USOO8323640B2

(12) United States Patent (10) Patent No.: US 8,323,640 B2 Sakuraba et al. (45) Date of Patent: *Dec. 4, 2012

(54) HIGHLY FUNCTIONAL HAVING OTHER PUBLICATIONS O-GALACTOSIDASE ACTIVITY Broun et al., Catalytic plasticity of fatty acid modification (75) Inventors: Hitoshi Sakuraba, Abiko (JP); Youichi underlying chemical diversity of plant lipids. Science, 1998, vol. 282: Tajima, Tokyo (JP); Mai Ito, Tokyo 1315-1317. (JP); Seiichi Aikawa, Tokyo (JP); Cameron ER. Recent advances in transgenic technology. 1997, vol. T: 253-265. Fumiko Aikawa, Tokyo (JP) Chica et al., Semi-rational approaches to engineering enzyme activ (73) Assignees: Tokyo Metropolitan Organization For ity: combining the benefits of directed evolution and rational design. Curr, opi. Biotechnol., 2005, vol. 16:378-384. Medical Research, Tokyo (JP); Altif Couzin et al. As Gelsinger case ends, therapy Suffers another Laboratories, Tokyo (JP) blow. Science, 2005, vol. 307: 1028. Devos et al., Practical Limits of Function prediction. Proteins: Struc (*) Notice: Subject to any disclaimer, the term of this ture, Function, and Genetics. 2000, vol. 41: 98-107. patent is extended or adjusted under 35 Donsante et al., AAV vector integration sites in mouse hapatocellular U.S.C. 154(b) by 0 days. carcinoma. Science, 2007. vol. 317: 477. Juengst ET., What next for human gene therapy? BMJ., 2003, vol. This patent is Subject to a terminal dis 326: 1410-1411. claimer. Kappel et al., Regulating in transgenic animals. Current Opinion in Biotechnology 1992, vol. 3: 548-553. (21) Appl. No.: 13/052,632 Kimmelman J. Recent developments in gene transfer: risk and eth ics. BMJ, 2005, vol. 350; 79-82. (22) Filed: Mar 21, 2011 Kisselev L., Polypeptide release factors in prokaryotes and eukaryotes: same function: same function, different structure. Struc (65) Prior Publication Data ture, 2002, vol. 10:8-9. Mullins et al., Transgenesis in the rat and larger mammals. J. Clin. US 2011 FO171198 A1 Jul. 14, 2011 Invest., 1996, vol. 97 (&); 1557-1560. Mullins et al., Transgenesis in nonmurine species. Hypertension, Related U.S. Application Data 1993, vol. 22 (4): 630-633. (62) Division of application No. 12/084,982, filed as Raper SE., Gene therapy: The good, the bad, and the ugly, Surgery, 2005, vol. 137 (5): 487-492. application No. PCT/JP2006/323509 on Nov. 17, Rosenberg et al., Gene therapist, healthy self. Science, 2000, vol. 2006, now Pat. No. 7,935,336. 287: 1751. Sen et al., Developments in directed evolution for improving enzyme (30) Foreign Application Priority Data functions. Appl. Biochem. Biotechnol., 2007, vol. 143: 212-233. Touchette et al., Gene therapy: Not ready for prime time. Nat Med., Nov. 18, 2005 (JP) ...... 2005-333660 1996, vol. 2 (1): 7-8. Whisstock et al., Prediction of protein function from protein sequence. Q. Rev. Biophysics., 2003, vol. 36 (3): 307-340. (51) Int. Cl. Wigley et al., Site-specific transgene insertion: an approach. Reprod. A6 IK38/43 (2006.01) Fert. Dev., 1994. vol. 6: 585-588. CI2N 9/40 (2006.01) Wishart et al. A single mutation converts a novel phosphotyrosine CI2N IS/00 (2006.01) binding domain into a dual-specificity phosphatase. J. Biol. Chem. 1995, vol. 270 (45): 26782-26785. CI2P2/06 (2006.01) Witkowski et al., Conversion of b-ketoacyl synthase to a Malonyl CI2P 19/34 (2006.01) Decarboxylase by replacement of the active cysteine with glutamine. C7H 2L/04 (2006.01) Biochemistry, 1999, vol. 38: 11643-1 1650. C07K L/00 (2006.01) Wolf JA. The “grand” problem of synthetic delivery. Nat. (52) U.S. Cl...... 424/94.1; 435/208; 435/69.1:435/91.1; Biotechnol., 2002. vol. 20:768-769. 435/320.1; 536/23.1536/23.2:530/350 (58) Field of Classification Search ...... 424/94.1; (Continued) 435/208, 69.1, 91.1, 320.1; 536/23.1, 23.2: Primary Examiner — Ganapathirama Raghu 530/350 (74) Attorney, Agent, or Firm — Westerman, Hattori, See application file for complete search history. Daniels & Adrian, LLP

(56) References Cited (57) ABSTRACT The present invention provides, as an enzyme which can be U.S. PATENT DOCUMENTS used for enzyme replacement therapy for Fabry disease, a 5,356,804 A 10, 1994 DeSnicket al. protein having O-galactosidase activity, which shows no 5,401,650 A 3, 1995 DeSnicket al. allergic adverse side effect, shows a high stability in blood, 6,887,696 B2 * 5/2005 Garger et al...... 435,208 and can be easily incorporated into a cell of an affected organ. 6,890,748 B2 * 5/2005 Garger et al...... 435,208 7,935,336 B2 * 5/2011 Sakuraba et al...... 424/94.61 The protein of the present invention is a protein which has 2003/OO77806 A1 4/2003 Selden et al. acquired C-galactosidase activity by changing the structure of 2010/0291059 A1* 11/2010 Sakuraba et al...... 424/94.61 the active site of wild-type human C.-N-acetylgalactosamini dase. FOREIGN PATENT DOCUMENTS JP 2001-504324 A 4/2001 12 Claims, 9 Drawing Sheets US 8,323,640 B2 Page 2

OTHER PUBLICATIONS Garman S.C. et al., “The 1.9 AStructure of alpha-N- Database UniProt Aug. 11, 2005, XP002512954 Database accession Acetylgalactosaminidase: Molecular Basis of Glycosidase Defi No. AEA27443. ciency Diseases'; Structure, vol. 10, No. 3, pp. 425-434, 2002. Garman S.C. et al., “The Molecular Defect Leading to Fabry Disease: Database UniProt, Dec. 1, 2001, XP0025 12955, Database accession Structure of Human alpha-galactosidase'; J. Mol. biol. vol. 337, No. No. P51569. 2, pp. 319-335, 2004. Supplemental European Search Report dated Feb. 17, 2009, issued in International Search Report of PCT/JP20061323509, date of maling corresponding European Patent Application No. 06833313.7. Feb. 13, 2007. Garman S.C. et al., “Structural basis of Fabry disease'; Molecular Genetics and , vol. 77, No. 1-2, pp. 3-11, 2002. * cited by examiner U.S. Patent Dec. 4, 2012 Sheet 1 of 9 US 8,323,640 B2

(31WHISGÎSHLIMX3idWOO)

U.S. Patent Dec. 4, 2012 Sheet 3 of 9 US 8,323,640 B2

!côzne

U.S. Patent Dec. 4, 2012 Sheet 5 Of 9 US 8,323,640 B2

G89N698NLOZ//|,N7Z||N.(SELISGTVIOI) SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSST\/0–?O4W N.(SELIS9TV|O|)G?ZNZ6||N69|| U.S. Patent Dec. 4, 2012 Sheet 6 of 9 US 8,323,640 B2

(3]*>{{SgnSHIMXHidWOO)

