WO2013118165A1

WO2013118165A1 - Pharmaceutical agents fused with atf for improved bioavailability

Info

Publication number: WO2013118165A1
Application number: PCT/JP2012/000805
Authority: WO
Inventors: Kenichi Takahashi; Eiji YODEN
Original assignee: JCR Pharmaceuticals Co Ltd
Current assignee: JCR Pharmaceuticals Co Ltd
Priority date: 2012-02-07
Filing date: 2012-02-07
Publication date: 2013-08-15
Anticipated expiration: 2014-08-07
Also published as: JP5957085B2; JP2015511932A

Description

PHARMACEUTICAL AGENTS FUSED WITH ATF FOR IMPROVED BIOAVAILABILITY

The present invention relates to pharmaceutical agents with improved bioavailability and a method for improving bioavailability of drugs using the amino terminal fragment of the urokinase-type plasminogen activator (ATF). More specifically, the present invention relates to delivery of pharmaceutical proteins fused to ATF into cells via binding of the ATF moiety to the urokinase type plasminogen activator receptor expressing itself on the surface of the cells.

In the human case, the urokinase-type plasminogen activator (uPA) is a protein consisting of two peptide chains linked to each other by a disulfide bond. These chains, A and B, are formed by enzymatic cleavage (with plasmin, kallikrein, cathepsin, etc.) between amino acids 178 and 179 of prourokinase, which in turn is formed by removal of the N-terminal signal peptide (amino acids 1-20) from a single-chain protein called prepro-urokinase (sc-uPA). uPA then is cleaved between amino acids 155 and 156 in vivo, thereby giving rise to the low molecular weight urokinase-type plasminogen activator (LMW-uPA) and the amino terminal fragment of urokinase type plasminogen activator (ATF)

uPA possesses a proteolytic activity and converts plasminogen to plasmin in physiological conditions to trigger a proteolytic cascade which results in thrombosis or extracellular matrix degradation. uPA comprises two domains, i.e. ATF and the protease domain. ATF is the region consisting of amino acids 1-135 on the amino terminal side, while the protease domain is the region consisting of amino acids 136 to 411 on the carboxy terminal side of uPA. ATF thus does not have a protease activity but an affinity for the urokinase type plasminogen activator receptor (uPAR)(NPL 1).

ATF itself is composed of two independent domains, i.e. the amino-terminal growth factor-like domain (1 to 43 amino acids from the N-terminus), with which uPA binds to uPAR, and a single kringle domain (43 to 135 amino acids from N-terminus) (NPL 2, 3). Therefore, ATF is itself able to bind to uPAR via the amino-terminal growth factor-like domain.

uPAR is a GPI-anchored membrane protein, consists of 335 amino acids, and has three extracellular domains, of which the first domain from N-terminus acts as the binding site of uPA(NPL 4). ATF binding to uPAR is not internalized into the cell, whereas uPA binding to uPAR is internalized if complexed with plasminogen activator inhibitors (NPL 5).

The uPAR associated protein (uPARAP), known also as ENDO180, is a lectin-like membrane protein belonging to the macrophage mannose receptor protein family (NPL 6). ENDO180 is an internalization receptor, which directs a ligand bound to it to the lysosome, where lysosomal enzymes exert their catalytic activities to degrade the ligand. ENDO180 is also known to form a triple complex with uPA and uPAR.

Lysosomal diseases, a group of rare inherited metabolic disorders, are caused by various genetic disorders in the genes encoding lysosomal enzymes. Lack of enzymatic activities of some of those enzymes results in accumulation of their substrates in cells, which then inflicts severe damages on the cells.

Lysosomal diseases include, for example, Pompe disease, Fabry's disease, Gaucher's disease, Hurler syndrome (mucopolysaccharidosis type I, MPS I), Hunter syndrome (MPS II), Maroteaux-Lamy syndrome (MPS VI), Niemann-Pick disease, and Morquio syndrome (MPS IV). Of those, several types of lysosomal diseases are treated by enzyme replacement therapies (ERT), in which the enzyme lacking in a patient is externally supplied to the patient.

Pompe disease, caused by acid alpha-glucosidase deficiency, is a lysosomal storage disorder clinically characterized by muscle weakness and cardiomyopathy. It has been well established that an enzyme replacement therapy (ERT) is effective in this disease to ameliorate weakening muscle strength and to improve heart functions, where recombinant acid alpha-glucosidase is administered to the patients. Efficacy of this therapy has been confirmed by several clinical trials in patients with different ages of onset and disease severity (NPL 7).

Fabry's disease, caused by alpha-galactosidase deficiency, is a lysosomal storage disorder clinically characterized by complication of pain, kidney failure, cardiomyopathy, and cerebrovascular events, which are responsible for morbidity and mortality of this disease. For patients with Fabry's disease, two different enzyme preparations are available. It has been well established that enzyme replacement therapy (ERT) using one of those preparations is effective for improving patient's quality of life (QOL). Efficacy of this therapy has been demonstrated, which includes amelioration of pain, reduction of heart size, improvement in hearing and sweating, stabilization of kidney functions, and retardation of renal failure in patients with end-stage renal disease (NPL 8).

Gaucher's disease, a genetic disease caused by an inherited disorder of glucosylceramidase, is a lysosomal storage disorder characterized by bruising, fatigue, anemia, low blood platelets, and enlargement of the liver and spleen. It has been well established that enzyme replacement therapy (ERT) is effective for improving patient' quality of life (QOL) (NPL 9).

Hurler syndrome, known as mucopolysaccharidosis type IH, is a genetic disease caused by an inherited disorder of alpha-L-iduronidase. Sheie's syndrome, mucopolysaccharidosis type IS, is a mild case of Hurler's syndrome. These diseases, including an intermediate type, are categorized as Hurler-Sheie's syndrome. Patients of Hurler syndrome accumulate dermatan sulphate and heparan sulphate in the lysosomes, and consequently show progressive strong body deformation and mental detergency which set in at the infantile stage, resulting in death by age 10. In the case of Sheie's syndrome, mental detergency is not, or only slightly, observed. In both cases, increase of dermatan sulphate and heparan sulphate in urine is typically observed. It has been well established that enzyme replacement therapy (ERT) with alpha-L-iduronidase significantly improves respiratory functions and the physical capacity and reduces glycosaminoglycan storage in patients with Hurler syndrome (NPL 10).

Hunter's syndrome is a genetic disease caused by an inherited disorder of alpha-L-iduronidase, iduronate-2-sulfatase (I2S) (NPL 11). I2S is a lysosomal enzyme having an activity to hydrolyze sulfate ester bonds in the molecules of glycosaminoglycans, such as heparan sulfate and dermatan sulfate. Lack of this enzyme results in the accumulation of its substrates in tissues, such as those of the liver and kidney, which then causes diverse symptoms as seen in patients suffering Hunter's syndrome, including skeletal deformities and severe mental retardation. It has been well established that the enzyme replacement therapy (ERT) improves physical capacity of patients (NPL 12).

Maroteaux-Lamy syndrome, mucopolysaccharidosis type VI, is an inherited disease caused by a genetic disorder of N-acetylgalactosamine-4-sulfatase (ASB). ASB, known as arylsulfatase B, is an enzyme which releases sulfate ion from chondroitin 4-sulfate, dermatan sulfate, and UDP-N-acetylgalactosamine 4-sulfate, by a hydrolytic reaction. It has been well established that enzyme replacement therapy (ERT) using this enzyme improves physical capacity of patients (NPL 13).

Niemann-Pick disease is categorized into types A to F according to the difference of their etiology and symptoms. Of those, types A and B are caused by an inherited deficiency of acidic sphingomyelinase (ASM). ASM is a lysosomal enzyme which hydrolyzes the sphingomyelin to the colin phosphoric acid and the ceramide. Enzymatic activity of ASM decreases remarkably in type A, whereas in type B, it remains comparatively high. Such difference comes from corresponding mutational patterns in the ASM gene. In the case of type A, the deficiency of the enzymatic activity of ASM causes accumulation of sphingomyelin in the central nervous system, the liver, the spleen, lungs, and marrow, which consequently leads to liver splenoma and a central nervous system disorder. As a result, the majority of type A patients die before reaching age four. On the other hand, the majority of type B patients don't suffer from a central nervous system disorder and are survivable to grow-up. Though efficacy of enzyme replacement therapy (ERT) with ASM has been reported, it has not been well established yet (PTL 1).

In enzyme replacement therapy, a patient receives intravenous infusion of the enzyme that he or she lacks. For the infused enzyme to exhibit its clinical effects, it is prerequisite that the enzyme be taken up into the cells where it should act. Further, for the enzyme to exhibit its clinical effect, it is also requisite that the enzyme be sorted into lysosomes where the enzyme is to exert its catalytic activity.

