Gene Therapy Approaches for Neurological Lysosomal Storage Diseases by Victoria Jane McCurdy A dissertation submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Doctor of Philosophy Auburn, Alabama May 6, 2014 Keywords: gangliosidosis, gene therapy, lysosomal storage disease, neurological disease, feline model, adeno-associated virus Approved by Douglas R. Martin, Associate Professor, Anatomy, Physiology and Pharmacology Eleanor M. Josephson, Associate Professor, Anatomy, Physiology and Pharmacology Nancy R. Cox, Professor, Pathobiology Bruce F. Smith, Professor, Pathobiology Aime K. Johnson, Associate Professor, Clinical Sciences ii Abstract GM1 and GM2 gangliosidosis are lysosomal storage diseases caused by deficiency of enzymes required for ganglioside catabolism. Enzyme deficiencies cause neuronal accumulation of ganglioside resulting in progressive neurodegeneration and premature death. There is no cure for these fatal diseases. Feline GM1 and GM2 gangliosidosis models are close replicas of the human juvenile- and infantile- onset disease forms, respectively, and provide ideal large animal models to test therapies for translation to humans. Prior to the initiation of this research project evaluating adeno-associated virus (AAV) gene therapy, no experimental treatment had altered the course of the feline gangliosidoses. Sixteen weeks after intracranial injection of AAV vectors in gangliosidosis cats, therapeutic enzyme activity was widely restored at near or above normal levels throughout the central and peripheral nervous system, and was also detected at lower levels in peripheral tissues. Restoration of enzyme activity resulted in substantial clearance of pathologic storage material and normalization of a secondary biomarker of lysosomal function. In long term treatment groups, mean survival of GM1 gangliosidosis cats treated before disease onset currently stands at >4.7 times that of untreated cats. The majority of treated cats remain alive and in good clinical condition, so mean survival continues to increase. Mean survival of GM2 gangliosidosis cats treated before disease onset was >4.3 times that of untreated. Enzyme activity was maintained throughout the current life span, but it was variable between subjects and between central nervous system regions, with several regions still demonstrating a substantial burden of storage iii material. The significant increase in life span of treated GM2 gangliosidosis cats resulted in the emergence of previously subclinical peripheral disease symptoms that were the ultimate cause of death in seven out of nine cats. AAV-gene therapy also produced significant survival gains in GM2 gangliosidosis cats treated during the early symptomatic disease stage, which is when most human patients are diagnosed. This research represents a significant advancement in the treatment of the feline gangliosidoses and supports the continued improvement and refinement of AAV gene therapy for translation to humans. Future studies should continue to investigate ways to maximize enzyme distribution throughout the central nervous system in order to further reduce storage material as well as investigate methods to treat peripheral disease pathology. iv Acknowledgments The author thanks her advisor Douglas R. Martin for providing the opportunity and guidance that made this enjoyable Ph.D. project possible. The author also thanks the other members of the graduate advisory committee, Nancy R. Cox, Bruce F. Smith, Eleanor M. Josephson, and Aime K. Johnson for their helpful advice and involvement. The author also greatly appreciates and thanks the following people: Heather L. Gray-Edwards and Allison M. Bradbury for their encouragement, support, and advice as well as for the provision of valuable MRI data and chapter 2 antibody titers, respectively; Misako Hwang for her ever helpful advice and willing assistance in the laboratory as well as for the provision of chapter 2 cellulose ion exchange chromatography data; Ashley N. Randle for her help and input with clinical rating scores and generation of neurological exam results; Nancy E. Morrison for her ever helpful assistance in the laboratory; Atoska S. Gentry for her kind assistance around the laboratory as well as for processing slides for histological analyses; and Brandon L. Brunson for chapters 1 and 2 histological analyses. The author also thanks the Auburn Cellular and Molecular Biosciences (CMB) Program who provided the opportunity and initial financial support to come to Auburn University to embark on a Ph.D. Finally, the author thanks Alan and Sheila Jones for their everyday love, support, and advice, and Jay McCurdy for his everyday love, support, and influence. v Table of Contents Abstract ........................................................................................................................................ .ii Acknowledgments........................................................................................................................ iv List of Tables .............................................................................................................................. vii List of Figures ............................................................................................................................... x List of Abbreviations ................................................................................................................. xiv Introduction ................................................................................................................................ 1 Methods....................................................................................................................................... 34 Chapter 1 Sustained normalization of neurological disease after intracranial gene therapy in a feline model of GM1 gangliosodosis ...................................................................... 43 Abstract .................................................................................................................. 43 Introduction ............................................................................................................ 44 Results ...................................................................................................................... 46 Discussion ................................................................................................................ 53 Figures and Tables ................................................................................................... 58 Chapter 2 Widespread correction of central nervous system disease following intracranial gene therapy in a feline model of Sandhoff disease ................................................ 78 Abstract .................................................................................................................... 78 Introduction .............................................................................................................. 79 Results ...................................................................................................................... 80 Discussion ................................................................................................................ 85 vi Figures and Tables ................................................................................................... 89 Chapter 3 Dramatic phenotypic improvement after adeno-associated virus gene therapy in a feline model of Sandhoff disease .......................................................................... 107 Abstract .................................................................................................................. 107 Introduction ............................................................................................................ 107 Results .................................................................................................................... 110 Discussion .............................................................................................................. 116 Figures and Tables ................................................................................................. 122 Chapter 4 Significant therapeutic benefit after post-symptomatic intracranial gene therapy in a feline model of Sandhoff disease .......................................................................... 138 Abstract .................................................................................................................. 138 Introduction ............................................................................................................ 139 Results .................................................................................................................... 141 Discussion .............................................................................................................. 146 Tables and Figures ................................................................................................. 152 Discussion ................................................................................................................................. 166 References ............................................................................................................................... 174 Appendix 1 Contribution of thalamus and DCN injection sites to overall therapeutic outcome following intracranial gene therapy ....................................................................... 194 Appendix 2 Widespread correction of central nervous system disease following thalamus and intracerebroventricular gene therapy in a feline model of Sandhoff disease ........ 202 Appendix 3 Intracisternal gene therapy in GM1 gangliosidosis cats ....................................... 217 vii List of Tables Introduction Table 1 Characteristic symptoms of the gangliosidoses .......................................................... 29 Chapter 1 Table 1 Treatment groups and current clinical status of AAV-treated GM1 gangliosidosis cats ............................................................................................................................ 58 Table 2 ?gal activity in brain, spinal cord, CSF, and liver of AAV-treated and untreated GM1 gangliosidosis cats ...................................................................................................... 62 Table 3 Vector copy number in brain, spinal cord, and liver of AAV-treated GM1 gangliosidosis cats ...................................................................................................... 64 Table S1 Vector copy number and ?gal activity in AAV-treated GM1 gangliosidosis cats and their offspring ............................................................................................................. 77 Chapter 2 Table 1 Hex activity in brain, spinal cord, CSF, sciatic nerve, and pituitary of AAV-treated and untreated Sandhoff disease (SD) cats. ................................................................. 91 Table S1 Vector copy number in brain, spinal cord, sciatic nerve, and pituitary of AAV-treated Sandhoff disease (SD) cats ....................................................................................... 101 Table S2 Quantification of storage material in the CNS of Sandhoff disease (SD) cats 16 weeks after treatment........................................................................................................... 104 Chapter 3 Table 1 Treatment groups and survival for long term AAV-treated Sandhoff disease (SD) cats ............................................................................................................................ 122 viii Table 2 HexA activity in the nervous system and periphery of long term AAV-treated Sandhoff disease (SD) cats ...................................................................................... 125 Table 3 Vector copy number in the brain and spinal cord of long term AAV-treated Sandhoff disease (SD) cats. ..................................................................................................... 130 Table 4 Peripheral disease symptoms and seizure episodes reported in AAV-treated Sandhoff disease (SD) cats, untreated SD cats, and normal cats ............................................. 134 Table 5 Cause of death for AAV-treated Sandhoff disease (SD) cats ................................... 136 Chapter 4 Table 1 Treatment groups for post-symptomatic AAV-treated Sandhoff disease (SD) cats . 152 Table 2 Hex activity in the nervous system, CSF, liver, muscle, and heart of AAV-treated and untreated Sandhoff disease (SD) cats ....................................................................... 159 Table S1 Peripheral disease symptoms and seizure activity in post-symptomatic AAV-treated Sandhoff disease (SD) cats and untreated controls .................................................. 164 Table S2 Vector copy number in nervous system, liver, muscle, and heart of post-symptomatic AAV-treated Sandhoff disease (SD) cats ................................................................. 165 Appendix 1 Table 1 HexA and AAV vector distribution in the nervous system, liver, and muscle of Sandhoff disease (SD) cats treated with thalamus or DCN only injections of AAV vectors ...................................................................................................................... 199 Appendix 2 Table 1 Hex activity in brain and spinal cord of thalamus and ICV AAV-treated Sandhoff disease (SD) cats and untreated controls .................................................................. 210 Table 2 Quantification of storage material in the CNS of Sandhoff disease (SD) cats 16 weeks after thalamus and ICV AAV-treatment .................................................................. 213 ix Appendix 3 Table 1 ?gal activity and AAV vector copy number in the brain and spinal cord of GM1 gangliosidosis cats 6 weeks after IC or thalamus and DCN injections of AAV vector .................................................................................................................................. 220 x List of Figures Introduction Figure 1 Structure of GM1 ganglioside ................................................................................... 23 Figure 2 GM1 and GM2 ganglioside catabolic pathway ........................................................ 24 Figure 3 Degradation of membrane bound GM1 ganglioside ................................................ 25 Figure 4 Hexosaminidase isozyme subunit composition and substrate specificity ................. 26 Figure 5 Posttranslational processing of Hexosaminidase ?- and ?-subunits. ......................... 27 Figure 6 Correlation between residual ?-gal activity and clinical disease onset ..................... 28 Figure 7 Characteristic pathological changes in gangliosidosis diseases ................................ 30 Figure 8 AAV genome and production of recombinant AAV vectors .................................... 31 Figure 9 Lysosomal enzyme trafficking and cross-correction ................................................. 32 Figure 10 Anterograde and retrograde transport after AAV injection ..................................... 33 Chapter 1 Figure 1 Therapeutic enzyme distribution in the CNS of GM1 gangliosidosis cats after AAVrh8 treatment .................................................................................................... 60 Figure 2 Storage in the CNS of GM1 gangliosidosis cats 16 weeks after treatment ............... 66 Figure 3 Normalization of lysosomal hexosaminidase activity in the CNS and liver of GM1 gangliosidosis cats 16 weeks after treatment ............................................................ 67 Figure 4 Survival and clinical progression of AAV-treated GM1 gangliosidosis cats ............ 68 Figure 5 ?gal activity in the CNS and CSF of GM1 gangliosidosis cats treated long term .... 70 xi Figure 6 MRI evaluation of GM1 gangliosidosis cats ............................................................. 72 Figure S1 ?gal distribution in the CNS of GM1 gangliosidosis cats 16 weeks after AAV1 treatment ................................................................................................................... 73 Figure S2 Clinical progression of AAV1-treated GM1 gangliosidosis cats .................................... 74 Figure S3 Hypereosinophilic neurons in the cortex of an AAV-treated GM1 gangliosidosis cat ................................................................................................................................... 