A GENE THERAPY APPROACH FOR ARGININOSUCCINIC ACIDURIA
Dr Julien Colomban Baruteau
A thesis submitted for the degree of Doctor of Philosophy
University College London
June 2017
DECLARATION
I, Julien Baruteau, confirm that the work presented in this thesis is my own. Where information has been derived from other sources, I confirm that this has been indicated in the thesis.
Where experimental work has been completed by others, this has been stated in the relevant section of this thesis. However, the instances are also listed below:
- Paraffin-embedding and cutting of peripheral organ samples of mice were performed by the Histopathology Department, Institute of Neurology, University College London, UK (Section 2.7.6).
- Periodic acid-Schiff, Oil red O and Masson trichrome staining of murine liver samples were performed by the Histopathology Department, Great Ormond Street Hospital for Children NHS Foundation Trust, London, UK (Section 2.7.9-12).
- Processing of liver samples for electron microscopy was performed by Dr Kerrie Verner, Electron Microscopy Department, Institute of Neurology, University College London, UK (Section 2.8).
Part of the work of this thesis has been published in the following articles for which copyright clearance has been obtained (see Appendices 10.5):
- Baruteau J, et al. Gene therapy for monogenic liver diseases: clinical successes, current challenges and future prospects. J Inherit Metab Dis. In Press.
- Baruteau J, et al. Delivering efficient liver-directed AAV-mediated gene therapy. Gene Ther. 2017 May;24(5):263-264.
- Baruteau J, et al. Expanding the phenotype in argininosuccinic aciduria: need for new therapies. J Inherit Metab Dis. 2017 May;40(3):357-368.
I confirm that these publications were written by me and may therefore partly overlap with my thesis.
Julien Baruteau
1
ACKNOWLEDGEMENTS
It has been a privilege to meet and work under the supervision of talented scientists Dr Simon Waddington, Prof Paul Gissen, Dr Philippa Mills and Dr Suzy Buckley. I would like to express my deep gratitude to my primary supervisor Simon, who has guided me through this work and has been an excellent academic lead. Extremely patient and available, he has paid lots of attention to the project with endless support, kindness and encouragement. I am very grateful to Paul for having believed in this project in its first days, having provided fruitful advices and encouragement for getting it funded and having largely contributed to these achievements. It is a pleasure for me to carry on working with him in developing translational projects of gene therapy for paediatric metabolic patients. I would like to thank Philippa who has demonstrated treasures of patience and kindness throughout these years for teaching me mass spectrometry, gene sequencing, and correcting manuscripts. A big thank you to Suzy and Dr Steve Howe for their teaching and support in this work. A huge thank you to my colleagues in the Gene Transfer Technology Group at the Institute for Women’s Health especially Dany Perocheau for his precious skills and lab tips, in the Translational Omics Group at the Institute of Child Health, to Dr Ahad Rahim and his group at the School of Pharmacy, to Paul Gissen’s group at the MRC Laboratory for Molecular Cell Biology for their friendship and willingness to help. I am confident they all have made good progress with their French skills.
I am indebted to the patients and the metabolic physicians who have taken part in this research. I am sincerely grateful to Action Medical Research for having entrusted me in funding this project and providing me this unique opportunity.
Many thanks to Kelly for her everlasting loving support and to our lovely bright stars, Beatriz Emmanuelle and Constance Louise, who were born during this PhD and have grown up in parallel with this work making it more challenging but so much more inspiring. I am extremely grateful to my parents Remi and Véronique, my brother and sisters and their in-laws Alban & Gabrielle, Florence & Albin, Marie and Alix, my extended French and Brazilian families and all my friends for their very kind and relentless support along these years.
2
ABSTRACT
Argininosuccinate lyase (ASL) is central to two metabolic pathways: i) the liver-based urea cycle, which detoxifies ammonia, ii) the citrulline-nitric oxide cycle, which synthesises nitric oxide from L-arginine. Patients deficient in argininosuccinate lyase present with argininosuccinic aciduria characterised by hyperammonaemia and a multi-organ disease with a severe neurological phenotype. Compared to other urea cycle disorders, argininosuccinic aciduria presents a low frequency of hyperammonaemic crises but a high frequency of cognitive impairment. This paradox questions the causative role of hyperammonaemia in the neuropathology.
An observational UK-wide study was designed to study the natural history. Data about clinical status, neuroimaging and hASL genotyping were collected from 56 patients. Six had molecular analysis performed in this work. A homogeneous neurological phenotype was observed in most patients. hASL sequencing was available in 19 patients and 20 mutations were found. A genotype-phenotype correlation showed that the prognosis was more likely related to genotype rather than severity of hyperammonaemia.
The hypomorph mouse model AslNeo/Neo mimicking the human disease was used to study the neuropathology in argininosuccinic aciduria and showed a neuronal disease with oxidative/nitrosative stress. To define the role of hyperammonaemia in this finding, a gene therapy approach using an adeno-associated viral vector (AAV) encoding the murine Asl gene was delivered in AslNeo/Neo mice. The long-term correction of both pathways was observed: i) the urea cycle after a single systemic injection in adult mice; ii) the citrulline-nitric oxide cycle in the brain after a single systemic injection at birth. The neuronal disease persisted if ammonaemia only was normalised but was dramatically reduced after correction of both ammonaemia and neuronal ASL activity. This demonstrated the key-role of a neuronal disease independent from hyperammonaemia in argininosuccinic aciduria.
This work provides new insight in the neuropathology of argininosuccinic aciduria and a proof of concept of successful AAV-mediated gene therapy.
3
TABLE OF CONTENTS
DECLARATION ...... 1
ACKNOWLEDGEMENTS ...... 2
ABSTRACT ...... 3
TABLE OF CONTENTS ...... 4
LIST OF TABLES ...... 15
LIST OF FIGURES ...... 17
ABBREVIATIONS ...... 25
1. BACKGROUND ...... 27
1.1 Urea cycle and related inherited human diseases ...... 27
1.1.1 The urea cycle ...... 27
1.1.2 Urea cycle defects ...... 28
1.2 Argininosuccinic aciduria ...... 31
1.2.1 Argininosuccinate lyase ...... 31
1.2.2 Pathophysiology ...... 35
1.2.3 Clinical phenotype ...... 39
1.2.4 Diagnosis ...... 45
1.2.5 Therapeutics ...... 46
1.2.6 Long-term outcome ...... 51
1.2.7 Phenotype-genotype correlation ...... 52
1.2.8 Newborn screening ...... 53
1.2.9 Animal models ...... 55
1.3 Gene therapy for monogenic disorders ...... 56
4
1.3.1 Overview of gene therapy development ...... 56
1.3.2 Strategies for gene transfer ...... 60
1.3.3 Liver-directed gene therapy: clinical applications ...... 64
1.3.4 Adeno-associated virus ...... 68
1.3.5 Recombinant adeno-associated viral vectors ...... 73
1.3.6 Main challenges of AAV-mediated gene therapy for liver diseases ...... 80
1.3.7 Urea cycle defects and gene therapy ...... 85
1.4 Hypothesis and aims of this thesis ...... 87
1.4.1 Hypothesis and aim 1 ...... 87
1.4.2 Hypothesis and aim 2 ...... 88
2. MATERIAL AND METHODS ...... 90
2.1 Reagents ...... 90
2.2 Characterisation of a new cohort of patients with argininosuccinic aciduria ...... 90
2.2.1 Clinical case recruitment ...... 90
2.2.2 Consent and ethics approval ...... 91
2.2.3 Clinical phenotyping ...... 91
2.3 hASL mutational analysis ...... 92
2.3.1 Design of primers ...... 92
2.3.2 PCR optimisation of hASL exons ...... 93
2.3.3 Sequencing and data analysis ...... 94
2.4 Phenotyping of AslNeo/Neo mice ...... 97
2.4.1 Study approval ...... 97
2.4.2 Animals ...... 97
5
2.4.3 DNA extraction from mouse tissue ...... 97
2.4.4 Genotyping ...... 98
2.4.5 Weight and survival analysis ...... 104
2.5 Behavioural studies ...... 105
2.5.1 Righting reflex ...... 105
2.5.2 Grid walking test ...... 105
2.5.3 Rotarod ...... 106
2.5.4 Open field test ...... 107
2.5.5 Novel object recognition test ...... 109
2.5.6 Tail suspension test ...... 109
2.6 Biomarkers ...... 110
2.6.1 Blood sampling and analysis ...... 110
2.6.2 Urine sampling and analysis ...... 111
2.6.3 Nitrite and nitrate measurements (Griess reaction) ...... 111
2.6.4 Cyclic guanosine monophosphate (cGMP) measurement ...... 113
2.6.5 Glutathione analysis...... 114
2.7 Histopathological assessment ...... 115
2.7.1 Perfusion of mice, organ collection and storage ...... 115
2.7.2 Sectioning of mice brains on freezing microtome ...... 116
2.7.3 Stereoscopic fluorescence microscopy ...... 116
2.7.4 Brain free-floating immunohistochemical staining ...... 116
2.7.5 Brain free-floating immunofluorescence ...... 117
2.7.6 Paraffin-embedded immunohistochemical staining for systemic organs . 118
2.7.7 Nissl staining ...... 119 6
2.7.8 Terminal Transferase-Mediated d-UTP Nick End-Labelling (TUNEL) staining ...... 120
2.7.9 Haematoxylin and eosin (H&E) staining ...... 120
2.7.10 Periodic Acid-Schiff (PAS) staining ...... 121
2.7.11 Masson trichrome staining ...... 121
2.7.12 Oil red O staining ...... 122
2.7.13 Microscopy and images ...... 122
2.7.14 Quantification of staining ...... 123
2.8 Electronic microscopy ...... 123
2.9 Western Blot of murine argininosuccinate lyase ...... 125
2.10 Green fluorescent protein enzyme linked immunosorbent assay ...... 126
2.11 Quantitative PCR ...... 127
2.12 Cloning of the AAV vector construct ...... 129
2.12.1 Plasmids ...... 129
2.12.2 DNA extraction ...... 130
2.12.3 Amplification of mAsl ...... 130
2.12.4 Agarose gel electrophoresis ...... 131
2.12.5 DNA gel extraction ...... 131
2.12.6 Restriction enzyme digestion ...... 131
2.12.7 Alkaline phosphatase treatment ...... 132
2.12.8 Ligation ...... 132
2.12.9 Bacterial transformation and expansion of colonies ...... 132
2.12.10 DNA extraction, digestion and sequencing ...... 133
2.13 AAV vector production ...... 135
7
2.13.1 Plasmid production ...... 136
2.13.2 Triple transfection ...... 136
2.13.3 Virus purification by affinity chromatography ...... 137
2.13.4 Vector titration ...... 140
2.13.5 Analysis of capsid proteins with Coomassie gel ...... 141
2.14 Quantification of urea cycle amino acids in dried blood spots and ASL enzyme activity in tissue samples ...... 142
2.14.1 Mass spectrometer parameters ...... 142
2.14.2 Quantification of analytes using stable isotopes ...... 143
2.14.3 Derivatisation of amino acids with 9-fluorenylmethyl chloroformate ...... 144
2.14.4 Optimised high-performance liquid chromatography parameters ...... 145
2.14.5 Analysis of dried blood spot amino acids ...... 146
2.14.6 Protein extraction of tissues for ASL enzymatic assay ...... 146
2.14.7 Optimised protocol for analysis of ASL activity ...... 147
2.15 Statistical analysis ...... 147
3. EXPANDING THE PHENOTYPE OF ARGININOSUCCINIC ACIDURIA ...... 149
3.1 Introduction ...... 149
3.2 Patients ...... 150
3.3 Neurological phenotype ...... 150
3.4 Systemic phenotype ...... 156
3.5 Biomarkers and therapies ...... 160
3.6 Genotype-phenotype correlation ...... 163
3.7 Discussion ...... 167
3.7.1 An expanding phenotype of argininosuccinic aciduria ...... 167
8
3.7.2 Pathophysiology of argininosuccinic aciduria ...... 170
3.7.3 Need for new therapies ...... 173
3.8 Conclusion ...... 174
4. ASSAY DEVELOPMENT FOR FUNCTIONAL CHARACTERISATION OF THE UREA CYCLE ...... 176
4.1 Introduction ...... 176
4.1.1 Detection of analytes by tandem mass spectrometry ...... 176
4.1.2 Separation of analytes by chromatography ...... 181
4.2 Development of a liquid chromatography - tandem mass spectrometry (LC-MS/MS) method for the quantification of urea cycle-related amino acids ...... 182
4.2.1 Detection of amino acids using tandem mass spectrometer ...... 182
4.2.2 Optimisation of mobile phase conditions used for LC-MS/MS analysis ... 184
4.2.3 Extraction of amino acids from dried blood spots ...... 187
4.2.4 Quantification of amino acids using stable isotopes ...... 192
4.3 Development of an LC-MS/MS based assay to measure ASL enzyme activity ...... 201
4.3.1 Previous published assays described for assessing argininosuccinate lyase activity ...... 201
4.3.2 Development and optimisation of argininosuccinate lyse activity assay .. 202
4.4 Discussion ...... 211
4.5 Conclusion ...... 213
5. PHENOTYPING OF AslNeo/Neo MICE ...... 214
5.1 Introduction ...... 214
5.2 Systemic phenotype ...... 215
9
5.2.1 Macroscopic phenotype ...... 215
5.2.3 Blood and urine biomarkers ...... 220
5.2.4 Liver phenotype ...... 224
5.3 Neurological phenotype ...... 231
5.3.1 Motor phenotype ...... 231
5.3.2 Neuropathology ...... 233
5.3.3 Nitric oxide metabolism in the brain ...... 241
5.4 Discussion ...... 243
5.4.1 Recapitulation of the human disease ...... 243
5.4.2 Neurological phenotype ...... 244
5.5 Conclusion ...... 246
6. VECTOR DESIGN FOR Asl GENE TRANSFER ...... 248
6.1 Introduction ...... 248
6.2 Vector design and delivery ...... 249
6.2.1 Rationale for vector design ...... 249
6.2.2 Target structures and cell types ...... 251
6.2.3 Timing of delivery ...... 252
6.2.4 Method of delivery ...... 252
6.3 EFS promoter study ...... 253
6.3.1 Experimental design ...... 253
6.3.2 Biodistribution of AAV vector constructs in peripheral organs ...... 254
6.3.3 Biodistribution of AAV vector constructs in the central nervous system .. 263
6.4 Discussion ...... 267
6.5 Conclusion ...... 269 10
7. Asl GENE THERAPY TO TREAT AslNeo/Neo MICE ...... 270
7.1 Introduction ...... 270
7.2 Experimental design ...... 270
7.3 Effect on the macroscopic phenotype ...... 274
7.3.1 Survival ...... 274
7.3.2 Growth ...... 275
7.3.3 Fur phenotype ...... 276
7.4 Effect on the urea cycle ...... 279
7.4.1 Hepatomegaly ...... 279
7.4.2 Biomarkers ...... 280
7.4.3 Hepatocyte transduction and ASL enzymatic activity ...... 285
7.5 Effect on the citrulline-nitric oxide cycle in the liver ...... 289
7.6 Effect on the citrulline-nitric oxide cycle in the brain ...... 289
7.7 Functional impact on motor and neurological functions ...... 293
7.7.1 Behaviour ...... 293
7.7.2 Cell death in the brain ...... 294
7.8 Discussion ...... 296
7.8.1 Correction of the urea cycle ...... 296
7.8.2 Correction of the citrulline-nitric oxide cycle ...... 299
7.8.3 Pathophysiology of the brain disease in argininosuccinic aciduria ...... 300
7.9 Conclusion ...... 305
8. SUMMARY AND FUTURE WORK ...... 307
8.1 Expansion of the clinical and genetic spectrum of argininosuccinic aciduria ...... 307
11
8.2 Oxidative/nitrosative stress not mediated by hyperammonaemia causes a neuronal disease ...... 308
8.3 ASL gene transfer, a new therapy for argininosuccinic aciduria ...... 309
8.3.1 Proof of concept of AAV-mediated gene therapy in argininosuccinic aciduria ...... 309
8.3.2 Considerations for clinical translation ...... 310
8.3.3 Application for neurohepatotropic diseases ...... 312
8.4 Overall conclusion ...... 312
9. REFERENCES ...... 314
10. APPENDICES ...... 370
10.1 Asl gene therapy with repeated injections in AslNeo/Neo mice: preliminary results ...... 370
10.1.1 Design of the experiment ...... 370
10.1.2 Preliminary results ...... 371
10.1.3 Discussion ...... 379
10.2 Development of a new model of brain conditional Asl knockout mouse: preliminary results ...... 380
10.2.1 The Cre-Lox system in the AslNeo/Neo mouse ...... 380
10.2.2 Titration of the AAV9.Cre vector ...... 382
10.2.3 Preliminary results ...... 385
10.2.4 Discussion ...... 392
10.3 Protocols for the preparation of general laboratory reagents ...... 394
10.3.1 Reagents for PCR amplification of hASL exons ...... 394
10.3.2 Reagents for sequencing reaction ...... 394
10.3.3 Reagents and buffers for DNA extraction for genotyping ...... 395
12
10.3.4 Reagents and buffers for DNA electrophoresis on agarose gel for genotyping ...... 396
10.3.5 Reagents for blood sampling ...... 397
10.3.6 Reagents and buffers for nitrite and nitrate measurements (Griess reaction) ...... 397
10.3.7 Method for gelatin coating slides ...... 399
10.3.8 Reagents and buffers for perfusion of mice and organ collection ...... 399
10.3.9 Reagents and buffers for brain sectioning on freezing microtome ...... 400
10.3.10 Reagents and buffers for immunostaining in free-floating sections ..... 401
10.3.11 Reagents and buffers for immunostaining in paraffin-embedded slides ...... 402
10.3.12 Reagents and buffers for NISSL staining ...... 404
10.3.13 Reagents and buffers for TUNEL staining ...... 405
10.3.14 Reagents and buffers for Haematoxylin and eosin (H&E) staining ...... 407
10.3.15 Reagents and buffers for Periodic Acid-Schiff (PAS) staining ...... 408
10.3.16 Reagents and buffers for Masson trichrome staining ...... 408
10.3.17 Reagents and buffers for Oil Red O staining ...... 409
10.3.18 Reagents and buffers for mASL western blot ...... 410
10.3.19 Reagents and buffers for GFP Enzyme linked immunosorbent assay (ELISA) ...... 413
10.3.20 Reagents and buffers for WPRE quantitative polymerase chain reaction (qPCR) ...... 414
10.3.21 Reagents and buffers for cloning the vector construct ...... 414
10.3.22 Method for making Agar plates ...... 415
13
10.3.23 Reagents and buffers for AAV vector production, concentration and titration ...... 416
10.3.24 Reagents for identification of amino acids by mass spectrometry ...... 419
10.3.25 Reagents for separation of amino acids by liquid chromatography ..... 420
10.3.26 Reagents and buffer for ASL assay in liver extracts ...... 421
10.3.27 Positive controls for liver staining ...... 422
10.4 Database for clinical study ...... 423
10.4.1 Systemic phenotype ...... 423
10.4.2 Neurological phenotype ...... 425
10.5 Publications related to this work and copyright clearance ...... 427
10.5.1 Baruteau et al. Gene therapy for monogenic liver diseases: clinical successes, current challenges and future prospects. Journal of Inherited Metabolic Diseases, May 2017; In Press ...... 427
10.5.2 Baruteau et al. Delivering efficient liver-directed AAV-mediated gene therapy. Gene Therapy, May 2017;24(5):263-264 ...... 449
10.5.3 Baruteau et al. Expanding the phenotype in argininosuccinic aciduria: need for new therapies. Journal of Inherited Metabolic Diseases, May 2017; 40(3):357-368 ...... 452
14
LIST OF TABLES
Table 1. Clinical symptoms of acute and chronic presentations of urea cycle defects ...... 29
Table 2. Prevalence and inheritance of urea cycle defects ...... 30
Table 3. Features of different strategies for gene delivery ...... 62
Table 4. Comparative study of cellular receptors/co-receptors and seroprevalence by adeno-associated virus serotype ...... 72
Table 5. Preferential tropism of the main adeno-associated virus serotypes ...... 72
Table 6. Sequential PCR optimisation for the 16 exons of hASL ...... 96
Table 7. PCR reaction mix for genotyping ...... 98
Table 8. Mastermix used for quantitative PCR ...... 128
Table 9. Parameters used for the detection and quantification of amino acids by multiple reaction monitoring ...... 143
Table 10. Mobile phase gradient elution profile used for high-performance liquid chromatography ...... 146
Table 11. Epidemiological and clinical data for the three analysed cohorts: early- onset, late-onset and screened patients ...... 152
Table 12. Pair comparison for statistical tests ...... 162
Table 13. Genotype-phenotype correlation of hASL ...... 164
Table 14. Allele frequency of novel mutations in Eva-ExAC and Ensembl databases ...... 166
Table 15. Optimised mass spectrometry parameters used for the detection of underivatised urea cycle amino acids in positive ion mode ...... 183
Table 16. Optimised mobile phase gradient profile ...... 185
Table 17. Linearity parameters for each analyte ...... 193
15
Table 18. Recovery, intra- and inter-batch coefficient of variations for each analyte ...... 196
Table 19. Limits of detection and quantification for each analyte ...... 198
Table 20. Detailed general phenotype of the patients included in the clinical study 424
Table 21. Detailed neurological phenotype of patients included in the clinical study ...... 426
16
LIST OF FIGURES
Figure 1. The urea cycle ...... 28
Figure 2. Metabolic pathways involving argininosuccinate lyase ...... 31
Figure 3. Arginine-derived metabolic pathways ...... 33
Figure 4. L-arginine remains an essential amino acid after liver transplantation in argininosuccinic aciduria ...... 34
Figure 5. Metabolic pathways affected by argininosuccinate lyase deficiency ...... 36
Figure 6. Glutamate-glutamine biosynthesis pathway ...... 37
Figure 7. Creatine biosynthesis pathway ...... 38
Figure 8. Median protein intake in argininosuccinic aciduria by age and country of origin in Western Europe ...... 48
Figure 9. The Gartner Hype cycle of gene therapy ...... 59
Figure 10. Factors influencing gene transfer ...... 60
Figure 11. Clinical trials of liver-directed gene therapy for monogenic liver disorders ...... 67
Figure 12. Genomic structure of AAV ...... 70
Figure 13. Phylogenetic tree of VP1 protein sequence of primate AAV serotypes ... 71
Figure 14. Synthesis of an AAV vector ...... 74
Figure 15. AAV vector uptake and in-cell processing ...... 76
Figure 16. Recombinant single-stranded (ss) and self-complementary (sc) recombinant AAV vector genomes ...... 77
Figure 17. The triple transfection method for adeno-associated virus vector production ...... 79
Figure 18. Species-related differences in transduction of the hepatic lobule by AAV vectors ...... 83 17
Figure 19. Liver mass according to age and gender ...... 84
Figure 20. Binding sites of primers for genotyping wild-type and AslNeo alleles...... 100
Figure 21. Optimisation of the annealing temperature for genotyping PCR ...... 102
Figure 22. Optimisation of the primer concentration for genotyping PCR ...... 102
Figure 23. Optimisation of MgCl2 concentration in the presence or absence of PCR enhancer for genotyping PCR ...... 103
Figure 24. Optimised genotyping PCR protocol ...... 104
Figure 25. Picture of a 2 week-old mouse performing the grid walking test ...... 106
Figure 26. Picture of rotarod test ...... 107
Figure 27. Open field test: mouse tracking using Vernier Video Physics software . 108
Figure 28. Open field test: mouse tracking using MouseLabTracker application .... 108
Figure 29. Tail suspension test ...... 110
Figure 30. Nitric oxide metabolism and principle of Griess reaction ...... 112
Figure 31. Design of the primers used for mAsl insert amplification before cloning.130
Figure 32. Bgl1 digestion of the ligated plasmid ...... 134
Figure 33. Steps for AAV vector production...... 135
Figure 34. Purification by affinity chromatography ...... 139
Figure 35. AAV vector titration by alkaline gel ...... 140
Figure 36. AAV vector purity assessed by Coomassie gel ...... 142
Figure 37. Neuroimaging features of argininosuccinic aciduria patients ...... 155
Figure 38. Natural history of argininosuccinic aciduria ...... 157
Figure 39. Linear regression of creatinine clearance versus age ...... 159
Figure 40. Observed levels of common biomarkers in screened, early- and late-onset ASA patients ...... 161
18
Figure 41. Evolutionary conservation of novel missense mutations for hASL ...... 166
Figure 42. Schematic of electrospray ionisation ...... 178
Figure 43. Schematic of multiple reaction monitoring analysis in a triple quadrupole mass spectrometer ...... 179
Figure 44. Formation of the 2 anhydrides of argininosuccinic acid ...... 180
Figure 45. Chromatogram of each analyte and their internal standards ...... 186
Figure 46. Optimisation of extraction of analytes from a 3.2 mm dried blood spot punch ...... 190
Figure 47. Final optimised protocol for the measurement of the concentration of the amino acids of interest in a dried blood spot...... 191
Figure 48. Calibration curves of urea cycle amino acids ...... 194
Figure 49. Study of impact of matrix effect for each analyte ...... 200
Figure 50. . Steps of the urea cycle measured previously in order to assess argininosuccinate lyase activity ...... 201
Figure 51. Structures monitored during ASL assay when D7-citrulline used as a substrate...... 203
Figure 52. Urea cycle amino acids synthesised during the ASL assay when D7- citrulline was used as a substrate ...... 205
Figure 53. Development of argininosuccinate lyase enzyme assay ...... 207
Figure 54. The effect of incubation time on the conversion of argininosuccinic acid ...... 209
Figure 55. Argininosuccinate lyase activity in liver and brain samples from AslNeo/Neo and wild-type mice ...... 210
Figure 56. Survival of wild-type, heterozygote and AslNeo/Neo mice ...... 215
Figure 57. Growth and growth velocity of wild-type, heterozygote and AslNeo/Neo mice ...... 216
19
Figure 58. Macroscopic phenotype of AslNeo/Neo and wild-type mice during the first 3 weeks of life ...... 217
Figure 59. Long-term fur pattern in a 3 month-old surviving AslNeo/Neo mouse ...... 218
Figure 60. Microscopic examination of AslNeo/Neo and wild-type mice hair ...... 219
Figure 61. General biochemistry parameters in plasma of wild-type, heterozygote and AslNeo/Neo mice ...... 221
Figure 62. Urea cycle-related biomarkers tested in wild-type, heterozygote and AslNeo/Neo mice ...... 223
Figure 63. Hepatomegaly in AslNeo/Neo mice compared to wild-type mice ...... 224
Figure 64. Liver stainings highlight increased glycogen deposits in AslNeo/Neo mice compared to wild-type mice ...... 226
Figure 65. Argininosuccinate lyase immunostaining in liver, skeletal muscle, heart and kidney in AslNeo/Neo and wild-type mice ...... 227
Figure 66. Electron microscopy of liver samples from AslNeo/Neo and wild-type mice 228
Figure 67. Western blot and densitogram of murine argininosuccinate lyase present in liver extracts of AslNeo/Neo and wild-type mice ...... 229
Figure 68. Argininosuccinate lyase activity in livers from 6 month-old AslNeo/Neo and wild-type mice ...... 229
Figure 69. Liver nitrate/nitrite and reduced glutathione levels in liver samples from 8 week-old AslNeo/Neo and wild type mice ...... 230
Figure 70. Behavioural tests in wil- type, heterozygote and AslNeo/Neo sex-matched mice ...... 232
Figure 71. Brain hypertrophy in AslNeo/Neo mice compared to wild-type mice ...... 233
Figure 72. Brain ASL activity in AslNeo/Neo mice compared to wild-type mice ...... 234
Figure 73. Morphology of wild-type and AslNeo/Neo mouse brains (low magnification) ...... 236
20
Figure 74. Morphology of wild-type and AslNeo/Neo mouse brains (high magnification) ...... 237
Figure 75. Argininosuccinate lyase immunostaining in wild-type and AslNeo/Neo mouse brains ...... 238
Figure 76. Immunostaining assessing the oxidative/nitrosative stress in wild-type and AslNeo/Neo mouse brains ...... 239
Figure 77. TUNEL staining of wild-type and AslNeo/Neo mouse brains ...... 240
Figure 78. Neuronal oxidative/nitrosative stress is a component of the neurological disease in AslNeo/Neo mice ...... 242
Figure 79. Schematic of the designed vector ...... 251
Figure 80. Promoter study design ...... 254
Figure 81. Effect of promoter and mode of delivery of vector on GFP fluorescence of brain and peripheral organs ...... 255
Figure 82. GFP immunostaining in liver ...... 257
Figure 83. GFP immunostaining in skeletal muscle, heart and skin after intravenous injection ...... 259
Figure 84. Quantification of GFP in peripheral organs using ELISA ...... 261
Figure 85. Quantitative PCR analysis of the number of vector genomes and the effect of CMV on GFP expression in liver samples ...... 262
Figure 86. Decreasing rostrocaudal gradient in EFS injected brains revealed by GFP immunostaining ...... 264
Figure 87. Specific transduction of neurons using the EFS promoter revealed by GFP immunostaining (high magnification) ...... 265
Figure 88. Specific transduction of neurons using the EFS promoter revealed by GFP and NeuN immunofluorescence ...... 266
Figure 89. Supportive treatment in AslNeo/Neo mice improves survival ...... 272
Figure 90. Growth of AslNeo/Neo mice with or without supportive treatment ...... 272 21
Figure 91. Supportive treatment given to AslNeo/Neo mice ...... 273
Figure 92. Survival curves of adult- and neonatally-injected AslNeo/Neo mice ...... 274
Figure 93. Macroscopic aspect of gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice ...... 275
Figure 94. Growth of gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice ...... 277
Figure 95. Correction of the fur pattern in adult-injected injected AslNeo/Neo mice .... 278
Figure 96. Hepatomegaly ...... 279
Figure 97. Biomarkers of the urea cycle ...... 283
Figure 98. Alanine aminotransferase activity and creatinine concentrations in wild- type and AslNeo/Neo mice ...... 284
Figure 99. Argininosuccinate lyase activity, vector genome copy number and immunostaining in liver samples ...... 286
Figure 100. Liver argininosuccinate lyase immunostaining ...... 287
Figure 101. Microscopic liver architecture ...... 288
Figure 102. Liver nitrate/nitrite levels ...... 289
Figure 103. Cortical argininosuccinate lyase activity and nitric oxide metabolism in brain ...... 291
Figure 104. Cortical nitrotyrosine immunostaining ...... 292
Figure 105. Behavioural testing ...... 293
Figure 106. TUNEL staining in the cortex and the cerebellum ...... 295
Figure 107. Proposed pathophysiology involved in the brain disease of AslNeo/Neo mice ...... 303
Figure 108. Experimental design for repeated injections of Asl gene therapy in AslNeo/Neo mice ...... 371
22
Figure 109. Survival curve of AslNeo/Neo mice receiving 2 sequential injections of Asl gene therapy at birth and D30 ...... 372
Figure 110. Macroscopic aspect of gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice ...... 373
Figure 111. Growth of gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice ...... 374
Figure 112. Hepatomegaly and brain/body weight ratio in gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice ...... 374
Figure 113. Biomarkers of the urea cycle ...... 377
Figure 114. Argininosuccinate lyase activity in gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice ...... 378
Figure 115. Behavioural testing ...... 378
Figure 116. Asl gene construct used to create the AslNeo/Neo mouse ...... 381
Figure 117. Experimental design for characterisation of a conditional neuronal Asl knock-out mouse model ...... 382
Figure 118. Survival of AslNeo/Neo mice injected intracranially with the escalating doses of AAV9.Cre vector compared to uninjected AslNeo/Neo mice ...... 383
Figure 119. Growth of wild-type and AslNeo/Neo mice injected intracranially with escalating doses of AAV9.Cre vector compared to uninjected controls ...... 384
Figure 120. Gross phenotype of Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups ...... 386
Figure 121. Survival curves of Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups ...... 387
Figure 122. Righting reflex in Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups ...... 387
Figure 123. Hepatomegaly and brain/body weight ratios in Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups ...... 388 23
Figure 124. Dried blood spot urea cycle-related amino acid analysis of Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups ...... 390
Figure 125. ASL activity in the liver and brain of Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups ...... 391
Figure 126. Positive controls for Periodic acid-Schiff, Masson trichrome and Oil red O stainings ...... 422
24
ABBREVIATIONS
1H-MRS Proton magnetic resonance spectroscopy
AAP Assembly-activating protein
AAV Adeno-associated virus
ADA-SCID Adenosine deaminase-severe combined immunodeficiency
ADHD Attention deficit hyperactivity disorder
AIPA (2S)-(+)-Amino-5-iodoacetamidopentanoic acid
ALP Alkaline phosphatase
ALT Alanine amino transferase
ASA Argininosuccinic aciduria
Asl Murine argininosuccinate lyase gene
ASL Argininosuccinate lyase protein
ASLD Argininosuccinate lyase deficiency
ASS Argininosuccinate synthase
AST Aspartate amino transferase
BCA Bicinchoninic acid
BH2 Dihydrobiopterin;
CAT1 Cationic amino acid transporter 1 cGMP Cyclic guanosine monophosphate
CLIC/GEEC Clathrin-independent carriers / Glycosylphosphatidylinositol- enriched
25
CMV Cytomegalovirus
CPS: Carbamylphosphate synthase
CREB Cyclic adenosine deaminase response-element binding protein
DAB 3,3’ Diaminobenzidine ddH2O Double distilled water dH2O Distilled water
DNA Deoxyribonucleic acid dNTP Deoxyribonucleotide triphosphate ddNTP Dideoxyribonucleotide triphosphate dUTP Deoxyuridine trisphosphate
EFS Elongation Factor 1 α short promoter
ELISA Enzyme linked immunosorbent assay
ESI Electrospray ionisation
EMA European Medicines Agency
FMOC 9-fluorenylmethyl chloroformate
G Universial gravitational constant
GAMT Guanidinoacetate methyltransferase
GAPDH Glyceraldehyde-3-phosphate dehydrogenase
GDH Glutamate dehydrogenase;
GFP Green fluorescent protein
GGT Gamma glutamyltranspeptidase
GS Glutamine synthase
GSH Reduced glutathione
GSSG Oxidised glutathione 26
GT Gene therapy-treated
GTMP Gene therapy medicinal product
H Hydrogen element
H2O2 Hydrogen peroxide
hASL Human argininosuccinate lyase gene
H&E Haematoxylin and eosin
HCC Hepatocellular carcinoma
HCl Chloridric acid
HDAC2 Histone deacetylase 2
HD-Ad Helper-dependent adenoviral vectors
HEK293T Human embryonic kidney 293T cells hHGF Human hepatocyte growth factor
HRP Horse Radish Peroxidase
HSP90 Heat Shock Protein 90
IEC Ion exchange chromatography
IQ Intelligence quotient
ITR Inverted terminal repeats
LOD Limit of detection
LOQ Limit of quantitation
LC-MS/MS Liquid chromatography/tandem mass spectrometry mAsl Murine argininosuccinate lyase gene
MHCI Major histocompatibility complex I
MPS VII Mucopolysaccharidosis type VII
27
MQH2O Milli-Q Ultra pure Water Purification Systems (Millipore, Brussels, Belgium)
MRI Magnetic resonance imaging
MRM Multiple reaction monitoring
MUT Mutant
N Nitrogen element
NAGS N-acetyl glutamate synthase
NKCC1 Na+K+Cl-cotransporter isoform 1
NADP Nicotinamide adenine dinucleotide phosphate
NH3 Ammonia
+ NH4 Ammonium, ionised molecule of ammonia
NMDA N-methyl-D-aspartate
- NO2 Nitrite
- NO3 Nitrate
NOS Nitric oxide synthase
O Oxygen element
- O2 Superoxide
- ONO2 Peroxynitrite
ORF Open reading frame
OTC: Ornithine transcarbamylase
PAS Periodic Acid-Schiff
PB Phosphate buffer
PBS Phosphate buffered saline
PCR Polymerase chain reaction
28
qPCR Quantitative polymerase chain reaction
PEI Polyethylenimine
PFA Paraformaldehyde
PVDF Polyvinylidene difluoride
RSNO Nitrosothiols
SD Standard deviation
SOC Super-optimal broth with catabolite repression
STN Signal-to-noise
RBE Replication binding elements
Rpm Revolutions per minute
RNA Ribonucleic acid
SCID –X1 X-linked severe combined immunodeficiency shRNA Short hairpin RNA siRNA Small interfering RNA
SV40 Simian virus 40
TdT Terminal deoxytransferase
TLR Toll like receptors
TNFα Tumor necrosis factor α tRNA Transfer RNA
TAE solution Tris base, acetic acid, EDTA
TBS solution Tris-buffered saline
TBS-T Tris-buffered saline-Triton
TLR Toll like receptor
TMB Tetramethylbenzidine 29
TUNEL Terminal Transferase-Mediated d-UTP Nick End-Labelling
UCD Urea cycle defect
UNG Uracil N-glycosylase
UTR Untranslated region
UV Ultraviolet
UK United Kingdom
WT Wild-type
WPRE Woodchuck hepatitis virus post-transcriptional regulatory element
30 Chapter 1 - Background
1. BACKGROUND
1.1 Urea cycle and related inherited human diseases
1.1.1 The urea cycle
Enzymes of the urea cycle have 2 aims: i) detoxification of ammonia (NH3), a highly neurotoxic molecule produced by deamination of amino acids during the catabolism of proteins, into urea, and ii) de novo biosynthesis of L-arginine (1).
The urea cycle is compartmentalised within the hepatic lobule in periportal hepatocytes, a phenomenon called metabolic zonation (2).
The urea cycle (Figure 1) relies on
- 5 catalytic enzymes: Carbamoylphosphate synthetase 1 (CPS1), Ornithine transcarbamylase (OTC), Argininosuccinic synthetase (ASS), Argininosuccinic acid lyase (ASL), Arginase (ARG), - One enzyme producing a co-factor N-acetylglutamate synthetase (NAGS), which acts as an allosteric activator of the CPS enzyme
- One mitochondrial transporter: citrin
CPS1, NAGS and OTC are located in the mitochondria and are the first 3 enzymes involved in the cycle; a deficiency in one of these enzymes is known as proximal urea cycle defect and is associated with severe hyperammonaemic crisis; ASS, ASL and ARG are cytosolic enzymes and their deficiencies are considered as distal urea cycle defects.
27 Chapter 1 - Background
MITOCHONDRIA CYTOSOL
CPS NH + Carbamoyl phosphate 4 UREA Acetyl-CoA + Ornithine
Glutamate N-acetyl-Glutamate ARG NAGS OTC
Citrin Citrulline Arginine
Aspartate ASL ASS Argininosuccinate
Figure 1. The urea cycle
ARG: Arginase, ASL: Argininosuccinic acid lyase, ASS: Argininosuccinic synthetase, CPS1: Carbamoylphosphate synthetase type 1, NAGS: N-acetylglutamate + synthetase, NH4 : Ionised molecule of ammonia, OTC: Ornithine transcarbamylase.
1.1.2 Urea cycle defects
If one of the urea cycle enzyme becomes deficient, the impaired clearance of ammonia will cause hyperammonaemia and provoke an endogenous intoxication with clinical symptoms. The onset of symptoms is correlated with the severity of the enzymatic deficiency. For example, severe forms of urea cycle defects can present as neonatal coma at the end of the first week of life. Patients can be divided into early-onset (symptoms identified at the age of 28 days or before) and late-onset (symptoms observed after 28 days of life) phenotypes (3). A genotype-phenotype correlation is often observed and helps predicting an early- or late-onset form (4).
Hyperammonaemic crises are associated with a severe outcome as the survival and the neurological prognosis are correlated with the level and the duration of hyperammonaemia (5). Without treatment, most of the patients will die (6, 7). A large spectrum of neurological sequelae can be observed: learning difficulties, neurocognitive impairment, behavioral, neurosensory and locomotor difficulties,
28 Chapter 1 - Background epilepsy (8) with various neuroimaging abnormalities (cortical atrophy, ventricular dilatation, lesions of basal ganglia and thalamus, acute ischemia, myelination defects (9-11)). Early recognition is critical to rapidly reduce hyperammonaemia. However symptoms are usually non-specific and heterogeneous with neurocognitive, psychiatric, liver, digestive symptoms or failure to thrive (Table 1) (12).
Acute presentation Chronic presentation
Alteration of consciousness (from Dizziness, Confusion, somnolence to coma), Tiredness, Lethargy, Neurological Acute encephalopathy, Headaches, Seizures, Ataxia, Stroke-like episodes Learning disability, Mental retardation Psychiatric Hallucinations, Paranoia Mood alteration, Behavioural changes
Digestive Vomiting, Loss of appetite Vomiting, Abdominal pain
Liver Liver failure Hepatomegaly, transaminitis
In neonates: sepsis-like presentation with Other hypothermia, hypo- or hyperventilation, Protein aversion multiorgan failure
Table 1. Clinical symptoms of acute and chronic presentations of urea cycle defects
The prevalence of urea cycle defects (UCDs) is estimated between 1/8,200 (13) to 1/35,000 (Table 2) (14). Argininosuccinic aciduria accounts for 15% of all urea cycle defects and is considered as the second most common urea cycle defect after ornithine transcarbamylase deficiency (Table 2). Argininosuccinic aciduria prevalence is evaluated by different groups to be between 1/49,000 (15), 1/70,000 (16), 1/95,600 (17) to 1/218,750 (14). The newborn screening implemented in some western countries increases the prevalence as asymptomatic (also named benign) or mild forms can be diagnosed.
29 Chapter 1 - Background
DEFECT PREVALENCE % UCDs* INHERITANCE GENE
NAGS Deficiency <1:2,000,000 ≤1% Recessive NAGS
CPS1 Deficiency 1/1,300,000 3-8% Recessive CPS1
OTC Deficiency 1/56,500 57-62% X-linked OTC
Citrullinemia 1 1/250,000 13-19% Recessive ASS
Argininosuccinic 1/218,750 11-18% Recessive ASL aciduria
Arginase 1/950,000 2-3% Recessive ARG1
All urea cycle defects 1/35,000 100% / /
Table 2. Prevalence and inheritance of urea cycle defects
* Percentage of total urea cycle defects from the 3 main international registries: the Urea Cycle Disorders Consortium (UCDC) funded by the National Institutes of Health, the European Registry and Network for Intoxication type Metabolic Diseases (E-IMD), and the National Urea Cycle Disorders Foundation (NUCDF) (14). ARG: Arginase; ASL: argininosuccinate lyase; ASS: argininosuccinate synthase; CPS1: Carbamylphosphate synthase type 1; NAGS: N-acetyl glutamate synthase; OTC: ornithine transcarbamylase.
30 Chapter 1 - Background
1.2 Argininosuccinic aciduria
1.2.1 Argininosuccinate lyase
Argininosuccinate lyase (ASL) is a cytosolic protein, which cleaves argininosuccinate into arginine and fumarate. This biochemical reaction belongs to 2 pathways (Figure 2): i) the urea cycle (discussed in Section 1.1), ii) the citrulline-nitric oxide cycle, which synthesises nitric oxide from L-arginine in multiple tissues via nitric oxide synthase. Three isoforms of nitric oxide synthase (NOS) exist: NOS1 or neuronal NOS (nNOS), NOS2 or inducible NOS (iNOS) and NOS3 or endothelial NOS (eNOS).
Figure 2. Metabolic pathways involving argininosuccinate lyase
Argininosuccinate lyase (ASL) belongs to the urea cycle, which detoxifies ammonia into urea, and the citrulline-nitric oxide cycle, which allows nitric oxide (NO) production from nitric oxide synthase (NOS). All enzymes required in the urea cycle are expressed in the liver. The citrulline-NO cycle is expressed in most organs. ASL elicits (i) an enzymatic function in catalysing the production of arginine and fumarate from argininosuccinic acid, and (ii) a structural role required to maintain a multiprotein complex, which includes NOS, argininosuccinate synthase (ASS), a cationic amino acid transporter (CAT-1) and heat shock protein 90 (HSP90).
31 Chapter 1 - Background
ASL is mainly expressed in the liver but is present in various other tissues e.g. fibroblasts, red blood cells, muscle, heart, kidney, small intestine and brain (18, 19).
The structure of the active enzyme is a homotetramer with 4 active enzymatic sites (20) and a molecular weight of 187 kDa (21). The ASL monomer encompasses 464 aminoacids (22). The ASL monomer has 3 distinct and highly conserved domains: domains 1 and 3 are similar with 2 helix-turn-helix in perpendicular arrangement. Domain 2 has 9 helices with 5 of them with up-down-up-down-up topology forming the central helix bundle (23). Three of these 5 central helices bound another ASL subunit to form a dimer with hydrophobic and ionic interactions between arginine and glutamic acid residues (24). Each monomer contributes with one helix of its central helix bundle to the formation of the core of the tetramer.
In humans, ASL is the only enzyme enabling the endogenous supply of L-arginine. In human metabolism, L-arginine supply is provided by exogenous nutritional intake and an ASL-dependent endogenous synthesis (18). L-arginine is a cardinal biochemical precursor for pivotal metabolic pathways: i) synthesis of creatine, a molecule critical for energy metabolism in muscle and brain in recycling adenosine diphosphate into adenosine triphosphate via donation of a phosphate group (25); ii) synthesis of agmatine and polyamines (putrescine, cadaverine, spermine and spermidine) involved in the regulation of gene expression and ion channels response (26); iii) synthesis of nitric oxide involved in vasodilatation, neurotransmission and cell signalling (27) (Figure 3). This could explain the importance of this enzyme and its ubiquitous location.
32 Chapter 1 - Background
Arginine
Glycine ARG1 AGAT ADC NOS Ornithine Ornithine CO2 Citrulline
Guanidinoacetate Agma*ne Nitric oxide Urea
SAM GAMT SAH Putrescine
Crea*ne Spermidine Polyamines
Spermine
Figure 3. Arginine-derived metabolic pathways
ADC: arginine decarboxylase; AGAT: arginine glycine aminotransferase; ARG1: arginase type 1; GAMT: guanidinoacetate methyltransferase; NOS: nitric oxide synthase; SAH: S-adenosylhomocysteine; SAM: S-adenosylmethionine.
Under physiological condition, the endogenous synthesis of L-arginine is sufficient to meet physiological L-arginine supply and L-arginine is a non-essential amino acid. This L-arginine endogenous production is mainly guaranteed by the kidney, which contains both argininosuccinate synthase and ASL enzymes. In argininosuccinic aciduria, L-arginine synthesis is impaired and L-arginine becomes an essential amino acid, even after liver transplantation as the renal production of L-arginine remains deficient (Figure 4) (28). The “arginine paradox” illustrates the observation that despite saturating intracellular L-arginine levels in endothelial cells, supplementation of exogenous L-arginine can increase the endothelial nitric oxide synthase (eNOS)- dependent nitric oxide production (29) up to 2 fold and increase the perivascular tissue perfusion by 35% (30). This suggests an intracellular compartmentalisation or channelling effect of L-arginine to reach the active site of endothelial nitric oxide
33 Chapter 1 - Background synthase. However, this paradox is not observed in argininosuccinic aciduria where a systemic nitric oxide deficiency is observed despite L-arginine supplementation. The discovery of a cytosolic multiprotein complex including the cationic amino acid transporter CAT1, heat shock protein 90, argininosuccinate synthase, ASL and nitric oxide synthase highlighted the pivotal role of ASL in maintaining this complex (Figure 2) (31, 32). Hence ASL has a dual or “moonlighting” effect (33): i) a catalytic function responsible for L-arginine production, ii) a structural role in maintaining a multiprotein complex containing nitric oxide synthase and channelling L-arginine to nitric oxide synthase for nitric oxide production (32).
NH3
Hepatocyte Enterocyte CPS1 CPS1 OTC OTC
ASS Citrulline
ASL Extrahepa/c /ssues (especially kidney) ASS ARG ASL
Arginine Urea
Figure 4. L-arginine remains an essential amino acid after liver transplantation in argininosuccinic aciduria
The major metabolic flux of arginine production relies on the ASS and ASL enzymatic activities of the kidney. In argininosuccinic aciduria, the restoration of a functional urea cycle after liver transplantation does not correct the impaired production of arginine from extrahepatic tissues, which explains why a persisting arginine supplementation is required. ARG: Arginase; ASL: argininosuccinate lyase; ASS: argininosuccinate synthase; CPS1: Carbamylphosphate synthase type 1; OTC: ornithine transcarbamylase.
34 Chapter 1 - Background
ASL is encoded by the human argininosuccinate lyase gene (hASL) (OMIM 608310). This gene was cloned in 1986 (34), is located in chromosome 7 (7q11.21) and contains 17,554 base pairs (bp) divided in 16 exons (21). An exon 0 has been described coding for the 5’ untranslated region (UTR) (35). hASL is a very conserved gene, identified in various species from bacteria and yeast (36), algae (37), vegetables (38), birds (23) and mammals and has been well conserved through evolution confirming the important role of argininosuccinate lyase.
1.2.2 Pathophysiology
Argininosuccinate lyase deficiency causes argininosuccinic aciduria (OMIM 207900), an autosomal recessive disorder. Although the pathophysiology in argininosuccinic aciduria remains partially elusive, various mechanisms summarised in Figure 5 have been hypothesized to play a role in the pathophysiology of the disease.
a) Hyperammonaemia
Ammonia acts in different ways to produce devastating neurological symptoms (39). In case of high ammonia levels, glutamate and glutamine synthesis acts as a short- lasting buffering mechanism (Figure 6). The conversion of α-ketoglutarate into glutamate buffers one molecule of ammonia but can cause cataplerosis of the Krebs cycle with secondary energy failure. Increase of intracellular glutamate stimulates N- methyl-D-aspartate (NMDA) receptors and apoptosis. Glutamine has an osmotic effect and high intracellular glutamine levels cause swelling of astrocytes and cerebral cytotoxic oedema (40). High ammonia levels impair osmoregulation and lead to astrocytic dysfunction, swelling and cerebral oedema, increased intracranial pressure and eventually brain herniation. Brain autopsies showed extensive neuronal loss, gliosis and astrocytes with water-clear, oval nuclei also called Alzheimer’s type II astrocytes (9, 41, 42).
35 Chapter 1 - Background
Figure 5. Metabolic pathways affected by argininosuccinate lyase deficiency
Argininosuccinate lyase (ASL) cleaves argininosuccinate into arginine and fumarate as part of the urea cycle in the liver and the citrulline-nitric oxide cycle in most organs. ASL is part of a multiprotein complex including argininosuccinate synthase (ASS) and nitric oxide synthase (NOS). ASL deficiency leads to reduced arginine, increased ammonia and reduced nitric oxide levels. Nitric oxide acts on the nitric oxide-cyclic guanosine monophosphate (cGMP) signalling pathway to mediate its effects in a wide range of tissues and generates nitrosothiols (RSNO) through protein nitrosylation. Nitric oxide acts on the nitric oxide-cyclic guanosine monophosphate signalling pathway to mediate its effects in a wide range of tissues and generates nitrosothiols (RSNO) through protein nitrosylation. Low arginine or tetrahydrobiopterin (BH4) cause uncoupling of the NOS dimer generating superoxide.
Superoxide forms other reactive oxygen (O2) species like hydrogen peroxide (H2O2) - or peroxynitrite (ONO2 ) involved in free radical damage and acting as signalling molecules. Arginine is a precursor for various pathways i.e creatine, agmatine and polyamine synthesis. Argininosuccinic acid could have a toxic effect by itself or
36 Chapter 1 - Background through the formation of the toxic guanidinosuccinic acid. ASS: Argininosuccinate synthase; ASL: Argininosuccinate lyase; BH2: dihydrobiopterin; BH4: tetrahydrobiopterin; cGMP: cyclic guanosine monophosphate; H2O2 hydrogen - - peroxide; NOS: Nitric oxide synthase; O2 Superoxide; ONO2 peroxynitrite; RSNO: nitrosothiols.
NH3 NH3 α-ketoglutarate Glutamate Glutamine
GDH GS
Figure 6. Glutamate-glutamine biosynthesis pathway
GDH: glutamnate dehydrogenase; GS: glutamine synthase; NH3: ammonia.
Chronic hyperammonaemia causes i) abnormal neurotransmission with increased brain uptake of large neutral amino acids (tryptophan, tyrosine, phenylalanine, methionine, histidine) (43). The imbalance of tryptophan, a precursor of serotonin, disturbs the serotoninergic neurotransmission via alteration of the downstream production of neurotransmitters (39). High cerebrospinal fluid serotonin levels are associated with anorexia, a common symptom in patients with urea cycle defects (44); ii) down-regulation of NMDA receptors with depression of excitatory transmission and impairment of downstream glutamate-nitric oxide-cyclic guanosine monophosphate (cGMP) signalling involved in learning and memory (45); iii) ammonia competitively impairs astrocyte potassium buffering which can provoke seizures and stimulates Na+K+Cl-cotransporter isoform 1 (NKCC1), which alters neurotransmission (46).
It has been proposed that brain sensitivity to ammonia could be age-dependent (47). Experiments in rats have shown a more pronounced impairment of energy metabolism and the mitochondrial respiratory chain caused by ammonia in adult compared to neonatal animals (48).
37 Chapter 1 - Background b) Arginine deficiency
In argininosuccinic aciduria, L-arginine becomes an essential amino acid. Severe hypoargininemia will impair the synthesis of arginine-derived metabolites such as agmatine, polyamine, creatine and nitric oxide (Figure 5). Hypoargininemia affects the synthesis of arginine-enriched tissues like hair (18).
c) Nitric oxide deficiency
Nitric oxide synthesis is catalysed by nitric oxide synthase from L-arginine. Nitric oxide synthase is a dimer, which can uncouple in certain conditions such as arginine or tetrahydrobiopterin (BH4) deficiencies. In case of uncoupling, nitric oxide synthase- dependent nitric oxide production is defective and cells rely on nitric oxide synthase- independent nitric oxide production via the nitrite-nitric oxide pathway used as a bypass (49). This uncoupling generates superoxide and reactive oxygen species causing oxidative stress (Figure 5).
d) Creatine deficiency and increased guanidinoacetate levels
A secondary creatine deficiency caused by low levels of the precursor arginine might cause cerebral energy deficiency (Figures 5 & 7), especially in neurons and alters the respiratory chain and ATP production (50). This inhibit Na,K-ATPase activities and cause cell apoptosis (47).
Glycine Ornithine SAM SAH ATP ADP
Arginine Guanidinoacetate Crea0ne Phosphocrea0ne AGAT GAMT CK (Kidney) (Liver) (Muscle, heart, brain)
H2O
Crea0nine
Figure 7. Creatine biosynthesis pathway
AGAT: arginine glycine aminotransferase; ADP: adenosine di-phisphate; ATP: adenosine triphosphate; CK: creatine kinase; GAMT: guanidinoacetate methyltransferase; SAH: S-adenosylhomocysteine; SAM: S-adenosylmethionine.
38 Chapter 1 - Background
Increased levels of guanidinoacetate in plasma and brain have been reported in argininosuccinic aciduria (51-53) and might correlate with L-arginine supplementation (51) similarly as in lysinuric protein intolerance (54) and arginase deficiency (55). Guanidinoacetate toxicity has been advocated in guanidinoacetate methyltransferase (GAMT) deficiency, a primary inherited creatine biosynthesis disorder causing a severe neurological phenotype with intellectual impairment and epilepsy (56).
1.2.3 Clinical phenotype
The disease was initially described in 2 children with a late-onset phenotype, who presented with mental retardation, ataxic gait and epilepsy, and was considered as “a disease, probably hereditary, characterised by a severe mental deficiency and a constant gross abnormality in amino acid metabolism” with “an unusual urinary amino acid” on the chromatogram (57), which was later characterised as argininosuccinic acid (58).
Similarly to other urea cycle defects, 2 phenotypes based on the age of onset of the disease are described: i) an early-onset phenotype with neonatal hyperammonaemic coma (i.e. patients ≤ 28 days of life) (59), ii) a late-onset phenotype diagnosed > 28 days because of either hyperammonaemic episod or chronic symptoms without acute decompensation.
The following organs can be affected in the course of argininosuccinic aciduria:
a) Brain
Developmental delay shares a broad spectrum of severity from severe mental retardation with an intelligence quotient (IQ) <50 (15, 60) to borderline IQ allowing satisfactory socioeconomic integration (61). Various neurocognitive difficulties are reported: gross and fine motor delay, speech delay, learning disabilities necessitating adapted schooling and sheltered work. This delay is usually diagnosed early, between 12 and 24 months, in untreated or treated patients whatever the phenotype early- or late-onset (61-63).
39 Chapter 1 - Background
Epilepsy has been described in 55% of patients with different types of seizures: tonic clonic, clonic, myoclonic or focal (63). Acute symptomatic seizures related to hyperammonaemia are not predictive for developing an epilepsy later (63). The neuropathology causing epilepsy is not well understood. Ammonia toxicity might not be the primary cause and other mechanisms such as argininosuccinic acid-related toxicity, increased guanidinoacetate or consequences of arginine deficiency have been mentioned as potential causes (63). Electroencephalography is abnormal and shows a non-specific pattern without (41, 64), or with, conventional treatment (63), with spikes, spike-waves, sharp-and-slow waves (63, 65).
Ataxia, hypotonia, and tremor are common features in patients with the late-onset phenotype, including cerebellar dysphonia and dysarthria (60).
Behavioural difficulties including hyperactivity (60, 61), autoagressivity with self- mutilation (53), autism (63) have been reported. Psychiatric presentations exist with paranoid ideas (60), psychotic symptoms (66) or schizophrenia (67).
Forensic studies of brains from patients affected by urea cycle disorders showed intracerebral haemorrhage, cerebral oedema (41, 68), cortical atrophy with ventriculomegaly, prominent cortical neuronal loss (41, 69, 70).
Brain imaging shows:
• With computer tomography: global atrophy, hypodense changes of white matter despite normal ammonia levels, which might reflect mild demyelination caused by low arginine levels (60).
• With magnetic resonance imaging, Takanashi et al have proposed a classification of lesions observed in urea cycle defects: type 1 in neonates with cerebral oedema followed by atrophy; type 2 and 3 in infancy/childhood with extensive parenchymal infarct-like changes and ischemic lesions in cerebral intervascular boundary zones respectively and type 4 in adulthood with symmetric reversible abnormalities of the cingulate gyri, temporal lobes, insular cortex sparing the perirolandic cortex (10). Bilateral microcystic periventricular leukomalacia was observed in the frontal and posterior peritrigonal white matter (63).
40 Chapter 1 - Background
• Cerebral proton magnetic resonance spectroscopy (1H-MRS) in treated patients with argininosuccinic aciduria has shown contradictory results with either increased (52, 53) or decreased (71) brain guanidinoacetate and creatine contents in both white and grey matter. L-arginine supplementation has been used to correct low systemic creatine levels observed in urea cycle defects (72) but this was claimed as causing elevation of brain guanidinoacetate levels (53).
b) Liver
Chronic hepatomegaly and/or elevated transaminases (up to 10 times above the normal range) are the most common reported signs (61, 73) and can be the first symptom leading to the diagnosis (74). In treated patients, high doses of L-arginine supplementation (500 mg/kg/d) have increased transaminases and plasma argininosuccinic acid levels suggesting an argininosuccinic acid-mediated liver toxicity (75). Also hepatomegaly can be increased during hyperammonaemic episodes, the conventional treatment aiming to decrease ammonaemia usually does not modify the enlarged volume of the liver in argininosuccinic aciduria (66).
Mild liver failure as defined by impairment of liver synthetic functions can be observed (76). Fibrosis (74, 77) and cirrhosis (76) have been reported. Hepatocellular carcinoma has been reported even in children (78).
Liver biopsies can be normal (73) or show non-specific changes similar to other urea cycle defects: pale or swollen hepatocytes (40, 74, 79), inflammation, cholestasis, micro- or macrovesicular steatosis (40, 79), glycogen deposits (73, 74), progressive fibrosis (74, 77), cirrhosis (76, 80) or acute pericentral necrosis after a fatal acute hyperammonaemia (42). Glycogen deposits in urea cycle defects are not well understood as enzyme activities of glycogen metabolism are normal (81). This could be caused by a carbohydrate-enriched diet and essential amino acid mixtures with high leucine content used in the treatment of urea cycle defects (81), as leucine triggers insulin secretion and glycogenogenesis (82). Electron microscopy shows enlarged mitochondria with dense matrix and tubular cristae suggesting mitochondrial dysfunction (73).
41 Chapter 1 - Background
The understanding of the frequency and, in some cases, the severity of the liver disease in argininosuccinic aciduria remains poor: ammonaemia or argininosuccinic acid-mediated toxicity (75), elevated citrulline, impaired reparative process after hepatocyte necrosis (79), nitric oxide deficiency associated with hepatic hypoperfusion and liver ischemia (76, 83) have been proposed. It is acknowledged that the progression of the liver disease in argininosuccinic aciduria is independent of the control of ammonia and can continue despite adequate treatment (74).
c) Systemic hypertension
High blood pressure has been observed in patients with either early- or late-onset phenotypes and without correlation between the age and the severity of arterial hypertension. For example, two reported patients with early- and late-onset phenotypes were diagnosed with high blood pressure at 17 and 6 year-old respectively. The first patient required a monotherapy but the second developed a refractory hypertension, which responded to nitrite therapy (84, 85). Arterial hypertension in argininosuccinic aciduria is likely to have a low prevalence (86). This arterial hypertension in argininosuccinic aciduria has been associated with impaired eNOS-dependent nitric oxide production in endothelial cells and subsequent abnormal relaxation of the main vessels such as aorta as observed in an ASL- deficient mouse model (85).
Neonatal pulmonary hypertension has been reported in neonates carrying a polymorphism (Thre1405Asn) in the carbamoyl-phosphate synthetase type 1 (CPS1) gene, affecting the N-acetylglutamate binding domain of the enzyme considered as a rate-limiting step in the urea cycle (87). Plasma levels of arginine and nitric oxide metabolites were decreased suggesting a disturbed vasoregulation of the pulmonary arteries caused by an impaired urea cycle metabolism leading to hypoargininaemia and nitric oxide deficiency.
d) Kidney
Recurrent electrolyte disturbances with hypokalaemia (18) and long-term impairment of renal function (86) have been reported. This could be associated with an increase
42 Chapter 1 - Background waste of potassium excretion and tubular dysfunction. Nephrolithiasis have been described (88).
e) Hair
Hair abnormalities have been described essentially in late-onset phenotype (89): i) Trichorrhexis nodosa is a morphological abnormality of the hair shaft with thickening and nodes, which can break easily. Children can present with brittle, dry and short brush-like hair, as young as 1 year-old (90), and do not require any hair cut since birth as reported for some 7-year old patients (64, 91). This “aminogenic alopecia” or “arginine-responsive alopecia” responds well to L-arginine supplementation (92). ii) Monilethrix (from the Greek “monile” meaning necklace and “thrix” meaning hair) is the morphological aspects of expansion and constriction of the hair with an aspect of strings of beads, iii) Pili torti with an aspect of flattened hair twisting on its axis (89).
The hair contains 10.5% of L-arginine. As arginine is an essential amino acid in argininosuccinia aciduria, it has been suggested that chronic hypoargininaemia might explain these hair features (18).
f) Miscellaneous
Like other urea cycle defects, protein aversion, poor appetite, frequent vomiting are common features (71, 93). Recurrent gastrointestinal symptoms including profuse diarrhoea have been described (94).
Increased frequency in caries have been noticed (64). Other symptoms such as severe dermatitis of the face and genital area (61) or L-arginine-responsive “dry, scaly skin” (95) have been mentioned. The symptoms affecting teeth or skin are rarely associated with other urea cycle defects. It is interesting to note that these symptoms could be at least partly explained by nitric oxide deficiency due to the losses of the protection against gastrointestinal infection, the antimicrobicidal effect of nitric oxide in the mouth, the anti-inflammatory properties of nitrated lipids (96).
43 Chapter 1 - Background g) Asymptomatic
Rüegger et al described a cohort of mild patients affected by argininosuccinic aciduria with 11/31 (36%) being asymptomatic (97).
h) Pregnancies in women affected with argininosuccinic aciduria
Various uneventful pregnancies in argininosuccinic aciduria have been reported (15, 88, 98). A careful monitoring of the dietary needs is crucial with progressive increase of the protein intake especially in the last trimester of the pregnancy. Plasma glutamine level is a reliable biomarker for monitoring the urea cycle function (88). No teratogenicity in argininosuccinic aciduria has been characterised although a dysmorphic neonate was reported but with doubts about the causing effect of a familial genetic trait (99).
44 Chapter 1 - Background
1.2.4 Diagnosis
Common biomarkers can help in suspecting the diagnosis:
a) Hyperammonaemia
A large spectrum of plasma ammonia levels can be observed at diagnosis from severe hyperammonaemia > 4,500 μM in neonatal-onset disease (63), up to normal levels in late-onset forms. Chronic hyperammonaemia and acute hyperammonaemic decompensations are less frequent in argininosuccinic aciduria compared to other proximal urea cycle defects like carbamoyl phosphate synthetase deficiency, ornithine transcarbamylase deficiency or argininosuccinate synthase deficiency as assessed by less frequent prescriptions of ammonia scavenger drugs (97, 100, 101), the longest “honeymoon period” defined as the time between discharge after the initial hyperammonaemic episode and readmission for a second one (59). Normal ammonia levels can be obtained with protein-restriction and L-arginine supplementation alone without the need of ammonia scavengers (76). This is probably explained by the fact that similarly to urea, the molecule of argininosuccinic acid contains 2 nitrogen moieties and contributes directly to nitrogen excretion (102). However this introduces the paradox of argininosuccinic aciduria: among urea cycle defects, this disease shares with arginase deficiency the particularity of presenting a lower rate of hyperammonaemic episodes and an easier control of chronic hyperammonaemia but displays higher rates of neurological sequelae (7, 15, 17, 59, 86, 88) and liver disease (74).
b) Argininosuccinic acid
Argininosuccinic acid is a pathognomonic biomarker of the disease in biological fluids. This explains why enzymatic studies and molecular analysis are less frequently performed in this disorder as identification of argininosuccinic acid in biological fluids is enough to confirm the diagnosis (59). Concentration of
45 Chapter 1 - Background argininosuccinic acid is twice as high as in cerebrospinal fluid than in blood (15). To detect small amount of argininosuccinate, liquid chromatography/tandem mass spectrometry (LC-MS/MS) has been reported more sensitive compared to ion exchange chromatography (IEC) (103).
c) Plasma amino acids profile shows raised citrulline, abnormal levels of argininosuccinic acid and decreased L-arginine levels (See Figure 5 and Section 1.2.2) (62). Raised glutamine and glutamate are biomarkers of hyperammonaemia (See Figure 6 and Section 1.2.2) (104).
iv) Diagnostic confirmation can be performed via argininosuccinate lyase enzymatic analysis in fibroblasts, liver or red cells or via hASL sequencing. Various methods have been reported: the study of the direct forward reaction i.e. the conversion of argininosuccinic acid into fumarate can be detected by fluorometric method or a combined assay with arginase and detection of urea by the Jung fluorometric method (85). The study of the direct reverse reaction studies the transformation of labelled 14C-Argininosuccinic acid formation after incubation of L- arginine with labelled 14C-fumarate (105) or 13C-L-Glutamine (106).
v) Prenatal diagnosis is possible by chorionic villus biopsy, amniocentesis or fetal blood sampling for argininosuccinate lyase enzymatic analysis or DNA sequencing (107).
1.2.5 Therapeutics
a) Acute hyperammonaemia
Acute management of a metabolic decompensation requires intravenous ammonia scavenger drugs (sodium benzoate and/or phenylbutyrate), L-arginine supplementation, stopping all protein intake and maintenance of anabolism with hypercaloric intake based on carbohydrate and lipids exclusively. An early and
46 Chapter 1 - Background careful management of electrolyte disturbances and prevention of intracranial hypertension is crucial. If ammonaemia is >500 μM, an extrarenal removal of the ammonia in excess needs to be actively pursued with continuous veno-venous hemodiafiltration or peritoneal dialysis (12). This requires central venous access and admission in intensive care units.
b) Conventional treatment
The best-accepted treatment aims to control ammonia levels and correct L-arginine deficiency. This is a lifelong treatment. i) A specific diet is the basis of the day-to-day management. This combines i) a protein restriction to decrease the production of ammonia from the catabolism of proteins and ii) a sufficient energy intake to prevent catabolism and proteolysis (108). National guidelines and practice vary slightly between European countries (Figure 8). This diet requires specific dietary supplements enriched in energy, vitamins, and trace elements based on standard age-related requirements in parallel with close monitoring of the nutritional status to avoid deficiencies. Excessive protein restriction may cause severe amino acid imbalance precipitating catabolism and hyperammonaemia. When protein intake is too low and compromise growth and metabolic control, a supplementation with essential amino acids is recommended (12). This mixture contains less amino acids involved as neurotransmitter precursors (tryptophan, phenylalanine, tyrosine) and a larger amount of branched chain amino acids (108). Indeed hyperammonaemia facilitates the crossing of the blood brain barrier to some large neutral amino acids, especially amino acid precursors, associated with increased cerebral levels of serotonin levels (See Section 1.2.2). High doses of sodium phenylbutyrate, one of the ammonia scavengers, is associated with a decrease of branched chain amino acids, which justifies the increased content of these branched chain amino acids in essential amino acids mixture for patients with urea cycle defects (109). This diet presents various difficulties. This allows survival, but does not prevent later decompensations and a poor quality of life (110). Compliance is sometimes difficult especially during the adolescence and in recently diagnosed patients with late-onset phenotype who have not been used to this specific diet in their early years. This diet is expensive as illustrated by the cost of the
47 Chapter 1 - Background lifetime management of patients with organic acidaemias, another inherited metabolic disease requiring a similar diet and recurrent hospitalisations for acute decompensations, reported to be $1,500,000 (111).
Figure 8. Median protein intake in argininosuccinic aciduria by age and country of origin in Western Europe
Data compiled from (108, 112-116)
ii) Ammonia scavenger drugs, such as sodium benzoate, sodium phenylbutyrate (117) or glycerol phenylbutyrate, can be necessary to control ammonia levels on a day-to-day basis. They are prescribed orally up to 4 times a day or intravenously for acute management. Carglumic acid is an allosteric activator of carbamoyl-phosphate synthase, the rate-limiting enzyme of the urea cycle, which can be prescribed in emergency (12). iii) L-arginine supplementation divided in several doses daily aims to compensate hypoargininemia. High doses (500 mg/kg/d versus 100 mg/kg/d) increase the levels of transaminases (75). The recommendation is to find the minimal effective dose to normalise L-arginine levels (118).
48 Chapter 1 - Background
When the patient is at risk of decompensation, for example experiencing an intercurrent illness, performing an unusual physical exercise, having a fasting period or a surgery, the patient needs to follow specific protocols to adapt his/her treatment accordingly. The protein content of the diet is reduced or stopped and/or the ammonia lowering medications increased to prevent an acute hyperammonaemic decompensation.
c) Liver transplantation
Liver transplantation is a potentially curative therapeutic strategy for urea cycle defects (119). Liver transplantation works as an enzyme replacement therapy and allows the restoration of a functional urea cycle with normalisation of ammonaemia. As a result, the diet can be normalised and ammonia scavengers can be stopped without risking acute decompensations. Transplants can improve the neurological development and better learning and memory capacities are achieved with a better metabolic control of ammonia; however neurologic sequelae persist (120) even when a transplantation has been performed as early as at 18 months of age (76). A liver transplant does not correct high plasma argininosuccinic acid and citrulline levels and L-arginine remains an essential amino acid as the main endogeneous arginine supply, synthetised by the kidney (See Figure 4 & Section 1.2.2), remains affected as the argininosuccinate lyase deficiency persists in extrahepatic tissues (76). Transplants can prevent long-term ammonia-related neurological complications but may not solve the long-term effects of the disease in extrahepatic tissues.
Therefore, indications for liver transplantation in argininosuccinic aciduria are restricted to: i) Severe phenotypes, ideally at early age between 3 and 12 months of life, before irreversible neurological damages caused by hyperammonaemia occur (12, 40). Liver transplantation can cure the urea cycle-associated symptoms but is another disorder in itself with its own morbidity, mortality, need for lifelong immunosuppression. Moreover there is a lack of donors. Therefore <8% of patients affected by urea cycle defects undergo liver transplantation with 88% and 78% of patient and graft survival respectively in < 2 year-old patients (121). Liver
49 Chapter 1 - Background transplantation in children <1 year-old is known to be associated with an increased risk of morbidity and mortality (122, 123). ii) Severe liver cirrhosis (18).
d) Cell therapy
Hepatocyte transplantation allows a transient metabolic control of the disease and can be used as a “bridge to liver transplantation” (124). This “bridge to transplant” approach is preferred for patients with severe phenotype and recurrent hyperammonaemia despite optimised medical treatment (125). This 2-steps strategy has been successfully tested in argininosuccinic aciduria with improvement of psychomotor development and increase of argininosuccinate lyase activity in the liver from undetectable level to 3% at 8 months after 3 series of infusions (105). Fresh or cryopreserved hepatocytes can be used and infused in the portal vein. The poor engraftment rate and viability of hepatocytes, the need of immunosuppression and the risk of metabolic decompensation associated with the procedure (126) are complications to balance when assessing the risk-benefit of this therapeutic option.
e) Nitric oxide supplementation
In case of impaired citrulline-nitric oxide cycle, the nitrate-nitrite-nitric oxide pathway is a rescue path for delivering nitric oxide to the argininosuccinate lyase-deficient cell (49). As argininosuccinic aciduria is associated with systemic nitric oxide deficiency, nitric oxide supplementation has been proposed. Organic nitrate later changed for sodium nitrite supplementation (127) was successfully reported in one patient suffering from refractory hypertension. This allowed a normalisation of the blood pressure and the progressive stop of all antihypertensive drugs. Some neurocognitive parameters were mildly improved (85). To validate the long-term benefit of nitric oxide supplementation, a randomised clinical trial versus placebo is currently recruiting (Clinical trials numbers NCT02252770 and03064048 from clinicaltrials.gov accessed on 06/04/2017). A dietary supplementation in nitrite- enriched food such as beetroot juice has been proposed as well (96).
50 Chapter 1 - Background
1.2.6 Long-term outcome
Any acute hyperammonaemic episode requires prompt management to avoid or reduce life-threatening concentrations of ammonia. Any delay in diagnosis and/or appropriate management increase the risk of neurologic sequelae or death. After the initial decompensation that leads to the diagnosis, these patients suffer recurrent hyperammonaemic crisis causing poor long-term outcomes with high rates of mortality and neurodisability.
Survival in historical cohorts was 75-100% in early-onset phenotype (90, 101, 128) and 50-100% in late-onset phenotype (6, 17, 128).
Neurological outcome is poor with developmental delay in 30-100% (3, 7, 59, 97, 129-131), abnormal fine motor movements (58%), muscular hypotonia (22%), ataxia (27%), attention deficit hyperactivity disorder (ADHD) (30%), epilepsy (19-55%) and abnormal electroencephalogram (54%) (7, 63, 86). Also a neurodevelopmental delay is usually diagnosed during the first 2 years, it can be sometimes challenging to prove it even in late childhood. Mild neurological symptoms can be difficult to recognise in young children until the first years of schooling. Therefore it is important to monitor carefully these patients during follow-up as some neurological or systemic symptoms can appear over time independently to the severity of hyperammonaemia. That could explain the apparent heterogeneous rate of neurological complications. For instance, neonatally screened and prospectively treated patients with argininosuccinic aciduria, who have never experienced acute hyperammonaemic decompensation develop in the long-term learning disabilities (35-38%), epilepsy (0- 31%), and EEG abnormalities (45-48%) (15, 17).
Overall it is now recognised that “novel therapeutic strategies are needed for severely affected patients with argininosuccinic aciduria to improve their currently bleak neurological prognosis” (21).
51 Chapter 1 - Background
1.2.7 Phenotype-genotype correlation
Compared with other urea cycle defects, few mutations have been reported so far likely because argininosuccinic acid is a pathognomonic marker of the disease and molecular analysis is not necessary for the diagnosis (35). 134 disease-causing mutations have been reported so far with missense mutations (n=92), nonsense mutations (n=14), splice site mutations (n=13), small deletions (n=8), small duplications (n=4), indel mutations (n=2) and one large mutation. These mutations have been described in all exons except exon 1. No hotspot mutations are observed but the exons 3, 4, 6, 7 and 9 are more frequently affected and represent 41% of the mutations for only 25% of the coding DNA sequence. Exons 4, 6 and 12 involved closely in the active site of the enzymatic protein may have more severe phenotypes (21). Three founder mutations have been reported: the Finnish mutation c.299T>C (61) and Saudi Arabian mutations c.1060C>T (18) and c.346C>T (132). The other mutations have been described with panethnic occurrence.
Some phenotype-genotype correlations are supported by in silico analysis of the 3- dimensional structure of argininosuccinate lyase. For example, the arginine in position 186 plays a key-role in the tetramer formation and any change is predicted to be deleterious. Reported p.Arg186Trp and p.Arg186Gln have both been associated with severe phenotypes. Same observation has been published for glutamine in position 286 affecting a loop in domain 3 involved in substrate entry and exit (21).
There is a large spectrum of clinical phenotypes in argininosuccinic aciduria from early-onset disease with severe hyperammonaemia to asymptomatic urinary excretion of argininosuccinic acid (15, 17). The molecular basis of the disease is not well understood and several hypothesis have been proposed: tissue-specific hASL expression (124, 133-135), intragenic complementation (136, 137), hASL expression regulated via methylation through development (138), alternative splicing events with partial exon deletions (139, 140), high frequency of hASL transcript variants underlying an increased formation of stable mutant homo or heterotetramer with dominant negative effect on argininosuccinate lyase activity (24).
52 Chapter 1 - Background
1.2.8 Newborn screening
Newborn screening for argininosuccinic aciduria has been reported either by measuring plasma citrulline in dried blood spots (18, 141, 142) or blood or urinary argininosuccinic acid (15, 143, 144).
Patients diagnosed by newborn screening and treated prospectively from birth without experiencing acute hyperammonaemia seem to develop an attenuated neurological disease with learning disability, mild developmental delay, seizures and/or abnormal electroencephalogram (15, 17): i) Margalith et al were the first to describe a patient neonatally screened who developed with a borderline IQ at 7 year-old and mild ataxia (145). ii) Ficioglu et al described 13 patients diagnosed after a screening performed from 4 to 6 weeks of age in > 630,000 infants born in Massachussetts between 1969 and 1978. These patients received a conventional treatment initiated within the first 6 months of life. They never developed acute hyperammonaemia; 9 of them developed mild neurological abnormalities including learning difficulties (5/13; 38%), seizures (4/13, 31%) or abnormal electroencephalogram (6/13, 48%) and 4 were asymptomatic after 13 to 33 years of follow-up. In comparison, patients with late- onset phenotype, who did not benefit from an early screening, were diagnosed between 1.5 and 15 years old, had all severe delay with an IQ of 40-60, epilepsy and ataxia (15). The assumption of a better control of ammonia levels and/or early correction of hypoargininemia and downstream metabolites was supposed to explain this clinical benefit. iii) A neonatal screening programme performed in Austria tested >1,400,000 neonates between 1973 and 1990. 12 patients had a diagnosis of argininosuccinic aciduria confirmed by detection of ASL activity in red blood cells. A conventional treatment was initiated within the first 6 months of life, which prevented any acute decompensation later in life. No developmental delay (IQ ranging from 92 to 132) and attendance of mainstream schooling were observed for all patients. However 5 patients (40%) had abnormal electroencephalogram pattern, some developed raised transaminases and trichorrhexis nodosa that corrected when L-arginine doses were
53 Chapter 1 - Background increased (95). This dramatic improvement of the long-term neurological outcome was balanced by a publication 18 years later, in which this cohort was expanded and showed 35% mental retardation (6/17; 35%) and abnormal electroencephalogram pattern (4/9 ; 45%) but no epilepsy (17). iv) In an European cohort of 31 patients affected by argininosuccinic aciduria with either a mild phenotype or asymptomatic, 13 (46%) and 3 (10%) were diagnosed by newborn screening or familial history respectively. These patients were treated early and presented a low frequency of acute hyperammonaemic crisis (23%), the lowest compared to all other urea cycle disorders. However the mental retardation in this cohort was by far higher (65%) compared to other proximal urea cycle defects like ornithine transcarbamylase deficiency (31%) and argininosuccinate synthase deficiency (16%) (97). A large proportion of patients with argininosuccinic aciduria diagnosed by newborn screening remained asymptomatic but some developed symptomatic hyperammonaemia despite some carrying a genotype predicting a « mild » course. Overall clear conclusions from this study were difficult to draw as all patients were presented in a same cohort whatever their symptoms of diagnosis and the age at the start of treatment, although half of the patients having been treated prospectively from birth and the other half not.
The neurological outcome of the patients diagnosed by newborn screening is significantly better compared with outcome from patients presenting symptomatically with a diagnostic delay. However newborn screening might diagnose asymptomatic or mildly symptomatic patients, who will not have been diagnosed otherwise. Consequently it is difficult to conclude that an early treatment modifies the neurological phenotype.
54 Chapter 1 - Background
1.2.9 Animal models
A severe argininosuccinate lyase-deficient mouse model was first described by Reid- Sutton et al (146). The phenotype showed early hyperammonaemia causing lethality within the first 48 hours of life. Liver ASL residual activity in mutants was 2 to 5% of wild-type ASL activity. No abnormal levels of nitrite or creatine metabolites were observed. As the mice died very early after birth, this might explain why some phenotypic features might not have been observed.
A hypomorphic mouse model of argininosuccinate lyase deficiency, the AslNeo/Neo mouse, was described recently with an insertion of the neomycin cassette in the 9th intron of the murine Asl gene. More than 75% of the mutant mice died within the first month of life displaying hyperammonaemia, a multi-organ disease, abnormal plasma levels of urea cycle-related amino acids and systemic nitric oxide deficiency. The residual argininosuccinate lyase activity in the liver was 16% (32).
55 Chapter 1 - Background
1.3 Gene therapy for monogenic disorders
The liver is a major metabolic organ responsible for manufacturing new proteins such as clotting factors, detoxification of by-products, for example, those from amino acid metabolism, energy production and many other functions. Impairment of these metabolic pathways lead to various inherited genetic/metabolic diseases (147). Although several disease-specific conventional therapies have been described, liver replacement therapies such as whole or partial liver transplantation (148) or hepatocyte infusions (149), represent a potential therapeutic approach for severe monogenic liver disorders in restoring the defective pathway (150). However the shortage of liver donors, the mortality/morbidity associated with these procedures and lifelong required immunosuppression challenge a broad application of this strategy and limit it to severely affected patients (12). In the past decades, liver- directed gene therapy has progressively achieved promising results and is now rising as a valid alternative to liver replacement therapies for monogenic liver disorders.
As argininosuccinic aciduria remains a life-threatening disease with poor long-term neurological outcome, new therapeutic strategies have to be explored. Liver-directed gene therapy has successfully reached early phases clinical trials and is a realistic approach for liver inherited metabolic diseases with unmet needs.
1.3.1 Overview of gene therapy development
Gene therapy consists in replacing a defective gene by its correct copy in the affected cell. The European Medicines Agency (EMA) defines a gene therapy medicinal product (GTMP) as “a biological medicinal product which fulfils the following two characteristics: (a) it contains an active substance which contains or consists of a recombinant nucleic acid used in or administered to human beings with a view to regulating, repairing, replacing, adding or deleting a genetic sequence; (b) its therapeutic, prophylactic or diagnostic effect relates directly to the recombinant nucleic acid sequence it contains, or to the product of genetic expression of this sequence.”
56 Chapter 1 - Background
(http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2015/ 06/WC500187744.pdf). Regulators differentiate germline gene therapy, in which the therapeutic gene will be passed to the next generation, and somatic gene therapy. Only somatic gene therapy is authorised by law.
Gene therapy has been investigated for more than 40 years to achieve the bedside as a novel therapeutic option for monogenic disorders (151). The principle initially considered was the supply of a non-mutated copy of the mutated gene, which could synthesise a deficient protein and subsequently restore a physiolocal pathway and correct the phenotype. With the development of gene-editing technology, the delivery of genome editing tools or non-coding RNA able to modulate gene expression widely open the possibilities of interfering with pathophysiological processes (152).
The Gartner Hype cycle is a tool developed to assess the maturity of novel technologies and this modelisation can be applied to gene therapy (Figure 9). Gene therapy reached its “peak of inflated expectation” in the mid-1990s driven by a rapid increase in clinical trials and the demonstration of proof-of-concept studies for genetic and acquired conditions such as adenosine-deaminase deficiency (ADA- SCID) (153, 154) and brain tumors, respectively (155). This excessive appeal was criticised in a report commissioned by the National Institute of Health, the Orkin- Motulsky report (156). This report emphasized the need for a better understanding of the mechanisms observed in gene transfer to improve the safety profile of this emerging technology. Soon after the death of a young adult in a clinical trial for ornithine transcarbamylase deficiency (OTCD), the field fell into its “trough of disillusionment” (157). The clinical success reported in the treatment of X-linked severe combined immunodeficiency (SCID-X1) (158) was soon overcome by the occurrence of leukaemia in 5 patients (out of 20) secondary to insertional mutagenesis causing the death of one patient (159-161). These concerns about insertional mutagenesis were further complicated by growing evidence of the challenges imposed by vector-induced immune responses (162). This led to a decline in financial investment (163). In parallel, scientific research was digging the ground to better understand the underlying causes of these adverse events, which
57 Chapter 1 - Background subsequently laid the foundations for the recent clinical successes in various inherited orphan diseases such as Leber’s congenital amaurosis (164-166), X-linked adrenoleukodystrophy (167), metachromatic leukodystrophy (168) and haemophilia B (169) and the first market authorisation granted by the European Medicines Agency (EMA) in 2012, to Glybera® for lipoprotein lipase deficiency (170). This shed a new light onto the field, which is now following a “slope of enlightenment” and progressively reaching a “plateau of productivity”. Various biotechnology companies have been created dedicating their efforts to gene therapy development and receiving substantial financial investment from venture capitalist funds (171). In accordance, the number of gene therapy clinical trials has risen rapidly with more than 2,400 clinical trials completed, ongoing or approved in 2017 in 37 different countries (172, 173) .
58 Chapter 1 - Background
Orkin report 1995
Inser;onal mutagenesis 1st liver-directed Secondary leukemia Success in trial in HFH in SCID-X1 Metachroma3c leukodystrophy 1994 Success in 1st success in Haemophilia B 2013 SCID-X1 Success in 2011 Leber’s amaurosis 2016 2012 1st trial in 2007 EMA approves ADA-SCID 2009 Strimvelis® 2003 (GSK) 1990 1999 Success in X-linked 2000 adrenoleukodystrophy 2004 Immune response EMA approves precludes long-term efficacy Glybera® OTC deficiency trial in Haemophilia B (Uniqure) Death of Jesse Gelsinger
Technology Peak of inflated Trough of Plateau of trigger expecta0ons disillusionment Slope of enlightenment produc0vity
Figure 9. The Gartner Hype cycle of gene therapy
ADA-SCID adenosine deaminase-severe combined immunodeficiency ; GSK: Glaxo-Smith-Kline; HFH Homozygous familial hypercholesterolaemia ; OTC ornithine transcarbamylase deficiency ; SCID-X1 severe combined immunodeficiency X1.
59 Chapter 1 - Background
1.3.2 Strategies for gene transfer
Various possibilities can support a successful gene transfer to the targeted organ or cell (Figure 10).
Vector Delivery • Transgene casse*e • Promoter • Dose • Regulatory elements • Route
Host immunity
• against the capsid • against the transgene product
Figure 10. Factors influencing gene transfer
Vector construct
The vector contains 2 elements: the transgene cassette and a facultative envelope, which influences the efficacy of transduction.
The transgene cassette contains i) a transgene, commonly a coding DNA sequence, which can be codon-optimised to achieve higher expression of the transgene product, ii) a promoter, which critically determines the cell-type restricted specificity of expression, the level of expression of the transgene, and is involved in insertional mutagenesis, iii) various pre- and/or post-regulatory elements to stabilise transgene messenger RNA and therefore increase the yield of transgene product e.g. addition 60
Chapter 1 - Background of an intron downstream the promoter containing a bacterial replication origin (174) or the woodchuck hepatitis post-transcriptional regulatory element (WPRE) (175), respectively.
An envelope is not necessary to deliver the transgene cassette to the target cell/organ (176). Therefore, various approaches have been developed with either non-enveloped DNA, non-viral or viral vectors. i) Naked DNA, also called plasmids (177, 178) or mini-circles (179-181), is a simple mode of transgene delivery. Mini-circles are devoid of plasmid backbone DNA, which enhance transgene expression by overcoming heterochromatin formation and avoiding inflammation triggered by bacterial DNA (182). The main advantages are a relatively non-immunogenic method and no risk of viral replication compared to some viral vector approaches (183), an easy production with a good safety profile and the ability to elicit long-lasting transgene expression in post-mitotic tissues (183, 184). However this approach hardly allows transgene expression to a level, which can achieve phenotypic correction in monogenic disorders, thus limiting currently translatable options. ii) Non-viral vectors use the chemical properties of engineered nanoparticles such as liposomes or polymers to form bioparticles encapsidating the transgene (185). An easy production, no restriction of the transgene size, and a reliable safety profile have made this approach appealing but the stability, cellular uptake and limited ability of the particles remain barriers preventing long-lasting transgene expression (186). iii) Viral vectors use the ability of viruses to easily transduce human cells. This is currently the most popular approach to deliver a therapeutic transgene. Several viral vectors have been developed and are now in clinical trials: adenoviral, retroviral, adeno-associated viral (AAV) and lentiviral vectors are among the most popular for clinical translation (Table 3).
61
Chapter 1 - Background
Non viral Retroviral / Adenoviral Adeno-associated vector lentiviral vector vector viral vector
Double-stranded Double-stranded Single-stranded Wild type virus None RNA (9.2 kb) DNA (36 kb) DNA (4.7 kb)
Derived from No Yes Yes No pathogenic virus Size of transgene No limit 14 kb 7.5 kb 4.9 kb Insertion to host No Yes No Rarely genome Achievable titre High Low High High Long-lasting No Yes No Yes transgene expression Target dividing and Yes Yes Yes Yes non-dividing cells Insertional Inflammatory ? Insertional Safety issues No mutagenesis response mutagenesis
Table 3. Features of different strategies for gene delivery
Adapted from Simonato et al (187)
Vector delivery
Parameters involved in the delivery are crucial for efficacy. i) Finding the minimal effective dose is essential. If low dose might prevent efficacy (188), high doses of vector generate more severe immune responses and might therefore have a limited benefit (157, 189). ii) The route of delivery is a fundamental consideration, which depends on the organ or cell type to transduce. A local injection will reach more easily the target, will limit the loss of vector caused by biodistribution and the initiation of immune response. For liver-directed gene therapy, peripheral intravenous injection provides a similar transduction compared to intraportal or intrahepatic routes for AAV vectors (190, 191). iii) Compared to in vivo delivery, ex vivo gene therapy relies on host cells collected and transduced in vitro before reinjection to the patient. This approach has
62
Chapter 1 - Background been very successful with integrating vectors i.e. retroviral and lentiviral vectors for X- linked severe combined immunodeficiency (SCID-X1) (158), and X-linked adrenoleukodystrophy (167) and metachromatic leukodystrophy (168), respectively. For liver monogenic disorders, this has been only reported once in a trial for homozygous familial hypercholesterolaemia with very limited efficacy (192).
Host immune responses
Innate or adaptive immune responses can be directed either against the viral capsid or the transgene product. Immune responses can generate severe symptoms from flu-like symptoms to multi-organ failure and death. In parallel to these safety concerns, immune reactions might preclude the therapeutic effect of gene therapy in neutralising the vector or in reacting against the transgene product before and after transduction, respectively (193-195).
The innate immune response is non-specific and triggered by antigen presenting cells such as dendritic cells or macrophages initiating the release of proinflammatory cytokines (interleukins 1 and 6, tumor necrosis factor α (TNFα), type I interferon α and β) via stimulation of Toll like receptors (TLR). Adenoviruses are known to be very immunogenic and trigger a severe innate immune response, which has been deemed responsible for severe adverse events observed in clinical trials for ornithine transcarbamylase deficiency (157) or haemophilia A (196).
Specific and long-lasting immune response generates B- and T-cell responses with secretion of neutralising antibodies by plasmocytes and CD8+ cytotoxic T lymphocytes, respectively. Whatever the viral vector considered, immune responses share various similarities (197-201). For example, it was suggested that T-cell mediated cytotoxicity could be the cause of the raise of transaminases observed 8 to 12 weeks after liver-directed gene therapy mediated by AAV vectors for haemophilia B (202-204). The recognition of antigenic epitopes from the AAV capsid presented at the surface of the transduced cell would trigger the immune response.
Any preexposure to a wild-type viral serotype, which has a similar antigenic motif to the capsid of a viral vector, might be responsible for a cross-reactivity of the immune 63
Chapter 1 - Background system of the host. This immune memory of pre-exposure to wild-type viruses can prevent efficient cell transduction by pre-existing neutralising antibodies and might potentiate an adaptive immune response caused by memory T cells. To prevent this, clinical trials using AAV vectors for instance are recruiting patients who don’t have neutralising antibodies against the vectorised capsid. This narrows the population, which might benefit from the gene therapy product.
After vector injection, patients will generate high levels of antibodies against the injected capsid (188, 204).
1.3.3 Liver-directed gene therapy: clinical applications
Various clinical trials using viral vector-mediated gene therapy have been conducted for liver monogenic disorders.
During the 1990s, retroviral vectors with ex vivo gene therapy have been used in homozygous familial hypercholesterolemia. The results showed a mild improvement of lipid profiles in 2 patients with a very low rate of stable engraftment at 4 months after gene therapy injection (192, 205). But the risks were by far outweighing the benefit of this approach with the need of two invasive procedures: i) the surgical resection of a liver lobe to harvest enough primary hepatocytes for transduction as hepatocytes cannot be expanded in culture and ii) the reinjection of transduced hepatocytes into the portal circulation via a local catheter, with the associated risks of venous thrombosis, catheter misplacement and haemorrhage. Moreover the efficiency with which hepatocytes were harvested and transduced was low, 30% and 10% respectively (205). Later retroviral vectors were used in vivo in haemophilia A patients: a single intravenously injection was well tolerated but did not show significant clinical benefits (206). Currently a lentiviral vector, which is a retroviral vector with a safer profile with a reduced risk of insertional mutagenesis, is currently in preclinical development for haemophilia B (207).
64
Chapter 1 - Background
Adenoviral vectors have been considered as the leading viral vector for liver gene therapy in the 1990s. For instance, a second-generation adenoviral vector of serotype 5, from which proviral DNA sequences E1- and E4- had been deleted, was tested in ornithine transcarbamylase deficiency. A young adult with late-onset ornithine transcarbamylase deficiency, Jesse Gelsinger, died after developing a fatal acute toxic reaction with fulminant inflammatory response and multi-organ failure hours after the injection of the vector (157). Innate immune response was hypothesised as the cause of the fatal outcome (157) although the reason for the severity of this immune response remained unclear as another patient injected with the same dose only exhibited mild flu-like symptoms. A genetic predisposition or an immune memory response caused by pre-exposure to adenoviruses might have potentially been responsible for this discrepancy (208). To limit immune responses, “gutless” or helper-dependent adenoviral vectors (HD-Ad) were developed by deletion of all coding regions of the proviral DNA except the inverted terminal repeat sequences (ITR) in 5’ and 3’ untranslated regions (UTR) and the packaging signal (Ψ) required for gene expression and encapsidation of the adenoviral genome, respectively (209). A trial using this helper-dependent vector was developed for haemophilia A. The trial was prematurily stopped as the first patient developed flu- like symptoms and increased levels of transaminases within a week after injection. The patients showed no increase of plasma factor IX (196, 210, 211).
Since 2004, adeno-associated viral vectors (AAV) vectors have emerged as the leading candidates for gene therapy in monogenic liver disorders with the best accepted benefit-risk ratio (207). In 2004, an AAV2 vector was tested in haemophilia B patients and administered via the hepatic artery. This demonstrated a transient increase of plasma FIX from <1% to 3-11% during 4 weeks followed by a gradual decline over 4 to 8 weeks concomitant with transient asymptomatic transaminitis (202), later recognised as T cell-mediated cytotoxicity (203). In 2009, Nathwani et al., injected an AAV8 vector via a peripheral vein and reported a long-lasting (up to 6 years) increase of plasma FIX from <1% to 1-7% (212). Transaminitis occurring 7 to 10 weeks post-injection resolved after an oral course of corticosteroids, but was associated with a decrease of 50-70% in plasma FIX levels attributable to a cellular 65
Chapter 1 - Background immune response against capsid epitopes (204). More recently, D’Avola et al. reported results of a trial with an AAV5 vector injected intravenously in acute intermittent porphyria. No vector-related safety issue was reported but no change in the levels of metabolic biomarkers was observed (188). This might be explained by a too low vector dose in order to modify the phenotype and the fact that AAV5 has a less efficient liver transduction profile compared to AAV8 (213). Several clinical trials are ongoing for haemophilia B confirming Nathwani’s promising results. After a single intravenous injection of AAV vectors with different capsids encoding the FIX gene, plasma FIX levels have ranged from 3-8% in the AMT-060 trial sponsored by “Uniqure” using AAV5 (214) and the DTX-101 trial sponsored by “Dimension Therapeutics” using AAVrh10 (http://www.dimensiontx.com) to 20-44% in the high- dose cohort of the BAX 335 trial sponsored by Shire using AAV8 (215) and in the SPK-9001 trial sponsored by Spark Therapeutics/Pfizer using an engineered AAV (216). In a haemophilia A gene therapy trial, BioMarin reported plasma factor VIII (FVIII) from 12-306% in the BMN 270 trial with AAV5 (217). Importantly, endogenous FVIII is primarily secreted by endothelial cells (218). All the AAV-based trials have so far involved only adult patients, who had no neutralising antibodies against the capsid. Many monogenic liver disorders in AAV-based gene therapy development pipelines of pharmaceutical companies include ornithine transcarbamylase deficiency, glycogen storage disease type Ia, citrullinemia type I, phenylketonuria, Wilson disease, methylmalonic acideamia and Crigler-Najjar syndrome (219).
Overall liver-directed gene therapy has achieved major milestones in the last 2 decades with the development of safe and efficient AAV-based viral vectors, which has led to recent clinical successes in haemophilia A and B (Figure 11).
66
Chapter 1 - Background
A" 10 Retrovirus Adenovirus 8 AAV 6
4 Trials / year Trials 2
0
1992 1998 2003 2004 2009 2012 2014 2015 2016 2017
B" 10 Haemophilia A and B HFH 8 OTCD 6 AIP CN 4 Trials / year Trials 2
0
1992 1998 2003 2004 2009 2012 2014 2015 2016 2017
Figure 11. Clinical trials of liver-directed gene therapy for monogenic liver disorders
Presentation according to (A) viral vector system or (B) liver disease The dotted line represents trials expected to start in 2017 and does not pretend to be exhaustive. AAV: Adeno-associated virus; AIP: Acute intermittent porphyria; CN: Crigler Najjar; HFH: Homozygous familial hypercholesterolaemia; OTCD: Ornithine transcarbamylase deficiency. If the date of the start of the trial was not available in ss.gov website (accessed 06/04/2017), the date of publication of the results is mentioned.
67
Chapter 1 - Background
1.3.4 Adeno-associated virus
AAV vectors have emerged as the leading technology for liver-directed gene therapy as illustrated by the rapidly increasing number of clinical trials and the publications demonstrating proof-of-concept in animal models for various liver monogenic disorders including urea cycle defects, organic acidurias, phenylketonuria, glycogen storage disease type Ia, long chain fatty acid oxidation disorders, homozygous familial hypercholesterolemia, primary hyperoxaluria type I, and progressive familial intrahepatic cholestasis (220, 221).
Adeno-associated viruses naturally have a very safe profile due to being non- pathogenic in humans and dependent on a helper virus for replication.
AAVs are non-enveloped, single-stranded DNA viruses that belong to the Dependovirus genus and the Parvoviridae family. Initially discovered in 1965 as contaminant of cultures of adenoviruses (222), the virus was later shown to require co-infection with a helper virus to replicate. In the absence of helper virus AAV can enter target cells and establish latent infection through genomic integration and/or formation of episomes. AAV is considered non-pathogenic and has yet to be definitively linked to disease causation. AAV virions consist of an icosahedral capsid of approximately 22 nm in diameter enclosing a 4.7 kilobases single-stranded genome. The genome is flanked by two 145 nucleotides-long palindromic sequences of inverted terminal repeats (ITRs) containing all of the necessary cis-acting functions for proviral rescue, genome replication and packaging. These ITRs present a secondary structure similar to a hairpin by base pairing self-complementarity of this 145 base pairs sequence and play a key-role in the initiation of replication via replication binding elements (RBE) binding termination resolution site.
The viral genome contains two main open reading frames (ORF) encoding functional proteins (Figure 12). The rep gene encodes 4 Rep proteins named according to their molecular weight: Rep40, Rep52, Rep68 and Rep78. These four proteins are 68
Chapter 1 - Background synthesised from messenger RNA initiated from the p5 or p19 promoters, each of which is either spliced or left intact. The two largest proteins Rep68 and Rep78 possess single-strand endonuclease, helicase and ATPase activities required for DNA replication. They are site-specific and bind with replication binding elements (223). The two smallest proteins Rep40 and Rep52 retain only the helicase activity and are required for packaging DNA into capsids.
The cap gene encodes for three capsid proteins VP1, VP2 and VP3 synthetised by alternative splicing initiated from the p40 promoter and assemble to form the capsid (Figure 12) (224). The two minor proteins VP1 and VP2 contain additional N-terminal sequences with nuclear localisation signals and phospholipase A2 activity required for infection.
The capsid assembly might require expression of an assembly-activating protein (AAP) encoded by an alternative third open reading frame of the Cap gene and providing scaffolding activity (225) in some but not all AAV serotypes (226). Once assembled, the AAV capsid is icosahedral, non-enveloped, and reaches 22 nm of diameter with a VP1:VP2:VP3 ratio of approximately 1:1:10 (224).
