Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy

YMGME-05908; No. of pages: 8; 4C: Molecular Genetics and Metabolism xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme

Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy☆ Heather L. Gray-Edwards a,⁎, Brandon L. Brunson b, Merrilee Holland c, Adrien-Maxence Hespel c, Allison M. Bradbury a,b,1, Victoria J. McCurdy a,b,2, Patricia M. Beadlescomb a,b, Ashley N. Randle a, Nouha Salibi d,e, Thomas S. Denney e,f, Ronald J. Beyers e, Aime K. Johnson c, Meredith L. Voyles c, Ronald D. Montgomery c, Diane U. Wilson a,c, Judith A. Hudson c, Nancy R. Cox a,g, Henry J. Baker a,g, Miguel Sena-Esteves h, Douglas R. Martin a,b a

Scott-Ritchey Research Center, College of Veterinary Medicine, Auburn University, Auburn, AL, USA Department of Anatomy, Physiology & Pharmacology, College of Veterinary Medicine, Auburn University, Auburn, AL, USA Department of Clinical Sciences, College of Veterinary Medicine, Auburn University, Auburn, AL, USA d MR R&D Siemens Healthcare, Malvern, PA, USA e Auburn University MRI Research Center, Auburn, AL, USA f Department of Electrical and Computer Engineering, Auburn University, Auburn, AL, USA g Department of Pathobiology, College of Veterinary Medicine, Auburn University, Auburn, AL, USA h Department of Neurology and Gene Therapy Center, University of Massachusetts Medical School, Worcester, MA, USA b c

a r t i c l e

i n f o

Article history: Received 23 March 2015 Received in revised form 6 May 2015 Accepted 6 May 2015 Available online xxxx Keywords: Sandhoff disease Lysosomal storage disease Gene therapy Bone Heart GM2 gangliosidosis

a b s t r a c t Sandhoff disease (SD) is a fatal neurodegenerative disease caused by a mutation in the enzyme β-Nacetylhexosaminidase. Children with infantile onset SD develop seizures, loss of motor tone and swallowing problems, eventually reaching a vegetative state with death typically by 4 years of age. Other symptoms include vertebral gibbus and cardiac abnormalities strikingly similar to those of the mucopolysaccharidoses. Isolated fibroblasts from SD patients have impaired catabolism of glycosaminoglycans (GAGs). To evaluate mucopolysaccharidosis-like features of the feline SD model, we utilized radiography, MRI, echocardiography, histopathology and GAG quantification of both central nervous system and peripheral tissues/fluids. The feline SD model exhibits cardiac valvular and structural abnormalities, skeletal changes and spinal cord compression that are consistent with accumulation of GAGs, but are much less prominent than the severe neurologic disease that defines the humane endpoint (4.5 ± 0.5 months). Sixteen weeks after intracranial AAV gene therapy, GAG storage was cleared in the SD cat cerebral cortex and liver, but not in the heart, lung, skeletal muscle, kidney, spleen, pancreas, small intestine, skin, or urine. GAG storage worsens with time and therefore may become a significant source of pathology in humans whose lives are substantially lengthened by gene therapy or other novel treatments for the primary, neurologic disease. © 2015 Published by Elsevier Inc.

1. Introduction

☆ Compliance with Ethics Guidelines: Heather L Gray-Edwards, Brandon L Brunson, Merrilee Holland, Adrien-Maxence Hespel, Allison M Bradbury, Victoria J McCurdy, Patricia M Beadlescomb, Ashley N Randle, Thomas S. Denney, Ronald J Beyers, Aime K Johnson, Ronald D Montgomery, Diane U Wilson, Judith A Hudson, Nancy R Cox, Henry J Baker, Miguel Sena-Esteves and Douglas R Martin declare that they have no conflict of interest.Nouha Salibi is an employee of Siemens Healthcare. ⁎ Corresponding author at: Auburn University College of Veterinary Medicine, ScottRitchey Research Center, 1265 HC Morgan Drive, Auburn University, AL 36849, USA. E-mail address: [email protected] (H.L. Gray-Edwards). 1 Allison Bradbury, current address: Department of Clinical Sciences, University of Pennsylvania School of Veterinary Medicine, USA. 2 Victoria McCurdy, current address: Department of Biological Sciences, Mississippi State University, USA.

Tay–Sachs and Sandhoff diseases are progressive neurologic diseases collectively known as the GM2 gangliosidoses, which can be classified into three clinical forms: infantile, juvenile and adult onset. The infantile form is typically diagnosed at the age of 13 months with a failure to achieve, or loss of previously acquired, developmental milestones. Rapid neurodegeneration causes seizures, loss of motor tone, swallowing difficulties and eventually a vegetative state, with death by 4 years of age [3]. The GM2 gangliosidoses result from a deficiency of the enzyme β-Nacetylhexosaminidase (Hex; EC 3.2.1.52), which exists as distinct functional isozymes including Hex A, a heterodimer of α and β subunits, and Hex B, a homodimer of β subunits. Hex α-subunit deficiency causes a

http://dx.doi.org/10.1016/j.ymgme.2015.05.003 1096-7192/© 2015 Published by Elsevier Inc.

Please cite this article as: H.L. Gray-Edwards, et al., Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy, Mol. Genet. Metab. (2015), http://dx.doi.org/10.1016/j.ymgme.2015.05.003

