EA039750B1 - Method for selection of high m6p recombinant proteins - Google Patents

Abstract

Described is a method for producing purified recombinant human acid alpha-glucosidase (rhGAA), wherein the purified rhGAA comprises an amino acid sequence that is at least 98% identical to SEQ ID NO: 2, the method comprising culturing in a bioreactor host cells that secrete the rhGAA; removing media from the bioreactor; filtering the media to provide a filtrate; loading the filtrate onto an anion exchange chromatography (AEX) column to capture the rhGAA; eluting a first protein product from the AEX column; loading the rhGAA eluted from the AEX column onto an immobilized metal affinity chromatography (IMAC) column; and eluting the rhGAA from the IMAC column.

Description

Technical field

The principles and embodiments of the present invention relate generally to the production of recombinant proteins, in particular lysosomal enzymes, which are characterized by a high content of mannose-6-phosphate.

Prerequisites

Lysosomal storage disorders are a group of autosomal recessive genetic disorders characterized by the accumulation of cellular glycosphingolipids, glycogen, or mucopolysaccharides in intracellular compartments called lysosomes. Individuals with these diseases carry mutant genes encoding enzymes that are defective in catalyzing the hydrolysis of one or more of these compounds, which then accumulate in lysosomes. For example, Pompe disease, also known as acid maltase deficiency or type II glycogen storage disease, is one of several lysosomal storage disorders. Other examples of lysosomal disorders include Gaucher's disease, GM1 gangliosidosis, fucosidosis, mucopolysaccharidoses, Hurler-Scheie disease, Niemann-Pick types A and B, and Fabry disease. Pompe disease is also classified as a neuromuscular disease or metabolic myopathy.

Pompe disease is estimated to occur in approximately 1 in 40,000 newborns and is caused by a mutation in the GAA gene, which codes for the enzyme lysosomal α-glucosidase (EC:3.2.1.20), also commonly known as acid α-glucosidase. Acid α-glucosidase is involved in the metabolism of glycogen, a branched polysaccharide that is the main form of glucose storage in animals, by catalyzing its hydrolysis to glucose in lysosomes. Because individuals with Pompe disease produce a mutant, defective acid α-glucosidase that is inactive or under-active, glycogen breakdown is slower or non-existent, and glycogen is stored in the lysosomes of various tissues, particularly striated muscle, leading to a wide range of clinical manifestations, including progressive muscle weakness and respiratory failure. In particular, tissues such as cardiac and skeletal muscles are affected.

Pompe disease can vary widely in terms of enzyme deficiency, severity, and age of onset, with more than 500 different mutations in the GAA gene identified, many of which cause disease symptoms of varying severity. The disease has been classified into main types: early onset or infantile form and late onset. Earlier onset of the disease and lower enzymatic activity are usually associated with a more severe course of the disease. The infantile form of Pompe disease is the most severe, being due to a complete or near complete deficiency of acid α-glucosidase, and presents with symptoms that include severe lack of muscle tone, weakness, enlarged liver and heart, and cardiomyopathy. The tongue may become enlarged and protrude, and swallowing may become difficult. The most severely affected children die of respiratory and cardiac complications before the age of two. Late onset Pompe disease can present at any age over 12 months and is characterized by no cardiac involvement and a better short-term prognosis. Symptoms are associated with progressive skeletal muscle dysfunction and include generalized muscle weakness and atrophy of the respiratory muscles in the trunk, proximal lower extremities, and diaphragm. Some adult patients do not have major symptoms or movement restrictions. The prognosis usually depends on the degree of damage to the respiratory muscles. Most subjects with Pompe disease eventually become physically debilitated, requiring the use of a wheelchair and assisted ventilation, with frequent premature death due to respiratory failure.

Current treatment options for Pompe disease include enzyme replacement therapy (ERT) using recombinant human acid α-glucosidase (rhGAA). Conventional rhGAA based products are known as alglucosidase-alpha, Myozyme® or Lumizyme® manufactured by Genzyme, Inc. ERT is a long-term treatment required throughout the life of the patient and involves the administration of a replacement enzyme by intravenous infusion. The replacement enzyme is then transported into the bloodstream and enters the lysosomes within the cells, where its action is to break down stored glycogen, which compensates for the deficiency in the activity of the endogenous defective mutant enzyme and thus reduces the severity of disease symptoms. In subjects with infancy-onset Pompe disease, treatment with alglucosidase-alpha has been shown to significantly improve survival compared to historical controls, and in late-onset Pompe disease, alglucosidase-alpha has been shown to be statistically significant, although not significant. , effect on 6-minute walk test (6MWT) and forced vital capacity (FVC) compared to placebo.

However, the condition of most subjects remains stable or continues to deteriorate despite being treated with alglucosidase-alpha. The reason for the marked suboptimal effect of ERT with alglucosidase-alpha is not clear, but this may be partly due to the progressive nature of the underlying muscle pathology or the weak targeting of existing ERT on tissues. For example, an infused enzyme is unstable at neutral pH, including plasma pH (approximately pH 7.4), and may be irreversibly inactivated in the circulation. In addition, infused alglucosidase-alpha exhibits insufficient uptake by major diseased muscles, possibly due to improper glycosylation by mannose-6-phosphate (M6P) residues. Such residues bind to cation-independent mannose-6-phosphate receptors (CIMPRs) on the cell surface, allowing the enzyme to enter the cell and its lysosomes. Therefore, high doses of enzyme may be necessary for effective treatment so that the proper amount of active enzyme can reach the lysosomes, making therapy costly and time consuming.

There are seven potential N-linked glycosylation sites in rhGAA. Because each glycosylation site is heterogeneous in the type of N-linked oligosaccharides (N-glycans) present, rhGAA is composed of a complex mixture of proteins with N-glycans having varying binding affinities for the M6P receptor and other carbohydrate receptors. RhGAA, which contains high-mannose N-glycans having one M6P group (mono-M6P), binds to CIMPR at low (~6000 nM) affinity, while rhGAA, which contains two M6P groups in the same N-glycan (6c-M6P), binds with high (~2 nM) affinity. Exemplary constructs for non-phosphorylated, mono-M6P and bis-M6P glycans are shown in FIG. 1A. The mannose-6-P group is shown in FIG. 1b. Once inside the lysosomes, rhGAA can enzymatically break down stored glycogen. However, traditional rhGAA have low total levels of glycans carrying M6P and 6c-M6P and thus are not sufficiently targeted to muscle cells resulting in poor delivery of rhGAA to lysosomes. Productive targeting of a rhGAA drug is shown in FIG. 2A. Most of the rhGAA molecules in these conventional products do not have phosphorylated N-glycans, resulting in a lack of affinity for CIMPR. The mannose receptor can also be cleared of high mannose non-phosphorylated glycans resulting in unproductive clearance of the ERT agent (FIG. 2B).

Also present in rhGAA are other types of N-glycans, complex carbohydrates that contain galactose and sialic acids. Because complex N-glycans are non-phosphorylated, they lack affinity for CIMPR. However, complex-type N-glycans with accessible galactose residues have a moderate to high affinity for the hepatic hepatocyte asialoglycoprotein receptor, resulting in rapid unproductive clearance of rhGAA (FIG. 2B).

Modern manufacturing methods used to make traditional rhGAA such as Myozyme®, Lumizyme® or alglucosidase-alpha have not significantly increased the content of M6P or bisM6P because the processing of cellular carbohydrates is inherently complex and extremely difficult to implement. Accordingly, there remains a need for further improvements in enzyme replacement therapy for the treatment of Pompe disease, such as new ways to produce, capture and purify rhGAA.

Likewise, other recombinant proteins that target lysosomes, such as other lysosomal enzymes, also bind to CIMPR. However, the current manufacturing methods used to produce other conventional lysosome-targeting recombinant proteins do not produce recombinant proteins high in M6P or 6c-M6P. Accordingly, there also remains a need for further improvement in methods for the production, capture and purification of these other recombinant proteins.

Short description

The present invention relates to a method for producing purified recombinant human acid alpha glucosidase (rhGAA), wherein the purified rhGAA contains an amino acid sequence that is at least 98% identical to SEQ ID NO: 2, the method comprising culturing in a bioreactor host cells that secrete the recombinant human lysosomal protein, removing media from the bioreactor, filtering the media to obtain a filtrate, loading the filtrate on an anion exchange chromatography (AEX) column to capture the lysosomal protein and eluting the first protein product from the AEX column, loading the rhGAA eluted from the AEX column onto the column for immobilized metal ion affinity chromatography (IMAC) and eluting rhGAA from the IMAC column.

In one or more embodiments, the purified rhGAA contains a first potential N-glycosylation site, a second potential N-glycosylation site, a third potential N-glycosylation site, a fourth potential N-glycosylation site, a fifth potential N-glycosylation site, a sixth potential N-glycosylation site and the seventh potential N-glycosylation site.

In one or more embodiments, the method further comprises loading the rhGAA eluted from the IMAC column onto a third chromatographic column and eluting the rhGAA from the third chromatographic column.

In one or more embodiments, the third chromatographic column is selected from a cation exchange chromatography (CEX) column and a size exclusion chromatography (SEC) column.

In one or more embodiments, media filtration is selected from variable tangential flow filtration (ATF) and tangential flow filtration (TFF).

In one or more embodiments, the method further comprises inactivating the viruses in one or more of rhGAA eluted from the AEX column, rhGAA eluted from the IMAC column, and rhGAA eluted from the third chromatographic column.

In one or more embodiments, the method further comprises filtering the rhGAA eluted from the IMAC column or the rhGAA eluted from the third chromatographic column to obtain a filtered product. In one or more embodiments, the method further comprises lyophilizing the filtered product.

In one or more embodiments, the host cells include Chinese hamster ovary (CHO) cells.

In one or more embodiments, (i) at least 90% of the purified rhGAA binds to the cation-independent mannose-6-phosphate receptor (CIMPR) or (ii) at least 90% of the purified rhGAA contains an N-glycan bearing mono-mannose-6 -phosphate (mono-M6P) or bismannose-6-phosphate (6uc-M6P).

Various embodiments are listed below. It will be understood that the following embodiments may be combined not only as indicated below, but also in other suitable combinations in accordance with the scope of the present invention.

Brief description of graphic materials

Additional features of the present invention will become apparent from the following written description and the accompanying figures, which show the following.

In FIG. 1A shows non-phosphorylated high mannose glycan, mono-M6P glycan and bisM6P glycan.

In FIG. 1B shows the chemical structure of the M6P group.

In FIG. 2A describes the productive targeting of rhGAA to target tissues (eg, muscle tissues of a subject with Pompe disease) via M6P-bearing glycans.

In FIG. 2B depicts non-productive drug clearance from non-target tissues (eg, liver and spleen) or binding of M6P-free glycans to non-target tissues.

In FIG. 3A is a graphical representation of the CIMPR receptor (also known as the IGF2 receptor) and the domains of this receptor.

In FIG. Table 3B shows the binding affinity (nanomolar) of bis- and mono-M6P-bearing glycans to CIMPR, the binding affinity of high-mannose type glycans to mannose receptors, and the binding affinity of the desialylated complex glycan to asialoglycoprotein receptors. RhGAA, which has glycans carrying M6P and 6c-M6P, can bind productively to CIMPR in muscle.

In FIG. 4A and 4B are respectively graphs showing CIMPR affinity chromatography results for Lumizyme® and Myozyme®. The dotted lines refer to the elution gradient with M6P. Elution with M6P displaces GAA molecules bound to CIMPR via an M6P-containing glycan. As shown in FIG. 2A, 78% of the active GAA form in Lumizyme® was eluted prior to the addition of M6P. In FIG. 2B shows that 73% of the active form of GAA in Myozyme® was eluted prior to the addition of M6P. Only 22% or 27% of rhGAA respectively in Lumizyme® or Myozyme® was eluted with M6P. These figures show that most of the rhGAA in these two traditional rhGAA based products lack the M6P glycans that are required to target CIMPR in target muscle tissues.

In FIG. 5 shows a DNA construct for transforming CHO cells with DNA encoding rhGAA. CHO cells were transformed with a DNA construct encoding rhGAA.

Fig. 6 is a schematic diagram of an exemplary method for the production, capture and purification of a recombinant lysosomal protein.

Fig. 7A and 7B respectively are graphs showing CIMPR affinity chromatography results for rhGAA as Myozyme® and ATB200. As shown in FIG. 7B, approximately 70% of the rhGAA molecules in rhGAA ATB200 contain M6P.

Fig. 8 is a graph showing CIMPR affinity chromatography results for rhGAA ATB200 with and without capture on an anion exchange column (AEX).

Fig. 9 is a graph showing elution profiles with Polywax

- 3 039750 for rhGAA Lumizyme® and ATB200.

Fig. 10 is a table that summarizes Lumizyme® N-glycan structures compared to three different rhGAA ATB200 formulations, designated as

BP-rhGAA, ATB200-1 and ATB200-2.

In FIG. 11A-11H show the results of a site-specific N-glycosylation analysis of rhGAA ATB200.

Fig. 12A is a graph comparing CIMPR binding affinity for rhGAA ATB200 (left curve) and for Lumizyme® (right curve).

Fig. 12B is a table comparing Bis-M6P content in rhGAA Lumizyme® and ATB200.

Fig. 13A is a graph comparing rhGAA ATB200 activity (left curve) with Lumizyme® rhGAA activity (right curve) in normal fibroblasts at various GAA concentrations.

Fig. 13B is a table comparing rhGAA ATB200 activity (left curve) with Lumizyme® rhGAA activity (right curve) in fibroblasts from a Pompe disease subject at various GAA concentrations.

Fig. 13C is a table comparing K _uptake and _U i of fibroblasts obtained from healthy subjects and those with Pompe disease.

Fig. 14A is a graph showing the amount of glycogen versus dose of human recombinant acid α-glucosidase in mouse heart muscle after contact with an inert medium (negative control), with 20 mg/ml alglucosidase-alpha (Lumizyme®) or with 5, 10 or 20 mg/kg ATB200.

Fig. 14B is a graph showing the amount of glycogen versus dose of human recombinant acid α-glucosidase in mouse quadriceps after contact with an inert medium (negative control), with 20 mg/ml alglucosidase-alpha (Lumizyme®) or with 5, 10 or 20 mg/kg ATB200.

Fig. 14C is a graph showing the amount of glycogen versus dose of human recombinant acid α-glucosidase in mouse triceps after contact with an inert medium (negative control), with 20 mg/ml alglucosidase-alpha (Lumizyme®) or with 5, 10 or 20 mg/kg ATB200.

Fig. 15 is a table showing that the combination of rhGAA ATB200 and the miglustat chaperone provided significantly better glycogen clearance in Gaa knockout mice compared to either rhGAA Lumizyme® or ATB200 treatments without the miglustat chaperone.

Fig. 16 is a series of electron micrographs of cardiac muscle, diaphragm, and soleus muscle of wild-type and Gaa knockout mice treated with inert medium, alglucosidase-alpha, and ATB200 in the presence and absence of miglustat, showing levels of lysosome-associated membrane protein. (LAMP-1).

Fig. 17 is a series of electron micrographs of cardiac muscle, diaphragm, and soleus muscle of wild-type and Gaa knockout mice treated with inert medium, alglucosidase-alpha, and ATB200 in the presence and absence of miglustat, showing glycogen levels when stained with Schiff's reagent/ iodic acid (PAS).

Fig. 18 is a series of electron micrographs (1000x) of the quadriceps muscles of wild-type and Gaa knockout mice treated with inert medium, alglucosidase-alpha and ATB200 in the presence and absence of miglustat, stained with methylene blue to show vacuoles (indicated by arrows). ).

Fig. 19 is a series of electron micrographs (40x) of the quadriceps muscles of wild-type and Gaa knockout mice treated with inert medium, alglucosidase alfa and ATB200 in the presence and absence of miglustat, showing the levels of autophagy markers, the microtubule-associated protein light chain 3 1A/1B, conjugated with phosphatidylethanolamine (LC3A II), and p62, insulin-dependent glucose transporter GLUT4 and insulin-independent glucose transporter GLUT1.

Fig. 20A and 20B, respectively, are graphs showing CIMPR affinity chromatography results for recombinant α-galactosidase A (rha-Gal A) enzyme before and after capture and purification on an anion exchange (AEX) column.

Fig. 21A and 21B are graphs showing wire hang and muscle grip strength data in WT and Gaa knockout mice treated with inert medium, alglucosidase-alpha and ATB200 in the presence of miglustat.

Fig. 22A-22G are graphs showing glycogen levels in quadriceps, triceps, and cardiac muscle cells of wild-type and Gaa knockout mice treated with inert medium, alglucosidase-alpha, and ATB200 in the presence and absence of miglustat.

- 4 039750

Fig. 23 is a series of photomicrographs (100x and 200x) of vastus lateralis (VL) muscle fibers of wild-type and Gaa knockout mice treated with inert medium, alglucosidase-alpha and ATB200 in the presence and absence of miglustat, showing signals dystrophin.

In FIG. Figure 24 shows the design of a phase 1/2, open-label, fixed-sequence, escalating-dose study, the first to be conducted in humans, to evaluate the safety, tolerability, PK, PD, and efficacy of intravenous infusions of ATB200 given in combination with oral miglustat. , in adults with Pompe disease.

Fig. 25A-25B are graphs showing plasma total GAA protein concentration versus time profiles in human subjects following a dose of 5, 10, or 20 mg/kg ATB200, 20 mg/kg ATB200, and 130 mg miglustat or 20 mg /kg ATB200 and 260 mg miglustat.

Fig. 25C is a graph showing plasma AUCs for total GAA protein in human subjects following a dose of 20 mg/kg ATB200, 20 mg/kg ATB200 and 130 mg miglustat, or 20 mg/kg ATB200 and 260 mg miglustat.

Fig. 25D is a graph showing plasma total GAA protein concentration versus time profiles in two separate human subjects following a dose of 20 mg/kg of ATB200 and 260 mg of miglustat.

Fig. 26 is a graph showing the time profiles of miglustat plasma concentration in human subjects following a dose of 130 or 260 mg of miglustat.