The enzymes which are used for ERT have sugar chains linked to them containing mannose-6-phosphate or mannose residues, and through binding to the mannose-6-phosphate receptor or the mannose receptor expressing itself on the surface of the cells, the enzymes are taken up by the cells and then sorted into the lysosomes, where they exert their enzymatic activities (NPL 14-16). However, as the mannose-6-phosphate receptor and the mannose receptor are molecules ubiquitously expressed in human body, the binding of the enzymes to these receptors is not enough for the enzymes to be sorted to the specific tissues where they should exert their activities. In addition, as these receptors are expressed in both of the liver and spleen, it is possible that infused enzymes containing mannose or mannose-6-phosphate residues in their sugar chains are rapidly removed by these organs (NPL 14).

Hence to achieve more efficient uptake of these enzymes into cells, the enzymes should be modified with some or other ligand which has affinity for other molecules on the cells than those of mannose-6-phosphate receptor or mannose receptor. Currently available enzymes for ERT, none of which contains such a ligand, are considered to be not effectively sorted to their target tissues, resulting in their limited effects.

Further, in the case of those enzymes which have relatively lower content of mannose-6-phosphate, such as acid alpha-glucosidase, binding of them to the receptor via mannose-6-phosphate is not enough for them to be taken up into the cells. To overcome this deficiency, an attempt has been made to fuse the enzyme to a receptor-associated protein, one of the ligands of the LDL receptor, reporting an improved cellular uptake of the enzyme(NPL 17).

US Patent 5686240

Blasi F., et. al., J Cell Biol. 104, 801-4 (1987) Novokhatny V., et. al., J Biol Chem. 267:3878-86 (1992). Mazar AP., et. al., Fibrinolysis. 6:49-55 (1992). Ploug M., et. al., J Biol Chem. 266:1926-33 (1991). Cubellis MV., et. al., EMBO J. 9:1079-85 (1990). Behrendt N., et. al., J Biol Chem. 275, 1993-2002 (2000) Nicolino, M., et. al., Genet Med. 11, 210-9 (2009) West, M., et. al., J Am. Soc Nephrol. 20, 1132-39 (2009) Kishnani PS, et. al., Mol Genet Metab. 96, 164-70 (2009) Wraith JE., J Pediatr. 144, 581-8 (2004) Wilson PJ et al., Proc Natl Acad Sci USA. 87: 8531-5, 1990. Muenzer J. et al., Genet Med. 8, 465-73 (2006). Harmatz P, et al., J Pediatr. 148, 533-539 (2006) Bielicki J. et. al., Biochem J. 289, 241-246 (1993) Grubb JH. et al., Rejuvenation Res. 13, 229-36 (2010) Kornfield S., FASEB J. 1, 462-8 (1987) Prince WS, J Biol Chem. 279 35037-46 (2004)

Against the above background, it is an objective of the present invention to provide a novel means for delivering such drugs into cells that are to exert their pharmacological activities only after taken up in the cells.

The present inventor found that human iduronate-2-sulfatase (hI2S) fused to human or mouse ATF can be taken up, via its binding to the urokinase-type plasminogen activator receptor (uPAR), into fibroblasts, where hI2S can exert its catalytic activity. The present invention was completed through further studies based on the finding.
Thus the present invention provides what follows:

1. A pharmaceutical agent comprising a pharmacologically active compound (A) and a peptide (B) fused thereto, wherein the peptide (B) comprises the receptor binding domain of ATF.
2. The pharmaceutical agent according to 1 above, wherein the pharmacologically active compound (A) is a protein.
3. The pharmaceutical agent according to 2 above, wherein the protein is an enzyme.
4. The pharmaceutical agent according to 3 above, wherein the enzyme is a lysosomal enzyme.
5. The pharmaceutical agent according to 4 above, wherein the lysosomal enzyme is selected from the group consisting of acid alpha-glucosidase, alpha-galactosidase, glucosylceramidase, alpha-L-iduronidase, alpha-L-iduronidase, iduronate-2-sulfatase, N-acetylgalactosamine-4-sulfatase, or acidic sphingomyelinase.
6. The pharmaceutical agent according to one of 1 to 5 above, wherein the peptide (B) is fused to the pharmacologically active compound (A) by an amide bond or an ester bond.
7. The pharmaceutical agent according to one of 1 to 5 above, wherein the peptide (B) is fused to the pharmacologically active compound (A) at the amino-terminal amino acid of the peptide (B)
8. The pharmaceutical agent according to one of 3 to 5 above, wherein the peptide (B) is fused to the protein by a peptide bond formed between the amino-terminal amino acid of the protein and the carboxy-terminal amino acid of the peptide (B).
9. The pharmaceutical agent according to one of 3 to 5 above, wherein the peptide (B) is fused to the protein by a peptide bond formed between the carboxy-terminal amino acid of the protein and the amino-terminal amino acid of the peptide (B).
10. The pharmaceutical agent according to one of 1 to 9 above, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ ID NO:33, or a derivative thereof in which 1 to 3 amino acids are added, substituted, deleted or inserted.
11. The pharmaceutical agent according to one of 1 to 9 above, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ ID NO:32, or a derivative thereof in which 1 to 10 amino acids are added, substituted, deleted or inserted.
12. The pharmaceutical agent according to 11 above, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ. ID NO:32, and wherein the peptide (B) is fused at the carboxy-terminal amino acid thereof to the pharmacologically active compound (A) through a linker consisting of at least one amino acid.
13. The pharmaceutical agent according to 12 above, wherein the linker is selected from the group consisting of -Lys-, Lys-Pro-, -Lys-Pro-Ser-, -Lys-Pro-Ser-Ser- (SEQ ID NO:41), and -Lys-Pro-Ser-Ser-Pro- (SEQ ID NO:42).
14. The pharmaceutical agent according to one of 1 to 9 above, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ ID NO:31, or a derivative thereof in which 1 to 5 amino acids are added, substituted, deleted or inserted.
15. The pharmaceutical agent according to one of 1 to 9 above, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ ID NO:30, or a derivative thereof in which 1 to 10 amino acids are added, substituted, deleted or inserted.
16. The pharmaceutical agent according to 15 above, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ. ID NO:30 and wherein the peptide (B) is fused at the carboxy-terminal amino acid thereof to the pharmacologically active compound (A) through a linker consisting of at least one amino acid.
17. The pharmaceutical agent according to 16 above, wherein the linker is selected from the group consisting of -Lys-, Lys-Pro-, -Lys-Pro-Ser-, -Lys-Pro-Ser-Ser- (SEQ ID NO:41), and -Lys-Pro-Ser-Ser-Pro- (SEQ ID NO:42).
18. A pharmaceutical composition for enzyme replacement therapy comprising the pharmaceutical agent according to one of 1 to 17 above.
19. A DNA encoding a fusion protein comprising an pharmacologically active protein (A) and a peptide (B) fused thereto, wherein the peptide (B) comprises the receptor binding domain of ATF, and wherein the peptide (B) is fused to the pharmacologically active protein (A) at the amino- or carboxy-terminus of the peptide (B) by a peptide bond.
20. The DNA according to 19 above, wherein the pharmacologically active protein is an enzyme.
21. The DNA according to 20 above, wherein the enzyme is a lysosomal enzyme.
22. The pharmaceutical agent according to 21 above, wherein the lysosomal enzyme is selected from the group consisting of acid alpha-glucosidase, alpha-galactosidase, glucosylceramidase , alpha-L-iduronidase, alpha-L-iduronidase, iduronate-2-sulfatase, N-acetylgalactosamine-4-sulfatase, acidic sphingomyelinase.
23. An expression vector comprising the DNA according to one of 19 to 22 above.
24. A mammalian cell transformed with the vector according to 23 above.

The present invention provides a means which serves to improve cellular uptake of pharmacologically active compounds through their binding to the urokinase type plasminogen activator receptor (uPAR) on the cells, and thus also provides pharmaceutical agents with improved uptake by cells expressing uPAR. In particular, as this improved cellular uptake is to be independent from any interaction of the pharmaceutical agent with the mannose-6-phosphate receptor or the mannose receptor on the cells, the present invention provides such pharmaceutical agents that, despite having no or only a small number of mannose or mannose-6-phosphate residues associated to their molecule, can still be taken up into cells