Typ-O?M t?lvalsansH}{MX3?dWOO) U.S. Patent US 8,323,640 B2

U.S. Patent Dec. 4, 2012 Sheet 9 Of 9 US 8,323,640 B2

US 8,323,640 B2 1. 2 HIGHLY FUNCTIONAL ENZYME HAVING from serious cases which are similar to those of hemizygotein O-GALACTOSIDASE ACTIVITY males to cases in which substantially no symptoms are observed exist. However, according to recent research, it has CROSS-REFERENCE TO RELATED become clear that most female heterozygote Fabry patients APPLICATIONS develop some symptoms when they become old. There is a viewpoint that they should be treated not as “carriers' but as This application is a Divisional of application Ser. No. “patients”. 12/084,982, filed May 14, 2008 now U.S. Pat. No. 7,935,336, Recently, enzyme replacement therapy for Fabry disease which is a U.S. National Stage of International Application has also been established, and a recombinant human O-galac PCT/JP2006/323509, filed on Nov. 17, 2006 and claiming 10 tosidase produced in a cell derived from mammals has been priority to Japanese Application No. 2005-333.660, filed on widely used as an active ingredient of a Fabry disease thera Nov. 18, 2005. peutic agent in the above therapy (refer to Eng C Metal. Am TECHNICAL FIELD J. Hum Genet, 68: 711–722 (2001); Eng C Met al., N Engl J 15 Med, 345: 9-16 (2001); and Schiffmann R et al., Proc Natl The present invention relates to a recombinant protein hav AcadSci USA,97: 365-370 (2000)). ing O-galactosidase activity. Furthermore, a method in which a recombinant human O-galactosidase produced using a cell (for example, yeast) BACKGROUND ART other than an animal cell as a host is used for the medical treatment (enzyme replacement therapy) of Fabry disease For hereditary enzyme deficiency, for which no radical (refer to Japanese Unexamined Patent Application Publica treatments have been known to date, enzyme replacement tion No. 2002-369692), a gene therapeutic method in which therapy in which an enzyme is produced by genetic engineer an enzyme is replaced by introducing a gene encoding human ing and is then administered in a blood vessel by intravenous O-galactosidase into a cell of an affected tissue to express the drip or the like has been developing. As an example of heredi 25 gene (refer to Japanese Unexamined Patent Application Pub tary enzyme deficiency whose prevalence is relatively high lication (Translation of PCT Application) No. 2002-522509), and which is designated as a specified disease (intractable and the like have also been proposed. disease), Fabry disease (hereditary C.-galactosidase defi ciency, which is one of a group of genetic diseases and also DISCLOSURE OF INVENTION called lysosomal disease), is well known (refer to Kenneth J. 30 Dean et al., Fabry Disease, “Practical Enzymology of the However, since an existing enzyme agent used in enzyme Sphingolipidoses”, U.S.A., Aln R. Liss, Inc., 1997, p. 173 replacement for treating Fabry disease is often administered 216). to patients who originally do not have an enzyme (human Fabry disease is a glycolipid metabolic disorder which O-galactosidase), the enzyme contained in the therapeutic develops as follows: As a result of a decrease in the activity of 35 agent is recognized as a foreign Substance in many patients an enzyme called "O-galactosidase, which is one of the administered with the enzyme agent, and thus, an antibody is enzymes present in a lysosome, which is one of the human produced. As a result, adverse side effects, mainly, allergic intracellular organelles, and deficiency of the enzyme, a gly reactions are expressed at a high frequency. This similarly colipid called globotriaosylceramide (also referred to as cera occurs in the case where the enzyme is replaced using a mide trihexoside), which is an in vivo substrate of the 40 method of gene therapy. enzyme, is not decomposed and accumulated in the body (for In addition, such an enzyme agent used in enzyme replace example, blood vessels, skins, cornea, nerves, kidneys, and ment is administered in blood vessels, but C-galactosidase heart). itself is unstable in blood. Accordingly, in the actual therapy, Since a gene encoding O-galactosidase lies on the X chro the enzyme agent must be frequently administered (once mosome, this disease has a mode of X-chromosomal inherit 45 every two weeks), and it may be necessary to increase the ance. Therefore, in this disease, a definite clinical feature is dosage per administration. Furthermore, human C-galactosi observed mainly in hemizygote in males. It is believed that dase has a relatively small number of Sugar chains (N-type “classic Fabry disease', which takes a typical clinical course, Sugar chains) to which mannose-6-phosphate (M6P) residue develops in about one out of 40,000 male children. Symptoms can be bonded, the Sugar chains being necessary for human Such as pain in the hand and the foot, hypohidrosis, angiok 50 O-galactosidase to be incorporated into a cell (more specifi eratoma, and corneal opacity appear during the juvenile term cally, into a lysosome in a cell) in an affected organ. There and adolescence; these symptoms progress and then cause fore, it is difficult for human C.-galactosidase to be taken from systemic organ damage Such as renal failure, heart failure, blood and incorporated into a cell. In particular, the incorpo and cerebrovascular disorder during middlescence and there ration efficiency in the kidney or heart, which is the main after, and these become the cause of death. In addition, a 55 affected organ in Fabry disease, is low, and thus the therapeu disease which does not take Such a typical clinical course as tic effect for nephropathy or cardiopathy is insufficient. “classic Fabry disease' and which develops late and takes a Accordingly, in order to allow a certain amount of enzyme to relatively moderate course, is “variant Fabry disease’. In be incorporated into a target cell in therapy, a large amount of patients having this type of disease, residual O-galactosidase enzyme is required. Consequently, it is necessary to admin activity is observed though it is low. As a variant Fabry dis 60 ister an enzyme agent used in enzyme replacement more ease, for example, “cardiac Fabry disease' is known. The frequently and in a larger amount. Such therapy places a large above-mentioned glycolipid accumulation mainly occurs in burden on patients physically, mentally, and economically, the heart. Thereby, cardiac hypertrophy occurs, and disorders and thus adversely affects the “quality of life (QOL)’. such as heart failure and arrhythmia are caused. On the other Accordingly, it is an object of the present invention to hand, in female heterozygote Fabry patients, various types of 65 provide, as an enzyme which can be used for enzyme replace clinical features are observed in accordance with the charac ment therapy for Fabry disease, a protein having O-galactosi teristics of the X . Specifically, cases ranging dase activity, which shows no allergic adverse side effect, US 8,323,640 B2 3 4 shows a high stability in blood (in plasma), and can be easily An example of the protein is a protein wherein the amino incorporated into a cell of an affected organ. acid other than serine is glutamic acid or aspartic acid. The present inventors conducted intensive studies in order In addition, another example of the protein is a protein to solve the above problems. As a result, the present inventors wherein the amino acid other than alanine is one selected focused on “C-N-acetylgalactosaminidase (C-NAGA), from the group consisting of leucine, Valine, isoleucine, phe which has a substrate specificity different from that of C-ga nylalanine, and methionine. lactosidase but which has a three-dimensional structure very Furthermore, another example of the protein is a protein similar to that of O-galactosidase as a whole. The present wherein the amino acid other than serine is glutamic acid, and inventors have found that when the substrate specificity of the amino acid other than alanine is leucine. C-NAGA is converted so as to have C-galactosidase activity 10 (4) A gene encoding the protein according to any one of items by changing the structure of the active site of C-NAGA using (1) to (3). a gene recombination technique, an excellent novel highly (5) A gene containing DNA described by (a) or (b): functional enzyme for treating Fabry disease that can be used (a) DNA containing any one of base sequences described to solve the above problems can be created. This finding 15 by (i) to (iii): resulted in completion of the present invention. (i) a base sequence composed of the 52nd to 1.236th bases More specifically, the present invention is as follows: included in a base sequence in which the 562nd to 564th bases (1) A protein which has acquired C-galactosidase activity by are replaced with bases representing a codon of an amino acid changing the structure of the active site of wild-type human other than serine in the base sequence shown in sequence No. Cl-N-acetylgalactosaminidase. 1; An example of the protein is a protein having the Substrate (ii) a base sequence composed of the 52nd to 1.236th bases specificity of C-galactosidase. included in a base sequence in which the 571st to 573rd bases (2) A protein having O-galactosidase activity, the protein are replaced with bases representing a codon of an amino acid being composed of an amino-acid sequence in which at other than alanine in the base sequence shown in sequence least one of the 188th amino acid and the 191st amino acid 25 No. 1; and in the amino-acid sequence of wild-type human Cl-N- (iii) a base sequence composed of the 52nd to 1.236th bases acetylgalactosaminidase is replaced with another amino included in a base sequence in which the 562nd to 564th bases acid, or an amino-acid sequence in which one or several are replaced with bases representing a codon of an amino acid amino acids except the 188th amino acid and the 191st other than serine and the 571st to 573rd bases are replaced amino acid included in the replaced amino-acid sequence 30 are deleted, replaced, or added. with bases representing a codon of an amino acid other than An example of the protein is a protein wherein the 188th alanine in the base sequence shown in sequence No. 1; or amino acid is replaced with glutamic acid or aspartic acid in (b) DNA which encodes a protein having O-galactosidase the replacement with another amino acid. activity and which hybridizes with DNA composed of a base In addition, another example of the protein is a protein 35 sequence complementary to DNA containing any one of base wherein the 191st amino acid is replaced with one selected sequences described by (i) to (iii) under a stringent condition, from the group consisting of leucine, Valine, isoleucine, phe wherein bases corresponding to the bases at the replacement nylalanine, and methionine in the replacement with another sites are identical to the bases at the replacement sites. amino acid. An example of the gene is a gene wherein the amino acid Furthermore, another example of the protein is a protein 40 other than serine is glutamic acid or aspartic acid. wherein the 188th amino acid is replaced with glutamic acid, In addition, another example of the gene is a gene wherein and the 191st amino acid is replaced with leucine in the the amino acid other than alanine is one selected from the replacement with another amino acid. group consisting of leucine, Valine, isoleucine, phenylala (3) A protein described by (a) or (b): nine, and methionine. (a) a protein containing any one of amino-acid sequences 45 Furthermore, another example of the gene is a gene described by (i) to (iii): wherein the amino acid other than serine is glutamic acid, and (i) an amino-acid sequence composed of the 18th amino the amino acid other than alanine is leucine. acid to the 411th amino acid included in an amino-acid (6) A recombinant vector containing the gene according to sequence in which the 188th amino acid is replaced with an item (4) or (5). amino acid other than serine in the amino-acid sequence 50 (7) A transformant containing the recombinant vector accord shown in sequence No. 2; ing to item (6). (ii) an amino-acid sequence composed of the 18th amino (8) A method of producing a protein having O-galactosidase acid to the 411th amino acid included in an amino-acid activity, wherein the structure of the active site of wild-type sequence in which the 191st amino acid is replaced with an human C.-N-acetylgalactosaminidase is changed so that a amino acid other than alanine in the amino-acid sequence 55 Substrate of C-galactosidase can be bound to the active site. shown in sequence No. 2; and (9) A method of producing a protein having O-galactosidase (iii) an amino-acid sequence composed of the 18th amino activity including a step of culturing the transformant acid to the 411th amino acid included in an amino-acid according to item (7), and a step of collecting the protein sequence in which the 188th amino acid is replaced with an having O-galactosidase activity from the resulting cultured amino acid other than serine and the 191st amino acid is 60 product. replaced with an amino acid other than alanine in the amino (10) A pharmaceutical composition containing the protein acid sequence shown in sequence No. 2; or according to any one of items (1) to (3). (b) a protein containing an amino-acid sequence in which (11) A therapeutic agent for Fabry disease containing the one or several amino acids other than the amino acid oramino pharmaceutical composition according to item (10) as an acids located at the replacement site or sites are deleted, 65 active ingredient. replaced, or added in any one of amino-acid sequences (12) A pharmaceutical composition containing the gene described by (i) to (iii), and having O-galactosidase activity. according to item (4) or (5). US 8,323,640 B2 5 6 (13) A genetherapeutic agent for Fabry disease containing the 1. Summary of the present invention pharmaceutical composition according to item (12) as an The present invention provides a recombinant protein serv active ingredient. ing as an excellent novel highly functional enzyme which can (14) A method of treating Fabry disease, wherein the phar be used for enzyme replacement therapy for Fabry disease. maceutical composition according to item (10) or (12) is For existing enzyme agents used in enzyme replacement administered to a Fabry patient. for treating Fabry disease, recombinant human C-galactosi dase produced in a cell derived from mammals, such as a BRIEF DESCRIPTION OF DRAWINGS (Chinese hamster ovary) CHO cell or a human fibroblast, is used. However, the use of the recombinant human O-galac FIG. 1 includes schematic views showing the three-dimen 10 tosidase causes problems such as allergic adverse side effects, sional overall structures of a subunit of wild-type C-GAL and instability in blood, and low incorporation efficiency into a a subunit of wild-type C-NAGA. cell of an affected organ, and thus places a very large burden FIG. 2A includes schematic views showing the structures on patients in the actual therapy. Accordingly, a solution to of the active site of wild-type C-GAL and the active site of these problems has been desired. wild-type C-NAGA. Note that the amino acids (shown by a 15 In order to solve these problems, the present inventors Stick model (except a Substrate)) shown in the figure are the studied whether or not an enzyme other than O-galactosidase same type of amino acids in the case where the amino-acid (C-GAL) can be used as an enzyme used in enzyme replace sequence of the active site of wild-type C-GAL is compared ment for treating Fabry disease. Specifically, the present with that of wild-type C-NAGA. inventors focused on "C-N-acetylgalactosaminidase FIG. 2B includes schematic views showing the structures (C.-NAGA), which is a lysosomal enzyme similarly to of the active site of wild-type C-GAL and the active site of C-GAL (that is, the localization of C.-NAGA in a cell being wild-type C-NAGA. Note that the amino acids (shown by a the same as that of O-GAL), which has a Substrate specificity Stick model (except a Substrate)) shown in the figure are different from that of O-GAL, but which has a three-dimen different types of amino acids in the case where the amino sional structure very similar to that of O-GAL as a whole. acid sequence of the active site of wild-type C-GAL is com 25 C-GAL used to be called C-galactosidase A. It was pared with that of wild-type C-NAGA. believed that an isozyme called C-galactosidase B having FIG. 2C includes schematic views showing amino acids biochemical properties which are very similar to those of constituting the active site of wild-type C-GAL and amino C-GAL existed. It was known that C-galactosidase B had acids constituting the active site of wild-type C-NAGA, and higher stability than that of C.-GAL, but did not have the interaction sites of the amino acids with a substrate. 30 ability to decompose globotriaosylceramide, which is accu FIG. 3 includes schematic views showing the comparison mulated in the body by Fabry disease. Afterward, it became between the number of N-glycosylation sites and locations clear that C.-galactosidase B is actually Cl-N-acetylgalac thereof in the subunit of wild-type C-GAL and those in the tosaminidase (C.-NAGA). It is known that C.-NAGA is subunit of wild-type O.-NAGA. encoded by a gene which is considered to be derived from the FIG. 4 includes schematic views in which the N-glycosy 35 same ancestor gene as that of a gene encoding O-GAL. The lation sites (shown by a stick model (except a Substrate)) in cDNA thereof has been cloned, and it is known that the gene the subunits of wild-type C-GAL and wild-type C-NAGA are encodes a protein composed of 411 amino acid residues con shown in the three-dimensional structures of the subunits. taining a signal peptide composed of 17 amino acid residues. FIG. 5 is a graph showing the results obtained by compar In addition, when the structure of human C.-NAGA is com ing the stability of O-GAL in blood to that of an O-NAGA 40 pared with that of human C-GAL, the structure of human mutant with time on the basis of the residual ratio of O-GAL C.-NAGA has the homology of 57.9% in terms of the base activity. sequence level, and 52.2% in terms of the amino-acid FIG. 6(a) is a schematic view showing a state in which sequence level. Furthermore, as in human C-GAL, human wild-type C-NAGA is bonded to a substrate of C.-NAGA, and C.-NAGA is an enzyme that exists in the form of a homodimer. FIG. 6(b) is a schematic view showing a state in which 45 On the basis of the above knowledge, the present inventors C.-NAGA(S188E/A191L), which is an O-NAGA mutant, is first constructed three-dimensional structural models of bonded to a substrate of O-GAL. C.-NAGA and C.-GAL, and compared the structures. More FIG. 7 includes schematic views showing the structures of specifically, a three-dimensional structural model of human the active site of wild-type C-NAGA and the active site of O-NAGA was constructed with reference to the structural C-NAGA(S188E/A191L), which is an O-NAGA mutant. 50 information of chicken C-NAGA (ID: 1KTC) registered in Protein Data Bank (PDB www.rcsb.org/pdb/)), and this struc BEST MODES FOR CARRYING OUT THE ture was compared with the three-dimensional structure of INVENTION human C-GAL (ID: 1R47) registered in PDB. As a result, it was found that the three-dimensional structure of human The present invention will now be described in detail, but 55 C.-NAGA was very similar to the three-dimensional structure the scope of the present invention is not restricted by the of human O-GAL in terms of both the entire structure and the descriptions below. In addition to exemplifications described active site. Regarding the active site, to be exact, only a few below, various modifications can be made to the present amino acid residues are different from each other. However, invention without departing from the purpose of the present among these amino acid residues, there are important amino invention. 60 acid residues which are presentina Substrate-binding site and This description includes the entirety of the specification of which affect the difference between the substrate specificity Japanese Patent Application No. 2005-333.660, which claims of O-GAL and the substrate specificity of C-NAGA. It was the priority of this application. In addition, all publications, found that, in this regard, there is a significant difference in for example, prior art documents, unexamined patent appli the three-dimensional structure between the active site of cation publications, patent publications, and other patent 65 C-GAL and the active site of C-NAGA. documents cited in this description are incorporated in this Thus, C-NAGA is an enzyme which differs from C-GAL in description as references. the structure of a part of the substrate-binding site in the active US 8,323,640 B2 7 8 site but is very similar to O-GAL in terms of the structure and chains per subunit (three locations of Asn139, Asn192, and in terms of properties regarding the other parts including the ASn215; six Sugar chains in the case where a dimer is formed) catalytic site (refer to FIGS. 1, 2A, and 2B). Therefore, the are present. In contrast, in C-NAGA, five Sugar chains per catalytic reaction mechanism of C.-NAGA is very similar to Subunit (ten Sugar chains in the case where a dimer is formed) the catalytic reaction mechanism of O-GAL in terms of, for 5 are present (refer to FIGS. 3 and 4). Among these sugar example, the types of reaction Substrate used and reaction chains, three Sugar chains (three locations of Asn124. product produced. Asn177, and Asn201) correspond to the locations of the sugar Consequently, as described above, the present inventors chains in C-GAL, and two other Sugar chains (two locations focused on C.-NAGA and found that when the substrate speci of Asn359 and Asn385) are specific to C.-NAGA. Accord ficity of C-NAGA is modified so as to have O-galactosidase 10 ingly, it is believed that C.-NAGA is taken from blood and activity by changing the structure of the active site (in par incorporated into a lysosome in a cell of an affected organ at ticular, the Substrate-binding site) by gene manipulation of a higher efficiency than that in the case of O-GAL. C-NAGA (for example, when, among amino acid residues From the above standpoints, the present inventors focused related to the substrate recognition of C-NAGA, key amino on the 188th amino acid (serine (Ser)) and the 191st amino acid residues are changed from C-NAGA-type amino acid 15 acid (Ala (alanine)) among an amino-acid residue group con residues to O-GAL-type amino acid residues), a novel excel stituting the substrate-binding site of C-NAGA. The present lent highly functional enzyme for treating Fabry disease can inventors prepared a recombinant enzyme (mutant enzyme) be created. in which the 188th serine (Ser) is replaced with glutamic acid The reasons why the present inventors focused on (Glu) and the 191st alanine is replaced with leucine (Leu) C.