WO2025164792A1 - High specific activity/high secretion expression-type factor viii variant, nucleic acid encoding high expression-type factor viii, and use thereof - Google Patents

Abstract

The present invention provides: a human factor VIII variant characterized in that at least one amino acid residue selected from the group consisting of K213, S367, and F2196 of a wild-type human factor VIII is substituted with another amino acid, and characterized by having a higher specific activity and/or secretion expression efficiency as compared to the wild type; a nucleic acid encoding the human factor VIII variant; and a therapeutic agent for hemophilia A using the same.

Description

High specific activity/high secretory expression type of modified factor VIII, highly expressed type of nucleic acid encoding factor VIII, and uses thereof

The present invention relates to human factor VIII variants with high specific activity and high secretory expression, nucleic acids encoding the same, highly expressible human factor VIII-encoding nucleic acids, and uses thereof. More specifically, the present invention relates to human factor VIII variants with enhanced secretory expression efficiency and/or specific activity compared to the wild-type, nucleic acids encoding the same, nucleic acids encoding human factor VIII (including variants) whose expression efficiency has been enhanced by modifying the base sequence, and the use thereof for the treatment of hemophilia A, etc.

Hemophilia is a disease caused by a genetic deficiency in blood clotting factors involved in hemostasis. According to the World Federation of Hemophilia (WFH), there are approximately 400,000 hemophilia patients worldwide, with approximately 6,000 patients in Japan. There are two types of hemophilia: hemophilia A, which is caused by a deficiency in factor VIII, and hemophilia B, which is caused by a deficiency in factor IX. Hemophilia A patients predominate, with a ratio of 5:1.

Traditionally, hemophilia A has been treated with factor VIII replacement therapy. However, the half-life of this clotting factor is short, and patients need to receive intravenous administration two to three times a week from childhood to prevent bleeding. Furthermore, when factor VIII is administered continuously to hemophilia A patients can develop antibodies against factor VIII, known as inhibitors (25-30%), which is known to reduce the effectiveness of replacement therapy.

In recent years, the antibody drug emicizumab, which acts as a substitute for factor VIII, has been launched as a treatment for hemophilia A. Emicizumab can treat hemophilia A with subcutaneous injections less frequently (once a month) than factor VIII preparations. However, it is no different from conventional coagulation factor preparations in that it requires continuous administration, and because it is produced using recombinant animal cells, the manufacturing costs are high, placing a significant burden on the medical economy.

Therefore, gene therapy using viral vectors, which are expected to maintain clotting factor activity for long periods with a single administration, is attracting attention as a new treatment for hemophilia. Hemophilia is a hereditary disease caused by a single gene deficiency. Because therapeutic effects can be easily confirmed by measuring clotting factor levels in the blood and because even a relatively small increase in blood levels can be expected to be effective, it has long been considered a good target for gene therapy. Furthermore, preclinical studies have shown that continuous expression of clotting factors through gene therapy can induce immune tolerance via regulatory T cells, potentially eliminating inhibitors. Recently, a series of partially successful human gene therapy cases using adeno-associated virus (AAV) vectors have been reported (see, for example, Non-Patent Documents 1 and 2). Clotting factors are produced in the liver, and AAV vectors can deliver genes to the liver with high efficiency via intravenous administration.

In a clinical trial conducted by BioMarin, high factor VIII activity was maintained even after one year in the high-dose administration group (Non-Patent Document 1), but a tendency for blood levels to decrease was observed after three years. One possible reason for this is that factor VIII expression from a high-dose AAV vector in liver cells may induce endoplasmic reticulum stress (Non-Patent Document 3).
On the other hand, in Spark's clinical trials, a serotype with high liver tropism was used and a lower dose of AAV vector than BioMarin's was administered, but hepatotoxicity was observed in the 2 x ¹⁰¹² vg/kg administration group (Non-Patent Document 2). Attempts have been made to reduce the dose and improve safety by achieving high expression through the use of liver-specific promoters and codon optimization of factor VIII, but these approaches are not sufficient in cases where high protein expression itself impairs the long-term stability of the therapeutic effect.

As one way to avoid the attenuation of therapeutic efficacy due to high-dose viral vector administration, it is desirable to develop a variant of factor VIII with high specific activity that can exert coagulation activity equivalent to or greater than that of wild-type factor VIII expressed at appropriate protein levels, as well as a variant of factor VIII with improved extracellular secretion efficiency to eliminate adverse effects on cells such as endoplasmic reticulum stress. Xiao et al. have reported that substituting 10 amino acid residues in the A1 domain with porcine residues significantly increases secretion efficiency, with five amino acid substitutions being particularly important (Patent Document 1 and Non-Patent Document 4). Furthermore, the same group has reported that substituting 12 amino acid residues in the light chain of human factor VIII with canine residues increases specific activity while maintaining wild-type secretion efficiency (Non-Patent Document 5). Furthermore, Sabatino et al. have disclosed that the specific activity can be increased by substituting or deleting one or more amino acids in the furin recognition site and a3 domain in the human factor VIII light chain with canine forms (Patent Document 2).

WO 2014/209942 US 2019/0144524

Rangarajan, S. et al., N. Eng. J. Med., 377: 2519-2530 (2017) Perrin, G.Q. et al., Blood, 133: 407-414 (2019) Zolotukhin, I. et al., Mol. Ther. Methods Clin. Dev., 3: 16063 (2016) Cao, W. et al., Mol. Ther. Methods Clin. Dev., 19: 486-495 (2020) Firrman, J. et al., Mol. Ther. Methods Clin. Dev., 17: 328-336 (2020)

The first object of the present invention is to provide a novel human factor VIII variant and the nucleic acid encoding it that can stably complement factor VIII over the long term in gene therapy for hemophilia A, thereby establishing a gene therapy for hemophilia A with superior long-term efficacy. The second object of the present invention is to provide a recombinant factor VIII preparation containing the variant and an mRNA drug encoding the variant.

As a result of extensive research aimed at solving the above-mentioned problems, the present inventors have found that various variants in which at least one amino acid residue selected from the group consisting of lysine at position 213 (K213), serine at position 367 (S367), and phenylalanine at position 2196 (F2196) of wild-type human factor VIII is substituted with another amino acid exhibit higher specific activity and/or higher secretory expression than the wild-type (sometimes abbreviated herein as "high specific activity/high secretory expression type"). (In this specification, the positions of amino acid residues in human factor VIII are represented by the amino acid numbering in the mature full-length single-chain polypeptide (2332 amino acids), and the nth amino acid residue X (X is the one-letter code for the amino acid) is abbreviated as "Xn." When n is a negative integer, it indicates the position of the amino acid residue in the signal peptide. When the nth amino acid residue X is substituted with another amino acid Y (Y is the one-letter code for the other amino acid), it is abbreviated as "XnY.")

Furthermore, in order to improve the expression efficiency of the variants containing the above amino acid substitutions in humans, the inventors optimized the nucleotide sequence of the nucleic acid encoding the variants for human use using a codon optimization algorithm. However, the resulting nucleotide sequence contained many CpG sequences. The inventors predicted that CpG sequences may induce an immune response in the host and inhibit continuous transgene expression, so they performed base substitutions to remove these CpG sequences without changing the amino acid sequence of the variants, and then performed codon optimization again. The resulting variant-encoding nucleic acid was ligated downstream of a promoter sequence from which CpG sequences had also been removed, and then incorporated into an AAV vector. When the resulting nucleic acid encoding the variants was administered to a human liver cell line and a hemophilia model mouse, factor VIII expression was significantly increased compared to the variant-encoding nucleic acid before CpG sequence removal. To demonstrate that the expression-enhancing effect of CpG sequence removal is not limited to the variant but is common to any human factor VIII, including wild-type factor VIII, the expression of factor VIII was similarly compared between a codon-optimized nucleic acid encoding wild-type human factor VIII and a nucleic acid that had been further codon-optimized after CpG sequence removal. As a result, a significant enhancement of expression was observed after CpG sequence removal.
Based on these findings, the present inventors have conducted further studies and have completed the present invention.

