MODIFIED LYSOSOMAL PROTEIN AND PRODUCTION THEREOF

Abstract

Disclosed herein are a modified lysosomal protein, methods for preparing a modified lysosomal protein and therapeutic use of such a modified protein. Further disclosed herein is a method of treating a mammal afflicted with a lysosomal storage disease. In particular, the present disclosure relates to a method of preparing a modified lysosomal protein, said method comprising reacting a glycosylated lysosomal protein with an alkali metal periodate and reacting said lysosomal protein with an alkali metal borohydride for a time period of no more than 2 h, thereby modifying glycan moieties of the lysosomal protein and reducing the activity of the lysosomal protein with respect to glycan recognition receptors.

Claims

1. A method of preparing a modified lysosomal protein, said method comprising: a) reacting a glycosylated lysosomal protein with an alkali metal periodate for a time of no more than 4 h; and b) reacting said lysosomal protein with an alkali metal borohydride for a time period of no more than 2 h; thereby modifying glycan moieties of the lysosomal protein and reducing the activity of the lysosomal protein with respect to glycan recognition receptors, provided that said protein is not sulfamidase.

2. The method of claim 1, wherein step b) is further characterized by at least one of: i) said alkali metal borohydride is sodium borohydride; ii) said borohydride is used at a concentration of no more than 4 times the concentration of said periodate; iii) said reaction is performed for a time period of no more than 2 h; and iv) said reaction is performed at a temperature of between 0 and 8 C.

3. The method according to claim 1, wherein step a) is further characterized by at least one of: i) said alkali metal periodate is sodium meta-periodate; ii) said periodate is used at a concentration of no more than 20 mM; iii) said reaction is performed at a temperature of between 0 and 22 C.; iv) said reaction is performed for a time period of no more than 3 h; and v) said reaction of step a) is performed at a pH of 3-7.

4. The method according to claim 1, wherein step a) is performed for a time period of no more than 3 h and step b) is performed for no more than 1 h, and said borohydride optionally is used at a concentration of no more than 4 times the concentration of said periodate.

5. The method according to claim 1, wherein step a) and step b) are performed in sequence without performing any dialysis, ultrafiltration, precipitation or buffer exchange.

6. A method of preparing a modified lysosomal protein, said method comprising: a) reacting a glycosylated lysosomal protein with an alkali metal periodate, and b) reacting said lysosomal protein with an alkali metal borohydride; thereby modifying glycan moieties of the lysosomal protein and reducing the activity of the lysosomal protein with respect to glycan recognition receptors, wherein the active site or functional epitope of said lysosomal protein is made inaccessible to oxidative and/or reductive reactions during at least one of steps a) and b).

7. The method according to claim 6, wherein step b) is further characterized by at least one of: i) said alkali metal borohydride is sodium borohydride; ii) said borohydride is used at a concentration of no more than 4 times the concentration of said periodate; iii) said reaction is performed for a time period of no more than 2 h; and iv) said reaction is performed at a temperature of between 0 and 8 C.

8. The method according to claim 6, wherein step a) is further characterized by at least one of: i) said alkali metal periodate is sodium meta-periodate; ii) said periodate is used at a concentration of no more than 20 mM; iii) said reaction is performed at a temperature of between 0 and 22 C.; iv) said reaction is performed for a time period of no more than 3 h; and v) said reaction of step a) is performed at a pH of 3-7.

9. The method according to claim 6, wherein step a) is performed for a time period of no more than 3 h and step b) is performed for no more than 1 h, and said borohydride optionally is used at a concentration of no more than 4 times the concentration of said periodate.

10. The method according to claim 6, wherein step a) and step b) are performed in sequence without performing any dialysis, ultrafiltration, precipitation or buffer exchange.

11. The method according to claim 1, wherein said modified lysosomal protein is a sulfatase; a glycoside hydrolase, or a protease.

12. The method according to claim 1, wherein the lysosomal protein is selected from deoxyribonuclease-2-alpha; beta-mannosidase; ribonuclease T2; lysosomal alpha-mannosidase; alpha L-iduronidase; tripeptidyl-peptidase 1; hyaluronidase-3; cathepsin L2; ceroid-lipofuscinosis neuronal protein 5; glucosylceramidase; tissue alpha-L-fucosidase; myeloperoxidase; alpha-galactosidase A; beta-hexosaminidase subunit alpha; cathepsin D; prosaposin; beta-hexosaminidase subunit beta; cathepsin L1; cathepsin B; beta-glucuronidase; pro-cathepsin H; non-secretory ribonuclease; lysosomal alpha-glucosidase; lysosomal protective protein; gamma-interferon-inducible lysosomal thiol reductase; tartrate-resistant acid phosphatase type 5; arylsulfatase A; prostatic acid phosphatase; N-acetylglucosamine-6-sulfatase; arylsulfatase B; beta-galactosidase; alpha-N-acetylgalactosaminidase; sphingomyelin phosphodiesterase; ganglioside GM2 activator; N(4)-(beta-N-acetylglucosaminyl)-L-asparaginase; iduronate 2-sulfatase; cathepsin S; N-acetylgalactosamine-6-sulfatase; lysosomal acid lipase/cholesteryl ester hydrolase; lysosomal Pro-X carboxypeptidase; cathepsin O; cathepsin K; palmitoyl-protein thioesterase 1; arylsulfatase D; dipeptidyl peptidase 1; alpha-N-acetylglucosaminidase; galactocerebrosidase; epididymal secretory protein E1; di-N-acetylchitobiase; N-acylethanolamine-hydrolyzing acid amidase; hyaluronidase-1; chitotriosidase-1; acid ceramidase; phospholipase B-like 1; proprotein convertase subtilisin/kexin type 9; group XV phospholipase A2; putative phospholipase B-like 2; deoxyribonuclease-2-beta; gamma-glutamyl hydrolase; arylsulfatase G; L-amino-acid oxidase; sialidase-1; legumain; sialate O-acetylesterase; thymus-specific serine protease; cathepsin Z; cathepsin F; prenylcysteine oxidase 1; dipeptidyl peptidase 2; lysosomal thioesterase PPT2; heparanase; carboxypeptidase Q; -glucuronidase, and sulfatase-modifying factor 1.

13. The method according to claim 1, wherein at least one of steps a) and b) of the method is/are performed in the presence of a protective ligand.

14. The method according to claim 1 wherein steps a) and b) of the method are performed while the lysosomal protein is immobilized on a resin.

15. A modified lysosomal protein having a reduced content of unmodified glycan moieties, characterized in that no more than 50% of the glycan moieties remain unmodified as compared to an unmodified form of the lysosomal protein, said protein thereby having a reduced activity for glycan recognition receptors, provided that said protein is not sulfamidase, -glucuronidase, tripeptidyl peptidase 1 (TPP1) or alpha L-iduronidase.

16. The modified lysosomal protein according to claim 15, said protein being selected from deoxyribonuclease-2-alpha; beta-mannosidase; ribonuclease T2; lysosomal alpha-mannosidase; hyaluronidase-3; cathepsin L2; ceroid-lipofuscinosis neuronal protein 5; glucosylceramidase; tissue alpha-L-fucosidase; myeloperoxidase; alpha-galactosidase A; beta-hexosaminidase subunit alpha; cathepsin D; prosaposin; beta-hexosaminidase subunit beta; cathepsin L1; cathepsin B; pro-cathepsin H; non-secretory ribonuclease; lysosomal alpha-glucosidase; lysosomal protective protein; gamma-interferon-inducible lysosomal thiol reductase; tartrate-resistant acid phosphatase type 5; arylsulfatase A; prostatic acid phosphatase; N-acetylglucosamine-6-sulfatase; arylsulfatase B; beta-galactosidase; alpha-N-acetylgalactosaminidase; sphingomyelin phosphodiesterase; ganglioside GM2 activator; N(4)-(beta-N-acetylglucosaminyl)-L-asparaginase; iduronate 2-sulfatase; cathepsin S; N-acetylgalactosamine-6-sulfatase; lysosomal acid lipase/cholesteryl ester hydrolase; lysosomal Pro-X carboxypeptidase; cathepsin O; cathepsin K; palmitoyl-protein thioesterase 1; arylsulfatase D; dipeptidyl peptidase 1; alpha-N-acetylglucosaminidase; galactocerebrosidase; epididymal secretory protein E1; di-N-acetylchitobiase; N-acylethanolamine-hydrolyzing acid amidase; hyaluronidase-1; chitotriosidase-1; acid ceramidase; phospholipase B-like 1; proprotein convertase subtilisin/kexin type 9; group XV phospholipase A2; putative phospholipase B-like 2; deoxyribonuclease-2-beta; gamma-glutamyl hydrolase; arylsulfatase G; L-amino-acid oxidase; sialidase-1; legumain; sialate O-acetylesterase; thymus-specific serine protease; cathepsin Z; cathepsin F; prenylcysteine oxidase 1; dipeptidyl peptidase 2; lysosomal thioesterase PPT2; heparanase; carboxypeptidase Q, and sulfatase-modifying factor 1.