-NAGA further include the following points (i) to (iii): (refer to Examples 1 and 2). Subsequently, this recombinant (i) C-NAGA is the responsible enzyme of Shindler disease enzyme was expressed using a fibroblast derived from a Fabry and Kanzaki disease (note that a disease that develops due to patient, collected, and analyzed. As a result, high C-GAL abnormality of the same enzyme as an enzyme that develops activity was observed (refer to Example 2). In addition, the Shindler disease and that has a clinical phenotype different stability of this recombinant enzyme in blood was signifi from that of Shindler disease is called Kanzaki disease), and 25 cantly higher than that of wild-type O-GAL (refer to Example deficiency of C.-NAGA is a cause of the development of 3 and FIG. 5). By using Such a recombinant enzyme having a Shindler disease and Kanzaki disease. In general, however, modified Substrate specificity, an enzyme agent used in the development of Shindler disease and Kanzaki disease is enzyme replacement for treating Fabry Disease which is very rare even in Fabry patients. Accordingly, almost all Superior to existing enzyme agents can be provided. Fabry patients have C-NAGA normally. Therefore, it is 30 Note that, in this description, unless otherwise stated, the believed that even when a protein in which only the substrate terms 'O-galactosidase' and "O-GAL' mean “human C-ga specificity of C.-NAGA is modified into the substrate speci lactosidase A, and the terms "C-N-acetylgalactosaminidase ficity of O-GAL is administered as an enzyme agent used in and “O-NAGA' mean “human C.-N-acetylgalactosamini enzyme replacement, the antigenicity thereof rarely appears dase'. The orders “the 188th' and “the 191st” represent loca as in the case where wild-type C-NAGA is administered, and 35 tions based on the amino-acid sequence of C-NAGA shown in thus there is substantially no probability that an adverse sequence No. 2. immune reaction Such as an allergic side effect is induced. 2. Protein (ii) C.-NAGA also functions in the form of a homodimer A protein of the present invention is a mutant enzyme of similarly to O-GAL, but in general, the stability in the dimer C.-N-acetylgalactosaminidase (C-NAGA). More specifically, form of C-NAGA is higher than that of O-GAL. The three 40 the protein of the present invention is a protein which has dimensional structural model constructed by the present acquired C-galactosidase (C-GAL) activity by changing the inventors also supports this stability. Specifically, it was con structure of the active site (in particular, the Substrate-binding firmed that, in the dimer of human C.-NAGA, two bonds due site) of wild-type C-NAGA, and preferably, a protein having to electrostatic interaction were observed between Asp45 and the substrate specificity of C.-GAL. Arg350 in two subunits, whereas such bonds were not 45 Herein, the phrase “has acquired C-GAL activity” means observed in C.-GAL. Accordingly, it is believed that, as in that, in the substrate-binding site of C-NAGA, the binding C.-NAGA, an O-NAGA mutant also has high stability in blood reactivity to a substrate of C.-GAL becomes relatively higher (in plasma), compared with C-GAL, and is very Suitable for than the binding reactivity to a substrate of C.-NAGA. Accord enzyme replacement therapy. In addition, if the existing ingly, the above structural change is not limited to a structural dimer proportion is increased because of the above stability, it 50 change which makes it impossible for C-NAGA to bind with is expected that the incorporation efficiency into a lysosome the substrate of C-NAGA. Alternatively, the structural change in a cell is also improved in relation to point (iii) below. may be a structural change that makes the binding reactivity Furthermore, it is advantageous in that, before administra to the substrate of C-GAL significantly higher than the bind tion, the effect can be maintained as an enzyme preparation ing reactivity to the substrate of C.-NAGA, which has origi for a long period. 55 nally been relatively significantly higher than the binding (iii) It is necessary that an enzyme used in enzyme replace reactivity to the substrate of O-GAL. Furthermore, the phrase ment therapy be incorporated into a lysosome in a cell of an “having the substrate specificity of O-GAL' means that the affected organ. In general, the transportation from a cell structure (in particular, the positions and the types of amino membrane to a lysosome is performed via a calcium-indepen acid residues which play an important role in the binding dent M6P receptor, which recognizes mannose-6-phosphate 60 reactivity to a substrate) of the active site of the protein is the (M6P) present in Sugar chain portions of the enzyme. Accord same as that of O-GAL. ingly, an enzyme having a large number of Sugar chains In the present invention, the term “substrate of O-GAL (N-type sugar chains) to which the residue of M6P residue means a natural compound Such as a glycolipid, e.g., globot can be bonded is preferable because a high incorporation riaosylceramide, which has a galactose residue bound to the efficiency into a lysosome can be achieved. Regarding the 65 non-reducing end by C-bonding, or a synthetic compound, number of the above Sugar chains, it has become clear that, e.g., 4-methylumbelliferyl-C-D-galactoside. The term “sub from X-ray crystal structure analysis, in O-GAL, three Sugar strate of C-NAGA' means a natural compound Such as an US 8,323,640 B2 10 oligosaccharide, , or glycolipid having an the amino-acid sequence (sequence No. 13) of the Subunit of N-acetylgalactosamine residue bound to the non-reducing wild-type C-GAL (homodimer) and information on the base end by C-bonding, or a synthetic compound, e.g., 4-methy sequence (sequence No. 12) encoding the above amino-acid lumbelliferyl-C-N-acetyl-D-galactosaminide. sequence are published, for example, as “accession number: Here, a catalytic reaction of wild-type C-GAL is shown by NP 000160 in GenBank, and registered as “entry name: reaction formula (1), and a catalytic reaction of wild-type AGAL HUMAN, accession number: PO6280" in Swiss C.-NAGA is shown by reaction formula (2). Prot (available from tw.expasy.org/uniprot/). Herein, examples of the above "amino-acid sequence in which one or several amino acids are deleted, replaced, or (1) 10 added preferably include amino-acid sequences in which RI about one to ten amino acids, and preferably about one to five 5 O amino acids are deleted, replaced, or added. 6 4 Furthermore, regarding “the protein composed of an HOHC O- 1 No:OH OH amino-acid sequence in which one or several amino acids are 15 deleted, replaced, or added, it is important that the protein can stably exhibit C.-GAL activity. Therefore, for example, all GA. of or some of (preferably, all of) the 28th to 31 stamino acids, OH the 77th to 81stamino acids, the 117th to 127th amino acids, HOHC OH the 150th to 158th amino acids, the 192nd amino acid, the O- OH + ROH 209th to 220th amino acids, and the 242nd to 254th amino OH acids (in particular, the 156th and 217th aspartic acids (Asp: D)), all of which are believed to be important for the binding In formula (1), when the Substrate is a natural compound, performance (Substrate-binding performance) with an O-ga R" represents "a group derived from a sugar complex”, and lactose residue in a substrate of O-GAL and the catalytic when the substrate is a synthetic compound, R' represents “a 25 reactivity to the substrate; the 45th aspartic acid (Asp: D) and 4-methylumbelliferyl group'. the 350th arginine (Arg: R), both of which are believed to be important for forming a homodimer; and the 124th, 177th, 201st, 359th, and 385th amino acids (all of which being (2) asparagine (ASn: N)), all of which are N-type-Sugar-chain 30 binding sites, are preferably amino acids which are not mutated (deleted, replaced, or added) from the amino-acid sequence of wild-type C-NAGA. Regarding the 188th amino acid residue, the other alterna tive amino acid residue is not particularly limited as long as 35 the amino acid residue is not serine (Ser: S). For example, glutamic acid (Glu: E) and aspartic acid (Asp: D) are prefer able, and glutamic acid is more preferable. Similarly, regard NAGA ing the 191st amino acid residue, the other alternative amino OH acid residue is not particularly limited as long as the amino HOHC OH 40 O- NH CH + ROH acid residue is not alanine (Ala: A). For example, leucine OH (Leu: L), valine (Val: V), isoleucine (Ile: I), phenylalanine r (Phe: F), and methionine (Met: M) are preferable, and leucine O is more preferable. Particularly preferably, among these, as the alternative amino acids, the 188th amino acid is glutamic In formula (2), when the Substrate is a natural compound, 45 acid and the 191 stamino acid is leucine. Note that, preferably, R represents "a group derived from a sugar complex”, and the amino acids after the replacement do not substantially when the substrate is a synthetic compound, R represents “a affect the structure composed of otheramino acids which are 4-methylumbelliferyl group'. not replaced. From this point of view, in a particularly pref Examples of the protein of the present invention preferably erable replacement embodiment, the 188th amino acid resi include proteins which are composed of an amino-acid 50 due is glutamic acid and the 191st amino acid residue is sequence in which at least one of the 188th amino acid and the leucine. 191st amino acid included in the amino-acid sequence of By replacing the 188th amino acid and the 191st amino wild-type C-NAGA is replaced with another amino acid acid, both of which are present in the Substrate-binding site, as (more preferably, an amino-acid sequence in which both the described above, the following effects can be achieved. Spe 188th amino acid and the 191st amino acid are replaced with 55 cifically, as exemplified in FIG. 6, regarding the 188th amino otheramino acids) oranamino-acid sequence in which one or acid residue, the interaction with the N-acetyl group (in par several amino acids except the 188th amino acid and the 191st ticular, the oxygen atom) in the substrate of C-NAGA can be amino acid included in the above replaced amino-acid removed, and in addition, a binding action with the hydroxyl sequence are deleted, replaced, or added and which have group in the Substrate of O-GAL can be generated. Regarding C-GAL activity. Information on the amino-acid sequence 60 the 191stamino acid residue, the interaction with the N-acetyl (sequence No. 2) of the subunit of wild-type C-NAGA (ho group (in particular, the methyl group) in the Substrate of modimer) and information on the base sequence (sequence C-NAGA can be removed, and in addition, the binding space No. 1) encoding the above amino-acid sequence are pub of the substrate (in particular, the space into which the lished, for example, as “accession number: NM 000262 in N-acetyl group is taken) can be restricted. As a result, the GenBank, and registered as “entry name: NAGAB 65 recombinant enzyme (recombinant protein) obtained after the HUMAN, accession number: P17050" in Swiss-Prot (avail replacement of amino acids has a binding reactivity to the able from tw.expasy.org/uniprot/). Similarly, information on substrate of C.-GAL higher than the binding reactivity to the US 8,323,640 B2 11 12 Substrate of C-NAGA, and thus can be an enzyme having sequence No. 13. The signal peptide of preprotrypsin is a C-GAL activity significantly higher than O-NAGA activity. peptide composed of the amino-acid sequence shown in At least the recombinant enzyme having an amino-acid sequence No. 15. sequence in which the 188th amino acid (serine) is replaced As the protein described by (a), among the proteins con with glutamic acid and the 191st amino acid (alanine) is taining an amino-acid sequence described by (i), (ii), or (iii), replaced with leucine is particularly preferable from the a protein containing the amino-acid sequence described by standpoint that the above-described effects can be satisfacto (iii) is particularly preferable. rily achieved. A preferable example of the protein described by (a) is a In addition, the protein of the present invention is prefer protein in which “the amino acid other than serine' described 10 in (i) and (iii) is glutamic acid or aspartic acid. Similarly, ably a protein described by (a) or (b): another preferable example of the protein described by (a) is (a) a protein containing any one of the amino-acid a protein in which “the amino acid other than alanine' sequences described by (i) to (iii): described in (ii) and (iii) is one selected from the group (i) an amino-acid sequence composed of the 18th amino consisting of leucine, Valine, isoleucine, phenylalanine, and acid to the 411th amino acid included in an amino-acid 15 methionine. sequence in which the 188th amino acid is replaced with an Furthermore, a particularly preferable example of the pro amino acid other than serine in the amino-acid sequence tein described by (a) is a protein in which “the amino acid shown in sequence No. 2; other than serine' described in (i) to (iii) is glutamic acid and (ii) an amino-acid sequence composed of the 18th amino “the amino acid other than alanine' described in (i) to (iii) is acid to the 411th amino acid included in an amino-acid leucine. A preferable example of the protein is a protein sequence in which the 191st amino acid is replaced with an (C.-NAGA(S188E/A191L)) in which, in the amino-acid amino acid other than alanine in the amino-acid sequence sequence of wild-type C-NAGA (sequence No. 2), the 188th shown in sequence No. 2; and serine is replaced with glutamic acid and the 191st alanine is (iii) an amino-acid sequence composed of the 18th amino replaced with leucine (refer to FIG. 7 and sequence No. 4). In acid to the 411th amino acid included in an amino-acid 25 general, regarding the alphabetical notation of amino acids, sequence in which the 188th amino acid is replaced with an an amino acid is denoted by three letters (e.g., “Ser') or one amino acid other than serine and the 191st amino acid is letter (e.g., “S”). The letter of the alphabet located before a replaced with an amino acid other than alanine in the amino number (e.g., "188”) representing the location of an amino acid sequence shown in sequence No. 2; or acid from the N-terminal represents one-latter notation of an (b) a protein containing an amino-acid sequence in which 30 amino acid before replacement, and the letter of the alphabet located after the number represents one-letter notation of an one or several amino acids other than the amino acid oramino amino acid after replacement. Accordingly, the notation acids located at the replacement site or sites are deleted, “S188E represents the case where the 188th Ser is replaced replaced, or added in any one of the amino-acid sequences with Glu. described by (i) to (iii), and having O-galactosidase activity. 35 The protein described by (b) is not limited as long as the The amino-acid sequence shown in sequence No. 2 is an protein contains an amino-acid sequence in which one or amino-acid sequence composed of 411 amino acids consti several (for example, about one to ten, and preferably, about tuting wild-type C-NAGA. one to five) amino acids other than the amino acid or amino Specifically, the protein described by (a) is a protein com acids located at the replacement site or sites are deleted, posed of an amino-acid sequence containing an amino-acid 40 replaced, or added in any one of amino-acid sequences sequence ranging from the 18th amino acid to the 411th described by (i) to (iii) included in the protein described by amino acid excluding the 1st amino acid to the 17th amino (a), and has C-galactosidase activity. Herein, the term “the acid, which constitute the signal peptide of wild-type replacement site' means, among the 394 amino acid residues C-NAGA, included in an amino-acid sequence in which at constituting the amino-acid sequences described by (i) to (iii), least one amino acid is replaced as described in (i) to (iii) in 45 the amino acid residue corresponding to the location of the the amino-acid sequence shown in sequence No. 2. As 188th amino acid in the amino-acid sequence shown in described above, each of the 188th amino acid residue and the sequence No. 2 (however, the amino-acid sequence being 191st amino acid residue is one of amino acids constituting limited to the amino-acid sequence described by (i) or (iii)), the Substrate-binding site. and the amino acid residue corresponding to the location of Here, a preferable example of the amino-acid sequence 50 the 191st amino acid in the amino-acid sequence shown in containing an amino-acid sequence ranging from the 18th sequence No. 2 (however, the amino-acid sequence being amino acid to the 411thamino acid is an amino-acid sequence limited to the amino-acid sequence described by (ii) or (iii)). in which a signal peptide is bound to the N-terminal of the More specifically, the former amino acid residue means the amino-acid sequence ranging from the 18th amino acid to the 171st amino acid residue in the amino-acid sequence 411th amino acid. The signal peptide is not limited as long as 55 described by (i) or (iii), and the latter amino acid residue the signal peptide can allow the protein to pass through a cell means the 174th amino acid residue in the amino-acid membrane of an affected organ. For example, a signal peptide sequence described by (ii) or (iii). of a lysosomal enzyme such as wild-type C-NAGA or wild Note that it is important that the protein described by (b) is type O-GAL, or a signal peptide of a secretase Such as pre a protein that can stably exhibit C.-GAL activity. Therefore, protrypsin is preferable. The signal peptide of wild-type 60 for example, amino acid residues which are believe to be C.-NAGA or wild-type C-GAL is more preferable. The signal important forbinding performance (Substrate-binding perfor peptide of wild-type C-NAGA is a peptide composed of the mance) with an O-galactose residue in a substrate of O-GAL 1st amino acid to the 17th amino acid included in the amino and the catalytic reactivity to the substrate are preferably acid sequence of wild-type C-NAGA shown in sequence No. amino acid residues which are not mutated (deleted, replaced, 2. The signal peptide of wild-type C-GAL is a peptide com 65 or added) from the amino-acid sequences described by (i) to posed of the 1st amino acid to the 31st amino acid included in (iii). Preferable examples of the amino acid residues include, the amino-acid sequence of wild-type C-GAL shown in among the amino acid residues constituting the amino-acid US 8,323,640 B2 13 14 sequences described by (i) to (iii), amino acid residues cor part thereof and may further contain other known base responding to the locations of the 28th to 31st amino acids, sequences (such as a transcriptional promoter, the SD the 77th to 81stamino acids, the 117th to 127th amino acids, sequence, the Kozak sequence, and a terminator) required for the 150th to 158th amino acids, the 192nd amino acid, the gene expression. Thus, the gene is not limited thereto. 209th to 220th amino acids, and the 242nd to 254th amino (a) DNA containing any one of base sequences described acids (in particular, the 156th and 217th aspartic acids (Asp: by (i) to (iii); D)) in the amino-acid sequence shown in sequence No. 2. (i) a base sequence composed of the 52nd to 1.236th bases Similarly, amino acid residues which are believed to be included in a base sequence in which the 562nd to 564th bases important for forming a homodimer are also preferably amino “agc' are replaced with bases representing a codon of an acid residues which are not mutated (deleted, replaced, or 10 added) from the amino-acid sequences described by (i) to amino acid other than serine in the base sequence shown in (iii). Preferable examples of the amino acid residues include, sequence No. 1; among the amino acid residues constituting the amino-acid (ii) a base sequence composed of the 52nd to 1.236th bases sequences described by (i) to (iii), amino acid residues cor included in a base sequence in which the 571st to 573rd bases responding to the locations of the 45th amino acid and the 15 'gcc are replaced with bases representing a codon of an 350th amino acid (specifically, the 45th aspartic acid (Asp: D) amino acid other than alanine in the base sequence shown in and the 350th arginine (Arg: R)) in the amino-acid sequence sequence No. 1; and shown in sequence No. 2. (iii) a base sequence composed of the 52nd to 1.236th bases Furthermore, amino acid residues which are N-type-Sugar included in a base sequence in which the 562nd to 564th bases chain-binding sites are also preferably amino acid residues “agc' are replaced with bases representing a codon of an which are not mutated (deleted, replaced, or added) from the amino acid other than serine and the 571st to 573rd bases are amino-acid sequences described by (i) to (iii). Preferable replaced with bases representing a codon of an amino acid examples of the amino acid residues include, among the other than alanine in the base sequence shown in sequence amino acid residues constituting the amino-acid sequences No. 1; or described by (i) to (iii), amino acid residues corresponding to 25 (b) DNA which encodes a protein having O-galactosidase the locations of the 124th, 177th, 201st, 359th, and 385th activity and which hybridizes with DNA composed of a base amino acids (all of which being asparagine (ASn: N)) in the sequence complementary to DNA containing any one of base amino-acid sequence shown in sequence No. 2. sequences described by (i) to (iii) under a stringent condition, In the present invention, C.-GAL activity can be measured wherein bases corresponding to the bases at the replacement as follows. For example, a target protein is allowed to be 30 sites described above are identical to the bases at the replace expressed with a cell derived from mammals, such as a CHO ment sites. cell or a human fibroblast, and is collected. The protein (en Zyme solution) is then mixed with 4-methylumbelliferyl-C.- In the present invention, the term “codon” means not only D-galactoside (a synthetic Substrate obtained from C-D-ga the triple base linkage (triplet) of an RNA sequence after lactose and 4-methylumbelliferone (fluorogenic substrate)), 35 transcription but also the triple base linkage of a DNA and the mixture is allowed to react under an acidic condition. sequence. Accordingly, codons of a DNA sequence are In this case, the amount of 4-methylumbelliferone released by denoted using thymine (T) instead of uracil (U). a unit quantity of the enzyme solution per unit time is detected The base sequence shown in sequence No. 1 is a base to measure C-GAL activity. sequence composed of 1.236 bases encoding wild-type O-NAGA activity can also be measured as in C.