Intracerebral injection of sulfamidase delays neuropathology in murine MPS-IIIA

Intracerebral injection of sulfamidase delays neuropathology in murine MPS-IIIA

Molecular Genetics and Metabolism 82 (2004) 273–285 www.elsevier.com/locate/ymgme Intracerebral injection of sulfamidase delays neuropathology in mur...

3MB Sizes 0 Downloads 5 Views

Molecular Genetics and Metabolism 82 (2004) 273–285 www.elsevier.com/locate/ymgme

Intracerebral injection of sulfamidase delays neuropathology in murine MPS-IIIA Peter S. Savas,a Kim M. Hemsley,b,¤ and John J. Hopwoodb a

Lysosomal Disorders Research Unit, Department of Chemical Pathology, Women’s and Children’s Hospital and University of Melbourne, Medical School 72 King William Road, North Adelaide, SA 5006, Australia b Lysosomal Disorders Research Unit, Department of Genetic Medicine, Women’s and Children’s Hospital and Department of Paediatrics, Adelaide University 72 King William Road, North Adelaide, SA 5006, Australia Received 17 March 2004; received in revised form 7 May 2004; accepted 10 May 2004 Available online 2 July 2004

Abstract Lysosomal storage disorders (LSD) are rare inherited metabolic diseases in which genetic alterations aVect lysosomal proteins. Mucopolysaccharidosis type IIIA (MPS-IIIA) is an LSD characterized by reduced activity of sulfamidase (heparan-N-sulfatase, EC3.10.1.1), which degrades the sulfated glycosoaminoglycan heparan sulfate. The central nervous system (CNS) is the main site of pathology in MPS-IIIA, resulting in reduced neurological function and neurocognitive decline. Neuropathological changes include lysosomal vacuolation of heparan sulfate and lipids in neurons, glia, and perivascular cells and the formation of axonal spheroids and ectopic dendrites. At present there is no eVective treatment for the CNS eVects of LSD as enzyme administered intravenously cannot cross the blood–brain barrier. We have previously established and characterized a mouse model of MPS-IIIA, and in the present study, we injected recombinant human sulfamidase directly into the brain at 6, 12 or 18 weeks of age. Treatment reduced vacuolation and gliosis and delayed the onset of ubiquitin-positive neurodegenerative changes in widespread areas of MPS-IIIA brain, assessed at 24 weeks of age. However, ubiquitin-positive axonal spheroids already detectable by 6 weeks of age were unaVected by treatment at any age, suggesting their irreversibility and thus indicating the importance of early detection of MPS-IIIA and instigation of therapy.  2004 Elsevier Inc. All rights reserved. Keywords: Lysosomal storage disorder; MPS-IIIA; Mucopolysaccharidosis; Sulfamidase; Mouse; Neurodegeneration; Neuropathology; Ubiquitin; GFAP; Brain

1. Introduction Lysosomal storage disorders (LSD) are inherited metabolic diseases in which genetic alterations aVecting lysosomal proteins disrupt the function of the lysosomal system. Individual LSD are rare disorders, however this group of around 50 diseases has an estimated prevalence in Australia of 1 per 7700 live births, not including the neuronal ceroid lipofuscinoses [1]. Mucopolysaccharidosis type IIIA (MPS-IIIA) otherwise known as SanWlippo type A is an LSD with a prevalence of around 1 per ¤

Corresponding author. Fax: +61-8-8161-7100. E-mail addresses: [email protected] (P.S. Savas), [email protected] (K.M. Hemsley), [email protected] adelaide.edu.au (J.J. Hopwood). 1096-7192/$ - see front matter  2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2004.05.005

114,000 live births [1] and is characterized by reduced activity of the lysosomal enzyme sulfamidase (heparanN-sulfatase or EC3.10.1.1), which degrades the sulfated glycosoaminoglycan heparan sulfate [2]. Although the somatic organs are aVected in MPS-IIIA, the dominant clinical feature of this disorder (and MPS types IIIB, IIIC, and IIID) is neurological dysfunction, resulting in neurocognitive decline. A mouse model of MPS-IIIA has been characterized [3]. As a result of a mis-sense mutation D31N (aspartic acid to asparagine) in the sulfamidase gene [4], the mouse possesses only »3% normal sulfamidase activity. Mice also excrete a disproportionate amount of heparan sulfate in their urine, and the brain and urine contain an excessive amount of longer chain heparan sulfate oligosaccharides normally amenable to digestion by

274

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

sulfamidase. These Wndings resemble data obtained from human MPS-IIIA patients [5,6] and other animal models of MPS-III [7–9]. MPS-IIIA mice exhibit somatic pathology consistent with human MPS-IIIA, including facial dysmorphism, skin thickening, coarse fur, mild skeletal deformity, and hepatosplenomegaly [3,10]. Mice exhibit hyperactivity from 4 to 10 weeks of age, gait disturbances from 10 weeks of age, aggressive behavior from »15 weeks of age (K. Hemsley and B. Gliddon, personal observations), and diminished performance in the Morris Water Maze by 20 weeks of age [10]. Females display poor maternal behavior involving rejection of, or killing of pups and both sexes become lethargic later in life. Possibly underpinning these behavioral disturbances, lysosomal storage similar to that found in human MPS-III patients is observed in a broad range of tissues, including extensive depositions in many areas of the brain. The prevailing paradigm for treatment of LSD caused by enzyme deWciency is enzyme replacement therapy (ERT). The nature of the mannose-6-phosphate receptor system provides a unique opportunity for therapy, since correctly glycosylated enzyme supplied to the extracellular space will be taken up and targeted to the lysosomes where it may act to clear stored substrates [11]. Intravenous administration of recombinant human enzymes has been shown to produce a therapeutic eVect on the somatic pathology of many LSD in humans and animal models [11]. However, intravenous administration of lysosomal enzymes has no eVect on CNS pathology due to the impermeability of the blood–brain barrier (BBB) to large molecules [12]. We have recently demonstrated that i.v. injection of sulfamidase does not alter the pathological or behavioral processes occurring in the murine MPS-IIIA brain when enzyme is supplied after the BBB has closed [10]. Bone marrow transplantation (BMT) has been used as a therapy in both neuronopathic and non-neuronopathic LSD. EYcacy is well established for improving somatic pathology in diseases such as Gaucher type I [13], however BMT has been unsuccessful in treating MPS-IIIA [14]. A variety of experimental therapies are presently under evaluation for LSD with CNS involvement. Some of the approaches currently in development include, cell mediated therapy [15,16], enzyme enhancement therapy [17], and viral gene therapy [18]. However, in the absence of wide-scale newborn screening for LSD, which would enable early detection before the disease is entrenched, the experimental therapies for LSD with CNS involvement may be of limited eYcacy. Thus far only one study has provided evidence suggesting the reversibility of functional neurological deWcits in the MPS group of disorders. Enzyme replacement in the CNS via viral gene therapy in adult MPS-VII mice produced a reversal in cognitive defects as assessed by continuous behavioral testing prior to and following treatment [19].