Fig. 1-1 shows a flow diagram illustrating the method for construction of pBSK(IRES-Hygr-mPGKpA) vector. Fig. 1-2 shows a flow diagram illustrating the method for construction of pBSK(NotI-IRES-Hygr-mPGKpA) vector. Fig. 1-3 shows a flow diagram illustrating the method for construction of pE-IRES-Hygr vector. Fig. 1-4 shows a flow diagram illustrating the method for construction of pPGK-neo vector. Fig. 1-5 shows a flow diagram illustrating the method for construction of pPGK-IRES-Hygr vector. Fig. 1-6 shows a flow diagram illustrating the method for construction of pPGK-IRES-GS-delta-polyA vector. Fig. 1-7 shows a flow diagram illustrating the method for construction of pE-puro vector. Fig. 1-8 shows a flow diagram illustrating the method for construction of pE-puro(XhoI) vector. Fig. 1-9 shows a flow diagram illustrating the method for construction of pE-IRES-GS-puro vector. Fig. 2 shows a flow diagram illustrating the method for construction of pE-mIRES-GS-puro vector. Fig. 3 shows a partial schematic diagram of pE-mIRES-GS-puro(mATF-hI2S) vector. Fig. 4 shows a partial schematic diagram of pE-mIRES-GS-puro(hATF-hI2S) vector. Fig. 5 shows a result of SDS PAGE analysis of purified mATF-fused hI2S. Fractions from HiTrap Q-Sepharose FF which contained hI2S activities were loaded. Fig. 6 shows a graph showing the result of measurement of cellular uptake of hATF-fused hI2S into human fibroblasts. The vertical axis shows the amount of hATF-fused hI2S or hI2S (ng) taken up per unit mass (mg) of the total cellular protein (ng/mg total cellular protein). The horizontal axis shows the concentration of hATF-fused hI2S or hI2S added to the medium. Filled circles indicate the values of cellular uptake of hATF-fused hI2S, an open circle indicates the value of cellular uptake of hATF-fused hI2S in the presence of 10 mM M6P, and an open triangle indicates the values of cellular uptake of hATF-fused hI2S in the presence of 10 mM M6P and 4.02 micrograms/mL hATF. Filled boxes indicate the values of cellular uptake of hI2S, and an open box indicates the value of cellular uptake of hI2S in the presence of 10 mM M6P. Fig. 7 shows the result of measurement of blockage by M6P of cellular uptake of hATF-fused hI2S into human fibroblasts. The vertical axis shows the cellular uptake ratio of hATF-fused hI2S or hI2S (%). The horizontal axis shows the concentration of M6P added to the medium (mM). Filled circles indicate the values of cellular uptake ratio of hATF-fused hI2S, and filled boxes indicate the values of cellular uptake ratio of hI2S. Fig. 8 shows the result of measurement of cellular uptake of mATF-fused hI2S into mouse fibroblast. The vertical axis shows the amount of mATF-fused hI2S or hI2S (ng) taken up per unit mass (mg) of the total cellular protein (ng/mg total cellular protein). The horizontal axis shows the concentration of mATF-fused hI2S or hI2S added to the medium. Filled circles indicate the values of cellular uptake of mATF-fused hI2S, and an open circle indicates the value of cellular uptake of mATF-fused hI2S in the presence of 10mM M6P. Filled boxes indicate the values of cellular uptake of hI2S, and an open box indicates the value of cellular uptake of hI2S in the presence of 10mM M6P. Fig. 9 shows the result of measurement of blockage by M6P of cellular uptake of mATF-fused hI2S into mouse fibroblasts. The vertical axis shows the cellular uptake ratio of mATF-fused hI2S or hI2S (%). The horizontal axis shows the concentration of M6P added to the medium (mM). Filled circles indicate the value of uptake ratio of mATF-fused hI2S, and filled boxes indicate the value of uptake ratio of hI2S. Fig. 10 shows the alignment of amino acid sequences of human ATF (top row) and mouse ATF (bottom row). Asterisks indicate the different amino acids between human ATF and mouse ATF.

In the present invention, the amino terminal fragment of urokinase type plasminogen activator (ATF) is preferably the mammalian ATF, more preferably the human or mouse ATF, and particularly preferably the human ATF. The amino acid sequences of the wild type mouse and human ATFs are shown as SEQ ID NO:30 and SEQ ID NO:32, respectively. In the present invention, ATF is preferably of the wild type, but mutated ATF, which has undergone addition, substitution, deletion, and/or insertion of one or more amino acids, is also preferably used. The number of amino acids added, substituted, deleted, and inserted in such a mutated ATF is preferably 1 to 10, more preferably 1 to 8, still more preferably 1 to 5, and further more preferably 1.

In the present invention, a peptide comprising the receptor binding domain of ATF is preferably a mammalian peptide, more preferably the human or mouse peptide, and particularly preferably the human peptide. The term "peptide comprising the receptor binding domain of ATF" herein does not include a compound comprising uPA. The amino acid sequences of the receptor binding domains of the wild type mouse and human ATFs are shown as SEQ ID NO:31 and SEQ ID NO:33, respectively. In the present invention, the receptor binding domain of ATF is preferably of the wild type, but a mutated receptor binding domain of ATF which has undergone addition, substitution, deletion, and/or insertion of one or more amino acids is also preferably used. The number of amino acids added, substituted, deleted, and inserted in such a mutated receptor binding domain of ATF is preferably 1 to 5, more preferably 1 to 3, and still more preferably 1.

The pharmaceutical agent according to the present invention is an agent to be administered to mammals, preferably human, livestock including horse, pig, cattle, and sheep, and pet animals including dog and cat. The agent is preferably administered intravenously. In the present invention, the term "pharmaceutical agent" does not include a compound comprising uPA.

In the present invention, since it is the receptor binding domain of ATF that plays the key role in leading a pharmacologically active compound to the binding to the uPAR on the surface of cells and then into them, there is no particular limitation as to the scope of pharmacologically active compounds to which the peptide comprising the receptor binding domain of ATF is to be fused. Preferred examples of pharmacologically active compounds include, but are not limited to, proteins, polysaccharides, lipids, and other kind of chemicals including antibiotics, anticancer agents, and anti-inflammatory agents. Among them, proteins are particularly preferred. The peptide comprising the receptor binding domain of ATF may be fused to a pharmacologically active compound by any kind of chemical bond as desired, of which amide bond and ester bond are preferred. The peptide comprising the receptor binding domain of ATF may be fused to a pharmacologically active compound by an amide bond or an ester bond formed on its carboxy- or amino-terminal amino acid of the peptide. While there is no particular limitation on the number of the peptide comprising the receptor binding domain of ATF to be fused per molecule of a pharmacologically active compound, the number is preferably 1 to 10, more preferably 1 to 5, and still more preferably 1 to 3, and further more preferably 1.

Where a pharmacologically active compound to which the peptide comprising the receptor binding domain of ATF is to be fused is a protein, the peptide is preferably fused to the protein by a peptide bond. More preferably, the peptide comprising the receptor binding domain of ATF may be fused to the protein, either at the protein's carboxy- or amino-terminal amino acid. Further the peptide comprising the receptor binding domain of ATF may be fused to the protein via linker. A linker for this purpose may be a single or more amino acids interposed between two adjacent termini of the protein and the peptide comprising the receptor binding domain of ATF. The linker may consist of preferably 1 to 20 amino acids, more preferably 1 to 10 amino acids, and still more preferably 1 to 5 amino acids. The amino acid sequence of such a linker may be preferably selected from the group consisting of -Lys-, Lys-Pro-, -Lys-Pro-Ser-, -Lys-Pro-Ser-Ser- (SEQ ID NO:41), and -Lys-Pro-Ser-Ser-Pro- (SEQ ID NO:42).

A fusion protein as a pharmaceutical agent made of the peptide comprising the receptor binding domain of ATF and a pharmacologically active protein fused between the carboxy-terminal amino acid of the former and the amino-terminal amino acid of the latter can be produced by letting a fusion gene express itself which comprises the gene encoding the peptide comprising the receptor binding domain of ATF and the gene encoding the pharmacologically active protein fused in frame in this order tandemly in the 5' to 3' orientation. Examples of such fusion genes are shown as SEQ ID NO:26, i.e., a fusion gene between the mouse ATF and the human I2S genes, and as SEQ ID NO:28, i.e., a fusion gene between the human ATF and the human I2S genes.

A fusion protein as a pharmaceutical agent made of the peptide comprising the receptor binding domain of ATF and a pharmacologically active protein fused between the amino-terminal amino acid of the former and the carboxy-terminal amino acid of the latter can be produced by letting a fusion gene express itself which comprises the gene encoding the pharmacologically active protein and the gene encoding the peptide comprising the receptor binding domain of ATF fused in frame in this order tandemly, in the 5' to 3' orientation.

In order for one of the above-mentioned fusion genes to be expressed, a gene chosen is inserted into an expression vector, and the expression vector thus obtained then is introduced into host cells. There is no particular limitation as to the species to which the host cells to be used above belongs, but they are preferably E. coli, yeast, insect cells, or mammalian cells, among which mammalian cells are more preferred, and CHO cells are particularly preferred. The host cells thus obtained, which have been transformed with the expression vector, then are incubated for the fusion protein to be expressed.

In the present invention, there is no particular limitation as to the scope of pharmacologically active proteins to which the peptide comprising receptor binding domain of ATF is to be fused. Preferred examples of pharmacologically active proteins include, but are not limited to, enzymes, among which particularly preferred are lysosomal enzymes, inter alia acid alpha-glucosidase, alpha-galactosidase, alpha-L-iduronidase, alpha-L-iduronidase, iduronate-2-sulfatase, N-acetylgalactosamine-4-sulfatase, or acidic sphingomyelinase. Pharmacologically active proteins to be employed in the present invention are preferably mammalian proteins, and more preferably human proteins.