Fig. 27A-27D are graphs showing changes in levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatine phosphokinase (CPK), and hexose tetrasaccharide (Hex4) in human patients after administration of increasing doses of ATB200 (5, 10, and 20 mg/kg ) followed by co-administration of ATB200 (20 mg/kg) and miglustat (130 and 260 mg).

Detailed description

Before describing several illustrative embodiments of the present invention, it should be understood that the present invention is not limited to the construction details or method steps set forth in the following description. The present invention is capable of other embodiments and may be practiced or carried out in various ways.

While specific reference is made to GAA, one of ordinary skill in the art will appreciate that the methods described herein can be used to generate, capture, and purify other recombinant proteins that target lysosomes, including, without limitation, the lysosomal enzyme α-galactosidase A.

Various aspects of the present invention relate to novel methods for the production, capture and purification of recombinant human lysosomal proteins, such as recombinant human acid α-glucosidase (rhGAA). Other aspects of the present invention relate to recombinant proteins obtained by the methods described herein, as well as to pharmaceutical compositions, methods of treatment and ways of using such recombinant proteins.

Definitions

The terms used in this description, as a rule, have their usual meanings in the art, in the context of the present invention and in the specific context in which each term is used. Certain terms are discussed below or elsewhere in this specification to provide additional guidance to the practitioner in describing the compositions and methods of the present invention and their preparation and use.

In the present description, except when the context otherwise requires due to language or necessary implied meaning, the word contain or variants such as contains or containing are used in the inclusive sense, i.e. to indicate the presence of the recited features, but without excluding the presence or addition of additional features in various embodiments of the present invention.

As used herein, the term lysosomal protein refers to any protein that targets lysosomes, such as a lysosomal enzyme. Examples of lysosomal enzymes and related diseases are shown in Table 1. 1 below.

- 5 039750

Lysosomal enzyme Disease Acid a-glucosidase Pompe disease α-G alactosidase A Fabry disease Acid β-glucosidase Gaucher disease a-L-iduronidase Hurler-Scheie disease Iduronate sulfatase Hunter disease β-galactosidase GM 1 - gangliosidosis; disease Morquio B β-glucuronidase Sly disease (MPS VI) a-fucosidase Fucosidosis Acid sphingomyelinase Niemann-Pick disease types A and B β-hexosaminidase A Tay-Sachs disease β-hexosaminidase B Sandhoff disease β-galactocerebrosidase Krabbe disease Acid ceramidase Farber's disease Heparan-X-sulfatase Sanfilippo disease type A (MPS Sha) α-N-acetylglucosaminidase Sanfilippo disease type B (MPS 1pb) a-glucosaminide-N acetyltransferase Sanfilippo disease type C (MPS Shs) N-acetylglucosamine-bsulfate sulfatase Sanfilippo D disease (MPS Illd) N-acetylgalactosamine bisulfate sulfatase Morquio disease A (MPS IVb) Arylsulfatase A Metachromatic leukodystrophy Arylsulfatase B Maroteau-Lami Syndrome (MPS VI) acid lipase wolf disease Acid a-mannosidase a-Mannosidosis Acid β-mannosidase β-Mannosidosis α-N-acetylneuraminidase Sialidosis α-N-acetylgalactosaminidase Schindler-Kanzaki disease X-aspartyl-C-glucosaminidase Aspartylglucosaminuria

Table 1

As used herein, the term Pompe disease, also referred to as acid maltase deficiency, glycogen storage disease type II (GSDII), and glycogenosis type II, is intended to refer to a genetic lysosomal storage disorder characterized by mutations in the gene

-6039750

GAA, which codes for the enzyme human acid α-glucosidase. The term includes, without limitation, forms of the disease with early and late onset, including, without limitation, the form of the disease.

Pompe with onset in infancy, adolescence, and adulthood.

As used herein, the term acidic α-glucosidase is intended to mean a lysosomal enzyme that hydrolyzes α-1,4 bonds between the D-glucose units of glycogen, maltose and isomaltose. Alternative names include, without limitation, lysosomal α-glucosidase (EC: 3.2.1.20); glucoamylase; 1,4-α-D-glucanglucohydrolase; amyloglucosidase; gammaamylase and exo-1,4-a-glucosidase. Human acid α-glucosidase is encoded by the GAA gene (Gene ID 2548 at the National Center for Biotechnology Information (NCBI)), which has been mapped to the long arm of chromosome 17 (location 17q25.2-q25.3). To date, more than 500 mutations have been identified in the human GAA gene, many of which are associated with Pompe disease. Mutations resulting in misfolding or misprocessing of the acid α-glucosidase enzyme include T1064C (Leu355Pro) and C2104T (Arg702Cys). In addition, GAA mutations that affect enzyme maturation and processing include Leu405Pro and Met519Thr. The activity of the acid α-glucosidase protein requires the presence of the conserved WIDMNE hexapeptide at amino acid residues 516-521. As used herein, the abbreviation GAA is meant to mean the enzyme acid α-glucosidase, while the italicized abbreviation GAA is meant to mean the human gene encoding the enzyme human acid α-glucosidase. The italicized abbreviation Gaa is meant to refer to non-human genes encoding non-human acid α-glucosidase enzymes, including but not limited to rat or mouse genes, and the abbreviation Gaa is meant to refer to non-human acid α-glucosidase enzymes. Thus, the abbreviation rhGAA is meant to mean the enzyme recombinant human acid α-glucosidase.

As used herein, the term alglucosidase-alpha is meant to mean recombinant human acid α-glucosidase identified as [199-arginine,223 histidine]prepro-α-glucosidase (human); registration number 420794-05-0 according to the Chemical Abstracts Service. Alglucosidase-alpha has been approved for marketing in the US since January 2016 under the names Lumizyme® and Myozyme® by Genzyme.

As used herein, the term ATB200 is intended to refer to the human recombinant acid α-glucosidase described in co-pending patent application PCT/US 2015/053252, the disclosure of which is incorporated herein by reference.

As used herein, the term glycan is intended to mean a polysaccharide chain covalently linked to an amino acid residue in a protein or polypeptide. As used herein, the term N-glycan or N-linked glycan is meant to mean a polysaccharide chain attached to an amino acid residue in a protein or polypeptide by forming a covalent bond with the nitrogen atom of the amino acid residue. For example, an N-glycan may be covalently attached to the side chain nitrogen of an aspartic residue. Glycans may contain one or more monosaccharide units, and the monosaccharide units may be covalently linked to form a straight chain or a branched chain. In at least one embodiment, the N-glycan units attached to ATB200 may contain one or more monosaccharide units, each independently selected from N-acetylglucosamine, mannose, galactose, or sialic acid. N-glycan units in a protein can be determined using any suitable analytical technique such as mass spectrometry. In some embodiments, N-glycan units can be determined by tandem mass spectrometry liquid chromatography (LC-MS/MS) using instruments such as the Orbitrap Velos Pro™ mass spectrometer from Thermo Scientific, the Orbitrap Fusion Lumos Tribid™ mass spectrometer from Thermo Scientific or Waters Xevo® G2-XS QTof mass spectrometer.

As used herein, the term high mannose N-glycan is intended to mean an N-glycan having one to six or more mannose units. In at least one embodiment, the high mannose N-glycan unit may comprise a 6uc(Nacetylglucosamine) chain linked to an aspartic residue and further linked to a branched polymannose chain. As used herein, the interchangeable terms M6P or mannose-6-phosphate are intended to mean a mannose unit phosphorylated at position 6; those. having a phosphate group linked to a hydroxyl group at position 6. In at least one embodiment, one or more mannose units of one or more N-glycan units are phosphorylated at position 6 to form mannose-6 phosphate units. In at least one embodiment, the terms M6P or mannose-6phosphate refer to a mannose phosphodiester having N-acetylglucosamine (GlcNAc) in

- 7 039750 as a cap in the phosphate group, as well as the mannose unit, which has an accessible phosphate group devoid of the GlcNAc cap. In at least one embodiment, the N-glycans of a protein may have multiple M6P groups, wherein at least one M6P group has a GlcNAc cap and at least one other M6P group lacks a GlcNAc cap.

As used herein, the term complex N-glycan is meant to mean an N-glycan containing one or more galactose and/or sialic acid units. In at least one embodiment, the N-glycan complex can be a high mannose N-glycan wherein one or more mannose units are further linked to one or more monosaccharide units each independently selected from N-acetylglucosamine, galactose, and sialic acid.

As used herein, the compound miglustat, also known as N-butyl-1deoxynojirimycin NB-DNJ, or (2R,3R,4R,5S)-1-butyl-2-(hydroxymethyl)piperidine-3,4,5-triol, is is a compound having the following chemical formula:

is he

is he

One miglustat-based formulation is marketed under the trade name Zavesca® as a monotherapy for type 1 Gaucher's disease.

As discussed below, pharmaceutically acceptable salts of miglustat can also be used in the present invention. In the case of the salt miglustat, the salt dosage is adjusted so that the dose of miglustat received by the patient is equivalent to the amount that would be obtained with the use of miglustat free base.

As used herein, the compound duvoglustat, also known as 1deoxynojirimycin or DNJ or (2R,3R,4R,5S)-2-(hydroxymethyl)piperidine-3,4,5-triol, is a compound having the following chemical formula:

is he

is he

If the duvoglustat salt is used, the dosage of the salt is adjusted so that the dose of duvoglustat received by the patient is equivalent to the amount that would be obtained with duvoglustat free base.

As used herein, the term pharmacological chaperone, or sometimes simply the term chaperone, is intended to mean a molecule that specifically binds to a lysosomal protein and has one or more of the following effects:

improves the formation of a stable molecular conformation of the protein;

improves proper protein transport from the endoplasmic reticulum to another location in the cell, preferably a native location in the cell, in order to prevent protein degradation associated with the endoplasmic reticulum;

prevents aggregation of conformationally unstable or misfolded proteins;

at least partially restores and/or improves the function, stability and/or activity of the wild-type protein and/or improves the phenotype or function of the cell containing the protein.

Thus, the pharmacological chaperone for acid α-glucosidase is a molecule that binds to acid α-glucosidase, which causes proper folding, transport, non-aggregation and activity of acid α-glucosidase. This term, as used herein, includes, without limitation, active site-specific chaperones (ASSCs) that bind to the active site of an enzyme, inhibitors or antagonists, and agonists. In at least one embodiment, the pharmacological chaperone may be an inhibitor or antagonist of acid α-glucosidase. As used herein, the term antagonist is intended to mean any molecule that binds to acid α-glucosidase and partially or completely blocks, inhibits, reduces or neutralizes acid α-glucosidase activity. In at least one embodiment, the pharmacological chaperone is miglustat. Another non-limiting example of a pharmacological chaperone for acid α-glucosidase is duvoglustat.

As used herein, the term active site is intended to mean a region of a protein that is associated with, and required for, a particular biological activity of the protein. In at least one embodiment, the active site may be a site that binds to a substrate or other binding partners and provides amino acid residues directly involved in the formation and breaking of chemical bonds.

As used herein, a therapeutically effective dose and effective amount are intended to mean an amount of recombinant human lysosomal protein (eg, rhGAA) and/or chaperone and/or combinations thereof that is sufficient to elicit a therapeutic response in a subject. A therapeutic response may be any response that the user (eg, consultant physician) recognizes as an effective response to therapy, encompassing any surrogate clinical markers or symptoms described herein and known in the art. Thus, in at least one embodiment, the therapeutic response may be a reduction in intensity or inhibition of one or more symptoms or markers of Pompe disease known in the art. Symptoms or markers of Pompe disease include, without limitation, reduced tissue acid α-glucosidase activity; cardiomyopathy; cardiomegaly; progressive muscle weakness, especially in the trunk or lower extremities; deep hypotension; macroglossia (and in some cases protrusion of the tongue); difficulty swallowing, sucking and/or eating; respiratory failure; hepatomegaly (moderate); lethargy of the muscles of the face; areflexia; intolerance to physical activity; shortness of breath on exertion; orthopnea; sleep apnea; morning headaches; drowsiness; lordosis and/or scoliosis; decreased deep tendon reflexes; low back pain and mismatch with key stages of motor development. It should be noted that the concentration of a chaperone (e.g., miglustat) that has an inhibitory effect on acid α-glucosidase may be an effective amount for the purposes of the present invention due to dilution (and subsequent shift in binding due to a change in equilibrium state), bioavailability and metabolism of the chaperone when administered in vivo.

As used herein, the term enzyme replacement therapy or ERT is intended to mean the administration of a non-native purified enzyme to an individual deficient in that enzyme. The protein to be administered can be obtained from natural sources or through recombinant expression. The term also refers to the administration of a purified enzyme to an individual otherwise requiring or benefiting from the administration of a purified enzyme. In at least one embodiment, such an individual suffers from an enzyme deficiency. The enzyme administered may be a purified in vitro recombinant enzyme, or a protein purified from isolated tissue or fluid, such as, for example, placenta or animal milk, or from plants.

As used herein, the term combination therapy is intended to mean any therapy in which two or more separate therapeutic agents are administered simultaneously or sequentially. In at least one embodiment, the results of combination therapy are improved compared to the effect of each type of therapy carried out separately. Enhancement may include any improvement in the effect of various therapies that may result in a favorable outcome compared to those achieved with those therapies alone. Enhanced effect or results may include synergistic enhancement, where the enhanced effect is greater than the additive effects of each of the therapies administered alone; additive amplification, where the enhanced effect is essentially equal to the additive effect of each of the therapies carried out separately; or a less synergistic effect, where the enhanced effect is weaker than the additive effect of each of the therapies administered alone, but still better than the effect of each of the treatments alone. Enhanced effect can be measured using any of the methods known in the art that measure the effectiveness or outcome of a treatment.

As used herein, the term pharmaceutically acceptable is intended to mean molecular entities and compositions that are physiologically tolerable and do not typically cause adverse reactions when administered to a human. As used herein, the term "pharmaceutically acceptable" preferably means approved by a regulatory agency of the federal or state government or referred to in the USP or other generally accepted pharmacopoeia for use in animals and, more specifically, in humans.

As used herein, the term carrier is intended to mean a diluent, adjuvant, excipient, or inert medium with which a compound is administered. Suitable pharmaceutical carriers are known in the art and in at least one embodiment are described in Remington's Pharmaceutical Sciences, E. W. Martin, 18th edition or other editions.

As used herein, the terms subject or patient are intended to mean a human or non-human animal. In at least one embodiment, the subject is a mammal. In at least one embodiment, the subject is a human.

- 9 039750

As used herein, the term anti-drug antibody is intended to mean an antibody that specifically binds to a drug administered to a subject and elicited by the subject as at least part of the humoral immune response to administration of the drug to the subject. In at least one embodiment, the drug is a therapeutic protein drug. An anti-drug antibody present in a subject can elicit immune responses ranging from mild to severe, including, but not limited to, life-threatening immune responses, which include, but are not limited to, an anaphylactic reaction, cytokine release syndrome, and cross-neutralization of endogenous proteins mediating vital functions. In addition, or alternatively, an anti-drug antibody present in a subject may decrease the effectiveness of the drug.

As used herein, the term neutralizing antibody is intended to mean an antibody to a drug that acts to neutralize the function of the drug. In at least one embodiment, the therapeutic protein drug is an analog of an endogenous protein whose expression in the subject is reduced or absent. In at least one embodiment, the effect of neutralizing antibodies may be to neutralize the function of an endogenous protein.

As used herein, the terms approximately and approximately are intended to mean an acceptable degree of error for the quantity being measured, taking into account the nature or accuracy of the measurements. For example, the degree of error may be indicated by the number of significant numbers provided for the measurement, as understood in the art, and includes, without limitation, a ± 1 change in the most accurate significant number provided for the measurement. Typical illustrative degrees of error are within 20%, preferably within 10%, and more preferably within 5% of the specified value or range of values. Alternatively, and in particular in biological systems, the terms about and approximately can mean values that are within an order of magnitude, preferably within a 5-fold and more preferably within a 2-fold change in said value. Numerical values given in this document are approximate unless otherwise indicated, which means that the term approximately or approximately may be indicative unless specifically stated.

The term simultaneously, as used herein, is intended to mean at the same time as, or within a reasonably short period of time before or after, as will be understood by one skilled in the art. For example, if two treatments are administered simultaneously with each other, then one treatment may be administered before or after the other treatment, making allowance for the time required to prepare the last of the two treatments. Therefore, simultaneous administration of two treatments includes, without limitation, administration of one treatment agent following the other 20 minutes or less, about 20 minutes, about 15 minutes, about 10 minutes, about 5 minutes, about 2 minutes, about 1 minute or less than 1 min.

The term pharmaceutically acceptable salt as used herein is meant to mean a salt that, under careful medical judgment, is suitable for use in contact with tissues of humans and lower animals without non-specific toxicity, morbidity, allergic reaction, etc., is reasonable the ratio of risk and benefit, is usually water- or fat-soluble or dispersible and is effective for the intended application. The term includes pharmaceutically acceptable acid addition salts and pharmaceutically acceptable base addition salts. Lists of suitable salts are found, for example, in S. M. Birge et al., J. Pharm. Sc., 1977, 66, pp. 1-19, incorporated herein by reference.

The term pharmaceutically acceptable acid addition salt as used herein is intended to mean those salts that retain the biological effectiveness and properties of free bases and that are not biologically or otherwise undesirable formed with inorganic acids, including without limitation with hydrochloric acid, hydrobromic acid, sulfuric acid, sulfamic acid, nitric acid, phosphoric acid, etc., and organic acids, including but not limited to acetic acid, trifluoroacetic acid, adipic acid, ascorbic acid, aspartic acid, benzenesulfonic acid , benzoic acid, butyric acid, camphoric acid, camphorsulfonic acid, cinnamic acid, citric acid, digluconic acid, ethanesulfonic acid, glutamic acid, glycolic acid, glycerophosphoric acid, hemisulphate acid, hexanoic acid, formic acid, fumaric acid, 2-hydroxyethanesulfonic acid (isethionic acid), lactic acid, hydroxymaleic acid, malic acid, malonic acid, mandelic acid, mesitylene sulfonic acid, methanesulfonic acid, naphthalene sulfonic acid, nicotinic acid, 2-naphthalene sulfonic acid, oxalic acid acid, pamoic acid, pectic acid,

- 10 039750 phenylacetic acid, 3-phenylpropionic acid, pivalic acid, propionic acid, pyruvic acid, salicylic acid, stearic acid, succinic acid, sulfanilic acid, tartaric acid, p-toluenesulfonic acid, undecanoic acid, and the like.