-GAL activ 40 O-NAGA. ity. A target protein is allowed to be expressed and is col More specifically, the DNA described by (a) is DNA com lected. The protein (enzyme solution) is then mixed with posed of a base sequence containing a base sequence ranging 4-methylumbelliferyl-C-N-acetyl-D-galactosaminide (a syn from the 52nd base to the 1,236th base excluding the 1st base thetic substrate obtained from C-N-acetyl-D-galactosamine to the 51st base, which encode a signal peptide of wild-type and 4-methylumbelliferone (fluorogenic substrate)), and the 45 C-NAGA, in a base sequence in which bases are replaced as mixture is allowed to react under an acidic condition. In this described in (i) to (iii) in the base sequence shown in sequence case, the amount of 4-methylumbelliferone which can be No. 1. released per unit time by a unit quantity of the enzyme solu Here, a preferable example of the base sequence containing tion is detected to measure C-NAGA activity. a base sequence ranging from the 52nd base to the 1.236th In the above methods of measuring O-GAL activity and 50 base is a base sequence in which a base sequence (polynucle C.-NAGA activity, various types of known detection methods otide) encoding a signal peptide is bound to the 5' side of the can be employed for detecting the fluorogenic substrate. For base sequence ranging from the 52nd base to the 1.236th base. example, a detection method using a fluorophotometer or the The signal peptide is not limited as long as the signal peptide like is preferable. The target protein can be expressed by can allow the protein to pass through a cell membrane of an incorporating a gene encoding the protein into a known 55 affected organ. For example, a signal peptide of a lysosomal expression vector or the like, and then introducing the vector enzyme such as wild-type C-NAGA or wild-type C-GAL, or into a cell. As the measurement methods, specifically, the a signal peptide of a secretase such as preprotrypsin is pref methods described in Example 3 and Example 4 below can be erable. A signal peptide of wild-type C-NAGA or wild-type preferably exemplified. C-GAL is more preferable. A base sequence encoding the 3. Recombinant Gene 60 signal peptide of wild-type C-NAGA is a base sequence com Preferable examples of a gene encoding the above-de posed of the 1st base to the 51st base in the base sequence of scribed protein of the present invention include, but are not wild-type C-NAGA shown in sequence No. 1. A base limited to, containing DNA described by (a) or (b) sequence encoding the signal peptide of wild-type C-GAL is below. The DNAs described by (a) and (b) are preferably a base sequence composed of the 1st base to the 93rd base in structural genes of the protein of the present invention. The 65 the base sequence of wild-type C-GAL shown in sequence gene containing any of these DNAS may entirely consist of No. 12. A base sequence encoding the signal peptide of pre the DNA. Alternatively, the gene may contain the DNA as a protrypsin is a base sequence shown in sequence No. 14. US 8,323,640 B2 15 16 As the DNA described by (a), among the DNAs containing which is designed such that a missense mutation is introduced the base sequence described by (i), (ii), or (iii), DNA contain to produce bases representing a codon of a desired amino ing the base sequence described by (iii) is particularly pref acid, and using, as a template, for example, DNA containing erable. a base sequence encoding wild-type C-NAGA. Preferable In addition, as the DNA described by (a), DNA in which 5 examples of such a PCR primer include synthetic oligonucle “the bases representing a codon of an amino acid other than otide primers shown in sequence Nos. 8 and 10, which are serine' described in (i) and (iii) are bases representing a described in Example 1 below. A DNA polymerase used for codon of glutamic acid or aspartic acid is preferable. Simi the PCR is not limited, but a DNA polymerase with high larly, as the DNA described by (a), DNA in which “the bases accuracy is preferable. Preferable examples thereof include representing a codon of an amino acid other than alanine 10 Pwo DNA (Polymerase Roche Diagnostics K.K.), Pfu DNA described in (ii) and (iii)are bases representing a codon of one polymerase (Promega), platinum Pfx DNA polymerase (In selected from the group consisting of leucine, Valine, isoleu vitrogen Corporation), KOD DNA polymerase (Toyobo Co., cine, phenylalanine, and methionine is also preferable. Ltd.), and KOD-plus-polymerase (Toyobo Co., Ltd.). Reac Herein, regarding bases representing a codon of each of the tion conditions for the PCR can be appropriately determined above amino acids (wherein the base disposed at the left end 15 in accordance with, for example, the optimum temperature of is defined as a base at the 5' side), bases representing a codon DNA polymerase used, and the length and the type of DNA to of glutamic acid are "gag or 'gaa' (preferably gag), and be synthesized. For example, under preferable cycle condi bases representing a codon of aspartic acid are "gat" or 'gac'. tions, a cycle consisting of “5 to 30 seconds at 90° C. to 98°C. Similarly, bases representing a codon of leucine are “ctc. (thermal denaturation and dissociation)->5 to 30 seconds at “ctt”, “cta', or “ctg” (preferably “ctic'), bases representing a 20 50° C. to 65° C. (annealing)->30 to 1,200 seconds at 65° C. to codon of valine are “gtt”, “gtc. “gta', or “gtg', bases repre 80°C. (synthesis and extension) is performed a total of 20 to senting a codon of isoleucine are “att”, “atc', or “ata”, bases 200 times. representing a codon of phenylalanine are “ttt” or “ttic', and The DNA described by (b) can be obtained as follows. A bases representing a codon of methionine are “atg'. Bases known hybridization method such as colony hybridization, representing a codon of serine include “agt” in addition to 25 plaque hybridization, or Southern blotting is performed using “agc' mentioned above. Bases representing a codon of ala DNA containing any one of base sequences described by (i) to nine include 'gct', 'gca', and 'gcg in addition to 'gcc. (iii) (i.e., DNA described by (a)), DNA composed of a base mentioned above. sequence complementary to this DNA, or a fragment thereof Furthermore, as the DNA described by (a) above, DNA in as a probe, and the DNA described by (b) can be obtained which “the bases representing a codon of an amino acid other 30 from a cDNA library or a genomic library. A library prepared than serine' described in (i) to (iii) are bases representing a by a known method may be used. Alternatively, a commer codon of glutamic acid, and “the bases representing a codon cially available cDNA library or genomic library may be ofanamino acid other thanalanine described in (i) to (iii) are used. The library is not limited thereto. bases representing a codon of leucine is particularly prefer Regarding a detailed procedure of the hybridization able. A preferable example of such DNA is DNA composed of 35 method, refer to, for example, Molecular Cloning, A Labora a base sequence (sequence No. 3) in which the 562nd to 564th tory Manual 2nd ed. (Cold Spring Harbor Laboratory Press bases, which are bases representing a codon of serine, (1989)) as needed. included in the base sequence of wild-type O.-NAGA (se The term “stringent condition during the performance of quence No. 1) are replaced with bases representing a codon of a hybridization method means a condition during washing glutamic acid (“agc’->"gag'), and the 571st to 573rd bases, 40 after hybridization, and specifically, a salt concentration of a which are bases representing a codon of alanine, included in buffer in the range of 15 to 330 mM and a temperature in the the base sequence of wild-type C-NAGA (sequence No. 1) are range of 25°C. to 65°C., and preferably, a salt concentration replaced with bases representing a codon of leucine in the range of 15 to 150 mM and a temperature in the range (“gcc’->"ctic'). In this exemplification, the 562nd to 564th of 45° C. to 55° C. More specifically, an example of the bases after replacement are not limited to "gag mentioned 45 stringent condition is a salt concentration of 80 mM and a above and may be other bases as long as the bases represent a temperature of 50° C. Furthermore, in addition to the salt codon of glutamic acid. Similarly, the 571st to 573rd bases concentration, the temperature, and the like, in consideration after replacement are not limited to “ctic' mentioned above of other conditions such as the probe concentration, the length and may be other bases as long as the bases represent a codon of the probe, and the reaction time, conditions for obtaining of leucine. 50 the DNA described by (b) can be appropriately determined. Such mutant DNA can be prepared in accordance with a The DNA to be hybridized has a base sequence having a site-directed mutagenesis method described in, for example, homology of preferably at least 40% or more, more prefer Molecular Cloning, A Laboratory Manual 2nd ed., Cold ably 60%, further preferably 90% or more, particularly pref Spring Harbor Laboratory Press (1989), Current Protocols in erably 95% or more, and most preferably 99% or more rela Molecular Biology, John Wiley & Sons (1987-1997). Specifi- 55 tive to the base sequence of the DNA described by (a). cally, such DNA can be prepared by a known method such as Furthermore, in the DNA described by (b), bases corre a Kunkel method or a Gapped duplex method using a kit for sponding to the bases at the replacement sites described above introducing a mutation utilizing the site-directed mutagenesis are the same as the bases at the replacement sites. method. Preferable examples of the kit include Quick Herein, the term “replacement sites' means sites of the ChangeTM Site-Directed Mutagenesis Kit (manufactured by 60 base replacements performed in any one of base sequences Stratagene), GenelailorTM Site-Directed Mutagenesis System described by (i) to (iii) contained in the DNA described by (a). (manufactured by Invitrogen Corporation), and TakaRa Site Specifically, the term “replacement sites’ means sites of Directed Mutagenesis System (e.g., Mutan-K or Mutan-Su bases (triplet) representing a codon after the change caused per Express Km: manufactured by Takara Bio Inc.). by the base replacements. More specifically, the term Alternatively, as described in Examples below, such DNA 65 “replacement sites' means, among 1,185 bases constituting can be prepared by performing a polymerase chain reaction the base sequences described by (i) to (iii), bases correspond (PCR) under an appropriate condition using a PCR primer ing to the locations of the 562nd to 564th bases in the base US 8,323,640 B2 17 18 sequence shown in sequence No. 1 (however, the base A particularly preferable example of the DNA described by sequences being limited to the base sequences described by (b) is DNA composed of a base sequence which is not com (i) and (iii)), and bases corresponding to the locations of the pletely the same as the base sequence of the DNA described 571st to 573rd bases in the base sequence shown in sequence by (a) but in which the amino-acid sequence after translation No. 1 (however, the base sequences being limited to the base is completely the same as the amino-acid sequence of the sequences described by (ii) and (iii)). More specifically, the DNA described by (a) (i.e., DNA obtained by performing a former bases mean the 511th to 513th bases in the base silent mutation on the DNA described by (a)). sequences described by (i) and (iii) above, and the latter bases Regarding agene encoding the protein of the present inven mean the 520th to 522nd bases in the base sequences tion, codons corresponding to each of the amino acids after described by (ii) and (iii) above. 10 translation are not particularly limited. Accordingly, the gene In addition, the term “bases corresponding . . . . in the encoding the protein of the present invention may contain phrase “bases corresponding to the bases at the replacement DNA representing codons which are generally used (prefer sites' means bases (triplet) which are located so as to face ably, codons whose frequency of use is high) in mammals, bases (triplet) complementary to the bases at the replacement 15 Such as the human, after transcription. Alternatively, the gene sites in a hybrid prepared by hybridizing the DNA described may contain DNA representing codons which are generally by (b) with a strand complementary to the DNA described by used (preferably, codons whose frequency of use is high) in, (a). For example, when the base sequence of the DNA for example, a microorganism, such as E. coli or yeast, or a described by (b) does not have a mutation such as deletion or plant, after transcription. addition, as compared with the DNA described by (a) (that is, 4. Recombinant Vector and Transformant when the length (the number of bases) of the DNA described In order to express the protein of the present invention, first, by (a) is the same as the length (the number of bases) of the it is necessary to incorporate the above-described gene of the DNA (b)), the 511th to 513th bases and/or the 520th to 522nd present invention into an expression vector to construct a bases in the base sequence of the DNA described by (b) are recombinant vector. In this step, as needed, a transcriptional the above “bases corresponding . . . . in the phrase “bases 25 promoter, the SD sequence (in the case where a host is a corresponding to the bases at the replacement sites’. prokaryotic cell), and the Kozak sequence (in the case where It is important that the DNA described by (b) be DNA a host is a eukaryotic cell) may be linked upstream of the gene encoding a protein having O-GAL activity. Therefore, for to be incorporated into the expression vector in advance. A example, bases representing a codon of an amino acid residue terminator may be linked downstream thereof in advance. which is believed to be important for the binding performance 30 (Substrate-binding performance) with an O-galactose residue Furthermore, an enhancer, a splicing signal, a poly-A addi in a substrate of O-GAL and the catalytic reactivity to the tion signal, a selective marker, and the like may also be linked substrate are preferably bases which are not mutated (deleted, in advance. The above elements, such as a transcriptional replaced, or added) from the base sequences described by (i) promoter, required for expressing a gene may be originally to (iii). Preferable examples of such bases of the base 35 contained in the gene. In the case where these elements are sequences described by (i) to (iii) include bases correspond originally contained in the expression vector, the elements ing to the locations of the 82nd to 93rd bases (4 codons), the contained in the expression vector may be utilized. Thus, the 229th to 243rd bases (5 codons), the 349th to 381st bases (11 form of use of these elements is not particularly limited. codons), the 448th to 474th bases (9 codons), the 574th to As a method of incorporating the gene into an expression 576th bases (1 codon), the 625th to 660th bases (12 codons), 40 vector, various types of methods using a known gene recom and the 724th to 762nd bases (13 codons) in the base sequence bination technique, for example, a method using a restriction shown in sequence No. 1 among the base sequences described enzyme, or a method using a topoisomerase, can be by (i) to (iii). Among these bases, bases corresponding to the employed. The expression vector is not limited as long as the 466th to 468th bases and 649th to 651st bases, which repre expression vector can maintain a gene encoding a protein of sent codons of amino acid residues of a catalytic site, are 45 the present invention. Examples of the expression vector particularly preferable. include plasmid DNA, bacteriophage DNA, retrotransposon Furthermore, in the DNA described by (b), bases represent DNA, a retrovirus vector, and artificial chromosome DNA. A ing a codon of an amino acid residue which is believed to be vector which can be suitably combined with a host cell used important for forming a homodimer are also preferably bases can be appropriately selected and used. which are not mutated (deleted, replaced, or added) from the 50 Subsequently, the constructed recombinant vector is intro base sequences described by (i) to (iii). Preferable examples duced into a host to obtain a transformant, and the transfor of such bases of the base sequences described by (i) to (iii) mant is cultured. Thus, the protein of the present invention include bases corresponding to the locations of the 133rd to can be expressed. The term “transformant’ used in the present 135th bases and the 1,048th to 1,050th bases in the base invention means a product in which a foreign gene is intro sequence shown in sequence No. 1 among the base sequences 55 duced into a host. For example, the transformant includes a described by (i) to (iii). product in which a foreign gene is introduced by introducing Furthermore, in the DNA described by (b), bases represent plasmid DNA or the like into a host (transformation) and a ing a codon of an amino acid residue which is an N-type product in which a foreign gene is introduced by infecting a Sugar-chain-binding site are also preferably bases which are host with a virus or a phage (transduction). not mutated (deleted, replaced, or added) from the base 60 The host is not limited as long as the host can express a sequences described by (i) to (iii). Preferable examples of protein of the present invention after the recombinant vector such bases of the base sequences described by (i) to (iii) is introduced thereinto, and can be appropriately selected. include bases corresponding to the locations of the 370th to Examples thereof include known hosts such as animal cells, 372nd bases, the 529th to 531st bases, the 601st to 603rd e.g., a human cell and amouse cell, plant cells, bacteria, yeast, bases, the 1,075th to 1,077th bases, and the 1,153rd to 1,155th 65 and plant cells. bases in the base sequence shown in sequence No. 1 among When an animal cell is used as a host, for example, a human the base sequences described by (i) to (iii). fibroblast, a CHO cell, a monkey COS-7 cell, Vero, a mouse US 8,323,640 B2 19 20 L cell, a rat GH3, a human FL cell, or the like is used. method used for culturing a host. The target protein is accu Alternatively, insect cells such as an Sf9 cell or an Sf21 cell mulated in the cultured product. can also be used. As a medium used for the culture, a known natural medium When a bacterium is used as a host, for example, E. coli or or a synthetic medium may be used as long as the medium Bacillus subtilis is used. 5 contains, for example, a carbon source, a nitrogen Source, and When yeast is used as a host, for example, Saccharomyces inorganic salts, all of which can be utilized by the host, and the cerevisiae or Schizosaccharomyces pombe is used. transformant can be cultured efficiently. When a plant cell is used as a host, for example, a tobacco In order to prevent detachment of a recombinant vector BY-2 cell is used. contained in the transformant and detachment of a gene The method of obtaining a transformant is not limited and 10 encoding a target protein, the culture may be performed in a can be appropriately selected in consideration of the combi nation of the types of host and expression vector used. Pref state in which a selective pressure is applied. Specifically, in erable examples of the method include an electroporation the case where a selective marker is a drug-resistance gene, a method, a lipofection method, a heat shock method, a poly corresponding drug can be added to the medium. In the case ethylene glycol (PEG) method, a calcium phosphate method, 15 where a selective marker is an auxotrophic complementary a diethylaminoethyl dextran (DEAE-dextran) method, and a gene, a corresponding nutritional factor can be removed from method of infecting a virus such as a DNA virus or an RNA the medium. For example, in the case where a human fibro virus. blast transduced with a vector containing the G418-resistant In the resulting transformant, the codon type of a gene gene is cultured, G418 (G418 sulfate) may be added during contained in the recombinant vector is not limited. The codon culturing, as needed. type may be identical to or different from the codon type of a In the case where, for example, a transformant transformed host which is actually used. with an expression vector in which an inducible promoter is 5. Method of Producing Protein used as a promoter is cultured, an appropriate inducer (for A protein of the present invention can be produced by example, isopropyl (3-D-thiogalactopyranoside (IPTG)) may changing the structure of the active site (in particular, the 25 be added to the medium, as needed. substrate-binding site) of wild-type C-NAGA so that a sub The culture conditions of the transformant are not particu strate of O-GAL can be bound thereto. If the substrate of larly limited as long as productivity of the target protein and C-GAL can be bound to the active site, the substrate can be growth of the host are not interrupted. In general, the culture hydrolyzed by a catalysis due to the catalytic site of wild-type is performed at a temperature in the range of 10°C. to 40°C., O-NAGA. 30 and preferably in the range of 20° C. to 37° C. for 5 to 100 Such a structural change is performed as follows. For hours. The pH can be adjusted using an inorganic or organic example, as described above, in the amino-acid sequence acid, an alkaline solution, or the like. Examples of the method constituting the active site (Substrate-binding site) of wild of culturing include Solid culture, static culture, shaking cul type C-NAGA, (i) the 188th serine is replaced with another ture, and aeration stirring culture. amino acid Such as glutamic acid or aspartic acid, (ii) the 35 When the target protein is produced in bacteria or in cells, 191st alanine is replaced with another amino acid Such as the target protein can be collected by disrupting the bacteria leucine, Valine, isoleucine, phenylalanine, or methionine, or or the cells. As a method of disrupting the bacteria or the cells, (iii) both the 188th serine and the 191st alanine are replaced as for example, a high-pressure treatment using a French press described in (i) and (ii) above by a gene recombination tech ora homogenizer, an ultrasonic treatment, a milling treatment nique. The structural change can be achieved by changing the 40 using glass beads or the like, an enzyme treatment using three-dimensional structure of the side chains of the amino , , pectinase, or the like, a freeze-thawing acids before and after the replacement. Consequently, the treatment, a treatment with a hypotonic Solution, or a bacte substrate specificity of wild-type C-NAGA can be changed. riolysis-inducing