Where a lysosomal enzyme is employed, it is preferably the wild type enzyme, though a mutated enzyme is also preferably used as far as it retains the catalytic activity. A lysosomal enzyme and the peptide comprising the receptor binding domain of ATF can be fused via a linker consisting of at least one amino acid, which may form peptide. Such a linker is covalently interposed between the amino- and carboxy-terminal amino acids of the two moieties to be linked, i.e., an enzyme employed and the peptide comprising receptor binding domain of ATF. A linker comprises preferably 1 to 20 amino acids, more preferably 1 to 10 amino acids, and still more preferably 1 to 5 amino acids.

In the present invention, a lysosomal enzyme fused to the peptide comprising the receptor binding domain of ATF can be used as a pharmaceutical agent in enzyme replacement therapy for lysosomal disease. For example, alpha-glucosidase, alpha-galactosidase, alpha-L-iduronidase, glucosylceramidase, alpha-L-iduronidase, iduronate-2-sulfatase, N-acetylgalactosamine-4-sulfatase, or acidic sphingomyelinase fused to the peptide comprising the receptor binding domain of ATF is used as pharmaceutical agent in enzyme replacement therapy for Pompe disease, Fabry's disease, Gaucher's disease, Hurler syndrome (mucopolysaccharidosis type I, MPS I), Hunter syndrome (MPS II), and Niemann-Pick disease, respectively.

The present invention is described in further detail below with reference to examples. However, it is not intended that the present invention be limited to the examples.

(Construction of pE-IRES-GS-puro vector)
pPGKIH expression vector (Miyahara M. et.al., J. Biol. Chem. 275,613-618 (2000)) was digested with XhoI and BamHI to cut out a DNA fragment containing Internal Ribosome Entry Site (IRES) derived from murine encephalomyelitis virus (EMCV), hygromycin resistance gene (Hygr gene), and polyadenilation signal (mPGKpA) of mouse phosphoglycerate kinase (mPGK). This fragment was designated IRES-Hygr-mPGKpA, whose nucleotide sequence is set forth as SEQ ID NO:1, in which (i) nucleotides 1-6 represent an XhoI site, (ii) nucleotides 7-119 an intermediate sequence (iii) nucleotides 120-718, including the "atg" at the 3' end of this region, represent a sequence containing IRES derived from 5' untranslated region of murine encephalomyelitis virus, (iv) nucleotides 716-1741 a hygromycin resistance gene (Hygr gene), (v) nucleotides 1742-1746 an intermediate sequence (vi) nucleotides 1747-2210 the sequence containing polyadenilation signal sequence of mouse phosphoglycerate kinase gene(mPGKpA), and (vii) nucleotides 2211-2216, i.e., the six nucleotides at the 3'-terminus, a BamHI site. The DNA fragment was inserted between XhoI and BamHI sites of pBSK vector (STRATAGENE), and the vector thus obtained was designated pBSK (IRES-Hygr-mPGKpA)(Fig. 1-1).

A DNA fragment containing part of EMCV-IRES was amplified by PCR performed with a set of primers, IRES5' (SEQ ID NO:2) and IRES3' (SEQ ID NO:3), and using pBSK (IRES-Hygr-mPGKpA) as a template. The PCR-amplified DNA fragment was digested with XhoI and HindIII, and subsequently inserted between XhoI and HindIII sites of pBSK vector, and the vector thus obtained was designated pBSK (NotI-IRES-Hygr-mPGKpA)(Fig. 1-2). pBSK (NotI-IRES-Hygro-mPGKpA) was digested with NotI and BamHI, and the DNA fragment obtained containing part of EMCV-IRES was inserted between NotI and BamHI sites of pE-hygr vector. pE-hygr vector was constructed as per the procedure disclosed in a patent literature (JP2011-172602). The resulting vector was designated pE-IRES-Hygr (Fig. 1-3).

PCR was performed with a set of primers, mPGKP5' (SEQ ID NO:4) and mPGKP3' (SEQ ID NO:5), and using pPGKIH as a template to amplify a DNA fragment containing mPGK promoter region (mPGKp), whose nucleotide sequence is set forth as SEQ ID NO:6, in which (i) nucleotides 1-3 represent an intermediate sequence (ii) nucleotides 4-9 a BamHI site, (iii) nucleotides 10-516 a sequence containing mPGKp, (iv) nucleotides 517-523 an intermediate sequence , (v) nucleotide 524-529 an EcoRI site, and (vi) nucleotides 530-532, the last three nucleotides, represent an intermediate sequence. The amplified DNA fragment then was digested with BglII and EcoRI, and inserted between BglII and EcoRI sites of pCI-neo (Promega). The vector thus obtained was designated pPGK-neo (Fig. 1-4).

pE-IRES-Hygr was digested with NotI and BamHI, and the DNA fragment thus obtained (IRES-Hygr) was inserted between NotI and BamHI sites of pPGK-neo. The vector thus obtained was designated pPGK-IRES-Hygr (Fig. 1-5).

A DNA fragment containing a glutamine synthetase gene (GS gene) was amplified by PCR performed with a set of primers, GS5' (SEQ ID NO:7) and GS3' (SEQ ID NO:8), and using a cDNA library prepared from mRNA of CHO-K1 cells as a template. The DNA fragment thus obtained was digested with BalI and BamHI, and inserted between BalI and BamHI sites of pPGK-IRES-Hygr. The vector obtained was designated pPGK-IRES-GS-delta-polyA (Fig. 1-6).

PCR was performed with a set of primers, puro5' (SEQ ID NO:9) and puro3' (SEQ ID NO:10), and using pCAGIPuro (Miyahara M. et.al., J. Biol. Chem. 275,613-618(2000)) as a template to amplify a DNA fragment containing puromycin resistance gene (Puro gene), whose nucleotide sequence is set forth as SEQ ID NO:11, in which (i) nucleotides 2-7 represent an AflII site, (ii) nucleotides 8-607 the sequence containing Puro gene, and (iii) nucleotides 608-619 a BstXI site. The DNA fragment obtained was digested with AflII and BstXI, and inserted between AflII and BstXI sites of pE-neo vector. pE-neo vector was constructed as per the procedure disclosed in a patent literature (JP2011-172602). The vector thus obtained was designated pE-puro (Fig. 1-7).

PCR was performed with a set of primers, SV40polyA5' (SEQ ID NO:12) and SV40polyA3' (SEQ ID NO:13), and using pE-puro as a template to amplify a DNA fragment containing the SV40 late polyadenilation region. The DNA fragment thus amplified was digested with NotI and HpaI, and inserted between NotI and HpaI sites of pE-puro. The vector obtained was designated pE-puro(XhoI) (Fig. 1-8).

pPGK-IRES-GS-delta-polyA was digested with NotI and XhoI, and inserted between NotI and XhoI sites of pE-puro(XhoI). The vector obtained was designated pE-IRES-GS-puro (Fig. 1-9).

(Construction of pE-mIRES-GS-puro vector)
PCR reaction was performed with a set of primers, mIRES-GS5' (SEQ ID NO:14) and mIRES-GS3' (SEQ ID NO:15), and using pE-IRES-GS-puro as a template to amplify a DNA fragment containing EMCV-IRES of which the second start codon was disrupted. Then, using the DNA fragment thus amplified and aforementioned IRES5' as primers and, as a template, pE-IRES-GS-puro, PCR reaction was performed to amplify the region from EMCV-IRES to GS gene. The DNA fragment thus amplified, containing the region from EMCV-IRES to GS gene, was digested with NotI and PstI and inserted between NotI and PstI sites of pE-IRES-GS-puro. The vector obtained was designated pE-mIRES-GS-puro (Fig. 2).

(Cloning of cDNA encoding human ATF (human amino terminal fragment of uPA; hATF) gene)
hATF gene was amplified by PCR performed with a set of primers, hATF-f (SEQ ID NO:16, in which nucleotides 4-9 represent an MluI site) and hATF-r (SEQ ID NO:17, in which nucleotides 4-9 represent an XhoI site), and using human kidney Quick Clone cDNA (Clontech) as a template. In the nucleotide sequence of hATF-f or hATF-r, nucleotides 4-9 represent MluI or XhoI site, respectively. The fragment thus amplified was digested with MluI and XhoI, and then inserted between MluI and XhoI sites of pCI-neo vector (Promega) to give pCI-neo (hATF) vector.

(Cloning of cDNA encoding mouse ATF (mouse amino terminal fragment of uPA; mATF) gene)
mATF gene was amplified by PCR performed with a set of primers, mATF-f (SEQ ID NO:18) and mATF-r (SEQ ID NO:19), and using mouse kidney Quick Clone cDNA (Clontech) as a template. In the nucleotide sequence of mATF-f or mATF-r, nucleotides 4-9 represent MluI or XhoI site, respectively. The fragment thus amplified was digested with the MluI and XhoI, and then inserted between MluI and XhoI sites of pCI-neo (Promega) to give pCI-neo (mATF) vector.