The term pharmaceutically acceptable base addition salt as used herein is intended to mean those salts that retain the biological effectiveness and properties of free acids and that are not biologically or otherwise undesirable, formed with inorganic bases, including without limitation with ammonia or hydroxide, ammonium carbonate or bicarbonate, or a metal cation such as sodium, potassium, lithium, calcium, magnesium, iron, zinc, copper, manganese, aluminum, and the like. Salts derived from pharmaceutically acceptable organic non-toxic bases include, but are not limited to, salts of primary, secondary, and tertiary amines, quaternary ammonium compounds, substituted amines, including naturally occurring substituted amines, cyclic amines, and basic ion exchange resins such as methylamine, dimethylamine, trimethylamine, ethylamine, diethylamine, triethylamine, isopropylamine, tripropylamine, tributylamine, ethanolamine, diethanolamine, 2dimethylaminoethanol, 2-diethylaminoethanol, dicyclohexylamine, lysine, arginine, histidine, caffeine, hydrabamine, choline, betaine, ethylenediamine, glucosamine, methylglucamine, theobromine, purines, piperazine, piperidine, N-ethylpiperidine, tetramethylammonium compounds, tetraethylammonium compounds, pyridine, N,N-dimethylaniline, N-methylpiperidine, N-methylmorpholine, dicyclohexylamine, dibenzylamine, N,Ndibenzylphenethylamine, 1-ephenamine, N,N'-dibenzylethylenediamine , polyamine resins, etc.

rhGAA ATB200

In at least one embodiment, a recombinant human lysosomal protein (e.g., rhGAA) is expressed in Chinese hamster ovary (CHO) cells and has an increased content of N-glycan units bearing one or more mannose-6-phosphate residues compared to the content of N -glycan units bearing one or more mannose-6-phosphate residues in a conventional recombinant human lysosomal protein such as alglucosidase-alpha. In at least one embodiment, the acid α-glucosidase is recombinant human acid α-glucosidase, referred to herein as ATB200, as described in co-pending International Patent Application PCT7US2015/053252. ATB200 has been shown to bind to cation-independent mannose-6-phosphate receptors (CIMPR) with high affinity (KD - 2-4 nM) and is efficiently internalized by Pompe disease fibroblasts and skeletal muscle myoblasts (K _uptake - 7-14 nM) . ATB200 has been characterized in vivo and shown to have a shorter apparent plasma half-life (t _1/2 - 45 min) than alglucosidase alpha (t _1/2 - 60 min).

In at least one embodiment, the recombinant human acid α-glucosidase is an enzyme having the amino acid sequence shown under SEQ ID NO: 1 or SEQ ID NO: 2.

SEQ ID NO: Met Gly Vai Arg His Pro Pro Cys Ser His Arg Leu Leu Ala Vai

Cys Ala Leu Vai Ser Leu Ala Thr Ala Ala Leu Leu Gly His He

Leu Leu His Asp Phe Leu Leu Vai Pro Arg Glu Leu Ser Gly

Ser Ser Pro Vai Leu Glu Glu Thr His Pro Ala His Gin Gin Gly

Ala Ser Arg Pro Gly Pro Arg Asp Ala Gin Ala His Pro Gly Arg

Pro Arg Ala Vai Pro Thr Gin Cys Asp Vai Pro Pro Asn Ser Arg

- 11 039750

Phe Asp Cys Ala Pro Asp Lys Ala He Thr Gin Glu Gin Cys Glu Ala Arg Gly Cys Cys Tyr He Pro Ala Lys Gin Gly Leu Gin Gly Ala Gin Met Gly Gin Pro Trp Cys Phe Phe Pro Pro Ser Tyr Pro Ser Tyr Lys Leu Glu Asn Leu Ser Ser Ser Glu Met Gly Tyr Thr Ala Thr Leu Thr Arg Thr Pro Thr Phe Phe Pro Lys Asp He Leu Thr Leu Arg Leu Asp Vai Met Met Glu Thr Glu Asn Arg Leu His Phe Thr He Lys Asp Pro Ala Asn Arg Arg Tyr Glu Vai Pro Leu Glu Thr Pro Arg Vai His Ser Arg Ala Pro Ser Pro Leu Tyr Ser Vai Glu Phe Ser Glu Glu Pro Phe Gly Vai He Vai His Arg Gin Leu Asp Gly Arg Vai Leu Leu Asn Thr Vai Ala Pro Leu Phe Phe Ala Asp Gin Phe Leu Gin Leu Ser Thr Ser Leu Pro Ser Gin Tyr He Thr Gly Leu Ala Glu His Leu Ser Pro Leu Met Leu Ser Thr Ser Trp Thr Arg He Thr Leu Trp Asn Arg Asp Leu Ala Pro Thr Pro Gly Ala Asn Leu Tyr Gly Ser His Pro Phe Tyr Leu Ala Leu Glu Asp Gly Gly Ser Ala His Gly Vai Phe Leu Leu Asn Ser Asn Ala Met Asp Vai Vai Leu Gin Pro Ser Pro Ala Leu Ser Trp Arg Ser Thr Gly Gly He Leu Asp Vai Tyr HeP he Leu Gly Pro Glu Pro Lys Ser Vai Vai Gin Gin Tyr Leu Asp Vai Vai Gly Tyr Pro Phe Met Pro Pro Tyr Trp Gly Leu Gly Phe His Leu Cys Arg Trp Gly Tyr Ser Ser Thr Ala He Thr Arg Gin Vai Vai Glu Asn Met Thr Arg Ala His Phe Pro Leu Asp Vai Gin Trp Asn Asp Leu Asp Tyr Met Asp Ser Arg Arg Asp Phe Thr Phe Asn Lys Asp Gly Phe Arg Asp Phe Pro Ala Met Vai Gin Glu Leu His Gin Gly Gly Arg Arg Tyr Met Met He Vai Asp Pro Ala He Ser Ser Ser Gly Pro Ala Gly Ser Tyr Arg Pro Tyr Asp Glu Gly Leu Arg Arg Gly Vai Phe He Thr Asn Glu Thr Gly Gin Pro Leu He Gly Lys Vai Trp Pro Gly Ser Thr Ala Phe Pro Asp Phe Thr Asn Pro Thr Ala Leu Ala Trp Trp Glu Asp Met Vai Ala Glu Phe His Asp Gin Vai Pro Phe Asp Gly Met Trp He Asp Met Asn Glu Pro Ser Asn Phe He Arg Gly Ser Glu Asp Gly Cys Pro Asn Asn Glu Leu Glu Asn Pro Pro Tyr Vai Pro Gly Vai Vai

- 12 039750

Gly Gly Thr Leu Gin Ala Ala Thr He Cys Ala Ser Ser His Gin Phe Leu Ser Thr His Tyr Asn Leu His Asn Leu Tyr Gly Leu Thr Glu Ala He Ala Ser His Arg Ala Leu Vai Lys Ala Arg Gly Thr Arg Pro Phe Vai He Ser Arg Ser Thr Phe Ala Gly His Gly Arg Tyr Ala Gly His Trp Thr Gly Asp Vai Trp Ser Ser Trp Glu Gin Leu Ala Ser Ser Vai Pro Glu He Leu Gin Phe Asn Leu Leu Gly Vai Pro Leu Vai Gly Ala Asp Vai Cys Gly Phe Leu Gly Asn Thr Ser Glu Glu Leu Cys Vai Arg Trp Thr Gin Leu Gly Ala Phe Tyr Pro Phe Met Arg Asn His Asn Ser Leu Leu Ser Leu Pro Gin Glu Pro Tyr Ser Phe Ser Glu Pro Ala Gin Gin Ala Met Arg Lys Ala Leu Thr Leu Arg Tyr Ala Leu Leu Pro His Leu Tyr Thr Leu Phe His Gin Ala His Vai Ala Gly Glu Thr Vai Ala Arg Pro Leu Phe Leu Glu Phe Pro Lys Asp Ser Ser Thr Trp Thr Vai Asp His Gin Leu Leu Trp Gly Glu Ala Leu Leu He Thr Pro Vai Leu Gin Ala Gly Lys Ala Glu Vai Thr Gly Tyr Phe Pro Leu Gly Thr Trp Tyr Asp Leu Gin Thr Vai Pro He Glu Ala Leu Gly Ser Leu Pro Pro Pro Pro Pro Ala Ala Pro Arg Glu Pro Ala He His Ser Glu Gly Gin Trp Vai Thr Leu Pro Ala Pro Leu Asp Thr He Asn Vai His Leu Arg Ala Gly Tyr He Pro Leu Gin Gly Pro Gly Leu Thr Thr Glu Ser Arg Gin Gin Pro Met Ala Leu Ala Vai Ala Leu Thr Lys Gly Gly Glu Ala Arg Gly Glu Leu Phe Trp Asp Asp Gly Glu Ser Leu Glu Vai Leu Glu Arg Gly Ala Tyr Thr Gin Vai He Phe Leu Ala Arg Asn Asn Thr He Vai Asn Glu Leu Vai Arg Vai Thr Ser Glu Gly Ala Gly Leu Gin Leu Gin Lys Vai Thr Vai Leu Gly Vai Ala Thr Ala Pro Gin Gin Vai Leu Ser Asn Gly Vai Pro Vai Ser Asn Phe Thr Tyr Ser Pro Asp Thr Lys Vai Leu Asp He Cys Vai Ser Leu Leu Met Gly Glu Gin Phe Leu Vai Ser Trp Cys

SEQID NO. Q _n Q _n Qty q1 _a $ _er p _ro p _ro q^ q _§ r q^ q^ _a

Pro Gly Arg Pro Arg Ala Vai Pro Thr Gin Cys Asp Vai Pro Pro

- 13 039750

Asn Ser Arg Phe Asp Cys Ala Pro Asp Lys Ala He Thr Gin Glu Gin Cys Glu Ala Arg Gly Cys Cys Tyr He Pro Ala Lys Gin Gly Leu Gin Gly Ala Gin Met Gly Gin Pro Trp Cys Phe Phe Pro Pro Ser Tyr Pro Ser Tyr Lys Leu Glu Asn Leu Ser Ser Ser Glu Met Gly Tyr Thr Ala Thr Leu Thr Arg Thr Thr Pro Thr Phe Phe Pro Lys Asp He Leu Thr Leu Arg Leu Asp Vai Met Met Glu Thr Glu Asn Arg Leu His Phe Thr He Lys Asp Pro Ala Asn Arg Arg Tyr Glu Vai Pro Leu Glu Thr Pro Arg Vai His Ser Arg Ala Pro Ser Pro Leu Tyr Ser Vai Glu Phe Ser Glu Glu Pro Phe Gly Vai He Vai His Arg Gin Leu Asp Gly Arg Vai Leu Leu Asn Thr Vai Ala Pro Leu Phe Phe Ala Asp Gin Phe Leu Gin Leu Ser Thr Ser Leu Pro Ser Gin Tyr He Thr Gly Leu Ala Glu His Leu Ser Pro Leu Met Leu Ser Thr Ser Trp Thr Arg He Thr Leu Trp Asn Arg Asp Leu Ala Pro Thr Pro Gly Ala Asn Leu Tyr Gly Ser His Pro Phe Tyr Leu Ala Leu Glu Asp Gly Gly Ser Ala His Gly Vai Phe Leu Leu Asn Ser Asn Ala Met Asp Vai Vai Leu Gin Pro Ser Pro Ala Leu Ser Trp Arg Ser Thr Gly Gly He Leu Asp Vai Tyr He Phe Leu Gly Pro Glu Pro Lys Ser Vai Vai Gin Gin Tyr Leu Asp Vai Vai Gly Tyr Pro Phe Met Pro Pro Tyr Trp Gly Leu Gly Phe His Leu Cys Arg Trp Gly Tyr Ser Ser Thr Ala He Thr Arg Gin Vai Vai Glu Asn Met Thr Arg Ala His Phe Pro Leu Asp Vai Gin Trp Asn Asp Leu Asp Tyr Met Asp Ser Arg Arg Asp Phe Thr Phe Asn Lys Asp Gly Phe Arg Asp Phe Pro Ala Met Vai Gin Glu Leu His Gin Gly Gly Arg Arg Tyr Met Met He Vai Asp Pro Ala He Ser Ser Ser Gly Pro Ala Gly Ser Tyr Arg Pro Tyr Asp Glu Gly Leu Arg Arg Gly Vai Phe He Thr Asn Glu Thr Gly Gin Pro Leu He Gly Lys Vai Trp Pro Gly Ser Thr Ala Phe Pro Asp Phe Thr Asn Pro Thr Ala Leu Ala Trp Trp Glu Asp Met Vai Ala Glu Phe His Asp Gin Vai Pro Phe Asp Gly Met Trp He Asp Met Asn Glu Pro Ser Asn Phe He Arg Gly Ser Glu Asp Gly Cys Pro Asn Asn Glu Leu Glu Asn Pro Pro Tyr

- 14 039750

Val Pro Gly Val Val Gly Gly Thr Leu Gin Ala Ala Thr He Cys Ala Ser Ser His Gin Phe Leu Ser Thr His Tyr Asn Leu His Asn Leu Tyr Gly Leu Thr Glu Ala He Ala Ser His Arg Ala Leu Val Lys Ala Arg Gly Thr Arg Pro Phe Val He Ser Arg Ser Thr Phe Ala Gly His Gly Arg Tyr Ala Gly His Trp Thr Gly Asp Val Trp Ser Ser Trp Glu Gin Leu Ala Ser Ser Val Pro Glu He Leu Gin Phe Asn Leu Leu Gly Val Pro Leu Val Gly Ala Asp Val Cys Gly Phe Leu Gly Asn Thr Ser Glu Glu Leu Cys Val Arg Trp Thr Gin Leu Gly Ala Phe Tyr Pro Phe Met Arg Asn His Asn Ser Leu Leu Ser Leu Pro Gin Glu Pro Tyr Ser Phe Ser Glu Pro Ala Gin Gin Ala Met Arg Lys Ala Leu Thr Leu Arg Tyr Ala Leu Leu Pro His Leu Tyr Thr Leu Phe His Gin Ala His Val Ala Gly Glu Thr Val Ala Arg Pro Leu Phe Leu Glu Phe Pro Lys Asp Ser Ser Thr Trp Thr Val Asp His Gin Leu Leu Trp Gly Glu Ala Leu Leu He Thr Pro Val Leu Gin Ala Gly Lys Ala Glu Val Thr Gly Tyr Phe Pro Leu Gly Thr Trp Tyr Asp Leu Gin Thr Val Pro He Glu Ala Leu Gly Ser Leu Pro Pro Pro Pro Ala Ala Pro ArgGl u Pro Ala He His Ser Glu Gly Gin Trp Val Thr Leu Pro Ala Pro Leu Asp Thr He Asn Val His Leu Arg Ala Gly Tyr He He Pro Leu Gin Gly Pro Gly Leu Thr Thr Thr Glu Ser Arg Gin Gin Pro Met Ala Leu Ala Val Ala Leu Thr Lys Gly Gly Glu Ala Arg Gly Glu Leu Phe Trp Asp Asp Gly Glu Ser Leu Glu Val Leu Glu Arg Gly Ala Tyr Thr Gin Val He Phe Leu Ala Arg Asn Asn Thr He Val Asn Glu Leu Val Arg Val Thr Ser Glu Gly Ala Gly Leu Gin Leu Gin Lys Val Thr Val Leu Gly Val Ala Thr Ala Pro Gin Gin Val Leu Ser Asn Gly Val Pro Val Ser Asn Phe Thr Tyr Ser Pro Asp Thr Lys Val Leu Asp He Cys Val Ser Leu Leu Met Gly Glu Gin Phe Leu Val Ser Trp Cys

In at least one embodiment, the human recombinant acid α-glucosidase has the wild type GAA amino acid sequence set forth under SEQ ID NO: 1 as described in US Pat. . In at least one embodiment, the recombinant human acid α-glucosidase is glucosidase alpha, a human acid α-glucosidase enzyme encoded by the most predominant of the nine observed GAA gene haplotypes.

In at least one embodiment, the recombinant human acid α-glucosidase is initially expressed as having the full-length 952 amino acid wild-type GAA sequence shown under SEQ ID NO: 1, and the recombinant human acid α-glucosidase is intracellularly processed to remove part of the amino acids, for example the first 56 amino acids. Accordingly, the recombinant human acid α-glucosidase secreted by the host cell may have a shorter amino acid sequence than the recombinant human acid α-glucosidase that is originally expressed in the cell. In at least one embodiment, the shorter protein may have an amino acid sequence shown under SEQ ID NO: 2, which differs from SEQ ID NO: 1 only in that the first 56 amino acids containing the signal peptide and the precursor peptide have been deleted from thus resulting in a protein having 896 amino acids. Other differences in the number of amino acids are also possible, such as the presence of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more deletions, substitutions and/or insertions at compared to the amino acid sequence described under SEQ ID NO: 1 or SEQ ID NO: 2. In some embodiments, the rhGAA-based product contains a mixture of human recombinant acid α-glucosidase molecules having different amino acid lengths.

In at least one embodiment, the recombinant human acid α-glucosidase

- 15 039750 undergoes post-translational and/or chemical modifications in one or more amino acid residues of the protein. For example, methionine and tryptophan residues can be oxidized. As another example, N-terminal glutamine can form pyroglutamate. As another example, aspartic residues can be deamidated to aspartic acid. As another example, aspartic acid residues can be isomerized to isoaspartic acid. As another example, unpaired cysteine residues in a protein can form disulfide bonds with free glutathione and/or cysteine. Accordingly, in some embodiments, the enzyme is initially expressed as having the amino acid sequence listed under SEQ ID NO: 1 or SEQ ID NO: 2, and the enzyme undergoes one or more of these post-translational and/or chemical modifications. Such modifications are also within the scope of the present invention.

