Method for producing antibody-lysosomal enzyme fusion proteins

Abstract

To provide methods for producing fusion proteins comprising an antibody fused to a lysosomal enzyme.SOLUTION: A method for producing an antibody-human lysosomal enzyme fusion protein comprises: (a) culturing mammalian cells producing the fusion protein in a serum free culture medium to allow secretion of the fusion protein into the culture medium; (b) harvesting the culture supernatant from the culture medium by removing the mammalian cells from the culture medium; (c) purifying the fusion protein from the culture supernatant by a column chromatography using a material having an affinity to the antibody as a stationary phase, an anionic ion exchange column chromatography, a cationic ion exchange column chromatography and a size exclusion column chromatography.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for producing a fusion protein obtained by fusing an antibody with a lysosomal enzyme, and more particularly to a method for purifying a recombinant fusion protein obtained by culturing host cells into which an expression vector incorporating the gene encoding the fusion protein has been introduced, to a purity suitable for use as a pharmaceutical. [Background technology]

[0002] Currently, many pharmaceuticals containing recombinant proteins as active ingredients are commercially available. These recombinant proteins are obtained in the culture supernatant by culturing host cells into which an expression vector containing the gene encoding the target protein has been introduced. Recombinant proteins obtained in the culture supernatant cannot be used as pharmaceuticals in their raw form because they contain impurities. To use them as pharmaceuticals, the recombinant proteins contained in the culture supernatant must be purified.

[0003] Methods have been reported for purifying recombinant proteins obtained from culture supernatants of mammalian cells, cultured from these host cells, to a level suitable for pharmaceutical use. For example, a method has been reported for expressing human erythropoietin (hEPO), a glycoprotein that acts on erythroblast progenitor cells to differentiate them into erythrocytes and promote erythrocyte production, as a recombinant protein using CHO cells as host cells, and purifying it from the culture supernatant using various chromatography methods, including dye affinity column chromatography, to a level suitable for pharmaceutical use (Patent Document 1). Another example is the expression of human follicle-stimulating hormone (hFSH), a type of gonadotropin that has the activity to promote estrogen production and secretion in the ovaries, as a recombinant protein using CHO cells as host cells, and purifying it from the culture supernatant using various chromatography methods, including cation exchange column chromatography, to a level suitable for pharmaceutical use (Patent Document 2). For example, a method has been reported in which human iduronic acid-2-sulfatase (hI2S), a type of lysosomal enzyme that has the activity to hydrolyze sulfate ester bonds present in glycosaminoglycan (GAG) molecules such as heparan sulfate and dermatan sulfate, is expressed as a recombinant protein using CHO cells as host cells, and then purified from the culture supernatant using various chromatography methods, including cation exchange column chromatography, to a level suitable for pharmaceutical use (Patent Document 3). For example, a method has also been reported in which human α-galactosidase A (hα-Gal A), a type of lysosomal enzyme that has the activity to hydrolyze terminal α-galactosyl bonds of glycolipids and glycoproteins, is expressed as a recombinant protein using CHO cells as host cells, and then purified from the culture supernatant using various chromatography methods, including anion exchange column chromatography, to a level suitable for pharmaceutical use (Patent Documents 4 and 5). Furthermore, for example, a method has been reported in which human DNase I, which has the activity of degrading DNA in a non-specific manner based on its base sequence, is expressed as a recombinant protein using CHO cells as host cells, and then purified from the culture supernatant using various chromatography methods, including anion exchange column chromatography and dye ligand affinity column chromatography, to a level suitable for pharmaceutical use (Patent Document 6).Furthermore, for example, an anti-hTfR antibody conjugated with human iduronate-2-sulfatase is expressed as a recombinant protein using CHO cells as host cells, and this is purified from the culture supernatant using protein A affinity column chromatography, hydroxyapatite column chromatography, and size exclusion column chromatography until it can be used in pharmaceuticals (Patent Document 7).

[0004] Thus, in order to obtain a recombinant protein that can be used as a pharmaceutical, an independent purification method has been developed for each recombinant protein.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Patent Document 5

Patent Document 6

Patent Document 7

Summary of the Invention

Problems to be Solved by the Invention

[0006] An object of the present invention is to provide a method for expressing a fusion protein obtained by fusing an antibody and a lysosomal enzyme as a recombinant protein and purifying it to a purity that enables it to be distributed on the market as a pharmaceutical.

Means for Solving the Problems

[0007] In research toward the above objective, the inventors, after diligent investigation, discovered that by culturing mammalian cells into which an expression vector containing a gene encoding a fusion protein of an anti-transferrin receptor antibody and human α-L-iduronidase (hIDUA) has been introduced in serum-free medium, the fusion protein obtained in the culture medium can be purified efficiently with high purity by a combination of column chromatography, anion exchange column chromatography, cation exchange column chromatography, and size exclusion column chromatography using a material conjugated with a substance having affinity for the antibody as the stationary phase. The present invention was completed by further investigation based on these findings. That is, the present invention provides the following. 1. A method for producing a fusion protein obtained by fusing an antibody with a human lysosomal enzyme, (a) A step of culturing mammalian cells that produce the fusion protein in serum-free medium to cause the fusion protein to be secreted into the culture medium, (b) A step of collecting the culture supernatant by removing the mammalian cells from the culture medium obtained in step (a) above, (c) From the culture supernatant obtained in step (b) above, the fusion protein is purified using column chromatography with a material conjugated with a substance having affinity for the antibody as the stationary phase, anion exchange column chromatography, cation exchange column chromatography, and size exclusion column chromatography. A manufacturing method comprising [the specified element]. 2. The method for producing the antibody described in 1, wherein in step (c), column chromatography using a material to which a substance having affinity for the antibody is bound is used as the stationary phase, anion exchange column chromatography, cation exchange column chromatography, and size exclusion column chromatography are used in this order. 3. The method for producing the antibody according to 1 or 2 above, wherein the substance having affinity for the antibody has affinity for the CH1 region of the heavy chain of the antibody. 4. A method for producing an anion exchanger used in the anion exchange column chromatography, wherein the anion exchanger used is a strong anion exchanger, any of the methods described in 1 to 3 above. 5. A method for producing any of the above 1 to 4, wherein the cation exchanger used in the cation exchange column chromatography is a weak cation exchanger. 6. A method for producing an antibody fused with the human lysosomal enzyme, wherein the antibody is a humanized antibody or a human antibody, as described in any of items 1 to 5 above. 7. A method for producing any of the above 1 to 5, wherein the antibody fused with the human lysosomal enzyme is a humanized antibody. 8. A method for producing any of the above 1 to 7, wherein the antibody fused with the human lysosomal enzyme uses a molecule present on the surface of vascular endothelial cells as an antigen. 9. The method for producing the above-mentioned method, wherein the molecule present on the surface of vascular endothelial cells is selected from the group consisting of transferrin receptor (TfR), insulin receptor, leptin receptor, lipoprotein receptor, IGF receptor, OATP-F, organic anion transporter, MCT-8, monocarboxylic acid transporter, and Fc receptor. 10. The method for producing the method described in item 8 above, wherein the vascular endothelial cells are cerebral vascular endothelial cells. 11. The method for producing the above 10, wherein the molecule present on the surface of the cerebral vascular endothelial cell is selected from the group consisting of transferrin receptor (TfR), insulin receptor, leptin receptor, lipoprotein receptor, IGF receptor, OATP-F, organic anion transporter, MCT-8, and monocarboxylic acid transporter. 12. A method for producing any of the above 8 to 11, wherein the vascular endothelial cells are human vascular endothelial cells. 13. The antibody is an anti-human transferrin receptor antibody, The antibody has a variable heavy chain in which CDR1 comprises the amino acid sequence of SEQ ID NO: 13 or SEQ ID NO: 14, CDR2 comprises the amino acid sequence of SEQ ID NO: 15 or SEQ ID NO: 16, and CDR3 comprises the amino acid sequence of SEQ ID NO: 17 or SEQ ID NO: 18, and A method for producing any of the above 1 to 12, wherein the antibody has a variable region of the light chain in which CDR1 comprises the amino acid sequence of SEQ ID NO: 8 or SEQ ID NO: 9, CDR2 comprises the amino acid sequence of SEQ ID NO: 10 or SEQ ID NO: 11, or the amino acid sequence Lys-Val-Ser, and CDR3 comprises the amino acid sequence of SEQ ID NO: 12. 14. The method for producing the antibody described in 13, wherein the framework region 3 of the heavy chain contains the amino acid sequence of SEQ ID NO: 19. 15. The method for producing the antibody described in 14, wherein the variable region of the heavy chain in the antibody contains the amino acid sequence of SEQ ID NO: 21. 16. The antibody comprises an amino acid sequence in the variable region of the heavy chain that has 80% or more identity with the amino acid sequence of SEQ ID NO: 21, and CDR1 comprises the amino acid sequence of SEQ ID NO: 13 or SEQ ID NO: 14, CDR2 comprises the amino acid sequence of SEQ ID NO: 15 or SEQ ID NO: 16, CDR3 comprises the amino acid sequence of SEQ ID NO: 17 or SEQ ID NO: 18, and The method for producing the method described in 13 or 14, wherein the framework region 3 comprises the amino acid sequence of SEQ ID NO: 19. 17. The antibody comprises an amino acid sequence in the variable region of the heavy chain that has 90% or more identity with the amino acid sequence of SEQ ID NO: 21, and CDR1 comprises the amino acid sequence of SEQ ID NO: 13 or SEQ ID NO: 14, CDR2 comprises the amino acid sequence of SEQ ID NO: 15 or SEQ ID NO: 16, CDR3 comprises the amino acid sequence of SEQ ID NO: 17 or SEQ ID NO: 18, and The method for producing the method described in 13 or 14, wherein the framework region 3 comprises the amino acid sequence of SEQ ID NO: 19. 18. The antibody comprises an amino acid sequence in which the variable region of the heavy chain has been modified by substitution, deletion, or addition of 1 to 5 amino acids relative to the amino acid sequence of SEQ ID NO: 21, and CDR1 comprises the amino acid sequence of SEQ ID NO: 13 or SEQ ID NO: 14, CDR2 comprises the amino acid sequence of SEQ ID NO: 15 or SEQ ID NO: 16, CDR3 comprises the amino acid sequence of SEQ ID NO: 17 or SEQ ID NO: 18, and The method for producing the method described in 13 or 14, wherein the framework region 3 comprises the amino acid sequence of SEQ ID NO: 19. 19. The antibody comprises an amino acid sequence in which the variable region of the heavy chain has been modified by substitution, deletion, or addition of 1 to 3 amino acids relative to the amino acid sequence of SEQ ID NO: 21, and CDR1 comprises the amino acid sequence of SEQ ID NO: 13 or SEQ ID NO: 14, CDR2 comprises the amino acid sequence of SEQ ID NO: 15 or SEQ ID NO: 16, CDR3 comprises the amino acid sequence of SEQ ID NO: 17 or SEQ ID NO: 18, and The method for producing the method described in 13 or 14, wherein the framework region 3 comprises the amino acid sequence of SEQ ID NO: 19. 20. A method for producing the antibody described in any of the above 13 to 19, wherein the variable region of the light chain in the antibody contains the amino acid sequence of SEQ ID NO: 20. 21. The antibody comprises an amino acid sequence in the variable region of its light chain that has 80% or more identity with the amino acid sequence of SEQ ID NO: 20, and CDR1 comprises the amino acid sequence of SEQ ID NO: 8 or SEQ ID NO: 9, CDR2 comprises the amino acid sequence of SEQ ID NO: 10 or SEQ ID NO: 11, or the amino acid sequence Lys-Val-Ser. A method for producing any of the above 13 to 19, wherein CDR3 comprises the amino acid sequence of SEQ ID NO: 12. 22. The antibody comprises an amino acid sequence in the variable region of its light chain that has 90% or more identity with the amino acid sequence of SEQ ID NO: 20, and CDR1 comprises the amino acid sequence of SEQ ID NO: 8 or SEQ ID NO: 9, CDR2 comprises the amino acid sequence of SEQ ID NO: 10 or SEQ ID NO: 11, or the amino acid sequence Lys-Val-Ser. A method for producing any of the above 13 to 19, wherein CDR3 comprises the amino acid sequence of SEQ ID NO: 12. 23. The antibody comprises an amino acid sequence in which the variable region of the light chain has been modified by substitution, deletion, or addition of 1 to 5 amino acids relative to the amino acid sequence of SEQ ID NO: 20, and CDR1 comprises the amino acid sequence of SEQ ID NO: 8 or SEQ ID NO: 9, CDR2 comprises the amino acid sequence of SEQ ID NO: 10 or SEQ ID NO: 11, or the amino acid sequence Lys-Val-Ser. A method for producing any of the above 13 to 19, wherein CDR3 comprises the amino acid sequence of SEQ ID NO: 12. 24. The antibody comprises an amino acid sequence in which the variable region of the light chain has been modified by substitution, deletion, or addition of 1 to 3 amino acids relative to the amino acid sequence of SEQ ID NO: 20, and CDR1 comprises the amino acid sequence of SEQ ID NO: 8 or SEQ ID NO: 9, CDR2 comprises the amino acid sequence of SEQ ID NO: 10 or SEQ ID NO: 11, or the amino acid sequence Lys-Val-Ser. A method for producing any of the above 13 to 19, wherein CDR3 comprises the amino acid sequence of SEQ ID NO: 12. 25. The method for producing the antibody according to the above 15, wherein the heavy chain of the antibody contains the amino acid sequence of SEQ ID NO: 23. 26. The method for producing the antibody according to the above 20 or 25, wherein the light chain of the antibody contains the amino acid sequence of SEQ ID NO: 22. 27. A method for producing any of the above 1 to 26, wherein the antibody is Fab,F(ab')2, or F(ab'). 28. A method for producing any of the above 1 to 27, wherein the human lysosomal enzyme in the fusion protein is bound to the C-terminal or N-terminal side of the light chain of the antibody. 29. The method for producing the above 28, wherein the human lysosomal enzyme in the fusion protein is directly or via a linker bound to the C-terminal or N-terminal side of the light chain. 30. The method for producing the above 28, wherein the human lysosomal enzyme in the fusion protein is linked to the heavy chain via a linker at the C-terminal or N-terminal end. 31. A method for producing the above 29 or 30, wherein the linker sequence is a peptide consisting of 1 to 50 amino acid residues. 32. The method for producing the above 31, wherein the linker is a peptide comprising one glycine molecule, one serine molecule, the amino acid sequence Gly-Ser, the amino acid sequence Gly-Gly-Ser, the amino acid sequence of SEQ ID NO: 1, the amino acid sequence of SEQ ID NO: 2, the amino acid sequence of SEQ ID NO: 3, the amino acid sequence of SEQ ID NO: 4, and an amino acid sequence consisting of 1 to 10 consecutive such amino acid sequences. 33. A method for producing any of the above 1 to 32, wherein the human lysosomal enzyme in the fusion protein is human α-L-iduronidase. 34. The antibody in the fusion protein is Fab, The light chain of the antibody consists of the amino acid sequence of SEQ ID NO: 22, and The method for producing the antibody described in 33, wherein the heavy chain of the antibody binds to the human α-L-idulonidase at its C-terminus via a linker consisting of the amino acid sequence of SEQ ID NO: 4, thereby causing the fusion protein to form the amino acid sequence of SEQ ID NO: 27. 35. The antibody in the fusion protein is Fab, The light chain of the antibody consists of the amino acid sequence of SEQ ID NO: 22, and The method for producing the antibody described in 33, wherein the heavy chain of the antibody consists of the amino acid sequence of SEQ ID NO: 23, and at its C-terminus, it binds to human α-L-iduronidase consisting of the amino acid sequence of SEQ ID NO: 6 via a linker consisting of the amino acid sequence of SEQ ID NO: 4. 36. The method for producing the human α-L-iduronidase described in 33, wherein the human α-L-iduronidase comprises the amino acid sequence of SEQ ID NO: 6 or 7. [Effects of the Invention]

[0008] According to the present invention, it is possible to provide a fusion protein of an anti-transferrin receptor antibody and a lysosomal enzyme, purified to a purity suitable for clinical use as a therapeutic agent for lysosomal storage diseases accompanied by central nervous system disorders. In particular, it is possible to provide a fusion protein of an anti-transferrin receptor antibody and human IDUA, purified to a purity suitable for clinical use as a therapeutic agent for Hurler syndrome accompanied by central nervous system disorders. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows the SE-HPLC chart of the purified humanized anti-hTfR antibody-hIDUA product obtained in Example 4. The vertical axis represents absorbance at 215 nm, and the horizontal axis represents retention time (minutes). (A) shows the peak derived from the monomer of humanized anti-hTfR antibody-hIDUA, and (B) and (C) show the peaks derived from the polymer of humanized anti-hTfR antibody-hIDUA, respectively. [Modes for carrying out the invention]

[0010] This invention relates to a method for producing a protein obtained by conjugating an anti-transferrin receptor antibody (anti-TfR antibody) with a human lysosomal enzyme. Here, the antibody conjugated with the lysosomal enzyme is not particularly limited in terms of animal species, as long as it has the property of specifically binding to the antigen; however, it is particularly preferred to be a human antibody or a humanized antibody. For example, the antibody may be an antibody from a mammal other than a human, or it may be a chimeric antibody of a human antibody and an antibody from another mammal other than a human.