75 Figure S4 Hexosaminidase activity in the liver of GM1 gangliosidosis cats treated long term 76 Chapter 2 Figure 1 Therapeutic enzyme distribution in the CNS of Sandhoff disease (SD) cats 16 weeks after AAV treatment ................................................................................................. 89 Figure 2 Ratio of Hex isozymes separated by DEAE cellulose ion exchange chromatography ................................................................................................................................... 90 Figure 3 Serum antibody titers to the AAVrh8 vectors in AAV-treated Sandhoff disease (SD) cats ............................................................................................................................ 93 Figure 4 HPTLC of glycosphingolipids in the CNS of Sandhoff disease (SD) cats 16 weeks after treatment ........................................................................................................... 94 Figure 5 Quantification of therapeutic enzyme activity and storage material in the CNS of Sandhoff disease (SD) cats 16 weeks after treatment ............................................... 95 Figure 6 Lysosomal ?-mannosidase activity in the CNS and pituitary of Sandhoff disease (SD) cats 16 weeks after treatment ........................................................................... 97 Figure 7 Clinical progression of AAV-treated and untreated Sandhoff disease (SD) cats ...... 98 Figure S1 Therapeutic enzyme distribution in the cervical spinal cord of Sandhoff disease (SD) cats 16 weeks after AAV-treatment ............................................................... 100 Figure S2 Histological abnormalities in the CNS of AAV-treated and untreated Sandhoff disease (SD) cats ..................................................................................................... 102 Figure S3 HPTLC of glycosphingolipids in the CNS of Sandhoff disease (SD) cats 16 weeks after treatment ......................................................................................................... 103 xii Figure S4 Quantification of myelin enriched lipids in the CNS of Sandhoff disease (SD) cats 16 weeks after treatment ......................................................................................... 105 Chapter 3 Figure 1 Therapeutic enzyme distribution in the CNS of long term AAV-treated Sandhoff disease (SD) cats ..................................................................................................... 123 Figure 2 HexA:total Hex ratio in the nervous system, liver, and muscle of long-term AAV- treated Sandhoff disease (SD) cats ......................................................................... 127 Figure 3 HexA activity in the nervous system and periphery of Sandhoff disease (SD) cats treated with the full dose of AAV-fHEXA and AAV-fHEXB ................................. 128 Figure 4 Correlation between Hex distribution and clearance of ganglioside storage material in the brain and spinal cord of long term AAV-treated Sandhoff disease (SD) cats .... ................................................................................................................................. 131 Figure 5 Survival and clinical progression of long term AAV-treated Sandhoff disease (SD) cats .......................................................................................................................... 132 Figure 6 MRI evaluation of Sandhoff disease (SD) cats treated with AAV long term ......... 137 Chapter 4 Figure 1 Survival and clinical progression of post-symptomatic AAV-treated Sandhoff disease (SD) cats and untreated SD cats.............................................................................. 153 Figure 2 Neurological exam performance of post-symptomatic AAV-treated Sandhoff disease (SD) cats versus untreated SD cats ............................................................. 155 Figure 3 MRI evaluation of a late post-symptomatic AAV-treated Sandhoff disease (SD) cat and untreated controls ............................................................................................. 156 Figure 4 Therapeutic enzyme distribution in the CNS of post-symptomatic AAV-treated Sandhoff disease (SD) cats ..................................................................................... 157 Figure 5 Storage in the CNS of post-symptomatic AAV-treated Sandhoff disease (SD) cats and untreated SD cats.............................................................................................. 161 Figure 6 Lysosomal ?-mannosidase activity in the CNS and liver of Sandhoff disease (SD) cats treated after disease onset ................................................................................ 163 xiii Appendix 1 Figure 1 Therapeutic enzyme distribution in the CNS of Sandhoff disease (SD) cats treated with thalamus or DCN injections only injections of AAV vectors ....................... 197 Figure 2 Survival of Sandhoff disease (SD) cats treated with thalamus only or DCN only injections of AAV vectors ...................................................................................... 201 Appendix 2 Figure 1 Therapeutic enzyme distribution in the CNS of Sandhoff disease (SD) cats after thalamus and ICV AAV treatment .......................................................................... 207 Figure 2 Therapeutic enzyme distribution in the cervical spinal cord of thalamus and ICV AAV-treated Sandhoff disease (SD) cats followed long term ................................ 209 Figure 3 HPTLC of glycosphingolipids in the CNS of Sandhoff disease (SD) cats 16 weeks after thalamus and ICV AAV-treatment ................................................................. 212 Figure 4 Lysosomal ?-mannosidase activity in the CNS of Sandhoff disease (SD) cats 16 weeks after thalamus and ICV AAV treatment ...................................................... 214 Figure 5 Clinical progression of short term thalamus and ICV AAV-treated Sandhoff disease (SD) cats ................................................................................................................. 215 Figure 6 Survival of thalamus and ICV AAV-treated Sandhoff disease (SD) cats followed long term ................................................................................................................ 216 Appendix 3 Figure 1 ?gal distribution in the CNS of GM1 gangliosidosis cats 6 weeks after IC or thalamus and DCN injections of AAV vector ........................................................ 219 xiv List of Abbreviations 4MU 4-methylumbelliferone AAV Adeno-associated virus Ab Antibody BBB Blood brain barrier ?gal ?-galactosidase CBA Hybrid cytomegalovirus enhancer/chicken ?-actin promoter CNS Central nervous system CSF Cerebrospinal fluid DCN Deep cerebellar nuclei DRG Dorsal root ganglia EPS Early post-symptomatic ER Endoplasmic reticulum ERT Enzyme replacement therapy GA1 Asialo GM1 ganglioside GA2 Asialo GM2 ganglioside GM1 GM1 ganglioside (ganglioside mono sialic acid one) GM2 GM2 ganglioside (ganglioside mono sialic acid two) GM2A GM2 activator protein gene GM2AP GM2 activator protein GLB1 ?-galactosidase gene xv H&E Hematoxylin and eosin Hex ?-N-acetyl-hexosaminidase HexA Hexosaminidase A isozyme HexB Hexosaminidase B isozyme HexS Hexosaminidase S isozyme HEXA Hexosaminidase ?-subunit gene HEXB Hexosaminidase ?-subunit gene HPTLC High performance thin layer chromatography HSCT Hematopoietic stem cell transplantation IC Intracisternal ICV Intracerebroventricular IT Intrathecal ITR Inverted terminal repeat IV Intravenous LOD Limit of detection LPS Late post-symptomatic LSD Lysosomal storage disease M6PR Mannose 6 phosphate MCB Membranous cytoplasmic bodies MRI Magnetic resonance imaging MPS Mucopolysaccharidosis MUG 4-methylumbelliferone-N-acetyl-?-D-glucosaminide MUGS 4-methylumbelliferone-6-sulfa-2-Acetoamido-2-Deoxy-?-D-Glucopyranoside xvi Naphthol naphthol AS-BI-N-acetyl-B-D-glucosaminide NHP Non-human primate ORF Open reading frame PAS Periodic Acid Schiff PN Peripheral nerve PNS Peripheral nervous system rAAV Recombinant AAV SD Sandhoff disease SRT Substrate reduction therapy TSD Tay-Sachs disease vg Vector genomes WPRE Woodchuck Hepatitis Virus posttranslational regulatory element Xgal 5-bromo-4-chloro-3-indolyl-?-D-galactopyranoside 1 Introduction Lysosomal Storage Diseases The lysosome catabolizes marcomolecules to their constituent monomeric components. Lysosomal enzymes responsible for catabolism are primarily soluble acid hydrolases located in the lumen and often require a non-enzymic activator protein for function. Lysosomal storage diseases (LSDs) comprise a group of over 50 distinct inherited metabolic diseases, which collectively have an estimated frequency of 1 in 7000-8000 live births1, 2. LSDs are characterized by accumulation of un-degraded macromolecules in lysosomes. The majority of LSDs result from genetic deficiency of a lysosomal enzyme or it?s activator protein, but a minority result from deficiency in the activation or trafficking of lysosomal enzymes3. Approximately 60% of LSDs are neuronopathic and are collectively a common cause of neurodegeneration in children. Non-neuronopathic LSDs are characterized by involvement of various visceral organs and tissues, but not the central nervous system (CNS). The gangliosidoses are neuronopathic LSDs and result from accumulation of ganglioside inside lysosomes: GM1 ganglioside (GM1) accumulation causes GM1 gangliosidosis and GM2 ganglioside (GM2) accumulation causes the GM2 gangliosidoses. History of the gangliosidoses The first clinical description of what is now known as GM2 gangliosidosis was in 1881 by the British ophthalmologist, Warren Tay, who described a cherry red-spot on the retina of a 1- year old child with psychomotor degeneration4. Further characterization came in 1896 when the 2 American neurologist, Bernhard Sachs, described distended cytoplasm in neurons from affected Jewish patients5, 6. In the 1930s the German biochemist, Ernst Klenk, identified the stored substrate in brain tissue of GM2 gangliosidosis patients as a sialic acid containing glycolipid, which was termed ganglioside due to a high concentration in ganglion cells7. The stored substrate was identified as GM2 in 1962 following isolation of the major storage product from the brain of an infantile GM2 gangliosidosis patient, and lysosomal Hexosaminidase (Hex) was suspected as the deficient enzyme8. Following the identification of different Hex isozymes in 1968, then the characterization of Hex isozyme subunit compositions in 1973, and the identification of a GM2 activator protein (GM2AP) in 1978, it is now known that GM2 gangliosidosis has three disease subtypes9-11. Tay-Sachs disease (TSD), Sandhoff disease (SD), and GM2AP deficiency result from defects in the Hex ?-subunit, Hex ?-subunit, and GM2AP, respectively. GM1 gangliosidosis was first described in 1959, but was not recognized as a distinct disease. Instead it was termed ?Tay-Sachs disease with visceral involvement?12. Between 1963 and 1965 biochemical analyses from several GM1 gangliosidosis patients revealed the disease to be a specific entity with GM1 as the stored substrate13, 14. ?-galactosidase (?gal) was identified as the deficient enzyme in 196515. Ganglioside Structure, Catabolism and Function Gangliosides are one of four classes of glycosphingolipids, the major glycolipid in mammals. Gangliosides are most abundant in neurons and are primarily located in the outer leaflet of the plasma membrane, particularly in regions of nerve endings and dendrites. The highest ganglioside content is in brain gray matter. A small proportion of ganglioside is also present in endoplasmic reticulum (ER), mitochondria, and glial cells. A hydrophobic ceramide 3 backbone anchors the molecule to the membrane and is linked via a terminal hydroxyl group to a hydrophilic oligosaccharide chain with or without sialic acid residues (Figure 1)16, 17. Ceramide is formed at the cytosolic face of the ER and is transported to the Golgi apparatus. Ceramide glucosyltransferase transfers a glucose residue from UDP-glucose to ceramide on the cytosolic face of the Golgi to form glucosylceramide. Lactosyl ceramide and higher gangliosides are synthesized on the luminal side of the Golgi through the action of membrane bound glycosyltransferases, which catalyze the stepwise addition of monosaccharides to the growing glycan chain16, 17. Ganglioside catabolism takes place in the lysosome (Figure 2). Following endocytosis from the plasma membrane, gangliosides become components of the lysosomal membrane through vesicular flow. Acid hydrolases in the lysosol sequentially cleave sugar residues from the non-reducing end of gangliosides. Oligosaccharide chains of one to four residues are not accessible to acid hydrolases, so activator proteins are required to mediate interaction between the membrane bound substrate and water-soluble enzyme (Figure 3)17. The structural pattern of gangliosides changes with cell development, differentiation, and ontogenesis. Although the function of individual gangliosides is poorly understood, there is evidence they are important in cell differentiation, recognition, and adhesion as well as in the regulation of growth factor receptors, regulation of protein kinases, synaptic transmission, and cellular calcium homeostasis18-20. Additionally, exogenously added gangliosides can promote survival and repair of damaged neurons21, 22. Due to a high concentration at nerve endings and dendrites, gangliosides are also implicated in neuritogenesis. Specifically, GM1 promotes neurite outgrowth in a number of cell culture systems, and concentration of GM1 and GM2 correlates with ectopic dendrite growth in ganglioside storage diseases19, 23. 4 Enzyme deficiency causing GM1 Gangliosidosis Lysosomal ?-gal (EC 3.2.1.23) hydrolyzes terminal ?-1,3- or ?-1,4-linked galactose residues from GM1, asialo GM1 (GA1), and numerous glycoconjugates24. Defective ?gal causes GM1 gangliosidosis, which has an estimated prevalence of 1 in 244, 000-384,000 conceptions in the general population1. A higher frequency is reported in the Maltese Islands (1 in 370025), the Gypsy population (1 in 10,00026), Southern Brazil (1 in 17,00027, 28), and Japan (high but unknown incidence of adult onset disease29). The 2.4 kilobase ?gal cDNA is composed of 16 exons encoded by GLB1 on chromosome 3p21.33. ?gal is synthesized as a 677 amino acid (84 kilodalton) precursor with a 23 amino acid N-terminal signal peptide. The signal peptide is cleaved off upon entry into the ER where the precursor is glycosylated with high-mannose oligosaccharides. In the Golgi, ?gal is phosphorylated on mannose residues and binds the mannose-6-phosphate receptor (M6PR) for targeting to the lysosome. The ?gal precursor is C-terminally processed in the endodomal/lysosomal compartment within two hours to a mature 64 kilodalton polypeptide. C- terminal processing and stability depend on the association of ?gal with cathepsin A. The majority of lysosomal ?gal exists in a 680 kilodalton complex containing four ?gal monomers and eight cathepsin A protomers. In the absence of cathepsin A, precursor ?gal is produced normally and is functionally identical to the mature form, but the half-life is reduced by one- tenth due to excessive intralysosomal degradation. A 1.27 megadalton ternary complex also exists, which additionally contains neuraminidase and N-acetylgalactosamine-6-sulfate-sulfatase. The two complexes are present in a 10:1 ratio in the lysosome30. ?gal has a pH optimum of 4.5 to 4.75 and is activated by chloride ions. Hydrolytic activity against GM1 requires the presence of the saposin B activator protein or the GM2AP (Figure 3)31. 5 At least 91 mutations resulting in GM1 gangliosidosis have been reported on GLB1. Mutation types include missense/nonsense, small deletions, small insertions, and splicing defects. Common mutations are found in certain ethnic groups, such as the R208C mutation in American infantile onset patients32 and the I51T mutation in Japanese adult onset patients29. The latter mutation results in an enzyme that is not phosphorylated, so only a small amount of ?gal reaches the lysosome. Enzyme Deficiency Causing GM2 Gangliosidosis Defective lysosomal ?-N-acetylhexosaminidase (Hex) (EC 3.2.1.52) causes GM2 gangliosidosis. HEXA and HEXB encode the Hex ?- and Hex ?-subunit, respectively. The Hex subunits dimerize to form two major isozymes, HexA (??) and HexB (??), as well as a minor isozyme, HexS (??). Hex removes terminal ?