Polynucleotide sequences encoding GAA and similar human GAA variants are also contemplated and can be used for recombinant expression of rhGAA in accordance with the present invention.

Preferably, no more than 70%, 65%, 60%, 55%, 45%, 40%, 35%, 30%, 25%, 20%, 15%, 10%, or 5% of the total recombinant human lysosomal protein (e.g., rhGAA) molecules lack N-glycan a unit bearing one or more mannose-6-phosphate residues, or lacked the ability to bind to the cation-independent mannose-6-phosphate receptor (CIMPR). Alternatively, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 99%, <100% or more of a recombinant human lysosomal protein (e.g., rhGAA) molecules contain at least one N-glycan unit bearing one or more mannose-6-phosphate residues or capable of binding to CIMPR.

Recombinant human lysosomal protein molecules (eg, rhGAA) may have 1, 2, 3, or 4 mannose-6-phosphate (M6P) groups in their glycans. For example, only one N-glycan in a recombinant human lysosomal protein molecule can carry M6P (monophosphorylated), a single N-glycan can carry two M6P groups (bisphosphorylated), or each of two different N-glycans in the same recombinant human lysosomal protein molecule can carry separate M6P groups. Recombinant human lysosomal protein molecules can also have N-glycans that do not carry M6P groups. In another embodiment, the N-glycans average greater than 3 mol/mol M6P and greater than 4 mol/mol sialic acid such that the recombinant human lysosomal protein contains an average of at least 3 mol of mannose-6-phosphate residues per mol of recombinant lysosomal protein human and at least 4 moles of sialic acid per mole of recombinant human lysosomal protein. On average, at least about 3%, 4%, 5%, 6%, 7%, 8%, 9%, or 10% of the total glycans in a recombinant human lysosomal protein may be in the mono-M6P glycan form, e.g., about 6.25% of the total amounts of glycans can carry one M6P group, and on average at least about 0.5, 1, 1.5, 2.0, 2.5, 3.0% of the total glycans in recombinant human lysosomal protein are in the form of bis -M6P-glycan, and on average less than 25% of the total recombinant human lysosomal protein molecules do not contain a phosphorylated glycan that binds to CIMPR.

A recombinant human lysosomal protein (e.g., rhGAA) may have an average content of M6P-bearing N-glycans ranging from 0.5 to 7.0 mol/mol lysosomal protein, or any value in between in the subrange, including 0.5, 1 .0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5 or 7.0 mol/mol lysosomal protein. The lysosomal protein can be fractionated to produce lysosomal protein preparations with different average numbers of M6P-bearing or bis-M6P-bearing glycans, thus allowing for further individual tailoring of the targeting of the lysosomal protein to lysosomes in target tissues by selecting a specific fraction or by selectively combining different fractions.

In some embodiments, a recombinant human lysosomal protein (eg, rhGAA) will carry an average of 2.0 to 8.0 moles of M6P per mole of recombinant human lysosomal protein (eg, rhGAA). This range includes all intermediate values and subranges, including 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6, 5, 7.0, 7.5 and 8.0 mol M6P/mol recombinant human lysosomal protein (eg rhGAA).

Up to 60% of the N-glycans in a recombinant human lysosomal protein (eg rhGAA) may be fully sialylated, for example up to 10, 20, 30, 40, 50 or 60% of the N-glycans may be fully sialylated. In some embodiments, 4 to 20% of the total N-glycans are fully sialylated. In other embodiments, no more than 5%, 10%, 20%, or 30% of the N-glycans in a recombinant human lysosomal protein (eg, rhGAA) carry sialic acid and a terminal galactose (Gal) residue. This range includes all intermediate values and sub-ranges, for example, from 7 to 30% of the total number of N-glycans in a recombinant human lysosomal protein can carry sialic acid and terminal galactose. In a few other embodiments, no more than 5%, 10%, 15%, 16%, 17%, 18%, 19%, or 20% of the N-glycans in the recombinant human lysosomal protein have only terminal galactose and do not contain sialic acid. The

- 16 039750 the range includes all intermediate values and sub-ranges, for example, from 8 to 19% of the total number of N-glycans in the recombinant human lysosomal protein in the composition may have only terminal galactose and do not contain sialic acid.

In other embodiments, 40%, 45%, 50%, 55-60% of the total N-glycans in a recombinant human lysosomal protein (eg, rhGAA) are complex-type N-glycans; or not more than 1, 2, 3, 4, 5, 6, 7% of the total N-glycans in a recombinant human lysosomal protein (eg, rhGAA) are N-glycans of the hybrid type; no more than 5, 10, or 15% of the high-mannose type N-glycans in a recombinant human lysosomal protein (eg, rhGAA) are unphosphorylated; at least 5% or 10% of the high-mannose type N-glycans in a recombinant human lysosomal protein (eg, rhGAA) are mono-M6P-phosphorylated; and/or at least 1 or 2% of the high-mannose type N-glycans in a recombinant human lysosomal protein (eg, rhGAA) are 6c-M6P phosphorylated. These values include all intermediate values and sub-ranges. The recombinant human lysosomal protein (eg, rhGAA) may fall within one or more of the content ranges described above.

In some embodiments, a recombinant human lysosomal protein (eg, rhGAA) will carry an average of 2.0 to 8.0 moles of sialic acid residues per mole of recombinant human lysosomal protein (eg, rhGAA). This range includes all intermediate values and subranges, including 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6, 5, 7.0, 7.5 and 8.0 mol residues/mol recombinant human lysosomal protein (eg rhGAA). Without wishing to be bound by any theory, it is believed that the presence of N-glycan units bearing sialic acid residues can prevent unproductive clearance of recombinant human lysosomal protein (eg, rhGAA) by asialoglycoprotein receptors.

In one or more embodiments, the recombinant human lysosomal protein (eg, rhGAA) has M6P and/or sialic acid units at certain N-glycosylation sites of the recombinant human lysosomal protein. For example, as stated above, there are seven potential N-linked glycosylation sites in rhGAA. These potential glycosylation sites are at the following positions of SEQ ID NO: 2: N84, N177, N334, N414, N596, N826 and N869. Similarly, for the full length amino acid sequence of SEQ ID NO: 1, these potential glycosylation sites are at the following positions: N140, N233, N390, N470, N652, N882 and N925. Other rhGAA variants may have similar glycosylation sites depending on the location of the aspartic residues. Generally, the sequences ASN-XSER or ASN-X-THR in the amino acid sequence of a protein indicate potential glycosylation sites, except that X cannot be HIS or PRO.

In various embodiments, rhGAA has a specific N-glycosylation profile. In one or more embodiments, at least 20% of rhGAA is phosphorylated at the first N-glycosylation site (eg, N84 in SEQ ID NO: 2 and N140 in SEQ ID NO: 1). For example, at least 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95% of rhGAA may be phosphorylated at the first N-glycosylation site. This phosphorylation may result from the presence of mono-M6P and/or 6c-M6P units. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95% of rhGAA carry a mono-M6P moiety in the first N-glycosylation site. In some embodiments, at least 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, or 95% of rhGAA carry the 6c-M6P moiety in the first N-glycosylation site.

In one or more embodiments, at least 20% of rhGAA is phosphorylated at the second N-glycosylation site (eg, N177 in SEQ ID NO: 2 and N223 in SEQ ID NO: 1).

For example, at least 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95% of rhGAA may be phosphorylated at the second N-glycosylation site. This phosphorylation may result from the presence of mono-M6P and/or 6c-M6P units. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95% of rhGAA carry a mono-M6P moiety in second N-glycosylation site. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95% of rhGAA carry the 6c-M6P moiety in second N-glycosylation site. In one or more embodiments, at least 5% of rhGAA is phosphorylated at the third N-glycosylation site (eg, N334 in SEQ ID NO: 2 and N390 in SEQ ID NO: 1). In other embodiments, less than 5%, 10%, 15%, 20%, or 25% of rhGAA are phosphorylated at the third N-glycosylation site. For example, the third N-glycosylation site may have a combination of non-phosphorylated high mannose glycans, two-, three-, and four-antennary complex glycans, and hybrid glycans as major forms. In some embodiments, at least 3%, 5%, 8%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, or 50% of rhGAA are sialylated at the third N-glycosylation site.

- 17 039750

In one or more embodiments, at least 20% of rhGAA is phosphorylated at the fourth N-glycosylation site (eg, N414 in SEQ ID NO: 2 and N470 in SEQ ID NO: 1). For example, at least 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95% of rhGAA may be phosphorylated at the fourth N-glycosylation site. This phosphorylation may result from the presence of mono-M6P and/or 6c-M6P units. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, or 95% of rhGAA carry a mono-M6P moiety in fourth site of N-glycosylation. In some embodiments, at least 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, or 95% of rhGAA carry the 6c-M6P moiety in fourth site of N-glycosylation. In some embodiments, at least 3%, 5%, 8%, 10%, 15%, 20%, or 25% of rhGAA are sialylated at the fourth N-glycosylation site.

In one or more embodiments, at least 5% of rhGAA is phosphorylated at the fifth N-glycosylation site (eg, N596 in SEQ ID NO: 2 and N692 in SEQ ID NO: 1). In other embodiments, less than 5%, 10%, 15%, 20%, or 25% of rhGAA are phosphorylated at the fifth N-glycosylation site. For example, the fifth N-glycosylation site may have fucosylated two-antennary complex glycans as the main forms. In some embodiments, at least 3, 5, 8, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 6θ, 65, 70, 75, 80, 85, 90, or 95% rhGAA are sialylated at the fifth N-glycosylation site.

In one or more embodiments, at least 5% of rhGAA is phosphorylated at the sixth N-glycosylation site (eg, N826 in SEQ ID NO: 2 and N882 in SEQ ID NO: 1). In other embodiments, less than 5%, 10%, 15%, 20%, or 25% of rhGAA are phosphorylated at the sixth N-glycosylation site. For example, the sixth N-glycosylation site may have a combination of two-, three-, and four-antennary complex glycans as major forms. In some embodiments, at least 3, 5, 8, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, or 95% rhGAA are sialylated at the sixth N-glycosylation site.

In one or more embodiments, at least 5% of rhGAA is phosphorylated at the seventh N-glycosylation site (eg, N869 in SEQ ID NO: 2 and N925 in SEQ ID NO: 1). In other embodiments, less than 5%, 10%, 15%, 20%, or 25% of rhGAA are phosphorylated at the seventh N-glycosylation site. In some embodiments, less than 40%, 45%, 50%, 55%, 60%, or 65% of rhGAA have any glycan at the seventh N-glycosylation site. In some embodiments, at least 30%, 35%, or 40% of rhGAA have a glycan at the seventh N-glycosylation site.

In various embodiments, rhGAA has an average fucose content of 05 moles per mole of rhGAA, a GlcNAc content of 10-30 moles per mole of rhGAA, a galactose content of 5-20 moles per mole of rhGAA, a mannose content of 10-40 moles per mol of rhGAA, an M6P content of 2-8 mol per mol of rhGAA, and a sialic acid content of 2-8 mol per mol of rhGAA. In various embodiments, rhGAA has an average fucose content of 2-3 moles per mole of rhGAA, a GlcNAc content of 20-25 moles per mole of rhGAA, a galactose content of 8-12 moles per mole of rhGAA, a mannose content of 22-27 mol per mol of rhGAA, an M6P content of 3-5 mol per mol of rhGAA, and a sialic acid content of 4-7 mol per mol of rhGAA.

The recombinant human lysosomal protein (e.g., rhGAA) is preferably produced using Chinese hamster ovary (CHO) cells, such as the CHO GA-ATB-200 or ATB200-001-X5-14 cell lines, or a subculture or derivative of such a CHO cell culture. . DNA constructs that express allelic acid α-glucosidase variants or amino acid sequences of other acid α-glucosidase variants, such as at least 90%, 95%, 98%, or 99% identical to SEQ ID NO: 1 or SEQ ID NO: 2, can be construct and express in CHO cells. These amino acid sequences of acid α-glucosidase variants may contain deletions, substitutions and/or insertions compared to SEQ ID NO: 1 or SEQ ID NO: 2, such as having 1, 2, 3, 4, 5, 6, 7 , 8, 9, 10, 11, 12, 13, 14, 15 or more deletions, substitutions and/or insertions compared to the amino acid sequence described under SEQ ID NO: 1 or SEQ ID NO: 2. Those skilled in the art can select alternative vectors suitable for transforming CHO cells to obtain such DNA constructs.

To calculate the identity of two sequences, various alignment algorithms and/or programs can be used, including FASTA or BLAST, which are available as part of the GCG sequence analysis software package (University of Wisconsin, Madison, Wisconsin) and can be used, for example, with default settings. For example, polypeptides are provided that are at least 90%, 95%, 98%, or 99% identical to the specific polypeptides described herein, and preferably exhibit substantially the same functions, as well as polynucleotides encoding such polypeptides. Unless otherwise noted, the similarity score calculation will be

- 18 039750 based on the application of BLOSUM62. When using BLASTP, the percent similarity is based on the BLASTP Positive score and the percent sequence identity is based on the BLASTP Identical score. Identical in BLASTP show the number of residues in pairs of sequences with a high similarity score that are identical and their proportion of the total number of residues; a Positive in BLASTP shows the number and proportion of residues for which alignment scores are positive and are similar to each other. The present invention provides for and embraces amino acid sequences having these degrees of identity or similarity, or any intermediate degree of identity or similarity, to the amino acid sequences disclosed herein. Polynucleotide sequences of similar polypeptides are derived from the genetic code and can be obtained using conventional methods, in particular by restoring their amino acid sequences using the genetic code.

The present inventors have found that recombinant human α-glucosidase, which has an exceptional ability to target cation-independent mannose 6-phosphate receptors (CIMPR) and cellular lysosomes, as well as glycosylation patterns that reduce its unproductive excretion in vivo, can be produced using cells Chinese hamster ovary (CHO). These cells can be induced to express human recombinant acid α-glucosidase with significantly higher levels of N-glycan units bearing one or more mannose-6-phosphate residues than traditional human recombinant acid α-glucosidase products such as alglucosidase. -alpha. The recombinant human acid α-glucosidase produced by these cells, exemplified by ATB200, has a significantly higher content of mannose-6-phosphate (M6P) and bis-mannose-6-phosphate N-glycan residues for targeting muscle cells than the traditional acidic α-glucosidase such as Lumizyme®. Without wishing to be bound by any theory, it is believed that this extensive glycosylation allows the ATB200 enzyme to be more efficiently taken up by target cells, and therefore undergo more efficient clearance from the circulation, than other recombinant human acid α-glucosidases such as, for example, alglucosidase-alpha, which has a significantly lower content of M6P and 6c-M6P. ATB200 has been shown to potently bind to CIMPR and is efficiently taken up by skeletal muscle and cardiac muscle, and has a glycosylation pattern that provides a favorable pharmacokinetic profile and reduces waste in vivo.

It is also contemplated that extensive glycosylation of ATB200 may contribute to a reduction in the immunogenicity of ATB200 compared to, for example, alglucosidase-alpha. As will be recognized by those skilled in the art, glycosylation of proteins with conservative mammalian sugars generally improves the solubility of the product and reduces the aggregation and immunogenicity of the product. Glycosylation indirectly alters the immunogenicity of a protein by minimizing protein aggregation as well as by shielding immunogenic epitopes of the protein from the immune system (Guidance for Industry - Immunogenicity Assessment for Therapeutic Protein Products, U.S. Department of Health and Human Services, Food and Drug Administration , Center for Drug Evaluation and Research, Center for Biological Product Evaluation and Research, August 2014). Therefore, in at least one embodiment, the administration of recombinant human acid α-glucosidase does not induce the production of antibodies to the drug. In at least one embodiment, administration of human recombinant acid α-glucosidase induces anti-drug antibodies in the subject at a lower frequency than the level of anti-drug antibodies induced by administration of alglucosidase-alpha.

As described in co-pending International Patent Application PCT/US 2015/053252, cells such as CHO cells can be used to produce the rhGAA described herein, and this rhGAA can be used according to the present invention. Examples of such a CHO cell line are GA-ATB-200 or ATB-200-001-X5-14 or a subculture thereof that produces the rhGAA-based composition described herein. Such CHO cell lines may contain multiple copies of a gene, such as 5, 10, 15, or 20 or more copies of a GAA-encoding polynucleotide.

rhGAA rich in M6P and 6c-M6P, such as rhGAA ATB200, can be obtained by transforming CHO cells with a DNA construct that encodes GAA. Although CHO cells have previously been used to produce rhGAA, it has not been recognized that transformed CHO cells can be cultured and selected to produce rhGAA having a high content of M6P and bis-M6P glycans that target CIMPR.

Surprisingly, it has been found that it is possible to transform CHO cell lines, select transformants that produce rhGAA high in glycans bearing M6P or bis

- 19 039750

M6Ps that provide CIMPR targeting and stably express this rhGAA high in M6P. Thus, the methods for obtaining these CHO cell lines are also described in the international patent application PCT7US 2015/053252, which is simultaneously pending. The method comprises transforming a CHO cell with DNA encoding GAA or a GAA variant, selecting a CHO cell whose chromosome(s) stably integrates DNA encoding GAA and which stably expresses GAA, and selecting a CHO cell that expresses GAA having a high content of glycans bearing M6P or 6c-M6P; and optionally selecting a CHO cell having N-glycans high in sialic acid and/or having a low content of non-phosphorylated high-mannose N-glycans.

These CHO cell lines can be used to produce rhGAA and rhGAA based compositions by culturing CHO cell lines and extracting said composition from CHO cell culture.

Production, capture and purification of recombinant human lysosomal protein

Various embodiments of the present invention relate to methods for producing and/or capturing and/or purifying a recombinant human lysosomal protein (eg, rhGAA). An exemplary method 600 for producing, capturing, and purifying a recombinant human lysosomal protein is shown in FIG. 6.