2

H.L. Gray-Edwards et al. / Molecular Genetics and Metabolism xxx (2015) xxx–xxx

non-functional Hex A protein resulting in Tay–Sachs disease (TSD), while Hex β-subunit deficiency affects both Hex A and Hex B, producing Sandhoff disease (SD). In humans, Hex A (αβ) cleaves the terminal Nacetyl galactosamine residue from charged and uncharged substrates and is solely responsible for the hydrolysis of GM2 ganglioside. Since GM2 ganglioside accumulation is most detrimental in neurons, the primary clinical phenotype of both TSD and SD is neurodegenerative. β-N-acetylhexosaminidase is capable of cleaving uncharged substrates, including asialo GM2, globoside, oligosaccharides and glycosaminoglycans (GAGs) in most tissues [20,38]. Human Hex A specifically cleaves some GAGs [30,50], and murine Hex B is also thought to break down GAGs, which do not accumulate in the presence of normal or supranormal levels of Hex B in the Tay–Sachs mouse model [54]. In the absence of all three isozymes (Hex A, Hex B and Hex S), the Hex α/β double knockout mouse stores copious amounts of GAGs, and consequently Hex S must also play an important role in keeping GAG levels down [26,43,54]. In conclusion, all three Hex isozymes likely contribute to the metabolism of GAGs. GAGs are the primary storage product in a related group of disorders known as mucopolysaccharidoses (MPS) [38], which cause deformities of bone and connective tissue, including heart valves. The functional redundancy of the Hex isozymes may explain partially why humans with GM2 gangliosidosis do not universally develop MPS-like pathology. However, some GM2 patients do develop vertebral gibbus deformities and cardiac abnormalities that closely resemble those observed in MPS patients [4,56]. Also, alterations of GAG levels were noted in one Tay–Sachs patient with elevations of heparan sulfate in urine and increased levels of dermatan sulfate in the liver, though total GAG concentrations were within normal limits [55]. Impaired catabolism of GAGs has been reported in cultured fibroblasts from SD humans [7] and mice [43] as well. MPS-like pathology is present in several other lysosomal storage disorders including Gaucher disease [61], Krabbe disease [10], metachromatic leukodystrophy [39] and α-mannosidosis [33]. The feline SD model, first reported in 1977, exhibits clinical, pathological and biochemical features similar to the infantile-onset form of human SD [1,11,12]. Recent preclinical results with gene therapy for the central nervous system (CNS) component of feline SD, support translation of an AAV-mediated approach to human clinical trials [36, 42]. CNS therapy has been remarkably successful, with profound attenuation of neurologic signs 16 weeks after treatment [36,42]. However, the CNS-targeted approach led to incomplete correction of extraneural tissues and the development of severe peripheral disease. Here we report an MPS-like pathology in the feline SD model and suggest it is a contributing mechanism of peripheral disease, which may not be adequately addressed by standard CNS-directed gene therapy approaches. The mechanistic basis of emergent peripheral disease is a crucial consideration for human clinical trials, currently in planning.

intracerebroventricular (ICV) injection. Surgeries were performed at ~ 1 month of age as previously described [5,19,42]. Total vector doses were as follows: Thal, ~ 4.5 × 1011 vg distributed equally between right and left thalami; DCN, ~1.6 × 1011 vg distributed equally between right and left DCN; ICV, ~6.4 × 1011 vg injected into the left lateral ventricle. At humane endpoint (4.5 ± 0.5 months in untreated SD cats) or a predetermined endpoint of 16 weeks after treatment, animals were euthanized by pentobarbital overdose (100 mg/kg) in accordance with the American Veterinary Medical Association guidelines. After cats were transcardially perfused with cold, heparinized saline, tissue samples were stored in neutral buffered formalin or flash frozen in liquid nitrogen and stored at −80 °C for analysis. 2.3. Imaging In untreated SD cats at humane endpoint (~4.5 months) and in agematched normal controls, lateral and ventrodorsal radiographs were performed using a Diagnost Radiography and Fluoroscopy unit (Philips Healthcare, Eindhoven, Netherlands) and MRIs were performed as previously described [21]. All MRIs were performed on a 3 T MAGNETOM Verio scanner (Siemens Healthcare, Erlangen, Germany) using an 8 channel human wrist coil (In vivo Corp, Gainesville, FL, USA). Anatomical 2D transverse T2w turbo spin echo (TSE) images were obtained with 0.3 × 0.3 × 1 mm3 resolution, TR/TE of 4630/107 ms, echo train length of 9, BW of 243 Hz/pix, and six averages. Echocardiograms were performed with a Phillips IE33 (Philips Healthcare, Andover, MA) using a P7-4 or C8-5 ultrasound probe in non-sedated, untreated SD cats at 3 months and humane endpoint. Age-matched normal controls were included. Measurements of the diameter of the aorta and left atria were made using a right parasternal short axis view. 2.4. Glycosaminoglycan (GAG) measurement Urine was frozen at −80 °C for later analysis. Upon thawing, samples were centrifuged at 4000× g, and urine creatinine was measured using a Cobas C311 chemistry analyzer (Roche Hitachi, Basel, Switzerland & Tokyo, Japan). Urinary GAG concentration was measured as previously described [40]. Tissue GAG measurement was performed as previously described [16]. 2.5. Histopathology Paraffin embedded tissue sections were stained with hematoxylin and eosin (H & E), Masson's trichrome [32] and alcian blue [46]. Alcian blue staining was performed using 1% alcian blue (8Gx Sigma, A5286) solution containing 3% acetic acid at pH 2.5 for 60 min. Sections were counterstained with 0.1% nuclear fast red (Kernechtrot, Sigma) solution containing 0.146 M aluminum sulfate (Sigma).

2. Materials and methods 2.6. Statistical analysis 2.1. AAV vectors Feline cDNAs for the Hex α and β subunits were cloned into separate, monocistronic AAV vector backbones as previously described [5]. Feline Hex transgene expression is driven by the hybrid CBA promoter including the cytomegalovirus immediate-early enhancer fused to the chicken β-actin promoter, and vectors include the woodchuck hepatitis virus post-transcriptional regulatory element (WPRE). AAVrh8 vectors were injected at a 1:1 ratio of α:β subunits. 2.2. Animals and surgery All animal procedures were approved by the Auburn University Institutional Animal Care and Use Committee. Treatments consisted either of bilateral injection of the thalamus (Thal) and deep cerebellar nuclei (DCN), or bilateral Thal injection combined with unilateral

