood.

[0078] INDUSTRIAL APPLICABILITY The present invention has industrial applicability in that it can provide novel cells for ex vivo gene therapy that produce FVIII variants.

Claims

1. Adipocytes for ex vivo gene therapy that produce a modified human FVIII consisting of amino acid sequences 20 to 2351 in the amino acid sequence shown in Sequence ID No. 1, wherein the B domain (amino acid sequences 760 to 1667) is modified, the number of N-linked glycosylation Asn residues is 5 counted from the N-terminus, the function of the furin recognition site is lost, and the 758th Pro residue is replaced with a polar uncharged amino acid residue selected from Thr, Ser, Asn, and Gln residues to produce an FVIII modified cell.

2. The adipocyte for ex vivo gene therapy according to claim 1, wherein the FVIII variant is any of (a) to (c) below. (a) A protein consisting of amino acid sequences 20 to 1652 in the amino acid sequence shown in Sequence ID No.

5. (b) A protein having the same biological activity as the protein in (a), comprising an amino acid sequence in which the amino acid sequence from position 20 to 1652 in the amino acid sequence shown in Sequence ID No. 5 includes the deletion, substitution, addition, or insertion of one or more amino acid residues, and having the same biological activity as the protein in (a). (c) A protein having at least 90% sequence identity with the amino acid sequence between positions 20 and 1652 in the amino acid sequence shown in Sequence ID No. 5, and having biological activity equivalent to that of the protein in (a).

3. The adipocyte for ex vivo gene therapy according to claim 1, wherein the FVIII variant is any of (a) to (c) below. (a) A protein consisting of amino acid sequences 20 to 1652 in the amino acid sequence shown in SEQ ID NO:

14. (b) A protein having the same biological activity as the protein in (a), comprising an amino acid sequence in which the amino acid sequence from position 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 14 includes the deletion, substitution, addition, or insertion of one or more amino acid residues, and having the same biological activity as the protein in (a). (c) A protein having at least 90% sequence identity with the amino acid sequence between positions 20 and 1652 in the amino acid sequence shown in Sequence ID No. 14, and having biological activity equivalent to that of the protein in (a).

4. The adipocyte for ex vivo gene therapy according to claim 1, wherein the FVIII variant is any of (a) to (c) below. (a) A protein consisting of amino acid sequences 20 to 1652 in the amino acid sequence shown in SEQ ID NO:

15. (b) A protein having the same biological activity as the protein in (a), comprising an amino acid sequence in which the amino acid sequence from position 20 to 1652 in the amino acid sequence shown in Sequence ID No. 15 includes the deletion, substitution, addition, or insertion of one or more amino acid residues, and having the same biological activity as the protein in (a). (c) A protein having at least 90% sequence identity with the amino acid sequence between positions 20 and 1652 in the amino acid sequence shown in Sequence ID No. 15, and having biological activity equivalent to that of the protein in (a).

5. The adipocyte for ex vivo gene therapy according to claim 1, wherein the FVIII variant is any of (a) to (c) below. (a) A protein consisting of amino acid sequences 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16 (b) A protein having the same biological activity as the protein in (a), comprising an amino acid sequence in which the amino acid sequence from position 20 to 1652 in the amino acid sequence shown in SEQ ID NO: 16 includes the deletion, substitution, addition, or insertion of one or more amino acid residues, and having the same biological activity as the protein in (a). (c) A protein having at least 90% sequence identity with the amino acid sequence between positions 20 and 1652 in the amino acid sequence shown in Sequence ID No. 16, and having biological activity equivalent to that of the protein in (a).

6. A method for producing adipocytes for ex vivo gene therapy, comprising introducing an expression vector into adipocytes that carries at least one polynucleotide encoding the amino acid sequence of a modified human FVIII, comprising the amino acid sequences from 20 to 2351 in the amino acid sequence shown in Sequence ID No. 1, wherein the B domain (amino acid sequences from 760 to 1667) is modified, the number of N-linked glycosylation Asn residues is 5 counted from the N-terminus, the function of the furin recognition site is lost, and the 758th Pro residue is replaced with a polar uncharged amino acid residue selected from Thr, Ser, Asn, and Gln residues.

7. A method for producing adipocytes for ex vivo gene therapy according to claim 6, wherein the expression vector is a viral vector.

8. A method for producing adipocytes for ex vivo gene therapy according to claim 7, wherein the viral vector is a lentiviral vector.

9. An expression vector for producing adipocytes for ex vivo gene therapy, comprising a modified human FVIII consisting of amino acid sequences 20 to 2351 in the amino acid sequence shown in Sequence ID No. 1, wherein the B domain (amino acid sequences 760 to 1667) is modified, the number of N-linked glycosylation Asn residues is 5 counted from the N-terminus, the function of the furin recognition site is lost, and the 758th Pro residue is replaced with a polar uncharged amino acid residue selected from Thr, Ser, Asn, and Gln residues, and carrying a polynucleotide encoding at least the amino acid sequence of the modified FVIII.

10. A modified human FVIII having amino acid sequences 20 to 2351 in the amino acid sequence shown in Sequence ID No. 1, wherein the B domain (amino acid sequences 760 to 1667) is modified, the number of N-linked glycosylation Asn residues is 5 counted from the N-terminus, the function of the furin recognition site is lost, and the 758th Pro residue is replaced with a polar uncharged amino acid residue selected from Thr, Ser, Asn, and Gln residues.

11. A polynucleotide that encodes at least the amino acid sequence of the FVIII variant described in claim 10.