17. The modified lysosomal protein according to claim 15, wherein no more than 45% of the glycan moieties remain unmodified compared to an unmodified form of the lysosomal protein.

18. The modified lysosomal protein according to claim 15, wherein unmodified glycan moieties of said lysosomal protein are disrupted by single bond breaks and double bond breaks, the extent of single bond breaks being at least 60% in oligomannose glycans.

19. The modified lysosomal protein according to claim 15, wherein said unmodified glycan moieties are absent from at least one N-glycosylation site of said lysosomal protein.

20. The modified lysosomal protein according to claim 15, wherein said lysosomal protein has retained catalytic activity of that of the corresponding unmodified lysosomal protein.

21. A modified lysosomal protein obtainable by the method according to claim 1, provided that said protein is not sulfamidase.

22. (canceled)

23. (canceled)

24. A method of treating a mammal afflicted with a lysosomal storage disease comprising administering to the mammal a therapeutically effective amount of a modified lysosomal protein according to claim 15.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0130] FIG. 1 is a picture outlining the differences between the methods for chemical modification developed by the inventors, disclosed in Example 3, and the known method, disclosed in WO 2008/109677.

[0131] FIG. 2A shows a SDS-PAGE gel of sulfamidase (lane 1), sulfamidase modified according to the known method (lane 2), iduronate 2-sulfatase (lane 3) and iduronate 2-sulfatase modified according to the known method (lane 4), alpha-L-iduronidase (lane 5) and alpha-L-iduronidase modified according to the known method (lane 6). Four protein bands, denoted 1-4, generated by the glycan modification procedure of sulfamidase were identified (lane 2).

[0132] FIG. 2B shows SDS-PAGE gels of sulfamidase, iduronate 2-sulfatase and alpha-L-iduronidase modified according to the known method (lane 1, 3 and 5) and modified according to the new methods disclosed herein (lane 2, 4, 6, 7 and 8).

[0133] FIG. 3A shows a SEC chromatogram of sulfamidase modified according to the known method.

[0134] FIG. 3B shows a SEC chromatogram of sulfamidase modified according to new method 1 as disclosed herein. Marked by an arrow is the peak of multimeric forms of modified sulfamidase.

[0135] FIG. 4A shows scattering intensity measured by dynamic light scattering of sulfamidase modified according to the known method.

[0136] FIG. 4B shows scattering intensity measured by dynamic light scattering of sulfamidase modified according to new method 1 as described herein.

[0137] FIG. 5 is a diagram visualizing the receptor mediated endocytosis in MEF-1 cells of unmodified recombinant sulfamidase, sulfamidase modified according to the known method and sulfamidase modified according to new method 1 and 4 as described herein.

[0138] FIG. 6A shows the results from in vivo treatment of MPS IIIA deficient mice. The diagram shows clearance of heparan sulfate storage in the brain of mice after i.v. dosing every other day (13 doses) of sulfamidase modified according to new method 1 at 30 mg/kg.

[0139] FIG. 6B shows the results from in vivo treatment of MPS IIIA deficient mice. The diagram shows clearance of heparan sulfate storage in the liver of mice after i.v. dosing every other day (13 doses) of sulfamidase modified according to new method 1 at 30 mg/kg.

[0140] FIG. 6C shows the results from in vivo treatment of MPS IIIA deficient mice. The diagram shows clearance of heparan sulfate storage in the brain of mice after i.v. dosing once weekly (10 doses) of sulfamidase modified according to new method 1 at 30 mg/kg and 10 mg/kg, respectively.

[0141] FIG. 7 is a schematic drawing of the three archetypal N-glycan structures generally present in proteins of mammalian origin and the typical N-glycan present in yeast proteins. The left glycan represents the oligomannose type, the second from the left the complex type, and the second from the right the hybrid type. The one on the far right is the polymannose type of yeast proteins. In the Figure the following compounds are depicted: black filled diamonds correspond to N-acetylneuraminic acid; black filled circles correspond to mannose; squares correspond to N-acetylglucosamine; black filled triangle corresponds to fucose; circle corresponds to galactose. Sugar moieties marked with an asterisk can be modified by periodate/borohydride treatment disclosed herein.

[0142] FIG. 8A is a schematic drawing illustrating predicted bond breaks on mannose after chemical modification.

[0143] FIG. 8B is a schematic drawing illustrating a model of a Man-6 glycan. The sugar moieties suscpetible to bond breaks upon oxidation with periodate are indicated. Grey circles correspond to mannose, black squares correspond to N-acetylglucosamine, T13 corresponds to the tryptic peptide NITR of sulfamidase (SEQ ID NO:44) with the N-glycosylation site N(131) included.

[0144] FIG. 9 represents mass spectra of doubly charged ions corresponding to tryptic peptide T13 of sulfamidase (SEQ ID NO:44) with Man-6 glycan attached to N(131) (T13+Man-6 glycan), prior to (A) and after chemical modification (B-D) according to previously known method (S.=single bond breaks; D.=double bond breaks; e.g. D.x3=3 double bond breaks).

[0145] FIG. 10A is a diagram visualizing the extent of bond breaking of the tryptic peptide T13+Man-6 glycan after chemical modification of sulfamidase (SEQ ID NO:44) according to the previously known method (black bar), new method 1 (black dots), new method 3 (white), and new method 4 (cross-checkered).

[0146] FIG. 10B is a diagram visualizing the relative abundance of single bond breaks in the tryptic peptide T13+Man-6 glycan after chemical modification of sulfamidase (SEQ ID NO:44) according to the previously known method (black bar), new method 1 (black dots), new method 3 (white), and new method 4 (cross-checkered).

[0147] FIG. 11 is a diagram showing the activity of iduronate 2-sulfatase as well as iduronate 2-sulfatase modified according to new method 10 and 11.

EXAMPLES

[0148] The Examples which follow disclose the development of modified lysosomal proteins, exemplified by sulfamidase, alpha-L-iduronidase and iduronate 2-sulfatase.

[0149] Materials and Methods

[0150] The recombinant alpha-L-iduronidase used in the Examples below was the medicinal product Aldurazyme whereas the recombinant iduronate 2-sulfatase was the medicinal product Elaprase. Both were purchased from a pharmacy (Apoteket farmaci, Sweden), stored according to the manufacturer's specifications and treated under sterile conditions.

[0151] The sulfamidase was produced by cloning and transient expression in HEK293 cells using the pcDNA3.1(+) vector and in CHO with the Quattromed Cell Factory (QMCF) episomal expression system (Icosagen AS) using the pQMCF1 vector. Sulfamidase was captured from medium by anion exchange chromatography (AlEX) on a Q sepharose column (GE Healthcare) equilibrated with 20 mM Tris, 1 mM EDTA, pH 8.0 and eluted by a NaCl gradient. Captured sulfamidase was further purified by 4-Mercapto-Ethyl-Pyridine (MEP) chromatography; sulfamidase containing fractions were loaded on a MEP HyperCel chromatography column and subsequently eluted by isocratic elution in 50 mM NaAc, 0.1 M NaCl, 1 mM EDTA, 1 mM DTT, pH 4.6. Final polishing was achieved by cation exchange chromatography (CIEX) on a SP Sepharose FF (GE Healthcare) column equilibrated in 25 mM NaAc, 2 mM DTT, pH 4.5. A NaCl gradient was used for elution.

Example 1

Chemical Modification of the Lysosomal Proteins Sulfamidase, Alpha-L-Iduronidase and Iduronate 2-Sulfatase According to Previously Known Method

[0152] Chemical Modification According to WO 2008/109677:

[0153] In order to modify glycan moieties, above mentioned lysosomal proteins were initially incubated with 20 mM sodium meta-periodate at 0 C. for 6.5 h in 20 mM sodium phosphate, 137 mM NaCl (pH 6.0). Glycan oxidation was quenched by addition of ethylene glycol to a final concentration of 192 mM. Quenching was allowed to proceed for 15 min at 0 C. before performing dialysis against 20 mM sodium phosphate, 137 mM NaCl (pH 6.0) over night at 4 C. Following dialysis, reduction was performed by addition of sodium borohydride to the reaction mixture at a final concentration of 100 mM. The reduction reaction was allowed to proceed over night. Finally, enzyme preparations were dialyzed against 20 mM sodium phosphate, 137 mM NaCl (pH 6.0). All incubations were performed in the dark.

Example 2

Analyses of the Lysosomal Proteins Sulfamidase, Alpha-L-Iduronidase and Iduronate 2-Sulfatase Modified According to Known Method

[0154] Material and Methods

[0155] SDS-PAGE Analysis:

[0156] The lysosomal enzymes modified according to the known method as described in Example 1 was subjected to SDS-PAGE analysis with protein loaded on NuPAGE 4-12% Bis-Tris gels. Seeblue 2 plus marker was used for molecular weight calibration and the gels were stained with Instant Blue (C.B.S Scientific).