Therefore, in this study we sought to determine whether supplying recombinant human sulfamidase to the MPSIIIA mouse CNS at various ages representing early, moderate, and advanced disease phenotypes could prevent, delay or even reverse the various pathological features of murine MPS-IIIA.

Materials and methods Production of recombinant human sulfamidase Recombinant human sulfamidase (rhNS) was manufactured and puriWed according to established procedures [20]. The Wnal eluate was concentrated to 17 g/l protein in 0.1 M Tris, pH 7.4. Sulfamidase activity was monitored throughout the process using the established tritiated tetrasaccharide substrate method [21]. Animals Normal/MPS-IIIA male mice were obtained from breeding colonies maintained at the WCH Animal House. Genotyping was carried out according to established methods [4]. Mice were provided with food and water ad libitum and maintained in a constant temperature and humidity environment with a 12 h light and dark cycle (7:00 am to 7:00 pm). Animals were grouphoused where possible, however some MPS-IIIA mice had to be separated due to Wghting from »15 weeks of age. All animal procedures were undertaken in accordance with the National and Medical Research Council of Australia (NHMRC) Code of Practice for the Care and Use of Animals for ScientiWc Purposes, and were approved by the WCH Animal Ethics Committee. Intracerebral injection of sulfamidase Three MPS-IIIA and normal mice received injections of enzyme/vehicle at each of three ages—6, 12, or 18 weeks of age. They were pretreated with acetaminophen (0.16 mg/ml, in drinking water) three days before and after surgery as a mild analgesic. Prior to anesthesia, glycopyrrolate (Robinul, 0.01 mg per mouse, Wyeth Ayerst, Baulkham Hills, NSW, Australia) was administered (intramuscularly) to suppress oral and pharyngeal secretions. A dosing regimen of 170 mg/kg ketamine (Parnell Laboratories, Alexandria, NSW, Australia) and 10 mg/kg xylazine (Xylazil-20, Troy Laboratories, Sydney, Australia) for normal mice and 170 mg/kg ketamine and 20 mg/ kg xylazine for MPS-IIIA mice produced anesthesia of adequate duration without complications. Once mice were unresponsive to a toe pinch, they were secured in a stereotaxic frame (David Kopf Instruments, California, USA). Following resection of the scalp, burr holes were drilled at each of the four sites using a hand drill Wtted

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

with a 0.5 mm drill bit (Flintware, Adelaide, Australia). Mice received bilateral injections into the dentate gyrus of the hippocampus (coordinates in reference to bregma and the midline: 2 mm posterior, 1.5 mm lateral, and 1.5 mm ventral) and the cerebellum (6.3 mm posterior, 1.5 mm lateral, and 2.3 mm ventral) [22]. Two microliters of enzyme or vehicle was infused at rate of 0.5 l/min using 27 G dental needles attached via polyethylene tubing to a syringe pump (World Precision Instruments, Florida, USA). The needles were left in place for 2 min before being slowly withdrawn and the scalp wound was sutured. Processing of tissues for electron microscopy Four mice (one MPS-IIIA + enzyme, one MPS-IIIA + vehicle, one normal + enzyme, and one normal + vehicle) were sacriWced from each treatment group (6, 12 or 18 weeks of age) at 24 weeks of age. They were perfused with 4% (w/v) paraformaldehyde/1.25% (v/v) glutaraldehyde/ 4% (w/v) sucrose in phosphate-buVered saline (PBS) and the brain was removed and immersed in the same Wxative overnight. Blocks (1 mm £ 2 mm £ 2 mm) taken from the dentate gyrus of the hippocampus and cerebellum were washed in 4% (w/v) sucrose in PBS, post-Wxed in 1% aqueous osmium tetroxide followed by dehydration through alcohol and inWltration with 50% propylene oxide/50% Epon–Araldite resin. Blocks were embedded in 100% Epon–Araldite and semi-thin (1 m) survey sections were cut and stained with toluidine blue. After areas of interest were identiWed, resin blocks were trimmed and ultrathin (80 nm) sections cut and stained with uranyl acetate and lead citrate. Sections were visualized with a Phillips CM100 transmission electron microscope. Processing of tissues for light microscopy Two MPS-IIIA mice (+enzyme), two MPS-IIIA mice (+vehicle), and two normal mice (at each treatment time point, i.e., 6, 12, and 18 weeks of age) were sacriWced at 24 weeks of age. Six additional mice (4 MPS-IIIA and 2 normal, all untreated) were sacriWced at 7, 15, 20, and 28 weeks of age to determine the extent of neurodegenerative changes over the course of the disorder. All mice received an overdose with isoXuorane (David Bull Laboratories, Melbourne, Australia) and were perfusion/Wxed with 4% (w/v) paraformaldehyde in PBS. The brain was left in the skull and immersed in Wxative overnight before being divided along the midsagittal Wssure. Brain halves were processed into paraYn and 5 m sagittal sections were taken between the midline and the injection sites using a rotary microtome.

275

(Dako #Z458, Glostrup, Denmark) and GFAP (Dako #Z334). The primary antibodies were polyclonal species raised in rabbit, diluted to 1:1000 in horse serum prior to use. Primary antibodies were labelled with an avidin– biotin peroxidase protocol (Vectorstain ABC kit, Vector Laboratories, California, USA), and sections stained with 3,3-diaminobenzidine tetrahydrochloride (Sigma, Sydney, Australia) followed by counterstaining with haematoxylin. Prior to immunolabelling, endogenous peroxidase activity was blocked by incubation with 1.7% (v/v) hydrogen peroxide in methanol for half an hour. Sections were also pretreated with 3% (v/v) normal horse serum to obscure non-speciWc epitopes.