Multiple AAV serotypes have been isolated from humans, non-human primates and other species (Figure 13). A diversity of cell surface receptors/co-receptors (Table 4) (227) will bind the viral capsid, which determines species and target cell tropism (Table 5) (228). For example, AAV3B uses the human hepatocyte growth factor (hHGF) receptor, which restricts transduction to primates and especially to the liver (229). If intracellular trafficking pathways remain incompletely understood, a multiserotype AAV receptor has been recently identified (230), but its precise role in uptake or trafficking has not yet been elucidated (231).
69
Chapter 1 - Background
p5 p19 p40 Poly(A)
rep cap
ITR ITR
Rep78
Rep68
Rep52
Rep40
VP1
VP2, VP3 AAP
Figure 12. Genomic structure of AAV
The 4.7 kb genome contains 2 open reading frames flanked by inverted terminal repeats (ITR). The rep gene encodes for 4 Rep proteins Rep78, Rep68, Rep52, Rep40 synthesised from mRNA transcripts initiated from p5 and p19 promoters, each of which is either left intact or spliced. The cap gene encodes 3 capsid proteins VP1, VP2, VP3 synthesised from a transcript initiated from p40 promoter. The same messenger RNA codes for VP2, VP3 and an assembly-activating protein (AAP). Compared to VP3, VP2 has N-terminal residues (white box) required for infectivity of the virion.
70
Chapter 1 - Background
VP1 homology Clades to AAV2 AAV2 100% B
AAV3 87% C AAV1 83% 99% A AAV6 83%
AAVrh10 84% 93% E AAV8 83%
AAV7 82% D
AAV9 82% F
AAV11 63% 84% AAV12 60% AAV4 61%
AAV5 58%
AAV isolated from Human cell lines Non-human primates cell lines
Figure 13. Phylogenetic tree of VP1 protein sequence of primate AAV serotypes
The phylogenetic relationships mentioned are based on the amino acid sequence of VP1 capsid protein compared with AAV2 serotype and presented in percentage based on data from Vance et al (113) and Gao et al (232).
71
Chapter 1 - Background
Primary receptor Co-receptor Seroprevalence AAV1 α2,3/α2,6 N-linked sialic acid ? 27-59% Fibroblast/Hepatocyte growth factor AAV2 Heparan sulfate receptor, Laminin receptor, Integrin 22-100% αVβ5/α5β1 AAV3 Heparan sulfate Hepatocyte growth factor receptor AAV4 2,3 O-linked sialic acid ? AAV5 2,3 N-linked sialic acid Platelet-derived growth factor receptor 3-50% α2,3/α2,6 N-linked sialic AAV6 Epidermal growth factor receptor 30-91% acid/heparan sulfate AAV7 ? ? AAV8 ? Laminin receptor 14-90% AAV9 Galactose Laminin receptor 33%
Table 4. Comparative study of cellular receptors/co-receptors and seroprevalence by adeno-associated virus serotype
Adapted from Asokan et al (228), Boutin et al (233), Louis Jeune et al (234), Calcedo et al (235).
Central Skeletal nervous Eye Liver Heart Kidney Lung muscle system AAV1 x x AAV2 x x x AAV3 x AAV4 x x AAV5 x x x
AAV6 x x AAV7 x x
AAV8 xxx xx x xxx x x
AAV9 xxx x xxx xx
Table 5. Preferential tropism of the main adeno-associated virus serotypes
Updated from Wu et al (236), Asokan et al (228). Organ tropism is marked as mild (X), moderate (XX) or high (XXX). 72
Chapter 1 - Background
1.3.5 Recombinant adeno-associated viral vectors
AAV vector generation
AAV2 is the most widely studied serotype and was the first among the AAV serotypes to be used as a vector. To generate a recombinant AAV vector, the native rep and cap genes are removed (making the vector gutless and replication-defective) and replaced by a transgene expression cassette retaining only the flanking ITRs (Figure 14A). Recombinant virus is produced by supplying plasmids containing rep and cap and necessary adenoviral helper functions. A major development in AAV vector technology was the demonstration that recombinant AAV2 genomes can be cross-packaged, or pseudo-serotyped, with the capsids from other AAV serotypes (237). This has dramatically broadened the cell types that can be efficiently targeted with AAV vectors taking advantage of the organ-specific improved transduction provided by AAV capsid selection (Table 5). For example, pseudo-serotyping a recombinant AAV2 vector genome with the AAV8 capsid (designated AAV2/8) enhances tropism for hepatocytes, particularly in the mouse (Table 5 & Figure 14B).
The defect of rep gene in recombinant AAV vectors makes the process of integration inefficient and is not targeting the AAVS1 site. Integration can occur randomly in the host genome creating small deletions <100 base pairs in the host genome (238). This integration process has been observed in up to 8% of hepatocytes in mouse models (239).
73
Chapter 1 - Background
A
Wild type AAV AAV1 AAV2 AAV3B AAV4 AAV5 AAV6 AAV7 AAV8 AAV9
Wild type AAV2 Capsid Proviral DNA
ITR Rep Cap ITR
Regulatory Promoter Transgene elements ITR ITR
Transgene expression cassette ITR Transgene expression cassette ITR
B
AAV2 ITRs Transgene ITR expression cassette ITR AAV8 Rep Rep Cap & Cap gene Adv helper plasmid Pseudotyped AAV8 Cap AAV2/8 vector
Figure 14. Synthesis of an AAV vector
(A) Initially, the single-stranded proviral DNA is excised to remove rep and cap genes from different wild-type AAV serotypes. The transgene expression cassette containing the promoter, the transgene and various regulatory elements is cloned between the two ITRs, which are the only wild type AAV sequences retained. (B) For vector synthesis, triple transfection of three plasmids is performed in a packaging cell with a plasmid containing the transgene cassette, a plasmid containing rep and cap and a helper plasmid. “Pseudotyped” AAV vectors contain ITRs from a specific AAV
74
Chapter 1 - Background serotype (usually AAV2) and a cap gene encoding viral proteins (VP1, 2 and 3) from a different serotype (e.g AAV8) in order to provide organ-specific transduction of the recombinant AAV vector named AAV2/8. AAV: Adeno-associated virus; Adv: Adenovirus; ITR: Inverted terminal repeat. Reproducted with permission from (240).
Cell processing of AAV vector
AAV vectors binds to the target cell via serotype-specific receptor and co-receptor that differ in a capsid-dependent manner and are taken up by endocytosis or macropinocytosis. Endosomal uptake relies on clathrin-coated vesicles or clathrin- independent carriers/GPI-enriched (CLIC/GEEC) (241). Within the first 2 hours post- infection, virions accumulate in a perinuclear location and are dependent on pH acidification to achieve a structural change in exposing the N-terminal end of the VP1 capsid protein at the surface of the capsid. This allows the rupture of the endosome and the escape of the virion into the cytoplasm where it can reach the nucleus (224, 242, 243). The trafficking to the nucleus involves the cytoskeleton via microtubules and the nuclear pore complex (244). Once in the nucleus the capsid uncoating occurs and uncoated single-stranded genomes are converted to double-stranded DNA (245) (Figure 15) (246). Conversion of single-stranded genomes to double- stranded transcriptionally active forms occurs with variable efficiency in different cell types. The viral genome persists either as a non-integrated single- or double- stranded episome or integrates into the host genome. Integration of genomic sequences of a wild-type AAV is a rare event < 0.1% (247). Some preferential regions into the human genome have been described as AAVS1 in the long arm of chromosome 19 (19q13) (248), a region with GCTC repeating motifs very similar to the invertal terminal repeat sequence of the AAV genome (249). Rep68 and Rep78 play a central role in site-specific integration binding. In that case, AAV sequences can be transmitted through cell division.
75
Chapter 1 - Background
Fenestrae endothelium of the liver sinusoids
Bloodstream A
Co-receptor Receptor
H+ Hepatocyte B Endosome
C
Nuclear pore complex
D
E G H
Capsid uncoating F
Figure 15. AAV vector uptake and in-cell processing
(A) Once reaching the target cell, the AAV vector binds an extracellular receptor and co-receptor specific to the capsid motifs. (B) After an uptake by endocytosis, the vector is trafficked in the cytoplasm in early then late endosome. (C) Acidification of the endosome modifies the capsid conformation and allows endosomal escape. (D) The AAV vector enters the nucleus via the nuclear pore complex. (E) Capsid uncoating and release of the proviral DNA precede (F) the synthesis of the 2nd strand of DNA. The viral genome then persists either as (G) a non-integrated single- or double-stranded episome or (H) integrates into the host genome.
76
Chapter 1 - Background
Capsid proteins are degraded by the proteasome. Then capsid antigenic motifs are charged onto major histocompatibility complex I (MHCI) for presentation at the surface of the transduced cell and trigger immune response. The proteasome inhibition by inhibitors such as bortezomib increases by 2-fold the transduction (250). Similarly, phosphorylation of tyrosine residues on capsid proteins mediates ubiquitin signals to target proteins to proteasome-mediated degradation. Directed tyrosine mutated capsids allow an escape from early degradation and promote a better transduction (251). This has been demonstrated in vivo with different serotypes AAV2, 8, 9 with a 3 log-increase of green fluorescent protein (GFP) intensity (252).
Self-complementary (sc) vectors differ from single-stranded vectors (ss) in that they contain a self-complementary transgene cassette that folds back on itself to form a double-stranded DNA thereby bypassing the requirement for second strand synthesis, which is considered as a rate-limiting step for transgene expression (Figure 16) (253). The transduction efficacy can be increased by 5 to 140 times (253). As a consequence the packaging size of the transgene cassette in scAAV is reduced by half (254). scAAV have been associated with higher innate immune response (255).
A Single-stranded AAV vector B Self-complementary AAV vector ITR
ITR ITR ITR ITR
Genome 4.7 kb 2.3 kb capacity
Promoter Transgene Poly(A) tail
Figure 16. Recombinant single-stranded (ss) and self-complementary (sc) recombinant AAV vector genomes
Kb: kilobases; ITR: inverted terminal repeat. 77
Chapter 1 - Background
AAV vector production
Vector production is commonly performed via a triple transfection with 3 different plasmids in packaging cells expressing adenovirus E1a and E1b genes, such as human embryonic kidney (HEK) 293 T cells or Hela cells. Adenovirus is replaced by a plasmid containing the adenoviral genes (E2a, E4, and virus-associated RNA) required for AAV replication and packaging which increases the safety (236, 256). Two other plasmids providing the transgene cassette and the rep and cap genes are transfected (Figure 17).
Another common approach uses a double baculovirus infection in which two re- engineered baculoviruses are providing the transgene cassette and the rep and cap genes under the control of baculovirus promoters in sf9 insect cells (224).
The purification can use several methods: density gradient centrifugation with cesium chloride or iodixanol, ion exchange or affinity chromatography. The result will be a mixture of both empty and full capsids which can range fro 20:1 to 0.05:1 respectively (224). If centrifugation-based methods are recommended to remove a maximum of empty capsids, this antigenic payload might have a positive effect when used as decoys to soak up pre-existing neutralising antibodies (199).
Titration methods use either quantitative PCR (257) or alkaline gels (258).
78
Chapter 1 - Background
AAV"vector"plasmid" Helper"plasmid" Transgene=containing"plasmid"
Rep" Cap" ITR" Promoter" Therapeu2c"gene" ITR"
4" 1"
5" 2" 2"1"
3" Packaging"cell"
Figure 17. The triple transfection method for adeno-associated virus vector production
A first plasmid carries the adeno-associated virus rep and cap genes encoding the replicative and capsid proteins respectively, the second helper adenoviral plasmid allows the adeno-associated virus replication, the third plasmid contains the therapeutic gene flanked by ITR regions (Inverted terminal repeats). 1) The three plasmids are transfected in a packaging cell. 2) The plasmids are expressed and transcribed. 3) Synthesized proteins are assembling to form a virion. 4) Virions accumulate in high numbers until the cell bursts or is scrapped. 5) Virions are collected and purified. AAV: adeno-associated virus.
79
Chapter 1 - Background
1.3.6 Main challenges of AAV-mediated gene therapy for liver diseases
Insertional mutagenesis
Although more than 170 AAV-mediated clinical trials are ongoing or have been completed (259), no safety issues regarding insertional mutagenesis have been reported. This is facilitated by the persistence of the transgene delivered by the AAV vector as a preferential non-integrated (or episomal) form. However publications have shown a small rate of integration in the host genome, especially in transcriptionally active genes, areas with damaged DNA or enriched CpG islands (247). Experiments in neonatal mice have identified an increased risk of hepatocellular carcinoma (HCC) after systemic injection, which was influenced by the enhancer/promoter activity, the younger age at time of injection and the vector dose (260-262). Analysis of integration sites identified a rodent-specific hotspot in the Rian locus, which was not observed in human trials, in which a genome wide integration pattern involving neither HCC-related genes nor the human Rian homolog, MEG8 in the Dlk1-Dio3 domain (263-265).
After the detection of wild-type AAV2 sequences in 11/193 HCCs within HCC-related genes (266), controversies have risen regarding the possible mutagenic role of wild- type AAV. Whether these findings were the cause or a consequence of the proliferation of the cancer cells, has been largely debated with no definitive conclusion as yet (259, 265, 267, 268).
Several arguments are in favour of the high safety profile of AAV vectors: i) the absence of carcinogenic events observed in a rapidly expanding number of trials; ii) the low rate of AAV integrations associated with HCC despite the high seroprevalence of wild type AAV in the human population (e.g. >80% for AAV2) (269); iii) studies have highlighted the protective role of AAV against cancer in
80
Chapter 1 - Background cervical carcinomas (259), likely by inhibition of c-Ha-ras, c-fos and c-myc proto- oncogenes by AAV2 Rep proteins. Nevertheless, the remaining interrogations and observations in neonatal rodents warrant close monitoring in ongoing human trials.
AAV Immunity
AAV vector injection triggers a non-specific innate immunity with the release of proinflammatory cytokines (270) after interactions with TLR receptors (271, 272). This innate immune response is weaker compared to other immunogenic vectors like adenoviral vectors, and is essential to initiate an adaptive response. Secondarily after presentation of antigenic motifs of the capsid primed on major histocompatibility complex I by antigen presenting cells, an adaptive immunity occurs. This highly- specific immune response expands B- and T-cells against the capsid and/or the transgene product.
Neutralising antibodies against the capsid prevent transduction after systemic delivery (273). CD8+ T-cell mediated immune response generates asymptomatic raise of transaminases in haemophilia B trials (202, 204). Liver presents tolerogenic properties with the proliferation of a specific subset of T cells, CD4+CD25+FoxP3+ T cells (known as “regulatory” T cells) (274). These cells can re-orientate the immune response in suppressing cytotoxicity and developing immune tolerance (275). The mechanism of tolerance requires continuous antigen presentation and has been successfully induced with transgenic proteins (275, 276) but not with capsid proteins rapidly eliminated and unlikely to induce tolerance (246).
To overcome these immune responses, which prevent cell transduction or destroy transduced cells, some strategies have been proposed: i) capsid modification targeting specific highly antigenic epitopes (277); ii) strategies optimising the transduction of the target organ to decrease the vector dose and the subsequent immune response e.g. optimised expression cassette design, or specific capsid modifications (278); iii) immunomodulation with either corticosteroids (204, 279, 280) or more complex protocols similar to immunosuppression prescribed after organ 81
Chapter 1 - Background transplantation e.g. monoclonal antiCD20 antibody (rituximab) (281, 282), non- depleting antiCD4 antibody (283), sirolimus (282), cyclosporine A (283, 284), tacrolimus with mycophenolate mofetil (279), or proteasome inhibitors (285).
Size contraints of AAV vectors
A single-stranded AAV vector can accommodate a transgene with limited size, usually with a maximum of 4600 to 5200 base pairs (286, 287). This capacity is divided by half (2.2 kilobase pairs) in self-complementary vectors carrying a transgene with double-stranded DNA (254). In comparison with other common viral vectors, this is a major inconvenient as lentiviral/retroviral vectors can accommodate transgenes up to 14 kb (288) and helper-dependent adenoviruses up to at least 38.9 kb (289).
To deliver oversized transgenes, approaches have been developed in either refining the design of the transgene cassette with mini-promoters or mini-genes of interest (290) or combining dual AAV co-transduction with a fragmented transgene (286), although this latter approach often reduces functional transduction efficiency.
Worldwide shortage of manufacturing capacity
The rapidly growing demand in supply of good-manufacturing practice (GMP)-graded AAV vectors for preclinical or clinical studies has generated a worldwide shortage in manufacturing capacities although recent improvements in bioreactor capacity and purification methods should progressively help in coping with the high demand (291, 292).
Metabolic zonation
Across the hepatic lobule, which represents the functional unit of the liver, metabolic pathways are differently regulated and expressed. This variability is called metabolic zonation. For example, the urea cycle activity mostly takes place in periportal
82
Chapter 1 - Background hepatocytes and glutamine synthesis preferentially in perivenous hepatocytes (Figure 18) (2). The pattern of liver transduction in the hepatic lobule with a same AAV serotype varies among different species. For example, using the AAV8 capsid, transgene expression is highest in perivenous hepatocytes in mice and dogs and predominants in periportal hepatocytes in non-human primates (Figure 18) (293). The control of hyperammonaemia in a mouse model of urea cycle defect using an AAV8 capsid might require a higher transgene expression compared to primates caused by the preferential transduction of hepatocytes, which do not express the enzymes of the urea cycle. This was suggested in a mouse model of OTC deficiency treated with an AAV8 vector (294).
Peripheral intravenous delivery of AAV8 Mice, dog Non human primates
Zona;on Perivenous hepatocytes Periportal hepatocytes Anatomic
Urea cycle Func;onal Glutamine synthesis
Figure 18. Species-related differences in transduction of the hepatic lobule by AAV vectors
Representation of the liver metabolic zonation for ammonia clearance: example of liver transduction with AAV8 vector.
Perinatal gene therapy
Albeit perinatal gene therapy has several advantages in correcting early the disease phenotype, preventing long-term sequelae and taking advantage of an immature 83
Chapter 1 - Background immune system, concerns have been highlighted with regards to tumorigenicity in the developing murine liver (260, 261) and the progressive loss of vector genomes over time in parallel with the liver growth (Figure 19). It is unlikely that a perinatal injection with a non-integrative AAV vector will be sufficient to provide lifelong correction of the phenotype in metabolic liver diseases, and a reinjection later in life will be necessary.
Alternative integrating approaches, using genome editing tools to design AAV vector delivering an integrating transgene, have been successfully reported as proof of concept in a neonatal mouse model of ornithine ranscarbamylase deficiency (Spfash mouse). A dual injection of AAV vectors was performed simultaneously with both the therapeutic transgene and the editing system for integration with Piggybac transposase (295) and CRISPR-Cas9 (296), the latter allowing site-specific integration.
2000
1500
1000
Weight (grams) Weight 500
0 0.5 1 5 10 25 65 Age (years)
Male and female Male Female
Figure 19. Liver mass according to age and gender
Data from (297, 298)
84
Chapter 1 - Background
1.3.7 Urea cycle defects and gene therapy
Urea cycle defects are life-threatening disorders with unmet therapeutic needs and represent a paradigm for the development of gene therapy. The four most common defects have been successfully cured in mouse models using viral vectors.
Ornithine transcarbamylase deficiency
A residual ornithine transcarbamylase activity of 8% (43) to 10% (299) is expected to significantly modify a severe phenotype into a minor one.
As being the most prevalent urea cycle defect, ornithine transcarbamylase deficiency has been extensively studied. Spf (sparse fur) and spfash (abnormal skin and hair) mice are two hypomorphic mouse models available with 10% and 5% of residual ornithine transcarbamylase activity, respectively. The phenotype is mild with excretion of orotic aciduria but without hyperammonaemia. An anti-ornitihine transcarbamylase short hairpin RNA (shRNA) encapsidated in AAV8 can silence the endogenous murine Otc expression and create a severe hyperammonaemic ornithine transcarbamylase deficient mouse model (294). No large animal model exists for this disorder.
Several groups have demonstrated the efficacy of a single AAV8-mediated gene therapy injection in spfash adult (300-302) and neonatal (303) mice in restoring ornithine transcarbamylase enzymatic activity, and restoring ureagenesis. However a single neonatal injection was not able to support long-term correction due to i) episomal loss during the liver growth, ii) a different transduction pattern irrespective of the physiological metabolic zonation of the hepatic lobule (See Figure 18 & Section 1.3.6) (303).
85
Chapter 1 - Background
Citrullinaemia type 1 (argininosuccinate synthase deficiency)
A knockout murine model of citrullinaemia type 1 has been successfully treated by four sequential systemic injections in utero (E16) and at birth (D0) and at day 14 (D14) and 28 (D28) of life in alternating injections with AAVrh10 (E16, D0) and AAV8 (D14, D28) capsids. Survival was normalised but hyperammonaemia was persisting (304).
Argininosuccinic aciduria (argininosuccinate lyase deficiency)
A helper-dependent adenoviral vector encoding the murine Asl gene under the control of a liver specific promoter was successfully administered to the ASL-deficient AslNeo/Neo mouse (See Section 1.2.8) showing normalisation of survival, improvement of growth, normalisation of plasma amino acids. However this did not correct ASL deficiency in extra-hepatic tissues as shown in aorta tissue (85).
Hyperargininaemia type 1 (arginase deficiency)
A hypomorphic mouse model of hyperargininaemia type 1 was treated by systemic injection of AAV8 vector. This showed a rescue of the survival and an improvement of growth (305), behavioural tests (306) but ammonaemia remained elevated. A minimal improvement of in vivo ureagenesis of 3.3% compared to wild-type mice was sufficient to improve significantly the survival (307).
N-acetylglutamate synthase deficiency
No animal model has been described (308).
Carbamoyl-phosphate synthase 1 deficiency
A knockout mouse model has been described but to our knowledge, the colony is not available anymore (309). 86
Chapter 1 - Background
1.4 Hypothesis and aims of this thesis
1.4.1 Hypothesis and aim 1
Argininosuccinic aciduria is the second most common urea cycle defect and is an inherited condition proven to cause systemic nitric oxide deficiency (32) as the disease is caused by mutations in argininosuccinate lyase (ASL), an enzyme involved in two metabolic pathways: i) the liver-based urea cycle which detoxifies ammonia, and ii) the citrulline-nitric oxide cycle, present in most organs, producing nitric oxide from L-arginine (See Section 1.2.1) (310). Compared to other urea cycle defects, argininosuccinic aciduria presents a lower rate of hyperammonaemic episodes but a higher rate of neurological and systemic complications (18), identified as the argininosuccinic aciduria paradox (See Section 1.2.4). Long-term survival and neurological outcomes of these patients remain poor (See Section 1.2.6) despite various therapeutic approaches (See Section 1.2.5), which are essentially focussed on controlling ammonia levels and correcting the urea cycle and do not target nitric oxide deficiency (18). As discussed in this chapter, there is increasing evidence that nitric oxide might be involved in the clinical phenotype explaining at least partially arterial hypertension (See Section 1.2.3) (85).
A hypomorphic AslNeo/Neo mouse model has been described reproducing the human phenotype with multi-organ disease, hyperammonaemia and early death and shows impairment of both urea and citrulline-NO cycles (See Section 1.2.8) (32).
Our first hypothesis is that nitric oxide plays a key-role in the multi-organ disease developed by patients and is involved in the neuropathology of the disease and the paradox of argininosuccinic aciduria.
87
Chapter 1 - Background
In testing this hypothesis, this thesis aims: i) To better decipher the natural history of patients suffering from argininosuccinic aciduria in assessing the long-term outcome of a British cohort with a particular attention to the genotype-phenotype correlation (Chapter 3). ii) To study the neuropathology of argininosuccinic aciduria in the AslNeo/Neo mouse model, for which the neurological phenotype has not been described yet (Chapter 5).
1.4.2 Hypothesis and aim 2
AAV-mediated gene therapy has achieved promising results in recent clinical trials in liver (204) and neurological (311) inherited diseases (See Section 1.3.6). This viral vector is emerging as the leading strategy for liver-directed gene therapy, which is a logical and necessary approach to control ammonia levels and curing urea cycle defects (See Section 1.3.7) such as argininosuccinic aciduria. The severe long-term outcome of patients affected by argininosuccinic aciduria demonstrate the urgent need for new therapeutic approaches (See Section 1.2.6).
A liver-directed adenovirus-mediated gene therapy in the AslNeo/Neo mouse model was able to correct the urea cycle but not the extra-hepatic features of the disease unrelated to hyperammonaemia (See Section 1.3.7). Adenoviral vectors have elicited safety concerns in trials for liver-directed gene therapy (See Section 1.3.2) (157) and are not currently a vector translatable into clinics.
Our second hypothesis is that AAV-mediated gene therapy is a suitable therapeutic approach for argininosuccinic aciduria.
In testing this hypothesis, this thesis aims:
88
Chapter 1 - Background iii) To develop a liquid chromatography-tandem mass spectrometry based method to accurately measure the amino acids involved in the urea cycle and argininosuccinate lyase enzymatic activity (Chapter 4). iv) To set up a proof-of-concept of AAV-mediated gene therapy in AslNeo/Neo mice in targeting both metabolic pathways impaired in argininosuccinic aciduria i.e urea and citrulline-nitric oxide cycles (Chapters 6 & 7).
89
Chapter 2 – Material and Methods
2. MATERIAL AND METHODS
2.1 Reagents
Unless stated otherwise, general laboratory reagents were purchased from Sigma- Aldrich (Sigma, St Louis, MO, USA). Protocols for standard laboratory solutions and specific buffer and reagents are listed in Appendix 10.2.
2.2 Characterisation of a new cohort of patients with argininosuccinic aciduria
2.2.1 Clinical case recruitment
Patients included had plasma argininosuccinic acid levels > 5 μM, and/or pathogenic mutations in hASL. Anonymised data were collected prospectively from March 2013 and retrospectively before, from five tertiary metabolic centres in the UK: Birmingham Children’s Hospital, Birmingham; Guy’s and St Thomas’ Hospital, London; Great Ormond Street Hospital for Children, London; the National Hospital for Neurology and Neurosurgery, London; the Royal Manchester Children Hospital, Manchester. Patients were considered lost to follow-up if no clinical assessment was performed during the last 3 years at the relevant metabolic centre. The database was closed on 31st December 2015.
90
Chapter 2 – Material and Methods
2.2.2 Consent and ethics approval
Molecular analysis of patients was approved by the National Research Ethics Service Committee London-Bloomsbury (13/LO/0168).
2.2.3 Clinical phenotyping
The neurological outcome was assessed via physical neurological examination and especially collected informations related to the following symptoms: intellectual disability, epilepsy, ataxia, myopathy-like symptoms and brain magnetic resonance imaging (MRI) features. The assessment was performed as follows: if neuropsychological assessment was unavailable, cognitive impairment was determined by clinical judgement of the metabolic specialist or neuropaediatrician or by the need for an additional support at school or subsequently at the workplace. Epilepsy was defined as the repetitive occurrence of two or more seizures without contemporary hyperammonaemia. Indication for neuroimaging was an unexplained and/or severe neurological disease. Brain MRIs were analysed by two neuroradiologists blinded to the report of each other with the help of Dr Kling Chong, Neuroradiology Department, Great Ormond Street Hospital, London and Dr Katharine Foster and Lesley MacPherson, Neuroradiology Department, Birmingham Children’s Hospital, Birmingham. Proton magnetic resonance spectroscopy (1H- MRS) protocols were performed as described (312). 1H-MRS studies were performed concurrently with clinically indicated MRI scans at 1.5 Tesla. Comparison was made with 1H-MRS metabolite data from a standard cohort of children with normal appearing MRI as described previously (312).
Systemic phenotype monitored involvement of liver, kidney, arterial hypertension, gut, pancreas, heart and hair. Liver involvement was considered using the following parameters: enlarged liver, increased levels of transaminases assessed by alanine
91
Chapter 2 – Material and Methods aminotransferase (ALT) above 50 IU/L. Nephromegaly was defined as renal length on ultrasound imaging above the 95th centile for the age and sex.
Biochemical data were assessed using the mean of at least the last 10 available results during compensated metabolic state. Plasma ammonia levels were considered elevated if > 100 µM before 28 days of life or > 45 µM subsequently. Hypokalaemia was defined as a plasma potassium level lower than 3.5 mM and judged as “transient” if observed in a single sample, or “persistent” if measured in ≥ 2 samples separated by ≥ 1 month. Plasma arginine and argininosuccinic acid reflected the last 10 measurements performed during follow-up in a compensated metabolic state on standard treatment. For comparative analysis, patients were divided into three groups: (i) early-onset form (hyperammonaemic symptoms started on/before 28 days of life), (ii) late-onset form (presentation after 28 days of life), (iii) perinatally screened patients diagnosed after a family proband and treated prospectively from birth. For the last group, the status (early- or late-onset) of the familial proband was investigated but missing data prevented inclusion of these index cases into the study.
2.3 hASL mutational analysis
2.3.1 Design of primers
Primers were designed for the amplification of human ASL transcript ASL-005 based on the reference sequence ENST00000395332 (www.ensembl.org). This transcript is 1,622 base pairs in length and results in the synthesis of the full protein (464 amino acids). Exons and introns were identified in the genomic sequence.
The primers were designed approximately 50 base pairs upstream and downstream of the intended exon to ensure that the splice sites and the exon-intron boundaries would be amplified for sequencing. Each primer was approximately 20 nucleotides in length and did not contain mutations or polymorphisms reported in www.ensembl.org. Each primer was screened with Basic Local Alignment Search 92
Chapter 2 – Material and Methods
Tool (BLAST) (http://www.ncbi.nlm.nih.gov/tools/primer-blast/), to search for paralogues. When designing the primers, pair considerations were made with regard to the annealing temperature (range of 60-65°C), GC content (40-60%) and self- complementarity to prevent the formation of self-complementary dimers or primer- dimer.
Tails were added at the 5’ end of the polymerase chain reaction (PCR) primers. These ‘tails’ were then incorporated into the PCR product in order to make downstream sequencing more rapid. The sequence of the universal tailing primers were as follows: forward - AGCTAAGCGCGAGAAGGC and reverse - GCTTTACCGCTCAACCGTT.
2.3.2 PCR optimisation of hASL exons
Reagents and buffers used are as detailed in the Appendix 10.3.1.
Primers were diluted in nuclease free water. Stock solutions of 100 µM were prepared prior to making a working dilution of 10 µM.
The standard reaction mix for a 25 µL PCR using the BIOTAQTM DNA polymerase Kit contained 0.2 µL (1 unit) of DNA Polymerase, 2.5 µL of 10x NH4 reaction buffer, 2.5 µL of 10 mM deoxyribonucleotide triphosphate (dNTPs), 1.25 µL of each primer diluted at 5 µM, 1 µL of DNA template (concentration 100 µg/µL approximately). Standard cycling conditions included an initial activation step of 95°C for 60 seconds followed by 35 cycles of denaturation at 95°C for 30 seconds, the optimised temperature for each set of primers for 30 seconds and an extension temperature of 72°C for 30 seconds. A final extension step was performed at 72°C for 10 minutes. PCR reactions were optimised for each pair of primers using a 96-well Fast Thermal Cycler (ThermoScientific, Kalamazoo, MI, USA). Assessed parameters included annealing temperatures, MgCl2 concentrations, presence of enhancer (Betaine) at 0.5 M. PCR products were visualised using a Ultraviolet transilluminator BioDoct-It Imaging system (UVP, Upland, CA, USA) with UVO TS software (UVP, Upland, CA,
93
Chapter 2 – Material and Methods
USA) after electrophoresis through a 1% agarose gel. The optimised PCR conditions for all sets of primers are summarised in Table 6.
2.3.3 Sequencing and data analysis
Reagents and buffers used are as detailed in the Appendix 10.3.2.
DNA was extracted from white blood cells in North East Thames Regional Genetics Service, Great Ormond Street Hospital, London, UK. The individual exons of hASL were amplified by PCR using the optimised conditions and the PCR products were visualised on a 1% agarose gel (See Section 2.3.2).
The amplified samples were treated with exonuclease I and Shrimp alkaline phosphatase to remove leftover primers and remaining deoxynucleotides triphosphate (dNTPs), respectively. 2.5 µL of a mix containing 0.75 µL of double distilled water (ddH2O), 0.25 µL of Shrimp alkaline phosphatase dilution buffer, 0.5 µL of exonuclease I and 1 µL Shrimp alkaline phosphatase was added to 10 µL of the PCR product in a 96-well plate and the plate was sealed with an adhesive lid. After a brief centrifugation to ensure all reagents were at the bottom of the well, the reaction was incubated at 37°C for 15 minutes and then the enzymes heat-inactivated by heating at 80°C for 15 minutes.
A sequencing reaction with Big Dye Terminator 1.1 was performed to amplify the sample DNA with dye-binding dideoxynucleotides triphosphate (ddNTPs). Each ddNTP (Adenosine, Thymidine, Cytosine, Guanosine) is associated with a different coloured dye (yellow, blue, red, green). The standard reaction mix for a 10 µL PCR contained 1.5 µL of buffer, 0.5 µL of Big Dye version 1.1, 1 µL of universal tailing primers (forward: AGCTAAGCGCGAGAAGGC; reverse: GCTTTACCGCTCAACCGTT) diluted at 5 μM, 3 µL of cleaned PCR product template. Standard cycling conditions used included an initial activation step of 95°C for 2 minutes followed by 35 cycles of denaturation at 95°C for 20 seconds, an annealing step at 50°C for 10 seconds and an extension step at 60°C for 3 minutes.
94
Chapter 2 – Material and Methods
The sequencing product was precipitated by adding 2 µL of 3 M sodium acetate, pH 5.2 and 50 µL of 100% ethanol per sample. These were left at room temperature for 20 minutes. The plate was then centrifuged at 3,000 g for 40 minutes and the supernatant thrown off. 50 µL of 70% ethanol per well was added to wash the DNA pellet and the plate was centrifuged again at 3,000 g for 10 minutes and the supernatant thrown off. The plate was turned upside down on a piece of tissue and centrifuged finally at 300 g for 1 minute to remove any remaining ethanol. The DNA pellet was resuspended in 10 µL of 1:10 TE buffer and the plate stored at -20°C before sequencing.
Plates were then sent to North East Thames Regional Genetics Service, Great Ormond Street Hospital, London, UK for sequencing on an ABI PRISM 3730 DNA Analyzer (Applied Biosystems/ThermoFisher Scientific) and data analysed using Sequencher 5.3 software (Gene Codes Corporation, Ann Arbor, MI, USA).
The prevalence of all variants identified was assessed in the Ensembl (http://ensembl.org) and ExAC databases (http://exac.broadinstitute.org), and their presence or absence in the Human Gene Mutation Database was examined (last accessed 11/12/2015; http://www.hgmd.cf.ac.uk). All protein alignments were performed using ClustalW2 software (http://www.ebi.ac.uk/Tools/msa/clustalw2/).
95
Chapter 2 – Material and Methods
Human&ASL&sequencing From&Transcript&ASL&N°5&ID&Ensembl&ENST00000395332 PCR&conditions Annealing& Mg& Expected& Annealing& Betaine& Primers& Exon s/as Sequence Tailed&sequence temperature& concentration& band&size time&(sec) 0.5&M dilution (Deg) (mM) 1 s ACACTATCCGTGCGGCCAGG AGCTAAGCGCGAGAAGGCACACTATCCGTGCGGCCAGG 313 64 30 1 No 1;20 as ACTCTCTCCTTTGGAGGCTA GCTTTACCGCTCAACCGTTACTCTCTCCTTTGGAGGCTA
2 s CCTGCTACCATCAGACTTGA AGCTAAGCGCGAGAAGGCCCTGCTACCATCAGACTTGA 384 64 30 1.5 No 1;20 as CAACACTGCACTGTTTGCTC GCTTTACCGCTCAACCGTTCAACACTGCACTGTTTGCTC
3 s AATCCCTGAGCAAACAGTGC AGCTAAGCGCGAGAAGGCAATCCCTGAGCAAACAGTGC 761 64 30 1.5 No 1;20 as GGGCAGTTTTCCTCATCAGC GCTTTACCGCTCAACCGTTGGGCAGTTTTCCTCATCAGC
4+5 s ACTGCCCTGCCTGGGTTGAC AGCTAAGCGCGAGAAGGCACTGCCCTGCCTGGGTTGAC 412 64 30 1.5 No 1;20 as GGCTTCCATCACACCTCTGT GCTTTACCGCTCAACCGTTGGCTTCCATCACACCTCTGT
6+7 s CATTCAGGTGGAGTGCTGC AGCTAAGCGCGAGAAGGCCATTCAGGTGGAGTGCTGC 466 68 10 1.5 No 1;20 as TAAGGCGGTTTTCTGTCCCA GCTTTACCGCTCAACCGTTTAAGGCGGTTTTCTGTCCCA
8 s CTTGGTTCTCTGTGTGTGCG AGCTAAGCGCGAGAAGGCCTTGGTTCTCTGTGTGTGCG 236 60 10 1.5 No 1;20 as AGGTGTGTCTGCAGGTCTTG GCTTTACCGCTCAACCGTTAGGTGTGTCTGCAGGTCTTG
9 s CTGTGCAAAAGATCCCTCCC AGCTAAGCGCGAGAAGGCCTGTGCAAAAGATCCCTCCC 246 68 30 1.5 No 1;20 as GCTGAAGGACTAAAGTGAGGC GCTTTACCGCTCAACCGTTGCTGAAGGACTAAAGTGAGGC
10 s GAAAATTTAGCCGGGTCCCC AGCTAAGCGCGAGAAGGCGAAAATTTAGCCGGGTCCCC 248 60 20 1.5 Yes as AGATGGAGGTGGCAGTTCAG GCTTTACCGCTCAACCGTTAGATGGAGGTGGCAGTTCAG 1;20
11+12 s GACGTGGCTGCCTTCCTC AGCTAAGCGCGAGAAGGCGACGTGGCTGCCTTCCTC 499 60 30 1.5 Yes 1;20 as AAGGGCAGAGAGTTCAAGGG GCTTTACCGCTCAACCGTTAAGGGCAGAGAGTTCAAGGG
13 s ATGGAAGGCAGTGGGGATG AGCTAAGCGCGAGAAGGCATGGAAGGCAGTGGGGATG 231 68 30 1.5 No 1;20 as CAGTGCCTGGGACCTAGG GCTTTACCGCTCAACCGTTCAGTGCCTGGGACCTAGG
14 s CAGCTTCAGATCCCAGGGTC AGCTAAGCGCGAGAAGGCCAGCTTCAGATCCCAGGGTC 212 68 10 1.5 Yes 1;20 as CCCATTCCTGAGCTCCCC GCTTTACCGCTCAACCGTTCCCATTCCTGAGCTCCCC
15 s GGGTGCAGGCAATGGAGGCA AGCTAAGCGCGAGAAGGCGGGTGCAGGCAATGGAGGCA 215 64 10 1.5 No 1;20 as CTAGGAGGCAGCATGGGG GCTTTACCGCTCAACCGTTCTAGGAGGCAGCATGGGG
16 s GAAGTGAGCCTGGGTGCC AGCTAAGCGCGAGAAGGCGAAGTGAGCCTGGGTGCC 300 64 30 1.5 No 1;20 as CAGCAACGAGGGAGCCAG GCTTTACCGCTCAACCGTTCAGCAACGAGGGAGCCAG Table 6. Sequential PCR optimisation for the 16 exons of hASL hASL transcript ASL-005 reference ENST000000395332 in ensembl.org
96 Chapter 2 – Material and Methods
2.4 Phenotyping of AslNeo/Neo mice
2.4.1 Study approval
All procedures were performed in compliance with UK Home Office regulations and within the guidelines of University College London ethical review committee. The experimets were performed under the Project License of Dr Simon Waddington (PPL licenses 70/6906 and 70/8030) and Personal License (Dr Julien Baruteau PIL 70/25779).
2.4.2 Animals
The AslNeo/Neo mice (B6.129S7-Asltm1Brle/J) were purchased from Jackson Laboratory (Bar Harbor, ME). Wild type or heterozygote mice were maintained on standard rodent chow (Harlan 2018, Teklab Diets, Madison, WI; protein content 18%) with free access to water. AslNeo/Neo mice were started on a supportive treatment including reduced-protein diet (5CR4, Labdiet, St Louis, MO; protein content 14.1%) from day 15 to day 50 and received daily intraperitoneal injections of sodium benzoate (0.1g/kg/d) and L-arginine (1g/kg/d) from day 10 to day 30.
Mice were weighed daily during the first 45 days then every 5 days until day 120 then every 15 days. Mice were checked daily for assessing behaviour and wellbeing.
2.4.3 DNA extraction from mouse tissue
Reagents and buffers used are as detailed in the Appendix 10.3.3.
DNA was extracted from an ear or tail clip using sodium hydroxide (NaOH) as follow:
97 Chapter 2 – Material and Methods i) The 2 - 3mm tissue clip was placed in 75 µL of 25 mM NaOH containing 0.2 mM EDTA, adjusted to pH 12. ii) The mix was heated to 95°C for 10 minutes then cooled to room temperature. iii) 75 µL of 40 mM Tris-HCl, adjusted to pH 5 was finally added. iv) 4 µL was used in a 20 µL PCR reaction.
2.4.4 Genotyping
a) PCR amplification
A commercially available PCR kit (Taq DNA Polymerase, Peqlab, Germany) was used as per manufacturer’s instructions. The PCR reaction mix is detailed in Table 7.
Reagent Volume (μL)
10x PCR reaction buffer 2 (containing 20 mM MgCl2)
dNTP mixture (10 mM) 2
Forward primer (25 µM) 1
Reverse 1 primer (25 µM) 1
Reverse 2 primer (25 µM) 1
Taq DNA polymerase (5 U/µL) 0.1
Nuclease-free water 8.9
DNA sample 4
Final volume 20
Table 7. PCR reaction mix for genotyping
98 Chapter 2 – Material and Methods
Two sets of primers were subsequently tested.
The first genotyping method relied on primers described in Erez et al (32): Forward primer: 5’-GTACCCAGAGCCTCAACCTC-3’; Reverse primer for WT allele: 5’- CAGATCTTACGAGGATGGAAGG-3’ and Reverse primer for AslNeo/Neo allele: 5’- CGTCCTGCAGTTCATTCAG-3’ and reaction cycles: 95°C 1’, 58°C 1’, 72°C 1’ for 35 cycles.
A second set of primers as advised by Jackson Laboratory (Bar Harbor, MA, USA) was tested: Forward primer: 5’-GGTTCTTGGTGCTCATGGAT-3’; Reverse primer for WT allele: 5’-GCCAGAGGCCACTTGTGTAG-3’ and Reverse primer for AslNeo/Neo allele: 5’-CATGACAGCTCCCATGAAGA-3’ binding in the 9th intron of the mouse Asl gene (Figure 20). The reaction cycle: 94°C for 30 seconds, 63°C for 30 seconds, 72°C for 1 minute. This reaction cycle was repeated 40 times. The same PCR reaction mix as detailed in Table 7 was used with a primer concentration of 25 µM. Optimisation of the following parameters were performed i) annealing temperature, ii) primers concentration, iii) magnesium chloride concentrations, iv) utility of PCR enhancer and is presented below.
b) DNA electrophoresis on agarose gel
Reagents and buffers used are as detailed in the Appendix 10.3.4.
A 2% agarose gel was made up by adding agarose powder to 1x TAE solution (Tris base, acetic acid, EDTA), and microwaving at 900 watts for 3 minutes. A Safeview nucleic acid stain (NBS Biologicals, Huntingdon, UK) was added (5 µL per 100 mL) before pouring the mixture into an appropriate tank with a 20-wells comb. The gel was allowed to cool at room temperature for 15 minutes.
The gel was then placed in a PCR tank filled with 1x TAE. 20 µL of PCR sample was mixed with 4 µL of PCR loading buffer and 10 µL of the resulting mix was loaded into each well. Electrophoresis of the samples was performed at 100 Volts towards the positive (red) electrode, for 60 minutes. The gel was imaged in a benchtop ultraviolet
99 Chapter 2 – Material and Methods transilluminator BioDoct-It Imaging system (UVP, Upland, CA, USA) with UVO TS software (UVP, Upland, CA, USA).
A
B
C
Figure 20. Binding sites of primers for genotyping wild-type and AslNeo alleles.
(A) Map of the genomic DNA of mouse Asl gene (NC000071.6) identifying exons (in blue), introns and set of primers used for identification of the wild-type allele (in red).
100 Chapter 2 – Material and Methods
(B) Detailed sequence of the gene identifying the 20-base pairs long primers (in red) in the 9th intron flanked by adjacent exons (in blue). (C) Partial sequence of the neomycin cassette aligned with the reverse primer used for detection of the AslNeo allele (neomycin and primer sequences provided by Jackson Laboratory). Alignment performed with Snapgene (GSL Biotech LLC, Chicago, IL, USA).
c) Optimisation of PCR conditions
A method for early genotyping in the first 24 hours of life of the animals was developed. Wild-type, heterozygote and AslNeo alleles were identified by PCR amplification targeting sequences specific to the WT or the AslNeo allele.
As a first PCR amplification using a set of primers published in Erez et al (32) couldn’t allow a suitable identification of the alleles, a second set of primers and adapted PCR conditions provided by Jackson Laboratory (Bar Harbor, MA, USA) were used and optimised as follows:
i) Annealing temperature
The AslNeo and WT sets of primers were tested at annealing temperatures varying between 60 to 65°C (Figure 21). 63°C was chosen as an acceptable consensus for both alleles in order to perform a unique PCR reaction testing both alleles.
101 Chapter 2 – Material and Methods
100 bp ladder 500 bp 400 bp 300 bp i 200 bp ii
Annealing 60 61 62 63 64 65 Negative Temperature Control (°C) (63 °C) Figure 21. Optimisation of the annealing temperature for genotyping PCR
Testing of a heterozygote sample: (i) AslNeo allele (250 bp; (ii) WT allele (205 bp).
ii) Primer concentration
Concentration of primers was tested at 50 and 25 µM (Figure 22). 25 µM was considered for the final genotyping protocol.
100#bp## Primers#25#μM# Primers#50#μM# ladder# 600#bp#
400#bp# i# 200#bp# ii#
Genotype# HET1# HET2# Nega@ve# WT# HET1# HET2# Nega@ve# WT# control# control#
Figure 22. Optimisation of the primer concentration for genotyping PCR
Two conditions were tested at 25 to 50 µM. (i) AslNeo allele (250 bp; (ii) WT allele (205 bp). WT: wild-type; HET: heterozygote.
102 Chapter 2 – Material and Methods
iii) MgCl2 concentration
Divalent cations are necessary for polymerase activity but the concentration can dramatically affect the activity. MgCl2 concentration was tested at various concentrations ranging between 2 to 4 mM in the PCR mix (Figure 23A). MgCl2 concentration at 2 mM was considered for the final genotyping protocol.
iv) PCR enhancer
The effect of a PCR enhancer provided by the commercial kit was tested (Figure 23B). GC-rich templates can form strong secondary structures, which can cause inefficient separation of the DNA strands or compete with primers preventing an efficient annealing. The PCR enhancer reduces the DNA base pairs interactions and the stability of hydrogen bonds. No enhancer was added in the final genotyping protocol.
Nega*ve% A" HET1% HET2% control% MgCl % 2 2% 2.5% 3% 3.5% 4% 2% 2.5% 3% 3.5% 4% 2% 2.5% 3% 3.5% 4% (mM)% 600%bp%
400%bp%
i% ii% 200%bp%
B" 600%bp%
400%bp% i% 200%bp% ii%
Figure 23. Optimisation of MgCl2 concentration in the presence or absence of PCR enhancer for genotyping PCR
103 Chapter 2 – Material and Methods
MgCl2 concentration varies from 2 to 4 mM, in absence (A) or presence (B) of PCR enhancer. (i) AslNeo allele (250 bp) ; (ii) WT allele (205 bp). WT: wild-type; HET: heterozygote.
v) Final genotyping PCR protocol
Once optimised, this protocol was successfully tested for all genotypes Asl+/+ (wild- type), Asl+/Neo (heterozygote) and AslNeo/Neo mice (Figure 24).
100 bp ladder 600 bp
400 bp
AslNeo allele, 250 bp 200 bp WT allele, 205 bp
Genotype WT WT WT HET Nega;ve AslNeo/Neo control
Figure 24. Optimised genotyping PCR protocol
(313): wild-type; HET heterozygote.
2.4.5 Weight and survival analysis
Mice were weighed daily during the first 45 days of life then every 5 days until day 120 then every 15 days. Behaviour and wellbeing of the mice was checked daily.
104 Chapter 2 – Material and Methods
2.5 Behavioural studies
For behavioural testing, mice were allowed to acclimatise to the procedure room for 30 to 60 minutes before the start of the experiment. Males were tested before females to avoid bias due to the females’ pheromones. The investigator was blinded to the genotype until the second week of life. At this age, AslNeo/Neo mice were easily distinguishable from wild-type or heterozygote mice by the fur pattern.
Righting reflex and grid walking tests were performed from day 1 and 14, respectively. Other tests were performed from the age of 2 months onwards.
2.5.1 Righting reflex
The mouse is placed in the supine position. The time in seconds to turn over to the prone position and place the 4 paws in contact with the surface is monitored.
2.5.2 Grid walking test
The mouse is placed on a stainless steel grid floor (20 x 40 cm with a mesh size of 3 cm2) elevated 30 cm above the benchtop (Figure 25). The number of foot faults (where a limb fell through the grid) over a 30 second time period and the number of steps are both recorded.
105 Chapter 2 – Material and Methods
Figure 25. Picture of a 2 week-old mouse performing the grid walking test
2.5.3 Rotarod
After a period of acclimatisation including 3 trials per day for 3 consecutive days, the test was performed on a Rotarod LE 8200 (Panlab, Harvad apparatus, Cambridge, UK) with 3 attempts per day for 5 consecutive days. The latency to fall from the rod under continuous acceleration from 4 to 40 revolutions per minute (rpm) over 5 minutes was recorded in accordance with previous published work (314) (Figure 26).
106 Chapter 2 – Material and Methods
Figure 26. Picture of rotarod test
A wild-type mouse has fallen in the far left lane. An AslNeo/Neo mutant is falling in the second lane from the left. Wild-type phenotypic mice are running in the other lanes.
2.5.4 Open field test
The animal was placed in the centre of a plastic box (40 x 40 cm floor) and video- recorded for 5 minutes. Computational analysis of video records to analyse the distance walked was automatically performed using i) either the Vernier Video Physics and Vernier Graphical analysis softwares (http://www.vernier.com/products/software), which allows a manual tracking on iPad. This was used for analysing results in Section 5.3.1 (Figure 27); ii) or the MouseLabTracker v0.2.9 application on Matlab software (Mathworks, Natick, MA, USA) (315), which performs automated analysis. This was used for analysing results in Section 7.7.1 (Figure 28).
107 Chapter 2 – Material and Methods
Figure 27. Open field test: mouse tracking using Vernier Video Physics software
Figure 28. Open field test: mouse tracking using MouseLabTracker application
108 Chapter 2 – Material and Methods
2.5.5 Novel object recognition test
During the first phase, two identical objects were placed in a plastic box (40 x 40 x 40 cm) and the mouse was free to explore for 3 minutes. Four hours later, a second phase was tested and the mouse was presented a familiar object already explored in the first phase and a novel object that they had not been exposed to before. An active exploration was defined as the nose of the animal approaching less than 2 cm from the object. The time of exploration (3 minutes) was recorded using a timer. This test was performed with a minimum of 48 hours interval after the open field test based on published guidelines (316).
2.5.6 Tail suspension test
The mouse was suspended by the tail with vinyl tape placed at 3/4 of the way from the base of the tail with a 10 cm distance from the head to the floor over 5 minutes (Figure 29). A 5 cm long plastic cylinder with soft edges was applied around the base of the tail to prevent any climbing behaviour usually observed with the C57BL/6 strain. The time of recording was 6 minutes. Initially the mouse struggles to face upward; when the animal becomes immobile and stops struggling, he is considered to have given-up. Prolonged immobility is considered as a depression-like state (317).
109 Chapter 2 – Material and Methods
Figure 29. Tail suspension test
A plastic cylinder around the tail prevents the mouse climbing back up.
2.6 Biomarkers
2.6.1 Blood sampling and analysis
Reagents used are as detailed in the Appendix 10.3.5.
Blood was collected by tail vein puncture. Blood volume was in accordance with the Home Office regulation i.e. either 10% of the mouse blood volume once a week or 15% of the mouse blood volume once a month.
Blood sampling was performed under isoflurane anaesthetics either by terminal bleeding with cardiac puncture or tail vein bleeding after a 10-30 minutes pre-heating stay in a heating chamber maintained at 37°C.
110 Chapter 2 – Material and Methods
For dried blood spot collection, whole blood was immediately spiked onto a Guthrie card provided by the Newborn Screening Department, Great Ormond Street Hospital, London. Guthrie cards were let drying at room temperature for 24 hours before being stored in a foil bag with desiccant at -20°C.
Whole blood was collected in a 0.2 mL microcentrifuge tube with 10 µL of sodium heparin (1,000 IU/mL), vortexed and centrifuged at 16,000 g for 5 minutes at room temperature. Red blood cells and plasma were separated forming a pellet and a supernatant, respectively. Plasma was then pipeted carefully, then immediately frozen and stored at -20°C until analysis.
Basic biochemistry parameters in plasma (potassium, urea, creatinine, albumin, liver function tests i.e. transaminases, gamma-glutamyl transpeptidase, alkaline phosphatase, and ammonaemia) were analysed in the Chemical Pathology Department, Great Ormond Street Hospital, London.
2.6.2 Urine sampling and analysis
Urine was collected daily for 3 to 5 consecutive days by gently holding the mice by the scruff of the neck to stimulate urination. Urine was pipeted and collected into pre- opened 0.2 mL microcentrifuge tube on wet ice, immediately frozen after collection and stored at -20°C until analysis.
Urine argininosuccinic acid and creatinine levels were measured at Chemical Pathology Department, Great Ormond Street Hospital, London.
2.6.3 Nitrite and nitrate measurements (Griess reaction)
- - Oxidative derivatives of nitric oxide, nitrite (NO2 ) and nitrate (NO3 ) are directly - proportional to the level of nitric oxide in tissues. NO2 is an unstable compound and - oxidises spontaneously in NO3 . This step can be reverted via the catalytic activity of nitrate reductase. The Griess reaction is based on a diazotization reaction. - Sulfanilamide reacts with NO2 in acidic conditions to form a diazonium salt, which
111 Chapter 2 – Material and Methods reacts with N-1-naphthylethylenediamine dihydrochloride to form a pink azo dye agent, which shows a maximal absorbance with a 550 nm wavelength. Nicotinamide adenine dinucleotide phosphate (NADPH) is required as a cofactor for the reduction as are glucose-6-phosphate dehydrogenase and glucose-6-phosphate, which allow recycling of NADP into NADPH (Figure 30).
Diet
- Nitrate (NO3 )
NADPH 6-Phosphoglucono-δ-lactone Nitrate Glucose-6-phosphate Reductase Dehydrogenase Reduc&on
NADP Glucose-6-phosphate
-
Nitrite (NO2 ) Griess reac9on Oxida&on
Nitric oxide (NO) Biologically ac9ve Nitric oxide molecule Synthase
L-arginine + O 2
Figure 30. Nitric oxide metabolism and principle of Griess reaction
NADP: Nicotinamide adenine dinucleotide; NADPH: Nicotinamide adenine dinucleotide phosphate.
Reagents and buffers used are as detailed in the Appendix 10.3.6.
- - Samples were collected carefully in order to minimise the risk of NO3 and NO2 contamination. For this purpose, all glassware and plasticware were first cleaned with double distilled water (ddH2O). Animals were anaesthetised and perfused on wet ice. Organs were collected on wet ice in less than 3 minutes in a 1.5 mL microcentrifuge tube according to Glowinsli and Iversen (318) and snap-frozen immediately on dry 112 Chapter 2 – Material and Methods
ice. Samples were homogenised in 2 volumes of ddH2O with a grinder (Tissue Master 125, OMNI International, Kennesaw, GA, USA) on ice then centrifuged at 16,000 g at room temperature for 10 minutes in 3,000 kDa cut-off filters (Merck Millipore, Darmstadt, Germany).
The Griess reaction protocol was adapted from Waddington et al (319). Enzymatic stocks were made and stored as follows: nitrate reductase was dissolved in 1:1 phosphate buffer:glycerol to a final specific activity of 59 units/mL and stored at -20°C. Glucose-6-phosphate dehydrogenase was dissolved in 1:1 nitrate assay phosphate buffer:glycerol to a final specific activity of 125 units/mL and stored at -20°C. The enzyme mastermix was made fresh for each experiment using a mixture of 1 mL of phosphate buffer, 10 μL of nitrate reductase stock solution, 10 μL of glucose-6-phosphate dehydrogenase stock solution, 1.14 mg of glucose-6- phosphate. A calibration curve, covering a range of 0 - 600 µM was prepared using serial dilutions of nitrite and nitrate standards. Analysis was performed in a 96-well plate and each sample was analysed in duplicate. 60 μL of sample was mixed with 10 μL of 10 mM NADPH and 40 μL of enzyme mastermix. Samples were incubated - for 1 hour at room temperature on a rotating shaker to allow conversion of NO3 to - NO2 . The Griess reaction was performed by adding 75 µL of 116 mM sulfanilamide, 5% phosphoric acid, then 75 µL of 7.7 mM N-1-naphthylethylenediamine dihydrochloride. The plate was read at 550 nm in a FLUOstar Omega spectrophotometer (BMG Labtech, Ortenberg, Germany). Concentrations of - - cumulative NO3 and NO2 were calculated from the standard curve. To calculate the - NO2 alone, the same Griess reaction was performed with 150 µL of sample without - - the nitrate reduction step and a NO2 standard curve. The NO3 levels were - - - extrapolated by subtracting NO2 from the total NO2 and NO3 value.
2.6.4 Cyclic guanosine monophosphate (cGMP) measurement
Brain samples from phosphate buffered saline (PBS) perfused mice were flash- frozen in dry ice and stored at -80°C before being ground in liquid nitrogen. Samples were weighed and diluted in 10 volumes of 0.1 M hydrochloric acid (HCl) prior to centrifugation at 600 g for 10 minutes. cGMP was measured using a cGMP complete 113 Chapter 2 – Material and Methods
ELISA kit (Enzo Life Sciences, Farmingdale, NY, USA) according to manufacturer’s instructions in a non-acetylated format reaction.
Samples were analysed in a 96-well plate in wells pre-coated with an affinity-purified antibody contained in the commercial kit, GxR IgG antibody. This antibody was binding a rabbit polyclonal antibody anti-cGMP administered in 50 µL of a pre-diluted mix available in the kit. This second antibody bound, in a competitive manner, either the cGMP contained in 100 µL of the tested sample or 50 µL of a blue solution containing cGMP conjugated to alkaline phosphatase. After having added the reagents, the plate was incubated on a rocking shaker (Mini gyratory rocker SSM3, Stuart, Stone, UK) at 500 rpm for 2 hours at room temperature. Wells were then emptied and washed 3 times with a wash buffer provided in the commercial kit and only the bound cGMP was left in the wells. 200 µL of a substrate solution p- Nitrophenyl phosphate (pNpp) was then added to the wells and the plate incubated for an hour at room temperature without shaking. pNpp was catalysed by the alkaline phosphatase binding cGMP; this generated a yellow color, the maximal absorbance of which could be read at 405 nm in a FLUOstar Omega spectrophotometer (BMG Labtech, Ortenberg, Germany). The signal read was indirectly proportional to the amount of cGMP in the sample. Quantification was performed using a standard curve of 5 calibrated points provided in the commercial kit, which ranged from 0.8 to 500 nM.
2.6.5 Glutathione analysis.
Reduced glutathione was measured as described previously by Kamencic et al (320). Snap-frozen samples were weighed, thawed and homogenized in 20 volumes of cold 50 mM Tris buffer pH 7.4 prior to sonication (Branson Ultrasonics Sonifier S-450, Branson, Danbury, CO, USA). Monochlorobimane (at a final concentration of 100 µM) and glutathione-S-transferase (1 IU/mL) were added to the samples prior to incubation at room temperature whilst protected from light for 30 minutes. Samples were analysed in a FLUOstar Omega spectrophotometer (BMG Labtech, Ortenberg, Germany) using an excitation wavelength of 360 nm and an emission wavelength of
114 Chapter 2 – Material and Methods
450 nm. Quantification was performed from a standard curve of reduced glutathione dissolved in 50 mM Tris buffer (pH 7.4) and ranging from 0 to 200 µM.
2.7 Histopathological assessment
Immunostaining was performed on gelatin double-coated slides prepared as described in the Appendix 10.3.7.
2.7.1 Perfusion of mice, organ collection and storage
Reagents used are as detailed in the Appendix 10.3.8.
Mice were anaesthetised by isoflurane in an anaesthetic chamber. When the animal was fully anaesthetised and did not respond when their toes were pinched, the animal was placed on the procedure area with an adapted anaesthetic mask. With scissors, the skin, the abdominal wal, the diaphragm and the rib cage were subsequently cut to expose the heart. A small incision was made in the right atrium. A 10 mL syringe of sterile PBS prepared in advance was inserted in the left ventricle and the PBS was injected slowly over 1 to 2 minutes. The exposed liver turned a light brown colour as long as the blood was removed and leaking from the right atrium. The organs were then collected in 1.5 mL microcentrifuge tube labelled appropriately. For paraffin embedding, peripheral organs were fixed in 10% formalin for 24 hours at 4°C then stored in 70% ethanol at 4°C until processing. The brain was extracted as follows: the head was cut from the body, the skin peeled from the skull. The skull was fixed for 24 hours at 4°C in 4% paraformaldehyde (PFA), then the skull was removed carefully and the brain itself fixed for the next 24 hours at 4°C in 4% PFA. Then the brain was stored in 30% sucrose at 4°C until processing.
115 Chapter 2 – Material and Methods
2.7.2 Sectioning of mice brains on freezing microtome
Reagents and buffers used are as detailed in the Appendix 10.3.9.
Dry ice was placed on the receptacle of a microtome (Carl Zeiss, Welwyn Garden City, UK) and a new blade (MX35 Premier Disposable Low-Profile Microtome Blades, Thermo Scientific, Loughborough, UK) put in place. An embedding matrix (Shandon embedding matrix, Thermo Scientific, Loughborough, UK) was placed on the receptacle and allowed to cool down and solidify. Before full solidification, the brain was placed with the olfactory bulbs facing up for coronal sections. The brain was then covered with a dry ice powder until frozen. The blade was then positioned to start sectioning the brain (40 μm thick slices) from the olfactory bulbs to the bulb. Each slice was carefully removed from the blade with a brush and placed in a 96-well plate with TBSAF (Tris-buffered saline, sodium azide). When sealed, the plate was stored at 4°C.
2.7.3 Stereoscopic fluorescence microscopy
GFP expression was assessed using a stereoscopic fluorescence microscope (MZ16F, Leica, Wetzlar, Germany).
2.7.4 Brain free-floating immunohistochemical staining
Reagents and buffers are detailed as in the Appendix 10.3.10.
Brain sections were placed in a petri dish containing 1x TBS (Tris-buffered saline) and a few drops of TBS-T (Tris-buffered saline-Triton) to break the surface tension, and then transferred to a 6-well plate. Each following step was separated by 3 washes of 5 minutes in TBS. Incubations were performed on a shaker (Mini gyratory rocker SSM3, Stuart, Stone, UK) at room temperature except where stated otherwise.
116 Chapter 2 – Material and Methods
Blocking of endogenous peroxidase in TBS/1% H2O2 for 30 minutes was followed by blocking of non-specific protein binding in Tris-buffered saline–Triton (TBST) /15% normal goat serum. An overnight incubation of primary antiserum diluted in TBST/10% normal serum was performed at 4°C with anti-GFP (1:10,000; Ab290, Abcam, Cambridge, UK), anti-nitrotyrosine (1:800; 06-284, Merck Millipore, Temecula, CA, USA), anti-nNOS (1:200; bs10197R, Bioss antibodies, Woburn, MA, USA), anti-iNOS (1:500; NBP1-50606, Novus Biologicals, Abingdon, UK), anti-eNOS (1:300; 610296, BD Transduction Lab), anti-GFAP (1:500; MAB3402, Merck Millipore, Temecula, CA, USA), and anti-CD68 (1:100; MCA1957, Biorad, Serotec) antibodies. The next day, a 2-hour incubation of the biotinylated antiserum (i.e. biotinylated goat anti-rabbit, anti-mouse and anti-rat secondary antibodies, Vector, Burlingame, CA, USA) diluted 1:1000 in TBST/10% normal serum, followed by a 2 hours incubation of the Vectastain (avidin-biotin) solution (avidin and biotin diluted 1:1000 in TBS, prepared 30 minutes in advance; ABC; Vector, Peterborough, UK). Detection was performed using a 0.05% 3,3’-diaminobenzidine (DAB) solution diluted in TBS prealably mixed and filtered through a 0.45 μm syringe filter 30 minutes before use. 6 µL of hydrogen peroxide (H2O2) was added to the resulting diluted DAB solution, which was then mixed and wrapped in foil before immediate use. 1 mL of this solution was added to sections, wrapped in foil to avoid photosensitivity, for 3-5 minutes and stain intensity was checked regularly during incubation. The reaction was stopped by the addition of ice cold TBS. The slides were dried at room temperature for 24 hours, dehydrated in 100% ethanol for 5 minutes, let dry, washed in Histoclear for 5 minutes, let dry and mounted with cover-slips using a blob of mounting agent DPX-new (Merck Millipore Corporation, Temecula, CA, USA) in a fume cupboard.
2.7.5 Brain free-floating immunofluorescence
As described previously (321), free-floating sections were blocked in TBST/15% normal goat serum and incubated overnight at 4°C with anti-GFP (1:4000; Ab290, Abcam, Cambridge, UK) and anti-NeuN (1:4000; Millipore, Billerica, MA, USA) diluted in TBST/10% normal goat serum. After 3 washes with TBS, samples were incubated with goat anti-rabbit Alexa 488 and goat anti-mouse Alexa 546 (1:1000; 117 Chapter 2 – Material and Methods
Invitrogen, Pailey, UK). After 3 further washes, sections were incubated with DAPI (1:2000; Invitrogen) and mounted on chrome-gelatin-coated slides and cover-slipped with Fluoromount (Southern Biotech, Birmingham, AL, USA).
2.7.6 Paraffin-embedded immunohistochemical staining for systemic organs
Reagents and buffers used are detailed as in the Appendix 10.3.1110.3.11 Reagents and buffers for immunostaining in paraffin-embedded slides.
Fixed peripheral organs were cut and paraffin-embedded in the Pathology lab, Institute of Neurology, University College London, London and stored in plastic boxes at room temperature.
Paraffin-embedded sections were dewaxed, dehydrated in an ethanol gradient (2 batches of Histoclear for 5 minutes each, 2 batches of 100% ethanol for 5 minutes each, 95% ethanol for 5 minutes, 75% ethanol for 5 minutes, distilled water for 5 seconds) before a 30-minute incubation in 1% H2O2 (diluted in methanol) to block any remaining endogenous peroxidase. An antigen retrieval step was performed using 10 mM sodium citrate buffer pH 7.4 as follows: the cuvette containing the slides was placed in a sink under running tap water for 5 minutes to remove methanol. Then, slides were placed in a plastic box containing citrate buffer and boiled for 20 minutes in a microwave. Slides were allowed to cool in the citrate buffer gradually at room temperature and prior to washing with TBS.
Non-specific binding was blocked by incubating the slides for 30 minutes with 15% normal goat serum (Vector, Burlingame, CA, USA). After 3 washes in PBS, sections were incubated with primary human anti-ASL (1:1000; Ab97370, Abcam, Cambridge, UK) and primary rabbit polyclonal anti-GFP (1:1000; Ab290, Abcam, Cambridge, UK) antibodies diluted in 10% goat serum in TBST overnight at 4 °C. Detection was performed with Polink-2 HRP Plus Rabbit Detection System for Immunohistochemistry (GBI labs, Mukilteo, WA, USA) as per manufacturer’s instructions described as follows. After 3 washes of 5 minutes with TBS, a ready-to- use rabbit antibody enhancer provided in the kit was added onto the sections and
118 Chapter 2 – Material and Methods incubated for 10 minutes in a moist chamber. Without any intermediary wash, the secondary goat anti-rabbit biotinylated IgG antiserum diluted in 10% goat serum in TBST was incubated for 45 minutes followed by 3 washes of 5 minutes with TBS. Then, the detection solution with DAB was prepared as follows: one drop of DAB chromogen was added to 1 mL of DAB substrate and protected from light. 100 μL per slice were incubated in a moist chamber for 5 minutes and colour development was monitored under microscope. The tissue slice was then rinsed with tap water, mounted and dehydrated. After dehydration in a gradient of ethanol and three washes in xylene, slices were cover-slipped with DPX-new (Merck Millipore Corporation, Temecula, CA, USA).
Initial studies did not use the Polink-2 HRP Plus Rabbit Detection System for Immunohistochemistry and used the following antibodies: green fluorescent protein immunostaining required a primary rabbit polyclonal anti-GFP antibody (Abcam 290, Abcam, Cambridge, UK) diluted 1:1,000 in 10% goat serum in TBST; a secondary goat anti-rabbit biotinylated IgG (Vector, Burlingame, CA, USA) diluted 1:200 in 10% goat serum in TBST with blocking goat serum (Vector, Burlingame, CA, USA). Immunostaining of argininosuccinate lyase was performed using a primary human anti-ASL antibody (Abcam 97370, Abcam, Cambridge, UK) diluted 1:1,000 (liver and muscle) or 1:100 (other peripheral organs) diluted in 10% goat serum in TBST; a secondary goat anti-rabbit biotinylated IgG (Vector, Burlingame, CA, USA) diluted 1:200 in 10% goat serum in TBST with blocking goat serum (Vector, Burlingame, CA, USA).
2.7.7 Nissl staining
Reagents and buffers used are detailed as in the Appendix 10.3.12.
Brain sections were fixed initially in 4% PFA for 24 hours then in 70% ethanol for 24 hours. On day 3, sections were incubated in Cresyl Violet solution (BDH, East Grinstead, West Sussex, UK) for 3 minutes followed by dehydration in an ethanol gradient (70%, 90%, 96%, 96% with acetic acid, 100%), isopropanol, and 3 washes in xylene before being cover-slipped with DPX-new (Merck Millipore Corporation, Temecula, CA, USA). 119 Chapter 2 – Material and Methods
2.7.8 Terminal Transferase-Mediated d-UTP Nick End-Labelling (TUNEL) staining
Reagents and buffers used are detailed as in the Appendix 10.3.13.
TUNEL staining was performed as described previously (322) using the Roche kit (Roche, Welwyn Garden City, Hertfordshire, UK). Briefly, sections were incubated in 3% hydrogen peroxide in methanol for 15 minutes and washed in 0.1 M phosphate buffer (PB) before incubation with a solution containing 0.1% terminal deoxytransferase (TdT), 0.15% deoxyuridine trisphosphate (dUTP) and 1% cacodylate buffer at 37°C for 2 hours. The reaction was stopped by incubating the section in TUNEL stop solution (300 mM NaCl, 300 mM sodium citrate) for 10 minutes. Sections were then washed in 3 x 0.1 M PB solution, incubated with avidin- biotinylated horse radish peroxidase (ABC; 1:100; Vector, Peterborough, UK) at room temperature for 1 hour, washed 4 times in 10 mM PB and visualised with DAB enhanced with cobalt nickel. The reaction was stopped by immersing the slide in 10 mM PB and washed twice in double-distilled (ddH2O) water.
Counting of TUNEL positive cells was performed as follows : ten random pictures per animal was observed at low magnification in the brain region studied. The number of TUNEL positive cells per animal was summed.
2.7.9 Haematoxylin and eosin (H&E) staining
Reagents and buffers used are detailed as in the Appendix 10.3.14.
This staining is the standard morphological staining for formalin-fixed paraffin- embedded sections.
The staining was performed according to the Standard Operating Procedure protocol used by the Histopathology laboratory, Great Ormond Street Hospital, London. Sections were dewaxed in xylene, hydrated down through graded alcohol solutions to water, stained in Harris Haematoxylin for 5 minutes, rinsed in tap water and
120 Chapter 2 – Material and Methods differentiated in 1% acid ethanol. Sections were then washed (5 minutes) in running tap water and counterstained in 1% eosin for 3 minutes, rinsed briefly in running tap water and dehydrated through ascending grades of alcohol. Sections were then cleared in xylene and mounted.
2.7.10 Periodic Acid-Schiff (PAS) staining
Reagents and buffers used are detailed as in the Appendix 10.3.15.
Periodic Acid-Schiff reaction enables the visualisation of glycogen granules and carbohydrates in tissues. Periodic acid oxidises carbohydrates resulting in the formation of aldehydes. The basic fuschin contained within the Schiff’s reagent (which also contains HCl and sodium metabosulfite) reacts with newly formed aldehyde groups to form a pink/red colour.
The staining was performed according to the Standard Operating Procedure protocol at the Histopathology laboratory, Great Ormond Street Hospital, London. Sections were dewaxed in xylene, hydrated down through graded alcohol solutions to water, rinsed in tap water then distilled water, incubated for 10 minutes in 0.5% periodic acid, rinsed in distilled water, stained (10 minutes) with Schiff reagent, then rinsed in distilled water. Sections were then washed for 5 minutes in running tap water as this develops the magenta staining and counterstained in 1% eosin for 1 minute, rinsed briefly in running tap water and dehydrated through ascending grades of alcohol. Sections were then cleared in xylene and mounted.
A positive control is presented in Appendix 10.3.27.
2.7.11 Masson trichrome staining
Reagents and buffers used are detailed as in the Appendix 10.3.16.
Masson trichrome staining is used to distinguish collagen structures and assess fibrotic changes.
121 Chapter 2 – Material and Methods
The staining was performed according to the Standard Operating Procedure protocol used at the Histopathology laboratory, Great Ormond Street Hospital, London. Sections were dewaxed in xylene, hydrated down through graded alcohol solutions to water, and pre-treated in 3% potassium dichromate for 1 hour at 60°C. Then, sections were stained with Celestin blue working solution for 1 minute then in Mayer’s haematoxylin for 3 minutes, washed in tap water, stained with Masson trichrome working solution for 10 minutes at 4°C, rinsed in distilled water, dehydrated through ascending grades of alcohol, cleared in xylene and mounted.
A positive control is presented in Appendix 10.3.27.
2.7.12 Oil red O staining
Reagents and buffers used are detailed as in the Appendix 10.3.17.
This staining demonstrates abnormal deposition of fat, with the dye dissolving in the lipid content present and being more soluble in the cellular lipid than in the dye solvent. Dye molecules will migrate from the solvent into the tissue lipid and stain the tissue.
The staining was performed according to the Standard Operating Procedure protocol used at the Histopathology laboratory, Great Ormond Street Hospital, London. Slides were rinsed in 70% ethanol then incubated for 2 hours in Oil Red O working solution in a coplin jar. Samples were then rinsed quickly in 70% ethanol then in water, stained for 1 minute in Mayer’s haematoxylin, washed in tap water and mounted in aqueous mounting medium.
A positive control is presented in Appendix 10.3.27.
2.7.13 Microscopy and images
Representative images were captured usiong a microscope camera (DFC420, Leica Microsystems, Milton Keynes, UK) and software (Image Analysis; Leica Microsystems).
122 Chapter 2 – Material and Methods
2.7.14 Quantification of staining
Ten random images per sample were captured with a microscope camera (DFC420; Leica Microsystems, Milton Keynes, UK) and software (Image Analysis; Leica Microsystems). Quantitative analysis was performed with thresholding analysis using the Image-Pro Premier 9.1 software (Rockville, MD, USA).
For TUNEL staining, quantification was performed manually from 10 random pictures of the same area for each sample. Staining positive cells were summed and the average was used for the analysis.
2.8 Electronic microscopy
Transmission electron microscopy is a microscopy technique based on the interactions of a beam of electrons as it passes through an ultrathin sample. An image is formed from the electrons transmitted through the specimen, is magnified and focused by an objective lens and displayed on an imaging screen. As the wavelength of the electrons is much smaller than the wavelength of light, the optimal resolution is up to 3 orders of magnitude better than that of a light microscope (323).
For visualisation, fixation and dehydration of the sample are essential so that the high energy caused by the electron beam and the vacuum inside the microscope, necessary for avoiding deflection of the electrons, do not destroy or affect it. The process of sample preparation should preserve the ultrastructure of the cells and tissues. Dehydration is followed by incubation in a transient solvent before infiltration and embedding in a liquid resin. Embedding allows the sample, now a resin block, to be cut in thin or ultra thin sections. Thin sections are stained and looked at with an optic microscope to ensure the absence of major artefact during preparation and select appropriate areas of the sample. Ultra thin sections are then cut with an ultramicrotome, collected on metal mesh grids and stained with electron dense stains, usually with heavy metals, to increase the contrast of the image before observation in the electron microscope. 123 Chapter 2 – Material and Methods
Liver samples were processed by Mrs Kerrie Verner, Electron microscopy unit, Institute of Neurology, University College London, London following recommended fixation and preparation methods for liver samples (323). I took the pictures under the supervision of Mrs Kerrie Verner.
Samples were fixed at time of collection and stored at 4°C in a solution of 3% glutaraldehyde, 0.1 M sodium cacodylate buffer pH 7.4, 5 mM CaCl2 kindly provided by Dr Alex Virasami from the Histopathology laboratory, Great Ormond Street Hospital, London.
Sample preparation consisted of an initial wash in distilled water then incubation in
1% aqueous osmium tetroxide (OsO4) for 3 hours at 4°C followed by a further wash in distilled water. The sample was then dehydrated in an increasing gradient of ethanol concentrations (70% ethanol for 10 minutes twice, 90% ethanol for 10 minutes twice, then 100% ethanol for 10 minutes 4 times). Further incubation in 1,2- epoxypropane for 15 minutes twice and then in an equal volume of 1,2- epoxypropane/araldite CY212 resin mixture for an hour was performed prior to leaving in pure araldite CY212 resin mixture overnight. Samples were then embedded in moulds using neat araldite CY212 resin mixture the next day and left for 3 days at 60°C to allow the resin to polymerise. Semi-thin sections of the sample of 950 nm width were then cut on a Reichert ultracut E ultramicrotome (Reichert- Jung, Wien, Austria) with a Diatome Histo diamond knife (Diatome, Hatfield, PA, USA). These were stained with 1% toluidine blue in 1% borax. Optic microscopy of these sections confirmed the absence of major artefact that may have occured during fixation and preparation and selected areas were targeted for further utra thin sectioning. Sections were cut on the ultramicrotome with a Diatome Ultra diamond knife (Diatome, Hatfield, PA, USA), and collected on 400 mesh copper grids (Agar Scientific, Stansted, UK). Samples were then stained with 25% uranyl acetate in methanol and Reynolds lead citrate (324). Images were captured with a Philips CM10 Transmission electron microscope (Philips Electron Optics, Eindhoven, The Netherlands) using the Megaview Olympus digital imaging system (EMSIS, Manchester, UK).
124 Chapter 2 – Material and Methods
2.9 Western Blot of murine argininosuccinate lyase
Reagents and buffers used are detailed as in the Appendix 10.3.18.
Samples were collected and immediately snap-frozen in dry ice. Samples were stored at -20°C before analysis. For processing, samples were crushed in a pre- cooled pestle and mortar with regular pouring of liquid nitrogen on the sample to ensure they remained frozen. The powder was then stored at -80°C. Cell lysis was performed by resuspending the samples in 1x NP-40 lysis buffer. The mixture was incubated with gentle rotation at 4°C for 30 minutes and after sonication the lysate was clarified by centrifugation at 14,000 g for 15 min at 4°C. The protein concentration for each sample lysate was measured using the PierceTM BCA protein assay kit (ThermoFisher Scientific, Rockford, IL, USA) according to manufacturers’ instructions and read at 570 nm in a FLUOstar Omega spectrophotometer (BMG Labtech, Ortenberg, Germany). 150 μg of protein from sample lysate were diluted in 5x Laemlli loading buffer and denatured for 5 minutes at 100°C before being loaded onto a NuPAGE Novex 4-12% Bis-Tris protein gel, 1.0 mm, 10 well (Invitrogen Thermo Scientific, Loughborough, UK). The gel was run in 1x Tris-glycine buffer for one hour (till the loading dye had reached the bottom of the gel) at 120 Volts. A protein ladder (Page Ruler; Fermentas, York, UK) was run alongside the samples, on the same gel, to evaluate the correct size of the protein bands. Proteins were then transferred to a polyvinylidene difluoride (PVDF) membrane using a wet transfer method. The PVDF membrane was first activated in methanol before the blotting sandwich was prepared in 1x Transfer Buffer in the following order: fibre pad, filter paper, PVDF membrane, gel, filter paper, fibre pad. The sandwich was then placed into the transfer device (Mini-PROTEAN® Tetra Cell Systems, Bio-Rad, Watford, UK) in order to allow the electrotransfer of proteins from the gel to the PVDF membrane. The transfer was performed in 1x Transfer Buffer at 250 mA for 1.5 hours, with an ice block to prevent overheating and temperature fluctuations during protein transfer. During the detection process the membrane was first blocked in 10 % dried non-fat milk powder in PBST, on a roller, for one hour at room temperature. The membrane was then incubated with the primary antibody diluted in 5 % milk solution for one hour at room temperature. The membrane was then washed three times on a shaker (200 rpm), for 10 minutes each time. After the third wash, the membrane was incubated
125 Chapter 2 – Material and Methods with secondary antibody diluted in 5% milk solution and incubated at room temperature for one hour. ASL was identified using a primary human anti- argininosuccinate lyase antibody (Abcam 97370, Abcam, Cambridge, UK) diluted 1:1,000 and the secondary polyclonal goat anti-rabbit HRP IgG (Dako, Glostrup, Denmark) diluted 1:2,000. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) protein was identified using the primary mouse anti-GAPDH antibody (Abcam 9484, Abcam, Cambridge, UK) diluted 1:10,000 and the secondary polyclonal rabbit anti- mouse HRP IgG (Dako, Glostrup, Denmark) diluted 1:15,000. The secondary antibodies were all conjugated to Horse Radish Peroxidase (HRP), which allowed detection once the membrane had been incubated for 2 minutes with the ECL+ western blotting detection system reagents (GE Healthcare) at a ratio of 1:40 for reagents A and B, respectively. Exposure of the autoradiography film (Kodak) was performed for variable periods of time depending on the signal strength of the protein of interest after subsequent development in an X-ray film developer (Konica Minolta). The band intensities were scanned and quantified using the densitometry function of Fiji (Fiji is just Image J) software (325).
2.10 Green fluorescent protein enzyme linked immunosorbent assay
Reagents and buffers used are detailed as in the Appendix 10.3.19.
Protein extraction was performed using a β-galactosidase enzyme linked immunosorbent assay (ELISA) (Roche, Mannheim, Germany). Lysis buffer was added to cover the sample and the sample was frozen and thawed a couple of times prior to grinding with a homogeneizer (Tissue Master 125, OMNI International, Kennesaw, GA, USA). The sample was then centrifuged for 2 minutes at 3,500 g and the supernatant collected. The protein concentration for each sample was measured using the PierceTM BCA protein assay kit (ThermoFisher Scientific, Rockford, IL, USA) according to manufacturers’ instructions and read at 570 nm in a FLUOstar Omega spectrophotometer (BMG Labtech, Ortenberg, Germany). Each step of the ELISA protocol was separated by 3 washes of a wash buffer (0.05% Tween20 in PBS). A monoclonal anti-GFP antibody (diluted 1:10,000; ab1218, Abcam, 126 Chapter 2 – Material and Methods
Cambridge, UK) was added to the plate overnight at 4°C followed by blocking with 1% bovine serum albumin in PBS for one hour at 37°C. Analysis was done in a sealed 96-well plate and all samples were measured in duplicate. GFP standards were serially diluted in wash buffer and incubated alongside a buffer blank and the samples for one hour at 37°C. An anti-GFP biotin-conjugated secondary antibody (diluted 1:5,000; Ab6658, Abcam, Cambridge, UK) followed by a streptavidin- horseradish peroxidase conjugate (diluted 1:20,000; SNN2004, Invitrogen, Camarillo, CA, USA) were added to the samples. Both were incubated for one hour at 37°C, successively. Tetramethylbenzidine (TMB) was then added for 2 minutes at room temperature before the reaction was stopped by the addition of 2.5 M H2SO4. The plate was read at 450 nm within 30 minutes in a FLUOstar Omega spectrophotometer (BMG Labtech, Ortenberg, Germany).
2.11 Quantitative PCR
Reagents and buffers used are detailed as in the Appendix 10.3.20.
Liver samples were frozen and stored at -20°C before DNA extraction with the DNeasy blood and tissue kit (QIAgen, Crawley, UK) according to manufacturer’s instructions.
In Chapter 6, specific detection of the WPRE sequence was performed using probe- based quantitative PCR and the following set of primers:
5'-FAM-CTTCTGCTACGTCCCTTCGGCCCT-3'-TAMRA (WPRE mut 6 PCR (Genethon) Probe); 5’-TGGATTCTGCGCGGGA-3’ (sense); 5’ GAAGGAAGGTCCGCTGGATT 3’ (antisense). The mouse titin gene was used as a housekeeping gene to normalise the data and was amplified using a set of primers specific for exon 5 of the gene: 5'-FAM-TGCACGGAAGCGTCTCGTCTCAGTC- 3'TAMRA (probe); 5’-AAAACGAGCAGTGACGTGAGC-3’ (sense); R 5’- TTCAGTCATGCTGCTAGCGC-3’ (antisense) (326). The Platinum® Quantitative PCR SuperMix was prepared following manufacturer’s instructions (Table 8) in a PCR workstation. Reactions were performed in triplicate for samples and serial 127 Chapter 2 – Material and Methods dilutions of plasmid standards containing the WPRE and titin. Samples were set up in 0.2 mL MicroAmp reaction tubes (Applied biosystems, Foster city, CA, USA) and pulsed in a centrifuged before running the reaction on an ABI 7000 SDS Thermal cycler (Applied biosystems, Foster city, CA, USA). Standard cycling conditions used included an initial step at 50°C for 2 minutes to activate the uracil N-glycosylase (UNG) enzyme included in the mastermix to prevent carry-over contamination, a Hot start activation step of 95°C for 10 minutes followed by 40 cycles of denaturation at 95°C for 15 seconds, annealing primers at 72°C for 1 second and extension at 60°C for 1 minute without a final extension step. Results were analysed with 7000 System SDS software using the automatic function to set baseline and cycle threshold (Ct). Ct values were used to calculate the amount of vector genomes (vg) and cells (on the assumption that there were 2 copies of titin per cell) from the standard curve. Values of WPRE vector genomes were compared to titin to calculate the transduction per cell.
Reagent Volume (μL) Final concentration (nM)
Universal Mastermix 12.5 1X
Forward primer 0.225 900
Reverse primer 0.225 900
Probe 0.05 200
Nuclease-free water 7
DNA sample 5
Final volume 25
Table 8. Mastermix used for quantitative PCR
In Chapter 7, a detection using a double stranded DNA binding dye (SYBR green) was used. WPRE sequence, amplified using 5’-TTCCGGGACTTTCGCTTTCC-3’ (sense) and 5’-CGACAACACCACGGAATTG-3’ (antisense), was detected and
128 Chapter 2 – Material and Methods normalised against glyceraldehyde 3-phosphate dehydrogenase, amplified using 5’- ACGGCAAATTCAACGGCAC-3’ (sense) and 5’-TAGTGGGGTCTCGCTCCTGG-3’ (antisense). Reactions (final volume 25 μL) were carried out with 5 μL of sample, 2.5 μM of each primer, and SYBR green master mix using the Quantitect SYBR Green PCR Kit (QIAgen, Crawley, UK). The amplification conditions were 95°C for 10 minutes followed by 40 cycles of 95°C for 15 seconds, 60°C for 1 minute, 72°C for 30 seconds. Data were processed with StepOneTM softwarev2.3 (ThermoFisher Scientific, Rockford, IL, USA).
2.12 Cloning of the AAV vector construct
Reagents and buffers used are detailed as in the Appendix 10.3.21.
2.12.1 Plasmids
A plasmid containing the elongation factor 1α short (EFS) promoter, a modified SV40 small t antigen intron upstream of a codon optimised human VPS33B coding DNA and a simian virus 40 late polyA (SV40 LpA) downstream (EFS-SV40-hVPS33Bco- WPRE-pA) flanked by AAV2 ITRs (recipient plasmid) was a kind gift from Joanna Hanley, post-doctoral researcher in Prof Paul Gissen’s lab. The VPS33B gene was flanked by Nhe1 and EcoRV restriction sites up- and downstream, respectively.
The murine non codon-optimised Asl gene (mAsl) (Transcript Asl-0003, http://vega.sanger.ac.uk/Mus_musculus/Transcript/Sequence_cDNA?g=OTTMUSG0 0000033763;r=5:130011504-130025019;t=OTTMUST00000085369) inserted in a pCMV-SPORT6 gateway was bought from Thermo Scientific (Loughborough, UK) and provided as an E. coli culture in LB Broth medium with 8% glycerol and ampicillin 100 mg/L.
129 Chapter 2 – Material and Methods
2.12.2 DNA extraction
2 µL of the E. coli culture containing mAsl plasmid was cultured in autoclaved LB Broth medium with ampicillin (100 mg/L) and placed in a shaking incubator (Orbital incubator SI500, Stuart, Stone, UK) at 37°C, 250 rpm for 24 hours. DNA was extracted using the QIAgen Spin miniprep kit (QIAgen, Crawley, UK) as per manufacturer’s instructions, and quantified by spectrophotometry on a FLUOstar Omega (BMG Labtech, Ortenberg, Germany).
2.12.3 Amplification of mAsl
A set of primers was designed to amplify the mAsl insert (Figure 31): Forward primer: 5’-CTTGCTAGCGCCACCATGGCATCAGAGA-3’ containing the Nhe1 restriction site and the GCC Kozak sequence, which plays a major role in the initiation of translation for eukaryotic messenger RNA and the 16 first base pairs of mAsl; Reverse primer: 5’-CTAAGATATCCTAGGGCTCCTGGGCTTGC-3’ containing the EcoRV restriction site and the complementary sequence of the last 19 base pairs of mAsl.
Nhe1,restric9on,site,, Kozak,sequence,, 16,first,bp,of,mAsl,,
5’@CTTGCTAGCGCCACCATGGCATCAGAGA@3’,,
5’@CTAAGATATCCTAGGGCTCCTGGGCTTGC@3’,,
EcoRV,restric9on,site,, complementary,sequence,of,the,last,19,bp,of,mAsl,,
Figure 31. Design of the primers used for mAsl insert amplification before cloning.
A set of 20 base pairs-long primers was designed to amplify the mAsl gene contained in the gateway vector.
130 Chapter 2 – Material and Methods
The standard reaction mix for a 25 µL PCR using Q5 High Fidelity Polymerase Kit contained 0.5 μL (0.25 units) of Q5 High-Fidelity DNA Polymerase, 5 μL of 5x Q5 buffer (for a final MgCl2 concentration of 2 mM in the standard reaction mix), 0.5 μL of 10 mM dNTPs, 1.25 μL of each primer diluted to 10 μM, 1 ng of DNA template (prepared as described in Section 2.12.2). Standard cycling conditions used included an initial activation step of 98°C for 30 seconds followed by 30 cycles of denaturation at 98°C for 5 seconds, annealing of primers at 62°C for 20 seconds and an extension at 72°C for 45 seconds. A final extension step was performed at 72°C for 2 minutes.
2.12.4 Agarose gel electrophoresis
The amplified mAsl plasmid (see Section 2.12.3) was run on a 1% agarose gel (see Section 2.4.4). The expected product of 1395 base pairs was visualised on a benchtop ultraviolet transilluminator BioDoct-It Imaging system (UVP, Upland, CA, USA) with UVO TS software (UVP, Upland, CA, USA). The DNA band containing the mAsl product was excised from the gel using a blade whilst viewing the gel under direct visualisation.
2.12.5 DNA gel extraction
The gel slice containing the amplified mAsl was extracted with the QIAquick Gel Extraction Kit (QIAgen, Crawley, UK) as per manufacturer’s instructions.
2.12.6 Restriction enzyme digestion
Restriction enzymes are used to cut plasmid DNA at specific sites allowing cloning experiments.
The gel-purified mAsl (see Section 2.12.5) and the plasmid AAV2 ITR-EFS-SV40- hVPS33Bco-WPRE-pA-AAV2 ITR were digested with the restriction enzyme EcoRV
131 Chapter 2 – Material and Methods and then Nhe1 for 2 hours at 37°C in both cases. Each reaction was performed in a final volume of 40 μL and contained 1 μg of DNA template, 1 unit of restriction enzyme, 4 μL of 10x NE buffer 3.1 and nuclease free water. After digestion, the products were run on a 1% agarose gel for gel purification before ligation. After confirmation of the band size, the EcoRV-digested mAsl (1395 bp) and the recipient plasmid without VPS33B (5356 bp) were extracted from the gel as described in Section 2.12.5.
2.12.7 Alkaline phosphatase treatment
After Nhe1 digestion, an additional phosphatase step was performed to avoid self- ligation of the recipient plasmid. This removed the 5’ phosphate group from the AAV backbone to prevent re-circularisation. This required 10 minutes incubation at 37°C then inactivation for 5 minutes at 75°C using Fast thermosensitive alkaline phosphatase in a final volume of 40 μL containing 1 μg of DNA template, 1 unit of alkaline phosphatase, 4 μL of 10x Fast AP buffer and nuclease free water.
2.12.8 Ligation
Ligation of the digested, de-phosphorylated recipient plasmid with the mAsl insert was performed using an approximate 3:1 ratio based on the band intensity on a 1% agarose gel. The reaction was incubated overnight at 4°C prior to inactivation. The reaction contained T4 DNA ligase (800 units) in a final volume of 20 μL containing 1 μL of the recipient plasmid, 3 μL of the mAsl insert, 2 μL of 10x T4 DNA ligase buffer and nuclease free water. A second mix with the recipient plasmid alone was used as negative control.
2.12.9 Bacterial transformation and expansion of colonies
DH5α competent E. coli cells (Thermo Scientific, Loughborough, UK) were thawed on ice and 50 µL mixed with either 5 μL of the recipient plasmid alone (used as a 132 Chapter 2 – Material and Methods negative control), or 5 μL of the recipient ligated plasmid and mAsl insert, or 1 μL of pUC19 plasmid producing ampicillin-resistant colonies (used as a positive control). After 30 minutes of incubation on ice, the cells were heat shocked at 42°C for 45 seconds and then were replaced on ice for 5 minutes. 250 μL of SOC outgrowth medium was added to each sample and all were incubated for one hour in a shaking incubator at 37°C at 250 rpm. 100 μµL of cells were then spread on Agar plates, allowed to dry, (made as detailed in Appendix 10.2.22) and incubated overnight at 37°C. Successful ligations were checked the day after by looking at colony formation and comparing to controls. 10 colonies were carefully picked using pipette tips and each tip placed in 5 mL of LB medium and incubated at 37°C for 24 hours shaking at 250 rpm.
2.12.10 DNA extraction, digestion and sequencing
For each clone (see Section 2.12.9), DNA was extracted from 2 mL of the mixture using the QIAgen Spin Miniprep Kit (QIAgen, Crawley, UK). The remaining 3 mL were stored at 4°C.
A digestion using the restriction enzyme Bgl1 was performed to check for homologous recombination. This required 2 hours incubation at 37°C then inactivation for 20 minutes at 65°C of a 10 μL reaction mix containing 5 μL of sample with DNA template, 1 μL of Bgl1 restriction enzyme, 1 μL of buffer 3.1 and 3 μL nuclease free water. Then the product of the reaction was run on a 1% agarose gel. The predicted bands of 619, 1130, 1164, 1418 and 2345 base pairs were appropriatey individualised in all clones (Figure 32).
133 Chapter 2 – Material and Methods
Figure 32. Bgl1 digestion of the ligated plasmid
As the digestion did show the predicted pattern of digestion thus making the possibility of large homologous recombination highly unlikely, the ligated plasmid was sent for sequencing to confirm the absence of mutations with the expected DNA template. Primers were designed to flank the mAsl insert (Forward: CAGAACACAGGTGTCGTGA; Reverse: GCAGCGTATCCACATAGC) and the purified plasmid was sent for sequencing to Source Bioscience (Nottingham, UK) to verify ligation. The sequencing data was aligned using Clone Manager Basic 9, Sci- Ed software (Morrisville, NC, USA) and the sequence of the construct, AAV2 ITR- EFS-SV40-mAsl-WPRE-pA-AAV2 ITR confirmed.
134 Chapter 2 – Material and Methods
2.13 AAV vector production
Reagents and buffers used are detailed as in the Appendix 10.3.23.
The steps involved in AAV vector production are summarised in Figure 33.
Figure 33. Steps for AAV vector production.
The triple transfection in HEK (Human embryionic kidney) 293T cells requires 3 plasmids, one encoding the transgene cassette flanked by the AAV2 inverted terminal repeats (ITRs), one encoding the rep and cap genes and one encoding the helper sequences (E2A, E4, and virus-associated RNA). After polyethylenimine (PEI) mediated transfection, AAV vectors were collected from the media and scrapped cells and purified via affinity chromatography. After concentration, a titration was performed by direct visualisation of DNA via alkaline gel. A Coomassie gel was performed to confirm the purity of the produced vector.
135 Chapter 2 – Material and Methods
2.13.1 Plasmid production
Transformation of DH5α competent E. coli cells was performed as described in Section 2.12.9 for each of the 3 plasmids required for the triple transfection and the bacteria were cultured in LB Broth medium in conical flasks in a shaking incubator (200 rpm; 37°C) for 48 hours. DNA was extracted using the QIAgen Spin Miniprep Kit (QIAgen, Crawley, UK) according to manufacturer’s instructions. DNA concentration was measured by looking at the 260:280 nm ratio in a FLUOstar Omega spectrophotometer (BMG Labtech, Ortenberg, Germany).
2.13.2 Triple transfection
Packaging HEK293T (Human Embryonic Kidney) cells were used to produce the AAV vector with the triple transfection method based on the transfection of 3 different plasmids carrying the necessary enzymatic activities to allow the synthesis of a replicative-defective virion i.e. a plasmid containing the transgene, a plasmid containing the AAV8 rep and cap genes provided by Penn Vector Core, University of Pennsylvania (Philadelphia, PA, USA) and a helper plasmid containing the adenoviral genes (E2A, E4, and virus-associated RNA) required for AAV replication and packaging provided by the Gene Transfer Vector Core, Harvard University (Boston, MA, USA) (327).
All the following steps were performed in a Class I laminar flow hood.
293T cells were cultured in 40 x 15 cm flasks at 37°C, 5% CO2 in Dulbecco’s Modified Eagle’s Medium (DMEM) + 10% fetal calf serum (FCS) and passaged using trypsin EDTA every 2 to 3 days once they had reached >70% confluence. Passage 3 onwards was considered for transfection.
For a transfection of 40 dishes of 15 cm diameter; i) On day 0,6 mL of 0.1% polyethylenimine (PEI) adjusted to pH 7.2 was mixed with 54 mL of DMEM. 300 μg of the transgene plasmid, 300 μg of the AAV8 rep/cap plasmid and 900 μg of the helper plasmid were added to 62 mL of DMEM and filtered
136 Chapter 2 – Material and Methods at 0.22 μm. The 2 solutions were mixed together and left at room temperature for 20 minutes. 3 mL of this solution were then added to each dish. ii) On day 1, DMEM + 10% FCS was replaced by 15 mL of DMEM + 2% FCS per dish. iii) On day 3 after 72 hours of incubation, the cells were scrapped off and collected into 50 mL falcons. After centrifugation for 10 minutes at 1,300 g, the supernatant was collected and stored at 4°C. The pellet was resuspended in PBS filtered at 0.22 μm and re-centrifugated for 10 minutes at 1,300 g. The pellet was then resuspended in a final volume of 40 mL of TD buffer and stored at -80°C. iv) The day before purification,
- the pellet was freeze-thawed 5 times to lyse the cells and release the viral particles. Cells were then incubated in 0.5% deoxycholic acid with benzonase (50 IU/mL) to digest all remaining plasmids. The mix was spun at 3,800 g for 30 minutes at 18°C. The supernatant was filtered at 0.22 μm and diluted 1:3 in filtered PBS.
- the supernatant was spun at 3,800 g for 30 minutes at 4°C. 25 IU/mL of benzonase was added to the supernatant. Then a 100 m M MgSO4 solution was added at a 1:1000 volume ratio. After 30 minutes incubation at 37°C, the solution was filtered at 0.22 μm and store in the fridge.
2.13.3 Virus purification by affinity chromatography
AAV vector particles were purified by protein affinity interaction with sepharose using different pH buffer solutions to trap or release the AAV particles (328, 329). AKTAprime FPLC plus chromatography system (GE Healthcare Life Sciences UK Ltd, Buckinghamshire, UK) with Primeview 5.0 software was connected to either a prepacked HiTrap AVB Sepharose High Performance column (GE Healthcare UK Ltd, Buckinghamshire, UK) or a Proteus 1 mL FliQ column (Generon, Maidenhead, UK). This column is designed specifically with a high affinity for different AAV serotypes, essentially binding the AVB resin via a specific epitope SPAKFA at residues 665-670 (330).
137 Chapter 2 – Material and Methods
For standard purification, the supernatant and all buffers were initially degassed with a ME4R Vaccubrand (Diaphragm vacuumpup, Wertheim, Germany) for 30 minutes. Then lines A and B were pre-washed with filtered 20% ethanol then line A with filtered PBS (pH 7.5) and line B with filtered 50 mM glycine with pH adjusted at 2.7. After attaching the column, the system was equilibrated with PBS for 20 minutes until a flat baseline was achieved at 260 and 280 nm. The supernatant was filtered through the system and the system washed with PBS briefly to remove unbound protein and DNA. Elution with 50 mM glycine was then performed with visualisation of a sharp absorbance peak indicating the release of the virus (Figure 34). The virus was collected in 1 mL fractions into tubes with 1 M Tris adjusted to pH 8.8 to neutralise glycine. The same process was performed for the pellet.
To remove glycine and restore a neutral pH, a dialysis was performed overnight at 4°C in PBS with a 10 kDa cut-off dialysis cassette (Slide-A-Lyzer Thermo Scientific, Loughborough, UK).