In FIG. 6, arrows indicate the direction of movement for various liquid phases containing recombinant human lysosomal protein. The bioreactor 601 contains a culture of cells, such as CHO cells, that express and secrete a recombinant human lysosomal protein (eg, rhGAA) into the surrounding liquid culture medium. The bioreactor 601 may be any suitable cell culture bioreactor, such as a perfusion bioreactor, a batch bioreactor, or a batch bioreactor. In various embodiments, the bioreactor has a volume of from about 1 to about 20,000 liters. Exemplary bioreactor volumes include about 1 L, about 10 L, about 20 L, about 30 L, about 40 L, about 50 L, about 60 L, about 70 L, about 80 L, about 90 L, about 100 L, about 150 l, approximately 200 l, approximately 250 l, approximately 300 l, approximately 350 l, approximately 400 l, approximately 500 l, approximately 600 l, approximately 700 l, approximately 800 l, approximately 900 l, approximately 1000 l, approximately 1500 l, approximately 2000 l, approximately 2500 l, approximately 3000 l, approximately 3500 l, approximately 4000 l, approximately 5000 l, approximately 6000 l, approximately 7000 l, approximately 8000 l, approx. l.

As shown in FIG. 6, the medium can be removed from the bioreactor. Such media removal may be continuous for a perfusion bioreactor or may be intermittent for a batch or batch reactor. The medium is filtered using a filtration system 603 to remove cells. In some embodiments, cells removed from the medium are reintroduced into the bioreactor and the medium containing the recovered recombinant lysosomal protein can be further processed. The filtration system 603 may be any suitable filtration system, including a variable tangential flow filtration (ATF) system, a tangential flow filtration (TFF) system, a centrifugal filtration system, and the like. In various embodiments, a filter having a pore size of from about 10 nm to about 2 µm is used in the filtration system. Exemplary filter pore sizes include approximately 10 nm, approximately 20 nm, approximately 30 nm, approximately 40 nm, approximately 50 nm, approximately 60 nm, approximately 70 nm, approximately 80 nm, approximately 90 nm, approximately 100 nm, approximately 150 nm, approximately 200 nm, approximately 250 nm, approximately 300 nm, approximately 350 nm, approximately 400 nm, approximately 500 nm, approximately 600 nm, approximately 700 nm, approximately 800 nm, approximately 900 nm, approximately 1 µm, approximately 1.5 and approximately 2 µm.

In various embodiments, the media removal rate is from about 1 to about 20,000 L/day. Illustrative media removal rates include about 1 L/day, about 10 L/day, about 20 L/day, about 30 L/day, about 40 L/day, about 50 L/day, about 60 L/day, about 70 l/day, approximately 80 l/day, approximately 90 l/day, approximately 100 l/day, approximately 150 l/day, approximately 200 l/day, approximately 250 l/day, approximately 300 l/day, approximately 350 l/day day, approximately 400 l/day, approximately 500 l/day, approximately 600 l/day, approximately 700 l/day, approximately 800 l/day, approximately 900 l/day, approximately 1000 l/day, approximately 1500 l/day, approximately 2000 l/day approximately 2500 l/day approximately 3000 l/day approximately 3500 l/day approximately 4000 l/day approximately 5000 l/day approximately 6000 l/day approximately 7000 l/day approximately l/day, approximately 9000 l/day, approximately 10000 l/d day, approximately 15,000 l/day and approximately 20,000 l/day. Alternatively, the media removal rate can be expressed as a function of

- 20 039750 from the volume of the bioreactor, such as from about 0.1 to about 3 reactor volumes per day. Illustrative media removal rates include about 0.1, about 0.2, about 0.3, about 0.4, about 0.5, about 0.6, about 0.7, about 0.8, about 0.9 , about 1, about 1.1, about 1.2, about 1.3, about 1.4, about 1.5, about 2, about 2.5, and about 3 reactor volumes per day.

For a continuous or batch process, the feed rate of fresh medium to the bioreactor can be from about 1 to about 20,000 liters/day. Exemplary medium infusion rates include about 1 L/day, about 10 L/day, about 20 L/day, about 30 L/day, about 40 L/day, about 50 L/day, about 60 L/day, about 70 L /day, approximately 80 L/day, approximately 90 L/day, approximately 100 L/day, approximately 150 L/day, approximately 200 L/day, approximately 250 L/day, approximately 300 L/day, approximately 350 L/day , approximately 400 L/day, approximately 500 L/day, approximately 600 L/day, approximately 700 L/day, approximately 800 L/day, approximately 900 L/day, approximately 1000 L/day, approximately 1500 L/day, approximately 2000 l/day, approx. 2500 l/day, approx. 3000 l/day, approx. 3500 l/day, approx. 4000 l/day, approx. 5000 l/day, approx. /day, approx. 9000 l/day, approx. 10000 l/day, approx. between 15,000 and approximately 20,000 l/day. Alternatively, the rate of introduction of the environment can be expressed depending on the volume of the bioreactor, such as from about 0.1 to about 3 reactor volumes per day. Illustrative values for the rate of introduction of the medium include about 0.1, about 0.2, about 0.3, about 0.4, about 0.5, about 0.6, about 0.7, about 0.8, about 0.9 , about 1, about 1.1, about 1.2, about 1.3, about 1.4, about 1.5, about 2, about 2.5, and about 3 reactor volumes per day.

After filtration, the filtrate is loaded into the protein capture system 605. The protein capture system 605 may include one or more chromatographic columns. If more than one chromatography column is used, then the columns can be placed in series so that loading of the next column can begin immediately after loading the first column. Alternatively, the process of deleting the environment can be stopped for the duration of the column switch.

In various embodiments, the protein capture system 605 comprises one or more anion exchange (AEX) columns for direct capture of a recombinant human lysosomal protein product, particularly a lysosomal protein having a high M6P content. Without wishing to be bound by any particular theory, it is believed that the use of AEX chromatography to capture recombinant human lysosomal protein from filter media ensures that the captured recombinant protein product will have a higher M6P content due to the more negative charge of the recombinant protein having one or more groups. M6P. As a result, the non-phosphorylated recombinant protein and the host cell impurities are not bound by the column resin, just as the highly phosphorylated recombinant protein and the non-phosphorylated recombinant protein and the host cell impurities pass through the column. Accordingly, AEX chromatography can be used to enrich the M6P content of a protein product (ie, to select for proteins having more M6P) due to the high affinity of M6P containing proteins for AEX resin.

In addition, without wishing to be bound by any particular theory, it is also believed that the presence of direct capture of the recombinant protein product by AEX chromatography ensures that recombinant proteins having a high M6P content are removed from the medium containing proteases and other enzymes, which can degrade the protein and/or dephosphorylate the protein. The result is a high quality product.

Suitable columns for AEX chromatography have functional chemical groups that bind negatively charged proteins. Illustrative functional groups include, without limitation, primary, secondary, tertiary, and quaternary ammonium or amino groups. These functional groups may be associated with membranes (eg cellulose membranes) or conventional chromatographic resins. Exemplary column media include SP, CM Q, and DEAE Sepharose® Fast Flow media from GE Healthcare Lifesciences.

The volume of the AEX column may be any suitable volume, such as 1 to 1000 L. Exemplary column volumes include about 1 L, about 2 L, about 3 L, about 4 L, about 5 L, about 6 L, about 7 L, about 8 L, about 9 L, about 10 L, about 20 L, about 30 l, approximately 40 l, approximately 50 l, approximately 60 l, approximately 70 l, approximately 80 l, approximately 90 l, approximately 100 l, approximately 150 l, approximately 200 l, approximately 250 l, approximately 300 l, approximately 350 l, approximately 400 liters and approximately 500 liters, approximately 600 liters, approximately 700 liters, approximately 800 liters, approximately 900 and approximately 1000 liters.

- 21 039750

Illustrative conditions for an anion exchange column are given in table. 2 below.

table 2

Procedure Buffer Flow rate (cm/h) Volume (CV) Temperature P "( ) Sanitizing before use 0.1-10 M NaOH <25- 2500 > 1-3 (> 10-120 min.) 15-25 niem Prebalancing 20-2000 mM phosphate buffer (PB), pH 6.9-7.3 <25- 2500 > 1-5 15-25 Balancing 4-400 mM RV, pH 6.9- 7.3 <25- 2500 > 1-5 2-15 Loading No data < ΙΟ- Ι000 No data 2-15 Rinse 1 4-400 mM RV, pH 6.9- 7.3 <25- 2500 >2-10 2-15 Rinse 2 4-400 mM RV, pH 6.9- 7.3 <25- 2500 >2-10 15-25 Eluted ie 4-400 mM RV, 20-2000 MMNaCl, pH 6.1-6.5 <25- 2500 No data 15-25 Desorbirov ing 4-400 mM RS, 0.1-10 M NaCl, pH 6.1-6.5 <25- 2500 > 1-5 15-25 Sanitary processing after use 0.1-10 M NaOH <25- 2500 > 1-3 (> 10-120 min.) 15-25 nia Storage 0.01-1.0 M NaOH <25- 2500 > 1-5 15-25

After the recombinant human lysosomal protein is loaded into the protein capture system 605, the recombinant human lysosomal protein is eluted from the column(s) by changing the pH and/or salt content of the column.

The eluted recombinant human lysosomal protein may be subjected to additional purification steps and/or quality assurance steps. For example, as shown in FIG. 6, the eluted recombinant human lysosomal protein can be subjected to a virus killing step 607 . Such killing of viruses 607 may include one or more of low pH killing, detergent killing, or other technique known in the art.

- 22 039750

The recombinant protein product after the virus killing step 607 can be introduced into the second chromatographic system 609 for further purification of the recombinant protein product. Alternatively, the eluted recombinant protein from the protein capture system 605 can be fed directly to the second chromatography system 609. In various embodiments, the second chromatography system 609 includes one or more immobilized metal ion affinity chromatography (IMAC) columns to further remove impurities. Exemplary metal ions include cobalt, nickel, copper, iron, zinc, or gallium ions.

The volume of the second chromatographic column (eg, IMAC column) may be any suitable volume, such as 0.1 to 100 L. Exemplary column volumes include about 0.1 L, about 0.2 L, about 0.3 L, about 0.4 L, about 0.5 L, about 0.6 L, about 0.7 L, about 0.8 l, approximately 0.9 l, approximately 1 l, approximately 1.5 l, approximately 2 l, approximately 2.5 l, approximately 3 l, approximately 3.5 l, approximately 4 l, approximately 4.5 l, approximately 5 l, approximately 6 l, approximately 7 l, approximately 8 l, approximately 9 l, approximately 10 l, approximately 15 l, approximately 20 l, approximately 25 l, approximately 30 l, approximately 35 l, approximately 40 l and approximately 50 l, approximately 60 liters, approximately 70 liters, approximately 80 liters, approximately 90 liters and approximately 100 liters.

Exemplary column conditions for IMAC are shown in Table 1. 3 below.

_____________________________________________________________ Table 3

Procedure Buffer Flow rate (cm/h) Volume (CV) Rinsing 4-400 mM RV, pH 6.3-6.7 < 25-2500 > 1-5 Sanitization before use 0.01-1.0 MNaOH < 25-2500 > 1-3 (10 - 30 min)

- 23 039750

Balancing 4-400 mM RV, pH 6.5 < 25-2500 > 1-5 Washing with WFI Water for injection (WFI) < 25-2500 > 1-3 Chelation 0.01-1.0 M copper acetate < 25-2500 > 1-5 Washing with WFI WFI < 25-2500 >2-10 Washing with acidic buffer 2-200 mM sodium acetate, 0.05-5 M NaCl, pH 3.5-4.5 < 25-2500 >2-10 Balancing 4-400 mM RV, pH 6.3-6.7 < 25-2500 > 1-5 Blank run with elution buffer 4-400 mMP, 15-1500 mM glycine, pH 6.1-6.5 < 25-2500 >2-20 Balancing 4-400 mM RV, pH 6.3-6.7 < 25-2500 > 1-5 Loading No data < 25-2500 > 1-5 Flushing 1 4-400 mM RV, pH 6.3-6.7 < 25-2500 >2-10 Flushing 2 4-400 mMP, 0.1-10 M NaCl, 5-30% propylene glycol, pH 6.36.7 < 25-2500 >2-10 Wash 3 4-400 mM RV, pH 6.3-6.7 < 25-2500 >2-10 Elution Desorption Sanitization after use 4-400 mMP, 15-1500 mM glycine, pH 6.1-6.5 4-400 mM RS, 50-5000 mM imidazole, pH 6.3-6.7 0.01-lMNaOH < 25-2500 < 25-2500 < 25-2500 No data > 1-5 > 1-3 (10 - 30 min) Rinsing 4-400 mM RV, pH 6.3-6.7 < 25-2500 > 1-5 Storage 5-30% ethanol < 25-2500 > 1-5

After loading the recombinant protein into the second chromatographic system 609, the recombinant protein is eluted from the column(s). As shown in FIG. 6, the eluted recombinant protein can be subjected to a virus killing step 611. Like the virus kill step 607, the virus kill step 611 may include one or more of low pH kill, detergent kill, or other technique known in the art. In some embodiments, only one of the virus kill steps 607 or 611 is used, or the virus kill procedures are performed at the same step in the purification process.

As shown in FIG. 6, the recombinant protein product after the virus killing step 611 can be introduced into the third chromatographic system 613 for further purification of the recombinant protein product. Alternatively, the eluted recombinant protein from the second chromatographic system 609 can be fed directly to the third chromatographic system.

- 24 039750

613. In various embodiments, the third chromatography system 613 comprises one or more cation exchange chromatography (CEX) and/or size exclusion chromatography (SEC) columns to further remove impurities. The recombinant protein product is then eluted from the third 613 chromatography system.

The volume of the third chromatographic column (eg CEX or SEC column) may be any suitable volume such as 0.1 to 200 L. Exemplary column volumes include about 0.1 L, about 0.2 L, about 0.3 L, about 0.4 L, about 0.5 L, about 0.6 L, about 0.7 L, about 0.8 l, approximately 0.9 l, approximately 1 l, approximately 1.5 l, approximately 2 l, approximately 2.5 l, approximately 3 l, approximately 3.5 l, approximately 4 l, approximately 4.5 l, approximately 5 l, approximately 6 l, approximately 7 l, approximately 8 l, approximately 9 l, approximately 10 l, approximately 15 l, approximately 20 l, approximately 25 l, approximately 30 l, approximately 35 l, approximately 40 l and approximately 50 l, approximately 60 liters, approximately 70 liters, approximately 80 liters, approximately 90 liters, approximately 100 liters, approximately 150 and approximately 200 liters.

Illustrative column conditions for CEX are given in table. 4 below.

Table 4

Speed Volume Procedure Buffer flow (cm/h) (CV) Sanitary 0.1-10 M NaOH < 25-2500 > 1-3 processing before (> 10-120 min) using Balancing 2-200 mM sodium citrate, pH 4.0-5.0 <30-3000 >2-10 Loading No data < 30-3000 No data Washing 2-200 mM sodium citrate, pH 4.0-5.0 <30-3000 >2-10 Elution 2-200 mM sodium citrate, 151500 mM NaCl, pH 4.0-5.0 <30-3000 >2-10 Desorption 2-200 mM sodium citrate, 0.110 M NaCl, pH 4.0-5.0 <30-3000 > 1-5 Sanitary > 1-3 processing after 0.1-10 M NaOH < 25-2500 use (> 10-120 min) Storage 0.01-1.0 M NaOH < 30-3000 > 1-5

The recombinant protein product can also be further processed. For example, another filtering system 615 can be used to remove viruses. In some embodiments, such filtration can use filters with pore sizes from 5 to 50 microns. Other product processing may include a product adjustment step 617, in which the recombinant protein product can be sterilized, filtered, concentrated, stored, and/or additional components added to it to obtain the final product formulation. For example, the recombinant protein product can be concentrated 2-10 times, such as from an initial protein concentration of about 2 to about 20 mg/ml to a final protein concentration of about 4 to about 200 mg/ml. This end product can be used to fill vials and can be lyophilized for future use.

Introduction of recombinant protein

The recombinant human lysosomal protein (eg, rhGAA) or a pharmaceutically acceptable salt thereof can be formulated according to conventional procedures into a pharmaceutical composition adapted for administration to humans. For example, in a preferred embodiment, the intravenous composition is a solution in sterile isotonic aqueous buffer. If necessary, the composition may also contain a solubilizing agent and a local anesthetic to reduce pain at the injection site. Typically, the ingredients are supplied individually or mixed together in a unit dosage form, for example, as a dry lyophilized powder or anhydrous concentrate in a hermetically sealed container such as an ampoule or sachet, indicating the amount of active agent. If the composition is to be administered by infusion, it may be dispensed into an infusion bottle containing sterile pharmaceutical grade water, saline, or a dextrose/water mixture. When the composition is administered by injection, an ampoule of sterile water for injection or saline may be provided so that the ingredients can be mixed prior to administration.

The recombinant human lysosomal protein (eg, rhGAA) (or a composition or drug product containing the recombinant human lysosomal protein) is administered by a suitable route. In one embodiment, the recombinant human lysosomal protein is administered intravenously. In other embodiments, a recombinant human lysosomal protein (e.g., rhGAA) is administered by direct injection into a target tissue such as cardiac or skeletal muscle (e.g., intramuscularly) or into the nervous system (e.g., direct injection into the brain; intraventricular; intrathecal ). More than one path can be used at the same time if necessary.