That is, the present invention provides the following.
[Section 1]
A human factor VIII variant characterized in that at least one amino acid residue selected from the group consisting of K213, S367 and F2196 of wild-type human factor VIII is replaced with another amino acid, and that the variant has higher specific activity and/or secretory expression efficiency compared to the wild-type variant.
[Section 2]
Item 2. The variant according to Item 1, wherein the amino acid substitution at K213 is K213N or K213H, and/or the amino acid substitution at S367 is S367P, S367N or S367Q, and/or the amino acid substitution at F2196 is F2196L or F2196M.
[Section 3]
Item 3. The variant according to Item 2, comprising the amino acid substitutions K213N, S367P, and F2196L.
[Section 4]
Item 4. The variant according to any one of Items 1 to 3, wherein at least one amino acid residue selected from the group consisting of R-5, P25, A28, L152, M217, W228, Q410, Y487, R489, F501, M539, I566, L603, I642, S727, A736, S1657, Q1659, E1661, I1668, D1681, R1776, H1859, A1993, H2007, N2019, K2085, K2207, F2275, S2296, V2314, Q2316, and M2321 is further substituted with another amino acid.
[Section 5]
R-5P, P25H, A28T, L152A, L152D, L152E, L152H, L152I, L152M, L152N, L152P, L152Q or L152V, M217T, W228Q, Q410L, Y487H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657 Item 5. The variant according to Item 4, comprising at least one amino acid substitution selected from the group consisting of P, Q1659E, E1661K, I1668F, D1681G, R1776K, H1859R, A1993V, H2007Q, N2019K, K2085M, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L.
[Section 6]
L152A, L152D, L152E, L152H, L152I, L152M, L152N, L152P, L152Q or L152V, and/or I1668F and D1681G
Item 4. The variant according to Item 3, further comprising an amino acid substitution of:
[Section 7]
the amino acid substitution at K213 is K213N;
M539L, and optionally at least one amino acid substitution in the A2 domain selected from the group consisting of I566M, L603P, and I642V;
Item 14. The variant of item 1, further comprising at least one amino acid substitution in the C2 domain selected from the group consisting of K2207Q, Q2316H and M2321L, and optionally F2275L, S2296A and V2314A.
[Section 8]
Item 8. The variant of Item 7, further comprising the amino acid substitutions I1668F and D1681G.
[Section 9]
Item 4. The variant of item 3, further comprising amino acid substitutions L152P, Y487H and L603P, and optionally further comprising at least one amino acid substitution selected from the group consisting of S727P, Q1659E, H1859R, A1993V, H2007Q, N2019K and K2085M.
[Section 10]
Item 4. The variant according to Item 3, further comprising the amino acid substitutions M539L, I566M, L603P, I642V, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L.
[Section 11]
Item 11. The variant of Item 10, further comprising the amino acid substitutions I1668F and D1681G.
[Section 12]
A human factor VIII variant having the following amino acid substitutions (a) to (n) in wild-type human factor VIII:
(a) K213N, S367P, and F2196L
(b) K213N, S367P, F2196L, I1668F, and D1681G
(c) K213N, M539L, I566M, I642V, K2207Q, F2275L, Q2316H, and M2321L
(d) K213N, M539L, I566M, L603P, I642V, F2196L, K2207Q, F2275L, Q2316H, and M2321L
(e) K213N, M539L, I566M, I642V, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L
(f) K213N, M539L, I566M, I642V, I1668F, D1681G, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L
(g) L152P, K213N, S367P and F2196L
(h) L152P, K213N, S367P, Y487H, L603P and F2196L
(i) L152P, K213N, S367P, Y487H, L603P, S727P, Q1659E, and F2196L
(j) L152P, K213N, S367P, Y487H, L603P, S727P, Q1659E, H1859R, A1993V, H2007Q, K2085M and F2196L
(k) K213N, S367P, M539L, I566M, L603P, I642V, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
(l) P25H, A28T, K213N, M217T, S367P, Q410L, Y487H, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P , Q1659E, E1661K, I1668F, D1681G, R1776K, N2019K, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
(m) R-5P, P25H, A28T, K213N, M217T, W228Q, S367P, Q410L, Y487H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736 V, S1657P, Q1659E, E1661K, I1668F, D1681G, R1776K, N2019K, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
(n) R-5P, P25H, A28T, L152P, K213N, M217T, W228Q, S367P, Q410L, Y487 H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P, Q1 659E, E1661K, I1668F, D1681G, R1776K, H1859R, A1993V, H2007Q, N201 9K, K2085M, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
[Section 13]
Item 13. The variant according to any one of Items 1 to 12, which lacks the B domain.
[Section 14]
Item 14. The variant according to Item 13, wherein the deletion of the B domain is a deletion of the amino acid sequence from positions 763 to 1656 in the amino acid sequence represented by SEQ ID NO:1.
[Section 15]
A nucleic acid encoding the variant according to any one of Items 1 to 14.
[Section 16]
Item 16. The nucleic acid of Item 15, which is codon-optimized for human use.
[Section 17]
17. The nucleic acid of claim 16, or a nucleic acid encoding wild-type human Factor VIII that has been codon-optimized for human use, which has been further modified to not contain CpG sequences without changing the amino acid sequence that it encodes.
[Section 18]
Item 18. An expression vector comprising the nucleic acid according to any one of Items 15 to 17.
[Section 19]
19. The expression vector of claim 18, wherein the nucleic acid is under the control of a liver-specific promoter.
[Section 20]
Item 20. The expression vector according to Item 18 or 19, wherein the promoter has been modified so as not to contain a CpG sequence.
[Section 21]
Item 21. The expression vector according to any one of Items 18 to 20, wherein the vector is a viral vector.
[Section 22]
Item 22. The expression vector according to Item 21, wherein the virus is an adeno-associated virus having liver tropism.
[Section 23]
A host cell into which the expression vector according to any one of Items 18 to 22 has been introduced.
[Section 24]
24. A method for producing human factor VIII or a variant thereof, comprising culturing the host cell according to Item 23 and recovering human factor VIII or a variant thereof from the resulting culture.
[Section 25]
A pharmaceutical comprising the variant according to any one of Items 1 to 14, the nucleic acid according to any one of Items 15 to 17, the expression vector according to any one of Items 18 to 22, or the host cell according to Item 23.
[Section 26]
Item 26. The pharmaceutical agent according to Item 25, which is for treating hemophilia A.
[Section 26a]
A method for treating hemophilia A, comprising administering to a patient with hemophilia A an effective amount of the variant of any one of Items 1 to 14, the nucleic acid of any one of Items 15 to 17, the expression vector of any one of Items 18 to 22, or the host cell of Items 23.
[Section 26b]
The variant of any one of Items 1 to 14, the nucleic acid of any one of Items 15 to 17, the expression vector of any one of Items 18 to 22, or the host cell of Item 23, for use in treating hemophilia A.
[Section 26c]
Use of the variant according to any one of Items 1 to 14, the nucleic acid according to any one of Items 15 to 17, the expression vector according to any one of Items 18 to 22, or the host cell according to Item 23 for the manufacture of a therapeutic agent for hemophilia A.

The present invention provides human factor VIII variants with high specific activity and/or secretory expression efficiency, and nucleic acids encoding the same, which, when used, enable the treatment of hemophilia A with stable factor VIII supplementation over a long period of time. Furthermore, the present invention provides human factor VIII-encoding nucleic acids with enhanced protein expression efficiency compared to other nucleic acids encoding the same amino acid sequence.

The amino acid substitutions of the variants of the present invention (Ver. 1 to Ver. 12) are shown below. The in vitro (left) and in vivo coagulation factor activities of Ver. 1 to Ver. 10 are shown. The in vitro coagulation factor activity of variants such as S3, S5, S7, S4, S6, S8, and S12 is shown (left: one-stage coagulation assay, right: synthetic substrate assay). 1 shows the effect of each amino acid substitution contained in the S8 variant on the coagulation factor activity. 1 shows the coagulation factor activity of XNPL variants when L152 is substituted with various other amino acids. 1 shows the coagulation factor activity of PXPL variants when K213 is substituted with various other amino acids. 1 shows the coagulation factor activity of PNXL variants when S367 is substituted with various other amino acids. 1 shows the coagulation factor activity of PNPX variants when F2196 is substituted with various other amino acids. 1 shows the in vivo coagulation factor activity and blood FVIII antigen levels of various variants. 1 shows the intracellular and extracellular localization of wild-type and various mutant FVIII. 1 shows the subcellular localization of wild-type and FG mutant FVIII. 1 shows the in vitro coagulation factor activity of variant Ver. 12. The amino acid substitutions of the variants of the present invention (F8-Tochigi-3, 5, 8, 10, 11, 13, 15) are shown. 1 shows the in vitro coagulation factor activity and antigen amount of the variants of the present invention (F8-Tochigi-3, 5, 8, 10, 11, 13, 15, 28). 4 shows the in vitro and in vivo expression-enhancing effect of the Ver. 4 variant due to removal of CpG sequences. 1 shows the effect of removing CpG sequences on enhancing the expression of wild-type hFVIII in vitro and in vivo. This shows that an AAV8 vector carrying the Ver. 4 variant with CpG sequences removed can significantly increase factor VIII activity in a cynomolgus monkey model at the dose used in clinical trials. This shows that an AAV5 vector carrying the Ver. 4 variant with CpG sequences removed can significantly increase factor VIII activity in a cynomolgus monkey model at a dose of 1/10 (top) and 1/30 (bottom) of the commercially available drug. 1 shows the effect of enhancing factor VIII activity by shortening the polyA addition signal, a sequence between the promoter and the human factor VIII coding sequence.

The single-chain full-length human factor VIII protein, after its 19-amino acid signal peptide is cleaved in the endoplasmic reticulum, consists of three A domains (A1, A2, A3), two C domains (C1, C2), a B domain, and three peptide regions rich in acidic amino acids (a1, a2, a3). From the amino terminus, these are arranged in the following order: A1 (positions 1-336), a1 (positions 337-372), A2 (positions 373-710), a2 (positions 711-740), B (positions 741-1648), a3 (positions 1649-1689), A3 (positions 1690-2019), C1 (positions 2020-2172), and C2 (positions 2173-2332). In the Golgi apparatus, R1313 and R1648 in the B domain are cleaved, resulting in a two-chain heavy and light chain. Since the B domain is deleted without significantly impairing coagulation activity, B domain-deleted (BDD) factor VIII is commonly used and is also preferably used in the present invention.

The present invention provides a human factor VIII variant (hereinafter also referred to as the "variant of the present invention") that has enhanced specific activity and/or secretory expression efficiency compared to the wild-type form. The variant of the present invention is characterized in that at least one amino acid residue selected from the group consisting of K213, S367, and F2196 of wild-type human factor VIII is substituted with another amino acid.

As used herein, the term "wild-type human factor VIII" refers to a protein comprising the amino acid sequence of naturally occurring normal human factor VIII, specifically:
Examples of such proteins include (a) a protein consisting of the amino acid sequence represented by SEQ ID NO: 1 (registered in UniProtKB under accession number P00451), or (b) a protein consisting of the amino acid sequence of an allelic variant (minor allele frequency (MAF) less than 1%) or genetic polymorphism (MAF 1% or more; e.g., SNP) of the protein of (a) (e.g., a variant not associated with disease listed in the "Variant" column of UniProtKB under accession number P00451; excluding the amino acid mutations at positions K213, S367, and F2196), which protein exhibits a specific activity and secretory expression efficiency equivalent to that of the protein of (a). Here, "equivalent" in specific activity or secretory expression efficiency means that the specific activity or secretory expression efficiency is 0.8 to 1.2 times, preferably 0.9 to 1.1 times, and more preferably 0.95 to 1.05 times that of the protein of (a).
In a preferred embodiment, wild-type human factor VIII is the protein of (a) above.

As used herein, "variant human factor VIII" refers to a protein having an amino acid sequence in which one or more amino acids have been substituted, deleted, inserted, or added in the amino acid sequence (reference sequence) of wild-type human factor VIII (a) or (b) above, and in which the specific activity and/or secretory expression efficiency has been substantially altered compared to human factor VIII consisting of the reference sequence. Here, "substantially altered" preferably means a statistically significant change, but also encompasses cases in which there is a tendency for an increase or decrease even if there is no significant difference.

The variants of the present invention are human factor VIII variants that have higher specific activity and/or secretory expression efficiency compared to the wild-type. The specific activity of human factor VIII can be evaluated by measuring the coagulation factor activity using the publicly known one-stage coagulation assay (OSA) or chromogenic assay (CSA), measuring the protein amount using a publicly known protein quantification method (e.g., immunoassay such as ELISA), and calculating the coagulation factor activity per unit protein amount. The secretory expression efficiency of human factor VIII can also be evaluated by measuring the concentration of human factor VIII in the culture supernatant of cells transfected with a nucleic acid encoding it or in the plasma of animals administered the nucleic acid using a publicly known method (e.g., immunoassay such as ELISA).