[0011] Human antibodies are antibodies whose entirety is encoded by genes derived from humans. However, antibodies encoded by genes that have been mutated from the original human gene for purposes such as increasing gene expression efficiency are also considered human antibodies. Furthermore, antibodies created by combining two or more genes that encode human antibodies, in which a part of one human antibody is replaced with a part of another human antibody, are also considered human antibodies. Human antibodies have three complementarity-determining regions (CDRs) in the immunoglobulin light chain and three complementarity-determining regions (CDRs) in the immunoglobulin heavy chain. The three CDRs in the immunoglobulin light chain are called CDR1, CDR2, and CDR3, in order from the N-terminus. The three CDRs in the immunoglobulin heavy chain are called CDR1, CDR2, and CDR3, in order from the N-terminus. Antibodies whose antigen specificity, affinity, etc., has been modified by replacing the CDR of one human antibody with the CDR of another human antibody are also considered human antibodies.

[0012] In the present invention, antibodies that have been modified by altering the gene of the original human antibody, thereby adding mutations such as substitutions, deletions, or additions to the amino acid sequence of the original antibody, are also called human antibodies. When amino acids in the amino acid sequence of the original antibody are substituted with other amino acids, the number of amino acids to be substituted is preferably 1 to 20, more preferably 1 to 10, even more preferably 1 to 5, and even more preferably 1 to 3. When amino acids are deleted from the amino acid sequence of the original antibody, the number of amino acids to be deleted is preferably 1 to 20, more preferably 1 to 10, even more preferably 1 to 5, and even more preferably 1 to 3. Furthermore, antibodies that have been modified by combining these amino acid substitutions and deletions are also human antibodies. When amino acids are added, preferably 1 to 20, more preferably 1 to 10, even more preferably 1 to 5, and even more preferably 1 to 3 amino acids are added to the amino acid sequence of the original antibody or to the N-terminus or C-terminus. Antibodies that have been modified by combining these amino acid additions, substitutions, and deletions are also human antibodies. The amino acid sequence of the mutated antibody preferably exhibits 80% or more identity with the amino acid sequence of the original antibody, more preferably 85% or more identity, even more preferably 90% or more identity, even more preferably 95% or more identity, and still more preferably 98% or more identity. In other words, in this invention, when we refer to "human-derived genes," we include not only the original human-derived genes but also genes obtained by modifying the original human-derived genes.

[0013] In the present invention, the term "humanized antibody" refers to an antibody in which the amino acid sequence of a part of the variable region (for example, all or part of the CDR) is derived from a mammal other than a human, and the remaining region is derived from a human. For example, a humanized antibody may be an antibody produced by replacing three complementarity-determining regions (CDRs) of the immunoglobulin light chain and three complementarity-determining regions (CDRs) of the immunoglobulin heavy chain that constitute a human antibody with CDRs of another mammal. The species of the other mammal from which the CDRs transplanted to the appropriate positions in the human antibody are derived is not particularly limited as long as it is a mammal other than a human, but is preferably a mouse, rat, rabbit, horse, or a primate other than a human, more preferably a mouse and a rat, for example a mouse.

[0014] In the present invention, the case where the antibody is a human antibody or a humanized antibody will be described in detail below. The light chain of human antibodies and humanized antibodies has a λ chain and a κ chain. The light chain constituting the antibody may be either a λ chain or a κ chain. Furthermore, the heavy chain of human antibodies and humanized antibodies has a γ chain, a μ chain, an α chain, a σ chain and an ε chain, which correspond to IgG, IgM, IgA, IgD and IgE, respectively. The heavy chain constituting the antibody may be any of the γ chain, μ chain, α chain, σ chain and ε chain, but a γ chain is preferred. Moreover, the γ chain of the antibody's heavy chain has γ1 chain, γ2 chain, γ3 chain and γ4 chain, which correspond to IgG1, IgG2, IgG3 and IgG4, respectively. When the heavy chain constituting the antibody is a γ chain, the γ chain may be any of the γ1 chain, γ2 chain, γ3 chain and γ4 chain, but a γ1 chain or a γ4 chain is preferred. If the antibody is a humanized antibody or a human antibody and is IgG, the light chain of the antibody may be either a lambda chain or a kappa chain, and the heavy chain of the antibody may be any of the gamma1, gamma2, gamma3, and gamma4 chains, but preferably gamma1 or gamma4. For example, one preferred embodiment of the antibody is one in which the light chain is a kappa chain and the heavy chain is a gamma1 chain, or one in which the light chain is a lambda chain and the heavy chain is a gamma1 chain.

[0015] In the present invention, the term "chimeric antibody" refers to an antibody formed by linking together fragments of two or more different antibodies originating from two or more different species.

[0016] A chimeric antibody is an antibody in which a portion of a human antibody is replaced by a portion of an antibody from a non-human mammal. The antibody consists of an Fc region, a Fab region, and a hinge region, as described below. A specific example of such a chimeric antibody is one in which the Fc region originates from a human antibody while the Fab region originates from an antibody from another mammal. The hinge region originates from either the human antibody or the antibody from the other mammal. Conversely, an example of a chimeric antibody is one in which the Fc region originates from another mammal while the Fab region originates from a human antibody. The hinge region may originate from either the human antibody or the antibody from the other mammal.

[0017] Furthermore, antibodies can be said to consist of a variable region and a constant region. Another specific example of a chimeric antibody is the constant region of the heavy chain (C H ) and the steady region of the light chain (C L ) is derived from human antibodies, while the variable region of the heavy chain (V H ) and the variable region of the light chain (V L ) are derived from antibodies of other mammals, or conversely, the constant region of the heavy chain (C H ) and the steady region of the light chain (C L ) is derived from antibodies of other mammals, while the variable region of the heavy chain (V H ) and the variable region of the light chain (V L ) also includes those derived from human antibodies. Here, the other mammalian species is not particularly limited as long as it is a mammal other than a human, but preferably it is a mouse, rat, rabbit, horse, or a primate other than a human, and more preferably it is a mouse.

[0018] A chimeric antibody of a human antibody and a mouse antibody is particularly referred to as a "human / mouse chimeric antibody". Examples of human / mouse chimeric antibodies include a chimeric antibody in which the Fc region is derived from a human antibody while the Fab region is derived from a mouse antibody, and conversely, a chimeric antibody in which the Fc region is derived from a mouse antibody while the Fab region is derived from a human antibody. The hinge region is derived from either a human antibody or a mouse antibody. As other specific examples of human / mouse chimeric antibodies, the constant region (C H ) of the heavy chain and the constant region (C L ) of the light chain are derived from a human antibody, while the variable region (V H ) of the heavy chain and the variable region (V L ) of the light chain are derived from a mouse antibody. Conversely, there are those in which the constant region (C H ) of the heavy chain and the constant region (C L ) of the light chain are derived from a mouse antibody, while the variable region (V H ) of the heavy chain and the variable region (V L ) of the light chain are derived from a human antibody.

[0019] An antibody originally has a basic structure consisting of a total of four polypeptide chains, namely two immunoglobulin light chains and two immunoglobulin heavy chains. However, in the present invention, when referring to an "antibody", in addition to those having this basic structure, (1) those consisting of a total of two polypeptide chains, one immunoglobulin light chain and one immunoglobulin heavy chain, and as will be described in detail below, (2) a single-chain antibody formed by binding a linker sequence to the C-terminal side of an immunoglobulin light chain and further binding an immunoglobulin heavy chain to the C-terminal side thereof, and (3) a single-chain antibody formed by binding a linker sequence to the C-terminal side of an immunoglobulin heavy chain and further binding an immunoglobulin light chain to the C-terminal side thereof are also included. Also, (4) those consisting of a Fab region in which the Fc region is deleted from the basic structure of an antibody in its original meaning, and those consisting of a Fab region and all or part of the hinge region (including Fab, F(ab’), and F(ab’)2) are also included in the "antibody" in the present invention.

[0020] Here, Fab refers to the variable region and CL A single light chain containing a region (the steady region of the light chain), and a variable region and C H This refers to a molecule in which one heavy chain, containing one region (part 1 of the constant region of the heavy chain), is linked to the cysteine ​​residues present in each region by disulfide bonds. In Fab, the heavy chain consists of a variable region and a C H In addition to one region (part 1 of the constant region of the heavy chain), a portion of the hinge region may also be included, but in this case, the hinge region lacks the cysteine ​​residues present in the hinge region that bind the heavy chains of the antibody together. In Fab, the light chain and the heavy chain are defined as the constant region of the light chain (C L Cysteine ​​residues located in the region and the constant region of the heavy chain (C H The bonds are formed by disulfide bonds between cysteine ​​residues located in the 1st region or the hinge region. The heavy chain that forms the Fab is called the Fab heavy chain. Since the Fab lacks cysteine ​​residues located in the hinge region that bind antibody heavy chains together, it consists of one light chain and one heavy chain. The light chain that makes up the Fab consists of a variable region and C L It includes a region. The heavy chains that make up Fab are variable regions and C H It may consist of one region, a variable region, C H It may also include a portion of the hinge region in addition to one region. However, in this case, the hinge region is selected so as not to contain cysteine ​​residues that connect the heavy chains, so that a disulfide bond is not formed between the two heavy chains in the hinge region. In F(ab'), the heavy chain consists of a variable region and C HIn addition to one region, it includes all or part of the hinge region containing cysteine ​​residues that link the heavy chains together. F(ab')2 refers to a molecule in which two F(ab') molecules are linked by disulfide bonds between cysteine ​​residues present in their hinge regions. The heavy chain that forms F(ab') or F(ab')2 is called the Fab' heavy chain. Furthermore, polymers such as dimers and trimers formed by the direct or linker linkage of multiple antibodies are also antibodies. Moreover, not limited to these, anything that includes a part of an immunoglobulin molecule and has the property of specifically binding to an antigen is included in the definition of "antibody" in this invention. That is, in this invention, when we refer to an immunoglobulin light chain, it includes those derived from an immunoglobulin light chain and having all or part of the amino acid sequence of its variable region. Similarly, when we refer to an immunoglobulin heavy chain, it includes those derived from an immunoglobulin heavy chain and having all or part of the amino acid sequence of its variable region. Therefore, as long as it has all or part of the amino acid sequence of the variable region, even if, for example, the Fc region is deleted, it is still an immunoglobulin heavy chain.

[0021] Furthermore, here, Fc or the Fc region refers to the C region in the antibody molecule. H This refers to a region containing a fragment consisting of two regions (part 2 of the steady-state region of the heavy chain) and a CH3 region (part 3 of the steady-state region of the heavy chain).

[0022] Furthermore, in this invention, when we refer to "antibody", (5) The present invention also includes scFab, scF(ab'), and scF(ab')2, which are single-chain antibodies obtained by linking the light chain and heavy chain constituting Fab, F(ab'), or F(ab')2 as shown in (4) above via a linker sequence. Here, in the case of scFab, scF(ab'), and scF(ab')2, the linker sequence may be attached to the C-terminus of the light chain, and the heavy chain may be attached to the C-terminus of the light chain, or the linker sequence may be attached to the C-terminus of the heavy chain, and the light chain may be attached to the C-terminus of the heavy chain. Furthermore, scFv, which is a single-chain antibody obtained by linking the variable region of the light chain and the variable region of the heavy chain via a linker sequence, is also included in the antibodies of the present invention. In the case of scFv, a linker array may be attached to the C-terminal side of the variable region of the light chain, and the variable region of the heavy chain may be attached to the C-terminal side of that linker array, or a linker array may be attached to the C-terminal side of the variable region of the heavy chain, and the variable region of the light chain may be attached to the C-terminal side of that linker array.

[0023] Furthermore, the term "antibody" as used herein includes not only full-length antibodies and those described in (1) to (5) above, but also a broader concept encompassing (4) and (5) above, which includes antigen-binding fragments (antibody fragments) in which a portion of a full-length antibody is missing.

[0024] The term "antigen-binding fragment" refers to a fragment of an antibody that retains at least some of its specific binding activity to an antigen. Examples of binding fragments include, for example, those shown in (4) and (5) above, Fab, Fab', F(ab')2, variable region (Fv), and variable region of the heavy chain (V H ) and the variable region of the light chain (V L A single-chain antibody (scFv) linked with a suitable linker, and a variable region (V) of the heavy chain. H ) and the variable region of the light chain (V L Diabody, scFv, is a polypeptide dimer containing a portion of the constant region (C) in its heavy chain (H chain). H3) This includes minibodies, which are dimers of the bound substance, and other low-molecular-weight antibodies. However, it is not limited to these molecules as long as they have the ability to bind to the antigen. Furthermore, these binding fragments include not only full-length antibody proteins treated with appropriate enzymes, but also proteins produced in appropriate host cells using genetically modified antibody genes.

[0025] In the present invention, "single-chain antibody" refers to a protein in which a linker sequence is bound to the C-terminus of an amino acid sequence containing all or part of the variable region of an immunoglobulin light chain, and further bound to the C-terminus of that amino acid sequence containing all or part of the variable region of an immunoglobulin heavy chain, and which can specifically bind to a specific antigen. Furthermore, a protein in which a linker sequence is bound to the C-terminus of an amino acid sequence containing all or part of the variable region of an immunoglobulin heavy chain, and further bound to the C-terminus of that amino acid sequence containing all or part of the variable region of an immunoglobulin light chain, and which can specifically bind to a specific antigen, is also considered a "single-chain antibody" in the present invention. For example, those described in (2) and (3) above are included in single-chain antibodies. In single-chain antibodies in which an immunoglobulin light chain is bound to the C-terminus of an immunoglobulin heavy chain via a linker sequence, the immunoglobulin heavy chain usually lacks an Fc region. The variable region of the immunoglobulin light chain has three complementarity-determining regions (CDRs) that are involved in the antigen specificity of the antibody. Similarly, the variable region of the immunoglobulin heavy chain also has three CDRs. These CDRs are the main regions that determine the antigen specificity of the antibody. Therefore, it is preferable that single-chain antibodies contain all three CDRs of the immunoglobulin heavy chain and all three CDRs of the immunoglobulin light chain. However, single-chain antibodies can also be made by deleting one or more CDRs, as long as the antigen-specific affinity of the antibody is maintained.

