Impact of high-dose, chemically modified sulfamidase on pathology in a murine model of MPS IIIA

Impact of high-dose, chemically modified sulfamidase on pathology in a murine model of MPS IIIA

Experimental Neurology 230 (2011) 123–130 Contents lists available at ScienceDirect Experimental Neurology j o u r n a l h o m e p a g e : w w w. e ...

769KB Sizes 11 Downloads 32 Views

Experimental Neurology 230 (2011) 123–130

Contents lists available at ScienceDirect

Experimental Neurology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y e x n r

Impact of high-dose, chemically modified sulfamidase on pathology in a murine model of MPS IIIA Tina Rozaklis a, Helen Beard a, Sofia Hassiotis a, Antony R. Garcia b, Matthew Tonini b, Amanda Luck a, Jing Pan b, Justin C. Lamsa b, John J. Hopwood a, Kim M. Hemsley a,⁎ a b

Lysosomal Diseases Research Unit, SA Pathology [Women's and Children's Hospital Campus], North Adelaide, SA 5006, Australia Shire Human Genetic Therapies Inc., 700 Main Street, Cambridge, MA 02139, USA

a r t i c l e

i n f o

Article history: Received 15 November 2010 Revised 29 March 2011 Accepted 7 April 2011 Available online 16 April 2011 Keywords: Mucopolysaccharidosis Sanfilippo Mouse Enzyme replacement therapy Brain

a b s t r a c t Mucopolysaccharidosis type IIIA (MPS IIIA) is a neurodegenerative lysosomal storage disorder that results from a deficiency of sulfamidase (N-sulfoglucosamine sulfohydrolase), with consequential accumulation of its substrate, partially degraded heparan sulfate. Conventional doses (e.g. 1 mg/kg) of intravenously delivered recombinant human sulfamidase (rhSGSH) do not improve neuropathology in MPS IIIA mice due to an inability to traverse the blood–brain barrier; however high-dose treatment or administration of enzyme that has been chemically modified to remove mannose-6-phosphate glycans has been shown to reduce neuropathology in related animal models. We have combined these approaches to evaluate the ability of 1, 5, 10 or 20 mg/kg of similarly chemically modified or unmodified rhSGSH to reduce neuropathology following repeated intravenous delivery to adult MPS IIIA mice. rhSGSH was detected in brain homogenates from mice treated with all doses of modified rhSGSH and those receiving the two higher doses of unmodified rhSGSH, albeit at significantly lower levels. Immunohistochemically, rhSGSH visualized in the brain was localized to the endothelium, meninges and choroid plexus, with no convincing punctate intra-neuronal staining seen. This presumably underlies the failure of the treatment to reduce the relative level of a heparan sulfate-derived oligosaccharide (GlcNS-UA), or secondarily stored substrates that accumulate in MPS IIIA brain cells. However, modification of rhSGSH significantly increased its effectiveness in degrading GlcNS-UA in non-CNS tissues, potentially as a result of its reduced plasma clearance. If this observation is generally applicable, chemical modification may permit the use of significantly lower doses of lysosomal enzymes in patients currently receiving intravenous enzyme replacement therapy. Crown Copyright © 2011 Published by Elsevier Inc. All rights reserved.

Introduction Sanfilippo syndrome or mucopolysaccharidosis type III (MPS III) is a lysosomal storage disorder (LSD) that consists of four subtypes (A, B, C and D), the most common of which is MPS IIIA. MPS IIIA results from a deficiency of the lysosomal sulfatase, sulfamidase (N-sulfoglucosamine sulfohydrolase; SGSH; EC3.10.1.1), which is responsible for cleaving glucosamine-N-sulfate bonds at the non-reducing end of heparan sulfate (HS) fragments (Hopwood and Morris, 1990;Scott et al., 1995). The consequence of SGSH deficiency is accumulation of HS-derived oligosaccharides and secondarily stored glycolipids, including GM2 and GM3 ganglioside and unesterified cholesterol in the lysosomal compartment (McGlynn et al., 2004), with the central nervous system

Abbreviations: MPS IIIA, Mucopolysaccharidosis IIIA; CNS, central nervous system; LSD, lysosomal storage disorder; SGSH, N-sulfoglucosamine sulfohydrolase; HS, heparan sulfate; ERT, enzyme replacement therapy; M6P, mannose-6-phosphate; M6PR, mannose-6-phosphate receptor. ⁎ Corresponding author. Fax: + 61 8 8161 7100. E-mail address: [email protected] (K.M. Hemsley).

(CNS) being the primary site of pathology. MPS IIIA is characterized by progressive CNS degeneration and early death in patients, and there is presently no treatment. Enzyme replacement therapy (ERT) strategies have been effective in treating non-neuronal pathology in lysosomal storage disorders such as MPS I (Kakkis et al., 2001), MPS II (Muenzer et al., 2002), MPS VI (Harmatz et al., 2006) and Gaucher (Barton et al., 1990) and Fabry (Schiffmann et al., 2001) diseases. This method relies on the delivery of intravenously (i.v.) administered enzyme to cell surface receptors of affected tissues. Recombinant human SGSH (rhSGSH) uptake by human skin fibroblasts has been shown to occur via the mannose-6phosphate (M6P) receptor (Bielicki et al., 1998). The transport of rhSGSH into the brain parenchyma of neonatal MPS IIIA mice is also mediated by M6P receptors (M6PR), however this up-take mechanism is developmentally down-regulated by 2 weeks of age (Urayama et al., 2008) and thereafter, i.v. ERT at conventional doses is unable to deliver sufficient enzyme to the CNS to reduce neuropathology (Gliddon and Hopwood, 2004). Recently, high-dose i.v. ERT has been successfully employed in several LSD animal models e.g. aspartylglycosaminuria mice (Dunder

0014-4886/$ – see front matter. Crown Copyright © 2011 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2011.04.004