Data are expressed as mean ± standard deviation and error bars represent standard deviation. Statistics were performed using a twotailed t-test assuming unequal variances (Microsoft Excel 2013, Redmond, Washington) and a P ≤ 0.05 was considered significant for all comparisons. No data were excluded from the analysis because outliers are hard to determine with small samples sizes. 3. Results The SD cat exhibits coarse facial features with a flattened skull, blunted nose, wide placed eyes and corneal clouding (Fig. 1). Radiographic analysis reveals diffuse osteopenia of the SD cat skeleton with the spinal column most severely affected (Fig. 2). The intervertebral disc spaces are widened throughout the length of the spine with malformation of most vertebrae (Fig. 2B & D). Cervical vertebral bodies are

Please cite this article as: H.L. Gray-Edwards, et al., Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy, Mol. Genet. Metab. (2015), http://dx.doi.org/10.1016/j.ymgme.2015.05.003

H.L. Gray-Edwards et al. / Molecular Genetics and Metabolism xxx (2015) xxx–xxx

3

Fig. 1. Physical appearance of the SD cat. (A) Normal feline facial features compared to the face of the untreated SD cat (B). The SD cat exhibits coarse facial features including a shortened muzzle, flattened skull and smaller ears. A clear normal feline cornea (C) is compared to the opaque SD cat cornea (D). The SD cat also may exhibit protrusion of the third eyelid (B and D).

severely narrowed (~5 mm; Fig. 2 D) in cranial to caudal length, measuring approximately one-half that of normal cats (9–10 mm, Fig. 2 C). There is also moderate narrowing of the thoracic vertebrae and mild narrowing of the lumbar vertebrae (not shown). The cranial and caudal epiphyses of the cervical and cranial thoracic vertebrae are widened, resulting in incomplete coverage of the metaphyses (not shown). On the appendicular skeleton, there is mild shortening of the long bones, flattening and poor coverage of the epiphyses (Fig. 2F). Multiple punctate radiolucent defects were noted in the epiphyses of the humerus, distal radius, femur, and tibia. Severe bilateral coxofemoral changes resulting in luxation and severe subluxation of the coxofemoral joints were noted (Fig. 2F). This was associated with mild varus of both tibiae.

Bilaterally, the joint space of the elbow was mildly widened laterally and there was mild subluxation (data not shown). MRI was utilized to determine the impact of cervical vertebral changes on the spinal cord. The spinal cord of the SD cat was consistently compressed both ventrally and dorsally at the level of C1–C2 (Fig. 2H). This resulted in complete attenuation of the hyperintense signal associated with epidural fat and CSF at the level of compression on T2-weighted sequences. Intensity changes of the gray and white matter of the brain and cerebellum were consistent with Sandhoff disease [47]. Unlike the linear appearance of the normal rib physis (Fig. 3A), untreated SD cats have a bowed physeal structure (Fig. 3B) with decreased

Fig. 2. Radiological findings in the SD cat The SD cat exhibits multiple bony abnormalities throughout the cervical spine (B & D) as compared to normal (A & C). The SD cat has severe bilateral coxofemoral changes (F) compared to normal (E). T2 weighted sagittal MRI shows spinal cord compression at the level of C1 (arrow) in the untreated SD cat (H) as compared to the normal cat (G).

Please cite this article as: H.L. Gray-Edwards, et al., Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy, Mol. Genet. Metab. (2015), http://dx.doi.org/10.1016/j.ymgme.2015.05.003

4

H.L. Gray-Edwards et al. / Molecular Genetics and Metabolism xxx (2015) xxx–xxx

Fig. 3. Histomorphologic comparison of bone in untreated SD and normal cats. The rib physis is linear in normal cats (A) but has a pronounced bow in untreated SD cats (B). Higher numbers of chondrocytes populate the growth plate in the normal (C) versus the SD cat (D). The normal cat humerus (E) has numerous, well developed trabeculae compared to the SD cat (F). Osteoblasts (arrows) are reduced in number and sporadically line the SD bone trabeculae (F). Compared to normal (G), chondrocytes in the SD cat (H) exhibit clear vacuolization of the cytoplasm consistent with storage material. Scale bars: A & B = 500 μM; C & D = 25 μM; E & F = 50 μM; G & H = 10 μM. H & E stain used in A–H.

numbers of chondrocytes in the zones of proliferation, hypertrophy and maturation (Fig. 3D). Bony changes are also observed in the humerus, where trabeculae are small, fewer in number, and contain a primarily cartilaginous core with a thin rim of osteoid matrix (Fig. 3F), consistent with osteopenia. Few, sporadically placed osteoblasts line the trabeculae in the SD cat. Chondrocytes exhibit clear vacuolation of the cytoplasm, consistent with storage material (Fig. 3H). Grossly, the cartilage of the SD cat is thin upon dissection with multiple erosive lesions (data not shown). Because cardiac change is common with GAG storage, echocardiograms were performed to assess heart function. The SD cat aortic valve is abnormally shaped and enlarged (Fig. 4B). Compared to the nearly transparent aortic valve leaflets in normal cats (Fig. 4A), SD cats have leaflets that are hyperechoic, which is consistent with leaflet thickening. The SD cat also exhibits aortic enlargement (Fig. 4D), which is represented as an aortic (AO) size relative to the left atrium (LA), with an AO to LA ratio of 1.03, compared to the normal (Fig. 4C) AO to LA ratio of 0.75. At 2 months of age, the SD cat has a normal sized aorta (Fig. 4E) with an AO to LA ratio of 0.75, but by 4 months the same SD cat has a similarly sized aorta and left atria (Fig. 4F), with a ratio of 0.94. Electrocardiograms were within normal limits (data not shown). Histologic examination of hearts from SD cats revealed irregularity and expansion of great vessel leaflets (Fig. 4H) by large fibroblasts with clear to lightly eosinophilic vacuoles (Fig. 4J). Vacuoles contained punctate alcianophilic staining

