Adeno-Associated Virus Gene Therapy in a Sheep Model of Tay

Abstract

Tay–Sachs disease (TSD) is a fatal neurodegenerative disorder caused by a deficiency of the enzyme hexosaminidase A (HexA). TSD also occurs in sheep, the only experimental model of TSD that has clinical signs of disease. The natural history of sheep TSD was characterized using serial neurological evaluations, 7 Tesla magnetic resonance imaging, echocardiograms, electrodiagnostics, and cerebrospinal fluid biomarkers. Intracranial gene therapy was also tested using AAVrh8 monocistronic vectors encoding the α-subunit of Hex (TSD α) or a mixture of two vectors encoding both the α and β subunits separately (TSD α + β) injected at high (1.3 × 10¹³ vector genomes) or low (4.2 × 10¹² vector genomes) dose. Delay of symptom onset and/or reduction of acquired symptoms were noted in all adeno-associated virus–treated sheep. Postmortem evaluation showed superior HexA and vector genome distribution in the brain of TSD α + β sheep compared to TSD α sheep, but spinal cord distribution was low in all groups. Isozyme analysis showed superior HexA formation after treatment with both vectors (TSD α + β), and ganglioside clearance was most widespread in the TSD α + β high-dose sheep. Microglial activation and proliferation in TSD sheep—most prominent in the cerebrum—were attenuated after gene therapy. This report demonstrates therapeutic efficacy for TSD in the sheep brain, which is on the same order of magnitude as a child's brain.

Introduction

The GM2 gangliosidoses Tay–Sachs disease (TSD) and Sandhoff disease (SD) are neurodegenerative and fatal lysosomal storage disorders. Individual storage disorders are rare, but collectively they consist of >40 distinct metabolic disorders with a combined prevalence of ∼1 in 7,700 live births,¹ similar to that of cystic fibrosis or hemophilia. GM2 gangliosidosis is caused by a deficiency of the enzyme β-N-acetylhexosaminidase (Hex; EC 3.2.1.52), which results in storage of GM2 ganglioside in the central nervous system (CNS). Hex exists as three isozymes: HexA, a heterodimer of α and β subunits; HexB, a ββ homodimer; and HexS, an αα homodimer. TSD is due to loss of function of the α-subunit of HexA, whereas SD results from loss of the β subunit of HexA and HexB. Since HexA (αβ) is solely responsible for degrading GM2 ganglioside in humans, loss of either subunit leads to GM2 storage and similar clinical phenotypes.

Both TSD and SD disease exist as three clinically recognized syndromes: infantile onset, juvenile onset, and late onset. Though less severe than the infantile form, juvenile-onset TSD (jTSD) exhibits a wider range of symptoms, with a large variation in symptom onset.^2,3 For example, the age of onset for a jTSD patient varies from 1.5 to 15 years, and the most common clinical sign is abnormal gait, which occurs in 88% of patients. Other common clinical signs of jTSD patients include proximal limb weakness (66%), distal limb weakness (52%), limb contracture (15%), and seizures (38%).³

Currently, there are three laboratory animals relevant for therapy development in GM2 gangliosidoses: the SD mouse,⁴ the SD cat,⁵ and the TSD sheep.^6,7 The mouse model of TSD (Hexa–/–) is largely asymptomatic, with only mild neuropathology and a lack of neurologic disease, which is attributed to an alternative catabolic pathway for GM2 ganglioside in mice.^4,8 First reported in 2010, the sheep TSD model is caused by a point mutation at the end of exon 11, resulting in a reduction of HexA activity to ∼6% of normal in the brain. Similar to TSD in patients, this decrease in HexA activity is accompanied by storage of GM2 in the CNS, with ensuing onset of neurological symptoms at ∼10 months of age.^6,7 Therefore, TSD sheep are the only animal model that reproduces the genetics, biochemical pathology, and progressive manifestation of neurological symptoms characteristic of TSD patients. Furthermore, with a brain weight of 100 g and a spinal cord that is ∼65 cm long, the sheep CNS is more similar in size and complexity to humans and may uncover new challenges experienced with a scale-up in size.

Great strides have been made in developing therapies for GM2 gangliosidoses using mouse and cat models, including adeno-associated virus (AAV) gene therapy.^9

–13 Intracranial gene therapy in the SD mouse resulted in extension of the life-span beyond 1 year of age compared to a <20-week survival of untreated animals,^12,14 and intravenous gene therapy markedly reduced neurologic disease at a predetermined endpoint at 43 weeks.¹⁵ In the SD cat, gene therapy has resulted in global distribution of Hex throughout the brain and spinal cord and a more than fourfold increase in life-span.^9,10,16 To date, testing of gene therapy has been limited to animal models of SD, but because treatment for both diseases is identical (i.e., co-administration of vectors encoding both α and β subunits), clinical trials are in the planning stages for both TSD and SD. Testing this therapy in a large-animal model of TSD is vital for translation to humans. This study performed an in-depth evaluation of the natural history of TSD in sheep and tested the clinical efficacy of intracranial AAV gene therapy.

Methods

AAV vectors

Ovine cDNAs for Hex α (HEXA) and Hex β (HEXB) subunits were cloned into separate monocistronic AAVrh8 vector backbones based on those tested in cats.^9,10 Sheep transgene expression was driven by a hybrid chicken β-actin promoter, which includes the cytomegalovirus immediate-early enhancer (conventionally called the CBA promoter). Vectors also included the woodchuck hepatitis virus post-transcriptional regulatory element (WPRE). Both vector constructs were packaged into AAVrh8 and injected either as the Hex α-subunit only or at a 1:1 ratio of both α:β subunits.

Animals and surgery

All animal procedures were approved by the Auburn University Institutional Animal Care and Use Committee. Sheep were pre-medicated with buprenorphine (0.25 mg/kg). Flunixin megalmine (2.0 mg/kg) was administered as an anti-inflammatory. Sheep were anesthetized using midazolam (2 mg/kg) and ketamine (2 mg/kg) and intubated, and anesthesia was maintained by isoflurane gas. Injections consisted of bilateral thalamic (Thal) and intracerebroventricular (ICV) injections, as previously described.¹⁷ Skull landmarks can vary dramatically depending on horn orientation and number (Jacob Sheep can have 0–6 horns), so stereotactic coordinates were based on presurgical magnetic resonance imaging (MRI) and/or intraoperative ultrasound images. An example of thalamic coordinates relative to bregma for a two-horned ewe are AP −0.1, ML ±0.7, and DV −3.2 cm. Vector was injected at a rate of 5 μL/min, delivered in ascending steps of 0.15 cm and a total volume of 175 μL/thalamus. ICV injections were performed using ultrasound guidance through a 1.5 cm × 1 cm elliptical craniotomy using an 8–5 MHz ultrasound probe on a Phillips IE33 ultrasound unit (Philips Healthcare, Andover, MA).