In the variants of the present invention, at least one, preferably two (e.g., K213 and S367, K213 and F2196, or S367 and F2196), more preferably three (K213, S367 and F2196) amino acid residues selected from the group consisting of K213, S367, and F2196 of wild-type human factor VIII are substituted with other amino acids. When the wild-type human factor VIII is the protein (b) above and the positions of the amino acid residues differ from those in SEQ ID NO: 1 due to the deletion or insertion/addition of one or more amino acids, the amino acid residues corresponding to K213, S367, and F2196 can be identified as the amino acid residues corresponding to K213, S367, and F2196 in SEQ ID NO: 1, respectively, when the amino acid sequence of (b) and the amino acid sequence of SEQ ID NO: 1 are appropriately aligned using any homology search algorithm.

The other amino acids are not particularly limited, so long as the human factor VIII variant obtained as a result of the amino acid substitution has enhanced specific activity and/or secretory expression efficiency compared to the wild-type. However, the amino acid substitution at K213 is preferably K213N or K213H, with K213N being more preferred. The amino acid substitution at S367 is preferably S367P, S367N or S367Q, with S367P being more preferred. The amino acid substitution at F2196 is preferably F2196L or F2196M, with F2196L being more preferred.

In a particularly preferred embodiment, the variant of the present invention comprises the amino acid substitutions K213 N , S367 P , and F2196 L in wild-type human factor VIII (hereinafter, sometimes referred to as the "NPL mutation" after the substituted amino acid residues). A variant having the NPL mutation is characterized by a higher specific activity than the wild-type, provided that there are no other mutations (combinations of mutations) that would negate the effect of the NPL mutation. As described above, a high-specific-activity variant can achieve coagulation activity equivalent to or greater than that achieved by overexpressing wild-type factor VIII at an appropriate protein expression level that does not impair the long-term stability of the therapeutic effect, and is therefore useful for gene therapy for hemophilia A.

As long as the variant of the present invention maintains the properties of higher specific activity and/or secretory expression efficiency compared to the wild type, in addition to at least one amino acid substitution selected from the group consisting of K213, S367 and F2196, amino acid residues at positions other than K213, S367 and F2196 may be substituted with other amino acids. Examples of such substitution sites include R-5, P25, A28, L152, M217, W228, Q410, Y487, R489, F501, M539, I566, L603, I642, S727, A736, S1657, Q1659, E1661, I1668, D1681, R1776, H1859, A1993, H2007, N2019, K2085, K2207, F2275, S2296, V2314, Q2316, and M2321. Accordingly, in one embodiment, the variant of the present invention is one in which at least one of the amino acid residues is further substituted with another amino acid.

The other amino acids are not particularly limited as long as the human factor VIII variant obtained as a result of the amino acid substitution still has enhanced specific activity and/or secretory expression efficiency compared to the wild-type, but are preferably R-5P, P25H, A28T, L152A, L152D, L152E, L152H, L152I, L152M, L152N, L152P, L152Q or L152V, M217T, W228Q, Q410L, Y487H, and at least one amino acid substitution selected from the group consisting of R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P, Q1659E, E1661K, I1668F, D1681G, R1776K, H1859R, A1993V, H2007Q, N2019K, K2085M, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L.
More preferably, the other amino acid substitution is:
(i) L152A, L152D, L152E, L152H, L152I, L152M, L152N, L152P, L152Q, or L152V, and/or (ii) I1668F and D1681G
Examples include:

A particularly preferred embodiment of the present invention is a variant that contains the amino acid substitution (i) above in addition to the NPL mutation (hereinafter, this may be referred to as an "XNPL mutation," and when L152 is substituted with proline (P), this may be referred to as a "PNPL mutation"). Variants having XNPL mutations, including PNPL mutations, not only have a higher specific activity than the wild-type, but can also have the property of being secreted at a higher level than the wild-type. Therefore, in addition to the advantages conferred by the NPL mutation, these variants may have the further advantage of suppressing the induction of endoplasmic reticulum stress by improving secretion efficiency, thereby realizing long-term stability of the therapeutic effects of human factor VIII.

Another preferred embodiment of the present invention is a variant that includes the amino acid substitution (ii) above (hereinafter sometimes referred to as "FG mutation") in addition to the NPL mutation or XNPL mutation. Compared to the corresponding human factor VIII lacking either the I1668F or D1681G mutation, the variant with the FG mutation exhibits significantly reduced intracellular retention of the light chain, and therefore further improvements in secretion efficiency and/or heavy and light chain reconstitution can be observed.

In a preferred embodiment, the variant of the present invention is
the amino acid substitution at K213 is K213N;
M539L, and optionally at least one amino acid substitution in the A2 domain selected from the group consisting of I566M, L603P, and I642V;
The variant further comprises at least one amino acid substitution in the C2 domain selected from the group consisting of K2207Q, Q2316H, and M2321L, and optionally F2275L, S2296A, and V2314A. By including at least the amino acid substitutions K213N, M539L, K2207Q, Q2316H, and M2321L, the variant can be conferred the property of high secretory expression compared to wild-type human factor VIII, even without the NPL mutation. The secretory expression efficiency of the variant can be further enhanced by further including the FG mutation.

In another preferred embodiment, the variant of the present invention further comprises, in addition to the NPL mutation, amino acid substitutions L152P, Y487H, and L603P, and optionally at least one amino acid substitution selected from the group consisting of S727P, Q1659E, H1859R, A1993V, H2007Q, N2019K, and K2085M. Because the variant contains the PNPL mutation, it has a higher specific activity and is expressed at a higher level of secretion compared to the wild-type.

In another preferred embodiment, the variant of the present invention further comprises the following amino acid substitutions in addition to the NPL mutation: M539L, I566M, L603P, I642V, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L. While the high specific activity characteristic of the NPL mutation may be attenuated by the presence of additional amino acid mutations including the 10 amino acid substitutions, the secretory expression efficiency of the variant is enhanced regardless of the presence or absence of an amino acid substitution at L152 (e.g., L152P) (i.e., even without the XNPL mutation). The secretory expression efficiency of the variant can be further enhanced by further including the FG mutation.

Preferred embodiments of the present invention include modified human factor VIII having the following amino acid substitutions (a) to (n) in wild-type human factor VIII:
(a) K213N, S367P, and F2196L
(b) K213N, S367P, F2196L, I1668F, and D1681G
(c) K213N, M539L, I566M, I642V, K2207Q, F2275L, Q2316H, and M2321L
(d) K213N, M539L, I566M, L603P, I642V, F2196L, K2207Q, F2275L, Q2316H, and M2321L
(e) K213N, M539L, I566M, I642V, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L
(f) K213N, M539L, I566M, I642V, I1668F, D1681G, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L
(g) L152P, K213N, S367P and F2196L
(h) L152P, K213N, S367P, Y487H, L603P and F2196L
(i) L152P, K213N, S367P, Y487H, L603P, S727P, Q1659E, and F2196L
(j) L152P, K213N, S367P, Y487H, L603P, S727P, Q1659E, H1859R, A1993V, H2007Q, K2085M and F2196L
(k) K213N, S367P, M539L, I566M, L603P, I642V, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
(l) P25H, A28T, K213N, M217T, S367P, Q410L, Y487H, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P , Q1659E, E1661K, I1668F, D1681G, R1776K, N2019K, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
(m) R-5P, P25H, A28T, K213N, M217T, W228Q, S367P, Q410L, Y487H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736 V, S1657P, Q1659E, E1661K, I1668F, D1681G, R1776K, N2019K, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
(n) R-5P, P25H, A28T, L152P, K213N, M217T, W228Q, S367P, Q410L, Y487 H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P, Q1 659E, E1661K, I1668F, D1681G, R1776K, H1859R, A1993V, H2007Q, N201 9K, K2085M, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L

The B domain of human factor VIII does not contain regions that interact with various proteins and does not contribute to cofactor activity in circulating blood; therefore, deletion of the B domain retains coagulation factor activity. Therefore, the variants of the present invention may lack the B domain. "B domain deletion" refers not only to the complete deletion of the amino acid sequence of the B domain (positions 741-1648), but also to the retention of an amino acid sequence of approximately 1-15 amino acid residues at the N-terminus and/or C-terminus of the B domain between the a2 and a3 regions. B domain-deleted (BDD) human factor VIII variants are advantageous when using AAV vectors in gene therapy for hemophilia A. Although AAV vectors can carry a small nucleic acid size of 4.7 kbp, full-length human factor VIII consists of 2351 amino acids, so the coding region alone exceeds 7 kbp and cannot be carried by AAV vectors. On the other hand, the coding sequence for BDD-human factor VIII is less than 4.4 kbp, and the entire expression cassette, including the promoter sequence, fits within the packaging size of an AAV vector.

In a preferred embodiment, the variant of the present invention lacks positions 744 to 1637 of the B domain (corresponding to positions 763 to 1656 in the amino acid sequence represented by SEQ ID NO: 1) (i.e., S743 is linked to Q1638; this may be referred to as "FVIII SQ" or "SQ"). BDD-human factor VIII that leads to efficient human factor VIII processing is known, other than FVIII SQ, including deletions of the B domain that leave portions of both termini of the B domain, or substitutions of the B domain with three or four arginine (R) residues (e.g., Lind et al., Eur. J. Biochem. 232, 19-27 (1995)), and these can also be used in the present invention.

On the other hand, since the B domain is known to be involved in the intracellular transport of factor VIII, the variants of the present invention may contain all or part of the B domain in order to improve secretion efficiency. For example, when using lentivirus (LV) vectors or adenovirus (AdV) vectors with larger packaging sizes in gene therapy, when using mRNA encoding the variants of the present invention as mRNA pharmaceuticals, or when recombinantly producing the variants of the present invention and using them as protein preparations, it is also preferable to use variants that contain the B domain.

The variants of the present invention may be expressed in the animal to be treated by administering a nucleic acid encoding them in vivo or ex vivo to the animal, or they may be recombinant proteins obtained by introducing the nucleic acid into suitable host cells, allowing it to be expressed, and culturing the host cells. In either case, the variants of the present invention are produced based on the nucleic acid that encodes them, and the present invention also provides nucleic acids encoding any of the above variants of the present invention.