The vector was then concentrated using Amicon Ultra 15_100kDa MWCO (Merck Millipore, Darmstadt, Germany) after priming with PBS. The purified virus preparations were centrifuged at 4,800 g for 5 minutes and the flow-through discarded. The concentrated virus was then collected in a 1.5 mL microcentrifuge tube. The membranes of the amicon filter were washed several times with 400 m L PBS to collect the virus.
The virus was then centrifuged in a 1.5 mL microcentrifuge tube containing a 0.22 μm filter with a cellulose acetate membrane (Corning Costar Spin-X, Sigma) at 16,000 g for 3 minutes. Under a laminar flow hood, the filter was removed and the sterilised virus was stored at 4°C.
138 Chapter 2 – Material and Methods
A
Ini/al wash and Filtra/on of the solu/on equilibra/on containing the virus Virus elu/on (supernatant or pellet )
B
Peak of virus Figure 34. Purification by affinity chromatography
Chromatograms from virus purification by affinity chromatography with ÄKTAprime plus (A) Full run with wash and equilibration of the column, filtration of the solution containing the virus, then elution; (B) Zoom from screen A focussed on the viral peak eluted.
139 Chapter 2 – Material and Methods
2.13.4 Vector titration
Titration of the AAV vector was performed by electrophoresis on an alkaline gel (258).
An alkaline electrophoresis buffer (1x) and an alkaline gel made with 0.8% agarose and an alkaline electrophoresis buffer (1x) were prepared. Two hours were allowed for the gel to set. Three viral samples containing 5, 10 and 25 μL of virus made up to a final volume of 25 μL using PBS were mixed with 8.5 μL of alkaline sample loading buffer at room temperature, then cooled on ice before loading on the gel. The alkaline loading buffer releases the DNA and there is no need to heat to denaturate the virus. The gel, the tank and the alkaline buffer were transferred to a cold room (4°C) and the samples loaded onto the gel with 5 μL of Hyperladder I added to wells on either side of the samples. The gel was run at 20 Volts overnight. The following day, the gel was placed in 3 gel volumes of 0.1 M Tris adjusted to pH 8.0 for one hour on a rotating incubator and then placed in 1 gel volume of 4x Gelred solution in 0.1 M NaCl for 2 hours and protected from light. The gel was rinsed twice in tap water and an image captured using a benchtop ultraviolet transilluminator BioDoct-It Imaging system (UVP, Upland, CA, USA) with UVO TS software (UVP, Upland, CA, USA) (Figure 35).
) Hyperladder) Nega8ve 5μL) 10μL) 25μL) control)
10)kb)
6)kb)
4)kb) 3)kb) 2.5)kb) 2944bp) 2)kb)
Figure 35. AAV vector titration by alkaline gel
The purified virus is run overnight on an alkaline gel at 4°C. Three different volumes are loaded. The concentration of full capsids is estimated by the DNA concentration.
140 Chapter 2 – Material and Methods
Hyperladder I shows a pattern of bands corresponding to different amount of DNA in ng. Band intensities from the Hyperladder and samples were analysed with the densitometry function of FiJi Image J software (325) and plotted in an excel spreadsheet to produce a standard curve and calculate the amount of DNA in ng for each sample. A conversion of amount of DNA (ng) into vector genome (vg) copies was performed using the www.endmemo.com website. An average of the 3 values obtained from the 3 vector samples was used to calculate the final titre in vg/μL.
2.13.5 Analysis of capsid proteins with Coomassie gel
This method allows a relative quantification of the AAV vector by band intensity of the capsid proteins VP1, VP2 and VP3.
20 μL of concentrated virus were mixed with an equal volume of Laemlli loading buffer and heated at 95°C for 5 minutes. 5 μL of Pageruler plus prestained protein ladder, 30 μL of both viral sample and a control AAV sample were loaded in a NuPAGE Novex 4-12% Bis-Tris protein gel, 1.0 mm, 10 well (Invitrogen Thermo Scientific, Loughborough, UK) and were run in 1x NuPAGE MOPS SDS running buffer diluted with ddH2O at 150 Volts for 30 minutes. At the end of the run, the gel was removed from the plastic frame and placed in a plastic box where the Coomassie blue dye had been poured. This was left for 2 hours on a platform rotating shaker then placed in a destain solution for 2 hours or until the background staining had been cleared. The gel was then imaged using a transilluminator BioDoct-It Imaging system (UVP, Upland, CA, USA) with UVO TS software (UVP, Upland, CA, USA) (Figure 36).
141 Chapter 2 – Material and Methods
Protein( ladder((
100(kDa( VP1(((90(kDa(
70(kDa( VP2(((72(kDa(
55(kDa( VP3(((60(kDa(
35(kDa(
Figure 36. AAV vector purity assessed by Coomassie gel
Identification of the 3 capsid proteins VP1, VP2, VP3 of the adeno-associated virus in line with the theoretical ratio 1:1:10.
2.14 Quantification of urea cycle amino acids in dried blood spots and ASL enzyme activity in tissue samples
2.14.1 Mass spectrometer parameters
Detection of metabolites was performed using a Micromass Quattro micro API (Micromass UK Ltd, Cheshire, UK) tandem mass spectrometer using multiple reaction monitoring in positive ion mode. The temperature of the source and for desolvation were 120°C and 350°C, respectively. The capillary and cone voltages were 3.7 kVolts and 35 Volts respectively. The cone gas flow was 50 L/h and the syringe pump flow set at 30 µL/min. The mass spectrometer vacuum was 4.3x10-3 mbar. The multiplier and extractor voltages were 650 Volts and 1 Volt, respectively. Dry and highly purified argon was used as the collision cell gas (331). Data were analysed using Masslynx 4.1 software (Micromass UK Ltd, Cheshire, UK).
Multiple reaction monitoring using the parameters summarised in Table 9 was used to identify the amino acids of interest.
142 Chapter 2 – Material and Methods
Data were analysed using Targetlynx (Masslynx 4.1 software; Micromass UK Ltd, Cheshire, UK).
Retention Precursor Cone Collision Product ion Analyte time ion voltage energy (m/z) (minutes) (m/z) (Volts) (Volts) L-Ornithine 5.73 132.93 69.94 17 17 L-Ornithine-d7 5.73 139.98 76.99 16 15 L-Arginine 5.79 175.10 69.94 28 20 L-Arginine-13C6 5.79 181.10 73.94 28 20 ASA anhydrides 6.07 273.12 69.87 40 29 Argininosuccinate 6.26 291.02 69.87 40 29 L-Glutamine 6.51 147.06 83.94 18 12 L-Glutamine-13C2 6.51 149.06 84.94 18 12 L-Glutamate 6.57 148.02 83.82 20 11 L-Glutamic acid-d5 6.57 153.06 88.86 20 11 L-Citrulline 6.60 176.10 69.94 30 18 L-Citrulline-d7 6.60 183.15 76.99 30 18
Table 9. Parameters used for the detection and quantification of amino acids by multiple reaction monitoring
2.14.2 Quantification of analytes using stable isotopes
Reagents and buffers used are as detailed in the Appendix 10.3.24.
Stable isotopes were added to the samples to enable quantification of the compounds of interest. A stable isotope is a natural isotope of a chemical element, which is non-radioactive and therefore is chemically stable and safe. A stable isotope is an atom whose nucleus contains the same number of protons but a different number of neutrons. For example, the atom hydrogen 1H contains one proton and has another natural stable isotope, deuterium 2D containing one proton and one neutron. Similarly carbon can occur as 12C, which is composed of 6 protons and 6 neutrons, or as a natural less abundant stable isotope 13C containing 6 protons and 7 neutrons. This difference in number of neutrons does not modify the biochemical property or biological activity of the compound but results in a slightly higher
143 Chapter 2 – Material and Methods molecular mass and m/z compared to that of the more common natural biochemical. These less abundant stable isotopes can be added to samples in a known amount, and can be detected by the mass spectrometer. For each analyte, a stable isotope was chosen (L-Arginine-13C, L-Citrulline-2,3,3,4,4,5,5-d7, L-Glutamic-2,3,3,4,4-d5 Acid, L-Ornithine-2,3,3,4,4,5,5-d7 HCl, L-Glutamine-1,2-13C2) and used as an internal standard for quantification. L-Arginine-13C and L-Glutamine-1,2-13C2 were sourced from CK isotopes, Ibstock, UK. L-Citrulline-2,3,3,4,4,5,5-d7, L-Glutamic- 2,3,3,4,4-d5 Acid, L-Ornithine-2,3,3,4,4,5,5-d7 HCl were bought from CDN istotopes, Pointe-Claire, Quebec, Canada. For argininosuccinic acid, no stable isotope from this molecule is commercially available and L-Citrulline-2,3,3,4,4,5,5-d7 was used as internal standard.
Stock solutions of 10 mM and 1 mM were made for each stable isotope by dilution in
Milli-Q Ultra pure water (MQ-H2O) (Millipore, Brussels, Belgium). Similarly stock solutions at identical concentrations were made for the urea cycle amino acids (L- arginine, L-citrulline, L-Glutamine, L-Glutamic Acid, L-Ornithine and L- Argininosuccinic acid) bought from Sigma, St Louis, MO, USA. These stock solutions were used for the optimisation of the mass spectrometry and liquid chromatography parameters described in Sections 4.2 and 4.3.
2.14.3 Derivatisation of amino acids with 9-fluorenylmethyl chloroformate
Amino acids are highly polar molecules. A pre-column derivatisation step, which tags the analyte with a reactive group, can increase the sensitivity (332) of detection of compounds of interest. 9-Fluorenylmethyl chloroformate is a molecule commonly used for the derivatisation of amino acids and binds to the amino group of the molecule (333). Derivatisation of amino acids with 9-fluorenylmethyl chloroformate was initially investigated to see if this would increase sensitivity as published previously (334). Derivatisation was performed by mixing 20 μL of 10 mM solution of each amino acid of interest with 250 μL of 5.8 mM 9-fluorenylmethyl chloroformate
(FMOC) in acetone, 250 μL of 0.1 M borate buffer, pH 10.4, 100 μL of MQ-H2O. The solution was left for 10 minutes at room temperature and applied to an Isolute C18 144 Chapter 2 – Material and Methods column (500 mg; reservoir 6 mL) (Biotage, Upssala, Sweden) that had been primed with 3 mL of 50% acetonitrile and then 3 mL of ddH2O. An initial wash with 3 mL of
MQ-H2O was applied to remove salts. Then 3 mL of an increasing percentage of organic solution was sequentially applied to the column (20% acetonitrile, 40% acetonitrile, 60% acetonitrile, 80% acetonitrile, 100% acetonitrile) and each eluant was collected individually.
2.14.4 Optimised high-performance liquid chromatography parameters
The stationary phase selected was a general purpose reversed-phase hybrid-based
C18 column, which is composed of octadecyl C18 aliphatic carbon chains attached to a silica support (Waters XTerra® RP18, 5 μm 3.9 x 150 mm column (Waters, Milford,
MA, USA) fitted with a Waters XTerra® RP18, 5 μm 3.9 x 20 mm guard column). The silica support contains both inorganic (silica) and organic (methylsiloxane) components, which have complementary binding properties. 18C silica has high efficiency and mechanical strength in binding but has a limited pH range (pH range between 2-8) and is chemically unstable. The organic polymer (methylsiloxane) is suitable over a wide range of pHs (pH range between 1 to 12) and is chemically more stable.
A Waters Alliance 2795 LC system (Waters, Midford, USA) high-performance liquid chromatography machine (HPLC) was used to separate the analytes of interest. Two mobile phases were used with the gradient elution profile presented in Table 10.
145 Chapter 2 – Material and Methods
3.7% Methanol Flow Time (min) Acetic Acid (%) (mL/min) (%) 0 0 100 0.2 1 15 85 0.2 6 15 85 0.2 8 25 75 0.2 9 95 5 0.2 15 95 5 0.2 16 0 100 0.2 17 0 100 0.5 25 0 100 0.5
Table 10. Mobile phase gradient elution profile used for high-performance liquid chromatography
2.14.5 Analysis of dried blood spot amino acids
Blood sampling was performed as described in Section 2.6.1. A 3.2 mm diameter punch from a dried blood spot was added to a 1.5 mL Eppendorf containing 90 μL of methanol and 10 μL of a mixture of internal standards containing 2 nM of L-Arginine- 13C, L-Citrulline-2,3,3,4,4,5,5-d7, L-Glutamic-2,3,3,4,4-d5 Acid, and 0.66 nM L- Glutamine-1,2-13C2 and L-Ornithine-2,3,3,4,4,5,5-d7. The sample was sonicated in a sonication bath for 15 minutes before adding 400 μL of methanol resulting in a 4:1 volume dilution prior to centrifugation at 16,000 g for 5 minutes. The supernatant was removed prior to injection of a volume of 3 µL into the liquid chromatography-tandem mass spectrometry system.
2.14.6 Protein extraction of tissues for ASL enzymatic assay
Reagents and buffers used are as detailed in the Appendix 10.2.26.
Liver and brain samples were snap-frozen in dry ice at time of collection after perfusion of the animal to remove the residual blood in tissues as detailed in Section 146 Chapter 2 – Material and Methods
2.7.1. As detailed in Section 2.10, proteins were extracted in lysis buffer (50 mM Tris, 150 mM NaCl, 1% Triton adjusted to pH 7.5), freeze thawed once before homogenising and subsequently centrifugated at 16,000 g for 20 minutes at 4°C. Protein quantification was performed using the bicinchoninic acid (BCA) assay using the PierceTM BCA protein assay kit (ThermoFisher Scientific, Rockford, IL, USA).
2.14.7 Optimised protocol for analysis of ASL activity
All samples were analysed in duplicates.
20 µg of liver lysate was incubated in a buffer solution containing 10 mM Tris, 0.5 mM argininosuccinic acid, 0.02 nM D7-citrulline, 0.02 nM 13C-arginine, pH 7.45, in a final volume of 100 µL. The reaction was stopped by adding 400 µL of methanol either immediately (time point 0) or for 2 hours at 37°C (time point 1). Samples were then centrifuged immediately at 16,000 g for 2 minutes and the supernatant analysed by LC-MS/MS. Enzyme activity was expressed as argininosuccinic acid and anhydride clearance per amount of protein per minute and was calculated by substracting the amount of argininosuccinic acid and anhydrides present at time points 0 from time point 1.
80 μg of brain lysate was added in a buffer solution of 10 mM Tris, 30 µM argininosuccinic acid, 0.02 nM D7-citrulline, 0.02 nM 13C-arginine in a final volume of 100 μL. Samples were then processed as described for liver samples.
2.15 Statistical analysis
Statistical analysis of data was performed using GraphPad Prism 5.0 software, San Diego, CA, USA and approved by Dr Deborah A. Ridout, statistician at the Population, Policy and Practice Programme, Great Ormond Street Institute of Child Health, University College London, London. p values < 0.05 were considered statistically significant.
147 Chapter 2 – Material and Methods
The statistical analysis described in Chapter 3 used Fisher’s exact t-test for investigating the association between categorical data and the patient groups (www.vassarstats.net). Continuous variables between groups were compared using the two-tailed unpaired Student’s t-test or one-way ANOVA with Bonferroni post-test for pairwise comparison. Kaplan-Meier survival curves were compared with the log- rank test. Patients 10, 18 and 52, who died during the first month of life, and patient 53, for whom very limited clinical information was available, were excluded from long- term analysis. For patients lost during follow-up (n=6), the assessments at their last follow-up visits were used for analysis. Figures show mean ± 95% confidence interval.
For Chapters 4 to 7, statistical analysis used two-tailed unpaired Student’s t-test for pairwise comparisons, or one-way ANOVA with Dunnett’s or Bonferroni post-tests for comparisons of continuous variables between two or more experimental groups. Figures show mean ± standard deviation. Standard deviation was chosen to describe the variability of results (335). Kaplan-Meier survival curves were compared with the log-rank test. Normality was tested with the D’Agostino & Pearson omnibus normality test. In data which failed normality testing, a log-linear regression was performed to normalise data. Gaussian distribution of log-transformed data was then confirmed by the same test. These transformed values were used for statistical analysis using the statistical tests mentioned above, which assume normal distribution.
148 Chapter 3 - Results
3. EXPANDING THE PHENOTYPE OF ARGININOSUCCINIC ACIDURIA
3.1 Introduction
The phenotype of patients with argininosuccinic aciduria differs from other urea cycle defects by the higher incidence of neurocognitive symptoms, liver fibrosis, renal impairment, and systemic hypertension (See Section 1.2.3) (85, 336). These symptoms are observed in patients with early- or late-onset forms and in those without documented episodes of hyperammonaemia (76). Among urea cycle defects, patients with argininosuccinic aciduria have the lowest frequency of hyperammonaemic crises (23%) but the second highest frequency of cognitive impairment (65-74%) after arginase deficiency (See Section 1.2.4) (8, 97). This paradox raises questions about the role of hyperammonaemia in causing the neurological problems.
Newborns screened and treated prospectively from birth have been reported to have a better neurological outcome (See Section 1.2.8) (15, 17, 95). As conventional treatment decreases ammonia levels, it was suggested that neurological complications were caused by unrecognised hyperammonaemic episodes (95). However, newborn screening programmes can capture a wide phenotypic spectrum, including patients who would remain asymptomatic without treatment. Some of these screened patients had high residual ASL activity (15), suggesting that the prospectively treated cohorts might have had an increased number of mild cases, introducing a bias into the comparison.
To better describe the natural history of the disease, we studied a United Kingdom (UK)-wide cohort of patients with argininosuccinic aciduria in order to better delineate the natural history of the disease and to report the long-term neurological outcomes with a special interest for neuroimaging and genotype-phenotype correlation. We aimed to better characterise the benefit of an early treatment without the bias of a newborn screening, which allows the diagnosis of a wide phenotypic spectrum from
149 Chapter 3 - Results severely symptomatic to asymptomatic patients. Patients clinically asymptomatic but excreting biomarkers might therefore be wrongly considered as good responders to an early therapeutic intervention. We studied the outcome in families where a proband had been diagnosed symptomatically, and compared the outcome with a sibling or first-degree relative treated prospectively from birth after perinatal screening.
3.2 Patients
Fifty-six patients were classified as early-onset (n=23/56), late-onset (n=23/56) or perinatally screened patients (n=10/56). Ethnic origins were White British (n=24; 43%), Pakistani (n=16; 28%), Chinese (n=6; 11%), Indian (n=5; 9%), Bangladeshi (n=3; 5%), other White European (n=1; 2%), and unknown (n=1; 2%). Screened patients, diagnosed either antenatally or neonatally, had an affected familial proband with early-onset (n=5), late-onset (n=3) or unknown (n=2) phenotype (See Appendix 10.4.1). One pair of siblings with early-onset phenotype (Patients 15 and 16) was included in the study. Mean follow-up was not significantly different between the early-onset (EO) or late-onset (LO) groups compared to the screened group (SCR) (p=0.19 and p=0.19 respectively) (Table 11).
3.3 Neurological phenotype
Data file is available in Appendix 10.4.2.
The frequency and median age of onset of the neurological features were not significantly different between the groups (Table 11).
Developmental impairment was reported in 48/52 patients (92%) and was the most common symptom. Only four patients were reported with normal neurocognitive function: three early-onset patients aged <6 months (Patient 22), 23 months (Patient 15) and 11 year-old at last assessment (Patient 11) and one patient screened at birth
150 Chapter 3 - Results
(Patient 47; sibling to a late-onset proband, aged 8 year-old at last assessment) (Appendix 10.4.1). The median age at diagnosis was 2 years (range 0.1 to 6 years) (Table 11) and when observed, developmental impairment was present before the age of 6 years in all but two patients. Developmental impairment was mild or moderate affecting predominantly speech and learning ability. Detailed information about schooling was available in 35 patients (Appendix 10.3.2). Only 6/35 patients (17%) attended mainstream school without the need for additional educational support (Patients 9, 11, 35, 36, 47 and 48 with last assessment at 25, 11, 22, 20, 8 and 16 year-old respectively). Most patients (20/35; 57%) required speech and language therapy.
The neuropsychological assessments identified behavioural difficulties with auto- or hetero-aggression (n=3) and learning disabilities in logic and reasoning. The intelligence quotients (IQs) were in the extremely low range (n=3; median IQ=54; range 48 to 52). Twenty-one adults (EO n=3; LO n=17; SCR n=1) with a median age of 22.3 years (range 18 to 57 years) were assessed for socioeconomic status. Five (25%) had semi-skilled employment. Independent living was reported in 1/12 (8%), and long-term relationships in 3/12 (25%). Patients not living independently were accommodated in the parental (9/11; 82%) or care (2/11; 18%) homes.
Epilepsy was observed in 22/52 patients (42%) with no significant difference of the median age of onset between groups (Table 11). Various seizure types were reported including generalised, partial and complex, febrile and afebrile seizures. Tonic-clonic seizures were most frequent (n=16), followed by absence seizures (n=5), myoclonic jerks (n=4) and atonic seizures (n=2). Status epilepticus was occasionally observed (n=1) (Appendix 10.4.2). No type of seizure was preferentially associated with a group of patients. In epileptic patients, the electroencephalogram frequently shows a general pattern of complex spikes and waves pattern from bilateral centrotemporal and occipital regions without identifiable focus. Electroencephalograms for three non-epileptic patients showed an abnormal pattern in two patients (early-onset n=1, late-onset n=1). 17/22 patients (77%) were treated with an average of 1.5 antiepileptic drugs (range 0 to 4).
151 Chapter 3 - Results
Early Onset Late Onset Screened Total Number (adult) 23 (3=10%) 23 (16=70%) 10 (1=10%) 56 (20 = 36%) Sex (M/F) 12/11 11/12 8/2 31/25 Epidemiology Consanguinity 12/23 (52%) 2/23 (9%) 5/8 (62%) 21/52 (40%) Patients still living 16/23 (73%) 21/23 (91%) 9/10 (90%) 46/56 (82%)
2 antenatally Median age 4 days (2-8) 2.75 years (0.25-12) 2 years (0-12) 8 neonatally At diagnosis Ammonia (RI <100µmol/L if <28 days of life; 861 ± 550 212 ± 230 84 ± 50 530 ± 542 <50µmol/L if >28 days of life) Median follow-up (years) 11 (1.9-25.7) 15.1 (1-53) 15.6 (8-18.2) 12.4 (0-53) Follow up Patients lost 2/23 (9%) 4/23 (18%) 2/10 (20%) 8/56 (14%) Age in December 2015 (years) 11 (1.9-25.7) 23 (16.7-57) 15.6 (8-18.2) 15.6 (1.9-57)
Developmental delay 18/21 (86%) 23/23 (100%) 7/8 (88%) 48/51 (94%) Age when first reported (years) 2 (0.1-4) 2.5 (1-6) 3.1 (2-4) 2 (0.1-6) Epilepsy 8/21 (38%) 11/23 (48%) 3/8 (38%) 22/52 (42%) Age when first reported (years) 9 (1.5-13) 2 (0.7-11) 8.5 (8-9) 5.5 (0.7-13) Neurology Ataxia 3/21 (14%) 6/23 (26%) 0/8 (0%) 9/52 (17%) Myopathic features 4/21 (19%) 3/23 (13%) 0/8 (0%) 7/52 (13%) Spasticity 1/21 (5%) 1/23 (4%) 1/8 (12%) 3/52 (6%) Dystonia 1/21 (5%) 0/23 (0%) 0/8 (0%) 1/52 (2%) Abnormal brain MRI 5/9 (56%) 5/10 (50%) 2/4 (50%) 12/23 (52%) Hepatomegaly 17/21 (81%) 3/23 (13%) 5/8 (62%) 25/51 (49%) Age when first reported (years) 2.5 (0-12) 13.00 7.5 (0.9-11) 2.5 (0-12) Liver Transaminitis 18/21 (86%) 4/23 (17%) 6/8 (75%) 28/51 (55%) Age when first reported (years) 0.15 (0-6) 23 (1-53) 7 (3-7) 1 (0-53) ALT (RI 20-50 IU/L) 238 ± 218 81 ± 69 181 ± 111 169 ± 163
Kidney appearance Enlargement (>95thcentile) 8/18 (44%) 2/9 (22%) 2/5 (40%) 12/32 (38%) (ultrasound)
Poor corticomedullar differentiation 2/7 (29%) 0/3 (0%) 1/2 (50%) 3/12 (25%) Phenotype Hypokalemia : total 16/23 (70%) 4/23 (17%) 6/10 (60%) 26/56 (46%) Persistent 3/23 (13%) 3/23 (13%) 1/10 (10%) 7/56 (12%) Intermittent 13/23 (56%) 1/23 (4%) 5/10 (50%) 19/56 (34%) Arterial hypertension 1/23 (4%) 0/23 (0%) 0/9 (0%) 1/55 (2%) Age when first reported (years) 11 / / 11 Trichorrhexis nodosa 0/23 (0%) 5/23 (22%) 0/10 (0%) 5/56 (9%) Age when first reported (years) / 5.7 (2.5-12) / Severe diarrhoea 10/21 (48%) 3/20 (15%) 4/10 (40%) 17/51 (33%) Miscellaneous Impaired growth 2/23 (9%) 0/23 (0%) 1/10 (10%) 3/55 (5%) Prematurity 2/23 (9%) 0/23 (0%) 1/10 (10%) 3/55 (5%)
Atrial flutter (n=1), Hypercholesterolaemia psoriasis arthritis (n=1), Others (n=2), 22q11 deletion (=1), Umbilical hernia (n=1) jejunal atresia (n=1) sensorineural deafness (n=1)
Argininaemia (RI 30-126µmol/L) 126 ± 56 102 ± 46 134 ± 34 116 ± 47
Plasma argininosuccinic acid (RI <5µmol/L) Biology 512 ± 243 234 ± 180 238 ± 292 356 ± 257
Frequency 20/20 (100%) 16/20 (80%) 6/7 (86%) 42/47 (89%) Protein restricted diet Daily protein allowance (g/kg/d) 1.2 ± 0.47 1.2 ± 0.41 1.4 ± 0.18 1.2 ± 0.38
L-Arginine Frequency 20/20 (100%) 20/20 (100%) 7/7 (100%) 47/47 (100%) supplementation L-Arginine (mg/kg/d) 239 ± 93 155 ± 97 251 ± 89 201 ± 107 Na benzoate Frequency 17/19 (89%) 2/22 (9%) 6/7 (72%) 25/48 (52%) supplementation Na benzoate (mg/kg/d) 215 ± 53 134 ± 58 167 ± 25 191 ± 57
Therapeutics Frequency 7/19 (37%) 2/22 (9%) 1/7 (14%) 10/48 (21%) Na phenylbutyrate supplementation Na phenylbutyrate (mg/kg/d) 200 ±870 57 ± 17 NA 143 ± 83
No alternative Frequency 0/19 (100%) 18/22 (82%) 1/7 (14%) 19/48 (40%) pathway drugs
Table 11. Epidemiological and clinical data for the three analysed cohorts: early-onset, late-onset and screened patients
“Age at diagnosis” means the first occurrence of symptom. “Age at diagnosis” and “duration of follow-up” are presented as median ± range. Other figures show mean ± standard deviation. Hypokalaemia-total includes patients with intermittent and 152 Chapter 3 - Results persistent hypokalaemia. ALT: alanine aminotransferase. NA: not available. RI: Range interval. Follow-up is considered until December 2015. Plasma arginine and argininosuccinic acid concentrations reflect the last 10 measurements performed during follow-up when patients were in a compensated metabolic state on their standard treatment. Reproduced from (337).
Cerebellar dysfunction was detected in early-onset and late-onset patients (n=3 and n=6 respectively) with the incidence of 9/52 (17%) (Table 11). Ataxia was first noticed at a median age of 8.5 years (range 1 to 12) with two main age groups at first observation, either early around the age of 1 year in two cases (one early-onset patient and one late-onset patient) or later as teenagers (n=7). Three patients had dyskinesia and tremor and one had nystagmus (Appendix 10.4.2). The inconvenience caused was mild and no specific medical or surgical treatment was required.
Episodes of myopathy-like symptoms were reported in 7/52 patients (13%) (Table 11). This included global hypotonia with a hypomimic facial expression, and unexplained recurrent episodes of general weakness persisting for several days before spontaneous recovery. One hypotonic patient developed a severe scoliosis. One patient (patient 28, currently aged 15.9 years) was reported with fatigable ptosis, proximal weakness in upper limbs and reduced deep tendon reflexes from 12 year- old onwards. Four patients were investigated with electromyogram, which were always normal, including a Tensilon test in one patient. One patient (Patient 33 currently aged 26 years) presented with an electrophysiologically confirmed episode of Guillain-Barré syndrome at 8 year-old (Appendix 10.4.2).
Some patients were reported with spasticity (n=2), hemiparesis (n=1), dystonia (n=1) and visual impairment with retinal atrophy (n=1; patient 29). All these symptoms were considered to be sequelae of hyperammonaemia except for one screened patient presenting with mild spasticity without having experienced hyperammonaemic episodes (Patient 55).
MRI brain was performed as part of the clinical work-up in twenty-one patients with unexplained or severe neurological features. The average age at the time of MRI was
153 Chapter 3 - Results
12 years (range 0-23). Twelve scans were reported as abnormal (52%) (Table 11). Only one MRI was performed at diagnosis and showed oedema in the cerebellum, central tegmental tract and internal capsule, and small diffuse haemorrhages in the white matter. Neuroimaging performed during follow-up showed small parenchymal infarcts (n=4), foci of white matter hyperintensity on T2-weighted sequences (n=4), nodular heterotopia (n=2), cortical atrophy (n=2), cerebellar atrophy (n=2), perirolandic gliosis (n=1), thalamic atrophy (n=1), hyperintensity of caudate head and posterior putamen (n=1) or isosignal between pallidi and putamen (n=1) (Figure 37A and Appendix 10.4.2). Spectroscopy of basal ganglia (n=8; 3 early-onset, 5 late- onset) indicated a significant decrease of N-acetylaspartate and choline in early- onset patients compared to controls (One-way ANOVA with Bonferroni post-test, p<0.01 and p<0.01, respectively; Table 12). Creatine and guanidinoacetate levels in the basal ganglia did not differ significantly between controls, early- or late-onset groups (Figure 37B). Spectroscopy of the white matter (n=4; 3 early-onset, 1 late- onset) showed a significant decrease in creatine levels and an increase of guanidinoacetate (One-way ANOVA with Bonferroni post-test compared to controls, p<0.01 and p<0.05, respectively; Table 12) (Figure 37C) in patients compared to controls. Mean glutamate and glutamine levels were not raised (data not shown). Myo-inositol was not significantly decreased (One-way ANOVA with Bonferroni post- test compared to controls, p>0.05) (Figure 37C).
Neurological outcomes of screened patients presented in Table 11 were compared to early-onset or late-onset patients with Fisher’s exact t-test (Table 12). No significant difference was observed regarding the prevalence of developmental delay, epilepsy, ataxia, myopathy features, spasticity, dystonia or abnormal neuroimaging. Median age of onset of developmental delay and epilepsy did not differ significantly between the cohorts (One way ANOVA with Bonferroni post-test compared to controls, p=0.07 and p=0.13 respectively).
154 Chapter 3 - Results
Figure 37. Neuroimaging features of argininosuccinic aciduria patients
(A) Morphological brain MRI features (a, b) T2-weighted axial images showing brain matter volume loss and mild ex vacuo dilatation of ventricles (a) and high signal in bilateral caudate heads and posterior putamina (b) (white arrows). (c) T2-weighted axial images with severe diffuse cerebral atrophy and ventricular dilatation. (d, h) T1- weighted coronal image with right periventricular heterotopia (white arrows). (e, f): T2-weighted axial (e) and coronal (f) images with evidence of right inferior frontal lobe infarct (white arrow). (g) T2-weighted axial image with bilateral high signal of the peritrigonal white matter (white arrow). (B) Proton magnetic resonance spectroscopy features in basal ganglia. Assessment in early-onset (n=5), late-onset (n=3) and control (n=63) patients analysed using a paired t test. (C) Proton magnetic resonance spectroscopy features in white matter. Patients affected by argininosuccinic aciduria (n=4) and controls (n=53) analysed with one-way ANOVA with Bonferroni post-test compared to controls (only significant results are represented). Graphs represent mean ± standard deviation. * p<0.05 ; ** p<0.01. Reproduced from (337).
155 Chapter 3 - Results
None of the neurological elements considered for long-term outcome were statistically different between groups regardless of the timing of treatment or occurrence of hyperammonaemia (Table 11).
3.4 Systemic phenotype
Data file is available in Appendix 10.4.1.
47/56 patients (84%) were alive at the time of assessment with no significant difference between groups, with a median follow-up of 12.4 years (range 0-53) (Figure 38A). Cause and age of death were hyperammonaemic decompensation at presentation (n=2; patients aged 3 and 4 days), sepsis (n=3; patients aged 7 days, 11 years and 20 years), extradural hematoma (n=1; patient aged 2 years), hepatocellular carcinoma (n=1; patient aged 4.5 years), acute pancreatitis (n=1; patient aged 12 years) and a possible arrhythmia (n=1; patient aged 52 years).
Initial presentation at diagnosis showed: i) Early-onset patients were diagnosed at a median age of 4 days (range 2 to 8 days). Clinical symptoms at admission were drowsiness, hypotonia, lethargy, encephalopathy (n=13), neonatal seizures (n=4), poor feeding and vomiting (n=8), tachypnea and grunting (n=7), hypertonia and brisk reflexes (n=2) and hepatomegaly (n=1). ii) Late-onset patients were diagnosed at a median age of 2.75 years (range 0.25 to 12 years) with chronic symptoms (n=19) and acute symptoms caused by hyperammonaemic decompensation (n=3) (Data not available in one patient). Symptoms at diagnosis were developmental delay (n=12), brittle hair (n=5), behavioural disturbances (n=3), drowsiness or coma during metabolic stress (n=3), recurrent vomiting (n=3), protein aversion or poor feeding (n=3), seizures (n=2), tiredness (n=1), metabolic alkalosis, hypoglycaemia, and hyperlacticacidaemia (n=1). Screened patients were diagnosed in the antenatal (n=2) or neonatal period (n=8) and all were asymptomatic at diagnosis. 156 Chapter 3 - Results
A 1.00
0.75
0.50 Survival (%) Survival 0.25
0.00 0 10 20 30 40 50 60 B Time (Years)
Developmental delay Raised ALT Hepatomegaly Epilepsy Brittle hair Ataxia Hypokalaemia High blood pressure 0 2 4 6 8 10 12 Time (Years)
Figure 38. Natural history of argininosuccinic aciduria
(A) Kaplan-Meier survival curves for all (solid line), early-onset (dashed line), late- onset (dashed dotted line) and screened (dotted line) patients. (B) Natural history of the systemic phenotype of argininosuccinic aciduria. Mean ± standard deviation of age of onset of each symptom from data of the whole cohort when information available: developmental delay (n=7), abnormal LFTs (n=8), hepatomegaly (n=18), epilepsy (n=15), brittle hair (n=4), ataxia (n=6), hypokalaemia (n=2), high blood pressure (n=1). Symptom frequency in the total population of patients studied is presented in brackets. ALT: plasma alanine aminotransferase activity. It was assumed that patients had normal blood pressure if hypertension was not specifically mentioned in medical records. Reproduced from (337).
157 Chapter 3 - Results
Natural history data included the age of onset of organ involvement or symptoms (Figure 38B). Various symptoms were identified with follow-up indicating a progressive disease with a growing phenotype. Among neurological symptoms, developmental delay was the first observed usually during the second or third year of life followed by epilepsy and ataxia, later in childhood.
The most frequent hepatic involvement was a persistent rise in plasma alanine transaminase activity in 28/51 (55%), usually accompanied by hepatomegaly in 25/51 (49%). These features were related to age of onset: liver abnormalities were significantly more frequent in early-onset versus late-onset patients and screened versus late-onset patients (Fisher’s exact t-test, p<0.00001 and p<0.005, respectively; Table 11). In screened patients, the likelihood of liver features segregated according to the age of onset of the disease in the familial proband: 4/4 of screened patients with an early-onset familial history had hepatic abnormalities compared to 1/3 with a familial late-onset phenotype (Appendix 10.4.1). The age of onset of liver abnormalities was significantly earlier in early-onset versus late-onset patients (One-way ANOVA with Bonferroni post-test compared to controls, p<0.05; Table 12). No cholestasis was observed. Liver biopsy was performed in 2 early-onset patients. One confirmed a hepatocellular carcinoma. Another child had 2 biopsies performed at 1 and 11 years: both samples showed glycogen deposits, mild fibrosis and inflammation with no progression and no features of steatosis, cholestasis or iron deposition.
Other miscellaneous systemic symptoms were noted. Nephromegaly and poor corticomedullary differentiation were observed in 12/32 (38%) and 3/12 (25%) of patients assessed by ultrasound, respectively Both features were significantly more frequently observed in early-onset compared to late-onset patients (Fisher’s exact t- test, p=0.02 and p=0.01, respectively). Both nephrocalcinosis and renal hyperechogenicity were noted once. No impairment of the creatinine clearance with age was observed from data collected from 6 patients and followed from birth to adolescence (Figure 39).
158 Chapter 3 - Results
Hypokalaemia was observed in 26/56 (46%) as either a transient (19/56 ; 34%) or persistent (7/56 ; 12%) feature, both occurring more frequently in early-onset versus late-onset patients (Fisher’s exact t-test, p<0.001 and p<0.04, respectively). Acute metabolic decompensation, gastroenteritis and acute diarrhoea were significantly associated with transient hypokalaemia (Fisher’s exact t-test, p<0.005).
180
160
140
120
100
80 Creatinine clearance (mL/min) 0 5 10 15 20 Age (years)
Figure 39. Linear regression of creatinine clearance versus age
Creatinine clearance for 8 patients at Great Ormond Street Hospital, London (age range; 1 to 15 years old). Linear regression with r2 =0.004, not significantly different from the X axis.
Trichorrhexis nodosa was observed only in late-onset patients (n=5) before diagnosis and normalised with treatment (Table 11).
Chronic profuse diarrhoea was observed in 17 patients (33%; including early-onset n=10, late-onset n=3, screened patients n=4; Table 11). This symptom was refractory to symptomatic and immunosuppressive treatments and caused nutritional difficulties in several early-onset patients. This caused chronic metabolic decompensation in some patients (Patient 6). Despite generally becoming less troublesome with time, this severely impaired the quality of life of children and families. Two patients had colonoscopies performed at 5 years of age and repeated
159 Chapter 3 - Results at ages 7 and 10, respectively. Intestinal biopsies showed non-specific mild inflammation. Chronic pancreatitis was observed in one early-onset patient.
Refractory arterial hypertension was diagnosed in one early-onset patient (Patient 6) at the age of 9 years and was sub-optimally controlled despite three antihypertensive medications. This patient died at 12 years old from acute pancreatitis. One late-onset patient developed atrial flutter at 60 years.
3.5 Biomarkers and therapies
Alanine aminotransferase levels were more frequently raised in early-onset and screened groups but the values were not significantly different between groups (One- way ANOVA with Bonferroni post-test, p>0.05) (Figure 40A).
None of the patients in the screened group suffered severe or prolonged hyperammonaemia. Their mean plasma ammonia concentration at diagnosis was 84 μM (Table 11). Three of these patients had an initial ammonia level >100 μM (133, 134 and 190 μM), which normalised in less than 24 hours. All early-onset patients had hyperammonaemia at diagnosis (mean 861 μM, median age at diagnosis 4 days) with values significantly higher than in the late-onset and screened groups (One-way ANOVA with Bonferroni post-test p<0.001; Table 11 & 12). Only 50% of the late-onset patients were hyperammonaemic at diagnosis (mean ammonia 212 μM, mean age at diagnosis 2.75 years) (Figure 40B).
A protein-restricted diet was used in 42/47 patients (89%) (20/20 (100%) of the early- onset group and 16/20 (80%) of the late-onset group; Table 11). All patients were treated with L-arginine with no significant difference in the dose between groups (Table 11). Ammonia scavenger drugs (sodium benzoate and phenylbutyrate) were prescribed significantly more often in the early-onset and screened groups than in the late-onset group (One-way ANOVA with Bonferroni post-test, p<0.001 and p<0.01 respectively) and at higher doses in the early-onset group (One-way ANOVA with Bonferroni post-test, p<0.05) (Table 11 & 12).
160 Chapter 3 - Results
Plasma argininosuccinate levels were higher in early-onset (512 ± 92 μM) compared to late-onset (234 ± 64 μM) (p=0.03) (Table 11 and Figure 40C).
Plasma arginine levels were in the normal range as patients were all on L-arginine supplementation (Table 11).
A B M)
800 µ 1000 *
600 800
600 400 400
200 200
0 0 Alanina aminotransferase (IU/L) aminotransferase Alanina Plasma argininosuccinic acid ( acid argininosuccinic Plasma Early onset Late onset Screened Early onset Late onset Screened
*** 2500 C ***
M) 2000 µ
1500 ns 1000
Ammonaemia ( Ammonaemia 500
0 Early onset Late onset Screened
Figure 40. Observed levels of common biomarkers in screened, early- and late-onset ASA patients
(A) Alanine aminotransferase activity, (B) ammonaemia concentration and (C) plasma argininosuccinic acid concentration in early-onset, late-onset and perinatally screened groups. Horizontal lines display the mean ± standard deviation. One-way ANOVA with Bonferroni post-test; * p<0.05; *** p<0.001, ns not significant; (A) n=5-9; (B) n=8-26 ; (C) 2-10.
161 Chapter 3 - Results
Early onset versus Early onset versus Late onset versus Statistical Late onset patients Screened patients Screened patients test Biomarkers and p<0.001 (EO) p<0.001 (EO) ns Ammonia at diagnosis therapeutics (95% CI 286.6 to 1045) (95% CI 328 to 1198) (-380.5 to 575.6) One way p<0.05 (EO) ns Plasma ANOVA (95% CI 26.26 to ns (95% CI -446.3 to argininosuccinate with 565.4) (95% CI -165.3 to 711.8) 401.1) Bonferroni's p<0.05 (EO) ns ns post test Dose of ammonia (95% CI 12.12 to (95% CI -88.95 to (95% CI -199.4 to lowering medications 188.7) 160.8) 70.36)
Developmental delay 0.1 1 0.26 Epilepsy 0.76 1 0.68 Ataxia 0.46 0.54 0.14 Neurological Fisher's Myopathy feature 0.67 0.28 1 features exact t test Spaticity 1 0.53 1 Dystonia 1 1 1 Abnormal imaging 1 1 1
ns p<0.01 (EO) ns N-acetylaspartate Spectroscopy in (95% CI -1.52 to 0.42) (95% CU -1.67 to -0.32) (95% CI -1.17 to 0.39) One way basal ganglia ns p<0.01 (EO) ns Choline (95% CI -0.47 to 0.17) (95% CI -0.46 to -0.05) (95% CI -0.37 to 0.15) ANOVA with Bonferroni's p<0.05 (EO) Age of onset of liver post test Liver symptoms (95% CI -29.94 to - ns ns symptoms 4.36) (95% CI -16.12 to 7.26) (95% CI -2.88 to 28.32) Table 12. Pair comparison for statistical tests
The direction of the pair comparison is mentioned in brackets (EO for Early-onset). CI: confidence interval; ns: not significant. Reproduced from (337).
162 Chapter 3 - Results
3.6 Genotype-phenotype correlation
The genotype was available for nineteen patients (Table 13). Twenty mutations (including eight novel) were identified: eleven were missense, five splice site, two nonsense mutations and two deletions. The deletions included a 13 base pair deletion (c.1045_1057del, p.(Val349Cysfs*72)) and one large deletion of approximately 2 kilobases which included exons 15 and 16 (c.1143+117_*1353del). Homozygous mutations observed with early onset disease included c.437G>A p.(Arg146Gln), c.749T>A p.(Met250Lys), c.1045_1057del p.(Val349Cysfs*72), c.1143+117_*1353del and c.1153C>T p.(Arg385Cys). Homozygous mutations observed with late-onset disease included c.35G>A p.(Arg12Gln), c.377G>A p.(Arg126Gln) and c.1138A>G p.(Lys380Glu). The c.1045_1057del deletion is predicted to cause a frameshift and introduction of a premature stop codon and the c.1143+117_*1353del deletion is predicted to cause the loss of exons 15 and 16, and both were associated with early-onset phenotype. Patients homozygous for c.1143+117_*1353del were younger brothers of a proband who was not genotyped but presented with the early-onset phenotype (Table 13).
Among the eight novel mutations, 4 mutations were missense (Arg126Gln, Arg146Gln, Met250Lys, Glu258Lys). An alignment of hASL genes of other species with that of human ASL using Clustal omega software (Figure 41) showed that these amino acids are highly conserved throughout evolution and strengthened the likelihood of a deleterious effect. Out of the eight novel mutations, only c.377G>A, p.(Arg126Gln) and c.749T>A, p.(Met250Lys) alleles were reported in Eva-ExAC and Ensemble databases (accessed on 01/12/2016) with an allele frequency <0.01% (Table 14).
163 Chapter 3 - Results
Patient Severity in Allele 1 Allele 2 Presumed effect on protein Reported severity in the literature number this study 1 c.35G>A c.35G>A p.(Arg12Gln) p.(Arg12Gln) Late onset Unknown (n=1) (Balmer et al., 2014) 2 c.348+1G>A c.532G>A Splicing effect p.(Val178Met) Late onset New genotype 3 c.349-1G>A c.532G>A Splicing effect p.(Val178Met) Early onset New genotype 4, 5 c.377G>A c.377G>A p.(Arg126Gln) p.(Arg126Gln) Late onset New genotype 6 c.437G>A c.437G>A p.(Arg146Gln) p.(Arg146Gln) Early onset New genotype 7 c.437G>A c.446+1G>A p.(Arg146Gln) Splicing effect Late onset New genotype 8 c.544C>T c.772G>A p.(Arg182*) p.(Glu258Lys) Late onset New genotype 9 c.719-2A>G c.857A>G Splicing effect p.(Gln286Arg) Early onset New genotype 10 c.721G>A c.918+5G>A p.(Glu241Lys) Splicing effect Early onset New genotype 11, 12 c.749T>A c.749T>A p.(Met250Lys) p.(Met250Lys) Early onset New genotype 13 c.1045_1057del c.1045_1057del p.(Val349Cysfs*72) p.(Val349Cysfs*72) Early onset New genotype 14 c.1138A>G c.1138A>G p.(Lys380Glu) p.(Lys380Glu) Late onset Unknown (n=1) (Balmer et al., 2014) Loss of exons Loss of exons 15, 16 # c.1143+117_*1353del c.1143+117_*1353del Early onset New genotype 15 and 16 15 and 16 Prenatal diagnosis (n=3, Keskinen et al., 2008, Balmer et al., 2014), Early onset (n=4, this study, Kleijer et al., 2002, Keskinen et al., 2008), 17, 18 c.1153C>T c.1153C>T p.(Arg385Cys) p.(Arg385Cys) Early onset Late onset (n=7, Kleijer et al., 2002, Keskinen et al., 2008), Unknown (n=3, Balmer et al., 2014) 19 c.1284G>A c.1366C>T p.(Trp428*) p.(Arg456Trp) Late onset New genotype
Table 13. Genotype-phenotype correlation of hASL
New genotype refers to combination of mutations not described before. Presumed protein effect is mentioned within brackets for novel mutations. For screened patients, the severity of the phenotype is deducted from the symptomatic familial proband. # Patients 15 and 16 are siblings. Reproduced from (337).
164 Chapter 3 - Results
Human ------MASESGKL 8 Chimpanzee ------MASESGKL Cow ------MASESGKL Rat ------MASESGKL Mouse ------MASESGKL Chicken ------MASEGDKL Frog ------MASEGNKL Zebrafish ------MASSEGNKL Staphylococcus_aureus ------MSNKA Sulfobacillus_acidophilus ------MPDL Corn MAFTSQSLIFRAPASPSCARITPSSGRVTLRDRSAVFPLVAAASTSMASSESEDKRETKL .
Human WGGRFVGAVDPIMEKFNASIAYDRHLWEVDVQGSKAYSRGLEKAGLLTKAEMDQILHGLD 68 Chimpanzee WGGRFVGAVDPIMEKFNASIAYDRHLWEVDVQGSKAYSRGLEKAGLLTKAEMDQILHGLD Cow WGGRFVGTVDPIMEKFNSSITYDRHLWEADVQGSKAYSRGLEKAGLLTKAEMDQILHGLD Rat WGGRFAGSVDPTMDKFNSSIAYDRHLWNVDLQGSKAYSRGLEKAGLLTKAEMQQILQGLD Mouse WGGRFVGAVDPIMEKFNSSISYDRHLWNVDVQGSKAYSRGLEKAGLLTKAEMQQILQGLD Chicken WGGRFSGSTDPIMEMLNSSIACDQRLSEVDIQGSMAYAKALEKAGILTKTELEKILSGLE Frog WGGRFVGSIDPIMEMFNCSVNYDQRMWSADIRGSQAYVKALEKAGLVSKSEMEQIISGLD Zebrafish WGGRFVGNTDPIMEKFNASISYDQRMWKADIKGSKAYVKALQKASLVTQNEMEQILTGLD Staphylococcus_aureus WGGRFEVQPEEWVDDFNASITFDQTLIDQDIEGSIAHATMLADQGIISQQDSEQIIQGLK Sulfobacillus_acidophilus WSGRLPGGLDPKARAFSSSLAVDWRLARYDIQGSLAHADMLAAVGLLTPEEHEAIQQGLE Corn WGGRFEEGVTDAVERFTESISYDWQLYKYDIMGSKAHASMLAGQGLITATDRDIILEGLD *.**: :. *: * : *: ** *: * .::: : : * **.
Human KVAEEWAQGTFKLNSNDEDIHTANERRLKELIGATAGKLHTGRSRNDQVVTDLRLWMRQT 128 Chimpanzee KVAEEWAQGTFKLNSNDEDIHTANERRLKELIGETAGKLHTGRSRNDQVVTDLRLWMRQT Cow KVAEEWAQGTFKLNPNDEDIHTANERRLKELIGETAGKLHTGRSRNDQVVTDLRLWMRQN Rat KVAEEWAQGIFKLYPNDEDIHTANERRLKELIGEAAGKLHTGRSRNDQVVTDLRLWMRQT Mouse KVAEEWAQGTFKLHPNDEDIHTANERRLKELIGEAAGKLHTGRSRNDQVVTDLRLWMRQT Chicken KISEEWSKGVFVVKQSDEDIHTANERRLKELIGDIAGKLHTGRSRNDQVVTDLKLLLKSS Frog KIHDEWSSGTFVLTKGDEDIHTANERRLKELIGEVAGKLHTGRSRNDQVVTDMRLWLRDS Zebrafish KVLDEWSKGEFEIKPGDEDIHTANERRLKELIGDAAGKLHTGRSRNDQVATDMRLWLRDG Staphylococcus_aureus SIQHDYHQDQIQFSASLEDIHLNIEHELIKRIGDAGGKLHTGRSRNDQVATDMHLYTKKQ Sulfobacillus_acidophilus ALLSEVEAGADPWDLMAEDVHSAVEQELTRRLGPVAGKLHTARSRNDQVALDLHLYVKDA Corn QIERLIQEGKFEWRKDREDVHMNIEAALIERVGEPAKKLHTARSRNDQIVTDLRLWCRDA : . **:* * * . :* . ****.******:. *::* :.
Human CSTLSGLLWELIRTMVDRAEAERDVLFPGYTHLQRAQPIRWSHWILSHAVALTRDSERLL 188 Chimpanzee CSTLSGLLWELIRTMVDRAEAERDVLFPGYTHLQRAQPIRWSHWILSHAVALTRDSERLL Cow CSMLSALLCELIRTMVDRAEAERDVLFPGYTHLQRAQPIRWSHWILSHAVALTRDSERLL Rat YSKLSTFLKVLIEAMVDRAEAECEVLFPGYTHLQRAQPIRWSHWILSHAVALTRDLERLK Mouse CSKLSALLRVLIGTMVDRAEAERDVLFPGYTHLQRAQPIRWSHWILSHAVALTRDSERLL Chicken ISVISTHLLQLIKTLVERAATEIDVIMPGYTHLQKALPIRWSQFLLSHAVALIRDSERLG Frog CSTLYTHLTRLIQTMVERAAIEVNILFPGYTHMQRAQPIRWSHWILSHAVALSRDAERLG Zebrafish IATLKELALQLINTMVERAAAEIEILCPGYTHMQRAQPIRWSHWLLSHVVAISRDVERLE Staphylococcus_aureus VQDIIALIKSLQSVIVDIASNNVDTIMPGYTHLQRAQPISFAHHIMTYFWMLQRDQQRFE Sulfobacillus_acidophilus GQKTIAALNHLVDTILGLAERWADLPLPGYTHMQPGQPVTVGHHLLAYVWMLLRDRSRIE Corn IDRILIRIKQLQVSLVILASKYVDLIVPGYTHLQRAQPVLLPHLLLSYVEQLERDAGRLV * :: * : *****:* . *: : :::: : ** *:
Human EVRKRINVLPLGSGAIAGNPLGVDRELLRAELNFGAITLNSMDATSERDFVAEFLFWASL 248 Chimpanzee EVRKRINVLPLGSGAIAGNPLGVDRELLRAELNFGAITLNSMDATSERDFVAEFLFWASL Cow EVRKRINVLPLGSGAIAGNPLGVDRELLRAELDFGAITLNSMDATSERDFVAEFLFWASL Rat EVQKRINVLPLGSGAIAGNPLGVDREFLCAELNFGAITLNSMDATSERDFVAEFLFWASL Mouse EVQKRINVLPLGSGAIAGNPLGVDRELLRAELNFGAITLNSMDATSERDFVAEFLFWASL Chicken EVKKRMSVLPLGSGALAGNPLEIDRELLRSELDFASISLNSMDAISERDFVVELLSVATL Frog EVKKRVNVLPLGSGAIAGNPLGVDRELLCK------AEFLFWASL Zebrafish DIRKRVNVMPLGSGAIAGNPFDIDRELLRQELSFDSISINSMDATGQRDFVAEFLFWGSM Staphylococcus_aureus DSLKRIDINPLGAAALSGTTYPIDRHETTALLNFGSLYENSLDAVSDRDYIIETLHNISL Sulfobacillus_acidophilus DMAARANVSPLGAGALAGTTLPIDPTRTAHHLGFAGPYQNSLDAVSDRDFAIEFLADLAL Corn NCRERMNFCPLGACALAGTGLPIDRFQTAKDLKFTAPMKNSIDAVSDRDFVLEFLAANSI : * .. ***: *::*. :* * * ::
Human CMTHLSRMAEDLILYCTKEFSFVQLSDAYSTGSSLMPQKKNPDSLELIRSKAGRVFGRCA 308 Chimpanzee CMTHLSRMAEDLILYCTKEFSFVQLSDAYSTGSSLMPQKKNPDSLELIRSKAGRVFGRCA Cow CMTHLSRMAEDLILYGTKEFSFVQLSDAYSTGSSLMPQKKNPDSLELIRSKAGRVFGRCA Rat CMTHLSRMAEDLILYGTKEFNFVQLSDAYSTGSSLMPQKKNPDSLELIRSKARRVFGRCA Mouse CMTHLSRMAEDLILYGTKEFSFVQLSDAYSTGSSLMPQKKNPDSLELIRSKAGRVFGRCA Chicken LMIHLSKLAEDLIIFSTTEFGFVTLSDAYSTGSSLLPQKKNPDSLELIRSKAGRVFGRLA Frog CMTHLSKMSEDLIIYSTKEFGFVTLSDSYSTGSSLMPQKKNPDSLELIRGKTGRVFGRCS Zebrafish CLTHLSKMAEDLILYSTKEFSFINLTDAYSTGSSLMPQKKNADSLELIRSKAGRVFGRCA Staphylococcus_aureus TMVHLSRFAEEIIFWSTDEAKFITLSDAFSTGSSIMPQKKNPDMAELIRGKVGRTTGHLM Sulfobacillus_acidophilus VAVHLSRLSEELILWSGREFGFITLSDHWATGSSMMPQKKNPDIAELIRGKSGVVIGQLT Corn AAVHLSRIGEEWVLWASEEFGFLIPSDKVSTGSSIMPQKKNPDPMELVRGKSARVVGDLM ***::.*: ::: * *: :* :****::***** * **:*.* . *
Human GLLMTLKGLPSTYNKDLQEDKEAVFEVSDTMSAVLQVATGVISTLQIHQENMGQALSPDM 368 Chimpanzee GLLMTLKGLPSTYNKDLQEDKEAVFEVSDTMSAVLQVATGVISTLQIHQENMGQALSPDM Cow GLLMTLKGLPSTYNKDLQEDKEAVFEVSDTMSAVLQVATGVISTLQIHRENMGRALSPDM Rat GLLMTLKGLPSTYNKDLQEDKEAVFEVSDTMTAVLQVATGVISTLQIHRENMAQALSPDM Mouse GLLMTLKGLPSTYNKDLQEDKEAVFEVSDTMIAVLQVATGVISTLQIHRENMKQALSPDM Chicken AVLMVLKGLPSTYNKDLQEDKEAVFDVVDTLTAVLQVATGVISTLQVNKENMEKALTPEL Frog GFLTTLKGLPSTYNKDLQEDKEAMFDVYDTVCAVLQVASGVIATLQINKEAMEKALSPDM Zebrafish GFLMTLKGLPSTYNKDLQEDKEAMFDTYDTVHAVLQVATGVISTLKVNQVKMEEALSPDM Staphylococcus_aureus SMLMTLKGLPLAYNKDMQEDKEGLFDAVHTIKGSLRIFEGMIQTMTINKERLNQTVKEDF Sulfobacillus_acidophilus AFLTLMKGLPLAYNRDLQEDKGLLFTGVDTVLASLDAMAGLLAGITVNPKRIQEDMGEDL Corn TVLTLCKGLPQAYNRDLQEDKEPLFDSVKAVLGMLEVCTEFAQNISFNSKRIQSSLPAGY .* **** :**:*:**** :* .:: . * . : .: : :
Human L-ATDLAYYLVRKGMPFRQAHEASGKAVFMAETKGVALNQLSLQELQTISPLFSGDVICV 427 Chimpanzee L-ATDLAYYLVRKGMPFRQAHEASGKAVFMAETKGVALNQLSLQELQTISPLFSGDVSCV Cow L-ATDLAYYLVRKGMPFRQAHEASGKAVFMAETKGVALNQLSLQELQTISPLFSGDVSHV Rat L-ATDLAYYLVRKGMPFRQAHEASGKAVVVAEMKGVALNQLSLQELQTVSPLFSSDVNLV Mouse L-ATDLAYYLVRKGMPFRQAHEASGKAVFMAETKGVALNLLSLQELQTISPLFSGDVSHV Chicken L-STDLALYLVRKGMPFRQAHVASGKAVHLAETKGIAINKLTLEDLKSISPLFASDVSQV Frog L-STDIAYYLVRKGMPFRQAHGVSGKVVQLAETKGMSVDKLTLENLKSISPLFSDDVSKV Zebrafish L-ATDLAYHLVRKGMPFREAHGCAGKAVYIAESKNIRLSQLTVEDLQTVSSLFDKDVSSV Staphylococcus_aureus SNATELADYLVTKNIPFRTAHEIVGKIVLECIQQGHYLLDVPLSTYQQHHSSIDADIYDY Sulfobacillus_acidophilus L-ATDWAEKLVQDGIPFREAHGQVAARFRDPQVKAASP------DAIR-- Corn LDATTLADYLVKKGVPFRTSHEIVGRSVALCVSKNCQLAELELADLKSVHPVFEDDVYEY :* * ** ..:*** :* . . : :
Human WDYGHSVEQYGALGGTARSSVDWQIRQVRALLQAQQA------464 Chimpanzee WDYGHSVEQYGALGGTARSSVDWQIRQVRALLQAQQA------Cow WDYGHSVEQYEALGGTARSSVDWQIGQLRALLRAQQTESPPHASPK Rat WDYSHSVEQYTALGGTAQSSVEWQISQVRALLQM------Mouse WDYSHSVEQYSALGGTAKSSVEWQIRQVRALLQAQEP------Chicken FNIVNSVEQYTAVGGTAKSSVTAQIEQLRELLKKQKEQA------Frog WNYTNSVEQYTAAGGTAKSSVLVQIEQLRTWMKTHRV------Zebrafish WDYVKSVEQYSAPGGTSKSSVTAQIQHFKTWLQAQKL------Staphylococcus_aureus LQPENCLKRRQSYGSTGQSSVKQQLDVAKQLLSQ------Sulfobacillus_acidophilus ----HSLTLRDRPMGPGPAHVREQIRAARQLL------Corn LGVENAVNKFISYGSTGSEQVKKQLEDWRIQLGVSS------:.: . . * *: : : 165 Chapter 3 - Results
Figure 41. Evolutionary conservation of novel missense mutations for hASL
Clustal Omega software (http://www.ebi.ac.uk/Tools/msa/clustalo/) (last accessed 01/05/2017) was used to generate the alignment of the protein sequence of different species. Residues identical to the human ASL gene are indicated by an asterisk (*). Conserved amino acids with strong or weak similarities in the biochemical properties are marked with a colon (:) and a dot (.), respectively. Amino acids affected by novel missense or truncating mutations are highlighted in yellow Arg126Gln, Arg146Gln, Met250Lys, Glu258Lys. Protein sequences used to generate this alignment are NP_000039 (human), H2RBX6 (chimpanzee), NP_001029600 (cow), NP_001025885 (chicken), NP_067588 (rat), NP_598529 (mouse), NP_001106586 (frog), NP_956745 (Zebrafish), (staphylococcus aureus), (sulfobacillus acidophilus) and (corn).
Allele frequency
Mutation Eva-ExAC Ensembl
c.348+1G>A 0/118,472 Novel c.349-1G>A 0/118,472 Novel c.377G>A 1/119,582 1/121,403 c.437G>A 0/118,472 Novel c.749T>A 1/117,096 1/121,409 c.772G>A 0/118,472 Novel c.1143+117_*1353del 0/118,472 Novel c.1284G>A 0/118,472 Novel
Table 14. Allele frequency of novel mutations in Eva-ExAC and Ensembl databases
Databases accessed on 01/12/2016. Reproduced from (337).
166 Chapter 3 - Results
3.7 Discussion
This study describes three groups of patients with argininosuccinic aciduria (early- onset, late-onset and perinatally screened after the diagnosis of a familial proband) with prolonged follow-up periods and compares the long-term outcome with regards to the time at initiation of treatment. In contrast to previously reported patients diagnosed by newborn screening, this study describes screened patients, who had a familial index case with a known phenotype.
3.7.1 An expanding phenotype of argininosuccinic aciduria
Neurological phenotype
The most common complications of patients affected by argininosuccinic aciduria were neurological. The comparison of the different groups demonstrates a homogeneous long-term neurological outcome, with no significant difference in frequency, severity and age of onset for all neurological features assessed (developmental delay, neuroimaging, myopathy, epilepsy, ataxia and other abnormal movements) (See Section 3.3). Aggressive behaviour and psychiatric problems such as psychosis and paranoid ideation have been previously reported in argininosuccinic aciduria (See Section 1.2.3) (53, 60, 66, 67), although these features were not observed in this cohort.
The early-onset group showed a tendency to have higher ammonia levels compared to the late-onset group, as evidenced by differences in the ammonia levels at diagnosis, and the subsequent need for ammonia scavenger medications and protein-restricted diet. The screened patients also needed more treatment to control their ammonia levels than patients in the late-onset group because most of them (5/8) had siblings with early-onset disease. Plasma argininosuccinic acid levels were higher in early-onset compared to late-onset and screened patients (See Section 3.5). However, the differences in ammonia levels and therapeutic requirements did not affect the neurological phenotype. These observations suggest that ammonaemia
167 Chapter 3 - Results and argininosuccinic acid levels are not the dominant factors causing the long-term neurological phenotype in argininosuccinic aciduria.
Compared to other urea cycle defects, patients affected by argininosuccinic aciduria have the lowest frequency of hyperammonaemic crises (23%) but the second highest frequency of cognitive impairment (65%) after arginase deficiency (97). As the neurological disease is progressive (See Section 3.4), duration of the follow-up is related to the outcome. At the end of the first year, the frequency of developmental impairment is similar to other urea cycle defects (64%) attributable to sequellae of neonatal hyperammonaemia in early-onset patients (338). This frequency increases to 50-100% with time (this study, (3, 6, 7, 59, 97, 129-131) with a median age at first report of developmental impairment in this study of 2 years (See Section 3.3).
Previous publications have suggested that neonatal screening and early initiation of treatment may prevent or ameliorate the neurological disease (15, 95, 145). For example a cohort of neonatally screened Austrian patients with late-onset phenotype were initially described to have a normal psychomotor development at the mean age of 6 years (95). However, another report describing an extended Austrian cohort of neonatally screened patients but tested at a median age of 13 years found that 35% of patients (6/17) had an IQ of less than 80. Although in this group none of the patients had seizures, 45% (4/9) had an abnormal electroencephalogram (17). Interestingly, none of the patients had recorded episodes of hyperammonaemia and none were treated with protein restricted diet or ammonia scavenger drugs. The authors hinted that argininosuccinic aciduria could inherently be a heterogeneous disorder and therefore lack of knowledge of genotype-phenotype correlation makes it difficult to conclude whether early treatment modifies the neurological phenotype (See Section 1.2.8). In our study all screened patients had a symptomatic sibling, which provided us with the information about probable disease subtype.
Systemic phenotype
In addition to the neurological complications, our study found a wide range of systemic problems in patients with argininosuccinic aciduria (See Section 3.4). All of the systemic complications were more frequent in the early-onset group, apart from
168 Chapter 3 - Results trichorrhexis nodosa, which was only seen in the late-onset patients (5/23). Liver disease is well recognised in argininosuccinic aciduria (74) and was found in 81% of the early-onset patients. No patients had cholestasis or developed cirrhosis or liver failure. Hypokalaemia occurred in 70% of our early-onset patients, usually as an intermittent problem during illness. Nephromegaly was also common, providing further evidence for renal involvement. Chronic diarrhoea has not been reported previously but it was a major problem in many of our patients with endoscopy showing non-specific mild inflammation. This has been observed in an enterocyte- specific conditional knockout mouse model, in which a loss of ASL was associated with necrotizing enterocolitis, a persisting proinflammatory state with increased levels of interleukin 6 (IL-6) and neutrophilic infiltration of the gut. This was associated with an activation of proapoptotic pathways in enterocytes (339).
As a similar pattern of systemic complications was found in the screened group, this suggests that prospective treatment has no preventative effect.
Genotype-phenotype correlation
Mutation analysis of this cohort of patients identified likely genotype-phenotype correlation for some of the mutations in agreement with the literature.
The frequently occurring mutation c.35G>A, p.(Arg12Gln) was associated with the late-onset phenotype in one patient in this study as previously suggested for homozygous and compound heterozygous patients (17, 21, 23). It has been reported that the arginine 12 residue on the N-terminal loop close to the catalytic site of ASL might influence the binding or exit of the substrate without affecting the catalytic site therefore explaining the milder form of argininosuccinic aciduria associated with this mutation (23).
The novel homozygous mutation c.749T>A, p.(Met250Lys) was observed in two unrelated patients with an early onset phenotype. This mutation involves changes in the protein sequence close to two other amino acid modifications also associated with an early-onset phenotype p.(Glu241Lys) and p.(Trp245fs) (21).
The homozygous mutation c.1153C>T, p.(Arg385Cys) has been associated previously with either early-onset (n=4) (61, 130), this study), or late onset 169 Chapter 3 - Results phenotypes (n=7) (61, 130). However, all patients with the late onset phenotype were diagnosed before 20 months of life. This mutation has a founder effect in the Finnish population (130) and is associated with very low argininosuccinate lyase activity affecting an amino acid near the catalytic site of the enzyme (340).
In this study c.857A>G, p.(Gln286Arg), when in compound heterozygous form with c.719-2A>G, resulted in an early-onset disease. Previously, when detected in homozygous form it was also reported to cause early-onset disease (21, 35, 130, 137, 341 5286). It is not surprising as the glutamine 286 residue is part of the 280’s loop (residues 270-290), which affects catalysis (23). However c.857A>G has also been found in combination with c.578G>A, another mutation seen in early-onset patients, in a patient with a late-onset phenotype (diagnosed at 15 months) and borderline ammonia levels (61). These inconsistencies suggest that a novel combination of two missense mutations may produce an unexpected phenotype.
Some similar genotypes can cause different phenotypes e.g. c.348+1G>A/c.532G>A and c.349-1G>A/c.532G>A causing late- and early-onset phenotypes, respectively. The cut-off age of 28 days of life differentiating early- from late-onset phenotypes is not as clear-cut in a clinical context and does not reflect the diversity of clinical presentation. This needs to be considered in a phenotypic continuum taking into account the age at diagnosis. Therefore a patient presenting with a late-onset phenotype diagnosed during paediatric age might have a genotype more similar to a patient with early-onset presentation rather than a patient with late-onset phenotype diagnosed at adulthood. Supporting this, the patient with genotype c.348+1G>A/c.532G>A was diagnosed at the age of 12 years.
3.7.2 Pathophysiology of argininosuccinic aciduria
Argininosuccinate lyase is not only involved catalytically in the urea cycle but is also required structurally to maintain a complex enabling the channelling of arginine to nitric oxide synthase for nitric oxide production. An early-onset phenotype with neonatal hyperammonaemia is associated with a more severe systemic phenotype, whilst the neurological phenotype is not affected by the age of onset.
170 Chapter 3 - Results
Various pathophysiological mechanisms have been proposed to account for the long- term complications of argininosuccinic aciduria (See Section 1.2.2).
Argininosuccinic acid may be toxic to the brain, either directly or via the formation of guanidino compounds i.e. guanidinosuccinic acid or guanidinoacetate. Raised guanidinoacetate was reported on brain spectroscopy of patients affected by argininosuccinic aciduria in the gray (3.63 ± 0.6 mM) and the white matter (3.52 ± 0.09 mM) (52, 53) and may be explained by L-arginine supplementation (53). In our study, levels of guanidinoacetate were similar to controls in basal ganglia but slightly elevated in white matter (1.05 ± 0.41 mM). Patients with guanidinoacetate methyltransferase (GAMT) deficiency have much higher guanidinoacetate concentrations in brain (3.4-3.6 mM) (342) and cerebro spinal fluid (11-12 μM) (343). There is also some evidence of raised guanidinoacetate in patients with hyperargininaemia due to arginase deficiency, with variable cerebrospinal fluid guanidinoacetate concentrations (up to 0.127 μM versus controls 0.049 μM) (344). Guanidinosuccinic acid could mimic nitric oxide and activate N-methyl-D-aspartate (NMDA) receptors (345-347) and there is also evidence that guanidinoacetate could be neurotoxic (348, 349). However, if argininosuccinic acid were to be the main cause of the neurological complications, one might expect more severe problems in patients with higher concentrations (such as patients in the early-onset group), which we did not see in our patients (See Section 3.3) (345-347).
In humans, argininosuccinate lyase is crucial for the synthesis of L-arginine, which is, therefore, an essential amino acid in patients with argininosuccinic aciduria. Arginine deprivation, associated with altered nitric oxide-mediated immune responses, can alter neuronal function and lead to site-specific neuronal loss in animal models of neurodegenerative diseases (350). Arginine is a precursor for several metabolic pathways including the synthesis of creatine, agmatine and polyamines (See Section 1.2.1). Creatine deficiency was observed on brain spectroscopy in this study (See Section 3.3) and other publications (51-53). Secondary creatine deficiency is a common feature in urea cycle defects (51) caused by either hypoargininaemia or defective transport of creatine (351) but its role in cerebral dysfunction has not been convincingly demonstrated (51). Agmatine, which is synthesized from L-arginine may
171 Chapter 3 - Results be involved in learning and memory processing (352, 353), neuroprotection against inflammation, ischemia, or traumatic injury (354). In addition agmatine is likely to be involved in other nitric oxide-modulated effects including anticonvulsant effects (355), mood-modification, lowering blood pressure and inhibiting sympathetic tone (356), improvement of renal function (357, 358). Thus secondary agmatine deficiency could explain some of the symptoms observed in argininosuccinic aciduria.
Finally, the neurological problems in argininosuccinic aciduria may be caused by impaired nitric oxide synthesis. Using an AslNeo/Neo hypomorphic argininosuccinic aciduria mouse model, Erez et al. showed that defective argininosuccinate lyase is responsible not only for the loss of the catalytic function of the enzyme, which cleaves argininosuccinate into L-arginine and fumarate, but also affects the structural integrity of a multi-protein complex composed of argininosuccinate lyase, argininosuccinate synthase, cationic transporter 1, HSP90 and nitric oxide synthase which is required for nitric oxide production (See Section 1.2.1) (359, 360). Defective intracellular arginine channelling disrupts the nitric oxide synthase-dependent nitric oxide synthesis pathway and leads to systemic nitric oxide deficiency (310). The production of nitric oxide by nitric oxide synthase is tightly regulated in tissues.
Hypoargininaemia or depletion of tetrahydrobiopterin (BH4) can lead to uncoupling of nitric oxide synthase (32), decreased nitric oxide production and increased generation of free radicals that damage tissues (See Section 1.2.2 & Figure 5) (85). - - Reactive oxygen species such as peroxynitrite (ONOO ), superoxide (O2 ) or hydrogen peroxide (H2O2) can damage DNA or lipids but interfere as well with nitric oxide synthase activity, nitric oxide production and regulation of the microcirculation (361). Nitric oxide synthesis occurs in most organs and nitric oxide mediates essential functions including vasodilatation (362, 363), macrophage-mediated cytotoxicity (364), gastro-intestinal smooth muscle relaxation (365), platelet aggregation (365, 366) and bronchodilation (367). In addition regulation of tissue nitric oxide levels affect progression of liver fibrosis (368), muscle strength and performance (369, 370) and kidney filtration rate (357, 371, 372), through activation of guanylate cyclase (373) and protein S-nitrosylation (374), which in turn regulates histone methylation and gene expression (375). In the brain, nitric oxide plays a key- role as a signalling molecule (376) involved in neurotransmission (377), regulation of
172 Chapter 3 - Results neuronal differentiation (378, 379) and migration (380, 381). The nitric oxide signalling pathway controls cyclic adenosine deaminase response-element binding protein (CREB)-dependent genes through S-nitrosylation of histone deacetylase 2 (HDAC2) (382). S-nitrosylated HDAC2 is released from chromatin inducing acetylation of histones that promotes transcription of genes (383) involved in neurodevelopment (376, 384, 385).
Nitric oxide deficiency might account for the neuropathology underlying the neuroimaging findings highlighted in this study, such as local parenchymal infarcts or nodular heterotopia, due to impaired microcirculation and abnormal neuronal migration during development, respectively. Besides neurological implications, several symptoms might be caused by systemic nitric oxide imbalance such as systemic hypertension, chronic hepatitis, renal (357, 371, 372), and gut dysfunction (386), nodular heterotopia due to abnormal neuronal migration during development, liver fibrosis (368), myopathy-like phenotype (369) and chronic diarrhoea (364, 387). Nitric oxide participates in the regulation of the microcirculation in the liver (388), is involved in liver ischemia reperfusion injury potentiating inflammation and release of reactive oxygen species (389). In liver fibrosis, nitric oxide bioavailability is reduced by down-regulation of the endothelial nitric oxide synthase (eNOS) and increased production of reactive oxygen species (368). Hence disturbed hepatic metabolism of nitric oxide could account for at least some of the symptoms observed in patients (See Section 1.2.2).
3.7.3 Need for new therapies
This study displays a persisting neurological and systemic disease in patients affected by argininosuccinic aciduria and treated conventionally, although not obviously linked to hyperammonaemia. Although some organs (liver, kidney, gut) are more frequently affected in early-onset patients, who have higher ammonia and argininosuccinic acid levels, this interestingly is not true for the brain. Our observation of parenchymal infarcts, nodular heterotopia and the recent report of mild
173 Chapter 3 - Results neurological improvement after nitric oxide supplementation (85) support the role of nitric oxide deficiency in the pathophysiology of the brain disease in argininosuccinic aciduria. Currently correction of nitric oxide deficiency is not considered in the conventional treatment of argininosuccinic aciduria although a clinical trial has recently been initiated to study the long-term outcome of nitric oxide supplementation (Clinical trials numbers NCT02252770 and NCT03064048 from clinicaltrials.gov accessed on 06/04/2017).
As argininosuccinate lyase deficiency impairs two essential metabolic pathways, metabolic consequences of the two pathways need to be considered in tailoring adequate therapeutic approaches. Nitric oxide deficiency is not affected by the conventional treatment which focuses on controlling hyperammonaemia and correcting hypoargininaemia. Similarly a liver-targeted therapy alone such as liver (76, 120, 390, 391) or hepatocyte transplantations (105) result in only mild neuropsychological improvement with persisting developmental delay in patients transplanted after 12 months of age and with follow-up after transplantation ranging from 1 to 5 years.
If the neuropathology is partially caused by impaired cerebral nitric oxide metabolism, liver transplantation, whilst curing the urea cycle, would not be expected to correct the systemic citrulline-nitric oxide cycle defect. Concordant with the above, successful liver-targeted gene therapy in the AslNeo/Neo mouse did not correct extra- hepatic features such as defective nitric oxide-mediated vascular relaxation (85). There is therefore a need for new therapeutic approaches to correct nitric oxide deficiency seen in patients affected by argininosuccinic aciduria. This could include the use of an enriched nitrate diet (27), nitrate therapy (85, 310) or multiorgan- targeted gene or cell therapy.
3.8 Conclusion
This study expands the phenotypic spectrum of the disease in the UK highlighting new neurological and systemic features that have not been reported previously. It also affords a better understanding of the genotype-phenotype correlation with the report of new pathogenic mutations. 174 Chapter 3 - Results
This study accurately describes the poor and homogeneous long-term neurological outcome in patients affected by argininosuccinic aciduria irrespective of their phenotype (early- or late-onset) and the age at the start of conventional treatment. This demonstrates the insufficient benefit of the current therapeutic approach for the preservation of the neurocognitive outcome. Neuroimaging data in this study and data from the literature support the role of nitric oxide in the pathophysiology of the disease. Although the neuropathology in argininosuccinic aciduria and the respective role of nitric oxide warrants further scientific work, the findings presented in this study offer avenues to harness alternative therapeutic strategies for argininosuccinic aciduria.
175 Chapter 4 - Results
4. ASSAY DEVELOPMENT FOR FUNCTIONAL CHARACTERISATION OF THE UREA CYCLE
4.1 Introduction
The development of alternative therapeutic strategies for argininosuccinic aciduria requires methods to accurately assess efficacy with relevant endpoints. In urea cycle defects, amino acids involved in the urea cycle are useful biomarkers commonly used diagnostically to define the deficient enzymatic step and for monitoring metabolic control during regular follow-up visits. Concentrations of these amino acids are either increased if breakdown is deficient or decreased if synthesis is inadequate, depending on their position, before or after the metabolic block, respectively. Measurement of enzymatic activities is another tool of paramount interest to evaluate the residual function of a rate-limiting step in the pathway.
Over the last 30 years, advances in mass spectrometry, a technology used to identify compounds based on their molecular weight, has led to regular applications in daily laboratory practice. For instance, when coupled to a chromatography based method additional separation of compounds can be achieved enabling very acurate identification and quantification of multiple metabolites in a reduced amount of tissue or biological fluid, which is now routinely used in pathology laboratories and for newborn screening (392).
4.1.1 Detection of analytes by tandem mass spectrometry
Mass spectrometry is an accurate method to identify and quantify analytes on the basis of their mass to charge ratio (m/z). The analyte molecule is converted into an ionised (charged) state and the resulting ions are analysed.
176 Chapter 4 - Results
The most common method of ionisation is electrospray ionisation (ESI) developed by Fenn, who was awarded the Nobel prize in 2002 for this pioneering work (393). It is well suited for the analysis of most mildly polar biological molecules. The liquid sample is pumped through a metal capillary maintained at 3 to 5 kiloVolts and nebulised to form a very fine spray of highly charged droplets, which rapidly evaporate in the flow of heated dry nitrogen. As the solvent evaporates, the droplet becomes smaller and will reach its Rayleigh limit when the electrostatic repulsion of charges becomes more powerful than the surface tension of the droplet. The droplet liberates further smaller droplets called the Coulomb fission, which will subsequently undergo the same process. This results in the formation of a gas-phase containing ionised analytes as the electrical charges on the droplet are transferred to the analytes (394). Due to their charge, ions will be attracted to the ion source and transferred into the vacuum of the mass spectrometer for analysis (Figure 42). Positive (mass + H+) or negative (mass – H+) ionisation occurs depending on the chemistry of the molecule and its ionisational groups. ESI is a “soft” ionisation method with little energy being transferred to the analyte and hence minimal fragmentation (395).
The ionised analytes present in the gas-phase are then focussed in a triple quadrupole mass spectrometer. A quadrupole contains 4 parallel metal rods, where varying radiofrequency and direct current voltages are applied producing an oscillating electric field, to convey a specific analyte along the quadrupole according to its m/z value. Varying voltages allow screening for different m/z values. A triple quadrupole system allows selection of a precursor or “parent” ion in the first quadrupole. This ion then undergoes fragmention by collision with an inert gas such as nitrogen or argon in the collision cell, and analysis of the product or “daughter” ions or fragments produced by the collision, are “focused” onto a detector by a second set of quadrupoles. This process is called collision-induced dissociation. Every precursor ion depending on its chemical formula will have a preferred and reproducible fragmentation pattern with specific product ions. The first and second quadrupoles can either be used in scan mode, allowing screening of any ion passing through, or fixed mode with detection limited to specific m/z values (395). The latter, a process referred to as multiple reaction monitoring (MRM) involves defining and optimising m/z transitions, which are applied in the first and second quadrupoles and
177 Chapter 4 - Results allows the analysis of various compounds of interest in the same sample run. Whilst identification of the parent ion using mass spectrometry in multiple reaction monitoring mode is highly sensitive, the probability of the presence of another biochemical in biological fluids sharing the same m/z is high. However adding a second step of m/z selection for fragmented ions arising from the initial “parent” ion adds a much higher level of sensitivity and specificity (Figure 43). The combination of this with liquid chromatography, which will initially separate the molecules present in the sample, further increases the specificity of the method.
Atmospheric pressure Vacuum
HPLC column Mass spectrometer Metal-coated Solvent droplets glass capillary Analyte ions released from solvent
Hea,ng nitrogen drying gas
3.5 kV Poten,al difference
Figure 42. Schematic of electrospray ionisation
The analytes are eluted from the HPLC column by liquid chromatography into the electrospray capillary, which has a high potential difference. The sample is subsequently sprayed as small charged droplets with a surface charge of the same polarity to the charge of the electrospray capillary. These droplets are repelled from the capillary and move towards the mass spectrometer due to the electric charges present. During this trajectory, the solvent evaporates and ionised molecules enter the mass spectrometer. A heated inert gas is applied to facilitate the evaporation. The electrospray ionisation transforms analytes in solution into gas-phase ions.
178 Chapter 4 - Results
Quadrupole 1 Collision cell Quadrupole 2 Detector
Argon
Figure 43. Schematic of multiple reaction monitoring analysis in a triple quadrupole mass spectrometer
The compound of interest is separated from other compounds (‘the matrix’) according to its m/z ratio in quadrupole 1. After this first selection, the ‘parent’ ion is fragmented in the collision cell. Fragment ion(s) or product ion(s) with specific m/z ratios are selected in the second quadrupole and passed onto the detector.
Hence liquid chromatography-tandem mass spectrometry (LC-MS/MS) is a method of choice to separate, identify and quantify biochemicals accurately and reproducibly and is routinely used in diagnostic laboratories. An example of the use of LC-MS/MS is in the analysis of amino acids which gives better sensitivity than methods such as ion-exchange liquid chromatography alone (103).
This chapter discusses the development and optimisation of a method for the analysis of the urea cycle-related amino acids: arginine, citrulline, glutamate, glutamine, ornithine and argininosuccinic acid and also the anhydride form(s) of argininosuccinic acid. Argininosuccinic acid has the possibility to create a structure having a five-membered ring when an acyl group binds an amine group located within the aliphatic chain of the molecule. This creates the loss of one molecule of water, creating an anhydride derivative. Two anhydrides of argininosuccinic acid have been identified (396) (Figure 44). These compounds have the same molecular mass and are readily interconvertible (397). This biochemical reaction happens spontaneously with formation being increased by heating. It has been estimated that these anhydride forms account for 1% (398) to 5-10% of the argininosuccinic acid 179 Chapter 4 - Results present in urine samples (397), although high percentages might reflect a false increase caused by experimental conditions (398).
Argininosuccinic acid (290) Anhydride 1 (272)
H2O
Anhydride 2 (272)
H2O
Figure 44. Formation of the 2 anhydrides of argininosuccinic acid
The molecular weight of each molecule is indicated in brackets. H: Hydrogen, N: Nitrogen, O: Oxygen.
180 Chapter 4 - Results
4.1.2 Separation of analytes by chromatography
Chromatography is an accurate and reproducible method used for the separation of compounds based on their physicochemical properties, which was developed for separation of plant pigments by Tswett, a Russian botanist in the 1900s (399). The term “chromatography” was invented from the association of two Greek words “Chrom” and “Graph” meaning “colour” and “writing” as the method initially was developed to separate green chlorophyll, orange carotenes and yellow xanthophylls. The separation is obtained with a system comprised of 2 “phases”: a mobile phase (liquid or gas) and a stationary phase as a fixed porous sorbent. The stationary phase can retain the analyte based on i) selective non-covalent interactions based on electromagnetic affinity called “affinity chromatography”, ii) covalent interactions and electric polarity between functional groups of molecules named “ion exchange chromatography” or iii) molecular size called “size-exclusion chromatography”. The modification of composition of the mobile phase through a progressive gradient allows binding of the compounds to the stationary phase then sequential elution based on physicochemical properties of each analyte. Each analyte, depending on the specific mobile phase, will bind differently to the stationary phase and will be eluted and detected at a specific reproducible time (retention time) for a particular method (400).
181 Chapter 4 - Results
4.2 Development of a liquid chromatography - tandem mass spectrometry (LC-MS/MS) method for the quantification of urea cycle-related amino acids
4.2.1 Detection of amino acids using tandem mass spectrometer
Derivitisation of amino acids with 9-fluorenylmethyl chloroformate (FMOC) and identification by LC-MS/MS
The FMOC derivatisation protocol is described in Section 2.14.3.
Each fraction was infused directly into the mass spectrometer and m/z transitions were calculated for each precursor (“parent”) and product (“daughter”) ions. Cone voltage and collision energies were optimised to maximise the detection of both precursor and product ions for each analyte (Table 15). Whilst this protocol resulted in good sensitivity for the majority of the amino acids, the sensitivity for argininosuccinate and its anhydrides was very low. Derivatisation was attempted at different pHs (pH range from 7 to 10.4) to see if sensitivity could be improved however no increase in sensitivity was achieved. An attempt was made to convert the argininosuccinate to its anhydride form(s) to see if this would increase the sensitivity by heating the solution for one hour at 100°C. This increased 5-fold the amount of the arginosuccinate anhydrides detected but decreased the sensitivity of the other amino acids. Hence it was decided not to use the FMOC derivatisation protocol but to analyse the amino acids underivatised as summarised in Section 2.14.5 and extensively described in Section 4.2.3.
Optimisation of parameters used for ESI-MS/MS analysis of underivatised amino acids
182 Chapter 4 - Results
Published reports have shown that it is possible to analyse underivatised amino acids using electrospray ionisation tandem mass spectrometry in plasma, urine, cerebrospinal fluids and leukocyte pellets (401). Some reports have also highlighted better precision using this methodology for detection of arginine, citrulline and ornithine in dried blood spots (402).
Similarly to the analysis of the derivatised amino acids (see Section 2.14.3), solutions of underivatised amino acids and deuterated analytes were infused into the mass spectrometer. Optimal m/z transitions for precursor and product ions for each analyte were identified by infusing a stock solution of 0.1 and 1 mM of each analyte into the mass spectrometer without and with the presence of collision gas, respectively in either positive or negative ion mode. The best detection was observed in positive ion mode. Using the multiple reaction monitoring mode, cone voltage and collision energies which resulted in maximal ion detection for both precursor and product ions were identified (Table 15). In order to calculate the theoretical m/z transition to monitor, one supplementary unit is added to the molecular weight of each analyte, which corresponds to the protonised molecule, to obtain the mass of the parental ion.
Retention Precursor Cone Collision Product ion Analyte time ion voltage energy (m/z) (minutes) (m/z) (Volts) (Volts) L-Ornithine 5.73 132.93 69.94 17 17 L-Ornithine-d7 5.73 139.98 76.99 16 15 L-Arginine 5.79 175.10 69.94 28 20 L-Arginine-13C6 5.79 181.10 73.94 28 20 ASA anhydrides 6.07 273.12 69.87 40 29 Argininosuccinate 6.26 291.02 69.87 40 29 L-Glutamine 6.51 147.06 83.94 18 12 L-Glutamine-13C2 6.51 149.06 84.94 18 12 L-Glutamate 6.57 148.02 83.82 20 11 L-Glutamic acid-d5 6.57 153.06 88.86 20 11 L-Citrulline 6.60 176.10 69.94 30 18 L-Citrulline-d7 6.60 183.15 76.99 30 18 Table 15. Optimised mass spectrometry parameters used for the detection of underivatised urea cycle amino acids in positive ion mode
183 Chapter 4 - Results
4.2.2 Optimisation of mobile phase conditions used for LC-MS/MS analysis
In this work, a technique known as “reverse phase chromatography” was used. The initial descriptions of the chromatography method developed by Tswett used mobile and stationary phases that were non polar and polar, respectively. Conversely, reverse phase chromatography uses a polar mobile phase and non-polar stationary phase, justifying the name. In this method, compounds are separated via i) hydrophobicity exacerbated by the polar mobile phase, with the more hydrophobic compound being retained by the solid phase for longer, and ii) an absorption process necessary for the sample molecules to bind via non-covalent interactions (i.e. van der Waals forces) with the non-polar sorbent of the column.
Thus the mobile phase plays a key-role acting both as a vehicle to move the compounds as well as contributing to the separation, with different compositions and gradients applied regulating chemical properties of the analytes. Variations in pH and polarity will modify the attachment of the molecules of interest to the stationary phase. Hence during method development for the LC-MS/MS analysis of the urea cycle amino acids described in this chapter, several mobile phases were tested with different gradients to obtain an optimal separation. The stationary phase selected was a reversed-phase C18 column as described in Section 2.14.4.
Because of the polarity of the amino acids, the initial starting mobile phase(s) for the separation required a high percentage of aqueous solution. Various aqueous mobile phases were tested: 4% ammonium acetate with or without 0.01-1% formic acid; 0.045% trifluoroacetic acid; 0.1% formic acid; 0.5 mM heptafluorobutyric acid; 3.7% acetic acid (403) with and without 1.3% heptafluorobutyric acid; 0.12% acetic acid (404). Amino acids were eluted from the column by increasing the percentage of the organic mobile phase. Various organic phases were investigated: acetonitrile with or without 0.01-0.1% formic acid; 80% acetonitrile containing 0.05% trifluoroacetic acid; methanol. Optimal separation and sensitivity were achieved using 3.7% acetic acid (pH 2.4) and methanol as aqueous and organic mobile phases, respectively. A 25 minute gradient profile modified from Piraud et al (401) was used (Table 16). Each gradient step was linear.
184 Chapter 4 - Results
The liquid chromatography-tandem mass spectrometry method developed showed reproducible separation and retention times for all analytes (Figure 45)
3.7% Methanol Flow Time (min) Acetic Acid (%) (mL/min) (%) 0 0 100 0.2 1 15 85 0.2 6 15 85 0.2 8 25 75 0.2 9 95 5 0.2 15 95 5 0.2 16 0 100 0.2 17 0 100 0.5 25 0 100 0.5 Table 16. Optimised mobile phase gradient profile
185 Chapter 4 - Results
5.79 Arginine (175.1 > 69.94)
5.79 13C6-Arginine (181.1 > 73.94)
6.26 Argininosuccinic acid (291.02 > 69.87)
6.14 Argininosuccinic acid anhydrides (273.12 > 69.87)
6.60 Citrulline (176.1 > 69.94)
6.60 D7-Citrulline (183.15 > 76.99)
6.57 Glutamate (148.02 > 83.82)
6.57 D5-Glutamate (153.06 > 88.86)
6.51 Glutamine (147.06 > 83.94)
6.51 1,2-13C2-Glutamine (149.06 > 84.94)
5.73 Ornithine (132.93 > 69.94)
5.73 D7-Ornithine (139.98 > 76.99)
Figure 45. Chromatogram of each analyte and their internal standards
The amount of each analyte injected was 2 nmol except for L-Glutamine-1,2-13C2 and L-Ornithine-2,3,3,4,4,5,5-d7,where 0.66 nmol were used. For each analyte, the
186 Chapter 4 - Results transition monitored is mentioned as follows: precursor masss > product mass. At the top of each peak, the retention time (presented in minutes) is indicated. In the L- citrulline channel, there is a small amount of cross-talk observed with a small peak of L-arginine detected: this can be explained as the parent ion transitions for these amino acids only differ by one mass unit. However as L-arginine and L-citrulline have different retention times (5.79 and 6.60, respectively), they are readily distinguished from each other.
4.2.3 Extraction of amino acids from dried blood spots
To facilitate mouse blood sampling, a method for the quantification of amino acids from dried blood spots was established to simplify collection and to minimise the amount of blood needed for analysis. It is recognised that amino acid values measured in dried blood spots differ from plasma values however this approach has been shown to be reliable for monitoring amino acids (402, 405).
The protocol for the extraction of amino acids from blood spots (as described in Section 2.14.5) was adapted from Schulze et al (406). Whole blood was spiked onto a Guthrie card at time of collection and allowed to dry at room temperature for 24 hours before being stored at -20°C in a foil bag with desiccant. On the day of analysis, a punch from the centre of the dried blood spot was placed in a 1.5 mL microcentrifuge tube, which contained methanol and a mixture of internal standards including 2 nM of L-Arginine-13C, L-Citrulline-2,3,3,4,4,5,5-d7, L-Glutamic-2,3,3,4,4- d5 Acid, and 0.66 nM L-Glutamine-1,2-13C2 and L-Ornithine-2,3,3,4,4,5,5-d7. The concentrations of the internal standards used were similar to the expected range of concentrations of the analytes to be analysed within the samples. The sample was sonicated in a sonication bath. The supernatant was removed and centrifuged at 16,000 g for 5 minutes prior to injection into the liquid chromatography-tandem mass spectrometry system. Dried blood spots from anonymised healthy human controls were used to optimise different parameters of the extraction method (Figures 46).
Glutamine can be converted into glutamate either spontaneously or via the catalytic activity of glutaminase (407). This spontaneous deamidation can occur during
187 Chapter 4 - Results storage (408). Taking into account this natural deamination, it was decided to measure glutamate and glutamine separately and present results together. This was considered as acceptable as glutamate and glutamine are part of the same buffering pathway to prevent hyperammonaemia (See Section 1.2.2).
As expected the levels of argininosuccinic acid and the associated anhydrides were not detectable in the dried blood spots from healthy individuals. As AslNeo/Neo mice exhibit a severe failure to thrive, a punch of 3.2 mm diameter corresponding to a blood volume of 3.4 μL (404) was chosen to facilitate blood collection.
The following parameters were investigated for optimisation of the extraction: i) Extraction volume of methanol. The effect of increasing volumes of methanol (100 to 500 μL) was tested but this did not affect the concentration of analytes detected (Figures 46A-F). Therefore the volume of extraction was minimised to 100 μL in order to concentrate the amino acids extracted from the dried blood spot and to increase the signal detected on the mass spectrometer. ii) Duration of extraction (Figures 46G-L). iii) Effect of sonication on extraction (Figures 46G-L).
Sample extraction was tested at different time points either with sonication (15 minutes to 1 hour) or without sonication (30 minutes to 7 hours). Although no obvious benefit of sonication was observed (Figures 46G-L) in contrast to what has been described in some publications (402), it was decided to maintain this step in the final protocol as this step has been incorporated in some protocols of newborn screening involving dried blood spots (409, 410). Therefore the minimum time of extraction with sonication tested (i.e. 15 minutes) was chosen.
188 Chapter 4 - Results
A B 50 40
40 30 mol/L mol/L) 30 µ (µ 20 20 10 Arginine ( Arginine 10 Citrulline
0 0 100 200 333 400 500 100 200 333 400 500 Volume of methanol (µL) Volume of methanol (µL)
C D 100 150
80 mol/L) mol/L) 100 60 (µ (µ 40 50
Ornithine 20 Glutamine
0 0 100 200 333 400 500 100 200 333 400 500 Volume of methanol (µL) Volume of methanol (µL)
E F 300 400
300 mol/L) 200 + mol/L) (µ (µ 200 100 Glutamine 100 glutamate Glutamate 0 0 100 200 333 400 500 100 200 333 400 500 Volume of methanol (µL) Volume of methanol (µL)
189 Chapter 4 - Results
G H 60 30
40
mol/L) 20 mol/L) µ (µ
20 10
( Arginine Citrulline
0 0.15 0.3 0.5 1 2 4 7 0 0.15 0.3 0.5 1 2 4 7 Time of extraction (hours) Time of extraction (hours) Punch 3mm_US Punch 3mm_No US Punch 3mm_US Punch 3mm_No US
I J 100 150
80 mol/L) mol/L) 100 60 (µ (µ 40 50
Ornithine Ornithine 20 Glutamine
0 0 0.15 0.3 0.5 1 2 4 7 0.15 0.3 0.5 1 2 4 7 Time of extraction (hours) Time of extraction (hours) Punch 3mm_US Punch 3mm_No US Punch 3mm_US Punch 3mm_No US
K L 300 400
300 + + mol/L)
200 mol/L) (µ (µ 200
100 Glutamine 100 glutamate Glutamate
0 0 0.15 0.3 0.5 1 2 4 7 0.15 0.3 0.5 1 2 4 7 Time of extraction (hours) Time of extraction (hours) Punch 3mm_US Punch 3mm_No US Punch 3mm_US Punch 3mm_No US
Figure 46. Optimisation of extraction of analytes from a 3.2 mm dried blood spot punch
The volume of methanol used to extract the analytes from the dried blood spots was varied (A-F) as was the time of extraction with and without sonication (G-L). (A,G) L- arginine, (B,H) L-citrulline, (C,I) L-ornithine, (D,J) L-glutamine, (E,K) L-glutamate, (F,L) summed L-glutamine and L-glutamate.
190 Chapter 4 - Results
The final method is summarised in Figure 47 and Section 2.14.5. A 3.2 mm diameter punch from the centre of the dried blood spot (corresponding to a blood volume of 3.4 μL (404)) was placed in a 1.5 mL microcentrifuge tube in a 100 μL volume containing 90 μL of methanol and 10 μL of a mixture of internal standards containing 2 nM of L-Arginine-13C, L-Citrulline-2,3,3,4,4,5,5-d7, L-Glutamic- 2,3,3,4,4-d5 Acid, and 0.66 nM L-Glutamine-1,2-13C2 and L-Ornithine-2,3,3,4,4,5,5- d7. The sample was sonicated in a sonication bath for 15 minutes before adding 400 μL of methanol resulting in a 4:1 volume dilution. The supernatant was removed and centrifuged at 16,000 g for 5 minutes prior to injection into the liquid chromatography- tandem mass spectrometry.
Figure 47. Final optimised protocol for the measurement of the concentration of the amino acids of interest in a dried blood spot.
191 Chapter 4 - Results
4.2.4 Quantification of amino acids using stable isotopes
Various different parameters can alter reproducibility of day-to-day quantification by mass spectrometry including cleanliness of the ion source, ion optics and collision cell, ion source flow rates, collision cell pressure and mass spectrometer vacuum (395). In order to correct for this variability, mass spectrometry methods routinely add internal standards of known concentrations to the sample to allow for accurate quantification. The stable isotopes L-Arginine-13C, L-Citrulline-2,3,3,4,4,5,5-d7, L- Glutamic-2,3,3,4,4-d5 Acid, L-Glutamine-1,2-13C2 and L-Ornithine-2,3,3,4,4,5,5-d7 were used as internal standards. 2 nM of L-Arginine-13C, L-Citrulline-2,3,3,4,4,5,5- d7, L-Glutamic-2,3,3,4,4-d5 Acid, and 0.66 nM L-Glutamine-1,2-13C2 and L- Ornithine-2,3,3,4,4,5,5-d7 were spiked in all samples and into the points of the standard curve. For argininosuccinic acid, no stable isotope was available so this analyte was quantified using the internal standard L-Citrulline-2,3,3,4,4,5,5-d7.
To validate the method, linearity, precision studies and assessment of matrix effect were performed.
Linearity studies
Linearity studies are performed to ensure that quantification of the compound of interest is accurate. In the case of saturation of detectors for example, linearity is not observed anymore and the sample needs to be diluted to obtain a signal, which is in the range of linearity.
The linearity of the response of the LC-MS/MS using the mobile phase described in Section 4.2.2 was investigated. Amino acids were prepared at a concentration of 0 to 500 µM in methanol with 13 intermediary points: 0, 0.1, 0.2, 0.5, 1, 2, 5, 10, 20, 50, 100, 200, 500 μM. This range was chosen to cover the expected concentrations of the amino acids of interest in wild-type and AslNeo/Neo mice. Linear standard curves were obtained over this range for every compound with r2 > 0.99 for all amino acids except L-glutamine (r2 > 0.98) (Table 17). For this compound, linearity was achieved between 0 to 200 μM before reaching a plateau suggesting saturation of the detector 192 Chapter 4 - Results
(Figure 48). Therefore every sample analysed was run twice, undiluted for quantification of all amino acids except glutamine and after 1:10 dilution in methanol for accurate measurement of glutamine.
Regresssion parameters Slope X-intercept r2 Arginine 0.3728 -0.0938 0.9960 Citrulline 0.6087 -0.0794 0.9983 Glutamate 2.5111 -0.3541 0.9959 Glutamine 0.4118 1.1826 0.8682 Ornithine 1.5327 0.0222 0.9951 Argininosuccinic acid 2.2607 -0.2273 0.9979 Argininosuccinic acid anhydrides 0.2573 0.0999 0.9979 Table 17. Linearity parameters for each analyte
Compounds were diluted in methanol.
