ENGINEERED ENZYME FOR ENZYME REPLACEMENT THERAPY

Abstract

An engineered enzyme, comprising an amino acid sequence that is at least 80% identical to the amino acid sequence of a human beta-glucuronidase, wherein the engineered enzyme exhibits a higher level of alpha-iduronidase enzymatic activity as compared to the human beta-glucuronidase.

Claims

1. An engineered enzyme, comprising an amino acid sequence that is at least 80% identical to the amino acid sequence of a human beta-glucuronidase, wherein the engineered enzyme exhibits a higher level of alpha-iduronidase enzymatic activity as compared to the human beta-glucuronidase.

2. The engineered enzyme of claim 1, wherein the human beta-glucuronidase comprises the amino acid sequence of SEQ ID NO:2.

3. The engineered enzyme of claim 2, wherein the engineered enzyme comprises a substitution at a residue that corresponds to residue T204, Q279, K438, N484, N502, S503, Y504, S506, Y508, H509, G542, T545, L565, W587, F592, T594, E595, P598, R600, G603, N604, K606, and/or P636 in the sequence of SEQ ID NO:2.

4. The engineered enzyme of claim 3, wherein the engineered enzyme comprises a substitution at N484, N502, S503, S506, H509, F592, E595, N604, and/or K606.

5. The engineered enzyme of claim 3, wherein residue 204 is T or K; residue 279 is Q or H; residue 438 is K or M; residue 484 is S, D, H, R, S, or C; residue 502 is N, D, or K; residue 503 is A, D, Y, P, H, or V; residue 504 is Y, G, or C; residue 506 is S or G; residue 508 is Y or D; residue 509 is H, A, or P; residue 542 is G or D; residue 545 is T or A; residue 565 is L or A; residue 587 is W or T; residue 592 is F or Y; residue 594 is T or L; residue 595 is L, V, Q, or G; residue 598 is P or D; residue 600 is R or A; residue 603 is G or E; residue 604 is Y, S, A, or T; residue 606 is Q, F, or L; and residue 636 is P or S.

6. The engineered enzyme of claim 1, wherein the engineered enzyme does not have a substitution at a residue that corresponds to residue S447, G542, L565, W587, R600, G603, and/or P636 of the sequence of SEQ ID NO:2.

7. The engineered enzyme of claim 3, wherein the engineered enzyme comprises residues S484, D502, A503, G506, A509, D542, A545, Y592, V595, S604, and/or F606.

8. The engineered enzyme of claim 3, wherein the engineered enzyme comprises residues H279, C484, K502, Y503, G504, G506, P509, A545, A565, L594, Q595, A604, and/or F606.

9. The engineered enzyme of claim 3, wherein the engineered enzyme comprises residues D484, K502, Y503, G506, D508, P509, A545, Y592, L594, G595, D598, T604, F606, and/or 5636.

10. A nucleic acid molecule comprising a nucleic acid sequence encoding the engineered enzyme of claim 1.

11. A host cell comprising the nucleic acid molecule of claim 10.

12. A pharmaceutical composition comprising the engineered enzyme of claim 1 and a pharmaceutically acceptable carrier.

13. A method of treating mucopolysaccharidosis in a subject, comprising administering to a subject in need thereof the engineered enzyme of claim 1.

14. A method of developing a candidate enzyme replacement therapy for treating a disorder associated with a deficient enzyme in a subject having the disorder, the method comprising: selecting a template enzyme, wherein the template enzyme is endogenous to the subject and expressed normally in the subject, and altering the template enzyme to obtain an engineered enzyme, wherein the engineered enzyme exhibits an increased target enzymatic activity as compared to that of the template enzyme, the target enzymatic activity being an enzymatic activity of the wild-type counterpart of the deficient enzyme; wherein the engineered enzyme is a candidate enzyme replacement therapy for treating the disorder.

15. A method of identifying a candidate enzyme replacement therapy for treating a disorder associated with a deficient enzyme in a subject having the disorder, the method comprising: providing a library of engineered enzymes, wherein each engineered enzyme in the library is a variant of a template enzyme, the template enzyme being endogenous to and expressed normally in the subject; and assaying the library for a target enzymatic activity, wherein the target enzymatic activity is an enzymatic activity of the wild-type counterpart of the deficient enzyme; wherein an engineered enzyme that exhibits an increased target enzymatic activity as compared to that of the template enzyme is a candidate enzyme replacement therapy for treating the disorder.

16. The method of claim 14, wherein the disorder is a lysosomal storage disorder and the deficient enzyme is a lysosomal enzyme.

17. The method of claim 16, wherein the deficient enzyme is a human alpha-iduronidase and the template enzyme is a human beta-glucuronidase.

18. The method of claim 17, wherein the template enzyme comprises the sequence of SEQ ID NO:2.

19. The method of claim 18, wherein the engineered enzyme comprises a substitution at a residue that corresponds to residue T204, Q279, K438, N484, N502, S503, Y504, S506, Y508, H509, G542, T545, L565, W587, F592, T594, E595, P598, R600, G603, N604, K606, or P636 in the sequence of SEQ ID NO:2.

20. The method of claim 17, wherein the target enzymatic activity is hydrolysis of 4-methylumbelliferyl-alpha-L-iduronide.

21. The method of claim 14, further comprising determining whether a candidate enzyme replacement therapy induces an immune response in an animal model or a subject having the disorder.

22. The method of claim 21, further comprising selecting a candidate enzyme replacement therapy that induces no immune response, or induces a lower level of the immune response as compared to the immune response induced by the deficient enzyme.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0009] FIG. 1 is an illustration of the concept of cloaked stealth enzyme for enzyme replacement therapy. Wild-type alpha-iduronidase (IDUA) can generate specific antibodies in mucopolysaccharidosis (MPS) type I patients which can inhibit the therapeutic efficacy and even cause life-threatening anaphylactic reactions. Beta-glucuronidase (G) variants which have been altered in their active pockets to display alpha-iduronidase activity may appear as normal immunotolerant self-proteins and thus be able to prevent and/or minimize antibody responses, thereby retaining therapeutic activity and reducing side effects. The sizes of human immunoglobulin G (150 kDa), alpha-iduronidase (83 kDa), and beta-glucuronidase (78 kDa) are not drawn to scale.

[0010] FIG. 2 is an illustration of an application of the enzyme cleavable surface tethered all-purpose screening system (ECSTASY). (1) A library of beta-glucuronidase variants was cloned into a retroviral vector, which codes for a signal peptide, an HA tag, a 6His tag, an enzyme variant, a myc tag, and the C-terminal 37 amino acids of human DAF. (2) Library cells expressing one GPI-anchored beta-glucuronidase variant each were generated by retroviral transduction of HEK293 cells at a low multiplicity of infection. (3, 4) The cells were immunofluorescence stained with an anti-beta-glucuronidase-FITC conjugate to identify cells expressing beta-glucuronidase on their surface. Fluorescence-activated cell sorting was employed to rapidly discard cells expressing misfolded or unstable enzyme variants as determined by low immunofluorescence staining. (5) Positive cells were individually sorted into 96-well microtiter plates and allowed to grow to full confluence. (6) Surface-tethered enzyme variants were released from the cells by PI-PLC cleavage. (7) The supernatant containing soluble enzyme variants were assayed for enzyme concentration by ELISA and enzyme activity by flourometric assays. (8) Selected mutants with higher activity can be used as templates for a new generation library by DNA shuffling or site-saturation mutagenesis.