Results Electron microscopy Wndings in vehicle-treated normal and MPS-IIIA mice SpeciWc cell types were evaluated qualitatively for lysosomal storage in the dentate gyrus and cerebellum. There was no storage pathology in normal mice and there were no apparent diVerences between normal enzyme treated and normal vehicle treated animals. Basic pathology in MPS-IIIA brains was similar to that observed in previous studies [3,10]. We observed individual neurons containing both zebra-body-like inclusions and vacuoles containing Xoccular (granular) storage material, whereas glial cells and peri-neuronal satellite cells displayed large electron-lucent vacuoles. Storage was cell-type speciWc, with cerebellar Purkinje cells tending to display a few large granular mixed-type inclusions, and glia an abundance of smaller vacuoles containing lammellated structures (zebra-bodies, whorls, and stacks). Ultrastructure of the dentate gyrus—eVect of enzyme treatment Lysosomal storage was examined in granule cells and glia that were either perivascular or free. Normal granule cells appeared to possess many electron lucent vesicles that were diVerentiated from storage vacuoles by their greater variability in shape and electron-dense limiting membrane, which at higher magniWcation is decorated on the cytoplasmic side with electron dense puncta (Fig. 1). The vehicle-treated MPS-IIIA mouse showed unexpectedly mild storage in the dentate gyrus. The Wndings are summarized in Table 1. Ultrastructure of the cerebellum—eVect of enzyme treatment

Immunohistochemistry Detection of ubiquitin and glial Wbrillary acidic protein (GFAP) was performed with antibodies to ubiquitin

Storage was evaluated in Purkinje cells and glia (vessel-associated and free). Pathology in both cell types was most severe in the vehicle-treated MPS-IIIA mouse, in

276

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285 Table 1 EVect of sulfamidase treatment at 6, 12 or 18 weeks of age on lysosomal storage material in the dentate gyrus of the hippocampus of MPS-IIIA/normal mice at 24 weeks of age Group

Granule cells

Glial cells

MPS-IIIA vehicle treatment

3–4 moderately sized electron lucent vacuoles 110 moderately sized electron lucent vacuoles 3–6 moderately sized electron lucent vacuoles 02 moderately sized electron lucent vacuoles

Semi-coalescent large to small vacuoles

MPS 6 weeks old at enzyme treatment MPS 12 weeks old at enzyme treatment MPS 18 weeks old at enzyme treatment

Large electron lucent coalescent vacuoles Large electron lucent coalescent vacuoles One moderately sized electron lucent vacuole

contrast to the EM observations in the dentate gyrus. Enzyme therapy appeared to ameliorate storage pathology in both cell types in the 6, 12, and 18 weeks old at treatment mice (Fig. 2E). In glial/perivascular cells, treatment at the latest time point (18 weeks of age) seemed to be most beneWcial with signiWcantly reduced numbers and sizes of vacuoles observed. A similar but subtler eVect was found in Purkinje cells. Cerebellar granule cells were devoid of storage in aVected mice regardless of treatment (not shown). Cerebellar astrocytes showed signs of storage but were not evaluated due to their low number and diYculties in associating vacuoles with one particular cell. The Wndings are summarized in Table 2. Light microscopic Wndings in vehicle-treated normal and MPS-IIIA mice Two vehicle-treated MPS-IIIA and normal mice were used to investigate the neuropathology present in the mouse at 24 weeks of age. We also examined tissue samples from untreated MPS-IIIA and normal mice that had been sacriWced at 7, 15, 20, and 28 weeks of age, to determine the progression of pathological changes over the course of the disorder. Ubiquitin Punctate cytoplasmic staining indicating neurodegeneration and dot-like structures (DLS) consistent with axonal spheroids were observed only rarely in the brains of normal mice. The highest level of immunoreactivity in normal mice was observed throughout the extent of the corpus callosum, but aVected mice showed a similar level of staining. In a few vehicle-treated normal mice, punctate cytoplasmic staining and DLS were associated with

277

the needle track in the hippocampus. Selected areas of MPS-IIIA brain showed ubiquitin positivity (Table 3). In sections near the injection site (»1.5 mm from the midline), the cortex, hippocampus, and cerebellum were almost entirely spared, however the core of the brainstem (pons and medulla) and the superior and inferior colliculi (Fig. 3) displayed the greatest density of dot-like structures. The internal capsule and cerebral peduncle also displayed elevated immunoreactivity. In more medial sections, hippocampus and cerebellum were again devoid of staining. The retrosplenial agranular and granular cortex, however displayed a horizontal band of cells with punctate cytoplasmic ubiquitin immunoreactivity indicative of neurodegeneration. The brainstem and colliculi again showed diVuse involvement. A striking pathological feature in the brainstem was the presence in the gracile and cuneate nuclei/fasciculi of gigantic round inclusions that stained variably with ubiquitin (Fig. 4). Some of these structures showed faint dot-like staining, some possessed incomplete halos of staining, and others stained very little. They were strongly eosinophilic and are presumed to be axonal spheroids. GFAP In 24-week-old vehicle-treated normal and MPS-IIIA brain, GFAP immunoreactivity (a marker of glial cells) was observed in several brain regions, most prominently in white matter tracts such as the corpus callosum and internal capsule. In both MPS-IIIA and normal mice, grey matter regions also displayed a diVuse distribution of glia and a focal increase was occasionally associated with the needle track in both hippocampus and cerebellum, however, as a rule, these structures remained largely immunonegative in both MPS-IIIA and normal mice. The diVerence in GFAP staining between normal and aVected mice was a matter of degree in many areas, however MPS-IIIA brains contained greater numbers of glia in all areas of the cortex, whereas there was little cortical staining in normal mice (Fig. 5). AVected mice also exhibited a proliferation of darkly staining glia in the superior colliculus, periaqueductal gray, lateral habenula, thalamus, and gracile and cuneate nuclei/fasciculi. Cell processes in the latter region formed dense rings around the edges of ubiquitin-positive axonal spheroidal structures (Fig. 4). Light microscopic evaluation of treatment Ubiquitin Enzyme treatment at either 6 or 12 weeks of age appeared to greatly diminish (but not completely elimi-

Fig. 1. Electron micrographs of granule cells (A,C,E,G,I) and perivascular cells (B,D,F,H,J) in the dentate gyrus of the hippocampus of 24-week-old normal mice (A,B), MPS-IIIA mice treated with vehicle (C,D), or MPS-IIIA mice treated with enzyme at 6 weeks (E,F), 12 weeks (G,H) or 18 weeks (I,J) of age. Arrows indicate storage vacuoles.