193 Chapter 4 - Results
40 20 Arginine Citrulline
30 15 y = 0.0609x - 0.0794 y = 0.0378x - 0.1348 R² = 0.99827 standard ra?o
standard ra?o R² = 0.996 20 10 PAR PAR internal internal / 10
/ 5
0 Analyte 0 Analyte 0 100 200 300 400 500 0 100 200 300 400 500 Concentra-on (μM) Concentra-on (µM)
150 Ornithine 100 Glutamate
80 y = 0.2511x - 0.3541 y = 0.1533x + 0.0222 standard ra?o 100 R² = 0.99591 standard ra?o 60 R² = 0.99507 PAR PAR
internal 40 50 internal / / 20 Analyte
0 Analyte 0 0 100 200 300 400 500 0 100 200 300 400 500 Concentra-on (µM) Concentra-on (µM ) 25 Glutamine 20 Glutamine 20 15 y = 0.0758x + 0.2782 15
standard ra?o R² = 0.98628 standard ra?o
PAR 10 10 PAR 2 internal y = -0.0001x + 0.0978x + 0.0935 internal
/ 5 5 R² = 0.9926 / 0 0 Analyte 0 100 200 300 400 500 Analyte 0 50 100 150 200 250 Concentra-on (µM) Concentra-on (µM)
150 Argininosuccinic acid Argininosuccinic acid 15 anhydrides y = 0.2261x - 0.2273
standard ra?o 100 R² = 0.99788 10 y = 0.0257x + 0.0999
standard ra?o R² = 0.99788 PAR PAR internal 50 5 / internal /
Analyte 0 0 0 100 200 300 400 500 Analyte 0 100 200 300 400 500 Concentra-on (µM) Concentra-on ((µM)
Figure 48. Calibration curves of urea cycle amino acids
Three different calibration curves in methanol per analyte are presented on the same graph for each analyte. Arginine, citrulline, glutamate, ornithine, and argininosuccinic 194 Chapter 4 - Results acid and argininosuccinic acid anhydrides were shown to be linear between 0 to 500 µM and glutamine from 0 to 200 µM. Slope and X-intercept are calculated after plotting the mean values for each intermediary point.
Precision studies
Precision studies were performed according to international guidelines (International conference on Harmonisation, Harmonised Tripartite Guideline, Validation of analytical procedures: text and methodology Q2(R1). 1994). These guidelines are available at http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Quality/Q2_ R1/Step4/Q2_R1_Guideline.pdf). Precision studies were adapted from Wang et al (409) and performed with whole blood spiked onto a Guthrie card with 3 different known concentrations for each amino acid of interest: low (50 μM), medium (300 μM) and high (1,500 μM). Unspiked blood from the same person taken at the same time was spotted onto a Guthrie card to measure the endogenous concentrations of amino acids. Studies included: i) Intrabatch coefficient of variance (CV). This was measured in 10 samples run on the same day. The coefficient of variance was calculated as follow: CV (%) = (standard deviation / mean) x 100. ii) Interbatch coefficient of variance. This was calculated from 5 measurements of the same sample analysed on different days with a minimum of a week interval between 2 repetitive analyses. iii) Recovery studies. Ten samples were run on the same day and the recovery study was calculated for each amino acid of interest at the low, medium and high concentration spiked.
Recovery (%) = (Measured concentration – endogenous concentration / theoretical concentration of spiked analyte) x 100.
Results are presented in Table 18. Intra and interbatch coefficients of variances were all below 10% except for argininosuccinic acid where higher values were seen. Precision in measurement of argininosuccinic acid showed values below 10% for
195 Chapter 4 - Results intrabatch coefficient of variation (also called within-run precision) but higher values for interbatch coefficient of variation (also called between-run precision). At medium and high concentrations of argininosuccinic acid (300 and 1500 μM, respectively), interbatch coefficient of variation was below 15%, which is acceptable for assays used for method development of investigational medicinal products for the European Medicines Agency (411) and Food and Drug Administration (412). However at low concentration (50 µM), the value was 27%, which is similar to values published (9 to 22%) for urea cycle amino acids in some other dried blood spot studies (402, 413).
Amino acid amound 50 µM 300 µM 1500 µM added to whole blood
Recovery Intrabatch Interbatch Recovery Intrabatch Interbatch Recovery Intrabatch Interbatch Parameter tested (%) CV (%) CV (%) (%) CV (%) CV (%) (%) CV (%) CV (%)
Arginine 99 5 5 98 2 4 97 6 5 Citrulline 104 2 9 116 3 2 106 7 3 Glutamate 101 2 6 107 2 2 102 4 4 Glutamine 99 1 0 99 2 2 96 2 1 Ornithine 99 5 3 101 3 2 99 3 3 Argininosuccinic 74 9 27 110 3 14 103 0 12 acid and anhydrides Table 18. Recovery, intra- and inter-batch coefficient of variations for each analyte
Three different concentrations were measured: low 50 μM, medium 300 μM and high 1500 μM. Amino acids were spiked onto a Guthrie card. CV: coefficient of variance.
iv) Limits of detection and quantification. 5 standard curves were run on the same day for each analyte. Two different methods of calculation were used (414), based on: a) signal-to-noise ratio calculating the standard deviation of the calibration curve (STN method). This method is based on comparison of signals from samples of a known concentration with the blank (i.e noise). The recommended signal-to-noise ratio for detection and quantitation are 3:1 and 10:1, respectively. b) the standard deviation, the response and the slope, based on calculation of the standard deviation of the intercept and slope (SD method). The 5 lowest
196 Chapter 4 - Results concentrations of the standard curve excluding the blank sample were used for calculation (range of 0.1 to 2 µM).
The limit of detection (LOD) is the lowest analyte concentration that can be distinguished reliably from a blank sample and was calculated as follows:
STN method: LOD = 3 x standard deviation of low concentrations / slope of the calibration curve
SD method: LOD = 3 x standard deviation of X-intercept / slope of the calibration curve
The limit of quantitation (LOQ) is the lowest analyte concentration that can be determined quantitatively with suitable precision and accuracy.
STN method: LOQ = 10 x standard deviation of low concentrations / slope of the calibration curve
SD method: LOQ = 10 x standard deviation of X-intercept / slope of the calibration curve
Limits of detection and quantitation for the amino acids of interest are presented in Table 19.
197 Chapter 4 - Results
Parameters tested Limit of detection Limit of quantification Method of calculation STN SD STN SD Arginine 2.70 0.39 8.99 1.30 Citrulline 1.75 0.24 5.82 0.81 Glutamate 2.67 0.37 8.88 1.23 Glutamine 16.41 2.32 54.70 7.74 Ornithine 2.93 0.41 9.76 1.35 Argininosuccinic acid 1.96 0.28 6.52 0.93 Argininosuccinic acid anhydrides 1.96 0.28 6.53 0.93 Table 19. Limits of detection and quantification for each analyte
Two methods were used: standard deviation of the calibration curve (STN) and standard deviation of the intercept and slope (SD).
Matrix effect
Other compounds co-eluting with the analytes of interest can have a matrix effect, also called ion suppression. If this occurs this “environment” might modify the ionisation and interactions between biochemicals, the matrix itself and analytes. A complex matrix like blood contains many biological compounds, which might interfere. Therefore it is important to ensure that the retention times of the analytes of interest are in regions of the chromatogram with little or no matrix ion suppression (401).
In this case, the matrix is the whole blood spiked on the Guthrie card. If the matrix contains ions with identical m/z transitions eluting at the same time as the molecule of interest, ion suppression can occur with competitive inhibition in the detector. For each amino acid, 2 standard curves with 11 different concentrations from 0.1 to 200 µM were run; i) in methanol and ii) mixed with whole blood and spiked on Guthrie cards.
The matrix effect was calculated as published previously (415):
Matrix effect (%) = ((peak area of the amino acid spiked in the DBS – peak area of the endogenous amino acid) / peak area of the amino acid in methanol) x 100.
198 Chapter 4 - Results
Matrix effects were negligible for L-arginine (85%), L-citrulline (98%), L-ornithine (104%) and L-glutamate (93%). The matrix effects however were particularly marked with regard to L-glutamine (205%) and argininosuccinate (33%) with potentiation and ion suppression, retrospectively (Figure 49). Although an important matrix effect was observed with argininosuccinate and L-glutamine, recovery experiments indicated that the internal standard dilution method was able to compensate for these matrix effects (415). However as L-citrulline is structurally different compared to argininosuccinic acid and its anhydrides and is eluted at a different time, this could affect the precision and explain partly why the matrix effect is higher for argininosuccinic acid and its derivatives. Another publication had previously used an internal standard derived from Lysine (401).
199 Chapter 4 - Results
L-Citrulline MATRIX EFFECT: LINEARITY 9" L-Arginine 14" y"="0.0624x"*"0.1224" 8" y"="0.0371x"*"0.0872" 12" R²"="0.98632" R²"="0.99338" STUDIES IN DBS 7" 10" 6" 8" 5" y"="0.0326x"+"0.1003" standard ra)o R²"="0.99118" standard ra)o 4" 6" y"="0.0635x"+"0.0562" R²"="0.99016" 3" 4" internal internal
/ 2" / 2" 1"
Analyte 0" Analyte 0" 0" 50" 100" 150" 200" 250" 0" 50" 100" 150" 200" 250" *1" *2" Concentra)on of analyte Concentra)on of analyte L-Ornithine 60" 40" L-Glutamate y"="0.2647x"+"0.6961" 50" R²"="0.98336" 35" y"="0.1687x"+"0.3672" R²"="0.98798" 30" 40" 25" 30"
standard ra)o 20" standard ra)o y"="0.255x"+"1.8981" 20" R²"="0.98654" 15"
internal y"="0.1803x"+"0.3442" internal 10" /
/ 10" R²"="0.98597" 5" 0" Analyte
Analyte 0" 0" 50" 100" 150" 200" 250" 0" 50" 100" 150" 200" 250" +10" +5" Concentra)on of analyte Concentra)on of analyte
40" Argininosuccinate and anhydrides 50" L-Glutamine
35" y"="0.1149x"+"11.468" y"="0.2273x"*"0.2615" R²"="0.83885" 40" R²"="0.99625" 30"
25" 30"
standard ra)o 20" standard ra)o 20" 15"
internal internal 10" y"="0.0757x"*"0.4966" / 10" / R²"="0.9764" y"="0.0759x"+"0.2678" 5" 0" R²"="0.99308" Analyte Analyte 0" 50" 100" 150" 200" 250" 0" 0" 50" 100" 150" 200" 250" *10" Concentra)on of analyte Concentra)on of analyte Figure 49. Study of impact of matrix effect for each analyte
Standard curves are run in methanol (blue) and spiked in blood then spotted onto a Guthrie card (red), respectively.
200 Chapter 4 - Results
4.3 Development of an LC-MS/MS based assay to measure ASL enzyme activity
4.3.1 Previous published assays described for assessing argininosuccinate lyase activity
Argininosuccinate lyase catalyses the transformation of argininosuccinic acid into L- arginine and fumarate (Figure 50). Several methods have been described previously for the analysis of the activity of argininosuccinate lyase (ASL): i) Direct forward reaction: study of conversion of argininosuccinic acid into fumarate and L-arginine using fluorometric detection of fumarate or using a combined assay with fluorometric detection of urea (85). ii) Direct reverse reaction: quantification of labelled 14C-argininosuccinic acid formation after incubation of L-arginine with labelled 14C-fumarate (105) or 13C-L- Glutamine (106).
Although these methods are published and validated, mass spectrometry is an alternative technology, which offers the advantages of being able to assess enzyme activity in minimal tissue samples. LC-MS/MS methods were therefore developed to enable us to accurately measure ASL activity in the small amounts of tissue that were available.
ASL L-Citrulline ASS L-Arginine Arginase Urea + Argininosuccinic acid + + Aspartate Fumarate L-Ornithine Figure 50. . Steps of the urea cycle measured previously in order to assess argininosuccinate lyase activity
ASS : Argininosuccinate synthase ; ASL : Argininosuccinate lyase.
201 Chapter 4 - Results
4.3.2 Development and optimisation of argininosuccinate lyse activity assay
Organs were perfused at culling then stored as described in Section 2.7.1. Proteins were extracted from each sample as described in Section 2.14.6.
The conditions used for the enzyme assay were similar to those described previously by Bush et al (416) and described in Section 10.2.26. All reactions were performed in a final volume of 100 µL with a known amount of protein. Internal standards L- Arginine-13C and L-Citrulline-2,3,3,4,4,5,5-d7 were added to the buffer at a final concentration of 2 nM. Samples were prepared on ice prior to the assay, which was performed in a water bath at 37°C. The reaction was stopped by the addition of a 4:1 ratio of methanol. Samples were then centrifuged immediately at 16,000 g for 2 minutes and the supernatant analysed by LC-MS/MS as described above in Section 4.2.
Direct forward reaction using D7-citrulline as a substrate
Deuterated argininosuccinic acid is not commercially available therefore the use of a direct forward coupled reaction was initially investigated using L-Citrulline- 2,3,3,4,4,5,5-d7 as a substrate, and monitoring the formation of deuterated arginine (Figures 50 & 51). L-Citrulline-2,3,3,4,4,5,5-d7 was added to the initial buffer at a final concentration of 5 mM. Transitions for d7-, d6-, d5- argininosuccinic acid, the argininosuccinic acid anhydrides, arginine and ornithine derivatives as shown in Figure 51 were monitored.
202 Chapter 4 - Results
L-Citrulline (175) D7-citrulline (182) Daughter ion of Citrulline (69) Daughter ion of D7-citrulline (74)
D7-Argininosuccinic acid (297) Argininosuccinic acid (290) Daughter ion of Daughter ion of D7-argininosuccinic acid argininosuccinic acid (69) (74 or 75)
( )
Arginine (174) D7-arginine (181)
Daughter ion of Daughter ion of arginine (69) D7-arginine (73, 74 or 75)
( )
( )
Figure 51. Structures monitored during ASL assay when D7-citrulline used as a substrate.
203 Chapter 4 - Results
The molecular weight of each molecule is indicated in brackets. Deuterated hydrogens are labelled with an orange asterisk. If the conservation of the deuterated form is uncertain from the predictive biochemistry of the reaction, this asterisk is in brackets. Multiple reaction monitoring was performed in positive ion mode, hence the ions monitored by LC-MS/MS are one supplementary unit greater than the corresponding masses depicted. m/z transitions monitored for D7-citrulline were 183>75, for D7-argininosuccinic acid 298>75 and 298>76, for D7-arginine 182>74, 182>75 and 182>76.
20 µg of liver protein extract was incubated over a 2-hour period in preliminary experiments. Under these conditions no significant increase in any of the argininosuccinic acid intermediates was observed. D7-arginine and D7-ornithine products were detected but these were not proportional compared to the clearance of D7-citrulline observed. It is likely that they were converted into other molecules that were not monitored. Surprisingly, upon analysis, it appeared that some D7-arginine was detected at the start of the incubation period however upon closer examination this was shown to have a different retention time to that expected and indeed this was shown to correspond to that of D7-citrulline (Figure 52). The mass of the precursor ions for D7-arginine and D7-citrulline only differ by 1 mass unit and these compounds have daughter ions with the same mass. It is likely at the concentrations of D7-citrulline used, that there is a small amount of cross-talk between channels which would explain this finding. Because of these troubleshooting findings, it was decided to consider a direct forward ASL reaction using argininosuccinic acid as a substrate.
204 Chapter 4 - Results
D7-Citrulline 40 D7-Argininosuccinic acid D7-Arginine 30 D7-Ornithine D7-Arginine+D7-Ornithine 20
(nmol) Amount 10
0 0 60 120 Time (minutes)
Figure 52. Urea cycle amino acids synthesised during the ASL assay when D7-citrulline was used as a substrate
20 µg of liver protein extract from a wild-type mouse were incubated over a 2-hour incubation period in a buffer containing 5 mM of D7-citrulline. The products of the reaction were analysed by LC-MS/MS.
Direct forward reaction using argininosuccinate as a substrate
Preliminary experiments using unlabelled argininosuccinic acid as the substrate for the ASL assay were also performed. Argininosuccinic acid was added to the reaction buffer in excess, at a final concentration of 10 mM and incubated with 20 µg of liver protein extract for 60 minutes. A 70% increase in the amount of arginine was observed relative to the endogenous arginine present at time 0. Different parameters were then optimised, monitoring both argininosuccinic acid clearance and arginine production. Preliminary experiments with a limited number of samples were initially undertaken (Figure 53). Then the method was further optimised with various time points tested as shown in Figure 54. i) The effect of increasing the liver protein concentration (from 20 to 1,500 µg) on ASL enzyme activity was studied in the presence of 10 mM argininosuccinic acid (Figure 53A). 150 µg was identified as the minimal amount of liver protein extract, which would allow measurable production of arginine. Due to argininosuccinic acid being present in excess, no decrease of argininosuccinic acid was observed.
205 Chapter 4 - Results ii) 150 µg of liver extract was incubated with various concentrations of argininosuccinic acid (ranging from 0.1 to 10 mM) and arginine production and argininosuccinic acid clearance was quantified (Figures 53B and 53C, respectively). At concentrations of 1 and 10 mM argininosuccinic acid, arginine production was detectable however accurate quantification of argininosuccinic acid clearance was only achievable at the lower concentration of 1 mM. At lower concentrations (0.1 and 0.2 mM), argininosuccinic acid clearance was observed but the limited signal detected made the precision of the quantification difficult.
Subsequent exploratory experiments investigating liver ASL activity in wild-type and AslNeo/Neo mice incubated 150 µg of liver lysate with 1 mM argininosuccinic acid for 2 hours. There was a marked difference in both L-arginine production (2.8 versus 0.3 pmol/µg proteins/minute for wild-type and AslNeo/Neo mice, respectively) and argininosuccinic acid clearance (12 versus 2.4 pmol/µg proteins/minute for wild-type and AslNeo/Neo mice, respectively) (Figures 53D and 53E, respectively). iii) Because the clearance of argininosuccinic acid and its anhydrides was 30-fold higher than arginine production in wild-type samples, it was hypothesised that arginine was being rapidly metabolised to ornithine or nitric oxide by arginase or nitric oxide synthase, respectively (Figure 53F). To prevent any conversion of L-arginine that was being produced, (2S)-(+)-amino-5-iodoacetamidopentanoic acid (AIPA) (Enzo Life Sciences, Farmingdale, NY, USA), an arginase inhibitor, was added at 0.1 mM and 1 mM as described previously (417) (Figure 53G). Adding AIPA did not result in increased concentrations of L-arginine at either the recommended or higher concentration (0.1 mM and 1 mM, respectively). No inhibitor of nitric oxide synthase was tested. It was therefore decided to assess argininosuccinate lyase activity based on argininosuccinic acid clearance rather than arginine synthesis.
206 Chapter 4 - Results
20 µg 3 A 150 µg 1500 µg 2
1 Arginine (nmol) Arginine
0 0 20 40 60 Time (minutes)
3 B C 3500
2 100
1 50 Arginine (nmol) Arginine + anhydrides (nmol) + anhydrides Argininosuccinic acid acid Argininosuccinic
0 0 0 20 40 60 0 20 40 60 Time (min) Time (minutes) ASA 10mM ASA 1mM ASA 0.2mM ASA 0.1mM ASA 10mM ASA 1mM ASA 0.2mM ASA 0.1mM
D 60 E 400
300 40
200
20
Arginine (nmol) Arginine 100 anhydrides (nmol) Argininosuccinic acid + acid Argininosuccinic 0 0 0 30 60 90 120 0 30 60 90 120 Time (minutes) Time (minutes) Neo/Neo WT Asl Neo/Neo Blank WT Asl Blank
F 500 15 G 300 Argininosuccinic acid + anhydrides 400 Arginine(nmol) Arginine 10 300 Ornithine 200 Arginine + Ornithine 200 5 nmol 100
+ anhydrides (nmol) + anhydrides 100 Argininosuccinic acid acid Argininosuccinic
0 0 0 20 40 60 Time (minutes) 0 0 20 40 60 Argininosuccinic acid + anhydrides Time (minutes) L-arginine
Figure 53. Development of argininosuccinate lyase enzyme assay
(A) L-arginine synthesis after incubation of different amounts of wild-type mouse liver lysate in the presence of 10 mM argininosuccinic acid; (B) arginine synthesis and (C) argininosuccinic acid clearance when wild-type liver lysate is incubated with various
207 Chapter 4 - Results argininosuccinic concentrations (0.1 to 10 mM). (D) L-arginine synthesis and (E) argininosuccinic acid clearance in wild-type and AslNeo/Neo mice after incubation of 150 µg of proteins from liver lysate with 1 mM argininosuccinic acid. (F) Comparison of argininocsuccinic acid clearance and arginine synthesis after incubation of 150 µg of proteins from liver lysate of wild-type mouse liver lysate with 1 mM argininosuccinic acid. (G) Argininosuccinic acid clearance and arginine and ornithine synthesis after incubation of 150 µg of proteins from liver lysate of wild-type mouse liver lysate with 1 mM argininosuccinic acid in the presence of arginase inhibitor (1 mM (2S)-(+)-Amino-5-iodoacetamidopentanoic acid (AIPA)).
In some cases, especially in young AslNeo/Neo mice, we were not able to obtain 150 µg of proteins from liver lysate per mouse. Previous published methods assessing argininosuccinate lyase have used 20 µg of protein (85). Hence additional experiments were performed to ascertain whether it was possible to accurately quantify the clearance of argininosuccinic acid by LC-MS/MS when incubations were performed using this lower protein concentration in the presence of a reduced concentration of argininosuccinic acid (0.5 mM).
Additional modifications considered included the removal of KCl, MgCl2 and aspartate from the enzyme buffer, which are essential for argininosuccinate synthase activity and therefore for a coupled assay but are not necessary for ASL activity as shown previously (85, 416, 418). The simplified buffer didn’t show any difference in ASL activity in the samples tested.
The disappearance of argininosuccinic acid was measured under these conditions and was found to be linear for the first 2 hours of the reaction (Figure 54). Internal standards (0.02 nM D7-citrulline and 0.02 nM L-Arginine-13C) were added to the incubation buffer at the start of the assay. Under the conditions of the assay used, no conversion of the internal standards was observed, enabling them to be used to accurately quantify ASL activity.
208 Chapter 4 - Results
Argininosuccinic acid + anhydrides
80 Arginine
60
40
Amount (nmol) Amount 20
0 0 60 120 180 240
Time (Minutes)
Figure 54. The effect of incubation time on the conversion of argininosuccinic acid
20 µg of proteins from liver lysate of a wild-type mouse was incubated with 0.5 mM argininosuccinic acid. The black line shows that ASL activity assessed by argininosuccinic acid clearance was linear for the first 2 hours.
The final protocol used for investigating argininosuccinate lyase activity in mouse liver samples was as follows: samples were analysed in duplicate. 20 µg of lysate was incubated in a buffer solution of 10 mM Tris, pH 7.45 containing 0.5 mM argininosuccinic acid, 0.02 nM D7-citrulline and 0.02 nM L-Arginine-13C, in a final volume of 100 µL for 2 hours at 37°C. The reaction was stopped by adding 400 µL of methanol. Samples were then centrifuged immediately at 16,000 g for 2 minutes and the supernatant analysed by LC-MS/MS (as described in Section 2.14.7). Enzyme activity was expressed as argininosuccinic acid clearance per amount of protein per minute. Results showed a significant decrease of argininosuccinate lyase activity in 2 month-old AslNeo/Neo mice compared to wild-type littermates with 13.5% residual activity of wild type samples being observed (Two-tailed unpaired t-test, p<0.0001) (Figure 55A). This was confirmed in a larger experiment assessing the liver phenotype and presented in Section 5.2.4.
209 Chapter 4 - Results
A similar protocol was used to investigate argininosuccinate lyase activity in brain extracts. As the amount of protein extract is much lower of that observed in liver, argininosuccinic acid concentrations were reduced to allow adequate sensitivity. Duplicate samples containing 80 µg of brain lysate were incubated in a 10 mM Tris, pH 7.45 buffer containing 30 µM argininosuccinic acid, 0.02 nM D7-citrulline and 0.02 nM L-Arginine-13C for 2 hours at 37°C (see Section 2.14.7). Results showed a significant decrease in argininosuccinate lyase activity in AslNeo/Neo mice compared to wild-type mice with 16% residual activity of wild-type samples being evident (Two- tailed unpaired t-test, p=0.02) (Figure 55B).
Wild type AslNeo/Neo
A B 500 300 * *** 400 200 300
200 100 100 ASL ASL activity (nmol/ng/min) 0 ASL activity (nmol/ng/min) 0
Figure 55. Argininosuccinate lyase activity in liver and brain samples from AslNeo/Neo and wild-type mice
Argininosuccinate lyase activity in (A) liver and (B) brain samples from 12 month-old sex-matched AslNeo/Neo and wild-type mice. Horizontal lines display the mean ± standard deviation. Two-tailed unpaired t-test, * p<0.05, *** p<0.001; n=4-7.
210 Chapter 4 - Results
4.4 Discussion
This chapter describes the development of a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method to accurately phenotype and monitor efficacy of the gene therapy protocol described in Chapter 7. This technology is routinely used for newborn screening and can screen various compounds in very limited volumes of biological fluids (404). The method did not involve derivatisation and was capable of detecting the amino acids involved in the urea cycle (L-ornithine, L-citrulline, argininosuccinic acid, L-arginine) as well as being able to monitor the physiological buffering pathway in case of hyperammonaemia (L-glutamine, L-glutamate) (Section 4.2.4). This method was adapted from Piraud et al (401) and showed superiority compared to the initial derivatisation protocol tested with FMOC for detection of argininosuccinic acid (Section 4.2). Methods that do not involve derivatisation have been reported to be more precise for the analysis of amino acids of the urea cycle y (402). Another advantage of non-derivatisation is the decreased time involved in sample preparation.
The liquid chromatography gradient used 3.7% acetic acid as used previously to detect carboxylic acid derivatives of vitamin B6 (403). The method showed good linearity over the physiological range of analytes expected (Section 4.2.4).
Using this method, we had sufficient sensitivity to analyse the urea cycle amino acids in dried blood spots, which was preferable to plasma collection simplifying animal sampling. With the method developed, a 3.2 mm punch, which corresponds to 3.4 µL of blood (404), was a sufficient volume for appropriate testing. Plasma collection, which requires an initial centrifugation step to remove erythrocytes, thereby reducing the amount of volume available for analysis, would have been very unlikely to have achieved similar results with the volumes available. Hyperammonaemic AslNeo/Neo mice have a phenotype with restricted growth (32), which makes the sampling harder and the animal more fragile in case of repeated blood collections. Hence, dried blood spots were considered the method of choice for this application. The extraction of the amino acids of interest from dried blood spots was optimised. The precision and reproducibility of the method, including extraction and analysis, showed an excellent
211 Chapter 4 - Results intrabatch (within-day) reproducibility for all compounds and a satisfactory interbatch (between-days) reproducibility for all of the compounds analysed with the exception of argininosuccinic aciduria when present at low concentrations (50 µM). However this remained within acceptable values published for dried blood spot analysis (402) and indeed the range of argininosuccinic acid observed in untreated AslNeo/Neo mice was greater than this lower concentration with a range of 120 to 180 µM of being observed (Section 4.2.4).
No matrix effect was observed with L-ornithine, L-citrulline, L-arginine and L- glutamate. Ion suppression and potentiation were observed with argininosuccinic acid and L-glutamine, respectively. However quantification using internal standards was able to correct for this matrix effect. Performance of the method was validated with regards to linearity observed over the physiological range expected, precision and recovery and was shown to have sufficient sensitivity to accurately measure analytes in the low nanomolar range.
No deuterated argininosuccinic acid is commercially available as internal standard. The use of a «secondary» internal standard as D7-citrulline, which has different biochemical properties and retention time for the quantification of argininosuccinic acid, may explain the variability observed in the method development i.e. higher interbatch reproducibility compared to other amino acids measured and matrix effect with ion suppression.
The method developed to measure argininosuccinate lyase activity was adapted from that of Nagamani et al (85) and monitors the conversion of the substrate argininosuccinic acid to arginine. This assay was optimised for liver and brain samples. Results showed a residual activity in AslNeo/Neo mice compared to wild type of 13.5% and 16%, respectively which is in accordance with the 14.5% liver residual activity reported in Erez et al (32). This is further discussed in Chapter 5.
212 Chapter 4 - Results
4.5 Conclusion
Liquid chromatography-tandem mass spectrometry allows the quantification of biochemicals in reduced amounts of tissue or small volumes of biological fluids in a reproducible and accurate way. A method was developed to measure amino acids involved in the urea cycle in dried blood spots, an easy form of sampling in potentially sick small animal models. This method was robust with regards to linearity and precision. A reliable enzymatic assay for argininosuccinate lyase was developed in liver and brain tissues. These tools will enable precise phenotyping of the animal model and aid in monitoring the efficacy of any therapeutic intervention.
213 Chapter 5 - Results
5. PHENOTYPING OF AslNeo/Neo MICE
5.1 Introduction
This thesis project aims to develop a new therapeutic strategy for argininosuccinic aciduria based on gene therapy.
For this, a hypomorphic mouse model AslNeo/Neo recapitulating the human disease has been described by Erez et al in 2011 (32). The insertion of a neomycin cassette in the 9th intron of the murine Asl gene reduces messenger RNA and protein expression by 25% and ASL enzymatic activity is reduced to 16% of wild-type ASL activity (32). This leads to a multi-organ disease with impaired growth, abnormal fur pattern, raised transaminases, impaired renal function, high blood pressure, hyperammonaemia, systemic nitric oxide deficiency and early death within the first month of life (32). This animal model was chosen for its relevance to the human disease and the milder phenotype compared to knockout models where the severity of the hyperammonaemic disease causes a death within 48 hours after birth (146). This milder presentation allows to better study the phenotype and the pathophysiology associated with nitric oxide deficiency, especially in the brain.
Once the AslNeo/Neo breeding colony had been established, the neurological and systemic phenotype was characterised. This has been an essential preliminary step to define the best parameters to assess the efficacy of gene therapy. Specific attention was paid to the neurological phenotype, which had not been described before.
214 Chapter 5 - Results
5.2 Systemic phenotype
5.2.1 Macroscopic phenotype
Survival
AslNeo/Neo mice showed a reduced survival compared to wild-types and heterozygotes with 75% of them dying between day 15 and day 25 around the weaning period (Log Log rank test, p<0.001) (Figure 56).
100
50
Survival (%) Survival
0 0 10 20 30 40 Age (Days)
Wild-type Heterozygote AslNeo/Neo
Figure 56. Survival of wild-type, heterozygote and AslNeo/Neo mice
Kaplan-Meyer survival curve of wild-type n=19; heterozygote n=16; AslNeo/Neo n=18.
Growth
AslNeo/Neo mice display impaired growth, which becomes obvious within the first 10 days of life and is sustained until at least 9 months of age (Figure 57A). The growth velocity shows an initial increase during the first 2 weeks followed by a decrease corresponding to the time of weaning then a second higher increase lasting for a further 2 weeks. Subsequently, the growth velocity remains parallel to baseline. This
215 Chapter 5 - Results physiological profile is observed in both wild-type and heterozygote mice. In AslNeo/Neo mice, these 2 peaks are much reduced and delayed to the second and fifth weeks, respectively (Figure 57B).
Neo/Neo Wild-type Heterozygote Asl
A 60
40
20 Weight (grams) Weight
0 1 14 42 77 112 270
Days
B 1.0
0.8
0.6
0.4
(gram/day) 0.2 Growth velocity Growth 0.0
7 14 21 35 49 63 100 300 -0.2 Days
Figure 57. Growth and growth velocity of wild-type, heterozygote and AslNeo/Neo mice
(A) Weight (grams) over the first 9 months of life. (B) Growth velocity. Horizontal lines display the mean ± standard deviation. Unpaired 2-tailed Student’s t-test ** p<0.01. Wild-type n=5; Heterozygote n=11; AslNeo/Neo n=5.
216 Chapter 5 - Results
Fur pattern
AslNeo/Neo mice present with an abnormal fur pattern with brittle and sparse hair (Figure 58). This feature, persists into adulthood (Figure 59). Compared to wild-type fur, microscopic examination of AslNeo/Neo hair shows pili torti (twisted hair) and monilethrix aspects (beaded appearance) (Figure 60).
Figure 58. Macroscopic phenotype of AslNeo/Neo and wild-type mice during the first 3 weeks of life
Pictures taken at (A) day 3, (B) day 8, (C) day 12, (D) day 15 and day 21 of life. The asterisk indicates the AslNeo/Neo mouse. Scale bars are (A, B) 0.5 cm, (C) 1 cm, (D) 1.5 cm and (E) 2 cm.
217 Chapter 5 - Results
Figure 59. Long-term fur pattern in a 3 month-old surviving AslNeo/Neo mouse
(A) Face, and (B) body aspects of the AslNeo/Neo mouse are compared to (C) wild-type mouse. Scale bars (A, B) 0.5 cm and (C) 1 cm.
218 Chapter 5 - Results
Figure 60. Microscopic examination of AslNeo/Neo and wild-type mice hair
Hair (A) and tip (B) of wild-type mouse. Pili torti hair (C, D) and tip (E) of AslNeo/Neo mouse. Hair of wild-type (F) and AslNeo/Neo (G-I) mice with monilethrix (G) and pili torti (H, I). Scale bars: (A-E) 500 µm; (F-I) 125 µm; n=3.
219 Chapter 5 - Results
5.2.3 Blood and urine biomarkers
General biochemistry
Biochemical parameters were compared in plasma of 2 month-old sex-matched wild-type, heterozygote and AslNeo/Neo mice using one-way ANOVA with Dunnett’s post-test (Figure 61). There was a significant increase in blood alanine aminotransferase (ALT) in AslNeo/Neo mice compared to wild-type and heterozygote mice (Figure 61A; p<0.01 and p<0.001, respectively). No other tested liver parameters (aspartate aminotransferase (AST), gamma glutamyltranspeptidase (GGT), alkaline phosphatase, albuminaemia) were found to be significantly different (Figures 61B-E); notably there was no evidence of cholestasis. Creatinine concentrations were elevated in AslNeo/Neo mice compared to that of wild type and heterozygote mice, respectively (Figure 61G; p<0.01 and p<0.001, respectively). Uremia and kalaemia did not differ (Figures 61F & 61H). Total bilirubinaemia was below the limit of quantification in all groups.
220 Chapter 5 - Results
** A 200 B ** 1000 150
100 500 AST (IU/L) ALT ALT (IU/L) 50
0 0 WT HET AslNeo/Neo WT HET AslNeo/Neo
300 C 40 D
30 200
20 100 GGT (IU/L) 10
0 (IU/L) phosphatase Alkaline 0 WT HET AslNeo/Neo WT HET AslNeo/Neo
35 E F 15
30 10
25 Urea (mM) Urea 5 Albuminemia (g/L) Albuminemia
20 0 WT HET AslNeo/Neo WT HET AslNeo/Neo
* 40 G *** H 8 30
mM) 6 (µ 20 4
10 Kalaemia (mM) Kalaemia 2 Creatinine
0 0 Neo/Neo WT HET AslNeo/Neo WT HET Asl
Figure 61. General biochemistry parameters in plasma of wild-type, heterozygote and AslNeo/Neo mice
(A) Alanine aminotransferase (ALT), (B) aspartate aminotransferase (AST), (C) gamma-glutamyl transpeptidase (GGT), (D) alkaline phosphatase, (E) albuminaemia, (F) uremia, (G) creatininemia and (H) kalaemia were tested in 2 month-old sex- matched wild-type, heterozygote and AslNeo/Neo mice; Horizontal lines display the
221 Chapter 5 - Results mean ± standard deviation. One-way ANOVA with Dunnett’s post-test compared to AslNeo/Neo mice; * p<0.05; *** p<0.001, ns not significant; n=8-9.
Urea cycle-related biomarkers
Clinical parameters used to assess the urea cycle function were compared in 2 months-old sex-matched wild type, heterozygote and AslNeo/Neo mice using one-way ANOVA with Dunnett’s post-test compared to AslNeo/Neo mice (Figure 62). Various parameters were significantly modified in AslNeo/Neo mice compared to wild-type and heterozygote mice: increased plasma ammonia levels (Figure 62A; p<0.001 and p<0.001, respectively), increased orotic aciduria (Figure 62B; p<0.001 and p<0.001, respectively). In dried blood spots of AslNeo/Neo mice, the following amino acids were compared to wild-type and heterozygote mice: glutamine and glutamate were increased (Figure 62C; p<0.001 and p<0.01, respectively), decreased arginine in dried blood spots (Figure 62D; p<0.001 and p<0.05, respectively), citrulline was increased (Figure 62E; p<0.001 and p<0.001, respectively), argininosuccinic acid was increased (Figure 62F; p<0.001 and p<0.001, respectively). Similarly urinary argininosuccinic acid was increased in AslNeo/Neo mice compared to wild-type and heterozygote mice (Figure 62G; p<0.05 and not significant, respectively) although ornithine in dried blood spots, plasma creatine and guanidinoacetate were not modified between genotypes (Figure 62H, 62I & 62J, respectively).
222 Chapter 5 - Results
1000 1000 *** 150 A *** B 150 ****** 750 ****** M) 750 ***
M) *** 100 (µ
(µ 100 acid
acid 500
Orotic 50
Orotic 50 Ammonia Ammonia 250 mol / mol creatinine mmol µ mol / mol creatinine mmol µ 0 0 0 Neo/Neo Neo/Neo WT HET Asl WTWT HETHET AslAslNeo/Neo WT HET AslNeo/Neo M)
(µ *** 80 *** 2000 M) 80 ***
(µ *** ** * 2000 M) 60 * 1500 M) C ** D (µ 60 glutamate (µ
1500 glutamate and
40 1000 40 and
1000 500 2020 Argininaemia 500 Argininaemia glutamine
glutamine 0 00 0 Neo/Neo WT HET Asl WTWT HETHET AslNeo/NeoAslNeo/Neo Plasma WT HET AslNeo/Neo Plasma *** *** M) 200 *** ***
*** (µ 250 *** M) 250 200 *** (µ
*** acid E F M) 200 150 µ acid M) 200 µ 150 150 150 100 100 100 100 50 argininosucinic
50
Citrullinaemia ( Citrullinaemia 50 argininosucinic
50 Citrullinaemia ( Citrullinaemia 0 0 0 0 WT HET AslNeo/Neo Plasma Neo/Neo
Plasma WT HET Asl WT HET AslNeo/Neo WT HET AslNeo/Neo
* ns 200000 * 4040 ns 200000 M) M) 3030 (µ G 150000 H (µ 150000
100000 2020 100000
10 5000050000 10 Ornithinaemia mol / mol creatinine mmol Ornithinaemia mol / mol creatinine mmol µ µ 0 Urinary argininosuccinic acid Urinary argininosuccinic acid 0 0 0 Neo/Neo WT HET AslNeo/Neo WT WT HET HET Asl AslNeo/Neo WT HET AslNeo/Neo
800 20002000 800 600 I 15001500 J 600 400 10001000 400 Creatine Creatine 200
500500 Guanidinoacetate 200 Guanidinoacetate mol / mol creatinine mmol mol / mol creatinine mmol mol / mol creatinine mmol mol / mol creatinine mmol µ µ µ µ 0 0 0 0 Neo/Neo WTWT HETHET AslAslNeo/Neo WTWT HETHET AslNeo/NeoAslNeo/Neo
Figure 62. Urea cycle-related biomarkers tested in wild-type, heterozygote and AslNeo/Neo mice
223 Chapter 5 - Results
(A) Plasma ammonia, (B) orotic aciduria, (C) glutamine and glutamate, (D) arginine, (E) citrulline, (F) ornithine, (G) argininosuccinic acid in dried blood spots (C-G), (H) urinary argininosuccinic acid, (I) plasma creatine and (J) plasma guanidinoacetate were examined in 2 month-old sex-matched wild-type, heterozygote and AslNeo/Neo mice. Horizontal lines display the mean ± standard deviation. One-way ANOVA with Dunnett’s post-test compared to AslNeo/Neo mice; * p<0.05; *** p<0.001, ns not significant; n=3-8.
5.2.4 Liver phenotype
The liver expresses all of the enzymes of the urea cycle and is one of the main organs involved in the pathophysiology of argininosuccinic aciduria.
Comparison of the macroscopic observation of livers from AslNeo/Neo mice with those of wild-type mice showed an enlarged liver with a loss of the physiological brown colour normally observed (Figure 63A). Although AslNeo/Neo mice have impaired growth and a reduced body weight compared to wild-type mice, their liver weight at 6 weeks was heavier with a significantly increased liver/body weight ratio (Figure 63B; Two-tailed unpaired t-test, p<0.0001).
A B *** 0.20
0.15
0.10
0.05 Ratio Liver / Body weight 0.00
WT Asl Neo/Neo
Figure 63. Hepatomegaly in AslNeo/Neo mice compared to wild-type mice
Experiment performed in 6 week-old sex-matched mice. Horizontal line displays the mean ± standard deviation. Two-tailed unpaired t-test, *** p<0.0001; n=22-28.
224 Chapter 5 - Results
Histopathology was performed to better understand the suspected liver disease, as suggested by elevated plasma alanine aminotransferase (ALT) and hepatomegaly. Various stainings were performed in 10 month-old wild-type and AslNeo/Neo mice to capture features of the liver disease, which might be progressive and would not be obvious in examination of young animals (Figure 64): i) Haematoxylin & eosin (H&E) staining (described in Section 2.7.9) revealed unstained large cytoplasmic vacuoles in hepatocytes of AslNeo/Neomice. ii) These vacuoles were stained with periodic acid-Schiff staining (PAS) (described in Section 2.7.10). This staining, specific for polysaccharides, demonstrated glycogen deposits. iii) No sign of fibrosis was observed with Masson trichrome staining, a staining used to assess connective tissue and fibrosis (described in Section 2.7.11). iv) A negative Oil red O staining (ORO) staining, a staining used to identify lipid deposits (described in Section 2.7.12) did not support a steatosis in these samples.
225 Chapter 5 - Results
WT AslNeo/Neo
Low Mag High Mag Low Mag High Mag H&E Masson trichrome PAS ORO
Figure 64. Liver stainings highlight increased glycogen deposits in AslNeo/Neo mice compared to wild-type mice
Representative images after staining with Haematoxylin & eosin (H&E), Masson trichrome, periodic acid-Schiff (PAS) and Oil red O (ORO) with vacuolar cytoplasmic deposits of glycogen in liver of 10 month-old wild-type and AslNeo/Neo mice. Scale bars of low and high magnification (Mag): 500 and 125 μm respectively; mice per group, n=3. WT Wild-type.
Murine argininosuccinate lyase (ASL) staining (described in Section 2.7.6) was studied in liver, skeletal muscle, heart, and kidney as these organs can be involved with the clinical phenotype of the disease in humans. The four organs showed reduced staining demonstrating the decreased amounts of murine argininosuccinate lyase in 3 month-old AslNeo/Neo mice (Figure 65).
226 Chapter 5 - Results
Figure 65. Argininosuccinate lyase immunostaining in liver, skeletal muscle, heart and kidney in AslNeo/Neo and wild-type mice
Representative images of argininosuccinate lyase immunostaining in various peripheral organs highlight the reduced staining of ASL in 3 month-old AslNeo/Neo compared to age-matched wild-type mice. Scale bars of low and high magnification: 500 and 125 μm respectively; mice per group, n=3. WT: Wild-type.
Compared to 3 month-old wild-type mice (Figures 66A, 66D), electron microscopy of hepatocytes of age-matched AslNeo/Neo mice (method described in Section 2.8) showed an excess of glycogen identified as discrete circular granules (Figures 66B, 66E) (419, 420). Some lipid droplets were present indicating steatosis (Figures 66C, 66F), which could not be detected by Oil red O staining (Figure 65). Clusters of cytoplasmic vesicles with an appearance of crystallised content were observed (Figures 66G-I). These observations were secondarily confirmed by the expertise of Dr Glenn Anderson, electron microscopist at Great Ormond Street Hospital, London.
227 Chapter 5 - Results
A B C
D E F
G H I
Figure 66. Electron microscopy of liver samples from AslNeo/Neo and wild- type mice
Liver samples from (A, D) wild-type and (B, C, E-I) AslNeo/Neo mice. Scale bars (A-C, G): 5 μm, (D-F, H): 2 μm, (I): 1 μm, respectively; n=3. (B, E) Pink arrows identify excess of glycogen; (C, F) Blue arrows identify lipid droplets; (G, I) Brown arrows identify cytoplasmic vesicles.
The amount of murine argininosuccinate lyase (mASL) present in liver extracts of AslNeo/Neo and wild-type mice was assessed by western blotting (described in Section 2.9). A ratio with glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which was used as a reference protein, showed that the ASL/GAPDH ratio in AslNeo/Neo liver (n=3) was 10% of wild-type levels (n=2) (Figure 67).
ASL activity in AslNeo/Neo mice (described in Section 4.3) was 13.5% of wild-type activity (Figure 68; Two-tailed unpaired t-test, p<0.0001).
228 Chapter 5 - Results
Figure 67. Western blot and densitogram of murine argininosuccinate lyase present in liver extracts of AslNeo/Neo and wild-type mice
ASL: argininosuccinate lyase; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; WT: wild type. The band intensities were quantified using the densitometry function of Fiji (Fiji is just Image J) software (325)
400 ***
300
200
100
activity (nmol/ng/min) Argininosuccinate lyase lyase Argininosuccinate 0
WT Asl Neo/Neo
Figure 68. Argininosuccinate lyase activity in livers from 6 month-old AslNeo/Neo and wild-type mice
Horizontal lines display the mean ± standard deviation. Two-tailed unpaired t-test, *** p<0.0001; n=4. WT: wild type.
229 Chapter 5 - Results
Nitrite/nitrate levels (described in Section 2.6.3) were found to be decreased in liver samples from AslNeo/Neo compared to wild-type mice (Figure 69A; Two-tailed unpaired t-test, p=0.03). Reduced glutathione levels in liver were also reduced (Figure 69B; Two-tailed unpaired t-test, p=0.03).
A * B 300 300 * M) M) / mg prot µ 200 µ 200
100 100 Nitrate + Nitrite (
0 0
WT Asl Neo/Neo ( glutathione Reduced WT Untr
Figure 69. Liver nitrate/nitrite and reduced glutathione levels in liver samples from 8 week-old AslNeo/Neo and wild type mice
(A) Nitrate/nitrite and (B) reduced glutathione levels levels in AslNeo/Neo and wild type mice. Horizontal lines display the mean ± standard deviation. Experiment performed in 8 week-old sex-matched mice. Two-tailed unpaired t-test, * p<0.05; n=3-5. WT: wild type.
230 Chapter 5 - Results
5.3 Neurological phenotype
5.3.1 Motor phenotype
Righting reflex, grid walking, rotarod and open field tests (described in Section 2.5) were used to assess motricity and endurance. The open field test assesses the ability of the mouse to cope with anxiety when it is placed in a new environment. The novel object recognition test was used to test memory, learning and exploratory behaviour whilst the tail suspension test was chosen to assess perseverance, mood and the tendency for depression (421).
Tail suspension, grid walking, righting reflex and novel object recognition tests showed no significant difference between wild-type, heterozygote and AslNeo/Neo mice (Figures 70A-D). The rotarod test showed reproducible and significant differences with impaired ability to stay on the rod in the AslNeo/Neo group compared to wild-types and heterozygotes (Figures 70E & 70F) and the open field test elicited significantly decreased distance walked in the AslNeo/Neo group in comparison to wild-types (Figure 70G). These two tests showed the reduced motricity of AslNeo/Neo mice when tested from 3 weeks onwards compared to other genotypes.
231 Chapter 5 - Results
A 300 B 15
200 10
100 5 (seconds) 0 0
Time remaining immobile remaining Time WT AslNeo/Neo Neo/Neo -100 Numbers of paw faults WT HET Asl -5
ns
C D 100 60 ns
40 50 20
0 Exploration time (seconds) time Exploration 0
Righting reflex (seconds) 2 4 5 6 7 8 10 -20 Old New Old New Days WT AslNeo/Neo
ns 200 E 200 F * 150 150 100 100
Time (seconds) Time 50
Time (seconds) Time 50 0 0 1 2 3 4 5 WT HET AslNeo/Neo Days
G 25 ** 20
15
10
5 Distance walked (m)
0 WT AslNeo/Neo
Figure 70. Behavioural tests in wil- type, heterozygote and AslNeo/Neo sex- matched mice
232 Chapter 5 - Results
(A) Tail suspension test with sex-matched mice aged from 3 weeks to 4 months, n=7; (B) Grid walking test at 3 weeks of age, n=3-8; (C) Righting reflex during the first 10 days of life, n=7-12; (D) Novel object recognition with 2 month-old mice, n=8; (E, F): Rotarod test with sex-matched mice aged 3 weeks to 4 months with (E) mean of each mouse over the five days of testing or (F) performance over time during the five days of testing, n=3-8; (G) Open field test, n=11-12. Horizontal lines display the mean ± standard deviation. (A, D, G) Two-tailed unpaired t-test, (B, E) One-way ANOVA with Dunnett’s post-test compared to AslNeo/Neomice; * p<0.05; ** p<0.01. WT: wild-type; HET: heterozygote.
5.3.2 Neuropathology
The following analysis presented in this section were performed in sex-matched mice aged 2 to 4 months.
Observation of the macroscopic aspect of the brain as brain weight did not differ between genotypes (Figure 71A; Two-tailed unpaired t-test; not significant). Taking into account the impaired growth, the brain/body weight ratio showed brain sparing with relative cerebral hypertrophy (Figure 71B & 71C; Two-tailed unpaired t-test, p=0.004 and p=0.004, respectively).
** 4 A B 50 C ** 0.6 40 3
30 0.5 2 20 Weight (grams) Weight Percentage (%) Percentage Weight (grams) Weight 1 0.4 10
0 0
WT Asl Neo/Neo
Figure 71. Brain hypertrophy in AslNeo/Neo mice compared to wild-type mice
(A) Brain and (B) body weight; (C) Brain/body weight ratio, n=3-8. Horizontal lines display the mean ± standard deviation. Two-tailed unpaired t-test; ** p<0.01; WT: wild type.
233 Chapter 5 - Results
ASL activity in brain homogenates showed a significant decrease in AslNeo/Neo mice with 14.3% of residual activity compared to wild-type mice (Figure 72; Two-tailed unpaired t-test, p=0.007).
300 **
200
100
(nmol/ng/min)
0 activity lyase Argininosuccinate WT Asl Neo/Neo
Figure 72. Brain ASL activity in AslNeo/Neo mice compared to wild-type mice
Horizontal lines display the mean ± standard deviation. Two-tailed unpaired t-test; ** p<0.01; WT: wild type.
Nissl staining (described in Section 2.7.7) was used to look at the morphology and cellularity between wild-type and AslNeo/Neo brains. Examination at low and high magnifications (Figures 73 and 74, respectively) did not elicit any difference in cellularity or structure. No abnormal migration pattern, also known as grey matter heterotopia, was observed.
Immunostaining against argininosuccinate lyase (described in Section 2.7.4) showed reduced staining in the cortex and brainstem of AslNeo/Neo mice compared to wild-type mice (Figure 75). Various immunostainings were performed to assess oxidative/nitrosative stress. Nitrotyrosine is formed by the nitrosylation of a tyrosine residue by reactive nitrogen species and is a marker of nitrosative stress. Nitrotyrosine staining was increased in the cortex of AslNeo/Neo mice compared to wild- type mice (Figure 76A). The staining was observed in cells morphologically identified
234 Chapter 5 - Results as neurons. This was not associated with a neuroinflammation mediated by astrocytic or microglial activation evaluated by glial fibrillary acid protein (GFAP) and CD68 stainings, respectively (Figures 76B, 76C). Immunohistochemical staining against the three nitric oxide synthase (NOS) isoforms was performed. All revealed increased staining in AslNeo/Neo mice with different sublocalisations: neuronal nitric oxide synthase (nNOS or NOS1) and inducible nitric oxide synthase (iNOS or NOS2) were highly expressed in cells morphologically identified as neurons (Figures 76D, 76E, respectively). Endothelial nitric oxide synthase (eNOS or NOS3) was highly expressed in cells identified morphologically as endothelial cells and easily visualised in cerebral capillaries (Figure 76F).
To better understand if this nitrosative stress was associated with cellular damage, a terminal deoxynucleotidyl transferase dUTP nick end-labeling (TUNEL) staining (described in Section 2.7.8) was performed to identify cell death. Cortex and cerebellum from AslNeo/Neo brains showed a higher number of TUNEL positive cells compared to wild-type although heterogeneity was observed between animals. TUNEL positive cells were particularly abundant in the subcortex and in the granular layer of cerebellum (Figure 77).
235 Chapter 5 - Results
Figure 73. Morphology of wild-type and AslNeo/Neo mouse brains (low magnification)
Nissl staining. WT: wild-type. n=3 mice per group. Scale bar 5mm.
236 Chapter 5 - Results
Figure 74. Morphology of wild-type and AslNeo/Neo mouse brains (high magnification)
Nissl staining of (A) superficial cortex, (B) cortico-subcortical junction, (C) granular layer, (D) Purkinje and molecular layers. WT: wild type. n=3 mice per group. Scale bar 500 μm.
237 Chapter 5 - Results
Figure 75. Argininosuccinate lyase immunostaining in wild-type and AslNeo/Neo mouse brains
Mag: magnification; WT: wild type. n=3 mice per group. Scale bars are 500 μm (high magnification) and 125 μm (low magnification).
238 Chapter 5 - Results
Figure 76. Immunostaining assessing the oxidative/nitrosative stress in wild-type and AslNeo/Neo mouse brains
Immunostaining of cortical sections of 1-3 month-old wild-type and AslNeo/Neo mice: (A) increased nitrotyrosine in morphologically- identified neurons, (B) no difference in the astrocytic GFAP or (C) microglial CD68. Increased nNOS (D) and iNOS (E) in morphologically-identified neurons and (F) eNOS in endothelial cells in sections of AslNeo/Neo mice compared to wild-type. n=3 mice per group. Mag; Magnification. Scale bars of low and high magnification: 500 and 125μm, respectively. WT: wild type. 239 Chapter 5 - Results
A WT1 WT2 WT3 B WT1 WT2 WT3
ASA1 ASA2 ASA3
ASA1 ASA2 ASA3
7.5 µm
500 µm
WT1 C WT3 D WT1 WT3
ASA1 ASA2 ASA3 ASA1 ASA2 ASA3
5 µm 500 µm
Figure 77. TUNEL staining of wild-type and AslNeo/Neo mouse brains
Representative coronal sections of (A) brain, (B) cortex, (C, D) cerebellum. Scale bars are 500 (A, C) and 125 μm (B, D), respectively. WT: wild type.
240 Chapter 5 - Results
5.3.3 Nitric oxide metabolism in the brain
In order to explore the nitric oxide metabolism of the brain, various biomarkers were assessed. Nitric oxide concentrations from wild-type and AslNeo/Neo mice were - - assessed by measurement of nitrite (NO2 ) and nitrate (NO3 ) ions, downstream metabolites of nitric oxide (422) as the half-life of nitric oxide is one second, which prevents a reliable direct measurement (377). Cyclic guanosine monophosphate (cGMP) is a signalling pathway physiologically upregulated by nitric oxide (423). Low tissue L-arginine is observed in argininosuccinic aciduria and can cause nitric oxide synthase uncoupling (85). This leads to the production of reactive oxygen species - - including superoxide ion (O2 ) or peroxynitrite (ONOO ) (See Section 1.2.2 & Figure 5). Peroxynitrite generates nitrotyrosine, a marker of oxidative/nitrosative stress (424). The detoxification of peroxynitrite by reduced glutathione (GSH) can generate - - nitrite via the reaction ONOO + 2GSH ® NO2 + GSSG + H2O (425).
Increased nitrotyrosine staining and increased expression of nitric oxide synthases (eNos, iNOS and nNOS) provide evidence of nitrosative stress in the AslNeo/Neo mouse brains. Nitrotyrosine is generated by reactive nitrogen species such as - peroxynitrite formed by superoxide ion (O2 ) and nitric oxide. This highly damaging compound is associated with oxidative stress and these two are usually combined as an oxidative/nitrosative stress.
Nitrite/nitrate concentrations in AslNeo/Neo mouse brains were found to be significantly elevated (Figure 78A; Two-tailed unpaired t-test, p=0.002). This was associated with a significant increase of cyclic guanosine monophosphate (cGMP), confirming the activation of the NO-cGMP signalling pathway (Figure 78B; Two-tailed unpaired t- test, p=0.005). Decreased concentrations of reduced glutathione (although not statistically significant) supported the hypothesis of oxidative stress as an ongoing pathophysiological mechanism in the AslNeo/Neo mouse brain (Figure 78C). Nitrite/nitrate concentrations were significantly increased in the cortex and the midbrain but not in the cerebellum (Figures 78D, 78E & 78F, respectively; Two-tailed unpaired t-test, p=0.02, p=0.02 and p=0.5, respectively).
241 Chapter 5 - Results
WT AslNeo/Neo
A ** B 800 M) µ 550 25 ** 20 300 200 15 10
100 5 Nitrate + Nitrite ( 0 cGMP (pmol/mg) / protein 0 -5
ns * C 150 D 50
M) 40 µ 100 30
20 50 10 Nitrate + Nitrite (
0 0 M reduced glutathione / mg protein glutathione M reduced
µ
* 20 ns E 20 F
M) µ
M) 15
µ 15
10 10
5
5 Nitrate + Nitrite ( Nitrate + Nitrite ( 0 0
Figure 78. Neuronal oxidative/nitrosative stress is a component of the neurological disease in AslNeo/Neo mice
(A) Nitrite/nitrate concentrations (n=12-15) (B) cyclic GMP (n=6) and (C) reduced glutathione (n=4-5) in brain homogenates. Nitrite/nitrate levels in (D) cortex, (E) midbrain and (F) cerebellum. Horizontal lines display the mean ± standard deviation. ns = not significant. Two-tailed unpaired t-test * p<0.05; ** p<0.01; ns not significant. (A) Graph displays not transformed data and illustrates the widespread of the values. The D’Agostino & Pearson omnibus normality test confirmed the absence of normal distribution for the values observed in AslNeo/Neo mice (p<0.0001). Log-transformed
242 Chapter 5 - Results data achieved a normal distribution with the same test and were used for statistical analysis to obtain Gaussian distribution. WT: wild type.
5.4 Discussion
5.4.1 Recapitulation of the human disease
The phenotype of the AslNeo/Neo mouse shows impaired growth, early death, abnormal fur pattern (See Section 5.2.1) and a multi-organ disease with liver, kidney, and brain involvement (See Sections 5.2.4, 5.2.3 and 5.3, respectively). This confirms the phenotype described by Erez et al (32). Our work did not evaluate the blood pressure but arterial hypertension has been confirmed in previous publications (32, 85).
Differences in various biomarkers in plasma: alanine aminotransferase, creatinine, urea cycle-related biomarkers such as ammonia, citrulline, arginine, plasma and urine argininosuccinic acid, and in urine, orotic aciduria were detected (See Section 5.2.3). Interestingly, alterations in biomarkers related to the citrulline-nitric oxide cycle including nitrite/nitrate levels and reduced glutathione in the liver were also observed (See Section 5.2.4). This was in line with systemic nitric oxide deficiency observed in previous publications related to the AslNeo/Neo mouse model (32, 85). Reduced glutathione levels was suggesting of oxidative stress as a pathophysiological mechanism of the liver disease.
This phenotype mimics the symptoms observed in patients with argininosuccinic aciduria (18) and in the natural history study presented in Sections 1.2.2 and 3.4, respectively.
The liver phenotype in the human disease presents with an enlarged liver, elevated alanine aminotransferase and glycogen deposits (See Section 1.2.2) as that observed in the AslNeo/Neo mouse. This phenotype of increased glycogen deposition, which is observed in glycogen storage diseases, is not well understood when observed in urea cycle defects (79). Enzymes involved in glycogenesis and
243 Chapter 5 - Results glycogenolysis pathways are usually in the normal range or slightly abnormal (81). The roles of a carbohydrate-enriched diet and essential amino acid mixtures with high leucine content used in the treatment of urea cycle defects have been discussed (81) as leucine triggers insulin secretion and glycogenogenesis (82). This cannot however explain the phenotype observed in the AslNeo/Neo mouse. Glutamine, increased in hyperammonaemia, is as well an important trigger for insulin release (426, 427) and might be an explanation (81).
Liver fibrosis has been observed in patients with argininosuccinic aciduria (79) (See Section 1.2.2) but was not observed in the AslNeo/Neo mouse (See Section 5.2.4). This symptom has been described in ornithine transcarbamylase deficient patients but the pathophysiology is poorly understood. Hyperammonaemia or citrulline accumulation might be involved (79).
Electron microscopy revealed cytoplasmic vesicles with “flocculent” or crystalline inclusions which lacked dense granules (See Section 5.2.4); this occurs in some patients with Reye-like syndrome (428). Reye syndrome is defined by an encephalopathy without cerebrospinal fluid inflammation, and liver dysfunction i.e. raised transaminases, hyperammonaemia or liver steatosis (429). This syndrome can be observed in various inherited metabolic diseases: mitochondrial fatty acid oxidation defects, organic acidurias, urea cycle defects and mitochondrial disorders.
5.4.2 Neurological phenotype
The neurological phenotype of the AslNeo/Neo mouse has not been described before. The demonstration of a significant reduction in the performance of behavioural tests assessing the motricity suggests a neurological or muscular effect of the disease (See Section 5.3.1).
Hyperammonaemia will affect the AslNeo/Neo mouse brain and should contribute significantly to the abnormal behavioural results. However the role of the impaired citrulline-nitric oxide cycle in the brain remains uncertain. The high rate of neurological complications in patients irrespective of the age of onset of the disease and of the time of treatment initiation confirms that hyperammonaemia alone cannot
244 Chapter 5 - Results account for this neurological phenotype (See Section 3.7.1). Therefore, the perspective of studying the nitric oxide metabolism in the AslNeo/Neo brain was attractive. These investigations supported the hypothesis that oxidative/nitrosative stress involving neurons contributes to the neuropathology of argininosuccinic aciduria. Hypoargininaemia, a common biological feature of the disease in patients and AslNeo/Neo mice (See Sections 1.2.3 and 5.2.3, respectively), is a known cause for uncoupling of nitric oxide synthase, which decreases nitric oxide production and increases generation of free radicals that damage tissues (85). Reactive oxygen species then interfere with nitric oxide production and regulation of the microcirculation (361). In addition decreased nitric oxide levels might affect protein nitrosation on tyrosine residues (374), which in turn regulates histone methylation and gene expression (375). In the brain, nitric oxide plays a key-role as a signalling molecule (376) involved in neurotransmission (377), regulation of neuronal differentiation (378, 379) and migration (380, 381). No morphological abnormalities were observed in the AslNeo/Neo mouse brain (See Section 5.3.2), which suggests that this neuronal migration process is not affected. However this might not be the case in humans, where impaired cerebral nitric oxide metabolism could explain unusual neuroimaging findings observed in the British cohort of patients described previously (See Section 3.3). In that case, local parenchymal infarcts could be caused by impaired nitric oxide-mediated vasoregulation of the microcirculation and nodular heterotopia could result of an abnormal neuronal migration during development due to impaired cerebral regulation of nitric oxide metabolism.
Overexpression of all isoforms of nitric oxide synthase (iNOS, eNOS and nNOS) were observed in AslNeo/Neo mice (See Section 5.3.2). Neuronal and endothelial nitric oxide synthases are both constitutive enzymes whereas inducible nitric oxide synthase is inducible (430). In the brain, endothelial nitric oxide synthase is observed almost exclusively within endothelial cells (377); this agrees with the overexpression of endothelial nitric oxide synthase observed, by immunostaining, in cerebral micro vessels and capillaries. Neuronal nitric oxide synthase is expressed in both immature and mature neurons, astrocytes, and in the adventitia of blood vessels (431). Inducible nitric oxide synthase can be expressed in astrocytes, microglia, endothelial cells and neurons (432). Glial expression is frequently observed in diseases like cerebral ischaemia (433), autoimmune encephalitis (434), amyotrophic lateral
245 Chapter 5 - Results sclerosis (435), and Alzheimer’s disease (436). It is interesting to note that inducible nitric oxide synthase and argininosuccinate synthase, the later being the enzymatic limiting step of nitric oxide production, co-localise in neurons and glial cells confirming the presence of the multiprotein complex involving argininosuccinate lyase in both cell types (437) (See Section 1.2.1).
Independent of the effect of arginine concentration on nitric oxide production, hyperammonaemia per se can cause brain toxicity through oxidative stress, which is associated with upregulation of neuronal and inducible nitric oxide synthase isoforms in neurons and astrocytes, respectively (39).
Arginine supply regulates nitric oxide metabolism in urea cycle defects. In argininosuccinic aciduria, arginine levels are low in patients (See Section 1.2.3) and AslNeo/Neo mice (See Section 5.2.3). Similarly in the ornithine transcarbamylase (438) deficient Spf mouse model (in which the high residual activity of ornithine transcarbamylase does not cause hyperammonaemia (308)), plasma arginine is low and cerebral nitric oxide production is decreased (439). However in arginase deficiency where plasma arginine is elevated, nitric oxide metabolites are elevated in patients (440).
To investigate if this oxidative/nitrosative stresss was a primary pathophysiological mechanism or secondary to hyperammonaemia, we designed a gene therapy approach to normalise ammonaemia and conditionally target the argininosuccinate lyase activity in neurons. This is described in Chapter 6.
5.5 Conclusion
Study of the hypomorphic AslNeo/Neo mouse model has confirmed data from literature, as this animal model presents a phenotype mimicking the human disease of argininosuccinic aciduria. Thus this is a suitable animal model for assessing new therapies. Several macroscopic, biological and histological parameters have been identified as relevant efficacy endpoints.
246 Chapter 5 - Results
Neurological study of the AslNeo/Neo mouse brain has revealed oxidative/nitrosative stress, which is prominent in neurons, especially in the cortex. This might be a secondary consequence of hyperammonaemia. However, the requirement of argininosuccinate lyase for nitric oxide production indicates the possibility of a primary pathophysiological mechanism exerting a direct effect on the nitric oxide metabolism in the AslNeo/Neo mouse brain. A gene therapy approach targeting the liver and conditionally the brain might be of use in clarifying this question.
247 Chapter 6 - Results
6. VECTOR DESIGN FOR Asl GENE TRANSFER
6.1 Introduction
Argininosuccinic aciduria is a complex inherited metabolic disease as the deficiency of argininosuccinate lyase affects 2 biochemical pathways, the urea cycle and the citrulline-nitric oxide cycle (See Section 1.2.1). Current therapeutic guidelines focus on lowering ammonia levels (See Section 1.2.5). However this does not prevent poor long-term neurological outcome (See Section 1.2.6). The AslNeo/Neo mouse model mimicks the human disease and the brain phenotype shows an oxidative/nitrosative stress affecting neurons (See Section 5.3.3).
This project aims to develop an innovative therapeutic strategy based on gene therapy for argininosuccinic aciduria, which could be translated to clinic. Designing a gene therapy will require transduction of hepatocytes in order to target the urea cycle. However as illustrated by previous work (85), this might not be sufficient to sustainably protect patients from neurocognitive impairment. Therefore a transgene capable of delivery to the brain would be of interest. This might correct the metabolic defect in situ and its metabolic consequences.
AAV vectors currently are the leading strategy for liver-directed gene therapy with the best benefit-risk balance (See Section 1.3.3) and have shown proof of concept of phenotypic correction in various mouse models of urea cycle defects (See Section 1.3.7). AAV vectors have been successfully used for clinical trials in neurological diseases in targeting the brain (441). Therefore, an AAV vector was considered for developing a gene therapy approach for argininosuccinic aciduria.
248 Chapter 6 - Results
6.2 Vector design and delivery
6.2.1 Rationale for vector design
The choice of the different elements involved in an AAV vector i.e. AAV serotype, single or double-stranded transgenic DNA, promoter, transgene, cis-acting regulatory elements (See Section 1.3.2) was made based on previous experimental evidence and is summarised in Figure 79.
i) The AAV8 serotype was chosen for its liver tropism in adult (442) and neonatal mice (239, 443, 444), its ability to transduce the mouse brain after either intracranial (445, 446) or intravenous injection (447), its success in clinical trials for liver inherited metabolic diseases like haemophilia B (448), its safety profile with a lower seroprevalence compared to other AAV serotypes in either children (449) or adults (233, 235).
ii) A single-stranded vector was chosen to increase the cloning capacity and to reduce the risk of immune response as suggested by experiment with the self- complementary counterpart (255). Self-complementary AAV vectors achieve 10 to 100 times better transduction compared to single-stranded AAV vectors. However compared to self-complementary AAV vector without WPRE, a single-stranded AAV vector containing the cis-acting woodchuck hepatitis virus post-transcriptional regulatory element (WPRE) inserted downstream the transgene has shown better gene expression in the central nervous system and peripheral organs after systemic delivery (321, 450). This DNA sequence is involved in the cytoplasmic accumulation of messenger RNA and increases gene expression in retroviral (451), adenoviral (452) and AAV (453) vectors. A link with tumorigenicity was proposed (454) but never confirmed and this DNA sequence is now used in clinical trials http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_Public_assessment _report/human/002145/WC500135476.pdf. The self-complementary AAV vector
249 Chapter 6 - Results cannot hardly accommodate this 600 base pairs-long WPRE sequence (455) because of the limited 2.2 kilobase pairs cloning capacity of scAAV (254).
iii) The choice of the promoter plays a critical role as it influences the type of cells transduced and the kinetics of transgene expression (456). The argininosuccinate lyase promoter was not chosen as this promoter has only been used previously in in vitro applications (457). The EFS promoter, a shorter version of the Elongation factor 1 α (EF1α) promoter, which lacks the intron (458), was chosen.
EF1α, a highly conserved protein throughout evolution (459), plays a key-role in transfer RNA (tRNA) export in eukaryote cells, protein elongation and protein export in mammalian cells (460). EF1α is a nucleotide-binding protein and part of the translational protein-ribosome complex (461, 462). The EF1α promoter contains the 5’ flanking region and the first intron of the EF1α gene (463).
The EF1α promoter targets the liver successfully and has shown 14 to 100 times greater transduction of murine liver compared to the CMV promoter when incorporated into an AAV2 vector and after systemic delivery (464, 465). Use of this promoter has allowed the correction of the liver phenotype in an adult mouse model of mucopolysaccharidosis VII after intrahepatic injection of AAV2 (466).
This promoter has a broad spectrum and is considered as an ubiquitous promoter (467). Its ability to successfully transduce hematopoietic stem cells (468) or embryonic stem cells (469) with retroviral vectors has been reported, with again a better achievement than the CMV promoter (469). The EFS promoter has several other advantages compared to other ubiquitous promoters: resistance to silencing (470), reduced potential risk of insertional mutagenesis compared to other ubiquitous promoters (458), already in use in clinical trials (471). In the brain, this promoter targets mainly neurons (443, 472)
iv) A modified simian virus 40 (SV40) small t antigen intron upstream of the transgene coding DNA was also incorporated into the vector as this has been proven to increase transcriptional efficacy by 10 to 100 times (473). The intron splicing is supposed to avoid supercoiling during transcription, and then prevent the cut and re- 250 Chapter 6 - Results ligation processes required by DNA topoisomerase I, which are identified as potential causes of insertional mutagenesis (474).
v) Argininosuccinate lyase plays a key structural role for the maintenance of a multiprotein complex, which includes nitric oxide synthase, argininosuccinate synthase and a cationic amino acid transporter CAT-1 (32). For this reason and to prevent any misformation of this critical complex, it was decided to clone the murine non codon-optimised Asl gene in the insert cassette.
vi) Simian virus 40 late polyadenylation tail (SV40 LpA) was added downstream of the transgene ccoding DNA. This increases gene expression by stabilising messenger RNA for nuclear export and translation (475).
EFS mAsl
Figure 79. Schematic of the designed vector
ITR Inverted terminal repeat; EFS: Elongation factor 1 alpha short promoter; SV40: Simian virus 40 small t antigen intron; mAsl: murine Asl gene; WPRE: Woodchuck hepatitis virus post-transcriptional regulatory element; SV40 pA: simian virus 40 late polyadenylation tail.
6.2.2 Target structures and cell types
In order to correct the urea cycle, the target cells will be periportal hepatocytes. However, because the citrulline-nitric oxide cycle is expressed in most tissues, ubiquitous distribution of the vector would ideally be required. As the main caveat of
251 Chapter 6 - Results clinical application of a liver-restricted gene therapy approach would be the high risk of persisting neurocognitive impairment, the brain will need to be one of the main targets of the vector. Investigation of the neuropathology of the AslNeo/Neo mouse (see Section 5.3) highlighted that neurons, especially those in the cortex, present with oxidative/nitrosative stress. Therefore, neuronal transduction will be required to target the correct cell type.
6.2.3 Timing of delivery
The liver growth significantly affects the efficacy of AAV gene therapy mediated by a non-integrating transgene delivery. Hence whilst early delivery of the vector might show limited efficacy for some days or even weeks it is unlikely to correct the phenotype as time progresses. Therefore a proof of concept of efficacy in the AslNeo/Neo mouse will require a later delivery in adulthood to evaluate the effect on the phenotype related to the urea cycle defect itself. Early delivery targeting the brain however will prevent potential neurological sequellae related to an impaired nitric oxide metabolism.
6.2.4 Method of delivery
Intraperitoneal or peripheral intravenous routes can be used to target the liver with AAV viral vectors (442). Previous experiments have shown similar efficacy in liver trandusction between intraportal and peripheral intravenous routes for liver-directed gene therapy with AAV vectors in mice or non-human primates (190, 191).
For brain delivery, a direct intracerebroventricular injection of AAV8 vector in mice has been reported to allow widespread transduction (443). Systemic delivery of AAV8 has also been shown to transduce the brain in certain circumstances if the injection is performed at a high dose before the age of 2 weeks (476). A later injection in adult mice did not allow successful brain transduction (442).
252 Chapter 6 - Results
6.3 EFS promoter study
6.3.1 Experimental design
EFS was considered as a suitable promoter to target liver and brain efficiently and able to specifically target neurons. EFS was compared to the CMV promoter, which was used as a positive control and characterised by a wide biodistribution in most of the organs. GFP was used as a reporter gene to assess biodistribution.
Two single-stranded AAV vectors AAV8.EFS.SV40i.GFP.WPRE.SV40pA and AAV9.CMV.PI.GFP.WPRE.bGH (PI : chimeric intron from Promega, Madison, WI, USA ; bGH : bovine growth hormone polyadenylation tail) were kindly provided by Dr Joanna Hanley from Prof Gissen’s laboratory and Dr Joanne Ng from Dr Simon Waddington’s laboratory, respectively.
The experimental design was as follows: two methods of injection were tested in neonatal pups of C57Bl/6 mice i.e. a single peripheral intravenous injection via the superficial temporal vein in a 20 μL volume or two symmetrical intracerebroventricular injections. Intracerebroventricular injection was performed 1 mm lateral to the sagittal suture, halfway between the lambda and the bregma intersections of the sutures as previously described (477) with injection of a 5 μL volume. 4x10e11 and 8.5x10e10 vector genomes (vg) per pup of AAV8.EFS.SV40i.GFP.WPRE.SV40LpA were injected intravenously or intracerebroventricularly, respectively. In littermates, 6x10e11 and 1.5x10e11 vg per pup of AAV9.CMV.PI.GFP.WPRE.bGH were injected intravenously or intracerebroventricularly, respectively. Control littermates were not injected. At 5 weeks of age, mice were culled for organ collection and biodistribution analysis. Vector injections were performed by Dr Simon Waddington.
Five experimental groups were compared: AAV vector with EFS promoter neonatally injected via either intravenous (EFS IV) or intracerebroventricular route (EFS IC), AAV vector with CMV promoter neonatally injected via either intravenous (CMV IV) or intracerebroventricular route (CMV IC) (n=4 in each group), uninjected control group (n=3). GFP expression was assessed by immunostaining, enzyme-linked
253 Chapter 6 - Results immunosorbent assay (ELISA) and quantitative polymerase chain reaction (qPCR) (Figure 80).
Intravenous rAAV2/8-EFS-GFP-WPRE, n=4 Intracranial rAAV2/8-EFS-GFP-WPRE, n=4 Intravenous rAAV2/9-CMV-GFP-WPRE, n=4 Intracranial rAAV2/9-CMV-GFP-WPRE, n=4 Injec&on Sacrifice Uninjected controls, n=3
Day 0 Week 5
Immunohistochemistry ELISA qPCR
Figure 80. Promoter study design
AAV: adeno-associated viral vector; CMV: cytomegalovirus; EFS: Elongation factor 1 alpha short promoter; ELISA: enzyme linked-immunosorbent assay; GFP: green fluorescent protein; PBS: phosphate buffered saline; qPCR quantitative polymerase chain-reaction; WPRE: Woodchuck hepatitis virus post-transcriptional regulatory element.
6.3.2 Biodistribution of AAV vector constructs in peripheral organs
Organs were observed under a fluorescence microscope in order to visualise the biodistribution of GFP. Brain, liver, heart, skeletal muscle and skin were well transduced in the EFS IV group. In the EFS IC group, brain was well transduced and a mild fluorescence was observed in the liver. Otherwise no other peripheral organ showed any fluorescence. All organs, including kidney and gut, showed diffuse fluorescence in the CMV IV group. In the CMV IC group, brain, liver, muscle and kidney were fluorescent. Controls did not show any transduction (Figure 81).
254 Chapter 6 - Results
EFS promoter CMV promoter Control Intravenous Intracranial Intravenous Intracranial
Brain
1s 0.3s
Liver
0.7s 0.7s
Heart
1s 0.7s
Muscle
0.3s 0.7s
Kidney
0.3s 0.7s
Skin
0.3s
Gut
1.1s
Aorta
1.1s Figure 81. Effect of promoter and mode of delivery of vector on GFP fluorescence of brain and peripheral organs
Representative pictures of brain and peripheral organs. CMV IC: AAV vector containing the CMV promoter encoding GFP injected via intracerebroventricular 255 Chapter 6 - Results route; CMV IV: AAV vector containing the CMV promoter injected via peripheral intravenous route; EFS IC: AAV vector containing the EFS promoter injected via intracerebroventricular route; EFS IV: AAV vector containing the EFS injected via peripheral intravenous route. n=4 per group, except controls n=3. Time in seconds in the bottom left corner of each picture is the time of exposure before the capture of the image.
GFP immunostaining was performed in peripheral organs and quantified by Image- Pro Premier 9.1 software by analysing 10 random pictures of a representative sample. The highest liver transduction was achieved in the EFS IV group compared to EFS IC, CMV IV, CMV IC and controls (One-way ANOVA with Dunnett’s post-test compared to EFS IV: p<0.001, p<0.01, p<0.001 and p<0.001, respectively) (Figure 82). The CMV IV group showed the highest transduction compared to EFS IV in skeletal muscle, heart and skin (One-way ANOVA with Dunnett’s post-test compared to EFS IV: p<0.001, p<0.001 and p<0.001, respectively) (Figure 83).
256 Chapter 6 - Results
EFS IV EFS IC
CMV IV CMV IC
Control 50 ***
40 *** ** 30 ***
20
% of transduction 10
0 EFS IV EFS IC CMV IV CMV IC Control
Figure 82. GFP immunostaining in liver
Representative liver sections of EFS IV: AAV vector containing the EFS promoter encoding GFP injected via peripheral intravenous route. n=4 animals per group; EFS IC: AAV vector containing the EFS promoter injected via intracerebroventricular route; CMV IV: AAV vector containing the CMV promoter injected via peripheral intravenous route; CMV IC: AAV vector containing the CMV promoter injected via intracerebroventricular route; uninjected controls n=3 animal. Quantification analysed by Image-Pro Premier 9.1 software from 10 random pictures of a representative
257 Chapter 6 - Results sample. Horizontal lines display the mean ± standard deviation. Scale bars are 500 (Main picture) and 125 μm (inset), respectively. One-way ANOVA with Dunnett’s post-test compared to EFS IV: ** p<0.01; *** p<0.001.
258 Chapter 6 - Results
EFS promoter CMV promoter Control Transduc)on
100 *** *** 80
60 muscle 40
% of transduction 20 Skeletal 0 EFS IV CMV IV Control
** 100 ***
80
60
40 Heart % of transduction 20
0 EFS IV CMV IV Control
80 *** *** 60
40 Skin 20 % of transduction
0 EFS IV CMV IV Control Figure 83. GFP immunostaining in skeletal muscle, heart and skin after intravenous injection
259 Chapter 6 - Results
Representative sections of skeletal muscle, heart and skin after GFP immunostaining. (A) EFS IV: AAV vector containing the EFS promoter encoding GFP injected via peripheral intravenous route; (B) CMV IV: AAV vector containing the CMV promoter injected via peripheral intravenous route; n=4 animals per group except (C) uninjected controls n=3 animals. (D) Quantification analysed by Image-Pro Premier 9.1 software from 10 random pictures of a representative sample. Horizontal lines display the mean ± standard deviation. Scale bars are 500 (Main picture) and 125 μm (inset), respectively. One-way ANOVA with Dunnett’s post-test compared to EFS IV: ** p<0.01; *** p<0.001.
260 Chapter 6 - Results
In order to better quantify GFP in peripheral organs, a GFP ELISA was set up. Irrespective of the method of injection, the CMV promoter showed the better transduction with 1 to 3-log higher transduction compared to the EFS promoter (Figure 84).
** 109 ***
*** 108 *** *** 107 ***
*** 6 ** 10 *** **
*** 105 *** pg GFP / mg protein
104
Liver Muscle Heart Skin Kidney Gut Spleen
EFS IV EFS IC CMV IV CMV IC Control
Figure 84. Quantification of GFP in peripheral organs using ELISA
CMV IC: AAV vector containing the CMV promoter encoding GFP injected via intracerebroventricular route; CMV IV: AAV vector containing the CMV promoter encoding GFP injected via peripheral intravenous route; EFS IC: AAV vector containing the EFS promoter encoding GFP injected via intracerebroventricular route; EFS IV: AAV vector containing the EFS promoter encoding GFP injected via peripheral intravenous route. n=4 animals per group, except controls n=3 animals. Horizontal lines display the mean ± standard deviation. One-way ANOVA with
261 Chapter 6 - Results
Dunnett’s post-test compared to EFS IV: ** p<0.01; *** p<0.001. Only significant comparisons are represented.
qPCR targeting the WPRE sequence was performed in liver samples from mice at 5 weeks of age after neonatal injections. Vector genomes were more than 8 times higher in CMV injected groups compared to EFS groups, respectively (. One-way ANOVA with Dunnet’s post-test compared to EFS IV; p=0.02) (Figure 85A). However the ratio comparing the amount of GFP protein synthesised to the vector genome copy number was higher in the EFS injected groups compared to the CMV groups (Figure 85B).
A B 64.00 * 105 16.00
4.00 104 1.00 0.25 103 0.06 0.02 pg GFP / vector genome vector genome / nucleus vector genome 102 EFS IV EFS IC CMV IV CMV IC Controls EFS IV EFS IC CMV IV CMV IC
Figure 85. Quantitative PCR analysis of the number of vector genomes and the effect of CMV on GFP expression in liver samples
(A) Vector genomes per nucleus presented on a log 2 scale; (B) ratio of the amount of GFP protein per vector genome. Analysis performed in triplicate. CMV IC: AAV vector containing the CMV promoter encoding GFP injected via intracerebroventricular route; CMV IV: AAV vector containing the CMV promoter injected via peripheral intravenous route; EFS IC: AAV vector containing the EFS promoter injected via intracerebroventricular route; EFS IV: AAV vector containing the EFS promoter injected via peripheral intravenous route. n=4 animals per group, except controls n=3 animals. Horizontal lines display the mean ± standard deviation. One-way ANOVA with Dunnet’s post-test compared to EFS IV; * p<0.05.
262 Chapter 6 - Results
6.3.3 Biodistribution of AAV vector constructs in the central nervous system
GFP immunostaining of brain tissue revealed a diffuse transduction of the brain in the EFS groups with a decreasing rostro-caudal gradient. Conversely, in mice injected with CMV, universal transduction of the whole brain was evident. As expected, a higher intensity of staining was observed in groups injected via the intracerebroventricular route. Controls did not show any staining (Figure 86).
Observation of brain sections stained for GFP revealed a neuronal transduction pattern in groups of mice injected with EFS whereas those injected with CMV had mixed transduction of neurons and astrocytes (Figure 87). This pattern of transduction was confirmed by immunofluorescence (Figure 88). Brains of mice injected intravenously with the EFS promoter had an exclusive pattern of neuronal transduction as shown by the co-localisation of GFP with the neuronal marker NeuN (Figure 88A). Brains of mice injected intravenously with the CMV promoter showed transduction of astrocytes and some neurons (Figure 88B).
263 Chapter 6 - Results
EFS IV EFS IC CMV IV CMV IC Controls
Pre-frontal cortex
Lateral ventricles
Hippocampus
Cerebellum
Brainstem
5 mm Figure 86. Decreasing rostrocaudal gradient in EFS injected brains revealed by GFP immunostaining
Representative sections of brain staining. CMV IC: AAV vector containing the CMV promoter injected via intracerebroventricular route; CMV IV: AAV vector containing the CMV promoter injected via peripheral intravenous route; EFS IC: AAV vector containing the EFS promoter injected via intracerebroventricular route; EFS IV: AAV vector containing the EFS promoter encoding GFP injected via peripheral intravenous route. n=4 animals per group; uninjected controls n=3 animals. Scale bar is 5 mm .
264 Chapter 6 - Results
Low Mag High Mag EFS IV EFS IC CMV IV CMV IC Control
Figure 87. Specific transduction of neurons using the EFS promoter revealed by GFP immunostaining (high magnification)
Representative sections of brain staining. CMV IC: AAV vector containing the CMV promoter injected via intracerebroventricular route; CMV IV: AAV vector containing the CMV promoter injected via peripheral intravenous route; EFS IC: AAV vector containing the EFS promoter injected via intracerebroventricular route; EFS IV: AAV vector containing the EFS promoter encoding GFP injected via peripheral intravenous route. n=4 animals per group; uninjected controls n=3 animals. Mag:
265 Chapter 6 - Results magnification. Scale bars are 500 (low magnification) and 125 μm (high magnification), respectively.
GFP NeuN DAPI Overlay EFS IV
GFP NeuN DAPI Overlay CMV IV
GFP NeuN DAPI Overlay Control
Figure 88. Specific transduction of neurons using the EFS promoter revealed by GFP and NeuN immunofluorescence
Representative sections of brains from CMV IV: AAV vector containing the CMV promoter encoding GFP injected via peripheral intravenous route; EFS IV: AAV vector containing the EFS promoter encoding GFP injected via peripheral intravenous route, and uninjected controls; n=4 animals per group, except controls n=3 animals. Scale bar is 125 μm. White arrows show neurons.
266 Chapter 6 - Results
6.4 Discussion
The 3 main parameters considered when designing a vector for gene therapy and which can influence biodistribution are the AAV serotype, the promoter and the route of delivery (See Section 1.3.2) (240). Experimental evidence demonstrated that use of the AAV8 serotype with the EFS promoter when delivered intravenously was successful in targeting both liver and neurons. Of note, neonatal injection allowed a high rate of hepatocyte transduction across the hepatic lobule at 5 weeks of age. This is of particular interest when considering treatment for a persisting metabolic defect such as a urea cycle disorder in infancy where the liver is still growing. GFP immunostaining confirmed a high rate of transduced hepatocytes across the hepatic lobule. GFP ELISA showed that the liver was the main peripheral organ transduced. GFP immunostaining of brain sections showed a clear pattern of neuronal transduction throughout the brain, which was strongest in the cortex and decreased rostro-caudally.
In mouse models, AAV8 can widely transduce the brain after intracerebroventricular administration (443). After a systemic injection, most organs especially the liver are successfully targeted (442) but the neurotropism is influenced by the age of infusion and the dose of vector. Successful brain transduction with AAV8 and a CMV promoter was reported until day 14 of life after systemic delivery (1.5x10e11 vg per mouse) (476). Brain transduction was however nearly undetectable in adult mice after a similar experiment (1x10e11 vg per mouse) (442). Increasing the dose of vector improves brain transduction in adult mice, but this is accompanied by a higher risk of immune response, which could compromise a potential re-injection of AAV vector with the same construct if required. Indeed intravenous injection in adult mice with AAV8 and the EF1α promoter showed mild brain transduction at 1x10e11 to 2x10e12 vg per mouse (438, 478), but widespread neuronal and astrocytic transduction at 7.2x10e12 vg per mouse (approximately 2.9x10e14 vg/kg) (479). The transient permeability of the blood brain barrier for AAV in the neonatal mouse brain is not well understood and could be due to immaturity or receptor-mediated transcytosis (476, 480). The cell types transduced differ depending on the time of
267 Chapter 6 - Results injection. A predominant neuronal transduction is observed when gene therapy is administered during the first 48 hours of life switching to a preferential astrocytic transduction from day 3 onwards (443). Our observations confirm this data.
Other AAV serotypes like AAV9 in mice (321, 481) and in non-human primates (482, 483) have demonstrated a long-standing transgene expression of up to 18 months in mice after systemic neonatal delivery (476). As observed in this study with the CMV promoter, AAV9 is a serotype that is capable of transducing a widespread spectrum of peripheral organs as well as the central nervous system. This has been further illustrated by the current use of AAV9-based vector delivered intravenously in a clinical trial for spinal muscular atrophy and mucopolysaccharidosis IIIA (Clinical trials numbers NCT02122952 and NCT027162246, respectively, from clinicaltrials.gov accessed on 06/04/2017) (441).
This promoter study has major caveats as the work compares the GFP transduction of an AAV8 serotype in combination the EFS promoter with an AAV9 serotype in combination with the CMV promoter, with the latter being administered at a 50% higher dose. Therefore findings need to be interpreted very carefully. However for the purpose of validating the design of our vector, this study allows pertinent conclusions to be drawn: the data suggest that CMV promoter might be subject to silencing. Indeed the amount of GFP protein per vector genome was higher in the EFS injected groups compared to CMV groups. Silencing of the CMV promoter has been well documented in the liver, as previously published (464, 465), but the EFS promoter has been found more resistant (470). The respective pattern of cell types transduced in the brain has not been proven to be affected by the AAV serotypes but instead appears to be more dependent on the promoter and age at time of injection as discussed above (442, 443, 476, 479). The neuronal specificity of the EFS promoter has already been reported (472). This neuronal specificity is of particular interest when considering treatment of the brain of AslNeo/Neo mice, which are affected by an oxidative/nitrosative stress particularly within neurons (See Section 5.3.2). The rate of transduced hepatocytes persisting at 5 weeks of age after a single neonatal injection is still particularly high and meets the requirement to treat a severe liver monogenic condition associated with early death. Moreover the EFS promoter allows a relatively ubiquitous expression despite some cell-type specificity as illustrated with the neuronal pattern and decreasing rostro-caudal gradient (467). This is of interest
268 Chapter 6 - Results for argininosuccinic aciduria where argininosuccinate lyase activity is expressed in most tissues. The EFS promoter is already used in patients, especially in clinical trials for severe combined immunodeficiency syndrome with retroviral vectors (471). These trials provide further evidence with regards to the safety profile of this promoter, already known for its lower risk of insertional mutagenesis compared to other ubiquitous promoters (458).
6.5 Conclusion
The vector design, which combines an AAV8 serotype and an EFS promoter, when delivered via a peripheral intravenous route at birth has demonstrated its ability to target the liver and neurons, especially those in the cortex, with transduction persisting in adult mice after neonatal delivery.
The reduced risk of silencing and the safer profile of EFS, which has already been exploited in clinical trials when compared to other promoters, are further convincing reasons to use EFS to test gene therapy in AslNeo/Neo mice.
269 Chapter 7 - Results
7. Asl GENE THERAPY TO TREAT AslNeo/Neo MICE
7.1 Introduction
Argininosuccinic aciduria is the inherited disorder caused by argininosuccinate lyase deficiency, an enzyme involved in the urea and the citrulline-nitric oxide cycles (See Section 1.2.1). Patients present with a high rate of neurological complications, irrespective of the severity of hyperammonaemia and the age when conventional treatment is initiated (See Sections 1.2.6 & 3.7.3). Neuroimaging and data from the literature suggests that impaired nitric oxide metabolism play a role in this (See Section 3.7.2). Assays using tandem mass spectrometry technology have been developed to accurately study the argininosuccinate lyase activity in tissues (Chapter 4). The AslNeo/Neo mouse recapitulates the human disease and shows an oxidative/nitrosative stress in the brain, predominantly in neurons, particularly in the cortex (Chapter 5).
When developing a gene therapy approach to treating this condition, targeting the liver to correct the urea cycle is necessary. This alone however is not sufficient to address the neurological disease. Therefore, a specific vector has been designed to target both the liver and the brain. In the brain, the promoter specificity has shown a restricted transduction to neurons of the cortex and midbrain (See Section 6.3.3). The efficacy of this vector will be tested in adult and neonatal AslNeo/Neo mice to see if it can improve the phenotype, especially in the brain.
7.2 Experimental design
Initial experiments were designed to observe the long-term effect of the AAV- mediated gene therapy on the urea cycle in the liver. Thirty day-old mice received a single intraperitoneal injection. The purpose of injecting adult animals was to avoid 270 Chapter 7 - Results the period of rapid liver growth in mice observed during the first 4 weeks of life. Indeed, as AAV vectors are non-integrating, episomal copies of the transgene are lost with hepatocyte division.
To keep AslNeo/Neo mice alive until adulthood, a supportive treatment was tested. Therapeutic measures based on conventional treatment administered to patients with argininosuccinic aciduria were tested to allow survival of mice until adulthood estimated at day 30. This supportive treatment was based on : i) a protein-restricted diet. Protein-restricted diet was considered as a critical measure as most of the AslNeo/Neo mice were dying at time of weaning when starting solid food meaning eating food with higher protein content. The protein content of mouse milk and the usual diet (Harlan 2018) are 12.5% (484) and 18%, respectively. A modified diet was introduced with 5CR4_LabDiet (protein content 14.1%). ii) a medical treatment modified from Perez et al (485). Daily intraperitoneal injections of the ammonia scavenger drug sodium benzoate (100 mg/kg/d), and a supplementation of L-arginine (1 g/kg/d) were initiated from day 15 and from day 10, respectively. Oral administration was initially tested with drugs diluted in drinking water at at concentration allowing the aimed daily dose but this did not show any improvement on survival, which required an alternative route of delivery e.g. intraperitoneal.
The treatment was well tolerated and significantly increased the survival of AslNeo/Neo mice from 25% to 75% at day 30 (Figure 89) but not the growth (Figure 90).
.
271 Chapter 7 - Results
100
75
50 *** Survival (%) Survival 25 p=0.0002
0 0 10 20 30 Days
AslNeo/Neo Untreated AslNeo/Neo with supportive treatment
Figure 89. Supportive treatment in AslNeo/Neo mice improves survival
Kaplan-Meyer survival curve of AslNeo/Neo mice with (n=19; dotted line) and without (n=24; solid line) supportive treatment. Log rank test *** p<0.001.
20
15
10
Weight (grams) Weight 5
0 1 5 10 17 28 42
Days
Untreated AslNeo/Neo (N=14) AslNeo/Neo with supportive
treatment (N=5)
Figure 90. Growth of AslNeo/Neo mice with or without supportive treatment
Horizontal lines display the mean ± standard deviation. 272 Chapter 7 - Results
Supportive treatment consisted of a protein-restricted diet from day 15 to day 50, combined with daily intraperitoneal administration of an ammonia scavenger drug (Sodium benzoate, 100 mg/kg/d) and arginine supplementation (1 g/kg/d) which was initiated at day 10 and halted at day 31, 24 hours after gene therapy injection (Figure 91). This was administered to all AslNeo/Neo mice, receiving gene therapy (i.e. adult- injected AslNeo/Neo mice) or not (i.e. untreated AslNeo/Neo mice). Five adult AslNeo/Neo mice received intraperitoneal AAV vector (2.5x10e11 vg per mouse or approximately 2x10e13 vg/kg in a 50 µL volume). All animals were monitored until sacrifice at 12 months after gene therapy injection i.e. aged 13 month-old.
A second experiment was conducted in neonatal pups, which received intravenous AAV via the superficial temporal vein (3.2x10e11 vg per pup or approximately 2x10e14 vg/kg in a 20 µL volume) within 24 hours after birth. This early neonatal injection was intended to target the central nervous system as described in Section 6.3.3. Additionally, it was intended to target the liver during the first weeks of life. These mice did not receive any supportive treatment and were monitored until sacrifice at 9 months after gene therapy injection i.e. aged 9 month-old.
For each experiment, wild-type and untreated AslNeo/Neo littermates served as controls. Results of these 2 experiments are presented in parallel. Vector injections were performed by Dr Simon Waddington.
Intraperitoneal gene therapy
Intraperitoneal Sodium benzoate + Arginine
D10 D15 D30 D31 D50
Protein restricted diet
Figure 91. Supportive treatment given to AslNeo/Neo mice
273 Chapter 7 - Results
Daily intraperitoneal injections of sodium benzoate (100 mg/kg/d) and arginine (1 g/kg/d) were performed from day 10 to day 31 and a protein restricted diet from day 15 to day 50. 14 and 38 uninjected AslNeo/Neo mice served as controls for experiment conducted in adult and neonatally AslNeo/Neo mice, respectively.
7.3 Effect on the macroscopic phenotype
7.3.1 Survival
Survival was significantly improved in both adult- and neonatally-injected AslNeo/Neo mice compared to untreated AslNeo/Neo mice at 12 months and 9 months after gene therapy injection, respectively (Figure 92). At 13 months of age, survival rates were 5/5 and 2/16 in adult-injected and untreated AslNeo/Neo mice, respectively (Log-rank test: p=0.002). At 9 months of age, survival rates were 5/7 and 2/21 in neonatally- injected and untreated AslNeo/Neo mice, respectively (Log-rank test: p=0.006).
WT Untreated AslNeo/Neo Neo/Neo Adult injected Asl Neonatally injected AslNeo/Neo A B
Figure 92. Survival curves of adult- and neonatally-injected AslNeo/Neo mice
(A) Adult-injected compared to untreated AslNeo/Neo mice (n=5 and 16, respectively); (B) Neonatally-injected compared to untreated AslNeo/Neo mice (n=7 and 21, respectively).
274 Chapter 7 - Results
7.3.2 Growth
Gross observations revealed a rapid and sustained improvement of the growth (Figure 93A-C) in adult-injected compared to untreated AslNeo/Neo mice. This was transiently observed in neonatally-injected compared to untreated AslNeo/Neo mice during the first month of life (Figure 93D-G). This growth improvement was associated with a modification of the fur pattern, which looked similar to wild-type animals during the whole experiment and for the first month after gene therapy injection in adult- and neonatally-injected mice, respectively (Figure 93A-G).
Figure 93. Macroscopic aspect of gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice
(A-C) Images of wild-type, untreated and adult-injected AslNeo/Neo mouse with gene therapy. (D-G) Images of wild-type, untreated and neonatally-injected AslNeo/Neo mouse with gene therapy. (A-G) Scale bar: 1 cm. WT: Wild-type, Untr: Untreated AslNeo/Neo mouse, GT: Gene therapy-injected AslNeo/Neo mouse.
275 Chapter 7 - Results
Weight measurement showed a rapid (Figure 94A) and long-lasting (Figure 94B) improvement in adult-injected compared to untreated AslNeo/Neo mice, although this never reached the weight of wild-type animals. In neonatally-injected AslNeo/Neo mice, the weight was similar to wild-type animals up until 2 weeks of age however this was not sustained and became similar to that of untreated animals from 5 weeks of life onwards (Figures 94C, 94D). A significant difference in weight between neonatally- injected and untreated AslNeo/Neo mice was observed between day 10 and day 30 of life (Figure 94C). Growth velocity showed a maintained peak of growth in the 2 weeks following the injection of gene therapy in adult-injected animals (Figure 94E). In neonatally-injected mice, the growth speed was similar to wild-type animals until day 15 of life (Figure 94E), consistent with measured weight (Figure 94C).
7.3.3 Fur phenotype
Sparse, brittle fur, called trichorrhexis nodosa, is a salient characteristic of untreated AslNeo/Neo mice. In adult-injected mice in parallel with growth improvement, the fur pattern dramatically and sustainably improved compared to untreated AslNeo/Neo mice (Figure 95). The correction of the fur phenotype was observed within 2 weeks after gene therapy injection (Figure 95A). An improvement was observed in the hair shaft becoming more straight and regular, the medulla becoming wider and the restoration of the ability to grow and form physiological tips (Figures 95B-D). In neonatally- treated AslNeo/Neo mice, improvement of the fur phenotype was also apparent however this was only transient ending at the end of the first month of life.
276 Chapter 7 - Results
WT Untreated AslNeo/Neo Neo/Neo Adult injected Asl Neonatally injected AslNeo/Neo A B 30 40
30 20
20 ## 10 # Weight (grams) Weight Weight (grams) Weight 10 ↑ 0 0 10 20 30 40 50 60 70 0 50 100 150 200 Days Days C D 40
20 30
20 10 ## ** Weight (grams) Weight Weight (grams) Weight 10 # ** *** *** 0 0 10 20 30 50 100 150 200 Days Days
E 1.0 0.8
0.6
0.4
0.2
Growth speed (g/d) speed Growth 0.0 3 5 7 9 15 25 35 45 75 -0.2 Age (Days)
Figure 94. Growth of gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice
(A, B) Mean growth of adult-injected AslNeo/Neo mice (n=5) compared to WT (n=11) and untreated AslNeo/Neo mice (n=19) over (A) the first 2 months of life and (B) 12 months. (C, D) Mean growth of neonatally injected AslNeo/Neo mice compared to WT (n=31) and untreated AslNeo/Neo mice (n=41) (C) during the first month of life (n=13)
277 Chapter 7 - Results and (D) over 9 months (n=7). (E) Growth velocity during the first 2.5 months of life in all groups. Horizontal lines display mean ± standard deviation. Two-tailed unpaired t- test ** p<0.01, *** p<0.001, neonatally-injected compared to untreated AslNeo/Neo mice. # 30% and ## <15% of untreated AslNeo/Neo mice still alive. Blue and purple arrows indicate adult- and neonatal-injections, respectively.
Figure 95. Correction of the fur pattern in adult-injected injected AslNeo/Neo mice
(A) 30-days old AslNeo/Neo mouse before (left panel) and 15 days after (right panel) a single intraperitoneal injection of AAV8.EFS.mASL vector. (B, D) Representative images of hair from a 3 month-old WT, untreated and adult-injected AslNeo/Neo mouse. Scale bars: (B, D) 500 μm; (C) 125 μm; n=3.