[0011] FIG. 3 is a set of graphs showing that human beta-glucuronidase displayed measurable activity toward the alpha-iduronidase substrate MUI. Recombinant human beta-glucuronidase was purified from human alpha-iduronidase deficient cells (34/2000, derived from Mucopolysaccharidosis type I patients) and incubated with 250 M -iduronidase substrate 4-methylumbelliferyl alpha-L-iduronide (MUI) in 0.2 M formate buffer, pH 3.5, for 1 or 17 h at 37 C. (A) The fluorescence was detected at an excitation wavelength of 355 nm and an emission wavelength of 460 nm. The hydrolysis of MUI into 4-methylumbelliferone (4-MU) was also detected by solid-phase extraction/high-performance liquid chromatography on a LiChroprep RP18 (40-63 m) column equilibrated with 20% methanol (pH 4). MUI alone (B) and MUI incubated with human -glucuronidase (C) was eluted with 25% methanol in double-distilled water (pH 4). Commercial MUI (unpurified) incubated without (D) or with alpha-iduronidase (E) was eluted with 25% methanol in double-distilled water (pH 4).

[0012] FIG. 4 is an illustration of the construction of a synthetic library. A full-length human beta-glucuronidase variant library was assembled from five DNA fragments (F0-F4) with eighteen overlapping nucleotides between each other. The F0 fragment contains an Apa I cloning site as well as a silent mutation to remove a unique Bgl II cutting site. F2 and F3 fragments, which contain variable amino acids at nineteen positions, were constructed by primer assembly followed by PCR amplification. The F4 fragment contains a Sal I cloning site. All fragments were mixed in equal molar ratios and amplified by PCR to obtain the full-length human beta-glucuronidase library. Wild-type DNA was removed by Bgl II restriction enzyme digestion after library ligation.

[0013] FIG. 5 is a set of graphs showing an example screen of a human beta-glucuronidase synthetic library by ECSTASY. A. 293 and 293/L1 cells were stained with 7G8-FITC and analyzed on a flow cytometer for surface human beta-glucuronidase expression (dashed gate, 16%). Cells exhibiting the highest human beta-glucuronidase expression level (solid gate, 6.8%) were sorted into 96-well microplates as single cells for subsequent screening. B. GPI-anchored human beta-glucuronidase variants were cleaved from individual 293/L1 clones by PI-PLC. The supernatant was assayed for hydrolysis of 4-methylumbelliferyl alpha-L-iduronide (MUI) at pH 3.5. Protein amounts were quantitated by sandwich ELISA. Wild-type and beta-glucuronidase variants are shown as closed and open circles, respectively. C. Specific activity of each beta-glucuronidase variant is presented as relative fluorescence unit (RFU) per ng protein.

[0014] FIG. 6 is a graph showing characterization of beta-glucuronidase variants displaying alpha-iduronidase activity. Recombinant wild-type and beta-glucuronidase variants were incubated with 4-methylumbelliferyl alpha-L-iduronide (MUI) at pH 3.5 and relative fluorescence was detected.

[0015] FIG. 7 is a set of graphs showing characterization of beta-glucuronidase variants displaying alpha-iduronidase activity. Human alpha-iduronidase deficient cells (34/2000, derived from a Mucopolysaccharidosis type I patient) were incubated with Na.sub.2.sup.35SO.sub.4 to radiolabel the glycosaminoglycans. Cells were then treated with 5 g/ml of recombinant enzymes for 72 h. A. The reduced radioactivity of cell lysates as compared to untreated cells is represented as meanSD (white bars). B. The increased radioactivity of supernatants as compared to untreated cells is represented as meanSD (black bars).

[0016] FIG. 8 is a graph showing characterization of beta-glucuronidase variants displaying alpha-iduronidase activity. Human alpha-iduronidase deficient cells were treated with 5 g/ml of recombinant enzymes for 72 h and stained with Lysotracker-Red DND-99 dye for lysosomes. Lysosomal fluorescence per cell is presented as meanSEM. Statistically significant differences to untreated cells in two-tailed t-test are indicated. *: p<0.05; **: p<0.01; ***: p<0.001; n.s.: non-significant (n=3).

[0017] FIG. 9 is a set of data showing establishment of human beta-glucuronidase transgenic mice. A. An illustration showing pCAG-hG for generation of transgenic mice expressing human beta-glucuronidase. B. Genotyping of newborn mice was performed by genomic PCR (230 bp). PC: positive control. NC: negative control. C. Total proteins were extracted from mouse tails, electrophoresed on a 10% SDS-PAGE, and immunoblotted with anti-human beta-glucuronidase (anti-hG) and anti--actin antibodies.

[0018] FIG. 10 is a set of graphs showing evaluation of engineered beta-glucuronidase immunogenicity in transgenic human beta-glucuronidase mice. A, B, and C. Groups of three human beta-glucuronidase transgenic mice were i.v. injected with 50 g mouse beta-glucuronidase, 50 g of human alpha-iduronidase or 50 g human beta-glucuronidase every three weeks. Weekly serum samples were assayed for antibodies against the respective protein by ELISA. D. Groups of three human beta-glucuronidase transgenic mice were i.v. injected with 50 g of the indicated recombinant protein every three weeks for a total of four injections. Serum samples were diluted 1000 fold and antibody responses were measured by ELISA. Significant differences between the antibody response in mice injected with human alpha-iduronidase and other proteins are shown. **, p<0.001.

DETAILED DESCRIPTION

[0019] Described herein are engineered enzymes useful for enzyme replacement therapies and methods of developing an enzyme replacement therapy.

[0020] It was discovered that the enzymatic activity or specificity of a template enzyme can be at least partially switched to that of another enzyme without dramatic alterations in the structure or sequence of the template enzyme.

[0021] Hence, described herein is an engineered enzyme, generated from a human beta-glucuronidase template (e.g., a wild-type beta-glucuronidase), that exhibits a higher level of alpha-iduronidase enzymatic activity than the template human beta-glucuronidase. The engineered enzyme shares high amino acid sequence identity (e.g., at least 80%, 85%, 90%, 95%, 97% or 99%) with the template beta-glucuronidase but very low sequence identity with a human alpha-iduronidase.

[0022] Exemplary nucleic acid and amino acid sequences of human beta-glucuronidase and human alpha-iduronidase are provided below.