The "nucleic acid encoding a human factor VIII variant" used in the present invention may be DNA or RNA, or may be a DNA/RNA chimera. The nucleic acid may be double-stranded or single-stranded. If double-stranded, it may be double-stranded DNA, double-stranded RNA, or a DNA:RNA hybrid. If single-stranded, it may be a sense strand (i.e., coding strand) or an antisense strand (i.e., non-coding strand). The type of nucleic acid can be appropriately selected depending on its application (such as the type of vector used), but is preferably DNA, more preferably double-stranded DNA. The nucleic acid may also be a physiologically acceptable salt with an acid or base, for example, a physiologically acceptable acid addition salt. Among such salts, salts with inorganic acids (e.g., hydrochloric acid, phosphoric acid, hydrobromic acid, sulfuric acid) or salts with organic acids (e.g., acetic acid, formic acid, propionic acid, fumaric acid, maleic acid, succinic acid, tartaric acid, citric acid, malic acid, oxalic acid, benzoic acid, methanesulfonic acid, benzenesulfonic acid) are used.
DNA encoding human factor VIII variants includes genomic DNA, cDNA (cRNA) derived from cells or tissues of humans or other mammals, and synthetic DNA (RNA).

More specifically, examples of "nucleic acids encoding variants of the present invention" include any nucleotide sequence encoding a variant amino acid sequence having any of the above-mentioned amino acid substitutions in the amino acid sequence of wild-type human factor VIII represented by SEQ ID NO: 1, or in the amino acid sequence of a natural allelic variant or genetic polymorphism (e.g., SNP, etc.) of a protein consisting of said amino acid sequence. Preferably, the nucleotide sequence is codon-optimized for expression in the host cell to be used, preferably a human cell. During gene expression, converting the nucleotide sequence to codons frequently used in the host organism can be expected to increase the amount of protein expression. Data on codon usage in the host to be used can be obtained, for example, from the genetic code usage database published on the website of the Kazusa DNA Research Institute (http://www.kazusa.or.jp/codon/index.html), or references listing codon usage in each host can be referenced. Alternatively, as used in the examples described below, codon optimization can be performed using a publicly known codon optimization algorithm (e.g., GeneArt Codon Optimizer). Such algorithms can take into account multiple parameters, such as GC content, removal of destabilizing RNA elements, removal of cryptic splice sites, removal of intragenic polyA sites, removal of repetitive sequences, avoidance of RNA secondary structures, and removal of IRES, in addition to frequency of codon usage in the host. DNA encoding the desired variant of the present invention can be constructed by chemically synthesizing a DNA strand with the codon-optimized sequence obtained as described above, or by connecting chemically synthesized, partially overlapping short oligo-DNA strands using PCR or Gibson Assembly.

In a preferred embodiment, the nucleic acid encoding the variant of the present invention is a wild-type human factor VIII consisting of the amino acid sequence represented by SEQ ID NO: 1, wherein (n) R-5P, P25H, A28T, L152P, K213N, M217T, W228Q, S367P, Q410L, Y487H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P, An example of such a nucleic acid is a nucleic acid consisting of the nucleotide sequence represented by SEQ ID NO: 2, which is codon-optimized for human use and encodes a variant having the amino acid substitutions Q1659E, E1661K, I1668F, D1681G, R1776K, H1859R, A1993V, H2007Q, N2019K, K2085M, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L.

A nucleotide sequence codon-optimized for a target host (humans in the case of use in gene therapy for hemophilia A or mRNA pharmaceuticals) may contain one or more CpG sequences within the sequence. However, it is predicted that CpG sequences may induce an immune response in the host and inhibit continuous transgene expression. Therefore, by making base substitutions to remove these CpG sequences without changing the amino acid sequence of the variant, and then re-optimizing the variant, the expression efficiency within host cells can be significantly increased. For example, a nucleic acid encoding a variant codon-optimized for humans, consisting of the nucleotide sequence represented by SEQ ID NO: 2, can be modified so that it does not contain any CpG sequences without changing the amino acid sequence it encodes. An example of such a nucleotide sequence from which CpG has been removed and which has been re-codon-optimized is the nucleotide sequence represented by SEQ ID NO: 3.

The improvement in expression efficiency achieved by removing CpG sequences is effective not only for the specific human factor VIII variants described above, but also for any variants, and even for nucleic acids encoding wild-type human factor VIII. Therefore, the present invention also provides a human factor VIII-encoding nucleic acid comprising a nucleotide sequence obtained by modifying a nucleic acid encoding human factor VIII that has been codon-optimized for a target host, preferably for humans, so that it does not contain CpG sequences without changing the amino acid sequence it encodes, and then further codon-optimizing the nucleotide sequence. For example, an example of a nucleotide sequence that has been codon-optimized for humans and encodes wild-type human factor VIII consisting of the amino acid sequence represented by SEQ ID NO: 1 is the nucleotide sequence represented by SEQ ID NO: 4. An example of a nucleotide sequence obtained by modifying the sequence so that it does not contain CpG sequences without changing the amino acid sequence it encodes, and then further codon-optimizing the nucleotide sequence for humans is the nucleotide sequence represented by SEQ ID NO: 5.

A nucleic acid encoding a variant of the present invention, or a nucleic acid encoding human factor VIII whose expression efficiency in host cells has been enhanced by codon optimization and removal of CpG sequences (hereinafter collectively referred to as the "nucleic acid of the present invention") can be linked downstream of a promoter functional in liver cells and inserted into a vector to construct an expression vector that can be expressed in liver cells.

The liver is composed of hepatic parenchymal cells (hepatocytes), which are responsible for the liver's main functions, such as bile production and metabolism, as well as non-parenchymal liver cells such as hepatic sinusoidal endothelial cells, Kupffer cells, hepatic stellate cells, pit cells, bile duct epithelial cells, and mesothelial cells. As used herein, "liver cells" refers to one or more of the above cell groups that make up the liver (including cancer cells and cell lines derived from them).

As used herein, a "promoter functional in liver cells" refers to a promoter capable of inducing transcription of a gene (a nucleic acid encoding a protein) linked downstream in liver cells. However, considering that the preferred use of the present invention is in vivo gene therapy, it is desirable that the promoter function specifically in liver cells (inducing transcription of a downstream gene). Here, the terms "specifically in liver cells" and "liver-specific" are used to encompass not only cases in which expression is limited to liver cells, but also cases in which expression in liver cells is significantly higher than expression in cells of other organs or tissues. Even when inducing gene expression in cells other than liver cells, for example, by combining it with a vector with high organ/tissue/cell tropism (e.g., various serotypes of viral vectors), substantially liver-specific gene expression (e.g., to the extent that it does not cause undesirable side effects) can be achieved.
As long as the promoter contains a minimal nucleotide sequence that exerts basal transcriptional activity, it may further contain other regulatory sequences (e.g., endogenous proximal or distal enhancer sequences, enhancer sequences derived from other genes, etc.).

Promoters functional in liver cells are not particularly limited as long as they can induce transcription of downstream genes in liver cells. However, they are preferably promoters of genes highly expressed in the liver, and more preferably liver-specific promoters. Here, "liver-specific" has the same meaning as above. Examples of promoters functional in liver cells that can be used in the expression vectors of the present invention include, but are not limited to, the transthyretin (TTR) promoter, the α1-antitrypsin (AAT) promoter, the albumin promoter, the α-fetoprotein promoter, and the thyroxine-binding globulin promoter. In a preferred embodiment, for example, when using an AAV vector for gene therapy of hemophilia A, the TTR promoter or the AAT promoter can be used as a promoter functional in liver cells. In another preferred embodiment, a chimeric promoter (HCRhAAT) composed of the human AAT promoter and the liver-type regulatory region (HCR) of the Apo E/C1 gene can also be used as a promoter functional in liver cells.

Promoters functional in liver cells can be obtained by known methods based on the sequence information of the gene (genomic DNA) from which each promoter is derived. Such sequence information is registered in publicly available databases such as NCBI, EMBL, FASTA, and DDBJ, and is also described in literature. Those skilled in the art can easily access this sequence information.

If the nucleotide sequence of a promoter functional in liver cells contains one or more CpG sequences, it is desirable to modify it to remove the CpG sequences, as in the case of the nucleic acid of the present invention, in order to enhance the expression efficiency in host cells of the nucleic acid of the present invention linked downstream. For example, while the wild-type mouse TTR promoter consists of the nucleotide sequence represented by SEQ ID NO: 6, a modified promoter can be used in which the CpG sequences in the wild-type mouse TTR promoter have been removed, and the nucleotide sequence represented by SEQ ID NO: 7.

The linking of a promoter functional in liver cells to the nucleic acid of the present invention can be carried out by methods known in the art. For example, if the promoter and nucleic acid fragments each have blunt ends, the two fragments can be ligated using DNA ligase. Alternatively, any adapter sequence can be added to the ends of both fragments, treated with an appropriate restriction enzyme to generate sticky ends, and then the two can be ligated.

In a preferred embodiment, the nucleic acid of the present invention linked downstream of a promoter functional in liver cells is preferably inserted into a vector in the form of an expression cassette further linked downstream of a transcription termination signal functional in liver cells, i.e., a polyA addition signal (e.g., an SV40 polyA addition signal).

For example, (i) downstream of a mouse TTR promoter (SEQ ID NO: 7) that has been modified so as not to contain a CpG sequence, (ii) in wild-type human factor VIII, (n) R-5P, P25H, A28T, L152P, K213N, M217T, W228Q, S367P, Q410L, Y487H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P, Q1659E, E1661K, I1668F, D1681G, R177 An expression cassette that encodes a variant having the amino acid substitutions 6K, H1859R, A1993V, H2007Q, N2019K, K2085M, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L, that has been codon-optimized for human use, further CpG sequence-removed, and re-codon-optimized (SEQ ID NO: 3), and (iii) is linked to an SV40 polyA additional signal sequence, can be preferably used. An example of such an expression cassette is a nucleic acid consisting of the nucleotide sequence (4814 bp) set forth in SEQ ID NO: 8. This nucleic acid contains a 17-nucleotide 5'-upstream sequence (positions 283-299) between the promoter sequence (positions 1-282) of SEQ ID NO: 7 and the coding sequence (positions 300-4673) of SEQ ID NO: 3. By deleting and shortening the 11 nucleotides at positions 283-293, the efficiency of secretory expression of human factor VIII can be improved. SEQ ID NO: 9 shows the nucleotide sequence (4803 bp) in which these 11 nucleotides have been deleted from the nucleotide sequence of SEQ ID NO: 8. Furthermore, SEQ ID NO: 10 shows the nucleotide sequence (4790 bp) in which 13 nucleotides from the 3' end of the SV40 polyA addition signal sequence (positions 4663-4803) of SEQ ID NO: 9 have been deleted, and SEQ ID NO: 11 shows the nucleotide sequence (4768 bp) in which 35 nucleotides from the 3' end have been deleted. By modifying (shortening) these polyA addition signals, the efficiency of secretory expression of human factor VIII can be further improved.