[0157] Glycan Analysis by LC/MS of Tryptic Fragments:

[0158] The glycosylation patterns were determined by LC/MS of tryptic fragments of the three lysosomal proteins of Example 1. Prior to glycopeptide analysis, proteins were reduced, alkylated and digested with trypsin. Reduction of the protein was done by incubation in 5 l DTT 10 mM in 50 mM NH.sub.4HCO.sub.3 at 60 C. for 1 h (70 C. for alpha-L-iduronidase). Subsequent alkylation with 5 l iodoacetamide 55 mM in 50 mM NH.sub.4HCO.sub.3 was performed at room temperature (RT) and in darkness for 45 min. Lastly, the tryptic digestion was performed by addition of 30 l of 50 mM NH.sub.4HCO.sub.3, 5 mM CaCl.sub.2, pH 8, and 0.2 g/l trypsin in 50 mM acetic acid (protease:protein ratio 1:20 (w/w)). Digestion was allowed to take place over night at 37 C.

[0159] Possible glycosylation variants of the tryptic peptide fragments were investigated by glycopeptide analysis. This was performed by liquid chromatography followed by mass spectrometry (LC-MS) on an Agilent 1200 HPLC system coupled to an Agilent 6510 Quadrupole time-of-flight mass spectrometer (Q-TOF-MS). Both systems were controlled by MassHunter Workstation. LC separation was performed by the use of a Waters XSELECT CSH 130 C18 column (1502.1 mm), the column temperature was set to 40 C. Mobile phase A consisted of 5% acetonitrile, 0.1% propionic acid, and 0.02% TFA, and mobile phase B consisted of 95% acetonitrile, 0.1% propionic acid, and 0.02% TFA. A gradient of from 0% to 10% B for 10 minutes, then from 10% to 70% B for another 25 min was used at a flow rate of 0.2 mL/min. The injection volume was 10 l. The Q-TOF was operated in positive-electrospray ion mode. During the course of data acquisition, the fragmentor voltage, skimmer voltage, and octopole RF were set to 90, 65, and 650 V, respectively. Mass range was between 300 and 2800 m/z.

[0160] The following analyses were conducted only for sulfamidase preparations.

[0161] Dynamic Light Scattering (DLS) Analysis of Sulfamidase:

[0162] The modified sulfamidase was degassed by centrifugation at 12000 rpm for 3 min at room temperature (RT). DLS experiments were performed on a DynaPro Titan instrument (Wyatt Technology Corp) using 25% laser power with 3 replicates of 75 L each.

[0163] Analysis by Size Exclusion Chromatography (SEC) of Sulfamidase:

[0164] The modified enzyme was analyzed by analytical size exclusion chromatography, performed on a AKTAmicro system (GE Healthcare). A Superdex 200 PC 3.2/30 column with a flow rate of 40 L/min of formulation buffer was used. The sample volume was 10 L and contained 10 g enzyme.

[0165] In-Gel Digestion and MALDI-TOF MS Analysis of Sulfamidase:

[0166] The SDS-PAGE analysis revealed some extra bands, which were excised, destained and processed by in-gel digestion with trypsin. Digestion was performed over night at 37 C. The supernatant was transferred to a new tube and extracted with 60% acetonitrile, 0.1% TFA (320 min) at RT. The resulting supernatants were evaporated in a Speed Vac to near dryness. The concentrated solution was mixed 1:1 with alpha-cyano-4-hydroxycinnamic acid solution (10 mg/mL) and 0.6 L was applied on a MALDI plate. Molecular masses of the tryptic peptide fragments were determined using a Sciex 5800 matrix-assisted laser desorption/ionization-time-of-flight mass spectrometer (MALDI-TOF/TOF MS). The analyses were performed in positive ion reflectron mode with a laser energy of 3550 and 400 shots.

[0167] Preservation of Sulfamidase Active Site:

[0168] Any effect of the chemical modification on the active site of sulfamidase was investigated by the use of LC-MS and LC-MS/MS analyses. The samples were prepared according to the LC-MS method described under section Glycosylation analysis. The resulting tryptic peptides containing cysteine 50 variants (cysteine50 (alkylated), oxidized cysteine 50, FGly50 and Ser50) were all semiquantified using peak area calculations from reconstructed ion chromatograms. The identity of the peptides was confirmed by MSMS sequencing. The MSMS parameters were as follows: the collision energies were set to 10, 15, and 20V, scan range 100-1800 m/z, and scan speed 1 scan/sec.

[0169] Results

[0170] As apparent by SDS-PAGE analysis, several major peptides of sizes distinct from that of the full length proteins were formed as a result of the chemical modification according to the known method (FIG. 2A). Peptide bands of lower molecular weight, representing peptide cleavage products are apparent for all three lysosomal proteins, although it was most prominent for sulfamidase. By MALDI-TOF MS analysis, four gel bands #1-4 observed on SDS-PAGE (FIG. 2A, lane 2) could be identified as fragments of sulfamidase generated by strand breaks during the chemical modification. The gel bands #1 and #2 were determined as two C-terminal truncations with molecular masses of 6 kDa and 30 kDa, gel band #3 as one 41 kDa N-terminal truncation.

[0171] It was also found that the chemical modification according the known method introduces oxidation on several methionine residues on sulfamidase, in particular on methionine 184 and methionine 443, which were almost completely oxidized. Methionine 226 (found in a tryptic peptide corresponding to amino acid residues 226-238) was oxidized to a much lower degree, but this oxidation appeared to give rise to a more unstable protein than unmodified sulfamidase as such, generating the 41 kDa N-terminal truncation. Thus, oxidation of methionine 226 and strand breaks seemed to be correlated, as observed in the MS analysis.

[0172] Notably, bands of higher molecular weight were apparent for all three lysosomal proteins indicating covalent multimerisation as a consequence of chemical modification according to the known method. For sulfamidase, the predominant band could be identified as a dimer of a molecular weight of 111 kDa (FIG. 2A, lane 2, band #4). Most severe multimerisation was seen for alpha-L-iduronidase (FIG. 2A, lane 6).

[0173] Thus, it was found that chemical modification of sulfamidase in accordance to the known method (WO 2008/109677) not only modifies glycans but also generates polypeptide strand breaks, covalent multimerisation and oxidation of amino acid residues crucial for structural integrity of the enzyme.

[0174] SDS-PAGE analysis also clearly showed a common lowering of the position of the main monomeric band for all three lysosomal proteins when compared to unmodified protein (FIG. 2A lane 1 vs lane 2; lane 3 vs lane 4; lane 5 vs lane 6). This suggests a loss of molecular weight of approximately 500-1500 Da and is expected for a chemical reaction where glycan moieties predominantly are modified by double bond breaks (FIG. 8).

[0175] Further analysis of sulfamidase by SEC revealed that the chemical modification procedure according to the known method promoted aggregation of sulfamidase, as demonstrated as a pre-peak in the chromatogram of FIG. 3A. The peak height of the pre-peak in the chromatogram was found to be approximately 3% of the height of the main peak. The DLS analysis moreover revealed that the same material contained 15-20% of protein of the total protein content in high molecular weight forms (i.e. above 10.sup.10 kDa) (FIG. 4A).

[0176] Moreover, by the use of LC-MSMS, the reduction step (FIG. 1A) was found to reduce the FGly residue at the active site position 50 of sulfamidase (SEQ ID NO:44) to Ser. Ser in this position is not compatible with efficient catalysis (Recksiek et al, J Biol Chem 273(11):6096-103 (1998)). The relative amount Ser produced from FGly was estimated based on peak area measurements of the doubly charged ions in the mass spectrum, corresponding to the two tryptic peptide fragments containing FGly50 and Ser50. The peak areas were based on MS response without correction for ionization efficiency. Table 2 below shows that the conversion of FGly to Ser is approximately 56% according to the known method (see also Example 4, Table 3).

TABLE-US-00002 TABLE 2 Conversion of FGly to Ser at active site Chemical modification of sulfamidase Ser formation (%) FGly/Ser ratio None 0 WO 2008/109677 56.0 0.3 (n = 3) ca 0.79

[0177] Thus, the known chemical modification procedure, in addition to the modifications mentioned above, causes reduction of an amino acid residue crucial for catalytic activity of sulfamidase. The FGly residue is present in all sulfatases and is crucial for enzymatic activity.

[0178] Glycan analysis by LC/MS of tryptic fragments, confirmed that no natural glycans were present in the lysosomal proteins studied after chemical modification, indicative of complete modification of the glycans.

Example 3

New Methods for Chemical Modification of the Lysosomal Enzymes Sulfamidase, Alpha-L-Iduronidase and Iduronate 2-Sulfatase

[0179] Chemical Modification According to New Method 1:

[0180] The above mentioned lysosomal proteins were initially incubated at 20 mM sodium meta-periodate at 0 C. in the dark for 120 min in phosphate buffers having a pH of 6.0. Glycan oxidation was quenched by addition of ethylene glycol to a final concentration of 192 mM. Quenching was allowed to proceed for 15 min at 6 C. before sodium borohydride was added to the reaction mixtures to a final concentration of 50 mM. After incubation at 0 C. for 120 min in the dark, the resulting protein preparations were ultrafiltrated against 20 mM sodium phosphate, 100 mM NaCl, pH 6.0. The new method 1 for chemical modification is depicted in FIG. 1B.