The recombinant human lysosomal protein (e.g., rhGAA) (or a composition or drug product comprising the recombinant human lysosomal protein) is administered in a therapeutically effective amount (e.g., at a dose level that, when administered at regular intervals, is sufficient to treat a disease, such as by reducing the intensity of symptoms associated with the disease, preventing or delaying the onset of the disease, and/or reducing the severity or frequency of the symptoms of the disease). The amount that will be therapeutically effective in the treatment of a disease will depend on the nature and extent of the effects of the disease and can be determined using standard clinical techniques. Additionally, in vitro or in vivo assays may optionally be used to assist in selecting optimal dose ranges. The exact dose to be used will also depend on the route of administration and the severity of the disease and should be determined according to the judgment of the practitioner and the condition of each patient. Effective doses can be extrapolated from dose-response curves obtained using in vitro test systems or animal models. In at least one embodiment, recombinant human α-glucosidase is administered by intravenous infusion at a dose of from about 1 to about 100 mg/kg, such as from about 5 to about 30 mg/kg, typically from about 5 to about 20 mg/kg. In at least one embodiment, recombinant human α-glucosidase is administered by intravenous infusion at a dose of about 5 mg/kg, about 10 mg/kg, about 15 mg/kg, about 20 mg/kg, about 25 mg/kg , approximately 30 mg/kg, approximately 35 mg/kg, approximately 40 mg/kg, approximately 50 mg/kg, approximately 50 mg/kg, approximately 60 mg/kg, approximately 70 mg/kg, approximately 80 mg/kg, approximately 90 or approximately 100 mg/kg. In at least one embodiment, recombinant human α-glucosidase is administered by intravenous infusion at a dose of about 20 mg/kg. The effective dose for a particular individual may vary (eg, increase or decrease) over time, depending on the needs of the individual. For example, the amount can be increased during times of medical illness or stress, or when anti-acid α-glucosidase antibodies appear or increase, or when the symptoms of the disease worsen.

A therapeutically effective amount of human recombinant acid α-glucosidase (or a composition or drug product containing recombinant human recombinant acid α-glucosidase) is administered at regular intervals depending on the nature and severity of the effects of the disease and on an ongoing basis. Administration at regular intervals, as used herein, indicates that a therapeutically effective amount is administered intermittently (as opposed to a single dose). The interval can be determined using standard clinical techniques. In preferred embodiments, human recombinant acid α-glucosidase is administered monthly, twice a month; once a week; twice a week or daily. The administration interval for an individual's need need not be a fixed interval and may

- 26 039750 vary over time depending on the needs of the individual. For example, the interval between doses can be reduced during times of medical illness or stress, or when antibodies to recombinant human recombinant acid α-glucosidase develop or increase, or when symptoms of the disease worsen. In some embodiments, a therapeutically effective amount of 5, 10, 20, 50, 100, or 200 mg of enzyme/kg of body weight is administered twice a week, once a week, or every two weeks with or without a chaperone.

The recombinant human lysosomal protein (eg, rhGAA) can be prepared for later use, such as in a single dose vial or syringe, or in an intravenous bottle or sachet. Kits containing a recombinant human lysosomal protein (e.g., rhGAA) and optional excipients or other active ingredients such as chaperones or other drugs may be enclosed in packaging material and provided with dilution, dilution, or dosing instructions for treatment of the subject, in need of treatment, such as a patient with Pompe disease.

Combination therapy with rhGAA and a pharmacological chaperone

In various embodiments, rhGAA (eg, ATB200) produced by the methods described herein can be used in combination therapy with a pharmacological chaperone such as miglustat or duvoglustat.

In at least one embodiment, the pharmacological chaperone (eg, miglustat) is administered orally. In at least one embodiment, miglustat is administered at an oral dose of about 200 to about 400 mg, or an oral dose of about 200 mg, about 250 mg, about 300 mg, about 350 mg, or about 400 mg. In at least one embodiment, miglustat is administered at an oral dose of about 233 to about 400 mg. In at least one embodiment, miglustat is administered at an oral dose of about 250 to about 270 mg, or an oral dose of about 250 mg, about 255 mg, about 260 mg, about 265 mg, or about 270 mg. In at least one embodiment, miglustat is administered as an oral dose of approximately 260 mg.

Those skilled in the art will appreciate that an oral dose of miglustat in the range of about 200 to 400 mg, or any lesser range therein, may be appropriate for an adult patient with an average body weight of about 70 kg. For patients having a significantly lower body weight than about 70 kg, including, but not limited to, underweight infants, children, or adults, a lower dosage may be considered appropriate by the attending physician. Therefore, in at least one embodiment, miglustat is administered as an oral dose of about 50 mg to about 200 mg, or as an oral dose of about 50 mg, about 75 mg, about 100 mg, 125 mg, about 150 mg, about 175 or about 200 mg. In at least one embodiment, miglustat is administered as an oral dose of about 65 to about 195 mg, or as an oral dose of about 65 mg, about 130, or about 195 mg.

In at least one embodiment, miglustat is administered in a pharmaceutically acceptable dosage form suitable for oral administration, which includes, without limitation, tablets, capsules, vaginal suppositories, tinctures, solutions or suspensions, gels, syrups, mouthwashes, or dry powder for oral administration. dilution with water or other suitable inert medium prior to use, optionally with flavoring and coloring agents for immediate, delayed, modified, sustained, intermittent or controlled release routes of use. Solid compositions such as tablets, capsules, lozenges, lozenges, pills, boluses, powders, pastes, granules, suppositories, dragees or premix preparations may also be used. In at least one embodiment, miglustat is administered as a tablet. In at least one embodiment, miglustat is administered as a capsule. In at least one embodiment, the dosage form contains from about 50 to about 300 mg of miglustat. In at least one embodiment, the dosage form contains approximately 65 mg of miglustat. In at least one embodiment, the dosage form contains approximately 130 mg of miglustat. In at least one embodiment, the dosage form contains approximately 260 mg of miglustat. It is contemplated that if the dosage form contains approximately 65 mg of miglustat, then miglustat can be administered as a dosage of four dosage forms or in a total dose of 260 mg of miglustat. However, for patients who are significantly lower than the average adult weight of 70 kg, including but not limited to infants, children, or underweight adults, miglustat can be administered as a single dose dosage form (for a total dose of 65 mg miglustat), two dosage forms (for a total dose of 130 mg miglustat) or three dosage forms (for a total dose of 195 mg miglustat).

- 27 039750

Solid and liquid compositions for oral use can be obtained in accordance with methods well known in the art. Such compositions may also contain one or more pharmaceutically acceptable carriers and excipients, which may be in solid or liquid form. Tablets and capsules can be prepared by conventional methods using pharmaceutically acceptable excipients, including but not limited to binders, fillers, lubricants, disintegrants, or wetting agents. Suitable pharmaceutically acceptable excipients are known in the art and include, without limitation, pregelatinized starch, polyvinylpyrrolidone, povidone, hydroxypropyl methylcellulose (HPMC), hydroxypropyl ethyl cellulose (HPEC), hydroxypropyl cellulose (HPC), sucrose, gelatin, gum arabic, lactose, microcrystalline cellulose, calcium hydrogen phosphate, stearate magnesium, stearic acid, glyceryl behenate, talc, silicon dioxide, corn, potato or cassava starch, sodium starch glycolate, sodium lauryl sulfate, sodium citrate, calcium carbonate, dibasic calcium phosphate, glycine, sodium croscarmellose and complex silicates. Tablets can be coated by methods well known in the art. In at least one embodiment, miglustat is administered as a formulation commercially available as Zavesca® (Actelion Pharmaceuticals).

In at least one embodiment, miglustat and recombinant human α-glucosidase are administered simultaneously. In at least one embodiment, miglustat and recombinant human acid α-glucosidase are administered sequentially. In at least one embodiment, the miglustat is administered prior to the administration of the recombinant human acid α-glucosidase. In at least one embodiment, the miglustat is administered less than three hours prior to administration of the human recombinant acid α-glucosidase. In at least one embodiment, the miglustat is administered approximately two hours prior to administration of the recombinant human α-acid α-glucosidase. In at least one embodiment, the miglustat is administered less than two hours prior to administration of the human recombinant acid α-glucosidase. In at least one embodiment, miglustat is administered approximately 1.5 hours prior to administration of the recombinant human α-acid α-glucosidase. In at least one embodiment, the miglustat is administered approximately one hour prior to administration of the human recombinant acid α-glucosidase. In at least one embodiment, the miglustat is administered about 50 to about 70 minutes prior to the administration of the human recombinant acid α-glucosidase. In at least one embodiment, the miglustat is administered about 55 to about 65 minutes prior to administration of the human recombinant acid α-glucosidase. In at least one embodiment, the miglustat is administered about 30 minutes prior to administration of the recombinant human α-acid α-glucosidase. In at least one embodiment, the miglustat is administered about 25 to about 35 minutes prior to administration of the human recombinant acid α-glucosidase. In at least one embodiment, miglustat is administered about 27 to about 33 minutes prior to administration of the human recombinant acid α-glucosidase.

In at least one embodiment, the miglustat is administered concurrently with the administration of recombinant human acid α-glucosidase. In at least one embodiment, miglustat is administered over a period of 20 minutes before or after administration of the human recombinant acid α-glucosidase. In at least one embodiment, miglustat is administered over a period of 15 minutes before or after administration of the human recombinant acid α-glucosidase. In at least one embodiment, miglustat is administered over a period of 10 minutes before or after administration of the human recombinant acid α-glucosidase. In at least one embodiment, the miglustat is administered over a period of 5 minutes before or after administration of the human recombinant acid α-glucosidase.

In at least one embodiment, miglustat is administered following administration of recombinant human acid α-glucosidase. In at least one embodiment, the miglustat is administered within a period of up to 2 hours after the administration of the recombinant human α-acid α-glucosidase. In at least one embodiment, the miglustat is administered about 30 minutes after administration of the recombinant human α-acid α-glucosidase. In at least one embodiment, the miglustat is administered approximately one hour after administration of the recombinant human α-glucosidase acid. In at least one embodiment, the miglustat is administered about 1.5 hours after administration of the human recombinant acid α-glucosidase. In at least one embodiment, the miglustat is administered about 2 hours after administration of the human recombinant acid α-glucosidase.

In another aspect of the present invention, a kit is provided for the combination therapy of Pompe disease in a patient in need thereof. The kit contains a pharmaceutically acceptable dosage form containing miglustat, a pharmaceutically acceptable dosage form containing human recombinant acid α-glucosidase as defined herein, and instruction

- 28 039750 for administering a pharmaceutically acceptable dosage form containing miglustat and a pharmaceutically acceptable dosage form containing recombinant acid α-glucosidase to a patient in need thereof. In at least one embodiment, the pharmaceutically acceptable dosage form containing miglustat is the oral dosage form described herein, including, without limitation, a tablet or capsule. In at least one embodiment, a pharmaceutically acceptable dosage form containing recombinant human acid α-glucosidase is a sterile injectable solution as described herein. In at least one embodiment, instructions for administering dosage forms include instructions for administering a pharmaceutically acceptable dosage form containing miglustat orally prior to administration of a pharmaceutically acceptable dosage form containing recombinant human α-glucosidase by intravenous infusion as described herein.

Without wishing to be bound by theory, it is believed that miglustat acts as a pharmacological chaperone for recombinant human α-glucosidase ATB200 and binds to its active site. For example, miglustat has been found to reduce the percentage of the unfolded ATB200 protein and stabilize the active conformation of ATB200, preventing its denaturation and irreversible inactivation at neutral plasma pH, and allowing it to withstand conditions in the bloodstream long enough to reach and be taken up by tissues. However, the binding of miglustat to the active site of ATB200 can also lead to inhibition of the enzymatic activity of ATB200 by interfering with the access of the natural glycogen substrate to the active site. It is believed that in cases where miglustat and recombinant human α-glucosidase are administered to a patient under the conditions described herein, plasma and tissue concentrations of miglustat and ATB200 are such that ATB200 stabilizes until it can be absorbed. tissues and target lysosomes, but due to the rapid clearance of miglustat, glycogen hydrolysis by ATB200 in lysosomes is not overly inhibited by the presence of miglustat, and the enzyme retains sufficient activity to be therapeutically useful.

All of the embodiments described above can be combined. This covers specific embodiments relating to the nature of the pharmacological chaperone, eg miglustat; and the active site for which it is specific;

dosage, route of administration of the pharmacological chaperone (eg, miglustat) and type of pharmaceutical composition, including the nature of the carrier and the use of commercially available compositions;

the nature of the drug, e.g., a therapeutic protein drug, which may be an analog of an endogenous protein whose expression is reduced or absent in the subject, suitably a recombinant human lysosomal protein (e.g., rhGAA), e.g., recombinant human acid α-glucosidase expressed in Chinese Hamster Ovary (CHO) cells and having an increased content of N-glycan units bearing one or more mannose-6-phosphate residues compared to the content of N-glycan units bearing one or more mannose-6-phosphate residues in alglucosidase -alpha; and suitably having the amino acid sequence shown under SEQ ID NO: 1 or SEQ ID NO: 2;

the number and type of N-glycan units in the recombinant human lysosomal protein (eg, rhGAA), such as N-acetylglucosamine, galactose, sialic acid, or N-glycan complex units formed by combining them attached to the recombinant human lysosomal protein;

the degree of phosphorylation of mannose units in a recombinant human lysosomal protein (eg, rhGAA) to form mannose-6-phosphate and/or bis-mannose-6-phosphate;

the dosage and route of administration (eg, intravenous administration, especially intravenous infusion, or direct administration to the target tissue) of the replacement enzyme (eg, human recombinant acid α-glucosidase) and formulation type, including carriers and the therapeutically effective amount;

dosing interval of pharmacological chaperone (miglustat) and recombinant human acid α-glucosidase;

the nature of the therapeutic response and the results of the combination therapy (eg, improved results compared to the effect of each therapy administered alone);

combination therapy, such as simultaneous administration of miglustat and recombinant human α-glucosidase, or sequential administration, for example, where miglustat is administered before recombinant human α-glucosidase or after recombinant human α-glucosidase, or at a specific time before or after administration of the recombinant human acid α-glucosidase; and

- 29 039750 the nature of the patient being treated (for example, a mammal, such as a human) and the condition from which the individual suffers (for example, enzyme deficiency).

Any of the embodiments listed above may be combined with one or more of the other embodiments in the list.

Examples

Other features of the present invention will become apparent from the following non-limiting examples, which illustrate the principles of the present invention by way of example.

Example 1 Limitations of Current rhGAA Based Myozyme® and Lumizyme® Products

To assess the potential of rhGAA in Myozyme® and Lumizyme®, the only currently approved treatments for Pompe disease, these rhGAA-based drugs were injected into a CIMPR column (which binds to rhGAA having M6P groups) and then eluted in a free M6 gradient. . Fractions were collected in a 96-well plate, and GAA activity was analyzed using a substrate representing 4MU-α-glucose. The relative amounts of bound and unbound rhGAA were determined based on GAA activity and reported as a percentage of total enzyme.

In FIG. 4A-B describe the problems associated with traditional ERT agents (Myozyme® and Lumizyme®): 73% rhGAA in Myozyme® (Fig. 4B) and 78% rhGAA in Lumizyme® (Fig. 4A) did not bind to CIMPR, see the leftmost peaks on each figure. Only 27% of rhGAA in Myozyme® and 22% of rhGAA in Lumizyme® contained M6P, which can efficiently target CIMPR in muscle cells.

An effective dose of Myozyme® and Lumizyme® corresponds to the amount of rhGAA containing M6P that targets CIMPR in muscle cells. However, most of the rhGAA in these two conventional products does not target the CIMPR receptor in the target muscle cells.

The introduction of traditional rhGAA, where most of the rhGAA does not target muscle cells, increases the risk of an allergic reaction or induction of an immune response to the untargeted rhGAA.

Example 2 Production of CHO Cells Producing rhGAA ATB200 High in N-Glycans Carrying Mono- or 6c-M6P

CHO cells were transfected with DNA that expresses rhGAA, followed by selection of transformants that produce rhGAA. A DNA construct for transforming CHO cells with DNA encoding rhGAA is shown in FIG. 5. CHO cells were transfected with DNA that expresses rhGAA, followed by selection of transformants that produce rhGAA.

After transfection, CHO-DG44 (DHFR-) cells containing a stably integrated GAA gene were selected on hypoxanthine/thymidine (-HT)-free medium. Upregulation of GAA expression in these cells was induced by treatment with methotrexate (MTX, 500 nM). Cell populations that expressed GAA at high levels were identified by GAA enzyme activity assays and used to create single clones that produce rhGAA. Individual clones were obtained in culture dishes with semi-liquid medium, collected using the ClonePix system and transferred to plates with 24 deep wells. Individual clones were analyzed for GAA enzyme activity to identify clones expressing GAA at a high level. In conditioned media, the substrate 4-MU-α-glucosidase was used to determine GAA activity. Clones producing GAA at higher levels, as measured by GAA enzyme assays, were further evaluated for viability, growth ability, GAA production productivity, N-glycan structure, and stable protein expression. CHO cell lines, including the CHO cell line GA-ATB-200 expressing rhGAA with increased levels of mono-M6P or bis-M6P-N-glycans, were isolated using this procedure.

Example 3 Capture and Purification of rhGAA ATB200

Several batches of rhGAA according to the present invention were prepared in shake flasks and perfusion bioreactors using the CHO GA-ATB-200 cell line and binding to CIMPR was measured. Purified rhGAA ATB200 from various production lots showed binding to the CIMPR receptor similar to that shown in FIG. 7B and 8 (~70%) indicating that rhGAA ATB200 can be stably produced. As shown in FIG. 4A, 4B, 7A and 7B, rhGAA Myozyme® and Lumizyme® showed significantly less binding to CIMPR than rhGAA ATB200.

Example 4 Analytical Comparison of ATB200 with Lumizyme®

Weak anion exchange liquid chromatography (WAX) was used to fractionate rhGAA ATB200 at terminal phosphate. Elution profiles were obtained by eluting the ERT agent with increasing amounts of salt. The profiles were controlled using UV radiation (A280 nm). rhGAA ATB200 was obtained from CHO cells and purified. Lumizyme® was obtained from a commercial source. Lumizyme® showed a high peak on the left side of his profile

- 30 039750 elution. rhGAA ATB200 showed four distinct elution peaks to the right of Lumizyme® (FIG. 9). This confirms that rhGAA ATB200 was phosphorylated to a greater extent than Lumizyme®, since this assessment was done by terminal charge and not by CIMPR affinity.