TABLE-US-00001 Ahumanbeta-glucuronidasenucleicacidsequence (SEQIDNO:1): ATGGCCCGGGGGTCGGCGGTTGCCTGGGCGGCGCTCGGGCCGTTGTTGTG GGGCTGCGCGCTGGGGCTGCAGGGCGGGATGCTGTACCCCCAGGAGAGCC CGTCGCGGGAGTGCAAGGAGCTGGACGGCCTCTGGAGCTTCCGCGCCGAC TTCTCTGACAACCGACGCCGGGGCTTCGAGGAGCAGTGGTACCGGCGGCC GCTGTGGGAGTCAGGCCCCACCGTGGACATGCCAGTTCCCTCCAGCTTCA ATGACATCAGCCAGGACTGGCGTCTGCGGCATTTTGTCGGCTGGGTGTGG TACGAACGGGAGGTGATCCTGCCGGAGCGATGGACCCAGGACCTGCGCAC AAGAGTGGTGCTGAGGATTGGCAGTGCCCATTCCTATGCCATCGTGTGGG TGAATGGGGTCGACACGCTAGAGCATGAGGGGGGCTACCTCCCCTTCGAG GCCGACATCAGCAACCTGGTCCAGGTGGGGCCCCTGCCCTCCCGGCTCCG AATCACTATCGCCATCAACAACACACTCACCCCCACCACCCTGCCACCAG GGACCATCCAATACCTGACTGACACCTCCAAGTATCCCAAGGGTTACTTT GTCCAGAACACATATTTTGACTTTTTCAACTACGCTGGACTGCAGCGGTC TGTACTTCTGTACACGACACCCACCACCTACATCGATGACATCACCGTCA CCACCAGCGTGGAGCAAGACAGTGGGCTGGTGAATTACCAGATCTCTGTC AAGGGCAGTAACCTGTTCAAGTTGGAAGTGCGTCTTTTGGATGCAGAAAA CAAAGTCGTGGCGAATGGGACTGGGACCCAGGGCCAACTTAAGGTGCCAG GTGTCAGCCTCTGGTGGCCGTACCTGATGCACGAACGCCCTGCCTATCTG TATTCATTGGAGGTGCAGCTGACTGCACAGACGTCACTGGGGCCTGTGTC TGACTTCTACACACTCCCTGTGGGGATCCGCACTGTGGCTGTCACCAAGA GCCAGTTCCTCATCAATGGGAAACCTTTCTATTTCCACGGTGTCAACAAG CATGAGGATGCGGACATCCGAGGGAAGGGCTTCGACTGGCCGCTGCTGGT GAAGGACTTCAACCTGCTTCGCTGGCTTGGTGCCAACGCTTTCCGTACCA GCCACTACCCCTATGCAGAGGAAGTGATGCAGATGTGTGACCGCTATGGG ATTGTGGTCATCGATGAGTGTCCCGGCGTGGGCCTGGCGCTGCCGCAGTT CTTCAACAACGTTTCTCTGCATCACCACATGCAGGTGATGGAAGAAGTGG TGCGTAGGGACAAGAACCACCCCGCGGTCGTGATGTGGTCTGTGGCCAAC GAGCCTGCGTCCCACCTAGAATCTGCTGGCTACTACTTGAAGATGGTGAT CGCTCACACCAAATCCTTGGACCCCTCCCGGCCTGTGACCTTTGTGAGCA ACTCTAACTATGCAGCAGACAAGGGGGCTCCGTATGTGGATGTGATCTGT TTGAACAGCTACTACTCTTGGTATCACGACTACGGGCACCTGGAGTTGAT TCAGCTGCAGCTGGCCACCCAGTTTGAGAACTGGTATAAGAAGTATCAGA AGCCCATTATTCAGAGCGAGTATGGAGCAGAAACGATTGCAGGGTTTCAC CAGGATCCACCTCTGATGTTCACTGAAGAGTACCAGAAAAGTCTGCTAGA GCAGTACCATCTGGGTCTGGATCAAAAACGCAGAAAATACGTGGTTGGAG AGCTCATTTGGAATTTTGCCGATTTCATGACTGAACAGTCACCGACGAGA GTGCTGGGGAATAAAAAGGGGATCTTCACTCGGCAGAGACAACCAAAAAG TGCAGCGTTCCTTTTGCGAGAGAGATACTGGAAGATTGCCAATGAAACCA GGTATCCCCACTCAGTAGCCAAGTCACAATGTTTGGAAAACAGCCTGTTT ACTTGA Ahumanbeta-glucuronidaseaminoacidsequence (SEQIDNO:2) MARGSAVAWAALGPLLWGCALGLQGGMLYPQESPSRECKELDGLWSFRAD FSDNRRRGFEEQWYRRPLWESGPTVDMPVPSSFNDISQDWRLRHFVGWVW YEREVILPERWTQDLRTRVVLRIGSAHSYAIVWVNGVDTLEHEGGYLPFE ADISNLVQVGPLPSRLRITIAINNTLTPTTLPPGTIQYLTDTSKYPKGYF VQNTYFDFFNYAGLQRSVLLYTTPTTYIDDITVTTSVEQDSGLVNYQISV KGSNLFKLEVRLLDAENKVVANGTGTQGQLKVPGVSLWWPYLMHERPAYL YSLEVQLTAQTSLGPVSDFYTLPVGIRTVAVTKSQFLINGKPFYFHGVNK HEDADIRGKGFDWPLLVKDFNLLRWLGANAFRTSHYPYAEEVMQMCDRYG IVVIDECPGVGLALPQFFNNVSLHHHMQVMEEVVRRDKNHPAVVMWSVAN EPASHLESAGYYLKMVIAHTKSLDPSRPVTFVSNSNYAADKGAPYVDVIC LNSYYSWYHDYGHLELIQLQLATQFENWYKKYQKPIIQSEYGAETIAGFH QDPPLMFTEEYQKSLLEQYHLGLDQKRRKYVVGELIWNFADFMTEQSPIR VLGNKKGIFTRQRQPKSAAFLLRERYWKIANETRYPHSVAKSQCLENSLF T Ahumanalpha-iduronidasenucleicacidsequence (SEQIDNO:3) ATGCGTCCCCTGCGCCCCCGCGCCGCGCTGCTGGCGCTCCTGGCCTCGCT CCTGGCCGCGCCCCCGGTGGCCCCGGCCGAGGCCCCGCACCTGGTGCATG TGGACGCGGCCCGCGCGCTGTGGCCCCTGCGGCGCTTCTGGAGGAGCACA GGCTTCTGCCCCCCGCTGCCACACAGCCAGGCTGACCAGTACGTCCTCAG CTGGGACCAGCAGCTCAACCTCGCCTATGTGGGCGCCGTCCCTCACCGCG GCATCAAGCAGGTCCGGACCCACTGGCTGCTGGAGCTTGTCACCACCAGG GGGTCCACTGGACGGGGCCTGAGCTACAACTTCACCCACCTGGACGGGTA CCTGGACCTTCTCAGGGAGAACCAGCTCCTCCCAGGGTTTGAGCTGATGG GCAGCGCCTCGGGCCACTTCACTGACTTTGAGGACAAGCAGCAGGTGTTT GAGTGGAAGGACTTGGTCTCCAGCCTGGCCAGGAGATACATCGGTAGGTA CGGACTGGCGCATGTTTCCAAGTGGAACTTCGAGACGTGGAATGAGCCAG ACCACCACGACTTTGACAACGTCTCCATGACCATGCAAGGCTTCCTGAAC TACTACGATGCCTGCTCGGAGGGTCTGCGCGCCGCCAGCCCCGCCCTGCG GCTGGGAGGCCCCGGCGACTCCTTCCACACCCCACCGCGATCCCCGCTGA GCTGGGGCCTCCTGCGCCACTGCCACGACGGTACCAACTTCTTCACTGGG GAGGCGGGCGTGCGGCTGGACTACATCTCCCTCCACAGGAAGGGTGCGCG CAGCTCCATCTCCATCCTGGAGCAGGAGAAGGTCGTCGCGCAGCAGATCC GGCAGCTCTTCCCCAAGTTCGCGGACACCCCCATTTACAACGACGAGGCG GACCCGCTGGTGGGCTGGTCCCTGCCACAGCCGTGGAGGGCGGACGTGAC CTACGCGGCCATGGTGGTGAAGGTCATCGCGCAGCATCAGAACCTGCTAC TGGCCAACACCACCTCCGCCTTCCCCTACGCGCTCCTGAGCAACGACAAT GCCTTCCTGAGCTACCACCCGCACCCCTTCGCGCAGCGCACGCTCACCGC GCGCTTCCAGGTCAACAACACCCGCCCGCCGCACGTGCAGCTGTTGCGCA AGCCGGTGCTCACGGCCATGGGGCTGCTGGCGCTGCTGGATGAGGAGCAG CTCTGGGCCGAAGTGTCGCAGGCCGGGACCGTCCTGGACAGCAACCACAC GGTGGGCGTCCTGGCCAGCGCCCACCGCCCCCAGGGCCCGGCCGACGCCT GGCGCGCCGCGGTGCTGATCTACGCGAGCGACGACACCCGCGCCCACCCC AACCGCAGCGTCGCGGTGACCCTGCGGCTGCGCGGGGTGCCCCCCGGCCC GGGCCTGGTCTACGTCACGCGCTACCTGGACAACGGGCTCTGCAGCCCCG ACGGCGAGTGGCGGCGCCTGGGCCGGCCCGTCTTCCCCACGGCAGAGCAG TTCCGGCGCATGCGCGCGGCTGAGGACCCGGTGGCCGCGGCGCCCCGCCC CTTACCCGCCGGCGGCCGCCTGACCCTGCGCCCCGCGCTGCGGCTGCCGT CGCTTTTGCTGGTGCACGTGTGTGCGCGCCCCGAGAAGCCGCCCGGGCAG GTCACGCGGCTCCGCGCCCTGCCCCTGACCCAAGGGCAGCTGGTTCTGGT CTGGTCGGATGAACACGTGGGCTCCAAGTGCCTGTGGACATACGAGATCC AGTTCTCTCAGGACGGTAAGGCGTACACCCCGGTCAGCAGGAAGCCATCG ACCTTCAACCTCITTGTGTTCAGCCCAGACACAGGTGCTGICTCTGGCTC CTACCGAGTTCGAGCCCTGGACTACTGGGCCCGACCAGGCCCCTTCTCGG ACCCTGTGCCGTACCTGGAGGTCCCTGTGCCAAGAGGGCCCCCATCCCCG GGCAATCCATGA Ahumanalpha-iduronidaseaminoacidsequence (SEQIDNO:4) MRPLRPRAALLALLASLLAAPPVAPAEAPHLVHVDAARALWPLRRFWRST GFCPPLPHSQADQYVLSWDQQLNLAYVGAVPHRGIKQVRTHWLLELVTTR GSTGRGLSYNFTHLDGYLDLLRENQLLPGFELMGSASGHFTDFEDKQQVF EWKDLVSSLARRYIGRYGLAHVSKWNFETWNEPDHHDFDNVSMTMQGFLN YYDACSEGLRAASPALRLGGPGDSFHTPPRSPLSWGLLRHCHDGTNFFTG EAGVRLDYISLHRKGARSSISILEQEKVVAQQIRQLFPKFADTPIYNDEA DPLVGWSLPQPWRADVTYAAMVVKVIAQHQNLLLANTTSAFPYALLSNDN AFLSYHPHPFAQRTLTARFQVNNTRPPHVQLLRKPVLTAMGLLALLDEEQ LWAEVSQAGTVLDSNHTVGVLASAHRPQGPADAWRAAVLIYASDDTRAHP NRSVAVTLRLRGVPPGPGLVYVTRYLDNGLCSPDGEWRRLGRPVFPTAEQ FRRMRAAEDPVAAAPRPLPAGGRLTLRPALRLPSLLLVHVCARPEKPPGQ VTRLRALPLTQGQLVLVWSDEHVGSKCLWTYEIQFSQDGKAYTPVSRKPS TFNLFVFSPDTGAVSGSYRVRALDYWARPGPFSDPVPYLEVPVPRGPPSP GNP