[0026] In single-chain antibodies, the linker sequence positioned between the light and heavy chains of immunoglobulin is preferably a peptide chain consisting of 2 to 50 amino acid residues, more preferably 8 to 50, even more preferably 10 to 30, and even more preferably 12 to 18 or 15 to 25, for example, 15 or 25 amino acid residues. Such linker sequences are not limited in their amino acid sequence, as long as the anti-hTfR antibody formed by linking the two chains maintains affinity for hTfR. Preferably, they consist of glycine alone or glycine and serine. For example, they may have the amino acid sequence Gly-Ser, Gly-Gly-Ser, Gly-Gly-Gly, Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 1), Gly-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 2), Ser-Gly-Gly-Gly-Gly-Gly (SEQ ID NO: 3), or sequences in which these amino acid sequences are repeated 2 to 10 times or 2 to 5 times. For example, when forming ScFV by attaching the variable region of an immunoglobulin light chain to the C-terminal side of an amino acid sequence consisting of the entire variable region of an immunoglobulin heavy chain via a linker sequence, a linker sequence consisting of a total of 15 amino acids (SEQ ID NO: 4), which corresponds to three consecutive Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 1), is preferably used.

[0027] In the present invention, the antigen specifically recognized by the antibody is, for example, a molecule present on the surface of vascular endothelial cells (surface antigen). Examples of such surface antigens include, but are not limited to, transferrin receptors (TfR), insulin receptors, leptin receptors, lipoprotein receptors, IGF receptors, organic anion transporters such as OATP-F, monocarboxylic acid transporters such as MCT-8, and Fc receptors. The antigen is preferably one of these molecules present on the surface of human vascular endothelial cells (surface antigen).

[0028] Among the surface antigens mentioned above, transferrin receptors (TfR), insulin receptors, leptin receptors, lipoprotein receptors, IGF receptors, organic anion transporters such as OATP-F, and monocarboxylic acid transporters such as MCT-8 are present on the surface of brain capillary endothelial cells (cerebral vascular endothelial cells) that form the blood-brain barrier. Antibodies that can recognize these antigens can bind to brain capillary endothelial cells via the antigen. Antibodies bound to brain capillary endothelial cells can then cross the blood-brain barrier and reach the central nervous system. Therefore, by binding the target protein to such an antibody, it can reach the central nervous system. Target proteins include proteins that have a function that should exert their pharmacological effect in the central nervous system. For example, lysosomal enzymes that are deficient or dysfunctional in patients with lysosomal storage diseases accompanied by central nervous system disorders are examples of target proteins. These lysosomal enzymes cannot reach the central nervous system on their own and therefore do not exert any therapeutic effect on central nervous system disorders in patients. However, by binding them with these antibodies, they can cross the blood-brain barrier, thereby improving central nervous system disorders seen in patients with lysosomal storage diseases.

[0029] In the present invention, the terms "human transferrin receptor" or "hTfR" refer to a membrane protein having the amino acid sequence shown in SEQ ID NO: 5. In one embodiment, the anti-hTfR antibody of the present invention specifically binds to the portion of the amino acid sequence shown in SEQ ID NO: 5 from the 89th cysteine ​​residue from the N-terminus to the phenylalanine residue at the C-terminus (the extracellular region of hTfR), but is not limited thereto.

[0030] The method for producing antibodies will be explained below, using antibodies against hTfR as an example. The general method for producing antibodies against hTfR involves creating recombinant human transferrin receptors (rhTfR) using cells into which an expression vector incorporating the hTfR gene has been introduced, and then immunizing animals such as mice with these rhTfR. By extracting antibody-producing cells against hTfR from the immunized animals and fusing them with myeloma cells, hybridoma cells capable of producing antibodies against hTfR can be produced.

[0031] Furthermore, cells that produce antibodies against hTfR can also be obtained by immunizing immune system cells obtained from animals such as mice with rhTfR using in vitro immunization. When immunizing by in vitro immunization, there are no particular limitations on the animal species from which the immune system cells originate, but preferably they are primates including mice, rats, rabbits, guinea pigs, dogs, cats, horses, and humans, more preferably mice, rats, and humans, and even more preferably mice and humans. As mouse immune system cells, for example, splenocytes prepared from the spleen of a mouse can be used. As human immune system cells, cells prepared from human peripheral blood, bone marrow, spleen, etc. can be used. When human immune system cells are immunized by in vitro immunization, human antibodies against hTfR can be obtained.

[0032] In the present invention, there are no particular limitations on the human lysosomal enzyme to be bound to the antibody, but examples include α-L-iduronidase, iduronic acid-2-sulfatase, glucocerebrosidase, β-galactosidase, GM2-activating protein, β-hexosaminidase A, β-hexosaminidase B, N-acetylglucosamine-1-phosphotransferase, α-mannosidase, β-mannosidase, galactosylceramidase, saposin C, arylsulfatase A, α-L-fucosidase, aspartylglucosaminidase, α-N-acetylgalactosaminidase, and acid sphingomyelinase. Examples include α-galactosidase A, β-glucuronidase, heparan N-sulfatase, α-N-acetylglucosaminidase, acetyl-CoA α-glucosaminide N-acetyltransferase, N-acetylglucosamine-6-sulfate sulfatase, acid ceramidase, amyl-1,6-glucosidase, sialidase, aspartylglucosaminidase, palmitoyl protein thioesterase-1 (PPT-1), tripeptidyl peptidase-1 (TPP-1), hyaluronidase-1, CLN1, CLN2, CLN3, CLN6, and CLN8, which are lysosomal enzymes.

[0033] When an antibody specifically recognizes molecules (surface antigens) present on the surface of vascular endothelial cells, the human lysosomal enzymes conjugated to the antibody are used as follows: α-L-iduronidase as a treatment for central nervous system disorders in Hurler syndrome, Hurler-Schaye syndrome, and Schayet syndrome; iduronate-2-sulfatase as a treatment for central nervous system disorders in Hunter syndrome; glucocerebrosidase as a treatment for central nervous system disorders in Gaucher disease; β-galactosidase as a treatment for central nervous system disorders in GM1 gangliosidosis types 1-3; and GM2 activating protein White matter is used as a treatment for central nervous system disorders in GM2-gangliosidosis AB variant, β-hexosaminidase A as a treatment for central nervous system disorders in Sandhoff disease and Tisachs disease, β-hexosaminidase B as a treatment for central nervous system disorders in Sandhoff disease, N-acetylglucosamine-1-phosphotransferase as a treatment for central nervous system disorders in I-cell diseases, α-mannosidase as a treatment for central nervous system disorders in α-mannosidosis, β-mannosidase as a treatment for central nervous system disorders in β-mannosidosis, and galactosidase Luceramidase is used as a treatment for central nervous system disorders in Krabbe disease, saposin C as a treatment for central nervous system disorders in Gaucher disease-like stocia, arylsulfatase A as a treatment for central nervous system disorders in metachromatic leukodystrophy, α-L-fucosidase as a treatment for central nervous system disorders in fucosidosis, aspartylglucosaminidase as a treatment for central nervous system disorders in aspartylglucosamineuria, and α-N-acetylgalactosaminidase as a treatment for central nervous system disorders in Schindler's disease and Kawasaki disease. Myelinase is used as a treatment for central nervous system disorders in Niemann-Pick disease, α-galactosidase A as a treatment for central nervous system disorders in Fabry disease, β-glucuronidase as a treatment for central nervous system disorders in Sly syndrome, heparan N-sulfatase, α-N-acetylglucosaminidase, acetyl-CoA α-glucosaminide N-acetyltransferase, and N-acetylglucosamine-6-sulfate sulfatase as treatments for central nervous system disorders in Sanfilippo syndrome, and acidic ceramidase as a treatment for central nervous system disorders in Faber disease.Amylo-1,6-glucosidase can be used as a treatment for central nervous system disorders in Forbes-Koli's disease, sialidase as a treatment for central nervous system disorders in sialidase deficiency, aspartylglucosaminidase as a treatment for central nervous system disorders in aspartylglucosamineuria, palmitoylprotein thioesterase-1 (PPT-1) as a treatment for central nervous system disorders in neuronal ceroid lipofuscinosis or Santavuori-Haltia disease, tripeptidyl peptidase-1 (TPP-1) as a treatment for central nervous system disorders in neuronal ceroid lipofuscinosis or Jansky-Bielschowsky disease, hyaluronidase-1 as a treatment for central nervous system disorders in hyaluronidase deficiency, and CLN1, CLN2, CLN3, CLN6, and CLN8 as treatments for central nervous system disorders in Batten disease.

[0034] When an antibody specifically recognizes molecules (surface antigens) present on the surface of vascular endothelial cells, human α-L-iduronidase (hIDUA) is a suitable lysosomal enzyme to bind to the antibody. hIDUA is a type of lysosomal enzyme that has the activity to hydrolyze iduronic acid bonds present in glycosaminoglycan (GAG) molecules such as heparan sulfate and dermatan sulfate. Mucopolysaccharidosis type I is a genetic disorder caused by a mutation in the gene encoding this enzyme. Mucopolysaccharidosis type I is classified into Hurler syndrome, Hurler-Scheille syndrome, and Schaye syndrome, with Hurler syndrome being the severe form, Hurler-Scheille syndrome being the intermediate form, and Schaye syndrome being the mild form. In these patients, heparan sulfate and dermatan sulfate accumulate in the tissues, resulting in various symptoms such as corneal opacity and intellectual disability. However, intellectual disability may not be observed in the mild form. The fusion protein of this antibody and hIDUA can degrade GAGs accumulated in brain tissue by crossing the blood-brain barrier (BBB), and therefore can be used as a treatment for central nervous system disorders when administered to patients with Hurler syndrome who exhibit intellectual disability.

[0035] In the present invention, the terms "human α-L-idulonidase" or "hIDUA" specifically refer to hIDUA having the same amino acid sequence as wild-type hIDUA. Wild-type hIDUA has an amino acid sequence consisting of 628 amino acids, as shown in SEQ ID NO: 6. A variant of hIDUA having an amino acid sequence consisting of 626 amino acids, as shown in SEQ ID NO: 7, is also considered hIDUA. However, it is not limited to these; as long as it has IDUA activity, any variant of wild-type hIDUA with mutations such as substitution, deletion, or addition to its amino acid sequence is also included as hIDUA. When substituting amino acids in the amino acid sequence of hIDUA with other amino acids, the number of amino acids to be substituted is preferably 1 to 10, more preferably 1 to 5, even more preferably 1 to 3, and even more preferably 1 to 2. When deleting amino acids from the amino acid sequence of hIDUA, the number of amino acids deleted is preferably 1 to 10, more preferably 1 to 5, even more preferably 1 to 3, and even more preferably 1 to 2. Mutations combining these amino acid substitutions and deletions can also be introduced. When adding amino acids to hIDUA, preferably 1 to 10, more preferably 1 to 5, even more preferably 1 to 3, and even more preferably 1 to 2 amino acids are added to the amino acid sequence of hIDUA or to the N-terminus or C-terminus. Mutations combining these amino acid additions, substitutions, and deletions can also be introduced. The amino acid sequence of the mutated hIDUA preferably has 80% or more identity with the original hIDUA amino acid sequence, more preferably 85% or more identity, even more preferably 90% or more identity, even more preferably 95% or more identity, and even more preferably 99% or more identity.

[0036] In this invention, the identity between the amino acid sequence of the original protein (including the antibody) and the amino acid sequence of the mutated protein can be easily calculated using well-known identity calculation algorithms. Examples of such algorithms include BLAST (Altschul SF. J Mol . Biol. 215. 403-10, (1990)), the Pearson and Lipman similarity search method (Proc. Natl. Acad. Sci. USA. 85. 2444 (1988)), and the Smith and Waterman local identity algorithm (Adv. Appl. Math. 2. 482-9 (1981)).

[0037] Furthermore, substitutions of amino acids in the amino acid sequence of the original protein (including antibodies) occur, for example, within amino acid families that are related by their side chains and chemical properties. Such amino acid families include the following: (1) Aspartic acid and glutamic acid, which are acidic amino acids, (2) Basic amino acids histidine, lysine, and arginine, (3) Aromatic amine acids such as phenylalanine, tyrosine, tryptophan, (4) Serine and threonine, which are hydroxy amino acids, (5) Hydrophobic amino acids methionine, alanine, valine, leucine, and isoleucine, (6) Neutral hydrophilic amino acids such as cysteine, serine, threonine, asparagine, and glutamine, (7) Glycine and proline, amino acids that affect the orientation of peptide chains, (8) Asparagine and glutamine, which are amide amino acids, (9) Aliphatic amino acids, alanine, leucine, isoleucine, and valine, (10) Alanine, glycine, serine, and threonine, which are amino acids with small side chains. (11) Alanine and glycine, which are amino acids with particularly small side chains, (12) Valine, leucine, and isoleucine, which are branched amino acids.

[0038] In this invention, when hIDUA is said to have IDUA activity, it means that when hIDUA is fused with an antibody to form a fusion protein, it has 3% or more of the activity inherently possessed by the native type of hIDUA. However, it is preferable that the activity be 10% or more, more preferably 20% or more, even more preferably 50% or more, and even more preferably 80% or more, compared to the activity inherently possessed by the native type of hIDUA. The same applies when the hIDUA fused with the antibody has been mutated. The antibody is, for example, an anti-hTfR antibody.

[0039] In this invention, "fusion protein" refers to a substance obtained by directly or via a non-peptide linker or a peptide linker by linking an antibody and a human lysosomal enzyme. The method for linking the antibody and the human lysosomal enzyme is described in detail below.

[0040] Methods for conjugating antibodies to human lysosomal enzymes include methods using non-peptide linkers or methods using peptide linkers. Non-peptide linkers can include polyethylene glycol, polypropylene glycol, copolymers of ethylene glycol and propylene glycol, polyoxyethylated polyols, polyvinyl alcohol, polysaccharides, dextran, polyvinyl ethers, biodegradable polymers, lipid polymers, chitins, and hyaluronic acid, or derivatives thereof, or combinations thereof. A peptide linker is a peptide chain or derivative thereof consisting of 1 to 50 amino acids linked by peptide bonds, and its N-terminus and C-terminus form covalent bonds with either the antibody or the human lysosomal enzyme, thereby conjugating the antibody to the human lysosomal enzyme.

[0041] When using biotin-streptavidin as a non-peptide linker, the antibody may be conjugated with biotin, and the human lysosomal enzyme may be conjugated with streptavidin, and the antibody and human lysosomal enzyme may be conjugated via the binding of biotin and streptavidin. Conversely, the antibody may be conjugated with streptavidin, and the human lysosomal enzyme may be conjugated with biotin, and the antibody and human lysosomal enzyme may be conjugated via the binding of biotin and streptavidin.