-1,4-N-acetyl-glucosamine and N-acetyl- galactosamine residues from GM2, asialo GM2 (GA2), and numerous other glycoconjugates. The GM2AP is required for Hex activity to facilitate interaction between the lipid based substrate and water soluble hydrolase. HexA and HexB have overlapping substrate specificities, but only HexA is capable of hydrolyzing charged substrates, including GM2 (Figure 3). HexS is an unstable isoform present at very low levels. It is considered to have negligible catalytic activity towards most substrates, but may play a role in the degradation of glycosaminoglycans (Figure 4). GM2 gangliosidosis has three disease forms: 1) TSD results from HEXA mutations and is associated with deficient HexA, but normal HexB, 2) SD results from HEXB mutations and is associated with deficient HexA and HexB, and 3) GM2AP deficiency results from mutations in GM2A, causing the inability to form a GM2/GM2AP complex11. TSD and SD have an estimated prevalence of 1 in 201,000-244, 000 and 1 in 322, 600-384,000 conceptions, 6 respectively1. Higher frequencies have been reported in some ethnic groups, such as the Ashkenazi Jewish population, with a TSD carrier frequency of 1 in 360033. However, most North American infantile onset TSD patients are now of non-Jewish ancestry due to successful carrier screening within the high risk population. HEXA and HEXB are located on chromosomes 15q23-q24 and 5q13, respectively. Both genes are composed of 14 exons and the gene products exhibit 57% homology in amino acid sequence. The ?-subunit is synthesized as a 529 amino acid pro-polypeptide (~68 kilodaltons) with a 23 amino acid signal sequence. The ?-subunit is synthesized as a 556 amino acid pro- polypeptide (~68 kilodaltons) with a 43 amino acid signal sequence. Figure 5 shows the post- translation modifications to Hex pro-polypeptides. Upon entry into the ER, the signal sequence is cleaved off and the pro-polypeptides are glycosylated on asparagine residues with high-mannose oligosaccharides. Dimerization of ?-subunits to form HexB occurs rapidly in the ER. As HexB is the most stable isozyme, ?-subunit monomers must accumulate for ~5 hours to provide a high enough concentration to force newly processed ?-subunit monomers to form the less stable HexA. In the Golgi, Hex isozymes are phosphorylated at mannose residues and directed to the lysosome by association with the M6PR. Hex isozyme precursors undergo final proteolytic processing in the lysosome to generate an ?-subunit of ~56.6 kilodaltons consisting of two polypeptides and a ?-subunit of ~55.9 kilodaltons consisting of three polypeptides33-35. At least 123 HEXA mutations leading to TSD and 36 HEXB mutations leading to SD have been characterized. Mutation types include small insertions, small deletions, gross deletions, missense/nonsense, base substitutions, and base insertions causing incorrect splicing. Common mutations occur in certain ethnic groups. For example, the 1278ins4 mutation, which introduces a premature stop codon, accounts for up to 80% of mutations in Ashkenazi Jewish 7 infantile onset TSD patients36. A 16 kilobase deletion extending from the promoter region of HEXB in to intron five is common in French and French-Canadian SD patients, and it may account for up to one third of HEXB mutations35. Clinical Progression Neuronal storage of ganglioside leads to progressive psychomotor degeneration. Infantile-, juvenile-, and adult-onset clinical disease forms are recognized and are characterized by the age of onset and progression of clinical symptoms, which are largely determined by the residual level of enzyme activity (Figure 6)35, 37. GM1 and GM2 gangliosidosis share many clinical features (Table 1). In the most common infantile-onset disease form, developmental arrest occurs between birth and 6 months old and progressive psychomotor deterioration ensues. Macular cherry red spots on the ocular fundus, seizures, increased startle response to sounds, and blindness are characteristic. Children become vegetative in late stage disease, and death typically occurs before age five. Additionally, dysmorphism, hepatosplenomegaly, and skeletal dysplasia (dystosis multiplex) are common features of infantile GM1 gangliosidosis. Peripheral tissue abnormalities are less commonly reported in GM2 gangliosidosis patients despite the storage of neutral substrates in peripheral tissues of SD patients. GM2AP deficiency has similar clinical characteristics to early onset TSD35, 38. Symptoms of juvenile-onset gangliosidosis are similar to those seen in infantile-onset patients, but progression is slower with death typically occurring at 10 to 15 years old. Adult-onset gangliosidosis occurs from childhood to adulthood and clinical phenotype varies widely. Gait and speech disturbances, abnormal posture, progressive dystonia, extrapyramidal signs, and motor neuron disease (muscle wasting, weakness, and fasciculation) 8 are common symptoms. Additionally, mild intellectual impairment and skeletal abnormalities (flattening of vertebral bodies) are common symptoms in GM1 gangliosidosis patients24, 29, 38, 39. Psychosis is common in GM2 gangliosidosis patients, but intelligence is not impaired35. Mild peripheral and autonomic nervous system dysfunction has been reported in late-onset SD patients and less commonly in TSD patients. Symptoms include: decreased sympathetic skin responses40, esophageal dysmotility41, chronic diarrhea42, urinary incontinence43, loss of libido41, cardiac dysrhythmia41, decreased nerve conduction velocity41, 43, 44, muscle denervation41, 44, 45, and storage in myenteric plexus neurons44, 45 as well as in tissues from appendix45, muscle44, and rectal submucosa41. Disease Pathology Pathologically the gangliosidoses are characterized by ganglioside accumulation in brain gray matter. GM1 and GA1 accumulate to approximately 10 times normal in GM1 gangliosidosis patients and GM2 and GA2 accumulate to 100 to 300 times normal and 20 to 100 times normal, respectively, in GM2 gangliosidosis patients. Total ganglioside storage is similar in both diseases because GM1 comprises a greater proportion of total ganglioside in the normal brain. Substrate accumulation is also evident in peripheral tissues. GM1 accumulates to approximately 20 to 50 times normal in liver of GM1 gangliosidosis patients. Also, storage vacuoles and foamy histiocytes are apparent in other peripheral tissues due to accumulation of galactose containing oligosaccharides24, 38. Visceral tissues of TSD patients often show no pathologic signs. SD patients have GA2 accumulation in the liver as well as storage vacuoles and foamy histiocytes in other peripheral tissues due to accumulation of globoside and neutral substrates35. 9 Computed tomography (CT) scans and magnetic resonance imaging (MRI) show diffuse atrophy of the CNS as well as myelin loss24, 38, 46. The neurodegeneration that characterizes gangliosidosis disease progression is likely a result of several mechanisms. Ganglioside accumulation causes neuronal hypertrophy with cytoplasmic vacuolation, loss of Nissl substance, and margination of nuclei. Neuronal and glial cytoplasmic inclusions consist of characteristic membranous cytoplasmic bodies (MCBs), which are composed of multiple concentric lamellae containing cholesterol, phospholipid and ganglioside (Figure 7a). Increased lysosomal storage may disrupt other catabolic processes, depriving the cell of precursors and leading to secondary dysfunction in other pathways. Also, there may be reduced entry of materials into swollen lysosomes, causing build-up elsewhere in the cell with disruptive consequences23. Soma-dendritic changes have been described in human6, 35 and feline gangliosidosis diseases. In feline models, the development of meganeurites- massive swellings of the axon hillock region- are related to disease severity (Figure 7b). Some meganeurites show growth of ectopic dendrites with evidence of aberrant synapse formation, which could alter normal electrophysiology and neuronal connections. Ectopic dendrite growth appears to be correlated with GM2 storage and can also occur at the axon hillock in the absence of meganeurite formation23, 47-52. In an induced ?-mannosidosis feline model, ectopic neutrites and their synaptic connections were not reversible even though other cellular changes were normalized following treatment53. Also apparent in feline models are axonal spheroids- swellings containing organelles- which develop distal to the axon hillock region and may affect action potential propagation (Figure 7c)23, 51, 54-56. Axonal spheroids may be caused by a block in retrograde transport due to disruption of the cell body, therefore depriving axons of important components. There is a 10 correlation between disease severity and the incidence of axonal spheroids in GABAergic neurons. GABAergic neurons, particularly Purkinje cells, likely have a high incidence of axonal spheroids because they have a high firing rate requiring fast turnover of axonal components. Loss of inhibitory circuits could promote the characteristic seizures in late stage disease. The block in axonal transport is also thought to contribute to myelin loss, as is loss of oligodendrocytes57. Increased neuronal apoptosis has been reported in human and mouse disease58. In mice, perturbation of calcium homeostasis causes an ER stress response as well as mitochondrial apoptotic signaling, which contribute to apoptosis20, 59, 60. In SD, mice neuroinflammation is high in areas of marked neuronal death58, 61, 62. It is hypothesized that macrophage/microglial cells initiate an inflammatory response as they engulf and destroy apoptotic neurons. However, the inability to degrade storage material results in a continued inflammatory response and further microglial expansion62. Other mechanisms thought to promote disease progression include activation of autophagy leading to mitochondrial dysfunction63 and neuronal accumulation of synucleins64. Thus, several different mechanisms likely contribute to disease progression. Feline Gangliosidosis Models Knock out mouse models of GM1 gangliosidosis65-67 and SD68 share many similarities with the human diseases and are invaluable for the initial testing of therapies. However, feline models provide important advantages over mice for translational research. The cat brain is 75 times larger than the rodent brain and only 10 to 20 times smaller than a human child?s brain, therefore providing a valid indication of enzyme distribution challenges that need to be overcome in a large, complex brain. Cats have a life span that allows long term measurements 11 and a body size that facilitates clinical manipulations and evaluations, surgical interventions, and frequent sampling of tissues and body fluids. Unlike mice, cats have a heterologous genetic background providing an indication of the variability in treatment outcome and immune responses that can be expected in humans69. The feline gangliosidoses are highly analogous to the human diseases, therefore providing an ideal large animal model to test therapeutic strategies. Feline GM1 gangliosidosis was first described in Siamese cats70 and in a domestic cat71. The Siamese model is maintained at Auburn University, but is now outbred and comprises a mixed breed colony. First described in 1971, the feline model is a close replica of human juvenile-onset GM1 gangliosidosis70, 72. The genetic mutation is G1448C causing an Arg483Pro substitution in exon 14 resulting in production of a mutant protein that does not reach the lysosome73. ?-gal activity is ?0.15-fold normal in the CNS, kidney, skin, and cultured fibroblasts. Initial reports described disease onset at 2 to 3 months old with tremors of the head and pelvic limbs followed by generalized dysmetria, severe balance disturbances, and spastic paraplegia, which defines the humane endpoint at 7 to 8 months old. Tissues show widespread neuronal degeneration characterized by varying degrees of swelling, cytoplasmic vacuolation, lack of Nissl substance, and margination of nuclei. Total ganglioside content is 2 to 3 times normal. GM1 ganglioside content is approximately 8 times normal and is stored in MCBs characteristic of the human disease. Gliosis and neuronophagia are evident in some areas of severe neuronal degeneration, but there is little neuronal loss. Peripherally, vacuolated macrophages and hepatocytes are evident, but cats do not display hepatosplenomegaly. Testes of pubescent cats are devoid of mature spermatozoa. Feline GM2 gangliosidosis has been described in American shorthair domestic74, Korat75, Japanese domestic76, and Burmese breeds77. The American domestic shorthair colony is 12 maintained at Auburn University. First described in 1977, it is a close replica of human infantile- onset SD74, 78. The genetic mutation is a 25 base pair inversion at the carboxyl terminus of exon 14 causing a translation stop codon 8 amino acids premature, which results in protein levels much lower than normal79. Less than 0.02-fold normal Hex activity is found in the CNS, liver, and fibroblasts of SD cats. Initial reports described the onset of head tremors at 4 to 10 weeks old, followed by ataxia and then paresis, which defines the humane endpoint at ~4.5 months old. Neurons throughout the CNS are distended, lack Nissl substance and contain MCBs characteristic of the human disease, with cerebellar Purkinje cells particularly affected. Total ganglioside content is 2 to 3 times normal and GM2 content is approximately 700 times normal. Unlike the human disease, there is little evidence of neuronal loss or gliosis. Peripherally, membrane bound storage vacuoles are present in visceral organs throughout the body. Treatment Options Although the first description of GM2 gangliosidosis was in 1881, the gangliosidoses as a group remain untreatable. Effective treatments will require the delivery of therapeutic enzyme to all cells with lysosomal enzyme deficiency. There are two observations that encourage the ongoing efforts to find an effective therapy. Firstly, a small proportion of lysosomal enzyme is secreted from the cell where it can be taken up by neighboring cells by the M6PR present in the plasma membrane80-82. Therefore, a small proportion of donor cells are able to restore lysosomal enzyme activity in a larger cell population (Figure 9). This cross-correction mechanism forms the basis of therapeutic approaches currently under investigation. Secondly, tissues of asymptomatic individuals heterozygous for gangliosidosis have as low as 0.11- to 0.20-fold 13 normal lysosomal enzyme activity, and adult onset patients can have activity at only 0.02- to 0.04-fold normal83. Therefore, very low level enzyme restoration may be therapeutic. Enzyme replacement therapy (ERT) ERT involves delivery of a functional enzyme via intravenous (IV) or cerebrospinal fluid (CSF) routes. ERT is approved for use to treat peripheral symptoms of six non-neuronopathic LSDs including Gaucher disease type I, Fabry disease, Mucopolysaccharidosis (MPS) types I, II and VI, and Pompe disease. ERT has also effected improved neurological function in a number of neuronopathic LSD animal models (reviewed in84), but not in gangliosidosis animals. Intracerebroventricular (ICV) administration of human HexA to SD mice only resulted in a 0.13- fold survival increase over untreated85, and IV administration of human HexA to SD cats did not result in survival benefit. However, HexA activity in SD cats was at least 0.44 fold normal in all 11 peripheral tissues analyzed with subsequent reductions in GM2 and GL4 globoside. Treatment also resulted in up to 1.74-fold normal brain HexA activity, but there was no substantial decrease in brain ganglioside content, possibly because the enzyme was not taken up by neurons86. ERT requires regular infusions, and the high cost of products makes it prohibitively expensive for rare diseases such as the gangliosidoses. Cell based therapies Hematopoietic stem cell therapy (HSCT) from healthy donors has been tested in a number of neuronopathic LSDs, but therapeutic benefit has largely been limited to a small subset of late-onset patients. Transfused donor derived HSCs provide metabolically competent progeny cells, such as macrophages and microglia, which migrate to affected tissues and provide a source 14 of functional enzyme. Better outcome is expected following early transplantation as donor engraftment can take several months, during which time disease progresses. HSCT success has been most notable in patients with Hurler?s syndrome and Globoid Cell Leukodystrophy. Outcome has been more variable in other LSDs with efficacy seen in some patients, but failure to improve symptoms despite engraftment in others87-89. HSCT showed moderate therapeutic efficacy in gangliosidosis mouse models90-92, resulting in survival increases of up to ~2-fold over untreated. However, HSCT failed to benefit a GM1 gangliosidosis Portuguese Water dog93, a SD cat94, and infantile onset GM195 and GM2 gangliosidosis patients96 transplanted before disease onset. The variable success of HSCT between different LSDs may depend on how efficiently leukocytes release a particular lysosomal enzyme. For example, brain microglia release more ?