T. Rozaklis et al. / Experimental Neurology 230 (2011) 123–130

Materials and methods Chemical modification of rhSGSH rhSGSH was manufactured and purified by Shire Human Genetic Therapies (Hemsley et al., 2007) and chemically modified as described by Grubb et al. (2008). Briefly, the enzyme was dialyzed against two changes of 20 mM Na2HPO4, 0.1 M NaCl (pH 6.0) at 4 °C and treated with 20 mM sodium meta-periodate prepared in the same buffer for 6.5 h in the dark. Periodate treatment will eliminate all carbohydratemediated clearance of enzyme, by either mannose- or mannose-6phosphate receptors. The reaction was quenched with 200 mM ethylene glycol for 15 min and after repeated dialysis as before, the enzyme was reduced in a final concentration of 0.1 M sodium borohydride (NaBH4) overnight at 4 °C, in the dark. The preparation was then dialyzed against two changes of 20 mM Na2HPO4, 0.1 M NaCl (pH 7.5) and was subsequently referred to as modified SGSH (modSGSH) and stored at 4 °C. As a control an additional aliquot of rhSGSH was exposed to identical pH changes by dialysis for equivalent times without periodate or borohydride treatment and is referred to as rhSGSH. The specific activity of modSGSH was reduced to 70% of the unmodified protein as a result of the modification procedure.

MPS IIIA mice and rhSGSH administration Congenic C57BL/6J MPS IIIA mice (Crawley et al., 2006b) or their unaffected littermates (wild-type +/+ or heterozygote +/−) of both genders were obtained from a breeding colony established at the Children's Youth and Women's Health Service (CYWHS) animal facility. Mixed strain MPS IIIA mice (Bhaumik et al., 1999) were used for one experiment (10 mg/kg rhSGSH every second day for 1 month). Mice were genotyped according to established methods (Gliddon and Hopwood, 2004) and all breeding and experimental procedures were undertaken according to the guidelines of the National Health and Medical Research Council of Australia with the approval of the Institutional Animal Ethics Committee (Adelaide, South Australia), or the Institutional Animal Care and Use Committee, Shire Human Genetic Therapies (Lexington, MA), an Association for Assessment and Accreditation of Laboratory Animal Care-accredited facility.

Evaluation of sulfamidase clearance from blood A single injection of 20 mg/kg of rhSGSH or modSGSH was administered i.v. via the tail vein of 6-week-old MPS IIIA mice. Three mice were euthanized at each of 4 h, 12 h, 24 h, 3 and 7 days post-injection. Age-matched MPS IIIA and unaffected untreated mice (n = 3) were included as controls. All mice were euthanized via CO2 asphyxiation and a blood sample was collected by cardiac puncture; mice were perfused with phosphate-buffered saline (PBS, pH 7.4) and brain and peripheral organs were collected and stored at −20 °C. Blood samples were allowed to clot at room temperature prior to centrifugation at 13,000 rpm. Serum was harvested and stored frozen until use. Evaluation of the effect of repeated i.v. delivery of sulfamidase on MPS IIIA-related pathology Groups of 6-week-old MPS IIIA mice received four i.v. injections of 1, 5, 10 or 20 mg/kg rhSGSH or modSGSH on a weekly basis. Mice received chlorpheniramine maleate (2.5 mg/kg) 30 min prior to injection to prevent hypersensitivity reactions and were euthanized after the fourth injection, at 9 weeks of age as described above. The total duration of the experiment was therefore 3 weeks and 1 day. A cohort of 19-week-old MPS IIIA mice received four weekly i.v. injections of 1 mg/kg rhSGSH (n = 5) or modSGSH (n = 4) and four mice received repeated injections of enzyme that had only undergone borohydride reduction (redSGSH). These mice were euthanized at 22 weeks of age. Finally, a cohort of 8- to12-week-old MPS IIIA mice received 10 mg/kg unmodified rhSGSH or vehicle (n = 6/group) i.v. every second day for 1 month. All mice were euthanized 24 h after the final injection. Two to three mice per group underwent fixation-

rhSGSH ng/mg serum protein

et al., 2000), MPS VII mice (Vogler et al., 2005), α-mannosidosis mice (Blanz et al., 2008) and guinea pigs (Crawley et al., 2006a), arylsulfatase A deficient mice (Matzner et al., 2009) and MPS II mice (Polito et al., 2010) to reduce lysosomal storage and in some cases, improve clinical function. Chemical modification of lysosomal βglucuronidase (P-GUS) by sodium meta-periodate has also been shown to increase its transport across the blood–brain barrier and mediate correction of neuronal storage in MPS VII mice (Grubb et al., 2008). Inactivation of carbohydrate-dependent uptake of P-GUS prolonged its clearance from circulation by 60-fold. Finally, another method by which to supply enzyme to the brain is via direct intracerebrospinal fluid injection (e.g. Hemsley et al., 2009), however this strategy results in somewhat regionalized enzyme delivery and is significantly more invasive than an intravenous injection. The best therapeutic strategy is likely to be one that delivers enzyme uniformly to the brain, with the bloodstream being the ideal conduit, given that each neuron has been estimated to be 15 micrometers (on average) μm from a blood vessel (Tsai et al. (2009)). Therefore in the present study we have combined the high-dose and chemically modified enzyme approaches described above and have administered high-dose modified or unmodified rhSGSH i.v. on a repeated basis to adult MPS IIIA mice. Enzyme clearance from blood was examined and efficacy of the two rhSGSH preparations in ameliorating MPS IIIA-related pathology was evaluated.

A 10000 1000 100 10 1 0.1 0.01

B SGSH activity pmol/min/mg serum protein

124

10000 1000 100 10 1 0.1 0.01

0h

4h

12h

1d

3d

7d

Time post injection Fig. 1. Clearance of i.v.-injected rhSGSH or modSGSH (20 mg/kg) from the circulation. A: The amount of modSGSH (solid squares) or rhSGSH (open circles) detected in MPS IIIA mouse serum at various times post-injection (n = 3/group; mean± SEM). B: Enzyme activity in the serum samples. The dotted line represents the average level of serum protein or activity (respectively) observed in normal mice (n = 3). SGSH protein and activity were undetectable in untreated MPS IIIA mouse serum (n = 3).