[0042] The antibody of the present invention, conjugated with a human lysosomal enzyme using PEG as a non-peptide linker, is specifically called antibody-PEG-human lysosomal enzyme. Antibody-PEG-human lysosomal enzyme can be produced by conjugating an antibody with PEG to create antibody-PEG, and then conjugating antibody-PEG with human lysosomal enzyme. Alternatively, antibody-PEG-human lysosomal enzyme can also be produced by conjugating a human lysosomal enzyme with PEG to create human lysosomal enzyme-PEG, and then conjugating human lysosomal enzyme-PEG with antibody. When conjugating PEG with antibodies and human lysosomal enzyme, PEG modified with functional groups such as carbonate, carbonylimidazole, active carboxylic acid esters, azulactone, cyclic imidothion, isocyanate, isothiocyanate, imidate, or aldehyde is used. These functional groups introduced into PEG react mainly with amino groups in the antibody and human lysosomal enzyme molecules, resulting in covalent bonding between PEG and the antibody and human lysosomal enzyme. There are no particular limitations on the molecular weight and shape of the PEG used in this case, but its average molecular weight (MW) is preferably MW = 300 to 60000, and more preferably MW = 500 to 20000. For example, PEG with average molecular weights of approximately 300, 500, 1000, 2000, 4000, 10000, and 20000 can be suitably used as a non-peptide linker.

[0043] For example, antibody-PEG can be obtained by mixing an antibody with polyethylene glycol having an aldehyde group as a functional group (ALD-PEG-ALD) so that the molar ratio of ALD-PEG-ALD to the antibody is 11, 12.5, 15, 110, 120, etc., and then adding a reducing agent such as NaCNBH3 and reacting the mixture. Next, antibody-PEG-human lysosomal enzyme can be obtained by reacting antibody-PEG with human lysosomal enzyme in the presence of a reducing agent such as NaCNBH3. Conversely, antibody-PEG-human lysosomal enzyme can also be obtained by first conjugating human lysosomal enzyme with ALD-PEG-ALD to produce human lysosomal enzyme-PEG, and then conjugating human lysosomal enzyme-PEG with an antibody.

[0044] Antibodies and human lysosomal enzymes can also be linked to the N-terminus or C-terminus of the N-terminus or C-terminus of the antibody's heavy or light chain via a linker sequence, either directly or by peptide bond. A fusion protein formed by linking an antibody and a human lysosomal enzyme in this way can be obtained by incorporating a DNA fragment in which the cDNA encoding the human lysosomal enzyme is positioned in-frame, either directly or with a DNA fragment encoding the linker sequence flanked by the 3' or 5' end of the cDNA encoding the antibody's heavy or light chain, into an expression vector for eukaryotes such as mammalian cells or yeast, and then culturing mammalian cells into which this expression vector has been introduced. In the case where the DNA fragment encoding the human lysosomal enzyme is linked to the heavy chain, an expression vector for mammalian cells incorporating the cDNA fragment encoding the antibody's light chain is also introduced into the same host cells. Similarly, in the case where the DNA fragment encoding the human lysosomal enzyme is linked to the light chain, an expression vector for mammalian cells incorporating the cDNA fragment encoding the antibody's heavy chain is also introduced into the same host cells. When the antibody is a single-chain antibody, a fusion protein formed by linking the antibody to a human lysosomal enzyme can be obtained by incorporating a DNA fragment into an expression vector for mammalian cells, yeast, or other eukaryotic cells. This DNA fragment consists of the single-chain antibody-encoding cDNA directly attached to the 5' or 3' end of the cDNA encoding the human lysosomal enzyme, or with a DNA fragment encoding a linker sequence sandwiched between them. The expression vector is then introduced into these cells.

[0045] A fusion protein of the type in which a human lysosomal enzyme is bound to the C-terminus of the antibody's light chain is one in which the antibody contains an amino acid sequence that includes all or part of the variable region of the light chain and an amino acid sequence that includes all or part of the variable region of the heavy chain, and the human lysosomal enzyme is bound to the C-terminus of the antibody's light chain. Here, the antibody's light chain and the human lysosomal enzyme may be bound directly or via a linker.

[0046] A fusion protein of the type in which a human lysosomal enzyme is bound to the C-terminus of the antibody's heavy chain is one in which the antibody contains an amino acid sequence that includes all or part of the variable region of the light chain and an amino acid sequence that includes all or part of the variable region of the heavy chain, and the human lysosomal enzyme is bound to the C-terminus of the antibody's heavy chain. Here, the antibody's heavy chain and the human lysosomal enzyme may be bound directly or via a linker.

[0047] A fusion protein of the type in which a human lysosomal enzyme is bound to the N-terminus of the antibody light chain is one in which the antibody contains an amino acid sequence that includes all or part of the variable region of the light chain and an amino acid sequence that includes all or part of the variable region of the heavy chain, and the human lysosomal enzyme is bound to the N-terminus of the antibody light chain. Here, the antibody light chain and the human lysosomal enzyme may be bound directly or via a linker.

[0048] A fusion protein of the type in which a human lysosomal enzyme is bound to the N-terminus of the antibody's heavy chain is one in which the antibody contains an amino acid sequence that includes all or part of the variable region of the light chain and an amino acid sequence that includes all or part of the variable region of the heavy chain, and the human lysosomal enzyme is bound to the N-terminus of the antibody's heavy chain. Here, the antibody's heavy chain and the human lysosomal enzyme may be bound directly or via a linker.

[0049] When a linker sequence is placed between the antibody and the human lysosomal enzyme, the sequence is preferably composed of 1 to 50 amino acids, more preferably 1 to 17, even more preferably 1 to 10, and even more preferably 1 to 5 amino acids. However, the number of amino acids constituting the linker sequence can be appropriately adjusted to 1, 2, 3, 1 to 17, 1 to 10, 10 to 40, 20 to 34, 23 to 31, 25 to 29, etc., depending on the human lysosomal enzyme to be bound to the antibody. Such linker sequences are not limited in their amino acid sequence, as long as the antibody linked by them maintains affinity to hTfR and the human lysosomal enzyme linked by the linker sequence exhibits the physiological activity of the protein under physiological conditions. However, they are preferably composed of glycine and serine, and for example, they may consist of one amino acid, either glycine or serine, the amino acid sequence Gly-Ser, Gly-Gly-Ser, Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 1), Gly-Gly-Gly-Gly-Gly-Ser (SEQ ID NO: 2), the amino acid sequence Ser-Gly-Gly-Gly-Gly-Gly (SEQ ID NO: 3), or sequences consisting of 1 to 50 amino acids formed by 1 to 10 or 2 to 5 consecutive such amino acid sequences, or sequences consisting of 2 to 17, 2 to 10, 10 to 40, 20 to 34, 23 to 31, or 25 to 29 amino acids. For example, sequences consisting of the amino acid sequence Gly-Ser, or sequences consisting of 15 amino acids with three consecutive Gly-Gly-Gly-Ser sequences (SEQ ID NO: 1) (SEQ ID NO: 4), can be suitably used as linker sequences. The same applies even if the antibody is a single-chain antibody.

[0050] In this invention, if a single peptide chain contains multiple linker sequences, for convenience, each linker sequence will be named sequentially from the N-terminus as the first linker sequence, the second linker sequence, and so on.

[0051] As a preferred embodiment of the antibody when the antibody is an anti-human transferrin receptor antibody, In the variable region of the light chain: (a) CDR1 comprises the amino acid sequence shown in SEQ ID NO: 8 or 9, (b) CDR2 comprises the amino acid sequence shown in SEQ ID NO: 10 or 11, and (c) CDR3 comprises the amino acid sequence shown in SEQ ID NO: 12; In the variable region of the heavy chain: (d) CDR1 comprises the amino acid sequence shown in SEQ ID NO: 13 or 14, (e) CDR2 comprises the amino acid sequence shown in SEQ ID NO: 15 or 16, and (f) An example of an antibody is one in which CDR3 comprises the amino acid sequence shown in SEQ ID NO: 17 or 18. Here, the antibody is preferably a human antibody or a humanized antibody.

[0052] The amino acid sequence combinations of each CDR shown in (a) to (f) above may be any combination, for example, those shown in Table 1.

[0053] [Table 1]

[0054] As a further preferred embodiment of the antibody, when the antibody is an anti-human transferrin receptor antibody, (x) An example of an antibody is one in which the variable region of the light chain has the amino acid sequence shown in SEQ ID NO: 20 and the variable region of the heavy chain includes the amino acid sequence shown in SEQ ID NO: 21. Here, the antibody is preferably a human antibody or a humanized antibody. Here, the amino acid sequence of the variable region of the light chain, indicated by SEQ ID NO: 20, includes the amino acid sequences indicated by SEQ ID NOs: 8 and 9 as CDR1, the amino acid sequences indicated by SEQ ID NOs: 10 and 11 as CDR2, and the amino acid sequence indicated by SEQ ID NO: 12 as CDR3. Furthermore, the amino acid sequence of the variable region of the heavy chain, indicated by SEQ ID NO: 21, includes the amino acid sequences indicated by SEQ ID NOs: 13 and 14 as CDR1, the amino acid sequences indicated by SEQ ID NOs: 15 and 16 as CDR2, and the amino acid sequences indicated by SEQ ID NOs: 17 and 18 as CDR3. In addition, the amino acid sequence indicated by SEQ ID NO: 19 is included as the heavy chain framework region 3.

[0055] However, the preferred embodiment of the antibody when the antibody is a humanized antibody and an anti-human transferrin receptor antibody is not limited to (x) above. For example, an antibody in which the amino acid sequence of the variable region of the light chain has 80% or more identity with the amino acid sequence of the variable region of the light chain in (x) above, and the amino acid sequence of the variable region of the heavy chain has 80% or more identity with the amino acid sequence of the variable region of the heavy chain in (x) above, can also be used in the present invention as long as it has affinity for hTfR.

[0056] Furthermore, an antibody in which the amino acid sequence of the variable region of the light chain has 85% or more identity with the amino acid sequence of the variable region of the light chain in (x) above, and the amino acid sequence of the variable region of the heavy chain has 85% or more identity with the amino acid sequence of the variable region of the heavy chain in (x) above, An antibody in which the amino acid sequence of the variable region of the light chain has 90% or more identity with the amino acid sequence of the variable region of the light chain in (x) above, and the amino acid sequence of the variable region of the heavy chain has 90% or more identity with the amino acid sequence of the variable region of the heavy chain in (x) above, and Antibodies in which the amino acid sequence of the variable region of the light chain has 95% or more identity with the amino acid sequence of the variable region of the light chain in (x) above, and the amino acid sequence of the variable region of the heavy chain has 95% or more identity with the amino acid sequence of the variable region of the heavy chain in (x) above, can also be used in the present invention, as long as they have affinity for hTfR.

[0057] As an antibody in which the amino acid sequence of the variable region of the light chain is identical to the amino acid sequence of the variable region of the light chain in (x) above, and the amino acid sequence of the variable region of the heavy chain is identical to the amino acid sequence of the variable region of the heavy chain in (x) above, (xa) The variable region of the light chain contains an amino acid sequence having 80% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 20, and CDR1 contains the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 contains the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 contains the amino acid sequence of SEQ ID NO: 12; the variable region of the heavy chain contains an amino acid sequence having 80% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 21, and CDR1 contains the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 contains the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 contains the amino acid sequence of SEQ ID NO: 17 or 18. (xb) The variable region of the light chain contains an amino acid sequence having 85% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 20, and CDR1 contains the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 contains the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 contains the amino acid sequence of SEQ ID NO: 12; the variable region of the heavy chain contains an amino acid sequence having 85% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 21, and CDR1 contains the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 contains the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 contains the amino acid sequence of SEQ ID NO: 17 or 18. (xc) The variable region of the light chain contains an amino acid sequence having 90% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 20, and CDR1 contains the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 contains the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 contains the amino acid sequence of SEQ ID NO: 12; the variable region of the heavy chain contains an amino acid sequence having 90% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 21, and CDR1 contains the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 contains the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 contains the amino acid sequence of SEQ ID NO: 17 or 18. (xd) The variable region of the light chain contains an amino acid sequence having 95% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 20, and CDR1 contains the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 contains the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 contains the amino acid sequence of SEQ ID NO: 12; the variable region of the heavy chain contains an amino acid sequence having 95% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 21, and CDR1 contains the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 contains the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 contains the amino acid sequence of SEQ ID NO: 17 or 18. These are some examples.

[0058] Furthermore, as a preferred embodiment of the antibody when the antibody is also an anti-human transferrin receptor antibody, In the amino acid sequence of the variable region of the light chain in (x) above, 1 to 5 amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, and in the amino acid sequence of the variable region of the heavy chain in (x) above, 1 to 5 amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, In the amino acid sequence of the variable region of the light chain in (x) above, 1 to 3 amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, and in the amino acid sequence of the variable region of the heavy chain in (x) above, 1 to 3 amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, and In the above (x), an amino acid sequence in the variable region of the light chain in which one or two amino acids are substituted, deleted, or added, and in the above (x), an amino acid sequence in the variable region of the heavy chain in which one or two amino acids are substituted, deleted, or added, can also be used in the present invention as long as it has affinity for hTfR. Here, the antibody is preferably a human antibody or a humanized antibody.

[0059] As an antibody in which the amino acid sequence of the variable region of the light chain of the antibody is such that the amino acids in the amino acid sequence constituting the variable region of the light chain in (x) above are substituted, deleted, or added, and the amino acid sequence of the variable region of the heavy chain in (x) above is such that the amino acids in the amino acid sequence constituting it are substituted, deleted, or added, (xe) In the variable region of the light chain, the amino acid sequence is such that 1 to 5 amino acids are substituted, deleted, or added, and CDR1 contains the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 contains the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 contains the amino acid sequence of SEQ ID NO: 12; and in the variable region of the heavy chain, the amino acid sequence is such that 1 to 5 amino acids are substituted, deleted, or added, and CDR1 contains the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 contains the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 contains the amino acid sequence of SEQ ID NO: 17 or 18. (xf) In the variable region of the light chain, the amino acid sequence is such that 1 to 3 amino acids are substituted, deleted, or added, and CDR1 contains the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 contains the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 contains the amino acid sequence of SEQ ID NO: 12; and in the variable region of the heavy chain, the amino acid sequence is such that 1 to 3 amino acids are substituted, deleted, or added, and CDR1 contains the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 contains the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 contains the amino acid sequence of SEQ ID NO: 17 or 18. (xg) In the variable region of the light chain, one or two amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, and CDR1 includes the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 includes the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 includes the amino acid sequence of SEQ ID NO: 12; and in the variable region of the heavy chain, one or two amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, and CDR1 includes the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 includes the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 includes the amino acid sequence of SEQ ID NO: 17 or 18. These are some examples.

[0060] In (xa) to (xg) above, the combination of amino acid sequences for each CDR can be any combination, for example, those shown in Table 2. The same applies to (ya) to (yg) below.

[0061] [Table 2]

[0062] In the present invention, as a preferred embodiment in which the antibody is a humanized antibody and is an anti-human transferrin receptor antibody, (y) An example of a Fab antibody is one in which the light chain contains the amino acid sequence shown in SEQ ID NO: 22 and the heavy chain contains the amino acid sequence shown in SEQ ID NO: 23. Here, the light chain contains the amino acid sequence shown in SEQ ID NO: 20 as a variable region, and the heavy chain contains the amino acid sequence shown in SEQ ID NO: 21 as a variable region. In the present invention, the heavy chain constituting the Fab is called the Fab heavy chain. That is, the heavy chain consisting of the amino acid sequence shown in SEQ ID NO: 23 is the Fab heavy chain.