The vector into which the nucleic acid of the present invention is inserted is not particularly limited as long as it is one generally used in gene therapy. Examples of suitable vectors include viral vectors such as AAV vectors, LV vectors, retroviral vectors, AdV vectors, Sindbis virus vectors, rabies virus vectors, Sendai virus vectors, and herpes simplex virus vectors, as well as non-viral vectors such as plasmids for animal cells. AAV and LV vectors are preferred from the viewpoints of high gene transfer and expression efficiency, ability to transfer to non-dividing cells, and long-term expression of the introduced gene. AAV vectors are more preferred from the viewpoints of low frequency of chromosomal integration, elimination of the risk of insertional mutagenesis, low immunogenicity, and high safety. AAV vectors are diluted in dividing cells as the cells proliferate, but adult hepatocytes rarely divide, so expression of the introduced gene can be maintained for a long period of time. However, because the gene size that can be carried is small at 4.7 kbp, depending on the size of the nucleic acid to be introduced, it may be preferable to use LV or AdV vectors, which can carry larger inserts.

For example, when using an AAV vector as a vector, it is desirable to use a vector derived from a serotype with high liver tropism. Examples of serotypes with liver tropism include AAV1, AAV2, AAV3, AAV5, AAV6, AAV6.2, AAV7, AAV8, AAV9, AAVrhlO, AAV-DJ, and AAV-DJ/8. Below, the procedure for constructing a vector will be explained using an AAV vector as an example. However, those skilled in the art will be able to construct other viral vectors and non-viral vectors (e.g., plasmid vectors) using methods known per se [e.g., Current Protocols in Molecular Biology, F. Ausubel et al. eds. (1994) John Wiley & Sons, Inc.; Molecular Cloning (A Laboratory Manual), 3rd ed. Volumes 1-3, Joseph Sambrook & David W. Russell, Cold Spring Harbor Laboratory Press (Cold Spring Harbor, New York) (2001); Culture of Animal Cells; A Manual of Basic The desired expression vector can be easily constructed using methods such as those described in "Adenovirus Technique," R. Freshney eds., 2nd ed. (1987), Wiley-Liss; Frank L. Graham, Manipulation of adenovirus vectors, Chapter 11, pp. 109-128; E.J. Murray eds., Methods in Molecular Biology, Vol. 7, Gene Transfer and Expression Protocols (1991); Chen, S-H. et al., Combination gene therapy for liver metastases of colon carcinoma in vivo, Proc. Natl. Acad. Sci. USA (1995) 92, 2477-2581.

In constructing an AAV expression vector, the synthetic promoter of the present invention is first inserted between the 5'- and 3'-ITRs of AAV into a plasmid that can be amplified in a suitable host cell (e.g., Escherichia coli, Bacillus subtilis, yeast, etc.). Examples of such plasmids include E. coli-derived plasmids (e.g., pBR322, pBR325, pUC12, pUC13), Bacillus subtilis-derived plasmids (e.g., pUB110, pTP5, pC194), and yeast-derived plasmids (e.g., pSH19, pSH15). While AAV ITRs derived from the desired serotype may be used, AAV2-derived ITRs are commonly used. While the synthetic promoter of the present invention alone can be inserted between the two ITR sequences, it is preferable to insert a polyA addition signal (e.g., an SV40 polyA addition signal) downstream thereof, and a woodchuck hepatitis virus post-transcriptional regulatory element (WPRE) can also be inserted between the synthetic promoter and the polyA addition signal. A shortened WPRE (0.25 kb; Mol. Brain, 7:17, 2014) can also be used. Inserting one or more restriction enzyme recognition sites, preferably a multicloning site (MCS), between the promoter of the present invention and the polyA addition signal (or between the promoter and WPRE if the WPRE is included) can facilitate insertion of the nucleic acid of the present invention (construction of an expression cassette for human factor VIII).

By inserting the nucleic acid of the present invention to be expressed in liver cells between the promoter and polyA addition signal of the resulting AAV vector expression plasmid (between the promoter and WPRE if WPRE is included), an expression cassette for human factor VIII is constructed in the plasmid.

If desired, the expression vector can further contain a 5'-UTR downstream of the promoter that can function in the host, and a 3'-UTR downstream of the DNA encoding human factor VIII that can function in the host. It may also contain an enhancer, a splicing signal, etc. Each of these components can be made using known components.

An AAV vector can be produced from an AAV vector expression plasmid containing the nucleic acid of the present invention downstream of a promoter functional in liver cells by methods known per se, such as the plasmid transfection method, recombinant baculovirus method, recombinant herpes virus vector method, yeast method, etc. For example, in the plasmid transfection method, the AAV vector expression plasmid, a plasmid containing the AAV Rep and Cap genes, and a pHelper plasmid containing the adenovirus-derived E2A, E4orf6, and VARNA genes are transfected into HEK293 cells or the like to produce AAV virus particles. The Rep gene does not need to be derived from the desired serotype; the Rep gene from AAV2 is generally used. On the other hand, the Cap gene must be derived from the desired serotype, and the cell tropism of the AAV serotype is determined by the capsid protein encoded by the Cap gene. That is, in the expression vector of the present invention that targets liver cells, Cap genes derived from liver-tropic viruses such as AAV1, AAV2, AAV3, AAV5, AAV6, AAV6.2, AAV7, AAV8, AAV9, AAVrhlO, AAV-DJ, and AAV-DJ/8 can be used as Cap genes.

On the other hand, when a non-viral vector is used as the expression vector of the present invention, the expression vector can be introduced using a polymeric carrier such as a poly-L-lysine-nucleic acid complex, or by encapsulation in a liposome. Liposomes are capsules made of phospholipids with a particle size of several tens to several hundreds of nanometers, and a plasmid vector can be encapsulated inside them.

An expression vector containing a nucleic acid of the present invention downstream of a promoter functional in liver cells can be administered to a subject suffering from a disease for which human factor VIII exerts a therapeutic effect, particularly hemophilia A, to treat the disease. Therefore, the present invention also provides a gene therapy agent for hemophilia A comprising the expression vector of the present invention.

The gene therapy agent of the present invention may be the expression vector of the present invention used as is, or, if necessary, may be mixed with a pharmacologically acceptable carrier to form various formulations, such as injections, and then used as a medicine.

Here, various organic or inorganic carrier substances commonly used as pharmaceutical ingredients are used as pharmacologically acceptable carriers, and are incorporated as solvents, solubilizers, suspending agents, isotonicity agents, buffers, soothing agents, etc. in liquid formulations. Furthermore, formulation additives such as preservatives, antioxidants, and coloring agents can also be used as needed.

Suitable examples of solvents include water for injection, physiological saline, Ringer's solution, alcohol, propylene glycol, polyethylene glycol, sesame oil, corn oil, olive oil, cottonseed oil, etc.

Suitable examples of solubilizing agents include polyethylene glycol, propylene glycol, D-mannitol, trehalose, benzyl benzoate, ethanol, trisaminomethane, cholesterol, triethanolamine, sodium carbonate, sodium citrate, sodium salicylate, and sodium acetate.

Suitable examples of suspending agents include surfactants such as stearyl triethanolamine, sodium lauryl sulfate, lauryl aminopropionic acid, lecithin, benzalkonium chloride, benzethonium chloride, and glycerin monostearate; hydrophilic polymers such as polyvinyl alcohol, polyvinylpyrrolidone, sodium carboxymethylcellulose, methylcellulose, hydroxymethylcellulose, hydroxyethylcellulose, and hydroxypropylcellulose; polysorbates; and polyoxyethylene hydrogenated castor oil.

Suitable examples of isotonic agents include sodium chloride, glycerin, D-mannitol, D-sorbitol, glucose, etc.

Suitable examples of buffering agents include buffer solutions such as phosphate, acetate, carbonate, and citrate.

Suitable examples of soothing agents include benzyl alcohol.

Suitable examples of preservatives include parahydroxybenzoic acid esters, chlorobutanol, benzyl alcohol, phenethyl alcohol, dehydroacetic acid, and sorbic acid.

Suitable examples of antioxidants include sulfites and ascorbates.

Suitable examples of coloring agents include water-soluble food tar dyes (e.g., food dyes such as Food Red No. 2 and No. 3, Food Yellow No. 4 and No. 5, and Food Blue No. 1 and No. 2), water-insoluble lake dyes (e.g., aluminum salts of the above-mentioned water-soluble food tar dyes), and natural dyes (e.g., β-carotene, chlorophyll, red iron oxide, etc.).

The dosage form of the pharmaceutical composition may be, for example, an injection (e.g., subcutaneous injection, intravenous injection, intramuscular injection, intraperitoneal injection, etc.), a drip infusion, or other parenteral preparation.

The gene therapy agent of the present invention can be produced by methods commonly used in the pharmaceutical technology field, such as the methods described in the Japanese Pharmacopoeia. The content of the viral vector, which is the active ingredient in the formulation, varies depending on the dosage form, the dose of the active ingredient, etc., but is, for example, about 0.1 to 100% by weight. The viral titer can be appropriately adjusted to, for example, about 1 x 10 ¹⁰ to 10 ¹³ vp/mL, but is not limited to this range.

Preparations suitable for parenteral administration (e.g., intravenous injection, subcutaneous injection, intramuscular injection, intraperitoneal administration, etc.) include aqueous and non-aqueous isotonic sterile injection solutions, which may contain antioxidants, buffers, bacteriostats, isotonicity agents, etc. Also included are aqueous and non-aqueous sterile suspensions, which may contain suspending agents, solubilizers, thickeners, stabilizers, preservatives, etc. The most preferred dosage form in the present invention is an injection solution.

The dosage of the formulation varies depending on the type of vector, promoter activity, administration route, severity of the disease, target animal species, drug tolerance, body weight, age, etc. of the target animal. For example, when an AAV vector (e.g., AAV8) with the HCRhAAT promoter as a basal promoter is used as a human factor VIII expression vector for hemophilia A, when the AAV vector is administered systemically, particularly via a peripheral vein, it can be administered at a single dose of, for example, about 5 × 10 ¹² to about 5 × 10 ¹³ vp/kg body weight. In recent clinical trials of AAV vectors, deaths due to severe liver damage were reported in the high-dose (3 × 10 ¹⁴ vp/kg body weight) administration group (Audentes Therapeutics. Letter to the MTM disease community. https://myotubulartrust.org/audentes-therapeutics-letter-23-june-2020/; Hum Gene Ther 2020; 31: 695-696), so it is desirable to administer at a dose of 10 ¹⁴ vp/kg body weight or less. For example, in one preferred embodiment, the gene therapy agent of the present invention can be systemically administered at a dose of about 1×10 ¹⁰ to about 5×10 ¹³ vp/kg body weight, preferably about 1×10 ¹¹ to about 1×10 ¹³ vp/kg body weight, more preferably about 5×10 ¹¹ to about 5×10 ¹² vp/kg body weight or about 1×10 ¹² to about 1×10 ¹³ vp/kg body weight, in terms of the amount of AAV vector.