278 Chapter 7 - Results
7.4 Effect on the urea cycle
7.4.1 Hepatomegaly
At time of sacrifice either for humane endpoints in sick mice or at the end of experiments, liver and body weight were recorded. Hepatomegaly, a feature of the liver disease in AslNeo/Neo mice was improved in 2 adult-injected mice (out of 5), with values of the ratio of liver/body weight being similar of those of wild-type animals. However this was not observed in the other adult- or neonatally-injected animals (Figure 96).
Figure 96. Hepatomegaly
Liver/body mass ratio in wild-type (n=18), untreated (n=23), adult-injected (n=5) and neonatally-injected AslNeo/Neo mice (n=4) of liver/body weight ratios at time of culling. Horizontal lines display the mean ± standard deviation. One-way ANOVA with Dunnett’s post-test compared to untreated AslNeo/Neo mice; *** p<0.001.
279 Chapter 7 - Results
7.4.2 Biomarkers
Various biomarkers of urea cycle activity demonstrated a robust long-term correction of the urea cycle in adult-treated but only partially in neonatally-treated AslNeo/Neo mice (Figure 97).
At one-month post gene therapy, plasma ammonia levels had normalised in both adult- and neonatally-injected groups compared to untreated AslNeo/Neo mice (Figure 97A) and persisted until sacrifice in both groups (Figure 97B). A progressive increase of ammonaemia in both treated groups likely reflects the progressive loss of transgene copies caused by hepatocyte turnover, however increasing ammonia levels remained in therapeutic values (< 100 μM). Plasma argininosuccinic acid levels were significantly decreased at one month post gene therapy in adult-injected but not in neonatally-injected mice compared to untreated AslNeo/Neo mice; this effect in adult-treated mice persisted over time (Figures 97C & 97D, respectively). Arginine levels in dried blood spots were significantly increased at one month post gene therapy in adult-injected but not in neonatally-injected mice compared with untreated AslNeo/Neo mice although plasma levels of the two groups had converged by the end of the study (Figures 97E & 97F, respectively). At one month post gene therapy, citrulline levels in dried blood spots were significantly lower in adult-injected compared with uninjected and neonatally-injected AslNeo/Neo mice although there was no difference between groups at the end of the study since citrulline levels in AslNeo/Neo mice progressively decreased (Figures 97E & 97F, respectively). As described in Section 1.2.2, plasma glutamate and glutamine represent a buffer system in case of elevated levels of ammonaemia. In adult-injected and neonatally- injected mice, these levels in dried blood spots were not significantly different from wild-type animals at one-month post gene therapy injection although these levels are significantly increased in untreated AslNeo/Neo mice (Figure 97I). This correction of glutamate and glutamine levels however was no longer observed in neonatally- injected mice at 3 months post-injection, although ammonia levels remained normal (Figure 97J). This indicates that the long-term ureagenesis in neonatally-treated animals was just sufficient to maintain the levels of ammonia within physiological ranges with the help of this glutamate-glutamine buffer pathway. In adult-treated animals, glutamate and glutamine levels in dried blood spots were not significantly
280 Chapter 7 - Results different from wild-type levels, except at 12 months after gene therapy injection, contemporary to the progressive mild increase of ammonia levels attributed to progressive loss of liver transgene copies (Figure 97J).
Urinary orotic acid levels were significantly increased in AslNeo/Neo mice at 3 and 10 weeks of age compared with wild-type mice (Unpaired two-tailed t-test, p<0.0001 and p=0.003, respectively; analysis of log-transformed data at 3 weeks of age due to non- normality of the data) (Figure 97K). It is noteworthy that values of orotic aciduria at 3 weeks of age in AslNeo/Neo mice segregated into 2 groups, one with high concentrations in mice which died in the following week before 1 month of age, and a second group of animals with lower concentrations, which survived for longer. Therefore values at 10 weeks of age appear to improve in untreated AslNeo/Neo mice as the sicker ones have already died (Two-tailed unpaired t-test, p=0.04) (Figure 97K). In both treated groups, although no difference in mean concentrations of orotic acid was observed compared to surviving untreated mice, it is interesting that this parameter normalised in 2 adult-treated animals (Figure 97K).
281 Chapter 7 - Results
WT Untreated AslNeo/Neo
Neo/Neo Neo/Neo Adult injected Asl Neonatally injected Asl **
** A ns B 600 ** 800 ns * mol/L) *** µ * * mol/L) 400
µ 600
400 200
200 Plasma ammonia ( ammonia Plasma
Plasma ammonia ( ammonia Plasma 0 0 M1 M2 M6 M9 M12
1200 *** *** ***
*** M) ns *** * *** * C D µ * *** *** 1500 *** *** *
M) ***
µ 800 *
1000
400 500 Argininosuccinic acid ( acid Argininosuccinic
Argininosuccinic acid ( acid Argininosuccinic 0 0 M1 M2 M3 M4 M6 M12
*** * ns ns *** ns 120 ** * * ns 150 *** ns * ns E F ns ns ns ) M µ M) 100 80 µ
50 Arginine (
Arginine ( Arginine 40
0 M1 M2 M3 M4 M6 M12 *** 600 *** G *** H *** *** ns ns ns *** 600 *** *** ***
M) * 400 (µ * M) *** 400 (µ ** ** 200
200 Citrulline Citrulline
0 0 M1 M2 M3 M4 M6 M12
282 Chapter 7 - Results
ns * I ns J ns *** ns *** ** 1500 M) *** M) µ ns * µ *** 1200 ns *** *** ns ** 1000 1000
800 500
600 Glutamate + Glutamine ( Glutamine + Glutamate Glutamate + Glutamine ( Glutamine + Glutamate 0 M1 M2 M3 M4 M6 M12
K 4500 *** 3000 1500 ns *** ns 300 Orotic acid Orotic 150 mol / mol creatinine mmol µ 0 3 weeks 10 weeks
Figure 97. Biomarkers of the urea cycle
Measurement of biomarkers in wild-type (n=5-18), untreated (n=6-14), adult-injected (n=4-5) and neonatally-injected AslNeo/Neo mice (n=3-6) of (A, B) Plasma ammonia levels (A) at 1 month post gene therapy injection and (B) over time; (C, D) argininosuccinic acid levels in dried blood spots (C) at 1 month post gene therapy injection and (D) over time; (E, F) arginine levels in dried blood spots (E) at 1 month post gene therapy injection and (F) over time; (G, H) citrulline levels in dried blood spots (G) at 1 month post gene therapy injection and (H) over time; (I, J) glutamate and glutamine levels in dried blood spots (I) at 1 month post gene therapy injection and (J) over time, (K) urine orotic acid at 3 and 10 weeks of age. Horizontal lines display the mean ± standard deviation. One-way ANOVA with Dunnett’s post correction compared to wild-type mice; ns: not significant, * p<0.05, ** p<0.01, *** p<0.001. M: months. (B, D, F, H, J) the timeline presented in axis corresponds to the time after gene theraoy injection (in months).
283 Chapter 7 - Results
Plasma alanine aminotransferase (ALT) levels were assessed at time of culling and were not significantly different from wild-type values in both adult- and neonatally- injected AslNeo/Neo mice (Figures 98A, 98B). Plasma creatinine levels, which were measured as a valuable endpoint in 2 month-old mice (See Section 5.2.3), did not show any differences between wild-type, untreated and both treated AslNeo/Neo mice, when measured in surviving older untreated AslNeo/Neo mice (Figures 98C, 98D).
WT Untreated AslNeo/Neo Neo/Neo Adult injected Asl Neonatally injected AslNeo/Neo
*** ns A B 250 * ns 400 ns ns *** 200 300 *** 150 200 100
100 50
0 0 Alanine aminotransferase (IU/L) aminotransferase Alanine Alanine aminotransferase (IU/L) aminotransferase Alanine M2 M9 M12
* ns C D 30 ns ns 30 ns ns M) µ
M) 20
µ 20
10 10 Creatinine ( Creatinine (
0 0 M2 M9 M12
Figure 98. Alanine aminotransferase activity and creatinine concentrations in wild-type and AslNeo/Neo mice
Measurement of alanine aminotransferase activity and creatinine in plasma of wild- type (n=13-18), untreated (n=5-12), adult-injected (n=4-5) and neonatally-injected AslNeo/Neo mice (n=4). (A, B) plasma alanine aminotransferase levels (A) at 9-12 months of age and (B) over time; (C, D) plasma creatinine levels (C) at 9-12 months
284 Chapter 7 - Results of age and (D) over time. Horizontal lines display the mean ± standard deviation. One-way ANOVA with Dunnett’s post-test compared to wild-type mice except for (B) at time point M2 where two-tailed unpaired t-test was used; ns: not significant, * p<0.05, *** p<0.001. (B, D) Colours indicate the outcome of the statistical test for untreated (red), adult-injected (blue) and neonatally-injected (purple) AslNeo/Neo mice compared to wild-types.
7.4.3 Hepatocyte transduction and ASL enzymatic activity
Residual liver argininosuccinate lyase activity was 14.5 % of wild-type activity in untreated AslNeo/Neo mice. This increased significantly to 47% and 18.5% in adult- and neonatally-injected groups, respectively, at the time of harvest (Figure 99A).
Vector genome copy number in liver samples was 0.99 ± 0.32, 0.69 ± 0.07 and 0.1 ± 0.09 in adult-injected, neonatally-injected and untreated AslNeo/Neo mice, respectively (mean ± standard deviation). The difference was not significant between both treated groups but significantly higher between both treated groups compared to untreated AslNeo/Neo mice (One-way ANOVA with Bonferronni post-test, p>0.05, p<0.001 and p<0.01, respectively) (Figure 99B). Whilst there was a large range of argininosuccinate lyase activity per vector genome in the adult-injected group, this was not statistically different to that of the neonatally treated AslNeo/Neo mice (Figure 99C).
Quantification of liver immunostaining against argininosuccinate lyase measured by an automated method of quantification (see Section 2.7.14) showed significantly greater staining in the wild-type and the adult-injected group but not in the neonatally- injected group relative to the untreated AslNeo/Neo mice (One-way ANOVA with Dunnett’s post-test, p<0.001, p<0.01 and p>0.05, respectively) (Figure 99D). Following adult injection, expression was predominant in perivenous hepatocytes whereas after neonatal injection more scattered and sparse expression was observed (Figure 100).
Haematoxylin and eosin (H&E) staining of liver samples showed cytoplasmic unstained vacuoles with a diffuse and regular pattern across the hepatic lobule.
285 Chapter 7 - Results
Periodic acid-Schiff (PAS) staining confirmed that these vacuoles were deposits of glycogen as previously discussed (See Section 5.2.4). This feature was markedly improved in the 3 adult-injected mice with the highest argininosuccinate lyase activity, but no improvement was seen in the 2 other adult-injected with lower residual ASL activity and all neonatally-injected AslNeo/Neo mice (Figure 101).
WT Untreated AslNeo/Neo Neo/Neo Adult injected Asl Neonatally injected AslNeo/Neo
A ns B *** ns 400 *** *** 1.5 **
300 1.0
200
0.5 100 activity (nmol/ng/min) vector genome / nucleus vector genome Argininosuccinate lyase lyase Argininosuccinate 0 0.0
C ns D 80 2 50 ***
40 60 ns 30 40 ** 20 20 10 ASL enzymatic activity ASL enzymatic % immunoreactivity / mm 0 0 (nmol/ng/min) per vector genome per (nmol/ng/min)
Figure 99. Argininosuccinate lyase activity, vector genome copy number and immunostaining in liver samples
(A) Argininosuccinate lyase activity, (B) vector genome copy number, (C) enzyme activity per vector genome in liver samples and (D) staining quantification of wild- type, untreated, adult-injected and neonatally-injected AslNeo/Neo mice. Horizontal lines display the mean ± standard deviation. One-way ANOVA with (A, B) Dunnett’s post-test compared to untreated AslNeo/Neo mice and (C) Bonferronni post-test; (D) Two-tailed unpaired t-test; ns: not significant, ** p<0.01, *** p<0.001.
286 Chapter 7 - Results
Figure 100. Liver argininosuccinate lyase immunostaining
Representative images of liver from (A) wild-type, (B) untreated, (C) adult-injected and (D) neonatally-injected AslNeo/Neo mice stained for argininosuccinate lyase. Scale bars: 500 μm; n=3 animals per group.
287 Chapter 7 - Results
Figure 101. Microscopic liver architecture
Representative images of liver from wild-type, untreated, adult-injected and neonatally-injected AslNeo/Neo mice stained with haematoxylin and eosin (H&E) highlighting vacuolar cytoplasmic deposits and periodic acid-Schiff (PAS) demonstrating glycogen deposits. Scale bars are 125 μm (Low Mag) and 500 μm (High Mag), respectively; n=3 animals per group. Mag: magnification.
288 Chapter 7 - Results
7.5 Effect on the citrulline-nitric oxide cycle in the liver
Liver nitric oxide levels, assessed by analysis of nitrite/nitrate levels, were significantly lower in untreated AslNeo/Neo mice and neonatally-injected mice but showed an improvement in adult-injected mice compared with wild-type mice (Figure 102). This suggests that the increase of argininosuccinate lyase activity can partially restore the nitric oxide production in this organ.
WT Untreated AslNeo/Neo Neo/Neo Adult injected Asl Neonatally injected AslNeo/Neo
ns ***
1500 ** M) µ 1000
500 Nitrate + Nitrite (
0
Figure 102. Liver nitrate/nitrite levels
Liver nitrite/nitrate levels of wild-type, untreated, adult-injected and neonatally- injected AslNeo/Neo mice. Horizontal lines display the mean ± standard deviation. One- way ANOVA with Dunnett’s post-test compared to wild-type; ns: not significant, ** p<0.01, *** p<0.001.
7.6 Effect on the citrulline-nitric oxide cycle in the brain
Argininosuccinate lyase activity was measured in the cortical tissue. Activity in both untreated and adult-injected Asl Neo/Neo mice were both significantly lower than that in wild-type mice (14.1% and 16.2% of wild-type activity, respectively; One-way ANOVA with Dunnett’s post-test, p<0.01 and p<0.01, respectively). However activity in neonatally-injected mice was dramatically higher than that of untreated or neonatally-
289 Chapter 7 - Results treated mice. Activity was lower than, but not significantly different from wild-type activity (64.8% of wild-type activity, p>0.05) (Figure 103A).
To assess the effect of this increase of argininosuccinate lyase activity on the nitric oxide metabolism in brains of neonatally-treated AslNeo/Neo mice, tissue nitrite/nitrate levels were measured. Compared to wild-type brains, nitrite/nitrate levels were higher in untreated and in adult-injected AslNeo/Neo mice (One-way ANOVA with Dunnett’s post-test analysis of log-tranformed data, p<0.01 and p<0.01, respectively), but were not significantly different in neonatally-injected mice (p<0.01) (Figure 103B). To investigate the physiological nitric oxide-cGMP pathway, cGMP levels in brain homogenates were quantified. No significant difference was seen between experimental groups although cGMP levels in untreated and adult-injected groups were increased compared to wild-type mice (One-way ANOVA with Dunnett’s post- test of log-tranformed data, p>0.05 and p>0.05, respectively) (Figure 103C). In the neonatally-injected group, amounts of cGMP detected were particularly variable ranging from near undetectable levels similar to those observed in wild-types to extremely high values, with no significance compared to wild-types (p>0.05) (Figure 103C).
To study if the decrease of nitrite/nitrate levels was correlated with a modification of the oxidative/nitrosative stress, cortical immunostaining against nitrotyrosine was performed (Figure 104). Quantification showed that, compared to wild-type mice, a significantly greater area of staining was observed in untreated and adult-injected Asl Neo/Neo mice (One-way ANOVA with Dunnett’s post-test, p<0.05 and p<0.01 respectively). In contrast, staining in neonatally-injected animals, which was reduced, was not statistically different compared to wild-types (One-way ANOVA with Dunnett’s post-test, p>0.05) (Figure 103D).
290 Chapter 7 - Results
WT Untreated AslNeo/Neo Neo/Neo Adult injected Asl Neonatally injected AslNeo/Neo
ns A ns B * 300 ** ** 150 ** M) µ 200 100
100 50 (nmol/ng/min) Nitrate + Nitrite (
0 0 Argininosuccinate lyase activity lyase Argininosuccinate C D ns 60 * 30 ** 2 *
40 ns 20 ns
20 10 cGMP (pmol/mg) / protein % immunoreactivity / mm 0 0
Figure 103. Cortical argininosuccinate lyase activity and nitric oxide metabolism in brain
(A) Cortical argininosuccinate lyase activity. (B) Nitrite/nitrate concentration and (C) cGMP levels in brain homogenates. (D) Computational quantification of nitrotyrosine immunostaining. Horizontal lines display the mean ± standard deviation. (A-D) One- way ANOVA with Dunnett’s post-test compared to wild-types: ns - not significant; * p<0.05; ** p<0.01. WT n=3-8; untreated AslNeo/Neo n=3-6; adult-injected AslNeo/Neo n=3- 5; neonatally-injected AslNeo/Neo n=3-5. cGMP: cyclic guanosine monophosphate.
291 Chapter 7 - Results
Figure 104. Cortical nitrotyrosine immunostaining
Representative images of nitrotyrosine-stained sections of cortex in (A, B) wild-type, (C, D) untreated, (E, F) adult-injected and (G, H) neonatally-injected AslNeo/Neo mice (n=3 animals per group). Scale bars of low and high magnification: 500 and 125 μm, respectively.
292 Chapter 7 - Results
7.7 Functional impact on motor and neurological functions
7.7.1 Behaviour
Two behavioural tests identified previously as discriminant between wild-type and AslNeo/Neo mice (see Section 5.3.1) were performed to assess the effect of gene therapy on both treated groups.
The open field test was assessed at 2 months of age for all groups. There was a significant reduction in the walking distance in the untreated AslNeo/Neo mice and an improvement in both treated groups with a greater effect evident in the adult-injected than in the neonatally-injected group (One-way ANOVA with Dunnett’s post-test, p<0.001, p<0.05 and p<0.01, respectively) (Figure 105A).
Latency to fall from an accelerating rotarod was also measured at 2 months of age. Compared to wild-type mice, latency was significantly reduced in untreated AslNeo/Neo mice but were not significantly different in both adult- and neonatally-injected groups (One-way ANOVA with Dunnett’s post-test, p<0.05, p>0.05 and p>0.05, respectively) (Figure 105B).
WT Untreated AslNeo/Neo Neo/Neo Adult injected Asl Neonatally injected AslNeo/Neo * A ns B ns 50 600 *** ns 40 ** 400 30
20 200 10 (seconds) Time Distance walked (m)
0 0
Figure 105. Behavioural testing
(A) Open field test and (B) accelerating rotarod performed in 2 month-old mice. Horizontal lines display the mean ± standard deviation. (A, B) One-way ANOVA with Dunnett’s post-test compared to wild-type mice: ns - not significant, * p<0.05, ** 293 Chapter 7 - Results p<0.01, *** p<0.001. Wild-type n=20-22; untreated AslNeo/Neo n=8-9; adult-injected AslNeo/Neo n=5-6; neonatally-injected AslNeo/Neo n=5-6. WT: Wild-type.
7.7.2 Cell death in the brain
Cell death was assessed by terminal deoxynucleotidyl transferase dUTP nick end- labeling (TUNEL) staining. Compared with wild-type mice, there was a dramatic and significant increase in the cortex of untreated AslNeo/Neo mice; there was a smaller, but nonetheless significant reduction in adult-treated AslNeo/Neo mice (One-way ANOVA with Dunnett’s post-test of log-tranformed data, p<0.001 and p<0.05, respectively) (Figures 106A, 106B). In neonatally-injected animals, there was no significant difference compared to wild-type mice (p>0.05) (Figure 106B).
In the cerebellum, a significant higher rate of cell death was noted in untreated AslNeo/Neo mice compared to wild-type mice whilst this parameter was not significantly elevated in both treated groups compared to wild-types (One-way ANOVA with Dunnett’s post-test of log-tranformed data, p<0.01, p>0.05 and p>0.05, respectively) (Figures 106C, 106D).
294 Chapter 7 - Results
WT Untreated AslNeo/Neo Neo/Neo A Adult injected Asl Neonatally injected AslNeo/Neo B ns * 30 ***
20
10 TUNEL Positive cells Positive TUNEL
0 C D ns 80 ns ** 60
40
20 TUNEL Positive cells Positive TUNEL
0
Figure 106. TUNEL staining in the cortex and the cerebellum
(A) Representative images of TUNEL positive cells in the cortex of an untreated 25 day-old AslNeo/Neo mouse. (B) Counting of TUNEL positive cells in the cortex at time of culling in wild-type (aged 9 - 12 months; n=3), untreated (aged 21 days - 12 months; n=9), adult- (aged 12 months; n=3) and neonatally-injected AslNeo/Neo mice (aged 9 months; n=3). (C) Representative images of TUNEL positive cells in the cerebellum of a 35 day-old untreated AslNeo/Neo mouse symptomatic of hyperammonaemia and culled for humane endpoint. (D) Counting of TUNEL positive cells at time of culling in the cerebellum of wild-type (aged 9 - 12 months; n=3), untreated (aged 21 days - 12 months; n=3), adult- (aged 12 months; n=3) and neonatally-injected AslNeo/Neo mice (aged 9 months; n=3). Horizontal lines display the mean ± standard deviation. (B, D) One-way ANOVA with Dunnett’s post-test compared to wild-type mice: ns - not significant, * p<0.05, ** p<0.01, *** p<0.001. Scale bars: 500 μm and 125 μm in main and inset pictures, respectively. WT: Wild- type. For each sample, counting of TUNEL positive cells was performed from 10 random pictures. 295 Chapter 7 - Results
7.8 Discussion
7.8.1 Correction of the urea cycle
Long-term correction of plasma ammonia levels was observed in mice injected as adults and neonates. This demonstrated the restoration of an ureagenesis pathway able to successfully detoxify ammonia and normalise or dramatically improve survival after adult and neonatal injection, respectively. This metabolic effect persisted for 12 and 9 months, respectively. Although plasma ammonia levels rose over time, rising quicker in the neonatally-injected group, this sustained critical outcome lasted for half the lifetime of these mice. This is consistent with previous studies demonstrating long-term correction after a single systemic injection of AAV-mediated gene therapy in animal models of liver inherited metabolic diseases including urea cycle defects, organic acidurias, phenylketonuria, glycogen storage disease type Ia, long chain fatty acid oxidation disorders, homozygous familial hypercholesterolemia, primary hyperoxaluria type I, and progressive familial intrahepatic cholestasis (220, 221, 240). Plasma alanine aminotransferase levels were sustainably improved in both treated groups suggesting that this control of ammonia levels had a direct impact on liver pathophysiology.
Adult-injected AslNeo/Neo mice exhibited a more complete correction of the urea cycle and the liver disease, although variability between animals was apparent in the long- term. Various parameters were much improved in all animals including growth, fur pattern, plasma amino acids but others like orotic aciduria and clearance of liver glycogen deposits were only normalised in some animals. Disease-specific biomarkers similarly to clinical signs depend on the level of residual enzymatic activity. For example, controlling orotic aciduria in the Spfash mouse model, a model of ornithine transcarbamylase deficiency, required 5 times more vector and a much higher residual ornithine transcarbamylase activity compared to that necessary to normalise ammonaemia (308). In this respect, our study adds more knowledge about the hierarchisation of biomarkers in the phenotype of argininosuccinic aciduria. Ammonaemia can be normalised with 18% of wild type liver activity of argininosuccinate lyase, which are the mean values in neonatally-treated mice. 296 Chapter 7 - Results
Plasma amino acid levels and orotic aciduria require a liver argininosuccinate lyase activity >39%, i.e. at least twice the level of activity required to control ammonaemia. The spontaneous reduction of orotic aciduria observed between 3 to 10 weeks in untreated Asl Neo/Neo mice might be explained by the fact that the 3 weeks time point is contemporary to a high metabolic stress caused by weaning. Weaning is associated with temporary reduction in food intake, which can precipitate a hyperammonaemic crisis. Orotic acid shows higher levels in early-onset patients compared to late-onset patients in urea cycle defects such as ornithine transcarbamylase deficiency (486). This suggests that orotic aciduria levels are correlated with the severity of hyperammonaemia. At 10 weeks, surviving animals should be at a more compensated metabolic state, therefore with reduced levels of orotic aciduria.
The fur phenotype observed in argininosuccinate lyase and argininosuccinate synthase deficient mice is likely caused by hypoargininaemia as arginine represents up to 10% of hair composition (18). The long-term phenotypic improvement of the fur in adult-injected mice is in accordance with measured arginine levels.
The pattern of hepatocyte transduction shows a preferential pattern of transduction in perivenous hepatocytes. This has been documented with systemic AAV8-mediated gene therapy in various species such as mice and dogs (See Section 1.3.6) (293, 300).
The variability of the phenotypic correction observed in adult-injected animals at 12 months post injection was correlated with the liver residual activity of argininosuccinate lyase ranging from level similar to those seen in untreated AslNeo/Neo mice to those seen in wild-type mice. This might be partly explained by difficulties in the reproducibility of delivery by intraperitoneal injection. Some work has highlighted the risk of error in the technique of intraperitoneal injection from 10 to 20% with misplacement of the needle causing intracaecal, intracystic, subcutaneous, intravascular, intragastric or intravascular injections (487, 488). Insufficient dilution can also cause a trapping in the peritoneal cavity, which prevents uniform uptake through the mesenteric veins (489). This is even more appropriate in this work as untreated AslNeo/Neo mice are smaller with a thick skin, which increases the risk of errors. Vector administered in the adult-injected group came from 2 different batches
297 Chapter 7 - Results of vectors, produced and titrated similarly as presented in Section 2.13.4. However an error margin might exist in the process of titration and could cause some differences in the titer administered.
The physiological activity of the urea cycle prevails in periportal hepatocytes according to the metabolic zonation of the liver (2). Although this pattern of liver transduction observed in the adult-injected group does not match this physiological activity, this does not prevent a successful correction of the pathway. As predicted from previous studies (169, 239), a single neonatal injection was not sufficient to fully correct the phenotype as the mice reached adulthood. Neonatally-injected AslNeo/Neo mice displayed a transient correction of the macroscopic phenotype including fur and growth within the first couple of weeks after injection only. Since the liver mass increases and multiplies by 30 fold during the first 8 weeks of life (239), this temporary correction is likely caused by the loss of the episomal transgene copies. It is acknowledged that systemic neonatal delivery allows the persistence of 4-8% of transduced hepatocytes (239). These results are in line with previous successful experiments using AAV8-derived vectors in neonatal murine models of urea cycle defects, which showed a loss of efficacy after the first weeks of life (300, 304, 305). However in this thesis, neonatally treatment normalised ammonaemia, which is intriguing. This is likely to be multifactorial. The AslNeo/Neo mouse model is hypomorphic with a high residual argininosuccinate lyase activity. Therefore the increase required to reach the threshold necessary to control ammonia levels is lower compared to a knockout animal model. Argininosuccinic acid as the accumulating biomarker of the disease enables the excretion of two nitrogen moities, which partly excretes nitrogen in bypassing the final steps of ureagenesis. This is also observed in another distal urea cycle defect, arginase deficiency, in which the excess of arginine allows the excretion of the 2 nitrogen residues. The EFS promoter has a strong promoter activity and is less subject to methylation and silencing (470). Its diffuse pattern of liver transduction driving a high persisting transgene expression 5 weeks after a single systemic neonatal injection is another advantage observed during the experiment validating the vector construct (See Section 6.3.2). Previous publications regarding urea cycle defects have shown that a modest increase in the residual activity of the enzyme can have a significant clinical impact on the phenotype. AAV-mediated gene therapy in an arginase-deficient mouse showed that
298 Chapter 7 - Results a minimal ureagenesis of 3.3% of wild-type levels was enough to restore survival and to reduce ammonia levels partially (307). Similarly, a small increase in liver enzyme activity (as little as 3%) has already been shown to result in a less severe phenotype in ornithine transcarbamylase deficiency, the most common urea cycle defect (490).
Extrapolating recently published results of a liver-directed clinical trial for acute intermittent porphyria (188), it has been suggested that non-integrated episomes may have a potentially reduced transgene expression compared to endogenous chromosomal alleles (213). This may also explain the difficulty to normalise orotic aciduria as observed in ornithine transcarbamylase deficient mice (294).
7.8.2 Correction of the citrulline-nitric oxide cycle
Adult-injected mice resulted in a small but significant increase in liver nitrite/nitrate levels, compared to untreated AslNeo/Neo mice. The argininosuccinate lyase activity required to achieve this is relatively high as this improvement was not observed in neonatally-injected mice. As low levels of reduced glutathione are observed in untreated AslNeo/Neo mice, it suggests that oxidative stress is a part of the liver pathophysiology. This finding is in line with previous reports assessing the pathophysiology of argininosuccinic aciduria in this AslNeo/Neo mouse demonstrating systemic nitric oxide deficiency (32) and increased oxidative stress (85). Improvement of nitric oxide deficiency should theoretically decrease this toxicity although parameters looking at oxidative stress have not been appropriately tested in liver samples of both gene therapy-injected groups. Besides correction of ureagenesis, this partial correction of nitric oxide deficiency in the liver is of particular interest as this shows the dual corrective effect of argininosuccinate lyase activity on both metabolic pathways, the urea and the citrulline-nitric oxide cycles.
In the brain, argininosuccinate lyase activity in the cortex of the neonatally-injected AslNeo/Neo mice was found to be significantly increased compared to both untreated and adul-injected groups. As this vector construct targets neurons specifically as shown in Section 6.3.3, we can conclude that a single systemic neonatal injection allows a sustained correction of the neuronal argininosuccinate lyase activity up to at
299 Chapter 7 - Results least 9 months of age. This enzymatic increase dramatically reduces the nitrosative/oxidative stress, displaying a massive reduction in nitrite/nitrate levels in the brain and nitrotyrosine staining in the cortex. However the disturbed nitric oxide metabolism observed in the AslNeo/Neo mouse was not modified in the adult-injected group, which achieved the most complete correction of the urea cycle. Therefore this oxidative/nitrosative stress in the brain is not caused by hyperammonaemia, but can be accounted for as a direct consequence of the disturbed in situ nitric oxide metabolism caused by cerebral argininosuccinate lyase deficiency. As observed in Section 5.3.2, staining for nitrotyrosine, and the neuronal and inducible nitric oxide synthases occurs essentially within neurons. The vector construct used was also shown to specifically target neurons (Section 6.3.3). Thus correcting the neuronal activity of argininosuccinate lyase can successfully overcome this oxidative/nitrosative stress.
7.8.3 Pathophysiology of the brain disease in argininosuccinic aciduria
A dual neuropathophysiological mechanism
Compared to other urea cycle defects, the neurological disease in argininosuccinic aciduria is a paradox as this disease shares with arginase deficiency the lowest rate of hyperammonaemic decompensations but has the highest rate of neurological complications, which include neurocognitive delay, abnormal neuroimaging, epilepsy, ataxia.
There is no doubt that hyperammonaemia causes severe neurological symptoms in AslNeo/Neo mice as observed for example at time of weaning when most of the untreated animals die. The mice are ataxic, easily falling down, exhibiting repetitive behaviour with redundant circle motion, and later will look prostrated, lethargic, and refuse to move. At this stage, these animals were culled for humane endpoints. Concomitant ammonaemia levels were systematically very high. This is in accordance with observations in patients affected with argininosuccinic aciduria. The correction of ureagenesis fully amended these neurological signs in treated AslNeo/Neo mice. 300 Chapter 7 - Results
In the brain, hyperammonaemia increases extracellular glutamate, which generates nitric oxide production and oxidative stress via expression of neuronal and inducible nitric oxide synthases (39). Glutamate-mediated neuronal toxicity is mediated by cGMP (491). In our study, oxidative stress observed in untreated AslNeo/Neo mice was associated with high levels of cGMP compared to wild-type mice. However as increased cGMP is persisting in the adult-treated group, which has normal ammonaemia and plasma glutamate/glutamine levels, this finding is very unlikely to be caused by glutamate toxicity.
However the oxidative/nitrosative stress, observed as a hallmark of the neuropathology in this mouse model, was not modified by correction of hyperammonaemia only, as observed in the adult-injected group. Conversely the neonatally-injected group displayed a long-term correction of plasma ammonia levels and neuronal argininosuccinate lyase activity in the cortex, which was concurrent with a marked decrease of the oxidative/nitrosative stress. Therefore this demonstrates that the oxidative/nitrosative stress seen in ASL deficiency is independent of hyperammonaemia.
Functional benefits in correcting the neuropathophysiology
Further experiments were performed to assess the functional effect of these pathophysiological mechanisms. Behavioural testing showed significantly reduced activity in untreated AslNeo/Neo mice for open field test and accelerating rotarod. In the open field test, both treated groups showed a significant improvement, which was slightly better in the adult-treated group. This suggests that the improvement of performance tested is closely linked to ammonia levels. On the rotarod, again both treated groups exhibited an advantage over untreated AslNeo/Neo mice, but this effect was significantly higher in the neonatally-treated group, suggesting that this test assessed a performance based on a combination of both neuropathological mechanisms, hyperammonaemia and cerebral oxidative/nitrosative stress.
Assessment of cell death in the cortex and cerebellum showed a significant reduction in both treated groups compared to untreated AslNeo/Neo mice, confirming the paramount role of hyperammonaemia in causing this feature. However, in the cortex,
301 Chapter 7 - Results a further significant reduction was observed between adult- and neonatally-injected groups suggesting an ancillary role of disturbed nitric oxide metabolism in this process. This was not observed in the cerebellum. This could be explained by the compartimentalisation of this pathophysiological mechanism. Indeed, nitrite/nitrate levels of untreated AslNeo/Neo mice in different areas of the brain did show increased levels in the cortex, the midbrain but not in the cerebellum when compared with wild- type mice (See Section 5.3.3), insinuating that nitric oxide metabolism is less affected in this area of the central nervous system. Moreover, our vector construct has shown very poor transduction of the cerebellum (See Section 6.3.3). Therefore, it might not be surprising that no beneficial effect is observed in cell death rates in the cerebellum of neonatally-treated group compared to the adult-treated one.
Neuropathophysiology in argininosuccinic aciduria
Previous clinical reports have hypothesised that disturbed nitric oxide metabolism is a potential cause of the neurological disease seen in patients with argininosuccinate lyase deficiency (85, 337). The neurological phenotype of the AslNeo/Neo mouse model had not been described before to help in clarifying this assumption.
In AslNeo/Neo mice brains, an increase of nitric oxide (assessed by nitrite/nitrate levels) and cGMP concentrations suggested a persisting physiological up-regulation of the nitric oxide/cGMP pathway, which is observed with appropriate coupling of nitric oxide synthase (See Section 5.3.3) (423). This nitric oxide/cGMP pathway is strongly neuroprotective especially with regards to oxidative stress (492, 493).
However this increase of nitric oxide levels can also be associated with oxidative stress and increased production of reactive oxygen species, especially the superoxide ion and peroxynitrite (See Section 1.2.2). Peroxynitrite nitrates tyrosine residues, resulting in the formation of 3-nitrotyrosine, a feature of nitrosative/oxidative stress (See Section 5.3.3) (424). This process is highly selective involving specific tyrosine residues in selected proteins and can significantly affect the protein structure and function resulting in decreased enzymatic activity, enabling protein aggregation or triggering immune response (494-496). The generation of these reactive oxygen species can occur in the case of uncoupling of nitric oxide synthase, a process
302 Chapter 7 - Results facilitated by reduced tissue arginine and associated with an increase of systemic biomarkers of oxidative stress (85). Peroxynitrite is a powerful damaging compound, which subsequently contributes to the decrease of antioxidants, the inhibition of the respiratory chain and the opening of the permeability transition pore and cell death (425). In this study, the increase of both nitrite/nitrate levels and neuronal nitrotyrosine staining account for this oxidative/nitrosative stress likely caused by low tissue arginine levels and nitric oxide synthase uncoupling. Previous publications have suggested that oxidative stress could play a role in the neuropathophysiology in urea cycle defects associated with reduced tissue arginine via a neuronal disease (497). In argininosuccinic aciduria, nitric oxide synthase coupling and uncoupling might co-exist in the brain and both the physiological pathway and oxidative stress contribute to the increase of nitrite/nitrate levels (Figure 107).
Figure 107. Proposed pathophysiology involved in the brain disease of AslNeo/Neo mice
(A) Coupled nitric oxide synthase (NOS) produces nitric oxide generating cGMP production. (B) Low L-arginine causes nitric oxide synthase uncoupling and produces
303 Chapter 7 - Results
- - superoxide ions (O2 ), which will generate the production of peroxynitrite (ONOO ). Peroxynitrite causes oxidative/nitrosative stress with production of nitrotyrosine and - nitrite (NO2 ) after detoxification of peroxynitrite by reduced glutathione (GSH). Oxidative/nitrosative stress will impair the respiratory chain, alter the energy production of the cell and lead to cell death. Neurons are particularly vulnerable as they cannot increase their production of glutathione adequately, and rely on an astrocytic supply. Red arrows represent modified measured parameters in AslNeo/Neo compared to wild-type mice. cGMP: cyclic guanosine monophosphate, GSH: reduced - glutathione, GSSG: oxidised glutathione, NO: nitric oxide, NO2 : superoxide ion, - ONO2 : peroxynitrite, NOS: nitric oxide synthase.
The reality might be more complex and depends on cell type (498, 499), brain region and microenvironment (500). Precise mapping of the different changes in nitric oxide metabolism in the brain of AslNeo/Neo mouse is likely to be a difficult task. In this study, differences in nitrite/nitrate levels between wild-type and untreated AslNeo/Neo mice were highlighted in some regions (cortex, midbrain) but not in the cerebellum (See Section 5.3.3). These differences will apply to different cell types as well. Peroxynitrite has been shown having physiological properties in the presence of reduced glutathione and using cGMP as a signalling molecule (501), being involved in the relaxation of vascular smooth muscle (502, 503) and inhibition of platelet aggregation (504). In this work, no evidence of neuroinflammation was observed in astrocytes and microglial cells, as assessed by immunostaining for GFAP and CD68, respectively (See Section 5.3.3). Thus it seems that neurons are the main cell type affected by argininosuccinate lyase deficiency and that these are involved in the pathophysiology, with overexpression of both neuronal and inducible nitric oxide synthases and showing nitrosative/oxidative stress with nitrotyrosine staining. Neurons are more vulnerable to oxidative stress compared to astrocytes in vitro, as they cannot overexpress gamma glutamylcysteine synthetase, an enzyme required to replenish the intracellular reduced glutathione pool (GSH). Therefore they rely on the paracrine glutathione supply from astrocytes when exposed to reactive oxygen species (498). This might explain the apparent massive neuronal staining observed for nitrotyrosine and the efficacy of neuron-targeted gene therapy. This peroxynitrite/nitrotyrosine pathway has been previously involved in various
304 Chapter 7 - Results neurodegenerative diseases: Parkinson’s disease (505), Alzheimer’s disease (506, 507) and amyotrophic lateral sclerosis (508, 509).
This dual or “ambiguous” role of nitric oxide has been highlighted in various studies regarding the physiology of the central nervous system (510, 511). The regulation of the shift from cGMP-mediated neuroprotection to oxidative stress-related neurotoxicity and back is a subtle balance between two opposing roles, which results in cell survival or damage/death. This might even be a useful target of pharmacological modulation (493). This dual role is illustrated in the pathophysiology of neurodegenerative diseases such as Alzheimer’s disease. Nitrosative stress, caused by increased expression of inducible nitric oxide synthase in astrocytes and microglia, facilitates neuroinflammation against the β-amyloid protein and induces the formation of tau proteins. This is balanced by nitric oxide-mediated neuroprotection by increasing neuronal excitability in modifying potassium channels activity and promoting synaptic plasticity (512).
As the nitrosative/oxidative stress in cortical neurons plays a role independent of hyperammonaemia in the brain disease of AslNeo/Neo mice, the preservation of long- term neurological outcome needs to target both neuropathophysiological mechanisms i.e. hyperammonaemia and impaired cerebral nitric oxide metabolism. This suggests that a therapy targeting only the urea cycle in the liver will not fully control the brain disease and is unlikely to prevent long-term neurological sequelae. This is what is observed in patients treated with conventional therapy, which currently aims to control ammonaemia and hypoargininaemia only (See Section 3.7.3).
7.9 Conclusion
The vector designed in this study has demonstrated its efficacy in correcting the urea cycle and restoring long-term ureagenesis. This was observed in all treated animals with normalisation of ammonia levels, although with partial correction in neonatally- injected AslNeo/Neo mice and some variability in the phenotypic correction over time in the adult-injected mice.
305 Chapter 7 - Results
Long-term correction of the neuronal disease associated with nitrosative/oxidative stress was achieved in neonatally-injected mice after systemic neonatal vector delivery but not in adult-injected mice. This has confirmed that the neuronal disease is independent from hyperammonaemia.
This provides a better understanding of the pathophysiology and the neurological phenotype observed in patients affected by argininosuccinic aciduria and confirms the neurotoxic role of nitric oxide in the neuropathology of the disease as hypothesised in Chapter 3 (See Section 3.7.2). This illustrates the importance of revising current therapeutic guidelines, which currently focus only on the urea cycle defect and do not take into account other pathophysiological mechanisms.
306 Chapter 8 - Conclusion
8. SUMMARY AND FUTURE WORK
As the second most common urea cycle defect, argininosuccinic aciduria, the inherited metabolic disease caused by argininosuccinate lyase deficiency, has been studied for nearly half a century now. The paradox of a high prevalence of neurological symptoms in parallel with a low rate of hyperammonaemic crises has been questioned. Various putative pathophysiological explanations (See Section 1.2.2) have been proposed. Of these, the deleterious effect of argininosuccinate lyase deficiency on nitric oxide production has been proven through studies of an argininosuccinate lyase deficient mouse model, the AslNeo/Neo mouse (32) and the clinical improvement observed in a patient after nitric oxide supplementation (85). However the role of hyperammonaemia in the neurometabolic phenotype of the disease had remained still largely controversial and unknown.
8.1 Expansion of the clinical and genetic spectrum of argininosuccinic aciduria
As discussed in Section 1.2.3, argininosuccinic aciduria is a multi-organ disease. The current therapeutic guidelines recommend treatment aimed to control hyperammonaemia and hypoargininaemia. A combination of protein-restricted diet, ammonia lowering drugs and arginine supplementation was suggested to provide a better neurological outcome when started in the first months of life after newborn screening (See Section 1.2.8) (15, 17, 95).
Chapter 3 further delineates the natural history of this disorder, studying a large cohort of 56 British patients with argininosuccinic aciduria and documents the long- term impact of early conventional treatment. This has expanded the neurological and systemic spectrum of the disease, detailing the main neurological symptoms: neurocognitive impairment, epilepsy, myopathy-like symptoms and ataxia.
307 Chapter 8 - Conclusion
Neuroimaging has highlighted the high rate of focal infarcts, abnormal white matter signal in T2 sequence and abnormal creatine and guanidinoacetate in white matter. Nodular heterotopia, a previously unreported feature, has also been documented. Overall, the long-term neurological outcome was not significantly different between patients with early- or late-onset argininosuccinic aciduria and patients screened neonatally and those treated early (See Section 3.3).
A broader systemic spectrum was described with unreported gut and kidney symptoms. Unlike the neurological aspects of the disease the natural history of the systemic features is more clearly defined with regard to age of onset of symptoms, which displayed the progressivity of the phenotype (See Section 3.4). Biomarkers like plasma ammonia, argininosuccinic acid, and alanine aminotransferase presented with higher levels in early- rather than in late-onset patients (See Section 3.5). hASL sequencing was performed for 6 patients and data were available for 19 patients. Twenty mutations (8 novel) were found, which allowed the assessment of a phenotype-genotype correlation. This was supported for some of the mutations by the structural or functional analysis reported in the literature (See Section 3.6).
The key finding of this study was the demonstration that the neurological outcome was independent from the severity of hyperammonaemia irrespective of the age at treatment onset (See Section 3.7.1).
8.2 Oxidative/nitrosative stress not mediated by hyperammonaemia causes a neuronal disease
This study provides new insight into the neuropathophysiology of the brain disease in argininosuccinic aciduria. As described in Chapter 5, the AslNeo/Neo mouse shows evidence of cerebral oxidative stress and nitrosative stress in cortical neurons. Increased levels of all isoforms of nitric oxide synthase are observed in neurons for neuronal and inducible nitric oxide synthases, and endothelial cells for endothelial nitric oxide synthase (See Section 5.3.2). This is consistent with high levels of nitric oxide metabolites promoting oxidative stress and generation of reactive oxygen 308 Chapter 8 - Conclusion species such as peroxynitrite (See Section 5.3.3). Peroxynitrite nitrates tyrosine residues, forming nitrotyrosine, a hallmark of nitrosative stress (See Section 5.3.2).
As hyperammonaemia pathophysiology causes cerebral oxidative stress (See Section 1.2.2), the brains of mice, in which ammonaemia had normalised due to treatment by AAV-mediated gene therapy, were analysed (Chapter 7). This revealed persistence of oxidative/nitrosative stress in neurons in mice with correction of hyperammonaemia only. However this neuronal disease was dramatically improved after normalisation of hyperammonaemia and significant improvement of neuronal argininosuccinate lyase activity (See Section 7.6). Therefore this neuropathological feature was proven not being mediated by hyperammonaemia.
Based on further experiments presented in Section 7.7, a balance between the physiological and neuroprotective nitric oxide/cGMP pathway and the nitric oxide- mediated neurotoxicity caused by oxidative stress was proposed to account for the neuropathology of the disease. This disease-causing mechanism is common to various neurodegenerative disorders (See Section 7.8.3).
8.3 ASL gene transfer, a new therapy for argininosuccinic aciduria
8.3.1 Proof of concept of AAV-mediated gene therapy in argininosuccinic aciduria
Acknowledging the need for alternative therapies in argininosuccinic aciduria (See Section 3.7.3), the recent achievements of gene therapy in liver inherited metabolic diseases (See Section 1.3.3) especially for urea cycle defects (See Section 1.3.7) supported the development of a similar approach.
As illustrated in Chapter 6, an AAV8-derived vector construct was designed to target hepatocytes and cerebral neurons and deliver a transgene cassette containing the murine Asl gene. Neonatal injections of this vector validated its ability to target the
309 Chapter 8 - Conclusion desired cell types efficiently until adulthood. A mass spectrometry-based method presented in Chapter 4 was developed to accurately measure efficacy endpoints for this gene therapy experiment, i.e. amino acid levels in dried blood spots and argininosuccinate lyase activity in liver and brain samples. The AAV vector was injected systemically in either adult or neonatal AslNeo/Neo mice and followed over 12 and 9 months, respectively. The results presented in Chapter 7 showed successful correction of the urea cycle and restoration of ureagenesis, although in some of the adult-treated mice only partial correction was observed, and minimal correction in neonatally-treated animals (See Section 7.4). Adult-treated animals had a mild but significant correction of the citrulline-nitric oxide cycle in the liver as well (See Section 7.5). In the brain of neonatally-treated AslNeo/Neo mice, a dramatic reduction of the neuronal disease was observed, but this was not observed in adult-treated animals (See Section 7.6). The improvement of behavioural testing and protection against cell death in the brain were observed in both adult- and neonatally-injected groups with the latter being more marked in the brain of neonatally- compared to adult-injected mutants (See Section 7.7). These data suggest that these beneficial effects are mediated by correction of both pathways, control of hyperammonaemia and in situ correction of nitric oxide metabolism.
This confirmed the proof-of-concept of AAV gene therapy in argininosuccinic aciduria and highlights the necessity of acting on both the urea cycle and the citrulline-nitric oxide cycle, in the liver and the cortical neurons, respectively.
8.3.2 Considerations for clinical translation
This work endorses the need for an alternative nitric oxide supply to the brain of argininosuccinate lyase deficient patients to preserve their neurocognitive and neurological functions. Any future drug development and new therapeutic guidelines will need to consider the effect on both the urea and citrulline-nitric oxide cycle in order to appropriately tackle all of the symptoms of this disease. Liver transplantation (513) or gene therapy restricted to the liver will only cure the urea cycle and will not act on the disturbed nitric oxide metabolism in extra-hepatic tissues (85).
310 Chapter 8 - Conclusion
Appropriately, clinical trials providing dietary nitric oxide therapy have recently been initiated (Clinical trials numbers NCT02252770 and NCT03064048 from clinicaltrials.gov accessed on 06/04/2017).
The promising results presented in Chapter 7 raise the possibility of performing either a single or two injections: i) an early injection to target the urea cycle in the liver and cerebral neurons, ii) if necessary, a second later injection targeting the liver for long-term correction of the urea cycle. When the first injection is performed in the first weeks of life in mice, the immaturity of the immune system might not generate a humoral response (275). It is unlikely however that this can be extrapolated to humans. Overcoming the humoral immune response generated by the first AAV injection will need to be considered for the second injection, by for example using an alternative serotype and testing for no cross-reactivity (197).
This work has some limitations. The vector encoded the murine Asl gene and studies with the human ASL gene are necessary. Progress on capsid re-engineering is expanding rapidly and testing alternative capsids able to better transduce liver (514) and brain (515, 516) might be particularly interesting. Experiments were performed in small cohorts of mice and data will need to be reproduced in larger cohorts of animals to obtain more homogeneous results. Only one dose was assessed in each cohort and further work on minimal effective dosing and dose–escalation will be required. Additionally, the NO-cGMP pathway was only assessed in brain homogenates and not in a specific brain region or cell type unlike the immunostaining studies which specifically assessed the cortex. This does not therefore reflect the complexity of nitric oxide metabolism, which includes cell-type particularities (498, 499) and microenvironmental interactions (500). Further experiments in isolated neurons and glial cells from treated and untreated AslNeo/Neo mice will be of paramount importance to better delineate the pathophysiology associated with this oxidative/nitrosative stress. It would also be of interest to modify the vector construct and test an astrocyte-specific promoter (like GFAP promoter (517) to study the impact of an astrocyte-specific correction of argininosuccinate lyase activity.
The neurological signs developed by AslNeo/Neo mice are limited and only 2 behavioural tests were proven as discriminant between wild-type and AslNeo/Neo mice. As observed in Chapter 7, hyperammonaemia and disturbed cerebral nitric oxide
311 Chapter 8 - Conclusion metabolism both display an impact in functional studies assessing behaviour and cell death. It would be of interest to develop other tools, including more complex behavioural tests especially challenging the memory, and biomarkers to better assess the benefit of cerebral gene therapy.
These results are limited with regards to efficacy data. Further collection of data documenting safety will be required before submission of any clinical trial authorisation. AAV8 and the EFS promoter have separately reached the clinics and published data are available to defend a safe profile of this gene therapy product. Although AAV8 and EFS have been tested in combination in small animal models (438, 478, 479, 518, 519), this has not been performed in humans yet. As the risk of insertional mutagenesis still remains a controversial question for wild-type AAV (266- 268) and have been observed after systemic neonatal injection of AAV vectors (520), the possibility of genotoxicity that may occur using AAV vectors needs to be carefully investigated with integration studies in animal models.
8.3.3 Application for neurohepatotropic diseases
Several other inherited metabolic diseases, which encompass both brain and liver symptoms, such as mitochondrial diseases caused by nuclear genetic defects (POLG, MPV17, DGUOK), some lysosomal storage disorders (Neuronopathic Gaucher disease, mucopolysaccharidosis type I, II, VII), metal transporter deficiencies like Menkes disease or manganese transporter deficiencies, some congenital disorders of glycosylation, might benefit from a similar sequential approach using two successive systemic deliveries of a transgene cassette conveyed by AAV vectors to optimally correct brain and liver phenotypes.
8.4 Overall conclusion
In conclusion, this work has demonstrated the need for alternative therapies for patients affected by argininosuccinic aciduria, as the current therapeutic guidelines do not prevent a debilitating neurological disease. This is supported by new insights 312 Chapter 8 - Conclusion into the neuropathophysiology of this disorder, highlighting a neuronal disease, which occurs as a result of oxidative/nitrosative stress and not mediated by hyperammonaemia. A proposed therapeutic approach with AAV-mediated gene therapy has shown proof-of-concept of long-term improvement of both brain and liver phenotypes with a single vector construct after systemic delivery. A sequential approach would allow a long-term correction of the urea cycle and the neuronal oxidative/nitrosative stress and ultimately prevent the neurological disease. This work paves the way for clinical translation and more broadly provides new hope for hepatocerebral inherited diseases.
313 Chapter 9 - References
9. REFERENCES
1. Wraith JE. Ornithine carbamoyltransferase deficiency. Arch Dis Child. 2001;84(1):84-8.
2. Gebhardt R, Matz-Soja M. Liver zonation: Novel aspects of its regulation and its impact on homeostasis. World journal of gastroenterology : Wotld J Gastroenterol. 2014;20(26):8491-504.
3. Nassogne MC, Heron B, Touati G, Rabier D, Saudubray JM. Urea cycle defects: management and outcome. J Inherit Metab Dis. 2005;28(3):407-14.
4. Caldovic L, Abdikarim I, Narain S, Tuchman M, Morizono H. Genotype- Phenotype Correlations in Ornithine Transcarbamylase Deficiency: A Mutation Update. J Genet Genomics. 2015;42(5):181-94.
5. Posset R, Garcia-Cazorla A, Valayannopoulos V, Teles EL, Dionisi-Vici C, Brassier A, et al. Age at disease onset and peak ammonium level rather than interventional variables predict the neurological outcome in urea cycle disorders. J Inherit Metab Dis. 2016;39(5):661-72.
6. Bachmann C. Outcome and survival of 88 patients with urea cycle disorders: a retrospective evaluation. Eur J Pediatr. 2003;162(6):410-6.
7. Tuchman M, Lee B, Lichter-Konecki U, Summar ML, Yudkoff M, Cederbaum SD, et al. Cross-sectional multicenter study of patients with urea cycle disorders in the United States. Mol Genet Metab. 2008;94(4):397-402.
8. Waisbren SE, Gropman AL, Members of the Urea Cycle Disorders C, Batshaw ML. Improving long term outcomes in urea cycle disorders-report from the Urea Cycle Disorders Consortium. J Inherit Metab Dis. 2016.
9. Yamanouchi H, Yokoo H, Yuhara Y, Maruyama K, Sasaki A, Hirato J, et al. An autopsy case of ornithine transcarbamylase deficiency. Brain Dev. 2002;24(2):91-4.
314 Chapter 9 - References
10. Takanashi J, Barkovich AJ, Cheng SF, Weisiger K, Zlatunich CO, Mudge C, et al. Brain MR imaging in neonatal hyperammonemic encephalopathy resulting from proximal urea cycle disorders. AJNR Am J Neuroradiol. 2003;24(6):1184-7.
11. Gropman A. Brain imaging in urea cycle disorders. Mol Genet Metab. 2010;100 Suppl 1:S20-30.
12. Haberle J, Boddaert N, Burlina A, Chakrapani A, Dixon M, Huemer M, et al. Suggested guidelines for the diagnosis and management of urea cycle disorders. Orphanet J Rare Dis. 2012;7:32.
13. Gropman AL, Batshaw ML. Cognitive outcome in urea cycle disorders. Mol Genet Metab. 2004;81 Suppl 1:S58-62.
14. Summar ML, Koelker S, Freedenberg D, Le Mons C, Haberle J, Lee HS, et al. The incidence of urea cycle disorders. Mol Genet Metab. 2013;110(1-2):179-80.
15. Ficicioglu C, Mandell R, Shih VE. Argininosuccinate lyase deficiency: longterm outcome of 13 patients detected by newborn screening. Mol Genet Metab. 2009;98(3):273-7.
16. Brusilow SW, Maestri NE. Urea cycle disorders: diagnosis, pathophysiology, and therapy. Adv Pediatr. 1996;43:127-70.
17. Mercimek-Mahmutoglu S, Moeslinger D, Haberle J, Engel K, Herle M, Strobl MW, et al. Long-term outcome of patients with argininosuccinate lyase deficiency diagnosed by newborn screening in Austria. Mol Genet Metab. 2010;100(1):24-8.
18. Nagamani SC, Erez A, Lee B. Argininosuccinate lyase deficiency. Genet Med. 2012;14(5):501-7.
19. Ratner S. Enzymes of arginine and urea synthesis. Adv Enzymol Relat Areas Mol Biol. 1973;39:1-90.
20. Turner MA, Simpson A, McInnes RR, Howell PL. Human argininosuccinate lyase: a structural basis for intragenic complementation. Proc Natl Acad Sci U S A. 1997;94(17):9063-8.
315 Chapter 9 - References
21. Balmer C, Pandey AV, Rufenacht V, Nuoffer JM, Fang P, Wong LJ, et al. Mutations and polymorphisms in the human argininosuccinate lyase (ASL) gene. Hum Mutat. 2014;35(1):27-35.
22. Simard L, O'Brien WE, McInnes RR. Argininosuccinate lyase deficiency: evidence for heterogeneous structural gene mutations by immunoblotting. Am J Hum Genet. 1986;39(1):38-51.
23. Sampaleanu LM, Yu B, Howell PL. Mutational analysis of duck delta 2 crystallin and the structure of an inactive mutant with bound substrate provide insight into the enzymatic mechanism of argininosuccinate lyase. J Biol Chem. 2002;277(6):4166-75.
24. Hu L, Pandey AV, Eggimann S, Rufenacht V, Moslinger D, Nuoffer JM, et al. Understanding the role of argininosuccinate lyase transcript variants in the clinical and biochemical variability of the urea cycle disorder argininosuccinic aciduria. J Biol Chem. 2013;288(48):34599-611.
25. Santacruz L, Jacobs DO. Structural correlates of the creatine transporter function regulation: the undiscovered country. Amino Acids. 2016;48(8):2049-55.
26. Pegg AE. Mammalian polyamine metabolism and function. IUBMB Life. 2009;61(9):880-94.
27. Lundberg JO, Gladwin MT, Ahluwalia A, Benjamin N, Bryan NS, Butler A, et al. Nitrate and nitrite in biology, nutrition and therapeutics. Nat Chem Biol. 2009;5(12):865-9.
28. Rabier D, Narcy C, Bardet J, Parvy P, Saudubray JM, Kamoun P. Arginine remains an essential amino acid after liver transplantation in urea cycle enzyme deficiencies. J Inherit Metab Dis. 1991;14(3):277-80.
29. Kurz S, Harrison DG. Insulin and the arginine paradox. J Clin Invest. 1997;99(3):369-70.
30. Vukosavljevic N, Jaron D, Barbee KA, Buerk DG. Quantifying the L-arginine paradox in vivo. Microvasc Res. 2006;71(1):48-54.
316 Chapter 9 - References
31. Li C, Huang W, Harris MB, Goolsby JM, Venema RC. Interaction of the endothelial nitric oxide synthase with the CAT-1 arginine transporter enhances NO release by a mechanism not involving arginine transport. Biochem J. 2005;386(Pt 3):567-74.
32. Erez A, Nagamani SC, Shchelochkov OA, Premkumar MH, Campeau PM, Chen Y, et al. Requirement of argininosuccinate lyase for systemic nitric oxide production. Nat Med. 2011;17(12):1619-26.
33. Jeffery CJ. An introduction to protein moonlighting. Biochem Soc Trans. 2014;42(6):1679-83.
34. O'Brien WE, McInnes R, Kalumuck K, Adcock M. Cloning and sequence analysis of cDNA for human argininosuccinate lyase. Proc Natl Acad Sci U S A. 1986;83(19):7211-5.
35. Trevisson E, Salviati L, Baldoin MC, Toldo I, Casarin A, Sacconi S, et al. Argininosuccinate lyase deficiency: mutational spectrum in Italian patients and identification of a novel ASL pseudogene. Hum Mutat. 2007;28(7):694-702.
36. Crosas E, Sumoy L, Gonzalez E, Diaz M, Bartolome S, Farres J, et al. The yeast zeta-crystallin/NADPH:quinone oxidoreductase (Zta1p) is under nutritional control by the target of rapamycin pathway and is involved in the regulation of argininosuccinate lyase mRNA half-life. FEBS J. 2015;282(10):1953-64.
37. Tanaka A, Ito H, Tanaka R, Tanaka NK, Yoshida K, Okada K. Chlorophyll a oxygenase (CAO) is involved in chlorophyll b formation from chlorophyll a. Proc Natl Acad Sci U S A. 1998;95(21):12719-23.
38. Xia J, Yamaji N, Ma JF. An appropriate concentration of arginine is required for normal root growth in rice. Plant Signal Behav. 2014;9(3):e28717.
39. Braissant O, McLin VA, Cudalbu C. Ammonia toxicity to the brain. J Inherit Metab Dis. 2013;36(4):595-612.
40. Guertin SR, Levinsohn MW, Dahms BB. Small-droplet steatosis and intracranial hypertension in argininosuccinic lyase deficiency. J Pediatr. 1983;102(5):736-40.
317 Chapter 9 - References
41. Lewis PD, Miller AL. Argininosuccinic aciduria. Case report with neuropathological findings. Brain. 1970;93(2):413-22.
42. Crome L, France NE. The pathological findings in a case of argininosuccinic aciduria. J Ment Defic Res. 1971;15 Pt 4(0):266-70.
43. Bachmann C. Mechanisms of hyperammonemia. Clin Chem Lab Med. 2002;40(7):653-62.
44. Bergeron M, Swain MS, Reader TA, Grondin L, Butterworth RF. Effect of ammonia on brain serotonin metabolism in relation to function in the portacaval shunted rat. J Neurochem. 1990;55(1):222-9.
45. Monfort P, Munoz MD, Felipo V. Molecular mechanisms of the alterations in NMDA receptor-dependent long-term potentiation in hyperammonemia. Metab Brain Dis. 2005;20(4):265-74.
46. Rangroo Thrane V, Thrane AS, Wang F, Cotrina ML, Smith NA, Chen M, et al. Ammonia triggers neuronal disinhibition and seizures by impairing astrocyte potassium buffering. Nat Med. 2013;19(12):1643-8.
47. Gropman AL, Summar M, Leonard JV. Neurological implications of urea cycle disorders. J Inherit Metab Dis. 2007;30(6):865-79.
48. Garcia MV, Lopez-Mediavilla C, Juanes de la Pena MC, Medina JM. Tolerance of neonatal rat brain to acute hyperammonemia. Brain Res. 2003;973(1):31-8.
49. Gladwin MT, Tejero J. Nitrite-NO bailout for a NOS complex too big to fail. Nat Med. 2011;17(12):1556-7.
50. Natesan V, Mani R, Arumugam R. Clinical aspects of urea cycle dysfunction and altered brain energy metabolism on modulation of glutamate receptors and transporters in acute and chronic hyperammonemia. Biomed Pharmacother. 2016;81:192-202.
51. Boenzi S, Pastore A, Martinelli D, Goffredo BM, Boiani A, Rizzo C, et al. Creatine metabolism in urea cycle defects. J Inherit Metab Dis. 2012;35(4):647-53.
318 Chapter 9 - References
52. van Spronsen FJ, Reijngoud DJ, Verhoeven NM, Soorani-Lunsing RJ, Jakobs C, Sijens PE. High cerebral guanidinoacetate and variable creatine concentrations in argininosuccinate synthetase and lyase deficiency: implications for treatment? Mol Genet Metab. 2006;89(3):274-6.
53. Sijens PE, Reijngoud DJ, Soorani-Lunsing RJ, Oudkerk M, van Spronsen FJ. Cerebral 1H MR spectroscopy showing elevation of brain guanidinoacetate in argininosuccinate lyase deficiency. Mol Genet Metab. 2006;88(1):100-2.
54. Barilli A, Rotoli BM, Visigalli R, Bussolati O, Gazzola GC, Gatti R, et al. Impaired phagocytosis in macrophages from patients affected by lysinuric protein intolerance. Mol Genet Metab. 2012;105(4):585-9.
55. Wishart DS, Tzur D, Knox C ea. HMDB: the Human Metabolome Database Nucleic Acids Res 2007(35 (Database issue)):D521-6.
56. Schulze A, Ebinger F, Rating D, Mayatepek E. Improving treatment of guanidinoacetate methyltransferase deficiency: reduction of guanidinoacetic acid in body fluids by arginine restriction and ornithine supplementation. Mol Genet Metab. 2001;74(4):413-9.
57. Allan JD, Cusworth DC, Dent CE, Wilson VK. A disease, probably hereditary characterised by severe mental deficiency and a constant gross abnormality of aminoacid metabolism. Lancet. 1958;1(7013):182-7.
58. Westall RG. Argininosuccinic aciduria: identification and reactions of the abnormal metabolite in a newly described form of mental disease, with some preliminary metabolic studies. Biochem J. 1960;77:135-44.
59. Ah Mew N, Krivitzky L, McCarter R, Batshaw M, Tuchman M, Urea Cycle Disorders Consortium of the Rare Diseases Clinical Research N. Clinical outcomes of neonatal onset proximal versus distal urea cycle disorders do not differ. J Pediatr. 2013;162(2):324-9 e1.
60. Lagas PA, Ruokonen A. Late onset argininosuccinic aciduria in a paranoid retardate. Biol Psychiatry. 1991;30(12):1229-32.
319 Chapter 9 - References
61. Kleijer WJ, Garritsen VH, Linnebank M, Mooyer P, Huijmans JG, Mustonen A, et al. Clinical, enzymatic, and molecular genetic characterization of a biochemical variant type of argininosuccinic aciduria: prenatal and postnatal diagnosis in five unrelated families. J Inherit Metab Dis. 2002;25(5):399-410.
62. Gerrits GP, Gabreels FJ, Monnens LA, De Abreu RA, van Raaij-Selten B, Niezen-Koning KE, et al. Argininosuccinic aciduria: clinical and biochemical findings in three children with the late onset form, with special emphasis on cerebrospinal fluid findings of amino acids and pyrimidines. Neuropediatrics. 1993;24(1):15-8.
63. Grioni D, Furlan F, Corbetta C, Barboni C, Lastrico A, Marzocchi GM, et al. Epilepsy and argininosuccinic aciduria. Neuropediatrics. 2011;42(3):97-103.
64. Hambraeus L, Hardell LI, Westphal O, Lorentsson R, Hjorth G. Argininosuccinic aciduria. Report of three cases and the effect of high and reduced protein intake on the clinical state. Acta Paediatr Scand. 1974;63(4):525-36.
65. Verma NP, Hart ZH, Kooi KA. Electroencephalographic findings in urea-cycle disorders. Electroencephalogr Clin Neurophysiol. 1984;57(2):105-12.
66. Odent S, Roussey M, Journel H, Betremieux P, David V, Le Marec B. [Argininosuccinic aciduria. A new case revealed by psychiatric disorders]. J Genet Hum. 1989;37(1):39-42.
67. von Wendt L, Simila S, Ruokonen A, Puukka M. Argininosuccinic aciduria in a Finnish woman presenting with psychosis and mental retardation. Ann Clin Res. 1982;14(3):145-7.
68. Baumgartner R, Scheidegger S, Stalder G, Hottinger A. [Argininosuccinic aciduria disease of the newborn with lethal course]. Helv Paediatr Acta. 1968;23(1):77-106.
69. Capistrano-Estrada S, Marsden DL, Nyhan WL, Newbury RO, Krous HF, Tuchman M. Histopathological findings in a male with late-onset ornithine transcarbamylase deficiency. Pediatr Pathol. 1994;14(2):235-43.
320 Chapter 9 - References
70. Michalak A, Butterworth RF. Ornithine transcarbamylase deficiency: pathogenesis of the cerebral disorder and new prospects for therapy. Metab Brain Dis. 1997;12(3):171-82.
71. Roze E, Azuar C, Menuel C, Haberle J, Guillevin R. Usefulness of magnetic resonance spectroscopy in urea cycle disorders. Pediatr Neurol. 2007;37(3):222-5.
72. Arias A, Garcia-Villoria J, Ribes A. Guanidinoacetate and creatine/creatinine levels in controls and patients with urea cycle defects. Mol Genet Metab. 2004;82(3):220-3.
73. Parsons HG, Scott RB, Pinto A, Carter RJ, Snyder FF. Argininosuccinic aciduria: long-term treatment with arginine. J Inherit Metab Dis. 1987;10(2):152-61.
74. Mori T, Nagai K, Mori M, Nagao M, Imamura M, Iijima M, et al. Progressive liver fibrosis in late-onset argininosuccinate lyase deficiency. Pediatr Dev Pathol. 2002;5(6):597-601.
75. Nagamani SC, Shchelochkov OA, Mullins MA, Carter S, Lanpher BC, Sun Q, et al. A randomized controlled trial to evaluate the effects of high-dose versus low- dose of arginine therapy on hepatic function tests in argininosuccinic aciduria. Mol Genet Metab. 2012;107(3):315-21.
76. Marble M, McGoey RR, Mannick E, Keats B, Ng SS, Deputy S, et al. Living related liver transplant in a patient with argininosuccinic aciduria and cirrhosis: metabolic follow-up. J Pediatr Gastroenterol Nutr. 2008;46(4):453-6.
77. Zimmermann A, Bachmann C, Baumgartner R. Severe liver fibrosis in argininosuccinic aciduria. Arch Pathol Lab Med. 1986;110(2):136-40.
78. Jameson E BC, Jennions L, Vijay S, Morris AA. The Manchester experience with ASA patients. J Inherit Metab Dis. 2012;35((Suppl 1)):S7.
79. Yaplito-Lee J, Chow CW, Boneh A. Histopathological findings in livers of patients with urea cycle disorders. Mol Genet Metab. 2013;108(3):161-5.
80. Lee B, Goss J. Long-term correction of urea cycle disorders. J Pediatr. 2001;138(1 Suppl):S62-71.
321 Chapter 9 - References
81. Miles L, Heubi JE, Bove KE. Hepatocyte glycogen accumulation in patients undergoing dietary management of urea cycle defects mimics storage disease. J Pediatr Gastroenterol Nutr. 2005;40(4):471-6.
82. Yang J, Chi Y, Burkhardt BR, Guan Y, Wolf BA. Leucine metabolism in regulation of insulin secretion from pancreatic beta cells. Nutr Rev. 2010;68(5):270-9.
83. Farzaneh-Far R, Moore K. Nitric oxide and the liver. Liver. 2001;21(3):161-74.
84. Brunetti-Pierri N, Erez A, Shchelochkov O, Craigen W, Lee B. Systemic hypertension in two patients with ASL deficiency: a result of nitric oxide deficiency? Mol Genet Metab. 2009;98(1-2):195-7.
85. Nagamani SC, Campeau PM, Shchelochkov OA, Premkumar MH, Guse K, Brunetti-Pierri N, et al. Nitric-oxide supplementation for treatment of long-term complications in argininosuccinic aciduria. Am J Hum Genet. 2012;90(5):836-46.
86. Kolker S, Cazorla AG, Valayannopoulos V, Lund AM, Burlina AB, Sykut- Cegielska J, et al. The phenotypic spectrum of organic acidurias and urea cycle disorders. Part 1: the initial presentation. J Inherit Metab Dis. 2015.
87. Pearson DL, Dawling S, Walsh WF, Haines JL, Christman BW, Bazyk A, et al. Neonatal pulmonary hypertension--urea-cycle intermediates, nitric oxide production, and carbamoyl-phosphate synthetase function. N Engl J Med. 2001;344(24):1832-8.
88. Reid L, Perreault E, Lafrance G, Clarke JT. Experience with the treatment of argininosuccinic aciduria during pregnancy. J Inherit Metab Dis. 2009;32 Suppl 1:S191-5.
89. Coulter DL, Beals TF, Allen RJ. Neurotrichosis: hair-shaft abnormalities associated with neurological diseases. Dev Med Child Neurol. 1982;24(5):634-44.
90. Chen BC, Ngu LH, Zabedah MY. Argininosuccinic aciduria: clinical and biochemical phenotype findings in Malaysian children. Malays J Pathol. 2010;32(2):87-95.
91. Schutgens RBH, Beemer FA, Tegelaers WHH. Mild variant of argininosuccinic aciduria. J Inherit Metab Dis. 1979;2:13-4.
322 Chapter 9 - References
92. Shelley WB, Rawnsley HM. Aminogenic alopecia loss of hair associated with argininosuccinic aciduria. Lancet. 1965;286(7426):1328-9.
93. Gardeitchik T, Humphrey M, Nation J, Boneh A. Early clinical manifestations and eating patterns in patients with urea cycle disorders. J Pediatr. 2012;161(2):328- 32.
94. Ibarra-Gonzalez I, Fernandez-Lainez C, Vela-Amieva M. Clinical and biochemical characteristics of patients with urea cycle disorders in a developing country. Clin Biochem. 2010;43(4-5):461-6.
95. Widhalm K, Koch S, Scheibenreiter S, Knoll E, Colombo JP, Bachmann C, et al. Long-term follow-up of 12 patients with the late-onset variant of argininosuccinic acid lyase deficiency: no impairment of intellectual and psychomotor development during therapy. Pediatrics. 1992;89(6 Pt 2):1182-4.
96. Gilchrist M, Winyard PG, Benjamin N. Dietary nitrate--good or bad? Nitric Oxide. 2010;22(2):104-9.
97. Ruegger CM, Lindner M, Ballhausen D, Baumgartner MR, Beblo S, Das A, et al. Cross-sectional observational study of 208 patients with non-classical urea cycle disorders. J Inherit Metab Dis. 2014;37(1):21-30.
98. Shih VE (1972) Urea cycle disorders and other congenital hypera;;onae;ic syndro;es. In: Stanbury JB WJ, Fredrickson DS, eds. The Metabolic Basis of Inherited Disease. New York: McGraw-Hill, 362-386.
99. Worthington S, Christodoulou J, Wilcken B, Peat B. Pregnancy and argininosuccinic aciduria. J Inherit Metab Dis. 1996;19(5):621-3.
100. Summar ML, Dobbelaere D, Brusilow S, Lee B. Diagnosis, symptoms, frequency and mortality of 260 patients with urea cycle disorders from a 21-year, multicentre study of acute hyperammonaemic episodes. Acta paediatrica. 2008;97(10):1420-5.
101. Kido J, Nakamura K, Mitsubuchi H, Ohura T, Takayanagi M, Matsuo M, et al. Long-term outcome and intervention of urea cycle disorders in Japan. J Inherit Metab Dis. 2012;35(5):777-85.
323 Chapter 9 - References
102. Brusilow SW, Batshaw ML. Arginine therapy of argininosuccinase deficiency. Lancet. 1979;1(8108):124-7.
103. De Biase I, Liu A, Yuzyuk T, Longo N, Pasquali M. Quantitative amino acid analysis by liquid chromatography-tandem mass spectrometry: implications for the diagnosis of argininosuccinic aciduria. Clin Chim Acta. 2015;442:73-4.
104. Maestri NE, McGowan KD, Brusilow SW. Plasma glutamine concentration: a guide in the management of urea cycle disorders.J Pediatr. 1992;121(2):259-61.
105. Stephenne X, Najimi M, Sibille C, Nassogne MC, Smets F, Sokal EM. Sustained engraftment and tissue enzyme activity after liver cell transplantation for argininosuccinate lyase deficiency. Gastroenterology. 2006;130(4):1317-23.
106. Zheng L, Mackenzie ED, Karim SA, Hedley A, Blyth K, Kalna G, et al. Reversed argininosuccinate lyase activity in fumarate hydratase-deficient cancer cells. Cancer Metab. 2013;1(1):12.
107. Kleijer WJ, Garritsen VH, van der Sterre ML, Berning C, Haberle J, Huijmans JG. Prenatal diagnosis of citrullinemia and argininosuccinic aciduria: evidence for a transmission ratio distortion in citrullinemia. Prenat Diagn. 2006;26(3):242-7.
108. Adam S, Almeida MF, Assoun M, Baruteau J, Bernabei SM, Bigot S, et al. Dietary management of urea cycle disorders: European practice. Mol Genet Metab. 2013;110(4):439-45.
109. Scaglia F. New insights in nutritional management and amino acid supplementation in urea cycle disorders. Mol Genet Metab. 2010;100 Suppl 1:S72-6.
110. Foschi FG, Morelli MC, Savini S, Dall'Aglio AC, Lanzi A, Cescon M, et al. Urea cycle disorders: a case report of a successful treatment with liver transplant and a literature review. World J Gastroenterol. 2015;21(13):4063-8.
111. Li M, Dick A, Montenovo M, Horslen S, Hansen R. Cost-effectiveness of liver transplantation in methylmalonic and propionic acidemias. Liver transplant. 2015.
324 Chapter 9 - References
112. WHO/FAO/UNU E, Consultation. Protein and amino acid requirements in human nutrition. WHO Technical Report Series 935, UNU/WHO, Geneva, Switzerland. 2007.
113. Vance AM, Mitchell A, Richard JS. AAV Biology, Infectivity and Therapeutic Use from Bench to Clinic. In: Science IO, editor. Gene therapy - Principles and Challenges2015. p. 120-43.
114. Grote V, Verduci E, Scaglioni S, Vecchi F, Contarini G, Giovannini M, et al. Breast milk composition and infant nutrient intakes during the first 12 months of life. Eur J Clin Nutr. 2016;70(2):250-6.
115. Syrad H, Llewellyn CH, van Jaarsveld CH, Johnson L, Jebb SA, Wardle J. Energy and nutrient intakes of young children in the UK: findings from the Gemini twin cohort. Br J Nutr. 2016;115(10):1843-50.
116. Rolland-Cachera MF, Bellisle F, Deheeger M. Nutritional status and food intake in adolescents living in Western Europe. Eur J Clin Nutr. 2000;54 Suppl 1:S41- 6.
117. Ucar SK, Ozbaran B, Altinok YA, Kose M, Canda E, Kagnici M, et al. One Year Experience of Pheburane((R)) (Sodium Phenylbutyrate) Treatment in a Patient with Argininosuccinate Lyase Deficiency. JIMD Rep. 2015;19:31-3.
118. Nagamani SC, Lee B, Erez A. Optimizing therapy for argininosuccinic aciduria. Mol Genet Metab. 2012;107(1-2):10-4.
119. Squires RH, Ng V, Romero R, Ekong U, Hardikar W, Emre S, et al. Evaluation of the pediatric patient for liver transplantation: 2014 practice guideline by the American Association for the Study of Liver Diseases, American Society of Transplantation and the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition. J Pediatr Gastroenterol Nutr. 2014;59(1):112-31.
120. Robberecht E, Maesen S, Jonckheere A, Van Biervliet S, Carton D. Successful liver transplantation for argininosuccinate lyase deficiency (ASLD). J Inherit Metab Dis. 2006;29(1):184-5.
325 Chapter 9 - References
121. Perito ER, Rhee S, Roberts JP, Rosenthal P. Pediatric liver transplantation for urea cycle disorders and organic acidemias: United Network for Organ Sharing data for 2002-2012. Liver transplant. 2014;20(1):89-99.
122. Vazquez J, Gamez M, Santamaria ML, Murcia J, Diaz MC, Camarena C, et al. Liver transplantation in small babies. J Pediatr Surg. 1993;28(8):1051-3.
123. Dunn SP, Weintraub W, Vinocur CD, Billmire DF, Falkenstein K. Is age less than 1 year a high-risk category for orthotopic liver transplantation? J Pediatr Surg. 1993;28(8):1048-50.
124. Puppi J, Tan N, Mitry RR, Hughes RD, Lehec S, Mieli-Vergani G, et al. Hepatocyte transplantation followed by auxiliary liver transplantation--a novel treatment for ornithine transcarbamylase deficiency. Am J Transplant. 2008;8(2):452- 7.
125. Meyburg J, Hoffmann GF. Liver, liver cell and stem cell transplantation for the treatment of urea cycle defects. Mol Genet Metab. 2010;100 Suppl 1:S77-83.
126. Stephenne X, Najimi M, Smets F, Reding R, de Ville de Goyet J, Sokal EM. Cryopreserved liver cell transplantation controls ornithine transcarbamylase deficient patient while awaiting liver transplantation. Am J Transplant. 2005;5(8):2058-61.
127. Srilatha B, Meng J, Adaikan PG. Hydrogen sulfide: in the aftermath of argininosuccinate lyase and nitric oxide deficiency. Genet Med. 2012;14(6):627; author reply -8.
128. Enns GM, Berry SA, Berry GT, Rhead WJ, Brusilow SW, Hamosh A. Survival after treatment with phenylacetate and benzoate for urea-cycle disorders. N Engl J Med. 2007;356(22):2282-92.
129. Saudubray JM, Touati G, Delonlay P, Jouvet P, Narcy C, Laurent J, et al. Liver transplantation in urea cycle disorders. Eur J Pediatr. 1999;158 Suppl 2:S55-9.
130. Keskinen P, Siitonen A, Salo M. Hereditary urea cycle diseases in Finland. Acta paediatrica. 2008;97(10):1412-9.
326 Chapter 9 - References
131. Martin-Hernandez E, Aldamiz-Echevarria L, Castejon-Ponce E, Pedron-Giner C, Couce ML, Serrano-Nieto J, et al. Urea cycle disorders in Spain: an observational, cross-sectional and multicentric study of 104 cases. Orphanet J Rare Dis. 2014;9:187.
132. Al-Sayed M, Alahmed S, Alsmadi O, Khalil H, Rashed MS, Imtiaz F, et al. Identification of a common novel mutation in Saudi patients with argininosuccinic aciduria. J Inherit Metab Dis. 2005;28(6):877-83.
133. Glick NR, Snodgrass PJ, Schafer IA. Neonatal argininosuccinic aciduria with normal brain and kidney but absent liver argininosuccinate lyase activity. Am J Hum Genet. 1976;28(1):22-30.
134. Perry TL, Wirtz ML, Kennaway NG, Hsia YE, Atienza FC, Uemura HS. Amino acid and enzyme studies of brain and other tissues in an infant with argininosuccinic aciduria. Clin Chim Acta. 1980;105(2):257-67.
135. Kobayashi K, Itakura Y, Saheki T, Nakano K, Sase M, Oyanagi K, et al. Absence of argininosuccinate lyase protein in the liver of two patients with argininosuccinic aciduria. Clin Chim Acta. 1986;159(1):59-67.
136. Howell PL, Turner MA, Christodoulou J, Walker DC, Craig HJ, Simard LR, et al. Intragenic complementation at the argininosuccinate lyase locus: reconstruction of the active site. J Inherit Metab Dis. 1998;21 Suppl 1:72-85.
137. Walker DC, Christodoulou J, Craig HJ, Simard LR, Ploder L, Howell PL, et al. Intragenic complementation at the human argininosuccinate lyase locus. Identification of the major complementing alleles. J Biol Chem. 1997;272(10):6777- 83.
138. Renouf S, Fairand A, Husson A. Developmental control of argininosuccinate lyase gene by methylation. Biol Neonate. 1998;73(3):190-7.
139. Linnebank M, Homberger A, Rapp B, Winter C, Marquardt T, Harms E, et al. Two novel mutations (E86A, R113W) in argininosuccinate lyase deficiency and evidence for highly variable splicing of the human argininosuccinate lyase gene. J Inherit Metab Dis. 2000;23(4):308-12.
327 Chapter 9 - References
140. Walker DC, McCloskey DA, Simard LR, McInnes RR. Molecular analysis of human argininosuccinate lyase: mutant characterization and alternative splicing of the coding region. Proc Natl Acad Sci U S A. 1990;87(24):9625-9.
141. Naylor EW. Newborn screening of urea cycle disorders. Pediatrics. 1981;68(3):453-7.
142. Burgard P, Rupp K, Lindner M, Haege G, Rigter T, Weinreich SS, et al. Newborn screening programmes in Europe; arguments and efforts regarding harmonization. Part 2. From screening laboratory results to treatment, follow-up and quality assurance. J Inherit Metab Dis. 2012;35(4):613-25.
143. Auray-Blais C, Cyr D, Drouin R. Quebec neonatal mass urinary screening programme: from micromolecules to macromolecules. J Inherit Metab Dis. 2007;30(4):515-21.
144. Wilcken B, Smith A, Brown DA. Urine screening for aminoacidopathies: is it beneficial? Results of a long-term follow-up of cases detected bny screening one millon babies.J Pediatr. 1980;97(3):492-7.
145. Margalith D, Crichton JU, Wong L, Davidson AG, Applegarth DA, Toone JR. Argininosuccinic aciduria. A developmental and biochemical case study. J Neurol Sci. 1983;60(2):217-33.
146. Reid Sutton V, Pan Y, Davis EC, Craigen WJ. A mouse model of argininosuccinic aciduria: biochemical characterization. Mol Genet Metab. 2003;78(1):11-6.
147. Clayton PT. Inborn errors presenting with liver dysfunction. Semin Neonatol. 2002;7(1):49-63.
148. Spada M, Riva S, Maggiore G, Cintorino D, Gridelli B. Pediatric liver transplantation. World J Gastroenterol. 2009;15(6):648-74.
149. Dhawan A, Mitry RR, Hughes RD. Hepatocyte transplantation for liver-based metabolic disorders. J Inherit Metab Dis. 2006;29(2-3):431-5.
328 Chapter 9 - References
150. Sokal EM. Liver transplantation for inborn errors of liver metabolism. J Inherit Metab Dis. 2006;29(2-3):426-30.
151. Wirth T, Parker N, Yla-Herttuala S. History of gene therapy. Gene. 2013;525(2):162-9.
152. Valdmanis PN, Kay MA. Future of rAAV Gene Therapy: Platform for RNAi, Gene Editing, and Beyond. Hum Gene Ther. 2017.
153. Blaese RM, Culver KW, Miller AD, Carter CS, Fleisher T, Clerici M, et al. T lymphocyte-directed gene therapy for ADA- SCID: initial trial results after 4 years. Science. 1995;270(5235):475-80.
154. Bordignon C, Notarangelo LD, Nobili N, Ferrari G, Casorati G, Panina P, et al. Gene therapy in peripheral blood lymphocytes and bone marrow for ADA- immunodeficient patients. Science. 1995;270(5235):470-5.
155. Puumalainen AM, Vapalahti M, Agrawal RS, Kossila M, Laukkanen J, Lehtolainen P, et al. Beta-galactosidase gene transfer to human malignant glioma in vivo using replication-deficient retroviruses and adenoviruses. Hum Gene Ther. 1998;9(12):1769-74.
156. Orkin SH, Motulsky AG. Report and recommendations of the panel to assess the NIH investment in research on gene therapy 1995 [Available from: http://osp.od.nih.gov/sites/default/files/resources/Orlin_Motulsky_Report.pdf.
157. Raper SE, Chirmule N, Lee FS, Wivel NA, Bagg A, Gao GP, et al. Fatal systemic inflammatory response syndrome in a ornithine transcarbamylase deficient patient following adenoviral gene transfer. Mol Genet Metab. 2003;80(1-2):148-58.
158. Cavazzana-Calvo M, Hacein-Bey S, de Saint Basile G, Gross F, Yvon E, Nusbaum P, et al. Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease. Science. 2000;288(5466):669-72.
159. Fischer A, Hacein-Bey-Abina S, Cavazzana-Calvo M. 20 years of gene therapy for SCID. Nat Immunol. 2010;11(6):457-60.
329 Chapter 9 - References
160. Mukherjee S, Thrasher AJ. Gene therapy for PIDs: progress, pitfalls and prospects. Gene. 2013;525(2):174-81.
161. Hacein-Bey-Abina S, Von Kalle C, Schmidt M, McCormack MP, Wulffraat N, Leboulch P, et al. LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1. Science. 2003;302(5644):415-9.
162. Mingozzi F, High KA. Immune responses to AAV in clinical trials. Curr Gene Ther. 2007;7(5):316-24.
163. Ledley FD, McNamee LM, Uzdil V, Morgan IW. Why commercialization of gene therapy stalled; examining the life cycles of gene therapy technologies. Gene Ther. 2014;21(2):188-94.
164. Bainbridge JW, Smith AJ, Barker SS, Robbie S, Henderson R, Balaggan K, et al. Effect of gene therapy on visual function in Leber's congenital amaurosis. N Engl J Med. 2008;358(21):2231-9.
165. Maguire AM, Simonelli F, Pierce EA, Pugh EN, Jr., Mingozzi F, Bennicelli J, et al. Safety and efficacy of gene transfer for Leber's congenital amaurosis. N Engl J Med. 2008;358(21):2240-8.
166. Cideciyan AV, Aleman TS, Boye SL, Schwartz SB, Kaushal S, Roman AJ, et al. Human gene therapy for RPE65 isomerase deficiency activates the retinoid cycle of vision but with slow rod kinetics. Proc Natl Acad Sci U S A. 2008;105(39):15112-7.
167. Cartier N, Hacein-Bey-Abina S, Bartholomae CC, Veres G, Schmidt M, Kutschera I, et al. Hematopoietic stem cell gene therapy with a lentiviral vector in X- linked adrenoleukodystrophy. Science. 2009;326(5954):818-23.
168. Biffi A, Montini E, Lorioli L, Cesani M, Fumagalli F, Plati T, et al. Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy. Science. 2013;341(6148):1233158.
169. Nathwani AC, Rosales C, McIntosh J, Rastegarlari G, Nathwani D, Raj D, et al. Long-term safety and efficacy following systemic administration of a self- complementary AAV vector encoding human FIX pseudotyped with serotype 5 and 8 capsid proteins. Mol Ther. 2011;19(5):876-85. 330 Chapter 9 - References
170. Bryant LM, Christopher DM, Giles AR, Hinderer C, Rodriguez JL, Smith JB, et al. Lessons learned from the clinical development and market authorization of Glybera. Hum Gene Ther Clin Dev. 2013;24(2):55-64.
171. Cassiday L. Medical research: Gene-therapy reboot. Nature. 2014;509(7502):651-3.
172. Ginn SL, Alexander IE, Edelstein ML, Abedi MR, Wixon J. Gene therapy clinical trials worldwide to 2012 - an update. J Gene Med. 2013;15(2):65-77.
173. http://www.abedia.com/wiley/. [
174. Lu J, Williams JA, Luke J, Zhang F, Chu K, Kay MA. A 5' Noncoding Exon Containing Engineered Intron Enhances Transgene Expression from Recombinant AAV Vectors in vivo. Hum Gene Ther. 2017;28(1):125-34.
175. Lipshutz GS, Titre D, Brindle M, Bisconte AR, Contag CH, Gaensler KM. Comparison of gene expression after intraperitoneal delivery of AAV2 or AAV5 in utero. Mol Ther. 2003;8(1):90-8.
176. Verma IM, Somia N. Gene therapy -- promises, problems and prospects. Nature. 1997;389(6648):239-42.
177. Oishi Y, Kakimoto T, Yuan W, Kuno S, Yamashita H, Chiba T. Fetal Gene Therapy for Ornithine Transcarbamylase Deficiency by Intrahepatic Plasmid DNA- Micro-Bubble Injection Combined with Hepatic Ultrasound Insonation. Ultrasound Med Biol. 2016;42(6):1357-61.
178. Doenecke A, Kromer A, Scherer MN, Schlitt HJ, Geissler EK. AAV plasmid DNA simplifies liver-directed in vivo gene therapy: comparison of expression levels after plasmid DNA-, adeno-associated virus- and adenovirus-mediated liver transfection. J Gene Med. 2010;12(10):810-7.
179. Hou X, Jiao R, Guo X, Wang T, Chen P, Wang D, et al. Construction of minicircle DNA vectors capable of correcting familial hypercholesterolemia phenotype in a LDLR-deficient mouse model. Gene Ther. 2016;23(8-9):657-63.
331 Chapter 9 - References
180. Wu X, Liu G, Mu M, Peng Y, Li X, Deng L, et al. Augmenter of Liver Regeneration Gene Therapy Using a Novel Minicircle DNA Vector Alleviates Liver Fibrosis in Rats. Hum Gene Ther. 2016;27(11):880-91.
181. Viecelli HM, Harbottle RP, Wong SP, Schlegel A, Chuah MK, VandenDriessche T, et al. Treatment of phenylketonuria using minicircle-based naked-DNA gene transfer to murine liver. Hepatology. 2014;60(3):1035-43.
182. Mayrhofer P, Schleef M, Jechlinger W. Use of minicircle plasmids for gene therapy. Methods Mol Biol. 2009;542:87-104.
183. Wolff JA, Budker V. The mechanism of naked DNA uptake and expression. Adv Genet. 2005;54:3-20.
184. Kay MA, He CY, Chen ZY. A robust system for production of minicircle DNA vectors. Nature Biotech. 2010;28(12):1287-9.
185. Chira S, Jackson CS, Oprea I, Ozturk F, Pepper MS, Diaconu I, et al. Progresses towards safe and efficient gene therapy vectors. Oncotarget. 2015;6(31):30675-703.
186. Elsabahy M, Nazarali A, Foldvari M. Non-viral nucleic acid delivery: key challenges and future directions. Curr Drug Deliv. 2011;8(3):235-44.
187. Simonato M, Bennett J, Boulis NM, Castro MG, Fink DJ, Goins WF, et al. Progress in gene therapy for neurological disorders. Nat Rev Neurol. 2013;9(5):277- 91.
188. D'Avola D, Lopez-Franco E, Sangro B, Paneda A, Grossios N, Gil-Farina I, et al. Phase I open label liver-directed gene therapy clinical trial for acute intermittent porphyria. J Hepatol. 2016;65(4):776-83.
189. Mingozzi F, High KA. Immune responses to AAV vectors: overcoming barriers to successful gene therapy. Blood. 2013;122(1):23-36.
190. Nathwani AC, Gray JT, McIntosh J, Ng CY, Zhou J, Spence Y, et al. Safe and efficient transduction of the liver after peripheral vein infusion of self-complementary
332 Chapter 9 - References
AAV vector results in stable therapeutic expression of human FIX in nonhuman primates. Blood. 2007;109(4):1414-21.
191. Sarkar R, Mucci M, Addya S, Tetreault R, Bellinger DA, Nichols TC, et al. Long-term efficacy of adeno-associated virus serotypes 8 and 9 in hemophilia a dogs and mice. Hum Gene Thery. 2006;17(4):427-39.
192. Grossman M, Raper SE, Kozarsky K, Stein EA, Engelhardt JF, Muller D, et al. Successful ex vivo gene therapy directed to liver in a patient with familial hypercholesterolaemia. Nat Genet. 1994;6(4):335-41.
193. Jooss K, Chirmule N. Immunity to adenovirus and adeno-associated viral vectors: implications for gene therapy. Gene Ther. 2003;10(11):955-63.
194. Zaiss AK, Muruve DA. Immunity to adeno-associated virus vectors in animals and humans: a continued challenge. Gene Ther. 2008;15(11):808-16.
195. Wold WS, Toth K. Adenovirus vectors for gene therapy, vaccination and cancer gene therapy. Curr Gene Ther. 2013;13(6):421-33.
196. Chuah MK, Collen D, VandenDriessche T. Clinical gene transfer studies for hemophilia A. Semin Thromb Hemost. 2004;30(2):249-56.
197. Calcedo R, Wilson JM. Humoral Immune Response to AAV. Front Immunol. 2013;4:341.
198. Annoni A, Goudy K, Akbarpour M, Naldini L, Roncarolo MG. Immune responses in liver-directed lentiviral gene therapy. Transl Res. 2013;161(4):230-40.
199. Basner-Tschakarjan E, Mingozzi F. Cell-Mediated Immunity to AAV Vectors, Evolving Concepts and Potential Solutions. Front Immunol. 2014;5:350.
200. Bessis N, GarciaCozar FJ, Boissier MC. Immune responses to gene therapy vectors: influence on vector function and effector mechanisms. Gene Ther. 2004;11 Suppl 1:S10-7.
333 Chapter 9 - References
201. Tang J, Olive M, Pulmanausahakul R, Schnell M, Flomenberg N, Eisenlohr L, et al. Human CD8+ cytotoxic T cell responses to adenovirus capsid proteins. Virology. 2006;350(2):312-22.
202. Manno CS, Pierce GF, Arruda VR, Glader B, Ragni M, Rasko JJ, et al. Successful transduction of liver in hemophilia by AAV-Factor IX and limitations imposed by the host immune response. Nat Med. 2006;12(3):342-7.
203. Mingozzi F, Maus MV, Hui DJ, Sabatino DE, Murphy SL, Rasko JE, et al. CD8(+) T-cell responses to adeno-associated virus capsid in humans. Nat Med. 2007;13(4):419-22.
204. Nathwani AC, Reiss UM, Tuddenham EG, Rosales C, Chowdary P, McIntosh J, et al. Long-term safety and efficacy of factor IX gene therapy in hemophilia B. N Engl J Med. 2014;371(21):1994-2004.
205. Grossman M, Rader DJ, Muller DW, Kolansky DM, Kozarsky K, Clark BJ, 3rd, et al. A pilot study of ex vivo gene therapy for homozygous familial hypercholesterolaemia. Nat Med. 1995;1(11):1148-54.
206. Powell JS, Ragni MV, White GC, 2nd, Lusher JM, Hillman-Wiseman C, Moon TE, et al. Phase 1 trial of FVIII gene transfer for severe hemophilia A using a retroviral construct administered by peripheral intravenous infusion. Blood. 2003;102(6):2038-45.
207. Dolgin E. Early clinical data raise the bar for hemophilia gene therapies. Nature Biotech. 2016;34(10):999-1001.
208. Wilson JM. Lessons learned from the gene therapy trial for ornithine transcarbamylase deficiency. Mol Genet Metab. 2009;96(4):151-7.
209. Alba R, Bosch A, Chillon M. Gutless adenovirus: last-generation adenovirus for gene therapy. Gene Ther. 2005;12 Suppl 1:S18-27.
210. White GI, Monahan PE. Gene therapy for hemophilia A. In: Lee CR, Berntrop E, Hoots Ke, editors. Textbook of Hemophilia. Oxford, UK: Blackwell Publishing 2005. p. 226-8.
334 Chapter 9 - References
211. Chandler RJ, Venditti CP. Gene Therapy for Metabolic Diseases. Transl Sci Rare Dis. 2016;1(1):73-89.
212. Nienhuis AW, Nathwani AC, Davidoff AM. Gene Therapy for Hemophilia. Hum Gene Ther. 2016;27(4):305-8.
213. Baruteau J, Waddington SN, Alexander IE, Gissen P. Delivering efficient liver- directed AAV-mediated gene therapy. Gene Ther. 2017;24(5):263-4.
214. Miesbach W, Tangelder M, Klamroth R. Updated results from a dose escalating study in adult patients with haemophilia B treated with AMT-060 (AAV5- hFIX) gene therapy. Haemophilia. 2016;22(Suppl 4.):151.
215. Monahan P, Walsh C, Powell J. Update on a phase 1/2 open-label trial of BAX335, an adeno-associated virus 8 (AAV8) vector-based gene therapy program for hemophilia B. J Thromb Haemost. 2015;13 (Suppl 2):87.
216. George L, Sullivan S, Giermasz A. Adeno-associated virus mediated gene transfer for Hemophilia B achieves sustained mean factor IX activity levels of >30% without immunosuppression. 58th annual meeting of the American Society of Hemophilia. 2016.
217. Pasi J, Wong W, Rangarajan S. Interim results of an open-label, phase 1/2 study of BMN 270, an AAV5-FVIII gene transfer in severe hemophilia A. Haemophilia. 2016;22(Suppl4.):151.
218. Fahs SA, Hille MT, Shi Q, Weiler H, Montgomery RR. A conditional knockout mouse model reveals endothelial cells as the principal and possibly exclusive source of plasma factor VIII. Blood. 2014;123(24):3706-13.
219. Kattenhorn LM, Tipper CH, Stoica L, Geraghty DS, Wright TL, Clark KR, et al. Adeno-Associated Virus Gene Therapy for Liver Disease. Hum Gene Ther. 2016;27(12):947-61.
220. Hastie E, Samulski RJ. Recombinant adeno-associated virus vectors in the treatment of rare diseases. Expert Opin Orphan Drugs. 2015;3(6):675-89.
335 Chapter 9 - References
221. Junge N, Mingozzi F, Ott M, Baumann U. Adeno-associated virus vector- based gene therapy for monogenetic metabolic diseases of the liver. J Pediatr Gastroenterol Nutr. 2015;60(4):433-40.
222. Atchison RW, Casto BC, Hammon WM. Adenovirus-Associated Defective Virus Particles. Science. 1965;149(3685):754-6.
223. Goncalves MA. Adeno-associated virus: from defective virus to effective vector. Virol J. 2005;2:43.
224. Samulski RJ, Muzyczka N. AAV-Mediated Gene Therapy for Research and Therapeutic Purposes. Annu Rev Virol. 2014;1(1):427-51.
225. Naumer M, Sonntag F, Schmidt K, Nieto K, Panke C, Davey NE, et al. Properties of the adeno-associated virus assembly-activating protein. J Virol. 2012;86(23):13038-48.
226. Earley LF, Powers JM, Adachi K, Baumgart JT, Meyer NL, Xie Q, et al. Adeno- associated Virus (AAV) Assembly-Activating Protein Is Not an Essential Requirement for Capsid Assembly of AAV Serotypes 4, 5, and 11. J Virol. 2017;91(3).
227. Schultz BR, Chamberlain JS. Recombinant adeno-associated virus transduction and integration. Mol Ther. 2008;16(7):1189-99.
228. Asokan A, Schaffer DV, Samulski RJ. The AAV vector toolkit: poised at the clinical crossroads. Mol Ther. 2012;20(4):699-708.
229. Vercauteren K, Hoffman BE, Zolotukhin I, Keeler GD, Xiao JW, Basner- Tschakarjan E, et al. Superior In vivo Transduction of Human Hepatocytes Using Engineered AAV3 Capsid. Mol Ther. 2016;24(6):1042-9.
230. Pillay S, Meyer NL, Puschnik AS, Davulcu O, Diep J, Ishikawa Y, et al. An essential receptor for adeno-associated virus infection. Nature. 2016;530(7588):108- 12.
231. Summerford C, Samulski RJ. AAVR: A Multi-Serotype Receptor for AAV. Mol Ther. 2016;24(4):663-6.
336 Chapter 9 - References
232. Gao G, Vandenberghe LH, Alvira MR, Lu Y, Calcedo R, Zhou X, et al. Clades of Adeno-associated viruses are widely disseminated in human tissues. J Virol. 2004;78(12):6381-8.
233. Boutin S, Monteilhet V, Veron P, Leborgne C, Benveniste O, Montus MF, et al. Prevalence of serum IgG and neutralizing factors against adeno-associated virus (AAV) types 1, 2, 5, 6, 8, and 9 in the healthy population: implications for gene therapy using AAV vectors. Hum Gene Ther. 2010;21(6):704-12.