TSD sheep were randomly allocated into treatment groups consisting of α-only (n = 2) at a dose of 6.3 × 10¹² vector genomes (vg), or a combination of two monocistronic vectors at a 1:1 ratio (α + β) at high (1.3 × 10¹³ vg total dose including both vectors; n = 2) and low (4.2 × 10¹² vg total dose including both vectors, n = 4) doses. Two animals in the low-dose α + β cohort were euthanized at a predetermined endpoint at an age equivalent to the untreated TSD humane endpoint, and are included in the analysis for the low-dose cohort (n = 4 total). Animals were followed long term until they were unable to stand.

In vivo evaluation of therapeutic benefit was performed using both neurologic examination by a veterinarian and video footage for comparison of non-concurrent experimental subjects. Neurologic changes are represented by symptomatic events (listed in Fig. 1), and due to the heterogeneity of disease in the TSD sheep, evaluation of therapeutic benefit was determined by a percentage of symptoms attained in any given animal over the total observed in all TSD sheep. Since TSD affects all cells in the body, physical examinations were performed monthly. Heart murmurs were classified and age of onset noted for individual animals. Echocardiograms were performed using a Vivid E9 cardiac ultrasound (GE Healthcare, Wauwatosa, WI), with a M5S probe in two normal and three TSD sheep.

Figure 1.

Symptoms in Tay–Sachs disease (TSD) sheep. (A) Normal distal limb posture of a sheep. Common symptoms in TSD sheep include (B) metacarpophalangeal hyperextension (weakening fetlock) and (C) forward rotation of the metacarpophalangeal joint (knuckling). (D) Table of clinical signs in TSD sheep and the effect of adeno-associated virus (AAV) gene therapy on acquisition of symptoms. Since not all possible neurologic symptoms are experienced by individual TSD sheep, symptomatic events are expressed as a percentage representing the neurologic symptoms experienced in the animals within a group relative to the total number possible. N/A, clinical signs not applicable because they did not occur in that cohort. TSD, n = 13; TSD α-only, n = 2; TSD α + β high dose, n = 2; TSD α + β low dose, n = 2.

Animals were humanely euthanized by an intravenous pentobarbital overdose (100 mg/kg) in accordance with American Veterinary Medical Association guidelines.

MRI and magnetic resonance spectroscopy

Ultra-high field MRI was performed on a 7 Tesla (7T) MAGNETOM scanner (Siemens Healthcare, Erlangen, Germany). A 32-channel head coil (Nova Medical, Boston, MA) was used for all scans. Sheep were anesthetized as described above and imaged in dorsal recumbency. MR scans were performed at 9–12 months for normal sheep (n = 4) and at the humane endpoint for untreated TSD (n = 6) or AAV-treated sheep (n = 3). Anatomical transverse 2D axial T2 turbo spin echo images were acquired with TR/TE of 4,000/12 ms and a resolution of 0.5 mm × 0.5 mm × 1 mm. Single voxel spectroscopy was then acquired using the stimulated echo acquisition mode with variable pulse power and optimized relaxation delays water suppression, TE/TR = 4.6/10,000 ms and 32 averages. In all animals, a 10 mm × 10 mm × 17 mm voxel was placed in the striatothalamic region that was well defined on the high-resolution anatomical images. MRI data were analyzed with EFilm 3.2 software (Merge Healthcare, Chicago). Magnetic resonance spectroscopy (MRS) data were processed with LC model and internal water scaling (www.s-provencher.com/pages/lcmodel.shtml).

Electrodiagnostics

Brain-stem auditory evoked responses (BAER) were tested in three TSD sheep at humane endpoint and two age-matched normal sheep using a Nicolet VikingQuest electrodiagnostic instrument (CareFusion, San Diego, CA). Sheep were sedated with midazolam (2 mg/kg) to facilitate analysis. BAER were performed using earplug and hollow tubing auditory transmission apparatus, a click stimulus of 11.4 Hz at 75 dB, and contralateral noise at 40 dB. Electrode placement was as previously described¹⁸ and is as follows: mastoid reference (negative electrode), vertex reference (positive electrode), and ground placed on the midline at the level of the occipital condyles. Response to click stimulus was acquired over a 1.5 ms sweep with an amplitude of 0.2 μV and 1,000 averages. Measurements were repeated three times per side. Peak identification was performed, as previously described.^19
–21

Electroencephalography (EEG) was performed on two awake TSD and one normal sheep using a Neurofax 1200 series EEG (Nihon Kohden, Tokyo, Japan). Electrode placement was modified from the International Federation of Societies for Electroencephalography and Clinical Neurophysiology's standardized 10–20 electrode placement system,²² and were placed under the skin above the right and left the frontal (F1 and F2), central (C3 and C4), and occipital (O1 and O2) brain areas with the reference electrode (Cz) placed on the vertex of the skull. Only impedances <5 KΩ were accepted, and the electrodes were referenced to Cz and by hemisphere. The ground electrode was placed on the left thorax, adjacent to the heart.

Tissue preparation

After euthanasia, sheep were perfused with 2 L of ice-cold heparinized saline, and the brains were transversely sectioned into twelve 8 mm blocks extending from the frontal pole through the cerebellum and then further subdivided by hemisphere. The right hemisphere was frozen in optimal cutting temperature (OCT) medium and used for analysis of Hex distribution by naphthol staining, Hex-specific activity by 4-methylumbelliferone (4MU) assays, and quantitative polymerase chain reaction (qPCR). The left hemisphere was further subdivided into 4 mm halves for storage at −80°C (sialic acid storage analysis) or in 10% neutral-buffered formalin (hematoxylin and eosin staining or immunohistochemistry/immunofluorescence).

Hex activity, distribution, and isoenzyme analysis

Brain and spinal cord frozen sections were cut (40 μm) from each transverse OCT block and extracted on ice in 50 mM of citrate phosphate buffer containing 0.1% Triton X-100 and 0.05% bovine serum albumin (BSA), as previously described.²³ HexA activity was measured, as previously described, using a synthetic 4MU fluorogenic substrate (4MU-6-sulfo-2-acetamido-2-deoxy-β-D-glucopyranoside; MUGS) and is expressed as nmol 4MU/mg/h.²³ Qualitative assessment of Hex distribution was performed on frozen sections (40 μM) after gluteraldehyde/paraformaldehyde fixation and staining with AS-BI-N-acetyl-β-D-glucosaminide (naphthol), as previously described.²⁴ Relative levels of each isoenzyme (HexA, HexB, and HexS) were measured by diethylaminoethyl cellulose chromatography, as previously described.¹¹ Fractions were eluted from the column with increasing concentrations of sodium chloride (0–400 mmol/L), and total Hex activity in each fraction was determined with the fluorogenic substrate MUGS. To resolve the HexA and HexS peaks better, additional steps of 104, 108, 112, and 116 mM of sodium chloride were added between the standard 100 and 120 mM elutions, for a total of 30 fractions.

Western blots

Chromatography fractions (0.45 mL) were concentrated with an Amicon Ultra 0.5 mL centrifugal filter device (Millipore, Billerica, MA) to a final volume of 50–60 μL. A 12% polyacrylamide gel was loaded with 25 μL of concentrate and run for 45 min at 150 V. Proteins were transferred to a nitrocellulose membrane and probed with a rabbit polyclonal antibody to the alpha subunit of human hexosaminidase (1:100; #91624; Abcam, Cambridge, MA). The blot was visualized with a secondary antibody conjugated to IRDye^® 800CW (1:5,000) using an Odyssey scanner (Li-Cor, Lincoln, NE).