The present invention also provides a method for expressing human factor VIII in liver cells, which comprises introducing into liver cells in vitro a gene therapy agent containing the expression vector of the present invention or the nucleic acid of the present invention. By transplanting liver cells transfected with the gene therapy agent of the present invention into a subject with hemophilia A, an ex vivo treatment for the disease can be provided. Therefore, the present invention also provides a pharmaceutical (cell preparation), particularly a therapeutic agent for hemophilia, containing liver cells transfected with the gene therapy agent of the present invention as an active ingredient.

When the nucleic acid of the present invention is single-stranded RNA, the nucleic acid can be formulated in the form of an mRNA pharmaceutical. For example, the expression vector is introduced into a suitable host (e.g., mammalian cells) and cultured, and the mRNA is recovered using a method known per se (e.g., the LiCl method), and the mRNA encoding human factor VIII can be purified to obtain the mRNA.

Alternatively, the mRNA can be obtained by excising the human factor VIII coding sequence (including the 5'- and 3'-UTRs in addition to the CDS) from the expression vector and using it as a template to convert it into mRNA encoding this human factor VIII using a known in vitro transcription system. More specifically, the mRNA coding region is excised using an appropriate restriction enzyme and a phage (T7, T3, SP6, etc.) promoter is ligated to the 5' end, or a fragment of the mRNA coding region linked to the phage promoter is obtained by using the expression vector as a template and performing PCR with primers containing the phage promoter sequence. The resulting DNA fragment can be used as a template to react with phage (T7, T3, SP6, etc.) RNA polymerase to synthesize mRNA encoding human factor VIII in vitro. In this case, by adding pseudouridine (Ψ) or N1-methylpseudouridine (N1mΨ) triphosphate instead of UTP as an RNA monomer (NTP) to the reaction solution, the resulting mRNA will contain Ψ or N1mΨ instead of U. This allows the mRNA to avoid attack by natural immunity, improving mRNA stability and translation efficiency. Similarly, modified NTPs with base substitutions that have been reported to avoid natural immunity (e.g., inosine triphosphate instead of ATP, 5-methylcytidine triphosphate instead of CTP, etc.) can also be used for other NTPs.

It is also desirable to add a 5'-cap structure and a polyA tail, which are necessary for stabilizing mRNA and improving translation efficiency. The 5'-cap structure can be achieved by adding a Cap 0 structure after mRNA synthesis using a capping enzyme, and then converting it to a Cap 1 structure using an mRNA 2'-O-methyltransferase. Alternatively, transcription and 5'-capping can be performed simultaneously by adding an RNA cap analog (e.g., 3'-O-Me- ^m7G (5')ppp( ⁵ ')G, ^m7G (5')ppp(5')G, ^3'- O-Me-m7G(5')ppp(5')A, m7G(5')ppp(5')A, etc.) to the transcription reaction solution. PolyA tailing can also be added to the 3' end of mRNA after synthesis using polyA polymerase, or it can be performed simultaneously with the transcription reaction by adding a polyA sequence to the transcription template in advance. The mRNA obtained as described above can be purified by removing the template DNA with DNase I and then by, for example, the LiCl method.

A nucleic acid transfer reagent may be used to promote the transfer of the nucleic acid of the present invention into target cells. Examples of such nucleic acid transfer reagents include atelocollagen; liposomes; nanoparticles; lipofectin, lipofectamine, DOGS (transfectam), DOPE, DOTAP, DDAB, DHDEAB, HDEAB, polybrene, and cationic lipids such as poly(ethyleneimine) (PEI).

In one embodiment, the nucleic acid of the present invention may be a pharmaceutical composition encapsulated in a liposome. Liposomes are small, closed vesicles with an internal phase surrounded by one or more lipid bilayers, and can typically hold a water-soluble substance in the internal phase and a lipid-soluble substance within the lipid bilayer. The term "encapsulated" used herein refers to the nucleic acid of the present invention being held in the internal phase of the liposome or within the lipid bilayer. The liposomes used in the present invention may be monolayer or multilayer membranes, and the particle size can be appropriately selected, for example, from the range of 10 to 1,000 nm, preferably 50 to 300 nm. Considering delivery to target tissues, the particle size may be, for example, 200 nm or less, preferably 100 nm or less.

Methods for encapsulating water-soluble compounds such as polynucleotides into liposomes include, but are not limited to, the lipid film method (vortex method), reverse phase evaporation, surfactant removal, freeze-thaw method, and remote loading method, and any known method can be selected as appropriate.

In a preferred embodiment, the nucleic acid of the present invention is carried on a lipid nanoparticle and is encapsulated within the particle.
As used herein, "lipid nanoparticles" (sometimes abbreviated as "LNPs" in the present specification) refer to particles having a membrane structure in which the hydrophilic groups of amphipathic lipids are aligned toward the aqueous phase side of the interface, and having a particle diameter of less than 1 μm, and "amphipathic lipid" refers to a lipid having both a hydrophilic group and a hydrophobic group.

The particle diameter of the lipid nanoparticles used in the present invention is preferably 10 nm to 500 nm, and more preferably 30 nm to 300 nm. Particle diameter can be measured using a particle size distribution analyzer such as a Zetasizer Nano (Malvern). The particle diameter of the lipid nanoparticles can be adjusted appropriately depending on the manufacturing method of the lipid nanoparticles. In this specification, "particle diameter" refers to the average particle diameter (zeta mean) measured by dynamic light scattering.

Amphipathic lipids include, for example, cationic lipids, ionic lipids, phospholipids, and PEG lipids. As used herein, "cationic lipid" refers to a lipid having a constitutively positively charged hydrophilic group. As used herein, "ionizable lipid" refers to a lipid that is neutral at physiological pH but becomes protonated and positively charged at low pH. As used herein, "PEG" refers to polyethylene glycol, and "PEG lipid" refers to a lipid modified with PEG, i.e., a lipid to which PEG is bound.

In a preferred embodiment, the lipid nanoparticles in which the nucleic acid of the present invention is encapsulated are:
(A) a cationic lipid or an ionic lipid,
(B) phospholipids,
(C) steroids, and (D) lipid nanoparticles containing PEG lipids. Each of the component lipids (A) to (D) includes known lipids that are commonly used in lipid nanoparticles, and a person skilled in the art can easily select an appropriate type of lipid and its composition.

The present invention also provides host cells into which an expression vector containing the nucleic acid of the present invention has been introduced; a method for producing human factor VIII or a modified factor thereof, which comprises culturing the host cells and recovering human factor VIII or a modified factor thereof from the resulting culture; and a pharmaceutical (recombinant protein preparation), particularly a therapeutic agent for hemophilia A, containing the modified human factor VIII obtained by the method.

An expression vector for producing recombinant human factor VIII can be produced by linking a nucleic acid of the present invention (i.e., a nucleic acid encoding a variant of the present invention, or a nucleic acid encoding a highly expressed form of human factor VIII) downstream of a promoter in an appropriate expression vector. Expression vectors that can be used include plasmids derived from Escherichia coli (e.g., pBR322, pBR325, pUC12, pUC13); plasmids derived from Bacillus subtilis (e.g., pUB110, pTP5, pC194); yeast-derived plasmids (e.g., pSH19, pSH15); insect cell expression plasmids (e.g., pFast-Bac); animal cell expression plasmids (e.g., pA1-11, pXT1, pRc/CMV, pRc/RSV, pcDNAI/Neo); bacteriophages such as λ phage; insect virus vectors such as baculovirus (e.g., BmNPV, AcNPV); and animal virus vectors such as retrovirus, lentivirus, vaccinia virus, adenovirus, adeno-associated virus, and herpes virus.

Any promoter may be used as long as it is appropriate for the host used to express the gene. For example, when the host is an animal cell, the SRα promoter, SV40 promoter, LTR promoter, CMV (cytomegalovirus) promoter, RSV (Rous sarcoma virus) promoter, MoMuLV (Moloney murine leukemia virus) LTR, HSV-TK (herpes simplex virus thymidine kinase) promoter, etc. may be used. For other hosts, a publicly known promoter may also be selected as appropriate.

In addition to the above, expression vectors can also contain, as desired, enhancers, splicing signals, polyA addition signals, selection markers, SV40 origin of replication (hereinafter sometimes abbreviated as SV40 ori), etc. Examples of selection markers include the dihydrofolate reductase gene, ampicillin resistance gene, and neomycin resistance gene.

Human factor VIII or a modified form thereof can be produced by transforming a host with an expression vector containing the nucleic acid of the present invention and culturing the resulting transformant. Examples of hosts that can be used include Escherichia bacteria, Bacillus bacteria, yeast, insect cells, insects, and animal cells. Examples of mammalian cells that can be used include monkey COS-7 cells, monkey Vero cells, Chinese hamster ovary cells (hereinafter abbreviated as CHO cells), dhfr gene-deficient CHO cells (hereinafter abbreviated as CHO(dhfr ⁻ ) cells), mouse L cells, mouse AtT-20 cells, mouse myeloma cells, rat GH3 cells, human FL cells, HeLa cells, HepG2 cells, and HEK293 cells. For other hosts, cells known per se can be appropriately selected.

Transformation can be carried out according to known methods depending on the type of host. Animal cells can be transformed, for example, according to the methods described in Cell Engineering Special Issue 8, New Cell Engineering Experimental Protocols, pp. 263-267 (1995) (published by Shujunsha) and Virology, Vol. 52, p. 456 (1973).

The transformant can be cultured according to a known method depending on the type of host. For example, when the host is an animal cell, a medium such as minimum essential medium (MEM) containing about 5 to about 20% fetal bovine serum, Dulbecco's modified Eagle's medium (DMEM), RPMI 1640 medium, 199 medium, or Ham's F-12 medium can be used. The pH of the medium is preferably about 6 to about 8. The culture is usually carried out at about 30°C to about 40°C for about 15 to about 60 hours. Aeration or stirring may be performed as necessary.
In this manner, human factor VIII can be produced intracellularly or extracellularly in the transformant.

Human factor VIII can be separated and purified from the culture obtained by culturing the transformant using methods known per se. Examples of such methods include methods that utilize solubility, such as salting out and solvent precipitation; methods that primarily utilize differences in molecular weight, such as dialysis, ultrafiltration, gel filtration, and SDS-polyacrylamide gel electrophoresis; methods that utilize differences in charge, such as ion exchange chromatography; methods that utilize specific affinity, such as affinity chromatography; methods that utilize differences in hydrophobicity, such as reversed-phase high-performance liquid chromatography; and methods that utilize differences in isoelectric point, such as isoelectric focusing. These methods can also be combined as appropriate.