[0023] The engineered enzyme can have an amino acid sequence that is at least 80% identical to the sequence of SEQ ID NO:2 and include amino acid substitution(s) at one or more position(s) corresponding to the group of position(s) in the sequence of SEQ ID NO:2 (e.g., as determined by a sequence alignment) consisting of: 204, 279, 438, 484, 502, 503, 504, 506, 508, 509, 542, 545, 565, 587, 592, 594, 595, 598, 600, 603, 604, 606 and 636. For example, the engineered enzyme can have, at one or more of said above-mentioned 23 positions, the corresponding wild-type residue (as set forth in SEQ ID NO:2), or any other amino acid (e.g., A, R, N, D, C, E, Q, G, H, I, L, K, M, F, P, S, T, W, Y, V, or an analog thereof).

[0024] In one embodiment, said engineered enzyme have an amino acid sequence that is at least 80%, 85%, 90%, 95%, 97% or 99% identical to the sequence as set forth in SEQ ID NO:2.

[0025] In one embodiment, the engineered enzyme has an amino acid sequence in which, as compared to the sequence of SEQ ID NO:2, corresponding position 204 is T or K; position 279 is Q or H; position 438 is K or M; position 484 is S, D, H, R, or C; position 502 is N, D, or K; position 503 is A, D, Y, P, or V; position 504 is Y, G, or C; position 506 is S or G; position 508 is Y or D; position 509 is H, A, or P; position 542 is G or D; position 545 is T or A; position 565 is L or A; position 587 is W or T; position 592 is F or Y; position 594 is T or L; position 595 is L, V, Q, or G; position 598 is P or D; position 600 is R or A; position 603 is G or E; position 604 is Y, S, A, or T; position 606 is Q, F, or L; and position 636 is P or S. In one embodiment, substitution of amino acid residue(s) of the engineered enzyme amino acids was carried out as according to Table 1 or Table 2 (see below).

[0026] For example, the engineered enzyme can have, as compared to the sequence of SEQ ID NO:2, the following altered/substituted residue(s): S484, D502, A503, G506, A509, D542, A545, Y592, V595, S604, and/or F606. Another exemplary engineered enzyme has the following altered residues: H279, C484, K502, Y503, G504, G506, P509, A545, A565, L594, Q595, A604, and/or F606. Yet another engineered enzyme can have the following altered residues: D484, K502, Y503, G506, D508, P509, A545, Y592, L594, G595, D598, T604, F606, and/or S636.

[0027] Certain residues in a wild-type human beta-glucuronidase (e.g., SEQ ID NO:2) may be particular targets for altering its enzymatic activity, e.g., N484, N502, S503, S506, H509, F592, E595, N604, and K606. Hence, in one embodiment the engineered enzyme can include amino acid substitution(s) at one or more of said nine amino acid positions. On the other hand, certain wild-type residues, e.g., S447, G542, L565, W587, R600, G603, and P636, may be preferred. The engineered enzyme can thus retain one or more of these seven wild-type residues (as set forth in SEQ ID NO:2). As such, in one embodiment, the engineered enzyme does not have a substitution at a residue that corresponds to residue S447, G542, L565, W587, R600, G603, and/or P636 of the sequence of SEQ ID NO:2.

[0028] In one embodiment, the present invention also provides an isolated polynucleotide encoding the engineered enzyme as described herein.

[0029] In another embodiment, the present invention provides an expression vector comprising said polynucleotide encoding the engineered enzyme as described herein.

[0030] Methods known in the art, e.g., recombinant techniques, can be employed to generate the engineered enzyme described herein.

[0031] The engineered enzyme can be used as an enzyme replacement therapy to treat a subject having a disorder associated with a defective alpha-iduronidase, i.e., mucopolysaccharidosis. As the engineered enzyme will appear to the immune system of the subject as a normal, non-immunogenic endogenous enzyme, it will not induce unwanted immune responses in the subject.

[0032] Gene therapy involving administration of a nucleic acid molecule encoding the engineered enzyme can also be used to treat mucopolysaccharidosis in a subject.

[0033] Also described herein is a method of developing or identifying a candidate enzyme replacement therapy for treating a disorder associated with a deficient enzyme in a subject having the disorder, e.g., a lysosomal storage disorder such as MPS type I, MPS II, MPS type IV, type I Gaucher, Pompe disease or Fabry disease. The deficiency can be due to a mutant enzyme (e.g., truncated enzyme) or a lower than normal level of a wild-type enzyme.

[0034] In the method, the enzymatic activity and/or specificity of a normal endogenous enzyme (i.e., a template enzyme) are altered to compensate for that of the defective enzyme. As the modified enzyme will appear as a normal endogenous protein, the modified enzyme will not be immunogenic in the subject.

[0035] A suitable template enzyme should be one that is endogenous to and expressed normally in a subject having the defective enzyme. The template enzyme and the wild-type counterpart of the defective enzyme can be similar in one or more aspects, e.g., catalytic mechanism, catalytic domain structure, tissue expression profile, size, and cellular localization.

[0036] The selected template enzyme is then altered in order to at least partially switch its enzymatic activity or specificity to that of the normal counterpart of the defective enzyme. Such alterations include amino acid substitutions, deletions and insertions. The alterations should not make the template enzyme appear as a foreign protein to the immune system of the subject. In other words, the modified enzyme should still share a high sequence identity (e.g., at least 80%, 85%, 90%, 95%, or 99% identity) with the template enzyme. The alterations can be rationally designed, random, or a combination thereof.

[0037] For example, an engineered enzyme can be designed based on the structures (e.g., the structures of the whole enzymes and the structures of the active sites) of the template enzyme and the normal counterpart of the defective enzyme. Various techniques and softwares available in the art can be used to compare the sequences and structures of the two enzymes to identify potential residues for alteration. Residues known or predicted to interact with a substrate may be particular targets for alteration. A library of variants each with substitutions at one or more of the identified residues can be generated for screening. Screening a library of randomly generated variants of the template enzyme can also be carried out to identify variants that exhibit the desired activity and/or specificity.