T. Rozaklis et al. / Experimental Neurology 230 (2011) 123–130

perfusion for light microscopic examination of tissues (4% paraformaldehyde in PBS, pH 7.4) and the remaining mice were perfused with PBS for biochemical evaluation. Age-matched untreated MPS IIIA and normal mice (n = 6/genotype) were also utilized in each study. rhSGSH protein and activity in tissue and serum Half-brain and organ samples were homogenized in 1 mL of 20 mM Tris, 0.5 M NaCl, pH 7.2 and then sonicated for 1 min on ice. Total protein was determined with a micro-BCA protein assay kit (Thermo Scientific, Rockford, IL, USA). rhSGSH protein was measured in homogenates using a previously published ELISA-based immunoquantification method (Lau et al., 2010a) and corrected for total protein. rhSGSH activity was determined using a heparin-derived tritiated tetrasaccharide substrate as previously described (Hopwood and Elliot, 1982), with homogenates dialyzed against 0.05 M NaAc buffer (pH 5) and 50 μg of dialysate protein incubated with 400 pmol of GlcNS-UA-GlcNS-UOA (in a final volume of 12 μL) at 60 °C for 2– 16 h, until such a time that ≤30% substrate conversion was achieved. Baseline SGSH protein and activity in serum was determined by spiking a given volume of age-matched MPS IIIA mouse blood sample with either rhSGSH or modSGSH at a dose calculated to be equivalent to 20 mg/kg mouse body weight, based on a mouse having 5.5 mL blood per 100 g body weight (Mouse Phenome Database; Jackson Laboratory).

125

2010b). Medial sagittal sections of paraffin-embedded brain tissue were cut on a rotary microtome (6 μm; Leica, Wetzlar, Germany) and stained for LIMP II according to previously published methods (Hemsley et al., 2009), using a monoclonal anti-lysosomal integral membrane protein (LIMP-II) antibody raised in mouse against a synthetic peptide (Mimotopes, Australia). The peptide maps the Cterminal region of human LIMP-II. This antibody was kindly donated by Dr. E. Parkinson-Lawrence and Professor Doug Brooks (UniSA). Paraffin-embedded liver and brain tissue was examined for SGSH protein using a monoclonal antibody to human sulfamidase, 2C7 (supplied by Shire Human Genetic Therapies, Cambridge, MA, USA). Sections were de-paraffinized followed by heat-induced epitope retrieval for 45 min at 95 °C (citrate-EDTA buffer, pH 6.2), cooled and blocked with 10% normal donkey serum (NDS; Jackson ImmunoResearch, West Grove, PA, USA) prior to incubation overnight with 2C7 antibody (1:50 dilution in 2% NDS). Endogenous peroxidase was quenched with 0.3% hydrogen peroxidase in PBS, a biotinylated donkey anti-mouse IgG antibody (1:1600 in PBS; Jackson ImmunoResearch, West Grove, PA, USA) was applied for 1 h before incubation with Vectastain ABC reagent (Vector Laboratories, California, USA) and visualization with DAKO diaminobenzidine (DAB) liquid chromogen system (DAKO, Glostrup, Denmark). Mouse brain regions were defined with reference to Paxinos and Franklin (2001).

Immunohistochemistry

Liquid chromatography (LC)-electrospray ionization tandem mass spectrometry (ESI-MS/MS) analysis of HS-derived oligosaccharides in tissues

Medial sagittal sections of frozen brain tissue were cut on a cryostat (6 μm; Shandon Cryotome E, ThermoFisher Scientific, Australia) and stained for unesterified cholesterol and GM3 ganglioside according to published methods (McGlynn et al., 2004, Lau et al.,

Liver, spleen, heart and brain tissues were homogenized as above and 50 or 100 μg of homogenate was derivatized with PMP, processed and analyzed by mass spectrometry according to previously described methods of Hemsley et al. (2007, 2009). The relative amount of the

A

Liver #

1 0.1

SGSH ng/mg protein

C

Spleen #

10

#

1 0.1 0.01

Liver #

10 1 0.1

100

D

Spleen #

10 1 0.1

0.01

E

Heart

100

GlcNS-UA ratio/mg protein

SGSH ng/mg protein

100

B

0.01

0.01 100

GlcNS-UA ratio/mg protein

#

10

100

GlcNS-UA ratio/mg protein

SGSH ng/mg protein

100

10 1 0.1

F

Heart

10 1 0.1

0.01

0.01 MPS vehicle

MPS MPS Normal rhSGSH modSGSH vehicle

MPS vehicle

MPS MPS Normal rhSGSH modSGSH vehicle

Fig. 2. SGSH protein and relative level of GlcNS-UA in organs from mice treated with 1 mg/kg rhSGSH or modSGSH for 4 weeks from 6 weeks of age. The amount of SGSH protein recorded in homogenates of liver (A), spleen (C) and heart (E) taken from enzyme-treated mice is shown (n = 3; mean ± SEM; ✧ c.f. MPS IIIA mice; # c.f. normal mice; p b 0.05). The relative levels of GlcNS-UA in the same tissue homogenates (B, D and F respectively) are shown. Data points represent individual mice, ✧ c.f. MPS IIIA mice; # c.f. normal mice; p b 0.05.

126

T. Rozaklis et al. / Experimental Neurology 230 (2011) 123–130

HS-derived disaccharide glucosamine N-sulfate [α-1, 4] uronic acid (GlcNS-UA) was calculated with respect to the level of an internal standard. The co-efficient of variation in each batch was determined by inclusion of replicate quality control (QC) MPS IIIA mouse brain tissues.

Measurement of serum antibodies to sulfamidase Serum antibody titers were determined according to Hemsley et al. (2009), with each plate coated with the respective enzyme preparation delivered i.v. Antibody complexes were determined at O.D. 405 nm and titers expressed as the dilution for an absorbance greater than two standard deviations above the blank. A positive control sample was included with each assay and consisted of MPS IIIA mouse serum containing high titer antibodies to rhSGSH (N1/51,200).