[0181] Chemical Modification According to New Method 2:

[0182] The above mentioned lysosomal proteins were initially incubated at 15 mM sodium meta-periodate at 0 C. for 0.5 h in 20 mM sodium phosphate, 137 mM NaCl (pH 6.0). Glycan oxidations were quenched by addition of ethylene glycol to a final concentration of 96 mM. Quenching was allowed to proceed for 15 min at 0 C. Thereafter sodium borohydride was added to the reaction mixtures to a final concentration of 38 mM and the resulting mixtures were held at 0 C. for 0.5 h. Finally, the enzyme preparations were ultrafiltrated against 20 mM sodium phosphate, 137 mM NaCl (pH 6.0). All incubations were performed in the dark. The new method 2 for chemical modification is depicted in FIG. 1C.

[0183] Chemical Modification According to New Method 3:

[0184] The above mentioned lysosomal proteins were initially incubated at 10 mM sodium meta-periodate at 0 C. for 0.5 h in 20 mM sodium phosphate, 137 mM NaCl (pH 6.0). Glycan oxidations were quenched by addition of ethylene glycol to a final concentration of 96 mM. Quenching was allowed to proceed for 15 min at 0 C. Thereafter sodium borohydride was added to the reaction mixtures to a final concentration of 15 mM and the resulting mixtures were held at 0 C. for 1 h. Finally, the enzyme preparations were ultrafiltrated against 20 mM sodium phosphate, 137 mM NaCl (pH 6.0). All incubations were performed in the dark. The new method 3 for chemical modification is depicted in FIG. 1D.

[0185] Here follow examples of new methods evaluated and exemplified with one specific lysosomal enzyme.

[0186] New Method 4:

[0187] Exemplified for sulfamidase. Performed as New method 1 with the exception that the concentration of sodium borohydride in the reduction step was 10 mM.

[0188] New Method 5:

[0189] Exemplified for sulfamidase. Sulfamidase was oxidized by incubation with 10 mM sodium meta-periodate at 0 C. in the dark for 180 min in acetate buffer having an initial pH of between 4.5 to 6. Glycan oxidation was quenched by addition of ethylene glycol to a final concentration of 192 mM. Quenching was allowed to proceed for 15 min at 6 C. before sodium borohydride was added to the reaction mixture to a final concentration of 25 mM. After incubation at 0 C. for 60 min in the dark, the resulting sulfamidase preparation was ultrafiltrated against 10 mM sodium phosphate, 100 mM NaCl, pH 7.4.

[0190] New Method 6:

[0191] Exemplified for sulfamidase. Sulfamidase was oxidized by incubation with 10 mM sodium meta-periodate at 8 C. in the dark for 60 min in acetate buffer having an intial pH of 4.5. Glycan oxidation was quenched by addition of ethylene glycol to a final concentration of 192 mM. Quenching was allowed to proceed for 15 min at 6 C. before sodium borohydride was added to the reaction mixture to a final concentration of 25 mM. After incubation at 0 C. for 60 min in the dark, the resulting sulfamidase preparation was ultrafiltrated against 10 mM sodium phosphate, 100 mM NaCl, pH 7.4.

[0192] New Method 7:

[0193] Exemplified for sulfamidase. Sulfamidase was oxidized by incubation with 10 mM sodium meta-periodate at 8 C. in the dark for 60 min in acetate buffer having an intial pH of 4.5. Glycan oxidation was quenched by addition of ethylene glycol to a final concentration of 192 mM. Quenching was allowed to proceed for 15 min at 6 C. before sodium borohydride was added to the reaction mixture to a final concentration of 25 mM. After incubation at 0 C. for 30 min in the dark, the resulting sulfamidase preparation was ultrafiltrated against 10 mM sodium phosphate, 100 mM NaCl, pH 7.4.

[0194] New Method 8:

[0195] Exemplified for alpha-L-iduronidase. Alpha-L-iduronidase was initially incubated at 15 mM sodium meta-periodate at 0 C. for 20 min in 20 mM sodium phosphate, 137 mM NaCl (pH 6.0). Glycan oxidation was quenched by addition of ethylene glycol to a final concentration of 96 mM. Quenching was allowed to proceed for 15 min at 0 C. Thereafter sodium borohydride was added to the reaction mixture to a final concentration of 37 mM and the resulting mixture was held at 0 C. for 1 h. Finally, the enzyme preparation was ultrafiltrated against 20 mM sodium phosphate, 137 mM NaCl (pH 6.0). All incubations were performed in the dark.

[0196] New Method 9:

[0197] Exemplified for alpha-L-iduronidase. Reaction conditions were as described for new method 8, with the single exception that periodate oxidation was performed in the presence of 100 M 4-methylumbeliferone iduronide, functioning as a protecting ligand during the oxidation step.

[0198] Results

[0199] As already accounted for elsewhere herein, sodium meta-periodate is an oxidant that converts cis-glycol groups of carbohydrates to aldehyde groups, whereas borohydride is a reducing agent that reduces the aldehydes to more inert alcohols. The carbohydrate structure is thus irreversibly destroyed.

[0200] In order to provide an improved method for chemical modification of glycans, in particular a procedure that provides a modified lysosomal protein with improved properties, a significant number of reaction conditions were evaluated. It could be concluded that both oxidation by sodium meta-periodate and reduction by sodium borohydride introduced polypeptide modifications and aggregation; properties that negatively impact on catalytic activity and immunogenic propensity.

[0201] Conditions were discovered for an improved chemical modification procedure (Exemplified by new method 1-9). Surprisingly, the structural integrity and activity of the lysosomal proteins could be retained given that the step of sodium borohydride reduction was following directly after quenching of the sodium meta-periodate oxidation and reactant concentrations and time for reactions were kept balanced and significantly lower/shorter as compared to the known method. The new methods omit buffer change and long exposure of the lysosomal protein to reactive aldehyde intermediates. Examples of the new chemical modification procedures are depicted in FIG. 1B-1D.

Example 4

Analyses of Sulfamidase, Alpha-L-Iduronidase and Iduronate 2-Sulfatase Modified According to New Methods

[0202] The experimental methods described in Example 2 were used to analyze lysosomal proteins modified according to the new methods.

[0203] Results

[0204] Peptide bands of lower molecular weight, representing peptide cleavage products were apparent also for material modified according to the new methods but at a significantly lower extent (FIG. 2B, lane 1 vs lane 2; lane 3 vs lane 4; lane 5 vs lane 6, 7 and 8). As for alpha-L-iduronidase modified according to new methods 8, only the monomeric band was apparent (FIG. 2B lane 7). Importantly, the use of a ligand protecting the active site (new method 9, FIG. 2B lane 8) was compatible with the procedure and resulted in modified alpha-L-iduronidase that by SDS-PAGE analysis was indistinguishable from that where the ligand was omitted (new method 8).

[0205] In conclusion, process related impurities, limiting the quality and safety of a medicament produced by the modification methods, are significantly reduced by the new methods as compared to the previously known method.

[0206] Glycan analysis of selected tryptic peptide fragment showed that no, or in some cases less than 5%, naturally occurring glycan structures were present after chemical modification, indicative of complete or close to complete modification of the glycans.

[0207] Further analysis of sulfamidase by SEC showed that the sulfamidase modified according to the new method 1 contained less aggregates compared to the sulfamidase modified by the known method. This is demonstrated in the chromatograms of FIG. 3, where the high molecular weight form is present in the chromatogram as a pre-peak. The peak height of the pre-peak in FIG. 3B is 0.5%, relative the main peak height, thus representing a decrease compared to peak height (3%) in FIG. 3A. This is also the case for sulfamidase modified by new method 5 and 6 (data not shown). The DLS analysis (FIG. 4B) confirmed the results from SEC analysis: the sulfamidase produced according to the new method contained 5% protein in high molecular weight forms (above 10.sup.10 kDa). It could thus be concluded that formation of aggregated sulfamidase is limited by the new method.

[0208] Sulfamidase was further studied by evaluation of degree of active site preservation: The reduction of FGly to Ser in position 50 at the active site of sulfamidase was determined by LC-MS/MS and the tryptic peptides containing FGly and Ser were positively identified. The relative amount of the peptide fragments was analyzed with LC-MS by measuring the peak areas from reconstructed ion chromatograms of the doubly charged ions (without correction for ionization efficiency). The samples generated by four of the new methods described in Example 3 for the chemical modification were prepared and analyzed in duplicates or triplicates (Table 3)

TABLE-US-00003 TABLE 3 Conversion of FGly to Ser at active site Chemical modification of sulfamidase Ser formation (%) FGly/Ser ratio None 0 New method 1 45.4 0.9 (n = 3) 1.2 New method 4 11.5 1 (n = 3) 7.7 New method 5 44.1 2 (n = 2) 1.2 New method 6 34.4 2 (n = 2) 1.9

[0209] Loss of active site FGly is limited considerably by the new methods. The four new methods of modifying glycans on sulfamidase significantly decreased the amount of Ser formation, from 56% using the procedure described in WO 2008/109677 (see Table 2, Example 2), to 45%, 44%, and 34% (new method 1, 5, and 6, respectively, Table 3). The Ser formation of the new method 4 was about 11%, thus indicating that the conversion of FGly to Ser was highly dependent on sodium borohydride concentration.