treatment using a phage can be employed. In particular, the above structural change is preferably per After the disruption, the disruption residue (which contains a formed by replacing the 188th serine with glutamic acid and 45 fraction insoluble in a cell extract) of the bacteria or the cells by replacing the 191st alanine with leucine. Thereby, the can be removed, as needed. Examples of the method of substrate specificity of O-GAL can be imparted to C-NAGA. removing the residue include centrifugal separation and fil In the above structural change, an amino acid replacement tration. Furthermore, the efficiency of the removal of residue which causes a significant three-dimensional structural can be increased using, for example, a flocculant or a filter aid, change is the replacement of the 191st alanine with, for 50 as needed. The supernatant obtained after the removal of example, leucine. More specifically, the side chain of the residue is a fraction soluble in the cell extract, and can be used 191st amino acid is changed from “—CH, which is the side as a crude protein solution. chain of alanine, to a bulky side chain, Such as “—CH2—CH When the target protein is produced in bacteria or in cells, (CH)—CH', which is the side chain of leucine. As a result, alternatively, the bacteria of the cells may be recovered by, for the space of the active site in which the N-acetyl group in a 55 example, centrifugal separation or membrane separation, and substrate of C-NAGA is incorporated is restricted, thereby thus, the protein can be used without disrupting the bacteria or decreasing the binding performance of the protein with the the cells. Substrate. Instead, the binding performance with a substrate In contrast, when the target protein is produced outside of O-GAL is increased accordingly. bacteria or outside cells, the culture solution is used without The protein of the present invention can be produced spe 60 further treatment, or the bacteria or the cells are removed by, cifically by a method including a step of culturing the above for example, centrifugal separation or filtration. Subse descried transformant and a step of collecting a protein hav quently, the target protein is collected from the cultured prod ing O-galactosidase activity from the resulting cultured uct by, for example, extraction by ammonium sulfate precipi product. Herein, the term “cultured product” means all of a tation, as needed. Furthermore, according to need, the target culture Supernatant, cultured cells, cultured bacteria, dis 65 protein may be isolated and purified by dialysis and chroma rupted cells, and disrupted bacteria. The culture of the trans tography (such as gel filtration, ion-exchange chromatogra formant can be performed in accordance with a normal phy, or affinity chromatography). US 8,323,640 B2 21 22 The production yield of a protein obtained by culturing a In the case where the pharmaceutical composition is used transformant or the like can be determined by, for example, as an enzyme agent used in enzyme replacement, the mixing Sodium dodecyl Sulfate-polyacrylamide gel electrophoresis ratio of the protein of the present invention, and the types and (SDS-PAGE) in terms of the units per culture solution, the mixing ratios of other components used can be appropriately units per wet weight or dry weight of bacteria, the units per 5 determined in accordance with a preparation method of a protein in a crude enzyme solution, or the like. known enzyme agent used in enzyme replacement (in par In addition to the protein synthesis system using a trans ticular, an enzyme agent used in enzyme replacement therapy formant described above, a target protein can be produced for Fabry disease). using a cell-free protein synthesis system in which no living The method of administrating the pharmaceutical compo cells are used. 10 sition is not limited. When the pharmaceutical composition is The cell-free protein synthesis system is a system in which an enzyme agent used in enzyme replacement, parenteral a target protein is synthesized in an artificial container Such as administration Such as intravenous drip is generally used. In a test tube using a cell extract. A cell-free protein synthesis the pharmaceutical preparation that can be used in various system which can be used also includes a cell-free transcrip 15 administration methods such as parenteral administration, for tion system in which RNA is synthesized using DNA as a example, an excipient, a filler, an extender, a binder, a humec template. tant, a disintegrant, a lubricant, a Surfactant, a dispersant, a In this case, the cell extract used is preferably derived from buffer, a preservative, a solubilizer, an antiseptic, a flavoring the host cell described above. Examples of the cell extract that agent, a Soothing agent, a stabilizing agent, and an isotonizing can be used include extracts derived from eukaryotic cells and agent, all of which are generally used in the production of extracts derived from prokaryotic cells. More specifically, medicine, may be appropriately selected and used, and thus, examples of the cell extract include extracts of a CHO cell, a the pharmaceutical composition can be prepared by an exist rabbit reticulocyte, a mouse L-cell, a HeLa cell, wheat germ, ing method. budding yeast, or E. coli. The cell extract may be concentrated The form of the pharmaceutical composition is not limited. or diluted for use. Alternatively, the cell extract may be used 25 When the pharmaceutical composition is an enzyme agent without further treatment. Thus, the method of using the cell used in enzyme replacement, an intravenous injection (in extract is not limited. cluding drip infusion) is generally used. For example, the The cell extract can be obtained by, for example, ultrafil pharmaceutical composition can be provided in the form of tration, dialysis, and polyethylene glycol (PEG) precipita for example, a single-dose ampule or a multi-dose container. tion. 30 In general, the dosage of the pharmaceutical composition Alternatively, such cell-free protein synthesis can be per can be appropriately determined in a wide range in consider formed using a commercially available kit. Examples of the ation of not only the mixing ratio of the active ingredient in kit include a reagent kit PROTEIOSTM (Toyobo Co., Ltd.), the pharmaceutical preparation but also the age and body TNTTM System (Promega), a synthesis device PG-MateTM weight of the subject (patient) to be administered with the (Toyobo Co., Ltd.), and RTS (Roche Diagnostics K.K.). 35 pharmaceutical composition, the type of disease, the State of The target protein produced by cell-free protein synthesis disease, the administration route, the number of administra can be purified by appropriately selecting means such as tions, the term of administration, and the like. In particular, in chromatography, as described above. the case where the therapeutic agent of the present invention 6. Pharmaceutical Composition is an enzyme agent used in enzyme replacement, the number (i) Pharmaceutical composition used as enzyme agent used 40 of times of administration is preferably about one in every two in enzyme replacement or the like to four weeks. In Such a case, the amount of enzyme agent As described above, the protein of the present invention administered each time is determined such that, for example, can achieve various excellent effects regarding the treatment preferably about 0.1 to 10 mg/kg, more preferably about 0.1 of Fabry disease, and thus can be used as an active ingredient to 5 mg/kg, and further preferably about 0.2 to 1 mg/kg of the of a therapeutic agent for Fabry disease. That is, the present 45 protein or the like (recombinant enzyme) of the present inven invention provides a therapeutic agent for Fabry disease con tion, which is an active ingredient, can be administered rela taining a pharmaceutical composition containing the above tive to the body weight of a patient. described protein of the present invention. A preferable spe In the present invention, the protein (recombinant enzyme) cific example of this therapeutic agentis an enzyme agent that of the present invention, which functions as an active ingre can be used for enzyme replacement therapy. 50 dient, has excellent stability in blood and high incorporation The protein of the present invention, which functions as an efficiency into a cell of an affected organ. Therefore, even active ingredient in the pharmaceutical composition, may be when the protein is used in an amount Smaller than that in used in the form of a salt, a hydrate, or the like, as needed. In known pharmaceutical compositions, the effect of enzyme addition, the protein of the present invention may be used in replacement which is the same as or stronger than that a state in which an appropriate chemical modification is per 55 achieved by the known pharmaceutical compositions can be formed in consideration of storage stability (in particular, achieved. In addition, allergic adverse side effects of the maintenance of activity) as a therapeutic agent. Thus, the protein of the present invention are negligible. Accordingly, form of the protein of the present invention is not limited. physical, mental, and economical burdens on patients can be The pharmaceutical composition may contain components markedly reduced. other than the protein of the present invention. Examples of 60 (ii) Pharmaceutical Composition Used as Gene Therapeutic the other components include pharmaceutical various com Agent ponents (such as various types of pharmaceutically accept As described above, the gene of the present invention able carriers) that are required in accordance with the usage encodes a protein of the present invention which can achieve (the form of usage) of the pharmaceutical composition. The various excellent effects regarding the treatment of Fabry other components can be appropriately contained as long as 65 disease, and thus can be used as an active ingredient of a the effects achieved by the protein of the present invention therapeutic agent (gene therapeutic agent) of Fabry disease. and the like are not impaired. That is, the present invention provides a gene therapeutic US 8,323,640 B2 23 24 agent for Fabry disease containing, as an active ingredient, a specificity similar to that of human C-GAL, sites (locations of pharmaceutical composition containing the above-described an amino acid) of mutation to be introduced into human gene of the present invention. C-NAGA were determined by comparison and study using In the case where the pharmaceutical composition is used as a gene therapeutic agent, a method of directly administer three-dimensional structural models of proteins. The proce ing by injection or a method of administering a vector into dure and results of the determination are specifically which a nucleic acid is incorporated is used. Examples of the described below. vector include an adenovirus vector, an adeno-associated 1. Used Data virus vector, a herpesvirus vector, a vaccinia virus vector, a The amino-acid sequence data of human C-NAGA and retrovirus vector, and a lentivirus vector. By using these virus human C.-GAL which are registered in Swiss-Prot as vectors, the gene therapeutic agent can be administered with 10 described below were used. The protein three-dimensional high efficiency. A commercially available gene transfer kit structural data of chicken C-NAGA and human O-GAL which (for example, product name: AdenoExpress, manufactured are registered in Protein Data Bank (PDB) as described below by Clontech) can also be used. were used. In addition, in the case where the pharmaceutical compo (1) Amino-Acid Sequence Data sition is used as a gene therapeutic agent, the composition 15 may be introduced into a phospholipid endoplasmic reticu Used database: Swiss-Prot (tw.expasy.org/uniprot/) lum such as liposome, and the endoplasmic reticulum may be administered. Specifically, an endoplasmic reticulum in which a gene of the present invention is held is introduced into a predetermined cell by a lipofection method. The resulting entry name accession number cell is then administered in, for example, a vein or an artery. Human C-NAGA NAGAB. HUMAN P17050 Alternatively, such a cell can be locally administered in an Human O-GAL AGAL HUMAN PO628O organ affected by Fabry disease. For example, in the case where the pharmaceutical composition is administered to an (2) Protein Three-Dimensional Structural Data adult, the dose is preferably about 0.1 ug/kg to 1,000 mg/kg, 25 Used database: Protein Data Bank (PDB) (www.rcsb.org/ and more preferably, about 1 ug/kg to 100 mg/kg per day pdb/) relative to the body weight of the patient. 7. Method of Treating Fabry Disease The present invention includes a method of treating Fabry disease including administering the pharmaceutical compo PDB ID sition to a Fabry patient. The present invention also includes 30 the use of the pharmaceutical composition for treating Fabry Chicken C-NAGA 1KTC (refer to *1 below) disease, and the use of the pharmaceutical composition or the Human Ci-GAL 1R47 (refer to *2 below) protein of the present invention for producing medicine for * 1: Garman SC et al., Structure (Camb), 2002, 10(3): 425-34. treating Fabry disease. *2: Garman SC et al., J. Mol. Biol., 2004, 19; 337(2): 319-35. The pharmaceutical composition used in the method of 35 treatment of the present invention may be a pharmaceutical 2. Construction of Three-Dimensional Structural Model of composition containing a protein of the present invention Human C-NAGA ('section 6 (i) above), a pharmaceutical composition con The construction of a three-dimensional structural model taining a gene of the present invention ('section 6 (ii) above), of human C-NAGA was performed on the basis of the three or a combination of these pharmaceutical compositions. The 40 dimensional structure of chicken C.-NAGA using a homology pharmaceutical composition is not limited thereto, and can be modeling method, which is an existing method (refer to Sut appropriately selected in consideration of the state of disease cliffe MJ et al., Prot. Eng., 1987, 1,377-84; and Sutcliffe M of the patient, the presence or absence of adverse side effects, Jet al., Prot. Eng., 1987, 1,385-92). The three-dimensional the administration effect, and the like. Here, each of the structure of chicken-derived C.-NAGA (the complex with a pharmaceutical compositions to be administered to a Fabry 45 substrate) registered in PDB was used as a template three patient can be administered in the form of usage of the enzyme agent used in enzyme replacement or the gene thera dimensional structure. The degree of matching (identity) of peutic agent described above. amino acids between human C-NAGA and chicken C-NAGA In particular, when the pharmaceutical compositions are is 75%, which satisfies the condition (identity 23.0%) for used in combination as described above, for example, the constructing a three-dimensional structural model by the proportion of the amount of administration, the number of homology modeling method. The construction of a three times of administration, and the term of administration of dimensional structural model by the homology modeling each of the pharmaceutical compositions can be appropri method was performed using MODELLER, which is existing ately determined in accordance with the conditions of each software (capable of being used by accessing MODELLER patient. For example, a preferable method of administration CBSU Web (cbsuapps.tc.cornell.edu/modeller.aspx)). Fur and a preferable amount of administration of each of the 55 thermore, a model of a complex with a Substrate of human pharmaceutical compositions and the like are as described C.-NAGA was constructed by fitting a substrate bound to above. chicken C-NAGA into the constructed three-dimensional The present invention will now be descried more specifi structural model of human C-NAGA in accordance with the cally using Examples, but the present invention is not limited position of the substrate bound to chicken O-NAGA. thereto. 60 3. Comparison of Three-Dimensional Structures Contribut EXAMPLE1 ing to Substrate Specificity of Human C-GAL and that of Human C-NAGA Selection of Mutation Sites to be Introduced into The three-dimensional structure of human O-NAGA is C-N-Acetylgalactosaminidase (C-NAGA) similar to that of human C-GAL, and catalytic domains of 65 both human C-NAGA and human O-GAL have a (BC)s-barrel In order to design a novel enzyme in which the Substrate structure. Amino acid residues (catalytic residues) required specificity of human C.-NAGA is modified into a substrate for a catalytic action existing in the active site (including a US 8,323,640 B2 25 26 catalytic site and a substrate-binding site) are localized at the 5. Points of Difference Between Human O-GAL and Human C-terminal side of each strand of the (BC)s-barrel structure. In O-NAGA FIGS. 1 and 2, the three-dimensional structures of human There are two residues which are different between human C.-NAGA and human O-GAL are shown by a ribbon model, C-GAL and human C-NAGA (refer to FIG. 2C). In O-GAL, and amino acid residues of the catalytic site and the Substrate the amino acid residues corresponding to Ser188 and Ala 191 binding site in each of the structures are shown by a stick of C-NAGA were Glu203 and Leu206, respectively. FIG. 2B model. In order to compare the positional relationships shows locations of the amino acid residues which are differ between a substrate and the residues of the catalytic site and ent between C-NAGA and O-GAL in the three-dimensional the Substrate-binding site interms of three-dimensional struc Structure. ture, the three-dimensional structure of C.-NAGA was super 10 As shown in FIG. 2C, in O-GAL, an “ -OH group (a imposed on the three-dimensional structure of C.-GAL by the hydroxyl group) is bonded to the carbon atom at the 2-posi superimposing method developed by Kabsch (refer to Kabsch tion of the Sugar (six-membered ring) in the Substrate of W. et al., Acta Crystallogr; 1976: A32, 827-828; and Kabsch C-GAL, and in C-NAGA, an NH COCH)—O group (an W. et al., Acta Crystallogr; 1978: A34, 922–923). Subse N-acetyl group) is boned to the carbonatomat the 2-position quently, in the human C-NAGA model, amino acid residues 15 related to the binding of the substrate were selected by of the sugar (six-membered ring) in the substrate of C-NAGA. extracting amino acid residues adjacent to the Substrate. The It is supposed that, in C-NAGA, the hydroxyl group of the results are shown in Table 1. The right column of Table 1 side chain of Ser188 is bonded to the oxygen atom of the shows 14 amino acid residues selected from human N-acetyl group in the Substrate by a hydrogen bond, and the O-NAGA, and the left column of Table 1 shows amino acids methyl group of the side chain of Alal 91 is bonded to the in human C.-GAL which positionally correspond to the 14 methyl group of the N-acetyl group in the Substrate by a amino acid residues. hydrophobic bond. On the basis of these suppositions, it was believed that Ser188 and Ala 191 of C-NAGA are important TABLE 1. residues for recognizing the N-acetyl group in the Substrate. 25 In contrast, it has been reported that, in human O-GAL, Human O-GAL Human C-NAGA Glu203 and Leu206, which are different from the correspond Trp47 Trp33 ing residues of C-NAGA, are important for recognizing a Asp92 Asp78 substrate of O-GAL (Garman SC et al., J. Mol. Biol., 2004, Asp93 Asp79 19; 337(2): 319-35). Furthermore, it has been cleared that, Tyr134 Tyr119 30 Cys142 Cys127 from X-ray crystal structure analysis, the carboxyl group of Lys168 Lys154 the side chain of Glu203 of O-GAL forms a hydrogen bond Asp170 (*) Asp156 (*) with the hydroxyl group of the substrate. In addition, Leu206 Cys172 Cys158 of O-GAL is a residuehaving a bulky side chain, and occupies Glu2O3 Ser188 a part of the space of the substrate-binding site of O-GAL. On Leu306 Ala191 Tyr207 Tyr192 35 the other hand, the hydroxyl group (at the 2-position) in the Arg227 Arg213 Substrate of O-GAL is a functional group which is not bulky. Asp231 (*) Asp217 (*) It is obvious that the hydroxyl group is smaller than, for Asp266 Asp252 example, the N-acetyl group in the substrate of C-NAGA. (*) catalytic residue Accordingly, it is believed that, in the binding between 40 C-GAL and the substrate, the size of the space of the sub These amino acid residues were compared with each other strate-binding site of O-GAL is suitable for the size of the by Superimposing the three-dimensional structure of human hydroxyl group in the Substrate. Consequently, it was O-NAGA on that of human O-GAL to detect residues that are believed that, in O-GAL, two residues of Glu203 and Leu206 identical to each other and residues that are different from highly contribute to the substrate specificity. each other. Thus, common points and points of difference in 45 6. Verification of Substrate Specificity by Three-Structural the amino-acid sequences of human C-GAL and human Models O-NAGA were cleared. Furthermore, in order to verify the interaction of C-NAGA 4. Common Points Between Human O-GAL and Human with a substrate and the interaction of O-GAL with a sub O-NAGA strate, models in which the substrates were exchanged with As a result, it was found that among the 14 extracted 50 each other, that is, (i) a complex model combining the Sub residues, 12 residues including Asp156 and Asp217, which strate of O-GAL with C.-NAGA, and (ii) a complex model are the catalytic site of human C-NAGA, are identical combining the substrate of C-NAGA with C.-GAL, were con between C-NAGA and O-GAL. Atoms in these amino acid structed to examine the influence of the two residues which residues in the Superimposed three-dimensional structures are different between C-NAGA and O-GAL on the substrates. are also satisfactorily Superimposed with each other, and thus, 55 According to the results, in the complex model in which the it was confirmed that the locations in the three-dimensional substrate of O-GAL was fitted into the C-NAGA model struc structures are also very similar to each other. FIG. 2A shows ture, the side chain of Ser188 of C-NAGA did not interact locations of the amino acid residues in the three-dimensional with the hydroxyl group at the 2-position of the substrate of structures, the amino acid residues being common to C-GAL. In addition, a clearance space was formed between O-NAGA and O-GAL. FIG. 2C shows the interaction 60 Ala191 and the substrate of O-GAL, and thus, interaction between each of the amino acid residues and a substrate. It is with the hydroxyl group was not observed. On the other hand, believed that all of these residues are related to the substrate in the complex model in which the substrate of C-NAGA was by a hydrogenbond or a hydrophobic bond. Note that, in FIG. fitted into the O-GAL structure, it was confirmed that the 2C, the amino acids which are not underlined are amino acids N-acetyl group at the 2-position of the substrate of O-NAGA common to O-NAGA and O-GAL, and the underlined amino 65 collides against Glu203 and Leu206 of O-GAL. Conse acids are amino acids different between C-NAGA and quently, it was predicted that binding of the Substrate was C-GAL. blocked by the presence of these two residues. US 8,323,640 B2 27 28 These predicted results supported the experimental results 9. Other Candidates of Human C-NAGA Amino-Acid described above. Thus, it was supported that Ser188 and Replacement Ala 191 of O-NAGA and Glu2O3 and Leu2O6 of O-GAL are The above-described modification of the substrate speci important for the substrate specificity of C-NAGA and ficity is achieved by two actions, i.e., a binding inhibition due C-GAL, respectively. to a steric hindrance to a substrate of C-NAGA and the for 7. Amino-Acid Residue Replacement for Modifying Sub mation of a hydrogen bond with a substrate of C.