Results Clearance of sulfamidase from circulation The data plotted in Fig. 1 indicate that rhSGSH is cleared more rapidly from circulation than modSGSH. At 4 h post-infusion, rhSGSH protein and activity levels in serum were found to have reduced to approximately 6–7-fold those seen in normal mice, suggesting extensive clearance of enzyme from circulation by the liver in this period. modSGSH protein and activity levels, however, were still 19.8% and 2.3% (respectively) of those recorded at 0 h (~9675-fold and 13840-fold normal mouse levels respectively). While SGSH could still be detected in the serum of modSGSH-treated mice 3 days after infusion, it was present at low, but supra-normal mouse levels (see dashed line in Fig. 1). Quantification of SGSH in tissue samples following repeated intravenous delivery

Statistical analysis Data from MPS IIIA enzyme-treated mice were compared to that from untreated MPS IIIA and normal mice using analysis of variance tests followed by post hoc testing with Bonferroni correction to adjust for multiple group comparisons. Data shown are mean ± standard error of the mean (SEM) and a p value of b0.05 was considered statistically significant. Histological data were log-transformed Y = (Log Y + 1) and then examined in the same manner.

A

Brain

100

GlcNS-UA ratio/mg protein

SGSH ng/mg protein

100 10 1 0.1 0.01

Brain

10 1 0.1

MPS MPS Normal rhSGSH modSGSH vehicle

C

MPS vehicle

Brain

100

GlcNS-UA ratio/mg protein

SGSH ng/mg protein

B

0.01 MPS vehicle

100

Similar amounts of rhSGSH and modSGSH protein were detected in liver and spleen samples taken from mice aged 9 and 22 weeks, following repeated i.v. injection of low dose (1 mg/kg) unmodified enzyme, with sulfamidase levels greatest in the spleen (Fig. 2 and Supp. Fig. 1, panels A, C). The level of modSGSH found in the heart was 5 and 10-fold higher in 9- and 22-week-old mice respectively, than the amount measured in rhSGSH-treated mice. While no rhSGSH was measured in the brain (Fig. 3A), administration of modSGSH at 1 mg/kg,

10 1 0.1

D

MPS

5

10

20

5

10

Brain

10 1 0.1

0.01

0.01

MPS

20 Normal

MPS rhSGSH MPS modSGSH

5

Brain

100

10 1 0.1

F

MPS vehicle

MPS rhSGSH

Normal vehicle

5

10

20 Normal

Liver

10 1 0.1 ND

0.01

0.01

20

Treatment (mg/kg)

GlcNS-UA ratio/mg protein

GlcNS-UA ratio/mg protein

E

10

MPS rhSGSH MPS modSGSH

Treatment (mg/kg) 100

MPS MPS Normal rhSGSH modSGSH vehicle

MPS vehicle

MPS rhSGSH

ND

Normal vehicle

Fig. 3. SGSH protein and relative level of GlcNS-UA in tissues from mice treated with rhSGSH or modSGSH. The amount of rhSGSH (A, C) and the relative level of GlcNS-UA (B, D, E, F) detected in brain (A–E) or liver (F) homogenates following i.v. treatment of MPS IIIA mice is shown. Mice received either 1 mg/kg rhSGSH/modSGSH (A, B) or 5, 10 or 20 mg/kg rhSGSH/modSGSH (C, D) administered weekly for 4 weeks. Mice in E and F received 10 mg/kg rhSGSH administered every other day for 1 month. N = 2–3; mean ± SEM; ✧ c.f. normal mice. ND = not detectable or b 0.01.

T. Rozaklis et al. / Experimental Neurology 230 (2011) 123–130

restored sulfamidase levels to 29% of normal. However, there was a dose-dependent increase in the amount of rhSGSH and modSGSH in brain (Fig. 3C) with modified enzyme always observed at higher concentrations than rhSGSH, at each dose level, with up to 188% of normal levels reached with 20 mg/kg modSGSH. When enzyme activity in brain tissue was assessed, we detected very low levels (0.07 pmol/min/mg, n = 3) in brain at 24 h following the final injection of 20 mg/kg mod SGSH. No enzyme activity was observed in the brain tissues of mice treated with any other preparation/dose. The activity detected in the 20 mg/kg mod SGSHtreated mice was 29% of normal levels (0.24 pmol/min/mg, n = 3).

Effect of SGSH on the relative level of a HS-derived disaccharide in tissue samples following repeated intravenous delivery modSGSH treatment of mice aged 6 to 9 weeks resulted in statistically superior reductions in GlcNS-UA in liver, spleen and heart (c.f. rhSGSH; Fig. 2B, D, F), with 6, 15 and 5-fold greater reductions in this stored substrate respectively, observed in 9-week-old mice (p b 0.001). Similarly, enhanced degradation of stored GlcNS-UA was observed in mice treated with modSGSH from 19 to 22 weeks of age, with 5-, 52- and 4-fold greater impact seen in liver, spleen and heart respectively (c.f. rhSGSH; p b 0.001; Supplementary Fig. 1B, D, F).

B

1.0

However, despite sulfamidase being detected in the brain at supranormal levels following 20 mg/kg modSGSH, treatment over this time-frame did not reduce GlcNS-UA levels in the brain of MPS IIIA mice treated with any dose of modSGSH or rhSGSH (Fig. 3B, D). Administration of unmodified rhSGSH (10 mg/kg) every other day failed to have any impact on GlcNS-UA levels in brain, despite normalization of this analyte in the liver (Fig. 3E, F).