[0063] When the antibody is a humanized antibody and an anti-human transferrin receptor antibody, the preferred embodiment of the antibody is not limited to (y) above. For example, an antibody in which the amino acid sequence of the light chain has 80% or more identity with the amino acid sequence of the light chain in (y) above, and the amino acid sequence of the heavy chain has 80% or more identity with the amino acid sequence of the heavy chain in (y) above, can also be used in the present invention, as long as it has affinity for hTfR.

[0064] Furthermore, an antibody in which the amino acid sequence of the light chain has 85% or more identity with the amino acid sequence of the light chain in (y) above, and the amino acid sequence of the heavy chain has 85% or more identity with the amino acid sequence of the heavy chain in (y) above, An antibody in which the amino acid sequence of the light chain has 90% or more identity with the amino acid sequence of the light chain in (y) above, and the amino acid sequence of the heavy chain has 90% or more identity with the amino acid sequence of the heavy chain in (y) above, and Antibodies having a light chain amino acid sequence that is 95% or more identical to the light chain amino acid sequence in (y) above, and a heavy chain amino acid sequence that is 95% or more identical to the heavy chain amino acid sequence in (y) above, can also be used in the present invention, as long as they have affinity for hTfR.

[0065] As an antibody in which the amino acid sequence of the light chain is identical to the amino acid sequence of the light chain in (y) above, and the amino acid sequence of the heavy chain is identical to the amino acid sequence of the heavy chain in (y) above, (ya) The light chain contains an amino acid sequence having 80% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 22, and comprises the amino acid sequence of SEQ ID NO: 8 or 9 as CDR1, the amino acid sequence of SEQ ID NO: 10 or 11 as CDR2, and the amino acid sequence of SEQ ID NO: 12 as CDR3, and the heavy chain contains an amino acid sequence having 80% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 23, and comprises the amino acid sequence of SEQ ID NO: 13 or 14 as CDR1, the amino acid sequence of SEQ ID NO: 15 or 16 as CDR2, and the amino acid sequence of SEQ ID NO: 17 or 18 as CDR3, (yb) The light chain contains an amino acid sequence having 85% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 22, and comprises the amino acid sequence of SEQ ID NO: 8 or 9 as CDR1, the amino acid sequence of SEQ ID NO: 10 or 11 as CDR2, and the amino acid sequence of SEQ ID NO: 12 as CDR3, and the heavy chain contains an amino acid sequence having 85% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 23, and comprises the amino acid sequence of SEQ ID NO: 13 or 14 as CDR1, the amino acid sequence of SEQ ID NO: 15 or 16 as CDR2, and the amino acid sequence of SEQ ID NO: 17 or 18 as CDR3, (yc) The light chain contains an amino acid sequence having 90% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 22, and comprises the amino acid sequence of SEQ ID NO: 8 or 9 as CDR1, the amino acid sequence of SEQ ID NO: 10 or 11 as CDR2, and the amino acid sequence of SEQ ID NO: 12 as CDR3, and the heavy chain contains an amino acid sequence having 90% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 23, and comprises the amino acid sequence of SEQ ID NO: 13 or 14 as CDR1, the amino acid sequence of SEQ ID NO: 15 or 16 as CDR2, and the amino acid sequence of SEQ ID NO: 17 or 18 as CDR3, (yd) A light chain comprising an amino acid sequence having 95% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 22, and comprising the amino acid sequence of SEQ ID NO: 8 or 9 as CDR1, the amino acid sequence of SEQ ID NO: 10 or 11 as CDR2, and the amino acid sequence of SEQ ID NO: 12 as CDR3, and a heavy chain comprising an amino acid sequence having 95% or more sequence identity with the amino acid sequence shown in SEQ ID NO: 23, and comprising the amino acid sequence of SEQ ID NO: 13 or 14 as CDR1, the amino acid sequence of SEQ ID NO: 15 or 16 as CDR2, and the amino acid sequence of SEQ ID NO: 17 or 18 as CDR3.

[0066] Furthermore, in the case where the antibody is a humanized antibody and is an anti-human transferrin receptor antibody, a preferred embodiment of the antibody is that the amino acid sequence of the antibody is In the amino acid sequence of the light chain in (y) above, 1 to 5 amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, and in the amino acid sequence of the heavy chain in (y) above, 1 to 5 amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, In the amino acid sequence of the light chain in (y) above, 1 to 3 amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, and in the amino acid sequence of the heavy chain in (y) above, 1 to 3 amino acids are substituted, deleted, or added to the amino acid sequence that constitutes it, and In the above (y), an amino acid sequence of the light chain in which one or two amino acids are substituted, deleted, or added, and in the above (y), an amino acid sequence of the heavy chain in which one or two amino acids are substituted, deleted, or added, can also be used in the present invention as long as it has affinity for hTfR.

[0067] As an antibody in which the amino acid sequence of the light chain is such that the amino acids in the amino acid sequence constituting the light chain in (y) above are substituted, deleted, or added, and the amino acid sequence of the heavy chain in (y) above is such that the amino acids in the amino acid sequence constituting the heavy chain are substituted, deleted, or added, (ye) In the light chain amino acid sequence, 1 to 5 amino acids are substituted, deleted, or added, and CDR1 contains the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 contains the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 contains the amino acid sequence of SEQ ID NO: 12; and in the heavy chain amino acid sequence, 1 to 5 amino acids are substituted, deleted, or added, and CDR1 contains the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 contains the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 contains the amino acid sequence of SEQ ID NO: 17 or 18. (yf) In the amino acid sequence of the light chain, 1 to 3 amino acids are substituted, deleted, or added, and CDR1 contains the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 contains the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 contains the amino acid sequence of SEQ ID NO: 12; and in the amino acid sequence of the heavy chain, 1 to 3 amino acids are substituted, deleted, or added, and CDR1 contains the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 contains the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 contains the amino acid sequence of SEQ ID NO: 17 or 18. (yg) In the light chain amino acid sequence, one or two amino acids are substituted, deleted, or added to the amino acid sequence constituting it, and CDR1 contains the amino acid sequence of SEQ ID NO: 8 or 9, CDR2 contains the amino acid sequence of SEQ ID NO: 10 or 11, and CDR3 contains the amino acid sequence of SEQ ID NO: 12; and in the heavy chain amino acid sequence, one or two amino acids are substituted, deleted, or added to the amino acid sequence constituting it, and CDR1 contains the amino acid sequence of SEQ ID NO: 13 or 14, CDR2 contains the amino acid sequence of SEQ ID NO: 15 or 16, and CDR3 contains the amino acid sequence of SEQ ID NO: 17 or 18. These are some examples.

[0068] When the antibody is a humanized anti-human transferrin receptor antibody and the human lysosomal enzyme is human α-L-iduronidase (hIDUA), the following are preferred forms of the fusion protein: A fusion protein comprising a light chain of a humanized anti-hTfR antibody having the amino acid sequence shown in SEQ ID NO: 22, and a Fab heavy chain of a humanized anti-hTfR antibody having the amino acid sequence shown in SEQ ID NO: 23, with human α-L-iduronidase, shown in SEQ ID NO: 6, bound to the C-terminal side via the linker sequence shown in SEQ ID NO: 4.

[0069] Furthermore, when the antibody is a humanized anti-human transferrin receptor antibody and the human lysosomal enzyme is human α-L-iduronidase (hIDUA), the following are considered preferred forms of the fusion protein. That is, The light chain of the humanized anti-hTfR antibody has the amino acid sequence shown in SEQ ID NO: 22, A fusion protein in which the Fab heavy chain of the humanized anti-hTfR antibody is bound at its C-terminus to human α-L-idulonidase shown in SEQ ID NO: 6 via a linker having the amino acid sequence shown in SEQ ID NO: 4, and the entire protein has the amino acid sequence shown in SEQ ID NO: 27.

[0070] In the present invention, a fusion protein of an antibody and a human lysosomal enzyme can be produced by culturing mammalian cells that have been artificially manipulated to produce the fusion protein by expressing or strongly expressing the gene encoding the fusion protein. In this case, the gene to be strongly expressed in the mammalian cells that produce the fusion protein is generally introduced into the mammalian cells by transforming them with an expression vector into which the gene is incorporated. There are no particular limitations on the mammalian cells used in this process, but human, mouse, and Chinese hamster-derived cells are preferred, and CHO cells derived from Chinese hamster ovary cells are particularly preferred. In the present invention, when referring to a fusion protein, it specifically refers to a recombinant fusion protein that is secreted into the culture medium when mammalian cells that produce such a fusion protein are cultured.

[0071] A fusion protein of an antibody and a human lysosomal enzyme can also be produced by first producing the antibody and the human lysosomal enzyme separately, and then linking them with a non-peptide linker or a peptide linker. In this case, the antibody and the human lysosomal enzyme can be produced as recombinant proteins by culturing mammalian cells that have been artificially manipulated to produce them by expressing or strongly expressing the genes that encode them.

[0072] Expression vectors used to incorporate and express genes encoding fusion proteins, antibodies, or human lysosomal enzymes are not particularly limited as long as they express the gene when introduced into mammalian cells. The gene incorporated into the expression vector is located downstream of a DNA sequence (gene expression regulatory site) that can regulate the frequency of gene transcription in mammalian cells. Examples of gene expression regulatory sites that can be used in the present invention include cytomegalovirus-derived promoters, SV40 initial promoters, human elongation factor-1 alpha (EF-1α) promoters, and human ubiquitin C promoters.

[0073] Mammalian cells into which such expression vectors have been introduced begin to express the desired protein incorporated into the expression vector, but the expression level varies from cell to cell and is not uniform. Therefore, in order to efficiently produce recombinant proteins, it is necessary to select cells with high levels of expression of the desired protein from among the mammalian cells into which the expression vector has been introduced. To perform this selection step, the expression vector incorporates a gene that acts as a selection marker.

[0074] The most common selection markers are enzymes that break down drugs such as puromycin and neomycin (drug resistance markers). Mammalian cells die in the presence of these drugs above a certain concentration. However, mammalian cells into which an expression vector has been introduced can break down the drugs using the drug resistance marker incorporated into the expression vector, rendering them harmless or attenuating them, thus allowing them to survive in the presence of the drugs. By introducing an expression vector incorporating a drug resistance marker as a selection marker into mammalian cells and continuing to culture them in a selective medium containing the drug corresponding to that drug resistance marker while gradually increasing the concentration of the drug, cells that can proliferate even in the presence of higher drug concentrations can be obtained. In cells selected in this way, the expression level of the gene encoding the desired protein incorporated into the expression vector generally increases along with the expression level of the drug resistance marker, so as a result, cells with high levels of expression of the desired protein are selected.

[0075] Furthermore, glutamine synthase (GS) can be used as a selection marker. GS is an enzyme that synthesizes glutamine from glutamic acid and ammonia. When mammalian cells are cultured in a selective medium containing a glutamine synthase inhibitor, such as methionine sulfoximine (MSX), but without glutamine, the cells die. However, when an expression vector incorporating glutamine synthase as a selection marker is introduced into mammalian cells, the expression level of glutamine synthase increases in those cells, allowing them to proliferate even in the presence of higher concentrations of MSX. By gradually increasing the concentration of MSX while continuing the culture, cells capable of proliferating even in the presence of higher concentrations of MSX can be obtained. In the cells selected in this way, the expression level of the gene encoding the desired protein incorporated into the expression vector generally increases along with glutamine synthase, resulting in the selection of cells with high levels of expression of the desired protein.

[0076] Dihydrofolate reductase (DHFR) can also be used as a selection marker. When DHFR is used as a selection marker, mammalian cells into which the expression vector has been introduced are cultured in a selective medium containing a DHFR inhibitor such as methotrexate or aminopterin. By continuing the culture while gradually increasing the concentration of the DHFR inhibitor, cells that can proliferate even in the presence of higher concentrations of the DHFR inhibitor can be obtained. In the cells selected in this way, the expression level of the gene encoding the desired protein incorporated into the expression vector generally increases along with DHFR, so as a result, cells with high levels of expression of the desired protein are selected.

[0077] Expression vectors are known in which glutamine synthase (GS) is positioned downstream of the gene encoding the desired protein via an internal ribosome entry site (IRES) as a selection marker (International Patent Publications WO2012 / 063799, WO2013 / 161958). The expression vectors described in these publications can be particularly suitably used in the production method of the present invention.

[0078] For example, an expression vector for expressing a protein, comprising a gene expression regulatory site, a gene encoding the protein downstream thereof, an internal ribosome binding site further downstream, and a gene encoding glutamine synthase further downstream, and further comprising a dihydrofolate reductase gene or a drug resistance gene downstream of the gene expression regulatory site or another gene expression regulatory site, can be suitably used in the manufacturing method of the present invention. In this expression vector, a cytomegalovirus-derived promoter, an SV40 initial promoter, a human elongation factor-1 alpha promoter (hEF-1α promoter), and a human ubiquitin C promoter are suitably used as the gene expression regulatory site or another gene expression regulatory site, but the hEF-1α promoter is particularly suitable.

[0079] Furthermore, as the internal ribosome binding site, it is preferable to use one derived from the 5' untranslated region of a virus or gene selected from the group consisting of Picornaviridae viruses, foot-and-mouth disease virus, hepatitis A virus, hepatitis C virus, coronavirus, bovine enterovirus, Syler's mouse encephalomyelitis virus, coxsackievirus B, human immunoglobulin heavy chain binding protein gene, Drosophila Antennapedia gene, and Drosophila Ultravitrachus gene, but the internal ribosome binding site derived from the 5' untranslated region of mouse encephalomyomyocarditis virus is particularly preferable. When using an internal ribosome binding site derived from the 5' untranslated region of mouse encephalomyomyocarditis virus, in addition to the wild type, a wild-type internal ribosome binding site in which some of the multiple start codons contained in the internal ribosome binding site have been disrupted can also be suitably used. Furthermore, the drug resistance gene suitably used in this expression vector is preferably a puromycin or neomycin resistance gene, and more preferably a puromycin resistance gene.

[0080] Furthermore, for example, an expression vector for expressing a protein, comprising an hEF-1α promoter, a gene encoding the protein downstream thereof, an internal ribosome binding site derived from the 5' untranslated region of mouse encephalomyocarditis virus further downstream, and a gene encoding glutamine synthase further downstream, and further comprising another gene expression regulatory site and a dihydrofolate reductase gene downstream thereof, wherein the internal ribosome binding site has some of the multiple start codons contained in the wild-type internal ribosome binding site disrupted, can be suitably used in the production method of the present invention. An example of such an expression vector is the expression vector described in WO2013 / 161958.

[0081] Furthermore, for example, an expression vector for expressing a protein, comprising an hEF-1α promoter, a gene encoding the protein downstream thereof, an internal ribosome binding site derived from the 5' untranslated region of mouse encephalomyocarditis virus further downstream, and a gene encoding glutamine synthase further downstream, and further comprising another gene expression regulatory site and a drug resistance gene downstream thereof, wherein the internal ribosome binding site has some of the multiple start codons contained in the wild-type internal ribosome binding site disrupted, can be suitably used in the production method of the present invention. Examples of such expression vectors include pE-mIRES-GS-puro described in WO2012 / 063799 and pE-mIRES-GS-mNeo described in WO2013 / 161958.

[0082] In the present invention, mammalian cells into which an expression vector incorporating a gene encoding a fusion protein, antibody, or human lysosomal enzyme has been introduced are selectively cultured in a selective medium to select cells with high expression levels of the fusion protein, antibody, or human lysosomal enzyme.