Example 5

Receptor Mediated Endocytosis of Chemically Modified Lysosomal Proteins In Vitro

[0210] Material and Methods

[0211] Sulfamidase was prepared as described and modified according to the known method and new methods 1 and 4 (Example 1 and 3). Endocytosis was evaluated in MEF-1 fibroblasts expressing M6P receptors. The MEF-1 cells were incubated for 24 h in DMEM medium supplemented with 75 nM of sulfamidase. The cells were washed twice in DMEM and once in 0.9% NaCl prior to cell lysis using 1% Triton X100. Lysate sulfamidase activity and total protein content were determined and lysate specific activity was calculated. Activity was monitored by fluorescence intensity at 460 nm using 0.25 mM 4-methylumbelliferyl-alpha-D-N-sulphoglucosaminide as substrate in 14.5 mM diethylbarbituric acid, 14.5 mM sodium acetate, 0.34% (w/v) NaCl, and 0.1% BSA. Total protein concentration was determined using the BCA kit (Pierce) with BSA as standard. Data are presented as mean+SD (n=4).

[0212] Results

[0213] Sulfamidase activity could be detected in cell homogenate for all preparations evaluated in the endocytosis assay. Modified sulfamidase prepared by the known method as well as the new methods 1 and 4 showed specific activities in cell homogenate below 10% of that obtained with unmodified recombinant sulfamidase (FIG. 5). The activity retained in cells first loaded with and then grown in the absence of sulfamidase for 2 days were comparable for all preparations showing that chemical modification do not negatively impact on lysosomal stability.

[0214] It can therefore be concluded that chemical modification render sulfamidase less prone to cellular uptake which is a consequence of removal of epitopes for glycan recognition receptors as M6PR. On a macroscopic level, this loss of molecular interactions translates into a reduced clearance from plasma when administrated intravenously. The reduced clearance of the protein could allow for less frequent dosing for the patients. Similar results were obtained with modified alpha-L-iduronidase and iduronate 2-sulfatase (Data not shown).

Example 6

In Vivo Plasma/Serum Clearance of Lysosomal Proteins Sulfamidase, Alpha-L-Iduronidase and Iduronate 2-Sulfatase Modified According to New Methods

[0215] Material and Methods

[0216] In Life Phase:

[0217] Plasma/Serum clearance (CL) was investigated for the unmodified and modified lysosomal proteins sulfamidase, alpha-L-iduronidase and iduronate 2-sulfatase in mice (C57BL/6J). Mice were given an intravenous single dose administration in the tail vein. Blood samples were taken at different time points up to 24 h post dose (3 mice per time point) and plasma/serum was prepared. The plasma/serum levels of lysosomal enzymes were analyzed by electrochemiluminescence (ECL) immunoassay. Plasma/serum clearance was calculated using WinNonlin software version 6.3 (Non-compartmental analysis, Phoenix, Pharsight Corp., USA). For sulfamidase and sulfamidase modified according to new method 1 the dose was 10 mg/kg formulated at 2 mg/mL and administered at 5 mL/kg. For iduronate 2-sulfatase and iduronate 2-sulfatase modified according to new method 2 the dose was 1 mg/kg formulated at 0.2 mg/mL and administered at 5 mL/kg. For alpha-L-iduronidase and alpha-L-iduronidase modified according to new method 3 the dose was 3 mg/kg formulated at 0.6 mg/mL and administered at 5 mL/kg.

[0218] Quantification of Sulfamidase and Modified Sulfamidase by ECL:

[0219] Sulfamidase and modified sulfamidase in plasma PK samples were determined by ECL immunoassay using the Meso Scale Discovery (MSD) platform. A Streptavidin coated MSD plate was blocked with 5% Blocker-A in PBS. The plate was washed and different dilutions of standard and PK samples were distributed in the plate. A mixture of a biotinylated anti-sulfamidase mouse monoclonal antibody and Sulfo-Ru-tagged rabbit anti-sulfamidase antibodies was added and the plate was incubated at RT. Complexes of sulfamidase and labelled antibodies will bind to the Streptavidin coated plate via the biotinylated mAb. After washing, the amount of bound complexes was determined by adding a read buffer to the wells and the plate was read in a MSD S12400 instrument. The recorded ECL counts were proportional to the amount of sulfamidase in the sample and evaluated against a relevant sulfamidase standard.

[0220] Quantification of Alpha-L-Iduronidase and Modified Alpha-L-Iduronidase by ECL:

[0221] Alpha-L-iduronidase and modified alpha-L-iduronidase in plasma PK samples were determined by ECL immunoassay using the Meso Scale Discovery (MSD) platform. The wells of a 96 well streptavidin gold plate (#L155A-1, MesoScaleDiscovery (MSD)) were blocked with 1% Fish Gelatin in Phosphate buffer saline (PBS), washed with wash buffer (PBS+0.05% Tween-20) and incubated with a biotinylated, affinity purified goat-a-human alpha-L-iduronidase polyclonal antibody (BAF2449, R&D) after washing different dilutions of standard and PK samples in sample diluent (1% Fish Gelatin in PBS+0.05% Tween 20+1% C57BL6 serum pool) were incubated in the plate at 700 rpm shake and RT for 2 h. The plate was washed and a alpha-L-iduronidase specific Rutenium (SULFO-TAG, MSD) tagged goat polyclonal antibody (AF2449, R&D) was added and allowed to bind to the captured alpha-L-iduronidase or chemically modified alpha-L-iduronidase. The plate was washed and 2 Read Buffer (MSD) was added. The plate content was analyzed using a MSD Sector 2400 Imager Instrument. The instrument applies a voltage to the plate electrodes, and the SULFO-TAG label, bound to the electrode surface via the formed immune complex, will emit light. The instrument measures the intensity of the emitted light which is proportional to the amount of alpha-L-iduronidase or chemically modified alpha-L-iduronidase in the sample. The amount of alpha-L-iduronidase or chemically modified alpha-L-iduronidase was determined against a relevant alpha-L-iduronidase or chemically modified alpha-L-iduronidase standard.

[0222] Quantification of Iduronate 2-Sulfatase and Modified Iduronate 2-Sulfatase by ECL:

[0223] Iduronate 2-sulfatase and modified iduronate 2-sulfatase in plasma PK samples were determined by ECL immunoassay using the Meso Scale Discovery (MSD) platform. The wells of a 96 well streptavidin gold plate (#L155A-1, MesoScaleDiscovery (MSD)) were blocked with 1 Fish Gelatin in Phosphate buffer saline (PBS), washed with wash buffer (PBS+0.05% Tween-20) and incubated with a biotinylated, affinity purified goat-a-human iduronate 2-sulfatase polyclonal antibody (BAF2449, R&D) after washing different dilutions of standard and PK samples in sample diluent (1% Fish Gelatin in PBS+0.05% Tween 20+1% C57BL6 serum pool) were incubated in the plate at 700 rpm shake and RT for 2 h. The plate was washed and a iduronate 2-sulfatase specific Rutenium (SULFO-TAG, MSD) tagged goat polyclonal antibody (AF2449, R&D) was added and allowed to bind to the captured iduronate 2-sulfatase or chemically modified iduronate 2-sulfatase. The plate was washed and 2 Read Buffer (MSD) was added. The plate content was analyzed using a MSD Sector 2400 Imager Instrument. The instrument applies a voltage to the plate electrodes, and the SULFO-TAG label, bound to the electrode surface via the formed immune complex, will emit light. The instrument measures the intensity of the emitted light which is proportional to the amount of iduronate 2-sulfatase or chemically modified iduronate 2-sulfatase in the sample. The amount of iduronate 2-sulfatase or chemically modified iduronate 2-sulfatase was determined against a relevant iduronate 2-sulfatase or chemically modified iduronate-2-sulfatase standard.

[0224] Results

[0225] The plasma/serum clearance in mice of modified sulfamidase, iduronate 2-sulfatase and alpha-L-iduronidase as compared to unmodified counterparts were reduced significantly, see Table 4 below. This is probably at least partly due to the inhibition of receptor mediated uptake in peripheral tissue following chemical modification.

TABLE-US-00004 TABLE 4 Plasma/Serum clearance of lysosomal proteins sulfamidase, alpha-L- iduronidase and iduronate 2-sulfatase Plasma/Serum Dose CL Test article (mg/kg) (mL/(h .Math. kg)) sulfamidase (SEQ ID NO: 44) 10 170 modified sulfamidase (New method 1) 10 14 iduronate 2-sulfatase (SEQ ID NO: 35) 1 60 modified iduronate-2-sulfatase (New method 2) 1 14 alpha-L-iduronidase (SEQ ID NO: 38) 3 130 modified alpha-L-iduronidase (new method 3) 3 45

Example 7

In Vivo Effect of Modified Sulfamidase on Brain Heparan Sulfate Storage

[0226] Materials and Methods

[0227] The effect of intravenously (i.v.) administrated modified sulfamidase produced as described in the general material and methods section, in Quattromed Cell Factory (QMCF) episomal expression system (Icosagen AS) and modified according to new method 1 of Example 3 on brain heparan sulfate storage in vivo was investigated.