Immuno- and histochemical staining of tissues Punctate (and therefore presumptively lysosomal) immuno-stained rhSGSH was observed in the liver of MPS IIIA mice treated with both enzyme preparations (Fig. 4A). Dose-dependent levels of rhSGSH and modSGSH were observed in liver (Fig. 4B) and all doses reduced levels of the late endosome/lysosomal marker (LIMP II) in liver by at least 54% (Fig. 4C). Brain tissues from mice treated with modified rhSGSH (20 mg/kg) exhibited punctate (presumptively lysosomal) immunostaining in the meninges, associated with endothelial cells of the vasculature and in choroid plexus (Fig. 4D, E, F). This staining was not observed in mice that did not receive enzyme treatment, nor where the primary antibody was omitted. Punctate, but less intense immunoreactivity was seen in the same areas in MPS IIIA mice treated with unmodified 20 mg/kg rhSGSH. However, using light microscopy, we

Liver #

% Thresholded Area Y = log[Y+1] LIMPII

1.2

% Thresholded Area Y = log[Y+1] SGSH

A

0.8 0.6 0.4 0.2 0.0

MPS

127

1.2

C

0.8 0.6 0.4 0.2 0.0 MPS

5 10 20 10 20 Normal 5 MPS rhSGSH MPS modSGSH

Treatment (mg/kg)

1.0 0.8 0.6 0.4 0.2 0.0 MPS

5 10 20 5 10 20 Normal MPS rhSGSH MPS modSGSH

Treatment (mg/kg)

F

H

I Superior Colliculus

0.4 0.3

#

0.2 0.1

Cholesterol score

Superior Colliculus

Normalized Y=Log+1 % TH Area of GM3 Staining

% Thresholded Area Y = log[Y+1] LIMPII

G 1.2

5 10 20 5 10 20 Normal MPS rhSGSH MPS modSGSH

Treatment (mg/kg)

E

D

Liver

1.0

Superior Colliculus +++ ++/+++ ++ +/++ + -/+

Superior colliculus

0.0 MPS

5 10 20 5 10 20 Normal MPS rhSGSH MPS modSGSH

Treatment (mg/kg)

MPS

5 10 20 5 10 20 Normal MPS rhSGSH MPS modSGSH

Treatment (mg/kg)

Fig. 4. Effect of SGSH on pathology. The photo in A demonstrates immunostaining of rhSGSH in liver, with quantification of immunostaining shown in B and the impact of treatment on LIMP-II immunostaining in C. Immunostaining of rhSGSH in mouse brain meninges, within endothelial cells in the wall of a blood vessel and within cells of the choroid plexus is shown in D, E, F, respectively. A quantitative assessment of immunohistochemical staining of LIMP-II and GM3 in a representative area of brain (superior colliculus; see arrow in brain diagram) is shown in G and H, respectively. Unesterified cholesterol stained with filipin was semi-quantified using a −/+/++/+++ scale and is shown in I. Data in B, C, G, H are mean ± SEM from n = 3 mice, ✧ c.f. MPS IIIA mice; # c.f. normal mice; p b 0.05. The scale bar = 20 μm.

128

T. Rozaklis et al. / Experimental Neurology 230 (2011) 123–130

Table 1 Anti-rhSGSH antibody titers in mouse sera collected at euthanasia. Titers are expressed as the serum dilution at which the absorbance (measured at 405 nm) was greater than 2 standard deviations above the blank. Age at euthanasia

Treatment group

Serum antibody titers

9 weeks

Unaffected; vehicle MPS IIIA; vehicle MPS IIIA; rhSGSH; 1 mg/kg MPS IIIA; modSGSH; 1 mg/kg Unaffected; vehicle MPS IIIA; vehicle MPS IIIA; rhSGSH; 1 mg/kg MPS IIIA; redSGSH; 1 mg/kg MPS IIIA; modSGSH; 1 mg/kg Unaffected; vehicle MPS IIIA; vehicle MPS IIIA rhSGSH 5 mg/kg MPS IIIA rhSGSH 10 mg/kg MPS IIIA rhSGSH 20 mg/kg MPS IIIA; modSGSH 5 mg/kg MPS IIIA; modSGSH 10 mg/kg MPS IIIA; modSGSH 20 mg/kg

N 25, N 25, 25 N 25, 25, 25 400, 800, 800 200 400, 400 N 25, 25, 25, 25, 25, 50 N 25, N 25, 25, 25, 25, 25 25, 50, 400, 800, 800 400, 1600, 1600, 3200 100, 200 400, 400 N 25, N 25, N 25, N 25, 25, 50 N 25, N 25, N 25, 25, 25, 25 400,800, 1600, 1600, 3200, 3200 100, 200, 800, 1600, 1600, 3200 400,800, 1600, 1600, 1600, 1600 N 25, 25, 100, 800, 1600, 1600 N 25, 25, 100, 200, 200, 800 25, 25, 100, 400, 800, 1600

22 weeks

9 weeks

were unable to determine with any degree of certainty whether rhSGSH was localized in the lysosomes of neurons or glia in either modified or unmodified rhSGSH-treated MPS IIIA mouse brain. There was no change in the level of LIMP II, secondarily stored GM3 or the relative level of unesterified cholesterol in brain following i.v. delivery of any dose of rhSGSH or modSGSH (Fig. 4G, H, I respectively). Anti-sulfamidase antibody measurement in serum Compared to sera from untreated mice most enzyme-treated mice developed antibodies to the respective preparations of sulfamidase (Table 1), with slightly higher titers seen in rhSGSH and redSGSHtreated mice c.f. modSGSH-treated mice. Dose did not appear to be a factor in determining the level of the antibody titer. We have previously shown that mice injected with rhSGSH on an on-going basis via the cerebrospinal fluid also produce anti-rhSGSH antibodies, but that they are non-neutralizing (Hemsley et al., 2009). Therefore, we undertook a similar assay on samples from the modified/high-dose rhSGSH-treated mice i.e. pre-incubation of either modified or unmodified rhSGSH with sera samples from mice treated with 1, 5, 10 or 20 mg/kg of the same enzyme, followed by determination of enzyme activity against a natural substrate. The activity in rhSGSH-treated mice was then expressed as a % of the mean value recorded in four vehicle-treated mice (2 MPS, 2 unaffected). While one 10 mg/kg mod rhSGSH-treated mouse exhibited 72% activity, all other mice (n = 17) exhibited 97–100% of the activity recorded in MPS IIIA/unaffected vehicle-treated mice. This indicates that the low-titer antibodies generated were not neutralizing.