278

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

279

Table 2 EVect of sulfamidase treatment at 6, 12 or 18 weeks of age on lysosomal storage material in the cerebellum of MPS-IIIA/normal mice at 24 weeks of age Group

Purkinje cells

Glial/perivascular cells

MPS-IIIA vehicle treatment

12 large vacuoles containing mixed granular and lamellated material, 120 moderate to small vacuoles with strongly granular material showing great variability in size 1–2 large vacuoles containing amorphous mixed granular and lamellated material, 10–20 moderate vacuoles with granular material 1–2 large vacuoles containing amorphous mixed granular and lamellated material, 10–20 moderate vacuoles with granular material 1 large vacuole containing amorphous mixed granular and lamellated material found in selected cells, moderate granular vacuoles present in fewer number, commonly 015 per cell

Large electron lucent coalescent vacuoles, containing round inclusions

MPS-IIIA 6 weeks old at enzyme treatment MPS-IIIA 12 weeks old at enzyme treatment MPS-IIIA 18 weeks old at enzyme treatment

Large electron lucent coalescent vacuoles

Large electron lucent coalescent vacuoles

1–2 cells exhibited large electron lucent coalescent vacuoles. Most cells had one moderate vacuole

Table 3 Ubiquitin immunopositivity in murine MPS-IIIA brain at various ages

Brainstem Cuneate nucleus Raphe nucleus Superior colliculus Cerebellum Lateral/medial habenula Mamillary nucleus Thalamus Hypothalamus Motor cortex Retrosplenial cortex Shell of the nucleus accumbens Caudate putamen Substantia nigra/subthalamic nucleus Periaqueductal gray Hippocampus

7 weeks of age

15 weeks of age

20 weeks of age

24 weeks of age

28 weeks of age

+ ++++ ¡/+ + ¡ ¡/+ ¡ ¡ ¡ ¡ ¡ ¡ ND ND ND ¡

ND ND ND ++/+++ ¡ ND ++ ¡ ND ¡ ¡ ND ND ND ++ ¡

++ ND +++ ++++ ¡/+ ++ ++ ¡ ++ + + ++ ¡ ND ND ¡

++ ++++ ++ ++++ ¡ + ++ + ++ + ++ ++ ¡/+ ¡ ++ ¡

+++ ++++ +++ ND ND +++ ++ ¡/+ ND + ND + + ¡ ++ ¡

(¡) no immunopositive cells seen, (+) few immunopositive cells seen, (++) moderate number of immunopositive cells, (+++) many immunopositive cells, (++++) large amount of immunopositive cells, and (ND) not determined.

nate) the neurodegeneration apparent in the 24-weekold untreated MPS-IIIA brain, however the prominent spheroid-like staining in the gracile and cuneate pathways did not diVer between MPS-IIIA mice treated with vehicle or enzyme at 6, 12 or 18 weeks of age. Reduced numbers of ubiquitin-positive DLS were observed in many areas of the treated brain at the time of sacriWce (Table 4), this reduction was most obvious in the brainstem and superior colliculus (Fig. 3). In contrast to the untreated MPS-IIIA mice, neurons in the retrosplenial cortices with punctate cytoplasmic staining were not observed in mice treated at 6 weeks of age. One of the mice treated at 12 weeks of age showed no cytoplasmic staining in these neurons, the other had diminished levels. Ubiquitin immunoreactivity was not signiWcantly reduced in any area of the brain in animals treated at 18 weeks of age.

GFAP The most reliable assessment of the eVect of treatment on the level of GFAP immunoreactivity could be made in the cortex and brainstem, as normal mice lack staining in these areas, whereas MPS-IIIA mice are highly immunopositive for the glial marker. A reduction in GFAP staining, indicative of a reduction in the reactive gliosis seen in untreated MPS-IIIA mouse brain was observed with treatment. However, the eVect of treatment on gliosis was dependent on the time between treatment and sacriWce at 24 weeks of age. There was no diVerence in the pattern of cortical staining between vehicle-treated MPS-IIIA mice and those treated at 6 weeks of age (Fig. 5). In mice treated at 12 weeks of age, however, diminished numbers of glia were observed in the brainstem and the cortex, particular in the superWcial laminae. One of the two mice receiving treatment at 18 weeks of age showed reduced numbers of

Fig. 2. Electron micrographs of purkinje cells (A,C,E,G,I) and perivascular cells (B,D,F,H,J) in the cerebellum of 24-week-old normal mice (A,B), MPS-IIIA mice treated with vehicle (C,D), or MPS-IIIA mice treated with enzyme at 6 weeks (E,F), 12 weeks (G,H) or 18 weeks (I,J) of age. Arrows indicate storage vacuoles.

280

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

Fig. 3. Ubiquitin immunostaining in the superior colliculus at 24 weeks of age in a normal mouse (A), an MPS-IIIA mouse receiving vehicle (B), and MPS-IIIA mice treated with enzyme at 6 weeks (C), 12 weeks (D) or 18 weeks (E) of age. Arrows indicate ubiquitin-positive inclusions, 40£ magniWcation.

Fig. 4. Ubiquitin (A) and glial Wbrillary acidic protein (GFAP, B) immunostaining in the cuneate nucleus at 24 weeks of age in an MPS-IIIA mouse. H&E staining (C) highlights eosinophilic inclusions, 40£ magniWcation.

glia in the cortex. Again, treatment had no eVect on the prominent spheroid-like staining in the gracile and cuneate pathways in MPS-IIIA animals. The hippocampus and

cerebellum, both of which do not display signiWcantly elevated GFAP immunopositivity in untreated MPS-IIIA mice, were unaVected by enzyme treatment at any age.

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

281

Fig. 5. GFAP immunostaining in the cerebral cortex at 24 weeks of age in a normal mouse (A), an MPS-IIIA mouse receiving vehicle (B), and MPSIIIA mice treated with enzyme at 6 weeks (C), 12 weeks (D) or 18 weeks (E) of age. Arrows indicate GFAP positive cells, 40£ magniWcation. Table 4 EVect of intracerebral injection of sulfamidase at 6, 12 or 18 weeks of age on ubiquitin immunoreactivity in MPS-IIIA mouse brain at 24 weeks age Genotype

MPS-IIIA

MPS-IIIA

MPS-IIIA

MPS-IIIA

Normal

Treatment

Enzyme at 6 weeks

Enzyme at 12 weeks

Enzyme at 18 weeks

Vehicle

Vehicle

Brainstem Cuneate nuclei Raphe nuclei Superior colliculus Periaqueductal gray Cerebellum Hippocampus Lateral/medial habenula Thalamus Mamillary nucleus Hypothalamus Motor cortex Retrosplenial cortex Shell of nucleus accumbens Caudate putamen Substantia nigra/subthalamic nucleus

¡/+ ++++ ¡/+ ¡/+ ¡/+ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡

+ ++++ ¡/+ +/++ + ¡ ¡ ¡ ¡ ¡ + ¡ ¡ /+ ¡/+ ¡ ¡

+/++ ++++ + +++ +/++ ¡ ¡ + ¡ + +/++ ¡ ¡ /++ ¡/+ ¡ +

++ ++++ ++ ++++ ++ ¡ ¡ + + ++ ++ + ++ ++ ¡/+ ¡

¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡

(¡) No immunopositive cells seen, (++) moderate number of immunopositive cells, (+++) many immunopositive cells, (++++) large amount of immunopositive cells.