[0083] In selective culture, when DHFR is used as a selection marker, the concentration of the DHFR inhibitor in the selective medium is increased in steps. The maximum concentration is preferably 0.25 to 5 μM, more preferably 0.5 to 1.5 μM, and even more preferably about 1.0 μM, when the DHFR inhibitor is methotrexate.

[0084] When using GS as a selection marker, the concentration of the GS inhibitor in the selection medium is increased in steps. The maximum concentration, when the GS inhibitor is MSX, is preferably 100 to 1000 μM, more preferably 200 to 500 μM, and even more preferably about 300 μM. In this case, a medium that does not contain glutamine is generally used as the selection medium.

[0085] When using an enzyme that degrades puromycin as a selection marker, the maximum concentration of puromycin in the selection medium is preferably 3 to 30 μg / mL, more preferably 5 to 20 μg / mL, and even more preferably about 10 μg / mL.

[0086] When using an enzyme that degrades neomycin as a selection marker, the maximum concentration of G418 in the selection medium is preferably 0.1 mg to 2 mg / mL, more preferably 0.5 to 1.5 mg / mL, and even more preferably about 1 mg / mL.

[0087] Furthermore, as a culture medium for mammalian cells, both the medium used in selective culture and the medium used to produce fusion proteins, antibodies, or human lysosomal enzymes (recombinant protein production medium) described later can be used without particular limitations as long as they are capable of culturing and growing mammalian cells, but serum-free medium is preferably used.

[0088] Cells with high expression levels of fusion proteins, antibodies, or human lysosomal enzymes, selected through selective culture, are used as these producing cells for their production. The production of fusion proteins, antibodies, or human lysosomal enzymes is carried out by culturing these producing cells in a recombinant protein production medium. This culture is called production culture.

[0089] In the present invention, a serum-free medium used as a culture medium for recombinant protein production is preferably one containing, for example, 3 to 700 mg / L of amino acids, 0.001 to 50 mg / L of vitamins, 0.3 to 10 g / L of monosaccharides, 0.1 to 10000 mg / L of inorganic salts, 0.001 to 0.1 mg / L of trace elements, 0.1 to 50 mg / L of nucleosides, 0.001 to 10 mg / L of fatty acids, 0.01 to 1 mg / L of biotin, 0.1 to 20 μg / L of hydrocortisone, 0.1 to 20 mg / L of insulin, 0.1 to 10 mg / L of vitamin B12, 0.01 to 1 mg / L of putrescine, 10 to 500 mg / L of sodium pyruvate, and a water-soluble iron compound. If desired, thymidine, hypoxanthine, conventional pH indicators, and antibiotics may be added to the culture medium.

[0090] Furthermore, DMEM / F12 medium (a mixed medium of DMEM and F12) may be used as a serum-free medium for recombinant protein production, and these media are well known to those skilled in the art. In addition, DMEM(HG)HAM modified (R5) medium, which contains sodium bicarbonate, L-glutamine, D-glucose, insulin, sodium selenite, diaminobutane, hydrocortisone, ferrous sulfate, asparagine, aspartic acid, serine, and polyvinyl alcohol, may be used as a serum-free medium. Furthermore, commercially available serum-free media, such as CD OptiCHO, may also be used. TM Culture medium, CHO-S-SFM II medium or CD CHO medium (Thermo Fisher Scientific), EX-CELL TM Medium 302 or EX-CELL TM325-PF medium (SAFC Biosciences) can also be used as a basic medium. For example, EX-CELL, a serum-free medium containing 16 μmol / L thymidine, 100 μmol / L hypoxanthine, and 4 mmol / L L-alanyl-L-glutamine. TM Advanced CHO Fed-batch medium (SAFC Biosciences) is suitable for culturing fusion protein-producing cells. Also suitable is, for example, CD OptiCHO, a serum-free medium containing 16 μmol / L thymidine, 100 μmol / L hypoxanthine, and 10 mg / L insulin. TM The culture medium (Thermo Fisher Scientific) can also be suitably used for culturing fusion protein-producing cells.

[0091] For the production culture of cells that produce fusion proteins, antibodies, or human lysosomal enzymes, the density of these producing cells in the culture medium for recombinant protein production is preferably 0.2 x 10 5 ~5X10 5 pieces / mL, comfortable1X10 5 ~4x10 5 pieces / mL, more preferably about 3 x 10 5 The start is adjusted to cells / mL.

[0092] Production culture is carried out while observing the cell viability (%) over time, so that the cell viability during the production culture period is preferably maintained at 80% or higher, more preferably at 85% or higher.

[0093] Furthermore, during the production culture period, the culture temperature is preferably maintained at 33.5 to 37.5°C, and the dissolved oxygen saturation in the production medium is preferably maintained at 28 to 32%, more preferably at about 30%. Here, dissolved oxygen saturation refers to the amount of oxygen dissolved under the same conditions, with the saturated oxygen solubility being defined as 100%.

[0094] Furthermore, during the production culture period, the production medium is stirred with an impeller. The rotation speed of the impeller at this time is preferably 50 to 100 revolutions per minute, more preferably 60 to 110 revolutions per minute, for example, 90 revolutions per minute, 100 revolutions per minute, 110 revolutions per minute, etc., but the rotation speed is appropriately changed depending on the shape of the impeller, etc.

[0095] As initial conditions for suitable culture conditions in production culture, for example, the density of the producing cells in the recombinant protein production medium at the start of production culture should be 3 x 10⁻¹⁴. 5 Examples include a culture medium with a density of cells / mL, a culture temperature of 34-37°C during the production culture period, a dissolved oxygen saturation of 30% in the production medium, and agitation of the medium by an impeller rotating at a speed of approximately 100 revolutions per minute.

[0096] After the production culture is complete, the culture medium is collected, and the culture supernatant can be obtained by removing cells and other contaminants from the collected medium by centrifugation or membrane filtration. The target fusion protein contained in this culture supernatant is purified by various chromatography processes. The purification process can be carried out at room temperature or in a low-temperature environment, but is preferably carried out in a low-temperature environment, and is particularly preferably carried out at a temperature of 1 to 10°C.

[0097] The following describes in detail one embodiment of a method for purifying a fusion protein of an antibody and a human lysosomal enzyme contained in the culture supernatant. This embodiment is particularly suitable for use when the antibody is a human IgG antibody of type Fab and the human lysosomal enzyme is hIDUA. For example, it is suitable as a method for purifying a fusion protein in which the light chain of a humanized anti-hTfR antibody has the amino acid sequence shown in SEQ ID NO: 22, and the Fab heavy chain of the humanized anti-hTfR antibody is bound at its C-terminus to human α-L-idulonidase shown in SEQ ID NO: 6 via a linker having the amino acid sequence shown in SEQ ID NO: 4, and the fusion protein as a whole has the amino acid sequence shown in SEQ ID NO: 27.

[0098] The first step of the purification process is column chromatography using a stationary phase that has a substance attached to the antibody. There are no particular limitations on the substance attached to the antibody used, but it is preferably a substance that has an affinity for Fab, and more preferably a substance that has an affinity for the CH1 region of the heavy chain of Fab. Antibodies against this region can be used as the substance attached to the antibody. If Fab is a human IgG antibody, an anti-human IgG-CH1 antibody is preferably used. If the substance attached to the antibody is an antibody, the stationary phase is a support to which the antibody is attached. There are no particular limitations on the support, but agarose is an example. By loading the culture supernatant, the fusion protein contained in the culture supernatant is bound to the column, and after washing the column, the fusion protein is eluted from the column. In this way, many of the contaminants can be removed.

[0099] Since the culture supernatant has a large volume, it is preferable to concentrate it before using it in the purification process, but the concentration operation is not essential. In the first step of the purification process, arginine is added to the culture supernatant or its concentrate before loading it onto the column. At this time, the arginine is added so that the concentration in the solution is preferably 0.1M to 1.0M, more preferably 0.25M to 1.0M, even more preferably 0.3M to 1.0M, even more preferably 0.3M to 0.5M, for example 0.4M. The stationary phase is equilibrated in advance with a buffer containing arginine before loading the culture supernatant. The arginine concentration of the buffer used at this time is preferably 0.1M to 1.0M, more preferably 0.2M to 0.8M, even more preferably 0.2M to 0.6M, even more preferably 0.3M to 0.5M, for example 0.4M. Furthermore, there are no particular limitations on the buffer solution used at this time, but it is preferably an MES buffer solution, and its pH is preferably 6.0 to 7.0, and more preferably about 6.5.

[0100] After washing the stationary phase to which the fusion protein is bound, the fusion protein is eluted with a salt-free acidic buffer to recover the fraction containing the fusion protein. The buffer used at this time is preferably a glycine buffer or an acetate buffer, and its pH is preferably 3.2 to 3.8, more preferably 3.5. The pH of the recovered solution containing the fusion protein is quickly adjusted to near neutral.

[0101] The second step of the purification process is anion exchange column chromatography. There are no particular limitations on the anion exchanger used at this time, but a strong anion exchanger is preferred, and in particular anion exchanger having selectivity based on electrostatic interaction, hydrophobic interaction and hydrogen bond formation. Multimodal anion exchangers can be suitably used in the present invention, and for example, a strong anion exchanger having an N-benzyl-N-methylethanolamine group as a functional group can be particularly suitably used. Capto adhere (GE Healthcare) is one such multimodal strong anion exchanger having an N-benzyl-N-methylethanolamine group.

[0102] Before being subjected to anion exchange column chromatography, the solution recovered in the first step of the purification process is given a buffer solution to adjust its pH to near neutral. There are no particular limitations on the buffer solution used at this time, but MES buffer is one example. The pH is preferably adjusted to 5.5 to 6.5, more preferably 5.7 to 6.3, for example to 6.0. The stationary phase is also equilibrated in advance with a buffer solution. There are no particular limitations on the buffer solution used at this time, but MES buffer is preferred, and its pH is preferably 5.5 to 6.5, for example to 6.0.

[0103] The column chromatography-recovered fraction from the first step of the purification process, whose pH has been adjusted as described above, is loaded onto the anion exchange column that has been pre-equilibriumized as described above. After recovering the non-adsorbed fraction containing the fusion protein, the column is washed with a buffer solution. The resulting wash solution is also recovered as the non-adsorbed fraction. The buffer solution used for washing the column is preferably MES buffer, and its pH is preferably 5.5 to 6.5, for example, 6.0.

[0104] The third step in the purification process is cation exchange chromatography. There are no particular limitations on which cation exchange resin to use in cation exchange chromatography, but weak cation exchangers are preferred, and more preferably, weak cation exchangers that exhibit selectivity based on both hydrophobic interactions and hydrogen bond formation. For example, weak cation exchangers such as Capto MMC (GE Healthcare), which possess phenyl groups, amide bonds, and carboxyl groups and exhibit selectivity based on both hydrophobic interactions and hydrogen bond formation, can be used.

[0105] In cation exchange chromatography, the column is pre-equilibriumized with a buffer. The buffer used is preferably MES buffer, and its pH is preferably 6.0 to 7.0, and more preferably about 6.5.

[0106] The fusion protein-containing fraction and washing solution recovered in the second step of the purification process are loaded onto a cation exchange column. After washing the column to which the fusion protein has been bound, the fusion protein is eluted. Elution can be performed using a buffer solution with a high salt concentration. The buffer solution used at this time is preferably MES buffer solution, and its pH is preferably 6.0 to 7.0, more preferably about 6.5. The salt added to the buffer solution is preferably sodium chloride, and the salt concentration is preferably in the range of 0.5 to 1.5 M, more preferably in the range of 0.8 to 1.2 M, and even more preferably about 1 M.

[0107] The pH of the fusion protein-containing fraction of the eluate obtained in the third step is preferably adjusted to 5.5 to 6.1, more preferably to about 5.8.

[0108] The fourth step in the purification process is size exclusion column chromatography. This step removes small molecular weight impurities such as endotoxins, fusion protein polymers, and degradation products, thereby obtaining a substantially pure fusion protein.

[0109] In addition, one or more further column chromatography methods may be introduced to the above-described purification method, which includes, in this order, column chromatography using a material conjugated with an antibody-affinity substance as the stationary phase, anion exchange column chromatography, cation exchange column chromatography, and size exclusion column chromatography. Such additional column chromatography methods include, for example, column chromatography using a solid phase having affinity for phosphate groups, column chromatography using a material conjugated with an antibody-affinity substance as the stationary phase, hydrophobic column chromatography, anion exchange column chromatography, cation exchange column chromatography, and dye affinity column chromatography. Preferably, dye affinity column chromatography using a blue triazine dye as the ligand is introduced before the first step of the purification process, or between the first and second steps of the purification process.

[0110] In the purification process of the fusion protein, a step to inactivate viruses that may be introduced from the culture supernatant can also be added. This virus inactivation step may be performed before the first step of the purification process, between each step of the purification process, or after the completion of the purification process, but for example, it may be performed before the first step of the purification process or between the first and second steps of the purification process.

[0111] The virus inactivation process is carried out by adding a nonionic surfactant to a solution containing the fusion protein and stirring at 20-60°C for 2-6 hours. Suitable nonionic surfactants used in this process include polysorbate 20, 80, and tri-n-butyl phosphate, or mixtures thereof.

[0112] The virus inactivation process can also be carried out using a virus removal membrane. By passing a solution containing humanized anti-hTfR antibody-hIDUA through a virus removal membrane with pore sizes of 35 nm and 20 nm, viruses contained in the solution can be removed.

[0113] The purified fusion protein obtained using the manufacturing method of the present invention is of a purity suitable for direct use as a pharmaceutical. In particular, when the fusion protein has a humanized anti-hTfR antibody light chain having the amino acid sequence shown in SEQ ID NO: 22, and the Fab heavy chain of the humanized anti-hTfR antibody is bound at its C-terminus to human α-L-iduronidase shown in SEQ ID NO: 6 via a linker having the amino acid sequence shown in SEQ ID NO: 4, and the fusion protein as a whole has the amino acid sequence shown in SEQ ID NO: 27 (humanized anti-hTfR antibody-hIDUA), a purified product of a purity suitable for direct use as a pharmaceutical is obtained.

[0114] The concentration of host cell-derived protein (HCP) in the purified product of the fusion protein obtained using the manufacturing method of the present invention, particularly the humanized anti-hTfR antibody-hIDUA, is less than 100 ppm, for example, less than 60 ppm, less than 40 ppm, for example, 1-50 ppm, 1-40 ppm.

[0115] The proportion of polymer in the total fusion protein, particularly in the purified product of the humanized anti-hTfR antibody-hIDUA obtained using the manufacturing method of the present invention, is less than 0.5%, for example, less than 0.2%, less than 0.1%, less than 0.05%, for example, 0.01-0.1% and 0.001-0.1%.

[0116] The fusion protein obtained using the manufacturing method of the present invention, particularly the purified humanized anti-hTfR antibody-hIDUA, can be supplied as an aqueous solution containing a suitable excipient or as a lyophilized preparation when supplied as a pharmaceutical product. When supplied as an aqueous solution, it may be in the form of a vial or as a pre-filled preparation in a syringe. In the case of a lyophilized preparation, it is dissolved in an aqueous medium before use. [Examples]

[0117] The present invention will be described in more detail below with reference to examples, but the present invention is not intended to be limited to these examples.

[0118] [Example 1] Construction of a vector for humanized anti-hTfR antibody-hIDUA fusion protein expression The vector for expressing the humanized anti-hTfR antibody-hIDUA fusion protein was constructed using genes encoding a light chain having the amino acid sequence shown in SEQ ID NO: 22 as the antibody portion, and a heavy chain Fab region having the amino acid sequence shown in SEQ ID NO: 23.