Discussion We have, for the first time, examined the impact of high-dose sulfamidase replacement therapy on MPS IIIA mouse pathology. Further, we have also determined the effectiveness of periodate-based chemical modification of sulfamidase in enabling brain delivery of modified protein. While high-dose modified ERT resulted in replacement of up to 22% of normal mouse brain SGSH protein, determined using tissue homogenates, microscopy-based examination of tissues 24 h after i.v. injection of either form of rhSGSH showed that this was likely due to measurement of rhSGSH in blood-associated compartments i.e. endothelial cells lining vessels, the meninges (arachnoid) and choroid plexus. There appeared to be no readily discernable modified or unmodified high-dose rhSGSH located in brain parenchyma using light microscopy-based techniques. The sensitivity of the immunostaining method and subsequent examination of tissues may have been insufficient to detect very low levels of rhSGSH present in neurons and glia, however, we observed no reduction in the storage of a HS-derived disaccharide, no change in LIMP-II levels, nor amelioration of GM3 or unesterified cholesterol accumulation. While rhSGSH delivered at a conventional dose (1 mg/kg) was undetectable by any method in treated mouse brain, it was found at high levels in spleen in particular, but also in liver and heart, where it mediated reductions in HS-derived GlcNS-UA. That said, levels still remained 885-, 9- and 174-fold in 9-week-old mice (respectively) above normal. Increasing the dose and frequency of rhSGSH administration completely normalized liver GlcNS-UA levels. Our observations made on brain tissue contrast with those described in the literature in other mouse models of neurodegenerative LSD, where high-dose ERT resulted in improved neuropathology and even improved clinical function (Table 2). This may indicate differences in the mode of uptake of lysosomal enzymes, or that the level of enzyme replacement needed in MPS IIIA tissues is significantly higher than that required in other conditions. The lack of detection of enzyme in neurons or glia in high-dose rhSGSH-treated MPS IIIA mice would tend to suggest the former is more likely. Alternatively, perhaps there are disease-based factors that have contributed to the uptake of lysosomal enzymes into the brain of other animal models e.g. blood–brain barrier leakiness, that are not present in MPS IIIA mice. Interestingly, MPS IIIB mice have recently been shown to exhibit ‘leakiness’ of the barrier to compounds such as Evan's blue dye, as early as 3 months of age (Garbuzova-Davis et al., 2011), however our studies indicate that the blood–brain barrier in MPS IIIA mice is not impaired, in terms of leakiness to i.v. delivered rhSGSH, at the ages used here. We confirmed that chemical modification of SGSH with periodate delays the clearance of modSGSH from the circulation compared to the unmodified enzyme. This mechanism was hypothesized by Grubb et al. (2008), with increased tissue availability being one reason for the improved neuronal architecture seen in treated MPS VII mice. It is not possible for us to calculate the factor by which we have increased the serum half-life of modSGSH, with levels in serum reduced to ~20%

Table 2 Summary of outcomes from high-dose ERT studies conducted on animal models of neurodegenerative lysosomal storage disorders. Nd = not determined. Model

Dose

Frequency

% Normal enzyme activity achieved

Reduced neuropathology

Improved clinical parameters

Reference

Aspartylglycosaminuria mice α-Mannosidosis mice MPS VII mice Krabbe mice α-Mannosidosis guinea pigs MLD mice MPS II mice MPS IIIA mice

10 mg/kg 500 units 20 mg/kg 6 mg/kg 10 mg/kg 50 mg/kg 1.2–10 mg/kg 20 mg/kg 10 mg/kg

Every other day Twice a week Weekly Weekly Once Twice a week Every 2–7 days Weekly Every other day

Nd 15% ≤ 2.5% 7% Nd Nd Up to ~ 5% 22% protein Nd

Yes Yes Yes Yes Yes Yes Yes No No

Nd Yes Nd Yes Nd Yes Yes Nd Nd

Dunder et al. (2000) Roces et al. (2004); Blanz et al. (2008) Vogler et al. (2005) Lee et al. (2005) Crawley et al., 2006a Matzner et al (2009) Polito et al., 2010 This study This study

T. Rozaklis et al. / Experimental Neurology 230 (2011) 123–130

of the amount injected by 4 h post-injection; the first time-point examined (compared to negligible levels of rhSGSH), however a significant extension is apparent with low levels of modSGSH protein still detectable in serum after 3 days. The plasma half-life of modified lysosomal β-glucuronidase was increased from 11.7 min to 18 h with modification, a greater than 90-fold increase (Grubb et al., 2008). In contrast to our findings, chemical modification of β-glucuronidase increased its transport across the blood–brain barrier, as evidenced by reduced neuronal storage following 12 weekly i.v. injections of 2 or 4 mg/kg modified enzyme (Grubb et al., 2008) to MPS VII mice. Mice were euthanized 1 week post-last injection. These authors measured enzyme activity in the brain of mice treated with 4 mg/kg modified β-glucuronidase at 48-h post-injection and found it to be 7.8% of normal levels. Immunohistochemical verification of GUSB localization within neurons was not performed, although the reduction in storage related lesions would tend to indicate that GUSB was internalized within neurons. We did not see evidence of intraneuronal or glial rhSGSH delivery, nor observe improvements in either biochemical measurements of primary stored substrate (GlcNSUA) in brain, or decreases in the amount of LIMP-II expression in brain or reduced levels of secondarily stored GM3 and unesterified cholesterol, as determined using immuno- or histochemistry. One significant difference between the Grubb et al. (2008) study and our own may be the time-course of the experiment—a total of 3 weeks and 1 day in MPS IIIA mice and 12 weeks in MPS VII mice. Additionally, the pharmacodynamics, and potentially, mechanism of β-glucuronidase uptake and degradation appears quite different to that of SGSH. Grubb et al. (2008) demonstrate that modified βglucuronidase has a cell half-life of 12.9 days, with unmodified enzyme being reduced to 50% by 18.9 days. However, they commented that almost one-third of the activity of the modified enzyme remained 3 weeks later. We have previously reported (Gliddon and Hopwood, 2004) that the tissue half-life of unmodified rhSGSH is b24 h and our circulation half-life data indicates a significantly shorter t½ interval with SGSH (b4 h) than was found for βglucuronidase (18.5 h). Less than one-third of the original enzyme activity was seen at 4 h. Therefore, β-glucuronidase remains in both circulation and within cells/tissues for much greater time periods, and therefore the ‘window’ for substrate degradation is much greater. Perhaps a comparable impact upon disease parameters in MPS IIIA will only be able to be achieved with more frequent administration of SGSH. This leads to the question of how much SGSH is required in order to reduce primary and secondarily stored substrates and maintain them at low levels. The dynamics of HS removal following treatment and its subsequent re-accumulation are not yet understood, however it is apparent from other studies in MPS IIIA mice, that several-fold normal levels of enzyme must be supplied (at least initially) in order to reduce the pathological burden to any significant degree. While all enzyme-treated mice developed non-neutralizing antibodies, the titers we recorded were much lower than those reported for mice receiving intra-CSF injections of rhSGSH for extended periods of time (Hemsley et al., 2007). In summary, repeated i.v. administration of 1 mg/kg of modSGSH to adult MPS IIIA mice with established disease significantly reduced lysosomal storage in somatic organs, but there was no effect on brain. Although higher doses of modSGSH resulted in significantly higher (supra-normal) enzyme levels in brain homogenates, rhSGSH appeared to be localized only in endothelial cells, the meninges and choroid plexus, with an absence of punctate staining observed in neurons and glia. It was therefore not surprising that there was no apparent reduction in neuropathology over the time-frame used here. rhSGSH and β-glucuronidase exhibit significant differences in tissue half-life and potentially, uptake mechanisms, which may mean that they are unlikely to be equi-effective in any given therapeutic intervention.