indicating GAG storage (Fig. 4L). Additionally, the extracellular matrix of the valvular stroma of SD cats appeared to have a higher density of alcianophilic staining compared to normal, indicating an increased presence of GAGs extracellularly (Fig. 4L). Trichrome stains revealed no net increase in collagen deposition (data not shown). AAV gene therapy was performed on three cats by Thal + DCN injection and four cats by Thal + ICV injection with a predetermined endpoint of 16 weeks post-treatment. To evaluate the effect of AAV gene therapy on GAG storage, GAG content was measured in the liver, kidney, muscle, spleen, heart, pancreas, small intestine, brain, skin and lung (Fig. 5). Untreated SD cats had significant increases in GAG content in all tissues measured. Intracranial AAV gene therapy reduced GAG content to normal levels in the cerebral cortex, and partially reduced GAG levels in the liver. AAV gene therapy failed to significantly reduce GAG content in the remaining peripheral tissues; however, reductions trend toward significance in the kidney, lung, heart and skin in both treatment groups, the pancreas for the Thal + DCN cohort and the small intestine for the Thal + ICV cohort. Urinary GAG levels were significantly elevated at 4.21 ± 1.02 fold above normal in untreated SD cats at humane endpoint (~ 4.5 months; Fig. 5). Intracranial AAV gene therapy did not reduce the level of urinary GAGs after 16 weeks, with GAG levels at 5.36 ± 3.02 fold normal in the Thal + DCN cohort and 4.31 ± 2.36 fold normal in the Thal + ICV cohort.

Please cite this article as: H.L. Gray-Edwards, et al., Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy, Mol. Genet. Metab. (2015), http://dx.doi.org/10.1016/j.ymgme.2015.05.003

H.L. Gray-Edwards et al. / Molecular Genetics and Metabolism xxx (2015) xxx–xxx

5

Fig. 4. Cardiac pathology in GM2 cats. Left panel: Right parasternal short axis views at level of the heart base illustrate an abnormally shaped aorta with greater echogenicity in an untreated SD cat at humane endpoint ~4.5 months (B) as compared to an age matched normal (A). The aorta of untreated SD cats is of similar size to the left atrium (D) whereas the aorta is much smaller than the left atria in the normal (C). At 2 months of age, the untreated SD cat aorta is less echogenic and has a normal relationship of the aorta atria (E). At 4 months the same SD cat aorta has increased echogenicity and is of similar size to the left atria (F). Right panel: SD cat great vessel valve leaflets are expanded and irregularly thickened (H) by swollen fibroblastic cells, which contain intracytoplasmic clear to pale eosinophilic, vacuoles (J) compared to a normal cat (I). Swollen SD fibroblasts contain vacuoles which exhibit alcianophilic staining (L) compared to a normal cat (K). Thickened valve leaflets in the SD cat also contain alcianophilic staining material within the extracellular matrix (L) compared to a normal cat (K). Panels G–J = H & E staining; panels K & L alcian blue staining. Scale bars: panels G & H = 100 μm; panels I–L = 25 μm; insets I–L 12.5 μm.

4. Discussion Although the SD cat has been well studied since its discovery in 1977 [12], bony abnormalities have not been previously reported. Visceral

tissue vacuolation is a well-known feature of SD [11], and in these studies we evaluate the contribution of GAG storage in both peripheral organs and the central nervous system. GAGS are normally found in bone, connective tissue, heart valves and other stromal elements,

Fig. 5. GAG levels in Normal, SD and AAV-treated SD cats. Left axis: GAG levels are increased in all tissues analyzed from the SD cat as compared to the normal cat. Reduction after AAV gene therapy was noted in the liver and cerebral cortex of both the Thal + DCN and Thal + ICV cohort. Partial reduction after gene therapy trended toward significance in the kidney, heart, skin and lung for both cohorts, in the pancreas for the Thal + DCN cohort, and in the small intestine (S.I.) for the Thal + ICV cohort. Right axis: Urine GAG levels are increased in the SD cat, but failed to be significantly reduced in either AAV cohort. * = p b 0.05 from normal ** = p b 0.01 from normal, = p b 0.05 from untreated and Ŧ = p b 0.01 from untreated. Error bars = standard deviation.

Please cite this article as: H.L. Gray-Edwards, et al., Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy, Mol. Genet. Metab. (2015), http://dx.doi.org/10.1016/j.ymgme.2015.05.003

6

H.L. Gray-Edwards et al. / Molecular Genetics and Metabolism xxx (2015) xxx–xxx

which may explain why these are the primary areas of pathology in feline SD. While it is unclear why MPS features are not more universal in SD humans and mice, it is possible that the Hex isozymes exhibit species-specific differences in GAG metabolism, as they do for gangliosides [1,44]. Also, neurologic demise may precede the emergence of clinical MPS disease in humans and mice. Bony abnormalities in feline SD concur with previous reports of impaired GAG metabolism in human SD fibroblasts [8,60], the hexa/hexb double knockout GM2 gangliosidosis mouse [43] and other lysosomal storage disorders [15,24,49]. Gaucher patients experience bony changes and widespread osteopenia [61], the Krabbe twitcher mouse exhibits stunted postnatal bone growth [10] and the majority of hepatic storage in the GM1 cat is GAG [27]. Widespread bony changes are also present in patients with metachromatic leukodystrophy [39] and in 90% of αmannosidosis patients [33] with hip and knee pathology similar to that observed in the SD cat [14,23]. Perhaps the most notable skeletal changes occur in the MPS family of diseases where widespread skeletal and joint abnormalities are commonplace (for an extensive review, see [31]). The pelvic and spine pathologies of MPS, which are strikingly comparable to those observed in the SD cat, limit the patient's quality of life and often treatment must occur at the beginning of clinical signs to prevent major complications [52]. Cervical myelopathy as shown here in the SD cat, is a well-known feature of MPS and is characterized by diffuse thickening of the dura and extradural soft tissues, atlanto-axial instability [29] and cervical bony stenosis [57]. Spinal canal narrowing worsens with age and clinical signs associated with spinal cord compression are usually observed in adult MPS patients [2,28]. MPS IV patients are especially susceptible to spinal cord compression at the bulbospinal junction, which is largely corrected by stabilization of this area by posterior occipitocervical fusion [53]. Although spinal cord compression is consistently present in the SD cat at the same location, clinical signs associated with spinal cord compression are difficult to evaluate at the time of humane euthanasia at such a young age (4.5 ± 0.5 months) because of severe cerebellar disease. We have previously reported spinal cord compression and mild hindlimb weakness in two SD cats, which lived to a later age because of neurologic improvement after AAV gene therapy [5]. This suggests that clinical signs associated with spinal cord compression are masked by overt neurologic disease in the untreated SD cat. Therefore, due to the progressive nature of GAG deposition, it is likely that increased survival of the SD cat after effective CNS therapy will result in more severe soft-tissue spinal cord compression. Within bone, we show fewer trabeculae and reduced osteoid matrix in the SD cat, which is consistent with reduced bone density visualized by radiography. Reductions in osteoblast cell number were also noted in the SD cat, and changes in osteoblast and osteoclast activity have been noted in other LSDs. In Gaucher patients, bony abnormalities are associated with abnormal cytokine profile [37] and, in the Krabbe mouse model, reduced osteoblast and osteoclast activity is secondary to a reduction in IGF-1 [10]. Osteopenia is also present in the mucopolysaccharidoses [41]. In MPS I, increased capacity to support osteoclastogenesis was noted in bone marrow stem cells of patients [17], and, in MPS VI and VII, there is reduced capacity for bone mineralization due to inappropriate proliferation/ maturation of chondrocytes at the growth plate [48]. While the mechanism for bony abnormalities in feline SD remains unknown, we have previously shown reduction in IGF-1 in the similar GM1 gangliosidosis cat [13]. Further experiments are underway to evaluate IGF-1 and the cytokine profile in the SD cat. Cardiac manifestations in Sandhoff disease include pronounced cardiomegaly, left ventricular hypertrophy, short, thickened mitral and aortic valves with insufficiency and shortened/fused chordae tendineae [4,56]. Cardiac changes in SD, both human and feline, are strikingly similar to those described in MPS, including aortic root dilation [51], valve leaflet thickening [45] and dysplasia [6]. Ultrastructural analysis of human SD valve leaflets shows storage of an unidentified granular