282

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

Discussion The aim of this study was to determine whether supplying recombinant human sulfamidase to the mouse CNS could prevent, delay or reverse the various pathological features of murine MPS-IIIA. Enzyme was delivered directly to the brain parenchyma at various ages (6, 12, and 18 weeks of age—representing early, moderate and advanced disease phenotypes) via intracerebral injection and the eVect of enzyme on neuropathology was determined at 24 weeks of age. We examined the eVect of treatment on two parameters, immunostaining for ubiquitin (a marker of neurodegeneration) and GFAP (a glial marker) in the mouse brain. These markers have been shown to be elevated in other MPS animal models that aVect the brain [23,24], although the eVect of therapy on these markers has not been demonstrated. This is the Wrst report describing the presence of neurodegeneration and gliosis in the murine MPS-IIIA brain, and the response of these pathological changes to enzyme treatment. Recombinant human sulfamidase was injected simultaneously into both the dentate gyrus of the hippocampus and the cerebellum of the MPS-IIIA mouse CNS. These structures demonstrate signiWcant substrate storage in the MPS-IIIA mouse, possibly contributing to the behavioral deWcits displayed by the MPS-IIIA mouse, i.e., reduced memory/learning skills [10] and altered motor control/co-ordination (manuscript in preparation). Intracerebrally injected sulfamidase traveled large distances from the site of injection, as it resulted in widespread eVects on underlying neuropathology. Axonal transport of lysosomal enzyme [25] accompanied by trans-synaptic transfer [26] rather than simple diVusion would be the most likely explanation for the pervasive eVect on pathology. As would be predicted by the Wndings of other researchers [10,16,19,26,27] supplying lysosomal enzyme directly to the MPS-IIIA brain reduced cytoplasmic vacuolation in neurons, glia, and perivascular cells. Following treatment, the reduction in storage in the granule cells of the hippocampus and the Purkinje cells of the cerebellum was greatest in mice receiving enzyme at times closest to sacriWce (i.e., at 4 weeks prior to sacriWce, where treatment was instigated at 18 weeks of age). The amount of stored substances was observed to be greater in mice treated at 6 weeks of age (and sacriWced 18 weeks later) than that observed in mice treated at 12 weeks of age (and sacriWced 12 weeks later). This was most evident in the granule cells of the hippocampal dentate gyrus. It is likely that a re-accumulation of substrates has occurred in mice receiving treatment at 6 and 12 weeks of age. This illustrates the importance of regular enzyme treatment for continued clinical eYcacy. A similar picture was observed in perivascular/glial cells. However, the large amount of stored materials

present at sacriWce (24 weeks of age) in glia/perivascular cells of mice treated at 6 and 12 weeks of age, compared with the ‘normalized’ glia/perivascular cells in mice treated at 18 weeks of age, suggests that storage in these cells re-accumulates rapidly once lysosomal enzyme activity drops below a certain level. A similarly rapid reaccumulation of vacuoles in perivascular cells was observed in MPS-IIIA mice that received neonatal enzyme therapy [10]. Three weeks after receiving treatment, these cells showed no storage, but had begun to reacquire it at 8 weeks from treatment. There were also signiWcant diVerences in the level of immunopositivity for ubiquitin and the glial marker GFAP in the MPS-IIIA and normal mouse brain. A large number of ubiquitin-positive ‘dot-like structures’ consistent with axonal spheroids [28], were observed in sub-cortical areas and brainstem in murine MPS-IIIA. The inability of cells to degrade ubiquitinated proteins may result from either changes in protein substrates such that they are unable to be degraded or a failure within the ubiquitin/proteosomal degradative pathway itself. This indicator of non-speciWc neurodegeneration has also been demonstrated in the MPS-VII mouse [23], sphingolipid knockout mice [29], Niemann–Pick disease type C [30], and in other neurodegenerative disorders, such as Alzheimer’s disease [31]. The MPS-IIIA mouse displayed a markedly diVerent pattern of ubiquitin reactivity to that described in the MPS-VII mouse. In the present study, a large amount of ubiquitin was deposited in the cells of the superior colliculus and brainstem (especially the cuneate/gracile nuclei), these areas were largely unaVected in the MPSVII mouse, with only scattered cells reported in the brainstem [23]. In contrast to the MPS-VII study, we did not observe any stained cells in the hippocampus, at least not before »6 months of age. The pattern of cortical staining was also diVerent in the two models, with the MPS-IIIA mouse demonstrating ubiquitin positivity in the cytoplasm of cells in the secondary motor cortex and the retrosplenial agranular and retrosplenial granular cortices from »20 weeks of age. The MPS-VII mouse, however, showed ubiquitin reactivity in the cerebral cortex lateral to the hippocampus (entorhinal/auditory cortices). Positive staining was also observed in the stria terminalis in MPS-VII and scattered thalamic ubiquitin-positive cells were reported [23]. These were not observed in the MPS-IIIA mouse. Axonal spheroids have been described as the most signiWcant pathological feature in LSDs that aVect the brain [28], and their regional distribution in the MPSIIIA mouse may explain the behavioral dysfunction that has been noted in the model including impaired Morris Water Maze (MWM) performance [10], gait disturbances (Hemsley, personal observations), and aggression (B. Gliddon, personal observations). Many of the aVected sub-cortical areas such as the nucleus accumbens,