[0119] [Construction of pE-neo vectors and pE-hygr vectors] The pEF / myc / nuc vector (Invitrogen) was digested with KpnI and NcoI, and the region containing the EF-1 promoter and its first intron was excised and blunt-ended with T4 DNA polymerase. Separately, the pCI-neo (Invitrogen) was digested with BglII and EcoRI, and the region containing the CMV enhancer / promoter and intron was excised, followed by blunt-ending with T4 DNA polymerase. The region containing the EF-1α promoter and its first intron (after blunt-ending) was inserted into this to construct the pE-neo vector. The pE-neo vector was digested with SfiI and BstXI, and a region of approximately 1 kbp containing the neomycin resistance gene was excised. Using pcDNA3.1 / Hygro(+) (Invitrogen) as a template, the hygromycin gene was amplified by PCR using primers Hyg-Sfi5' (SEQ ID NO: 13) and Hyg-BstX3' (SEQ ID NO: 14). The amplified hygromycin gene was digested with SfiI and BstXI and inserted into the above-mentioned pE-neo vector to construct the pE-hygr vector. The construction of the pE-neo vector and pE-hygr vector was carried out with reference to patent document (Japanese Patent No. 6279466).

[0120] [Construction of pE-IRES-GS-puro] The expression vector pPGKIH (Miyahara M. et.al., J. Biol. Chem. 275,613-618(2000)) was digested with restriction enzymes (XhoI and BamHI), and the internal ribosome binding site (IRES) and hygromycin resistance gene (Hyg) derived from mouse encephalomyocarditis virus (EMCV) were extracted. r A DNA fragment containing the gene and the polyadenylated region (mPGKpA) of mouse phosphoglycerate kinase (mPGK) was excised. This DNA fragment was inserted between the XhoI and BamHI sites of pBluescript SK(-) (Stratagene), and this was named pBSK(IRES-Hygr-mPGKpA).

[0121] Using pBSK (IRES-Hygr-mPGKpA) as a template, a DNA fragment containing a portion of the EMCV IRES was amplified by PCR using primers IRES5' (SEQ ID NO: 29) and IRES3' (SEQ ID NO: 30). This DNA fragment was digested with restriction enzymes (XhoI and HindIII) and inserted between the XhoI and HindIII sites of pBSK (IRES-Hygr-mPGKpA) to obtain pBSK (NotI-IRES-Hygr-mPGKpA). pBSK (NotI-IRES-Hygr-mPGKpA) was digested with restriction enzymes (NotI and BamHI) and inserted between the NotI and BamHI sites of the pE-hygr vector to obtain plasmid pE-IRES-Hygr.

[0122] The expression vector pPGKIH was digested with EcoRI, and a DNA fragment containing the mPGK promoter region (mPGKp) was excised. This DNA fragment was inserted into the EcoRI site of pBluescript SK(-) (Stratagene), and this was designated as mPGK promoter / pBS(-). Using mPGK promoter / pBS(-) as a template, the DNA fragment containing the mPGK promoter region (mPGKp) was amplified by PCR using primers mPGKP5' (SEQ ID NO: 31) and mPGKP3' (SEQ ID NO: 32). This DNA fragment was digested with restriction enzymes (BglII and EcoRI) and inserted between the BglII and EcoRI sites of pCI-neo (Promega), and this was designated as pPGK-neo. pE-IRES-Hygr was digested with restriction enzymes (NotI and BamHI) to excavate DNA fragments (IRES-Hygr), which were then inserted between the NotI and BamHI sites of pPGK-neo to form pPGK-IRES-Hygr.

[0123] cDNA was prepared from CHO-K1 cells and used as a template. Using primers GS5' (SEQ ID NO: 33) and GS3' (SEQ ID NO: 34), a DNA fragment containing the GS gene was amplified by PCR. This DNA fragment was digested with restriction enzymes (BalI and BamHI) and inserted between the BalI and BamHI sites of pPGK-IRES-Hygr to form pPGK-IRES-GS-ΔpolyA.

[0124] Using pCAGIPuro (Miyahara M. et.al., J. Biol. Chem. 275,613-618(2000)) as a template, and primers puro5' (SEQ ID NO: 35) and puro3' (SEQ ID NO: 36), the puromycin resistance gene (puro r A DNA fragment containing a gene was amplified. This DNA fragment was inserted into a pT7Blue T-Vector (Novagen), and this was designated as pT7-puro. pT7-puro was digested with restriction enzymes (AflII and BstXI) and inserted between the AflII and BstXI sites of the expression vector pE-neo to create pE-puro.

[0125] Using pE-puro as a template, a DNA fragment containing the late polyadenylation region of SV40 was amplified by PCR using primers SV40polyA5' (SEQ ID NO: 37) and SV40polyA3' (SEQ ID NO: 38). This DNA fragment was digested with restriction enzymes (NotI and HpaI) and inserted between the NotI and HpaI sites of the expression vector pE-puro to form pE-puro(XhoI). pPGK-IRES-GS-ΔpolyA was digested with restriction enzymes (NotI and XhoI) to excise a DNA fragment containing the IRES-GS region, and this was inserted between the NotI and XhoI sites of the expression vector pE-puro(XhoI) to form pE-IRES-GS-puro. The construction of pE-IRES-GS-puro was performed with reference to patent document (Japanese Patent No. 6279466).

[0126] [Construction of pE-mIRES-GS-puro(ΔE)] Using the expression vector pE-IRES-GS-puro as a template, the region from IRES to GS of EMCV was amplified by PCR using primers mIRES-GS5' (SEQ ID NO: 39) and mIRES-GS3' (SEQ ID NO: 40). A DNA fragment was then amplified by disrupting the EMCV by adding a mutation to the start codon (ATG) located second from the 5' end of IRES. Using the expression vector pE-IRES-GS-puro as a template, a DNA fragment containing the above region from IRES to GS was amplified by PCR using this DNA fragment and the aforementioned primer IRES5'. This DNA fragment was digested with restriction enzymes (NotI and PstI), and the excised DNA fragment was inserted between the NotI and PstI sites of pBluescript SK(-) (Stratagene), resulting in mIRES / pBlueScript SK(-).

[0127] The expression vector pE-IRES-GS-puro was digested with SphI, and the SV40 enhancer region was excised. The remaining DNA fragment was self-ligated, and this was obtained as pE-IRES-GS-puro(ΔE). mIRES / pBlueScript SK(-) was digested with NotI and PstI, and a region containing the modified IRES (mIRES) and a portion of the GS gene was excised. Separately, pE-IRES-GS-puro(ΔE) was digested with NotI and PstI, and the above-mentioned region containing a portion of the mIRES and GS gene was inserted to construct pE-mIRES-GS-puro(ΔE).

[0128] [Construction of pEM-hygr(LC3) and pE-mIRES-GSp-Fab-IDUA] A DNA fragment (CMVE-EF-1αp-IFNβMAR) containing β-Globin MAR (Matrix Attachment Region), CMV enhancer, human EF-1α promoter, MluI and BamHI cleavage sites, and interferon β Mar was artificially synthesized (SEQ ID NO: 41). The HindIII sequence was introduced into the 5' end of this DNA fragment, and the EcoRI sequence into the 3' end. This DNA fragment was digested with HindIII and EcoRI, and inserted between the HindIII and EcoRI sites in the pUC57 vector to form JCR69 in pUC57. A DNA fragment (IRES-HygroR-mPGKpA) containing MluI and BamHI cleavage sites, IRES, hygromycin resistance gene, and mPGK polyadenylation signal was artificially synthesized (SEQ ID NO: 42). This DNA fragment was inserted into the MluI and BamHI sites of JCR69 in pUC57 to form pEM hygro.

[0129] A DNA fragment (SEQ ID NO: 26) containing the gene encoding the full length of the light chain of a humanized anti-hTfR antibody having the amino acid sequence shown in SEQ ID NO: 22 was artificially synthesized and inserted into pUC57-Amp, resulting in JCR131 in pUC57-Amp. A MluI sequence was introduced into the 5' end of this DNA fragment, and a NotI sequence into the 3' end. This plasmid DNA was digested with MluI and NotI and incorporated between the MluI and NotI sequences in the expression vector pEM hygro. The resulting vector was designated as pEM-hygr(LC3), an expression vector for the light chain of the humanized anti-hTfR antibody.

[0130] A DNA fragment containing the base sequence shown in SEQ ID NO: 28 was artificially synthesized. This fragment consisted of a human IDUA having the amino acid sequence shown in SEQ ID NO: 6, linked to the C-terminal side of the Fab region of the heavy chain of a humanized anti-hTfR antibody having the amino acid sequence shown in SEQ ID NO: 27, via a linker sequence shown in SEQ ID NO: 4. The entire DNA fragment then encoded a protein with the amino acid sequence shown in SEQ ID NO: 27. A MluI sequence was introduced to the 5' end of this DNA fragment, and a NotI sequence to the 3' end. This DNA fragment was digested with MluI and NotI and incorporated between MluI and NotI in pE-mIRES-GS-puro(ΔE). The resulting vector was designated as pE-mIRES-GSp-Fab-IDUA, an expression vector for a protein in which hIDUA is linked to the C-terminal side of the Fab heavy chain of a humanized anti-hTfR antibody.

[0131] [Example 2] Preparation of a cell line that highly expresses a humanized anti-hTfR antibody-hIDUA fusion protein CHO cells (CHO-K1: obtained from the American Type Culture Collection) were transformed with pEM-hygr(LC3) and pE-mIRES-GSp-Fab-IDUA constructed in Example 1 using NEPA21 (NEPAGENE) by the following method.

[0132] Cell transformation was generally carried out using the following method: 2 × 10⁶ CHO-K1 cells 7 CD OptiCHO TM The cells were suspended in a 1:1 mixture of culture medium (Thermo Fisher Scientific) and PBS. A 50 μL cell suspension was taken and added to it, and CD OptiCHO TM 50 μL of pEM-hygr(LC3) plasmid DNA solution, diluted to 200 μg / mL, was added to a 1:1 mixture of culture medium and PBS. Electroporation was performed using NEPA21 (NEPAGENE) to introduce pEM-hygr(LC3) plasmid DNA into the cells. After culturing overnight at 37°C and 5% CO2, the cells were treated with CD OptiCHO with 0.5 mg / mL of hygromycin. TMThe cells were selectively cultured in culture medium. The resulting cells were then introduced with pE-mIRES-GSp-Fab-IDUA plasmid DNA (digested with AhdI and linearized) using the same method. After culturing overnight at 37°C and 5% CO2, the cells were treated with CD OptiCHO by adding 0.5 mg / mL hygromycin and 10 μg / mL puromycin. TM Cells were selectively cultured in culture medium. During selective culture, the concentration of MSX was gradually increased until the final MSX concentration was 300 μM, which selectively proliferated cells exhibiting drug resistance.

[0133] Next, using the limiting dilution method, the cells selected in selective culture were seeded onto a 96-well plate so that no more than one cell per well proliferated, and the cells were cultured for approximately two weeks until each cell formed a monoclonal colony. The culture supernatant was collected from the wells in which monoclonal colonies had formed, and the humanized antibody content was examined by ELISA to select cell lines that highly expressed humanized anti-hTfR antibody-hIDUA fusion protein.

[0134] The ELISA method used in this study was generally as follows: 100 μL of chicken anti-IDUA polyclonal antibody solution, diluted to 5 μg / mL in 0.05 M bicarbonate buffer, was added to each well of a 96-well microtiter plate (Nunc). The plates were allowed to stand at room temperature or 4°C for at least 1 hour to allow the antibody to adsorb to the plate. Next, each well was washed three times with Tris-buffered saline (pH 8.0) with 0.05% Tween20 added (TBS-T). Finally, 300 μL of Tris-buffered saline (pH 8.0) with 1% BSA added was added to each well, and the plates were allowed to stand at room temperature for 1 hour. Next, each well was washed three times with TBS-T, and then 100 μL of culture supernatant or purified humanized anti-hTfR antibody-hIDUA fusion protein, diluted to an appropriate concentration with Tris-buffered saline (pH 8.0) with 0.1% BSA and 0.05% Tween20 (TBS-BT), was added to each well, and the plate was left to stand at room temperature for 1 hour. Next, the plate was washed three times with TBS-T, and then 50 μL of HRP-labeled anti-human IgG polyclonal antibody solution diluted with TBS-BT was added to each well, and the plate was left to stand at room temperature for 1 hour. After washing each well three times with TBS-T, color development was performed using the ELISA POD substrate TMB kit (Nakalai Tesque). Next, 50 μL of 1 mol / L sulfuric acid was added to each well to stop the reaction, and the absorbance at 450 nm was measured for each well using a 96-well plate reader. Cells corresponding to wells showing high measurement values ​​were selected as cell lines that highly express the humanized anti-hTfR antibody-hIDUA fusion protein. These cell lines that thus obtained highly express the humanized anti-hTfR antibody-hIDUA fusion protein were designated as humanized anti-hTfR antibody-hIDUA expression cells. The fusion protein of the humanized anti-hTfR antibody and hIDUA expressed by this cell line was designated as the humanized anti-hTfR antibody-hIDUA fusion protein (humanized anti-hTfR antibody-hIDUA).

[0135] The resulting humanized anti-hTfR antibody-hIDUA expression strain was mixed with CD OptiCHO containing 10 mg / L insulin, 16 μmol / L thymidine, 100 μmol / L hypoxanthine, 500 μg / mL hygromycin B, 10 μg / mL puromycin, 300 μmol / L MSX, and 10% (v / v) DMSO. TM The cells were suspended in culture medium, then dispensed into cryotubes and stored in liquid nitrogen as seed cells.

[0136] [Example 3] Culture of humanized anti-hTfR antibody-hIDUA expressing strain To obtain humanized anti-hTfR antibody-hIDUA, humanized anti-hTfR antibody-hIDUA-expressing strains were cultured using the following method. The humanized anti-hTfR antibody-hIDUA-expressing strains obtained in Example 2 were cultured to a cell density of approximately 3 x 10⁶ 5 Approximately 170 L of serum-free medium (CD OptiCHO), adjusted to pH 6.9, containing 10 mg / L insulin, 16 μmol / L thymidine, and 100 μmol / L hypoxanthine, to achieve a concentration of 10 cells / mL. TM The cells were suspended in culture medium (ThermoFisher SCIENTIFIC). 170 L of this cell suspension was transferred to a culture vessel. The medium was stirred with an impeller at approximately 100 rpm to maintain a dissolved oxygen saturation of approximately 30%, and the cells were cultured at a temperature range of 34-37°C for approximately 10 days. During the culture period, cell density, cell viability, glucose concentration in the medium, and lactate concentration were monitored. If the glucose concentration in the medium fell below 3.0 g / L, glucose solution was immediately added to the medium to bring the glucose concentration to 3.5 g / L. During the culture period, feed solution (EFFICIENTFEED A+) was used. TM ThermoFisher SCIENTIFIC added appropriate amounts to the culture medium. After the culture was complete, the medium was collected. The collected medium was filtered through Millistak+HC Pod Filter grade D0HC (Merck) and then further filtered through Millistak+ HC grade X0HC (Merck) to obtain a culture supernatant containing humanized anti-hTfR antibody-hIDUA. This culture supernatant was then processed using Pellicon TM3 Cassette w / Ultracel PLCTK Membrane (pore size: 30 kDa, membrane area: 1.14m 2 The solution was ultrafiltered using a Merck filter and concentrated until the volume was reduced to approximately 1 / 14th. Next, this concentrate was filtered using an Opticap XL600 (0.22 μm, Merck). The resulting solution was used as the concentrated culture supernatant.