-GAL. Fur strate Specificity of Human O-NAGA to Substrate Specificity thermore, regarding the above-described amino-acid replace Similar to that of Human O-GAL ments, the presence or absence of the possibility of replacement to other amino acids was studied. As described above, between human O-GAL and human First, for the above action of the binding inhibition, a C-NAGA, the amino acid sequences are completely identical 10 replacement with Leu, which is the same amino acid as that in including the catalytic site except for the two residues which C-GAL, was performed as a first candidate. Furthermore, as a recognize the functional group bonded to the carbon atom at replacement which achieved the same action, a replacement the 2-position of the Sugar (six-membered ring) in each of the with Val, Ile, Phe, or Met, which is a hydrophobic amino-acid substrates. This indicates it is possible to retain the catalytic residue, was also possible. activity as is before replacement and to change only the 15 In addition, for the above action of the formation of a substrate specificity from C-NAGA specific to O-GAL spe hydrogen bond, a replacement with Glu, which is the same cific or vice versa by replacing these two residues which amino acid as that in C-GAL, was performed as a first candi highly contribute to the substrate specificity. In order that the date. Furthermore, as a replacement which achieved the same Substrate specificity of human C.-NAGA is changed and action, a replacement with Asp, which also has a carboxyl C-GAL activity is expressed by C-NAGA, an amino-acid group as Glu, was also possible. replacement at these two positions is important. By replacing 10. Amino-Acid Sequence of Wild-Type Human C-NAGA Ser188 of human C-NAGA with Glu, the recognition by a and Amino-Acid Sequence of Modified C-NAGA hydrogen bond with the N-acetyl group of the substrate of The amino-acid sequence of wild-type human C-NAGA is C-NAGA can be removed, and an interaction by a hydrogen shown in “sequence No. 2', and the amino-acid sequence of bond to a hydroxyl group of the substrate of O-GAL can be 25 the mutant O-NAGA (C-NAGA S188E/A191L) is shown in introduced. Furthermore, by replacing Ala 191 of human “sequence No. 4'. C.-NAGA with Leu, the space in which an N-acetyl group is to be incorporated in the binding of a substrate of C.-NAGA is EXAMPLE 2 occupied by the bulky side chain of Leu, and thus, the binding of the substrate is blocked by this steric hindrance. It was 30 1. Preparation of Cl-N-acetylgalactosaminidase predicted that, in C-NAGA, the original recognition of a (C.-NAGA) Retrovirus Vector substrate of C.-NAGA could be removed and a high specificity with a substrate of O-GAL could be provided by the above An O-NAGA clNA clone (Homo sapiens N-acetylgalac effects. tosaminidase, alpha, m-RNA, Gene Bank Accession: 8. Evaluation of Human O-NAGA Amino-Acid Replacement 35 BC000095, IMAGE: 3504221) was purchased from Open Model Biosystems. The coding sequence of C-NAGA was amplified In the case where Ser188 of C.-NAGA was replaced with by PCR with a reaction mixture composition and under a Glu and Ala 191 thereof was replaced with Leu, in order to reaction condition described below, using the purchased confirm the effect on the peripheral three-dimensional struc C.-NAGA cDNA as a template with primers described below ture, a mutant C-NAGA (C-NAGA S188E/A191L) model 40 and KOD-plus-polymerase (Toyobo Co., Ltd.). was constructed, and the three-dimensional structure of the mutant C.-NAGA model was compared with that of wild C.-NAGA. As a result, it was confirmed that the above replace NAGA-5' primer: ments did not affect the three-dimensional structure com 5 - GATGCTGCTGAAGACAGTGCTCTT-3 (sequence No. 5) posed of peripheral amino acid residues. Accordingly, it was 45 NAGA-3' primer: supposed that the mutant C.-NAGA in which these mutations s" - TCACTGCTGGGACATCTCCAGGTT-3' (sequence No. 6) were introduced into human C.-NAGA can exist without prob lems in terms of the three-dimensional structure. In addition, a complex model in which a Substrate of C-GAL was fitted into the structure of the mutant C.-NAGA 50 was constructed. As a result, it was confirmed that the side Template (10 ng LL): 1 LL chain of Glu188 of the mutant C.-NAGA exists within a dis 10 x buffer: 10 IL tance in which the side chain of Glu188 can form a hydrogen 2.5 nM (NTP: 10 IL 25 mM MgSO: 4 I.L. bond with the hydroxyl group at the 2-position of the sub KOD-plus-polymerase: 1 LL strate (refer to FIG. 6(b)). Furthermore, in a complex model in 55 NAGA-5' primer (10 M): 1 LL which a substrate of C-NAGA is fitted into the structure of the NAGA-3' primer (10 M): 1 LL mutant C.-NAGA, it was Supposed that the N-acetyl group at Sterilized water: 68 IL the 2-position of the substrate causes a steric hindrance with the side chain of Leu191, and thus, the complex model has a Total: 100 IL structure to which the substrate cannot be bound. 60 According to the above results, it was expected that the mutant C.-NAGA loses the specificity to the original substrate The reaction mixture was heated at 94°C. for two minutes. of C.-NAGA and acquires a high specificity to the substrate of Subsequently, a cycle consisting of “thermal denaturation and C-GAL (that is, the mutant C.-NAGA substantially loses dissociation: 94° C. (15 seconds)->annealing: 60° C. (30 C.-NAGA activity and acquires C.-GAL activity). 65 seconds)->synthesis and extension: 68°C. (90 seconds) was The structure of the constructed mutant C.-NAGA performed a total of 35 times, and the reaction mixture was (C-NAGAS188E/A191L) is shown in FIG. 7. then cooled at 4°C. US 8,323,640 B2 29 30 The prepared C-NAGA DNA fragment was purified by serine (S) was replaced with glutamic acid (E) was intro agarose gel electrophoresis. duced, a NAGA 5188E-GT-3' primer, and KOD-plus-poly An O-NAGA DNA fragment whose ends were phosphory CaS. lated with T4 polynucleotide kinase (NEB) was ligated with a retrovirus vector pCX4Neo prepared by cleaving with a 5 NAGA S188E-GT-5' primer: restriction enzyme HpaI(Blant end) (NEB) and then dephos (sequence No. 8) phorylating using Alkaline Phosphatase, Calf Intestine s' - CCCATCGCCTTCTCCTGCGAGTGGCCAGCCTATGA-3' (NEB) (Tsuyoshi Akagi et al., Proc. Natl. Acad. Sci. USA, 100, 13567-13572 (2003)). C-NAGA pCX4Neo obtained by NAGA S188E-GT-3' primer: the ligation reaction was transformed into DH5a competent 10 (sequence No. 9) cells (Invitrogen Corporation) and seeded on an amplicillin 5 " - GCAGGAGAAGGCGATGGGGCGGCCTGTG-3' containing LB plate. Ampicillin-resistant colonies were then obtained. The resulting each of the resistant colonies was suspended in an LB medium. A colony PCR was performed with a 15 reaction mixture composition and under a reaction condition Template (6 ng LL): 1 LL 10 x buffer: 5 L described below, using the bacterial Suspension as a template 2.5 nM (NTP: 5 L with primers below and PCR Master Mix (manufactured by 25 mM MgSO: 2 LL Promega). KOD-plus-polymerase: 1 LL NAGAS188E-GT-5' primer (10 M): 1 LL NAGAS188E-GT-3' primer (10 M): 1 LL NAGA-5' primer: Sterilized water: 34 I.L. 5 - GATGCTGCTGAAGACAGTGCTCTT-3 (sequence No. 5) Total: 50 L pCX4-3' primer 25 s' - AAACCGTTGCTAGCTTAAGTT-3' (sequence No. 7) The reaction mixture was heated at 94°C. for two minutes. Subsequently, a cycle consisting of “thermal denaturation and 30 dissociation: 94° C. (15 seconds)->annealing: 60° C. (30 Template (1 colony/10 L): 1 LL seconds)->synthesis and extension: 68°C. (8 minutes) was PCR Master Mix: 10 IL NAGA-5'primer (10 M): 0.5 L performed a total of 35 times, and the reaction mixture was pCX4-3' primer (10 M): 0.5 L then cooled at 4°C. Sterilized water: 8 IL 35 The amplified DNA fragment (O-NAGA S188E Total: 20 IL pCX4Neo) was transformed into DH5O-T1 competent cells (Invitrogen Corporation) having McrBC endonuclease which cleaves methylated DNA. Since C.-NAGA pCX4Neo, which The reaction mixture was heated at 95°C. for two minutes. was used as a template, had been methylated, C-NAGA Subsequently, a cycle consisting of “thermal denaturation and 40 pCX4Neo was cleaved by McrBC endonuclease and could dissociation: 95° C. (30 seconds)->annealing: 55° C. (30 not form colonies. On the other hand, since a plasmid having seconds)->synthesis and extension: 72°C. (90 seconds) was an S188E mutation had not been methylated, the plasmid was performed a total of 40 times, and the reaction mixture was not cleaved and could form colonies. The formed several then cooled at 4°C. colonies were then cultured, and the plasmid DNA was then A clone in which C-NAGA DNA was incorporated in the 45 extracted and purified. The introduction of the S188E muta forward direction was selected from the resulting amplified tion was confirmed by a known method of determining a base product. More specifically, an E. coli template in which an sequence using a sequencer. amplified fragment of 1.4 kb was obtained was selected as a Next, an O-NAGAS188E/A191L mutant was prepared by clone in which the O.-NAGA DNA was incorporated in the amplification by PCR with a reaction mixture composition forward direction. The selected E. Coli clone of C-NAGA 50 pCX4Neo was cultured to obtain a large amount, i.e., 1 mg or and under a reaction condition described below, using the more (1 mg/mL), of C-NAGA pCX4Neo plasmid DNA. purified C.-NAGA S188E pCX4Neo as a template with a 2. Preparation of C-NAGA mutant NAGA A191L-GT-5' primer (a portion into which an A191L Regarding C-NAGAS188E/A191L, which is an O-NAGA missense mutation was introduced being underlined) which mutant, first, C-NAGA S188E was prepared, and O-NAGA 55 was designed such that the missense mutation (A191L) in S188E/A191L was then prepared using C-NAGA S188E. which the 191st alanine (A) was replaced with leucine (L) C-NAGA S188E/A191L was prepared with reference to the was introduced, a NAGA A 191L-GT-3' primer, and KOD instruction manual of the GeneTailor Site-Directed Mutagen plus-polymerase. esis System (Invitrogen Corporation), as needed. First, C-NAGA pCX4Neo (100 ng) was methylated with 60 NAGA A191L-GT-5' primer: DNA Methylase (4 U). C-NAGA S188E was prepared by (sequence No. 10) amplifying the DNA with a reaction mixture composition and s' - TTCTCCTGCGAGTGGCCACTCTATGAAGGCGGCCT-3' under a reaction condition described below, using the methy lated C-NAGApCX4Neo as a template with a NAGA5188E NAGA A191L-GT-3' primer: (sequence No. 11) GT-5' primer (a portion into which an S188E missense muta 65 tion was introduced being underlined) which was designed 5'-TGGCCACTCGCAGGAGAAGGCGATGGGG-3' such that the missense mutation (S188E) in which the 188th US 8,323,640 B2 31 32 (iv) The virus solution was removed. Subsequently, 5 mL of a culture medium was added, and culturing was performed for one night. (v) Culturing was performed with a selective medium in Template (6 ng LL): 1 IL 10 x buffer: 5 L which G418 (250 ug/mL) was added to a culture medium. 2.5 nM (NTP: 5 L Thus, G418-resistant F377 cells were established. The selec 25 mM MgSO: 2 LL tive medium was changed once every three days for 14 days KOD-plus-polymerase: 1 IL or more. Whether or not the established cell expressed the NAGA A191L-GT-5' primer (10 M): 1 IL NAGA A191L-GT-3' primer (10 M): 1 IL target protein was confirmed by the enzyme activity and a Sterilized water: 34 I.L. 10 Western blotting method (section 5 below). 5. Confirmation of Expression of Target Protein by Western Total: 50 L Blotting Method In order to examine whether or not the C-NAGA- or C-NAGA S188E/A191L-expressing F377 cells which were The reaction mixture was heated at 94°C. for two minutes. 15 established using retrovirus expressed the target protein, Subsequently, a cycle consisting of “thermal denaturation and Western blotting was performed. An anti-O-NAGA poly dissociation: 94° C. (15 seconds)->annealing: 60° C. (30 clonal antibody obtained by a known method of preparing an seconds)->synthesis and extension: 68°C. (8 minutes) was antibody was prepared as an antibody used in this Western performed a total of 35 times, and the reaction mixture was blotting. then cooled at 4°C. Samples for the Western blotting were prepared as follows. The amplified DNA fragment (C-NAGA S188E/A191L C-NAGA- or C-NAGAS188E/A191L-expressing F377 cells pCX4Neo) was transformed into DH5O-T1 competent cells. cultured in a 60-mm dish were temporarily recovered, and The plasmid DNA was then extracted and purified. The intro resuspended in a Triton-X lysis buffer (50 mM Tris-HCl pH duction of the A191L mutation in addition to the S188E 7.4, 150 mM NaCl, and 1% Triton-X). The suspension was mutation was confirmed by a known method of determining a 25 then Subjected to an ultrasonic treatment, and centrifuged at base sequence using a sequencer. 12,000 rpm for five minutes. The supernatant was recovered 3. Preparation of O-GAL-, C-NAGA-, and C-NAGAS188E/ and used as a sample. SDS-PAGE was performed as follows. A 191L-Expressing Recombinant Retroviruses The concentration of a protein of the sample was measured. Packaging cells of retrovirus (Phoenix Ampho Batchi: Subsequently, an equivalent volume of 2xSDS sample buffer F-14727 Transformed Human Embryonic Kidney HEK293) were purchased from American Type Culture Collection 30 (62.5 mM Tris-HCl pH 6.8, 4% SDS, 30% glycerol, and 0.2% (ATCC) (Coligan, J. E. et al., Curr. Protocols Immunol. bromophenol blue (BPB)) was added to the sample contain Suppl. 31, 10.28.1-10.28.17 (1999)). The Phoenix Ampho ing 5 ug of the protein. The mixture was boiled for five cells were cultured in a Dulbecco's Modified Eagle Medium minutes, and the resulting sample was then applied to a 4% to (DMEM) (High glucose)+10% heat-inactivated fetal bovine 20% gel (PAG mini: Daiichi Pure Chemicals Co., Ltd.). Elec serum (FBS) culture solution at 37° C. and at a CO, concen 35 trophoresis was performed at a constant current of 30 mA for tration of 5%. two hours. In order to prepare recombinant retroviruses, C.-GAL After the electrophoresis, in order to transfer the protein to pCX4Neo-, C-NAGA pCX4Neo-, or C-NAGA S188E/ a polyvinyl difluoride (PVDF) membrane (Immobilon-P, A 191L pCX4Neo-retrovirus vector was transfected into the MILLIPORE), the gel was immersed in a blotting buffer (25 Phoenix Ampho cells. In this transfection, 2 mL of OPTI mM Tris-HCl pH 8.3, 192 mM glycan, and 20% methanol) MEM culture solution (Invitrogen Corporation) containing 1 40 for 20 minutes, and placedonia PVDF membrane equilibrated ug of the retrovirus vector, 1 g of pCLAMP (RK Naviaux et with the blotting buffer. Transfer was then performed using a al., J. Virol.., 70,5701-5705 (1996)), and 18 ul of Dofect-GT1 Hoefer TE 70 semi-dry transfer unit (Amersham Bio (transfection reagent, Dojindo Laboratories) was added to the Sciences) at a constant current of 60 mA for one hour. Phoenix Ampho cells (5x10/60-mm dish), and the mixture After the completion of transfer, the membrane was was incubated at 37° C. for four hours. Subsequently, the 45 blocked with a blocking buffer (5% skim milk in Tris-Buff culture medium was changed to a normal culture medium, ered Saline (TBS) (50 mM Tris-HCl pH 7.4 and 100 mM and the resulting mixture was cultured for 48 hours. After the NaCl)) for 30 minutes. An anti-NAGA polyclonal antibody culturing, the Supernatant was collected and centrifuged at (primary antibody) diluted by 500 times with the blocking 1,000 rpm for 10 minutes. Recombinant retrovirus contained buffer was then added thereto, and incubation was performed in the supernatant was dispensed and stocked at -80°C. 50 at 4°C. for one night. 4. Establishment of Cell Strain which Stably Expresses Fabry The membrane obtained after the incubation with the pri Patient-Derived Fibroblasts (F377) which Express C-GAL-, mary antibody was washed with TBS for five minutes. This O-NAGA-, or O-NAGAS188E/A191L washing was performed three times. An anti-rabbit IgG HRP Each of the O-GAL-, C-NAGA-, and C-NAGA S188E/ labeled antibody (secondary antibody: Amersham Bio A191L-expressing recombinant retroviruses prepared in sec 55 sciences) diluted by 5,000 times with the blocking buffer was tion 3 above was infected in human fibroblasts (F377 cells) then added thereto, and incubation was performed at room derived from a Fabry patient to establish stably expressing temperature for one hour. cells. Specifically, the establishment was achieved by per The membrane obtained after the incubation with the sec forming the following steps (i) to (V): ondary antibody was washed with TBS for five minutes. This (i) 1x10 F377 cells were seeded on a 60-mm dish and washing was performed three times. An enhanced chemilu cultured at 37° C. for one night. 60 minescence (ECL) coloring reagent (Nacalai Tesque, Inc.) (ii) Polybrene (Sigma H-9266, Hexadimethrine Bromide) was added thereto, and reaction was performed at room tem was added to the culture medium So that the final concentra perature for two minutes. Subsequently, the membrane was tion of Polybrene was 2 Lig/mL, and culturing was performed developed by bringing into contact with HyperfilmTM ECL in at 37° C. for 30 minutes. a darkroom for one minute. (iii) The culture medium was removed. Subsequently, 1 65 According to the results, it was confirmed that the estab mL of a virus solution was added and was adsorbed at 37° C. lished C-NAGA-expressing F377 cells and C-NAGAS188E/ for 60 minutes. A 191L-expressing F377 cells expressed wild-type O-NAGA US 8,323,640 B2 33 34 with a molecular weight of about 45 kD and a mutant TABLE 2-continued C-NAGA (C-NAGAS188E/A191L), respectively. Transition of enzyme activity of Cl-NAGA mutant EXAMPLE 3 C-GAL CL-NAGA transfection mutation activity activity Transition of Enzyme Activity of C-NAGA Mutant F377 Fabry disease 2 233 O-NAGA wild-type 13 1295 The fact that the mutant C.-NAGA (C-NAGA S188E/ C-NAGA S188E/A191L, 552 105 A 191L) had acquired the substrate specificity of O-GAL was confirmed by the following procedure. 10 (nmolihrimg protein) First, a gene of wild-type O-NAGA or a gene of C.-NAGA S188E/A191L was introduced into fibroblasts (F377) derived from a Fabry patient, and C-GAL activity and C-NAGA activ EXAMPLE 4 ity were measured. F377 cells were used as a negative control of O-GAL activity, and fibroblasts (F652) derived from a 15 Stability of C.-NAGA Mutant in Blood (in Plasma) patient with Kanzaki disease (C-NAGA deficiency) were used as a negative control of C-NAGA activity. In addition, fibroblasts (F592) derived from a normal subject were used as The stability of the O.-NAGA mutant in blood (in plasma) a positive control of endogenous C.-GAL activity and was confirmed by a procedure below. C.-NAGA activity. First, an enzyme solution of C-NAGAS188E/A191L was The cells cultured in a 60-mm dish were recovered and prepared as in Example 3. In addition, as a control, another resuspended in Milli-Q water. The suspension was then sub enzyme solution was prepared in the same manner as the jected to an ultrasonic treatment to prepare a cell homogenate. above enzyme solution using cells prepared by introducing a This homogenate was used as a sample of an enzyme solu gene of wild-type O-GAL into F377. Plasma (50 uL) of a tion. The enzyme activity was determined using a synthetic 25 normal subject was added to each of the enzyme solutions (50 substrate composed of a 4-methylumbelliferone (4-MU) uL) and mixed. A reaction was then started at 37°C., and 10 derivative, which is a fluorogenic Substrate, by measuring the ul of the reaction mixture was sampled at intervals to measure amount of 4-MU released by 1 mL of the enzyme solution per C-GAL activity. The enzyme activity was measured as in hour as a fluorescence intensity. More specifically, 4-MU-C.- Example 3. The C-GAL activity of a sample sampled at the D-galactoside (4-MU-C-GAL: Calbiochem, CA) was used as 30 time of mixing of the enzyme solution with plasma was the synthetic substrate of C-GAL.4-MU-C-N-acetyl-D-ga defined as the standard (100%), and a decrease in the enzyme lactosaminide (4-MU-O-NAGA. Moscerdam Substrates, activity with time was represented as a percentage. Rotterdam) was used as the synthetic substrate of C-NAGA. In the measurement of O-GAL activity, as an inhibitor of According to the results, C-NAGAS188E/A191L had an O-NAGA, which reacts with 4-MU-O-GAL at the same time, excellent C.-GAL-activity-maintaining ability with time in N-acetyl-D-galactosamine (Sigma, Mo.) was added to the 35 blood (in plasma), as compared with wild-type C-GAL. This Substrate solution in advance so that the final concentration result showed that C.-NAGAS188E/A191L had a high stabil thereof was 117 mM. ity in blood. A Mcllvain's buffer (citric acid/phosphoric acid, pH 4.6, The results are shown in Table 3. In addition, plots of the 60LL) containing 5 mM4-MU-C-GAL or a Mcllvain's buffer results are shown in FIG. 5. (citric acid/phosphoric acid, pH 4.7, 40 LL) containing 1 mM 40 4-MU-O-NAGA was added to an enzyme solution (10 uL) and mixed, and a reaction was performed at 37° C. for 30 TABLE 3 minutes. The reaction was terminated by adding a 0.2 M Tests of stability of C-NAGA mutant in blood Glycine buffer (pH 10.7, 700 uL). In order to detect the amount of 4-MU released, the amount of 4-MU was mea Incubation time Relative C-GAL activity (% sured at an excitation wavelength of 365 nm and at a fluores 45 cence wavelength of 450 nm using a spectrofluorophotom (hr) C-GAL(wild-type) C-NAGA (S188E/A191L) eter. The specific activity of O-GAL or C.-NAGA was O 100 100 determined by dividing by the protein concentration (mg/mL) O.25 23 75 of the enzyme solution. The specific activity was defined as an O.S 7 60 enzyme activity in cell. 50 1 2 42 According to the results of the measurement of the enzyme 2 O 21 activity, C-NAGAS188E/A191L, which had been subjected 3 O 13 to a double mutation of S188E/A191L, exhibited high C-GAL activity. This result showed that C.-NAGA S188E/ A 191L acquired the substrate specificity of O-GAL. 55 The results are shown in Table 2. INDUSTRIAL APPLICABILITY