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

thalamus, and basal ganglia are important for normal function in the water maze [32], and neurodegeneration in these areas in the MPS-IIIA mouse may explain the memory and learning deWcits observed in this model [10]. The retrosplenial cortex, site of apparent neurodegeneration (but not axonal spheroid formation) in the MPSIIIA mouse brain, also seems to be relevant to water maze function, since rats with lesions in the caudal retrosplenial cortex display mildly impaired acquisition in the MWM [33,34]. The aggressive behavior that is observed both in humans with MPS-IIIA [35] and also in the mouse model may be linked to the neurodegeneration occurring in the hypothalamus and periaqueductal gray. These ubiquitin-positive regions are involved in defensive rage [36]. The superior colliculus, core of the brainstem and gracile and cuneate nuclei seem to be particularly aVected by spheroid-like structure formation in MPSIIIA mice. The superior colliculus has a role in controlling eye movements, an important aspect of visual attention and general visual function [37]. This too could produce impaired MWM performance, since spatial navigation using visual cues is expected to require a high level of visual functioning in itself. Exceptionally large spheroid-like structures were observed as early as 7 weeks of age (the earliest time point studied) in the gracile/cuneate nuclei of the MPS-IIIA mouse and were resistant to enzyme replacement therapy in the present study. These nuclei are involved in the dorsal–medial lemniscal pathway that transmits information regarding tactile sensation and limb proprioception [38]. Dysfunction in this pathway would be expected to produce poor motor coordination particularly due to impaired proprioceptive input. The early appearance of pathological changes in these nuclei correlates temporally with the functional deWcits in gait that have been observed in the MPS-IIIA mouse (K. Hemsley, personal observations). Interestingly, ubiquitin deposition has been linked to a triggering of the inXammatory response [39] and signiWcant gliosis is observed in the MPS-IIIA mouse, often associated with neurodegenerative changes. Whilst the increase in GFAP immunoreactivity in murine MPSIIIA was more subtle than that observed with ubiquitin, MPS-IIIA mice clearly had a greater number of glia overall. This was most evident in the brainstem, superior colliculus and in the superWcial laminae of the cortex, a region in which normal mice show a complete absence of GFAP positivity. The gracile and cuneate nuclei also exhibited increased staining with GFAP. The link between neurodegeneration and gliosis becomes less apparent, however when the eVect of therapy on the MPS-IIIA brain is considered. In mice treated at 12 weeks of age, reduced numbers of glia were observed in the brain 12 weeks later at sacriWce, accompanying the reduction in ubiquitin immunoreactivity.

283

However, in mice treated at 6 weeks of age, despite the near complete absence of neurodegenerative changes, the gliosis had returned in the 18-week period from treatment to sacriWce. We hypothesize that the pathological gliosis was improved by enzyme replacement therapy, but returned fairly quickly once the enzyme activity lapsed. This is supported by the observation that treatment at 18 weeks of age (and sacriWce at 24 weeks of age) resulted in a near complete elimination of reactive glia in one of the mice. The extent and pattern of gliosis in MPS-IIIA (and in the response to therapy) may actually correlate better with substrate storage, rather than neurodegeneration in the MPS-IIIA mouse. This would explain the extensive gliosis in the cerebral cortices, which presumably accompanies the widespread substrate storage observed in this region [10]. Increased gliosis (astrocytosis), particularly in the cortex, was observed in the MPS-IIIB mouse brain [24], presumably in response to neuronal damage and death. The inXammatory process of which astrocytosis is the terminal stage has been implicated as an important pathological process in MPS-IIIB [40] and Tay Sachs and SandhoV diseases [41]. A model for the pathophysiology of the GM2 gangliosidoses involving microglial and astrocytic activation has been proposed [41]. Astrocytic activation occurs in response to neuronal damage or as a result of accumulation of storage materials by these cells. This results in increased expression of inXammatory cytokines and other proteins, further increasing the level of neuronal damage, ultimately resulting in cell death. A similar mechanism may be occurring in MPS-IIIA in response to stored substrates. It is interesting to note that there was less of a correlation between the areas of the brain that store large amounts of GAG (e.g., cerebral/cerebellar cortices) [3], and those areas displaying neurodegeneration indicated by ubiquitin reactivity. This has also been observed in the MPS-VII mouse [23]. The clinical presentation of neurological disease may actually correlate better with the neurodegenerative changes than with substrate storage (however, see [42]). It has been postulated that neurodegeneration may result from structural alterations in the diseased brain and given the importance of heparan sulfate (the primary stored substrate in MPS-IIIA) in the development of the CNS, subtle yet critical developmental alterations may well be present [23]. In MPS-IIIA mice treated at 6 and 12 weeks of age, enzyme treatment delayed the appearance of neurodegenerative change that appears in the untreated MPSIIIA retrosplenial cortex between 15 and 20 weeks of age. In addition, the formation of spheroid-like bodies in the colliculi that is observed from 7 to 15 weeks of age in the untreated MPS-IIIA mouse was also delayed by treatment. However, the large spheroid-like structures seen in the gracile and cuneate nuclei at 7 weeks of age in

284

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285

the untreated mouse were not improved with enzyme replacement. The formation of spheroids in these structures has occurred earlier than the spheroid formation/ degenerative changes in other brain regions, e.g., superior colliculus, cortex, mamillary nucleus, and periaqueductal gray, which appear between 7 and 15 weeks of age onwards. That the early-forming spheroid-like structures in the gracile and cuneate nuclei were unaVected by enzyme treatment at any age, indicates their irreversibility. Where MPS-IIIA mice received treatment at 18 weeks of age, the eVect of treatment on ubiquitin-positive neuropathology at 6 weeks post-injection was mild and there was no change in spheroidal staining in the cuneate/gracile nuclei, thus adding some weight to the hypothesis that the degenerative changes take more than 4 weeks to resolve, and that axonal spheroids, once formed are irreversible [43]. The hypothesis was formed on the basis of data from an inducible model of -mannosidosis in the guinea pig, where axonal spheroids along with ectopic dendrites persisted two years after restoration of normal enzyme levels, suggesting that spheroids, once formed, cannot be cleared. The present study utilized a single application of lysosomal enzyme to the CNS and the eVect of repeated or on-going enzyme replacement on neurodegenerative change/ axonal spheroids may actually be quite diVerent. Near complete clearance of cortical/hippocampal dot-like structures (indicative of neurodegeneration) and a reduction in axonal spheroids in the stria terminalis was observed following gene therapy in MPS-VII mice [23]. Studies are planned to examine the eVect of long-term enzyme-replacement therapy for MPS-IIIA on neurodegeneration, axonal spheroid formation, substrate storage, inXammation, and behavioral changes. This will allow us to determine whether the axonal spheroids observed in the cuneate/gracile nuclei at 7 weeks of age in the MPS-IIIA mouse contribute to the functional deWcits observed in this model and if so, whether they are able to be reversed. In summary, this study aimed to investigate whether the neuropathology present in the MPS-IIIA mouse could be delayed, reversed, or prevented with direct injection of enzyme into the cerebrum. Data regarding the time of onset of neuropathological changes in various region of the untreated murine MPS-IIIA brain and their susceptibility to treatment suggest that enzyme administered to SanWlippo type A mice is able to delay the appearance of neurodegenerative changes when it is administered to younger mice (6–12 weeks of age) and reduce the reactive gliosis and substrate storage observed, when administered at 6, 12 or 18 weeks of age, but is unable to ameliorate axonal spheroid-like structures present prior to treatment onset. The Wndings of this study thus illustrate the importance of early detection and timely instigation of treatment in preventing the

development of irreversible pathological (and potentially functional) changes in the MPS-IIIA brain.