[0228] Test Article Preparation:

[0229] Modified sulfamidase was formulated at 6 mg/mL, sterile filtrated and frozen at 70 C. until used. Frozen modified sulfamidase and corresponding vehicle solution were thawed on the day of injection at RT for minimum one hour up to two hours before use. Chlorpheniramine was dissolved in isotonic saline to a concentration of 0.5 mg/mL, and stored at 20 C.

[0230] Animals:

[0231] Male mice having a spontaneous homozygous mutation at the mps3a gene, B6.Cg-Sgsh.sup.mps3a/PstJ (MPS IIIA)(Jackson Laboratories, ME, USA), were used. The animals were housed singly in cages at 231 C. and 40-60% humidity, and had free access to water and standard laboratory chow. The 12/12 h light/dark cycle was set to lights on at 7 pm. The animals were conditioned for at least two weeks before initiating the study. Wild-type siblings from the same breeding unit were also included as controls. In study A, mice were 23-24 weeks old whereas mice were 9-10 weeks old in study B.

[0232] Experimental Procedure Study A:

[0233] Modified sulfamidase at 30 mg/kg (n=8) and vehicle (n=7) were administered intravenously to MPS IIIA mice every other day for twenty-five days (13 injections). Chlorpheniramine was dosed (2.5 mg/kg) subcutaneously 30-45 min before administration of modified sulfamidase or vehicle. Dosing started approximately at 07.00 in the morning. The test article and vehicle were administered at 5 mL/kg. The final administration volume was corrected for the actual body weight at each dosing occasion. This scheme was repeated for vehicle. The study was finished 2 h after the last injection. Untreated age-matched wild-type mice (n=5) were included in conjunction with the test article-treated groups. The mice were anaesthetized by isoflurane. Blood was withdrawn from retro-orbital plexus bleeding. Perfusion followed by flushing 20 mL saline through the left ventricle of the heart. Tissues were dissected (brain, liver, spleen, lung, and heart), weighed and frozen rapidly in liquid nitrogen. The tissues and blood were prepared to measure hexosamine N-sulfate [-1,4] uronic acid (HNS-UA) levels using LC-MS/MS. HNS-UA is a disaccharide marker of heparan sulfate storage, and thus a decrease in HNS-UA levels reflects degradation of heparan sulfate. The HNS-UA data were calculated in relative units vs. internal standard, expressed per mg tissue and normalized to the average of the control group. The data were analyzed by one-way ANOVA test and if overall significance was demonstrated also by Bonferroni's multiple comparison post-hoc test for test of significance between groups (*P<0.05, **P<0.01, ***P<0.001).

[0234] Experimental Procedure Study B:

[0235] Modified sulfamidase at 30 mg/kg (n=6), 10 mg/kg (n=6) and vehicle (n=6) were administered intravenously to MPS IIIA mice once weekly for 10 weeks (10 injections). Chlorpheniramine was dosed (2.5 mg/kg) subcutaneously 30-45 min before administration of modified sulfamidase or vehicle. The final administration volume was corrected for the actual body weight at each dosing occasion. This scheme was repeated for vehicle. The study was finished 24 h after the last injection. Untreated age-matched wild-type mice (n=6) were included in conjunction with the test article-treated groups. The mice were anaesthetized by isoflurane. Blood was withdrawn from retro-orbital plexus bleeding. Perfusion followed by flushing 20 mL saline through the left ventricle of the heart. Tissues were dissected (brain, liver, spleen), weighed and frozen rapidly in liquid nitrogen. The tissues and blood were prepared to measure HNS-UA levels using LC-MS/MS. The HNS-UA data were calculated in relative units vs. internal standard, expressed per mg tissue and normalized to the average of the control group. The data were analyzed by one-way ANOVA test and if overall significance was demonstrated also by Bonferroni's multiple comparison post-hoc test for test of significance between groups (*P<0.05, **P<0.01, ***P<0.001).

[0236] LC-MS/MS Analysis of HNS-UA in Tissue Samples:

[0237] Liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis of hexosamine N-sulfate [-1,4] uronic acid (HNS-UA) in tissue samples was conducted partly according to methods described by Fuller et al (Pediatr Res 56: 733-738 (2004)) and Ramsay et al (Mol Genet Metab 78:193-204 (2003)). The tissues (90-180 mg) were homogenized in substrate buffer (29 mM diethylbarbituric acid, 29 mM sodium acetate, 0.68% (w/v) NaCl, 100 mL water, pH 6.5) using a Lysing Matrix D device (MP Biomedicals, LLC, Ohio, US). Homogenization was performed for 25 s in a Savant FastPrep FP120/Bio101 homogenizer (LabWrench, ON, Canada) and the homogenate was subsequently centrifuged in an Eppendorf centrifuge 5417R at 10000 rcf. The supernatant was evaporated to near dryness. 150 L derivatizing solution (250 mM 3-methyl-1-phenyl-2-pyrazolin-5-one (PMP), 400 mM NH.sub.3, pH 9.1) and 5 L of the internal standard Chondroitin disaccharide di-4S sodium (UA-GalNAc4S, 0.1 mg/mL) stock solution was added. The derivatization was performed at 70 C. for 90 min under agitation and then the solutions were acidified with 200 L of 800 mM formic acid. Deionized water was added to the samples to a final volume of 500 L, and extraction was performed with chloroform (3500 L) to remove excess PMP. Centrifugation was performed at 13000g for 5 min and the upper phase was transferred to a new vial. To remove any excess of formic acid and NH.sub.4COOH, the aqueous phase was evaporated to dryness in a speed vac (Savant Instruments Inc., Farmingdale, N.Y.). The samples were reconstituted to a total of 100 L of 5% acetonitrile/0.1% acetic acid/0.02% TFA.

[0238] LC-MS/MS analysis was performed on Waters Ultra Performance Liquid Chromatography (UPLC), coupled to Sciex API 4000 triple quadrupole mass spectrometer. Instrument control, data acquisition and evaluation were done with Analyst software.

[0239] LC separation was performed by the use of an Acquity C18 CSH column (502.1 mm, 1.7 m). Mobile phase A consisted of 5% acetonitrile/0.5% formic acid, and mobile phase B consisted of 95% acetonitrile/0.5% formic acid. A gradient from 1% to 99% B in 7 min was used at a flow rate of 0.35 mL/min. The injection volume was 10 L. The API 4000 was operated in electrospray negative ion multiple reaction monitoring (MRM) mode. The ion spray voltage was operated at 4.5 kV, and the source temperature was 450 C. Argon was used as collision gas. Collision energy was 34 V. The MRM transitions were 764.4/331.2 (PMP-HNS-UA) and 788.3/534.3 (PMP-internal standard). The relative amount of the HNS-UA was calculated with respect to the level of the internal standard.

[0240] Results

[0241] The results from study A shown in FIG. 6A illustrates that sulfamidase modified according to the new method 1 decreased the levels of HNS-UA in the brain by 30% following repeated intravenous administration every other day for 25 days (13 doses) at 30 mg/kg.

[0242] In addition, treatment with the modified sulfamidase totally abolished HNS-UA levels in liver (FIG. 6B) and lung (not shown).

[0243] The results from study B are shown in FIG. 6C and illustrates that modified sulfamidase according to the new method 1 decreased the levels of HNS-UA in the brain by 48% and 14% following repeated intravenous administration once weekly for 10 weeks at 30 mg/kg and 10 mg/kg, respectively.

[0244] These results thus demonstrate that a sulfamidase protein modified according to the new method 1 described herein causes, after long-term treatment, a robust reduction of HNS-UA levels in brain as well as an essentially complete reduction of HNS-UA levels in peripheral organs.

Example 8

Optimization of Sulfamidase Modification

[0245] The chemical modification process can generally be divided into two parts where the oxidation step is the first step, denoted R1 hereinafter, and the reduction is the second step, denoted R2. To optimize the two steps a full factor design of experiment (DoE) investigating the effect of temperature, concentration and time for the two steps was set up.

[0246] Materials and Methods

[0247] Sulfamidase produced as described in Example 1 in Quattromed Cell Factory (QMCF) episomal expression system (Icosagen AS) were modified essentially as described in Example 3 for new method 1, however parameters subjected to investigation were varied in accordance with Table 5 (below). The investigation of R1 was carried out with the same reduction and parameters work-up as described in Example 3 (new method 1). The end-points for the analysis are degree of oxidation of glycans described in Example 2, and the level of cell uptake of the modified protein, described in Example 5.