Example 5 Characterization of oligosaccharides in rhGAA ATB200

Purified rhGAA ATB200 and Lumizyme® glycans were evaluated by MALDI-TOF to determine the structures of the individual glycans found in each ERT agent (FIG. 10). ATB200 samples were found to contain lower amounts of non-phosphorylated high-mannose type N-glycans than Lumizyme®. The higher content of M6P-glycans in ATB200 than in Lumizyme® allows rhGAA ATB200 to target muscle cells more efficiently. The high percentage of monophosphorylated and bisphosphorylated structures as determined by MALDI is consistent with the CIMPR binding profiles, which illustrate a significantly greater degree of ATB200 binding to the CIMPR receptor. Analysis of N-glycans by MALDI-TOF mass spectrometry confirmed that, on average, each ATB200 molecule contains at least one natural bis-M6P-N-glycan structure. This higher content of 6uc-M6P-Nglycans in rhGAA ATB200 directly correlates with high affinity CIMPR binding in plate M6P receptor binding assays (KD was approximately 2-4 nM), FIG. 12A.

rhGAA ATB200 was also analyzed for site-specific N-glycan profiles using two different LC-MS/MS based analytical techniques. In the first assay, the protein was denatured, reduced, alkylated, and cleaved prior to LC-MS/MS analysis. During protein denaturation and reduction 200 µg protein sample, 5 µl 1 mol/l TrisHCl (final concentration 50 mM), 75 µl 8 mol/l guanidine-HCl (final concentration 6 M), 1 µl 0.5 mol/l EDTA (final concentration 5 mM), 2 μl of 1 mol/l DTT (final concentration 20 mM) and Milli-Q® water were added to a 1.5 ml tube to give a total volume of 100 μl. The sample was mixed and incubated at 56°C for 30 min in a dry heat bath. During alkylation, the denatured and reduced protein sample was mixed with 5 µl of 1 mol/l iodoacetamide (IAM, 50 mM final concentration), then incubated at 10-30°C in the dark for 30 min. After alkylation, 400 µl of pre-chilled acetone was added to the sample, and the mixture was refrigeration-frozen at -80°C for 4 hours. The sample was then centrifuged for 5 minutes at 13,000 rpm and 4°C, and the supernatant was removed. 400 µl of pre-chilled acetone was added to the pellet, which was then centrifuged for 5 min at 13,000 rpm and 4° C., and the supernatant was removed. The sample was then air dried on ice in the dark to remove residual acetone. To dissolve the protein, 40 µl of 8 M urea and 160 µl of 100 mM NH4HCO3 were added to the sample. During trypsin digestion, trypsin digestion buffer was then added to 50 μg of protein to a final volume of 100 μl and 5 μl of 0.5 mg/ml trypsin was added (protein to enzyme ratio 20/1 w/w). The solution was thoroughly mixed and incubated overnight (16±2 hours) at 37°C. 2.5 μl of 20% TFA was added to quench the reaction (0.5% final concentration). The sample was then analyzed using an Orbitrap Velos Pro™ mass spectrometer from Thermo Scientific.

In a second LC-MS/MS analysis, an ATB200 sample was prepared following a similar denaturation, reduction, alkylation, and cleavage procedure, except that iodoacetic acid (IAA) was used as the alkylating reagent instead of IAM, and then analyzed by mass Thermo Scientific Orbitrap Fusion Lumos Tribid™ spectrometer.

The results of the first and second analyzes are shown in figures 11A-11H. In FIG. 11A-11H, the results of the first analysis are represented by the bars on the left (dark gray), and the results of the second analysis are represented by the bars on the right (light gray). In FIG. 11B11G, the symbol nomenclature for representing glycans is according to Varki, A., Cummings, R.D., Esko J.D., et al., Essentials of Glycobiology, 2nd edition (2009).

In FIG. 8A-8I, it can be seen that the two assays gave similar results, although there was some difference between the results. This difference may be due to a number of factors, including the instrument used and the completeness of the N-glycan analysis. For example, if some forms of phosphorylated glycans have not been identified and/or quantified, then the total number of phosphorylated glycans may be underrepresented, and the percentage of rhGAA bearing phosphorylated glycans at a given site may be underrepresented. As another example, if some forms of non-phosphorylated glycans have not been identified and/or quantified, then the total number of non-phosphorylated glycans may be underrepresented, and the percentage of rhGAA carrying phosphorylated glycans at a given site may be overrepresented. quantity. In FIG. 11A shows N-glycosylation site occupancy in ATB200. As can be seen in FIG. 11A, the first, second, third, fourth, fifth, and sixth N-glycosylation sites were predominantly occupied, with both analyzes revealing that at each potential site, more than 90% to about 100% of the ATB200 enzyme molecules had a glycan. However, the seventh potential N-glycosylation site was glycosylated in approximately half of the cases.

- 31 039750

In FIG. 11B shows the N-glycosylation profile for the first site, N84. In FIG. 11B it can be seen that the main form of the glycans were bis-M6P-glycans. Both the first and second analyzes revealed that more than 75% of ATB200 had a bis-M6P glycan at the first site.

In FIG. 11C shows the N-glycosylation profile for the second site, N177. As can be seen in FIG. 11C, the main forms of glycans were mono-M6P-glycans and non-phosphorylated high-mannose glycans. Both the first and second analyzes revealed that more than 40% of ATB200 had a monoM6P glycan at the second site.

In FIG. 11D shows the N-glycosylation profile for the third site, N334. In FIG. 11D it can be seen that the main forms of glycans were non-phosphorylated high mannose glycans, two-, three- and four-antennary complex glycans, and hybrid glycans. Both the first and second analyzes revealed that more than 20% of ATB200 had a sialic acid residue at the third site.

In FIG. 11E shows the N-glycosylation profile for the fourth site, N414. In FIG. 11E it can be seen that the main forms of glycans were 6c-M6P and mono-M6P-glycans. Both the first and second analyzes revealed that more than 40% of ATB200 had a bis-M6P-glycan at the fourth site. Both the first and second analyzes also revealed that more than 25% of ATB200 had a mono-M6P-glycan at the fourth site.

In FIG. 11F shows the N-glycosylation profile for the fifth site, N596. In FIG. 11F, it can be seen that the main forms of glycans were fucosylated two-antennary complex glycans. Both the first and second analyzes revealed that more than 70% of ATB200 had a sialic acid residue at the fifth site.

In FIG. 11G shows the N-glycosylation profile for the sixth site, N826. As can be seen in FIG. 11G, the main forms of glycans were two-, three-, and four-antennary complex glycans. Both the first and second analyzes revealed that more than 80% of ATB200 had a sialic acid residue at the sixth site.

Analysis of glycosylation at the seventh site, N869, showed approximately 40% glycosylation, with the most abundant glycans being A4S3S3GF (12%), A5S3G2F (10%), A4S2G2F (8%), and A6S3G3F (8%).

In FIG. 11H shows a summary of phosphorylation information for each of the seven potential N-glycosylation sites. In FIG. 11H, it can be seen that both the first and second assays showed high levels of phosphorylation at the first, second and fourth sites. In both assays, more than 80% of ATB200 was mono- or diphosphorylated at the first site, more than 40% of ATB200 was monophosphorylated at the second site, and more than 80% of ATB200 was mono- or diphosphorylated at the fourth site.

Another analysis of ATB200 glycans was performed according to the hydrophilic interaction liquid chromatography/fluorescence detection mass spectrometry (HILIC FLD MS) method.

The results of the analysis by the method HILIC FLD MS are presented below in table. 5. In the table. 5, the first digit in the three-digit number indicates the number of branches in the glycan, the second digit indicates the number of core fucose units, and the third digit indicates the number of terminal sialic acid units. Using this nomenclature, 303 is a three-antennary glycan (first 3) with 0 core fucose units (2nd 0) and 3 terminal sialic acid units (last 3), 212 is a two-antennary glycan with 1 core fucose unit and 2 terminal sialic acid units, 404 is a four-antennary glycan with 0 core fucose units and 4 sialic acid terminal units, etc.

- 32 039750

Table 5

Peak number in FLD Peak number in MS RT (min.) Glycan structure % peak area one one BisPMan 8 2.83% 2 2 13.41 BisP Man 7 17.58% 3 14.30 BisP Man 6 1.02% 3 4 20.89 MonoPMan 6 2.34% 4 5 21.65 MonoPMan 5 1.16% 5 6 23.51 MonoPMan 8 1.28% 6 7 24.33 MonoPMan 7 4.35% 7 eight 25.61 MonoP Man 7_(+)GlcNAc 0.50% eight nine 28.76 MonoP_hMan6_ 101 0.48% nine ten 30.54 MonoP Man 6_(+)GlcNAc 0.68% ten eleven 33.50 Man 6 3.97% 12 33.65 303 0.74% eleven thirteen 34.97 Man 7 0.20% 12 fourteen 35.64 403 0.39% thirteen fifteen 36.61 302 0.36% fourteen sixteen 38.07 302 0.61% fifteen 17 38.53 man 5 1.85% sixteen eighteen 39.57 302 0.48% nineteen 39.78 hMan5_101 0.42% 20 40.05 hMan 5_100_(-)Gal 0.30%

-33 039750

17 21 40.77 301_(-)Gal 0.52% 22 40.58 301 0.50% eighteen 23 41.47 300_(-)Gal 0.80% nineteen 24 42.17 301_(-)Gal 0.11% 25 42.13 301 0.58% 20 26 42.89 301_(-)Gal 0.07% 27 42.79 301 0.80% 21 28 43.41 300 0.85% 29 43.28 101 0.39% 22 thirty 43.94 202 0.63% 23 31 44.45 401 0.39% 24 32 45.04 MonoP_hMan6_l 11 0.36% 25 33 45.69 MonoP_hMan6_l 11 1.45% 34 45.90 100 0.23% 35 45.90 400 0.19% 26 36 46.87 201 0.49% 37 47.15 202 0.34% 27 38 48.19 414 0.37% 28 39 48.94 202 1.97%

- 34 039750

29 40 50.79 MonoPMan 6110_(-)Gal 1.31% 41 51.37 414 0.62% thirty 42 52.22 313 0.74% 43 52.42 201_(-)Gal 0.46% 44 52.42 201 1.18% 45 53.11 hMan6_l 11 0.20% 31 46 53.83 200_(-)Gal 0.80% 47 54.23 201 1.27% 48 54.75 413 0.30% 32 49 55.47 200 1.30% 33 fifty 57,45,58,34 414_(+)GlcNAcGal 0.14% 51 56,62,56,91,57,99 413 0.94% 52 56,11,57,26,57,99 312 0.98% 34 53 60.19 413 0.33% 54 59.39 413_(+)GlcNAcGal 0.42% 55 59.80 312 0.52% 56 59.49 412 0.18% 35 57 60.75 413 0.78% 58 60.89 413_(+)GlcNAcGal 0.07%

-35 039750

- 36 039750

78 71.23 211_(-)Gal 0.19% 44 79 72.46 212 3.66% 45 80 74.82 221_(-)Gal(+)GalNAc 0.38% 81 74,43,74,96 411_(+)GlcNAcGal 0.66% 46 82 75.92 410 0.42% 47 83 76,73,77,87 211_(-)Gal 1.24% 84 77.23 211 3.64% 48 85 79.05 211 1.52% 86 79.38 210_(-)2Gal 0.45% 49 87 80.11 210_(-)Gal 1.58% fifty 88 81.15 210 2.41% 51 89 84.22-87.15 311 1.26% 52 90 95.35 Mono_Acetyl_NANA_212 0.99% 53 91 96.23 Mono_Acetyl_NANA_211 0.76% 54 92 97.37 Bis_Acetyl_NANA_212 0.42%

Based on this HILIC FLD MS analysis, the ATB200 tested is expected to have an average fucose content of 2-3 moles per mole of ATB200, a GlcNAc content of 20-25 moles per mole of ATB200, a galactose content of 8-12 moles per mole of ATB200, mannose content of 22-27 moles per mole of ATB200, M6P content of 3-5 moles per mole of ATB200, and sialic acid content of 4-7 moles per mole of ATB200.

Example 6 Affinity Characterization of ATB200 for CIMPR

In addition to having a higher percentage of rhGAA that can bind to CIMPR, it is important to understand the quality of this interaction. Binding of Lumizyme® and rhGAA ATB200 to receptors was determined using a CIMPR binding assay on plates. Briefly, CIMPR coated plates were used to capture GAA. Various concentrations of rhGAA were applied to the immobilized receptor, and unbound rhGAA was washed out. The amount of rhGAA remaining was determined by GAA activity. In FIG. 12A shows that rhGAA ATB200 bound to CIMPR significantly better than Lumizyme®.

In FIG. 12B shows the relative content of bis-M6P-glycans in Lumizyme®, traditional rhGAA and ATB200 in accordance with the present invention. In the case of Lumizyme®, on average only 10% of the molecules have a bisphosphorylated glycan. In contrast, in the case of ATB200, on average, each rhGAA molecule has at least one bisphosphorylated glycan.

Example 7 rhGAA ATB200 was more efficiently internalized by fibroblasts than Lumizyme

The relative uptake of rhGAA ATB200 and Lumizyme® cells was compared using a normal fibroblast cell line and a Pompe disease fibroblast cell line. Comparisons included 5-100 nM rhGAA ATB200 according to the present invention and 10-500 nM conventional rhGAA Lumizyme®. After a 16 hour incubation, the extrinsic rhGAA was inactivated with TRIS base and the cells were washed 3 times with PBS before being harvested.

Internalized GAA was measured by 4MU-α-glucoside hydrolysis and the data plotted against total cellular protein level and the results are shown in FIG. 13A-B.

It was also shown that rhGAA ATB200 was efficiently internalized by cells (FIGS. 13A and 13B, which respectively show that rhGAA ATB200 is internalized by both normal and Pompe fibroblast cells, and that it is internalized to a greater extent than traditional rhGAA Lumizyme®). rhGAA ATB200 saturates cellular receptors at approximately 20 nM, while Lumizyme® requires approximately 250 nM. The absorption efficiency constant (K _Π0ΓΛ0A e _HIA ) extrapolated from these results is 2-3 nM for ATB200 and 56 nM for Lumizyme® as shown in FIG. 13C. These results suggest that rhGAA ATB200 is a well-targeted treatment for Pompe disease.

Example 8 Decreased Glycogen in Gaa Knockout Mice

In FIG. 14A-14C show the effects of alglucosidase-alpha (Lumizyme®) and ATB200 administration on glycogen clearance in Gaa knockout mice. Animals received two IV bolus injections (once every two weeks); tissues were collected two weeks after the last dose and analyzed for acid α-glucosidase activity and glycogen content.

As seen in FIG. 14A-14C, ATB200 was found to deplete tissue glycogen stores in a dose-dependent manner in acid α-glucosidase (Gaa) gene knockout mice. A dose of 20 mg/kg of ATB200 consistently removed more stored glycogen in Gaa knockout mice than doses of 5 and 10 mg/kg. However, as seen in FIG. 14A-14C, ATB200 administered at a dose of 5 mg/kg showed a decrease in glycogen levels in mouse cardiac and skeletal muscles (quadriceps and triceps) similar to that of Lumizyme® administered at a dose of 20 mg/kg, while ATB200, administered at 10 and 20 mg/kg showed significantly better reduction in skeletal muscle glycogen levels than Lumizyme®.

In FIG. 15 shows the effects of alglucosidase-alpha (Lumizyme®) and ATB200 administration on glycogen clearance in Gaa knockout mice, and the effect of combined administration of ATB200 and miglustat on glycogen clearance. Twelve week old KO GAA mice were treated with 4 injections of 20 mg/kg Lumizyme® or ATB200 IV once every two weeks; miglustat was co-administered at a dose of 10 mg/kg orally 30 minutes prior to rhGAA administration as indicated. Tissues were collected 14 days after the last dose of enzyme to measure glycogen levels. In FIG. 15 shows the relative reduction in skeletal muscle glycogen levels in the quadriceps and triceps muscles, with ATB200 providing a greater glycogen reduction than Lumizyme® and the ATB200/miglustat combination providing an even greater glycogen reduction.

Example 9 Physiological and Morphological Characteristics of Muscles in Gaa Knockout Mice

Gaa knockout mice received two IV boluses of human recombinant acid α-glucosidase (alglucosidase-alpha or ATB200) at 20 mg/kg once every two weeks. Miglustat was administered orally at a dosage of 10 mg/kg to a subset of animals treated with ATB200 30 min prior to ATB200 administration. Control mice were treated with inert medium only. Soleus, quadriceps, and diaphragm tissue was collected two weeks after the last dose of human recombinant acid α-glucosidase. Soleus and diaphragm tissue were analyzed for glycogen levels by Schiff/Piodic Acid (PAS) staining and for lysosome proliferation by measuring levels of a lysosome-associated membrane marker protein (LAMP1), which is overexpressed in Pompe disease. Semi-thin sections of quadriceps embedded in epoxy resin (Epon) were stained with methylene blue and examined by electron microscopy (1000x) to determine the degree of presence of vacuoles. Quadriceps muscle samples were analyzed by immunohistochemistry to determine the levels of autophagy markers - light chain 3 associated with microtubule protein 1A/1B conjugated with phosphatidylethanolamine (LC3A II), and p62, insulin-dependent glucose transporter GLUT4 and insulin-independent glucose transporter GLUT 1.

In a similar study, Gaa knockout mice received four IV boluses of human recombinant acid α-glucosidase (alglucosidase-alpha or ATB200) at 20 mg/kg once every two weeks. Miglustat was administered orally at a dosage of 10 mg/kg to a subset of animals treated with ATB200 30 min prior to ATB200 administration. Control mice were treated with inert medium only. Cardiac muscle tissue was harvested two weeks after the last dose of human recombinant acid α-glucosidase and analyzed for glycogen levels by Schiff/Niodic acid (PAS) staining and for lysosome proliferation by measuring LAMP1 levels.

As seen in FIG. 16, a reduction in lysosome proliferation in cardiac muscle tissue, diaphragm tissue, and skeletal muscle tissue (soleus muscle tissue) was demonstrated when ATB200 was administered compared to conventional treatment with alglucosidase-alpha, and when miglustat was co-administered with ATB200, a significantly greater reduction in lysosome proliferation was demonstrated approaching levels seen in wild-type (WT) mice. Moreover, as seen in FIG. 17, with the introduction of ATB200, a decrease in glycogen levels in the form of pinpoint deposits in cardiac and skeletal muscle tissue (soleus tissue) was demonstrated compared with conventional treatment with alglucosidase-alpha, and when miglustat was co-administered with ATB200, a significantly greater decrease was shown, in this case also approaching to levels seen in wild-type (WT) mice.