If the human factor VIII thus obtained is in the free form, the free form can be converted into a salt by a method known per se or a method similar thereto. If the human factor VIII is obtained as a salt, the salt can be converted into the free form or another salt by a method known per se or a method similar thereto.

Human factor VIII is low in toxicity and can be administered to humans or other mammals parenterally (e.g., intravascular administration (intravenous administration, intraarterial administration, etc.), subcutaneous administration, intradermal administration, intraperitoneal administration, intramuscular injection, topical administration, etc.) either as a liquid preparation or as a pharmaceutical composition in an appropriate dosage form.

Compositions for parenteral administration include, for example, injections and suppositories, and injections may include dosage forms such as intravenous injections, subcutaneous injections, intradermal injections, intramuscular injections, and drip infusion injections. Such injections can be prepared according to known methods. For example, injections can be prepared by dissolving, suspending, or emulsifying one or more active ingredients in a sterile aqueous or oily liquid typically used for injections. Examples of aqueous solutions for injection include saline, isotonic solutions containing glucose and other adjuvants, and may be used in combination with appropriate solubilizers, such as alcohols (e.g., ethanol), polyalcohols (e.g., propylene glycol, polyethylene glycol), and nonionic surfactants (e.g., polysorbate 80, HCO-50 (polyoxyethylene (50 mol) adduct of hydrogenated castor oil)). As the oily liquid, for example, sesame oil, soybean oil, etc. are used, and solubilizing agents such as benzyl benzoate and benzyl alcohol may be used in combination. The prepared injection solution is preferably filled into appropriate ampoules. Suppositories for rectal administration may be prepared by mixing the active ingredient with a standard suppository base.

The above-mentioned parenteral pharmaceutical compositions are conveniently prepared in dosage unit forms that correspond to the dosage of the active ingredient. Examples of such dosage unit forms include tablets, pills, capsules, injections (ampoules), and suppositories, and each dosage unit typically contains 100 to 5,000 units, preferably 250 to 3,000 units.

The dosage of human factor VIII varies depending on the recipient, symptoms, and route of administration. For example, when used to treat hemophilia A, a single dose of approximately 10-100 units/kg body weight, preferably approximately 20-50 units/kg body weight, is typically administered intravenously or intraperitoneally, preferably two to three times a week. Similar amounts can also be administered for other parenteral administrations. When symptoms are particularly severe, the dosage may be increased accordingly.

The present invention will be explained in more detail below using examples, but these are merely illustrative and do not in any way limit the scope of the present invention.

In the following examples, all animal experimental protocols were approved by the Jichi Medical University Animal Care and Related Institutional Committee, and animal care followed the committee's guidelines. Mice were purchased from Japan SLC. Factor VIII-deficient mice were obtained from Jackson Laboratory (USA).

Example 1: Preparation of nucleic acids encoding B domain-deleted human factor VIII (hFVIIISQ) and its variants, and coagulation factor activity of each variant
hFVIISQ
The DNA sequence of human coagulation factor VIII (hFVIIISQ) with most of the B domain removed was designed based on the NCBI reference sequence (NM_000132.4), and codon-optimized and gene-synthesized using GeneArt (Thermo Scientific) (SEQ ID NO: 4).
Ver. 1-Ver. 12
A variant, Ver. 1, was prepared by gene synthesis and mutagenesis PCR based on the sequences of hFVIIISQ, canine FVIIISQ, porcine FVIIISQ, ovine FVIIISQ, and bovine FVIIISQ, altering 72 amino acid sites to those of canine FVIIISQ (codon-optimized for human use). Furthermore, variants Ver. 2 and Ver. 3 were prepared by reducing the number of substitutions to 58 and 50, respectively. Ver. 4, which contains only the amino acid substitutions common to Ver. 2 and Ver. 3, was also prepared. Ver. 5, which contains 56 substitutions, was prepared by removing the overlap with the X5 and X10 mutations previously reported (WO 2014/209942). Ver. 6 and Ver. 7, which contain 35 substitutions, were prepared by removing the overlap with the X5 mutation from Ver. 4. Ver. 7, which has 51 substitutions, was created by removing overlaps with the JF12 mutation previously reported (Mol Ther Methods Clin Dev. 2020 Jan 15; 17:328-336) from Ver. 2; Ver. 8, which has 32 substitutions, was created by removing overlaps with the JF12 mutation from Ver. 4; Ver. 9, which has 49 substitutions, was created by removing overlaps with the X5, X10, and JF12 mutations from Ver. 2; Ver. 10, which has 31 substitutions, was created by removing overlaps with the X5 and JF12 mutations from Ver. 4; Ver. 11, which has 28 substitutions, was created by reducing the number of substitutions by three from Ver. 10; and Ver. 12, which has 13 substitutions, was created by reducing the number of substitutions by 15 from Ver. 11 (the amino acid substitutions of Ver. 1 to Ver. 12 are summarized in Figure 1). For comparison, a variant containing the previously reported X5 mutation (but codon-optimized for humans) was also prepared.

After cloning, each hFVIIISQ was introduced into a plasmid (p1.1HCRhAAT) carrying the HCRhAAT promoter, which combines HCR, a liver-specific enhancer of the ApoE gene, with the human α1-antitrypsin promoter.

To evaluate FVIII expression following transfection of each FVIII gene (Ver. 1 to Ver. 10), Huh7 cells (a human hepatoma cell line) were transfected with the plasmid, and FVIII transiently secreted into the culture supernatant was measured. Huh7 cell culture medium was RPMI-1640 (Sigma) supplemented with 10% FBS, GlutaMAX (Thermo Scientific), and penicillin-streptomycin. Huh7 cells were seeded at 1.5 x ¹⁰ cells/well in a 24-well plate, and transfection was performed the following day using Lipoefectamin 3000 (Thermo Scientific). The medium was replaced the day after transfection, and the medium was collected the following day after 24 hours of culture. The collected medium was centrifuged, and the supernatant was used as a sample for measuring the activity and antigen of factor VIII.

Factor VIII activity was measured using the chromogenic assay with the fully automated blood coagulation analyzer CS-1600 (Sysmex).

AAV8 vectors carrying each FVIII gene (Ver. 1, 2, 4, 10) were administered to factor VIII-deficient mice at a vector dose of 1 x ¹⁰⁸ vg per gram of mouse body weight. Factor VIII activity in mouse plasma was measured by the synthetic substrate method.

The results are shown in Figure 2. All variants had increased coagulation factor activity compared to wild-type hFVIIISQ. In particular, Ver. 4 showed high activity in both in vitro and in vivo experiments.

Example 2: Identification of Mutation Sites Important for Increased Activity of Ver. 4 The number of substitutions was reduced from Ver. 4, and a variant S12 with 12 substitutions was identified using the one-stage clotting assay. Further reductions in the number of substitutions led to the identification of PNPL (L152P, K213N, S367P, F2196L) using the synthetic substrate assay. S7, with 7 substitutions, was created by removing five substitutions from S12 that overlap with the X5 and JF12 mutations. S5 and S3, with 5 and 3 substitutions, respectively, were further reduced, and S8, S6, and S4, with the addition of the L152P substitution to S7, S5, and S3, respectively, were created. These were transfected with plasmids as in Example 1, and the coagulation factor activity in the culture supernatant was measured using the one-stage clotting assay and the synthetic substrate assay. These results are shown in Figure 3 . S12, S8, and S6, which contain the PNPL mutation, exhibited high activity. Furthermore, Ver. 10, with the number of substitutions reduced to 28, was also developed. Ver. 11 also showed coagulation factor activity equivalent to that of Ver. 4.

Next, the effect of each amino acid substitution in S8 was evaluated individually using the one-stage clotting assay and the synthetic substrate assay. The results are shown in Figure 4. L152P, K213N, S367P, S727P, and F2196L increased clotting factor activity compared to the wild type.

Example 3: Substitution of PNPL variants with other amino acids A series of variants (XNPL, PXPL, PNXL, PNPX) were prepared by substituting any of the substitution sites in PNPL with 19 other amino acids. These variants were transfected with plasmids in the same manner as in Example 1, and the coagulation factor activity in the culture supernatant was measured by the one-stage coagulation assay and the synthetic substrate assay. The results are shown in Figures 5-1 to 5-4. It was revealed that substitutions of L152 with Q, V, M, etc., K213 with N, H, etc., S367 with N, Q, etc., and F2196 with M, etc., also showed high activity similar to that of PNPL.

Example 4: Administration Experiment to FVIII KO Mice The Ver. 4, 10, and 11, S12, S8, and PNPL genes, which showed particularly high activity in Examples 1 and 2, were loaded onto AAV8 vectors and administered to FVIII knockout hemophilia A model mice. Four weeks later, blood samples were collected and coagulation factor activity was measured using a one-stage clotting assay and a synthetic chromogenic assay. Antigen levels were also measured using ELISA. The results are shown in Figure 6. S12 and PNPL, which contain the PNPL mutation, almost matched the activity of Ver. 4, but the antigen levels were significantly lower. This suggests that these variants have increased specific activity.

Example 5: Localization of FVIII in and outside FVIII-expressing cells. Huh7 cells were transfected with Ver. 4, S12, and PNPL, and the amount of FVIII present in the culture supernatant and cell extract was analyzed by Western blot. The results are shown in Figure 7-1. Intracellular retention of the light chain was observed with wild-type hFVIIISQ, whereas retention was abolished with Ver. 4. S12 and PNPL did not improve intracellular retention of the light chain. The inventors predicted that the improvement in intracellular retention of the light chain in Ver. 4 was due to the two mutations, I1668F and D1681G, and prepared FVIIISQ variants containing only these two substitutions, and similarly examined the amount of FVIII present in cell extracts. As shown in Figure 7-2, the FG mutation demonstrated improved intracellular retention of the light chain.

Example 6 Identification of mutation sites important for high secretion and high activity of Ver. 4 (2)
Based on Ver. 11, Ver. 12 (FIG. 1) was identified using the synthetic substrate method, in which the number of substitutions was reduced to 13. The results are shown in FIG.
The number of substitutions was further reduced based on Ver. 12 to create highly secreted variants with 8 to 15 substitutions (F8-Tochigi-3 (NPL), 5, 8, 10, 11, 13, and 15) (the amino acid substitutions of these variants are shown in Figure 9). Plasmid transfection was performed in the same manner as in Example 1, and the coagulation factor activity in the culture supernatant was measured by the one-stage clotting assay and the synthetic substrate assay. The antigen amount in the culture supernatant was also measured using ELISA. The results are shown in Figure 10. In the figure, F8-Tochigi-28 represents the same variant as Ver. 11.