-mannosidase versus Hex in culture97. It may also depend on the relative disruption of the blood brain barrier (BBB) between different LSDs, which would affect infiltration of HSC derived cells. Transplantation of genetically modified HSCs proved successful in human patients with early onset Metachromatic Leukodystrophy (onset in the second year) transplanted pre- symptomatically98. Stable engraftment was demonstrated beginning 1 month after transplant, and therapeutic enzyme activity in CSF was comparable to healthy donors for up to 2 years. At 39 months old, patient one required a single aid to walk, but otherwise showed continuous motor development, normal IQ, and improvements in existing peripheral neuropathy. Patients two and three remained fully asymptomatic at 30 and 25 months old. The outcome is highly encouraging for the prospect of treating neurodegenerative LSDs. 15 Substrate reduction therapy (SRT) SRT aims to decrease the amount of deficient enzyme substrate synthesized by cells. Two imino sugars, N-butyldeoxynojirimycin (NB-DNJ;Miglustat) and N- butyldeoxygalactonojirimycin (NB-DGJ), are competitive inhibitors of ceramide-specific glucosyltransferase (GlcT), the enzyme that catalyzes the first step in glycosphingolipid biosynthesis, and have demonstrated moderate therapeutic benefit in gangliosidosis mouse models, resulting in up to a 1.36-fold survival increase versus untreated99-102. Miglustat is in clinical use as an effective treatment for peripheral symptoms of non-neuronopathic Gaucher disease type I and Niemann-Pick type C, but has not proven effective to treat neuronopathic symptoms of Gaucher disease type III. Miglustat did not provide any measurable benefit for infantile-onset GM2 gangliosidosis patients103. Outcome was variable in juvenile onset GM2 gangliosidosis patients, with some continuing to show neurological deterioration104, 105 and others demonstrating disease stabilization106, 107. However, stable periods are part of the natural disease course, so it is hard to assess treatment benefit in such a heterogeneous group with short follow up periods. In mouse models, SRT used in combination with cell therapy provided a synergistic therapeutic effect over the monotherapies91, 108, 109. Chaperone therapy Pharmacological chaperones are small molecules that bind and stabilize the native folded conformation of mutant enzyme that retains partial catalytic function. Once inside the acidic lysosome, the enzyme substrate will displace the chaperone. However, mutations in gangliosidosis patients are heterogeneous and complex. It is estimated that only 30% of GM1 gangliosidosis patients have unstable, but catalytically active enzyme that would be a candidate 16 for chaperone therapy110. The galactose derivative, N-octyl-4-epi-?-valienamine (NOEV), has shown moderate efficacy in GM1 gangliosidosis mouse models resulting in up to a 1.3 fold survival increase versus untreated110, 111. Additionally, NOEV can increase ?gal activity in fibroblasts from juvenile-onset patients as well as 30% of infantile-onset patients111, 112. N-nonyl- DGJ (NN-DGJ), a long alkyl-chain iminosugar derivative can also act as a chaperone in fibroblasts from GM1 gangliosidosis cats and human patients113. However, cell based studies may not be indicative of in vivo effect. Cell studies on human GM2 gangliosidosis fibroblasts showed that Pyrimethamine (PYR) could enhance HexA activity in some mutation types, but two subsequent phase I/II clinical trials did not report any notable success in late-onset patients114 115, 116. In one trial HexA activity did not approach the theoretical 0.1-fold normal level thought to be therapeutic, and the other trial was suspended due to the extent of side effects experienced by most patients115. Adeno-associated virus gene therapy Gene therapy has the potential to provide a permanent source of deficient enzyme following cellular transduction. Adeno-associated virus (AAV) vectors have become the vectors of choice for gene delivery to the CNS due to their efficiency at transducing neurons leading to long term expression of therapeutic genes with no apparent toxicity and limited inflammation at the site of injection117, 118. AAV is a non-pathogenic, single stranded DNA virus belonging to the Dependovirus genus of the Parvoviridae family. The AAV life cycle consists of a latent phase and a lytic phase, which requires co-infection of adenovirus or herpes simplex virus for replication. The 4.7 kilobase AAV genome consists of two open reading frames (ORFs), flanked by two inverted terminal repeats (ITRs). The rep gene encodes proteins required for excision, 17 replication and integration, and the cap gene encodes proteins for capsid formation (Figure 8a). The ITRs are the only cis elements required for viral replication and packaging. Recombinant AAV gene therapy vectors are produced by replacing rep and cap with cDNA encoding a therapeutic transgene expression cassette. The recombinant genome is packaged into an AAV capsid by providing Rep and Cap proteins as well as adenovirus helper functions in trans (Figure 8b). Following AAV attachment to cell surface receptors, the virus is endocytosed and subject to endosomal processing, which is an important step for viral nuclear transport. It is thought that AAV escapes from the late endosome prior to lysosomal maturation and traffics to the nucleus, but little is known about the process that controls nuclear translocation. Inside the nucleus, AAV uncoats prior to genome conversion from a single stranded DNA to a double stranded DNA intermediate capable of expressing proteins119. Through the action of Rep proteins, wild type AAV integrates in a site specific manner in chromosome 19120. As recombinant AAVs are devoid of viral ORFs, they lose capacity to integrate and persist primarily in episomal form121 122, 123. Random integration events occur at a frequency significantly lower than the generally accepted rate for spontaneous mutation in human genes121. Therefore, the risk of insertion mutagenesis is extremely low. A unique AAV serotype is one that cannot be efficiently neutralized by serum generated against viruses of other known serotypes. AAV1 was discovered as a contaminant in simian Adenovirus stock in 1965124. The discovery of AAV2 followed in 1966, and AAV2 became the first serotype to be cloned into a bacterial plasmid in 1982125. AAV serotypes differ in their cellular tropism due to the composition of capsid proteins, which facilitate binding to cell surface receptors. AAV1 and AAVrh8 are the serotypes utilized in this dissertation research project. The 18 cellular receptor for AAV1 is N-linked sialic acid, and AAV1 has a natural tropism for CNS, heart, and skeletal muscle126. AAVrh8 was isolated from the rhesus monkey in 2004 by screening tissues for endogenous AAV sequences127. Its cellular receptors are unknown, but it has tropism for hepatocytes, muscle, and CNS, therefore displaying similar tropism to AAV1128. AAV gene therapy in LSD animal models AAV intracranial gene therapy has been tested in many different LSD mouse models with varying success. It is hard to compare outcomes between different LSD models because there are many variables between studies which could affect the extent of therapeutic enzyme and vector distribution. Following injection of AAV vectors in the CNS, enzyme and vector are disseminated from injection sites by diffusion129, axonal transport to interconnected areas (Figure 9)130-133, and CSF flow130, 134. Therefore, reports of widespread distribution have followed injection of vectors into highly interconnected structures including the thalamus135-137, deep cerebellar nuclei (DCN)137-140, lateral ventricles141, striatum137, 140, 142-144, lumbar cistern145, 146, and cisterna magna141, 146, 147. Different vector serotypes have different cellular tropisms depending on their capsid properties. As AAV2 was the first serotype to be cloned it was the predominant one used in early gene therapy studies. However, comparisons have since shown that serotypes differ in their CNS distribution capability, and that other serotypes can achieve more widespread CNS distribution than AAV2137, 139, 148-150. Different serotypes have different propensities to undergo either anterograde or retrograde axonal transport (Figure 10). For example, AAV2 favors anterograde axonal transport whereas AAV6 favors retrograde transport, which affects the distribution characteristics of each vector when injected in the striatum and thalamus151. Axonal transport of 19 vector/enzyme is affected by efficiency of uptake at the plasma membrane, so higher vector doses may promote axonal transport152. Vector purification method can impact transduction efficiency when comparing between different serotypes, and the choice of regulatory elements can affect therapeutic transgene expression153. For example, the hybrid cytomegalovirus enhancer/chicken ?-actin (CBA) promoter was shown to promote stronger expression of green fluorescent protein (GFP) compared to the cytomegalovirus (CMV) promoter154, and inclusion of the Woodchuck hepatitis virus posttranscriptional regulatory element (WPRE) increases transcript expression145. Following translation, the use of non-species specific cDNA can limit protein distribution due to immune responses against foreign sequences141. Therefore, there are many variables between different gene therapy studies that can determine outcome. Due to mechanical injury, the intracranial injection procedure in na?ve animals produces an innate local inflammatory response, which subsides over time. Following vector delivery, a neutralizing antibody (Ab) response to the AAV capsid is usually encountered, but does not correlate with CNS vector transduction efficiency, protein expression, or clinical outcome141, 155, 156. The presence of pre-existing serum Abs against AAV2, AAVrh8, and AAV9 was shown to reduce CNS vector transduction when the same serotype vector was administered in mice137, 155 and non-human primates147. However, in other studies pre-existing Abs against AAV9 did not have a substantial effect on subsequent CNS AAV9 vector transduction in dogs141 and non- human primates146. Therefore, the presence of pre-existing neutralizing Abs may or may not affect CNS vector transduction. Ab responses against the therapeutic transgene can have a detrimental effect on protein distribution if non species specific transgenes are used141, 157, or if animals do not generate any natural protein (that is, negative for cross-reactive material, or CRM-)158, 159. 20 AAV gene therapy has been highly successful in gangliosidosis mouse models. GM1 gangliosidosis mice treated neonatally with bilateral ICV injections of AAV1-mouse (m) ?gal showed several-fold normal ?gal activity throughout the brain, and normalization of GM1 storage at the 3-month pre-determined endpoint160. In a later study, all but one GM1 gangliosidosis mouse treated at 6 to 8 weeks old with bilateral injections of AAV2/1-m?gal in the thalamus and DCN survived until the 52-week experiment endpoint (untreated endpoint, 32 weeks old). ?gal activity was restored to the presumed therapeutic level (>0.1-fold normal) throughout the brain and spinal cord, with near normalization of substrate storage in the brain, and 50% clearance in the spinal cord135. SD mice injected bilaterally in the striatum and DCN or hippocampus and DCN with AAV1-HEXA + AAV1-HEXB (1:1 vector formulation) survived to a median of 681 days and 609 days, respectively (untreated SD lifespan, 131 days). In both groups histochemical staining demonstrated widespread Hex activity and striking absence of storage material throughout the entire CNS140, which was irrespective of treatment age161. However, clinical benefit decreased with increasing treatment age due to pathological changes that could not be reversed, such as demyelination and neuronal apoptosis161. Other studies have also demonstrated that clinical efficacy decreases with increasing treatment age162, 163. Whilst mouse models are invaluable for initial testing of therapeutic approaches, gene therapy must also be proven in large animal models with closer resemblance to human patients. Intracranial AAV gene therapy has been reported in the ?-mannosidosis feline model164 and MPS type I and III canine models158, 165. Vector delivery to these large animal brains has demonstrated extensive enzyme distribution and clearance of storage material, although there was often much inter animal variation. Although the previous studies in LSD animal models were not designed to 21 determine clinical efficacy, treated ?-mannosidosis cats demonstrated reversal of clinical progression and displayed only mild neurological disease at the 18-week study endpoint (untreated lifespan, 18 weeks). Of two cats followed-up long term, one reached humane endpoint at 30 weeks old and the other was euthanized at the 52-week experimental endpoint, at which time there had been no deterioration in clinical status since the 18-week time-point164. AAV gene therapy in humans Intracranial AAV gene therapy has so far been reported in human phase I and II trials for Parkinson?s disease166-168 and two neuronopathic LSDs: Canavan disease169, 170 and Batten disease (Late Infantile Neuronal Ceroid Lipofuscinosis)171. There have been no adverse safety consequences from any of the trials that could be definitively attributed to the injection procedure or AAV vectors, indicating that the approach is safe for human application. The initial stages of human clinical trials are concerned with safety, so patients are administered a dose per kilogram that is much lower than the effective dose in animals. Therefore, clinical efficacy cannot be concluded from early stage trials. In the phase I trial for Batten disease, AAV2 vectors were administered to six cortical locations in ten patients aged 3.4 to 10 years old (disease onset 2 to 4 years). Only a mild transient humoral immune response was reported in four of ten patients. One subject died of status epilepticus 49 days following vector administration. However, seizures are part of the natural disease progression and the subject had no signs of CNS inflammation, so it is likely that the seizures were not a result of vector administration. Clinical data from the 36-month follow-up period suggested that decline per year was slower compared to the untreated control group, but outcome was not significant. 22 The phase I trial for Canavan was designed to investigate long-term safety of AAV2 vector administration to six cortical locations in ten patients treated between 4 and 83 months old (disease onset 3 to 9 months old). Only a low level transient immune response was reported in three of ten patients, and no long term adverse events were reported in up to 10 years follow-up. Over a minimum follow-up period of 60 months, MRI detected decreased substrate storage, normalization of myelination in some areas, and reversal or stabilization of brain atrophy in some areas with regional and subject-specific variation. Significant clinical improvements occurred for some parameters, such as gross motor function; alertness; spasticity of the lower extremities; and seizure frequency, but not in visual tracking; motor skills; self-care; mobility; social function; and cognitive function. As expected, patients treated at an earlier age experienced the greatest magnitude of clinical improvement. This dissertation research project describes the outcome of extensive preclinical gene therapy studies in GM1 gangliosidosis and SD cats. The long term goal is to optimize an AAV gene therapy approach that can be translated to human clinical trials. Results demonstrate the effective translation of intracranial gene therapy from gangliosidosis mice to gangliosidosis cats with an ~70-fold-scale-up in brain size, which is indicative of successful scale up to the larger human brain with minimal modification of approach. 23 Figure 1. Structure of GM1 ganglioside. The hydrophobic ceramide backbone is the anchor portion of the molecule, and it consists of a sphingosine base linked via an amide linkage to a long chain fatty acid. Ceramide is linked via the terminal hydroxyl group to a hydrophilic oligosaccharide chain containing a sialic acid residue. GM1 is named as such because it contains a mono sialic acid and it is the first molecule in the mono sialic acid series. Cleavage of the terminal galactose residue by ?gal yields GM2, the second molecule in the ganglioside mono sialic acid series. 