TABLE 2 According to the present invention, a protein which has O-galactosidase activity and which is advantageous in that no Transition of enzyme activity of C-NAGA mutant 60 allergic adverse side effect is shown, the stability in blood is C-GAL CL-NAGA high, and the protein can be easily incorporated into a cell of transfection mutation activity activity an affected organ can be provided. This protein is very useful as excellent novel highly functional enzyme for the therapy F592 Normal 112 252 control for Fabry disease. F652 Kanzaki 40 O 65 In addition, the present invention provides a gene which disease can encode the above protein, a recombinant vector contain ing the gene, a transformant or transductant containing the US 8,323,640 B2 35 36 recombinant vector, and a method of producing the protein. Sequence No. : synthetic DNA Furthermore, atherapeutic agent for Fabry disease containing Sequence No. : synthetic DNA the protein can be provided. Sequence No. : synthetic DNA Sequence No. : synthetic DNA SEQUENCE LISTING FREETEXT Sequence No. 9: synthetic DNA Sequence No. 3: recombinant DNA Sequence No. 10: synthetic DNA Sequence No. 4: recombinant protein Sequence No. 11: synthetic DNA

SEQUENCE LISTING

NUMBER OF SEO ID NOS: 15

SEO ID NO 1 LENGTH: 1236 TYPE: DNA ORGANISM: homo sapiens FEATURE; NAME/KEY: CDS LOCATION: (1) . . (1236)

< 4 OOs SEQUENCE: 1.

atg Ctg Ctg aag aca gtg citc. ttg Ctg gga Cat gtg gcc cag gtg Ctg 48 Met Lell Lell Lys Thir Wall Luell Lell Lell Gly His Wall Ala Glin Wall Lieu 1. 1O 15

atg Ctg gac aat 999 citc. Ctg Cag a Ca C Ca cc c atg ggc tgg Ctg gcc 96 Met Lell Asp Asn Gly Luell Luell Glin Thir Pro Pro Met Gly Trp Lieu Ala 2O 25 3 O

tgg gaa cgc ttic cgc tgc aac att aac gat gag gac C Ca aag aac 144 Trp Glu Arg Phe Arg Cys ASn Ile Asn Asp Glu Asp Pro Lys Asn 35 4 O 45

tgc ata gala cag citc. titc atg gag atg gct gac cgg atg gca Cag 192 Cys Ile Ser Glu Glin Luell Phe Met Glu Met Ala Asp Arg Met Ala Glin SO 55 60

gat gga tgg cgg gac atg ggc tac a Ca tac Cta aac att gat gac tec 24 O Asp Gly Trp Arg Asp Met Gly Tyr Thir Luell Asn Ile Asp Asp Cys 65 70 7s 8O

tgg atc. ggc ggt cgc gat gcc ggc cgc Ctg atg CC a gat c cc aag 288 Trp Ile Gly Gly Arg Asp Ala Ser Gly Arg Luell Met Pro Asp Pro Llys 85 90 95

cgc tto cott Cat ggc att cott tto Ctg gct gac tac gtt CaC tcc ctd 336 Arg Phe Pro His Gly Ile Pro Phe Lell Ala Asp Wall His Ser Luell 1OO 105 110

ggc Ctg aag ttg ggt at C tac gcg gac atg ggc aac ttic a CC tgc atg 384 Gly Lell Lys Luell Gly Ile Ala Asp Met Gly Asn Phe Thir 115 12O 125

ggt tac C Ca ggc acc aca Ctg gac aag gtg gt C cag gat gct cag acc 432 Gly Tyr Pro Gly Thir Thir Luell Asp Wall Wall Glin Asp Ala Gn. Thir 13 O 135 14 O

titc gcc gag tgg aag gta gac atg citc. aag Ctg gat ggc tgc titc. tcc. Phe Ala Glu Trp Lys Wall Asp Met Lell Lys Luell Asp Gly Cys Phe Ser 145 15 O 155 16 O

acc c cc gag gag cgg gcc cag 999 tac c cc aag atg gct gct gcc Ctg 528 Thir Pro Glu Glu Arg Ala Glin Gly Tyr Pro Met Ala Ala Ala Lieu 1.65 17 O 17s

aat gcc a Ca ggc cgc cc c atc. gcc tto t cc agc tgg C Ca gcc tat 576 ASn Ala Thir Gly Arg Pro Ile Ala Phe Ser Ser Trp Pro Ala Tyr 18O 185 190

gaa ggc ggc citc. cc c CC a agg gtg aac tac Ctg Ctg gcg gac atc 624 Glu Gly Gly Luell Pro Pro Arg Wall Asn Tyr Luell Luell Ala Asp Ile 195 2 OO 2O5

tgc aac citc. tgg aac tat gat gac atc. cag gac to c tgg tgg agc 672 Asn Lell Trp Arg Asn Asp Asp Ile Glin Asp Ser Trp Trp Ser

US 8,323,640 B2 39 40 - Continued

115 12 O 125

Gly Tyr Pro Gly Thir Thir Lell Asp Wall Wall Glin Asp Ala Glin Thir 13 O 135 14 O