material in addition to laminated lipid [4]. Considering impaired GAG metabolism in SD fibroblasts, which are the primary cell type within the heart valve, this granular material may represent GAG storage and requires further investigation. Gene therapy has been used to treat MPS bone lesions. MPS VII dogs, after systemic retroviral gene therapy, show exceptional improvement in clinical signs [58]. Growth plate normalization was noted despite no detectable restoration of the missing enzyme (β-glucuronidase) in chondrocytes and only a slight reduction in lysosomal storage material. Facial morphology, long bone lengths, articular cartilage thickness and cartilage erosions were improved, but lysosomal storage within chondrocytes and joint disease persisted at reduced levels. Complete correction of cartilage abnormalities may require intra-articular injection of AAV [34]. Cervical spine pathology also persists after gene therapy in the MPS VII dog; and, despite many therapeutic attempts, a method for complete correction of the cervical spinal column has yet to be elucidated [9,15]. Children with MPS VI treated with enzyme replacement therapy (ERT) experienced spinal cord compression that was hypothesized to be due to an unexpected increase in joint instability after treatment [28]; therefore combined surgical and ERT/gene therapy may be necessary for successful management of patients. In Gaucher disease, enzyme replacement must be started early, at high doses and administered routinely to have effect on bony changes, and even then bone disease may persist [18]. While intracranial gene therapy corrects brain GAG storage in the SD cat, it only clears GAG accumulation in a minority of peripheral organs; therefore systemic treatments may be needed to augment therapeutic effectiveness. Partial correction of peripheral tissue pathology after intracranial injection is notable, but is insufficient to correct storage in most tissues. AAV transduction of peripheral tissues after CSF injection has been reported in mice, dogs and non-human primates [22,25]. Also, we detected AAV vector in the liver of GM1 gangliosidosis cats after intracranial injection [35]. In our experiments, the injection needle transects the lateral ventricle during thalamic injections, allowing vector backflow through the needle tract to enter CSF for eventual absorption into the vasculature. Transduction efficiency in the periphery is likely a function of both tissue type and exposure to circulation [25,35]. This hypothesis is supported by the observation that the greatest correction of GAG storage was in the liver, which is exposed to a large circulatory volume and is among the most highly transduced peripheral tissues after CSF [22,25] or direct intravascular injection [59]. Because CSF ultimately is absorbed into the circulatory system, it was anticipated that correction of peripheral GAG storage would be superior after Thal + ICV versus That + DCN injection. However, both injection routes resulted in similar levels of GAG clearance in most tissues. Compromise of the blood brain barrier by the injection procedure or by primary disease, or vector leakage into CSF during surgery could account for this observation. Further study with larger sample sizes will be required before definitive conclusions can be made. Due to the primary neurologic nature of SD, it is likely that future therapies will be brain targeted, at least initially [36,42]. While the translational significance of GAG storage in SD remains unknown, it may become a source of pathology in humans once the application of novel therapeutics, as described herein, extend life. Acknowledgments This study was funded by the NIH grant U01NS064096, the National Tay–Sachs and Allied Diseases, the Cure Tay–Sachs Foundation and the Scott-Ritchey Research Center. References [1] R.C. Baek, D.R. Martin, N.R. Cox, T.N. Seyfried, Comparative analysis of brain lipids in mice, cats, and humans with Sandhoff disease, Lipids 44 (2009) 197–205. [2] M.E. Blaw, L.O. Langer, Spinal cord compression in Morquio-Brailsford's disease, J. Pediatr. 74 (1969) 593–600.