Acknowledgments This work was supported by funding from the National Health and Medical Research Council (NHMRC) of Australia (J.J.H). We wish to thank Dr. Peter Clements and Ms. Liz Melville for preparing the sulfamidase. We thank Dr. Jim Manavis for provision of and assistance with the immunohistochemical methods, we appreciate the help of Ms. Lynn Waterhouse with the EM studies and thank Dr. Briony Gliddon for providing us with some of the MPS-IIIA tissue sections. Further, we thank Drs. Dyane Auclair and Allison Crawley for their comments on the manuscript.

References [1] P.J. Meikle, J.J. Hopwood, A.E. Clague, W.F. Carey, Prevalence of lysosomal storage disorders, JAMA 281 (1999) 249–254. [2] J.J. Hopwood, C.P. Morris, The mucopolysaccharidoses. Diagnosis, molecular genetics and treatment, Mol. Biol. Med. 7 (1990) 381–404. [3] M. Bhaumik, V.J. Muller, T. Rozaklis, L. Johnson, K. Dobrenis, R. Bhattacharyya, S. Wurzelmann, P. Finamore, J.J. Hopwood, S.U. Walkley, P. Stanley, A mouse model for mucopolysaccharidosis type III A (SanWlippo syndrome), Glycobiology 9 (1999) 1389–1396. [4] R. Bhattacharyya, B. Gliddon, T. Beccari, J.J. Hopwood, P. Stanley, A novel missense mutation in lysosomal sulfamidase is the basis of MPS III A in a spontaneous mouse mutant, Glycobiology 11 (2001) 99–103. [5] S. Byers, T. Rozaklis, L.K. BrumWeld, E. Ranieri, J.J. Hopwood, Glycosaminoglycan accumulation and excretion in the mucopolysaccharidoses: characterization and basis of a diagnostic test for MPS, Mol. Genet. Metab. 65 (1998) 282–290. [6] J.J. Hopwood, J.R. Harrison, High-resolution electrophoresis of urinary glycosaminoglycans: an improved screening test for the mucopolysaccharidoses, Anal. Biochem. 119 (1982) 120–127. [7] A. Fischer, K.P. Carmichael, J.F. Munnell, P. Jhabvala, J.N. Thompson, R. Matalon, P.F. Jezyk, P. Wang, U. Giger, Sulfamidase deWciency in a family of Dachshunds: a canine model of mucopolysaccharidosis IIIA (SanWlippo A), Pediatr. Res. 44 (1998) 74–82. [8] R.D. Jolly, F.J. Allan, M.G. Collett, T. Rozaklis, V.J. Muller, J.J. Hopwood, Mucopolysaccharidosis IIIA (SanWlippo syndrome) in a New Zealand Huntaway dog with ataxia, N. Z. Vet. J. 49 (2000) 144–148. [9] M.Z. Jones, J. Alroy, P.J. Boyer, K.T. Cavanagh, K. Johnson, D. Gage, J. Vorro, J.A. Render, R.S. Common, R.A. Leedle, C. Lowrie, P. Sharp, S.S. Liour, B. Levene, H. Hoard, R. Lucas, J.J. Hopwood, Caprine mucopolysaccharidosis-IIID: clinical, biochemical, morphological and immunohistochemical characteristics, J. Neuropathol. Exp. Neurol. 57 (1998) 148–157. [10] B.L. Gliddon, J.J. Hopwood, Enzyme replacement therapy in a murine model of mucopolysaccharidosis type IIIA, Pediatr. Res. 56 (2004) 1–8. [11] E.F. Neufeld, J. Muenzer, The mucopolysaccharidoses, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), Metabolic Basis of Inherited Disease, seventh ed., McGraw-Hill, New York, 1995, pp. 2465–2494.