TABLE-US-00005 TABLE 5 Parameters varied in R1 and R2 Variable R1 R2 T ( C.) 0, 8, 22 0, 8, 22 t (min) 30, 60, 120 30, 60, 120 c (mol/L) 10, 20, 40 1.2x (c in R1), 2.5x (c in R1), 5x (c in R1)

[0248] The number of parameters and the type of design selected yields ten experiments for each step, the results of which were evaluated using the MODDE 10 software (Umetrics AB).

[0249] In addition the influence of the second quenching step was tested on sulfamidase produced with the R1 parameters 8 C., 60 min and 20 mM sodium meta-periodate. Two additional reactions were run in parallel to the DoE experiment and quenched using 0.1 M acetone or by addition of acetic acid until a pH of 6.0 or lower was obtained. The final work-up followed the scheme for the other reactions. The sulfamidase thus produced was evaluated using the SDS-PAGE method described in Example 2.

[0250] The R2 experiments were conducted with sulfamidase modified according to the parameters found to be optimal after the analysis of the DoE of R1.

[0251] Results

[0252] The R1 results are summarized in table 6 below:

TABLE-US-00006 TABLE 6 R1 experiments and results Cell uptake Remaining original % of (natural) N-Glycan unmodified Varied parameters (%) sulfamidase T ( C.) t (min) c (mmol/L) N (21) N (131) N (244) N (393) uptake 0 30 10 0 0.9 0 0 12 0 120 10 0 0.4 0 0 9.5 22 30 10 0 0.2 0 0 7.1 22 120 10 0 0.1 0 0 8.5 0 30 40 0 0.3 0 0 3.8 0 120 40 0 0.1 0 0 3.5 22 30 40 0 0.1 0 0 2.7 22 120 40 0 0.01 0 0 3.5 8 60 20 0 0.2 0 0 6.1 8 60 20 0 0.2 0 0 6.5

[0253] In addition, a glycosylation analysis according to Example 2 was conducted for sulfamidase modified according to the known method. No remaining original N-glycans were detected at the N-glycosylation sites N(21), N(131), N(244), and N(393).

[0254] The MODDE evaluation of R1 (oxidation) showed that an optimum for R1 at a temperature of around 8 C., a reaction duration of around 1 h and a concentration of around 10 mmol/L of sodium meta-periodate. The overall protein health (e.g. structural integrity) seems to benefit from the lowest oxidant concentration as possible that still limits the cellular uptake via glycan recognition receptors to the level of new method 1 (see Example 5 for details).

[0255] Among the various conditions disclosed for R1 reaction time was considered as an important parameter for degree of glycan modification. In addition, periodate concentration may influence degree of glycan modification.

[0256] The R2 (reduction) design thus used the above identified preferred parameters for R1, i.e. used for oxidation of sulfamidase. The critical end-point for R2 is FGly content since it was found to influence the activity of the modified sulfamidase (cf Examples 2 and 4). See Table 7 below for results. The relative amount of the peptide fragments containing FGly50 and Ser50 was analyzed with LC-MS by measuring the peak areas from reconstructed ion chromatograms (without correction for ionization efficiency).

TABLE-US-00007 TABLE 7 Summary of DoE for R2 and confirmatory experiments Active site Varied parameters Ser formation FGly/Ser t (min) T ( C.) c (mmol/L) (%) Ratio 30 0 12 10 9.0 90 0 12 11 8.1 30 22 12 15 5.7 90 22 12 17 4.9 30 0 50 40 1.5 90 0 50 50 1.0 30 22 50 64 0.6 90 22 50 72 0.4 60 8 25 42 1.4 30 0 20 25 3.0 30 0 50 45 1.2 60 0 15 15 5.7 60 0 25 33 2.0 60 8 12 15 5.7 60 8 50 62 0.6

[0257] The DoE for R2 showed that the Ser formation is related to concentration of sodium borohydride and temperature. Taking into account Ser formation and the presence of high molecular weight forms (data not shown, the results are analogous with the ones received for new method 4 in Example 3), the preferred conditions for R2 are a temperature of around 0 C., a reaction duration of around 1 h or less, and a sodium borohydride concentration of more than 12 mmol/L and up to and including 50 mmol/L.

[0258] It was confirmed on SDS-PAGE (data not shown) that the sulfamidase produced in a reaction where the reduction step was quenched was comparable with the sulfamidase produced without quenching. This indicates that the introduction of the second quenching step do not negatively affect the quality of the material by either quenching with 0.1 M acetone or by lowering the pH to below 6 by addition of acetic acid.

Example 9

Analysis of Glycan Structure after Chemical Modification of Sulfamidase According to Previously Known Method

[0259] Material and Methods

[0260] Chemical Modification According to the Known Method:

[0261] The chemical modification of sulfamidase according to the known method was performed as described in Example 1.

[0262] Glycosylation Analysis:

[0263] The analysis of glycan structure on sulfamidase after chemical modification was performed according to the LC-MS method described in Example 2.

[0264] Resulting modifications on the glycan moieties on the four tryptic peptide fragments containing the N glycosylation sites N(21), N(131), N(244) and N(393) described in Example 2 were investigated by LC-MS analysis.

[0265] Results

[0266] Glycosylation Analysis:

[0267] The type of glycosylation found on the four glycosylation sites prior to the chemical modification was predominantly complex glycans on N(21) and N(393), and oligomannose type of glycans on N(131) and N(244).

[0268] After the chemical modification, detailed characterization of the modified glycan structure was performed on the most abundant chemically modified glycopeptides (less abundant glycans were not detectable due to significant decrease in sensitivity as a result of increased heterogeneity of the glycans after chemical modification). In this Example, the modification on Man-6 glycan after chemical modification according to the known method is investigated.

[0269] Periodate treatment of glycans cleaves carbon bonds between two adjacent hydroxyl groups of the carbohydrate moieties and alter the molecular mass of the glycan chain. FIG. 8A illustrates an example of predicted bond breaks on mannose after chemical modification. FIG. 8B depicts a model of Man-6 glycan showing the theoretical bond breaks that may take place after oxidation with sodium periodate.

[0270] In FIG. 9 are shown mass spectra of the tryptic peptide NITR with Man-6 glycan attached to N(131) (T13+Man-6 glycan), prior to and after chemical modification according to the previously known method. Ions corresponding to the chemically modified glycopeptide with various degree of bond breaking were identified. For Man-6 glycan, there can theoretically be a maximum of 3 double bond breaks and one single bond break. When the modification was performed according to the known method, the most intense ion signal in the mass spectrum was found to be corresponding to 2 double bond breaks and 2 single bond breaks, while the second most intense ion signal corresponded to 3 double bond breaks and one single bond break, which is the most extensive bond breaks possible. A diagram visualizing the extent of bond breaking found on T13+Man-6 glycan after chemical modification according to the known method is shown in FIG. 10A (due to isotopic distribution from the ions observed, the results are approximative but comparable). The reproducibility of the chemical modification was tested by using three different batches of chemically modified sulfamidase produced according to the previously known method. The ions corresponding to different degree of bond breaking showed very similar distribution in the MS spectra from the three different batches.

Example 10

Analysis of Glycan Structure after Chemical Modification of Sulfamidase According to New Methods 1, 4, and 5

[0271] New Methods 1, 4, and 5:

[0272] The chemical modifications of sulfamidase according to the new methods were performed as described in Example 3.

[0273] Glycosylation Analysis:

[0274] The glycosylation analysis was performed according to the LC-MS method described in Example 2. Resulting modifications on the glycan variants of the four tryptic peptide fragments containing the N glycosylation sites N(21), N(131), N(244) and N(393) were investigated by LC-MS analysis.

[0275] Results

[0276] Glycosylation Analysis:

[0277] Detailed characterization of the modified glycan profile on sulfamidase, chemically modified according to new methods 1, 4, and 5, was performed on the most abundant chemically modified glycopeptides. In this Example 10, the modification on Man-6 glycan after chemical modification according to new methods 1, 4, and 5, was investigated.

[0278] Ions corresponding to the chemically modified glycopeptide T13+Man-6 glycan with various degree of bond breaking were identified. Theoretically there can be a maximum of 3 double bond breaks and one single bond break (see FIG. 8B a model of Man-6 glycan showing the bond breaks possible to occur after oxidation with sodium periodate). When the modification was performed according to the new method 1, the most intense ion signal in the mass spectrum was found to be corresponding to one double bond break and 3 single bond breaks, while the second most intense ion signal corresponded to 2 double bond breaks and 2 single bond breaks. When the modification was performed according to new methods 3 and 4, the bond breaks on Man-6 glycan were even further shifted to preferentially single bond breaks. In FIG. 10A is shown a diagram visualizing the extent of bond breaking of the tryptic peptide T13+Man-6 glycan after chemical modification.

[0279] The reproducibility of the chemical modification was tested by using triplicates (new method 1) or duplicates (new methods 3) of chemically modified sulfamidase.

[0280] When comparing the Man-6 glycan modifications resulting from sulfamidase chemically modified according to the known method with the Man-6 glycan modifications resulting from sulfamidase chemically modified according to the new methods 1, 4, and 5, there was a large difference in degree of bond breaking. This is illustrated in FIG. 10A, where the distribution of the different degrees of bond breaking is plotted for the four methods (due to isotopic distribution from the ions observed, the results are approximative, but comparable).