24 Figure 2. GM1 and GM2 ganglioside catabolic pathway. The ganglioside catabolic pathway takes place in the lysosome. Acid hydrolase enzymes and their activator proteins sequentially cleave monosaccharide residues and sialic acid from the non-reducing end of the molecule to leave the ceramide backbone. Ceramide is cleaved by ceramidase to sphingosine and fatty acid. 25 Figure 331. Degradation of membrane bound GM1 ganglioside. The saposin B activator protein or GM2 activator protein is required for in vivo cleavage of GM1 by ?gal. In the figure, ?gal binds to the negatively charged BMP-containing membrane surface. Saposin B acts as a physiological detergent that facilitates the interaction between membrane bound substrate and water soluble hydrolase enzyme, likely by disrupting the membrane surface and lifting the GM1 side chain into the ?gal active site. In a similar manner the GM2AP facilitates interaction between HexA and GM2. 26 Figure 435. Hexosaminidase isozyme subunit composition and substrate specificity. The Major Hex isozymes, HexA and HexB, have overlapping substrate specificities, but only HexA, with the aid of the GM2AP, is capable of GM2 degradation. HexS has negligible catalytic activity toward most Hex substrates, but may be important in the degradation of glycosaminoglycans as SD patients only show modest N-acetyl-galactosamine linked glycosaminoglycan storage despite the absence of the major Hex isozymes. HexA can cleave neutral or charged synthetic substrates, but HexB can only cleave neutral synthetic substrates. HexS can cleave charged synthetic substrates, but with a much lower specific activity then HexA. 27 Figure 535. Posttranslational processing of Hexosaminidase ?- and ?-subunits. Bars represent polypeptides, and filled bars represent signal sequences. Numbers below bars represent the amino acids of the termini. Numbers in bars represent asparagine residues modified by oligosaccharides. Circles represent oligosaccharides, and filled circles indicate preferred phosphorylation sites. The fragments of the mature subunits are held together by disulfide bonds (not shown). Half circles indicate oligosaccharide structures degraded in lysosomes without any effect on activity. 28 Figure 6. Correlation between residual ?-gal activity and clinical disease onset. Residual enzyme activity is positively correlated with the age of clinical disease onset. At least 0.1-fold normal enzyme activity is necessary for normal catabolism of ganglioside. The age of onset in patients expressing enzyme activity above this level is theoretically beyond the human life span. This figure is based on enzyme assay results using cultured skin fibroblasts and the synthetic fluorogenic substrate 4-methylumbelliferyl ?-galactopyranoside. Figure taken from: Chaperone therapy update: Fabry disease, GM1-gangliosidosis and Gaucher disease. Brain and development. (2013) 35: 517. 29 Table 1. Characteristic symptoms of the gangliosidoses GM1 Gangliosidosis GM2 Gangliosidosis Characteristic Infantile Juvenile Adult Infantile Juvenile Adult Onset 3-6 mo. 7 mo.- 3 yrs 3-30 yrs 3-6 mo. 2-6 yrs Child- to adulthood Age at death (years) < 2 5-15 Variable course < 5 5-15 Variable course CNS involvement Generalized Generalized Localized Generalized Generalized Localized Mental regression + + +/- + + +/- Major motor system affected Pyramidal Pyramidal Extrapyramidal A Pyramidal Pyramidal Extrapyramidal CardiomyopathyB +/- +/- +/- NR NR - Muscle atrophy - - +/- - - +/- Ocular cherry red spots + +/- - + +/- +/- Hepatosplenomegaly + +/- - - - - Dysmorphism +/- +/- - - - - Skeletal dysplasia Generalized Localized Localized - - - Visual loss + + - + + - Startle response to sound + + - + + - Seizures + + - + + - Macrocephaly rarely - - + - - Glomerular epithelial ballooning + + NR - - - Foam cells in marrow + + +/- - - - Oligosaccharide storage + + + + (SD) + (SD) + (SD) Neuronal lipidosis + + + + + + (Localized) Visceral histiocytosis + + + mild (SD) -/+ (SD) - Keratan sulfate storage + + NR - - - Globoside storage - - - + (SD) + (SD) + (SD) APyramidal signs are also present in a minority of patients. BCardiomyopathy has been reported in all GM1 gangliosidosis disease forms. Abbreviations: NR = not reported; SD = reported in Sandhoff disease patients; mo. = months 30 Figure 723, 35. Characteristic pathological changes in gangliosidosis diseases. Images are from feline gangliosidosis models. (a) Storage material accumulates in MCBs composed of concentric lamellae. (b) Meganeurites are swellings of the axon hillock region that sometimes show growth of ectopic dendrites. (c) Axonal spheroids are swellings containing organelles, which develop distal to the axon hillock region and are likely caused by a block in retrograde transport. 31 Figure 8117. AAV genome and production of recombinant AAV vectors. (a) The AAV genome consists of rep, encoding proteins for replication and packaging; cap, encoding proteins for capsid formation; and ITRs. Arrows indicate promoters for transcription of Rep and Cap genes. (b) For AAV gene therapy, rep and cap are replaced by a therapeutic transgene cassette. The viral ITRs are the only elements required in cis for replication and packaging of the rAAV genome. Rep and Cap genes are provided on a plasmid in trans along with adenovirus helper proteins. 32 Figure 9. Lysosomal enzyme trafficking and cross-correction. In the Golgi, lysosomal enzymes bind the mannose 6-phosphate receptor (M6PR). The majority (heavy arrows) of the enzymes are trafficked to the mature lysosome (Lys), but a minority (fine arrows) are secreted from the cell. Extracellular phosphorylated or non-phosphorylated enzyme can bind the plasma membrane- localized M6PR or the mannose receptor (ManR), respectively, for lysosomal targeting. Expression of ManR is limited to fixed tissue macrophages. Figure taken from: Gene Therapy for Lysosomal Storage Diseases. Molecular Therapy. (2006) 13: 841. 33 Figure 10151. Anterograde and retrograde transport after AAV injection. AAV vectors can be transported from the site of injection by anterograde and retrograde transport to connected regions. (a) Axonal anterograde transport requires the transport of viral particles via an axon projecting from the site of vector injection to a distal area with subsequent transduction of cells located within the brain region where the axon ends. (b) Retrograde transport of AAV vectors occurs when viral particles are taken up by axonal terminals in the injection site and are then transported back to the neuronal cell soma where they subsequently transduce the neuron. 34 Methods Surgeries (chapters 1-4 and appendices 1-3) General anesthesia was induced with ketamine (10 mg/kg) and dexmedetomidine (0.04 mg/kg) through an intravenous catheter and maintained using isoflurane (0.5 to 1.5%) in oxygen delivered through an endotracheal tube. For intracranial injections, cats were positioned sternally using a Horsley-Clark stereotaxic apparatus. Cats received injections of adeno-associated virus (AAV) vectors either 1) bilaterally in the thalamus and deep cerebellar nuclei (DCN), 2) bilaterally in the thalamus and the left lateral ventricle, or 3) intracisternally (IC). Craniotomy sites were made with a 20 gauge (G) needle at the following distances (in cm) from bony landmarks on the skull: thalamus (relative to bregma), anterior-posterior (AP) -0.7, mediolateral (ML) ?0.4, dorsoventral (DV) -1.6 (from meninges); DCN (relative to lambda), AP 0.0, ML ?0.4, DV -1.25 (from meninges); left lateral ventricle, caudo-thalamic notch located by magnetic resonance imaging (MRI) and ultrasound guidance through the bregma or median suture after a 1.5 cm incision of the scalp. Vector was delivered using a Hamilton syringe with a non-coring needle (22 to 25 G). A total of 70 ?l was injected into each thalamus in 10 to 20 ?l boluses, a total of 24 ?l was injected into each DCN in a 10 ?l and 14 ?l bolus, and a total of 200 ?l was injected into the left lateral ventricle at a rate of 15 ?l/min. Injection rate was 2 ?l/min and the needle was raised 0.15 cm between boluses. For IC injection, the atlanto-occipital joint was accessed with a 25G, 1.5 inch spinal needle then a 1 ml syringe was attached for injection of 200 ?l of vector over approximately 2 minutes. 35 Tissue Preparation (chapters 1-4 and appendices 1-3) For biochemical analysis, brains were divided into coronal blocks of 0.6 cm from the frontal pole through the cerebellum (chapter 1, Figure 1A). Coronal blocks from the right hemisphere were frozen in optimum cutting temperature medium and used for the following analyses: lysosomal enzyme distribution by histochemical staining, lysosomal enzyme specific activity by 4-methylumbelliferone (4MU) enzyme assays (performed in duplicate for each block), AAV vector distribution by SYBR Green quantitative PCR (performed in triplicate for each block), and sialic acid storage by periodic acid Schiff staining. Coronal blocks from the left hemisphere were halved to 0.3 cm and fixed in 10% formalin for routine hematoxylin and eosin (H&E) staining, or stored at -80 ?C for analysis of lipid storage by high performance thin layer chromatography (HPTLC) or resorcinol based assay (performed in duplicate). AAV vector design and preparation for GM1 cats (chapter 1 and appendix 1) The feline ?-galactosidase (?gal) cDNA sequence has the GenBank accession number AF006749. AAV vectors were produced as previously described by triple transfection of 293T cells with vector plasmid (derived from the plasmid pAAV-CBA-EGFP-W by replacing enhanced green fluorescent protein (EGFP) with the cDNA for feline ?gal)73, a mini adenovirus helper plasmid pF?6, and AAVrh8 helper plasmid pAR8, or AAV1 helper plasmid pXR1149. The inverted terminal repeats (ITRs) in the vectors are derived from AAV2. Transgene expression is controlled by a hybrid cytomegalovirus enhancer/chicken ?-actin promoter (CBA). All vectors carry the Woodchuck hepatitis virus posttranslational regulatory element (WPRE). 36 AAV vector design and preparation for GM2 cats (chapters 2-4 and appendices 2-3) The feline hexosaminidase (Hex) ?- and ?- subunit cDNA sequences have GenBank accession numbers JF899596 and JF899597, respectively. AAV vectors were produced as previously described by triple transfection of 293T cells with vector plasmid (derived from the plasmid pAAV-CBA-MGB-W by replacing MGB with cDNA for feline HEXA or feline HEXB157), a mini adenovirus helper plasmid pF?6, and AAVrh8 helper plasmid pAR8149. The ITRs in the vectors are derived from AAV2. Transgene expression is controlled by CBA, and vectors carry the WPRE element. Determination of lysosomal enzyme activity (chapters 1-4 and appendices 1-3) For brain and spinal cord tissue extraction, several frozen sections (50 ?m) were cut from each coronal block (Chapter 1, Figure 1A). For peripheral nervous system (PNS) and visceral tissue extraction, a sample was taken from tissue stored at -80 ?C. Tissue was homogenized manually in 50 mM citrate phosphate buffer, pH 4.4 (50 mM citric acid, 50 mM Na2HPO4, 10 mM NaCl) containing 0.1% TritonX 100 and 0.05% bovine serum albumin (BSA), followed by two freeze-thaw cycles and centrifugation at 15,700 g for 5 minutes at 4 ?C. Cerebrospinal fluid (CSF) samples were analyzed directly from -80 ?C. The activity of ?gal, Hex A, total Hex and ?- mannosidase activity were measured as previously described using synthetic 4MU fluorogenic substrates157, 172, but for ?gal we measured 10 ?l of sample for brain and 30 ?l for spinal cord. ?gal, 4-MU-?-D-galactopyranoside, pH 3.8; HexA, 1mM 4-MU-6-sulfa-2-Acetoamido-2- Deoxy-?-D-Glucopyranoside (MUGS), pH 4.2; total Hex, 1mM 4-MU-N-acetyl-?-D- glucosaminide (MUG); ?-mannosidase, 2 mM 4-MU-?-D-manopyranoside, pH 4.2. Total Hex activity in 8 mm punch biopsy samples was measured as previously described using 40 ?l of 37 sample supernatant and 20 ?l of MUG substrate108. Specific activity was expressed as nmol 4MU/mg/hr after normalization to protein concentration by the Lowry method. Histochemical staining for Hex activity (chapters 2-4 and appendices 2-3) Frozen sections at 40 ?M were thawed and fixed for 15 minutes in the presence of 0.1% glutaraldehyde in 4% paraformaldehyde in 0.1M phosphate buffer, pH 7.25. Sections were washed in phosphate buffered saline and then in citrate phosphate buffer (50 mM Na2HPO4.7H2O, 50 mM citric acid monohydrate), pH 4.5. Tissue sections were then incubated for 3 hours at 4 ?C in citrate phosphate buffer, pH 4.5 containing 0.20 mM naphthol AS-BI-N- acetyl-B-D-glucosaminide (naphthol) followed by a 2 hour incubation at room temperature in citrate phosphate buffer, pH 5.2, containing 0.25mM naphthol in the presence of 4% hexazotized pararosalinine (containing 4% pararosaniline hydrochloride and 4% NaNO2) as previously described173. Sections were washed in ddH2O, dehydrated and mounted. Histochemical staining for ?gal activity (chapter 1 and appendix 1) Frozen sections at 40 ?M were thawed and fixed in 0.5% glutaraldehyde in citrate phosphate buffer (50 mM Na2HPO4.7H2O, 50 mM citric acid monohydrate, 10 mM NaCl), pH 4.2 (brain) or pH 5.2 (spinal cord) for 10 minutes followed by washes in citrate phosphate buffer. Tissue sections were then incubated at 37 ?C overnight in citrate phosphate buffer, pH 4.2 (brain) or pH 5.2 (spinal cord) containing 20 mM K4Fe(CN)6, 20 mM K3Fe(CN)6, 2 mM MgCl2 ,0.02% IGEPAL, 0.01% deoxycholic acid and 2 mg/ml 5-bromo-4-chloro-3-indolyl-?-D- galactopyranoside (Xgal). The next day, sections were washed, dehydrated and mounted. 38 Periodic Acid Schiff (PAS) staining for ganglioside storage material (chapters 1, 3, and 4) Storage material was assessed qualitatively with PAS staining, which detects the oligosaccharide side chain of ganglioside. Frozen sections (20 ?M) were washed in phosphate buffered saline (PBS), fixed for 7 minutes with 3.7% paraformaldehyde in 95% ethanol/5% ddH2O (pH 7.4), washed again with PBS, incubated for 3 minutes in 0.5% periodic acid, washed in double distilled H2O and incubated in Schiff reagent for 45 seconds (brain) or 60 seconds (spinal cord). Lipid extraction and sialic acid quantification (Chapters 1 and 4) Punch biopsies of 8 mm diameter were taken from representative areas of the brain and spinal cord and lyophilized overnight. The next day, 10-25 mg of dry weight sample was rehydrated in 0.5 ml of water and total lipids extracted in 5 ml of chloroform: methanol (1:1 by volume). Samples were stirred at room temperature for at least 4.5 hours and centrifuged at 850 g for 20 minutes. Supernatant was collected and the pellet extracted again in 2 ml of chloroform: methanol for 15 minutes before another centrifugation. Supernatants for each sample were combined to measure sialic acid by the method of Svennerholm174 with modification from Miettinen175. A 0.25 ml aliquot was evaporated, re-dissolved in 0.5 ml of H2O, dissolved in 0.5 ml of resorcinol reagent (5 ml of distilled H2O, 40 ml of HCl, 0.125 ml of 0.1M copper sulfate, 5 ml of 2% resorcinol stock in distilled H2O), and boiled for 15 minutes. After cooling for 10 minutes, 1 ml of butyl acetate:butanol solution (85:15 v/v) was added and each sample was mixed vigorously. The upper phase was removed and read at 580 nm using a Shimadzu UV 160U spectrophotometer. Sialic acid concentration was expressed as nmol/mg. 39 Lipid extraction and sialic acid quantification (chapter 2 and appendix 2) Samples were homogenized in water and lyophilized overnight. Lipid isolation and purification has been detailed previously102. Briefly, dried tissue was rehydrated in 0.5 ml of H2O and lipids were extracted in chloroform (CHCl3):methanol (MeOH) (1:1 v/v). Neutral and acidic lipids/gangliosides were separated using ion exchange chromatography on a DEAE-Sephadex column as previously described. Neutral lipids and cholesterol were eluted with (CHCl3:CH3OH:dH2O, 30:60:8 v/v/v) and acidic lipids and gangliosides were eluted with CHCl3:CH3OH:0.8 M CH3COONa (30:60:8 v/v/v). Acidic lipids and gangliosides were dried by rotary evaporation, separated by Folch partitioning, base treated and desalted. Total ganglioside content was quantified before and after desalting using a resorcinol assay. Neutral lipids were dried by rotary evaporation and re-suspended in CHCl3:CH3OH (2:1 v/v). To further purify GA2, an aliquot was evaporated under a stream of nitrogen, treated with 1 N of NaOH and Folch partitioned. High performance thin layer chromatography (HPTLC) (chapter 2 and appendix 2) All lipids were analyzed qualitatively by HPTLC as detailed previously102. To enhance precision, oleoyl alcohol was added as an internal standard to the neutral and acidic lipid standards and samples. Lipids were spotted on 10 x 20 cm Silica gel HPTLC plates using a Camag Linomat V semi-automatic TLC spotter. Purified lipid standards were purchased from Matreya or were a gift from Dr. Robert Yu (Medical College of Georgia, Augusta, GA). Each lane was spotted with: 1.5 ?g of sialic acid for gangliosides, 70 ?g of dry weight for neutral lipids and 200 ?g of dry weight for acidic lipids and GA2. Ganglioside and GA2 HPTLC plates were developed in a single ascending run for 90 minutes with CHCL3: CH3OH: dH2O (55:45:10 40 v/v/v for gangliosides; CHCL3:CH3OH:dH2O 65:35:8 v/v/v for GA2) containing 0.02% CaCl2 (aq). Plates were sprayed with resorcinol-HCl reagent and heated at 95 ?C face down for 10 minutes and face up for 1 minute to visualize gangliosides, or face up for 5 minutes to visualize GA2. Neutral and acidic lipid HPTLC plates were developed to a height of 4.5 or 6 cm, respectively, in CHCL3:CH3OH:CH3COOH:CHOOH:dH2O (35:15:6:2:1 v/v/v/v/v). Plates were dried and developed to the top in C6H14:C6H14O:CH3COOH (65:35:2 v/v/v). Bands were visualized by charring with 3% cupric acetate in 8% phosphoric acid solution for 7 minutes. Quantification of individual lipids (chapter 2 and appendix 2) The percent distribution and density of individual bands was determined as previously described108. Total brain ganglioside distribution was normalized to 100% and the percentage distribution values were used to calculate sialic acid concentration of GM2 (?g/100 mg dry weight). Density values of neutral lipids, acidic lipids and GA2 were fit to a standard curve and used to calculate individual concentrations. Serum antibody titers (chapter 2) 100 ng of AAV vector was coated onto enzyme linked immunosorbent assay (ELISA) plates and incubated overnight at 4 ?C. Plates were washed then blocked with 5% non-fat powdered milk in phosphate-buffered saline for 90 minutes at room temperature. Next, 100 ?l of two-fold serial dilutions of feline serum samples were added. Goat anti-feline IgG:HRP (1:20,000) was used for color development with tetramethylbenzidine. 