234. Louis Jeune V, Joergensen JA, Hajjar RJ, Weber T. Pre-existing anti-adeno- associated virus antibodies as a challenge in AAV gene therapy. Hum Gene Ther Methods. 2013;24(2):59-67.
235. Calcedo R, Vandenberghe LH, Gao G, Lin J, Wilson JM. Worldwide epidemiology of neutralizing antibodies to adeno-associated viruses. J Infect Dis. 2009;199(3):381-90.
236. Wu Z, Asokan A, Samulski RJ. Adeno-associated virus serotypes: vector toolkit for human gene therapy. Mol Ther. 2006;14(3):316-27.
237. Rabinowitz JE, Rolling F, Li C, Conrath H, Xiao W, Xiao X, et al. Cross- packaging of a single adeno-associated virus (AAV) type 2 vector genome into multiple AAV serotypes enables transduction with broad specificity. J Virol. 2002;76(2):791-801.
238. Smith RH. Adeno-associated virus integration: virus versus vector. Gene Ther. 2008;15(11):817-22.
239. Cunningham SC, Dane AP, Spinoulas A, Logan GJ, Alexander IE. Gene delivery to the juvenile mouse liver using AAV2/8 vectors. Mol Ther. 2008;16(6):1081-8.
240. Baruteau J, Waddington SN, Alexander IE, Gissen P. Gene therapy for monogenic liver diseases: clinical successes, current challenges and future prospects. J Inherit Metab Dis. 2017;40(4):497-517.
337 Chapter 9 - References
241. Nonnenmacher M, Weber T. Adeno-associated virus 2 infection requires endocytosis through the CLIC/GEEC pathway. Cell Host Microbe. 2011;10(6):563- 76.
242. Bartlett JS, Wilcher R, Samulski RJ. Infectious entry pathway of adeno- associated virus and adeno-associated virus vectors. J Virol. 2000;74(6):2777-85.
243. Douar AM, Poulard K, Stockholm D, Danos O. Intracellular trafficking of adeno-associated virus vectors: routing to the late endosomal compartment and proteasome degradation. J Virol. 2001;75(4):1824-33.
244. Hansen J, Qing K, Srivastava A. Infection of purified nuclei by adeno- associated virus 2. Mol Ther. 2001;4(4):289-96.
245. Nakai H, Yant SR, Storm TA, Fuess S, Meuse L, Kay MA. Extrachromosomal recombinant adeno-associated virus vector genomes are primarily responsible for stable liver transduction in vivo. J Virol. 2001;75(15):6969-76.
246. Berry GE, Asokan A. Cellular transduction mechanisms of adeno-associated viral vectors. Curr Opin Virol. 2016;21:54-60.
247. McCarty DM, Young SM, Jr., Samulski RJ. Integration of adeno-associated virus (AAV) and recombinant AAV vectors. Annu Rev Genet. 2004;38:819-45.
248. Kotin RM, Menninger JC, Ward DC, Berns KI. Mapping and direct visualization of a region-specific viral DNA integration site on chromosome 19q13-qter. Genomics. 1991;10(3):831-4.
249. Weitzman MD, Kyostio SR, Kotin RM, Owens RA. Adeno-associated virus (AAV) Rep proteins mediate complex formation between AAV DNA and its integration site in human DNA. Proc Natl Acad Sci U S A. 1994;91(13):5808-12.
250. Nathwani AC, Cochrane M, McIntosh J, Ng CY, Zhou J, Gray JT, et al. Enhancing transduction of the liver by adeno-associated viral vectors. Gene Ther. 2009;16(1):60-9.
338 Chapter 9 - References
251. Zhong L, Li B, Jayandharan G, Mah CS, Govindasamy L, Agbandje-McKenna M, et al. Tyrosine-phosphorylation of AAV2 vectors and its consequences on viral intracellular trafficking and transgene expression. Virology. 2008;381(2):194-202.
252. Petrs-Silva H, Dinculescu A, Li Q, Min SH, Chiodo V, Pang JJ, et al. High- efficiency transduction of the mouse retina by tyrosine-mutant AAV serotype vectors. Mol Ther. 2009;17(3):463-71.
253. McCarty DM, Fu H, Monahan PE, Toulson CE, Naik P, Samulski RJ. Adeno- associated virus terminal repeat (TR) mutant generates self-complementary vectors to overcome the rate-limiting step to transduction in vivo. Gene Ther. 2003;10(26):2112-8.
254. McCarty DM. Self-complementary AAV vectors; advances and applications. Mol Ther. 2008;16(10):1648-56.
255. Martino AT, Suzuki M, Markusic DM, Zolotukhin I, Ryals RC, Moghimi B, et al. The genome of self-complementary adeno-associated viral vectors increases Toll- like receptor 9-dependent innate immune responses in the liver. Blood. 2011;117(24):6459-68.
256. Xiao X, Li J, Samulski RJ. Production of high-titer recombinant adeno- associated virus vectors in the absence of helper adenovirus. J Virol. 1998;72(3):2224-32.
257. D'Costa S, Blouin V, Broucque F, Penaud-Budloo M, Francois A, Perez IC, et al. Practical utilization of recombinant AAV vector reference standards: focus on vector genomes titration by free ITR qPCR. Mol Ther Methods Clin Dev 2016;5:16019.
258. Fagone P, Wright JF, Nathwani AC, Nienhuis AW, Davidoff AM, Gray JT. Systemic errors in quantitative polymerase chain reaction titration of self- complementary adeno-associated viral vectors and improved alternative methods. Hum Gene Ther Methods. 2012;23(1):1-7.
259. Srivastava A, Carter BJ. AAV Infection: Protection from Cancer. Hum Gene Ther. 2016.
339 Chapter 9 - References
260. Donsante A, Miller DG, Li Y, Vogler C, Brunt EM, Russell DW, et al. AAV vector integration sites in mouse hepatocellular carcinoma. Science. 2007;317(5837):477.
261. Chandler RJ, LaFave MC, Varshney GK, Trivedi NS, Carrillo-Carrasco N, Senac JS, et al. Vector design influences hepatic genotoxicity after adeno-associated virus gene therapy. J Clin Invest. 2015;125(2):870-80.
262. Chandler RJ, Sands MS, Venditti CP. rAAV integration and genotoxicity: insights from animal models. Hum Gene Ther. 2017;28(4):314-322.
263. Gil-Farina I, Schmidt M. Interaction of vectors and parental viruses with the host genome. Curr Opin Virol. 2016;21:35-40.
264. Kaeppel C, Beattie SG, Fronza R, van Logtenstein R, Salmon F, Schmidt S, et al. A largely random AAV integration profile after LPLD gene therapy. Nat Med. 2013;19(7):889-91.
265. Riordan JD. Letter to the Editor regarding manuscript doi:10.1089/hum.2016.147. Hum Gene Ther. 2017.
266. Nault JC, Datta S, Imbeaud S, Franconi A, Mallet M, Couchy G, et al. Recurrent AAV2-related insertional mutagenesis in human hepatocellular carcinomas. Nat Genet. 2015;47(10):1187-93.
267. Berns KI, Byrne BJ, Flotte TR, Gao G, Hauswirth WW, Herzog RW, et al. Adeno-Associated Virus Type 2 and Hepatocellular Carcinoma? Hum Gene Ther. 2015;26(12):779-81.
268. Buning H, Schmidt M. Adeno-associated Vector Toxicity-To Be or Not to Be? Mol Ther. 2015;23(11):1673-5.
269. Blacklow NR, Hoggan MD, Sereno MS, Brandt CD, Kim HW, Parrott RH, et al. A seroepidemiologic study of adenovirus-associated virus infection in infants and children. Am J Epidemiol. 1971;94(4):359-66.
270. Jayandharan GR, Aslanidi G, Martino AT, Jahn SC, Perrin GQ, Herzog RW, et al. Activation of the NF-kappaB pathway by adeno-associated virus (AAV) vectors
340 Chapter 9 - References and its implications in immune response and gene therapy. Proc Natl Acad Sci U S A. 2011;108(9):3743-8.
271. Zhu J, Huang X, Yang Y. The TLR9-MyD88 pathway is critical for adaptive immune responses to adeno-associated virus gene therapy vectors in mice. J Clin Invest. 2009;119(8):2388-98.
272. Hosel M, Broxtermann M, Janicki H, Esser K, Arzberger S, Hartmann P, et al. Toll-like receptor 2-mediated innate immune response in human nonparenchymal liver cells toward adeno-associated viral vectors. Hepatology. 2012;55(1):287-97.
273. Jiang H, Couto LB, Patarroyo-White S, Liu T, Nagy D, Vargas JA, et al. Effects of transient immunosuppression on adenoassociated, virus-mediated, liver-directed gene transfer in rhesus macaques and implications for human gene therapy. Blood. 2006;108(10):3321-8.
274. Cooper M, Nayak S, Hoffman BE, Terhorst C, Cao O, Herzog RW. Improved induction of immune tolerance to factor IX by hepatic AAV-8 gene transfer. Hum Gene Ther. 2009;20(7):767-76.
275. Shi Y, Falahati R, Zhang J, Flebbe-Rehwaldt L, Gaensler KM. Role of antigen- specific regulatory CD4+CD25+ T cells in tolerance induction after neonatal IP administration of AAV-hF.IX. Gene Ther. 2013;20(10):987-96.
276. Perrin GQ, Zolotukhin I, Sherman A, Biswas M, de Jong YP, Terhorst C, et al. Dynamics of antigen presentation to transgene product-specific CD4+ T cells and of Treg induction upon hepatic AAV gene transfer. Mol Ther Methods Clin Dev. 2016;3:16083.
277. Tseng YS, Agbandje-McKenna M. Mapping the AAV Capsid Host Antibody Response toward the Development of Second Generation Gene Delivery Vectors. Front Immunol. 2014;5:9.
278. Mingozzi F, High KA. Immune responses to AAV in clinical trials. Curr Gene Ther. 2011;11(4):321-30.
279. Chicoine LG, Montgomery CL, Bremer WG, Shontz KM, Griffin DA, Heller KN, et al. Plasmapheresis eliminates the negative impact of AAV antibodies on 341 Chapter 9 - References microdystrophin gene expression following vascular delivery. Mol Ther. 2014;22(2):338-47.
280. Flanigan KM, Campbell K, Viollet L, Wang W, Gomez AM, Walker CM, et al. Anti-dystrophin T cell responses in Duchenne muscular dystrophy: prevalence and a glucocorticoid treatment effect. Hum Gene Ther. 2013;24(9):797-806.
281. Mingozzi F, Chen Y, Edmonson SC, Zhou S, Thurlings RM, Tak PP, et al. Prevalence and pharmacological modulation of humoral immunity to AAV vectors in gene transfer to synovial tissue. Gene Ther. 2013;20(4):417-24.
282. Corti M, Elder M, Falk D, Lawson L, Smith B, Nayak S, et al. B-Cell Depletion is Protective Against Anti-AAV Capsid Immune Response: A Human Subject Case Study. Mol Ther Methods Clin Dev. 2014;1.
283. McIntosh JH, Cochrane M, Cobbold S, Waldmann H, Nathwani SA, Davidoff AM, et al. Successful attenuation of humoral immunity to viral capsid and transgenic protein following AAV-mediated gene transfer with a non-depleting CD4 antibody and cyclosporine. Gene Ther. 2012;19(1):78-85.
284. Mingozzi F, Chen Y, Murphy SL, Edmonson SC, Tai A, Price SD, et al. Pharmacological modulation of humoral immunity in a nonhuman primate model of AAV gene transfer for hemophilia B. Mol Ther. 2012;20(7):1410-6.
285. Monahan PE, Lothrop CD, Sun J, Hirsch ML, Kafri T, Kantor B, et al. Proteasome inhibitors enhance gene delivery by AAV virus vectors expressing large genomes in hemophilia mouse and dog models: a strategy for broad clinical application. Mol Ther. 2010;18(11):1907-16.
286. Hirsch ML, Wolf SJ, Samulski RJ. Delivering Transgenic DNA Exceeding the Carrying Capacity of AAV Vectors. Methods Mol Biol. 2016;1382:21-39.
287. McIntosh J, Lenting PJ, Rosales C, Lee D, Rabbanian S, Raj D, et al. Therapeutic levels of FVIII following a single peripheral vein administration of rAAV vector encoding a novel human factor VIII variant. Blood. 2013;121(17):3335-44.
342 Chapter 9 - References
288. Counsell JR, Asgarian, Z., Meng, J., Ferrer, V., Vink, C., A., Howe, S., J., Waddington, S., N., Thrasher, A. J., Muntoni, F., Morgan, J. E., and Danos, O. . Lentiviral vectors can be used for full-length dystrophin gene therapy. Sci Rep In Press. 2017.
289. Suzuki T, Sasaki T, Yano K, Sakurai F, Kawabata K, Kondoh M, et al. Development of a recombinant adenovirus vector production system free of replication-competent adenovirus by utilizing a packaging size limit of the viral genome. Virus Res. 2011;158(1-2):154-60.
290. Yan Z, Sun X, Feng Z, Li G, Fisher JT, Stewart ZA, et al. Optimization of Recombinant Adeno-Associated Virus-Mediated Expression for Large Transgenes, Using a Synthetic Promoter and Tandem Array Enhancers. Hum Gene Ther. 2015;26(6):334-46.
291. Grieger JC, Soltys SM, Samulski RJ. Production of Recombinant Adeno- associated Virus Vectors Using Suspension HEK293 Cells and Continuous Harvest of Vector From the Culture Media for GMP FIX and FLT1 Clinical Vector. Mol Ther. 2016;24(2):287-97.
292. Clement N, Grieger JC. Manufacturing of recombinant adeno-associated viral vectors for clinical trials. Mol Ther Methods Clin Dev. 2016;3:16002.
293. Bell P, Wang L, Gao G, Haskins ME, Tarantal AF, McCarter RJ, et al. Inverse zonation of hepatocyte transduction with AAV vectors between mice and non-human primates. Mol Genet Metab. 2011;104(3):395-403.
294. Cunningham SC, Kok CY, Dane AP, Carpenter K, Kizana E, Kuchel PW, et al. Induction and prevention of severe hyperammonemia in the spfash mouse model of ornithine transcarbamylase deficiency using shRNA and rAAV-mediated gene delivery. Mol Ther. 2011;19(5):854-9.
295. Cunningham SC, Siew SM, Hallwirth CV, Bolitho C, Sasaki N, Garg G, et al. Modeling correction of severe urea cycle defects in the growing murine liver using a hybrid recombinant adeno-associated virus/piggyBac transposase gene delivery system. Hepatology. 2015;62(2):417-28.
343 Chapter 9 - References
296. Yang Y, Wang L, Bell P, McMenamin D, He Z, White J, et al. A dual AAV system enables the Cas9-mediated correction of a metabolic liver disease in newborn mice. Nature Biotech. 2016;34(3):334-8.
297. Coppoletta JM, Wolbach SB. Body Length and Organ Weights of Infants and Children: A Study of the Body Length and Normal Weights of the More Important Vital Organs of the Body between Birth and Twelve Years of Age. Am J Pathol. 1933;9(1):55-70.
298. Sunderman FW, Boerner F. Normal value in clinical medicine. In: Saunders. PWB, editor.1949.
299. Ban K, Sugiyama N, Sugiyama K, Wada Y, Suzuki T, Hashimoto T, et al. A pediatric patient with classical citrullinemia who underwent living-related partial liver transplantation. Transplantation. 2001;71(10):1495-7.
300. Cunningham SC, Spinoulas A, Carpenter KH, Wilcken B, Kuchel PW, Alexander IE. AAV2/8-mediated correction of OTC deficiency is robust in adult but not neonatal Spf(ash) mice. Mol Ther. 2009;17(8):1340-6.
301. Wang L, Morizono H, Lin J, Bell P, Jones D, McMenamin D, et al. Preclinical evaluation of a clinical candidate AAV8 vector for ornithine transcarbamylase (OTC) deficiency reveals functional enzyme from each persisting vector genome. Mol Genet Metab. 2012;105(2):203-11.
302. Moscioni D, Morizono H, McCarter RJ, Stern A, Cabrera-Luque J, Hoang A, et al. Long-term correction of ammonia metabolism and prolonged survival in ornithine transcarbamylase-deficient mice following liver-directed treatment with adeno- associated viral vectors. Mol Ther. 2006;14(1):25-33.
303. Cunningham SC, Kok CY, Spinoulas A, Carpenter KH, Alexander IE. AAV- encoded OTC activity persisting to adulthood following delivery to newborn spf(ash) mice is insufficient to prevent shRNA-induced hyperammonaemia. Gene Ther. 2013;20(12):1184-7.
344 Chapter 9 - References
304. Kok CY, Cunningham SC, Carpenter KH, Dane AP, Siew SM, Logan GJ, et al. Adeno-associated virus-mediated rescue of neonatal lethality in argininosuccinate synthetase-deficient mice. Mol Ther. 2013;21(10):1823-31.
305. Lee EK, Hu C, Bhargava R, Rozengurt N, Stout D, Grody WW, et al. Long- term survival of the juvenile lethal arginase-deficient mouse with AAV gene therapy. Mol Ther. 2012;20(10):1844-51.
306. Lee EK, Hu C, Bhargava R, Ponnusamy R, Park H, Novicoff S, et al. AAV- based gene therapy prevents neuropathology and results in normal cognitive development in the hyperargininemic mouse. Gene Ther. 2013;20(8):785-96.
307. Hu C, Tai DS, Park H, Cantero G, Chan E, Yudkoff M, et al. Minimal ureagenesis is necessary for survival in the murine model of hyperargininemia treated by AAV-based gene therapy. Gene Ther. 2015;22(2):111-5.
308. Alexander IE, Kok C, Dane AP, Cunningham SC. Gene therapy for metabolic disorders: an overview with a focus on urea cycle disorders. J Inherit Metab Dis. 2012;35(4):641-5.
309. Deignan JL, Cederbaum SD, Grody WW. Contrasting features of urea cycle disorders in human patients and knockout mouse models. Mol Genet Metab. 2008;93(1):7-14.
310. Erez A. Argininosuccinic aciduria: from a monogenic to a complex disorder. Genet Med. 2013;15(4):251-7.
311. Muramatsu S, Fujimoto K, Kato S, Mizukami H, Asari S, Ikeguchi K, et al. A phase I study of aromatic L-amino acid decarboxylase gene therapy for Parkinson's disease. Mol Ther. 2010;18(9):1731-5.
312. Davison JE, Davies NP, Wilson M, Sun Y, Chakrapani A, McKiernan PJ, et al. MR spectroscopy-based brain metabolite profiling in propionic acidaemia: metabolic changes in the basal ganglia during acute decompensation and effect of liver transplantation. Orphanet J Rare Dis. 2011;6:19.
313. Emery JL, Howat AJ, Variend S, Vawter GF. Investigation of inborn errors of metabolism in unexpected infant deaths. Lancet. 1988;2(8601):29-31. 345 Chapter 9 - References
314. Carter RJ, Morton J, Dunnett SB. Motor coordination and balance in rodents. Curr Protoc Neurosci. 2001;Chapter 8:Unit 8 12.
315. Tort AB, Neto WP, Amaral OB, Kazlauckas V, Souza DO, Lara DR. A simple webcam-based approach for the measurement of rodent locomotion and other behavioural parameters. J Neurosci Methods. 2006;157(1):91-7.
316. Leger M, Quiedeville A, Bouet V, Haelewyn B, Boulouard M, Schumann-Bard P, et al. Object recognition test in mice. Nat Protoc. 2013;8(12):2531-7.
317. Bergner CL, Smolinsky AN, Hart PC, Dufour BD, Egan RJ, Laporte JL, et al. Mouse models for studying depression-like states and antidepressant drugs. Methods Mol Biol. 2010;602:267-82.
318. Glowinski J, Iversen L. Regional studies of catecholamines in the rat brain. 3. Subcellullar distribution of endogenous and exogenous catecholamines in various brain regions. Biochem Pharmacol. 1966;15(7):977-87.
319. Waddington SN, Mosley K, Cattell V. Induced nitric oxide (NO) synthesis in heterologous nephrotoxic nephritis; effects of selective inhibition in neutrophil- dependent glomerulonephritis. Clin Exp Immunol. 1999;118(2):309-14.
320. Kamencic H, Lyon A, Paterson PG, Juurlink BH. Monochlorobimane fluorometric method to measure tissue glutathione. Anal Biochem. 2000;286(1):35-7.
321. Rahim AA, Wong AM, Hoefer K, Buckley SM, Mattar CN, Cheng SH, et al. Intravenous administration of AAV2/9 to the fetal and neonatal mouse leads to differential targeting of CNS cell types and extensive transduction of the nervous system. FASEB J. 2011;25(10):3505-18.
322. Rocha-Ferreira E, Phillips E, Francesch-Domenech E, Thei L, Peebles DM, Raivich G, et al. The role of different strain backgrounds in bacterial endotoxin- mediated sensitization to neonatal hypoxic-ischemic brain damage. Neuroscience. 2015;311:292-307.
346 Chapter 9 - References
323. Wisse E, Braet F, Duimel H, Vreuls C, Koek G, Olde Damink SW, et al. Fixation methods for electron microscopy of human and other liver. World J Gastroenterol. 2010;16(23):2851-66.
324. Reynolds ES. The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J Cell Biol. 1963;17:208-12.
325. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9(7):676-82.
326. Charrier S, Stockholm D, Seye K, Opolon P, Taveau M, Gross DA, et al. A lentiviral vector encoding the human Wiskott-Aldrich syndrome protein corrects immune and cytoskeletal defects in WASP knockout mice. Gene Ther. 2005;12(7):597-606.
327. Nathwani AC, Davidoff A, Hanawa H, Zhou JF, Vanin EF, Nienhuis AW. Factors influencing in vivo transduction by recombinant adeno-associated viral vectors expressing the human factor IX cDNA. Blood. 2001;97(5):1258-65.
328. Davidoff AM, Ng CY, Sleep S, Gray J, Azam S, Zhao Y, et al. Purification of recombinant adeno-associated virus type 8 vectors by ion exchange chromatography generates clinical grade vector stock. J Virol Methods. 2004;121(2):209-15.
329. Smith RH, Levy JR, Kotin RM. A simplified baculovirus-AAV expression vector system coupled with one-step affinity purification yields high-titer rAAV stocks from insect cells. Mol Ther. 2009;17(11):1888-96.
330. Wang Q, Lock M, Prongay AJ, Alvira MR, Petkov B, Wilson JM. Identification of an adeno-associated virus binding epitope for AVB sepharose affinity resin. Mol Ther Methods Clin Dev. 2015;2:15040.
331. Micromass Quattro API Mass spectrometer Operator Guide. http://www.ecs.umass.edu/eve/facilities/equipment/Quattromicro/Quattro_Micro_API_ guide_rev_A.pdf; page 200 [
347 Chapter 9 - References
332. Kaspar H, Dettmer K, Gronwald W, Oefner PJ. Advances in amino acid analysis. Anal Bioanal Chem. 2009;393(2):445-52.
333. Jambor A, Molnar-Perl I. Amino acid analysis by high-performance liquid chromatography after derivatization with 9-fluorenylmethyloxycarbonyl chloride Literature overview and further study. J Chromatogr A. 2009;1216(15):3064-77.
334. Bank RA, Jansen EJ, Beekman B, te Koppele JM. Amino acid analysis by reverse-phase high-performance liquid chromatography: improved derivatization and detection conditions with 9-fluorenylmethyl chloroformate. Anal Biochem. 1996;240(2):167-76.
335. Cumming G, Fidler F, Vaux DL. Error bars in experimental biology. J Cell Biol. 2007;177(1):7-11.
336. Kolker S, Valayannopoulos V, Burlina AB, Sykut-Cegielska J, Wijburg FA, Teles EL, et al. The phenotypic spectrum of organic acidurias and urea cycle disorders. Part 2: the evolving clinical phenotype. J Inherit Metab Dis. 2015;38(6):1059-74.
337. Baruteau J, Jameson E, Morris AA, Chakrapani A, Santra S, Vijay S, et al. Expanding the phenotype in argininosuccinic aciduria: need for new therapies. J Inherit Metab Dis. 2017;40:357-68.
338. Burgard P, Kolker S, Haege G, Lindner M, Hoffmann GF. Neonatal mortality and outcome at the end of the first year of life in early onset urea cycle disorders- review and meta-analysis of observational studies published over more than 35 years. J Inherit Metab Dis. 2016;39(2):219-29.
339. Premkumar MH, Sule G, Nagamani SC, Chakkalakal S, Nordin A, Jain M, et al. Argininosuccinate lyase in enterocytes protects from development of necrotizing enterocolitis. Am J Physiol Gastrointest Liver Physiol. 2014;307(3):G347-54.
340. Hu L, Pandey AV, Balmer C, Eggimann S, Rufenacht V, Nuoffer JM, et al. Unstable argininosuccinate lyase in variant forms of the urea cycle disorder argininosuccinic aciduria. J Inherit Metab Dis. 2015;38(5):815-27.
348 Chapter 9 - References
341. Barbosa P, Cialkowski M, O'Brien WE. Analysis of naturally occurring and site- directed mutations in the argininosuccinate lyase gene. J Biol Chem. 1991;266(8):5286-90.
342. Stockler S, Holzbach U, Hanefeld F, Marquardt I, Helms G, Requart M, et al. Creatine deficiency in the brain: a new, treatable inborn error of metabolism. Pediatr Res. 1994;36(3):409-13.
343. Stockler-Ipsiroglu S, van Karnebeek C, Longo N, Korenke GC, Mercimek- Mahmutoglu S, Marquart I, et al. Guanidinoacetate methyltransferase (GAMT) deficiency: outcomes in 48 individuals and recommendations for diagnosis, treatment and monitoring. Mol Genet Metab. 2014;111(1):16-25.
344. Deignan JL, De Deyn PP, Cederbaum SD, Fuchshuber A, Roth B, Gsell W, et al. Guanidino compound levels in blood, cerebrospinal fluid, and post-mortem brain material of patients with argininemia. Mol Genet Metab. 2010;100 Suppl 1:S31-6.
345. D'Hooge R, Pei YQ, Marescau B, De Deyn PP. Convulsive action and toxicity of uremic guanidino compounds: behavioral assessment and relation to brain concentration in adult mice. J Neurol Sci. 1992;112(1-2):96-105.
346. Aoyagi K. Inhibition of arginine synthesis by urea: a mechanism for arginine deficiency in renal failure which leads to increased hydroxyl radical generation. Mol Cell Biochem. 2003;244(1-2):11-5.
347. Aoyagi K, Shahrzad S, Iida S, Tomida C, Hirayama A, Nagase S, et al. Role of nitric oxide in the synthesis of guanidinosuccinic acid, an activator of the N-methyl-D- aspartate receptor. Kidney Int Suppl. 2001;78:S93-6.
348. Schulze A. Creatine deficiency syndromes. Mol Cell Biochem. 2003;244(1- 2):143-50.
349. Hanna-El-Daher L, Braissant O. Creatine synthesis and exchanges between brain cells: What can be learned from human creatine deficiencies and various experimental models? Amino Acids. 2016.
349 Chapter 9 - References
350. Kan MJ, Lee JE, Wilson JG, Everhart AL, Brown CM, Hoofnagle AN, et al. Arginine deprivation and immune suppression in a mouse model of Alzheimer's disease. J Neurosci. 2015;35(15):5969-82.
351. Braissant O, Cagnon L, Monnet-Tschudi F, Speer O, Wallimann T, Honegger P, et al. Ammonium alters creatine transport and synthesis in a 3D culture of developing brain cells, resulting in secondary cerebral creatine deficiency. Eur J Neurosci. 2008;27(7):1673-85.
352. Leitch B, Shevtsova O, Reusch K, Bergin DH, Liu P. Spatial learning-induced increase in agmatine levels at hippocampal CA1 synapses. Synapse. 2011;65(2):146-53.
353. Utkan T, Gocmez SS, Regunathan S, Aricioglu F. Agmatine, a metabolite of L- arginine, reverses scopolamine-induced learning and memory impairment in rats. Pharmacol Biochem Behav. 2012;102(4):578-84.
354. Molderings GJ, Haenisch B. Agmatine (decarboxylated L-arginine): physiological role and therapeutic potential. Pharmacol Ther. 2012;133(3):351-65.
355. Demehri S, Homayoun H, Honar H, Riazi K, Vafaie K, Roushanzamir F, et al. Agmatine exerts anticonvulsant effect in mice: modulation by alpha 2-adrenoceptors and nitric oxide. Neuropharmacology. 2003;45(4):534-42.
356. Uzbay TI. The pharmacological importance of agmatine in the brain. Neurosci Biobehav Rev. 2012;36(1):502-19.
357. Satriano J, Cunard R, Peterson OW, Dousa T, Gabbai FB, Blantz RC. Effects on kidney filtration rate by agmatine requires activation of ryanodine channels for nitric oxide generation. Am J Physiol Renal Physiol. 2008;294(4):F795-800.
358. El-Kashef DH, El-Kenawi AE, Rahim MA, Suddek GM, Salem HA. Agmatine improves renal function in gentamicin-induced nephrotoxicity in rats. Can J Physiol Pharmacol. 2015:1-9.
359. Klatt P, Schmidt K, Lehner D, Glatter O, Bachinger HP, Mayer B. Structural analysis of porcine brain nitric oxide synthase reveals a role for tetrahydrobiopterin
350 Chapter 9 - References and L-arginine in the formation of an SDS-resistant dimer. EMBO J. 1995;14(15):3687-95.
360. Lin MI, Fulton D, Babbitt R, Fleming I, Busse R, Pritchard KA, Jr., et al. Phosphorylation of threonine 497 in endothelial nitric-oxide synthase coordinates the coupling of L-arginine metabolism to efficient nitric oxide production. J Biol Chem. 2003;278(45):44719-26.
361. Shu X, Keller TCt, Begandt D, Butcher JT, Biwer L, Keller AS, et al. Endothelial nitric oxide synthase in the microcirculation. Cell Mol Life Sci. 2015;72(23):4561-75.
362. Cosby K, Partovi KS, Crawford JH, Patel RP, Reiter CD, Martyr S, et al. Nitrite reduction to nitric oxide by deoxyhemoglobin vasodilates the human circulation. Nat Med. 2003;9(12):1498-505.
363. Green DJ, Dawson EA, Groenewoud HM, Jones H, Thijssen DH. Is flow- mediated dilation nitric oxide mediated?: A meta-analysis. Hypertension. 2014;63(2):376-82.
364. Bogdan C. Nitric oxide synthase in innate and adaptive immunity: an update. Trends Immunol. 2015;36(3):161-78.
365. Wan Saudi WS, Halim MA, Rudholm-Feldreich T, Gillberg L, Rosenqvist E, Tengholm A, et al. Neuropeptide S inhibits gastrointestinal motility and increases mucosal permeability through nitric oxide. Am J Physiol Gastrointest Liver Physiol. 2015;309(8):G625-34.
366. Radomski MW, Palmer RM, Moncada S. Characterization of the L- arginine:nitric oxide pathway in human platelets. Br J Pharmacol. 1990;101(2):325-8.
367. Perez-Zoghbi JF, Bai Y, Sanderson MJ. Nitric oxide induces airway smooth muscle cell relaxation by decreasing the frequency of agonist-induced Ca2+ oscillations. J Gen Physiol. 2010;135(3):247-59.
351 Chapter 9 - References
368. Diesen DL, Kuo PC. Nitric oxide and redox regulation in the liver: part II. Redox biology in pathologic hepatocytes and implications for intervention. J Surg Res. 2011;167(1):96-112.
369. De Palma C, Morisi F, Pambianco S, Assi E, Touvier T, Russo S, et al. Deficient nitric oxide signalling impairs skeletal muscle growth and performance: involvement of mitochondrial dysregulation. Skelet Muscle. 2014;4(1):22.
370. Percival JM, Anderson KN, Huang P, Adams ME, Froehner SC. Golgi and sarcolemmal neuronal NOS differentially regulate contraction-induced fatigue and vasoconstriction in exercising mouse skeletal muscle. J Clin Invest. 2010;120(3):816- 26.
371. Chien SJ, Lin KM, Kuo HC, Huang CF, Lin YJ, Huang LT, et al. Two different approaches to restore renal nitric oxide and prevent hypertension in young spontaneously hypertensive rats: l-citrulline and nitrate. Transl Res. 2014;163(1):43- 52.
372. Baylis C. Nitric oxide deficiency in chronic kidney disease. Am J Physiol Renal Physiol. 2008;294(1):F1-9.
373. Arnold WP, Mittal CK, Katsuki S, Murad F. Nitric oxide activates guanylate cyclase and increases guanosine 3':5'-cyclic monophosphate levels in various tissue preparations. Proc Natl Acad Sci U S A. 1977;74(8):3203-7.
374. Jaffrey SR, Erdjument-Bromage H, Ferris CD, Tempst P, Snyder SH. Protein S-nitrosylation: a physiological signal for neuronal nitric oxide. Nat Cell Biol. 2001;3(2):193-7.
375. Nott A, Riccio A. Nitric oxide-mediated epigenetic mechanisms in developing neurons. Cell Cycle. 2009;8(5):725-30.
376. Riccio A. Dynamic epigenetic regulation in neurons: enzymes, stimuli and signaling pathways. Nat Neurosci. 2010;13(11):1330-7.
377. Garthwaite J. Concepts of neural nitric oxide-mediated transmission. Eur J Neurosci. 2008;27(11):2783-802.
352 Chapter 9 - References
378. Peunova N, Enikolopov G. Nitric oxide triggers a switch to growth arrest during differentiation of neuronal cells. Nature. 1995;375(6526):68-73.
379. Lameu C, Trujillo CA, Schwindt TT, Negraes PD, Pillat MM, Morais KL, et al. Interactions between the NO-citrulline cycle and brain-derived neurotrophic factor in differentiation of neural stem cells. J Biol Chem. 2012;287(35):29690-701.
380. Bredt DS, Snyder SH. Transient nitric oxide synthase neurons in embryonic cerebral cortical plate, sensory ganglia, and olfactory epithelium. Neuron. 1994;13(2):301-13.
381. Nott A, Nitarska J, Veenvliet JV, Schacke S, Derijck AA, Sirko P, et al. S- nitrosylation of HDAC2 regulates the expression of the chromatin-remodeling factor Brm during radial neuron migration. Proc Natl Acad Sci U S A. 2013;110(8):3113-8.
382. Riccio A, Alvania RS, Lonze BE, Ramanan N, Kim T, Huang Y, et al. A nitric oxide signaling pathway controls CREB-mediated gene expression in neurons. Mol Cell. 2006;21(2):283-94.
383. Nott A, Watson PM, Robinson JD, Crepaldi L, Riccio A. S-Nitrosylation of histone deacetylase 2 induces chromatin remodelling in neurons. Nature. 2008;455(7211):411-5.
384. Dulac C. Brain function and chromatin plasticity. Nature. 2010;465(7299):728- 35.
385. Guan JS, Haggarty SJ, Giacometti E, Dannenberg JH, Joseph N, Gao J, et al. HDAC2 negatively regulates memory formation and synaptic plasticity. Nature. 2009;459(7243):55-60.
386. Kang JJ, Shu L, Park JL, Shayman JA, Bodary PF. Endothelial nitric oxide synthase uncoupling and microvascular dysfunction in the mesentery of mice deficient in alpha-galactosidase A. Am J Physiol Gastrointest Liver Physiol. 2014;306(2):G140-6.
387. Vallance P, Charles I. Nitric oxide as an antimicrobial agent: does NO always mean NO? [ comment]. Gut. 1998;42(3):313-4.
353 Chapter 9 - References
388. Vollmar B, Menger MD. The hepatic microcirculation: mechanistic contributions and therapeutic targets in liver injury and repair. Physiol Rev. 2009;89(4):1269-339.
389. Siriussawakul A, Zaky A, Lang JD. Role of nitric oxide in hepatic ischemia- reperfusion injury. World J Gastroenterol. 2010;16(48):6079-86.
390. Newnham T, Hardikar W, Allen K, Wellard RM, Hamilton C, Angus P, et al. Liver transplantation for argininosuccinic aciduria: clinical, biochemical, and metabolic outcome. Liver transplant. 2008;14(1):41-5.
391. Yankol Y, Mecit N, Kanmaz T, Acarli K, Kalayoglu M. Argininosuccinic Aciduria-A Rare Indication for Liver Transplant: Report of Two Cases. Exp Clin Transplant. 2016.
392. McHugh D, Cameron CA, Abdenur JE, Abdulrahman M, Adair O, Al Nuaimi SA, et al. Clinical validation of cutoff target ranges in newborn screening of metabolic disorders by tandem mass spectrometry: a worldwide collaborative project. Genet Med. 2011;13(3):230-54.
393. Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM. Electrospray ionization for mass spectrometry of large biomolecules. Science. 1989;246(4926):64- 71.
394. Kebarle P, Verkerk UH. Electrospray: from ions in solution to ions in the gas phase, what we know now. Mass Spectrom Rev. 2009;28(6):898-917.
395. Pitt JJ. Principles and applications of liquid chromatography-mass spectrometry in clinical biochemistry. Clin Biochem Rev. 2009;30(1):19-34.
396. Ratner S, Kunkemueller M. Separation and properties of argininosuccinate and its two anhydrides and their detection in biological materials. Biochemistry. 1966;5(6):1821-32.
397. Lee CR, Pollitt RJ. New derivatives of argininosuccinic acid in the urine of a patient with argininosuccinicaciduria. Biochem J. 1972;126(1):79-87.
354 Chapter 9 - References
398. Burns SP, Iles RA. An investigation of argininosuccinic acid anhydrides in argininosuccinic acid lyase deficiency by 1H-NMR spectroscopy. Clin Chim Acta. 1993;221(1-2):1-13.
399. Debenedetti E. [Michele Tsvet, the discoverer of chromatography]. Minerva Med. 1956;47(25):536-8.
400. Wolfe R, editor. Chromatography, in Radioactive and stable isotope tracers in biomedicine, Willey-liss editions: NY, USA; 1992.
401. Piraud M, Vianey-Saban C, Petritis K, Elfakir C, Steghens JP, Bouchu D. Ion- pairing reversed-phase liquid chromatography/electrospray ionization mass spectrometric analysis of 76 underivatized amino acids of biological interest: a new tool for the diagnosis of inherited disorders of amino acid metabolism. Rapid Commun Mass Spectrom. 2005;19(12):1587-602.
402. Wuyts B, Stove V, Goossens L. Critical sample pretreatment in monitoring dried blood spot citrulline. Clin Chim Acta. 2007;386(1-2):105-9.
403. Footitt EJ, Clayton PT, Mills K, Heales SJ, Neergheen V, Oppenheim M, et al. Measurement of plasma B6 vitamer profiles in children with inborn errors of vitamin B6 metabolism using an LC-MS/MS method. J J Inherit Metab Dis. 2013;36(1):139- 45.
404. Wang C, Zhu H, Zhang W, Song F, Liu Z, Liu S. Second-tier test for quantification of underivatized amino acids in dry blood spot for metabolic diseases in newborn screening. Amino Acids. 2013;44(2):661-71.
405. Canepa A, Filho JC, Gutierrez A, Carrea A, Forsberg AM, Nilsson E, et al. Free amino acids in plasma, red blood cells, polymorphonuclear leukocytes, and muscle in normal and uraemic children. Nephrol Dial Transplant. 2002;17(3):413-21.
406. Schulze A, Lindner M, Kohlmuller D, Olgemoller K, Mayatepek E, Hoffmann GF. Expanded newborn screening for inborn errors of metabolism by electrospray ionization-tandem mass spectrometry: results, outcome, and implications. Pediatrics. 2003;111(6 Pt 1):1399-406.
355 Chapter 9 - References
407. Serra A, Gallart-Palau X, Wei J, Sze SK. Characterization of Glutamine Deamidation by Long-Length Electrostatic Repulsion-Hydrophilic Interaction Chromatography-Tandem Mass Spectrometry (LERLIC-MS/MS) in Shotgun Proteomics. Anal Chem. 2016;88(21):10573-82.
408. Darmaun D, Manary MJ, Matthews DE. A method for measuring both glutamine and glutamate levels and stable isotopic enrichments. Anal Biochem. 1985;147(1):92-102.
409. Wang C, Zhang W, Song F, Liu Z, Liu S. A simple method for the analysis by MS/MS of underivatized amino acids on dry blood spots from newborn screening. Amino Acids. 2012;42(5):1889-95.
410. Moretti F, Birarelli M, Carducci C, Pontecorvi A, Antonozzi I. Simultaneous high-performance liquid chromatographic determination of amino acids in a dried blood spot as a neonatal screening test. J Chromatogr. 1990;511:131-6.
411. Guideline on bioanalytical method validation, EMEA/CHMP/EWP/192217/2009
412. Guidance for industry - bioanalytical method validation. http://www.fda.gov/.downloads/Drugs/GuidanceComplianceRegulatoryInformation/ Guidances/ucm070107.pdf.
413. Zytkovicz TH, Fitzgerald EF, Marsden D, Larson CA, Shih VE, Johnson DM, et al. Tandem mass spectrometric analysis for amino, organic, and fatty acid disorders in newborn dried blood spots: a two-year summary from the New England Newborn Screening Program. Clin Chem. 2001;47(11):1945-55.
414. Kruve A, Rebane R, Kipper K, Oldekop ML, Evard H, Herodes K, et al. Tutorial review on validation of liquid chromatography-mass spectrometry methods: part I. Anal Chim Acta. 2015;870:29-44.
415. Kruve A, Leito I, Herodes K. Combating matrix effects in LC/ESI/MS: the extrapolative dilution approach. Anal Chim Acta. 2009;651(1):75-80.
356 Chapter 9 - References
416. Bush JA, Wu G, Suryawan A, Nguyen HV, Davis TA. Somatotropin-induced amino acid conservation in pigs involves differential regulation of liver and gut urea cycle enzyme activity. J Nutr. 2002;132(1):59-67.
417. Moroz LL, Chen D, Gillette MU, Gillette R. Nitric oxide synthase activity in the molluscan CNS. J Neurochem. 1996;66(2):873-6.
418. Mukarram Ali Baig M, Habibullah CM, Swamy M, Hassan I, Taher uz Z, Ayesha Q, et al. Studies on urea cycle enzyme levels in the human fetal liver at different gestational ages. Pediatr Res. 1992;31(2):143-5.
419. Revel JP, Napolitano L, Fawcett DW. Identification of glycogen in electron micrographs of thin tissue sections. J Biophys Biochem Cytol. 1960;8:575-89.
420. Bozzola JJ. Conventional Specimen Preparation Techniques for Transmission Electron Microscopy of Cultured Cells In: J. Kuo © Humana Press Inc. T, NJ, editor. Methods in Molecular Biology Electron Microscopy: Methods and Protocols, Third Edition 3692014. p. 2.
421. Cryan JF, Mombereau C, Vassout A. The tail suspension test as a model for assessing antidepressant activity: review of pharmacological and genetic studies in mice. Neurosci Biobehav Rev. 2005;29(4-5):571-625.
422. Salter M, Duffy C, Garthwaite J, Strijbos PJ. Ex vivo measurement of brain tissue nitrite and nitrate accurately reflects nitric oxide synthase activity in vivo. J Neurochem. 1996;66(4):1683-90.
423. Garthwaite J. New insight into the functioning of nitric oxide-receptive guanylyl cyclase: physiological and pharmacological implications. Mol Cell Biochem. 2010;334(1-2):221-32.
424. Ischiropoulos H, Zhu L, Chen J, Tsai M, Martin JC, Smith CD, et al. Peroxynitrite-mediated tyrosine nitration catalyzed by superoxide dismutase. Arch Biochem Biophys. 1992;298(2):431-7.
357 Chapter 9 - References
425. Heales SJ, Bolanos JP. Impairment of brain mitochondrial function by reactive nitrogen species: the role of glutathione in dictating susceptibility. Neurochem Int. 2002;40(6):469-74.
426. Li C, Buettger C, Kwagh J, Matter A, Daikhin Y, Nissim IB, et al. A signaling role of glutamine in insulin secretion. J Biol Chem. 2004;279(14):13393-401.
427. Modi H, Cornu M, Thorens B. Glutamine stimulates biosynthesis and secretion of insulin-like growth factor 2 (IGF2), an autocrine regulator of beta cell mass and function. J Biol Chem. 2014;289(46):31972-82.
428. Iancu TC, Manov I. Electron Microscopy of Liver Biopsies, Liver Biopsy. In: HirokazuTakahashi D, editor.2011.
429. Reye RD, Morgan G, Baral J. Encephalopathy and Fatty Degeneration of the Viscera. A Disease Entity in Childhood. Lancet. 1963;2(7311):749-52.
430. Villanueva C, Giulivi C. Subcellular and cellular locations of nitric oxide synthase isoforms as determinants of health and disease. Free Radic Biol Med. 2010;49(3):307-16.
431. Zhou L, Zhu DY. Neuronal nitric oxide synthase: structure, subcellular localization, regulation, and clinical implications. Nitric Oxide. 2009;20(4):223-30.
432. Heneka MT, Feinstein DL. Expression and function of inducible nitric oxide synthase in neurons. J Neuroimmunol. 2001;114(1-2):8-18.
433. del Zoppo G, Ginis I, Hallenbeck JM, Iadecola C, Wang X, Feuerstein GZ. Inflammation and stroke: putative role for cytokines, adhesion molecules and iNOS in brain response to ischemia. Brain Pathol. 2000;10(1):95-112.
434. Willenborg DO, Staykova MA, Cowden WB. Our shifting understanding of the role of nitric oxide in autoimmune encephalomyelitis: a review. J Neuroimmunol. 1999;100(1-2):21-35.
435. Almer G, Vukosavic S, Romero N, Przedborski S. Inducible nitric oxide synthase up-regulation in a transgenic mouse model of familial amyotrophic lateral sclerosis. J Neurochem. 1999;72(6):2415-25.
358 Chapter 9 - References
436. Wallace MN, Geddes JG, Farquhar DA, Masson MR. Nitric oxide synthase in reactive astrocytes adjacent to beta-amyloid plaques. Exp Neurol. 1997;144(2):266- 72.
437. Heneka MT, Wiesinger H, Dumitrescu-Ozimek L, Riederer P, Feinstein DL, Klockgether T. Neuronal and glial coexpression of argininosuccinate synthetase and inducible nitric oxide synthase in Alzheimer disease. J Neuropathol Exp Neurol. 2001;60(9):906-16.
438. Inagaki K, Piao C, Kotchey NM, Wu X, Nakai H. Frequency and spectrum of genomic integration of recombinant adeno-associated virus serotype 8 vector in neonatal mouse liver. J Virol. 2008;82(19):9513-24.
439. Ratnakumari L, Qureshi IA, Butterworth RF, Marescau B, De Deyn PP. Arginine-related guanidino compounds and nitric oxide synthase in the brain of ornithine transcarbamylase deficient spf mutant mouse: effect of metabolic arginine deficiency. Neurosci Lett. 1996;215(3):153-6.
440. Scaglia F, Lee B. Clinical, biochemical, and molecular spectrum of hyperargininemia due to arginase I deficiency. Am J Med Genet C Semin Med Genet. 2006;142C(2):113-20.
441. Hocquemiller M, Giersch L, Audrain M, Parker S, Cartier N. Adeno-Associated Virus-Based Gene Therapy for CNS Diseases. Hum Gene Ther. 2016;27(7):478-96.
442. Zincarelli C, Soltys S, Rengo G, Rabinowitz JE. Analysis of AAV serotypes 1-9 mediated gene expression and tropism in mice after systemic injection. Mol Ther. 2008;16(6):1073-80.
443. Chakrabarty P, Rosario A, Cruz P, Siemienski Z, Ceballos-Diaz C, Crosby K, et al. Capsid serotype and timing of injection determines AAV transduction in the neonatal mice brain. PLoS One. 2013;8(6):e67680.
444. Holehonnur R, Luong JA, Chaturvedi D, Ho A, Lella SK, Hosek MP, et al. Adeno-associated viral serotypes produce differing titers and differentially transduce neurons within the rat basal and lateral amygdala. BMC Neurosci. 2014;15:28.
359 Chapter 9 - References
445. Broekman ML, Comer LA, Hyman BT, Sena-Esteves M. Adeno-associated virus vectors serotyped with AAV8 capsid are more efficient than AAV-1 or -2 serotypes for widespread gene delivery to the neonatal mouse brain. Neuroscience. 2006;138(2):501-10.
446. Klein RL, Dayton RD, Tatom JB, Henderson KM, Henning PP. AAV8, 9, Rh10, Rh43 vector gene transfer in the rat brain: effects of serotype, promoter and purification method. Mol Ther. 2008;16(1):89-96.
447. Gray SJ, Blake BL, Criswell HE, Nicolson SC, Samulski RJ, McCown TJ, et al. Directed evolution of a novel adeno-associated virus (AAV) vector that crosses the seizure-compromised blood-brain barrier (BBB). Mol Ther. 2010;18(3):570-8.
448. Nathwani AC, Tuddenham EG, Rangarajan S, Rosales C, McIntosh J, Linch DC, et al. Adenovirus-associated virus vector-mediated gene transfer in hemophilia B. N Engl J Med. 2011;365(25):2357-65.
449. Li C, Narkbunnam N, Samulski RJ, Asokan A, Hu G, Jacobson LJ, et al. Neutralizing antibodies against adeno-associated virus examined prospectively in pediatric patients with hemophilia. Gene Ther. 2012;19(3):288-94.
450. Mattar CN, Wong AM, Hoefer K, Alonso-Ferrero ME, Buckley SM, Howe SJ, et al. Systemic gene delivery following intravenous administration of AAV9 to fetal and neonatal mice and late-gestation nonhuman primates. FASEB J. 2015.
451. Hlavaty J, Schittmayer M, Stracke A, Jandl G, Knapp E, Felber BK, et al. Effect of posttranscriptional regulatory elements on transgene expression and virus production in the context of retrovirus vectors. Virology. 2005;341(1):1-11.
452. Xu ZL, Mizuguchi H, Mayumi T, Hayakawa T. Woodchuck hepatitis virus post- transcriptional regulation element enhances transgene expression from adenovirus vectors. Biochim Biophys Acta. 2003;1621(3):266-71.
453. Loeb JE, Cordier WS, Harris ME, Weitzman MD, Hope TJ. Enhanced expression of transgenes from adeno-associated virus vectors with the woodchuck hepatitis virus posttranscriptional regulatory element: implications for gene therapy. Hum Gene Ther. 1999;10(14):2295-305.
360 Chapter 9 - References
454. Kingsman SM, Mitrophanous K, Olsen JC. Potential oncogene activity of the woodchuck hepatitis post-transcriptional regulatory element (WPRE). Gene Ther. 2005;12(1):3-4.
455. Choi JH, Yu NK, Baek GC, Bakes J, Seo D, Nam HJ, et al. Optimization of AAV expression cassettes to improve packaging capacity and transgene expression in neurons. Mol Brain. 2014;7:17.
456. Chen P, Tian J, Kovesdi I, Bruder JT. Promoters influence the kinetics of transgene expression following adenovector gene delivery. J Gene Med. 2008;10(2):123-31.
457. Matsubasa T, Takiguchi M, Matsuda I, Mori M. Rat argininosuccinate lyase promoter: the dyad-symmetric CCAAT box sequence CCAATTGG in the promoter is recognized by NF-Y. J Biochem. 1994;116(5):1044-55.
458. Zychlinski D, Schambach A, Modlich U, Maetzig T, Meyer J, Grassman E, et al. Physiological promoters reduce the genotoxic risk of integrating gene vectors. Mol Ther. 2008;16(4):718-25.
459. Kamikawa R, Brown MW, Nishimura Y, Sako Y, Heiss AA, Yubuki N, et al. Parallel re-modeling of EF-1alpha function: divergent EF-1alpha genes co-occur with EFL genes in diverse distantly related eukaryotes. BMC Evol Biol. 2013;13:131.
460. Khacho M, Mekhail K, Pilon-Larose K, Pause A, Cote J, Lee S. eEF1A is a novel component of the mammalian nuclear protein export machinery. Mol Biol Cell. 2008;19(12):5296-308.
461. Riis B, Rattan SI, Clark BF, Merrick WC. Eukaryotic protein elongation factors. Trends Biochem Sci. 1990;15(11):420-4.
462. Negrutskii BS, El'skaya AV. Eukaryotic translation elongation factor 1 alpha: structure, expression, functions, and possible role in aminoacyl-tRNA channeling. Prog Nucleic Acid Res Mol Biol. 1998;60:47-78.
361 Chapter 9 - References
463. Wakabayashi-Ito N, Nagata S. Characterization of the regulatory elements in the promoter of the human elongation factor-1 alpha gene. J Biol Chem. 1994;269(47):29831-7.
464. Xu L, Daly T, Gao C, Flotte TR, Song S, Byrne BJ, et al. CMV-beta-actin promoter directs higher expression from an adeno-associated viral vector in the liver than the cytomegalovirus or elongation factor 1 alpha promoter and results in therapeutic levels of human factor X in mice. Hum Gene Ther. 2001;12(5):563-73.
465. Nakai H, Herzog RW, Hagstrom JN, Walter J, Kung SH, Yang EY, et al. Adeno-associated viral vector-mediated gene transfer of human blood coagulation factor IX into mouse liver. Blood. 1998;91(12):4600-7.
466. Sferra TJ, Backstrom K, Wang C, Rennard R, Miller M, Hu Y. Widespread correction of lysosomal storage following intrahepatic injection of a recombinant adeno-associated virus in the adult MPS VII mouse. Mol Ther. 2004;10(3):478-91.
467. Qin JY, Zhang L, Clift KL, Hulur I, Xiang AP, Ren BZ, et al. Systematic comparison of constitutive promoters and the doxycycline-inducible promoter. PLoS One. 2010;5(5):e10611.
468. Varma N, Janic B, Ali M, Iskander A, Arbab A. Lentiviral Based Gene Transduction and Promoter Studies in Human Hematopoietic Stem Cells (hHSCs). J Stem Cells Regen Med. 2011;7(1):41-53.
469. Kim S, Kim GJ, Miyoshi H, Moon SH, Ahn SE, Lee JH, et al. Efficiency of the elongation factor-1alpha promoter in mammalian embryonic stem cells using lentiviral gene delivery systems. Stem Cells Dev. 2007;16(4):537-45.
470. Zhang F, Frost AR, Blundell MP, Bales O, Antoniou MN, Thrasher AJ. A ubiquitous chromatin opening element (UCOE) confers resistance to DNA methylation-mediated silencing of lentiviral vectors. Mol Ther. 2010;18(9):1640-9.
471. Hacein-Bey-Abina S, Pai SY, Gaspar HB, Armant M, Berry CC, Blanche S, et al. A modified gamma-retrovirus vector for X-linked severe combined immunodeficiency. N Engl J Med. 2014;371(15):1407-17.
362 Chapter 9 - References
472. Tsuchiya R, Yoshiki F, Kudo Y, Morita M. Cell type-selective expression of green fluorescent protein and the calcium indicating protein, yellow cameleon, in rat cortical primary cultures. Brain Res. 2002;956(2):221-9.
473. Brinster RL, Allen JM, Behringer RR, Gelinas RE, Palmiter RD. Introns increase transcriptional efficiency in transgenic mice. Proc Natl Acad Sci U S A. 1988;85(3):836-40.
474. Niu DK, Yang YF. Why eukaryotic cells use introns to enhance gene expression: splicing reduces transcription-associated mutagenesis by inhibiting topoisomerase I cutting activity. Biol Direct. 2011;6:24.
475. Powell SK, Rivera-Soto R, Gray SJ. Viral expression cassette elements to enhance transgene target specificity and expression in gene therapy. Discov Med. 2015;19(102):49-57.
476. Miyake N, Miyake K, Yamamoto M, Hirai Y, Shimada T. Global gene transfer into the CNS across the BBB after neonatal systemic delivery of single-stranded AAV vectors. Brain Res. 2011;1389:19-26.
477. Kim JY, Ash RT, Ceballos-Diaz C, Levites Y, Golde TE, Smirnakis SM, et al. Viral transduction of the neonatal brain delivers controllable genetic mosaicism for visualising and manipulating neuronal circuits in vivo. Eur J Neurosci. 2013;37(8):1203-20.
478. Inagaki K, Fuess S, Storm TA, Gibson GA, McTiernan CF, Kay MA, et al. Robust systemic transduction with AAV9 vectors in mice: efficient global cardiac gene transfer superior to that of AAV8. Mol Ther. 2006;14(1):45-53.
479. Nakai H, Fuess S, Storm TA, Muramatsu S, Nara Y, Kay MA. Unrestricted hepatocyte transduction with adeno-associated virus serotype 8 vectors in mice. J Virol. 2005;79(1):214-24.
480. Merkel SF, Andrews AM, Lutton EM, Mu D, Hudry E, Hyman BT, et al. Trafficking of adeno-associated virus vectors across a model of the blood-brain barrier; a comparative study of transcytosis and transduction using primary human brain endothelial cells. J Neurochem. 2017;140(2):216-30.
363 Chapter 9 - References
481. Foust KD, Nurre E, Montgomery CL, Hernandez A, Chan CM, Kaspar BK. Intravascular AAV9 preferentially targets neonatal neurons and adult astrocytes. Nature Biotech. 2009;27(1):59-65.
482. Mattar CN, Waddington SN, Biswas A, Johana N, Ng XW, Fisk AS, et al. Systemic delivery of scAAV9 in fetal macaques facilitates neuronal transduction of the central and peripheral nervous systems. Gene Ther. 2013;20(1):69-83.
483. Gray SJ, Matagne V, Bachaboina L, Yadav S, Ojeda SR, Samulski RJ. Preclinical differences of intravascular AAV9 delivery to neurons and glia: a comparative study of adult mice and nonhuman primates. Mol Ther. 2011;19(6):1058-69.
484. Boumahrou N, Andrei S, Miranda G, Henry C, Panthier JJ, Martin P, et al. The major protein fraction of mouse milk revisited using proven proteomic tools. J Physiol Pharmacol. 2009;60 Suppl 3:113-8.
485. Perez CJ, Jaubert J, Guenet JL, Barnhart KF, Ross-Inta CM, Quintanilla VC, et al. Two hypomorphic alleles of mouse Ass1 as a new animal model of citrullinemia type I and other hyperammonemic syndromes. Am J Pathol. 2010;177(4):1958-68.
486. Choi JH, Lee BH, Kim JH, Kim GH, Kim YM, Cho J, et al. Clinical outcomes and the mutation spectrum of the OTC gene in patients with ornithine transcarbamylase deficiency. J Hum Genet. 2015;60(9):501-7.
487. Miner NA, Koehler J, Greenaway L. Intraperitoneal injection of mice. Appl Microbiol. 1969;17(2):250-1.
488. Steward JP, Ornellas EP, Beernink KD, Northway WH. Errors in the technique of intraperitoneal injection of mice. Appl Microbiol. 1968;16(9):1418-9.
489. Turner PV, Brabb T, Pekow C, Vasbinder MA. Administration of substances to laboratory animals: routes of administration and factors to consider. J Am Assoc Lab Anim Sci. 2011;50(5):600-13.
364 Chapter 9 - References
490. Tuchman M, Jaleel N, Morizono H, Sheehy L, Lynch MG. Mutations and polymorphisms in the human ornithine transcarbamylase gene. Hum Mutat. 2002;19(2):93-107.
491. Li Y, Maher P, Schubert D. Requirement for cGMP in nerve cell death caused by glutathione depletion. J Cell Biol. 1997;139(5):1317-24.
492. Ditlevsen DK, Kohler LB, Berezin V, Bock E. Cyclic guanosine monophosphate signalling pathway plays a role in neural cell adhesion molecule- mediated neurite outgrowth and survival. J Neurosci Res. 2007;85(4):703-11.
493. Kohgami S, Ogata T, Morino T, Yamamoto H, Schubert P. Pharmacological shift of the ambiguous nitric oxide action from neurotoxicity to cyclic GMP-mediated protection. Neurol Res. 2010;32(9):938-44.
494. Ischiropoulos H. Protein tyrosine nitration--an update. Arch Biochem Biophys. 2009;484(2):117-21.
495. Souza JM, Daikhin E, Yudkoff M, Raman CS, Ischiropoulos H. Factors determining the selectivity of protein tyrosine nitration. Arch Biochem Biophys. 1999;371(2):169-78.
496. Chou SM, Wang HS, Taniguchi A. Role of SOD-1 and nitric oxide/cyclic GMP cascade on neurofilament aggregation in ALS/MND. J Neurol Sci. 1996;139 Suppl:16-26.
497. Heales SJ, Bolanos JP, Brand MP, Clark JB, Land JM. Mitochondrial damage: an important feature in a number of inborn errors of metabolism? J Inherit Metab Dis. 1996;19(2):140-2.
498. Gegg ME, Beltran B, Salas-Pino S, Bolanos JP, Clark JB, Moncada S, et al. Differential effect of nitric oxide on glutathione metabolism and mitochondrial function in astrocytes and neurones: implications for neuroprotection/neurodegeneration? J Neurochem. 2003;86(1):228-37.
499. Bolanos JP, Heales SJ, Land JM, Clark JB. Effect of peroxynitrite on the mitochondrial respiratory chain: differential susceptibility of neurones and astrocytes in primary culture. J Neurochem. 1995;64(5):1965-72. 365 Chapter 9 - References
500. Dormanns K, Brown RG, David T. The role of nitric oxide in neurovascular coupling. J Theor Biol. 2016;394:1-17.
501. Mayer B, Schrammel A, Klatt P, Koesling D, Schmidt K. Peroxynitrite-induced accumulation of cyclic GMP in endothelial cells and stimulation of purified soluble guanylyl cyclase. Dependence on glutathione and possible role of S-nitrosation. J Biol Chem. 1995;270(29):17355-60.
502. Wu M, Pritchard KA, Jr., Kaminski PM, Fayngersh RP, Hintze TH, Wolin MS. Involvement of nitric oxide and nitrosothiols in relaxation of pulmonary arteries to peroxynitrite. Am J Physiol. 1994;266(5 Pt 2):H2108-13.
503. Zou MH, Hou XY, Shi CM, Nagata D, Walsh K, Cohen RA. Modulation by peroxynitrite of Akt- and AMP-activated kinase-dependent Ser1179 phosphorylation of endothelial nitric oxide synthase. J Biol Chem. 2002;277(36):32552-7.
504. Moro MA, Darley-Usmar VM, Goodwin DA, Read NG, Zamora-Pino R, Feelisch M, et al. Paradoxical fate and biological action of peroxynitrite on human platelets. Proc Natl Acad Sci U S A. 1994;91(14):6702-6.
505. Good PF, Hsu A, Werner P, Perl DP, Olanow CW. Protein nitration in Parkinson's disease. J Neuropathol Exp Neurol. 1998;57(4):338-42.
506. Smith MA, Richey Harris PL, Sayre LM, Beckman JS, Perry G. Widespread peroxynitrite-mediated damage in Alzheimer's disease. J Neurosci. 1997;17(8):2653- 7.
507. Butterfield DA, Reed TT, Perluigi M, De Marco C, Coccia R, Keller JN, et al. Elevated levels of 3-nitrotyrosine in brain from subjects with amnestic mild cognitive impairment: implications for the role of nitration in the progression of Alzheimer's disease. Brain Res. 2007;1148:243-8.
508. Beal MF, Ferrante RJ, Browne SE, Matthews RT, Kowall NW, Brown RH, Jr. Increased 3-nitrotyrosine in both sporadic and familial amyotrophic lateral sclerosis. Ann Neurol. 1997;42(4):644-54.
366 Chapter 9 - References
509. Drechsel DA, Estevez AG, Barbeito L, Beckman JS. Nitric oxide-mediated oxidative damage and the progressive demise of motor neurons in ALS. Neurotox Res. 2012;22(4):251-64.
510. Kang YC, Kim PK, Choi BM, Chung HT, Ha KS, Kwon YG, et al. Regulation of programmed cell death in neuronal cells by nitric oxide. In Vivo. 2004;18(3):367-76.
511. Bobba A, Atlante A, Moro L, Calissano P, Marra E. Nitric oxide has dual opposite roles during early and late phases of apoptosis in cerebellar granule neurons. Apoptosis. 2007;12(9):1597-610.
512. Balez R, Ooi L. Getting to NO Alzheimer's Disease: Neuroprotection versus Neurotoxicity Mediated by Nitric Oxide. Oxid Med Cell Longev. 2016;2016:3806157.
513. Mazariegos G, Shneider B, Burton B, Fox IJ, Hadzic N, Kishnani P, et al. Liver transplantation for pediatric metabolic disease. Mol Genet Metab. 2014;111(4):418- 27.
514. Lisowski L, Dane AP, Chu K, Zhang Y, Cunningham SC, Wilson EM, et al. Selection and evaluation of clinically relevant AAV variants in a xenograft liver model. Nature. 2014;506(7488):382-6.
515. Castle MJ, Turunen HT, Vandenberghe LH, Wolfe JH. Controlling AAV Tropism in the Nervous System with Natural and Engineered Capsids. Methods Mol Biol. 2016;1382:133-49.
516. Choudhury SR, Fitzpatrick Z, Harris AF, Maitland SA, Ferreira JS, Zhang Y, et al. In Vivo Selection Yields AAV-B1 Capsid for Central Nervous System and Muscle Gene Therapy. Mol Ther. 2016;24(7):1247-57.
517. Wang CY, Wang S. Astrocytic expression of transgene in the rat brain mediated by baculovirus vectors containing an astrocyte-specific promoter. Gene Ther. 2006;13(20):1447-56.
518. Pascual-Lucas M, Viana da Silva S, Di Scala M, Garcia-Barroso C, Gonzalez- Aseguinolaza G, Mulle C, et al. Insulin-like growth factor 2 reverses memory and synaptic deficits in APP transgenic mice. EMBO Mol Med. 2014;6(10):1246-62.
367 Chapter 9 - References
519. Jung SC, Park ES, Choi EN, Kim CH, Kim SJ, Jin DK. Characterization of a novel mucopolysaccharidosis type II mouse model and recombinant AAV2/8 vector- mediated gene therapy. Mol Cells. 2010;30(1):13-8.
520. Chandler RJ, Sands M, Venditti CP. rAAV integration and genotoxicity: insights from animal models. Hum Gene Ther. 2017.
521. Gaveriaux-Ruff C, Kieffer BL. Conditional gene targeting in the mouse nervous system: Insights into brain function and diseases. Pharmacol Ther. 2007;113(3):619- 34.
522. Thyagarajan B, Guimaraes MJ, Groth AC, Calos MP. Mammalian genomes contain active recombinase recognition sites. Gene. 2000;244(1-2):47-54.
523. Zhan X, Cao M, Yoo AS, Zhang Z, Chen L, Crabtree GR, et al. Generation of BAF53b-Cre transgenic mice with pan-neuronal Cre activities. Genesis. 2015;53(7):440-8.
524. Loonstra A, Vooijs M, Beverloo HB, Allak BA, van Drunen E, Kanaar R, et al. Growth inhibition and DNA damage induced by Cre recombinase in mammalian cells. Proc Natl Acad Sci U S A. 2001;98(16):9209-14.
525. Forni PE, Scuoppo C, Imayoshi I, Taulli R, Dastru W, Sala V, et al. High levels of Cre expression in neuronal progenitors cause defects in brain development leading to microencephaly and hydrocephaly. J Neurosci. 2006;26(37):9593-602.
526. Schuster DJ, Dykstra JA, Riedl MS, Kitto KF, Belur LR, McIvor RS, et al. Biodistribution of adeno-associated virus serotype 9 (AAV9) vector after intrathecal and intravenous delivery in mouse. Front Neuroanat. 2014;8:42.
527. Haurigot V, Marco S, Ribera A, Garcia M, Ruzo A, Villacampa P, et al. Whole body correction of mucopolysaccharidosis IIIA by intracerebrospinal fluid gene therapy. J Clin Invest. 2013.
528. Samaranch L, Salegio EA, San Sebastian W, Kells AP, Foust KD, Bringas JR, et al. Adeno-associated virus serotype 9 transduction in the central nervous system of nonhuman primates. Hum Gene Ther. 2012;23(4):382-9.
368 Chapter 9 - References
529. McLean JR, Smith GA, Rocha EM, Hayes MA, Beagan JA, Hallett PJ, et al. Widespread neuron-specific transgene expression in brain and spinal cord following synapsin promoter-driven AAV9 neonatal intracerebroventricular injection. Neurosci Lett. 2014;576:73-8.
530. Lukashchuk V, Lewis KE, Coldicott I, Grierson AJ, Azzouz M. AAV9-mediated central nervous system-targeted gene delivery via cisterna magna route in mice. Mol Ther Methods Clin Dev. 2016;3:15055.
531. Ng J, Rahim AA, Kurian M, Waddington S. The synapsin promoter drives expression in liver and other viscera following neontala AAV9 injection. Hum Gene Ther. 2015;26:A25.
369 Chapter 10 - Appendices
10. APPENDICES
10.1 Asl gene therapy with repeated injections in AslNeo/Neo mice: preliminary results
10.1.1 Design of the experiment
Experiments with single systemic injection administered in adult or neonatal AslNeo/Neo mice have demonstrated a therapeutic effect with a dramatic improvement of the liver and the brain phenotypes as shown in Chapter 7. Another experiment was considered to evaluate the benefit of combining both injections at birth and in adulthood. It has been suggested that neonatal mice in their first week of life have immunotolerance to injection of AAV vector and do not generate a humoral immune response (275). This finding allows a second injection of AAV-mediated gene therapy with the same capsid at a later stage. The benefit of repeated injections of gene therapy will allow a protection of the brain by the AAV vector and a transient correction of the urea cycle for the first weeks of life after the neonatal injection. Then the rapid growth of the liver will generate a progressive loss of episomal transgenes and a decrease efficacy on the liver-based urea cycle. To avoid brain damage caused by high ammonia levels, the need of a supportive treatment with sodium benzoate and L-arginine injections and protein-restricted diet (See Section 7.2) was transiently required. Then a second injection of gene therapy was performed at adulthood when the liver had reached its adult mass providing a long-term correction of the urea cycle (Figure 108).
Six AslNeo/Neo mice were injected intravenously within the first 24 hours of life via the superficial temporal vein with the AAV8.EFS.mAsl.WPRE vector (Described in Section 6.2) with a dose of 3.2x10e11 vector genomes (vg) per mouse in a 20 µL volume. These mice received a second injection of the same vector at day 30 of life
370 Chapter 10 - Appendices delivered intraperitoneally at a dose of 1.85x10e12 vg per mouse in a 50 µL volume. Vector injections were performed by Dr Simon Waddington.
Nine wild-type and 16 untreated AslNeo/Neo mice were used as controls for the experiment. AslNeo/Neo mice not injected with gene therapy did not receive any supportive treatment.
Intravenous Intraperitoneal gene therapy gene therapy
Intraperitoneal Sodium benzoate + Arginine
D0 D15 D30 D31 D50
Protein restricted diet
Figure 108. Experimental design for repeated injections of Asl gene therapy in AslNeo/Neo mice
AslNeo/Neo mice injected with Asl gene therapy received a first peripheral intravenous injection within the first 24 hours of life and a second intraperitoneal injection at day 30. Daily intraperitoneal injections of sodium benzoate (100 mg/kg/d) and arginine (1 g/kg/d) were performed from day 15 to day 31 and a protein-restricted diet from day 15 to day 50 as presented in Section 7.2.
10.1.2 Preliminary results
All AslNeo/Neo mice, which received 2 sequential injections of Asl gene therapy at birth and day 30, survived until 10 weeks of age when the experiment was terminated, whereas no untreated AslNeo/Neo mice survived past 5 weeks of age (Log rank test, p=0.0001) (Figure 109). The gross phenotype showed a sustained improvement of the growth and the fur (Figures 110 and 111A), with
371 Chapter 10 - Appendices a growth velocity similar to the wild-type littermates during the first 3 weeks of life, then showing a mild persisting benefit over untreated surviving AslNeo/Neo mice. The second injection caused an increase in the growth velocity compared to wild-type mice during the 10 days following the second injection of Asl gene therapy (Figure 111B).
100
75
50
(%) Survival 25
0 0 20 40 60 80 Days
Untreated AslNeo/Neo D0+D30 injected AslNeo/Neo
Figure 109. Survival curve of AslNeo/Neo mice receiving 2 sequential injections of Asl gene therapy at birth and D30
Kaplan-Meyer survival curve of AslNeo/Neo mice receiving repeated injectiosn of Asl gene therapy or untreated (n=6 and 11, respectively).
Hepatomegaly assessed by the liver/body weight ratio was normalised in 4 (out of 6) AslNeo/Neo mice treated with Asl gene therapy but without significant improvement compared to untreated AslNeo/Neo mice (One-way ANOVA with Dunett’s post-test compared to wild-types, p<0.01 and p<0.05, respectively) (Figure 112A). The brain/body weight ratio was elevated in untreated AslNeo/Neo mice but normalised in all gene therapy-injected AslNeo/Neo mice (One-way ANOVA with Dunett’s post-test compared to wild-types, p<0.001 and p>0.05, respectively) (Figure 112B).
372 Chapter 10 - Appendices
A B C
WT Untr GT GT GT WT
D
GT
WT
Figure 110. Macroscopic aspect of gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice
(A) Images of wild-type, and gene therapy-injected AslNeo/Neo mice at day 20 of life. (B, C) Images of wild-type, untreated and gene therapy-injected AslNeo/Neo mice at day 30 of life, (D) Images of wild-type, and gene therapy-injected AslNeo/Neo mice) at day 60 of life. GT: gene therapy-injected AslNeo/Neo mice; Untr: untreated AslNeo/Neo mice; WT: Wild-type.
373 Chapter 10 - Appendices
A 30 B 1.5
1.0 20 0.5
10 0.0
Weight (grams) Weight -0.5 0 5 10 15 20 25 30 35 40 0 (g/d) velocity Growth Time (Days) 0 15 30 45 60 -1.0 Time (Days) Neo/Neo WT Untreated AslNeo/Neo WT Untreated Asl D0+D30 injected AslNeo/Neo Neo/Neo D0+D30 injected Asl
Figure 111. Growth of gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice
(A) Mean growth of gene therapy-injected AslNeo/Neo mice (n=6) compared to WT (n=9) and untreated AslNeo/Neo mice (n=16) over 10 weeks of life; (B) Growth velocity during the first 45 days of life in all groups. Horizontal line displays the mean ± standard deviation. Blue arrows indicate injections of gene therapy at day 0 and 30. WT: Wild-type.
A B * ns 0.15 0.100 ** *** 0.075 0.10 0.050 0.05 0.025 Brain / Body weight ratio / weight Body Brain Liver / Body weight ratio 0.00 0.000
Neo/Neo WT Untreated Asl WT Untreated AslNeo/Neo
D0+D30 injected AslNeo/Neo D0+D30 injected AslNeo/Neo
Figure 112. Hepatomegaly and brain/body weight ratio in gene therapy- injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice
(A) Liver/body weight and (B) brain/body weight ratios of gene therapy-injected AslNeo/Neo mice (n=6) compared to wild-type (n=5-7) and untreated AslNeo/Neo mice (n=11). Horizontal lines display the mean ± standard deviation. One-way ANOVA
374 Chapter 10 - Appendices with Dunnett’s post-test compared to wild-types; ns: not significant, * p<0.05, ** p<0.01, *** p<0.001. WT: Wild-type.
Various biomarkers were assessed in wild-type, untreated and gene therapy-injected AslNeo/Neo mice at day 30, and in gene therapy-injected AslNeo/Neo mice at day 38. This allowed assessment of the benefit of the second injection of Asl gene therapy either just before the second injection and one week after (Figure 113). Plasma ammonia levels were elevated in untreated AslNeo/Neo mice but normalised in gene therapy- injected AslNeo/Neo mice before and after the second injection (One-way ANOVA with Dunett’s post-test compared to wild-types, p<0.001, p>0.05 and p>0.05, respectively) (Figure 113A). Arginine levels in dried blood spot were decreased in untreated AslNeo/Neo mice but normalised in gene therapy-injected AslNeo/Neo mice before and after the second injection (One-way ANOVA with Dunett’s post-test compared to wild-types, p<0.05, p>0.05 and p>0.05, respectively) (Figure 113B). Citrulline levels in dried blood spot were increased in untreated and in gene therapy-injected AslNeo/Neo mice before the second injection and normalised after the second injection of Asl gene therapy (One-way ANOVA with Dunett’s post-test compared to wild- types, p<0.001, p<0.001 and p>0.05, respectively) (Figure 113C). Argininosuccinic acid levels in dried blood spots were increased in untreated and in gene therapy- injected AslNeo/Neo mice before the second injection (although with reduced levels compared to untreated AslNeo/Neo mice) and were dramatically decreased after the second injection of Asl gene therapy (One-way ANOVA with Dunett’s post-test compared to wild-types, p<0.001, p<0.05 and p>0.05, respectively) (Figure 113D). Glutamine and glutamate levels in dried blood spot were increased in untreated and in gene therapy-injected AslNeo/Neo mice before the second injection (although with reduced levels compared to untreated AslNeo/Neo mice) and normalised after the second injection of Asl gene therapy (One-way ANOVA with Dunett’s post-test compared to wild-types, p<0.001, p<0.01 and p>0.05, respectively) (Figure 113E).
At time of culling at 10 weeks of life, ASL activity in liver showed significantly reduced ASL activities in untreated and in gene therapy-injected AslNeo/Neo mice with 13.3% and 56.5% of wild-type residual activities, respectively (One-way ANOVA with Bonferroni post-test compared to wild-types, p<0.001 and p<0.001, respectively). A
375 Chapter 10 - Appendices significant increase of ASL activity was observed between untreated and in gene therapy-injected AslNeo/Neo mice (One-way ANOVA with Bonferroni post-test compared to untreated AslNeo/Neo mice, p<0.001) (Figure 114).
Behavioural testing was performed in gene therapy-injected AslNeo/Neo mice before the second injection of Asl gene therapy at day 30 and one month after the second injection at 2 months of age. As few untreated AslNeo/Neo mice were alive at day 30 but none survived up to 2 months of age, data from wild-type and untreated AslNeo/Neo mice generated in previous experiments detailed in Chapters 5 and 7 (See Sections 5.3.1 & 7.7.1) were used as controls at both time points. A significantly reduced performance in the open field test was observed in untreated AslNeo/Neo mice at one and 2 months of life, which was normalised at both time points in the gene therapy- injected AslNeo/Neo mice (One-way ANOVA with Dunett’s post-test compared to wild- types, p<0.01 and p>0.05, respectively at one month of age ; p<0.001 and p>0.05, respectively at 2 months of age) (Figure 115A). With the rotarod testing, no significant difference was observed in untreated or gene therapy-injected AslNeo/Neo mice at one and 2 months of life (One-way ANOVA with Dunett’s post-test compared to wild-types) (Figure 115B).
Brains of different groups were stored but not processed and analysed due to time constraints.
376 Chapter 10 - Appendices
ns A B ns ns 800 ns 200 *** * 600 150 mol/L) mol/L) µ µ 400 100
200 50 Arginine ( Arginine Ammonia ( Ammonia
0 0
WT Untreated AslNeo/Neo WT Untreated AslNeo/Neo D0 injected AslNeo/Neo D0+D30 injected AslNeo/Neo D0 injected AslNeo/Neo D0+D30 injected AslNeo/Neo
ns ns
C *** D * 500 1500 *** ***
400
mol/L) 1000 mol/L) 300 µ (µ
200 500
Citrulline 100 anhydrides (
0 and acid Argininosuccinic 0
WT Untreated AslNeo/Neo WT Untreated AslNeo/Neo D0 injected AslNeo/Neo D0+D30 injected AslNeo/Neo D0 injected AslNeo/Neo D0+D30 injected AslNeo/Neo
E ns ** 1500 ***
1000 mol/L) µ ( 500
0 Plasma Glutamine + Glutamate + Glutamate Glutamine Plasma WT Untreated AslNeo/Neo D0 injected AslNeo/Neo D0+D30 injected AslNeo/Neo
Figure 113. Biomarkers of the urea cycle
Comparison of biomarkers in gene therapy-injected AslNeo/Neo mice (n=6) at day 30 just before the second injection (D0 injected AslNeo/Neo mice) and at day 40, one week after the second injection (D0+D30 injected AslNeo/Neo mice), with wild-type (n=6-9 and untreated (n=5-9) AslNeo/Neo mice at 30 days of life. (A) Plasma ammonia levels, (B) Dried blood spot arginine level, (C) Dried blood spot citrulline levels, (D) Dried blood spot argininosuccinic acid levels and (E) Dried blood spot glutamate and glutamine levels. Horizontal lines display the mean ± standard deviation. One-way 377 Chapter 10 - Appendices
ANOVA with Dunnett’s post-test compared to wild-types; ns: not significant, * p<0.05, ** p<0.01, *** p<0.001.
***
*** *** 400
300
200
100
0 WT Untreated AslNeo/Neo Argininosuccinate lyase activity(nmol/ng/min) lyase Argininosuccinate D0+D30 injected AslNeo/Neo
Figure 114. Argininosuccinate lyase activity in gene therapy-injected AslNeo/Neo mice compared to wild-type and untreated AslNeo/Neo mice
Argininosuccinate lyase activity in liver samples of wild-type (n=8), untreated (n=6), and gene therapy-injected AslNeo/Neo mice (n=6) at 10 weeks of life. Horizontal lines display the mean ± standard deviation. One-way ANOVA with Bonferronni post-test, *** p<0.001.
A ns B 800 ns 50 *** ns ns
40 600 ns ns 30 ** 400 20
Time (seconds) Time 200 10 Distance walked (m) 0 0 WT Untreated AslNeo/Neo WT Untreated AslNeo/Neo D0+D30 injected AslNeo/Neo D0+D30 injected AslNeo/Neo Age Age 2 Months 1 Month 2 Months at testing 1 Month at testing
Figure 115. Behavioural testing
(A) Open field test and (B) accelerating rotarod performed in 1 and 2 month-old gene therapy-injected (n=3-6) and untreated (n=8-10) AslNeo/Neo mice and wild-type mice
378 Chapter 10 - Appendices
(n=20-22). As all untreated AslNeo/Neo mice had died by 35 days of life, data from wild- type and untreated AslNeo/Neo mice from previous experiments were used as controls. Horizontal lines display the mean ± standard deviation. (A, B) One-way ANOVA with Dunnett’s post test compared to wild-types: ns not significant, ** p<0.01, *** p<0.001.
10.1.3 Discussion
Although the absence of results related to brain samples prevent any assessment of the long-term efficacy of this repeated approach of Asl gene therapy on the neurological disease in AslNeo/Neo mice, this experiment has shown some interesting findings. This experiment was designed to assess the long-term benefit of a second Asl gene therapy injection on the urea cycle at adulthood. As already shown in Section 7.4, this experiment confirmed that a single neonatal gene therapy injection normalises plasma ammonia levels. However, the benefit of a second Asl gene therapy injection on the urea cycle is confirmed with a normalisation of various biomarkers such as citrulline, argininosuccinic acid and glutamate/glutamine levels in dried blood spots similarly to the effect of a single adult Asl gene therapy injection as observed in Section 7.4. Residual ASL activity in the liver at 10 weeks of age was 56.5% so closer to the theoretical residual activity of a heterozygote mouse which displays a gross phenotype undistinguishable from a wild-type mouse. Similarly human carriers of severe hASL mutations are asymptomatic. Liver ASL activity in heterozygote mice has not been measured but might be close to 65% of ASL activity (as AslNeo/Neo mice have a 13.5 to 14.5% residual liver ASL activity (See Sections 5.2.4 & 7.4.3). AslNeo/Neo mice having received repeated injections of Asl gene therapy do not display a full correction of the growth with a failure to thrive obvious from 3 weeks of age onwards (See Figure 111). This might be caused by an insufficient correction of the urea cycle after the first days following the neonatal gene therapy injection which would require an earlier second injection of gene therapy, e.g. at 20 days of life. Another explanation might be the non-correction of all extra-hepatic tissues and a persisting mild multi-organ disease which would prevent a full correction of the growth and the general phenotype.
379 Chapter 10 - Appendices
10.2 Development of a new model of brain conditional Asl knockout mouse: preliminary results
10.2.1 The Cre-Lox system in the AslNeo/Neo mouse
The Cre-LoxP system is based on the specific catalytic activity of Cre recombinase, a site-specific DNA recombinase isolated from bacteriophage P1, which catalyses genomic DNA between two specific 34 base pair-long LoxP recognition sites (521). Whilst LoxP sites are naturally present in the mammalian genome (522), these sites can also be inserted by homologous recombination in embryonic stem cells in introns flanking an essential exon of a gene of interest. LoxP sites do not disturb gene transcription and the “floxed” animal (e.g. a mouse) will have a wild-type phenotype.
A second transgenic animal expressing Cre recombinase activity is required to breed with the floxed animal, to excise the genomic DNA flanked by LoxP sites in order for a phenotype to be displayed in the offspring. A Cre mouse can be created by pronuclear injection of the Cre gene under the transcriptional control of a promoter of interest, which can display restricted Cre expression in some tissues or cell-types. The Cre transgene integrates randomly in the genome, and after characterisation of the Cre expression in the desired restricted tissue or cells, the Cre mouse can be bred with the floxed mouse.
In the AslNeo/Neo mouse, LoxP sites are located in introns 6 and 9 flanking exons 7, 8 and 9 of mAsl (Figure 116).
The residual brain ASL activity in AslNeo/Neo mice is 14.3% of wild-type activity (See Section 5.3.2). AslNeo/Neo mice display a nitrosative/oxidative stress in the brain that is restricted to neurons (See Section 5.3.2), which is corrected by gene therapy which predominantly targets neurons in the brain (See Section 7.6). We hypothesised that neuronal knockout of ASL activity in the brain of AslNeo/Neo mice would display a more severe neurological phenotype and affect survival, behaviour and nitric oxide metabolism.
380 Chapter 10 - Appendices
5’ 3’ 1 6 7 9 10 17
Exon LoxP site Neomycin casse4e
Figure 116. Asl gene construct used to create the AslNeo/Neo mouse
The mAsl gene encodes 17 exons. Two LoxP sites are located in introns 6 and 9, respectively. A neomycin cassette is inserted in intron 9, upstream of the second LoxP site. The AslNeo/Neo knockdown mouse has 14% residual Asl activity compared to that of wild-type mice. In the presence of Cre recombinase, 3 exons are deleted and cause a knockout of Asl as reported previously (339).
Generation of a conditional neuronal ASL knockout mouse model by breeding the AslNeo/Neo mouse with a Cre mouse line, which displays a restricted Cre recombinase activity in cerebral neurons such as the BAF53b-Cre transgenic mouse (523), was considered to be too time-consuming to be investigated in the last year of this PhD. An alternative approach was therefore investigated i.e. delivery of the Cre- recombinase via an AAV vector injected intracranially. AAV9, a neurotropic AAV serotype, and Cre recombinase expression transcriptionally controlled by a neuronal promoter (human synapsin, hSyn) were used. An AAV9.hSyn.HI.eGFP- Cre.WPRE.SV40 (HI: chimeric intron from Promega, Madison, WI, USA ; SV40 : Simian virus 40 late polyadenylation tail) (AAV9.Cre) was bought from the Penn Vector Core, University of Pennsylvania (Philadelphia, PA, USA). In addition, the AAV8.EFS.SV40i.GFP.WPRE.SV40LpA vector (AAV8.mAsl), described in Section 6.3.1, was injected in one experimental group to treat the neuronal ASL knockout mouse model.
The experimental design is represented in Figure 117 with 5 different groups of either wild-type or AslNeo/Neo mice. Mice were either uninjected, or injected with the AAV9.Cre vector or with a combination of the AAV9.Cre and AAV8.mAsl vector.
381 Chapter 10 - Appendices
AAV9.Cre and AAV8.mAsl were injected intracranially and intravenously (via the superficial temporal vein as described in Section 6.3.1), respectively. All vector injections were performed by Dr Simon Waddington.
WT Cre-WT AslNeo/Neo Cre-AslNeo/Neo mAsl-Cre-AslNeo/Neo
Figure 117. Experimental design for characterisation of a conditional neuronal Asl knock-out mouse model
Five experimental groups (n=3-4 animals per group) were considered. Wild-type mice were either uninjected (WT), injected with the AAV9.Cre vector (Cre-WT). AslNeo/Neo mice were either uninjected (AslNeo/Neo), injected with the AAV9.Cre vector (Cre- AslNeo/Neo) or injected with the AAV9.Cre and AAV8.mAsl vectors (mAsl-Cre- AslNeo/Neo). AAV9.Cre and AAV8.mAsl were injected intracranially (red arrows) and intravenously via the superficial temporal vein (blue arrow) as described in Section 6.3.1.
10.2.2 Titration of the AAV9.Cre vector
As LoxP sites are naturally present in the mammalian genome, escalating doses of AAV9.Cre vector were administered to define a dose of AAV9.Cre, which would not cause aspecific brain damage not related to severe ASL deficiency. Indeed growth inhibition and genotoxicity in vitro (524), and hydrocephaly and microcephaly in mouse embryos have been previously reported after expression of high doses of Cre recombinase (525).
For titration, different doses of AAV9.Cre (6x10e10, 1.8x10e11, 6x10e11 and 6x10e12 vector genomes (vg) per mouse) were injected in the first 24 hours of life in
382 Chapter 10 - Appendices the Cre-WT and Cre-AslNeo/Neo groups. Uninjected littermates were used as controls. The experiment was stopped at day 8 of life.
In AslNeo/Neo mice, pups injected with intermediary doses (1.8x10e11 and 6x10e11 vg per mouse) were cannibalised by the dam before the age of 2 days. No interpretable data was therefore available for these groups.
All wild-type mice injected with AAV9.Cre and uninjected littermates survived. All AslNeo/Neo mice injected with the high dose (6x10e12 vg per mouse) died by day 6 of life. All AslNeo/Neo mice injected with the low dose (6x10e10 vg per mouse) and 16 out of 18 uninjected were alive at day 8 (Figure 118).
100
80
60 40
survival Percent 20
0 Days 0 2 4 6 8
AslNeo/Neo AslNeo/Neo_6x10e12
AslNeo/Neo_6x10e10
Figure 118. Survival of AslNeo/Neo mice injected intracranially with the escalating doses of AAV9.Cre vector compared to uninjected AslNeo/Neo mice
Experiment was stopped at day 8. AslNeo/Neo n=18, AslNeo/Neo_6x10e12 n=4, AslNeo/Neo_6x10e10 n=5. 6x10e10 to 6x10e12 represent the dose of vector injected in vg per mouse.
Wild-type mice had a lower weight at day 8 when injected with an intermediary dose of AAV9.Cre (6x10e11 vg per mouse) (Figure 119A). All wild-type mice injected with intermediary and high doses of AAV9.Cre (1.8x10e11, 6x10e11 and 6x10e12 vg per mouse) showed abnormal behaviour with difficulties in walking and ataxia with frequent falls.
383 Chapter 10 - Appendices
Both groups of AslNeo/Neo mice injected with low and high doses of AAV9.Cre (6x10e10 and 6x10e12 vg per mouse, respectively) displayed a significant weight loss at day 8 (One-way ANOVA with Dunnett’s post-test compared to wild-types, p<0.01 and p<0.001, respectively) (Figure 119B). AslNeo/Neo mice injected with high doses of AAV9.Cre displayed a severely modified behaviour similar initially to that observed in the wild-type group injected with intermediary and high doses of AAV9.Cre but which worsened rapidly to a pseudo-paralysis with tremor and rigidity with limited mobility. These pups were culled in accordance with humane endpoint. No behavioural abnormality was observed in the AslNeo/Neo mice injected with the low dose of AAV9.Cre.
Collectively these data showed that the lowest injected dose of AAV9.Cre (6x10e10 vg per mouse) had the safest profile and this titre was used for subsequent experiments. This dose was in accordance with a previous publication using AAV vectors delivering Cre recombinase to target mouse brains (477).
A B 6 6
4 4
2 2 Weight (grams) Weight Weight (grams) Weight
0 0 0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8 Time (Days) Time (Days)
WT WT_6x10e11 vg AslNeo/Neo AslNeo/Neo_6x10e10 vg WT_6x10e12 vg WT_1.8x10e11 vg AslNeo/Neo_6x10e12 vg WT_6x10e10 vg
Figure 119. Growth of wild-type and AslNeo/Neo mice injected intracranially with escalating doses of AAV9.Cre vector compared to uninjected controls
384 Chapter 10 - Appendices
10.2.3 Preliminary results
The design of the experiment is presented in Figure 117. Wild-type and AslNeo/Neo mice injected with the AAV9.Cre received a total dose of 6x10e10 vg per mouse within the first 24 hours of life in two symmetric intracranial injections. AslNeo/Neo mice concomitantly injected with AAV8.mAsl intravenously received 8x10e11 vg per mouse.
The experiment was stopped at day 8 of life, when the rapid liver growth was supposed to have significantly reduced the number of episomal mAsl transgene copies in the liver after systemic injection on the first day of life. This time point was chosen to ensure that the therapeutic effect on the urea cycle was still persisting and no hyperammonaemic episodes could bias results obtained from brain analysis in the AslNeo/Neo mice injected with the AAV8.mAsl.
No mouse died during this experiment. Compared to uninjected wild-type mice, the gross phenotype of uninjected and AAV9.Cre-injected AslNeo/Neo mice showed failure to thrive, which was more evident in the later group (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p<0.01 and p<0.001, respectively) (Figures 120, 121). An abnormal skin phenotype with sparse hair was seen at day 8 in the uninjected and AAV9.Cre-injected AslNeo/Neo mice (Figure 120). No difference in growth or fur was observed in the AAV8.mAsl,AAV9.Cre-injected AslNeo/Neo mice compared to wild-type mice suggesting a therapeutic effect of the systemic AAV8.mAsl injection (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p>0.05) (Figures 120, 121). No difference in growth was observed between the uninjected and AAV9.Cre-injected wild-type mice (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p>0.05) (Figures 120, 121). No mouse displayed any abnormal behaviour and the righting reflex tested at day 8 of life was not significantly different between groups (Figure 122). Compared to uninjected wild-type mice, liver/body weight ratio did not show any significant differences although a trend towards an increase in uninjected and AAV9.Cre-injected AslNeo/Neo mice was observed (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p>0.05) (Figure 123A). Brain/body weight ratio displayed a significant increase in uninjected and AAV9.Cre-injected
385 Chapter 10 - Appendices
AslNeo/Neo mice compared to uninjected wild-type mice, which was not observed in AAV8.mAsl,AAV9.Cre-injected AslNeo/Neo mice and AAV9.Cre-injected wild-type mice (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p<0.001, p<0.01, p>0.05 and p>0.05, respectively) (Figure 123B).
Cre-AslNeo/Neo WT mAsl-Cre-AslNeo/Neo
Cre-WT AslNeo/Neo
Figure 120. Gross phenotype of Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups
386 Chapter 10 - Appendices
8 WT Cre-WT 6 AslNeo/Neo Cre-AslNeo/Neo 4 mAsl-Cre-AslNeo/Neo
(grams) Weight 2
0 0 2 4 6 8 Days
Figure 121. Survival curves of Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups
WT (n=9), Cre-WT (n=5), AslNeo/Neo (n=7), Cre-AslNeo/Neo (n=4), mAsl-Cre-AslNeo/Neo (n=4). Horizontal lines display the mean ± standard deviation.
12
WT 9 Cre-WT AslNeo/Neo 6 Cre-AslNeo/Neo mAsl-Cre-AslNeo/Neo 3 Righting reflex (seconds) 0
Figure 122. Righting reflex in Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups
WT (n=7), Cre-WT (n=5), AslNeo/Neo (n=13), Cre-AslNeo/Neo (n=4), mAsl-Cre-AslNeo/Neo (n=4). Horizontal lines display the mean ± standard deviation.
387 Chapter 10 - Appendices
A B ns 0.05 0.09 * ** 0.08 ns 0.04 0.07
0.06 0.03 0.05 Liver / Body weight ratio Brain / Body weight ratio / weight Body Brain 0.02 0.04
WT Cre-WT AslNeo/Neo Cre-AslNeo/Neo mAsl-Cre-AslNeo/Neo
Figure 123. Hepatomegaly and brain/body weight ratios in Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups
WT (n=4), Cre-WT (n=3), AslNeo/Neo (n=4), Cre-AslNeo/Neo (n=4), mAsl-Cre-AslNeo/Neo (n=4). Horizontal lines display the mean ± standard deviation. One-way ANOVA with Dunnett’s post correction compared to uninjected wild-types; ns: not significant, * p<0.05, ** p<0.01.
At 8 days of life, concentrations of urea cycle amino acids measured in dried blood spots showed no significant difference between uninjected and AAV9.Cre-injected wild-type mice (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p>0.05 for all amino acids tested). Whilst significant differences (i.e. decreased arginine levels, increased citrulline, argininosuccinic acid and glutamine/glutamate levels) were observed in uninjected AslNeo/Neo mice compared to uninjected wild-types as observed in Section 5.2.3 (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p<0.001, p<0.01, p<0.001 and p<0.05, respectively) differences were more pronounced in AAV9.Cre-injected AslNeo/Neo mice compared to uninjected wild-types (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, arginine levels : p<0.001, citrulline levels : p<0.001, argininosuccinic acid levels: p<0.001 and glutamine/glutamate levels p<0.001, respectively). In AAV8.mAsl,AAV9.Cre-injected AslNeo/Neo mice, these abnormalities were improved compared to levels observed in the AAV9.Cre-injected AslNeo/Neo mice to levels similar to those observed in the uninjected AslNeo/Neo mice
388 Chapter 10 - Appendices although levels, except for those of glutamine/glutamate, were still significantly different to those of uninjected wild-type mice (One-way ANOVA with Dunett’s post- test compared to uninjected wild-types, arginine levels : p<0.01, citrulline levels : p<0.01, argininosuccinic acid levels: p<0.001 and glutamine/glutamate levels p>0.05, respectively) (Figure 124).
ASL activity in the liver and the brain showed no significant difference between uninjected and AAV9.Cre-injected wild-type mice (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p>0.05). Residual ASL activities in the liver and the brain of uninjected AslNeo/Neo mice were significantly reduced to 16.2% and 17.5% of uninjected wild-type activity, respectively (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p<0.001 and p<0.01, respectively). In the liver and the brain of uninjected AAV9.Cre-injected AslNeo/Neo mice, dramatic reductions of ASL activities to 1.5% and 4.3% of that of uninjected wild-type mice, respectively, were observed (One-way ANOVA with Dunett’s post- test compared to uninjected wild-types, p<0.001 and p<0.001, respectively). Residual ASL activities in the liver and the brain of uninjected AAV8.mAsl,AAV9.Cre-injected AslNeo/Neo mice were 62% and 93% of that of uninjected wild-type mice, respectively (One-way ANOVA with Dunett’s post-test compared to uninjected wild-types, p<0.05 and p>0.05, respectively) (Figure 125).
389 Chapter 10 - Appendices
A B ** 800 ** *** *** *** 400 ns 600 **
mol/L) ns
300 (µ 400 mol/L) µ 200 200 100 Citrulline Arginine ( Arginine
0 0
D C *** ns 1500 *** 1500 *** *** ns * mol/L) 1000 µ ns 1000
500 mol/L) µ ( 500 anhydrides ( Argininosuccinic acid and 0 glutamate + Glutamine 0
Figure 124. Dried blood spot urea cycle-related amino acid analysis of Cre- induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups
Concentration of (A) Arginine, (B) Citrulline, (C) Argininosuccinic acid and anhydrides and (D) glutamine and glutamate in dried blood spots from 8 day-old Cre-induced cerebral Asl knock out AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups. WT (n=5), Cre-WT (n=3), AslNeo/Neo (n=4), Cre-AslNeo/Neo (n=4), mAsl-Cre-AslNeo/Neo (n=4). Horizontal lines display the mean ± standard deviation. One-way ANOVA with Dunnett’s post-test compared to uninjected wild- types; ns: not significant, * p<0.05, ** p<0.01, *** p<0.001.
390 Chapter 10 - Appendices
* A B ns *** *** *** 500 200 ** ns ns 400 150 300 100 200 (nmol/ng/min) (nmol/ng/min) 50 100 ASL enzymatic activity ASL enzymatic ASL enzymatic activity ASL enzymatic 0 0 WT Cre-WT AslNeo/Neo Cre-AslNeo/Neo mAsl-Cre-AslNeo/Neo
Figure 125. ASL activity in the liver and brain of Cre-induced cerebral ASL knockout AslNeo/Neo mice with and without neonatal Asl gene therapy compared to control groups
ASL activity in (A) the liver and (B) the brain of WT (n=4), Cre-WT (n=4), AslNeo/Neo (n=9 and 3, respectively), Cre-AslNeo/Neo (n=4), mAsl-Cre-AslNeo/Neo (n=4). Horizontal lines display the mean ± standard deviation. One-way ANOVA with Dunnett’s post- test compared to uninjected wild-types; ns: not significant, * p<0.05, ** p<0.01, *** p<0.001.
391 Chapter 10 - Appendices
10.2.4 Discussion
The site-specific recombination facilitated by the Cre-LoxP system is a powerful technology to specifically knockdown the expression of a gene in some selected tissues or cell types (521). Using intracranial delivery of an AAV9 vector encoding Cre recombinase activity under the transcriptional control of the neuronal promoter human synapsin (hSyn), a set of experiments was conducted to study the cerebral effect of neuronal ASL knockout in AslNeo/Neo mice.
Initially an AAV9.Cre titration was performed to ensure the absence of deleterious off-target effect. Survival, growth and behaviour were assessed during the week following injection, which allowed determining a dose considered as non-deleterious and in accordance with previous reports (477).
The phenotype of AAV9.Cre-injected AslNeo/Neo mice was studied. A severe phenotype was observed with failure to thrive, abnormal fur and levels of urea cycle- related amino acids and decreased levels of ASL activities in the liver and the brain compared to uninjected AslNeo/Neo mice. The ASL brain activity was more severely decreased in AAV9.Cre-injected AslNeo/Neo mice compared to uninjected AslNeo/Neo mice (4.3% versus 17.5% of wild-type brain ASL activity). However the dramatic reduction of liver ASL activity (1.5% of wild-type liver ASL activity in AAV9.Cre- injected AslNeo/Neo mice versus 16.2% in uninjected AslNeo/Neo mice) was not expected. This was associated with a worsening of the phenotype related to the urea cycle defect illustrated by more pronounced abnormal levels of urea cycle amino acids. This was especially obvious for glutamine/glutamate levels indirectly reflecting ammonia levels (See Section 1.2.2). AAV8.mAsl injection concomitant with AAV9.Cre corrected the ASL knockout effect in both the liver and the brain, with an improvement in urea cycle amino acids, especially with regards to glutamine/glutamate concentrations for which there was no significant difference observed between AAV8.mAsl,AAV9.Cre-injected AslNeo/Neo mice and uninjected wild- types, suggesting that ammonia levels had been corrected.