Tissue histopathology

Paraffin-embedded sections (6 μM) were deparaffinized and treated with 0.3% hydrogen peroxide in methanol for 30 min. Then antigen retrieval was performed by heat treatment in Tris-EDTA buffer (10 mmol/L of Tris, 1 mmol/L of EDTA, 0.05% Tween 20, pH 9.0) for 20 min. Sections were then blocked with 5% normal horse serum in phosphate-buffered saline (PBS) with 0.5% BSA. The primary Iba1 antibody (1:100; Biocare, Concord, CA) was incubated for 1 h at room temperature. Slides were then washed with PBS containing 0.1% BSA before being stained with anti-rabbit IgG labeled with HRP (Vector Laboratories, Burlingame, CA). Color development was performed using Vectastain RTU ABC reagent (Vector Laboratories). Iba1staining pixel intensity was quantified using Image J (National Institutes of Health, Bethesda, MD) from four 20 × photomicrographs of the cerebral cortex, including superficial and deep layers of the parietal cortex, temporal lobe, corona radiata, and caudal thalamus. For the cerebellum, pixel intensity was quantified in four images including the vermis, brain stem, and folia. Comparable numbers of blood vessels per image were included. Counts were normalized for total area and averaged across animals within the group.

Immunofluorescence was performed, as previously described,¹¹ and slides were blocked with 5% donkey serum in PBS containing 0.05% Tween 20. Sections were stained with a monoclonal antibody to Hex β-subunit (1:1,000) followed by 488-congugated donkey anti-mouse IgG (1:100; Jackson ImmunoResearch, West Grove, PA).

Ganglioside storage quantification

Ganglioside storage was estimated by sialic acid quantification, as previously described.²³ Punch biopsies (8 mm) were taken from discrete areas of the brain and spinal cord (as shown in Fig. 4) and lyophilized. The next day, samples were pulverized and rehydrated in water at a ratio of 10 mg/mL of water. Total lipids were then extracted twice in 5 mL of chloroform and methanol (1:1 v/v), stirred at room temperature for 6 h, and centrifuged at 850 g for 20 min. Supernatant was evaporated and re-suspended in 1 mL of chloroform:methanol solution (1:1) to measure sialic acid using a modified Svennerholm method,^23,25 read at 580 nm with a UV160U spectrophotometer (Shimadzu, Kyoto, Japan), and values were expressed as nmol/mg.

qPCR for vector biodistribution

Vector genomes were quantified by qPCR with SYBR green (SSo Advanced SYBR Supermix; Bio-Rad Laboratories, Hercules, CA) with primers specific for WPRE (forward: 5′-AGTTGTGGCCCGTTGTCA-3′; reverse, 5′-GAGGGGGAAAGCGAAAGT-3′). Genomic DNA (37 ng for all sections) was measured in duplicate on a CFX96 real-time system (Bio-Rad Laboratories) and compared to a standard curve for WPRE (1 × 10⁸ to 1 × 10⁰). Cycle parameters were as follows: 50°C for 2 min followed by 39 cycles at 98°C for 15 s and 60°C for 30 s.

Statistics

Statistical analyses were performed using Microsoft^® Excel (Microsoft, Redmond, WA). A two-sided, paired Student's t-test assuming unequal variances was utilized for statistical comparisons between groups. p-Values of <0.05 and <0.01 are indicated from normal (* and **, respectively) and untreated TSD sheep (ł and Ŧ, respectively). Error bars represent standard deviation. For MRS, only spectra with a Cramér-Rao minimum variance bounds of <30 were included in the analysis.

Results

Natural history of sheep TSD

TSD sheep develop a myriad of clinical signs that are not mutually exclusive or sequential (Fig. 1). The most common and earliest detectable clinical symptoms in TSD sheep are proprioceptive deficits of the foreleg known as knuckling, that is, forward rotation of the metacarpophalangeal joint so that the leg rests on its dorsal surface (Fig. 1C). Later disease signs include gait disturbances (ataxia, hypermetria), distal limb paresis resulting in a palmar or plantar stance (Fig. 1B), limb contracture (data not shown), widened base stance, vision deficits (menace or blinking reflex diminished/absent, vision loss and/or blindness with normal electroretinography), proximal limb paresis, and tremors (Supplementary Video S1; normal sheep is Supplementary Video S2; Supplementary Data are available online at www.liebertpub.com/hum). Audition is retained in the TSD sheep, and brain-stem auditory evoked responses were within normal limits (data not shown). TSD sheep with advanced disease also exhibit subjective clinical signs, including disorientation and delayed or reduced response to stimuli such as presentation of feed, loud noises, or separation from the flock. Though disease progression is not exactly the same in all sheep, an untreated sheep will acquire on average 79% of all possible symptoms, and all TSD sheep experience neurologic decline and a humane endpoint by 9.4 ± 0.8 months (n = 14).

Several TSD sheep with advanced neurologic disease experience episodes of sudden loss of consciousness (Supplementary Video S3). To determine the etiology for this phenomenon, EEG was performed in two awake TSD sheep near the humane endpoint. The TSD sheep EEG showed paroxysmal discharges of sharp waves with variable duration consistent with seizure activity (Supplementary Fig. S1). However, as previously reported,⁶ a grade 4/6 holosystolic murmur with a point of maximal intensity over the mitral valve was auscultated in 25% of TSD sheep, including several that coincided with episodes of lost consciousness. No evidence of systolic dysfunction was noted (data not shown), but atrioventricular valve dysplasia with mild regurgitation were evident (Supplementary Fig. S2). However, due to the mild severity of the regurgitation, it is unlikely that the echocardiographic abnormalities were significant enough to suspect a cardiogenic cause for the loss of consciousness. A 24 h ambulatory electrocardiographic evaluation (via Holter monitor) revealed no clinically significant arrhythmias, despite multiple episodes of lost consciousness during the monitoring period, and the rhythm was normal during those times (data not shown). Furthermore, no significant differences were noted in arterial blood pressure (while anesthetized) in two TSD sheep with loss of consciousness episodes and four TSD sheep without the episodes (data not shown).

Gene therapy survival and biodistribution

TSD sheep were treated intracranially at 2–4 months of age with AAVrh8 vectors injected into a single lateral ventricle and into the thalamus bilaterally. Thalamic injections were selected as a mode of therapeutic delivery to the cerebral cortex because of its high level of interconnectivity. A single ventricular injection was selected to distribute to the cerebellum, since downstream from the third ventricle, distribution is identical to bilateral injections. Sheep were treated with the Hex α-subunit alone (TSD α; n = 2) or with both subunits expressed from separate vectors administered simultaneously in equal doses (TSD α + β). TSD α + β sheep were treated at either a high (1.3 × 10¹³ vg; n = 2) or low dose (4.2 × 10¹² vg; n = 4). All AAV-treated sheep had delayed onset or prevention of clinical signs (Fig. 1 and Supplementary Video S4). Survival trended toward significance in all treatment groups, with the greatest mean life-span of 13.6 ± 2.5 months in the TSD α + β low-dose cohort (p = 0.1), an increase of ∼50% compared to untreated sheep. Seizures were not observed in any AAV-treated sheep, and all treated sheep retained some degree of normal behaviors, including alertness, awareness of surroundings, excitement during presentation of feed, and vocalization in the presence of flock members. Such behaviors are lost or substantially reduced in untreated TSD sheep at advanced stages of disease. Vision was retained in all AAV-treated sheep. However, menace/blink response was lost in most AAV-treated sheep.