Please cite this article as: H.L. Gray-Edwards, et al., Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy, Mol. Genet. Metab. (2015), http://dx.doi.org/10.1016/j.ymgme.2015.05.003

H.L. Gray-Edwards et al. / Molecular Genetics and Metabolism xxx (2015) xxx–xxx [3] A.E. Bley, O.A. Giannikopoulos, D. Hayden, K. Kubilus, C.J. Tifft, F.S. Eichler, Natural history of infantile G(M2) gangliosidosis, Pediatrics 128 (2011) e1233–e1241. [4] L.C. Blieden, R.J. Desnick, J.B. Carter, W. Krivit, J.H. Moller, H.L. Sharp, Cardiac involvement in Sandhoff's disease. Inborn error of glycosphingolipid metabolism, Am. J. Cardiol. 34 (1974) 83–88. [5] A.M. Bradbury, J.N. Cochran, V.J. McCurdy, A.K. Johnson, B.L. Brunson, H. GrayEdwards, S.G. Leroy, M. Hwang, A.N. Randle, L.S. Jackson, N.E. Morrison, R.C. Baek, T.N. Seyfried, S.H. Cheng, N.R. Cox, H.J. Baker, M.B. Cachon-Gonzalez, T.M. Cox, M. Sena-Esteves, D.R. Martin, Therapeutic response in feline Sandhoff disease despite immunity to intracranial gene therapy, Mol. Ther. 21 (2013) 1306–1315. [6] E.A. Braunlin, P.R. Harmatz, M. Scarpa, B. Furlanetto, C. Kampmann, J.P. Loehr, K.P. Ponder, W.C. Roberts, H.M. Rosenfeld, R. Giugliani, Cardiac disease in patients with mucopolysaccharidosis: presentation, diagnosis and management, J. Inherit. Metab. Dis. 34 (2011) 1183–1197. [7] M. Cantz, H. Kresse, Sandhoff disease: defective glycosaminoglycan catabolism in cultured fibroblasts and its correction by beta-N-acetylhexosaminidase, Eur. J. Biochem. 47 (1974) 581–590. [8] M. Cantz, J.F. O'Brien, H. Kresse, Sandhoff disease: impaired catabolism of sulfated glycosaminoglycans in cultured fibroblasts, Birth Defects Orig. Artic. Ser. 11 (1975) 261–267. [9] J.A. Chiaro, P. O'Donnell, E.M. Shore, N.R. Malhotra, K.P. Ponder, M.E. Haskins, L.J. Smith, Effects of neonatal enzyme replacement therapy and simvastatin treatment on cervical spine disease in mucopolysaccharidosis I dogs, J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res. 29 (2014) 2610–2617. [10] M.A. Contreras, W.L. Ries, S. Shanmugarajan, G. Arboleda, I. Singh, A.K. Singh, Factors that affect postnatal bone growth retardation in the twitcher murine model of Krabbe disease, Biochim. Biophys. Acta 1802 (2010) 601–608. [11] L.C. Cork, J.F. Munnell, M.D. Lorenz, The pathology of feline GM2 gangliosidosis, Am. J. Pathol. 90 (1978) 723–734. [12] L.C. Cork, J.F. Munnell, M.D. Lorenz, J.V. Murphy, H.J. Baker, M.C. Rattazzi, GM2 ganglioside lysosomal storage disease in cats with beta-hexosaminidase deficiency, Science 196 (1977) 1014–1017. [13] N.R. Cox, N.E. Morrison, J.L. Sartin, F.C. Buonomo, B. Steele, H.J. Baker, Alterations in the growth hormone/insulin-like growth factor I pathways in feline GM1 gangliosidosis, Endocrinology 140 (1999) 5698–5704. [14] A.S. Desai, A. Dramis, T.N. Board, P.R. Kay, Cemented total hip arthroplasty in a patient with alpha-mannosidosis: a case report, Hip Int. 19 (2009) 151–154. [15] P.I. Dickson, S. Hanson, M.F. McEntee, C.H. Vite, C.A. Vogler, A. Mlikotic, A.H. Chen, K.P. Ponder, M.E. Haskins, B.L. Tippin, S.Q. Le, M.B. Passage, C. Guerra, A. Dierenfeld, J. Jens, E. Snella, S.H. Kan, N.M. Ellinwood, Early versus late treatment of spinal cord compression with long-term intrathecal enzyme replacement therapy in canine mucopolysaccharidosis type I, Mol. Genet. Metab. 101 (2010) 115–122. [16] S.B. Frazier, K.A. Roodhouse, D.E. Hourcade, L. Zhang, The quantification of glycosaminoglycans: a comparison of HPLC, carbazole, and alcian blue methods, Open Glycosci. 1 (2008) 31–39. [17] F. Gatto, D. Redaelli, A. Salvade, S. Marzorati, B. Sacchetti, C. Ferina, V.D. Roobrouck, F. Bertola, M. Romano, G. Villani, L. Antolini, A. Rovelli, C.M. Verfaillie, A. Biondi, M. Riminucci, P. Bianco, M. Serafini, Hurler disease bone marrow stromal cells exhibit altered ability to support osteoclast formation, Stem Cells Dev. 21 (2012) 1466–1477. [18] O. Goker-Alpan, Therapeutic approaches to bone pathology in Gaucher disease: past, present and future, Mol. Genet. Metab. 104 (2011) 438–447. [19] D. Golebiowski, A.M. Bradbury, C.S. Kwon, I.M.J. van der Bom, L. Stoica, A.K. Johnson, D.U. Wilson, H.L. Gray-Edwards, J.A. Hudson, J.A. Johnson, A.N. Randle, B.K. Whitlock, J.L. Sartin, A.L. Kühn, M. Gounis, W. Asaad, D.R. Martin, M. Sena-Esteves, AAV gene therapy strategies for lysosomal storage disorders with central nervous system involvement, in: J.V. Xuenong Bo (Ed.), Neuromethods: Gene Delivery and Therapy for Neurological Disorders, 2014 (accepted for publication). [20] R.A. Gravel, J.T. Clarke, M.M. Kaback, D. Mahuran, K. Sandhoff, K. Suzuki, The GM2 gangliosidoses, in: C.R. Scriver, Arthur L. Beaudet, William S. Sly, David Valle (Eds.), The Metabolic and Molecular Bases of Inherited Disease, McGraw-Hill, New York, 1995. [21] H.L. Gray-Edwards, N. Salibi, E.M. Josephson, J.A. Hudson, N.R. Cox, A.N. Randle, V.J. McCurdy, A.M. Bradbury, D.U. Wilson, R.J. Beyers, T.S. Denney, D.R. Martin, High resolution MRI anatomy of the cat brain at 3 Tesla, J. Neurosci. Methods 227 (2014) 10–17. [22] S.J. Gray, S. Nagabhushan Kalburgi, T.J. McCown, R. Jude Samulski, Global CNS gene delivery and evasion of anti-AAV-neutralizing antibodies by intrathecal AAV administration in non-human primates, Gene Ther. 20 (2013) 450–459. [23] S.S. Hale, J.G. Bales, S. Rosenzweig, P. Daroca, J.T. Bennett, Bilateral patellar dislocation associated with alpha-mannosidase deficiency, J. Pediatr. Orthop. B 15 (2006) 215–219. [24] M.E. Haskins, Animal models for mucopolysaccharidosis disorders and their clinical relevance, Acta Paediatr. Suppl. 96 (2007) 56–62. [25] V. Haurigot, S. Marco, A. Ribera, M. Garcia, A. Ruzo, P. Villacampa, E. Ayuso, S. Anor, A. Andaluz, M. Pineda, G. Garcia-Fructuoso, M. Molas, L. Maggioni, S. Munoz, S. Motas, J. Ruberte, F. Mingozzi, M. Pumarola, F. Bosch, Whole body correction of mucopolysaccharidosis IIIA by intracerebrospinal fluid gene therapy, J. Clin. Invest. 12 (2013) 3254–3271. [26] S.T. Hepbildikler, R. Sandhoff, M. Kolzer, R.L. Proia, K. Sandhoff, Physiological substrates for human lysosomal beta-hexosaminidase S, J. Biol. Chem. 277 (2002) 2562–2572. [27] E.W. Holmes, J.S. O'Brien, Hepatic storage of oligosaccharides and glycolipids in a cat affected with GM1 gangliosidosis, Biochem. J. 175 (1978) 945–953. [28] D.D. Horovitz, S. Magalhaes Tde, A. Pena e Costa, L.E. Carelli, D. Souza e Silva, A.P. de Linhares e Riello, J.C. Llerena Jr., Spinal cord compression in young children with type VI mucopolysaccharidosis, Mol. Genet. Metab. 104 (2011) 295–300.