Although the correction of ASL activities confirm the efficacy of this systemic gene therapy in both the liver and the brain as observed in Chapter 7, this AAV9.Cre- injected AslNeo/Neo mouse model could not be used further for studying the cerebral
392 Chapter 10 - Appendices effect of neuronal ASL knockout as the liver ASL knockout created a more severe urea cycle defect depicted by urea cycle amino acids levels which is likely to lead to severe hyperammonaemia, although the latter was not measured due to insufficient plasma samples being available. This would have had consequences on brain metabolism (See Section 1.2.2) and would have biased studies regarding neuronal nitrosative/oxidative stress (discussed in Section 5.4.2).
Previous studies have reported the systemic off-target biodistribution after AAV delivery in the central nervous system either by intracisternal or by intrathecal routes in various species: rat (526), dog (527) and non-human primates (528). This systemic biodistribution in peripheral viscera is preferentially observed in the liver with a diffuse transduction of the organ, which might be caused by intravenous redistribution. The working hypothesis that a highly-specific neuronal promoter as reported in the literature (529, 530) will avoid such a systemic off-target effect, was invalidated by in vivo experiments. Previous publications reporting transduction activity of the human synapsin promoter focussed their work in brain studies and did not consider off-target effects and systemic biodistribution. The transcriptional activity of the human synapsin promoter in peripheral viscera including the liver has been subsequently confirmed (531).
393 Chapter 10 - Appendices
10.3 Protocols for the preparation of general laboratory reagents
10.3.1 Reagents for PCR amplification of hASL exons
Reagents
• Ultrapure agarose (Invitrogen, Camarillo, CA, USA) • Betaine (Sigma, St Louis, MO, USA) • Custom primers (Integrated DNA Technologies, Leuven, Belgium) • Dinucleotides triphosphate (dNTPs) (Bioline, London, UK) • Invitrogen 100 bp DNA ladder (Invitrogen, Camarillo, CA, USA) • Water for cell culture (Sigma, St Louis, MO, USA) • BIOTAQTM DNA polymerase (Bioline, London, UK) • SYBR safe DNA gel stain (Invitrogen, Camarillo, CA, USA)
10.3.2 Reagents for sequencing reaction
Reagents
• Big Dye Terminator version 1.1 cycle sequencing kit (ThermoFisher scientific, Loughborough, UK) • Ethanol (Fisher scientific, Loughborough, UK) • Exonuclease I (New England Biolabs UK Ltd, Hitchin, UK) • Shrimp alkaline phosphatase (New England Biolabs UK Ltd, Hitchin, UK) • TE buffer (Sigma, St Louis, MO, USA) •
394 Chapter 10 - Appendices
10.3.3 Reagents and buffers for DNA extraction for genotyping
Reagents
• NaOH (Sigma, St Louis, MO, USA) • EDTA (Sigma, St Louis, MO, USA) • Tris (Sigma, St Louis, MO, USA)
Buffers and solutions
NaOH solution
Final concentration
NaOH 0.5 M 2.5 mL 25 mM
EDTA Na2 0.5 M 20 µL 0.2 mM dH2O 47.48 mL
Final volume 50 mL
No need to pH, this solution is pH 12.
Tris-HCl Buffer
Final concentration
Tris-HCl 1 M 2 mL 40mM adjusted to pH 8.0 dH2O 48 mL
Total 50 mL adjusted to pH 5.0
395 Chapter 10 - Appendices
10.3.4 Reagents and buffers for DNA electrophoresis on agarose gel for genotyping
Reagents
• Agarose powder (Sigma, St Louis, MO, USA) • Safeview nucleic acid stain (NBS Biologicals, Huntingdon, UK) • Tris-base (Sigma, St Louis, MO, USA) • Glacial acetic acid (Sigma, St Louis, MO, USA) • EDTA (Sigma, St Louis, MO, USA) • Orange G (Sigma, St Louis, MO, USA) • Glycerol (Sigma, St Louis, MO, USA)
Buffers and solutions
50xTAE
Tris-base 242 g
Glacial acetic acid 57.1 mL
EDTA 18.6 g dH2O up to 1L
PCR loading buffer
Orange G 0.1 g
30% Glycerol 10 mL
396 Chapter 10 - Appendices
10.3.5 Reagents for blood sampling
Reagents
• Sodium heparin (Sigma, St Louis, MO, USA)
10.3.6 Reagents and buffers for nitrite and nitrate measurements (Griess reaction)
Reagents
• Sulfanilamide (Sigma, St Louis, MO, USA) • N-1-naphtylethylenediamine dihydrochloride (Sigma, St Louis, MO, USA) • Phosphoric acid (Sigma, St Louis, MO, USA) • Nitrate reductase from Aspergillus niger (Sigma, St Louis, MO, USA) • Glycerol (Sigma, St Louis, MO, USA) • Glucose-6-phosphate dehydrogenase (Sigma, St Louis, MO, USA) • Glucose-6-phosphate (Sigma, St Louis, MO, USA) • Nicotinamide adenine dinucleotide phosphate (NADPH) (Sigma, St Louis, MO, USA) • Sodium phosphate dibasic (Sigma, St Louis, MO, USA) • Sodium phosphate monobasic monohydrate (Sigma, St Louis, MO, USA)
397 Chapter 10 - Appendices
Buffers and solutions
Nitrate assay phosphate buffer
Solution A
Na2HPO4.2H20 6.23 g ddH2O 1 L
Solution B
NaH2PO4.2H20 5.46 g ddH2O 1 L
Solutions A and B were mixed at a ratio 1:2 adjusted to pH 7.4
Griess reaction – reagents
Reagent A
Sulfanilamide 2 g
Phosphoric acid 5 mL ddH2O 100 mL
Reagent B
N-(1-Naphthyl)ethylaminediamine 200 mg ddH2O 100 mL
398 Chapter 10 - Appendices
10.3.7 Method for gelatin coating slides
Reagents
• Gelatine powder (Sigma, St Louis, MO, USA) • Chromium (III) potassium sulphate 12 hydrate (Sigma, St Louis, MO, USA)
Coating slides allows the slice of tissue to remain fixed onto the slide with the adherence of the gelatine and electric charge interaction between Chromium (III) potassium sulphate 12 (positively charged) and cell membranes (negatively charged).
Slides are placed in metal racks. 1 Liter of dH2O is poured in a plastic bowl and microwaved for 3 minutes at 1,500 Watts to reach 45°C. 5 g of gelatine and 0.5 g of Chromium (III) potassium sulphate 12 hydrate are added in hot water. Do not let the temperature drop below 42°C otherwise heat again in microwave. Immerse racks briefly and put in oven at 56°C to dry overnight. Repeat this step a second time the following day.
10.3.8 Reagents and buffers for perfusion of mice and organ collection
Reagents
• Isoflurane (Abbott, Abbott Park, IL, USA) • Paraformaldehyde (Sigma, St Louis, MO, USA) • Sucrose (Sigma, St Louis, MO, USA) • Formalin solution neutral buffered 10% (Sigma, St Louis, MO, USA) • Ethanol (Merck, Darmstadt, Germany)
399 Chapter 10 - Appendices
10.3.9 Reagents and buffers for brain sectioning on freezing microtome
Reagents
• Shandon M-1 Embedding matrix (ThermoScientific, Kalamazoo, MI, USA) • Trisma base (Sigma, St Louis, MO, USA) • NaCl (Sigma, St Louis, MO, USA) • HCl (Sigma, St Louis, MO, USA) • Sodium azide (Sigma, St Louis, MO, USA) • Ethylene glycol (Sigma, St Louis, MO, USA) • Sucrose (Sigma, St Louis, MO, USA)
Buffers and solutions
10xTBS
Trisma Base 64 g
NaCl 85 g
HCl 32 mL dH2O up to 1 L adjusted to pH 7.6
TBSA
10% Na azide 2 mL
TBS 198 mL
400 Chapter 10 - Appendices
TBSAF
Sucrose 75 g
Ethylene glycol 150 mL
TBSA 350 mL
10.3.10 Reagents and buffers for immunostaining in free-floating sections
Reagents
• 30% hydrogene peroxidase (Sigma, St Louis, MO, USA) • Trisma base (Sigma, St Louis, MO, USA) • NaCl (Sigma, St Louis, MO, USA) • HCl (Sigma, St Louis, MO, USA) • Sodium azide (Sigma, St Louis, MO, USA) • Ethylene glycol (Sigma, St Louis, MO, USA) • Sucrose (Sigma, St Louis, MO, USA) • Triton X-100 (Sigma, St Louis, MO, USA) • Vectastain ABC Kit (Vector, Burlingame, CA, USA) • 3,3’-diaminobenzidine (DAB) (Sigma, St Louis, MO, USA) • Histoclear (National diagnostics, Atlanta, GE, USA) • DPX-new (MerckMillipore, Darmstadt, Germany)
401 Chapter 10 - Appendices
Buffers and solutions
10xTBS
Trisma Base 64 g
NaCl 85 g
HCl 32 mL dH2O up to 1 L adjusted to pH 7.6
TBS-T
Triton X-100 300 µL
1xTBS 100 mL
10.3.11 Reagents and buffers for immunostaining in paraffin- embedded slides
Reagents
• Histoclear (National diagnostics, Atlanta, GE, USA) • IMS (Acquascience, Uckfield, UK) • Methanol (Merck, Darmstadt, Germany) • 30% hydrogene peroxidase (Sigma, St Louis, MO, USA) • Sodium citrate (Sigma, St Louis, MO, USA) • HCl (Sigma, St Louis, MO, USA) • Polink-2 HRP Plus rabbit DAB Detection System for immunohistochemistry (GBI Labs, Mukilteo, WA, USA)
402 Chapter 10 - Appendices
• Trisma base (Sigma, St Louis, MO, USA) • NaCl (Sigma, St Louis, MO, USA) • Sodium azide (Sigma, St Louis, MO, USA) • Ethylene glycol (Sigma, St Louis, MO, USA) • Sucrose (Sigma, St Louis, MO, USA) • Triton X-100 (Sigma, St Louis, MO, USA)
Buffers and solutions
10xTBS
Trisma Base 64 g
NaCl 85 g
HCl 32 mL dH2O up to 1 L adjusted to pH 7.6
TBS-T
Triton X-100 300 µL
1xTBS 100 mL
Citrate buffer
Final concentration
Sodium citrate 14.7 g 0.01 M
HCl 27 mL 1 M
Adjusted to pH 6.0
403 Chapter 10 - Appendices
10.3.12 Reagents and buffers for NISSL staining
Reagents
• Certistain (Cressyl violet) (Merck Millipore, Darmstadt, Germany) • Ethanol (Merck, Darmstadt, Germany) • Acetic acid (Sigma, St Louis, MO, USA) • Xylene (Sigma, Kalamazoo, MI, USA) • Isopropanol (Fisher, Loughborough, UK) • Paraformaldehyde (Sigma, St Louis, MO, USA)
Buffers and solutions
1% cressyl violet solution
Certistain (Cressyl violet) 4 g
Ethanol 100% 40 mL ddH2O 360 mL
Put on stirrer fot 20 minutes and heat then filter and let cool down overnight in fume cupboard.
404 Chapter 10 - Appendices
10.3.13 Reagents and buffers for TUNEL staining
Reagents
• Bovine serum albumine (Sigma, St Louis, MO, USA) • 30% hydrogene peroxidase (Sigma, St Louis, MO, USA) • PFA (Sigma, St Louis, MO, USA)
• K2HPO4 (Sigma, St Louis, MO, USA)
• KH2PO4 (Sigma, St Louis, MO, USA) • TDT enzyme (Roche, Mannheim, Germany) • Biotinilated dUTP (Roche, Mannheim, Germany) • Sodium cacodylate trihydrate (Sigma, St Louis, MO, USA) • Cobalt chloride (Sigma, St Louis, MO, USA) • Cobalt sulphate (Sigma, St Louis, MO, USA) • Nickel chloride (Sigma, St Louis, MO, USA) • NaCl (Sigma, St Louis, MO, USA) • Sodium citrate (Sigma, St Louis, MO, USA)
Buffers and solutions
0.1 M Phosphate buffer
K2HPO4 17.4 g
KH2PO4 13.6 g dH2O 1 L
10 mM Phosphate buffer
0.1 mL Phosphate buffer 0.1 L ddH2O 1 L 405 Chapter 10 - Appendices
PB/BSA buffer
0.1 M phosphate buffer 500 mL
0.1% BSA 500 mL
TUNEL solution
TDT enzyme 1 uL
Biotynilated dUTP 1.5 uL
Cacodylate buffer 100 uL ddH2O 897.5 uL
To be made and kept on ice, use immediately
Cacodylate buffer
Tris 3.64 g
Sodium cacodylate trihydrate 29.96 g ddH2O 80 mL adjusted to pH 7.5, then
Cobalt chloride 0.24 g adjusted to pH 7.2 made up to 100 mL with ddH2O and store at 4°C
TUNEL stop solution
Final concentration
NaCl 1,753.2 mg 300 mM
Sodium citrate 882.3 mg 30 mM ddH2O 100 mL
406 Chapter 10 - Appendices
DAB cobalt nickel solution
# Solution 1
Nickel chloride 2 g ddH2O 100 mL
Mix
# Solution 2
Cobalt sulfate 2.5 g ddH2O 100 mL
Mix
# For DAB nickel cobalt solution
DAB 25 mg
10 mM phosphate buffer 50 mL
Solution 1 0.5 mL
Solution 2 0.5 mL
Stirr and filter before use
10.3.14 Reagents and buffers for Haematoxylin and eosin (H&E) staining
Reagents
• Xylene (Leica Biosystems, Nanterre, France) • Harris Haematoxylin (Leica Biosystems, Nanterre, France) • Aqueous eosin 1% (Leica Biosystems, Nanterre, France) • HCl (Sigma, St Louis, MO, USA) • IMS (Acquascience, Uckfield, UK)
407 Chapter 10 - Appendices
Buffers and solutions
1% acid ethanol
HCl 1 mL
Ethanol 70 mL dH2O 30 mL
10.3.15 Reagents and buffers for Periodic Acid-Schiff (PAS) staining
Reagents
• Xylene (Leica Biosystems, Nanterre, France) • Harris Haematoxylin (Leica Biosystems, Nanterre, France) • Amylase from hog pancreas (Sigma, St Louis, MO, USA) • Periodic acid (Sigma, St Louis, MO, USA) • Commercial Fuelgen stain (5) (TCS Biosciences Ltd, Buckingham, UK) • IMS (Acquascience, Uckfield, UK)
10.3.16 Reagents and buffers for Masson trichrome staining
Reagents
• Xylene (Leica Biosystems, Nanterre, France) • Mayer’s Haematoxylin (Leica Biosystems, Nanterre, France) • Potassium diachromate (BDH chemicals, Radnor, PA, USA) • Chromotrope 2R (ThermoFisher Scientific, Loughborough, UK)
408 Chapter 10 - Appendices
• Fast Green FCF (Sigma, St Louis, MO, USA) • Phosphotungstic acid (BDH chemicals, Radnor, PA, USA) • Glacial acetic acid (Sigma, St Louis, MO, USA) • Celestine blue (Sigma, St Louis, MO, USA) • Sulphuric acid (Sigma, St Louis, MO, USA) • 2.5% aqueous ammonium iron (III) sulphate (Sigma, St Louis, MO, USA) • Glycerol (Sigma, St Louis, MO, USA) • IMS (Acquascience, Uckfield, UK)
Buffers and solutions
Masson trichrome working solution
Chromotrope 2R 0.6 g
Fast green FCF 0.3 g
Phosphotungstic acid 0.6 g
Glacial acetic acid 1 mL dH2O 100 mL
10.3.17 Reagents and buffers for Oil Red O staining
Reagents
• Xylene (Leica Biosystems, Nanterre, France) • Mayer’s Haematoxylin (Leica Biosystems, Nanterre, France) • Oil Red O (Sigma, St Louis, MO, USA) • Acetone (Acquascience, Uckfield, UK) • Formaldehyde (Sigma, St Louis, MO, USA) • Aquatex (Merck Millipore, Darmstadt, Germany) • IMS (Acquascience, Uckfield, UK) 409 Chapter 10 - Appendices
Buffers and solutions
Oil Red O working solution
Oil Red O 1 g
Acetone 5 mL
Dissolve
70% Ethanol 100 mL
Mix and let the solution settle for 24 hours then filter in a coplin jar before use
10.3.18 Reagents and buffers for mASL western blot
Reagents
• NP-40 (Sigma, St Louis, MO, USA) • Hepes (Sigma, St Louis, MO, USA) • KCl (Sigma, St Louis, MO, USA)
• MgCl2 (Sigma, St Louis, MO, USA) • DNase (Sigma, St Louis, MO, USA) • Protease inhibitors Complete Mini EDTA-free protease inhibitor cocktail (Roche, Mannheim, Germany) • Tris/HCl (Sigma, St Louis, MO, USA) • Glycerol (Sigma, St Louis, MO, USA) • Sodium dodecyl sulphate (SDS) (Sigma, St Louis, MO, USA) • Bromophenol blue (Sigma, St Louis, MO, USA) • Dithiothreitol (DTT) (Sigma, St Louis, MO, USA) • PBS tablets (Sigma, St Louis, MO, USA) • Tween20 (Sigma, St Louis, MO, USA) • Protein ladder (Page Ruler; Fermentas, York, UK)
410 Chapter 10 - Appendices
• Polyvinylidene difluoride (PVDF) (GE Healthcare, UK) • Non-fat milk powder Cow & Gate (Trowbridge, UK) • Glycine (Sigma, St Louis, MO, USA) • Trisma-Base (Sigma, St Louis, MO, USA) • Sodium dodecyl sulphate (SDS) (Sigma, St Louis, MO, USA) • Methanol (Merck, Darmstadt, Germany) • Amersham ECL western blotting detection reagent (GE Healthcare UK Ltd, Buckinghamshire, UK) • Enhanced chemo-luminescence plus (ECL+) (GE Healthcare UK Ltd, Buckinghamshire, UK)
Buffers and solutions
1x NP-40 lysis buffer
Final concentration
NP-40 0.3%
Hepes 2.38 g 10 mM adjusted to pH 8.5
KCl 740 mg 10 mM
MgCl2 476 mg 5 mM
DNase 100 mg 1 mg/mL
Dithiothreitol (DTT) 771 mg 5mM
Protease inhibitors dH2O adjusted for final volume 100 mL
5x Laemmli loading buffer
Final concentration
Tris/HCl 1.02 g 65 mM adjusted to pH 8
Glycerol 10 mL 10 % (v/v)
SDS 2.3 g 2.3 % (w/v)
411 Chapter 10 - Appendices
Bromophenol blue 10 mg 0.01%
Dithiothreitol (DTT) 1 g 1% dH2O adjusted for final volume 100 mL
1% PBST
PBS 100 mL
Tween20 1 mL
1x Tris-glycine buffer
Final concentration
Glycine 15.1 g 192 mM
Tris-Base 94 g 250 mM
SDS 10% 10 mL 0.1 % dH2O adjusted for final volume 1 L
1x Transfer buffer
Final concentration
Glycine 14.4 g 192 mM
Tris-Base 30.3 g 250 mM
Methanol 200 mL 20 % dH2O adjusted for final volume 1 L
412 Chapter 10 - Appendices
10.3.19 Reagents and buffers for GFP Enzyme linked immunosorbent assay (ELISA)
Reagents
• b gal ELISA kit (Sigma, St Louis, MO, USA) • Bovine serum albumin (Sigma, St Louis, MO, USA) • Tetramethylbenzidine (TMB) (Sigma, St Louis, MO, USA) • GFP protein (Sigma, St Louis, MO, USA)
• H2SO4 (Sigma, St Louis, MO, USA) • Monoclonal anti-GFP antibody (Abcam ab1218, Abcam, Cambridge, UK) • Biotin-conjugated polyclonal anti-GFP Abcam 6658, Abcam, Cambridge, UK) • Streptavidin - horseradish peroxidase (HRP) (Invitrogen, Camarillo, CA, USA) • Bicarbonate buffer (see Appendix I – Solutions and buffers) • Wash buffer (see Appendix I – Solutions and buffers) • Block solution (see Appendix I – Solutions and buffers)
Buffers and solutions
Bicarbonate buffer (ELISA protocol)
Na2CO3 0.8 g
NaHCO3 1.465 g dH2O adjusted for final volume 500 mL
Wash buffer (ELISA protocol)
Tween 20 200 µL
PBS 400 mL
413 Chapter 10 - Appendices
Block solution (ELISA protocol)
Bovine serum albumin 4 g
PBS 400 mL
10.3.20 Reagents and buffers for WPRE quantitative polymerase chain reaction (qPCR)
Reagents
• QIAgen DNeasy kit (QIAgen, Crawley, UK) • Platinum® Quantitative PCR SuperMix-UDG with ROX (Lifetechnologies, Foster city, CA, USA) • Custom primers (Lifetechnologies) • Dual labelled probes (Eurofins)
10.3.21 Reagents and buffers for cloning the vector construct
Reagents
• Custom primers (Integrated DNA Technologies, Leuven, Belgium) • Q5 High-Fidelity DNA Polymerase (New England Biolabs UK Ltd, Hitchin, UK) • Restriction enzymes Nhe1, EcoRV, Bgl1 (New England Biolabs UK Ltd, Hitchin, UK) • NE Buffer 3.1 (New England Biolabs UK Ltd, Hitchin, UK) • Molecular biology water (Sigma, St Louis, MO, USA) • Agarose (Sigma, St Louis, MO, USA) • QIAgen Spin Miniprep Kit (QIAgen, Crawley, UK) • QIAquick Gel Extraction Kit (QIAgen, Crawley, UK) • Fast alkaline phosphatase (Thermo Scientific, Loughborough, UK)
414 Chapter 10 - Appendices
• T4 DNA Ligase (New England Biolabs UK Ltd, Hitchin, UK) • Max efficiency DH5a TM Competent Cells (Invitrogen Thermo Scientific, Loughborough, UK) • SOC Outgrowth Medium (Invitrogen Thermo Scientific, Loughborough, UK) • LB Broth (Sigma, St Louis, MO, USA) • Ampicillin (Sigma, St Louis, MO, USA) • Invitrogen 1000 bp DNA ladder (Invitrogen, Camarillo, CA, USA)
Buffers and solutions
LB Broth medium
LB Broth 20 g dH2O 1 L
Ampicillin (100mg/mL) 1 mL
10.3.22 Method for making Agar plates
Reagents
• LB Agar (Sigma, St Louis, MO, USA) • Ampicillin (Sigma, St Louis, MO, USA)
LB Agar 8.75 g dH2O 250 mL
Mix well and autoclave at 121°C for 20 minutes. Add 1mL of Ampicillin diluted at 100 mg/mL per 1 L of Agar solution for a final ampicillin concentration of 100 ug/mL. Seal 10 mL of solution per plate, wait for cooling at room temperature.
415 Chapter 10 - Appendices
10.3.23 Reagents and buffers for AAV vector production, concentration and titration
Reagents
• QIAgen Spin Maxiprep Kit (QIAgen, Crawley, UK) • Dulbecco’s Modified Eagle Medium (DMEM) (GIBCO, Paisley, UK) • Fetal calf serum (FCS) (GIBCO, Paisley, UK) • Dulbecco’s PBS (GIBCO, Paisley, UK) • Trypsin EDTA (Sigma, St Louis, MO, USA) • Benzonase nuclease (250 units/uL) (Sigma, St Louis, MO, USA) • Deoxycholic acid (Sigma, St Louis, MO, USA) • Glycine (Sigma, St Louis, MO, USA) • Tris base (Sigma, St Louis, MO, USA) • Phosphate buffered saline (PBS) tablets (Sigma, St Louis, MO, USA) • Polyethylenimine (Polysciences Inc, Northampton, UK) • Glycerol (Sigma, St Louis, MO, USA) • Sodium dodecyl sulphate (SDS) (Sigma, St Louis, MO, USA) • Xylene cyanol (Sigma, St Louis, MO, USA) • 10000x Gel red (Cambridge Bioscience, Cambridge, UK) • Hyper Ladder I (Bioline, London, UK) • 2x Laemmli buffer (Sigma, St Louis, MO, USA) • Coomassie Brilliant blue R-250 (VWR International Ltd, Lutterworth, UK) • Acetic acid (Sigma, St Louis, MO, USA) • Methanol (Merck, Darmstadt, Germany) • Page ruler prestained protein ladder, broad range 10-250 kDa (Life Technologies, Paisley, UK) • NuPAGE MOPS SDS 20x running buffer (Invitrogen Thermo Scientific, Loughborough, UK).
416 Chapter 10 - Appendices
Buffers and solutions
TD buffer (Vector production)
Final concentration
NaCl 8.181 g 140 mM
KCl 372.8 mg 5 mM
K2HPO4 95.26 mg 0.7 mM
MgCl2 333.23 mg 3.5 mM
Tris 3.0285 g 25 mM ddH2O 1 L adjusted to pH 7.5, heated on a benchtop autoclave at 121°C for 20 minutes
PBS (Vector purification)
PBS 5 tablets ddH2O 1 L filtered at 0.22 uM, pH 7.5
Glycine buffer (Vector purification)
Final concentration
Glycine 1.875 g 50 mM ddH2O 500 mL filtered at 0.22 uM, adjusted to pH 2.7
417 Chapter 10 - Appendices
1M Tris buffer (Vector purification and titration)
Final concentration
Tris base 121.14 g 1 M ddH2O 1 L filtered at 0.22 uM, adjusted to pH 8.8
50x Alkaline electrophoresis buffer
NaOH 30 g
EDTA 7.31 g ddH2O 500 mL adjusted to pH 8.0
4x Sample loading buffer
Glycerol 200 µL
50X alkaline electrophoresis buffer 80 µL
20% w/v SDS 60 µL
Xylene cyanol powder arbitrary amount ddH2O 1 mL
0.1 M NaCl
NaCl 5.84 g ddH2O 1 L heated on a benchtop autoclave at 121°C for 20 minutes
418 Chapter 10 - Appendices
4x Gel red solution:
Final concentration
10000X Gel red 40 µL
NaCl 100 mL 0.1 M
Coomassie blue dye
Methanol 500 mL
Acetic acid 100 mL ddH2O 400 mL
Coomassie Brilliant blue R-250 2.5 g
Destain solution
Methanol 250 mL
Acetic acid 70 mL ddH2O 680 mL
10.3.24 Reagents for identification of amino acids by mass spectrometry
Reagents
• L-Arginine (>99%) (Sigma, St Louis, MO, USA) • L-Arginine-13C (99% atom %d) (CDN isotopes, Thaxted, UK) • L-Glutamine (>99%, recommended for HPLC) (Sigma, St Louis, MO, USA) • L-Glutamine-2,3,3,4,4-d5 (>98% atom %d) (QMX, Thaxted, UK) • L-Glutamic acid (>99%, recommended for HPLC) (Sigma, St Louis, MO, USA) 419 Chapter 10 - Appendices
• L-Glutamic-2,3,3,4,4-d5 acid (98% atom %d) (QMX, Thaxted, UK) • L-Ornithine HCl (>98%) (Sigma, St Louis, MO, USA) • L-Ornithine-2,3,3,4,4,5,5-d7 HCl (98% atom %d) (QMX, Thaxted, UK) • L-Citrulline (>98%) (Sigma, St Louis, MO, USA) • L-Citrulline-2,3,3,4,4,5,5-d7 (65% atom %d) (QMX, Thaxted, UK) • Argininosuccinic acid disodium saly hydrate (>80%) (Sigma, St Louis, MO, USA) • 9-Fluorenylmethyl chloroformate (Sigma, St Louis, MO, USA) • Acetone (Sigma, St Louis, MO, USA) • Boric acid (Sigma, St Louis, MO, USA)
• MQH2O, Milli-Q Ultra pure Water Purification Systems (Millipore, Brussels, Belgium)
10.3.25 Reagents for separation of amino acids by liquid chromatography
Reagents
• Methanol (HPLC grade) (Fisher scientific, Loughborough, UK) • Ethanol (HPLC grade) (Fisher scientific, Loughborough, UK) • Trifluoroacetic acid (HPLC grade) (Sigma, St Louis, MO, USA) • Heptafluorobutyric acid (HPLC grade) (Sigma, St Louis, MO, USA) • Trifluorobutyric acid (HPLC grade) (Sigma, St Louis, MO, USA) • Acetic acid (HPLC gradsepare) (Sigma, St Louis, MO, USA)
420 Chapter 10 - Appendices
10.3.26 Reagents and buffer for ASL assay in liver extracts
Reagents
• Adenosine triphosphate (Sigma, St Louis, MO, USA) • Potassium chloride (KCl) (Sigma, St Louis, MO, USA)
• Magnesium chloride (MgCl2) (Sigma, St Louis, MO, USA) • Aspartate (Sigma, St Louis, MO, USA) • TrisHCl (Sigma, St Louis, MO, USA) • Argininosuccinate disodium salt hydrate (Sigma, St Louis, MO, USA) • (2S)-(+)-Amino-5-iodoacetamidopentanoic acid (Enzo, Farmingdale, NY, USA)
Buffers and solutions
ASL enzymatic buffer
Pre-buffer A Final concentration
TrisHCl 121 mg 10 mM
KCl 149 mg 20 mM
MgCl2 47.6 mg 5 mM
Aspartate 65 mg 5 mM
MQH2O 88.1 mL
then pre-buffer B
Pre-buffer A 8 mL
Adenosine triphosphate (ATP) 27.5 mg 5 mM adjusted to pH 7.45
then final ASL enzymatic buffer
421 Chapter 10 - Appendices
Pre-buffer B 256.5 µL
10 mM argininosuccinate 13.5 µL 0.5 mM
10.3.27 Positive controls for liver staining
A
acid-Schiff
Periodic
B Masson trichrome
C O red
Oil
Figure 126. Positive controls for Periodic acid-Schiff, Masson trichrome and Oil red O stainings
(A) Periodic acid-Schiff staining of liver sample, (B) Masson trichrome of gut sample and (C) Oil red O staining of muscle sample. White arrows indicate positive staining. Scale bars are (A) 500 μm, (B) 1 cm and (C) 1 mm, respectively.
422 Chapter 10 - Appendices
10.4 Database for clinical study
10.4.1 Systemic phenotype
At diagnosis Long term phenotype Survival Higher Follow up Neurology Liver High Form Sex Ethnicity (Age at Enlarged Trichorrhexis Age ammonia (months) Developmental Abnormal Myopathic Abnormal Hypokalaemia blood Patient Patient
number death) Epilepsy Hepatomegaly Transaminitis kidneys nodosa (µmol/L) delay movements features brain MRI pressure 1 Early F White British 14 days NA Alive 244 Y N Y N Y Y Y N N N N 2 Early M White British 3 days 749 Alive 51 Y N N Y Y Y Y N Y N N 3 Early F White British 4 days 384 Alive 43 Y N N N NP Y Y N Y N N 4 Early M White Lituanian 4 days 520 Alive 149 Y Y Y Y N Y Y Y Y N N 5 Early M White British 7 days 368 Alive 150 Y Y N Y N Y Y Y N N N Died 6 Early F Indian 4 days 1370 186 Y N N N NP Y Y Y Y Y N (12 years) 7 Early M White British 4 days 308 Alive 137 Y Y N N NP N N NA N N N 8 Early M Chinese <28 days 190 Alive 217 Y Y N Y Y Y Y Y Y N N 9 Early M Chinese <28 days 800 Alive 308 Y Y Y N Y Y Y NA Y N N Died 10 Early M Indian <28 days 1941 0 NA (<28 days) 11 Early F Pakistani 7 days NA Alive 139 N N Y N NP N N N N N N 12 Early M White British 8 days 715 Alive 201 Y Y N N NP Y Y N Y N N 13 Early M Pakistani 4 days 650 Alive 111 Y N Y N NP NA Y Y Y N N 14 Early F Pakistani 2 days 800 Alive 151 Y N N N NP Y Y N Y N N 15 Early F Pakistani 4 days 668 Alive 23 N N N N NP Y Y N N N N Died 16 Early F Pakistani 4 days 1038 54 Y N N N NP Y Y Y Y N N (4 years) 17 Early F Pakistani 3 days 390 Alive 77 Y N N N NP Y Y Y Y N N Died 18 Early F Pakistani 2 days 2000 0 NA NA NA NA NP Y Y NA N N N (4 days) Died 19 Early F Pakistani 4 days 373 23 Y Y N N NP Y NA Y Y N N (23 months) 20 Early F Pakistani 12 days 2006 Alive 46 Y N N N NP Y Y N Y N N Died 21 Early M Bangladeshi 4 days 1020 132 Y Y N N NP N Y NA Y N N (11 years) 22 Early M White British 3 days 791 Alive Lost N N N N N Y Y N Y N N 23 Early M White British 3 days 1000 Alive 32 NA NA NA NA NP N N N Y N N Died 24 Late M White British 4 years NA 637 Y Y Y Y NP N Y NA Y N N (53 years) 25 Late M White British 2 years 278 Alive 295 Y Y Y N N N N NA N N N 26 Late M White British 3.5 years 456 Alive 419 Y N N N NP N N NA N N N 27 Late F Pakistani >1 year NA Alive 337 Y N N N NP Y Y NA N N N 28 Late M White British 1 year 723 Alive 191 Y Y Y Y N N N Y N N N 29 Late F Pakistani 13 months NA Alive 56 Y Y Y N Y Y Y N N N N 30 Late F Indian <10 years NA Alive Lost Y N N N NP N N NA N N Y
423 Chapter 10 - Appendices
At diagnosis Long term phenotype Survival Higher Follow up Neurology Liver High Form Sex Ethnicity (Age at Enlarged Trichorrhexis Age ammonia (months) Developmental Abnormal Myopathic Abnormal Hypokalaemia blood Patient Patient
number death) Epilepsy Hepatomegaly Transaminitis kidneys nodosa (µmol/L) delay movements features brain MRI pressure 31 Late M Indian <10 years NA Alive Lost Y N N N NP N N NA N N N 32 Late M Indian <10 years 323 Alive Lost Y N N N NP N N NA N N N 33 Late F White British 12 years 77 Alive 316 Y Y Y Y NP N N NA N N Y 34 Late F Banglageshi 35 days 14 Alive 138 Y Y N N Y N N Y N N N 35 Late F White British 5.5 years 18 Alive 262 Y N Y N Y N N NA N N Y 36 Late M White British 4 years 9 Alive 246 Y N Y N Y N N NA N N N 37 Late F White British 2 years 351 Alive 165 Y Y N N N N N NA N N N 38 Late M White British 3 months 34 Alive 90 Y Y N N N Y Y N N N N Died 39 Late F Chinese 1.5 years NA 275 Y Y N N N N N NA N N N (20 years) 40 Late F White British 6 years <45 Alive 444 Y N N N NP N N NA N N Y 41 Late F White British 5 years NA Alive 216 Y Y N N NP N N N N N Y 42 Late M White British 2.5 years NA Alive 307 Y N N N NP N N N Y N N 43 Late F White British 5 months NA Alive 280 Y Y N N Y N N N Y N N 44 Late F NA 3.5 years NA Alive 227 Y N N N NP N N N Y N N 45 Late M White British 11 months NA Alive 243 Y Y N N NP N N N N N N
46 Late M White British 2 years Alive 24 Y N N N NP N N NA N N N NA 47 Screened F Pakistani Neonatal 43 Alive 96 N N N N NP N N NA N N N
48 Screened M White British Neonatal <75 Alive 187 Y N N N N Y Y N N N N
49 Screened M Chinese Neonatal 133 Alive 155 Y Y N N Y Y Y Y N N N 50 Screened M Chinese Neonatal 68 Alive 218 Y N N N N N N N N N N 51 Screened M Chinese Antenatal 49 Alive 193 Y Y N N Y Y Y NA Y N N Died 52 Screened F Pakistani Neonatal NA 0 NA NA NA NA NP NA NA NA Y N N (7 days) 53 Screened M Pakistani Antenatal NA Lost Lost NA NA NA NA NP NA NA NA Y N N 54 Screened M Pakistani Neonatal 133 Lost 164 Y N N N NP Y Y NA Y N N 55 Screened M Pakistani Neonatal 92 Alive 123 Y Y Y N NP Y Y N Y N N 56 Screened M Bangladeshi Neonatal 190 Alive 211 Y N N N NP N Y Y Y N N
Table 20. Detailed general phenotype of the patients included in the clinical study
Database related to Sections 3.2 & 3.4. M: Male; F: Female; NP: Not performed; Y: Yes; N: No; NA: Not available; cDNA: coding DNA. It was assumed that patients had normal blood pressure if hypertension was not specifically mentioned in medical records
424 Chapter 10 - Appendices
10.4.2 Neurological phenotype
Neurological phenotype Abnormal Developmental delay Epilepsy Myopathic features Abnormal brain MRI movements Form Age at Severity (IQ) Mainstream Age at Type of Abnormal 2015 (Years) Y/N Y/N Y/N Features Y/N Features Y/N Features Patient number Patient diagnosis
Ataxia, 1* 20.3 Early Y Moderate (48) N N Y Y N Y Mild cerebellar atrophy tremor Cerebral oedema, small haemorrages in white matter, edema in the 2 4.25 Early Y 2 years Mild N NA N Y Hypotonia Y central tegmental tract and internal capsid 3 3.6 Early Y 23 months Mild N N NA N N NP Hypotonia, Ataxia, 4 12.4 Early Y 9 months Moderate N Y 5.5 years TC, M Y Y Y myopathic face, N tremor EMG Hypotonia, episods 5 12.5 Early Y 20 months Mild (54) N Y 9 years TC, Ab Y N Y N of weakness, EMG 6 15.5 Early Y 2 years Mild N N NA N N NP 7 11.4 Early Y 10 months Mild <24/40> N Y 18 months TC NA N N NP Punctual hyperintensity in peritrigonal white matter suggesting local 8* 18.1 Early Y 4 years Mild Y 10 years M Y N Y Hypotonia Y infarct on T2-weighted sequence Perirolandic gliosis, cortical atrophy, hyperintensity in caudate head 9* 25.7 Early Y 2 years Mild Y Y 13 years TC Y Y Dystonia N Y and posterior putamen 10 0 Early NA NA NA NA NA NA 11 11.6 Early N Y N NA Y Spasticity N NP 12 16.7 Early Y Mild N Y NA Ab NA N N NP 13 9.2 Early Y 3 years Mild N N NA Y N NP 14 12.6 Early Y Mild N N NA Y Ataxia N NP 15 1.9 Early N N NA N N NP 16 4.5 Early Y 1 month Mild N NA N N NP 17 6.4 Early Y Mild N N NA N N NP 18 0 Early NA NA NA NA NA NP 19 1.9 Early Y Mild Y NA N N NP 20 3.8 Early Y 1 month Mild N NA N N NP 21 11 Early Y Mild N Y NA M NA N N NP 22 Lost Early N N NA N N N 23 2.7 Early NA N NA NA NA NP Ataxia, 24* 53.1 Late Y Mild N Y 5 years A, focal NA Y Y NP nystagmus Ataxia, 25* 24.6 Late Y Mild N Y 3 years Febrile sz NA Y N N tremor 26* 34.9 Late Y Mild N N NA N N NP 27* 28.1 Late Y Mild N N NA N N NP Tiredness, fatigable ptosis, proximal 28 15.9 Late Y 2 years Mild N Y 1 year TC Y Y Ataxia Y weakness in arms, N TR difficult to elicit. EMG Moderate Left Cortical and cerebellar atrophy with grey and white matter 29 4.7 Late Y 1 year Y 20 months A, TC, M Y Y N Y <12/33> hemiparesis involvement. Thalami atrophy. Multiple bilateral infarcts. 30* Lost Late Y Mild N N NA N N NP
425 Chapter 10 - Appendices
Neurological phenotype Developmental delay Epilepsy Abnormal Myopathic features Abnormal brain MRI Form
2015 Age at Severity (IQ) Mainstream Age at Type of Abnormal Age in Patient Patient
number Y/N Y/N Y/N Features Y/N Features Y/N Features
December December diagnosis
Table 21. Detailed neurological phenotype of patients included in the clinical study
Database related to Section 3.3. * Adult patient at end of follow-up in December 2015. For screened patients, the phenotype of the disease affected the familial proband is specified: (E) Early-onset, (L) Late-onset. A: Atonic; Ab: Absence;
.
426 Chapter 10 - Appendices
10.5 Publications related to this work and copyright clearance
10.5.1 Baruteau et al. Gene therapy for monogenic liver diseases: clinical successes, current challenges and future prospects. Journal of Inherited Metabolic Diseases, May 2017; In Press
This article is distributed with Springer Open Access Policy under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
427 Chapter 10 - Appendices
J Inherit Metab Dis DOI 10.1007/s10545-017-0053-3
SSIEM 2016
Gene therapy for monogenic liver diseases: clinical successes, current challenges and future prospects
1,2,3 3,4 5,6 1,2,7 Julien Baruteau & Simon N. Waddington & Ian E. Alexander & Paul Gissen
Received: 12 January 2017 /Revised: 27 April 2017 /Accepted: 28 April 2017 # The Author(s) 2017. This article is an open access publication
Abstract Over the last decade, pioneering liver-directed gene care of a subset of severe inherited genetic/metabolic liver therapy trials for haemophilia B have achieved sustained clin- diseases in the relatively near term. In this review, we aim to ical improvement after a single systemic injection of adeno- summarise the milestones in the development of gene therapy, associated virus (AAV) derived vectors encoding the human present the different vector tools and their clinical applications factor IX cDNA. These trials demonstrate the potential of for liver-directed gene therapy. AAV-derived vectors are AAV technology to provide long-lasting clinical benefit in emerging as the leading candidates for clinical translation of the treatment of monogenic liver disorders. Indeed, with more gene delivery to the liver. Therefore, we focus on clinical than ten ongoing or planned clinical trials for haemophilia A applications of AAV vectors in providing the most recent up- and B and dozens of trials planned for other inherited genetic/ date on clinical outcomes of completed and ongoing gene metabolic liver diseases, clinical translation is expanding rap- therapy trials and comment on the current challenges that the idly. Gene therapy is likely to become an option for routine field is facing for large-scale clinical translation. There is clearly an urgent need for more efficient therapies in many severe monogenic liver disorders, which will require careful Presented at the Annual Symposium of the Society for the Study of risk-benefit analysis for each indication, especially in Inborn Errors of Metabolism, Rome, Italy, September 6–9, 2016 paediatrics. Communicated by: Johannes Häberle
* Julien Baruteau Introduction [email protected] The liver is a key-regulator of multiple complex metabolic 1 Genetics and Genomic Medicine Programme, Great Ormond Street pathways and the hepatocyte is a primary cell type affected Institute of Child Health, University College London, London, UK in numerous inherited genetic/metabolic diseases (Clayton 2 Metabolic Medicine Department, Great Ormond Street Hospital for 2002). Despite a wide range of disease-specific conventional Children NHS Foundation Trust, London, UK therapies, liver replacement therapies remain a valid strategy 3 Gene Transfer Technology Group, Institute for Women’s Health, and even a potential cure for many monogenic liver disorders University College London, London, UK due to the ability to restore the defective pathway (Sokal 4 Wits/SAMRC Antiviral Gene Therapy Research Unit, Faculty of 2006). Liver replacement options include whole or partial or- Health Sciences, University of the Witwatersrand, gan (Spada et al 2009), or hepatocytes transplantation Johannesburg, South Africa (Dhawan et al 2006). The shortage of donors, the associated 5 Gene Therapy Research Unit, The Children’s Hospital at Westmead mortality/morbidity and need for immunosuppression, how- and Children’s Medical Research Institute, Westmead, Australia ever, often limit this option to severely affected patients and 6 Discipline of Child and Adolescent Health, University of Sydney, those aged more than 3 months or weighing greater than 5 kg Sydney, Australia (Haberle et al 2012). In the past decade, liver-directed gene 7 MRC Laboratory for Molecular Cell Biology, University College therapy has emerged as a promising alternative to transplan- London, London, UK tation in monogenic liver disorders.
428 Chapter 10 - Appendices
JInheritMetabDis
Overview of gene therapy development: reaching Strategies for hepatocyte-directed gene transfer maturity A growing toolbox is available for gene transfer, which has Gene therapy, by providing additional functional gene copies, been the preferred approach in recent human trials targeting has been considered for decades as an attractive option for hepatocytes. Various elements in the choice of transgene ex- treatment of monogenic disorders (Wirth et al 2013). pression cassette design, mode of delivery and the subset of According to the Gartner hype cycle, a graphical representation patients targeted influence the efficacy of gene therapy depicting the maturity of novel technologies, gene therapy (Fig. 1). reached its Bpeak of inflated expectation^ in the mid-1990s which was paralleled by a rapid rise in clinical trial activity Parameters of vector delivery and the publication of early proof-of-concept studies for genetic and acquired conditions such as adenosine-deaminase deficien- The mode of transgene delivery is crucial: i) local injection cy (ADA-SCID) (Blaese et al 1995;Bordignonetal1995)and allows highly selective expression, but in a limited area. brain tumours, respectively (Puumalainen et al 1998). This pe- Conversely an intravenous injection allows a broad distribu- riod of inflated expectation was critiqued in the Orkin- tion balanced by non-specificity. Injections in the hepatic ar- Motulsky report commissioned by the National Institute of tery or the portal vein improve the selectivity but require can- Health (Orkin and Motulsky 1995). While acknowledging the nulation with its associated risks (Fumoto et al 2013). extraordinary promise of gene therapy, the report emphasised Peripheral intravenous delivery provides similar transduction the need for greater focus on gene transfer technology and the compared to intrahepatic or intraportal routes for AAVvectors basic science of gene transfer. Soon after, the field plunged into (Sarkar et al 2006;Nathwanietal2007); ii) higher doses of its Btrough of disillusionment^ following the death of a young vector achieve greater transduction, but may generate more adult, Jesse Gelsinger, in a clinical trial for ornithine severe immune responses (Raper et al 2003; Mingozzi and transcarbamylase (OTC) deficiency (Raper et al 2003). High 2013). Optimism arising from the subsequent clinical success in the treatment of X-linked severe combined immunodeficiency Host pre-sensitisation or acquired immune responses (SCID-X1) (Cavazzana-Calvo et al 2000)wassoondampened by the occurrence of leukaemia in five out of 20 patients sec- The immune response against the vector and/or the transgene ondary to insertional mutagenesis (Hacein-Bey-Abina et al, product might preclude the expected therapeutic effect (Jooss 2003a, b;Fischeretal2010;MukherjeeandThrasher2013), and Chirmule 2003; Zaiss and Muruve 2008;WoldandToth causing the death of one participant (Mukherjee and Thrasher 2013). 2013). Resultant concerns over gene therapy were further Immune memory of pre-exposure to wild-type viruses can compounded by growing awareness of the challenges imposed prevent efficient hepatocyte transduction by pre-existing by vector-induced immune responses (Mingozzi and High neutralising antibodies and might account for differences in 2007). Disbelief and doubt followed, leading to a decline in the severity of immune responses observed after systemic in- financial investment (Ledley et al 2014). In parallel, these ad- jection. Pre-immunisation against the transgene product can verse events motivated researchers to seek a better understand- occur when the recombinant transgenic protein has been ad- ing of the challenges posed by disease pathophysiology and to ministered. For example, in haemophilia B patients, this can develop safer and more efficient vectors. Recent clinical suc- result in the generation of anti-factor IX antibodies when treat- cesses in various inherited orphan diseases such as Leber’s ed by recombinant factor IX (Armstrong et al 2014). congenital amaurosis (Bainbridge et al 2008;Cideciyanetal Accordingly prior immunisation needs to be carefully consid- 2008; Maguire et al 2008), X-linked adrenoleukodystrophy ered in clinical trial enrolment criteria. (Cartier et al 2009), metachromatic leukodystrophy (Biffi et al Acquired immune responses after systemic gene delivery 2013)andhaemophiliaB(Nathwanietal2011)andthefirst are common. Innate immune responses are triggered by anti- market authorisation granted by the European Medicines gen presenting cells such as dendritic cells or macrophages Agency (EMA) in 2012, to Glybera® for lipoprotein lipase initiating the release of proinflammatory cytokines (interleu- deficiency (Bryant et al 2013), are driving the field up the kins 1 and 6, tumour necrosis factor α (TNFα), type I inter- Bslope of enlightenment^ and onto the Bplateau of feron α and β) via stimulation of Toll-like receptors (TLR). productivity^.Asaresultofthissuccess, various biotechnology Specific and long-lasting antigen-specific immune responses companies dedicated to gene therapy development have been are mediated by B- and T-cellsandinvolvesecretionof created and received substantial financial investment (Cassiday neutralising antibodies and CD8+ cytotoxic T lymphocytes, 2014). In parallel, the number of gene therapy-based clinical respectively, regulated by the recruitment of helper and regu- trials has risen rapidly in recent years (http://www.abedia.com/ latory CD4+ T cells. Whatever the viral vector considered, wiley/(accessed 2017 Jan 06); Ginn et al 2013). immune responses share various similarities, and must always
429 Chapter 10 - Appendices
J Inherit Metab Dis
Fig. 1 The triad to consider for successful gene therapy VECTOR Vector type VECTOR DELIVERY CONSTRUCT (If virus-derived, serotype)
Dose of vector Regulatory Promoter Transgene elements Route of administration
HOST
Immune responses against
Vector Transgene product
be carefully considered (Bessis et al 2004; Tang et al 2006; non-immunogenic compared to some viral vector approaches Annoni et al 2013;CalcedoandWilson2013;Basner- (Wolff and Budker 2005). Mini-circles are devoid of plasmid Tschakarjan and Mingozzi 2014). backbone DNA; this may enhance transgene expression by overcoming heterochromatin formation and avoiding inflam- Design and selection of the gene transfer vector mation triggered by bacterial DNA (Mayrhofer et al 2009). These approaches allow easy production of therapeutic mate- The transgene expression cassette contains i) a transgene, rial with a good safety profile and capable of eliciting long- which is commonly a cDNA and may be codon-optimised lasting transgene expression in post-mitotic tissues (Wolff and to achieve higher expression of the transgene product, ii) an Budker 2005; Kay et al 2010). These approaches have been enhancer/promoter, the selection of which determines the lev- employed in ∼17% of gene therapy trials so far (http://www. el of transgene expression, cell-type restricted specificity of abedia.com/wiley/(accessed 2017 Jan 06)). The expression and also influences the risk of insertional mutagen- hydrodynamic injection technique, developed in small esis, iii) various pre- and/or post-regulatory elements to stabi- animal models, consists of injecting DNA plasmids or mini- lise transgene mRNA and therefore increase the yield of trans- circles in a large vehicle volume to flood the liver with gene product, e.g. addition of an intron downstream of the pressurised DNA solution; this disrupts vascular endothelium, promoter containing a bacterial replication origin (Lu et al and allows high levels of transgene expression in small or 2017) or the woodchuck hepatitis virus post-transcriptional large animal models (Liu et al 1999). Although hydrodynamic regulatory element (WPRE) (Lipshutz et al 2003), injections are difficult to translate to humans, intravascular respectively. hydrodynamic procedures with partial catheterisation for liver-directed gene delivery have shown some success in large animals (Sendra et al 2016;Yokooetal2016). However, Several options are available to deliver the transgene current translatable options of non-viral approaches remain expression cassette to the target cell/organ limited. Therefore viral vectors, acting as Trojan horses to increase transduction efficiency, are frequently considered. Injection of naked DNA, either as plasmids (Doenecke et al Non-viral vectors are synthetically produced biological 2010; Oishi et al 2016)ormini-circles (Viecelli et al 2014; particles, in which the transgene is encapsulated, or com- Hou et al 2016;Wuetal2016), is a simple mode of transgene plexed, and released at the target site. Various engineered delivery which lends itself to local delivery and is relatively nanoparticles exist, e.g. liposomes or/and polymers (Chira
430 Chapter 10 - Appendices
JInheritMetabDis et al 2015). These options have several advantages: easy pro- Protocols for lentiviral-mediated gene therapy have improved duction, no restriction of the transgene size, and a reliable with in vitro transduction efficacy reaching 90% in non- safety profile. Limitations are the stability of these particles, human and human hepatocytes (Nguyen et al 2009). cellular uptake and a limited ability to achieve long-lasting Hepatocyte transplantation, however, remains relatively inef- transgene expression (Elsabahy et al 2011). These are there- ficient and variable, likely due to poor engraftment, limited fore suboptimal delivery vehicles for liver-directed clinical persistence of engrafted hepatocytes and the lack of a prolif- trials. erative advantage (Gramignoli et al 2015). Virus-derived vectors represent an attractive approach based Retroviral-mediated in vivo gene therapy was well tolerat- on their relatively efficient transduction human cells. The main ed when vector was administered intravenously in vectors that have been used in clinical trials are derived from haemophilia A patients, but no significant clinical benefits adenoviruses (21%), retroviruses (excluding lentiviruses) were observed (Powell et al 2003). Preclinical studies failed (19%), adeno-associated viruses (7%) and lentiviruses (6%) to accurately predict the therapeutic dose and it was suggested (http://www.abedia.com/wiley/(accessed 2017 Jan 06)). that retroviral vectors were unable to transduce non-dividing hepatocytes (Chuah et al 2004). Therefore, alternative viral vectors, including lentiviral vectors, have been developed for Retroviral vectors treatment of hemophilias. Indeed two promising approaches with lentiviral vectors are in preclinical development: i) sys- Gamma-retroviruses such as murine leukaemia viruses temic lentiviral-mediated liver-restricted gene therapy in a dog (MLVs) are RNA viruses encoding gag, pol and env genes model of haemophilia B, which showed long-term efficacy, flanked by long terminal repeats (LTRs), which carry induction of liver tolerogenic properties in stimulating enhancers/promoter elements and are required for integration. CD4+CD25+FoxP3+ regulatory T cells and no evidence of After transduction of the target cell, reverse-transcription gen- genotoxicity in mice (Cantore et al 2015). The immune toler- erates a double-stranded DNA copy of the proviral genome, ance is of particular interest for patients with anti-FIX inhibi- which then integrates in the host genome providing long-term tors, who are currently excluded from gene therapy trials; ii) transgene expression. Gamma-retroviruses are unable to trans- ex vivo transduction of haematopoietic stem cells (HSC) in duce non-dividing cells as the nuclear membrane prevents mice with successful FIX gene expression to target cells of the retroviral vectors from entering the nucleus (Miller et al erythroid (Chang et al 2008) or the megakaryocyte lineage 1990). This explains why this vector is more often considered (Chen et al 2014). A lentiviral vector is currently in preclinical for ex vivo gene therapy in which cultured target cells are development for haemophilia B (Dolgin 2016). Lentiviral stimulated to replicate and then transduce. Lentiviruses are a vectors are able to accommodate large transgenes such as class of retroviruses, the most widely known of which is FVIII gene for haemophilia A (Kuether et al 2012). HIV1. Lentiviral vectors are able to transduce dividing and Lentiviral-mediated liver-directed ex vivo gene therapy has non-dividing cells, which broadens their application. been successfully reported in a pig model of tyrosinaemia 1 Retroviral vectors have relatively large transgene capacities using the selective advantage of Fah+/+ modified hepatocytes (7.5 kilobases (kb)) (Verma and Somia 1997). In fact payloads (Hickey et al 2016). exceeding 14 kilobases have been packaged into lentiviral The risk of insertional mutagenesis has been reported with vectors (Counsell et al, 2017). retroviral vectors (Cavazzana-Calvo et al 2000; Mukherjee Clinical applications and limitations: In an early gene ther- and Thrasher 2013), however, there is evidence that the risk apy trial, a γ-retroviral vector was used in an ex vivo approach is lower with lentiviral vectors (Kotterman et al 2015). To with autologous hepatocytes in five patients with homozygous improve safety, self-inactivating (SIN) vectors have been de- familial hypercholesterolemia. This showed a mild improve- veloped in which LTR enhancer/promoter elements in the U3 ment of lipid profiles in two patients with a very low rate of region have been deleted (Miyoshi et al 1998; Zufferey et al stable engraftment at 4 months after gene therapy (Grossman 1998). This, however, does not completely eliminate the risk et al 1994;Grossmanetal1995). Several limitations emerged of insertional mutagenesis as heterologous enhancer-promoter from this trial: i) the need for two invasive procedures, i.e. the elements still need to be included in vector constructs. surgical resection of a liver lobe to obtain sufficient primary Genotoxicity has been observed after foetal injections of hepatocytes for transduction as hepatocytes cannot be expand- non-primate and primate SIN-lentiviral vectors (Nowrouzi ed in culture and reinjection of transduced hepatocytes into the et al 2013; Condiotti et al 2014). So far, more than 125 patients portal circulation via a local catheter, with the associated risks over 14 years have been treated with haematopoietic stem of venous thrombosis, catheter misplacement and haemor- cells or T cells transduced by lentiviral vectors with no onco- rhage (Grossman et al 1994; Raper et al 1996). The efficiency genic event reported (Cartier et al 2009;Biffietal2013; with which hepatocytes were harvested and transduced was McGarrity et al 2013; Booth et al 2016). Another approach low, 30% and 10% respectively (Grossman et al 1995). relies on the mutation of the integrase protein to generate non-
431 Chapter 10 - Appendices
J Inherit Metab Dis 12) 57) – 13) – 53) – 12) – 6) – – 53) – 5/56 (9%) 116 ± 9 1(0 11 2.5 (0 48/52 (92%) 22/52 (42%) 1 7/52 (13%) 25/51 (49%) 10/48 (21%) 12.4 (0 2(0.1 5.5 (0.7 9/52 (17%) 12/23 (52%) 8/56 (14%) 15.6 (1.9 28/51 (55%) 169 ± 37 26/56 (46%) 7/56 (12%) 19/56 (34%) 56 (20 = 36%) 21/52 (40%) 530 ± 85 17/51 (33%) 12/32 (38%) 2years(0 3/12 (25%) 1.2 ± 0.07 356 ± 62 143 ± 31 46/56 (82%) 201 ± 20 25/48 (52%) 191 ± 15 42/47 (89%) 47/47 (100%) 31/25 11) – ia-total includes patients with 18.2) 18.2) 4) 9) – – – – 7) – 0/10 (0%) 0/8 (0%) 7(3 0/8 (0%) 134 ± 15 3/8 (38%) 5/8 (62%) / 2/4 (50%) 5/8 (62%) 07.5 (0.9 6/8 (75%) 3.1 (2 8.5 (8 0 1/7 (14%) 2antenatally 7/8 (88%) / 2/5 (40%) 2/10 (20%) 1/10 (10%) 84 ± 18 4/10 (40%) 181 ± 50 6/7 (86%) 6/7 (72%) NA 10 (1 = 10%) 8neonatally 15.6 (8 15.6 (8 6/10 (60%) 5/10 (50%) 1/2 (50%) 238 ± 206 1.4 ± 0.08 251 ± 45 167 ± 17 8/2 9/10 (90%) 7/7 (100%) 12) – sreflectthelasttenmeasurementsperformedduring 57) 53) – 6) 11) – 53) – – – 10 (50%) /23 (13%) 5/23 (22%) 3/23 (13%) 6/23 (26%) 23 (1 2/23 (9%) 102 ± 12 11/23 (48%) / 2/22 (9%) 13.0 4/23 (18%) 23/23 (100%) 2.5 (1 2(0.7 0 3/23 (13%) 3 4/23 (17%) 23 (16.7 3/20 (15%) 5/ 0/3 (0%) 15.1 (1 2/22 (9%) 212 ± 67 2/9 (22%) 2.75 years (0.25 4/23 (17%) 1/23 (4%) 7.3 ± 2.2 81 ± 24 16/20 (80%) 57 ± 12 234 ± 64 20/20 (100%) 23 (16 = 70%) 155 ± 25 134 ± 33 11/12 21/23 (91%) 1.2 ± 0.1 er figures show mean ± standard error. Hypokalaem 8) – 6) 25.7) 25.7) – 12) 4) 13) – – – – – 19 (89%) /21 (86%) /19 (37%) 12 ± 92 12/23 (52%) 2.5 (0 0.15 (0 126 ± 19 5 10/21 (48%) 11 0/23 (0%) 9(1.5 17/21 (81%) 18/21 (86%) 8/18 (44%) 2/7 (29%) 1 7 8/21 (38%) 17/ 4days(2 11 (1.9 2(0.1 11 (1.9 4/21 (19%) 5/9 (56%) 861 ± 120 3/21 (14%) 16/23 (70%) / 238 ± 77 20/20 (100%) 200 ± 50 23 (3 = 10%) 18 13/23 (56%) Early onset Late onset Screened Total 2/23 (9%) 3/23 (13%) 12/11 239 ± 28 215 ± 18 20/20 (100%) 16/23 (73%) 1.2 ± 0.14 nine and argininosuccinic acid concentration nted as median ± range. Oth onset and screened patients mol/L μ mol/L μ 50 IU/L) – y if <28 days of life;if <50 >28 days of life) Severe diarrhoea ALT (RI 20 Intermittent Trichorrhexis nodosa Poor corticomedullar differentiation Persistent Arterial hypertension Age when first reported (years) Age when first reported (years) Abnormal brain MRI Age when first reported (years) Raised ALT Age when first reported (years) Myopathic features Enlargement (>95th centile) Hepatomegaly Mean follow-up (years) Developmental delay Age when first reported (years) Epilepsy Age when first reported (years) Ataxia Hypokalaemia: total Patients lost Frequency Age Age (years) Frequency Na benzoate (mg/kg/day) Ammonia (RI < 100 L-arginine (mg/kg/day) Na phenylbutyrate (mg/kg/day) Patients still living Sex (M/F) Consanguinity Number (adult) Daily protein allowance (g/kg/day) Frequenc Frequency dcohorts:early-onset,late- range interval etabolic state on their standard treatment RI w-up is considered until December 2015. Plasma argi currence of symptom and duration of follow-up are prese not available, mol/L) μ NA mol/L) μ (RI <5 Plasma argininosuccinic acid Plasma arginine (RI 30-126 Kidney appearance (ultrasound) Miscellaneous Liver Neurology Na benzoate supplementation Protein restricted diet Na phenylbutyrate supplementation L-arginine supplementation Epidemiological and clinical data for the three analyse alanine aminotransferase, Biology Phenotype Follow up At diagnosis Therapeutics Table 1 Epidemiology Age at diagnosis, currently, at firstintermittent oc and persistent hypokalaemia. Follo follow-up when patients were in aALT compensated m
432 Chapter 10 - Appendices
JInheritMetabDis
BandPompedisease(Rastalletal2016)innonhuman and the Parvoviridae family. Initially identified as a contam- primates. inant of an adenoviral preparation (Atchison et al 1965), the However, the acute innate immune response directed against virus was later shown to require co-infection with a helper capsid proteins is not abolished in HD-Ad vectors (Muruve et al virus to replicate. In the absence of helper virus, AAV can 2004). For both first-generation and helper-dependent adenoviral enter target cells and establish latent infection through geno- vectors, this acute immune response is dose-dependent and can mic integration and/or formation of episomes. AAVis widely be lethal at high doses in non human primates (Morral et al 2002; considered non-pathogenic and has yet to be definitively Brunetti-Pierri et al 2004). Differences in the severity of the linked to disease causation. AAVs can transduce dividing immune response between species have emerged due to variable and non-dividing cells. The seroprevalence against the most interactions between blood cells and hepatic microarchitecture common serotype AAV2 is 40–60% (Louis Jeune et al 2013). such as size of liver sinusoidal fenestration (Piccolo and AAVvirions consist of an icosahedral capsid of approximately Brunetti-Pierri 2014). The activation of this innate immunity is 22 nm in diameter enclosing a 4.7 kb single-stranded genome. multifactorial. Adenoviral particles can trigger the immune re- The genome is flanked by two 145 nucleotide inverted termi- sponse by binding to Toll-like receptors (TLR2, TLR9) at the nal repeats (ITRs) containing all of the necessary cis-acting surface of antigen presenting cells, and/or activate the comple- functions for proviral rescue, genome replication and packag- ment cascade in the bloodstream (Kiang et al 2006;Zhuetal ing. The viral genome encodes 4 Rep proteins required for 2007). Kupffer cells recognise the adenoviral capsid either via proviral rescue and genome replication, and three viral pro- antibody-mediated opsonisation or in binding complement fac- teins VP1, VP2 and VP3, which assemble to form the capsid tors. Kupffer cells develop a pro-inflammatory state with necrotic (Fig. 2A) (Samulski and Muzyczka 2014). Some, but not all, death, which further disseminate the immune response AAV capsid serotypes (Earley et al 2017) require expression (Schiedner et al 2003). of an assembly-activating protein (AAP) encoded by an alter- In vivo HD-Ad mediated gene therapy has been performed native reading frame of the Cap gene and providing scaffold- in one patient in a phase I trial for haemophilia A. After a ing activity (Naumer et al 2012). Numerous AAV serotypes single intravenous low-dose injection, the patient developed have been isolated from humans, non-human primates and flu-like symptoms with transient fever, chills, back pain, head- other species, with the viral capsid determining species and ache and transient biological abnormalities including throm- target cell tropism through interaction with a diversity of cell bocytopenia, laboratory features of disseminated intravascular surface receptors/co-receptors and intracellular trafficking coagulopathy, increase in interleukin 6 levels, and elevated pathways that remain incompletely understood. A multi- liver transaminase levels peaking at 7 days (marked as grade serotype AAV receptor has been recently identified (Pillay 3 liver toxicity) (Chuah et al 2004;WhiteandMonahan2005; et al 2016), but its precise role in uptake and trafficking has Chandler and Venditti 2016). The patient expressed 1% FVIII yet to be elucidated (Summerford and Samulski 2016). For for some months but the trial was halted for safety reason example, AAV3B uses the human hepatocyte growth factor although biological abnormalities came back to normal within (hHGF) receptor, which restricts transduction to primates and 19 days. Unfortunately, this trial has not been published in a especially to the liver (Vercauteren et al 2016). peer-reviewed format and few details are available (Piccolo Since 2004, AAV vectors have emerged as the leading and Brunetti-Pierri 2014). The cause of these symptoms re- candidates for gene therapy in monogenic liver disorders mains unclear although, as supposed for the OTC trial involv- with the best accepted benefit-risk ratio (Dolgin 2016). ing Jesse Gelsinger, the innate immune response against the Therefore, the following sections focus on this gene adenoviral capsid and its subsequent release of cytokines has transfer approach detailing clinical successes and current been suspected (Chuah et al 2004). Contamination by an ad- limitations. enoviral helper virus remains a possible explanation (Chuah AAV2 is the most widely studied serotype and was the et al 2004). first to be vectorized. To generate a recombinant AAV vec- Despite limited application for liver monogenic disorders, tor the native Rep and Cap genes are removed and replaced adenoviral vectors have been successfully used for oncolytic by a transgene expression cassette with only the flanking virotherapy (Rosewell Shaw and Suzuki 2016) and vaccina- ITRs retained (Fig. 2A). Recombinant virus is produced by tion (Majhen et al 2014), which exploits adenoviral supplying Rep and Cap and necessary adenoviral helper immunogenicity. functions in trans.AmajordevelopmentinAAVvector technology was the demonstration that recombinant AAV2 genomes can be cross-packaged, or pseudo- Adeno-associated viral vectors serotyped, with the capsids from other AAV serotypes (Rabinowitz et al 2002). This has dramatically broadened Adeno-associated viruses (AAV) are non-enveloped, single- the cell types that can be efficiently targeted with AAV stranded DNA viruses that belong to the Dependovirus genus vectors. For example, pseudo-serotyping a recombinant
433 Chapter 10 - Appendices
J Inherit Metab Dis
Fig. 2 Synthesis of an AAV a vector. (a) Initially, the single- stranded proviral DNA is excised Wild type to remove Rep and Cap genes AAV from different wild type AAV AAV1 AAV2 AAV3B AAV4 AAV5 AAV6 AAV7 AAV8 AAV9 serotypes. The transgene expression cassette containing the promoter, the transgene and various regulatory elements is Wild type AAV2 cloned between the 2 ITRs, which Capsid Proviral are the only wild type AAV DNA sequences retained. (b)Forvector synthesis, triple transfection of three plasmids is performed in a packaging cell with proviral plasmid encoding the recombinant viral genome, a ITR Rep Cap ITR plasmid containing Rep and Cap and a helper plasmid. BPseudotyped^ AAV vectors Regulatory Promoter Transgene contain ITRs from a specific AAV elements ITR ITR serotype (usually AAV2) and a Cap gene encoding viral proteins Transgene (VP1, 2 and 3) from a different expression serotype (e.g AAV8) in order to cassette ITR Transgene expression cassette ITR provide organ-specific transduction of the recombinant AAV vector named AAV2/8. AAV: adeno-associated virus; b Adv: adenovirus; ITR: inverted terminal repeat Transgene AAV2 ITRs ITR expression ITR cassette AAV8 Rep Rep Cap & Cap genes Adv helper plasmid Pseudotyped AAV8 Cap AAV2/8 vector
AAV2 vector genome with the AAV8 capsid (designated Clinical successes of liver-directed AAV-mediated AAV2/8) enhances tropism for hepatocytes, particularly in gene therapy the mouse (Fig. 2B). AAV vectors bind to target cells via specific receptors and co-receptors that differ in a capsid- A rapidly increasing number of publications have reported dependent manner and are taken up by endocytosis or proof-of-concept for AAV-based gene therapy in animal macropinocytosis, before being trafficked to the nucleus models for various inherited liver disorders including urea for capsid uncoating. The uncoated genomes can remain cycle defects, organic acidurias, phenylketonuria, glycogen in the nucleus in single-stranded form, be converted to storage disease type Ia, long chain fatty acid oxidation disor- double-stranded episomes or undergo genomic integration ders, homozygous familial hypercholesterolemia, primary (Fig. 3)(BerryandAsokan2016). Conversion of input hyperoxaluria type I and progressive familial intrahepatic cho- single-stranded genomes to double-stranded transcription- lestasis (Hastie and Samulski 2015; Junge et al 2015). ally active forms occurs with variable efficiency in differ- In parallel, pioneering trials have been conducted since the ent cell types. Self-complementary (sc)vectorsdifferfrom 2000s, two of which targeted haemophilia B. This disease is single-stranded vectors (ss)inthattheycontainaself- an attractive target for gene therapy as an increase in plasma complementary transgene cassette that folds back on itself factor IX (FIX) of as little as 1% can confer significant phe- to form double-stranded DNA thereby bypassing the re- notypic improvement. Haemophilia B is a burden in public quirement for second strand synthesis, which is considered healthcare systems with an annual cost of $300,000 for se- as a rate-limiting step for transgene expression. As a con- verely affected patients (Angelis et al 2015). sequence the packaging size of the transgene cassette in In 2004, a ssAAV2.ApoE/hAAT.hFIX vector, administered scAAV is reduced by half (McCarty 2008). via the hepatic artery, showed a transient increase of plasma
434 Chapter 10 - Appendices
JInheritMetabDis
Fig. 3 AAV vector uptake, in-cell processing and initiation of the integrated single- or double-stranded episome (99%) or (small immune response. Fenestrated endothelium of hepatic sinusoids allows percentage) integrates into the host genome (1%). Expression of the the AAV vector to freely reach the hepatocyte. Once reaching the target transgene is followed by synthesis of the protein of interest. Cell- cell, the vector binds an extracellular receptor and co-receptor specific to mediated immune responses are initiated by the degradation of capsid the capsid motifs. After an uptake by endocytosis, the vector is trafficked or the transgene product (protein) in the proteasome and presentation at in the cytoplasm in early then late endosome. Acidification of the the surface of the transduced cell via the major histocompatibility endosome modifies the capsid conformation. After endosomal escape, complex I. CD8+ T cells recognise the antigen at the cell surface and the AAV vector enters the nucleus via the nuclear pore complex. Capsid initiate the immune cascade. Neutralising antibodies bind to the vector uncoating and release of the proviral DNA precede the synthesis of the in the bloodstream and impair or prevent successful transduction of the 2nd strand of DNA. The viral genome then persists either as a non- organ target. MHC1: major histocompatibility complex I
FIX from <1% to 3–11% over 4 weeks followed by a gradual corticosteroids, but were associated with a decrease of 50– decline over 4–8 weeks concomitant with transient asymp- 70% in plasma FIX levels attributable to a cellular immune tomatic rise in transaminases levels (Manno et al 2006), later response against capsid epitopes (Nathwani et al 2014). recognised as T cell-mediated cytotoxicity (Mingozzi et al D’Avola et al, recently reported results of a trial of 2007). scAAV2/5.hAAT.hcoPBGD vector in acute intermittent por- In 2009, Nathwani et al, injected a scAAV2/8.LP1.hcoFIX phyria with peripheral intravenous delivery. No vector-related vector via a peripheral intravenous route and elicited a long- safety issues were reported and the rate of disease-related lasting (>5 years) increase of plasma FIX from <1% to 1–8% hospitalisation decreased, potentially as a consequence of (Nienhuis et al 2016). Elevated transaminases occurring 7– closer metabolic follow-up. No change was observed in the 10 weeks post-injection resolved after an oral course of levels of metabolic biomarkers (D’Avola et al 2016). This
435 Chapter 10 - Appendices
J Inherit Metab Dis might be explained by a less efficient liver transduction me- Controversies remain regarding the possible insertional diated by AAV5relative to AAV8and a reduced expression of mutagenic effects of wild type AAV. Detection of a clonal the episomal transgene compared to the endogenous gene of expansion of wild type AAV2 sequences in 11/193 HCCs interest (Baruteau et al 2017). However, no liver biopsy data within HCC-related genes (Nault et al 2015) initiated a pas- was available to address this assumption. sionate and unresolved debate about Bdriver^ or Bpassenger^ Preliminary reports from ongoing clinical trials have con- cancer-related genetic modifications (Berns et al 2015; firmed Nathwani’spromisingresultsforhaemophiliaB.Aftera Buning and Schmidt 2015). The cumulative safety experience single intravenous injection of AAV vectors with different cap- with the rapidly growing number of AAV-based trials sids encoding the FIX gene or its Padua FIX variant, which targeting the human liver, combined with the low rate of contains a gain-of-function mutation, reported stabilised plasma HCC-associated AAV integrations despite the high seroprev- FIX levels have ranged from 3 to 8% in the AMT-060 trial alence of wild type AAVin the human population (e.g. >50% sponsored by Uniqure (Miesbach et al 2016)andtheDTX-101 for AAV2) (Thwaite et al 2015)areconsistentwitha sponsored by Dimension Therapeutics (http://dimensiontx.com) favourable safety profile of AAV vectors. Nevertheless the to 20–44% in the high-dose cohort of the BAX 335 trial spon- findings of Nault et al warrant further studies and mandate sored by Shire (Monahan et al 2015)andintheSPK-9001trial close monitoring in ongoing human trials. sponsored by Spark Therapeutics/Pfizer (George et al 2016). In a haemophilia A gene therapy trial, BioMarin reported plasma fac- Immune response tor VIII (FVIII) from 4 to 60% in the high-dose group of the BMN 270 trial (Pasi et al 2016)(Table1). Importantly, endoge- After vector delivery, non-specific innate immunity trig- nous FVIII is primarily secreted by endothelial cells (Fahs et al gers both type I interferon signalling involved in trans- 2014). All the AAV-based trials have so far involved only adult gene silencing (Suzuki et al 2013)andthereleaseofpro- patients, who had an undetectablebaselinetitreofneutralising inflammatory cytokines (Jayandharan et al 2011). Highly- antibodies to the capsid (usually accepted cut-off of 1/5 serum specific and long-lasting adaptive immunity generates B- dilution). Monogenic liver disorders in AAV-based gene therapy and T-cell responses against the capsid and/or the trans- development pipelines of pharmaceutical companies include gene product (Fig. 3). Neutralising antibodies against the OTCD, glycogen storage disease type Ia, citrullinemia type I, capsid, even at low titers, inhibit transduction after sys- phenylketonuria, Wilson disease, methylmalonic acideamia and temic delivery (Jiang et al, 2006a, b). This barrier is of Crigler-Najjar syndrome (Kattenhorn et al 2016). substantial concern to gene therapy development and the ongoing liver-directed trials are recruiting only seronega- tive patients without neutralising antibodies against the Current challenges AAV capsid. This narrows the target population as the seroprevalence against liver-specific AAV serotypes Insertional mutagenesis ranges from 20 to 30% for AAV5, 6 and 8 to 50–60% for AAV2 (Louis Jeune et al 2013). Cross-reactivity be- Despite more than 170 AAV-based human trials approved, tween serotypes is commonly >50% (Boutin et al 2010). ongoing or completed (http://www.abedia.com/wiley/ This seroprevalence varies depending on geographic ori- (accessed 2017 Jan 06)), no tumorigenic events have been gin (Calcedo et al 2009)andage.Neonatesreceivema- reported so far. AAV vector genome mainly persists as ternal antibodies by transplacental transfer and acquired episome in the transduced cell with a relatively low with maternal milk, which are lost over the first months proportion of vector genomes undergoing integration of life. Thereafter, seroprevalence remains negligible until preferentially in transcriptionally active genes, damaged 3yearsofageafterwhichtheseroconversionratepro- DNA or enriched CpG islands (McCarty et al 2004). gressively increases until adulthood (Calcedo et al 2011; Experiments in neonatal mice have identified an increased risk Li et al 2012). of hepatocellular carcinoma (HCC) after systemic injection. Human CD8+ T-cell mediated immune responses are in- This risk increased with the enhancer/promoter activity, youn- volved in AAV hepatotoxicity and were initially encountered ger age at time of injection and vector dose (Donsante et al during the first haemophilia B trial (Manno et al 2006). Capsid 2007;Chandleretal2015;Chandleretal2017). Analysis of epitopes, presented via the major histocompatibility complex I integration sites identified a rodent-specific hotspot in the (MHC1), were shown to drive expansion of a pre-existing Rian locus. Integration studies from human trials have not pool of CD8+ memory T cells acquired during a previous shown such hotspots, but rather a genome wide integration co-infection of wild type AAV and helper virus (adenovirus pattern involving neither HCC-related genes nor the human or herpes virus for example). This response was dose- Rian homologue, Dlk1-Dio3 (Kaeppel et al 2013; Gil-Farina dependent (Mingozzi and High 2013) and could be stimulated and Schmidt 2016). by alternate capsids (Mingozzi et al 2007).
436 Chapter 10 - Appendices
JInheritMetabDis identifier Clinicaltrial. gov NA NCT00076557 NA NA NCT02082860 NCT02618915 NCT01054339 NCT01687608 NCT00430768 NCT00377416 NCT00004809 NCT00004498 NCT02484092 NCT00979238 NCT02396342 2015 2016 1994 1995 2014 2016 2006 2003 2004 2000 2003 et al et al 2011 2004 2006 2009 et al et al 1996 2003 2016 et al et al Avola ’ Manno et al Brantly et al Grossman Raper et al Status Reference Recruiting of treated patients 13 Terminated Powell et al 7TerminatedMannoetal 8TerminatedKayetal 6TerminatedMonahan 8TerminatedD 9TerminatedFlotteetal 9TerminatedBrantlyetal 12 Terminated Flotte et al 5TerminatedGrossman 18 Terminated Raper et al 7RecruitingGeorgeetal 10 Recruiting Nathwani 5RecruitingMiesbach Number website, the date of publication of the results is mentioned. 10e9 10e12vg/kg 10e13vg/kg 10e13vg 10e8vg/kg × 10e13vg 10e12vg/kg 10e12vg/kg 10e12vg/kg 10e11vg/kg 10e12vg/kg 10e13vg/kg × × × × 10e10vg/kg 1 Terminated Chuah et al 10e12 to 10e12 to 10e7 to × × × × × × × 10e11 to 10e11 to 10e11 to 10e10 to 10e12 to 10e9 to 10e11 to 10e11 vg/kg 10e12 to × × × × 2 1.8 4.4 1.8 6 3 6 6.9 hepatocytes 6 (Low dose) 2 2 × × × × × × × × × 4.3 8 2 2.8 5 6 2 6.9 2.1 1to3.3 2 5 2 5 clinicaltrials.gov (accessed 06/01/2017) and company websites. If the date of the start of the trial atic cular ous ous ous cular ous cular cular al (Ex vivo ap- proac- h) atic ous ous ous Intraven- Intrahep- Intramus- Intramus- Intramus- Intraport- Intrahep- Intraven- Intraven- Intraven- Route Dose OTC: ornithine transcarbamylase;clinicaltrial.gov PBGD: porphobilinogen deaminase.was not Information available sources: in The list of trials announced for 2017 is indicative and does not pretend to be exhaustive sHospital ’ Genetic Technologies Massachussets Massachussets Pennsylvania Michigan Ann Arbour Children Biopharma Applied University Digna Biotech Intraven- Shire University Spark Therapeutics Intraven- GenStar peutic Sponsor tor IX gene NA Intramus- factor IX gene gene Factor IX gene Uniqure AFactorIXgeneStJude NA PBGD gene NA LDL receptor gene University of NA OTC gene University of Product Thera BBAX335Paduamutant BNAFactorIXgeneAvigen apy products for liver monogenic disorders. Intraportal and porphyria familial hypercholesterolaemia transcarbamylase deficiency 1-antitrypsin NA1-antitrypsin hAAt NA hAAT 1-antitrypsin NA hAAT α Haemophilia A NAα FVIII gene Chiron Intraven- Haemophilia B SPK-9001 Padua mutant factor IX α Homozygous Retrovirus AAV Retrovirus 5 Viral vector Disease AAV2/5 Haemophilia B AMT-060 AAV2/rh10 Haemophilia B DTX101 Factor IX gene Clinical trials of gene ther 10 AAV1 intrahepatic routes of administrationrespectively. relate FVIII: to factor injection VIII;LDL: on FIX: low the factor density portal IX; lipoprotein; vein HD-adenovirus: MoML: and helper-dependent Moloney adenovirus; the murine hepaticYear leukaemia artery virus; NA: not(start of applicable; trial) 2006 AAV1 2004 AAV2 2004 HD-Adenovirus Haemophilia A NA FVIII gene 2003 MoMLV 2015 Engineered 2014 AAV2/5 Acute intermittent 2012 AAV2/8 Haemophilia 2009 AAV2/8 Haemophilia B N 2004 AAV2 Haemophilia 1999 AAV2 Haemophilia B NA Fac 1998 Adenovirus 5 Ornithine Table 1 1992 MoMLV 20
437 Chapter 10 - Appendices
J Inherit Metab Dis identifier Clinicaltrial. gov NCT02695160 NCT02991144 NCT02576795 NCT02651675 NCT03003533 NCT03001830 gov 2016 gov gov gov gov gov clinicaltrials. clinicaltrials. clinicaltrials. clinicaltrials. clinicaltrials. clinicaltrials. recruiti- ng Recruiting Recruiting Recruiting Recruiting Status Reference osed osed osed osed osed Undiscl- Undiscl- Undiscl- Undiscl- 9RecruitingPasietal of treated patients NA Not yet 10e12 to × 10e12vg/kg 10e11 to 10e13vg/kg 10e12vg/kg 10e13vg/kg 10e13vg/kg 10e13vg/kg 10e12vg/kg × 10e12 to 10e12 to × × × × × × × 10e12 to 10e12 to 10e11 to × × 4 and ZFN 5 5 2 1 6 7.5 6 × × × cDNA 4 2 1.6 6 2.5 Undisclosed Undiscl- 6 ous ous ous ous ous ous ous Intraven- Intraven- Intraven- Intraven- es Intraven- sciences apeutics Intraven- Therapeutics Pharmaceuticals Pennsylvania/Regen- xbio London Therapeutics/Bayer Dimension Sangamo Bioscin Sangamo Bio NA III gene University College 1gene AudentesTherapeutics peutic Sponsor Route Dose Number tor VIII gene Spark Ther integrating in the albumin locus via Zinc-finger-nuclease in the albumin locus via Zinc-finger-nuclease ctor VIII integrating actor VIII gene BioMarin Undisclosed Freeline Therapeutics FcoVIgn Shire 8FactorVIIIgene RGX-501 LDL receptor geneDTX301 University of OTC gene Dimension Therapeutics Intraven- MeuSIX ARSB gene AT342 UGT1A BSB-FIXFactorIX familial hypercholesterolaemia transcarbamylase deficiency VI Haemophilia A SPK-8011 Fac Haemophilia B FLT-180 V2/8 Crigler Najjar AAV AAV Viralvector Disease Product Thera AAV2/5 Haemophilia AAAV2/8 BMN 270 F AAV2/8 Homozygous Engineered Ornithine AAV2/8 Haemophilia A GO-8 Factor V Engineered AAV2/8AVVAAV2/6 Haemophilia A HaemophiliaA Haemophilia AAAV2/8 BAX-88 AA Mucopolysaccharidosis SB-525 DTX201 Fa FactorVIIIgene Dimension (continued) for 2017 Table 1 Year (start of trial) 2016 AAV2/6 Haemophilia Announced
438 Chapter 10 - Appendices
JInheritMetabDis
Hepatic lobule
Zonation
Pericentral Periportal Anatomic
Glutamine Central Urea cycle Functional synthesis vein
Mouse, Dog Non-human primate human
Preferentially transduced area after systemic delivery of AAV2/8 Portal triad Fig. 4 Species-related differences in transduction of the hepatic lobule by AAV vector compared with metabolic zonation for ammonia clearance: example with AAV2/8 vector
Hepatic tolerogenic properties involve the proliferation of a et al 2014). One currently controversial strategy to blunt specific T cell subset, CD4+CD25+FoxP3+ Treg cells (Cooper anti-capsid immune responses, is to co-inject Bfull^ et al 2009) interacting with Kupffer cells (Breous et al 2009). AAV vectors and Bempty^ capsids as decoys, in an at- Expansion of these cells suppresses cytotoxic immunity tempt to competitively bind existing antibodies (Tse against AAV-transduced hepatocytes and induces et al 2015). immunotolerance, e.g. after neonatal injection (Shi et al 2013). Regulatory tolerance requires continuous antigen pre- Optimised targeting sentation and has been successfully induced with transgenic proteins (Shi et al 2013;Perrinetal2016). In contrast, capsid The ideal AAV vector would exclusively transduce the proteins are rapidly eliminated in the proteasome (Berry and desired target cells. This would limit unwanted immune Asokan 2016) and therefore very unlikely to induce tolerance. responses, avoid ectopic transgene expression and fur- Various approaches aim to overcome these unwanted ther reduce the already low risk of germline transmis- immune responses in order either to treat seropositive sion. Extensive capsid-focused research based on high- patients or to prevent sensitisation against the AAV cap- throughput in vitro and in vivo screening of Blibraries^ sid, which would allow reinjection in the future. Capsid of new capsid variants is ongoing in order to optimise modification targeting specific epitopes can evade host vectors for specific applications. These Baccelerated immunity (Tseng and Agbandje-McKenna 2014). Any evolution^ libraries are generated using strategies such strategy optimising the transduction of the target organ as error-prone PCR (Kotterman and Schaffer 2014; such as optimised expression cassette design, or capsid Deverman et al 2016)andBcapsid shuffling^ with ran- modifications will in turn decrease the amount of vector dom cut-paste sequences of wild type cap genes (Kay required for a similar effect. The decrease in vector 2011;LouisJeuneetal2013;Choudhuryetal2016). dose will further reduce the immune response This approach is paying off by generating re-engineered (Mingozzi and High 2011). Various protocols involving AAV variants with increased transduction efficiency in transient immunosuppression have been proposed in primary human hepatocytes (Lisowski et al 2014). large animal models and humans. These include plasma- pheresis (Monteilhet et al 2011;Chicoineetal2014), Germline transmission monoclonal antiCD20 antibody (rituximab) (Mingozzi et al 2013;Cortietal2014), non-depleting antiCD4 The risk for this phenomenon is difficult to quantify. antibody (McIntosh et al 2012), sirolimus (Corti et al Insertion of AAV sequences into the genome of a gam- 2014), cyclosporine A (McIntosh et al 2012;Mingozzi ete could potentially interfere with normal foetal devel- et al 2012), tacrolimus with mycophenolate mofetil opment or promote tumorigenicity in the progeny. So (Chicoine et al 2014), proteasome inhibitors, e.g. far adult patients enrolled in AAV trials with systemic bortezomib (Monahan et al 2010)andcorticosteroids delivery have been required to use contraception. Vector (Flanigan et al 2013;Chicoineetal2014;Nathwani sequences have been detected transiently in semen of
439 Chapter 10 - Appendices
J Inherit Metab Dis treated patients in AAV2 (Manno et al 2006)orAAV8 AAV3B-derived vectors (AAV3-ST, AAV-LK03) are able (Nathwani et al 2014)trialswiththelatestclearanceof to transduce human hepatocytes approximately 10 times the vector observed at 12 weeks post-injection, quicker more efficiently than AAV2/8 whilst transduction of mu- in younger men, but not in an AAV5 trial (D’Avola rine hepatocytes is minimal (Lisowski et al 2014). et al 2016). Vector was observed in seminal fluid but not in motile sperm and spermatogonia aligned with previous studies (Arruda et al 2001;Coutoetal2004). The FRG mouse has a combination of tyrosinaemia type I and immunodeficiency phenotypes and is an attractive model Suboptimal animal models to study human hepatocytes in vivo with the intention of over- coming limitations due to species-specificity (Azuma et al The following examples demonstrate why animal experiments 2007). Human Fah+/+ hepatocytes have a selective growth provide limited value for predicting effects in human trials: advantage relative to the Fah-deficient native mouse hepato- cytes allowing human cell engraftment up to 90% of the liver – T-cell mediated cytotoxicity observed in the first mass (Azuma et al 2007). Moreover, this model can address haemophilia B trial was not predicted by experiments in disease-specific questions if engrafted with hepatocytes from mice, dogs or non-human primates (Pien et al 2009). patients with liver specific disorders. Recently, the even more Unlike research animals, humans are exposed to wild complex FRGN (Fah−/−/Rag2−/−/Il2rg−/−/NOD)mouse type AAV infections generating anti-AAV memory T model has been described in which FRG mice, developed in cells, which are reactivated at the time of vector exposure a non-obese diabetic (NOD) mouse strain, are simultaneously (Mingozzi et al 2007). co-transplanted with human hepatocytes and human – An over 80% rate of HCC was observed in a mouse haematopoietic stem cells (Wilson et al 2014). model of methylmalonic acidaemia injected neonatally with AAV2/8. Integration occurred in the Rian hotspot, Limited capacity of AAV vectors a rodent-specific locus absent in other vertebrate genomes (Chandler et al 2015). The single-stranded AAV vector can accommodate a trans- – The patterns of liver transgene expression in the hepatic gene cassette of approximately 4.6 to 5 kb (Hirsch et al lobule varies among different species. For example, using 2016). This capacity is reduced by half (2.3 kb) in self- the AAV8 capsid, transgene expression is predominantly complementary (i.e double-stranded) vectors. This is a major pericentral in mice and dogs and periportal in non-human limitation compared to non-viral or other common viral vec- primates (Fig. 4) (Bell et al 2011). This is of particular tors like lentiviral/retroviral vectors (up to at least 14 kb importance for liver diseases where metabolic zonation (Counsell et al 2017)) or helper-dependent adenoviruses (up underpins that certain metabolic functions occur predom- to at least 38.9 kb (Suzuki et al 2011)). To deliver oversized inantly in certain areas of the liver lobule, which is the transgenes, several approaches have been developed. functional unit of the liver. For example, the urea cycle Designing mini-promoters or mini-genes of interest can be activity mostly takes place in periportal hepatocytes successful (Yan et al 2015). Alternatively, dual AAV co- (Gebhardt and Matz-Soja 2014). To achieve adequate transduction has been successfully tested either with split control of severe hyperammonaemia in the OTCD mouse AAV or fragment AAV (Hirsch et al 2016). Split AAVs use model, therefore, requires a much higher than expected the inherent tendency for intermolecular genome association dose of AAV2/8 vector carrying OTC transgene, which observed with AAV genomes via either homologous recom- might be explained by the non-physiological pattern of bination (HR) or non-homologous end joining (NHEJ) to pro- liver transduction (Cunningham et al 2011). Thus, it is duce concatemers. The overlapping approach uses vectors A difficult to reliably extrapolate vector doses for human and B, which display a homology sequence to promote inter- translation from studies in mice in liver diseases with molecular homologous recombination (Duan et al 2001;Koo metabolic zonation like OTCD. et al 2014). In the trans-splicing approach, two splice sites in 3′ – AAV2/8 vectors are capable of transducing 100% hepa- cDNA of vector A and 5′ cDNA of vector B are recognised in tocytes in adult mice (Cunningham et al 2008), but data concatemerized provirus to generate the single DNA molecule from human trials in haemophilia B have shown an in- of the oversized gene of interest (Duan et al 2001). A combi- crease in plasma FIX only 2 to 8% (Nienhuis et al 2016), nation of the two approaches is known as the hybrid trans- suggesting much less efficient AAV2/8-mediated hepato- splicing technique (Trapani et al 2014). In fragment AAV, the cyte transduction in humans. Interestingly, most studies in transgene is not entirely encapsidated but only fragments of achimericFRG(Fah−/−/Rag2−/−/Il2rg−/−)mouse- different size, which can recombine on overlapping regions human liver model (Lisowski et al 2014;Wangetal (Hirsch et al 2016). A limitation common to these approaches 2015; Vercauteren et al 2016)showedthatAAV3Band is reduced functional transduction efficiency.