More than 1 year post injection, Hex activity was present throughout the brain at near-normal levels in the TSD α group and at supra-normal levels (up to 80-fold normal) in the TSD α + β high-dose cohort in almost all coronal blocks. The TSD α + β low-dose cohort had intermediate Hex levels, but remained supra-physiologic near the thalamic injection site (Fig. 2C and D). In the spinal cord, a maximum of 0.2-fold normal Hex activity was measured in the cervical spinal cord in the TSD α + β high-dose cohort. Beyond the cervical region of the cord, there was little enzyme activity in any treatment group, demonstrating perhaps a primary therapeutic barrier to long-term efficacy in TSD sheep (Fig. 2C and D). Isozyme analysis of the thalamus by liquid chromatography of untreated TSD sheep showed normal formation of HexB (ββ) but a lack of HexA due to absence of the α-subunit. Treatment by the α-subunit only resulted in preferential formation of Hex S (αα). Treatment with the same dose of α-subunit vector combined with an equal ratio of β-subunit vector (high-dose cohort) produced superior formation of HexA (Fig. 3). In animals of the same treatment cohort, vector genome content generally correlated with Hex activity (Supplementary Fig. S3), with the greatest amounts found near the injection site in the TSD α + β groups. The TSD α-only group had reduced vector levels, perhaps explained by delivery of only a single AAV vector.

Figure 2.

Hexosaminidase (Hex) activity and biodistribution after gene therapy. (A) Dorsoventral view of the sheep brain, with white lines outlining corresponding brain sections. White circles indicate the bilateral thalamic injection sites. The left lateral ventricle was also injected, usually through the craniotomy site for the thalamic injection. (B) Transverse brain sections from rostral to caudal (a–j). Spinal cord sections include rostral cervical (l), mid cervical (m), cervical intumescence (n), mid thoracic (o), thoracolumbar (p), mid lumbar (q), and lumbar intumescence (r; spinal cord image not to scale). (C) Hex activity (fold normal) after gene therapy in brain or spinal cord section (denoted at bottom) in TSD, TSD α, TSD α + β high-dose, and TSD α + β low-dose cohorts. Hex activity was the highest in section f, corresponding to the thalamic injection site. (D) Histochemical staining for Hex activity (red) of the TSD α + β high-dose cohort showed global Hex distribution throughout the brain, with an area of increased intensity at the thalamic injection site (block f). Hex distribution was minimal in the spinal cord, with mild staining in the gray matter of the rostral (k) and mid cervical (l) regions. *p < 0.05 from normal; **p < 0.01 from normal; ł, p < 0.05 from TSD sheep; Ŧ, p < 0.01 from TSD sheep. Normal, n = 5; TSD, n = 4; TSD α-only, n = 2; TSD α + β high dose, n = 2; TSD α + β low dose, n = 4.

Figure 3.

TSD sheep isozyme analysis. (A) Normal sheep (gray squares) have Hex activity peaks at fractions 3 and 15, corresponding to HexB (ββ) and HexA (αβ) isozymes, respectively. TSD sheep (black diamonds) express HexB but do not express HexA due to deficiency of the α subunit. Treatment with vectors encoding α + β subunits separately (gray triangles) results in robust Hex A production. The ratio of HexB:HexA is similar in both the high-dose cohort (shown) and low-dose cohort (not shown). Sheep treated with a vector encoding the α subunit (open circles) produce predominantly the unstable isoform, Hex S (αα, ∼fraction 18), and low levels of HexA. The thalamic injection site was the brain region analyzed. (B) To distinguish the HexA peak better from the HexS peak in sheep treated with the α-subunit only, four elution steps were added between the standard 100 and 120 mM NaCl fractions (see Methods section). The modified protocol better resolved HexA and HexS. (C) To confirm the integrity of the chromatography protocol, a Western blot was performed on eluted fractions from panel (B) using an anti-peptide antibody to the Hex α-subunit that is predicted to have minimal cross-reactivity with the β-subunit. (The full-length α- and β-subunits share >50% homology, but there are areas of the proteins with lower homology.) Prominent bands for the α-subunit were detected in fractions corresponding to HexA (αβ) and HexS (αα), with only a faint band in the HexB (ββ) fraction. Relevant fractions from chromatography in panel (B) were used for the Western blot: fraction 3, HexB; fraction 8, negative control; fraction 15, HexA; fraction 24, HexS. Also included is cerebral cortex from a normal sheep (N cortex). Molecular weight standards are depicted on the right side of the blot.

As a constituent of the oligosaccharide chain of GM2 ganglioside, sialic acid was measured throughout the brain and spinal cord of TSD sheep. For cohorts treated with α + β vectors at both high and low doses, sialic acid was reduced to near normal levels at the thalamic injection site and other adjacent or interconnected cortical areas, including the parietal cortex, temporal lobe, and midbrain (Fig. 4). In the brain stem, only the high dose of α + β vectors cleared storage, while neither dose was effective at treating the cerebellum or spinal cord, in which sialic acid was elevated above untreated levels in some samples. Animals treated with the α-subunit alone cleared storage less effectively than those treated with α + β vectors in almost every region of the CNS, with the exception of the midbrain (Fig. 4).

Figure 4.

Storage quantitation. The sialic acid component of GM2 ganglioside was measured in the parietal cortex, thalamus, temporal lobe, midbrain, cerebellum, brainstem, cervical intumescence, and lumbar intumescence of sheep. Black circles represent sample sites in transverse CNS blocks from normal, TSD, TSD α, TSD α + β high-dose, and TSD α + β low-dose sheep. Sialic acid concentration was increased in all blocks in untreated TSD sheep. Clearance of storage was best in the TSD α + β high-dose group, followed by the TSD α + β low-dose group. In the TSD α group, effective clearance occurred only in the thalamus and midbrain. In the spinal cord, sialic acid levels were near—or even exceeded—that of untreated sheep in all treatment groups. *p < 0.05 from normal; **p < 0.01 from normal; ł, p < 0.05 from TSD sheep, Ŧ, p < 0.01 from TSD sheep. Normal, n = 4; TSD, n = 4; TSD α-only, n = 2; TSD α + β high dose, n = 2; TSD α + β low, n = 4.