(Cloning of cDNA encoding human iduronate-2 sulfatase (hI2S) gene)
Using a human placenta cDNA library (Takara Bio) as a template, a nested PCR reaction was performed with two sets of primers, composed of a set of outer primers, hI2S-f (SEQ ID NO:20) and hI2S-r (SEQ ID NO:21) for the 1st reaction, and a set of inner primers, hI2S-f2 (SEQ ID NO:22) and hI2S-r2 (SEQ ID NO:23) for the 2nd reaction to amplify the DNA fragment containing hI2S cDNA. The PCR fragment thus amplified was inserted into EcoRV site (blunt cloning site) of pT7Blue vector (Novagen) to give pT7Blue(hI2S) vector. Then, cDNA encoding hI2S lacking signal sequence was amplified by PCR performed with a set of primers, hI2S-f3 (SEQ ID NO:24) and hI2S-r3 (SEQ ID NO:25), and using pT7Blue(hI2S) vector as a template. In the nucleotide sequence of hI2S-f3, nucleotides 4-9 represent a XhoI site, and in that of hI2S-r3, nucleotides 4-11 represent a NotI site. The PCR fragment thus amplified was digested with the XhoI and NotI, and then inserted between XhoI and NotI sites of pCI-neo (Promega) to give pCI-neo(delta-S-hI2S) vector.

(Construction of the expression vectors for ATF-fused hI2S)
pCI-neo (mATF) vector was digested with MluI and XhoI to cut out the cDNA encoding mATF. pCI-neo (delta-S-hI2S) was digested with XhoI and NotI to cut out the cDNA encoding hI2S lacking the signal sequence. Then both cDNAs were tandemly inserted between MluI and NotI sites of pE-mIRES-GS-puro to give pE-mIRES-GS-puro (mATF-delta-S-hI2S) (Fig. 3). The DNA sequence of mATF-fused hI2S gene embedded in the above vector is set forth as SEQ ID NO:26. The amino acid sequence of mATF-fused hI2S translated from the gene is set forth as SEQ ID NO:27, in which the portion consisting of the first 160 amino acids correspond to part of mouse sc-uPA, of which the first 21 amino acids correspond to the leader sequence, with the 135 amino acids that follow (amino acids 22 to 156) corresponding to mouse ATF (mATF, SEQ ID NO:30). The first 43-amino acid sequence of mATF comprises the receptor binding domain of mATF (SEQ ID NO:31).When expressed in CHO cells, mATF-fused hI2S, after removal of the leader sequence, is secreted from the cells.

pCI-neo (hATF) vector was digested with MluI and XhoI to cut out the cDNA encoding hATF. pCI-neo (delta-S-hI2S) was digested with XhoI and NotI to cut out the cDNA encoding hI2S lacking the signal sequence. Then both cDNAs were tandemly inserted between MluI and NotI sites of pE-mIRES-GS-puro to give pE-mIRES-GS-puro (hATF-delta-S-hI2S) (Fig. 4). The DNA sequence of hATF-fused hI2S gene embedded in the above vector is set forth as SEQ ID NO:28. The amino acid sequence of hATF-fused hI2S translated from the gene is set forth as SEQ ID NO:29, in which the portion consisting of the first 159 amino acids corresponds to part of human sc-uPA, of which the first 20 amino acids correspond to the leader sequence, with the 135 amino acids that follow (amino acids 21 to 155) corresponding to mouse ATF (hATF, SEQ ID NO:32). The first 43-amino acid sequence of hATF comprises the receptor binding domain of hATF (SEQ ID NO:33). When expressed in CHO cells, hATF-fused hI2S, after removal of the leader sequence, is secreted from the cells after the leader sequence is removed.

(Production of recombinant cells for expression of ATF-fused hI2S)
CHO-K1 cells (purchased from American Type Culture Collection) was transfected with one of the above-mentioned expression vectors, pE-mIRES-GS-puro(mATF-hI2S) or pE-mIRES-GS-puro(hATF-hI2S), using Lipofectamine2000 (Invitrogen) according to the following method. Briefly, on the day before transfection, 1 x 10⁶CHO-K1 cells were seeded in a 3.5-cm culture dish containing 3 mL of D-MEM/F12 medium containing 5% FCS (D-MEM/F12/5%FCS), and the cells were cultured overnight at 37 deg C in a humidified atmosphere of 5% CO₂ and 95% air. On the following day, the cells were transfected with 300 microliters of a 1:1 mixture solution consisting of Lipofectamine 2000 solution diluted 25 times with Opti-MEM I medium (Invitrogen) and a plasmid DNA solution diluted with Opti-MEM I medium to 13.2 micrograms/mL, at 37 deg C in a humidified atmosphere of 5% CO₂ and 95% air over night.

After transfection, the medium was replaced with a selection medium (the CD Opti CHO culture medium (Invitrogen) supplemented with 10 micrograms/mL insulin, 100 micromole/L hypoxanthine, 16 micromole/L thymidine and 30 micromole/L methionine sulfoximine (MSX, Sigma)), and a selective culture was carried out at 37 deg C in a humidified atmosphere of 5% CO₂ and 95% air. Cells that had grown in the selection medium were subjected to several successive rounds of subculture in the medium to give recombinant cells. The concentration of MSX and puromycin were escalated from 30 to 100 micromole/L and from 0 to 10 micrograms/ml, respectively, in the course of the selective culture. The cells survived in the course of the selective culture above were used as recombinant cells for expression of ATF-fused hI2S.

(Culture of recombinant cells for expression of ATF-fused hI2S)
The above recombinant cells were diluted to a cell density of 2 x 10⁵ cells/mL and cultured for 7 days in CD Opti CHO culture medium supplemented with 4 mM L-glutamine, 10 micrograms/mL insulin, 100 micromole/L hypoxanthine, 16 micromole/L thymidine and 10 micrograms/ml puromycin at 37 deg C in a humidified atmosphere of 5% CO₂ and 95% air. Then the culture supernatant was collected by centrifugation.

(Purification of ATF-fused hI2S)
To the culture supernatant collected above was added acetic acid to adjust the pH of the culture supernatant to 5.0. A precipitate formed by this was removed by filtration through a 0.22-micrometer membrane (Millipore). The culture supernatant thus recovered was loaded and adsorbed on a Capto MMC column (GE Healthcare), a cation-exchange column having a selectivity based both on hydrophobic interaction and on hydrogen bond formation, which had been equilibrated with a threefold column volume of 100 mM acetate buffer (pH 5.0). Then the column was washed with a fourfold column volumes of the same buffer. Subsequently the adsorbed ATF-fused hI2S was eluted and fractionated with a linear gradient of 20 column volumes of a buffer, from 100 % of 100 mM acetate buffer (pH 5.0) to 100% of 100mM HEPES buffer (pH 8.0) containing 1.5 M sodium chloride. After measuring the I2S activity by a method as described below, fractions having hI2S activity were collected as the eluate of interest. This eluate from the Capto MMC column above was diluted 10 times with 25 mM sodium phosphate buffer (pH 7.4) and loaded and adsorbed on a HiTrap Q-Sepharose FF 5 mL column (GE health care company), which had been equilibrated with the same buffer. Then the column was washed with a fourfold column volumes of the same buffer. Subsequently the adsorbed ATF-fused hI2S was eluted and fractionated with a linear gradient of 20 column volumes of a buffer, from 100% of 25 mM sodium phosphate buffer (pH7.4) to 100% of 25 mM sodium phosphate buffer (pH7.4) containing 1.5 M sodium chloride. After measuring the I2S activity by a method as described below, fractions having hI2S activity were collected and analyzed by SDS-PAGE, which revealed a single band having a molecular weight corresponding to that of ATF-fused hI2S (approximately 92kD) (Fig. 5). The collected fractions were used for further analysis as purified ATF-fused hI2S.

(Measurement of the Activity of rhI2S)
Samples were desalted by membrane filtration using vertical polyethersulfone membrane (VIVASPIN 2 5,000 MWCO PES; Sartorius) as ultrafilter membrane, and then desalted samples were diluted to approximately 100 ng/mL with Reaction Buffer (5 mM sodium acetate, 0.5 mg/L BSA, 0.1% Triton X-100, pH 4.45). To each well of a 96-well microtiter plate (FluoroNunc Plate, Nunc) 10 microliters of each rhI2S sample was added and pre-incubated for 15 minutes at 37 deg C. Substrate solution was prepared by dissolving 4-methyl-umbelliferyl sulfate (SIGMA) in Substrate Buffer (5 mM sodium acetate, 0.5 mg/mL BSA, pH 4.45) to a final concentration of 1.5 mg/mL. 100 microliters of Substrate solution was added to each well containing ATF-fused hI2S sample and the plate was let stand for 1 hour at 37 deg C in the dark. After the incubation, 190 microliters of Stop Buffer (0.33 M glycine, 0.21 M sodium carbonate, pH 10.7) was added to each well containing the sample. 150 microliters of 0.4 micromole/L 4-methylumbelliferone (4-MUF, Sigma) solution and 150 microliters of Stop Buffer was added to a well as the standard, then the plate was read on a 96-well plate reader with excitation light at the wavelength of 330 nm and fluorescent light at the wavelength of 440 nm.