Phe Ala Glu Trp Lys Wall Asp Met Luell Lys Luell Asp Gly Phe Ser 145 150 155 160

Thir Pro Glu Glu Arg Ala Glin Gly Pro Met Ala Ala Ala Luell 1.65 17O

Asn Ala Thir Gly Arg Pro Ile Ala Phe Ser Ser Trp Pro Ala Tyr 18O 185 19 O

Glu Gly Gly Luell Pro Pro Arg Wall Asn Ser Lell Lell Ala Asp Ile 195

Asn Luell Trp Arg Asn Tyr Asp Asp Ile Glin Asp Ser Trp Trp Ser 21 O 215

Wall Luell Ser Ile Lell Asn Trp Phe Wall Glu His Glin Asp Ile Luell Glin 225 23 O 235 24 O

Pro Wall Ala Gly Pro Gly His Trp Asn Asp Pro Asp Met Luell Luell Ile 245 250 255

Gly Asn Phe Gly Lell Ser Lell Glu Glin Ser Arg Ala Glin Met Ala Luell 26 O 265 27 O

Trp Thir Wall Luell Ala Ala Pro Luell Luell Met Ser Thir Asp Luell Arg Thir 285

Ile Ser Ala Glin Asn Met Asp Ile Luell Glin ASn Pro Lell Met Ile Lys 29 O 295 3 OO

Ile Asn Glin Asp Pro Lell Gly Ile Glin Gly Arg Arg Ile His Glu 3. OS 310 315

Ser Luell Ile Glu Wall Met Arg Pro Luell Ser Asn Ala Ser 3.25 330 335

Ala Luell Wall Phe Phe Ser Arg Thir Asp Met Pro Arg His 34 O 345 35. O

Ser Ser Luell Gly Glin Lell Asn Phe Thir Gly Ser Wall Ile Glu Ala 355 360 365

Glin Asp Wall Ser Gly Asp Ile Ile Ser Gly Lell Arg Asp Glu Thir 37 O 375

Asn Phe Thir Wall Ile Ile Asn Pro Ser Gly Wall Wall Met Trp Luell 385 390 395 4 OO

Pro Ile Asn Lell Glu Met Ser Glin Glin 4 OS 41O

SEQ ID NO 3 LENGTH: 1236 TYPE: DNA ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: recombinant DNA where 562nd to 564th bases of Wild type alpha-NAGA are replaced with bases representing a codon of glutamic acid and where 571st to 573rd bases of wild type alpha- NAGA are replaced with bases representing a codon of leucine FEATURE: NAME/KEY: CDS LOCATION: (1) ... (1236)

<4 OOs, SEQUENCE: atg ctg. Ctg aag aca gtg citc. ttg Ctg gga Cat gtg gcc cag gtg Ctg 48 Met Lieu. Lieu Lys Thr Wall Lell Luell Luell Gly His Wall Ala Glin Wall Lieu. 1. 5 1O 15 atg Ctg gac aat ggg citc. Ctg cag aca CC a coc atg ggc tgg Ctg gcc 96 Met Lieu. Asp Asn Gly Lell Lell Glin Thir Pro Pro Met Gly Trp Lieu Ala 25

US 8,323,640 B2 43 44 - Continued tcc toc Ctt ggc cag ctgaac ttic acc ggg tot gtg at a tat gag gCC 1104 Ser Ser Leu Gly Glin Lieu. Asn Phe Thr Gly Ser Val Ile Tyr Glu Ala 355 360 365

Cag gac git C tact caggt gac at C at C agt ggc ctic cqa gat gala acc 1152 Glin Asp Val Tyr Ser Gly Asp Ile Ile Ser Gly Lieu. Arg Asp Glu Thir 37 O 375 38O aac ttic aca gtg atc atc aac cct tca ggg gta gtg atg tdg tac Ctg 12 OO Asn Phe Thr Val Ile Ile Asn Pro Ser Gly Val Val Met Trp Tyr Lieu. 385 390 395 4 OO tat coc at C aag aac Ctg gag atg tcc cag cag ta 1236 Tyr Pro Ile Lys Asn Lieu. Glu Met Ser Glin Gln 4 OS 41O

<210s, SEQ ID NO 4 &211s LENGTH: 411 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: recombinant protein which is the amino-acid sequence of the mutant alpha-NAGA (alpha-NAGA S188E/A191L)

<4 OOs, SEQUENCE: 4 Met Lieu. Lieu Lys Thr Val Lieu. Lieu. Lieu. Gly His Val Ala Glin Val Lieu. 1. 5 1O 15 Met Lieu. Asp Asin Gly Lieu. Lieu. Glin Thr Pro Pro Met Gly Trp Lieu Ala 2O 25 3O Trp Glu Arg Phe Arg Cys Asn. Ile Asn. Cys Asp Glu Asp Pro Lys Asn 35 4 O 45 Cys Ile Ser Glu Glin Lieu. Phe Met Glu Met Ala Asp Arg Met Ala Glin SO 55 6 O Asp Gly Trp Arg Asp Met Gly Tyr Thr Tyr Lieu. Asn. Ile Asp Asp Cys 65 70 7s 8O Trp Ile Gly Gly Arg Asp Ala Ser Gly Arg Lieu Met Pro Asp Pro Llys 85 90 95 Arg Phe Pro His Gly Ile Pro Phe Leu Ala Asp Tyr Val His Ser Leu 1OO 105 11 O Gly Lieu Lys Lieu. Gly Ile Tyr Ala Asp Met Gly Asn. Phe Thr Cys Met 115 12 O 125 Gly Tyr Pro Gly Thr Thr Lieu. Asp Llys Val Val Glin Asp Ala Glin Thr 13 O 135 14 O Phe Ala Glu Trp Llys Val Asp Met Lieu Lys Lieu. Asp Gly Cys Phe Ser 145 150 155 160 Thr Pro Glu Glu Arg Ala Glin Gly Tyr Pro Llys Met Ala Ala Ala Lieu. 1.65 17O 17s Asn Ala Thr Gly Arg Pro Ile Ala Phe Ser Cys Glu Trp Pro Leu Tyr 18O 185 19 O Glu Gly Gly Lieu Pro Pro Arg Val Asn Tyr Ser Lieu. Lieu Ala Asp Ile 195 2OO 2O5 Cys Asn Lieu. Trp Arg Asn Tyr Asp Asp Ile Glin Asp Ser Trp Trp Ser 21 O 215 22O Val Lieu. Ser Ile Lieu. Asn Trp Phe Val Glu. His Glin Asp Ile Lieu. Glin 225 23 O 235 24 O Pro Val Ala Gly Pro Gly His Trp Asn Asp Pro Asp Met Lieu. Lieu. Ile 245 250 255 Gly Asn. Phe Gly Lieu. Ser Lieu. Glu Glin Ser Arg Ala Gln Met Ala Lieu. 26 O 265 27 O

Trp Thr Val Lieu Ala Ala Pro Lieu. Lieu Met Ser Thr Asp Lieu. Arg Thr US 8,323,640 B2 45 46 - Continued

27s 28O 285 Ile Ser Ala Glin Asn Met Asp Ile Lieu. Glin Asn Pro Lieu Met Ile Llys 29 O 295 3 OO Ile Asin Glin Asp Pro Lieu. Gly Ile Glin Gly Arg Arg Ile His Lys Glu 3. OS 310 315 32O Llys Ser Lieu. Ile Glu Val Tyr Met Arg Pro Lieu. Ser Asn Lys Ala Ser 3.25 330 335 Ala Leu Val Phe Phe Ser Cys Arg Thr Asp Met Pro Tyr Arg Tyr His 34 O 345 35. O Ser Ser Leu Gly Glin Lieu. Asn Phe Thr Gly Ser Val Ile Tyr Glu Ala 355 360 365 Glin Asp Val Tyr Ser Gly Asp Ile Ile Ser Gly Lieu. Arg Asp Glu Thir 37 O 375 38O Asn Phe Thr Val Ile Ile Asn Pro Ser Gly Val Val Met Trp Tyr Lieu. 385 390 395 4 OO Tyr Pro Ile Lys Asn Lieu. Glu Met Ser Glin Gln 4 OS 41O

<210s, SEQ ID NO 5 &211s LENGTH: 24 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Chemically Synthesized

<4 OOs, SEQUENCE: 5 gatgctgctgaagacagtgc tictt

<210s, SEQ ID NO 6 &211s LENGTH: 24 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Chemically Synthesized

<4 OOs, SEQUENCE: 6 t cactgctgg gaCatctoca ggtt

<210s, SEQ ID NO 7 &211s LENGTH: 21 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Chemically Synthesized

<4 OO > SEQUENCE: 7 aaac cqttgc tagcttaagt t

<210s, SEQ ID NO 8 &211s LENGTH: 35 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: Chemically Synthesized

<4 OOs, SEQUENCE: 8 cc catcgcct tct cotg.cga gtggc.cagcc tatga

<210s, SEQ ID NO 9 &211s LENGTH: 28 &212s. TYPE: DNA <213> ORGANISM: Artificial Sequence 22 Os. FEATURE:

US 8,323,640 B2 51 52 - Continued

Arg Phe Luell Ala Lell Wall Ser Trp Asp Ile Pro Gly Ala Arg Ala Luell 2O 25 3O

Asp Asn Gly Luell Ala Arg Thir Pro Thir Met Gly Trp Lell His Trp Glu 35 4 O 45

Arg Phe Met Asn Lell Asp Glin Glu Glu Pro Asp Ser Ile SO 55 6 O

Ser Glu Luell Phe Met Glu Met Ala Glu Luell Met Wall Ser Glu Gly 65 70

Trp Asp Ala Gly Tyr Glu Tyr Luell Cys Ile Asp Asp Trp Met 85 90 95

Ala Pro Glin Arg Asp Ser Glu Gly Arg Luell Glin Ala Asp Pro Glin Arg 105 11 O

Phe Pro His Gly Ile Arg Glin Luell Ala Asn Tyr Wall His Ser Gly 115 12 O 125

Lell Lys Luell Gly Ile Ala Asp Wall Gly ASn Lys Thir Ala Gly 13 O 135 14 O

Phe Pro Gly Ser Phe Gly Asp Ile Asp Ala Glin Thir Phe Ala 145 150 155 160

Asp Trp Gly Wall Asp Lell Lell Phe Asp Gly Asp Ser 1.65 17O 17s

Lell Glu Asn Luell Ala Asp Gly Tyr Lys His Met Ser Lell Ala Luell Asn 18O 185 19 O

Arg Thir Gly Arg Ser Ile Wall Tyr Ser Glu Trp Pro Luell Met 195 2O5

Trp Pro Phe Glin Pro Asn Tyr Thir Glu Ile Arg Glin Asn 21 O 215

His Trp Arg Asn Phe Ala Asp Ile Asp Asp Ser Trp Ser Ile Lys 225 23 O 235 24 O

Ser Ile Luell Asp Trp Thir Ser Phe Asn Glin Glu Arg Ile Wall Asp Wall 245 250 255

Ala Gly Pro Gly Gly Trp Asn Asp Pro Asp Met Lell Wall Ile Gly Asn 26 O 265 27 O

Phe Gly Luell Ser Trp Asn Glin Glin Wall Thir Glin Met Ala Luell Trp Ala 285

Ile Met Ala Ala Pro Lell Phe Met Ser Asn Asp Lell Arg His Ile Ser 29 O 295 3 OO

Pro Glin Ala Ala Lell Lell Glin Asp Asp Wall Ile Ala Ile Asn 3. OS 310 315

Glin Asp Pro Luell Gly Lys Glin Gly Tyr Glin Luell Arg Glin Gly Asp Asn 3.25 330 335

Phe Glu Wall Trp Glu Arg Pro Luell Ser Gly Luell Ala Trp Ala Wall Ala 34 O 345 35. O

Met Ile Asn Arg Glin Glu Ile Gly Gly Pro Arg Ser Tyr Thir Ile Ala 355 360 365

Wall Ala Ser Luell Gly Gly Wall Ala ASn Pro Ala Phe Ile 37 O 375

Thir Glin Luell Luell Pro Wall Arg Luell Gly Phe Glu Trp Thir 385 390 395 4 OO

Ser Arg Luell Arg Ser His Ile Asn Pro Thir Gly Thir Wall Luell Luell Glin 4 OS 41O 415

Lell Glu Asn Thir Met Glin Met Ser Luell Asp Lell Lell 42O 425

<210s, SEQ ID NO 14 US 8,323,640 B2 53 54 - Continued

&211s LENGTH: 45 &212s. TYPE: DNA <213> ORGANISM: homo sapiens 22 Os. FEATURE: <221s NAME/KEY: CDS <222s. LOCATION: (1) ... (45)

<4 OOs, SEQUENCE: 14 atgtct gca citt Ctg atc cta gct Ctt gtt gga gct gca gtt gct 45 Met Ser Ala Lieu Lieu. Ile Lieu Ala Lieu Val Gly Ala Ala Val Ala 1. 5 1O 15

SEO ID NO 15 LENGTH: 15 TYPE PRT ORGANISM: homo sapiens

< 4 OOs SEQUENCE: 15 Met Ser Ala Lieu. Lieu. Ile Lieu Ala Lieu Val Gly Ala Ala Wall Ala 1. 5 1O 15

The invention claimed is: 5. An isolated transformed host cell comprising the recom 1. An isolated protein, comprising: binant vector according to claim 4. the amino-acid sequence of SEQ NO:2 except that the 25 6. A method of producing a protein having O-galactosidase 188th amino acid is substituted with glutamic acid or activity comprising culturing the trans formant according to claim 5, and aspanic acid and the 191st amino acid is Substituted, collecting the protein having O-galactosidase activity from with leueine, Valine or isoleueinc, the resulting cultured product. wherein, in the amino-acid sequence, one to ten amino 30 7. A pharmaceutical composition comprising the protein acids other than the 188th amino acid and the 191st according to claim 1. amino acid are deleted, replaced, or added, 8. A therapeutic agent for Fabry disease comprising the wherein said Isolated protein has C-galactosidase activity, composition according to claim 7 as an active ingredient. and 9. A pharmaceutical composition comprising the gene wherein at least one of the following is not deleted, 35 according to claim 3. replaced or added: 10. A gene therapeutic agent for Fabry disease comprising the amino acid at the 45th position, and the composition according to claim 9 as an active ingredient. the amino acid at the 350th position. 11. A method of treating Fabry disease, wherein the phar 2. The isolated protein according to claim 1, wherein the maceutical composition according to claim 7 is administered 188th amino acid is substituted with glutamic acid, and the 40 to a Fabry patient. 191st amino acid is substituted with leucine. 12. The isolated protein of claim 1, wherein one to five 3. A isolated gene encoding the protein according to claim amino acids other than the 188th amino acids and the 191st 1. 4. A recombinant vector comprising the isolated gene amino acid are depleted, replaced, or added. according to claim 3. k k k k k UNITED STATES PATENT AND TRADEMARK OFFICE CERTIFICATE OF CORRECTION PATENT NO. : 8,323,640 B2 Page 1 of 1 APPLICATIONNO. : 13/052632 DATED : December 4, 2012 INVENTOR(S) : Hitoshi Sakuraba et al. It is certified that error appears in the above-identified patent and that said Letters Patent is hereby corrected as shown below:

On the Title page, Item (73): Change “(73) Assignee: Tokyo Metropolitan Organization For Medical Research (Tokyo, JP) Altif Laboratories (Tokyo, JP)

To be

--(73) Assignee: Tokyo Metropolitan 9rganizatieh Fer-Institute of Medical-Researeh Science (Tokyo, JP) Altif Laboratories (Tokyo, JP)--.

Signed and Sealed this Sixteenth Day of April, 2013

Teresa Stanek Rea Acting Director of the United States Patent and Trademark Office UNITED STATES PATENT AND TRADEMARK OFFICE CERTIFICATE OF CORRECTION PATENT NO. : 8,323,640 B2 Page 1 of 1 APPLICATIONNO. : 13/052632 DATED : December 4, 2012 INVENTOR(S) : Hitoshi Sakuraba et al. It is certified that error appears in the above-identified patent and that said Letters Patent is hereby corrected as shown below:

In the Claims

Column 53, Claim 1, Line 29: Change

“with leueine, Valine or isoleueinc. To be --with leucine, Valine or isoleucine.--

Column 53, Claim 1, Line 33: Change "wherein said Isolated protein has 0-galactosidase activity, To be --Wherein Said isolated protein has O-galactosidase activity.--

Signed and Sealed this Nineteenth Day of November, 2013

Teresa Stanek Rea Deputy Director of the United States Patent and Trademark Office