[0038] A candidate enzyme replacement therapy can be further tested (e.g., in an animal model) to determine whether it induces unwanted immune responses. In one embodiment, a candidate enzyme that induces no immune response, or induces a lower level of the immune response as compared to the immune response induced by the deficient enzyme, is selected for enzyme replacement therapy.

[0039] The method can be applied to a wild-type human beta-glucuronidase template (e.g., SEQ ID NO:2) to generated engineered enzymes that exhibit an alpha-iduronidase enzymatic activity. Residues within the catalytic domain of human beta-glucuronidase can be altered. As described above, positions 204, 279, 438, 484, 502, 503, 504, 506, 508, 509, 542, 545, 565, 587, 592, 594, 595, 598, 600, 603, 604, 606 and 636 in the sequence of SEQ ID NO:2 are each a target for alteration. Each of the specific engineered enzymes described herein can be further altered (e.g., addition of substituted residues or different substituted residues) to develop more engineered enzymes.

[0040] The screening method can be performed using techniques or systems known in the art. An exemplary technique is the Enzyme Cleavable Surface Tethered All-purpose Screen sYstem (ECSTASY). See Chen, C. P., et al., Protein Eng Des Sel, 2012. 25(7): p. 367-75; and also FIG. 2.

[0041] A subject refers to a human and a non-human animal. Examples of a non-human animal include all vertebrates, e.g., mammals, such as non-human primates (particularly higher primates), dog, rodent (e.g., mouse or rat), guinea pig, cat, and non-mammals, such as birds, amphibians, etc. In a preferred embodiment, the subject is a human. In another embodiment, the subject is an experimental animal or animal suitable as a disease model.

[0042] The term treating as used herein refers to the application or administration of a composition including one or more active agents to a subject, who has a disease, a symptom of the disease, or a predisposition toward the disease, with the purpose to cure, heal, alleviate, relieve, alter, remedy, ameliorate, improve, or affect the disease, the symptoms of the disease, or the predisposition toward the disease. An effective amount as used herein refers to the amount of each active agent required to confer therapeutic effect on the subject, either alone or in combination with one or more other active agents. Effective amounts vary, as recognized by those skilled in the art, depending on route of administration, excipient usage, and co-usage with other active agents.

[0043] The specific example below is to be construed as merely illustrative, and not limitative of the remainder of the disclosure in any way whatsoever. Without further elaboration, it is believed that one skilled in the art can, based on the description herein, utilize the present disclosure to its fullest extent. All publications cited herein are incorporated by reference herein in their entirety.

Example

[0044] Described below is an alternative strategy in which the enzymatic specificity of a normal endogenous enzyme is altered to compensate for the defective enzyme to help alleviate the antibody response. See FIG. 1. Because the modified enzyme will appear overall as a normal endogenous protein, this enzyme will be less immunogenic in LSD patients.

[0045] We employed human beta-glucuronidase as a template to generate alpha-iduronidase analogs. The expression of beta-glucuronidase is normal in MPS I patients, so recombinant beta-glucuronidase should be well tolerated and non-immunogenic. Beta-glucuronidase and alpha-iduronidase share a similar TIM ((3/a)8-barrel structure in their catalytic domains and belong to the same clan of glycoside hydrolase (GH-A). They also have similar catalytic mechanism which hydrolyzes substrates via a pair of glutamic acid residues, E451 and E540 for beta-glucuronidase, and E182 and E299 for alpha-iduronidase in a retentive fashion. In addition, in common with alpha-iduronidase, beta-glucuronidase can be targeted to lysosomes by receptor-mediated endocytosis via mannose-6-phosphate receptors present on the surface of deficient cells. Due to these similarities, beta-glucuronidase was selected as a candidate for specificity switching.

[0046] We constructed a beta-glucuronidase library and screened it for alpha-iduronidase activity using the Enzyme Cleavable Surface Tethered All-purpose Screen sYstem (ECSTASY) previously developed in our lab. See FIG. 2 and also Chen, C. P., et al., Protein Eng Des Sel, 2012. 25(7): p. 367-75. The beta-glucuronidase library was surface expressed by fusion to a C-terminal GPI-anchor signal sequence from human decay accelerating factor (DAF). See Caras, I. W., et al., Science, 1987. 238(4831): p. 1280-3. The membrane-bound beta-glucuronidase library cells were screened by high through-put fluorescence-activated cell sorting (FACS) followed by phosphatidylinositol-specific phospholipase C (PI-PLC) cleavage to facilitate screening for alpha-iduronidase activity under defined conditions. After screening, the in vitro effect of the beta-glucuronidase variants in alpha-iduronidase deficient cells was investigated.

[0047] We successfully isolated beta-glucuronidase variants that displayed significant alpha-iduronidase activity and exhibited phenotypic effects on MPS I cells. The data demonstrated that the specificity of a normally-expressed endogenous human enzyme can be shifted to compensate for a separate defective enzyme.

[0048] Wild-Type Human Beta-Glucuronidase Displays Detectable Alpha-Iduronidase Activity

[0049] Due to the similarity between human beta-glucuronidase and alpha-iduronidase, we sought to determine if human beta-glucuronidase displayed endogenous alpha-iduronidase activity.

[0050] We expressed and purified recombinant human beta-glucuronidase from human alpha-iduronidase deficient fibroblasts derived from MPS type I patients to eliminate possible contamination of the recombinant beta-glucuronidase with endogenous alpha-iduronidase. Recombinant beta-glucuronidase bearing a polyhistidine (6His) tag was purified by ammonium sulfate precipitation and Ni.sup.2+-nitrilotriacetic acid affinity chromatography.

[0051] In vitro assay showed that human beta-glucuronidase exhibited measurable activity against the alpha-iduronidase substrate, 4-methylumbelliferyl alpha-L-iduronide (MUI), corresponding to approximately 0.002% of the activity of wild-type human alpha-iduronidase. The hydrolysis of MUI was proportional to human beta-glucuronidase amount and incubation time. See FIG. 3, panel A. This activity was also confirmed by HPLC which showed the hydrolysis of MUI and an increase in the product 4-methylumbelliferone 4-MU. See FIG. 3, panels B and C. Note that commercial MUI contains 4-methylumbelliferyl beta-D-glucuronide (MUG) contamination ranging from 0.1 to 2% as stated in the datasheet provided as well as indicated by HPLC. See FIG. 3, panels D and E. Thus, the commercial MUI was purified as described in Materials and Methods before assaying to avoid signals from hydrolysis of MUG

[0052] Identification of Human Beta-Glucuronidase Variants Displaying Elevated Alpha-Iduronidase Activity

[0053] We screened human beta-glucuronidase variants for clones with higher alpha-iduronidase activity by ECSTASY. See FIG. 2. Briefly, a human beta-glucuronidase library with mutations at nineteen residues for a total diversity of 310.sup.9 was designed by structural analysis and literature review. See Table 1.

TABLE-US-00002 TABLE 1 Amino acid frequency of beta-glucuronidase variants with high alpha-iduronidase activity (n = 9) Expected frequency of each Frequency after ECSTASY Position position WT A.A. Mutant A.A. S E 447 0.50 9 0 N D H R G S C Y 484 0.125 0 3 1 2 0 2 1 0 N E D K 502 0.25 2 0 1 6 S A D Y P H F L V 503 0.11 0 1 1 4 2 0 0 0 1 Y G D C 504 0.25 4 4 0 1 S G R 506 0.33 2 7 0 Y D 508 0.50 3 5 H S D A P Y 509 0.16 2 0 0 2 5 0 G D 542 0.50 7 2 T A 545 0.50 5 4 L A 565 0.50 7 2 W T 587 0.50 7 2 F Y 592 0.50 2 7 T L 594 0.50 6 3 E L V Q G R 595 0.16 0 3 1 1 4 0 P D 598 0.50 3 6 R A 600 0.50 7 2 N Y S A T D 604 0.16 0 1 4 1 3 0 K P Q F L* 606 0.25 0 0 1 7 1 Underlined positions are putative hot spots identified from different prediction methods. *unexpected mutation

[0054] Mutations were introduced at nineteen positions in the human beta-glucuronidase gene by primer assembly followed by PCR amplification (see FIG. 4), resulting in 1.210.sup.7 bacterial colonies. Retroviral transduction of 293 cells with library DNA at a multiplicity of infection of 0.1 to ensure only a single beta-glucuronidase variant gene in each cell resulted in 310.sup.6 stable clones (293/L1 cells).