A standard curve was produced by measuring fluorescence intensity at various concentrations of 4-MUF in solution. The fluorescence intensity of each sample was extrapolated to the standard curve. Results were calculated as activity in Units/mL where one Unit of activity was equal to 1 micromole of 4-MUF produced per minute at 37 deg C. A published US patent application (publication No. 2004-0229250) was referred to for conducting this measurement.

(Measurement of mannose-6-phosphate (M6P) content)
Desalted samples containing about 0.1 mg protein were dried under reduced pressure, dissolved in 50% trifluoroacetic acid, and heated at 100 deg C for 2 hours. Then the samples were dissolved in 100 mM boric acid solution adjusted the pH to 9.0 with NaOH. Twenty mg of sodium mannose-6-phosphate was dissolved in water to make 100 mL and this was used as the M6P standard solution. The M6P standard stock solution was frozen and stored in 1 mL aliquots at -20 deg C, which was thawed before use, and diluted twofold with purified water and used as the M6P standard solution.

To purified water was added 6.2 g of boric acid and allowed to dissolve, and after adjusting the pH of this solution to 9.0 with 2 N sodium hydroxide, purified water was added to make the total volume of 1000 mL, which then was suction-filtered through a 0.22 micrometer membrane filter (Millipore). The solution thus obtained was designated solution A (100 mM boric acid solution (pH 9.0)). To purified water were added 6.2 g of boric acid and 11.7 g of sodium chloride and allowed to dissolve, and after adjusting the pH of this solution to 9.0 with 2 N sodium hydroxide, purified water was added to make the total volume of 1000 mL, which then was suction-filtered through a 0.22 micrometer membrane filter. The solution thus obtained was designated solution B (100 mM boric acid-200 mM sodium chloride solution (pH 9.0)).

To purified water were added 10 g of L-arginine and 30 g of boric acid and allowed to dissolve, and the total volume was adjusted to 1000 mL, suction-filtered through a membrane filter with the pore size of not more than 0.22 micrometer, and the solution thus obtained was used as the reagent solution for reaction.

An anion exchanger column, Shimpack ISA-07/S2504 (4.0 mm I.D. x 250 mm)(base material: polystyrene gel, stationary phase: quaternary ammonium group) was attached to Shimazu HPLC System LC-10Avp (reducing sugar analysis system), and further, Shim-pack guard column ISA (4.0 mm I.D. x 50 mm) was set as a column oven used to heat the column. A heat block (ALB-221, mftd. by Asahi Techno Glass) for heat reaction was set downstream of the outlet of the column. The column was heated in the column oven at 65 deg C, and the heat block was set at 150 deg C. A fluorescence detector system was installed downstream of the heat block, and adjusted so that it would irradiate excitation light at the wavelength of 320 nm and detect fluorescent light at the wavelength of 430 nm.

The solutions A and B were set on the autosampler of the reducing sugar analysis system, which then was set so that the reagent solution for reaction was supplied downstream of the outlet of the column (upstream of the heat block). After the column was equilibrated with the mobile phase (solution A) with which chromatography was to be started, the M6P standard solution or the samples were loaded onto the column.

After the column was loaded with the standard solution of M6P or the sample solutions, a first mobile phase prepared by mixing solution A and solution B at a volume ratio of 90:10 (thus, containing 100 mM boric acid and 20 mM sodium chloride) was let flow through the column at a flow rate of 0.3 mL/min for 35 min; then the volume ratio between solution A and solution B was changed in a linear fashion to 25:75 over 25 min at the same flow rate (thus, containing 100 mM boric acid, while sodium chloride being increased up to 150 mM), and further the volume ratio of solution B was set at 100% (thus, containing 100 mM boric acid and 200 mM sodium chloride) and the solution was let flow at the same flow rate for 10 min, and then solution A and solution B were let flow at a volume ratio of 90:10 (thus, containing 100 mM boric acid and 20 mM sodium chloride) as was the case of the first mobile phase. The reagent solution for reaction was supplied to the flow path downstream of the outlet from the column at a flow rate of 0.2 mL/min.

The peak areas corresponding to M6P were compared between standard solution and the samples, and the average number of M6P contained per protein molecule was calculated as mol/mol ratio.

The average number of M6P contained in hATF-fused hI2S and mATF-fused hI2S was calculated to be 2.6 mol/mol and 2.2 mol/mol, respectively. On the other hand the average number of M6P contained in recombinant human I2S which had been provided as pharmaceutical medicine (hereunder designated rhI2S) was calculated as 4.4 mol/mol.

(Production of His-tagged-hMPR9)
For measurement of the affinity of ATF-fused hI2S for the human mannose-6-phasphate receptor (hM6PR), His-tagged-hMPR9, which contained domain 9 of the human M6P receptor, to which ATF binds, was produced as described below.

A plasmid containing cDNA encoding the human mannose-6-phosphate receptor (human M6P receptor), which is essential for M6P's binding to the cells, was obtained from ATCC (ATCC No.95660). A DNA fragment encoding the domain 9 of the human M6P receptor (hMPR9) was amplified from the plasmid by PCR using a set of primers, hMPR9-f (SEQ ID NO:35) and hMPR9-r (SEQ ID NO:36).

The amplified DNA fragment was digested with NcoI and NotI , and inserted between NcoI and NotI sites of pET26 vector (Novagen). The resultant plasmid was designated pET26-MPR9. Two-step PCR was conducted for amplification of a DNA fragment whose sequence is set forth as SEQ ID NO:34, which encoded hMPR9 with added sequences at both of its 5'- and 3'-termini. This amplified DNA fragment, encoding hMPR9 having Bip signal at its N-terminus and His-tag at its C-terminus, was designated Bip- tagged-hMPR9.

In the above, The first reaction of the two-step PCR was conducted using pET26-MPR9, as template, and a set of primers: MPR9-f2 (SEQ ID NO:37) and MPR9-r2 (SEQ ID NO:38). Subsequently, using the amplified DNA fragment as template, the second PCR was conducted using a set of primers: MPR9-f3 (SEQ ID NO:39), and MPR9-r3 (SEQ ID NO:40).

Then the resultant DNA fragment was digested with EcoRV and NotI, and ligated into the Eco47III-NotI site of the pIB/V5-His-DEST vector (Invitrogen). The resultant plasmid was designated pXBi-MPR9, with which High Five cells were transfected to obtain cells expressing His-tagged-hMPR9 which was derived from Bip-tagged hMPR9 through removal of Bip signal sequence.

High Five cells (Invitrogen) were grown in a 24-well plate until 50% confluent using Express Five medium (Invitrogen), and transfected with the pXBi-MPR9 using the Hily Max transfection reagent (Dojin chemical, Japan). The cells were cultured in the presence of 30 micrograms/mL blasticidin to select stable transfectant. Stably transfected cells then were expanded and cultured in a Erlenmeyer flask (100 mL) for 4 days. The culture then was harvested and centrifuged at 3000 rpm for 30 min, and the supernatant was collected. The supernatant was filtrated though a 0.22 micrometer filter (Millipore), then diluted 5-fold with an equilibration buffer (10 mM phosphate buffer containing 300 mM NaCl (pH 7.2)). The diluted supernatant was applied to the chromatography column with Profinity IMAC Ni-charged Resin (bed volume: 1 mL, Bio-Rad) which had been equilibrated with the equilibration buffer, and the column then was washed with 5 bed volumes of the equilibration buffer. Then the His-tagged-hMPR9 bound to the resin was eluted by applying to the column 5 bed volumes of 10 mM NaPO₄, 300 mM NaCl and 10 mM imidazole (pH 7.2), and subsequently 5 bed volumes of 10 mM NaPO₄, 300 mM NaCl and 300 mM imidazole (pH 7.2). Fractions containing His-tagged-hMPR9 were collected and concentrated by Amicon 3K (Millipore) with the buffer exchanged to 20mM Tris buffer containing 150 mM NaCl (pH 7.4). The concentration of the His-tagged-hMPR9 was determined by measuring absorbance at 280 nm.

(Measurement of affinity of ATF-fused hI2S for hM6PR and huPAR)
Binding affinity of ATF-fused hI2S for hM6PR (human mannose-6-phosphate receptor) and huPAR (human urokinase type plasminogen activator) was measured using Biacore T100 (GE Healthcare) equipped with nitrilotriacetic acid fixed sensor chip (Series S Sensor Chip NTA "BR-1005-32"). Biacore T100 is measuring apparatus based on surface plasmon resonance (SPR), where a sample containing a ligand is sent at a constant flow rate onto the surface of the sensor chip on which a receptor is fixed. If the ligand binds to the receptor surface, the mass of the sensor chip is increased due to the ligand bound to the receptor, and a shift of the SPR signal can be detected as a change in the resonance unit (RU) in proportion to the amount of the ligand bound. For proteins, in general, 1 RU corresponds to approximately 1 pg/mm².