[0281] FIG. 10B shows the relative abundance of single bond breaks for the methods used. The previously known method provides a modified sulfamidase having 45% single bond breaks in the investigated Man-6-glycan, while the new methods 1, 3, and 4 have 70, 80, and 82% single bond breaks, respectively, after chemical modification.

Example 11

Analyses of Enzymatic Activity of Iduronate 2-Sulfatase Modified According to Known Method

[0282] Material and Methods

[0283] Catalytic activity of iduronate 2-sulfatase modified according to known method as described in Example 1 was assessed by incubating preparations of iduronate 2-sulfatase with the substrate 4-Methylumbeliferone iduronide-sulphate. The concentration of substrate in the reaction mixture was 50 M and the assay buffer was 50 mM sodium acetate, 0.005% Tween 20, 0.1% BSA, 0.025% Anapoe X-100, 1.5 mM sodium azide, pH 5. After the incubation, further desulphation was inhibited by addition of a stop buffer containing 0.4 M sodium phosphate, 0.2 M citrate pH 4.5. A second 24 hour incubation with iduronate 2-sulfatase (assay concentration 0.83 g/mL) was performed to hydrolyze the product (4-methylumbeliferone iduronide) and release 4-Methylumbeliferone, which was monitored by fluorescence at 460 nm after quenching the reaction with 0.5 M sodium carbonate, 0.025% Triton X-100, pH 10.7.

[0284] Results

[0285] The activity of iduronate 2-sulfatase modified according to the known method was below 50% of that of unmodified iduronate 2-sulfatase (results not shown).

Example 12

Analyses of Enzymatic Activity of Iduronate 2-Sulfatase Modified According to New Methods

[0286] Material and Methods

[0287] Iduronate 2-sulfatase was modified according to new methods 10 and 11, which are as Example 3 but with the difference that the sodium borohydride reaction mixtures were held at 0 C. for 0.5 h. In new method 11, further the periodate oxidation was performed in the presence of 0.5 mg/mL heparin. Catalytic activity of iduronate 2-sulfatase modified according to new methods 10 and 11 was determined according to the procedure described in Example 11.

[0288] Results

[0289] Iduronate 2-sulfatase prepared according to new method 10 and 11 showed an activity that was comparable to that of unmodified iduronate 2-sulfatase (FIG. 11).

Example 13

Chemical Modification of Alpha-L-Iduronidase in the Presence of an Active Site Protecting Ligand

[0290] As described in Example 3 new method 9, the oxidation (step a)) was performed in the presence of different ligands. The ligands used were 4-methylumbeliferone iduronide, 5-fluoro--l-idopyranosyluronic acid fluoride, heparin, heparin sulphate and D-Saccaric acid 1.4-lactone, respectively.

[0291] Enzymatic activity was measured as described in Standardization of -L-iduronidase Enzyme Assay with Michaelis-Menten Kinetics. Ou L, Herzog T L, Carrie M. Wilmot CM3, and Chester B. Whitley C B. Mol Genet Metab. 2014 111: 113-115.

[0292] Results

[0293] When 5-fluoro--l-idopyranosyluronic acid fluoride was used as a protecting ligand during step a) a 52% lower catalytic activity was obtained for the modified alpha-L-iduronidase compared to when step a) was performed without a protecting ligand i.e. according to new method 8. When other inhibitors known in the literature such as D-Saccaric acid 1.4-lactone was used a 25% decrease in catalytic activity was obtained for the modified alpha-L-iduronidase. A similar trend of decrease in catalytic activity was noted for substrates such as 4-MU-iduronide, heparin or heparin sulphate (data not shown).

Example 14

Chemical Modification of Alpha-L-Iduronidase Immobilized on a Gel Matrix

[0294] The modification method as described herein, and in particular, new method 3 of Example 3, was performed while alpha-L-iduronidase was immobilized on a gel matrix. Alpha-L-iduronidase was immobilized by loading the SOURCE 15S Strong Cation Exchange column with a 20 mM sodium phosphate buffer with 20 mM NaCl and a pH of 6.7.

[0295] Aldurazyme was incubated with 250 L Source 15S gel matrix for 1 hour. After that the gel matrix was gently pelleted and concentration of protein in supernatant was determined to be below 10% of that before incubation with gel. One sample was stored stored in a refrigerator one day before proceeding with chemical modification. A second incubation was made just prior to chemical modification.

[0296] Following loading of alpha-L-iduronidase, the column was equilibrated with solutions for step a), quenching of step a), step b), and quenching of step b) in a consecutive fashion. Elution of chemically modified alpha-L-iduronidase is performed by washing the column with a buffer containing 100 mM sodium phosphate and 700 mM sodium chloride with a pH of 5.6.

[0297] Enzymatic activity was measured as described in Standardization of -L-iduronidase Enzyme Assay with Michaelis-Menten Kinetics. Ou L, Herzog T L, Carrie M. Wilmot CM3, and Chester B. Whitley C B. Mol Genet Metab. 2014 111: 113-115.

[0298] Results Binding in Batch Mode to Source 15S

[0299] Performing the chemical modification while Aldurazyme was immobilized on a gel matrix gave an 8% increased catalytic activity of the resulting modified alpha-L-iduronidase compared to when the modification was performed in solution.

Example 15

Chemical Modification of Alpha-L-Iduronidase Immobilized on a Gel Matrix and in the Presence of a Protecting Ligand

[0300] The modification method as described herein, and in particular, new method 3 of Example 3, was performed while alpha-L-iduronidase was immobilized on a gel matrix and in the presence of a ligand. The ligands used were 5-fluoro--l-idopyranosyluronic acid fluoride and D-Saccaric acid 1.4-lactone, respectively.

[0301] Alpha-L-iduronidase was immobilized by loading the Source 15S Strong Cation Exchange column with a 20 mM sodium phosphate buffer with 20 mM NaCl and a pH of 6.7. Aldurazyme was incubated with 250 L Source 15S gel matrix for 1 hour. After that the gel matrix was gently pelleted and concentration of protein in supernatant was determined to be below 10% of that before incubation with gel. One sample was stored stored in a refrigerator one day before proceeding with chemical modification. A second incubation was made just prior to chemical modification. Following loading of alpha-L-iduronidase, the column was equilibrated with solutions for step a), quenching of step a), step b), and quenching of step b) in a consecutive fashion. Elution of chemically modified alpha-L-iduronidase was performed by washing the column with a buffer containing 100 mM sodium phosphate and 700 mM sodium chloride with a pH of 5.6.

[0302] Enzymatic activity was measured as described in Standardization of -L-iduronidase Enzyme Assay with Michaelis-Menten Kinetics. Ou L, Herzog T L, Carrie M. Wilmot CM3, and Chester B. Whitley C B. Mol Genet Metab. 2014 111: 113-115.

[0303] Results

[0304] The combined approach of using a inhibitor to protect the active site in combination with immobilization of aldurazyme on a gel matrix gave the surprising finding that 5-fluoro--l-idopyranosyluronic acid fluoride in combination of immobilization on a Source 15S Strong Cation Exchange column yielded an increase of 37% of catalytic activity of the resulting modified alpha-L-iduronidase compared to when the modification was performed in solution without a protective ligand. The corresponding result when using the inhibitor D-Saccaric acid 1.4-lactone was a 25% decrease in catalytic activity compared to when the modification was performed in solution without a protective ligand.

Example 16

Distribution of Modified Iduronate 2-Sulfatase to Brain of Iduronate 2-Sulfatase Deficient Mice

[0305] Materials and Methods

[0306] The distribution of intravenously (iv) administrated modified iduronate 2-sulfatase produced according to new method 2 of Example 4 to brain in vivo was investigated.

[0307] Test Article Preparation:

[0308] Modified iduronate 2-sulfatase was formulated at 2 mg/mL, sterile filtrated and frozen at 70 C. until used.

[0309] Animals:

[0310] Male mice, IDS-KO (B6N.Cg-Idstm1Muen/J)(Jackson Laboratories, ME, USA), were used. The animals were housed singly in cages at 231 C. and 40-60% humidity, and had free access to water and standard laboratory chow. The 12/12 h light/dark cycle was set to lights on at 7 pm. The animals were conditioned for at least two weeks before initiating the study. The mice were given an intravenous administration in the tail vein of 10 mg/kg modified iduronate 2-sulfatase. The study was finished 24 h after the last injection. The mice were anaesthetized by isoflurane. Blood was withdrawn from retro-orbital plexus bleeding. Perfusion followed by flushing 20 mL saline through the left ventricle of the heart. Brain was dissected weighed and frozen rapidly in liquid nitrogen. Brain homogenates where prepared and activity was assessed using the method described in example 2 with addition of 10 mM lead acetate in the assay buffer as adjustment to the protocoll.

[0311] Results: Activity of modified iduronate 2-sulfatase in perfused brain homogenates of IDS-KO mice could be confirmed. An average activity of 1.80.4 M/min (n=4) was determined under the assay conditions used.