Example 7 Construction of a High-Expression hFVIII Gene The nucleotide sequences of the codon-optimized wild-type hFVIII DNA (SEQ ID NO: 4) and the Ver. 4 variant DNA (SEQ ID NO: 2) contain numerous CpG sequences. The inventors predicted that these CpG sequences may induce an immune response in the host, impairing FVIII expression. Therefore, they removed all of the CpG sequences without changing the amino acid sequence of the FVIII they encode. Then, they re-optimized the codons, taking into account other parameters such as GC content, and designed the nucleotide sequences using GeneArt Codon Optimizer (Thermo Scientific). The nucleotide sequences of the CpG-removed and re-optimized wild-type hFVIII DNA and the Ver. 4 variant DNA are shown in SEQ ID NOs: 5 and 3, respectively. These DNAs were ligated downstream of a mouse transthyretin (mTTR) promoter sequence (SEQ ID NO: 7) from which the CpG sequence had also been removed, and then loaded into an AAV6 vector. The vector was then plasmid transfected and introduced into Huh7 cells in the same manner as in Example 1. The expression cassette was also loaded into an AAV8 vector and administered to mice. The coagulation factor activity in the culture supernatant and blood was measured by the one-stage clotting assay and the synthetic substrate assay. The results are shown in Figures 11-1 and 11-2. For both Ver. 4 and the wild-type, removal of the CpG sequence significantly increased the expression efficiency.

Example 8: Administration to a Cynomolgus Monkey Model The DNA prepared in Example 7, in which the CpG-deleted and re-optimized Ver. 4 variant coding sequence was ligated downstream of the CpG-deleted mTTR promoter sequence, was inserted into an AAV8 vector and administered to a cynomolgus monkey hemophilia model treated with immunosuppressants (prednisolone 1 mg/kg/day and tacrolimus 0.05 mg/kg/day) at the vector dose (2 x 10 ¹² vg/kg) used in a clinical trial conducted by the University of London (NCT0300183). Factor VIII activity and antigen levels were measured every one or two weeks after administration. The results are shown in Figure 12-1. Although the time course differed between individuals, factor VIII activity significantly increased in all individuals after vector administration.
Furthermore, the above DNA was inserted into an AAV5 vector and administered to a similarly treated cynomolgus monkey model at vector doses 1/10 (6 x 10 ¹² vg/kg) and 1/30 (2 x 10 ¹² vg/kg) of those used with Rectavian (registered trademark) (a hemophilia A treatment drug marketed by BioMarin). Factor VIII activity and antigen levels were measured every one or two weeks after administration. The results are shown in Figure 12-2. Factor VIII activity significantly increased after vector administration at both doses.

Example 9 Optimization of Expression Cassette Sequence Huh7 cells were transfected with the AAV6 vector (original) prepared in Example 7, in which DNA was inserted into the CpG-deleted and re-optimized Ver. 4 variant coding sequence ligated downstream of the CpG-deleted mTTR promoter sequence (which further contained an SV40 polyA addition signal downstream of the coding sequence and contained the nucleotide sequence shown in SEQ ID NO: 8 as an expression cassette), the AAV6 vector (v1mTTRp) containing, as an expression cassette, a nucleotide sequence (SEQ ID NO: 9) in which the sequence between the promoter and the coding sequence was shortened, and the AAV6 vector (v1-v1, v1-v2, and v1-v3, respectively) containing, as expression cassettes, three nucleotide sequences (SEQ ID NOs: 10, 11, and 12) in which the 3'-end of the polyA addition signal was shortened by different lengths. The medium was replaced after 24 hours, and the culture supernatant and cells were collected after another 24 hours. The Factor VIII activity and antigen content in the supernatant, as well as the AAV genome content per cell, were measured. The results are shown in Figure 13. Activity increased when the sequence between the promoter and coding sequence was shortened, and activity was further improved by shortening the polyA addition signal in combination. v1-v2 showed the highest activity. However, when the 3' end of the polyA addition signal was deleted up to v1-v3, Factor VIII activity was significantly reduced.

The variants of the present invention, the nucleic acids encoding them, expression vectors containing the nucleic acids, and liver cells transfected with the expression vectors are useful for in vivo or ex vivo gene therapy, mRNA medicines, etc. for hemophilia A. Furthermore, recombinant human factor VIII variants obtained by culturing host cells transfected with the expression vectors are useful as biological preparations for the treatment of hemophilia A. Furthermore, highly expressed human factor VIII-encoding nucleic acids are useful for gene therapy and mRNA medicines for hemophilia A, and are also useful for producing hosts that highly produce recombinant human factor VIII.

Claims (26)

A modified human factor VIII in which at least one amino acid residue selected from the group consisting of K213, S367, and F2196 of wild-type human factor VIII is substituted with another amino acid, and which is characterized by having higher specific activity and/or secretory expression efficiency compared to the wild-type. The variant according to claim 1, wherein the amino acid substitution at K213 is K213N or K213H, and/or the amino acid substitution at S367 is S367P, S367N or S367Q, and/or the amino acid substitution at F2196 is F2196L or F2196M. The variant described in claim 2, comprising the amino acid substitutions K213N, S367P, and F2196L. The variant of any one of claims 1 to 3, wherein at least one amino acid residue selected from the group consisting of R-5, P25, A28, L152, M217, W228, Q410, Y487, R489, F501, M539, I566, L603, I642, S727, A736, S1657, Q1659, E1661, I1668, D1681, R1776, H1859, A1993, H2007, N2019, K2085, K2207, F2275, S2296, V2314, Q2316, and M2321 is further substituted with another amino acid. R-5P, P25H, A28T, L152A, L152D, L152E, L152H, L152I, L152M, L152N, L152P, L152Q or L152V, M217T, W228Q, Q410L, Y487H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P 5. The variant of claim 4, comprising at least one amino acid substitution selected from the group consisting of Q1659E, E1661K, I1668F, D1681G, R1776K, H1859R, A1993V, H2007Q, N2019K, K2085M, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L. L152A, L152D, L152E, L152H, L152I, L152M, L152N, L152P, L152Q or L152V, and/or I1668F and D1681G
The variant of claim 3 further comprising an amino acid substitution of: the amino acid substitution at K213 is K213N;
M539L, and optionally at least one amino acid substitution in the A2 domain selected from the group consisting of I566M, L603P, and I642V;
2. The variant of claim 1, further comprising at least one amino acid substitution in the C2 domain selected from the group consisting of K2207Q, Q2316H and M2321L, and optionally F2275L, S2296A and V2314A. The variant of claim 7 further comprising the amino acid substitutions I1668F and D1681G. The variant of claim 3 further comprises amino acid substitutions L152P, Y487H, and L603P, and optionally at least one amino acid substitution selected from the group consisting of S727P, Q1659E, H1859R, A1993V, H2007Q, N2019K, and K2085M. The variant of claim 3 further comprising the amino acid substitutions M539L, I566M, L603P, I642V, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L. The variant of claim 10 further comprising the amino acid substitutions I1668F and D1681G. A human factor VIII variant having the following amino acid substitutions (a) to (n) in wild-type human factor VIII:
(a) K213N, S367P, and F2196L
(b) K213N, S367P, F2196L, I1668F, and D1681G
(c) K213N, M539L, I566M, I642V, K2207Q, F2275L, Q2316H, and M2321L
(d) K213N, M539L, I566M, L603P, I642V, F2196L, K2207Q, F2275L, Q2316H, and M2321L
(e) K213N, M539L, I566M, I642V, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L
(f) K213N, M539L, I566M, I642V, I1668F, D1681G, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H, and M2321L
(g) L152P, K213N, S367P and F2196L
(h) L152P, K213N, S367P, Y487H, L603P and F2196L
(i) L152P, K213N, S367P, Y487H, L603P, S727P, Q1659E, and F2196L
(j) L152P, K213N, S367P, Y487H, L603P, S727P, Q1659E, H1859R, A1993V, H2007Q, K2085M and F2196L
(k) K213N, S367P, M539L, I566M, L603P, I642V, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
(l) P25H, A28T, K213N, M217T, S367P, Q410L, Y487H, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P , Q1659E, E1661K, I1668F, D1681G, R1776K, N2019K, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
(m) R-5P, P25H, A28T, K213N, M217T, W228Q, S367P, Q410L, Y487H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736 V, S1657P, Q1659E, E1661K, I1668F, D1681G, R1776K, N2019K, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L
(n) R-5P, P25H, A28T, L152P, K213N, M217T, W228Q, S367P, Q410L, Y487 H, R489G, F501M, M539L, I566M, L603P, I642V, S727P, A736V, S1657P, Q1 659E, E1661K, I1668F, D1681G, R1776K, H1859R, A1993V, H2007Q, N201 9K, K2085M, F2196L, K2207Q, F2275L, S2296A, V2314A, Q2316H and M2321L The variant described in any one of claims 1 to 12, which lacks the B domain. The variant described in claim 13, wherein the deletion of the B domain is a deletion of the amino acid sequence from positions 763 to 1656 in the amino acid sequence represented by SEQ ID NO: 1. A nucleic acid encoding the variant described in any one of claims 1 to 14. The nucleic acid of claim 15, which is codon-optimized for human use. The nucleic acid of claim 16, or a nucleic acid encoding wild-type human factor VIII that has been codon-optimized for human use, which has been further modified to eliminate CpG sequences without changing the amino acid sequence it encodes. An expression vector comprising the nucleic acid described in any one of claims 15 to 17. The expression vector of claim 18, wherein the nucleic acid is under the control of a liver-specific promoter. The expression vector described in claim 18 or 19, wherein the promoter has been modified so as not to contain CpG sequences. The expression vector described in any one of claims 18 to 20, wherein the vector is a viral vector. The expression vector of claim 21, wherein the virus is an adeno-associated virus with liver tropism. A host cell into which the expression vector described in any one of claims 18 to 22 has been introduced. A method for producing human factor VIII or a variant thereof, comprising culturing the host cell of claim 23 and recovering human factor VIII or a variant thereof from the resulting culture. A pharmaceutical comprising the variant described in any one of claims 1 to 14, the nucleic acid described in any one of claims 15 to 17, the expression vector described in any one of claims 18 to 22, or the host cell described in claim 23. The pharmaceutical described in claim 25, which is for the treatment of hemophilia A.