As the domain in hM6PR to which M6P binds is the domain 9, His-tagged domain 9 of hM6PR (His-tagged-hMPR9) prepared as above was used as the receptor for M6P. And as the receptor for ATF, a recombinant huPAR (rhuPAR) purchased from R&D Systems (Minneapolis, MN) was used in this experiment.

To activate the sensor chip, 10 mM HEPES (pH 7.4) containing 500 micromole/L NiCl₂, 150 mM NaCl, 50 micromole/L EDTA and 0.05% Surfactant P20 was run at the flow rate of 10 microliters/min for 60 sec, and then, (i) approximately 2.1 microliters of 2.8 micrograms/mL His-tagged-hMPR9 dissolved in 10 mM HEPES (pH 7.4) containing 150 mM NaCl, 50 micromole/L EDTA and 0.05% Surfactant P20 or (ii) approximately 2.1 microliters of 5 micrograms/mL rhuPAR (R&D Systems) dissolved in the same buffer was applied, and subsequently the same buffer was run at the flow rate of 10 microliters/min for 60 sec to fix His-tagged-hMPR9 or rhuPAR on the activated sensor chip.

Each sample was diluted so as to adjust the concentration of ATF-fused hI2S, or of rhI2S, to 12.5, 6.25, 3.125 and 1.5625 nmol/L, with 10 mM HEPES (pH7.4) containing 150 mM NaCl, 50 micromole/L EDTA and 0.05% Surfactant P20. Each dilution then were applied at 50 microliters/min for 300 sec for the ligand (ATF-fused hI2S or rhI2S) to be bound to the corresponding receptor (His-tagged-hMPR9 or rhuPAR) on the sensor chip. Then, 10 mM HEPES (pH 7.4) containing 150 mM NaCl, 50 micromole/L EDTA and 0.05% Surfactant P20 was run at the flow rate of 50 microliters/min for 180 sec while constantly monitoring the dissociation status between the ligand (ATF-fused hI2S or rhI2S) and the receptor (His-tagged-hMPR9 or rhuPAR). Subsequently 10 mM HEPES (pH 8.3) containing 150 mM NaCl, 350 mM EDTA and 0.05% Surfactant P20 was run at the flow rate of 50 microliters/min for 60 sec to regenerate the sensor chip. The dissociation constant (Kd) was automatically calculated by Biacore T100 Evaluation Software based on the dissociation status monitored above.

As a result, Kd between hATF-fused hI2S and rhuPAR was calculated to be 1.16 nM, and that between mATF-fused hI2S and rhuPAR to be 5.59 nM. Thus the affinity of hATF-fused hI2S for rhuPAR was 4.8-fold higher than that of mATF-fused hI2S for rhuPAR. Specific binding between hI2S and rhuPAR was not detected. These data show that hATF fused to hI2S is able to bind to huPAR efficiently, and mATF, 31% of whose amino acids are different from hI2S's (42 of 135 amino acids), is still able to bind to huPAR efficiently (Fig. 10).

Further, Kd between hATF-fused hI2S and hM6PR was calculated to be 0.40 nM, and that between mATF-fused hI2S and hM6PR (domain9 of hM6PR) to be 0.43 nM. These values were comparable to the Kd value between hI2S and hM6PR (Kd of 0.47 nM). These data indicate that either of the murine or human ATF-fused hI2S has a binding affinity for hM6PR, which is similar to that of hI2S, despite their lower contents of M6P.

(Measurement of cellular uptake of hATF-fused hI2S using human fibroblast)
Cultured human fibroblasts (CCD-1076SK, purchased from DS Pharma Biomedical Co., Ltd.), which expressed both of M6P receptor and urokinase type plasminogen activator receptor (uPAR), were suspended in MEM-Eagle's medium (Gibco) containing 10% heat-inactivated FBS and 2 mM L-glutamine, and the cell density was adjusted to 8.0 x 10⁴cells/mL. One hundred microliters of this cell suspension were seeded into each well of a 96-well microplate and cultured for 2 days at 37 deg C in a humidified atmosphere of 5% CO₂ and 95% air. Samples containing hATF-fused hI2S or rhI2S were serially diluted with the medium to make various concentrations of them from 4.88 ng/mL to 20 micrograms/mL. Then, the medium in the 96-well microplate where the fibroblasts had been seeded was removed with a micropipette, and 100 microliters of each sample diluted above was added in a duplicate manner to the wells, and incubation was done for 18 hrs, where the final concentrations of hATF-fused hI2S or rhI2S were 4.88 ng/mL to 20 micrograms/mL. To confirm the specific binding of hATF-fused hI2S or rhI2S to the mannose 6-phosphate (M6P) receptor in this assay, samples containing 10 mmol/L of mannose 6-phosphate (M6P) as an antagonist were prepared and added to the 96-well microplate and incubated in the same manner as described above. The assay was made in a duplicate manner.

After washed with ice cold PBS thrice, the cells were lysed with M-PER Mammalian protein extraction reagent (Thermo Scientific) supplemented with a 0.5% protease inhibitor cocktail (Sigma). Then the amount of total cellular protein was measured by Pierce BCA^TM Protein Assay kit (Pierce, IL, USA), and the amount of hATF-fused hI2S and rhI2S taken up in the cells was determined by ELISA as described below. The amount of hATF-fused hI2S and rhI2S taken up was calculated per unit mass (mg) of the amount of total cellular protein, and plotted on a graph.

Compared with rhI2S, the level of cellular uptake of hATF-fused hI2S into normal human fibroblast was low (Fig 6). That may be because the number of M6P contained in hATF-fused hI2S (2.2 per protein molecule) was lower than that in rhI2S (4.4 per protein molecule), which lowered the efficiency of cellular uptake through M6P receptor in hATF-fused hI2S compared with rhI2S. On the other hand, cellular uptake of hATF-fused hI2S was higher than rhI2S in the presence of 10 mM M6P, which antagonized the binding of the ligand to M6P receptor. In the presence of 10 mM M6P, the uptake of rhI2S was blocked by more than 99%, whereas more than 20% of hATF-fused hI2S was taken up by fibroblast. This uptake of hATF-fused hI2S was blocked by further addition of 4.02 micrograms/mL human ATF.

The blockage of cellular uptake of hATF-fused hI2S and hI2S by addition of M6P was further analyzed as demonstrated in Fig. 7, where 20 micrograms/mL of hATF-fused hI2S or hI2S was added to the medium with various concentration of M6P for measurement of cellular uptake. In the presence of 3.33 mM M6P, the blocking efficacy of M6P appeared almost saturated, with more than 99 % of the uptake of rhI2S blocked, while more than 15% of hATF-fused hI2S was still taken up by the fibroblasts.

These data indicate that hATF-fused hI2S is able to be taken up by cells not only through M6P receptor but also in a M6P receptor-independent manner, possibly through binding of the hATF moiety to uPAR.

(Measurement of cellular uptake of mATF-fused hI2S using normal mouse fibroblast)
Measurement of cellular uptake of mATF-fused hI2S was conducted following the procedure described in Example 13 above, where primary mouse fibroblasts were substituted for human fibroblasts. The primary mouse fibroblasts were purchased from Kitayama Labes Co., Ltd., Japan.

Compared with rhI2S, the level of cellular uptake of mATF-fused hI2S into mouse fibroblasts was low (Fig 8). That may be because the number of M6P contained in mATF-fused hI2S (2.2 per protein molecule) was lower than that in rhI2S (4.4 per protein molecule), which lowered the efficiency of cellular uptake through M6P receptor in mATF-fused hI2S compared with rhI2S. On the other hand, cellular uptake of mATF-fused hI2S was higher than rhI2S in the presence of 10 mM M6P, which antagonized the binding of the ligand to M6P receptor. In the presence of 10 mM M6P, the uptake of rhI2S was blocked by about 91%, whereas more than 20% of mATF-fused hI2S was taken up by the fibroblasts.

The blockage of cellular uptake of mATF-fused hI2S and hI2S by addition of M6P was further analyzed as demonstrated in Fig. 9, where 20 micrograms/mL of mATF-fused hI2S or hI2S was added to the medium with various concentration of M6P for measurement of cellular uptake. In the presence of 3.33 mM M6P, the blocking efficacy of M6P appeared almost saturated, with more than 90 % of the uptake of rhI2S was blocked, while more than 20% of hATF-fused hI2S was still taken up by the fibroblasts.