7

[29] E. Kachur, R. Del Maestro, Mucopolysaccharidoses and spinal cord compression: case report and review of the literature with implications of bone marrow transplantation, Neurosurgery 47 (2000) 223–228 (discussion 228–229). [30] H. Kresse, W. Fuchs, J. Glossl, D. Holtfrerich, W. Gilberg, Liberation of Nacetylglucosamine-6-sulfate by human beta-N-acetylhexosaminidase A, J. Biol. Chem. 256 (1981) 12926–12932. [31] R. Lachman, K.W. Martin, S. Castro, M.A. Basto, A. Adams, E.L. Teles, Radiologic and neuroradiologic findings in the mucopolysaccharidoses, J. Pediatr. Rehabil. Med. 3 (2010) 109–118. [32] L.G. Luna, Manual of Histologic Staining Methods; of the Armed Forces Institute of Pathology, Blakiston Division, McGraw-Hill, 1968. [33] D. Malm, H.M. Riise Stensland, O. Edvardsen, O. Nilssen, The natural course and complications of alpha-mannosidosis—a retrospective and descriptive study, J. Inherit. Metab. Dis. 37 (2014) 79–82. [34] R.L. Mango, L. Xu, M.S. Sands, C. Vogler, G. Seiler, T. Schwarz, M.E. Haskins, K.P. Ponder, Neonatal retroviral vector-mediated hepatic gene therapy reduces bone, joint, and cartilage disease in mucopolysaccharidosis VII mice and dogs, Mol. Genet. Metab. 82 (2004) 4–19. [35] V.J. McCurdy, A.K. Johnson, H.L. Gray-Edwards, A.N. Randle, B.L. Brunson, N.E. Morrison, N. Salibi, J.A. Johnson, M. Hwang, R.J. Beyers, S.G. Leroy, S. Maitland, T.S. Denney, N.R. Cox, H.J. Baker, M. Sena-Esteves, D.R. Martin, Sustained normalization of neurological disease after intracranial gene therapy in a feline model, Sci. Transl. Med. 6 (2014) 231ra248. [36] V.J. McCurdy, H.E. Rockwell, J.R. Arthur, A.M. Bradbury, A.K. Johnson, A.N. Randle, B.L. Brunson, M. Hwang, H.L. Gray-Edwards, N.E. Morrison, J.A. Johnson, H.J. Baker, N.R. Cox, T.N. Seyfried, M. Sena-Esteves, D.R. Martin, Widespread correction of central nervous system disease after intracranial gene therapy in a feline model of Sandhoff disease, Gene Ther. 22 (2015) 181–189. [37] P. Mikosch, Miscellaneous non-inflammatory musculoskeletal conditions. Gaucher disease and bone, Best Pract. Res. Clin. Rheumatol. 25 (2011) 665–681. [38] E. Neufeld, Joseph Muenzer, The mucopolysaccharidosis, in: C.R. Scriver, Arthur L. Beaudet, William S. Sly, David Valle (Eds.), The Metabolic and Molecular bases of Inherited Disease, McGraw-Hill, New York, 1995. [39] S. Nevsimalova, M. Elleder, F. Smid, M. Zemankova, Multiple sulphatase deficiency in homozygotic twins, J. Inherit. Metab. Dis. 7 (1984) 38–40. [40] G. Panin, S. Naia, R. Dall'Amico, L. Chiandetti, F. Zachello, C. Catassi, L. Felici, G.V. Coppa, Simple spectrophotometric quantification of urinary excretion of glycosaminoglycan sulfates, Clin. Chem. 32 (1986) 2073–2076. [41] L.E. Polgreen, W. Thomas, E. Fung, D. Viskochil, D.A. Stevenson, J. Steinberger, P. Orchard, C.B. Whitley, K.E. Ensrud, Low bone mineral content and challenges in interpretation of dual-energy X-ray absorptiometry in children with mucopolysaccharidosis types I, II, and VI, J. Clin. Densitom. 17 (2014) 200–206. [42] H. Rockwell, V. McCurdy, S. Eaton, D. Wilson, A. Johnson, A. Randle, A. Bradbury, H. Gray-Edwards, H. Baker, J. Hudson, N. Cox, M. Sena-Esteves, T. Seyfried, D. Martin, AAV-mediated gene delivery in a feline model of Sandhoff disease corrects lysosomal storage in the feline central nervous system, ASN Neuro 7 (2015). [43] K. Sango, M.P. McDonald, J.N. Crawley, M.L. Mack, C.J. Tifft, E. Skop, C.M. Starr, A. Hoffmann, K. Sandhoff, K. Suzuki, R.L. Proia, Mice lacking both subunits of lysosomal beta-hexosaminidase display gangliosidosis and mucopolysaccharidosis, Nat. Genet. 14 (1996) 348–352. [44] K. Sango, S. Yamanaka, A. Hoffmann, Y. Okuda, A. Grinberg, H. Westphal, M.P. McDonald, J.N. Crawley, K. Sandhoff, K. Suzuki, R.L. Proia, Mouse models of Tay– Sachs and Sandhoff diseases differ in neurologic phenotype and ganglioside metabolism, Nat. Genet. 