129

Supplementary materials related to this article can be found online at doi:10.1016/j.expneurol.2011.04.004. Conflict of interest statement Role of funding sources: The study was primarily funded by the Sanfilippo Children's Research Foundation (SCRF; grant awarded to KHM and JJH). The SCRF did not have any role in study design, data collection, data analysis, interpretation of data, writing of the report or in the decision to submit the paper for publication. The every-otherday study was conducted and funded by Shire Human Genetic Therapies, who provided the mouse tissues to KMH and JJH for analysis. Shire HGT also provided rhSGSH for all experiments outlined in this manuscript. Disclosure statement: An international patent is held by JJH and others for mammalian sulfamidase and genetic sequences encoding it, for use in the investigation, diagnosis and treatment of subjects suspected of suffering from sulfamidase deficiency (US Patent # 5,863,782). Acknowledgments Prof. Bill Sly is acknowledged for kindly providing the method for chemical modification of enzyme. We thank Drs. Allison Crawley and Adeline Lau for assistance with some i.v. injections, Leanne Winner and Hanan Hannouche for genotyping the mice, Barbara King and Dr Maria Fuller for assistance with mass spectrometry, Lynn Scarman and Leslie Jenkins-White in the CYWHS Animal Care facility for daily animal care, and acknowledge Shire Human Genetic Therapies for providing the unmodified enzyme used in the study. This study would not have been possible without the generous funding provided for it by the Sanfilippo Children's Research Fund (SCRF) of Canada (to KMH and JJH). References Barton, N.W., Furbish, F.S., Murray, G.J., Garfield, M., Brady, R.O., 1990. Therapeutic response to intravenous infusions of glucocerebrosidase in a patient with Gaucher disease. Proc. Natl Acad. Sci. U S A 87, 1913–1916. Bhaumik, M., Muller, V.J., Rozaklis, T., Johnson, L., Dobrenis, K., Bhattacharyya, R., Wurzelmann, S., Finamore, P., Hopwood, J.J., Walkley, S.U., Stanley, P., 1999. A mouse model for mucopolysaccharidosis type III A (Sanfilippo syndrome). Glycobiology 9, 1389–1396. Bielicki, J., Hopwood, J.J., Melville, E.L., Anson, D.S., 1998. Recombinant human sulphamidase—expression, amplification, purification and characterization. Biochem. J. 329, 145–150. Blanz, J., Stroobants, S., Lullmann-Rauch, R., Morelle, W., Ludemann, M., D'Hooge, R., Reuterwall, H., Michalski, J.C., Fogh, J., Andersson, C., Saftig, P., 2008. Reversal of peripheral and central neural storage and ataxia after recombinant enzyme replacement therapy in alpha-mannosidosis mice. Hum. Mol. Genet. 17, 3437–3445. Crawley, A.C., King, B., Berg, T., Meikle, P.J., Hopwood, J.J., 2006a. Enzyme replacement therapy in alpha-mannosidosis guinea-pigs. Mol. Genet. Metab. 89, 48–57. Crawley, A.C., Gliddon, B.L., Auclair, D., Brodie, S.L., Hirte, C., King, B.M., Fuller, M., Hemsley, K.M., Hopwood, J.J., 2006b. Characterization of a C57BL/6 congenicmouse strain of mucopolysaccharidosis type IIIA. Brain Res. 1104, 1–17. Dunder, U., Kaartinen, V., Valtonen, P., Vaananen, E., Kosma, V.M., Heisterkamp, N., Groffen, J., Mononen, I., 2000. Enzyme replacement therapy in a mouse model of aspartylglycosaminuria. FASEB J. 14, 361–367. Garbuzova-Davis, S., Louis, M.K., Haller, E.M., Derasari, H.M., Rawls, A.E., Sanberg, P.R., 2011. Blood–brain barrier impairment in an animal model of MPS IIIB. PLoS One 6 (3), 216601. doi:10.1371/journal.pone.0016601. Gliddon, B.L., Hopwood, J.J., 2004. Enzyme-replacement therapy from birth delays the development of behavior and learning problems in mucopolysaccharidosis type IIIA mice. Pediatr. Res. 56, 65–72. Grubb, J.H., Vogler, C., Levy, B., Galvin, N., Tan, Y., Sly, W.S., 2008. Chemically modified {beta}-glucuronidase crosses blood–brain barrier and clears neuronal storage in murine mucopolysaccharidosis VII. Proc. Natl Acad. Sci. USA 105, 2616–2621. Harmatz, P., Giugliani, R., Schwartz, I., Guffon, N., Teles, E.L., Miranda, M.C., Wraith, J.E., Beck, M., Arash, L., Scarpa, M., Yu, Z.F., Wittes, J., Berger, K.I., Newman, M.S., Lowe, A.M., Kakkis, E., Swiedler, S.J., 2006. Enzyme replacement therapy for mucopolysaccharidosis VI: a phase 3, randomized, double-blind, placebo-controlled, multinational study of recombinant human N-acetylgalactosamine 4-sulfatase (recombinant human arylsulfatase B or rhASB) and follow-on, open-label extension study. J. Pediatr. 148, 533–539. Hemsley, K.M., King, B., Hopwood, J.J., 2007. Injection of recombinant human sulfamidase into the CSF via the cerebellomedullary cistern in MPS IIIA mice. Mol. Genet. Metab. 90, 313–328.