[0137] [Example 4] Purification of humanized anti-hTfR antibody-hIDUA To the concentrated culture supernatant obtained in Example 3, 0.25 times the volume of 2 M arginine solution (pH 7.0) was added. This solution was equilibrated with 25 mM MES buffer (pH 6.5) containing 400 mM arginine in a volume four times the column volume. Capture Select TM Humanized anti-hTfR antibody-hIDUA was adsorbed onto a CH1-XL column (column volume: approx. 3.1 L, bed height: approx. 20 cm, Thermo Fisher Scientific) at a constant flow rate of 100 cm / hour. TM The CH1-XL column is an affinity column in which a ligand that specifically binds to the CH1 domain of IgG antibodies is immobilized on a support.

[0138] Next, the column was washed by supplying five times the volume of the column buffer at the same flow rate. Then, the column was further washed by supplying three times the volume of the column buffer (25 mM MES buffer, pH 6.5) at the same flow rate. Next, the humanized anti-hTfR antibody-hIDUA adsorbed on the column was eluted with five times the volume of the column buffer (10 mM sodium acetate-HCl buffer, pH 3.5). The eluate was received in a container pre-filled with 250 mM MES buffer (pH 6.0) and immediately neutralized.

[0139] The eluate from the affinity column described above was diluted with 250 mM MES buffer (pH 6.5) to adjust the pH of the eluate to 6.0. This eluate was then filtered using an Opticap XL600 filter (pore size: 0.22 μm, Merck). The filtered solution was then loaded onto a Capto adhere column (column volume: approximately 1.5 L, bed height: approximately 10 cm, GE Healthcare), a multimodal anion exchange column equilibrated with 50 mM MES buffer (pH 6.0) containing 15 mM NaCl in a volume five times the column volume, at a constant flow rate of 300 cm / hour. The loaded solution containing the humanized anti-hTfR antibody-hIDUA was collected. Capto adhere is a strong anion exchanger with N-benzyl-N-methylethanolamine as its ligand, exhibiting selectivity based on electrostatic interactions, hydrogen bonding, hydrophobic interactions, etc.

[0140] Next, the column was washed by supplying five times the volume of the column buffer at the same flow rate, and this washing solution was collected.

[0141] The above loading solution and washing solution were equilibrated with 25 mM MES buffer (pH 6.5) containing 300 mM NaCl in a volume four times the column volume. A Capto MMC column (column volume: approximately 3.1 L, bed height: approximately 20 cm, GE Healthcare), a multimodal weak cation exchange column, was then loaded at a constant flow rate of 200 cm / hour. Capto MMC is a weak cation exchanger that exhibits selectivity based on hydrophobic interactions, hydrogen bond formation, etc.

[0142] Next, the column was washed by supplying five times the volume of the buffer solution at the same flow rate. Then, the humanized anti-hTfR antibody-hIDUA adsorbed on the weak cation exchange column was eluted with 25 mM MES buffer (pH 6.5) containing 1 M NaCl in ten times the volume of the column.

[0143] The eluate from the weak cation exchange column described above was adjusted to pH 5.8 by adding 20 mM citrate buffer (pH 5.5) containing 0.5 times the volume of 0.8 mg / mL NaCl and 75 mg / mL sucrose. Then, Pellicon TM 3 Cassette w / Ultracel PLCTK Membrane (pore size: 30 kDa, membrane area: 0.57 m 2 The solution was ultrafiltered using a Merck filter and concentrated until the concentration of humanized anti-hTfR antibody-hIDUA in the solution was approximately 30 mg / mL. This concentrate was then filtered using an Opticap XL600 (0.22 μm, Merck).

[0144] The above concentrate was loaded onto a BioSEC column (column volume: approximately 9.4 L, bed height: 30 cm, Merck), a size exclusion column, at a constant flow rate of 40 cm / hour, equilibrated with 20 mM citrate buffer (pH 5.5) containing 0.8 mg / mL NaCl and 75 mg / mL sucrose in a volume 1.5 times the column volume. The same buffer was then supplied at the same flow rate. At this time, a spectrophotometer was placed in the flow path of the eluate from the size exclusion column to continuously measure the absorbance of the eluate, and the absorbance at 280 nm was monitored. The fraction showing an absorption peak at 280 nm was collected as the fraction containing humanized anti-hTfR antibody-hIDUA, and this was used as the purified humanized anti-hTfR antibody-hIDUA product.

[0145] [Example 5] Measurement of the recovery rate of humanized anti-hTfR antibody-hIDUA in each purification step The amount of humanized anti-hTfR antibody-hIDUA loaded in the affinity column purification step and the amount recovered in the eluate were measured using the ELISA method described in Example 2. Furthermore, the amount of humanized anti-hTfR antibody-hIDUA loaded and recovered in the eluate in each of the other purification steps were calculated by measuring the absorbance at 280 nm using a spectrophotometer. The results are shown in Table 3. Initially, 17.5 g of humanized anti-hTfR antibody-hIDUA, equivalent to approximately 74.0% of the 23.7 g of humanized anti-hTfR antibody-hIDUA contained in the culture supernatant, was recovered as the purified product. These results demonstrate that the purification method described in Example 4 is a highly efficient method for purifying humanized anti-hTfR antibody-hIDUA. In Table 3, the process recovery rate (%) refers to the ratio of the amount of humanized anti-hTfR antibody-hIDUA recovered in each purification step to the amount of humanized anti-hTfR antibody-hIDUA loaded in each purification step, and the total recovery rate (%) refers to the ratio of the amount of humanized anti-hTfR antibody-hIDUA recovered in each purification step to the initial amount of humanized anti-hTfR antibody-hIDUA subjected to the purification process.

[0146] [Table 3]

[0147] [Example 6] Analysis of purified humanized anti-hTfR antibody-hIDUA product (quantification of HCP) The amount of host cell-derived protein (HCP) contained in the purified humanized anti-hTfR antibody-hIDUA product obtained in Example 4 was quantified by ELISA. First, 100 μL of anti-CHO cell-derived protein antibody was added to each well of a 96-well plate (Nunc), and the plate was allowed to stand overnight to adsorb the antibody. After washing each well three times, 200 μL of blocking solution containing casein was added to each well, and the mixture was shaken at 25°C for 60 minutes. After washing each well three times, 100 μL of either a solution containing the purified humanized anti-hTfR antibody-hIDUA product (sample solution) or an HCP standard solution was added to each well, and the mixture was shaken at 25°C for 2 hours. After washing each well three times, 100 μL of biotinylated anti-CHO cell-derived protein antibody was added to each well, and the mixture was shaken at 25°C for 60 minutes. After washing each well three times, 100 μL of HRP-conjugated streptavidin (Jackson Immuno Research Laboratories) was added and shaken at 25°C for 60 minutes. After washing each well three times, 100 μL of TMB substrate solution was added to each well and allowed to develop color at 25°C. The TMB substrate solution used was an equal mixture of TMB peroxidase substrate and Peroxidase substrate solution B from the TMB microwell peroxidase substrate system (KPL). After color development, 100 μL of 6.75% phosphate was added to each well to stop the enzymatic reaction, and the absorbance at 450 nm of each well was measured using a 96-well plate reader. A calibration curve was created from the measured values ​​of the HCP standard solution, and the values ​​of the sample solution were interpolated to quantify the HCP contained in the purified humanized anti-hTfR antibody-hIDUA product. The amount of HCP contained in the purified humanized anti-hTfR antibody-hIDUA product was quantified using the quantitative values ​​of HCP obtained in this way and the quantitative values ​​of the purified humanized anti-hTfR antibody-hIDUA product measured using the ELISA method described in Example 2. As a result, it was found that the amount of HCP contained in the purified humanized anti-hTfR antibody-hIDUA product was approximately 35 ppm (i.e., approximately 35 ng of HCP in 1 mg of purified humanized anti-hTfR antibody-hIDUA product).

[0148] [Example 7] Analysis of purified humanized anti-hTfR antibody-hIDUA product (SE-HPLC analysis) TSKgel G3000SW XL A column (7.8 mm inner diameter x 30 cm height, Tosoh Corporation) was set in the LC-20A system, SPD-20AV UV / VIS detector (Shimadzu Corporation). The column was equilibrated with 200 mM phosphate buffer containing 5% 2-propanol and 20 mM NaCl. A 10 μL solution containing the purified humanized anti-hTfR antibody-hIDUA obtained in Example 4 at a concentration of 1.0 mg / mL was loaded onto this column at a constant flow rate of 0.6 mL / min, and the same buffer was then supplied at the same flow rate. The elution profile prepared by measuring the absorbance at 215 nm is shown in Figure 1. The obtained profile showed almost only a single peak derived from the humanized anti-hTfR antibody-hIDUA. However, a peak derived from the polymer of the humanized anti-hTfR antibody-hIDUA (polymer peak in the figure) was observed, detected before the main peak (monomer peak in the figure). Based on the ratio of the total peak area to the polymerized peak area, the proportion of polymerized material in the total humanized anti-hTfR antibody-hIDUA was calculated to be approximately 0.02%.

[0149] [Example 8] Analysis of purified humanized anti-hTfR antibody-hIDUA product (summary) The analysis results of the purified humanized anti-hTfR antibody-hIDUA product described above indicate that the purified humanized anti-hTfR antibody-hIDUA product obtained in Example 4 contains almost no impurities including HCP, and that the proportion of polymers is extremely low. In other words, the purified humanized anti-hTfR antibody-hIDUA product is of a quality that can be used as is as a pharmaceutical agent, for example, for intravenous, intramuscular, subcutaneous, intraperitoneal, intra-arterial, or intra-focal administration. [Industrial applicability]

[0150] According to the present invention, for example, it is possible to provide a fusion protein formed by conjugating an antibody and a lysosomal enzyme, which has been purified to a purity level suitable for direct use as a pharmaceutical, and in particular, a fusion protein formed by conjugating an antibody and hIDUA. [Sequence Listing Free Text]

[0151] Sequence ID 1: Amino acid sequence of linker example 1 Sequence ID 2: Amino acid sequence of linker example 2 Sequence ID 3: Amino acid sequence of linker example 3 Sequence ID 4: Amino acid sequence of linker example 4 Sequence ID 5: Amino acid sequence of the human transferrin receptor Sequence ID 6: Amino acid sequence 1 of human IDUA Sequence ID 7: Amino acid sequence 2 of human IDUA SEQ ID NO: 8: Amino acid sequence 1 of the light chain CDR1 of the anti-hTfR antibody SEQ ID NO: 9: Amino acid sequence 2 of the light chain CDR1 of the anti-hTfR antibody SEQ ID NO: 10: Amino acid sequence 1 of the light chain CDR2 of the anti-hTfR antibody SEQ ID NO: 11: Amino acid sequence 2 of the light chain CDR2 of the anti-hTfR antibody SEQ ID NO: 12: Amino acid sequence 1 of the light chain CDR3 of the anti-hTfR antibody SEQ ID NO: 13: Amino acid sequence 1 of the heavy chain CDR1 of the anti-hTfR antibody SEQ ID NO: 14: Amino acid sequence 2 of the heavy chain CDR1 of the anti-hTfR antibody SEQ ID NO: 15: Amino acid sequence 1 of the heavy chain CDR2 of the anti-hTfR antibody SEQ ID NO: 16: Amino acid sequence 2 of the heavy chain CDR2 of the anti-hTfR antibody SEQ ID NO: 17: Amino acid sequence 1 of the heavy chain CDR3 of the anti-hTfR antibody SEQ ID NO: 18: Amino acid sequence 2 of the heavy chain CDR3 of the anti-hTfR antibody SEQ ID NO: 19: Amino acid sequence of heavy chain framework region 3 of anti-hTfR antibody SEQ ID NO: 20: Amino acid sequence of the variable region of the light chain of an anti-hTfR antibody SEQ ID NO: 21: Amino acid sequence of the variable region of the heavy chain of an anti-hTfR antibody Sequence ID 22: Amino acid sequence of the light chain of the anti-hTfR antibody SEQ ID NO: 23: Amino acid sequence of the Fab heavy chain of the anti-hTfR antibody Sequence ID 24: Primer Hyg-Sfi5', synthetic sequence Sequence ID 25: Primer Hyg-BstX3', synthetic sequence Sequence ID 26: Base sequence and synthetic sequence encoding the amino acid sequence of the light chain of the anti-hTfR antibody. Sequence ID 27: Amino acid sequence of the fusion protein of the Fab heavy chain of the humanized anti-hTfR antibody and human IDUA. Sequence ID 28: Base sequence and synthetic sequence containing the gene encoding the amino acid sequence of the fusion protein of the Fab heavy chain of the humanized anti-hTfR antibody and human IDUA. Sequence ID 29: Primer IRES5', synthetic sequence Sequence ID 30: Primer IRES3', synthetic sequence Sequence ID 31: Primer mPGKP5', synthetic sequence Sequence ID 32: Primer mPGKP3', synthetic sequence Sequence ID 33: Primer GS5', synthetic sequence Sequence ID 34: Primer GS3', synthetic sequence Sequence ID 35: Primer puro5', synthetic sequence Sequence ID 36: Primer puro3', synthetic sequence Sequence ID 37: Primer SV40polyA5', synthetic sequence Sequence ID 38: Primer SV40polyA3', synthetic sequence Sequence ID 39: Primer mIRES-GS5', synthetic sequence Sequence ID 40: Primer mIRES-GS3', synthetic sequence Sequence ID 41: CMVE-EF-1αp-IFNβMAR, synthetic sequence Sequence ID 42: IRES-HygroR-mPGKpA, synthetic sequence

Claims

1. A method for producing a fusion protein obtained by fusing an antibody with a human lysosomal enzyme; (a) Mammalian cells that produce the fusion protein are cultured in serum-free medium to cause the fusion protein to be secreted into the culture medium. (b) Remove the mammalian cells from the culture medium to collect the culture supernatant. (c) The fusion protein is purified from the culture supernatant by using column chromatography with a material conjugated to a substance having affinity for the antibody as the stationary phase, anion exchange column chromatography, cation exchange column chromatography, and size exclusion column chromatography in this order. A manufacturing method which includes the following: The human lysosomal enzyme in the fusion protein is human α-L-iduronidase. The antibody in the fusion protein is Fab, The light chain of the antibody contains the amino acid sequence of SEQ ID NO: 22, and A method for producing the antibody, wherein the heavy chain of the antibody binds to human α-L-iduronidase at its C-terminus via the amino acid sequence of SEQ ID NO: 4, thereby forming the amino acid sequence of SEQ ID NO: 27 in the fusion protein.

2. The method for producing the antibody according to claim 1, wherein the substance having affinity for the antibody has affinity for the CH1 region of the heavy chain of the antibody.

3. The manufacturing method according to claim 1 or 2, wherein the anion exchanger used in the anion exchange column chromatography is a strong anion exchanger.

4. The manufacturing method according to any one of claims 1 to 3, wherein the cation exchanger used in the cation exchange column chromatography is a weak cation exchanger.

5. The manufacturing method according to any one of claims 1 to 4, wherein the anion exchanger used in the anion exchange column chromatography is a multimodal strong anion exchanger, and the cation exchanger used in the cation exchange column chromatography is a multimodal weak cation exchanger.

Images