41 DEAE cellulose anion exchange chromatography (chapter 2) Several frozen sections (50 ?m) were cut from the coronal block containing the thalamus. Frozen brain tissue was extracted as previously described23 in a 10 nmol/l sodium phosphate buffer (pH 6.0) containing 0.1% Triton-X 100. Supernatant was applied to a diethylaminoethyl (DEAE) cellulose column and 0.5 ml each of fractions 1 to 26 were collected as previously described23 by washing the column with buffer (above) containing sodium chloride at increasing concentrations ranging from 0 to 400 mmol/l. Quantitative PCR for vector genomes (chapters 1-4 and appendices 1-3) Vector was measured by quantitative PCR using SYBR?Green-based reactions with primers specific for WPRE in the vector (forward 5?-AGTTGTGGCCCGTTGTCA-3?; reverse 5?-GAGGGGGAAAGCGAAAGT-3?). Samples were incubated for 2 minutes at 50 ?C, 10 minutes at 95 ?C, and then amplified for 40 cycles of 95 ?C for 30 seconds, 62 ?C for 30 seconds, and 72 ?C for 45 seconds. Genomic DNA samples (20 ng for spinal cord, 50 ng for brain and 25- 250 ng for peripheral tissue) were measured in triplicate on a BioRad CFX96 Real-Time System and compared to a standard curve generated from a plasmid containing WPRE (1 ? 108 to 1 ? 100 copies). The assay limit of detection (LOD) above background was 20 AAV copies/reaction. Background amplification was determined on DNA from untreated normal and untreated GM1 gangliosidosis or SD cats. Magnetic resonance imaging (MRI) (chapters 1, 3 and 4) Data were acquired on a 3T MAGNETOM Verio scanner with an eight channel phased array wrist coil. Whole-brain anatomical images were acquired with 3D MP RAGE (three- 42 dimensional magnetization-prepared rapid gradient echo) with 0.4 mm isotropic resolution and repetition time (TR)/ echo time (TE) of 1900/3.3 ms, followed by whole-brain multi slice 2D axial T2 TSE (turbo-spin echo) images with TR/TE of 4630/107 ms, turbo factor of 9 and a resolution of 0.3?0.3?1 mm3. MRI data were analyzed with eFilm 3.2 software. Clinical rating scores (chapters 1-4) Cats were assigned a clinical rating score (CRS) using two separate, independent readouts: (1) neurological exams performed by a veterinarian at two to four week intervals and (2) video footage at approximately 1 month intervals, which was retrospectively analyzed by the author. Statistical Analysis Data are expressed as means ? standard deviation throughout the text and figures. Statistics were performed with SAS 9.2 software, and P ? 0.05 was considered significant for all tests. The Wilcoxon signed rank test was used for pairwise comparisons of lysosomal enzyme activity, AAV vector distribution, and sialic acid levels. One-sided testing is reported throughout the text and figures to determine directional significance (that is, significantly higher or lower values), which would often not be realized using two-sided testing because of low animal numbers when using a feline model. The logrank test was used for survival comparison between groups. No data were excluded from analysis and, because outliers are hard to determine with small sample sizes, no data points in the study were defined as outliers. 43 Chapter 1: Sustained normalization of neurological disease after intracranial gene therapy in a feline model of GM1 gangliosidosis Published as: Sustained Normalization of Neurological Disease After Intracranial Gene Therapy in a Feline Model. In: Science Translational Medicine. Volume 6, Issue 231 (2014). Abstract Progressive debilitating neurological defects characterize feline GM1 gangliosidosis, a lysosomal storage disease caused by deficiency of lysosomal ?-galactosidase. No effective therapy exists for affected children, who often die before age 5 years. An adeno-associated viral vector carrying the therapeutic gene was injected bilaterally into two brain targets (thalamus and deep cerebellar nuclei) of a feline model of GM1 gangliosidosis. Gene therapy normalized ?- galactosidase activity and storage throughout the brain and spinal cord. The mean survival of 12 treated GM1 gangliosidosis animals was >38 months compared to 8 months for untreated animals. Seven of the eight treated animals remaining alive demonstrated normalization of disease, with abrogation of many symptoms including gait deficits and disequilibrium. Sustained correction of the GM1 gangliosidosis disease phenotype after limited intracranial targeting by gene therapy in a large animal model suggests that this approach may be useful for treating the human version of this lysosomal storage disorder. 44 Introduction GM1 gangliosidosis is an autosomal recessive lysosomal storage disease (LSD) caused by deficiency of ?-galactosidase (?gal, EC 3.2.1.23), the enzyme that hydrolyzes terminal galactose residues from numerous molecules. ?gal deficiency leads to neuronal storage of GM1 ganglioside and its asialo derivative (GA1), resulting in progressive neurodegeneration and death death15, 24. Three clinical forms exist and are thought to result from differing levels of residual enzyme activity37. Infantile- and juvenile-onset forms are fatal by ages 5 and 15 years, respectively, whereas the adult-onset phenotype varies considerably, with some patients living into the sixth decade24. No effective therapy exists for GM1 gangliosidosis. Because ?gal deficiency affects the central nervous system (CNS) globally, successful treatment strategies must target widespread areas of the brain and spinal cord. Whereas the circulatory system provides global access to the CNS for select molecules, lysosomal enzymes are excluded by the blood-brain barrier (BBB), making the task of enzyme replacement challenging. Therapeutic challenges posed by the BBB may be overcome by several key characteristics of lysosomal enzymes. First, even low concentrations of enzyme may be therapeutic, with activity in adult-onset LSD patients and clinically normal carriers reported at 2 to 4% and 11% of homozygous normal, respectively83. Also, diseased cells are able to endocytose and use normal lysosomal enzymes, so a locus of treated cells may ?cross-correct? a broad sphere of untreated neighboring tissue80, 81. Finally, lysosomal enzymes are distributed in the CNS by multiple mechanisms such as diffusion129, axonal transport130-132 and cerebrospinal fluid (CSF) flow132, 140, 141, 146, 160, which also apply to gene therapy vectors. Recently, adeno-associated viral (AAV) vectors have achieved widespread CNS distribution of lysosomal enzymes after injection into highly interconnected brain structures such as the ventral tegmental area, striatum, thalamus, and deep cerebellar nuclei (DCN)131, 135, 139, 140, 45 142. Most GM1 gangliosidosis mice treated by AAV injection of the thalamus and deep cerebellar nuclei survived for 52 weeks (the experimental endpoint) compared to a median survival of 38 weeks in untreated animals. ?gal activity was restored to a presumed therapeutic level (>10% wildtype mice) throughout the CNS, with normalization of substrate storage in the brain and a 50% reduction of substrate storage in the spinal cord135. Because the mouse brain is 1000 times smaller than that of a human infant, we tested AAV-based gene therapy in a feline model of GM1 gangliosidosis. First described in 1971, naturally occurring feline GM1 gangliosidosis is an accurate model of the human juvenile-onset disease in terms of enzymatic deficiency (<10% normal ?gal activity), storage levels, and CNS and peripheral organ pathology46, 70, 72, 176. Clinical disease progression is remarkably stereotypical, with affected cats reaching a humane endpoint at 8.0 (? 0.6) months. The naturally occurring feline mutation is analogous to a well-characterized human mutation that generates normal amounts of enzymatically defective ?gal protein (that is, positive for cross-reactive material, or CRM+)73, 177-179. The feline brain is ~20 times smaller than an infant?s brain and provides a good approximation of enzyme and vector distribution challenges that need to be overcome for human gene therapy. Here, GM1 gangliosidosis cats were treated by intracranial injection of AAV vectors encoding feline ?gal. Short- and long-term assessments of therapeutic effect included clinical and MRI-based analyses accompanied by postmortem assays to evaluate normalization of biochemical defects. 46 Results Treatment groups Twenty three GM1 gangliosidosis cats were treated before the average age of clinical disease onset by bilateral injection of the thalamus and deep cerebellar nuclei with AAV vectors expressing feline ?gal from a hybrid cytomegalovirus enhancer/chicken ?-actin (CBA) promoter [serotype AAV1 (n = 8) or AAVrh8 (n = 15)]. Cats were treated with either a traditional serotype already in human use (AAV1) or a new serotype that has shown promise in mouse experiments (AAVrh8). Outcomes were assessed at 16 weeks after injection (short term, n = 7) or at the humane endpoint (long term, n = 16), defined by the inability to stand on two consecutive days. Table 1 summarizes the study design. Widespread distribution of ?gal and AAV Sixteen weeks after a single surgery lasting ~90 minutes, ?gal activity exceeded normal (homozygous for the wildtype allele) concentrations throughout the brain and spinal cord of GM1 gangliosidosis cats (Figure 1A-C). Activity was highest at the injection sites (Figure 1B) and decreased with distance from the thalamus and deep cerebellar nuclei, yet ?gal staining was normal or above normal in most brain regions. Specific activity measured with a synthetic fluorogenic substrate ranged from 1.1- to 4.1-fold that of normal across all coronal brain blocks for cats treated with AAVrh8 (Table 2). In AAV1-treated cats, ?gal brain activity ranged from 1.4- to 4.2-fold that of normal, and there was no statistical difference in ?gal concentrations when compared to the AAVrh8 cohort (P ? 0.22 for each block; Table 2 and Figure S1). ?gal activity in the spinal cord at 16 weeks after injection also was near- or above- normal, with no statistical difference between the two AAV vectors (P ? 0.22 for each block; 47 Table 2). When measured in seven coronal blocks from the cervical, thoracic and lumbar spinal cord, ?gal activity ranged from 1.6- to 4.8-fold that of normal in the AAVrh8 cohort and from 0.9- to 4.2-fold that of normal in the AAV1 cohort (Figure 1C, Figure S1, and Table 2). Residual ?gal activity in untreated GM1 gangliosidosis cats ranged from 0.0- to 0.1-fold and 0.0- to 0.04-fold that of normal in the brain and spinal cord, respectively. Both vector cohorts demonstrated statistically significant increases in ?gal activity versus untreated animals (P ? 0.026 for each block). Vector genomes (vg) were detected in all brain and spinal cord blocks, demonstrating widespread dissemination from the injection sites, although vector levels did not always correlate with ?gal activity (Table 3). CSF ?gal activity was 28-fold that of normal in the AAVrh8 cohort and 42-fold that of normal in the AAV1 cohort, which was significantly higher than in untreated GM1 gangliosidosis cats (P = 0.026 for each cohort; Table 2). Additionally, ?gal activity and vector genomes were present in the liver of both AAV-treated cohorts demonstrating dissemination of enzyme and vector to peripheral tissues after intraparenchymal brain injection (Tables 2 and 3). Liver ?gal activity measured 0.38- and 0.24-fold that of normal in the AAVrh8 and AAV1 cohort, respectively, which was significantly higher than in untreated GM1 gangliosidosis cats (AAVrh8, P = 0.015; AAV1, P = 0.026). Clearance of storage material As glycosphingolipids with complex oligosaccharide side chains containing one or more sialic acid residues16, gangliosides can be detected colorimetrically by a variety of methods. To evaluate the clearance of stored substrate by vector-generated ?gal, we performed periodic acid Schiff (PAS) staining in the CNS in GM1 gangliosidosis cats 16 weeks after treatment. Storage 48 levels were substantially normalized throughout most of the brain and spinal cord, but clearance was incomplete in focal areas of the temporal lobe and cervical spinal cord (Figure 2A) that corresponded with sites of minimal ?gal activity in Figure 1. When quantitative sialic acid assays were performed on 15 samples throughout the CNS (Figure 2B, C), concentrations in untreated GM1 gangliosidosis cats were significantly higher than that of normal in the cerebrum (2.5- to 3.9-fold normal; P ? 0.015), cerebellum (1.8- to 1.9- fold normal; P ? 0.015) and spinal cord (1.9- to 2.5-fold normal; P ? 0.015). Treatment with AAV1 or AAVrh8 significantly reduced storage in all CNS samples (P ? 0.026), with the exception of the caudal cerebellum from AAV1-treated cats, for which samples from only two animals were available (Figure 2B, C, block h, P = 0.053). Although reduced by gene therapy, storage remained above normal in 4 of 15 samples including the temporal lobe (sample d3 in Figure 2B), rostral and middle cerebellum (samples f and g1 in Figure 2B), and cervical spinal cord (sample K in Figure 2B). In one sample from the frontal pole of AAVrh8-treated cats, sialic acid was reduced to 0.83-fold of normal (sample a in Figure 2B; P = 0.033). Normalized activity of other lysosomal enzymes In ?gal deficiency, the activity of other lysosomal enzymes is above normal70, 180. For example, lysosomal ?-N-acetylhexosaminidase (EC 3.2.1.52) activity was elevated up to 3.7-fold that of normal in the CNS and 2.0-fold that of normal in the liver of untreated GM1 gangliosidosis cats (P ? 0.015 for each CNS block and for liver; Figure 3). Treatment for 16 weeks with either AAV serotype reduced hexosaminidase activity significantly in the brain (P ? 0.026 for each block), spinal cord (P ? 0.026 for each block) and liver (P ? 0.026 ), demonstrating normalization of a secondary lysosomal biomarker after gene therapy (Figure 3). 49 Long-term clinical benefit of AAV treatment Long term clinical benefit in GM1 gangliosidosis cats was assessed after treatment with AAV1 (n = 5) or AAVrh8 (n = 7) serotypes. As shown in Table 1, animals were treated with 3 ? 1012 to 4 ? 1012 vector genomes (vg) (?full dose,? n = 9) or 1.2 ? 1013 vg when AAV1 titers made this ?high dose? possible (n = 3). Statistically significant increases in survival have been achieved for both the AAV1 (P = 0.0004) and AAVrh8 (P < 0.0001) cohorts (Figure 4A). Currently, the mean survival of treated GM1 gangliosidosis cats is >4.7 times that of untreated cats: 39.1 (? 10.1) months for the AAV1 cohort and 37.7 (? 