After AAV treatment, histopathology was largely corrected throughout the cerebral cortex, including the motor cortex, somatosensory parietal cortex, caudolateral thalamus, and hippocampus, as shown by normalization of neuronal size and morphology (Fig. 5). Morphologic abnormalities persisted in the cerebellum and more prominently in the spinal cord of AAV-treated sheep. The only potential adverse finding on histopathology was the presence of strongly eosinophilic neurons and, at higher magnification, neurons exhibited a granular cytoarchitecture (Fig. 6). Immunofluorescence with a monoclonal antibody to the Hex β subunit demonstrated that the granular inclusions contain Hex protein (Fig. 6D). Distribution of eosinophilic neurons was greatest in the TSD α + β high-dose sheep with Hex-positive cells in 17/20 brain blocks (Fig. 6E). Eosinophilic neurons were present at lower levels in the TSD α + β low-dose group and in the TSD α-only group (Fig. 6E).

Figure 5.

Histopathologic evaluation of the central nervous system of TSD sheep. Hematoxylin and eosin–stained sections of the motor cortex, parietal cortex, thalamus, ventral hippocampus, cerebellum, and cervical intumescence of the spinal cord were evaluated in normal, TSD, TSD α, TSD α + β low-dose, and TSD α + β high-dose AAV-treated sheep. TSD sheep neurons throughout the brain and spinal cord are vacuolated/distended with storage material. After gene therapy, there was improvement of neuronal morphology throughout the cortex, with greatest restoration in the TSD α + β low-dose group. The cerebellar folia and ventral horn cervical intumescence had persistent neurodegenerative change after gene therapy, though some improvement was noted in the purkinje cells of the cerebellum after gene therapy. Brain areas analyzed: cruciate gyrus of the motor cortex, suprasylvian gyrus of the parietal cortex, caudodorsal thalamus, ventral hippocampus, cerebellar vermis, and cervical intumescence of the spinal cord. Scale bars = 10 μM. Normal, n = 4; TSD, n = 4; TSD α-only, n = 2; TSD α + β low dose, n = 4; TSD α + β high dose, n = 2.

Figure 6.

Distribution of Hex-positive neurons. Absent in normal (not shown) or untreated TSD brains (A), eosinophilic granular inclusions were found in TSD sheep treated with both the low dose (B) and the high dose (C) of AAV gene therapy. Immunofluorescence for Hex showed positive staining within neurons containing eosinophilic material (D). Sections were then evaluated/scored throughout the brain based on the number of inclusion-positive neurons and the abundance of those inclusions within the cytoplasm of neurons. Panel (E) shows the treatment cohorts with identification numbers for each animal. Brain regions from each animal were scored for severity with the following symbols: Ŧ, marked number of eosinophilic inclusions; ◊, moderate number of inclusions; ○, mild number of inclusions. Inclusions were most abundant in the high-dose α + β group, followed by the low-dose α + β group. The absence of a symbol indicates the absence of eosinophilic inclusions. Scale bar for panels (A–C) = 25 μm; inset scale bars = 10 μm. Normal, n = 4 (not shown); TSD, n = 4; TSD α-only, n = 2; TSD α + β high dose, n = 2; TSD α + β low, n = 4.

To characterize neuroinflammation in sheep TSD partially, microglial activation was evaluated using Iba1 immunohistochemistry. Microglia changed from a resting, ramified morphology in normal animals (Fig. 7A, thalamus, and Fig. 7F, cerebellar vermis) to an activated, amoeboid shape in the cerebrum and cerebellum of untreated TSD sheep (Fig. 7B and G). Microglial proliferation was present in both brain areas of untreated TSD sheep, with the most pronounced staining in the internal capsule (data not shown). After AAV gene therapy, microglial activation and proliferation appeared attenuated in the thalamus and vermis of both AAV α + β low- (Fig. 7C and H) and high-dose (Fig. 7D and I) groups. In the α-only cohort, cerebellar microgliosis increased relative to animals treated with both vectors but was less severe than untreated TSD sheep (Fig. 7J). Microglial density across the cerebrum and cerebellum was quantified in normal, TSD, and TSD α + β low-dose sheep. The microglial response, as indicated by Iba1 staining, was increased in the cerebrum and cerebellum of untreated TSD sheep. In TSD sheep that were treated with the low dose of α + β vectors, microgliosis was intermediate to normal and untreated TSD sheep in the cerebrum and was normal in the cerebellum (Fig. 7K).

Figure 7.

Microglial activation and proliferation. Immunohistochemistry for Iba1 was performed in the cerebrum (A–E) and cerebellum (F–J). As expected, microglia in normal sheep were ramified, with small cell bodies and long, thin processes (A and F). In untreated TSD sheep, microglia proliferated and were activated, as indicated by an amoeboid morphology in which cell bodies were large and round with small projections (B and G). After gene therapy, microglial numbers and morphology were intermediate to normal in TSD sheep (C–E and H–J). To quantify microgliosis, Iba1 pixel densities were counted in the cerebrum and cerebellum of normal, TSD, and TSD α + β low-dose groups (K). Regions counted in the cerebrum consisted of superficial and deep layers of the parietal cortex, temporal lobe, corona radiata, and caudal thalamus, while regions counted in the cerebellum included vermis, brain stem, and folia. **p < 0.01 from normal; Ŧ, p < 0.01 from TSD sheep. Scale bars for (A–F) = 10 μm. Normal, n = 4; TSD, n = 4; TSD α-only, n = 2; TSD α + β high, n = 2; TSD α + β low, n = 4.

Biomarkers

Ultra-high field (7 Tesla) MRI was performed in normal sheep, TSD sheep, and TSD α + β low-dose sheep (Fig. 8A–F). MRI was not performed in the other cohorts because the MRI was not yet installed during those experiments. T2-weighted MRIs showed cortical white-matter isointensity relative to gray matter in untreated TSD sheep (Fig. 8B) compared to normal (Fig. 8A). Also present were areas of hyperintensity surrounding white matter of the corona radiata and cortical atrophy, apparent from increased cerebrospinal fluid (CSF) surrounding the brain (Fig. 8B). AAV treatment with the low dose of α + β vectors resulted in normalized white:gray matter intensities, although cortical atrophy was apparent at 11 months of age (Fig. 8C). Additionally, several treated sheep had hyperintensities in the thalamic injection site (Fig. 8F; arrow), a finding that has been reported previously with intraparenchymal injection of AAV.^11,23

Figure 8.

Ultra-high field (7 Tesla) magnetic resonance imaging (MRI) and cerebrospinal fluid (CSF) biomarkers in TSD sheep after gene therapy. The outermost aspect of the white matter (corona radiata) or innermost layer of the gray matter is hyperintense to gray matter in the untreated TSD sheep (arrows) (B and E) compared to normal (A and D). Areas of isointensity between gray and white matter (arrow) and hyperintensity surrounding gray matter (arrow) are present in several brain areas, and cortical atrophy is noted in the untreated sheep, with increased CSF surrounding the brain (B and E). In the TSD α + β low-dose cohort, white to gray matter intensities were largely normalized, but brain atrophy persisted (C and F). Additionally, bilateral thalamic hyperintensities were present at the level of the injection site (arrow) (F). Sheep ages at MRI: normal, 10.0 months; TSD, 9.0 months; TSD + AAV low dose, 15.4 months. (G) MR spectroscopy (7 Tesla) in the striatum and thalamus of TSD sheep showed a significant reduction in glutamate, N-acetylaspartate (NAA), and NAA + N-acetylglutamate, with increases in taurine, glycerophosphocholine (GPC), and phosphocholine (PCh). Metabolite normalization varied in AAV-treated sheep (striped bars), with complete correction of GPC + PCh. (H) Representative normal sheep MR spectrum with residuals (the data minus the data not fit to the data; top) from LC model software. CSF levels of aspartate aminotransferease (AST) (I) and lactate dehydrogenase (LDH) (J) increased in TSD sheep by 2 months of age and remained elevated through the humane endpoint (∼9.4 months). Significant reductions were measured in all treatment cohorts, with the exception of LDH from sheep treated with the high dose of α + β vectors. *p < 0.05 from normal; **p < 0.01 from normal; ł, p < 0.05 from TSD sheep; Ŧ, p < 0.01 from TSD sheep. Normal, n = 4; TSD, n = 8; TSD α + β low, n = 3.