11 (1995) 170–176. [45] L. Schroeder, P. Orchard, C.B. Whitley, J.M. Berry, J. Tolar, W. Miller, E.A. Braunlin, Cardiac ultrasound findings in infants with severe (Hurler phenotype) untreated mucopolysaccharidosis (MPS) type I, JIMD Rep. 10 (2013) 87–94. [46] J.E. Scott, J. Dorling, Differential staining of acid glycosaminoglycans (mucopolysaccharides) by alcian blue in salt solutions, Histochemie 5 (1965) 221–233. [47] R. Seshadri, R. Christopher, H.R. Arvinda, Teaching NeuroImages: MRI in infantile Sandhoff disease, Neurology 77 (2011) e34. [48] C.M. Simonaro, M. D'Angelo, M.E. Haskins, E.H. Schuchman, Joint and bone disease in mucopolysaccharidoses VI and VII: identification of new therapeutic targets and biomarkers using animal models, Pediatr. Res. 57 (2005) 701–707. [49] W.L. Simpson, G. Hermann, M. Balwani, Imaging of Gaucher disease, World J. Radiol. 6 (2014) 657–668. [50] J. Singh, G.V. Coppa, N. Di Ferrante, Beta-N-acetylhexosaminidase active on dermatan sulfate, Enzyme 19 (1975) 15–23. [51] M.M. Sleeper, C.M. Kusiak, F.S. Shofer, P. O'Donnell, C. Bryan, K.P. Ponder, M.E. Haskins, Clinical characterization of cardiovascular abnormalities associated with feline mucopolysaccharidosis I and VI, J. Inherit. Metab. Dis. 31 (2008) 424–431. [52] G.A. Solanki, K.W. Martin, M.C. Theroux, C. Lampe, K.K. White, R. Shediac, C.G. Lampe, M. Beck, W.G. Mackenzie, C.J. Hendriksz, P.R. Harmatz, Spinal involvement in mucopolysaccharidosis IVA (Morquio-Brailsford or Morquio A syndrome): presentation, diagnosis and management, J. Inherit. Metab. Dis. 36 (2013) 339–355. [53] J.M. Stevens, B.E. Kendall, H.A. Crockard, A. Ransford, The odontoid process in Morquio-Brailsford's disease. The effects of occipitocervical fusion, J. Bone Joint Surg. 73 (1991) 851–858. [54] K. Suzuki, K. Sango, R.L. Proia, C. Langaman, Mice deficient in all forms of lysosomal beta-hexosaminidase show mucopolysaccharidosis-like pathology, J. Neuropathol. Exp. Neurol. 56 (1997) 693–703. [55] L. Toma, W. Pinto, V.C. Rodrigues, C.P. Dietrich, H.B. Nader, Impaired sulphated glycosaminoglycan metabolism in a patient with GM-2 gangliosidosis (Tay–Sachs disease), J. Inherit. Metab. Dis. 13 (1990) 721–731. [56] P. Venugopalan, S.N. Joshi, Cardiac involvement in infantile Sandhoff disease, J. Paediatr. Child Health 38 (2002) 98–100.

Please cite this article as: H.L. Gray-Edwards, et al., Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy, Mol. Genet. Metab. (2015), http://dx.doi.org/10.1016/j.ymgme.2015.05.003

8

H.L. Gray-Edwards et al. / Molecular Genetics and Metabolism xxx (2015) xxx–xxx

[57] S.L. Wald, H.H. Schmidek, Compressive myelopathy associated with type VI mucopolysaccharidosis (Maroteaux–Lamy syndrome), Neurosurgery 14 (1984) 83–88. [58] E.M. Xing, V.W. Knox, P.A. O'Donnell, T. Sikura, Y. Liu, S. Wu, M.L. Casal, M.E. Haskins, K.P. Ponder, The effect of neonatal gene therapy on skeletal manifestations in mucopolysaccharidosis VII dogs after a decade, Mol. Genet. Metab. 109 (2013) 183–193. [59] B. Yang, S. Li, H. Wang, Y. Guo, D.J. Gessler, C. Cao, Q. Su, J. Kramer, L. Zhong, S.S. Ahmed, H. Zhang, R. He, R.C. Desrosiers, R. Brown, Z. Xu, G. Gao, Global CNS

transduction of adult mice by intravenously delivered rAAVrh.8 and rAAVrh.10 and nonhuman primates by rAAVrh.10, Mol. Ther. 22 (2014) 1299–1309. [60] T. Yutaka, S. Okada, T. Kato, H. Yabuuhi, Degradation of keratan sulfate by beta-Nacetylhexosaminidases in GM2-gangliosidosis, Clin. Genet. 21 (1982) 196–202. [61] I. Zeevi, Y. Anavi, I. Kaplan, Y. Zadik, Jaws features in Type 1 Gaucher disease, J. Oral Maxillofac. Surg. 71 (2013) 694–701.

Please cite this article as: H.L. Gray-Edwards, et al., Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy, Mol. Genet. Metab. (2015), http://dx.doi.org/10.1016/j.ymgme.2015.05.003