[0055] To remove the human beta-glucuronidase variants which cannot properly fold or be expressed on the surface of cells, we first stained live 293/L1 cells with mAb 7G8-FITC which binds to human beta-glucuronidase, and collected the cells which displayed relatively high levels of human beta-glucuronidase protein on their surface. Flow cytometry results indicated that 16% of 293/L1 cells expressed GPI-anchored human beta-glucuronidase on their surface (dashed gate, See FIG. 5, panel A). Cells exhibiting the highest human beta-glucuronidase expression level were sorted as single cells into 96-well microplates (6.8% of the population, solid gate, FIG. 5, panel A) and allowed to expand to near confluence. Each colony was then treated with PI-PLC to cut the GPI anchor and the concentration and activity of each solubilized beta-glucuronidase variant was assayed as described in Materials and Methods. For example, screening of several 96-well microtiter plates revealed three beta-glucuronidase variants that exhibited significantly higher alpha-iduronidase activity than wild-type beta-glucuronidase. See FIG. 5, panels B and C. In total, we screened fifty 96-well plates of sorted 293/L1 library cells and identified 1.3% (73/4800) of beta-glucuronidase variants with elevated alpha-iduronidase activity as compared to wild-type beta-glucuronidase.

[0056] Characterization of Human Beta-Glucuronidase Variants Displaying Alpha-Iduronidase Activity

[0057] Several human beta-glucuronidase variants which exhibited high alpha-iduronidase activity were randomly selected and cloned into a mammalian expression vector to produce greater amounts of recombinant soluble beta-glucuronidase from BALB/3T3 fibroblasts and 34/2000 cells (human alpha-iduronidase deficient fibroblasts derived from a MPS type I patient). All soluble human beta-glucuronidase variants displayed enhanced alpha-iduronidase activity as compared to wild-type beta-glucuronidase and the sequences were analyzed. See Table 1.

[0058] The amino acid sequences of selected clones are shown in Table 2. Three beta-glucuronidase variants, 102H1, 101C7, and 70H1, were further characterized. The recombinant human beta-glucuronidase variants showed similar molecular weights as wild-type human beta-glucuronidase as determined by immunoblotting with anti-6His tag antibody. The beta-glucuronidase variants also exhibited increased activity against MUI as compared to wild-type human beta-glucuronidase. See FIG. 6.

TABLE-US-00003 TABLE 2 Amino acid sequences of high alpha-iduronidase activity variants Amino acid residue Clones 204 279 438 447 484 502 503 504 506 508 509 542 Wild-type T Q K S N N S Y S Y H G 102H1 S D A G A D 101C7 H C K Y G G P 70H1 D K Y G D P 3D12 R K P G G D A D 1H9 H P G D 4G1 K M D K Y G D P 5A6 S V G G D 7H3 R K D C P 1A8 M D K Y G D P Amino acid residue Clones 545 565 587 592 594 595 598 600 603 604 606 636 Wild-type T L W F T E P R G N K P 102H1 A Y V S F 101C7 A A L Q A F 70H1 A Y L G D T F S 3D12 T Y G A Y Q 1H9 Y L D A E S L 4G1 A Y L G D T F 5A6 T Y L D S F 7H3 A L D S F 1A8 A Y L G D T F

[0059] The kinetic properties of human beta-glucuronidase variants against MUI were measured and analyzed. See Table 3. The substrate affinity K.sub.M to MUI of the human beta-glucuronidase variants 102H1, 101C7, and 70H1 were 36.93.2, 28.22.4, and 24.51.6 M, respectively. Compared to wild-type human beta-glucuronidase, these variants showed 19, 25, and 29-fold enhanced affinity to MUI, respectively. The enzyme turnover number k.sub.cat of the human beta-glucuronidase variants 102H1, 101C7, and 70H1 were 0.00990.0009, 0.0130.0011, and 0.00390.0003, which correspond to 11, 14, and 4-fold improvement as compared to wild-type beta-glucuronidase, respectively. The overall alpha-iduronidase activity of the three beta-glucuronidase variants were increased from 100 to 290-fold as compared to wild-type beta-glucuronidase. See Table 4. The enzyme specificity was shifted from beta-glucuronidase to alpha-iduronidase by a factor ranging from 7900 to 24500-fold. The beta-glucuronidase variants exhibited low but significant alpha-iduronidase activity ranging from 0.3 to 0.9% of wild-type alpha-iduronidase.

TABLE-US-00004 TABLE 3 Kinetic parameters of wild-type alpha-iduronidase (IDUA), beta- glucuronidase (G), and beta-glucuronidase variants for hydrolysis of 4-methylumbelliferyl alpha-L-iduronide (MUI) at pH 3.5 K.sub.m k.sub.cat k.sub.cat/K.sub.M (M) (s.sup.1) (s.sup.1 M.sup.1) IDUA 203 21 10.2 0.4 50100 3100 Wild-type G 705 7.0 0.0009 0.0002 1.22 0.35 102H1 36.9 3.2 0.0099 0.0009 270 24 101C7 28.2 2.4 0.013 0.0011 470 40 70H1 24.5 1.6 0.0039 0.0003 160 11 Results are presented as mean value SD of triplicate determinations.

TABLE-US-00005 TABLE 4 Relative enzyme activity and specificity change of wild-type and beta-glucuronidase (G) variants Activity relative Relative Relative Specificity to wild-type IDUA activity G activity shift IDUA Wild-type 1 1 1 0.002% 102H1 170 0.021 7900 0.514% 101C7 290 0.015 19200 0.895% 70H1 100 0.004 24500 0.306% The alpha-iduronidase (IDUA) and beta-glucuronidase (G) activity was assayed with 4-methylumbelliferyl alpha-L-iduronide (MUI) and 4-methylumbelliferyl beta-D-glucuronide (MUG), respectively. The relative enzyme activity was presented in fold increases of k.sub.cat/K.sub.M as compared to wild-type alpha-iduronidase and beta-glucuronidase. The specificity shift was presented in fold change of IDUA activity over G activity compared to wild-type beta-glucuronidase.

[0060] To address whether recombinant human beta-glucuronidase variants could alter the phenotype of MPS I cells, cellular GAG accumulation was measured by a SO.sub.4.sup.35 incorporation assay. MPS I cells were incubated with Na.sub.2.sup.35SO.sub.4 to radiolabel GAG before the cells were exposed to 5 g/ml recombinant enzyme for 72 h. Cell lysates and culture medium were then collected and the .sup.35S radioactivity was measured. Cells treated with wild-type alpha-iduronidase or the beta-glucuronidase variants 102H1 and 70H1 exhibited significantly reduced radioactivity in cell lysates as compared to untreated cells. See FIG. 7, panel A. The culture medium of MPS I cells which were treated with wild-type alpha-iduronidase and the three beta-glucuronidase variants exhibited significantly increased radioactivity as compared to cells treated with wild-type beta-glucuronidase (see FIG. 7, panel B), indicative of enhanced digestion and excretion of cellular GAG product. In summary, as compared to untreated cells and cells treated with wild-type beta-glucuronidase, cells treated with beta-glucuronidase variants 102H1 and 70H1 revealed significantly reduced cellular GAG and increased GAG excretion. Cells treated with beta-glucuronidase variant 101C7 displayed a non-statistically significant decrease of cellular GAG but significantly increased GAG excretion. Although the beta-glucuronidase variants were not as effective as wild-type alpha-iduronidase, these results indicate partial correction of GAG storage in MPS I cells.