MRS (representative spectrum shown in Fig. 8H) detected changes in several metabolites in the untreated TSD sheep brain (Fig. 8G). The neurotransmitter glutamate, and markers of neuroaxonal health N-acetyl aspartate (NAA) and NAA + N-acetyl glutamate, were significantly reduced in the TSD sheep brain at the humane endpoint. Elevations were present in the amino acid taurine and markers of demyelination (glycerophosphocholine [GPC] + phosphocholine [PCh]). Creatine + phosphocreatine, which reflect cerebral metabolism, trended toward an increase in untreated TSD sheep but did not reach statistical significance compared to normal (p = 0.07). In the AAV-treated sheep brain, GPC + PCh levels were normal, thus supporting the normalization of white matter intensity on MRI. NAA, NAA + N-acetyl glutamate, and taurine levels were intermediate to that of normal and TSD sheep, whereas no improvement in glutamate levels was detected after gene therapy. No significant changes were noted in the gliosis marker myoinositiol (p = 0.22) in untreated or AAV-treated sheep (p = 0.29).

Aspartate aminotransferase (AST) in CSF was elevated in untreated TSD sheep at 2 months of age and remained elevated at approximately the same level until the humane endpoint (Fig. 8I). AAV gene therapy reduced AST levels in all treatment groups compared to untreated TSD sheep, and AST levels were the lowest in the TSD α cohort. Lactate dehydrogenase (LDH) in CSF followed the same pattern as AST, with elevations in untreated TSD sheep by 2 months of age that were sustained until the humane endpoint. At the humane endpoint, reduction of LDH was only present in the TSD α and TSD α + β low-dose groups (Fig. 8J).

Discussion

TSD and SD are fatal neurodegenerative disorders for which only palliative care is currently available to patients. Hope for an effective treatment comes from recent studies that show profound, sustained efficacy after intracranial injection of AAV gene therapy in animal models. Because the TSD mouse has limited storage in the CNS and almost no clinical symptoms, the SD mouse is the model of choice for therapeutic studies. Intracranial AAV gene therapy in the knockout mouse model of SD has produced dramatic results, including a more than fivefold extension of life-span that allowed some SD animals to live as long as wild-type mice.^12,14 Similarly, at an age equivalent to the humane endpoint of untreated SD cats, AAV-treated SD cats had pronounced preservation of neurologic function and, in long-term studies, a more than fourfold extension of life-span.^9,10,16 Success of gene therapy in SD animal models has prompted consideration of human clinical trials, and because both SD and TSD share a deficiency of HexA, it is thought that both diseases can be treated with the same vectors. Testing AAV gene therapy in sheep allowed an authentic model with the same subunit deficiency as TSD patients to be used, while also evaluating biodistribution and efficacy in a large brain only 5–10 times smaller than that of a patient.

To establish a baseline of disease progression in ovine TSD against which treated animals can be compared, further characterization of the model was performed in a controlled research setting. Affected sheep exhibit heterogeneous disease progression that includes gait abnormalities, limb contracture, distal limb paresis, spasticity, and seizures, as reported in both juvenile and late-onset TSD patients.³ Though the sequence of symptom acquisition varies among TSD sheep, they reach the humane endpoint, defined by the inability to stand, within a narrow time interval at 9.4 ± 0.8 months of age (n = 14), which facilitates survival studies of experimental therapies, since relatively small numbers of animals are sufficient to detect an effect.

Untreated TSD sheep experienced episodes of lost consciousness, and even when awake and unsedated, they had epileptiform waves in the EEG, suggestive of seizure activity. The sheep model of neuronal ceroid lipofuscinosis (CLN5) also exhibits epileptiform discharges with no outward signs of seizure,²⁶ suggesting subclinical neurophysiological changes in sheep with lysosomal storage diseases. In TSD sheep, cardiovascular disease was ruled out as an etiology for loss of consciousness (by echocardiography and Holter monitoring), though murmurs ⁶ and knobby valves were noted. Cardiac valvular dysplasia has been reported in human SD^13,27,28 but not in TSD, suggesting a possible species difference.

Ultra-high field MRI (7 Tesla) provides unprecedented resolution of brain structure and metabolites for quantitative, non-invasive measures of TSD pathology. The ovine model faithfully represents numerous structural and metabolite abnormalities reported in human patients, as follows. Demyelination, visualized as increased signal intensity of the white matter on T2-weighted MRI and increases in GPC + PCh by MRS, are hallmark findings in TSD patients^29
–31 and affected sheep. Similarly, NAA reduction in affected sheep is a feature of TSD (and GM1 gangliosidosis) patients,^29,31,32 reflecting compromised neuro-axonal health. Reduction in glutamate, like that observed here, has been reported in other neurodegenerative diseases^33

–37 and may also reflect neuronal loss/dysfunction. An elevation in the amino acid taurine in affected sheep may represent taurine-conjugated GM2 ganglioside, isolated from the brain of a TSD patient and hypothesized to result from a compensatory mechanism to increase GM2 solubility and removal from the neuron.³⁸ The only brain metabolite with discordant findings in sheep and human patients was myoinositol, a marker of gliosis.^31,39 Clearly elevated in TSD patients, myoinositol was not significantly increased in affected sheep, despite microglial proliferation and activation. This discrepancy is likely due to differences in disease severity between affected humans and sheep at the end stage, as TSD sheep are humanely euthanized at a point of moderate neurologic disease compared to patients. For a review of MRS metabolites and potential implications, see Panigrahy et al.³⁰

AAV treatment in TSD sheep led to delayed symptom onset and disease progression with enhanced quality of life. Treated TSD sheep retained normal behavior and vocalization patterns in the presence of flock mates, and they retained alert mentation, even upon reaching the humane endpoint (inability to stand). Other indicators also point to a beneficial effect of gene therapy in the sheep model. For example, MR-based evaluation of brain architecture and metabolites demonstrated a clear improvement in treated TSD sheep. Significant decreases after treatment were documented in the CSF biomarkers AST and LDH, previously reported to correlate with therapeutic efficacy in feline models of gangliosidosis.¹⁹ Furthermore, microgliosis was attenuated in treated TSD sheep. With such positive effects on behavior, clinical biomarkers, and neuroinflammation in AAV-treated TSD sheep, the question remains as to why gene therapy did not increase life-span significantly in the current study.