130

T. Rozaklis et al. / Experimental Neurology 230 (2011) 123–130

Hemsley, K.M., Luck, A.J., Crawley, A.C., Hassiotis, S., Beard, H., King, B., Rozek, T., Rozaklis, T., Fuller, M., Hopwood, J.J., 2009. Examination of intravenous and intraCSF protein delivery for treatment of neurological disease. Eur. J. Neurosci. 29, 1197–1214. Hopwood, J.J., Elliott, H., 1982. Diagnosis of Sanfilippo type A syndrome by estimation of sulfamidase activity using a radiolabelled tetrasaccharide substrate. Clin. Chim. Acta. 123, 241–250. Hopwood, J.J., Morris, C.P., 1990. The mucopolysaccharidoses. Diagnosis, molecular genetics and treatment. Mol. Biol. Med. 7, 381–404. Kakkis, E.D., Muenzer, J., Tiller, G.E., Waber, L., Belmont, J., Passage, M., Izykowski, B., Phillips, J., Doroshow, R., Walot, I., Hoft, R., Neufeld, E.F., 2001. Enzyme-replacement therapy in mucopolysaccharidosis I. N. Engl. J. Med. 344, 182–188. Lau, A.A., Hopwood, J.J., Kremer, E.J., Hemsley, K.M., 2010a. SGSH gene transfer in mucopolysaccharidosis type IIIA mice using canine adenovirus vectors. Molec. Genet. Metab. 100, 168–175. Lau, A.A., Hannouche, H., Rozaklis, T., Hassiotis, S., Hopwood, J.J., Hemsley, K.M., 2010b. Allogeneic stem cell transplantation does not improve neurological deficits in mucopolysaccharidosis type iiia mice. Exp. Neurol. 225, 445–454. Lee, W.C., Courtenay, A., Troendle, F.J., Stallings-Mann, M.L., Dickey, C.A., DeLucia, M.W., Dickson, D.W., Eckman, C.B., 2005. Enzyme replacement therapy results in substantial improvements in early clinical phenotype in a mouse model of globoid cell leukodystrophy. FASEB J. 19, 1549–1551. Matzner, U., Lullmann-Rauch, R., Stroobants, S., Andersson, C., Weigelt, C., Eistrup, C., Fogh, J., D'Hooge, R., Gieselmann, V., 2009. Enzyme replacement improves ataxic gait and central nervous system histopathology in a mouse model of metachromatic leukodystrophy. Mol. Ther. 17, 600–606. McGlynn, R., Dobrenis, K., Walkley, S.U., 2004. Differential subcellular localization of cholesterol, gangliosides, and glycosaminoglycans in murine models of mucopolysaccharide storage disorders. J. Comp. Neurol. 480, 415–426.

Muenzer, J., Lamsa, J.C., Garcia, A., Dacosta, J., Garcia, J., Treco, D.A., 2002. Enzyme replacement therapy in mucopolysaccharidosis type II (Hunter syndrome): a preliminary report. Acta Paediatr. Suppl. 91, 98–99. Paxinos, G., Franklin, K.B.J., 2001. The Mouse Brain in Stereotaxic Coordinates. Academic Press, San Diego, CA. Polito, V.A., Abbondante, S., Polishchuk, R.S., Nusco, E., Salvia, R., Cosma, M.P., 2010. Correction of CNS defects in the MPS II mouse model via systemic enzyme replacement therapy. Hum. Mol. Genet. 19, 4871–4885. Roces, D.P., Lullmann-Rauch, R., Peng, J., Balducci, C., Andersson, C., Tollersrud, O., Fogh, J., Orlacchio, A., Beccari, T., Saftig, P., von Figura, K., 2004. Efficacy of enzyme replacement therapy in alpha-mannosidosis mice: a preclinical animal study. Hum. Mol. Genet. 13, 1979–1988. Schiffmann, R., Kopp, J.B., Austin, H.A.r., Sabnis, S., Moore, D.F., Weibel, T., Balow, J.E., Brady, R.O., 2001. Enzyme replacement therapy in Fabry disease: a randomized controlled trial. JAMA 285, 2743–2749. Scott, H.S., Blanch, L., Guo, X.-H., Freeman, V.C., Orsborn, A., Baker, E., Sutherland, G.R., Morris, P., Hopwood, J.J., 1995. Cloning of the sulphamidase gene and identification of mutations in Sanfilippo A syndrome. Nat. Genet. 11, 465–467. Tsai, P.S., Kaufhold, J.P., Blinder, P., Friedman, B., Drew, P.J., Karten, H.J., Lyden, P.D., Kleinfeld, D., 2009. Correlations of neuronal and microvascular densities in murine cortex revealed by direct counting and colocalization of nuclei and vessels. J. Neurosci. 29, 14553–14570. Urayama, A., Grubb, J.H., Sly, W.S., Banks, W.A., 2008. Mannose 6-phosphate receptormediated transport of sulfamidase across the blood–brain barrier in the newborn mouse. Mol. Ther. 16, 1261–1266. Vogler, C., Levy, B., Grubb, J.H., Galvin, N., Tan, Y., Kakkis, E., Pavloff, N., Sly, W.S., 2005. Overcoming the blood–brain barrier with high-dose enzyme replacement therapy in murine mucopolysaccharidosis VII. Proc. Natl Acad. Sci. USA 102, 14777–14782.