[0061] We also employed a qualitative lysosomal staining method to visualize the phenotypic change in MPS I cells. MPS I cells were incubated with 5 g/ml of recombinant enzymes for 72 h and then stained with Lysotracker-red DND-99 dye (Invitrogen, Carlsbad, Calif., USA) to visualize the lysosomes (data not shown). The lysosome fluorescence was quantitated as mean fluorescence intensity per cell. High lysosomal staining was observed in non-treated MPS I cells. As expected, treatment of the cells with wild-type beta-glucuronidase did not affect lysosome fluorescence. By contrast, the cells treated with alpha-iduronidase or beta-glucuronidase variants (102H1, 101C7, and 70H1) displayed significantly reduced lysosomal staining as compared to non-treated MPS I cells (see FIG. 8), indicating at least partial normalization of lysosomes in the deficient cells. In summary, these results revealed that beta-glucuronidase variants displayed beneficial effects toward correction of the phenotype of MPS I cells.

[0062] The beta-glucuronidase variants are expected to display reduced immunogenicity as compared to alpha-iduronidase in MPS I patients because only several amino acids are changed from the wild-type beta-glucuronidase sequence. For example, the selected beta-glucuronidase variants 102H1, 101C7, and 70H1 possess 11, 13, and 13 amino acid changes, which corresponds to 1.7, 2, and 2% of the total amino acids. Besides, these mutations are mostly buried in the interior active pocket and may be inaccessible to antibodies.

[0063] Immunogenicity of Human Beta-Glucuronidase Variants in Human Beta-Glucuronidase Transgenic Mice

[0064] To investigate the immunogenicity of the beta-glucuronidase variants, an appropriate animal model such as a human beta-glucuronidase transgenic mouse is very useful. Human beta-glucuronidase transgenic mice can mimic MPS I patients, who express normal human beta-glucuronidase but not human alpha-iduronidase. These mice can be used to investigate the immunogenicity of human beta-glucuronidase variants as well as to test if host autoimmune responses to endogenous human beta-glucuronidase is induced. We thus generated transgenic mice that express human beta-glucuronidase. See FIG. 9.

[0065] To determine whether human beta-glucuronidase is tolerant in the transgenic mice, 50 g of recombinant proteins (i.e., human beta-glucuronidase, mouse beta-glucuronidase, human alpha-iduronidase, human beta-glucuronidase variant 101C7, and human beta-glucuronidase variant 70H1) were intravenously injected into transgenic mice every three weeks for a total of 4 injections. Serum antibodies against the administered proteins were determined by ELISA.

[0066] As expected, the mice tolerated repeated injections of mouse beta-glucuronidase (see FIG. 10, panel A), but developed antibodies against human alpha-iduronidase (see FIG. 10, panel B). By contrast, the mice did not generate antibodies against human beta-glucuronidase (see FIG. 10, panel C), indicating that human beta-glucuronidase appeared as a self-antigen. Two human beta-glucuronidase variants that display alpha-iduronidase activity (101C7 and 70H1) displayed significantly less immunogenicity in the transgenic mice than wild-type alpha-iduronidase (see FIG. 10, panel D), suggesting that the cloaked stealth approach may help reduce the immunogenicity of enzyme replacement therapy.

[0067] In summary, the transgenic mice express human beta-glucuronidase and were well tolerant to administration of wild-type human beta-glucuronidase. This animal model readily mimicked MPS I patients who express normal beta-glucuronidase but not alpha-iduronidase.

[0068] Materials and Methods

[0069] Reagents and Antibodies

[0070] Phosphatidylinositol-specific phospholipase C (PI-PLC), Lysotracker-Red DND-99 dye and Hoechst 33342 nuclear dye were from Invitrogen (Carlsbad, Calif., USA). 4-methylumbelliferyl beta-D-glucuronide (MUG) was from Sigma-Aldrich (St. Louis, Mo., USA). 4-methylumbelliferyl alpha-L-iduronide (MUI) was from USB Corporation (Cleveland, Ohio, USA). Trace MUG contamination (2%) in commercial MUI was removed by solid-phase extraction/high-performance liquid chromatography on a LiChroprep RP18 (40-63 m) column equilibrated with 20% methanol (pH 4). MUI was eluted with 25% methanol in double-distilled water (pH 4) and condensed in a rotavapor. Mouse anti-human beta-glucuronidase monoclonal antibody (mAb) 7G8 was directly labelled with FITC or biotin as described [34, 35]. Streptavidin-horse radish peroxidase (HRP) was from Jackson ImmunoResearch (West Grove, Pa., USA).

[0071] Cell Culture

[0072] GP293V cells (derived from human embryonic kidney 293 cells) were kindly provided by Dr. Andre Lieber, University of Washington, Seattle, Wash. 34/2000 cells (human alpha-iduronidase deficient fibroblasts derived from a MPS type I patient) were a kind gift from Dr. Mirella Filocamo, Istituto G Gaslini, Genova, Italy. BALB/3T3 fibroblasts and HEK293 cells were obtained from ATCC (Manassas, Va., USA). Cells were cultured in Dulbecco's minimal essential medium (DMEM) supplemented with 10% fetal bovine serum, 2.98 g/L HEPES, 2 g/L NaHCO.sub.3, 100 U/ml penicillin, and 100 g/ml streptomycin.

[0073] Structure Analysis of Human Beta-Glucuronidase and Alpha-Iduronidase

[0074] The protein structural alignment of human beta-glucuronidase, human alpha-iduronidase and Thermoanaerobacterium saccharolyticum beta-xylosidase indicated that they have conserved catalytic glutamic acid residues. 3D structures of human beta-glucuronidase (PDB ID: 1BHG), a human alpha-iduronidase homology model deduced from Thermoanaerobacterium saccharolyticum beta-xylosidase (PDB ID: 1Y24), and Thermoanaerobacterium saccharolyticum beta-xylosidase (PDB ID: 1PX8) were also used in the analysis. The catalytic TIM (/).sub.8-barrel domains were superimposed and analyzed by PyMOL and OPAAS, respectively. Although the whole protein structures of human beta-glucuronidase and alpha-iduronidase do not closely resemble each other, these proteins share a common TIM (/).sub.8-barrel structure and conserved glutamic acid residues in their catalytic pockets. Residues predicted to contact substrates were selected for mutation.

[0075] Synthetic Library Construction

[0076] A human beta-glucuronidase library with mutations at nineteen residues for a total diversity of 310.sup.9 was designed. Fifteen residues (S447, N484, N502, S503, Y504, Y508, H509, G542, W587, F592, T594, E595, R600, N604, and K606) in the beta-glucuronidase catalytic domain were identified as surrounding the active pocket in which substrates were accommodated. Previous research also reported several beta-glucuronidase residues associated with enzyme activity and specificity (N484, S503, S506, H509, T545, L565, E595, P598, N604, and K606). A total of nineteen amino acid residues were mutated. See Table 1. The six underlined residues (N484, S503, H509, E595, N604, and K606) were considered as hot spots because they were identified by both structure analysis and a review of the literature. The six hot spots were mutated to variable amino acids to enrich the library diversity. For example, we employed degenerate primers at S503 to mutate serine into amino acids with side chains which are positively charged (histidine), negatively charged (aspartic acid), aromatic (tyrosine and phenylalanine), hydrophobic (alanine, leucine and valine), and special in conformation (proline). The other residues were mutated to the corresponding amino acids which were identified from the structural comparison or in previous studies (Table 1). Primer assembly was used to generate the human beta-glucuronidase library (see FIG. 4). Briefly, the full-length beta-glucuronidase sequence was divided into five fragments with 18 overlapping nucleotides between each other (F0-4). The F0 fragment contained an Apa I cloning site as well as a silent mutation (nucleotide G741A) to remove a unique Bgl II cutting site in the human beta-glucuronidase gene to remove wild-type DNA contamination after library construction. The wild-type fragment F1 simply served as a bridge between the F0 and F2 fragments. The F2 and F3 fragments, which contained variable amino acids at nineteen positions, were constructed by primer assembly followed by PCR amplification. The wild-type F4 fragment contained a Sal I cloning site. All fragments were mixed in the same molar ratio and amplified by PCR to obtain a full-length beta-glucuronidase library which was digested with Apa I and Sal I and ligated into the same sites in pLNCX-hG-DAF to append a sequence coding for a GPI anchor to the C-terminus of the enzyme variants. The DNA library was digested by Bgl II to remove any wild-type DNA contamination and then transformed into DH5 competent cells by electroporation. Transformed bacteria were selected on 15-cm carbenicillin-containing LB agar plates for 16 h at 37 C. Plasmid DNA was purified from single colonies and sequenced to determine the mutation rates at selected residues. All expected mutations were present in the synthetic library. Colonies from multiple plates were collected and expanded in carbenicillin-containing LB medium. The plasmid was amplified by addition of chloramphenicol to a final concentration 170 g/ml when OD.sub.600 was 0.5. After overnight culture, plasmid DNA was purified by centrifugation in a CsCl-ethidium bromide density gradient at 60,000 rpm in a Ti 70.1 rotor for 24 h at 4 C. using a Beckman Optima L-90K ultracentrifuge (Beckman Coulter, Fullerton, Calif., USA).