With increased animal numbers, the life-span of AAV-treated sheep may reach statistical significance. However, it is important to understand why gains in survival in this study were relatively modest. In animals treated with both subunits, most areas of the CNS had above-normal Hex activity and normal levels of ganglioside. Exceptions include the cerebellum and spinal cord, in which storage remained equal to or surpassed that of untreated TSD sheep, regardless of treatment, likely constituting the primary therapeutic limitation in the current study. It is possible that because survival was increased by effective treatment of the cerebrum, areas of the CNS that received less therapeutic benefit continued to accumulate storage, surpassing untreated levels. Histopathologic evaluation supports a need of improved targeting of cerebellum and spinal cord, with modest improvements in neuronal morphology and microgliosis in the cerebellum after gene therapy. This suggests that low levels of enzymatic restoration do provide some level of therapeutic efficacy. SD cats treated by the same injection routes, dose, capsid (AAVrh8), and vector design (with species-specific cDNAs) had Hex activity in the cerebellum and spinal cord that ranged from 5.0- to 14.7-fold normal, with corresponding reductions in GM2 ganglioside to near-normal levels.⁹ Thus, improved treatment of the cerebellum and spinal cord is a major priority for future work. Direct injection of the deep cerebellar nuclei and alternative routes of CSF delivery are currently under investigation in TSD sheep. Intrathecal or cisterna magna delivery of AAV has resulted in widespread transgene delivery to the cerebellum and/or spinal cord of nonhuman primates,^40,41 wild-type mice,⁴² mice and dogs with mucopolysaccharidosis (MPS) III,^43,44 dogs with MPS VII,⁴⁵ and cats with MPS I.⁴⁶ Substantial efficacy was shown in each of the MPS models treated by CSF-mediated delivery of AAV.

Though unlikely, toxicity of the AAV treatment cannot be ruled out as an explanation for relatively modest gains in life-span in this study. In both the low- and high-dose TSD α + β groups, HexA activity reached supra-physiologic levels at the thalamic injection sites, but no overt toxicity was observed. Interestingly, a trending decrease in vector genomes was noted in the high-dose cohort at the injection site, which could indicate a loss of transduced neurons. While clinical neurological toxicity could have been masked by the innate neurodegenerative process in TSD sheep, histopathologic analysis of treated sheep brains was unremarkable, with no overt neuronal loss or immune cell infiltration, and the only abnormality was a subtle looseness of the neuropil (data not shown), MRI hyperintensities, and the presence of granular, eosinophilic neurons expressing Hex (Fig. 6). Such neurons and MRI hyperintensities were also found in asymptomatic wild-type cats followed for ≥18 months after bilateral injection of both the thalamus and deep cerebellar nuclei using the same vector backbone, capsid, and approximate dose (scaled down for the cat brain size) as the current study. AAV-treated wild-type cats showed no overt toxicity during the entire 18 months of the study.¹¹

In summary, combination parenchymal and ICV delivery of AAV vectors results in widespread Hex distribution and clearance of storage in the majority of the TSD sheep brain. Further studies are necessary to improve biodistribution to the cerebellum and spinal cord, which may become an important factor in future clinical trials.

Footnotes

Acknowledgments

The authors wish to acknowledge the outstanding contributions and commitment of Fred and Joan Horak (Gettysburg, PA), Dave and Laurie Cellela (Terrell, TX) and Cheryl Terrano (Buckhannon, WV), without whom this research would not have been possible. We would also like to acknowledge Florian Eichler (Massachusetts General Hospital Boston, MA) and Swati Sathe (Rutgers, Newark, NJ) for their clinical insight into neurologic signs experienced by the TSD sheep. Studies were funded by the National Tay–Sachs and Allied Diseases Association, Inc., the Cure Tay–Sachs Foundation, Ms. Sophia Pesotchinsky, Ms. Lisa Robaut and Ms. Bonnie Pastor, to whom the authors are also grateful.

Author Disclosure

N.S. is an employee of Siemens Healthcare. D.M. and M.S.-E. are shareholders in Lysogene (Neuilly-sur-Seine, France). No competing interests exist for the remaining authors.

References

Meikle

, Hopwood

, Clague

, et al. Prevalence of lysosomal storage disorders. JAMA, 1999; 281:249–254.

Bley

, Giannikopoulos

, Hayden

, et al. Natural history of infantile G(M2) gangliosidosis. Pediatrics, 2011; 128:e1233–1241.

Maegawa

, Stockley

, Tropak

, et al. The natural history of juvenile or subacute GM2 gangliosidosis: 21 new cases and literature review of 134 previously reported. Pediatrics, 2006; 118:e1550–1562.

Phaneuf

, Wakamatsu

, Huang

, et al. Dramatically different phenotypes in mouse models of human Tay–Sachs and Sandhoff diseases. Hum Mol Genet, 1996; 5:1–14.

Cork

, Munnell

, Lorenz

, et al. GM2 ganglioside lysosomal storage disease in cats with beta-hexosaminidase deficiency. Science, 1977; 196:1014–1017.

Porter

, Lewis

, Edwards

, et al. Pathology of GM2 gangliosidosis in Jacob sheep. Vet Pathol, 2011; 48:807–813.

Torres

, Zeng

, Porter

, et al. Tay–Sachs disease in Jacob sheep. Mol Genet Metab, 2010; 101:357–363.

Sango

, Yamanaka

, Hoffmann

, et al. Mouse models of Tay–Sachs and Sandhoff diseases differ in neurologic phenotype and ganglioside metabolism. Nat Genet, 1995; 11:170–176.

Rockwell

, McCurdy

, Eaton

, et al. AAV-mediated gene delivery in a feline model of Sandhoff disease corrects lysosomal storage in the central nervous system. ASN Neuro, 2015; 7.

10.

McCurdy

, Rockwell

, Arthur

, et al. Widespread correction of central nervous system disease after intracranial gene therapy in a feline model of Sandhoff disease. Gene Ther, 2015; 22:181–189.

11.

Bradbury

, Cochran

, McCurdy

, et al. Therapeutic response in feline Sandhoff disease despite immunity to intracranial gene therapy. Mol Ther, 2013; 21:1306–1315.

12.

Cachon-Gonzalez

, Wang

, McNair

, et al. Gene transfer corrects acute GM2 gangliosidosis—potential therapeutic contribution of perivascular enzyme flow. Mol Ther, 2012; 20:1489–1500.

13.

Gray-Edwards

, Brunson

, Holland

, et al. Mucopolysaccharidosis-like phenotype in feline Sandhoff disease and partial correction after AAV gene therapy. Mol Genet Metab, 2015; 116:80–87.

14.

Cachon-Gonzalez

, Wang

, Lynch

, et al. Effective gene therapy in an authentic model of Tay–Sachs-related diseases. Proc Natl Acad Sci U S A, 2006; 103:10373–10378.

15.

Walia

, Altaleb

, Bello

, et al. Long-term correction of Sandhoff disease following intravenous delivery of rAAV9 to mouse neonates. Mol Ther, 2015; 23:414–422.

16.