P.S. Savas et al. / Molecular Genetics and Metabolism 82 (2004) 273–285 [12] G. Altarescu, S. Hill, E. Wiggs, N. JeVries, C. Kreps, C.C. Parker, R.O. Brady, N.W. Barton, R. SchiVmann, The eYcacy of enzyme replacement therapy in patients with chronic neuronopathic Gaucher’s disease, J. Pediatr. 138 (2001) 539–547. [13] R. SchiVmann, R.O. Brady, New prospects for the treatment of lysosomal storage diseases, Drugs 62 (2002) 733–742. [14] E.G. Shapiro, L.A. Lockman, M. Balthazor, W. Krivit, Neuropsychological outcomes of several storage diseases with and without bone marrow transplantation, J. Inherit. Metab. Dis. 18 (1995) 413–429. [15] J.E. Barker, S. Deveau, M. Lessard, N. Hamblen, C. Vogler, B. Levy, In utero fetal liver cell transplantation without toxic irradiation alleviates lysosomal storage in mice with mucopolysaccharidosis type VII, Blood Cells Mol. Dis. 27 (2001) 861–873. [16] H.K. Jin, J.E. Carter, G.W. Huntley, E.H. Schuchman, Intracerebral transplantation of mesenchymal stem cells into acid sphingomyelinase-deWcient mice delays the onset of neurological abnormalities and extends their life span, J. Clin. Invest. 109 (2002) 1183–1191. [17] A.R. Sawkar, W.C. Cheng, E. Beutler, C.H. Wong, W.E. Balch, J.W. Kelly, Chemical chaperones increase the cellular activity of N370S -glucosidase: a therapeutic strategy for Gaucher disease, Proc. Natl. Acad. Sci. USA 99 (2002) 15428–15433. [18] K.P. Ponder, J.R. Melniczek, X. Lu, M.A. Weil, T.M. O’Malley, P.A. O’Donnell, V.W. Knox, G.D. Aguirre, H. Mazrier, N.M. Ellinwood, M. Sleeper, A.M. Maguire, S.W. Volk, R.L. Mango, J. Zweigle, J.H. Wolfe, M.E. Haskins, Therapeutic neonatal hepatic gene therapy in mucopolysaccharidosis VII dogs, Proc. Natl. Acad. Sci. USA 99 (2002) 13102–13107. [19] A.I. Brooks, C.S. Stein, S.M. Hughes, J. Heth, P.M. McCray Jr., S.L. Sauter, J.C. Johnston, D.A. Cory-Slechta, H.J. FederoV, B.L. Davidson, Functional correction of established central nervous system deWcits in an animal model of lysosomal storage disease with feline immunodeWciency virus-based vectors, Proc. Natl. Acad. Sci. USA 99 (2002) 6216–6221. [20] J. Bielicki, J.J. Hopwood, E.L. Melville, D.S. Anson, Recombinant human sulphamidase: expression, ampliWcation, puriWcation and characterization, Biochem. J. 329 (Part 1) (1998) 145–150. [21] J.J. Hopwood, H. Elliott, Diagnosis of SanWlippo type A syndrome by estimation of sulfamidase activity using a radiolabelled tetrasaccharide substrate, Clin. Chim. Acta 123 (1982) 241–250. [22] G. Paxinos, K.B.J. Franklin, The Mouse Brain in Stereotaxic Coordinates, Academic Press, California, 2001. [23] G.G. Heuer, M.A. Passini, K. Jiang, M.K. Parente, V.M. Lee, J.Q. Trojanowski, J.H. Wolfe, Selective neurodegeneration in murine mucopolysaccharidosis VII is progressive and reversible, Ann. Neurol. 52 (2002) 762–770. [24] H.H. Li, H.Z. Zhao, E.F. Neufeld, Y. Cai, F. Gomez-Pinilla, Attenuated plasticity in neurons and astrocytes in the mouse model of SanWlippo syndrome type B, J. Neurosci. Res. 69 (2002) 30–38. [25] M.A. Passini, E.B. Lee, G.G. Heuer, J.H. Wolfe, Distribution of a lysosomal enzyme in the adult brain by axonal transport and by cells of the rostral migratory stream, J. Neurosci. 22 (2002) 6437–6446. [26] A.K. Hennig, B. Levy, J.M. Ogilvie, C.A. Vogler, N. Galvin, S. Bassnett, M.S. Sands, Intravitreal gene therapy reduces lysosomal storage in speciWc areas of the CNS in mucopolysaccharidosis VII mice, J. Neurosci. 23 (2003) 3302–3307. [27] W.A. Frisella, L.H. O’Connor, C.A. Vogler, M. Roberts, S. Walkley, B. Levy, T.M. Daly, M.S. Sands, Intracranial injection of

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

285

recombinant adeno-associated virus improves cognitive function in a murine model of mucopolysaccharidosis type VII, Mol. Ther. 3 (2001) 351–358. S.U. Walkley, H.J. Baker, M.C. Rattazzi, M.E. Haskins, J.Y. Wu, Neuroaxonal dystrophy in neuronal storage disorders: evidence for major GABAergic neuron involvement, J. Neurol. Sci. 104 (1991) 1–8. Y. Oya, H. Nakayasu, N. Fujita, K. Suzuki, K. Suzuki, Pathological study of mice with total deWciency of sphingolipid activator proteins (SAP knockout mice), Acta Neuropathol. (Berl.) 96 (1998) 29–40. Y. Higashi, S. Murayama, P.G. Pentchev, K. Suzuki, Cerebellar degeneration in the Niemann–Pick type C mouse, Acta Neuropathol. (Berl.) 85 (1993) 175–184. D.W. Dickson, A. Wertkin, Y. Kress, H. Ksiezak-Reding, S.H. Yen, Ubiquitin immunoreactive structures in normal human brains, distribution and developmental aspects, Lab. Invest. 63 (1990) 87–99. R. D’Hooge, P.P. De Deyn, Applications of the Morris water maze in the study of learning and memory, Brain Res. Brain Res. Rev. 36 (2001) 60–90. S.D. Vann, L.A. Kristina Wilton, J.L. Muir, J.P. Aggleton, Testing the importance of the caudal retrosplenial cortex for spatial memory in rats, Behav. Brain Res. 140 (2003) 107–118. K.T. Harker, I.Q. Whishaw, Impaired spatial performance in rats with retrosplenial lesions: importance of the spatial problem and the rat strain in identifying lesion eVects in a swimming pool, J. Neurosci. 22 (2002) 1155–1164. F.D. NidiVer, T.E. Kelly, Developmental and degenerative patterns associated with cognitive, behavioural and motor diYculties in the SanWlippo syndrome: an epidemiological study, J. Ment. DeWc. Res. 27 (Part 3) (1983) 185–203. T.R. Gregg, A. Siegel, Brain structures and neurotransmitters regulating aggression in cats: implications for human aggression, Prog. Neuropsychopharmacol. Biol. Psychiatry 25 (2001) 91–140. M.E. Goldberg, The control of gaze, in: E.R. Kandel, J.H. Schwartz, T.M. Jessel (Eds.), Principles of Neural Science, McGrawHill, New York, 2000, p. 782. E.P. Gardner, J.H. Martin, T.M. Jessell, The bodily senses, in: E.R. Kandel, J.H. Schwartz, T.M. Jessel (Eds.), Principles of Neural Science, McGraw-Hill, New York, 2000, p. 430. Z. Li, M. Jansen, S.-R. Pierre, M.E. Figueiredo-Pereira, Neurodegeneration: linking ubiquitin/proteasome pathway impairment with inXammation, Int. J. Biochem. Cell Biol. 35 (2003) 547–552. K. Ohmi, D.S. Greenberg, K.S. Rajavel, S. Ryazantsev, H.H. Li, E.F. Neufeld, Activated microglia in cortex of mouse models of mucopolysaccharidoses I and IIIB, Proc. Natl. Acad. Sci. USA 100 (2003) 1902–1907. R. Myerowitz, D. Lawson, H. Mizukami, Y. Mi, C.J. TiVt, R.L. Proia, Molecular pathophysiology in Tay–Sachs and SandhoV diseases as revealed by gene expression proWling, Hum. Mol. Genet. 11 (2002) 1343–1350. S.S. Zhan, K. Beyreuther, H.P. Schmitt, Neuronal ubiquitin and neuroWlament expression in diVerent lysosomal storage disorders, Clin. Neuropath. 11 (1992) 251–255. S.U. Walkley, S. Wurzelmann, D.A. Siegel, Ectopic axon hillockassociated neurite growth is maintained in metabolically reversed swainsonine-induced neuronal storage disease, Brain Res. 410 (1987) 89–96.