[0077] Generation of Stable Library Cells

[0078] To generate stable cell libraries, library plasmid DNA was cotransfected with pVSV-G (Clontech, Mountain View, Calif., USA) into GP293V cells to produce recombinant retroviral particles. Two days after transfection, the culture medium was filtered, mixed with 8 g/ml polybrene, and incubated with 293 cells at a multiplicity of infection of 0.1. Stable cell lines were selected in medium containing 0.5 mg/ml G418 (Calbiochem, San Diego, Calif., USA). The resulting synthetic library cells were denoted as 293/L1 cells.

[0079] Flow Cytometer Analysis and Library Cell Selection

[0080] Human beta-glucuronidase surface expression was determined by staining 293/L1 cells with 7G8-FITC, which binds to human beta-glucuronidase, and measuring immunofluorescence of viable cells with a FACScaliber flow cytometer (BD Biosciences, Franklin Lakes, N.J., USA). Generally, 210.sup.7 cells were washed and suspended in 1 ml HBSS (5.4 mM KCl, 0.3 mM Na.sub.2HPO.sub.4, 0.4 mM KH.sub.2PO.sub.4, 4.2 mM NaHCO.sub.3, 1.3 mM CaCl.sub.2, 0.5 mM MgCl.sub.2, 0.6 mM MgSO.sub.4, 137 mM NaCl, 5.6 mM D-glucose, pH 7.4) containing 0.5% BSA and 20 g/ml 7G8-FITC for 30 min at 4 C. The cells were washed with ice-cold HBSS containing 0.5% BSA and suspended in 0.5% BSA/HBSS containing 5 g/ml propidium iodide. Cells were sorted on a FACSAria cell sorter (BD Biosciences, Franklin Lakes, N.J., USA). Dead cells (propidium iodide positive, high FL3 fluorescence) were gated out before 7G8-FITC immunofluorescence was detected at excitation/emission wavelengths of 488/515 nm (FL1). Single cells expressing surface human beta-glucuronidase were sorted into 96-well microplates in Dulbecco's minimal essential medium supplemented with 10% bovine serum.

[0081] Surface Enzyme Release and Enzyme Activity Screening

[0082] 293/L1 cells in 96-well microplates were washed once with PBS and incubated with 100 l PBS containing 50 mU/mL phosphatidylinositol phospholipase C (PI-PLC) at 37 C. for 1 h to cleave GPI-anchored beta-glucuronidase variants from surface of the cells. Alpha-iduronidase activity of the released beta-glucuronidase was assayed by mixing 20 l samples of cleaved enzyme with 80 l of 50 M 4-methylumbelliferyl alpha-L-iduronide (MUI) in 0.2 M formate buffer, pH 3.5 for 17 h at 37 C. The reaction was stopped by adding 100 l stop buffer (1 M glycine, 0.5 M sodium bicarbonate, pH 10.7) and the 4-methylumbelliferone (4-MU) fluorescence in the wells was measured at an excitation wavelength of 355 nm and an emission wavelength of 460 nm. To reduce the systematic error of manual volumetric transfers during large scale MUI assay and sandwich ELISA, an automated liquid handling system, MicroLab MPH-96 (Hamilton Robotics, Reno, Nev., USA), was employed. Kinetic parameters against MUI were determined by hydrolysis of serial diluted substrate (400 M) with defined concentrations of enzymes. The reaction was terminated at various time points and the fluorescence was measured. The acquired fluorescence was converted to product concentration by comparison with a 4-MU standard curve. Lineweaver-Burk plots were used to determine K.sub.M and k.sub.cat.

[0083] Sandwich Enzyme-Linked Immunosorbent Assay (ELISA)

[0084] The concentration of soluble beta-glucuronidase generated by PI-PLC cleavage of surface enzyme from individual colonies of the sorted 293-L1 cells was measured by sandwich ELISA. 0.1 g mAb 7G8 in 50 l coating buffer (50 mM Na.sub.2CO.sub.3, 50 mM NaHCO.sub.3, pH 8) was incubated in each well of 96-well ELISA plates at room temperature for 1 h. The plates were washed 3 times with PBS and then blocked with 2.5% skim milk in PBS at room temperature for 1 h. The plates were washed 3 times with PBS and a 20 l human beta-glucuronidase variant sample diluted to 50 l with PBS was transferred to each well for 1 h at room temperature. The plates were washed 3 times with PBS containing 0.05% Tween 20 before 20 ng 7G8-biotin and 50 ng streptavidin-HRP in 50 l PBS containing 2.5% skim milk were each subsequently added at room temperature for 1 h. After each step, the plates were washed 3 times with PBS containing 0.05% Tween 20. 150 l freshly prepared 2, 2-azino-bis (3-ethylbenzthiazoline-6-sulphonic acid) ABTS substrate was added at room temperature for 30 mM and the absorbance of each well was measured at 405 nm.

[0085] Lysosome Staining and Image Acquisition

[0086] A lysosomal staining method was employed to visualize the enzyme effect in MPS I cells. Briefly, MPS I cells were plated in 96-well microplates and incubated with 5.0 g/ml of recombinant enzymes for 72 h. Cells were washed with PBS and live stained with 100 l medium containing 100 nM Lysotracker-red DND-99 dye and 1 g/mL Hoechst 33342 for 30 min at 37 C. The cells were washed twice with PBS, replenished with 200 l DMEM without phenol red and live imaged on an ImageXpress Micro XL High-Content Screening System (Molecular Devices, CA, USA). The Lysotracker and Hoechst staining were visualized using TRITC (Em=54520, Ex=59320 nm) and DAPI (Ex=35050, Em=45550 nm) filters, respectively. Nine sites of images per well were recorded and analyzed by MetaXpress High Content Image Acquisition & Analysis Software (Molecular Devices, CA, USA).

[0087] Statistical Analysis

[0088] The two-tailed student t-test was used to calculate the significant differences between wild-type and beta-glucuronidase variants by Graphpad Prism 5.0 (GraphPad Software Inc., San Diego, Calif., USA). Data were considered significant at P values less than 0.05.

OTHER EMBODIMENTS

[0089] All of the features disclosed in this specification may be combined in any combination. Each feature disclosed in this specification may be replaced by an alternative feature serving the same, equivalent, or similar purpose. Thus, unless expressly stated otherwise, each feature disclosed is only an example of a generic series of equivalent or similar features.

[0090] From the above description, one skilled in the art can easily ascertain the essential characteristics of the described embodiments, and without departing from the spirit and scope thereof, can make various changes and modifications of the embodiments to adapt it to various usages and conditions. Thus, other embodiments are also within the claims.