Bradbury

, McCurdy

, Johnson

, et al. Intracranial AAV gene therapy extends the life span of GM2 gangliosidosis cats >4 four-fold with no clinical evidence of vector toxicity. Mol Ther, 2013; 21s1:226.

17.

Golebiowski

, Bradbury

, Kwon

, et al. AAV gene therapy strategies for lysosomal storage disorders with central nervous system involvement. In: Xuenong

, Verhaagen

, eds. Neuromethods: Gene Delivery and Therapy for Neurological Disorders. New York: Humana Press, 2015:265–295.

18.

Steiss

, Baker

, Braund

, et al. Profile of electrodiagnostic abnormalities in cats with GM1 gangliosidosis. Am J Vet Res, 1997; 58:706–709.

19.

Bradbury

, Gray-Edwards

, Shirley

, et al. Biomarkers for disease progression and AAV therapeutic efficacy in feline Sandhoff disease. Exp Neurol, 2015; 263:102–112.

20.

Buchwald

, Huang

. Far-field acoustic response: origins in the cat. Science, 1975; 189:382–384.

21.

Sims

. Electrodiagnostic evaluation of auditory function. Vet Clin North Am Small Anim Pract, 1988; 18:913–944.

22.

Jasper

. The ten-twenty electrode system of the International Federation. Electroencephalogr Clin Neurophysiol, 1958; 371–375.

23.

McCurdy

, Johnson

, Gray-Edwards

, et al. Sustained normalization of neurological disease after intracranial gene therapy in a feline model. Sci Transl Med, 2014; 6:231ra248.

24.

Lacorazza

, Jendoubi

. In situ assessment of beta-hexosaminidase activity. Biotechniques, 1995; 19:434–440.

25.

Svennerholm

. Quantitative estimation of sialic acids. II. A colorimetric resorcinol-hydrochloric acid method. Biochim Biophys Acta, 1957; 24:604–611.

26.

Perentos

, Martins

, Watson

, et al. Translational neurophysiology in sheep: measuring sleep and neurological dysfunction in CLN5 Batten disease affected sheep. Brain, 2015; 138:862–874.

27.

Blieden

, Desnick

, Carter

, et al. Cardiac involvement in Sandhoff's disease. Inborn error of glycosphingolipid metabolism. Am J Cardiol, 1974; 34:83–88.

28.

Venugopalan

, Joshi

. Cardiac involvement in infantile Sandhoff disease. J Paediatr Child Health, 2002; 38:98–100.

29.

Imamura

, Miyajima

, Ito

, et al. Serial MR imaging and 1H-MR spectroscopy in monozygotic twins with Tay–Sachs disease. Neuropediatrics, 2008; 39:259–263.

30.

Panigrahy

, Nelson

Jr , Bluml

. Magnetic resonance spectroscopy in pediatric neuroradiology: clinical and research applications. Pediatr Radiol, 2010; 40:3–30.

31.

Aydin

, Bakir

, Tatli

, et al. Proton MR spectroscopy in three children with Tay–Sachs disease. Pediatr Radiol, 2005; 35:1081–1085.

32.

Regier

, Kwon

, Johnston

, et al. MRI/MRS as a surrogate marker for clinical progression in GM1 gangliosidosis. Am J Med Genet Part A, 2016; 170:634–644.

33.

Antuono

, Jones

, Wang

, et al. Decreased glutamate + glutamine in Alzheimer's disease detected in vivo with (1)H-MRS at 0.5 T. Neurology, 2001; 56:737–742.

34.

Lin

, Shic

, Enriquez

, et al. Reduced glutamate neurotransmission in patients with Alzheimer's disease—an in vivo (13)C magnetic resonance spectroscopy study. MAGMA, 2003; 16:29–42.

35.

Abdel-Aziz

, Schneider

, Solanky

, et al. Evidence for early neurodegeneration in the cervical cord of patients with primary progressive multiple sclerosis. Brain, 2015; 138:1568–1582.

36.

Plaitakis

, Constantakakis

, Smith

. The neuroexcitotoxic amino acids glutamate and aspartate are altered in the spinal cord and brain in amyotrophic lateral sclerosis. Ann Neurol, 1988; 24:446–449.

37.

Ernst

, Chang

, Melchor

, et al. Frontotemporal dementia and early Alzheimer disease: differentiation with frontal lobe H-1 MR spectroscopy. Radiology, 1997; 203:829–836.

38.

, Maskos

, Chou

, et al. Presence of an unusual GM2 derivative, taurine-conjugated GM2, in Tay–Sachs brain. J Biol Chem, 2003; 278:35286–35291.

39.

Assadi

, Baseman

, Janson

, et al. Serial 1H-MRS in GM2 gangliosidoses. Eur J Pediatr, 2008; 167:347–352.

40.

Gray

, Nagabhushan Kalburgi

, McCown

, et al. Global CNS gene delivery and evasion of anti-AAV-neutralizing antibodies by intrathecal AAV administration in non-human primates. Gene Ther, 2013; 20:450–459.

41.

Samaranch

, Salegio

, San Sebastian

, et al. Strong cortical and spinal cord transduction after AAV7 and AAV9 delivery into the cerebrospinal fluid of nonhuman primates. Hum Gene Ther, 2013; 24:526–532.

42.

Guo

, Wang

, Qiao

, et al. A single injection of recombinant adeno-associated virus into the lumbar cistern delivers transgene expression throughout the whole spinal cord. Mol Neurobiol, 2016; 53:3235–3248.

43.

Haurigot

, Marco

, Ribera

, et al. Whole body correction of mucopolysaccharidosis IIIA by intracerebrospinal fluid gene therapy. J Clin Invest, 2013 Jul 1 [Epub ahead of print]; DOI:10.1172/JCI66778.

44.

Ribera

, Haurigot

, Garcia

, et al. Biochemical, histological and functional correction of mucopolysaccharidosis type IIIB by intra-cerebrospinal fluid gene therapy. Hum Mol Genet, 2015; 24:2078–2095.

45.

Gurda

, De Guilhem De Lataillade

, Bell

, et al. Evaluation of AAV-mediated gene therapy for central nervous system disease in canine mucopolysaccharidosis VII. Mol Ther, 2016; 24:206–216.

46.

Hinderer

, Bell

, Gurda

, et al. Intrathecal gene therapy corrects CNS pathology in a feline model of mucopolysaccharidosis I. Mol Ther, 2014; 22:2018–2027.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.16 MB

0.13 MB

0.05 MB

4.56 MB

5.55 MB

35.86 MB

32.74 MB

Adeno-Associated Virus Gene Therapy in a Sheep Model of Tay–Sachs Disease

Abstract

Introduction

Methods

AAV vectors

Animals and surgery

MRI and magnetic resonance spectroscopy

Electrodiagnostics

Tissue preparation

Hex activity, distribution, and isoenzyme analysis

Western blots

Tissue histopathology

Ganglioside storage quantification

qPCR for vector biodistribution

Statistics

Results

Natural history of sheep TSD

Gene therapy survival and biodistribution

Biomarkers

Discussion

Footnotes

Acknowledgments

Author Disclosure

References

Supplementary Material