12.4) months for the AAVrh8 cohort. Eight of 12 treated cats remain alive, most with subtle or no disease signs. The oldest cats from each vector cohort remain alive and well at 52.8 (AAV1) and 50.5 (AAVrh8) months, or >6.3 times the life span of untreated cats. Of the four cats no longer living, two had normal gait at 29.0 months (9-1515) and 47.9 months (8-1397) but were euthanized because of abnormal recovery from anesthesia. Disease onset in treated cats was delayed versus untreated GM1 gangliosidosis controls, whose stereotypical disease progression began at 4.1 (? 0.6) months of age with fine tremors of the head and tail. Clinical symptoms in untreated animals progressed to generalized muscle weakness, wide-based stance, ataxia, carpal hyperextension (in 80% of cats), instability with occasional falling, loss of ambulation, and finally the inability to stand, which defined the humane endpoint at 8.0 (? 0.6) months. Of 12 treated GM1 gangliosidosis cats, disease onset is yet to be determined in four animals ranging from 34.6 to 44.9 months of age. Also, three cats whose disease onset occurred at 10.2, 36.6 and 37.0 months currently have only mild disease at 44.8, 50.5 and 52.8 months, respectively (8-1435, 8-1364 and 9-1356). 50 Quality of life for treated GM1 gangliosidosis cats has improved markedly, as measured by a clinical rating scale that reflects an animal?s departure from normal function (Figure 4B, C). Clinical rating scores decrease from 10 (normal function) to 1 (lateral recumbency) as animals become more debilitated by gait defects and balance disturbances/instability. Currently, minimal disease progression has been documented for five of seven animals in the AAVrh8 cohort (Figure 4B, C) and four of five animals in the AAV1 cohort (Figure S2), whose composite clinical rating scores at 40 months of age were 9.5 (? 0.6) and 9.7 (? 0.6), respectively. Whereas untreated GM1 gangliosidosis cats have severe dysequilibrium and gait defects by 7 months of age, most treated GM1 gangliosidosis cats remain overtly normal or have only subtle gait abnormalities. No difference in therapeutic benefit is discernible between the AAV1 and AAVrh8 cohorts. In 3 of 12 AAV-treated cats, therapeutic response was clearly positive but less marked. One cat that was symptomatic at the time of surgery (8-1626) has moderately progressive disease at 29.4 months of age, although current symptom acquisition has been delayed about threefold. Two animals had disease onset at 6.1 (8-1378) and 4.6 (9-1545) months of age, ultimately reaching humane endpoint at 17.5 and 25.3 months, respectively. Conclusions regarding why 3 of 12 treated animals responded less robustly than others await tissue analysis from more cats in the long term cohort. However, diminished enzyme activity is a reasonable hypothesis because both cats that reached the humane endpoint (8-1378 and 9-1545) had low ?gal concentrations in the spinal cord (?0.4-fold normal and ?0.14-fold that of samples from the 16 week time point). In contrast, two cats that had normal gait at 29.0 and 47.9 months of age (9-1515 and 8-1397, respectively) had above-normal ?gal activity in the spinal cord that was comparable to the 16 week cohort (Figure 5). ?gal activity in CSF was above normal (2.5 to 63.1 fold) in all cats from 51 the long term cohort, perhaps explaining why there was no correlation between CSF activity and clinical outcome/life span (Figure 5). Half of all treated cats in the long term cohort (6 of 12) experienced seizure activity, well-controlled by medication, with a mean onset of 20.1 (? 7.4) months (Table 1). Common symptoms included blank stares, salivation, and facial twitches, but progression to undirected running and tonic-clonic seizures did occur in some animals. In at least three of six animals, seizures were demonstrably inducible by sounds such as running water. No seizures occurred in GM1 gangliosidosis cats treated for only 16 weeks (n = 7). Seizures are a known feature of late- stage feline GM1 gangliosidosis46, 72, although no untreated GM1 gangliosidosis cats in the current study had seizures because they occur well beyond the humane endpoint used. Normalization of MRI brain architecture On T2-weighted MRI, cortical white matter is hypointense to (that is, darker than) gray matter in normal cats but is hyperintense to (that is, lighter than) gray matter in untreated GM1 gangliosidosis cats with moderate to severe disease. Also, the deep cerebellar nuclei area is hypointense to cerebellar gray matter in normal cats but becomes isointense with disease progression in untreated GM1 gangliosidosis cats. Therefore, MRI provides an easily discernible, non-invasive biomarker of disease progression (Figure 6). In AAV-treated GM1 gangliosidosis cats, hypointensity of white matter to gray matter is substantially normalized in the cortex to at least 32 months of age (cat 8-1364), or more than five times longer than in untreated cats. Additionally, hypointensity of the deep cerebellar nuclei area to surrounding gray matter is largely preserved after AAV treatment. Overall brain architecture of treated GM1 gangliosidosis cats is normal, although slight widening of sulci is appreciated in some animals. 52 To date, the only abnormality associated with AAV treatment is an irregularly-shaped locus of T2 hyperintensity in the thalamus of 5 of 11 cats evaluated (Figure 6). In previous studies of gene therapy in feline GM2 gangliosidosis (Sandhoff disease), similar hyperintense loci were thought to be associated with eosinophilic, botryoid neurons in hematoxylin and eosin sections from the dorsal thalamic nuclei157. In treated GM1 gangliosidosis cats in the current study, fine eosinophilic granules were found in scattered cortical and hippocampal neurons dorsal to the injection site (Figure S3), but not in thalamic regions corresponding to T2 hyperintensities. The underlying cause of hyperintense thalamic loci remains to be defined. Detected in AAV-treated cats by 6 months of age, T2 hyperintensities were not present in untreated cats at any age, including the humane endpoint of ~8 months. No histological or MRI anomalies were found in three normal cats injected with saline and followed for >18 months. Dose response When GM1 gangliosidosis cats were treated with one-tenth of the full AAVrh8 dose, mean survival was 17.1 (? 3.6) months, significantly lower than the full dose cohort (P = 0.0046, Figure 4A). Of four cats treated with the one-tenth dose, two reached the humane endpoint at 14.2 and 19.1 months of age, one was euthanized at 13.9 months due to chronic weight loss and lethargy, and one was euthanized at 21.2 months because of dysphagia, also reported in juvenile- onset GM1 gangliosidosis patients181. Nevertheless, the mean survival of the one-tenth dose cohort was 2.1 times greater than that of the untreated GM1 gangliosidosis cats (P = 0.0012). ?gal activity and vector copies for the one-tenth dose cohort are detailed in Tables 2 and 3. Liver ?gal activity of only 0.17-fold that of normal was sufficient to fully normalize hexosaminidase activity (P = 0.015 versus untreated GM1 gangliosidosis cats; Figure S4). 53 Restoration of breeding function In >40 years of study, no female GM1 gangliosidosis cat has become pregnant and no GM1 gangliosidosis male has been fertile. After AAV injection, two treated males bred two treated females and four untreated carrier females, producing a total of nine litters (27 kittens). Treated GM1 gangliosidosis cats bred successfully from 9.8 to at least 28.6 months of age. Fertility onset for carrier or normal colony cats is ~7 months. GM1 ? GM1 matings produced litters composed entirely of GM1 gangliosidosis kittens, which had stereotypical disease progression and life span. In agreement with previous reports182, 183, vector was detected only transiently in gonads and no evidence of germ line gene transmission was found in treated cats or their offspring (Table S1). Discussion As a group, LSD prevalence is ~1 in 7,700 live births1, similar to that of cystic fibrosis or hemophilia. Like most lysosomal diseases, GM1 gangliosidosis is neurodegenerative and fatal, with only palliative measures currently available to patients. Promising results have been achieved with intracranial AAV injections in rodent models of gangliosidosis. Treated GM2 gangliosidosis mice demonstrate near normalization of life span, representing a more than fivefold survival increase versus untreated GM2 gangliosidosis mice140. Also encouraging are data from large animal models of neurodegenerative lysosomal diseases. In dogs with mucopolysaccharidosis I or IIIb, intracranial gene therapy with AAV vectors improved storage and histological lesions throughout most of the brain, although no clinical or survival benefit was demonstrated158, 165. Cats with ?-mannosidosis treated by intracranial injections of an AAV1 vector through 14 burr holes had only mild neurological disease at the untreated humane 54 endpoint of 18 weeks, with restoration of ?-mannosidase activity to ~4% normal in the brain. Because of the study design, improved survival benefit was not achieved, although one of two treated cats followed long term had only mild disease when euthanized at 56 weeks of age164. The current report documents above-normal ?gal activity throughout the GM1 gangliosidosis cat brain and spinal cord after intracranial injections of AAV vectors carrying the therapeutic gene through only four burr holes in a surgery lasting ~90 minutes. The successful translation of a treatment approach from GM1 gangliosidosis mice to GM1 gangliosidosis cats, with an approximate 70-fold scale-up of brain size suggests that this approach may be beneficial in human patients. Although the human brain is ~20 times larger than the cat brain, the thalamus and deep cerebellar nuclei also are proportionally larger, allowing an equivalent AAV dose per kilogram of brain weight. In 12 treated animals followed long term, mean life span was extended greater than 4.7-fold and continues to increase because most of the treated cats are disease-free or have only subtle symptoms at present. Quality of life improved profoundly, with only 2 of 12 treated animals having progressed to the most debilitating symptoms such as dysequilibrium and severe ataxia. Clinical outcomes were similar for the two vector serotypes tested: AAV1, currently in human clinical trials184-187 or approved for human use188, 189 and AAVrh8, a new serotype190 that showed promise in a related feline model157. Little toxicity was apparent from the vector. The eosinophilic granules in brain neurons of treated GM1 gangliosidosis animals may represent a less extreme version of the large, botryoid hypereosinophilic inclusions previously reported in Sandhoff and normal cats treated with AAVrh8 vectors that overexpressed feline hexosaminidase, often at ?50 times normal157. In our treated GM1 gangliosidosis cats, typical ?gal expression is less than or equal to fourfold normal, so it is possible that fine eosinophilic granules represent inclusions with mildly over- 55 expressed ?gal. Additionally, there was limited toxicity from the injection procedure. A thalamic hemorrhage documented by MRI in one animal (8-1435) caused ataxia, head tilt and cortical blindness that resolved after 1 week of treatment with mannitol and steroids. Now ?44 months of age, 8-1435 has only slight hind limb gait deficits and is the first fertile male in the 40-year history of the research colony. Although about half of AAV-treated GM1 gangliosidosis cats had hyperintense thalamic loci on T2-weighted images, no histopathological or clinical correlates were identified. In related ongoing studies, similar loci were found in the thalamus of four of six normal cats treated with AAVrh8 vectors expressing feline hexosaminidase, none of which had seizures in ?18 months of follow-up. Because no seizure activity was observed in three of five treated GM1 cats with T2 anomalies, a causal relationship between hyperintense thalamic loci and seizures is unlikely. Known to occur in late-stage GM1 gangliosidosis disease in both humans181, 191, 192 and cats46, 72, seizures in treated animals may result from progressive disease in the temporal lobe, one of the few brain regions with incomplete restoration of ?gal activity (Figure 1B) and only partial normalization of storage (Figure 2A, C). Seizures in treated GM1 gangliosidosis cats often resemble temporal lobe epilepsy, with short duration, initial motionless stare, impaired awareness, involuntary motor behaviors, and salivation193-195. Improved treatment and abrogation of seizure activity may be achieved by adding injection sites that target areas of suboptimal therapeutic response. Above-normal ?gal activity in the spinal cord after injection of the brain parenchyma was among the surprising findings of this study, even with injection targets (thalamus and deep cerebellar nuclei)135, 139 chosen for a high degree of interconnectivity with other CNS structures. ?gal activity and vector genomes were detected throughout the spinal cord, up to 14 cm from the 56 deep cerebellar nuclei injection site at the time of surgery. We hypothesize that a portion of vector transport to the spinal cord occurred through CSF, known to be an effective medium for AAV transfection of the CNS140, 141. For our thalamic injections, the needle transects the lateral ventricle, which may provide a path of least resistance for vector backflow through the needle tract. Similarly for cerebellar injections, vector may travel into the subarachnoid space dorsal to the cerebellum. Varying degrees of vector leakage into the CSF because of slight differences in needle placement or anatomy may explain the reduced ?gal activity in the spinal cord of cats that responded less robustly to treatment. Vector likely gained access to peripheral organs via blood vessels disrupted during the injection procedure and via CSF, which ultimately is reabsorbed into the circulatory system196. To date, there have been no apparent peripheral disease symptoms in AAV-treated GM1 gangliosidosis animals. It is likely that residual ?gal activity (~5% normal) in GM1 gangliosidosis cats delays development of clinical peripheral disease, but the increased ?gal activity in peripheral tissues of AAV-treated GM1 gangliosidosis cats also may contribute to therapeutic success. Normalized liver activity of lysosomal hexosaminidase in all AAV-treated groups indicates that even partial restoration of liver ?gal activity is beneficial (Figure S4). Although it was not possible to blind investigators to the treatment status of cats in the present study, clinical rating scores were assigned independently by two investigators to attempt to limit bias. Survival and MRI data provided objective clinical measures, and biochemical analyses from the deceased long term cats supported their clinical scores. The promising therapeutic results in this study after intraparenchymal injections through only four needle tracts, with little evidence of toxicity, support the initiation of AAV-based clinical trials for GM1 gangliosidosis in human patients. Although deep cerebellar nuclei injection was effective in 57 restoring enzyme activity to the cerebellum, a similar strategy in human patients must weigh the benefit of direct cerebellar treatment versus the risk of hemorrhage in the posterior fossa. Safety studies in non-human primates will assess potential vector and procedural toxicity in human patients. Should the risk of deep cerebellar nuclei injection be considered too great, CSF- mediated approaches to treating the cerebellum may be incorporated, such as those deemed successful in mice and dogs141, 160. When predicting outcomes in human patients it must be considered that GM1 gangliosidosis cats in this study were treated pre-symptomatically, whereas most human patients are diagnosed after disease onset, so gene therapy may be less efficacious when administered later in the disease course. Nevertheless, as evidence for the benefits of AAV gene therapy continues to build in animal models, addition of future clinical trials to those already under way for human LSDs170, 171 will determine whether this therapeutic potential is realized for GM1 gangliosidosis and for other LSDs. 58 Table 1. Treatment groups and current clinical status of AAV-treated GM1 gangliosidosis cats DurationA Serotype DoseB Cat Gender Tx age (mos.) Endpoint or current age (mos.) Clinical description ? Seizures* Long term AAV1 High 9-1356 F 1.9 52.8 Mild hind limb weakness, carpal hyperextension + 8-1483 F 1.9 41.4 Normal + 8-1485 F 2.0 41.2 Normal - Full 9-1545 F 2.3 25.3 Deceased - 8-1551 F 2.1 34.6 Normal + AAVrh8 Full 8-1364 M 1.6 50.5 Mild hind limb weakness - 8-1378 M 1.3 17.5 Deceased - 8-1397 F 1.7 47.9 Deceased# + 9-1424 F 2.9 44.9 Normal - 8-1435 M 2.8 44.8 Mild hind limb weakness + 9-1515 M 1.9 29.0 Deceased# - 8-1626 M 3.0 29.4 Weakness of carpi and limbs (hind/fore), pelvic ataxia, tremors + Tenth 8-1574 M 1.8 21.2 Deceased + 8-1578 F 1.8 13.9 Deceased - 8-1576 F 1.8 14.2 Deceased - 9-1620 F 2.3 19.1 Deceased - Short term (16 weeks) AAV1 Full 9-1553 F 2.3 N/A Normal - 9-1555 F 2.3 N/A Normal - 8-1617 M 2.5 N/A Normal - AAVrh8 Full 9-1494 M 1.9 N/A Normal - 9-1502 M 1.8 N/A Normal - 8-1525 F 1.9 N/A Normal - 8-1526 F 1.9 N/A Normal - 59 A Cats in the long term cohort were euthanized at clinical humane endpoint. Short term cats were euthanized 16 weeks after injection. Cats received bilateral injections of the thalamus and deep cerebellar nuclei. B Doses: High, 1.2 ? 