- 38 039750

Also, as seen in FIG. 18, co-administration of miglustat with ATB200 significantly reduced the number of vacuoles in quadriceps muscle fiber in Gaa knockout mice compared to untreated and alglucosidase-alpha treated mice. As seen in FIG. 19, both LC3 II and p62 levels were increased in Gaa knockout mice compared to wild-type mice, however, they were significantly reduced by treatment with ATB200 and miglustat, indicating that the increase in autophagy associated with acid α deficiency -glucosidase, decreases with the combined administration of ATB200 and miglustat. In addition, the levels of the insulin-dependent glucose transporter GLUT4 and the non-insulin-dependent glucose transporter GLUT1 were increased in Gaa knockout mice compared to wild-type mice, but were also significantly reduced in this case when treated with ATB200 and miglustat. Elevated levels of GLUT4 and GLUT1, associated with acid α-glucosidase deficiency, may contribute to increased glucose uptake by muscle fibers and increased glycogen synthesis both in the basal state and after a meal. Thus, combined treatment with ATB200 and miglustat was found to improve skeletal muscle morphological and physiological characteristics in a mouse model of Pompe disease.

Example 10 Muscle Function in Gaa Knockout Mice

In longer-term studies with 12 biweekly doses, the combination of 20 mg/kg ATB200 with 10 mg/kg miglustat gradually increased functional muscle strength in KO Gaa mice from baseline as measured by both grip strength tests and tests hanging on a wire (Fig. 21A-21B). Mice treated with alglucosidase-alpha (Lumizyme®) that received a similar dose of ERT agent (20 mg/kg) experienced a reduction in performance under identical conditions for most of the study (FIGS. 21A-21B). As with the shorter term study, the ATB200/miglustat combination had significantly better glycogen clearance after 3 months (FIGS. 22A-22C) and 6 months (FIGS. 22D-22G) of treatment than alglucosidase-alpha. The ATB200/miglustat combination also reduced autophagy and intracellular accumulation of LAMP1 and dysferlin after 3 months of treatment (Figure 23) compared to alglucosidase-alpha. In FIG. 21A * indicates a statistically significant result compared to Lumizyme® alone (p < 0.05, 2-tailed t-test). In FIG. 22A-22G * indicates a statistically significant result compared to Lumizyme® alone (p < 0.05, Dunnett's multiple comparison, one-way ANOVA).

Taken together, these data indicate that the ATB200/miglustat combination was effective in targeting muscle, reversing cellular dysfunction and improving muscle function. Importantly, apparent improvements in muscle structure, as well as reduced autophagy and intracellular accumulation of LAMP1 and dysferlin, may be good surrogates for improved muscle physiology, which correlates with improvements in functional muscle strength. These results suggest that monitoring autophagy and the levels of these key muscle proteins, which may prove to be useful biomarkers in muscle biopsies in clinical trials, may be a viable practical way to assess the effectiveness of therapeutic treatments for Pompe disease in KO Gaa mice.

In FIG. 23 shows that administration of ATB200 with or without miglustat for 6 months reduced intracellular accumulation of dystrophin in KO Gaa mice. A more significant reduction in dystrophin accumulation was observed with the ATB200 ± miglustat combination than with Lumizyme®.

Example 11 Capturing rha-Gal A

The CIMPR binding profile for recombinant human α-galactosidase A (rha-Gal A) in spent cell culture medium was measured before product capture by AEX chromatography (FIG. 20A) and after product capture by AEX chromatography (FIG. 20B). The dotted line in both graphs refers to the elution gradient with M6P. Prior to product capture in AEX, 80% rha-Gal A can bind to CIMPR. After the capture of the product in AEX, the total binding of rha-Gal A increases to 96%.

Example 12 Pharmacokinetic and safety data for recombinant acid α-glucosidase ATB200 co-administered with miglustat in ERX treated and ERX naive patients with Pompe disease

This study was designed primarily to evaluate the safety, tolerability and pharmacokinetic parameters (PK) of ATB200 co-administered with miglustat. Using a PK/pharmacodynamic (PD) translational model in Gaa knockout mice, the combination of 20mg/kg ATB200 with a high dose (eg 260mg) of miglustat in humans was predicted to provide optimal glycogen reduction.

In the description below, a high dose of miglustat refers to a dose of approximately 260 mg and a low dose of miglustat refers to a dose of approximately 130 mg.

The aim was to evaluate preliminary PK data for total protein GAA, ATB200 and miglustat and safety markers from 10 patients in this Phase 1/2 study.

- 39 039750

It is an open-label Phase 1/2, fixed-sequence, escalating-dose study, the first in humans, to evaluate the safety, tolerability, PK, PD, and efficacy of intravenous infusions of ATB200 given in combination with oral miglustat in adults with Pompe disease (Fig. 24). Mean PK study results for total GAA protein and miglustat were assessed in the first 8 patients in cohort 1 to visit 9 and in the first 2 patients in cohort 3.

a Analyzed safety data from 2 indicator patients from cohort 1 at each pre-dose level in cohorts 2 and 3.

^b During steps 2 and 3, miglustat was administered orally prior to the initiation of intravenous infusions of ATB200. All doses of ATB200 were administered by intravenous infusion over 4 hours.

^e The first 2 patients in cohorts 2 and 3 served as indicator patients for their respective cohorts.

Key inclusion criteria:

Men and women aged 18-65 years who have been diagnosed with Pompe disease based on documented GAA enzyme deficiency or GAA genotyping results;

Received ERT using alglucosidase-alpha within 2-6 years (or >2 years for cohort 2) before the start of the trial (cohort 1);

Currently on biweekly alglucosidase-alpha who have had their last 2 infusions without drug-related adverse events leading to dosing interruption (cohorts 1 and 2);

Should be able to walk 200 to 500 m on a 6-minute walk test (cohorts 1 and 3);

Forced standing vital capacity should be 30-80% of the predicted normal value (cohorts 1 and 3);

Must be wheelchair bound and unable to walk unaided (cohort 2).

PK analysis

Blood samples were collected to analyze the concentration and activity of total GAA protein in blood plasma.

Stage 1: Before starting ATB200 infusion and 1, 2, 3, 3.5, 4, 4.5, 5, 6, 8, 10, 12, and 24 hours after starting the infusion.

Stages 2 and 3: 1, 2, 3, 4, 4.5, 5, 6, 7, 9, 11, 13 and 25 hours after oral administration of miglustat.

Blood samples for determination of miglustat plasma concentrations were taken immediately before oral administration of miglustat (time 0) and 1, 1.5, 2, 2.5, 3, 4, 5, 6, 9, 11 and 25 hours after oral administration. administration of miglustat. The amount of miglustat in blood plasma was determined by validated analysis according to the LC-MS/MS method.

Plasma total GAA protein concentrations for 5, 10, and 20 mg/kg ATB200 were determined by validated LC-MS/MS quantitation for the rhGAA specific signature(s) peptide(s).

A preliminary analysis was completed in 8 patients in cohort 1 who completed stages 1 and 2 and in 2 patients in cohort 3 who started stage 3.

Patients transitioning from initial ERT are representative of the Pompe disease population, with a median of 5.02 years per ERT (Table 6).

Table 6. Characteristics at baseline

Hlrzhi-riski no baseline (X IT) Amo saddlery. io.k chatyaie 'ERI (n=10) Not treated (n 2) The period of receipt of Sh1 (l.umizynw "Myozyme M. average hne.m fSTIJV) 5.02 N .21 X Λ In<нр;кт μ years. mean value (.ihhhuyun) 47.7(8.191 33 l 112.73) t lo.i. M7K. % 80.20 0 nude 6MW1. m<try. average hsh’knie ί^ΤΓΑΊ »8.4 (95.·»?) 432.1 (b7.8i GAS n shmozheshsh sieve. average value n %. nrsdsya “other <8’1 DY) 51.9 (13.84) 51.0(26.8?)

HMWl ieci with b-mshp shop walking: I’YC forsshptshnya zhishinnia emish κιknh. Ν Λ dgnshme oieyiewyroi: S ] DY

^a n=10 from cohort 1 (outpatients transferred from ERT) of the interim data analysis;

n=2 from cohort 3 (not treated).

- 40 039750

Total GAA protein

When administered alone, exposure to ATB200 increases slightly in a dose-dependent fashion (Table 7 and FIGS. 25A-25D). Variability appears to increase with dose of miglustat (Fig. 25C). Co-administration of 20mg/kg ATB200 with a high dose of miglustat (260mg) increased total GAA protein (AUC) exposure by approximately 25% compared to ATB200 alone at 20mg/kg. The half-life (α-phase) was increased by 45%, suggesting that a high dose of miglustat stabilizes plasma ATB200. An increase in the half-life is accompanied by an increase in AUC from the time the maximum plasma concentration is reached to about 12 hours after a dose. Increases in AUC and half-life can be observed on a logarithmic scale throughout the terminal phase of elimination (FIG. 25B). ATB200 showed a relatively large volume of distribution. Plasma distribution of total GAA protein appears similar between ERT-naive patients (cohort 3) and ERT-treated patients (cohort 1) (FIGS. 25A and 25D).

Table 7 Total GAA Protein

Cohort .................................................. ....... Treatment ties mon ml g Vn S'li' Al -'C, , CPU Of ml f A1. He's 'hmag fl "and : CL] (-Sh v„ 5 μι κι to 61Wh FOR ¢3.0G‘| 215 (W7> IHiUdi 1 V 116.7) LI i io.2i 2.1 (ID) 4.62 112.7) one ! (ι mg kg to igdsdypn in’’ Yd ¢16.6) 4-. 5 t <20% 581 <20. b 1.it doll L.l 1IO.5i 1.5 9 (25 4) 3.k? (16..5) one 20 μι ΚΙ μ Ol.tc 1NGOC1 It' ¹ '45 pcs 4L‘ < L5- 1m 1508 (14.51 142 < 14.4) .............2 1 <24.% 1.5 W5) ID (21.7) 3,?2 112.4) one 20 mi κι A l ID W tansy to ia Mill ICC 111 I'll St)' ¹ 344 (P..21 4.0 (.Ο- Ι 1o94 (17.7) (-'ll 1 117.5) 2.4 (НМ0 1A < Yu.2. 1 _(M (22.% 3.'6 (LU) one III Ml KI Λ (BUM ntm loda MHi.rxcraia MD ⁴ 353<1x7, 4.UCL5- 1804 <1?-% 18(18 <15.s) 2.5 id) 1A (21 AI 1.02 ¢21.4) 3.73 1 12.3! one in mi kg Wiki high to in mi.psgaga SIT 349 L.UTS 4 H¢3.5i . 1878 (P.5) 1m < 17.51 2.7 2 ¹ 1 1.4.91 0 -lx (26.5) 3. .'4 112.3) one 2U MI κι Λ l V21Yu high .yush msh.ga-ysh MP ⁴ 356 <2i.2> i ΧΝ(ι W1 W) “015) 2.i (1(..11 0.9S (27.% 3.6 3 2(1 Ml and L1V21Yu supreme, ιοκι Mitju-wu Ml) 20! (21,Sch 4.3 (J.u- one % 1I ('4% 1%III 13-1.9) 2.4 i5.% 2 114.51 (MH 2.1'1 ΙΠ.3) 3 2d Mi κι MBI'KI high do and mi glee eider MD ^f <27% 4.0 ¢4.0- HAI <2% |yu' (32.6) 2.0 id? H") 0.66 ,2h 2.33 (23.2)

: LSS nlchiad juu mpshoy: CI : total clearance sn <411 nw s: maximum action

-ΊΙ-Κ Prel ven mn and srsdsny; CV to "'ffishp-'SH SHMSCHSHIIOs gi; MP multifold LP sm; SI)

l. piksch I share .NIH t| _: first, I pl vykeleshy; (max krsma dostzheii maxmandyuy konpei ι early leutschtennok "medium tx V,. - the volume of rice N|HLs.'1ashya in c-|ationar1yum cosshinni 'Cp" 1g <.h 1 ewicipmwKiK (CW).' Me.nmm H1SHP1mal1yue- shke1YM.noL 'CpoHA'^fmfmtgaldl'e'ns.'nm'n 2.

PK data for miglustat

Miglustat showed a dose-proportional dependence of kinetic parameters (Table 8 and Fig. 26). Plasma levels for miglustat appear to be similar for single and multiple doses.

Table 8. Summary of PK data for miglustat

Treatment tmax (ng/ml) ⁱ tmax (h) ^b AUCo-t (w^h/ml)* And GS. (ng*h/ml) ⁱ Ta(h) ^s CL/F -leagues V _z /F (l G Low dose SD I486 (29.9) 3.5 (1.5- 3.5) 11807 (25.6) 12565 (26.8) 5.6 (11.7) 10.6 (23) 85.8 (23.4) Low dose MD 1518 (27.6) 3.0 (1.5- •3.5) 12254 (26.4) 13094 (28.3) 5.9 (32.1) 10.2 (23.9) 86.7 (43.9) High dose SD 3059 (36.1) 3.5 (1.5- 5) 23999 (35) 25859 (34.4) 5.7 (29.9) 10.6 (33) 86.3 (45.7) High dose MD 3569 (25.5) 3.0 (1.04.0) 24970 (24.1) 25972 (23) 5.3 (15.6) 10.3 (26.4) 81 (41.8) V _z = apparent volume of distribution in the final state. ^a Geometric mean (CV%). ^b Median (minimum-maximum). ^c Arithmetic mean (SU%).

Pharmacodynamic parameters

By visit 11, ERT-treated patients from cohort 1 (FIGS. 27A and 27B):

alanine aminotransferase (ALT) levels decreased in 5 of 8 patients; normalized in 4/4 patients with elevated baseline levels;

aspartate aminotransferase (AST) levels decreased in 6 of 8 patients; normalized in 3/4 patients with elevated baseline levels;

creatine phosphokinase (CPK) levels decreased in 6 of 8 patients; returned to normal in 2/6 patients

-41 039750 with elevated baseline levels;

urinary tetrasaccharide glucose (HEX4) levels decreased in 8 of 8 patients.

By week 4, levels of all 4 biomarkers decreased in 2 patients in the untreated cohort (cohort 3) (FIGS. 2?C and 27D).

In FIG. 27A-27D, data are presented as mean ± standard deviation.

Security

After a total of 155+ infusions, no serious adverse events (AE) or infusion-related reactions were reported in all patients.

Post-treatment AE, which occurred in 11/13 (84%) patients, was generally mild and transient.

Treatment-related AE reported in 7/13 (53%) patients: nausea (n=1), fatigue (n=1), headache (n=1), tremor (n=2), acne (n =1), tachycardia (n=1) and hypotension (n=1).

findings

ATB200 alone and in combination with miglustat was safe and well tolerated, with no infusion-related reactions to date.

ATB200 alone showed a superproportional dose-dependent increase in exposure, which was further enhanced by miglustat, suggesting a stabilizing effect of the chaperone on ATB200.

After switching from standard treatment to treatment with the ATB200/miglustat combination, patients generally showed improvement in muscle damage biomarker levels, with many patients showing normalization by week 18.

Initially untreated patients treated with the ATB200/miglustat combination showed sustained reductions in all biomarkers of muscle damage.

The embodiments described herein are intended to be illustrative of the compositions and methods of the present invention and are not intended to limit the scope of the present invention. Various modifications and changes that are consistent with the description as a whole and obvious to a person skilled in the art are meant to be included. The appended claims should not be limited to the specific embodiments set forth in the examples, but should provide the broadest possible interpretation consistent with the description as a whole.

Throughout this application, citations are made of patents, patent applications, publications, product descriptions, GenBank access numbers, and protocols, the disclosures of which are incorporated herein by reference in their entirety in all respects.

CLAIM

Claims (10)

CLAIM 1. A method for producing purified recombinant human acid alpha-glucosidase (rhGAA), wherein the purified rhGAA contains an amino acid sequence that is at least 98% identical to SEQ ID NO: 2, the method comprising culturing host cells that secrete rhGAA in a bioreactor ; removing the medium from the bioreactor; filtering the medium to obtain a filtrate; loading the filtrate onto an anion exchange chromatography (AEX) column to capture rhGAA; eluting the first protein product from the AEX column; loading the rhGAA eluted from the AEX column onto an immobilized metal ion affinity chromatography (IMAC) column; and eluting the rhGAA from the IMAC column. 2. The method of claim 1, wherein the purified rhGAA contains a first potential N-glycosylation site, a second potential N-glycosylation site, a third potential N-glycosylation site, a fourth potential N-glycosylation site, a fifth potential N-glycosylation site, a sixth potential N-glycosylation site, and a seventh potential N-glycosylation site. 3. The method of claim 1 or 2, further comprising loading the rhGAA eluted from the IMAC column into a third chromatographic column and eluting the rhGAA from the third chromatographic column. 4. The method of claim 3, wherein the third chromatographic column is selected from a cation exchange chromatography (CEX) column and a size exclusion chromatography (SEC) column. 5. The method according to any one of claims 1 to 4, wherein the media filtration is selected from variable tangential flow filtration (ATF) and tangential flow filtration (TFF). 6. The method of any one of claims 1 to 5, further comprising inactivating viruses in one or more of rhGAA eluted from the AEX column, rhGAA eluted from the IMAC column, and rhGAA eluted from the third chromatographic column. - 42 039750 7. The method of claim 3 further comprising filtering the rhGAA eluted from the column for IMAC, or rhGAA, eluted from the third chromatographic column to obtain a filtered product. 8. The method of claim 7 further comprising lyophilizing the filtered product. 9. The method of any one of claims 1 to 8, wherein the host cells include Chinese hamster ovary (CHO) cells. 10. The method according to any one of claims 1-9, where: (i) at least 90% of the purified rhGAA binds to the cation-independent mannose-6-phosphate receptor (CIMPR) or (ii) at least 90% of the purified rhGAA contains an N-glycan bearing mono-mannose-6-phosphate (mono-M6P ) or bis-mannose-6-phosphate (6c-M6P). Structure of high-mannose and phosphorylated oligosaccharides and their affinity