(Analysis of Human I2S by ELISA)
To each well of a 96-well microtiter plate (Nunc) was added 100 microliters of mouse anti-human monoclonal antibody diluted to 4 micrograms/mL with 0.05 M Carbonate-Bicarbonate buffer (pH 9.6), and the plate was let stand for at least 1 hour at room temperature to let the antibody be absorbed by the wells. Then, after each well was washed three times with Phosphate Buffered Saline, pH 7.4 containing 0.05% Tween 20 (PBS-T), 200 microliters of Starting Block(PBS) Blocking Buffer (Thermo Fisher Scientific) was added, and the plate was let stand for at least 30 minutes at room temperature. Then, after each well was washed three times with PBS-T, 100 microliters of the test sample or human I2S standard, which had been diluted as desired with PBS containing 0.5% BSA and 0.05% Tween 20 (PBS-BT), was added to the well, and the plate was let stand for at least one hour at room temperature. Then, after each well was washed three times with PBS-T, 100 microliters of biotin-labeled anti-human I2S monoclonal antibody diluted with PBS-BT was added and the plate was let stand for at least 1 hour. Then, after each well was washed three times with PBS-T, 100 microliters of streptavidin-HRP(R&D SYSTEMS) diluted with PBS-BT was added and the plate was let stand for at least 30 minutes. Then, after each well was washed three times with PBS-T, 100 microliters of 0.4 mg/mL o-phenylendiamine with phosphate-citrate buffer (pH 5.0) was added to the well, and the plate was stand for 8 to 20 minutes at room temperature. Then 0.1 mL of 1 mol/L H₂SO₄ was added to each well to stop the reaction, and the optical density at 490 nm was measured for the well on a 96-well plate reader.

The present invention can be used as a means to deliver various pharmacologically active compounds to organs, tissues and cells where the urokinase-type plasminogen activator receptor (uPAR) is expressed. Thus, the present invention can also be used to provide various pharmaceutical agents which are directed to organs, tissues and cells where the receptor is expressed.

1 LacZ promoter
2 mPGKpromoter
3 Wild type murine encephalomyelitis virus IRES (EMCV-IRES)
3a Mutated type murine encephalomyelitis virus (EMCV-mIRES)
4 mPGK polyadenilation signal (mPGKpA)
5 Sequence containing EF-1 promoter and its first intoron
6 SV40 late polyadenilation signal
7 SV40 early promoter
8 Synthesized polyadenilation signal
9 Cytomegalovirus promoter
10 Glutamine synthetase gene
11 Sequence encoding mouse ATF
12 Human I2S gene
13 Sequence encoding human ATF

SEQ ID NO:1 = DNA sequence encoding IRES-Hygr-mPGKpA
SEQ ID NO:2 = IRES5'
SEQ ID NO:3 = IRES3'
SEQ ID NO:4 = mPGKP5'
SEQ ID NO:5 = mPGKP3'
SEQ ID NO:6 = DNA sequence containing mPGK promoter region
SEQ ID NO:7 = GS5'
SEQ ID NO:8 = GS3'
SEQ ID NO:9 = puro5'
SEQ ID NO:10 = puro3'
SEQ ID NO:11 = DNA sequence containing puromycin resistance gene
SEQ ID NO:12 = SV40polyA5'
SEQ ID NO:13 = SV40polyA3'
SEQ ID NO:14 = mIRES-GS5'
SEQ ID NO:15 = mIRES-GS3'
SEQ ID NO:16 = hATF-f
SEQ ID NO:17 = hATF-r
SEQ ID NO:18 = mATF-f
SEQ ID NO:19 = mATF-r
SEQ ID NO:20 = hI2S-f
SEQ ID NO:21 = hI2S-r
SEQ ID NO:22 = hI2S-f2
SEQ ID NO:23 = hI2S-r2
SEQ ID NO:24 = hI2S-f3
SEQ ID NO:25 = hI2S-r3
SEQ ID NO:26 = The DNA sequence of mATF-fused hI2S gene
SEQ ID NO:27 = Amino acid sequence of mATF-fused hI2S
SEQ ID NO:28 = The DNA sequence of hATF-fused hI2S gene
SEQ ID NO:29 = Amino acid sequence of hATF-fused hI2S
SEQ ID NO:30 = Amino acid sequence of of mATF
SEQ ID NO:31 = Amino acid sequence of receptor binding domain of mATF
SEQ ID NO:32 = Amino acid sequence of hATF
SEQ ID NO:33 = Amino acid sequence of receptor binding domain of hATF
SEQ ID NO:34 = Synthetic DNA sequence encoding hMPR9 having Bip signal at its N-terminus and His-tag at the C-terminus
SEQ ID NO:35 = hMPR9-f
SEQ ID NO:36 = hMPR9-r
SEQ ID NO:37 = MPR9-f2
SEQ ID NO:38 = MPR9-r2
SEQ ID NO:39 = MPR9-f3
SEQ ID NO:40 = MPR9-r3

Claims

A pharmaceutical agent comprising a pharmacologically active compound (A) and a peptide (B) fused thereto, wherein the peptide (B) comprises the receptor binding domain of ATF.
The pharmaceutical agent according to claim 1, wherein the pharmacologically active compound (A) is a protein.
The pharmaceutical agent according to claim 2, wherein the protein is an enzyme.
The pharmaceutical agent according to claim 3, wherein the enzyme is a lysosomal enzyme.
The pharmaceutical agent according to claim 4, wherein the lysosomal enzyme is selected from the group consisting of acid alpha-glucosidase, alpha-galactosidase, glucosylceramidase, alpha-L-iduronidase, alpha-L-iduronidase, iduronate-2-sulfatase, N-acetylgalactosamine-4-sulfatase, or acidic sphingomyelinase.
The pharmaceutical agent according to one of claims 1 to 5, wherein the peptide (B) is fused to the pharmacologically active compound (A) by an amide bond or an ester bond.
The pharmaceutical agent according to one of claims 1 to 5, wherein the peptide (B) is fused to the pharmacologically active compound (A) at the amino-terminal amino acid of the peptide (B)
The pharmaceutical agent according to one of claims 3 to 5, wherein the peptide (B) is fused to the protein by a peptide bond formed between the amino-terminal amino acid of the protein and the carboxy-terminal amino acid of the peptide (B).
The pharmaceutical agent according to one of claims 3 to 5, wherein the peptide (B) is fused to the protein by a peptide bond formed between the carboxy-terminal amino acid of the protein and the amino-terminal amino acid of the peptide (B).
The pharmaceutical agent according to one of claims 1 to 9, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ ID NO:33, or a derivative thereof in which 1 to 3 amino acids are added, substituted, deleted or inserted.
The pharmaceutical agent according to one of claims 1 to 9, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ ID NO:32, or a derivative thereof in which 1 to 10 amino acids are added, substituted, deleted or inserted.
The pharmaceutical agent according to claims 11, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ. ID NO:32, and wherein the peptide (B) is fused at the carboxy-terminal amino acid thereof to the pharmacologically active compound (A) through a linker consisting of at least one amino acid.
The pharmaceutical agent according to claim 12, wherein the linker is selected from the group consisting of -Lys-, Lys-Pro-, -Lys-Pro-Ser-, -Lys-Pro-Ser-Ser- (SEQ ID NO:41), and -Lys-Pro-Ser-Ser-Pro- (SEQ ID NO:42).
The pharmaceutical agent according to one of claims 1 to 9, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ ID NO:31, or a derivative thereof in which 1 to 5 amino acids are added, substituted, deleted or inserted.
The pharmaceutical agent according to one of claims 1 to 9, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ ID NO:30, or a derivative thereof in which 1 to 10 amino acids are added, substituted, deleted or inserted.
The pharmaceutical agent according to claim 15, wherein the peptide (B) consists of an amino acid sequence set forth as SEQ. ID NO:30 and wherein the peptide (B) is fused at the carboxy-terminal amino acid thereof to the pharmacologically active compound (A) through a linker consisting of at least one amino acid.
The pharmaceutical agent according to claim 16, wherein the linker is selected from the group consisting of -Lys-, Lys-Pro-, -Lys-Pro-Ser-, -Lys-Pro-Ser-Ser- (SEQ ID NO:41), and -Lys-Pro-Ser-Ser-Pro- (SEQ ID NO:42).
A pharmaceutical composition for enzyme replacement therapy comprising the pharmaceutical agent according to one of claims 1 to 17.
A DNA encoding a fusion protein comprising an pharmacologically active protein (A) and a peptide (B) fused thereto, wherein the peptide (B) comprises the receptor binding domain of ATF, and wherein the peptide (B) is fused to the pharmacologically active protein (A) at the amino- or carboxy-terminus of the peptide (B) by a peptide bond.
The DNA according to claim 19, wherein the pharmacologically active protein is an enzyme.
The DNA according to claim 20, wherein the enzyme is a lysosomal enzyme.
The pharmaceutical agent according to claim 21, wherein the lysosomal enzyme is selected from the group consisting of acid alpha-glucosidase, alpha-galactosidase, glucosylceramidase , alpha-L-iduronidase, alpha-L-iduronidase, iduronate-2-sulfatase, N-acetylgalactosamine-4-sulfatase, acidic sphingomyelinase.
An expression vector comprising the DNA according to one of claims 19 to 22.
A mammalian cell transformed with the vector according to claim 23.