1013 vg; Full, 3 ? 1012 to 4 ? 1012 vg; one-tenth dose, 3 ? 1011 vg. *Animal developed seizures that were well controlled by medication. The mean age of seizure onset was 20.1 ? 7.4 months. #Cat 8-1397 had normal gait at 47.9 months of age but did not recover from an anesthetic procedure and was euthanized. Cat 9-1515 was clinically normal at 29.0 months of age but did not recover properly from anesthesia for MRI and was euthanized. Abbreviations: Tx = treatment; mos. = months; N/A = not applicable. 60 Figure 1. Therapeutic enzyme distribution in the CNS of GM1 gangliosidosis cats after AAVrh8 treatment. GM1 gangliosidosis cats were injected bilaterally in the thalamus and deep cerebellar nuclei with AAVrh8-CBA-?gal-WPRE (3 ? 1012 to 3 ? 1012 vector genomes total). (A) Injection sites (white circles) and 0.6 cm coronal blocks of the brain (a-h) and spinal cord (i-o) collected at necropsy. Blocks were halved and analyzed for sialic acid concentration (left) or for enzyme activity (right). Lysosomal ?gal activity (blue) detected with Xgal at acidic pH was visualized throughout the brain (B) and spinal cord (C) of a representative, treated short term cat (GM1 + AAV; 8-1526). Lysosomal ?gal activity was visualized throughout the brain (D) and spinal cord (E) of a treated long term cat (GM1 + AAV; 9-1515) at 29 months old. Corresponding ?gal activity is shown below each block as fold increase over concentrations in untreated normal healthy cats (fold N). Similar distribution occurred with AAV1 (Figure S1). Representative 61 control sections are shown from untreated normal healthy cats along with untreated GM1 gangliosidosis cats, which express ?0.10-fold normal ?gal activity in brain blocks a-h (B,d) and ?0.04-fold normal ?gal activity in spinal cord blocks i-o (C, o). The ranges of specific activities for normal control blocks were: brain, 12.5 (a) - 42.1 (e); spinal cord, 3.2 (l) - 8.8 (k) nmol 4MU/mg/hr. 62 * ?gal activity was not significantly different between AAVrh8 (n = 4) and AAV1 (n = 3) cohorts at the 16 week time point (a-h and i-o, P ? 0.22 for each block; CSF, P = 0.19; liver, P = 0.43). A n = 4; ?gal specific activity was significantly higher than that of untreated GM1 gangliosidosis cats (n = 4) in a-h and i-o (P ? 0.015 for each block), CSF (P = 0.026, n = 3, no sample available for 8-1526), and liver (P = 0.015). Table 2. ?gal activity in brain, spinal cord, CSF, and liver of AAV-treated and untreated GM1 gangliosidosis cats Fold-normal ?gal specific activity, mean (s.d) Region Block Short term, full doseA AAVrh8* Short term, full doseB AAV1* Long term, tenth doseC AAVrh8 Long term, full doseD AAVrh8 Long term, full doseE AAV1 GM1 no tx Cerebrum A 2.7 (0.85) 4.1 (4.4) 0.48 (0.24) 0.96 (0.37) 0.48 0.00 (0.00) B 1.8 (0.62) 2.3 (2.4) 0.42 (0.11) 0.71 (0.30) 0.42 0.00 (0.00) C 1.7 (0.56) 2.2 (2.0) 0.63 (0.57) 0.56 (0.17) 0.61 0.01 (0.01) D 1.7 (0.70) 2.6 (1.1) 0.62 (0.40) 0.78 (0.27) 1.2 0.02 (0.07) E 1.1 (0.56) 1.4 (1.1) 0.33 (0.16) 0.78 (0.35) 0.82 0.05 (0.02) Cerebellum F 1.5 (1.2) 3.0 (1.2) 0.55 (0.90) 2.9 (1.5) 0.40 0.10 (0.04) G 2.2 (1.8) 4.2 (2.1) 0.33 (0.23) 2.5 (0.96) 0.33 0.04 (0.02) H 4.1 (2.6) 1.7 (1.1) 0.27 (0.34) 1.4 (1.6) 0.30 0.04 (0.03) Spinal cord I 2.0 (1.5) 1.0 (0.38) 0.00 (0.00) 0.74 (0.65) 0.07 0.00 (0.00) J 1.6 (1.4) 0.85 (0.60) 0.03 (0.06) 1.2 (0.93) 0.04 0.00 (0.00) K 1.9 (1.0) 0.89 (0.28) 0.05 (0.08) 1.5 (1.1) 0.10 0.03 (0.05) L 3.2 (2.5) 1.0 (0.46) 0.07 (0.09) 0.93 (0.62) 0.00 0.03 (0.06) M 2.1 (1.4) 2.2 (1.9) 0.07 (0.13) 0.75 (0.68) 0.05 0.00 (0.00) N 2.3 (1.1) 1.8 (1.5) 0.09 (0.15) 1.3 (0.98) 0.05 0.04 (0.07) O 4.8 (3.1) 4.2 (3.2) 0.38 (0.61) 6.6 (8.5) 0.17 0.01 (0.02) CSF N/A 28 (20) 42 (28) 2.7 (1.3) 32 (42) 8.3 0.03 (0.08) Liver N/A 0.38 (0.25) 0.24 (0.16) 0.17 (0.11) 0.20 (0.19) 0.08 0.05 (0.01) 63 B n = 3; ?gal specific activity was significantly higher than that of untreated GM1 gangliosidosis cats (n = 4) in a-h and i-o (P ? 0.026 for each block), CSF (P = 0.026), and liver (P = 0.026). C n = 4; ?gal specific activity was significantly higher than that of untreated GM1 gangliosidosis cats (n = 4) in a-e, g, and h (P ? 0.021 for each block), CSF (P = 0.015), and liver (P = 0.015), but not in f (P = 0.15) or i-o (P ? 0.20 for each block). D n = 3; ?gal specific activity was significantly higher than that of untreated GM1 gangliosidosis cats (n = 4) in a-h, j-l, and n-o (P ? 0.025 for each block), as well in CSF (P = 0.025), but not in blocks i and m (P = 0.061), or liver (P = 0.054). E Not enough subjects currently available for statistical analysis. Abbreviations: no tx = no treatment; N/A = not applicable; CSF = cerebrospinal fluid 64 Table 3. Vector copy number in brain, spinal cord, and liver of AAV-treated GM1 gangliosidosis cats Vector copy number per ?g genomic DNA, mean (s.d) Region Block Short term full dose Short term full dose Long term tenth dose Long term full dose Long term full dose AAVrh8A, B AAV1A AAVrh8C AAVrh8B AAV1 Cerebrum A 14,000 (12,000) 28,000 (14,000) 630 (110) 6,900 (3,500) 6,000 B 8,700 (6,900) 15,000 (7,500) 630 (520) 4,800 (4,800) 1,600 C 11,000 (6,700) 14,000 (9,400) 380 (180) 5,500 (6,900) 1,700 D 60,000 (55,000) 100,000 (130,000) 6,500 (5,700) 7,400 (1,700) 1,900 E 40,000 (35,000) 15,000 (8,700) 6,400 (11,000) 27,000 (24,000) 13,000 Cerebellum F 4,400 (1,800) 16,000 (9,500) 6,700 (13,000) 59,000 (91,000) 4,600 G 89,000 (170,000) 75,000 (11,000) 570 (510) 140,000 (230,000) 550 H 170,000 (170,000) 26,000 (24,000) 20,000 (39,000) 9,000 (13,000) 950 Spinal cord I 4,500 (2,200) 76,000 (80,000) 170 (180) 3,700 (2,100) 15,000 J 4,800 (1,300) 57,000 (37,000) 540 (820) 2,300 (2,800) 10,000 K 4,600 (3,000) 65,000 (58,000) 140 (200) 3,600 (2,100) 3,100 L 1,600 (1,300) 54,000 (41,000) 69 (110) 2,300 (800) 13,000 M 2,700 (1,900) 60,000 (31,000) 210 (260) 3,000 (2,900) 8,700 N 4,200 (1,800) 68,000 (49,000) 130 (170) 3,100 (2,000) 8,800 O 3,500 (2,900) 60,000 (67,000) 290 (430) 1,900 (1,100) 4,200 Liver N/A 230,000 (230,000) 23,000 (14,000) 13,000 (16,000) 9,700 (8,200) 18,000 A Vector copy number was not significantly different between AAVrh8 (n = 4) and AAV1 (n = 3) cohorts at the 16 week time point, probably because of high standard deviations between cats (a-h and i-o, P ? 0.052 for each block; liver, P = 0.19). 65 B Vector copy number was not significantly different between the short term (n = 4) and long term (n = 3) full dose AAVrh8 cohorts in a-h and i-o (P ? 0.056 for all blocks), but was significantly lower in the liver of the long term cohort versus the short term cohort (P = 0.026). C n = 4; vector copy number was significantly lower than in the long term full dose AAVrh8 cohort (n = 3) in a-c, g, i, and l-o (P ? 0.026 for each block). Block lettering corresponds to Figure 1A. Abbreviations: N/A = not applicable. 66 Figure 2. Storage in the CNS of GM1 gangliosidosis cats 16 weeks after treatment. (A) Storage in untreated GM1 gangliosidosis cats was visualized by dark PAS staining in the gray matter and thalamus. In treated brains, residual ganglioside storage was present in focal areas (black arrows) of the temporal lobe [block d in (B)] and cervical spinal cord [block k in (B)]. (B) Sample sites for sialic acid quantitation (circles) in brain (a-h) and spinal cord (k and o correspond to Figure 1A; half of each block was used). (C) Sialic acid concentrations were measured in untreated GM1 gangliosidosis cats (n = 4) and after treatment with AAVrh8 (n = 4) or AAV1 (n = 3) for comparison to normal healthy cats (n = 4). *, Concentration in all samples from untreated GM1 gangliosidosis cats was significantly higher than in normal cats (P ? 0.015 for each block); **, concentration in all samples from treated GM1 gangliosidosis cats was significantly lower than in untreated GM1 gangliosidosis cats (P ? 0.026 for each block) except AAV1 block h, because only two samples were available (P = 0.053); ?, samples from treated GM1 gangliosidosis cats in which concentration was significantly higher than in normal cats (P ? 0.035); ?, a sample from treated GM1 gangliosidosis cats in which concentration was significantly lower than in normal cats (P = 0.033). The Wilcoxon signed rank test was used for statistical comparisons. 67 Figure 3. Normalization of lysosomal hexosaminidase activity in the CNS and liver of GM1 gangliosidosis cats 16 weeks after treatment. GM1 gangliosidosis cats were injected bilaterally in the thalamus and deep cerebellar nuclei with 3 ? 1012 to 4 ? 1012 vector genomes of AAVrh8 (n = 4, dark gray bars) or AAV1 (n = 3, black bars). Tissues were collected 16 weeks after treatment and hexosaminidase activity compared to untreated GM1 gangliosidosis cats (n = 4, light gray bars) and normal healthy cats (n = 4) in the brain (a-h), spinal cord (i-o), and liver. Lettering of brain and spinal cord blocks corresponds to Figure 1A. *, Activity in all samples from untreated GM1 gangliosidosis cats was significantly higher than in normal cats (P ? 0.015 for each sample). **, Activity in all samples from treated GM1 gangliosidosis cats was significantly lower than in untreated GM1 gangliosidosis cats (P ? 0.026 for each sample). ?, a sample from treated cats that had significantly higher activity than in normal cats (P = 0.026); ?, samples from treated cats that had significantly lower activity than in normal cats (P ? 0.030). The Wilcoxon signed rank test was used for statistical comparisons. 68 Figure 4. Survival and clinical progression of AAV-treated GM1 gangliosidosis cats. (A) Kaplan-Meier survival curves for untreated GM1 gangliosidosis cats [No treatment (No Tx), n = 12] and GM1 gangliosidosis cats treated long term with gene therapy, which have significantly increased survival compared to untreated animals using the logrank test: *P = 0.0004, AAV1 (n = 5); **P < 0.0001, AAVrh8 (n = 7); ***P = 0.0012, one-tenth dose AAVrh8 group (n = 4). Survival of AAVrh8 treated cats was significantly higher for full dose versus one-tenth dose (P = 0.0046). (B) Full dose AAVrh8-treated cats were scored on a scale based on disease progression in untreated GM1 gangliosidosis cats (No Tx). A composite of five cats that responded robustly 69 to treatment is shown (+AAV composite), whereas two cats that responded less favorably are depicted separately (8-1626 and 8-1378). Solid lines signify living cats; dashed lines represent deceased cats. Open symbols denote whole-body tremors, which occurred in all untreated cats but only one treated animal. (C) Age of symptom onset is shown for untreated GM1 gangliosidosis cats (mean ? SD, n = 12). The scale is based on gait defects, which ultimately defined the humane endpoint, with initial deficits appearing at 4.9 months. However, disease onset begins at 4.1 months with fine head and tail tremors that progress to whole-body tremors at 6.2 months. For clarity, only whole-body tremors are depicted in (B). 70 Figure 5. ?gal activity in the CNS and CSF of GM1 gangliosidosis cats treated long term. Cats were treated with the full dose of AAV1 (9-1545) or AAVrh8 (8-1378, 9-1515 and 8-1397). (A) Animals whose disease progressed to the humane endpoint (8-1378 and 9-1545) had substantially lower activity in the spinal cord than cats with normal gait that were euthanized due to abnormal recovery from anesthesia (9-1515 and 8-1397). Reduced ?gal activity cannot be attributed solely to a generalized decrease in expression over time, since the highest activities 71 were found in animals that lived the longest (9-1515, 29.0 months; 8-1397, 47.9 months). Lettering of brain and spinal cord blocks correspond to Figure 1A. 9-1545, black bars; 8-1378, medium gray bars; 9-1515, light gray bars; 8-1397, dark gray bars. (B) ?gal activities in CSF were variable and there was no correlation between activity level and clinical therapeutic effect. However, it may not be possible to assess the correlative value of ?gal activity in CSF since all cats had above-normal levels. 72 Figure 6. MRI evaluation of GM1 gangliosidosis cats. T2-weighted MRI images (3T) were taken at the level of the caudate nucleus (A-D), thalamus (E-H) and deep cerebellar nuclei (I-L). Cortical white matter is hypointense to (that is, darker than) gray matter in normal healthy cats, but hyperintense to (that is, lighter than) gray matter in untreated GM1 gangliosidosis cats [white arrow in (F)]. Also, the deep cerebellar nuclei area is hypointense to surrounding gray matter in normal healthy cats [outlined black arrowhead in (I)], but becomes isointense with disease progression in untreated GM1 gangliosidosis cats. In AAV-treated GM1 gangliosidosis cats (GM1+AAV), hypointensity of cortical white to gray matter and deep cerebellar nuclei area to cerebellar gray matter was largely preserved, indicating reduced myelin loss after treatment. Both AAV-treated cats were clinically normal at the time of imaging. In 5 of 11 treated cats tested, a locus of hyperintensity was noted in the thalamus [white arrowhead in (G)], although no clinical or histopathological correlates have been identified to date. Ages in months are shown for each cat. MRI images were acquired by Heather L. Gray-Edwards. 73 Figure S1. ?gal distribution in the CNS of GM1 gangliosidosis cats 16 weeks after AAV1 treatment. GM1 gangliosidosis cats were injected bilaterally in the thalamus and deep cerebellar nuclei with AAV1-CBA-?gal-WPRE (3 ? 1012 to 4 ? 1012 vector genomes total) and tissues were collected 16 weeks later. Block lettering corresponds to Figure 1A. Blocks were halved and analyzed for sialic acid concentration (left) or for enzyme activity (right). Lysosomal ?gal activity (blue) detected with Xgal at acidic pH was visualized throughout the brain (A) and spinal cord (B) of a representative, treated GM1 gangliosidosis cat (GM1+AAV; 9-1553). Corresponding ?gal activity is shown below each block as fold normal level (fold N). Representative control sections are shown from untreated normal cats along with untreated GM1 gangliosidosis cats, which express ?0.10-fold normal ?gal activity in the brain and ?0.04-fold normal ?gal activity in the spinal cord. The ranges of specific activities for normal control blocks were: brain, 13 (a) - 42 (e); spinal cord, 3.2 (l) - 8.8 (k) nmol 4MU/mg/hr. 74 Figure S2. Clinical progression of AAV1-treated GM1 gangliosidosis cats. AAV1-treated cats were scored on the clinical scale shown in Figure 4C. A composite of four cats that responded robustly to treatment is shown (+AAV composite), while one cat that responded less favorably is depicted separately (9-1545). Solid lines signify living cats; dashed lines represent deceased cats. Open symbols denote whole-body tremors, which occurred in all untreated cats but only one treated animal. 75 Figure S3. Hypereosinophilic neurons in the cortex of an AAV-treated GM1 gangliosidosis cat. Cat 9-1515 was treated with the full dose of AAVrh8. Fine eosinophilic granules (arrows) were found in scattered cortical and hippocampal neurons dorsal to the injection site. Histological analysis was provided by Brandon L. Brunson. 76 Figure S4. Hexosaminidase activity in the liver of GM1 gangliosidosis cats treated long term. Cats were treated with the full dose of AAV1 (n = 1) or AAVrh8 (n = 3), or with a one-tenth dose of AAVrh8 (n = 4). Liver was collected at humane endpoint and analyzed for hexosaminidase activity. * = activity was significantly higher in untreated GM1 gangliosidosis cats (n = 4) versus normal cats (n = 4) (P = 0.015); **, activity was significantly lower than in untreated GM1 gangliosidosis cats (P = 0.015). Hex activity in the full dose AAVrh8 group was not significantly different from that of untreated GM1 gangliosidosis cats (P = 0.056). Insufficient cats were available in the full dose AAV1 group for statistical analysis. 77 Table S1. Vector copy number and ?gal activity in AAV-treated GM1 gangliosidosis cats and their offspring Cohort Number & gender of cats Tissue Number of positive samples Vector copy number/?g DNA Number ?gal activity (fold normal, range)E 6 weeksA 3F Gonad 3/3 630-2,800 0.00-0.03 16 weeksA 4F, 3M Gonad 0/7