440 Chapter 10 - Appendices
JInheritMetabDis
Limited manufacturing capability as lifespan and ability to work (Orkin and Reilly 2016). Vouchers systems with longer financial incentives might be For the last couple of years, the rise in demand for good- another option (Schimmer and Breazzano 2016). manufacturing practice (GMP) AAV vectors for preclinical In parallel, the regulatory framework is evolving with the and clinical studies has created a bottleneck, delaying a num- progress of technology and the increasing experience being ber of projects. The industry is progressively taking up the gathered from human trials. The Food and Drug gauntlet and developing improved and innovative methods Administration (FDA) and the EMA have published recom- for vector production in larger bioreactors with optimised re- mendations for gene therapy products (Narayanan et al 2014). agents, purification techniques and packaging cell lines The need for shorter and less expensive paths to clinical trials (Clement and Grieger 2016; Grieger et al 2016). and conditional approval relying more on human data for safety and efficacy has now been recognised, as exemplified Gene therapy requires an innovative economic model by the FDA’s Breakthrough Therapies programme and the for success in modern healthcare EMA’s Adaptive Pathways and Priority Medicines (PRIME) schemes which were launched in 2012, 2014 and 2016, re- Patients with inherited metabolic disorders individually im- spectively (Mullard 2016). These more flexible pathways will pose a much heavier financial burden on the healthcare system need to be agreed by government funding bodies (Macaulay compared to the average person. For example, the lifetime 2015). cost for methylmalonic/propionic acidaemias and Gaucher disease is $1.5 and 5 million, respectively (Li et al 2015; Orkin and Reilly 2016). Gene therapy has the potential to Considerations for paediatric application achieve substantial savings. For example, a haemophilia B trial has shown that single injection of the gene therapy prod- Paediatric administration of gene therapy has several theoret- uct in a cohort of ten patients can save more than $2.5 million ical advantages (McKay et al 2011). These include prevention over three years for the healthcare system in the UK of early death or irreversible neurological sequellae, transduc- (Nathwani et al 2014). tion of stem/progenitor cells and possible avoidance of im- It is likely that to recover the investment in product devel- mune response as neonates have an immature immune system, opment, companies will need pricing gene therapy treatments while infants and children have lower rates of pre-existing ambitiously when their products reach market. However, the anti-AAV immunity. The potential for even earlier gene ther- cost of treatment will need to be affordable for public apy intervention has been explored in late gestation foetal healthcare systems. The first gene therapy product approved macaques by intrahepatic injection of scAAV2/8 and by the European Medicines Agency (EMA), Glybera®, is scAAV2/5 vectors. Plasma human FIX levels of 8–112% were marketed by Uniqure. This drug has been in development observed during a median follow-up period of 14 months for 8 years, with the initial developer going bankrupt and a without evidence hotspot integration and HCC (Mattar et al currently proposed market cost of $1.2 million per patient 2011). (Bryant et al 2013). GlaxoSmithKline‘s Strimvelis® is the There is a theoretical risk, however, of increased tumorige- second gene therapy product to reach the market and was nicity in the developing liver as observed in experiments per- approved by the EMA in 2016. This gene therapy product, formed in neonatal mice (Donsante et al 2007;Chandleretal which targets ADA-SCID has been in development for over 2015). A further challenge in the paediatric liver is the likely 16 years and the proposed cost is $665,000 per patient progressive loss of vector genomes over time in concert with (Hoggatt 2016; Schimmer and Breazzano 2016). hepatocellular proliferation. More than 90% of the AAV- Over the last 20 years, greater than $4.3 billion have been delivered transgene cassettes exists as non-integrated epi- spent on development of gene therapy technology and return somes (Cunningham et al 2008). The human liver weight on this investment is still awaited (Ledley et al 2014). doubles at 4 months, 16 months, 6 years and 12 years of Although most of the gene therapy programmes remain in age, which means that the adult liver is 16 times heavier than the early stages of development, healthcare economists are the neonatal liver (Coppoletta and Wolbach 1933;Sunderman generating models to cost treatments, which might provide and Boerner 1949). Therefore, it is unlikely that a neonatal lifelong cures. A pay-for-performance system has been pro- injection will be sufficient to provide lifelong correction of the posed with yearly-capped annuity paid to the pharmaceutical phenotype in metabolic liver diseases, with reinjection during company if criteria of a metabolic control of the disease are met the phase of rapid liver growth likely to be necessary. (Touchot and Flume 2015). These criteria might reflect cost- An alternative approach to reinjection could be the use of effectiveness and not only cost-saving. This approach values integrating vectors and or locus-specific genome engineering. the gain in quality of life estimated by quality-adjusted life Sangamo Therapeutics Inc. is developing tools for transgene years (QALY) analysis, which includes many parameters such integration into the albumin locus and uses zinc finger
441 Chapter 10 - Appendices
J Inherit Metab Dis nucleases coupled with AAV technology. Other genome injected (412 to 694 nM in the highest dose group) but far editing tools have been successfully tested with AAV vectors from the targeted protective level of 11 μM (Flotte et al 2011). in a neonatal OTC deficiency mouse model. In these experi- ments, two AAV vectors were injected simultaneously, one of which encoded the transgene and the other the enzymatic sys- Conclusion tem for integration or site specific cutting by Piggybac transposase (Cunningham et al 2015)andCRISPR-Cas9 Over the last decade, major discoveries in the understanding of (Yang et al 2016), respectively. viral vector biology have generated promising results in pioneering clinical trials for haemophilia B using AAV vectors. This has paved the way for a wider development of AAV- Muscle-directed gene therapy for liver monogenic mediated gene therapy for monogenic liver disorders. Although disorders several clinical, manufacturingandeconomicchallengesremain, this approach to treatment for severely debilitating diseases gen- Muscle-directed gene therapy has been developed for liver erated widespread enthusiasm shared by clinicians, researchers monogenic disorders with secreted protein such as and investors alike. Gene transfer technologies are reaching an haemophilia A and B and α1-antitrypsin deficiency. exciting threshold of efficacy and promise to revolutionise the Intramuscular injections might circumvent some caveats ob- management of many currently untreatable diseases. served with systemic injection, the common route of delivery for liver-targeted gene therapy: limited biodistribution with reduced risk of germline transmission, minimal exposure to Compliance with ethical standards circulating neutralising antibodies, reduced dose of vector for a similar effect. Conflict of interest J. Baruteau, S. N. Waddington, I. E. Alexander, and A proof of concept using AAV2 vectors in mice (Herzog P. Gissen declare that they have no conflict of interest. et al 1997)anddogs(Herzogetal1999)affectedby haemophilia B paved the way for a clinical trial (Kay et al Funding J.B. is funded by Action Medical Research for Children 2000; Manno et al 2003), which showed a safe profile with Charity (grant GN2137) and Great Ormond Street Hospital (GOSH) Children’s Charity Clinical Research Starter Grant. J.B., S.N.W. and long-standing expression in some patients (Jiang et al, 2006a, P.G. are in receipt of the UK Medical Research Council (MRC) grant b;Buchlisetal2012) but only a mild increase in plasma factor MR/N019075/1. S.N.W. is funded by the MRC grants MR/N026101/1 IX around 1% (Manno et al 2003). Depending on the dose and MR/P026494/1. considered, dozens to hundreds of intramuscular injections are Open Access This article is distributed under the terms of the Creative necessary, which makes this route particularly impractical. For Commons Attribution 4.0 International License (http:// instance, Manno et al administered between 10 to 90 injec- creativecommons.org/licenses/by/4.0/), which permits unrestricted use, tions per patients in lower limbs (Manno et al 2003). Similarly, distribution, and reproduction in any medium, provided you give appro- a proof of concept in mice with haemophilia A has been re- priate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. ported (Mah et al 2003). Three AAV-mediated clinical trials have been conducted for α1-antitrypsin deficiency. In this disorder, a plasma level of wild-type (M) α1-antitrypsin above 11 μMisconsidered References reducing the risk of developing emphysema. A first trial based on an AAV2 capsid showed an acceptable safety profile with Alba R, Bosch A, Chillon M (2005) Gutless adenovirus: last-generation adenovirus for gene therapy. Gene Ther 12(Suppl 1):S18 S27 mild local reactions at the site of intramuscular injection (red- – Andrews JL, Kadan MJ, Gorziglia MI, Kaleko M, Connelly S (2001) ness, tenderness, bruising) and a seroconversion against Generation and characterization of E1/E2a/E3/E4-deficient adeno- AAV2. Unfortunately, only one out of 12 patients demonstrat- viral vectors encoding human factor VIII. Mol Ther 3:329–336 ed a minimal increase of plasma M α1-antitrypsin at 82 nM Angelis A, Tordrup D, Kanavos P (2015) Socio-economic burden of rare (Flotte et al 2004; Brantly et al 2006). Two other trials (phase I diseases: a systematic review of cost of illness evidence. Health Policy 119:964–979 then phase II) were conducted with a vector based on AAV1 Annoni A, Goudy K, Akbarpour M, Naldini L, Roncarolo MG (2013) capsid known for its better muscle transduction compared to Immune responses in liver-directed lentiviral gene therapy. Transl AAV2 (Flotte et al 2011). In both trials, minor side effects and Res 161:230–240 a seroconversion against AAV1 were observed (Brantly et al Apolonia L, Waddington SN, Fernandes C et al (2007) Stable gene trans- 2009; Flotte et al 2011). A moderate infiltration of reactive T fer to muscle using non-integrating lentiviral vectors. Mol Ther 15: 1947–1954 lymphocytes in muscle biopsies was noticed (Flotte et al Armstrong EP, Malone DC, Krishnan S, Wessler MJ (2014) Costs and 2011). Plasma levels of M α1-antitrypsin were still mild, al- utilization of hemophilia a and B patients with and without inhibi- though improving in the latest trial due to higher doses tors. J Med Econ 17:798–802
442 Chapter 10 - Appendices
JInheritMetabDis
Arruda VR, Fields PA, Milner R et al (2001) Lack of germline transmis- dependent adenoviral vectors into nonhuman primates. Hum Gene sion of vector sequences following systemic administration of re- Ther 15:35–46 combinant AAV-2 vector in males. Mol Ther 4:586–592 Brunetti-Pierri N, Nichols TC, McCorquodale S et al (2005) Sustained Atchison RW, Casto BC, Hammon WM (1965) Adenovirus-associated phenotypic correction of canine hemophilia B after systemic admin- defective virus particles. Science 149:754–756 istration of helper-dependent adenoviral vector. Hum Gene Ther 16: Azuma H, Paulk N, Ranade A et al (2007) Robust expansion of human 811–820 hepatocytes in fah−/−/Rag2−/−/Il2rg−/− mice. Nat Biotechnol 25: Brunetti-Pierri N, Ng T, Iannitti DA et al (2006) Improved hepatic trans- 903–910 duction, reduced systemic vector dissemination, and long-term Bainbridge JW, Smith AJ, Barker SS et al (2008) Effect of gene therapy transgene expression by delivering helper-dependent adenoviral on visual function in Leber’s congenital amaurosis. N Engl J Med vectors into the surgically isolated liver of nonhuman primates. 358:2231–2239 Hum Gene Ther 17:391–404 Baruteau J, Waddington SN, Alexander IE, Gissen P (2017) Delivering Bryant LM, Christopher DM, Giles AR et al (2013) Lessons learned from efficient liver-directed AAV-mediated gene therapy. Gene Ther. doi: the clinical development and market authorization of Glybera. Hum 10.1038/gt.2016.90 Gene Ther Clin Dev 24:55–64 Basner-Tschakarjan E, Mingozzi F (2014) Cell-mediated immunity to AAV Buchlis G, Podsakoff GM, Radu A et al (2012) Factor IX expression in vectors, evolving concepts and potential solutions. Front Immunol 5:350 skeletal muscle of a severe hemophilia B patient 10 years after AAV- Belalcazar LM, Merched A, Carr B et al (2003) Long-term stable expres- mediated gene transfer. Blood 119:3038–3041 sion of human apolipoprotein A-I mediated by helper-dependent Buning H, Schmidt M (2015) Adeno-associated vector toxicity-to be or adenovirus gene transfer inhibits atherosclerosis progression and not to be? Mol Ther 23:1673–1675 remodels atherosclerotic plaques in a mouse model of familial hy- Calcedo R, Wilson JM (2013) Humoral immune response to AAV. Front percholesterolemia. Circulation 107:2726–2732 Immunol 4:341 Bell P, Wang L, Gao G et al (2011) Inverse zonation of hepatocyte trans- Calcedo R, Vandenberghe LH, Gao G, Lin J, Wilson JM (2009) duction with AAV vectors between mice and non-human primates. Worldwide epidemiology of neutralizing antibodies to adeno- Mol Genet Metab 104:395–403 associated viruses. J Infect Dis 199:381–390 Benihoud K, Yeh P, Perricaudet M (1999) Adenovirus vectors for gene Calcedo R, Morizono H, Wang L et al (2011) Adeno-associated virus delivery. Curr Opin Biotechnol 10:440–447 antibody profiles in newborns, children, and adolescents. Clin Berns KI, Byrne BJ, Flotte TR et al (2015) Adeno-associated virus type 2 Vaccine Immunol 18:1586–1588 and hepatocellular carcinoma? Hum Gene Ther 26:779–781 Cantore A, Ranzani M, Bartholomae CC et al (2015) Liver-directed Berry GE, Asokan A (2016) Cellular transduction mechanisms of adeno- lentiviral gene therapy in a dog model of hemophilia B. Sci Transl associated viral vectors. Curr Opin Virol 21:54–60 Med 7:277ra228 Bessis N, GarciaCozar FJ, Boissier MC (2004) Immune responses to Cartier N, Hacein-Bey-Abina S, Bartholomae CC et al (2009) gene therapy vectors: influence on vector function and effector Hematopoietic stem cell gene therapy with a lentiviral vector in X- mechanisms. Gene Ther 11(Suppl 1):S10–S17 linked adrenoleukodystrophy. Science 326:818–823 Biffi A, Montini E, Lorioli L et al (2013) Lentiviral hematopoietic stem Cassiday L (2014) Medical research: Gene-therapy reboot. Nature 509: cell gene therapy benefits metachromatic leukodystrophy. Science 651–653 341:1233158 Castello R, Borzone R, D’Aria S, Annunziata P, Piccolo P, Brunetti-Pierri Blaese RM, Culver KW, Miller AD et al (1995) T lymphocyte-directed N (2016) Helper-dependent adenoviral vectors for liver-directed gene therapy for ADA-SCID: initial trial results after 4 years. gene therapy of primary hyperoxaluria type 1. Gene Ther 23:129– Science 270:475–480 134 Booth C, Gaspar HB, Thrasher AJ (2016) Treating immunodeficiency Cavazzana-Calvo M, Hacein-Bey S, de Saint BG et al (2000) Gene ther- through HSC Gene therapy. Trends Mol Med 22:317–327 apy of human severe combined immunodeficiency (SCID)-X1 dis- Bordignon C, Notarangelo LD, Nobili N et al (1995) Gene therapy in ease. Science 288:669–672 peripheral blood lymphocytes and bone marrow for ADA- immu- Cerreto M, Mehdawy B, Ombrone D et al (2012) Reversal of metabolic nodeficient patients. Science 270:470–475 and neurological symptoms of phenylketonuric mice treated with a Boutin S, Monteilhet V,Veron P et al (2010) Prevalence of serum IgG and PAH containing helper-dependent adenoviral vector. Curr Gene neutralizing factors against adeno-associated virus (AAV) types 1, 2, Ther 12:48–56 5, 6, 8, and 9 in the healthy population: implications for gene therapy Chandler RJ, Venditti CP (2016) Gene therapy for metabolic diseases. using AAV vectors. Hum Gene Ther 21:704–712 Transl Sci Rare Dis 1:73–89 Brantly ML, Spencer LT, Humphries M et al (2006) Phase I trial of Chandler RJ, LaFave MC, Varshney GK et al (2015) Vector design in- intramuscular injection of a recombinant adeno-associated virus se- fluences hepatic genotoxicity after adeno-associated virus gene ther- rotype 2 alphal-antitrypsin (AAT) vector in AAT-deficient adults. apy. J Clin Invest 125:870–880 Hum Gene Ther 17:1177–1186 Chandler RJ, Sands MS, Venditti CP (2017) rAAV integration and Brantly ML, Chulay JD, Wang L et al (2009) Sustained transgene expres- genotoxicity: insights from animal models. Hum Gene Ther 28: sion despite T lymphocyte responses in a clinical trial of rAAV1- 314–322 AAT gene therapy. Proc Natl Acad Sci USA 106:16363–16368 Chang AH, Stephan MT, Lisowski L, Sadelain M (2008) Erythroid- Breous E, Somanathan S, Vandenberghe LH, Wilson JM (2009) Hepatic specific human factor IX delivery from in vivo selected hematopoi- regulatory T cells and Kupffer cells are crucial mediators of systemic T etic stem cells following nonmyeloablative conditioning in hemo- cell tolerance to antigens targeting murine liver. Hepatology 50:612–621 philia B mice. Mol Ther 16:1745–1752 Brown BD, Shi CX, Powell S, Hurlbut D, Graham FL, Lillicrap D (2004) Chen Y, Schroeder JA, Kuether EL, Zhang G, Shi Q (2014) Platelet gene Helper-dependent adenoviral vectors mediate therapeutic factor VIII therapy by lentiviral gene delivery to hematopoietic stem cells re- expression for several months with minimal accompanying toxicity stores hemostasis and induces humoral immune tolerance in in a canine model of severe hemophilia a. Blood 103:804–810 FIX(null) mice. Mol Ther 22:169–177 Brunetti-Pierri N, Ng P (2011) Helper-dependent adenoviral vectors for Chicoine LG, Montgomery CL, Bremer WG et al (2014) Plasmapheresis liver-directed gene therapy. Hum Mol Genet 20:R7–13 eliminates the negative impact of AAV antibodies on Brunetti-Pierri N, Palmer DJ, Beaudet AL, Carey KD, Finegold M, Ng P microdystrophin gene expression following vascular delivery. Mol (2004) Acute toxicity after high-dose systemic injection of helper- Ther 22:338–347
443 Chapter 10 - Appendices
J Inherit Metab Dis
Chira S, Jackson CS, Oprea I et al (2015) Progresses towards safe and Donsante A, Miller DG, Li Y et al (2007) AAV vector integration sites in efficient gene therapy vectors. Oncotarget 6:30675–30703 mouse hepatocellular carcinoma. Science 317:477 Choudhury SR, Fitzpatrick Z, Harris AF et al (2016) In vivo selection Duan D, Yue Y, Engelhardt JF (2001) Expanding AAV packaging capac- yields AAV-B1 capsid for central nervous system and muscle Gene ity with trans-splicing or overlapping vectors: a quantitative com- therapy. Mol Ther 24:1247–1257 parison. Mol Ther 4:383–391 Chuah MK, Collen D, VandenDriessche T (2004) Clinical gene transfer Earley LF, Powers JM, Adachi K et al (2017) Adeno-associated virus studies for hemophilia a. Semin Thromb Hemost 30:249–256 (AAV) assembly-activating protein is not an essential requirement Cideciyan AV,Aleman TS, Boye SL et al (2008) Human gene therapy for for capsid assembly of AAV serotypes 4, 5, and 11. J Virol 91: RPE65 isomerase deficiency activates the retinoid cycle of vision e01980-16 but with slow rod kinetics. Proc Natl Acad Sci USA 105:15112– Ehrhardt A, Kay MA (2002) A new adenoviral helper-dependent vector 15117 results in long-term therapeutic levels of human coagulation factor Clayton PT (2002) Inborn errors presenting with liver dysfunction. Semin IX at low doses in vivo. Blood 99:3923–3930 Neonatol 7:49–63 Ehrhardt A, Xu H, Dillow AM, Bellinger DA, Nichols TC, Kay MA Clement N, Grieger JC (2016) Manufacturing of recombinant adeno- (2003) A gene-deleted adenoviral vector results in phenotypic cor- associated viral vectors for clinical trials. Mol Ther Methods Clin rection of canine hemophilia B without liver toxicity or thrombocy- Dev 3:16002 topenia. Blood 102:2403–2411 Condiotti R, Goldenberg D, Giladi H et al (2014) Transduction of fetal Elsabahy M, Nazarali A, Foldvari M (2011) Non-viral nucleic acid de- mice with a feline lentiviral vector induces liver tumors which ex- livery: key challenges and future directions. Curr Drug Deliv 8:235– hibit an E2F activation signature. Mol Ther 22:59–68 244 Cooper M, Nayak S, Hoffman BE, Terhorst C, Cao O, Herzog RW (2009) Improved induction of immune tolerance to factor IX by hepatic Fahs SA, Hille MT, Shi Q, Weiler H, Montgomery RR (2014) A condi- tional knockout mouse model reveals endothelial cells as the prin- AAV-8 gene transfer. Hum Gene Ther 20:767–776 cipal and possibly exclusive source of plasma factor VIII. Blood Coppoletta JM, Wolbach SB (1933) Body length and organ weights of 123:3706 3713 infants and children: a study of the Body length and normal weights – of the more important vital organs of the Body between birth and Fischer A, Hacein-Bey-Abina S, Cavazzana-Calvo M (2010) 20 years of twelve years of age. Am J Pathol 9:55–70 gene therapy for SCID. Nat Immunol 11:457–460 Corti M, Elder M, Falk D et al (2014) B-cell depletion is protective Flanigan KM, Campbell K, Viollet L et al (2013) Anti-dystrophin T cell against anti-AAV capsid immune response: a human subject case responses in Duchenne muscular dystrophy: prevalence and a glu- study. Mol Ther Methods Clin Dev 1:14033 cocorticoid treatment effect. Hum Gene Ther 24:797–806 Counsell JR, Asgarian Z, Meng J, Ferrer V, Vink CA, Howe SJ, Flotte TR, Brantly ML, Spencer LT et al (2004) Phase I trial of intramus- Waddington SN, Thrasher AJ, Muntoni F, Morgan JE, Danos O cular injection of a recombinant adeno-associated virus alpha 1- (2017) Lentiviral vectors can be used for full-length dystrophin gene antitrypsin (rAAV2-CB-hAAT) gene vector to AAT-deficient adults. therapy. Sci Rep 7:79 Hum Gene Ther 15:93–128 Couto L, Parker A, Gordon JW (2004) Direct exposure of mouse sper- Flotte TR, Trapnell BC, Humphries M et al (2011) Phase 2 clinical trial of matozoa to very high concentrations of a serotype-2 adeno-associ- a recombinant adeno-associated viral vector expressing alpha1- ated virus gene therapy vector fails to lead to germ cell transduction. antitrypsin: interim results. Hum Gene Ther 22:1239–1247 Hum Gene Ther 15:287–291 Fumoto S, Kawakami S, Hashida M, Nishida K (2013) Targeted Crane B, Luo X, Demaster A et al (2012) Rescue administration of a gene delivery: importance of administration routes. In: Good helper-dependent adenovirus vector with long-term efficacy in dogs D, Wei M (eds.) Novel gene therapy approaches. doi: with glycogen storage disease type Ia. Gene Ther 19:443–452 10.5772/54741 Cunningham SC, Dane AP, Spinoulas A, Logan GJ, Alexander IE (2008) Gao GP, Yang Y, Wilson JM (1996) Biology of adenovirus vectors with Gene delivery to the juvenile mouse liver using AAV2/8 vectors. E1 and E4 deletions for liver-directed gene therapy. J Virol 70:8934– Mol Ther 16:1081–1088 8943 Cunningham SC, Kok CY, Dane AP et al (2011) Induction and preven- Gebhardt R, Matz-Soja M (2014) Liver zonation: novel aspects of its tion of severe hyperammonemia in the spfash mouse model of orni- regulation and its impact on homeostasis. World J Gastroenterol thine transcarbamylase deficiency using shRNA and rAAV- 20:8491–8504 mediated gene delivery. Mol Ther 19:854–859 George L, Sullivan S, Giermasz A (2016) Adeno-associated virus medi- Cunningham SC, Siew SM, Hallwirth CV et al (2015) Modeling correc- ated gene transfer for Hemophilia B achieves sustained mean factor tion of severe urea cycle defects in the growing murine liver using a IX activity levels of >30% without immunosuppression. 58th annual hybrid recombinant adeno-associated virus/piggyBac transposase meeting of the American Society of Hemophilia gene delivery system. Hepatology 62:417–428 Gil-Farina I, Schmidt M (2016) Interaction of vectors and parental viruses D’Avola D, Lopez-Franco E, Sangro B et al (2016) Phase I open label with the host genome. Curr Opin Virol 21:35–40 liver-directed gene therapy clinical trial for acute intermittent por- Ginn SL, Alexander IE, Edelstein ML, Abedi MR, Wixon J (2013) Gene phyria. J Hepatol 65:776–783 therapy clinical trials worldwide to 2012 - an update. J Gene Med Deverman BE, Pravdo PL, Simpson BP et al (2016) Cre-dependent se- 15:65–77 lection yields AAVvariants for widespread gene transfer to the adult brain. Nat Biotechnol 34:204–209 Gramignoli R, Vosough M, Kannisto K, Srinivasan RC, Strom SC (2015) Dhawan A, Mitry RR, Hughes RD (2006) Hepatocyte transplantation for Clinical hepatocyte transplantation: practical limits and possible so- liver-based metabolic disorders. J Inherit Metab Dis 29:431–435 lutions. Eur Surg Res 54:162–177 Doenecke A, Kromer A, Scherer MN, Schlitt HJ, Geissler EK (2010) Grieger JC, Soltys SM, Samulski RJ (2016) Production of recombinant AAV plasmid DNA simplifies liver-directed in vivo gene therapy: adeno-associated virus vectors using suspension HEK293 cells and comparison of expression levels after plasmid DNA-, adeno- continuous harvest of vector from the culture media for GMP FIX associated virus- and adenovirus-mediated liver transfection. J and FLT1 clinical vector. Mol Ther 24:287–297 Gene Med 12:810–817 Grossman M, Raper SE, Kozarsky K et al (1994) Successful ex vivo gene Dolgin E (2016) Early clinical data raise the bar for hemophilia gene therapy directed to liver in a patient with familial hypercholesterol- therapies. Nat Biotechnol 34:999–1001 aemia. Nat Genet 6:335–341
444 Chapter 10 - Appendices
JInheritMetabDis
Grossman M, Rader DJ, Muller DW et al (1995) A pilot study of ex vivo Kay MA, He CY, Chen ZY (2010) A robust system for production of gene therapy for homozygous familial hypercholesterolaemia. Nat minicircle DNA vectors. Nat Biotechnol 28:1287–1289 Med 1:1148–1154 Kiang A, Hartman ZC, Everett RS et al (2006) Multiple innate inflam- Haberle J, Boddaert N, Burlina A et al (2012) Suggested guidelines for matory responses induced after systemic adenovirus vector delivery the diagnosis and management of urea cycle disorders. Orphanet depend on a functional complement system. Mol Ther 14:588–598 Journal of Rare Diseases 7:32 Kim IH, Jozkowicz A, Piedra PA, Oka K, Chan L (2001) Lifetime cor- Hacein-Bey-Abina S, von Kalle C, Schmidt M et al (2003a) A serious rection of genetic deficiency in mice with a single injection of adverse event after successful gene therapy for X-linked severe com- helper-dependent adenoviral vector. Proc Natl Acad Sci USA 98: bined immunodeficiency. N Engl J Med 348:255–256 13282–13287 Hacein-Bey-Abina S, Von Kalle C, Schmidt M et al (2003b) LMO2- Koeberl DD, Sun B, Bird A, Chen YT, Oka K, Chan L (2007) Efficacy of associated clonal T cell proliferation in two patients after gene ther- helper-dependent adenovirus vector-mediated gene therapy in mu- apy for SCID-X1. Science 302:415–419 rine glycogen storage disease type Ia. Mol Ther 15:1253–1258 Hastie E, Samulski RJ (2015) Recombinant adeno-associated virus vec- Koo T, Popplewell L, Athanasopoulos T, Dickson G (2014) Triple trans- tors in the treatment of rare diseases. Expert Opin Orphan Drugs 3: splicing adeno-associated virus vectors capable of transferring the 675–689 coding sequence for full-length dystrophin protein into dystrophic Herzog RW, Hagstrom JN, Kung SH et al (1997) Stable gene transfer and mice. Hum Gene Ther 25:98–108 expression of human blood coagulation factor IX after intramuscular Kotterman MA, Schaffer DV (2014) Engineering adeno-associated virus- injection of recombinant adeno-associated virus. Proc Natl Acad Sci es for clinical gene therapy. Nat Rev Genet 15:445–451 USA 94:5804–5809 Kotterman MA, Chalberg TW, Schaffer DV (2015) Viral vectors for gene Herzog RW, Yang EY, Couto LB et al (1999) Long-term correction of therapy: translational and clinical outlook. Annu Rev Biomed Eng canine hemophilia B by gene transfer of blood coagulation factor IX 17:63–89 mediated by adeno-associated viral vector. Nat Med 5:56–63 Kuether EL, Schroeder JA, Fahs SA et al (2012) Lentivirus-mediated Hickey RD, Mao SA, Glorioso J et al (2016) Curative ex vivo liver- platelet gene therapy of murine hemophilia a with pre-existing an- directed gene therapy in a pig model of hereditary tyrosinemia type ti-factor VIII immunity. J Thromb Haemost 10:1570–1580 1. Sci Transl Med 8:349ra399 Ledley FD, McNamee LM, Uzdil V, Morgan IW (2014) Why commer- Hirsch ML, Wolf SJ, Samulski RJ (2016) Delivering transgenic DNA cialization of gene therapy stalled; examining the life cycles of gene exceeding the carrying capacity of AAV vectors. Methods Mol therapy technologies. Gene Ther 21:188–194 Biol 1382:21–39 Li C, Narkbunnam N, Samulski RJ et al (2012) Neutralizing antibodies Hoggatt J (2016) Gene therapy for "bubble boy" disease. Cell 166:263 against adeno-associated virus examined prospectively in pediatric Hou X, Jiao R, Guo X et al (2016) Construction of minicircle DNA patients with hemophilia. Gene Ther 19:288–294 vectors capable of correcting familial hypercholesterolemia pheno- Li M, Dick A, Montenovo M, Horslen S, Hansen R (2015) Cost- type in a LDLR-deficient mouse model. Gene Ther 23:657–663 effectiveness of liver transplantation in methylmalonic and Hu C, Cela RG, Suzuki M, Lee B, Lipshutz GS (2011) Neonatal helper- propionic acidemias. Liver Transpl 21:1208–1218 dependent adenoviral vector gene therapy mediates correction of Lipshutz GS, Titre D, Brindle M, Bisconte AR, Contag CH, Gaensler KM hemophilia a and tolerance to human factor VIII. Proc Natl Acad (2003) Comparison of gene expression after intraperitoneal delivery Sci USA 108:2082–2087 of AAV2 or AAV5 in utero. Mol Ther 8:90–98 Jayandharan GR, Aslanidi G, Martino AT et al (2011) Activation of the NF-kappaB pathway by adeno-associated virus (AAV) vectors and Lisowski L, Dane AP, Chu K et al (2014) Selection and evaluation of its implications in immune response and gene therapy. Proc Natl clinically relevant AAV variants in a xenograft liver model. Nature Acad Sci USA 108:3743–3748 506:382–386 Jiang H, Couto LB, Patarroyo-White S et al (2006a) Effects of transient Liu F, Song Y, Liu D (1999) Hydrodynamics-based transfection in ani- immunosuppression on adenoassociated, virus-mediated, liver- mals by systemic administration of plasmid DNA. Gene Ther 6: directed gene transfer in rhesus macaques and implications for hu- 1258–1266 man gene therapy. Blood 108:3321–3328 Louis Jeune V, Joergensen JA, Hajjar RJ, Weber T (2013) Pre-existing Jiang H, Pierce GF, Ozelo MC et al (2006b) Evidence of multiyear factor anti-adeno-associated virus antibodies as a challenge in AAV gene IX expression by AAV-mediated gene transfer to skeletal muscle in therapy. Hum Gene Ther Methods 24:59–67 an individual with severe hemophilia B. Mol Ther 14:452–455 Lozier JN, Metzger ME, Donahue RE, Morgan RA (1999) Adenovirus- Jooss K, Chirmule N (2003) Immunity to adenovirus and adeno- mediated expression of human coagulation factor IX in the rhesus associated viral vectors: implications for gene therapy. Gene Ther macaque is associated with dose-limiting toxicity. Blood 94:3968– 10:955–963 3975 Junge N, Mingozzi F, Ott M, Baumann U (2015) Adeno-associated virus Lu J, Williams JA, Luke J, Zhang F, Chu K, Kay MA (2017) A 5′ vector-based gene therapy for monogenetic metabolic diseases of noncoding exon containing engineered intron enhances transgene the liver. J Pediatr Gastroenterol Nutr 60:433–440 expression from recombinant AAV vectors in vivo. Hum Gene Kaeppel C, Beattie SG, Fronza R et al (2013) A largely random AAV Ther 28:125–134 integration profile after LPLD gene therapy. Nat Med 19:889–891 Macaulay R (2015) How ready are european payers for ema adaptive Kalyuzhniy O, Di Paolo NC, Silvestry M et al (2008) Adenovirus sero- pathways? Value Health 18:A341 type 5 hexon is critical for virus infection of hepatocytes in vivo. Maguire AM, Simonelli F, Pierce EA et al (2008) Safety and efficacy of Proc Natl Acad Sci USA 105:5483–5488 gene transfer for Leber’s congenital amaurosis. N Engl J Med 358: Kattenhorn LM, Tipper CH, Stoica L et al (2016) Adeno-associated virus 2240–2248 Gene therapy for liver disease. Hum Gene Ther 27:947–961 Mah C, Sarkar R, Zolotukhin I et al (2003) Dual vectors expressing Kay MA (2011) State-of-the-art gene-based therapies: the road ahead. Nat murine factor VIII result in sustained correction of hemophilia a Rev Genet 12:316–328 mice. Hum Gene Ther 14:143–152 Kay MA, Manno CS, Ragni MV et al (2000) Evidence for gene transfer Majhen D, Calderon H, Chandra N et al (2014) Adenovirus-based vac- and expression of factor IX in haemophilia B patients treated with an cines for fighting infectious diseases and cancer: progress in the AAV vector. Nat Genet 24:257–261 field. Hum Gene Ther 25:301–317
445 Chapter 10 - Appendices
J Inherit Metab Dis
Manno CS, Chew AJ, Hutchison S et al (2003) AAV-mediated factor IX Morral N, O’Neal W, Rice K et al (1999) Administration of helper- gene transfer to skeletal muscle in patients with severe hemophilia dependent adenoviral vectors and sequential delivery of different B. Blood 101:2963–2972 vector serotype for long-term liver-directed gene transfer in ba- Manno CS, Pierce GF, Arruda VR et al (2006) Successful transduction of boons. Proc Natl Acad Sci USA 96:12816–12821 liver in hemophilia by AAV-factor IX and limitations imposed by the Morral N, O’Neal WK, Rice K et al (2002) Lethal toxicity, severe endo- host immune response. Nat Med 12:342–347 thelial injury, and a threshold effect with high doses of an adenoviral Mattar CN, Nathwani AC, Waddington SN et al (2011) Stable human FIX vector in baboons. Hum Gene Ther 13:143–154 expression after 0.9G intrauterine gene transfer of self- Mukherjee S, Thrasher AJ (2013) Gene therapy for PIDs: progress, pit- complementary adeno-associated viral vector 5 and 8 in macaques. falls and prospects. Gene 525:174–181 Mol Ther 19:1950–1960 Mullard A (2016) European regulators launch "breakthrough" equivalent Mayrhofer P, Schleef M, Jechlinger W (2009) Use of minicircle plasmids programme. Nat Rev Drug Discov 15:223 for gene therapy. Methods Mol Biol 542:87–104 Muruve DA, Cotter MJ, Zaiss AK et al (2004) Helper-dependent adeno- McCarty DM (2008) Self-complementary AAV vectors; advances and virus vectors elicit intact innate but attenuated adaptive host immune applications. Mol Ther 16:1648–1656 responses in vivo. J Virol 78:5966–5972 McCarty DM, Young SM Jr, Samulski RJ (2004) Integration of adeno- Narayanan G, Cossu G, Galli MC et al (2014) Clinical development of associated virus (AAV) and recombinant AAV vectors. Annu Rev gene therapy needs a tailored approach: a regulatory perspective Genet 38:819–845 from the European Union. Hum Gene Ther Clin Dev 25:1–6 McCormack WM Jr, Seiler MP, Bertin TK et al (2006) Helper-dependent Nathwani AC, Gray JT, McIntosh J et al (2007) Safe and efficient trans- adenoviral gene therapy mediates long-term correction of the duction of the liver after peripheral vein infusion of self- clotting defect in the canine hemophilia a model. J Thromb complementary AAV vector results in stable therapeutic expression Haemost 4:1218–1225 of human FIX in nonhuman primates. Blood 109:1414–1421 McGarrity GJ, Hoyah G, Winemiller A et al (2013) Patient monitoring Nathwani AC, Rosales C, McIntosh J et al (2011) Long-term safety and and follow-up in lentiviral clinical trials. J Gene Med 15:78–82 efficacy following systemic administration of a self-complementary McIntosh JH, Cochrane M, Cobbold S et al (2012) Successful attenuation AAV vector encoding human FIX pseudotyped with serotype 5 and of humoral immunity to viral capsid and transgenic protein follow- 8 capsid proteins. Mol Ther 19:876–885 ing AAV-mediated gene transfer with a non-depleting CD4 antibody Nathwani AC, Reiss UM, Tuddenham EG et al (2014) Long-term safety and cyclosporine. Gene Ther 19:78–85 and efficacy of factor IX gene therapy in hemophilia B. N Engl J McKay TR, Rahim AA, Buckley SM et al (2011) Perinatal gene transfer Med 371:1994–2004 to the liver. Curr Pharm Des 17:2528–2541 Nault JC, Datta S, Imbeaud S et al (2015) Recurrent AAV2-related inser- Mian A, Lee B (2002) Urea-cycle disorders as a paradigm for inborn tional mutagenesis in human hepatocellular carcinomas. Nat Genet errors of hepatocyte metabolism. Trends Mol Med 8:583–589 47:1187–1193 Miesbach W, Tangelder M, Klamroth R (2016) Updated results from a Naumer M, Sonntag F, Schmidt K et al (2012) Properties of the adeno- dose escalating study in adult patients with haemophilia B treated associated virus assembly-activating protein. J Virol 86:13038– with AMT-060 (AAV5-hFIX) gene therapy. Haemophilia 22:151 13048 Miller DG, Adam MA, Miller AD (1990) Gene transfer by retrovirus Nguyen TH, Mainot S, Lainas P et al (2009) Ex vivo liver-directed gene vectors occurs only in cells that are actively replicating at the time therapy for the treatment of metabolic diseases: advances in hepato- of infection. Mol Cell Biol 10:4239–4242 cyte transplantation and retroviral vectors. Curr Gene Ther 9:136– Mingozzi F, High KA (2007) Immune responses to AAVin clinical trials. 149 Curr Gene Ther 7:316–324 Nienhuis AW, Nathwani AC, Davidoff AM (2016) Gene therapy for Mingozzi F, High KA (2011) Immune responses to AAVin clinical trials. hemophilia. Hum Gene Ther 27:305–308 Curr Gene Ther 11:321–330 Nightingale SJ, Hollis RP, Pepper KA et al (2006) Transient gene expres- Mingozzi F, High KA (2013) Immune responses to AAV vectors: over- sion by nonintegrating lentiviral vectors. Mol Ther 13:1121–1132 coming barriers to successful gene therapy. Blood 122:23–36 Nomura S, Merched A, Nour E, Dieker C, Oka K, Chan L (2004) Low- Mingozzi F, Maus MV, Hui DJ et al (2007) CD8(+) T-cell responses to density lipoprotein receptor gene therapy using helper-dependent adeno-associated virus capsid in humans. Nat Med 13:419–422 adenovirus produces long-term protection against atherosclerosis Mingozzi F, Chen Y, Murphy SL et al (2012) Pharmacological modula- in a mouse model of familial hypercholesterolemia. Gene Ther 11: tion of humoral immunity in a nonhuman primate model of AAV 1540–1548 gene transfer for hemophilia B. Mol Ther 20:1410–1416 Nowrouzi A, Cheung WT, Li T et al (2013) The fetal mouse is a sensitive Mingozzi F, Chen Y, Edmonson SC et al (2013) Prevalence and pharma- genotoxicity model that exposes lentiviral-associated mutagenesis cological modulation of humoral immunity to AAV vectors in gene resulting in liver oncogenesis. Mol Ther 21:324–337 transfer to synovial tissue. Gene Ther 20:417–424 Oishi Y, Kakimoto T, Yuan W, Kuno S, Yamashita H, Chiba T (2016) Miyoshi H, Blomer U, Takahashi M, Gage FH, Verma IM (1998) Fetal Gene therapy for ornithine Transcarbamylase deficiency by Development of a self-inactivating lentivirus vector. J Virol 72: intrahepatic plasmid DNA-micro-bubble injection combined with 8150–8157 hepatic ultrasound Insonation. Ultrasound Med Biol 42:1357–1361 Monahan PE, Lothrop CD, Sun J et al (2010) Proteasome inhibitors Orkin SH, Motulsky AG (1995) Report and recommendations of the enhance gene delivery by AAV virus vectors expressing large ge- panel to assess the NIH investment in research on gene therapy. nomes in hemophilia mouse and dog models: a strategy for broad National Institute of Health, Bethesda clinical application. Mol Ther 18:1907–1916 Orkin SH, Reilly P (2016) MEDICINE. Paying for future success in gene Monahan P, Walsh C, Powell J (2015) Update on a phase 1/2 open-label therapy. Science 352:1059–1061 trial of BAX335, an adeno-associated virus 8 (AAV8) vector-based Pasi J, Wong W, Rangarajan S (2016) Interim results of an open-label, gene therapy program for hemophilia B. J Thromb Haemost phase 1/2 study of BMN 270, an AAV5-FVIII gene transfer in se- 13(Suppl 2):87 vere hemophilia a. Haemophilia 22:151 Monteilhet V, Saheb S, Boutin S et al (2011) A 10 patient case report on Perrin GQ, Zolotukhin I, Sherman A et al (2016) Dynamics of antigen the impact of plasmapheresis upon neutralizing factors against presentation to transgene product-specific CD4+ T cells and of Treg adeno-associated virus (AAV) types 1, 2, 6, and 8. Mol Ther 19: induction upon hepatic AAV gene transfer. Mol Ther Methods Clin 2084–2091 Dev 3:16083
446 Chapter 10 - Appendices
JInheritMetabDis
Philippe S, Sarkis C, Barkats M et al (2006) Lentiviral vectors with a Shi Y, Falahati R, Zhang J, Flebbe-Rehwaldt L, Gaensler KM (2013) defective integrase allow efficient and sustained transgene expres- Role of antigen-specific regulatory CD4+CD25+ Tcells in tolerance sion in vitro and in vivo. Proc Natl Acad Sci USA 103:17684–17689 induction after neonatal IP administration of AAV-hF.IX. Gene Ther Piccolo P, Brunetti-Pierri N (2014) Challenges and prospects for helper- 20:987–996 dependent adenoviral vector-mediated gene therapy. Biomedicine 2: Sokal EM (2006) Liver transplantation for inborn errors of liver metabo- 132–148 lism. J Inherit Metab Dis 29:426–430 Pien GC, Basner-Tschakarjan E, Hui DJ et al (2009) Capsid antigen Spada M, Riva S, Maggiore G, Cintorino D, Gridelli B (2009) Pediatric presentation flags human hepatocytes for destruction after transduc- liver transplantation. World J Gastroenterol 15:648–674 tion by adeno-associated viral vectors. J Clin Invest 119:1688–1695 Stephen SL, Montini E, Sivanandam VG et al (2010) Chromosomal Pillay S, Meyer NL, Puschnik AS et al (2016) An essential receptor for integration of adenoviral vector DNA in vivo. J Virol 84:9987–9994 adeno-associated virus infection. Nature 530:108–112 Summerford C, Samulski RJ (2016) AAVR: a multi-serotype receptor for Powell JS, Ragni MV, White GC 2nd et al (2003) Phase 1 trial of FVIII AAV. Mol Ther 24:663–666 gene transfer for severe hemophilia a using a retroviral construct Sunderman FW, Boerner F (1949) Normal value in clinical medicine. administered by peripheral intravenous infusion. Blood 102:2038– Saunders, Philadelphia 2045 Suwanmanee T, Hu G, Gui T et al (2014) Integration-deficient lentiviral Puumalainen AM, Vapalahti M,AgrawalRSetal(1998)Beta- vectors expressing codon-optimized R338L human FIX restore nor- galactosidase gene transfer to human malignant glioma in vivo using mal hemostasis in hemophilia B mice. Mol Ther 22:567–574 replication-deficient retroviruses and adenoviruses. Hum Gene Ther Suzuki T, Sasaki T, Yano K et al (2011) Development of a recombinant 9:1769–1774 adenovirus vector production system free of replication-competent Rabinowitz JE, Rolling F, Li C et al (2002) Cross-packaging of a single adenovirus by utilizing a packaging size limit of the viral genome. adeno-associated virus (AAV) type 2 vector genome into multiple Virus Res 158:154–160 AAV serotypes enables transduction with broad specificity. J Virol Suzuki M, Bertin TK, Rogers GL et al (2013) Differential type I 76:791–801 interferon-dependent transgene silencing of helper-dependent ade- Rahim AA, Wong AM, Howe SJ et al (2009) Efficient gene delivery to noviral vs. adeno-associated viral vectors in vivo. Mol Ther 21:796– the adult and fetal CNS using pseudotyped non-integrating lentiviral 805 vectors. Gene Ther 16:509 520 – Tang J, Olive M, Pulmanausahakul R et al (2006) Human CD8+ cyto- Raper SE, Grossman M, Rader DJ et al (1996) Safety and feasibility of toxic T cell responses to adenovirus capsid proteins. Virology 350: liver-directed ex vivo gene therapy for homozygous familial hyper- 312–322 cholesterolemia. Ann Surg 223:116–126 Thaci B, Ulasov IV, Wainwright DA, Lesniak MS (2011) The challenge Raper SE, Haskal ZJ, Ye X et al (1998) Selective gene transfer into the for gene therapy: innate immune response to adenoviruses. liver of non-human primates with E1-deleted, E2A-defective, or E1- Oncotarget 2:113–121 E4 deleted recombinant adenoviruses. Hum Gene Ther 9:671–679 Thwaite R, Pages G, Chillon M, Bosch A (2015) AAVrh.10 immunoge- Raper SE, Yudkoff M, Chirmule N et al (2002) A pilot study of in vivo nicity in mice and humans. Relevance of antibody cross-reactivity in liver-directed gene transfer with an adenoviral vector in partial orni- human gene therapy. Gene Ther 22:196–201 thine transcarbamylase deficiency. Hum Gene Ther 13:163–175 Toietta G, Mane VP, Norona WS et al (2005) Lifelong elimination of Raper SE, Chirmule N, Lee FS et al (2003) Fatal systemic inflammatory hyperbilirubinemia in the Gunn rat with a single injection of response syndrome in a ornithine transcarbamylase deficient patient helper-dependent adenoviral vector. Proc Natl Acad Sci USA 102: following adenoviral gene transfer. Mol Genet Metab 80:148–158 3930 3935 Rastall DP, Seregin SS, Aldhamen YA et al (2016) Long-term, high-level – hepatic secretion of acid alpha-glucosidase for Pompe disease Touchot N, Flume M (2015) The payers’ perspective on gene therapies. achieved in non-human primates using helper-dependent adenovi- Nat Biotechnol 33:902–904 rus. Gene Ther 23:743–752 Trapani I, Colella P, Sommella A et al (2014) Effective delivery of large Reddy PS, Sakhuja K, Ganesh S et al (2002) Sustained human factor VIII genes to the retina by dual AAV vectors. EMBO Mol Med 6:194– expression in hemophilia a mice following systemic delivery of a 211 gutless adenoviral vector. Mol Ther 5:63–73 Tse LV, Moller-Tank S, Asokan A (2015) Strategies to circumvent hu- Rosewell Shaw A, Suzuki M (2016) Recent advances in oncolytic ade- moral immunity to adeno-associated viral vectors. Expert Opin Biol novirus therapies for cancer. Curr Opin Virol 21:9–15 Ther 15:845–855 Samulski RJ, Muzyczka N (2014) AAV-mediated gene therapy for re- Tseng YS, Agbandje-McKenna M (2014) Mapping the AAV capsid host search and therapeutic purposes. Annu Rev Virol 1:427–451 antibody response toward the development of second generation Sarkar R, Mucci M, Addya S et al (2006) Long-term efficacy of adeno- Gene delivery vectors. Front Immunol 5:9 associated virus serotypes 8 and 9 in hemophilia a dogs and mice. Unzu C, Sampedro A, Mauleon I et al (2013) Helper-dependent adeno- Hum Gene Ther 17:427–439 viral liver gene therapy protects against induced attacks and corrects Schiedner G, Bloch W, Hertel S et al (2003) A hemodynamic response to protein folding stress in acute intermittent porphyria mice. Hum Mol intravenous adenovirus vector particles is caused by systemic Genet 22:2929–2940 Kupffer cell-mediated activation of endothelial cells. Hum Gene Vercauteren K, Hoffman BE, Zolotukhin I et al (2016) Superior in vivo Ther 14:1631–1641 transduction of human hepatocytes using engineered AAV3 capsid. Schimmer J, Breazzano S (2016) Investor outlook: solving gene therapy Mol Ther 24:1042–1049 pricing...with a cures voucher? Hum Gene Ther Clin Dev 27:132– Verma IM, Somia N (1997) Gene therapy - promises, problems and 136 prospects. Nature 389:239–242 Schmitt F, Pastore N, Abarrategui-Pontes C et al (2014) Correction of Viecelli HM, Harbottle RP, Wong SP et al (2014) Treatment of phenyl- hyperbilirubinemia in gunn rats by surgical delivery of low doses of ketonuria using minicircle-based naked-DNA gene transfer to mu- helper-dependent adenoviral vectors. Hum Gene Ther Methods 25: rine liver. Hepatology 60:1035–1043 181–186 Vigant F, Descamps D, Jullienne B et al (2008) Substitution of hexon Sendra L, Miguel A, Perez-Enguix D et al (2016) Studying closed hy- hypervariable region 5 of adenovirus serotype 5 abrogates blood drodynamic models of "in vivo" DNA perfusion in pig liver for factor binding and limits gene transfer to liver. Mol Ther 16:1474– Gene therapy translation to humans. PLoS One 11:e0163898 1480
447 Chapter 10 - Appendices
J Inherit Metab Dis
Waddington SN, McVey JH, Bhella D et al (2008) Adenovirus serotype 5 Yanez-Munoz RJ, Balaggan KS, MacNeil A et al (2006) Effective gene hexon mediates liver gene transfer. Cell 132:397–409 therapy with nonintegrating lentiviral vectors. Nat Med 12:348–353 Wang L, Bell P, Somanathan S et al (2015) Comparative study of liver Yang Y, Ertl HC, Wilson JM (1994) MHC class I-restricted cytotoxic T Gene transfer with AAV vectors based on natural and engineered lymphocytes to viral antigens destroy hepatocytes in mice infected AAV capsids. Mol Ther 23:1877–1887 with E1-deleted recombinant adenoviruses. Immunity 1:433–442 White GI, Monahan PE (2005) Gene therapy for hemophilia a. In: Lee Yang Y, Wang L, Bell P et al (2016) A dual AAV system enables the CR, Berntrop E, Hoots K (eds) Textbook of hemophilia. Blackwell, Cas9-mediated correction of a metabolic liver disease in newborn Oxford, pp 226–228 mice. Nat Biotechnol 34:334–338 Wilson JM (2009) Lessons learned from the gene therapy trial for orni- Yokoo T, Kamimura K, Abe H et al (2016) Liver-targeted hydrodynamic thine transcarbamylase deficiency. Mol Genet Metab 96:151–157 gene therapy: recent advances in the technique. World J Wilson EM, Bial J, Tarlow B et al (2014) Extensive double humanization of Gastroenterol 22:8862–8868 both liver and hematopoiesis in FRGN mice. Stem Cell Res 13:404–412 Zaiss AK, Muruve DA (2008) Immunity to adeno-associated virus vec- Wirth T, Parker N, Yla-Herttuala S (2013) History of gene therapy. Gene tors in animals and humans: a continued challenge. Gene Ther 15: 525:162–169 808–816 Wold WS, Toth K (2013) Adenovirus vectors for gene therapy, vaccina- Zhang HG, Zhou T, Yang P, Edwards CK 3rd, Curiel DT, Mountz JD tion and cancer gene therapy. Curr Gene Ther 13:421–433 (1998) Inhibition of tumor necrosis factor alpha decreases inflam- Wolff JA, Budker V (2005) The mechanism of naked DNA uptake and mation and prolongs adenovirus gene expression in lung and liver. expression. Adv Genet 54:3–20 Hum Gene Ther 9:1875–1884 Wu X, Liu G, Mu M et al (2016) Augmenter of liver regeneration gene therapy using a novel minicircle DNA vector alleviates liver fibrosis Zhu J, Huang X, Yang Y (2007) Innate immune response to adenoviral in rats. Hum Gene Ther 27:880–891 vectors is mediated by both Toll-like receptor-dependent and - Yan Z, Sun X, Feng Z et al (2015) Optimization of recombinant adeno- independent pathways. J Virol 81:3170–3180 associated virus-mediated expression for large transgenes, using a Zufferey R, Dull T, Mandel RJ et al (1998) Self-inactivating lentivirus synthetic promoter and tandem array enhancers. Hum Gene Ther 26: vector for safe and efficient in vivo gene delivery. J Virol 72:9873– 334–346 9880
448 Chapter 10 - Appendices
10.5.2 Baruteau et al. Delivering efficient liver-directed AAV- mediated gene therapy. Gene Therapy, May 2017;24(5):263-264
449 Chapter 10 - Appendices
Gene Therapy (2017), 1–2 © 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved 0969-7128/17 www.nature.com/gt
COMMENT Delivering efficient liver-directed AAV-mediated gene therapy
J Baruteau1,2,3, SN Waddington3,4,8, IE Alexander5,6,8 and P Gissen1,2,7,8
Gene Therapy advance online publication, 12 January 2017; apeutic-Effect.html. In.), which is similar to the level observed in doi:10.1038/gt.2016.90 the high-dose group of the AAV8 trial receiving 2 × 1012 vg kg − 1 (plasma FIX of 5.1%, range 2.9–7.1%; n = 6) 4 months post infusion.3 These results suggest that, to obtain similar plasma FIX levels to those achieved in AAV8 trial, administration of 2.5-fold Adeno-associated virus vectors (AAV) have become the leading more AAV5 vector is necessary. technology for liver-directed gene therapy.1 After the pioneering Although this assumption is made on the basis of a small trials using AAV2(ref. 2) and AAV8(ref. 3) to treat haemophilia B, number of treated subjects, and confounded by different methods D’Avola et al.4 recently reported the first-in-human clinical trial of of production, titration and purification, it supports data obtained adeno-associated virus vector serotype 5 (AAV5) in acute after intravenous injection in different animal models: intermittent porphyria (AIP). Treatment was reported as safe, but the main surrogate biomarkers of AIP, porphobilinogen (PBG) and i In murine models of AIP, AAV5 resulted in 10-fold less liver fi delta-aminolevulinate (ALA) were unchanged. This lack of ef cacy transduction compared with AAV8.8 contrasts with results from the haemophilia B trial using AAV8 ii In Gunn rats, AAV5 vector was inefficient at restoring 3 fi capsid by Nathwani et al., which showed a signi cant and long- metabolic activity and achieved three times lower copy lasting improvement of the clinical phenotype. Haemophilia B is number compared with AAV8.6 an amenable target for successful gene therapy as raising iii In Rhesus macaques, AAV5 vector produced slightly lower expression of plasma factor IX (FIX) level above 1% can modify plasma FIX in adult animals with slower kinetics compared the phenotype from severe to moderate.3 Development of a with AAV8,9 lower hepatocyte transduction after fetal variety of capsids for clinical application is useful to overcome pre- fi intrahepatic venous injection and lower plasma FIX 2 months existing neutralising antibodies. The differences in cell-speci c − 1 − 1 transduction by different AAV serotypes are primarily owing to post injection (o1 μg ml (n = 3) versus 5 μg ml fi (n = 1)).10 speci cities in cellular uptake or post cell-entry processing. Indeed − / − − / − − / − AAV5 presents several theoretical advantages as an alternative iv In Fah /Rag2 /Il2rg (FRG) mice, AAV5 achieved capsid to AAV8 for liver-directed gene therapy: suitable liver transduction of 10-fold fewer of human hepatocytes than 11 tropism, less off-target biodistribution,5 low seroprevalence in AAV8 (0.1 versus 1.1%, respectively). humans and minimal cross-reactivity with other serotypes.6 Further results from larger human trials will provide further information on the reliability of animal data, which will accelerate RELIABILITY OF ANIMAL MODELS IN CAPSID TESTING the development of liver-directed gene therapy. The reliability of the available animal models for comparison of transduction of the liver by different AAV serotypes has been questioned.7 In the AIP trial,4 the high-dose group received EPISOMAL VERSUS ENDOGENOUS GENE EXPRESSION 1.8 × 1013 vg kg− 1, which is equivalent to the therapeutic thresh- D’Avola et al.4 are the first to report data from human liver old needed to achieve a correction of the murine phenotype biopsies after AAV treatment. Interestingly, the liver vector copy (1.25 × 1013 vg kg − 1),8 but lower than that required for supra- number 1 year post injection did not correlate with the escalating physiological enzymatic activity in Rhesus macaques doses of vector received. This finding is in contrast with the (5 × 1013 vg kg − 1).5 AAV5 is currently used in a clinical trial for studied tissues from animal models5,8 or plasma FIX levels in haemophilia B with the same transgene cassette used by haemophilia B trial.3 In liver biopsies with high vector copy Nathwani et al. (http://www.uniqure.com/news/320/182/uniQure- number of the transgene codon-optimised PBG deaminase Presents-Updated-Clinical-Data-in-Patients-with-Severe-Hemophi (coPBGD) (patients 2, 5 and 7), coPBGD mRNA expression lia-B-Demonstrating-up-to-9-Months-of-Sustained-Levels-of-Fac compared with endogenous PBGD (normalised by DNA copy tor-IX-Activity-and-Therapeutic-Effect.html. In.). Nine months number) was lower by 45%, 76% and 36%, respectively.4 In AAV- post infusion, the low-dose group, who received 5 × 1012 vg kg − 1, mediated gene therapy, most of the transgene DNA copies persist showed a plasma FIX of 5.4% (range 3.1–6.7%; n = 5) (http://www. as non-integrated episomes. Different episomal expression uniqure.com/news/320/182/uniQure-Presents-Updated-Clinical- compared with the endogenous gene of interest underpins Data-in-Patients-with-Severe-Hemophilia-B-Demonstrating-up- results observed in an ornithine transcarbamylase12-deficient to-9-Months-of-Sustained-Levels-of-Factor-IX-Activity-and-Ther Spfash mouse model. Untreated Spfash mice with a 5–7% wild-
1Genetics and Genomic Medicine Programme, Great Ormond Street Institute of Child Health, University College London, London, UK; 2Department of Metabolic Medicine, Great Ormond Street Hospital for Children NHS Foundation Trust, London, UK; 3Maternal and Fetal Medicine, Gene Transfer Technology Group, Institute for Women’s Health, University College London, London, UK; 4Antiviral Gene Therapy Research Unit, Faculty of Health Sciences, University of the Witswatersrand, Johannesburg, South Africa; 5Gene Therapy Research Unit, The Children’s Hospital at Westmead and Children’s Medical Research Institute, Sydney, New South Wales, Australia; 6Discipline of Child and Adolescent Health, The University of Sydney, Sydney, New South Wales, Australia and 7MRC Laboratory for Molecular Biology, University College London, London, UK. Correspondence: Dr J Baruteau, Genetics and Genomic Medicine Programme, Great Ormond Street Institute of Child Health, University College London, 30 Guilford Street, London WC1N1EH, UK. E-mail: [email protected] 8Co-authors. Received 21 November 2016; accepted 14 December 2016
450 Chapter 10 - Appendices
Comment 2 type residual ornithine transcarbamylase activity become hyper- REFERENCES ammonaemic after a short hairpin RNA (shRNA)-mediated knock- 1 Dolgin E. Early clinical data raise the bar for hemophilia gene therapies. Nat down of the endogenous ornithine transcarbamylase activity to 0– Biotechnol 2016; 34: 999–1001. ash 2.5%. In shRNA-injected Spf mice, the level of AAV-encoded 2MannoCS,PierceGF,ArrudaVR,GladerB,RagniM,RaskoJJet al. Successful ornithine transcarbamylase activity required to normalise ammo- transduction of liver in hemophilia by AAV-Factor IX and limitations imposed by naemia was threefold higher than the residual ornithine the host immune response. Nat Med 2006; 12: 342–347. transcarbamylase activity in untreated Spfash mice.13 An AAV 3 Nathwani AC, Reiss UM, Tuddenham EG, Rosales C, Chowdary P, McIntosh J et al. pattern of transduction not reproducing the physiological Long-term safety and efficacy of factor IX gene therapy in hemophilia B. N Engl J metabolic zonation of the liver might have had an additional Med 2014; 371:1994–2004. role. Although these findings rely on a small cohort and require 4 D'Avola D, Lopez-Franco E, Sangro B, Paneda A, Grossios N, Gil-Farina I et al. Phase caution in interpretation, various explanations might account for a I open label liver-directed gene therapy clinical trial for acute intermittent por- phyria. J Hepatol 2016; 65: 776–783. different episomal expression such as inadequate chromatinisa- 5 Paneda A, Lopez-Franco E, Kaeppel C, Unzu C, Gil-Royo AG, D'Avola D et al. Safety tion, incomplete circularisation of the AAV genome altering the and liver transduction efficacy of rAAV5-cohPBGD in nonhuman primates: a constitution of the open reading frame for transgene expression potential therapy for acute intermittent porphyria. Hum Gene Ther 2013; 24: or inverted terminal repeats recombination. The exact mechanism 1007–1017. fi for this phenomenon is yet to be identi ed. 6 Montenegro-Miranda PS, Paneda A, ten Bloemendaal L, Duijst S, de Waart DR, Gonzalez- Aseguinolaza G et al. Adeno-associated viral vector serotype 5 poorly transduces liver in rat models. PLoS One 2013; 8: e82597. FUNCTIONAL METABOLIC ASSAYS AS EFFICACY END POINTS 7 Wang L, Bell P, Somanathan S, Wang Q, He Z, Yu H et al. Comparative study of IN CLINICAL TRIALS liver gene transfer with AAV vectors based on natural and engineered AAV Finally, the use of metabolite levels as primary end point for trials capsids. Mol Ther 2015; 23: 1877–1887. in metabolic diseases can be questioned. These surrogate markers 8 Unzu C, Sampedro A, Mauleon I, Alegre M, Beattie SG, de Salamanca RE et al. often reflect a static picture, and remain indirect assessments of Sustained enzymatic correction by rAAV-mediated liver gene therapy protects the metabolic flux and its environmental or epigenetic regulation. against induced motor neuropathy in acute porphyria mice. Mol Ther 2011; 19: Indeed, haem biosynthesis is mainly regulated by haem-mediated 243–250. fi inhibitory feedback of the transcription of ALA-synthetase, but 9NathwaniAC,GrayJT,McIntoshJ,NgCY,ZhouJ,SpenceYet al. Safe and ef cient other parameters can exert an influence such as glucose intake, transduction of the liver after peripheral vein infusion of self-complementary AAV 14 vector results in stable therapeutic expression of human FIX in nonhuman pri- stress, drugs, circadian rhythm, and may potentially affect ALA mates. Blood 2007; 109: 1414–1421. and PBG results. Thus, whenever feasible, stable isotope studies 10 Mattar CN, Nathwani AC, Waddington SN, Dighe N, Kaeppel C, Nowrouzi A et al. would be better indicators of the in vivo dynamics of the pathway. Stable human FIX expression after 0.9G intrauterine gene transfer of self- 15 For example, oral administration of N -labelled glycine can complementary adeno-associated viral vector 5 and 8 in macaques. Mol Ther monitor the biosynthesis of haem and its intermediate com- 2011; 19:1950–1960. 15 pounds in physiology and patients with inherited porphyrias. 11 Vercauteren K, Hoffman BE, Zolotukhin I, Keeler GD, Xiao JW, Basner-Tschakarjan E This approach has been successfully used in other metabolic et al. Superior in vivo transduction of human hepatocytes using engineered pathways like the urea cycle to assess ureagenesis utilising either AAV3 capsid. Mol Ther 2016; 24:1042–1049. N15-labelled urea in animal models after AAV-mediated gene 12 Inagaki K, Piao C, Kotchey NM, Wu X, Nakai H. Frequency and spectrum of therapy16,17 or C13-labelled acetate in humans for accurately genomic integration of recombinant adeno-associated virus serotype 8 vector in stratifying the disease severity in ornithine transcarbamylase neonatal mouse liver. J Virol 2008; 82:9513–9524. deficiency.18 Furthermore, the use of clinically relevant end points 13 Cunningham SC, Kok CY, Dane AP, Carpenter K, Kizana E, Kuchel PW et al. would not only provide better assessment of the effect of therapy, Induction and prevention of severe hyperammonemia in the spfash mouse model fi but may be viewed more favourably by regulatory bodies. of ornithine transcarbamylase de ciency using shRNA and rAAV-mediated gene delivery. Mol Ther 2011; 19:854–859. 14 Besur S, Hou W, Schmeltzer P, Bonkovsky HL. Clinically important features of CONFLICT OF INTEREST porphyrin and heme metabolism and the porphyrias. Metabolites 2014; 4: 977–1006. The authors declare no conflict of interest. 15 London IM. The use of stable isotopes in biological and medical research. J Clin Invest 1949; 28(6 Pt 1): 1255–1270. ACKNOWLEDGEMENTS 16 Cunningham SC, Spinoulas A, Carpenter KH, Wilcken B, Kuchel PW, Alexander IE. AAV2/8-mediated correction of OTC deficiency is robust in adult but not neonatal JB is supported by a Clinical Starter Research Grant from Great Ormond Street Spf(ash) mice. Mol Ther 2009; 17:1340–1346. Hospital for Children Charity. 17 Hu C, Tai DS, Park H, Cantero G, Chan E, Yudkoff M et al. Minimal ureagenesis is necessary for survival in the murine model of hyperargininemia treated by AAV- based gene therapy. Gene Ther 2015; 22: 111–115. AUTHOR CONTRIBUTIONS 18 Opladen T, Lindner M, Das AM, Marquardt T, Khan A, Emre SH et al. In vivo JB wrote the manuscript. SNW, IEA and PG contributed and revised the monitoring of urea cycle activity with (13)C-acetate as a tracer of ureagenesis. Mol manuscript. All authors read and approved the final manuscript. Gene Metab 2016; 117:19–26.
Gene Therapy (2017) 1 – 2 © 2017 Macmillan Publishers Limited, part of Springer Nature.
451 Chapter 10 - Appendices
10.5.3 Baruteau et al. Expanding the phenotype in argininosuccinic aciduria: need for new therapies. Journal of Inherited Metabolic Diseases, May 2017; 40(3):357-368
This article is distributed with Springer Open Access Policy under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
452 Chapter 10 - Appendices
J Inherit Metab Dis DOI 10.1007/s10545-017-0022-x
ORIGINAL ARTICLE
Expanding the phenotype in argininosuccinic aciduria: need for new therapies
1,2,3 4 4 Julien Baruteau & Elisabeth Jameson & Andrew A. Morris & 2,5 5 5 1 Anupam Chakrapani & Saikat Santra & Suresh Vijay & Huriye Kocadag & 6 2 7 2 Clare E. Beesley & Stephanie Grunewald & Elaine Murphy & Maureen Cleary & 8 2 2,4 7 Helen Mundy & Lara Abulhoul & Alexander Broomfield & Robin Lachmann & 9 10 11 11 Yusof Rahman & Peter H. Robinson & Lesley MacPherson & Katharine Foster & 12 13 14 W. Kling Chong & Deborah A. Ridout & Kirsten McKay Bounford & 1,15 3 2,3,16 2 Simon N. Waddington & Philippa B. Mills & Paul Gissen & James E. Davison
Received: 7 November 2016 /Revised: 24 January 2017 /Accepted: 25 January 2017 # The Author(s) 2017. This article is published with open access at Springerlink.com
Abstract Results Fifty-six patients were defined as early-onset (n =23)if Objectives This UK-wide study defines the natural history of symptomatic < 28 days of age, late-onset (n =23)ifsymptomatic argininosuccinic aciduria and compares long-term neurologi- later, or selectively screened perinatally due to a familial proband cal outcomes in patients presenting clinically or treated pro- (n =10).Themedianfollow-upwas12.4years(range0–53). spectively from birth with ammonia-lowering drugs. Long-term outcomes in all groups showed a similar neurological Methods Retrospective analysis of medical records prior to phenotype including developmental delay (48/52), epilepsy (24/ March 2013, then prospective analysis until December 2015. 52), ataxia (9/52), myopathy-like symptoms (6/52) and abnormal Blinded review of brain MRIs. ASL genotyping. neuroimaging (12/21). Neuroimaging findings included
Communicated by: Matthias Baumgartner Electronic supplementary material The online version of this article (doi:10.1007/s10545-017-0022-x) contains supplementary material, which is available to authorized users.
* Julien Baruteau Stephanie Grunewald [email protected] [email protected] Elaine Murphy Elisabeth Jameson [email protected] [email protected] Maureen Cleary Andrew A. Morris [email protected] [email protected] Helen Mundy Anupam Chakrapani [email protected] [email protected] Lara Abulhoul Saikat Santra [email protected] [email protected] Alexander Broomfield Suresh Vijay [email protected] [email protected] Huriye Kocadag Robin Lachmann [email protected] [email protected] Clare E. Beesley Yusof Rahman [email protected] [email protected]
453 Chapter 10 - Appendices
JInheritMetabDis parenchymal infarcts (4/21), focal white matter hyperintensity ASL Argininosuccinate lyase (4/21), cortical or cerebral atrophy (4/21), nodular heterotopia CSF Cerebro spinal fluid (2/21) and reduced creatine levels in white matter (4/4). 4/21 adult NO Nitric oxide patients went to mainstream school without the need of additional NOS Nitric oxide synthase educational support and 1/21 lives independently. Early-onset pa- UCD Urea cycle defects tients had more severe involvement of visceral organs including liver, kidney and gut. All early-onset and half of late-onset patients presented with hyperammonaemia. Screened patients had normal Introduction ammonia at birth and received treatment preventing severe hyperammonaemia. ASL was sequenced (n =19)and20muta- In the central nervous system, nitric oxide (NO) is involved in tions were found. Plasma argininosuccinate was higher in early- crucial processes including neurotransmission (Garthwaite onset compared to late-onset patients. 2008), neuronal differentiation (Peunova and Enikolopov Conclusions Our study further defines the natural history of 1995) and migration (Nott et al. 2013). Argininosuccinate argininosuccinic aciduria and genotype–phenotype correla- lyase (ASL) cleaves argininosuccinate into arginine and fuma- tions. The neurological phenotype does not correlate with the rate as part of the NO-citrulline cycle that regulates NO pro- severity of hyperammonaemia and plasma argininosuccinic ac- duction in multiple tissues (e-Figure 1)(Nagamanietal. id levels. The disturbance in nitric oxide synthesis may be a 2012a). ASL deficiency causes argininosuccinic aciduria contributor to the neurological disease. Clinical trials providing (ASA; OMIM 207900), the only inherited condition proven nitric oxide to the brain merit consideration. to cause systemic NO deficiency (Erez et al. 2011). ASL is also required for the liver-based urea cycle, which detoxifies Abbreviations ammonia produced by amino acid catabolism. ASA is the ALT Alanine aminotransferase second most common urea cycle defect (UCD) with an inci- ASA Argininosuccinic aciduria dence of 1:70,000 live-births (Nagamani et al. 2012a) and
3 Peter H. Robinson Genetics and Genomic Medicine Programme, Great Ormond Street [email protected] Institute of Child Health, University College London, London, UK 4 Lesley MacPherson Metabolic Medicine Department, Royal Manchester Children [email protected] Hospital NHS Foundation Trust, Manchester, UK 5 Metabolic Medicine Department, Birmingham Children’s Hospital Katharine Foster NHS Foundation Trust, Birmingham, UK [email protected] 6 North East Thames Regional Genetic Services, Great Ormond Street W. Kling Chong Hospital NHS Foundation Trust, London, UK [email protected] 7 Charles Dent Metabolic Unit, National Hospital for Neurology and Deborah A. Ridout Neurosurgery, London, UK [email protected] 8 Metabolic Medicine Department, Evelina Children’sHospital, Kirsten McKay Bounford London, UK [email protected] 9 Metabolic Medicine Department, St Thomas Hospital, London, UK Simon N. Waddington 10 Paediatric Metabolic Medicine, Royal Hospital for Sick Children, [email protected] Glasgow, UK Philippa B. Mills 11 Neuroradiology Department, Birmingham Children’s Hospital NHS [email protected] Foundation Trust, Birmingham, UK 12 Paul Gissen Neuroradiology Department, Great Ormond Street Hospital NHS [email protected] Foundation Trust, London, UK 13 Population, Policy and Practice Programme, UCL Institute of Child James E. Davison Health, London, UK [email protected] 14 West Midlands Regional Genetic Laboratory, Birmingham Women’s Hospital, Birmingham, UK 1 Gene Transfer Technology Group, Institute for Women’s Health, 15 Wits/SAMRC Antiviral Gene Therapy Research Unit, Faculty of University College London, London, UK Health Sciences, University of the Witwatersrand, Johannesburg, South Africa 2 Metabolic Medicine Department, Great Ormond Street Hospital for Children NHS Foundation Trust, Great Ormond Street, WC1N 16 MRC Laboratory for Molecular Cell Biology, University College 3JH London, UK London, London, UK
454 Chapter 10 - Appendices
J Inherit Metab Dis presents clinically either as an early neonatal-onset (<28 days London; Great Ormond Street Hospital for Children, of age) hyperammonaemic coma, or a later-onset London; the National Hospital for Neurology and hyperammonaemic crisis (Nagamani et al. 2012a). A chronic Neurosurgery, London; the Royal Manchester Children phenotype with neurocognitive, gastrointestinal and liver Hospital, Manchester. Molecular analysis of patients was symptoms without severe hyperammonaemia is also approved by the National Research Ethics Service recognised (Nagamani et al. 2012a). Conventional treatment Committee London-Bloomsbury (13/LO/0168). Patients in- aims to decrease ammonia by use of a protein-restricted diet cluded had plasma argininosuccinic acid levels > 5 μmol/L, and ammonia scavenger drugs (sodium benzoate and and/or pathogenic mutations in ASL. Patients were considered phenylbutyrate) and to correct arginine deficiency by L- lost to follow-up if no clinical assessment was performed dur- arginine supplementation (Haberle et al. 2012). ing the last 3 years at the relevant metabolic centre. The data- The phenotype in ASA differs from other UCD by the base was closed on 31st December 2015. higher incidence of neurocognitive symptoms, liver fibro- Neurological outcome was assessed with physical neuro- sis, renal impairment and systemic hypertension (Nagamani logical examination regarding developmental impairment, ep- et al. 2012a; Kolker et al. 2015). These symptoms are ob- ilepsy, ataxia, myopathy-like symptoms and brain MRI fea- served in patients with early- or late-onset forms and in those tures and was performed as follows: if neuropsychological without documented episodes of hyperammonaemia (Marble assessment was unavailable, cognitive impairment was deter- et al. 2008). Among UCD, ASA patients have the lowest mined by clinical judgement of the metabolic specialist or frequency of hyperammonaemic crises (23%) but the second neuropaediatrician or by the need for additional support in highest frequency of cognitive impairment (65–74%) after school or subsequently at the workplace. Epilepsy was de- arginase deficiency (Ruegger et al. 2014;Waisbrenetal. fined as the occurrence of two or more seizures without ac- 2016). This paradox raises questions about the role of companying hyperammonaemia. MR spectroscopy was per- hyperammonaemia in causing the neurological problems. formed as described previously (Davison et al. 2011). Newborns screened and treated prospectively from birth Indication for neuroimaging was an unexplained and/or se- have been reported to have a better neurological outcome vere neurological disease. Brain MRIs were analysed by (Widhalm et al. 1992;Ficiciogluetal.2009;Mercimek- two neuroradiologists blinded to the report of each other. Mahmutoglu et al. 2010). As conventional treatment de- Magnetic resonance spectroscopy (MRS) studies were creases ammonia levels, it was suggested that neurological performed concurrently with clinically indicated MRI complications were caused by unrecognised scans at 1.5 T. Comparison was made with MRS metab- hyperammonaemic episodes (Widhalm et al. 1992). olite data from a standard cohort of children with normal However, newborn screening programmes can capture a appearing MRI as described previously. Liver involve- wide phenotypic spectrum, including patients who would ment was considered using the following parameters: he- remain asymptomatic without treatment. Some of these patomegaly, increased levels of transaminases (alanine amino- screened patients had high residual ASL activity transferase ALT > 50 IU/L). Nephromegaly was defined as (Ficicioglu et al. 2009), suggesting that the prospectively renal length on ultrasound imaging above the 95th centile treated cohort might have had an increased number of mild for the age and sex. Biochemical data were assessed using cases, introducing a bias into the comparison. the mean of at least the last ten results available during com- We describe a United Kingom (UK) wide cohort of pensated metabolic state. Plasma ammonia levels were con- ASA patients expanding the disease natural history, sidered elevated if >100 μmol/L before 28 days of life or reporting long-term neurological outcomes with a focus >45 μmol/L subsequently. Hypokalaemia was defined as a on neuroimaging and genotype–phenotype correlations. plasma potassium level lower than 3.5 mmol/L and judged The outcomes in patients treated prospectively (10/56) as Btransient^ if observed in a single sample, or Bpersistent^ were compared with those who presented with symptoms if measured in ≥ 2 samples separated by ≥ 1 month. Plasma before diagnosis (46/56). arginine and argininosuccinic acid reflect the last ten measure- ments performed during follow-up in a compensated metabol- ic state on the patient’s standard treatment. For analysis, pa- Material and methods tients were divided into three groups: (i) early-onset form (hyperammonaemic symptoms started on/before 28 days of Patients life), (ii) late-onset form (presentation after 28 days of life), (iii) perinatally screened patients diagnosed after a family pro- Anonymised data were collected prospectively from band and treated prospectively from birth. For the last group, March 2013 and retrospectively before, from five tertiary the status (early- or late-onset) of the familial proband was metabolic centres in the UK: Birmingham Children’s investigated but missing data prevented inclusion of these Hospital, Birmingham; Guy’sandStThomas’ Hospital, index cases into the study.
455 Chapter 10 - Appendices
JInheritMetabDis
ASL sequencing Developmental impairment was reported in 48/52 patients (92%) and was the most common symptom. The median age The 16 coding exons of ASL (NM_001024943.1; Ensembl at diagnosis was 2 years (range 0.1–6years)(Table1)and ENST00000395332) and the intron-exon boundaries were when observed, developmental impairment was present be- PCR-amplified. Sequencing performed with the Big Dye fore the age of 6 years in all but two patients. Only four Terminator Cycle Sequencing System version 1.1 (Applied patients were reported with normal neurocognitive function: Biosystems/ThermoFisher Scientific) was run on an ABI three early-onset patients aged <6 months (patient 22), PRISM 3730 DNA Analyzer (Applied Biosystems/ 23 months (patient 15) and 11 years old at last assessment ThermoFisher Scientific) (e-Methods). (patient 11) and one patient screened at birth (patient 47; sib- ling to a late-onset proband, aged 8 years old at last assess- Statistical analysis ment) (e-Table 2). Developmental impairment was mild or moderate affecting predominantly speech and learning ability. Statistical analyses used Fisher’s exact test for investigating Detailed information about schooling was available in 35 pa- the association between categorical data and the patient tients (e-Table 2). Only 6/35 patients (17%) attended main- groups (www.vassarstats.net). Continuous variables between stream school without the need for additional educational sup- groups were compared using the Student’s t test or one-way port (patients 9, 11, 35, 36, 47 and 48 with last assessment at ANOVA with Bonferroni correction for pairwise comparison 25, 11, 22, 20, 8 and 16 years old respectively). Most patients (p values detailed in e-Table 1) (GraphPad Prism 5.0, San (20/35; 57%) required speech and language therapy. The neu- Diego, CA, USA). p values ≤ 0.05 were considered statistical- ropsychological assessments identified behavioural difficul- ly significant. Kaplan-Meier survival curves were compared ties with auto- or hetero-aggression (n = 3) and learning dis- with the log-rank test. Patients 10, 18 and 52, who died during abilities in logic and reasoning. Twenty-one adults (EO n =3; the first month of life, and patient 53, for whom very limited LO n = 17; SCR n = 1) with a median age of 22.3 years (range clinical information was available, were excluded from long- 18–57 years) were assessed for socioeconomic status. Five term analysis. For patients lost during follow-up (n = 6), the (25%) had semi-skilled employment. Independent living was assessments at their last follow-up visits were used for reported in 1/12 (8%), and long-term relationships in 3/12 analysis. (25%). Patients not living independently were accommodated in the parental (9/11; 82%) or care (2/11; 18%) homes. Epilepsy was observed in 22/52 patients (42%) with no significant difference of the median age of onset between Results groups (Table 1). Various seizure types were reported includ- ing generalised, partial and complex, febrile and afebrile sei- Patients zures. Tonic-clonic seizures were most frequent (n =16), followed by absence seizures (n =5),myoclonicjerks Fifty-six patients were classified as early-onset (n = 23/56), (n = 4), atonic seizures (n = 2) and occasionally status epilep- late-onset (n =23/56)orscreenedpatients(n =10/56). ticus (n =1)(e-Table 2). Electroencephalogram was per- Ethnic origins were White British (n = 24; 44%), Pakistani formed in three non-epileptic patients and showed an abnor- (n =16;29%),Chinese(n =6;11%),Indian(n =5;9%), mal pattern in two patients. 17/22 patients (77%) were treated Bangladeshi (n =3;5%),otherWhiteEuropean(n =1;2%) with an average of 1.5 antiepileptic drugs (range 0–4). and missing data (n = 1; 2%) (e-Table 3). Screened patients, Cerebellar dysfunction was detected in early-onset and diagnosed either antenatally or neonatally, had an affected late-onset patients (n = 3 and n = 6 respectively) with the inci- familial proband with early-onset (n = 5), late-onset (n =3) dence of 9/52 (17%) (Table 1). Ataxia was first noticed at a or unknown (n = 2) phenotype (e-Table 3). One pair of sib- median age of 8.5 years (range 1–12) with two main age lings with early-onset form (patients 15 and 16) was included groups at first observation, early around the age of 1 year in the study. Mean follow-up was not significantly different (n = 2) or later as teenagers (n = 7). Three patients had dyski- between the early-onset (EO) or late-onset (LO) groups com- nesia and tremor and one had nystagmus (e-Table 2). The mild pared to the screened group (SCR) (p =0.19and p = 0.19 inconvenience caused did not require any specific medical or respectively) (Table 1). surgical treatment. Episodes of myopathy-like symptoms, reported in 7/52 pa- Neurological phenotype tients (13%) (Table 1), included global hypotonia with a hypomimic facial expression, and unexplained recurrent epi- The frequency and median age of onset of the neurological sodes of general weakness persisting several days before features were not significantly different between the groups spontaneous recovery. One patient (patient 28, currently aged (e-Table 1). 15.9 years) was reported with fatigable ptosis from 12 years
456 Chapter 10 - Appendices
J Inherit Metab Dis 12) 57) – 13) – 53) – 12) – 6) – – 53) – 5/56 (9%) 116 ± 9 1(0 11 2.5 (0 48/52 (92%) 22/52 (42%) 1 7/52 (13%) 25/51 (49%) 10/48 (21%) 12.4 (0 2(0.1 5.5 (0.7 9/52 (17%) 12/23 (52%) 8/56 (14%) 15.6 (1.9 28/51 (55%) 169 ± 37 26/56 (46%) 7/56 (12%) 19/56 (34%) 56 (20 = 36%) 21/52 (40%) 530 ± 85 17/51 (33%) 12/32 (38%) 2years(0 3/12 (25%) 1.2 ± 0.07 356 ± 62 143 ± 31 46/56 (82%) 201 ± 20 25/48 (52%) 191 ± 15 42/47 (89%) 47/47 (100%) 31/25 11) – ia-total includes patients with 18.2) 18.2) 4) 9) – – – – 7) – 0/10 (0%) 0/8 (0%) 7(3 0/8 (0%) 134 ± 15 3/8 (38%) 5/8 (62%) / 2/4 (50%) 5/8 (62%) 07.5 (0.9 6/8 (75%) 3.1 (2 8.5 (8 0 1/7 (14%) 2antenatally 7/8 (88%) / 2/5 (40%) 2/10 (20%) 1/10 (10%) 84 ± 18 4/10 (40%) 181 ± 50 6/7 (86%) 6/7 (72%) NA 10 (1 = 10%) 8neonatally 15.6 (8 15.6 (8 6/10 (60%) 5/10 (50%) 1/2 (50%) 238 ± 206 1.4 ± 0.08 251 ± 45 167 ± 17 8/2 9/10 (90%) 7/7 (100%) 12) – sreflectthelasttenmeasurementsperformedduring 57) 53) – 6) 11) – 53) – – – 10 (50%) /23 (13%) 5/23 (22%) 3/23 (13%) 6/23 (26%) 23 (1 2/23 (9%) 102 ± 12 11/23 (48%) / 2/22 (9%) 13.0 4/23 (18%) 23/23 (100%) 2.5 (1 2(0.7 0 3/23 (13%) 3 4/23 (17%) 23 (16.7 3/20 (15%) 5/ 0/3 (0%) 15.1 (1 2/22 (9%) 212 ± 67 2/9 (22%) 2.75 years (0.25 4/23 (17%) 1/23 (4%) 7.3 ± 2.2 81 ± 24 16/20 (80%) 57 ± 12 234 ± 64 20/20 (100%) 23 (16 = 70%) 155 ± 25 134 ± 33 11/12 21/23 (91%) 1.2 ± 0.1 er figures show mean ± standard error. Hypokalaem 8) – 6) 25.7) 25.7) – 12) 4) 13) – – – – – 19 (89%) /21 (86%) /19 (37%) 12 ± 92 12/23 (52%) 2.5 (0 0.15 (0 126 ± 19 5 10/21 (48%) 11 0/23 (0%) 9(1.5 17/21 (81%) 18/21 (86%) 8/18 (44%) 2/7 (29%) 1 7 8/21 (38%) 17/ 4days(2 11 (1.9 2(0.1 11 (1.9 4/21 (19%) 5/9 (56%) 861 ± 120 3/21 (14%) 16/23 (70%) / 238 ± 77 20/20 (100%) 200 ± 50 23 (3 = 10%) 18 13/23 (56%) Early onset Late onset Screened Total 2/23 (9%) 3/23 (13%) 12/11 239 ± 28 215 ± 18 20/20 (100%) 16/23 (73%) 1.2 ± 0.14 nine and argininosuccinic acid concentration nted as median ± range. Oth onset and screened patients mol/L μ mol/L μ 50 IU/L) – y if <28 days of life;if <50 >28 days of life) Severe diarrhoea ALT (RI 20 Intermittent Trichorrhexis nodosa Poor corticomedullar differentiation Persistent Arterial hypertension Age when first reported (years) Age when first reported (years) Abnormal brain MRI Age when first reported (years) Raised ALT Age when first reported (years) Myopathic features Enlargement (>95th centile) Hepatomegaly Mean follow-up (years) Developmental delay Age when first reported (years) Epilepsy Age when first reported (years) Ataxia Hypokalaemia: total Patients lost Frequency Age Age (years) Frequency Na benzoate (mg/kg/day) Ammonia (RI < 100 L-arginine (mg/kg/day) Na phenylbutyrate (mg/kg/day) Patients still living Sex (M/F) Consanguinity Number (adult) Daily protein allowance (g/kg/day) Frequenc Frequency dcohorts:early-onset,late- range interval etabolic state on their standard treatment RI w-up is considered until December 2015. Plasma argi currence of symptom and duration of follow-up are prese not available, mol/L) μ NA mol/L) μ (RI <5 Plasma argininosuccinic acid Plasma arginine (RI 30-126 Kidney appearance (ultrasound) Miscellaneous Liver Neurology Na benzoate supplementation Protein restricted diet Na phenylbutyrate supplementation L-arginine supplementation Epidemiological and clinical data for the three analyse alanine aminotransferase, Biology Phenotype Follow up At diagnosis Therapeutics Table 1 Epidemiology Age at diagnosis, currently, at firstintermittent oc and persistent hypokalaemia. Follo follow-up when patients were in aALT compensated m
457 Chapter 10 - Appendices
JInheritMetabDis old onwards. Four patients were investigated with electro- cortical atrophy (n =2),cerebellaratrophy(n =2),perirolandic myogram, which were always normal, including a Tensilon gliosis (n =1),thalamicatrophy(n =1),hyperintensityofcau- test in one patient. One patient (patient 33 currently aged date head and posterior putamen (n =1)orisosignalbetween 26 years) presented with an electrophysiologically confirmed pallidi and putamen (n =1)(Fig. 1A and e-Table 2). episode of Guillain-Barré syndrome at 8 years old (e-Table 2). Spectroscopy of basal ganglia (n =8;3early-onset,5late-onset) MRI brain was performed as part of the clinical work-up in indicated a significant decrease of N-acetylaspartate and choline 21 patients with unexplained or severe neurological features. in early-onset patients compared to controls (p =0.001andp = The average age at the time of MRI was 12 years (range 0– 0.008, respectively). Creatine and guanidinoacetate levels in the 23). Twelve scans were reported as abnormal (52%) (Table 1). basal ganglia did not differ significantly between controls, early- Neuroimaging performed during follow-up showed small pa- or late-onset groups (Fig. 1B). Spectroscopy of the white matter renchymal infarcts (n =4),fociofwhitematterhyperintensity (n =4;3early-onset,1late-onset)showedasignificantdecrease on T2-weighted sequences (n =4),nodularheterotopia(n =2), in creatine levels (p =0.003) and an increase of
Fig. 1 Neuroimaging. A: Morphological brain MRI features. A, B:T2- axial image with bilateral high signal of the peritrigonal white matter weighted axial images showing brain matter volume loss and mild ex (arrow). B 1H MR spectroscopy features in basal ganglia. Assessment vacuo dilatation of ventricles (A) and high signal in bilateral caudate in early-onset (n = 5), late-onset (n = 3) and control (n = 63) patients heads and posterior putamina (arrows). C: T2-weighted axial images with analysed using a paired t test. c 1H MR spectroscopy features in white severe diffuse cerebral atrophy and ventricular dilatation. D, H: matter. Patients affected by argininosuccinic aciduria (n = 4) and controls T1-weighted coronal image with right periventricular heterotopia (n = 53) analysed with one way ANOVA. Graphs represent mean ± 95% (arrowheads). E, F: T2-weighted axial (E)andcoronal(F) images with confidence interval. * p < 0.05; ** p <0.01 evidence of right inferior frontal lobe infarct (arrow). G:T2-weighted
458 Chapter 10 - Appendices
J Inherit Metab Dis
The commonest hepatic involvement was a persistent rise in plasma alanine transaminase activity, usually accompanied by hepatomegaly. These were significantly more frequent in early-onset and screened patients than in late-onset patients (p <0.00001and p <0.005,respectively;e-Table 1). In screened patients, the likelihood of hepatic abnormalities depended on the age of onset of the disease in the familial proband: 4/4 of screened patients with an early-onset familial history had hepatic abnormalities compared to 1/3 with a fa- milial late-onset phenotype (Table 1). There was no evidence of differences between groups for nephromegaly and poor corticomedullary differentiation assessed by ultrasound (Table 1). Transient or persistent hypokalaemia occurred more frequently in early-onset versus late-onset pa- tients (p < 0.03) (Table 1). Acute metabolic decompensation, gastroenteritis and acute diarrhoea were significantly as- sociated with transient hypokalaemia (p <0.005). Trichorrhexis nodosa was observed only in late-onset pa- tients before diagnosis and normalised with treatment (Table 1). Chronic profuse diarrhoea was observed in 17 patients (33%; including early-onset n = 10, late-onset n =3,screened patients n =4;Table 1). This symptom was refractory to Fig. 2 Natural history of argininosuccinic aciduria. A Kaplan-Meier sur- symptomatic and immunosuppressive treatments and caused vival curves for all (solid line), early-onset (dashed line), late-onset nutritional difficulties in several early-onset patients. Two pa- (dashed dotted line)andscreened(dotted line) patients. B Natural history of the systemic phenotype of argininosuccinic aciduria. Mean ± standard tients had colonoscopies performed at 5 years of age and re- error of age of onset of each symptom from data of the whole cohort when peated at ages 7 and 10. Intestinal biopsies showed non- information available: developmental delay (n = 7), abnormal LFTs (n = specific mild inflammation. Chronic pancreatitis was ob- 8), hepatomegaly (n = 18), epilepsy (n = 15), brittle hair (n =4),ataxia served in one early-onset patient. (n = 6), hypokalaemia (n = 2), high blood pressure (n = 1). Symptom fre- quency in the total population of patients studied is presented in brackets. Refractory arterial hypertension was diagnosed in one ALT: plasma alanine aminotransferase activity. It was assumed that pa- early-onset patient (patient 6) at the age of 9 years and was tients had normal blood pressure if hypertension was not specifically sub-optimally controlled despite three antihypertensive medi- mentioned in medical records cations. This patient died at 12 years old from acute pancrea- guanidinoacetate (p =0.01)(Fig. 1C)inpatientscomparedto titis. One late-onset patient developed atrial flutter at 60 years. controls.
Ammonaemia, ASA levels and therapies Systemic phenotype None of the patients in the screened group suffered severe or 47/56 patients (84%) were alive at the time of assessment with prolonged hyperammonaemia. Three of these patients had an no significant difference between groups, with a median follow- initial ammonia level >100 μmol/L (133, 134 and 190 μmol/ up of 12.4 years (range 0–53) (Fig. 2A). Cause and age of death L), which normalised in less than 24 hours. All early-onset were hyperammonaemic decompensation at presentation (n =2; patients had hyperammonaemia at diagnosis with values sig- patients aged day 3 and 4 of life), sepsis (n =3;patientsaged nificantly higher than in the late-onset and screened groups 7days,11yearsand20years),extraduralhematoma(n =1; (p < 0.001; e-Table 1). Only 50% of the late-onset patients patient aged 2 years), hepatocellular carcinoma (n =1;patient were hyperammonaemic at diagnosis. aged 4.5 years), acute pancreatitis (n =1;patientaged12years) Aprotein-restricteddietwasused in 89% of patients (100% and a possible arrhythmia (n =1;patientaged52years). of the early-onset group and 80% of the late-onset group; Natural history data included the age of onset of organ Table 1). All patients were treated with L-arginine with no involvement or symptoms (Fig. 2B). Among neurological significant difference in the dose between groups (Table 1). symptoms, developmental delay was the first observed usual- Ammonia scavenger drugs (sodium benzoate and ly during the second or third year of life followed by epilepsy phenylbutyrate) were prescribed significantly more often in and ataxia. the early-onset and screened groups than in the late-onset group
459 Chapter 10 - Appendices
JInheritMetabDis ) ) ), 2014 2014 2014 =7, =3, n n =3,Keskinen ) ;Keskinen ;Keskinen n =4,thisstudy, 2014 2002 2002 n =1)(Balmeretal. =1)(Balmeretal. ;Balmeretal. ), late onset ( ), unknown ( n n 2008 2008 2008 et al. early onset ( Kleijer et al. et al. Kleijer et al. et al. Balmer et al. New genotype New genotype New genotype New genotype New genotype Prenatal diagnosis ( Unknown ( New genotype New genotype New genotype New genotype study Reported severity in the literature screened patients, the severity of the phenotype is Late onset Late onset Early onset Early onset Early onset Late onset Late onset Early onset Early onset ithin brackets for novel mutations. For icing effect Arg385Cys) licing effect (Val349Cysfs*72) Early onset .(Met250Lys) p Sp Spl p.(Arg12Gln) Late onset Unknown ( p.(Val178Met)p.(Val178Met)p.(Arg126G1n) Late onset Early onset Late onset New genotype New genotype New genotype p.(Glu258Lys) p.(Arg456Trp) p.(Gln286Arg) p.(Lys380Glu) p.( ) ) ) * * p.(Arg146G1n) p.(Met250Lys) p.(Lys380Glu p.(Glu241Lys) Presumed effect on protein Severity in this p.(Arg182 p.(Arg385Cys) p.(Trp428 Splicing effect e. Presumed protein effect is mentioned w 1353del Loss of exons 15 and 16 Loss of exons 15 and 16 Early onset * c.1138A>G c.446+1G>A c.918+5G>A c.772G>A c.1153C>T c.749T>A c.857A>G c.1366C>T hASL utations not described befor 1353del c.1143+117_ * phenotype correlation of – c.437G>A c.544C>T c.719-2A>G c.1143+117_ c.1153C>T c.721G>A c.749T>A c.1045_1057delc.1138A>G c.1045_1057del p.(Val349Cysfs*72) p. c.1284G>A Genotype a Patients 15 and 16 are siblings 1c.35G>Ac.35G>Ap.(Arg12Gln) 2c.348+1G>Ac.532G>ASplicingeffect 3c.349-1G>Ac.532G>ASplicingeffect 4, 56c.437G>Ac.437G>Ap.(Arg146G1n)p.(Arg146G1n) c.377G>A c.377G>A p.(Arg126G1n) 7 8 deducted from the symptomatic familial proband New genotype refers to combination of m Table 2 a Patient number Allele 1 Allele 2 17, 18 9 15, 16 13 14 10 11, 12 19
460 Chapter 10 - Appendices
J Inherit Metab Dis
(p <0.001andp <0.01respectively)andathigherdosesinthe The early-onset group had higher ammonia levels compared early-onset group (p =0.03)(e-Table1). to the late-onset group, as evidenced by differences in the am- Plasma argininosuccinate levels were higher in early-onset monia levels at diagnosis, and need for ammonia scavenger (512 ± 92 μmol/L) compared to late-onset (234 ± 64 μmol/L) medications and protein restriction. The screened patients also (p = 0.03) (e-Table 1). needed more treatment to control their ammonia levels than patients in the late-onset group because most of them (5/8) Genotype–phenotype correlation had siblings with early-onset disease. Plasma ASA levels were higher in early-onset compared to late-onset and screened pa- The genotype was available for 19 patients (Table 2). Twenty tients. However, these differences did not affect the neurological mutations (including eight novel) were identified: 11 were phenotype. These observations show that hyperammonaemia missense, five splice site, two nonsense mutations and two and ASA levels are not the dominant factors causing the long- deletions. A sequence alignment of ASL showed that all mis- term neurological phenotype in ASA. Previous publications sense mutations affect amino acid residues that are highly have suggested that neonatal screening and early treatment conserved across species. The deletions included a 13 base may prevent or ameliorate the neurological disease (Widhalm pair deletion (c.1045_1057del, p.(Val349Cysfs*72)) and one et al. 1992; Ficicioglu et al. 2009). However, an extended large deletion of approximately 2 kb which included exons 15 Austrian cohort of neonatally screened patients, initially report- and 16 (c.1143+117_*1353del). Homozygous mutations ob- ed with normal neurocognitive outcome at a mean age of 6 years served with early onset disease included c.437G>A (Widhalm et al. 1992), showed that 35% of patients (6/17; me- p.(Arg146Gln), c.749T>A p.(Met250Lys), c.1045_1057del dian age 13 years) had an IQ of less than 80 (Mercimek- p.(Val349Cysfs*3), c.1143+117_*1353del and c.1153C>T Mahmutoglu et al. 2010). As the neurological disease is pro- p.(Arg385Cys). Homozygous mutations observed with late- gressive, duration of the follow-up is essential to determine the onset disease included c.35G>A p.(Arg12Gln), c.377G>A outcome objectively. At the end of the first year, the frequency p.(Arg126Gln) and c.1138A>G p.(Lys380Glu). The of developmental impairment is similar to other UCD (64%) c.1045_1057del deletion is predicted to cause a frameshift attributable to sequelae of neonatal hyperammonaemia in early- and introduction of a premature stop codon and the c.1143+ onset patients (Burgard et al. 2016). This frequency increases to 117_*1353del deletion is predicted to cause the loss of exons 65–100% with time (this study; Keskinen et al. 2008;Tuchman 15 and 16, and both were associated with early-onset pheno- et al. 2008;AhMewetal.2013;Rueggeretal.2014). type. Patients homozygous for c.1143+117_*1353del were Aggressive behaviour and psychiatric problems such as younger brothers of a proband who was not genotyped but psychosis and paranoid ideation were previously reported in presented with the early onset phenotype (Table 2). ASA (von Wendt et al. 1982; Odent et al. 1989; Lagas and Ruokonen 1991;Sijensetal.2006), although these features were not observed in this cohort. Discussion Systemic phenotype This study describes three groups of ASA patients (early-on- set, late-onset and perinatally screened after a familial pro- Our study found a wide range of systemic complications. band) with prolonged follow-up periods and compares the All of them were more frequent in the early-onset group, long-term outcome with regard to the time at initiation of apart from trichorrhexis nodosa, although blood pressure treatment. In contrast to previously reported patients diag- was not systematically investigated. Chronic diarrhoea, nosed by newborn screening, this study describes screened not reported previously, was a major problem in many patients, who had a familial index case with a known patients with endoscopy showing mild inflammation. phenotype. This has been observed in an enterocyte-specific condi- tional knockout mouse model, in which a loss of ASL was Neurological outcome associated with necrotizingenterocolitis(Premkumar et al. 2014). A similar pattern of systemic complications The most common complications of ASA were neurologi- was found in the screened group, suggesting that prospec- cal. Comparison between groups demonstrates a homoge- tive treatment has no preventative effect. neous long-term neurological outcome, with no significant difference in frequency, severity and age of onset for all Genotype–phenotype correlation neurological features assessed. Previously unreported neu- roimaging findings such as focal infarcts or heterotopia Mutation analysis of our cohort identified genotype/ might be related to impaired NO-dependent neuronal mi- phenotype correlation for some of the mutations in agreement gration or microcirculation. with the literature.
461 Chapter 10 - Appendices
JInheritMetabDis
The frequently occurring mutation c.35G>A p.(Arg12Gln) acid can be neurotoxic (D’Hooge et al. 1992)andactivatesN- was associated with the late-onset phenotype in one patient as methyl-D-aspartate receptors (Aoyagi et al. 2001). However, previously suggested for homozygous and compound hetero- this hypothesis is not supported by the observation of the zygous patients (Sampaleanu et al. 2002;Mercimek- early-onset group, which has higher levels of argininosuccinic Mahmutoglu et al. 2010; Balmer et al. 2014). It has been acid but neurological outcomes similar to the other groups. reported that the arginine 12 residue on the N-terminal loop In humans, ASL is crucial for the synthesis of L-arginine, close to the catalytic site of ASL might influence the binding/ which becomes an essential amino acid in ASA. Arginine exit of the substrate without affecting the catalytic site deprivation, associated with altered NO-mediated immune re- explaining the milder phenotype (Sampaleanu et al. 2002). sponses, can lead to site-specific neuronal loss in animal The novel homozygous mutation c.749T>A p.(Met250Lys) models of neurodegenerative diseases (Kan et al. 2015). was observed in two unrelated patients with early-onset pheno- Arginine is a precursor for the synthesis of creatine and type. This mutation involves changes in the protein sequence agmatine (e-Figure 1). Brain spectroscopy showed creatine close to two other amino acid modifications also associated deficiency (this study; Sijens et al. 2006; van Spronsen et al. with early-onset phenotype p.(Glu241Lys) and p.(Trp245fs) 2006; Boenzi et al. 2012). However, the role of secondary (Balmer et al. 2014). creatine deficiency in cerebral dysfunction has not been con- The homozygous mutation c.1153C>T p.(Arg385Cys) vincingly demonstrated (Boenzi et al. 2012). Agmatine is in- has been associated previously with both early-onset (n = volved in learning (Leitch et al. 2011), neuroprotection 4) (Kleijer et al. 2002;Keskinenetal.2008)orlate-onset (Molderings and Haenisch 2012) and anticonvulsant effect phenotypes (n =7)(Kleijeretal. 2002;Keskinenetal. (Demehri et al. 2003). Thus, secondary agmatine deficiency 2008). However, all patients with late-onset phenotype were could explain some of the neurological symptoms. diagnosed before 20 months of life. p.(Arg385Cys) has been Finally, several symptoms may be caused by impaired reported as a founder mutation in the Finnish population NO synthesis. Using an AslNeo/Neo mouse model, Erez et al. (Keskinen et al. 2008)andisassociatedwithverylowASL (2011)showedthatdefectiveASLisresponsibleforthe activity affecting an amino acid near the catalytic site (Hu loss of the catalytic function of the enzyme and affects et al. 2015). the structure of a multi-protein complex incorporating NOS. This disrupts the NOS-dependent NO synthesis and Pathophysiology of ASA leads to systemic NO deficiency. Hypoargininaemia can lead to uncoupling of NOS, decreased NO production and in- Various pathophysiological mechanisms have been pro- creased generation of free radicals that damage tissues (e- posed to account for the long-term complications of Figure 1) (Nagamani et al. 2012b). Reactive oxygen species ASA. Argininosuccinic acid may be toxic to the brain, interfere with NO production and regulation of the microcir- either directly or via the formation of guanidino com- culation (Shu et al. 2015). In addition decreased NO levels pounds. Raised guanidinoacetate was reported on brain spec- might affect protein S-nitrosylation (Jaffrey et al. 2001), troscopy of ASA patients in the grey (3.63 ± 0.6 mmol/L) and which in turn regulates histone methylation and gene expres- the white matter (3.52 ± 0.09 mmol/L) (Sijens et al. 2006;van sion (Nott and Riccio 2009). In the brain, NO plays a key-role Spronsen et al. 2006) and may be explained by L-arginine as a signalling molecule (Riccio 2010)involvedinneurotrans- supplementation (Sijens et al. 2006). In our study, levels of mission (Garthwaite 2008), regulation of neuronal differenti- guanidinoacetate were similar to controls in basal ganglia but ation (Peunova and Enikolopov 1995; Lameu et al. 2012)and slightly elevated in white matter (1.05 ± 0.41 mmol/L). migration (Bredt and Snyder 1994;Nottetal.2013). In hu- Patients with guanidinoacetate methyltransferase (GAMT) de- man, NO therapy was reported to have mild neurocognitive ficiency have much higher guanidinoacetate concentrations in benefit (Nagamani et al. 2012b). In this study, NO deficiency brain (3.4-3.6 mmol/L) (Stockler et al. 1994)andCSF(11– might account for the neuropathology underlying the neuro- 12 μmol/L) (Stockler-Ipsiroglu et al. 2014). There is also imaging findings such as local parenchymal infarcts or nodu- some evidence of raised guanidinoacetate in patients with lar heterotopia due to impaired microcirculation and abnormal hyperargininaemia due to Arginase deficiency, with variable neuronal migration during development respectively. Besides CSF guanidinoacetate concentrations (up to 0.127 μmol/L neurological implications, NO is involved in various physio- versus controls 0.049 μmol/L) (Deignan et al. 2010), although logical processes such as vasodilatation (Cosby et al. 2003), the spectral peak of guanidinoacetate at 3.8 ppm was not seen liver fibrosis (Diesen and Kuo 2011), muscle strength and in a cohort of adult patients with hyperargininaemia (Carvalho performance (De Palma et al. 2014), kidney filtration rate et al. 2014), while in a 3 year old patient a prominent peak at (Satriano et al. 2008)andgutphysiology(Vallanceand 3.8 ppm was ascribed to arginine, which in vitro has resonances Charles 1998;Premkumaretal.2014;Bogdan2015). at 3.75 and 3.23 ppm (Wishart et al. 2007) and may have Therefore, NO deficiency might be involved at least partially masked any detectable guanidinoacetate. Guanidinosuccinic in various symptoms highlighted in this study including
462 Chapter 10 - Appendices
J Inherit Metab Dis chronic hepatitis, myopathy-like phenotype, chronic diarrhoea Aoyagi K, Shahrzad S, Iida S et al (2001) Role of nitric oxide in the and systemic hypertension. synthesis of guanidinosuccinic acid, an activator of the N-methyl- D-aspartate receptor. Kidney Int Suppl 78:S93–S96 Balmer C, Pandey AV, Rufenacht V et al (2014) Mutations and polymor- Optimising therapeutics in ASA phisms in the human argininosuccinate lyase (ASL) gene. Hum Mutat 35:27–35 Boenzi S, Pastore A, Martinelli D et al (2012) Creatine metabolism in This study demonstrates persisting neurological and systemic urea cycle defects. J Inherit Metab Dis 35:647–653 disease not obviously related to hyperammonaemia in ASA Bogdan C (2015) Nitric oxide synthase in innate and adaptive immunity: patients on conventional treatment. Although some organs an update. Trends Immunol 36:161–178 (liver, kidney, gut) are more frequently affected in early- Bredt DS, Snyder SH (1994) Transient nitric oxide synthase neurons in onset patients, who have higher ammonia and ASA levels, embryonic cerebral cortical plate, sensory ganglia, and olfactory epithelium. Neuron 13:301–313 this is strikingly not the case for the brain. Our observation Burgard P, Kolker S, Haege G, Lindner M, Hoffmann GF (2016) of parenchymal infarcts, nodular heterotopia and the report of Neonatal mortality and outcome at the end of the first year of life mild neurological improvement after NO therapy (Nagamani in early onset urea cycle disorders—review and meta-analysis of et al. 2012b) support the role of NO deficiency in the patho- observational studies published over more than 35 years. J Inherit Metab Dis 39:219–229 physiology of the brain disease in ASA. Currently correction Carvalho DR, Farage L, Martins BJ, Brum JM, Speck-Martins CE, of NO deficiency is not considered in the conventional treat- Pratesi R (2014) Brain MRI and magnetic resonance spectroscopy ment of ASA. Liver transplantation (Marble et al. 2008; findings in patients with hyperargininemia. J Neuroimaging 24:155– Newnham et al. 2008) cures the urea cycle but would not be 160 expected to correct the systemic NO-arginine cycle defect. Cosby K, Partovi KS, Crawford JH et al (2003) Nitrite reduction to nitric Neo/Neo oxide by deoxyhemoglobin vasodilates the human circulation. Nat Similarly, successful liver-targeted gene therapy in Asl Med 9:1498–1505 mouse did not correct extra-hepatic features such as defective Davison JE, Davies NP, Wilson M et al (2011) MR spectroscopy-based NO-mediated vascular relaxation (Nagamani et al. 2012b). brain metabolite profiling in propionic acidaemia: metabolic chang- Future therapeutic approaches in ASA might consider es in the basal ganglia during acute decompensation and effect of liver transplantation. Orphanet J Rare Dis 6:19 targeting the NO deficiency, which could include the use of De Palma C, Morisi F, Pambianco S et al (2014) Deficient nitric oxide an enriched nitrate diet, nitrate therapy (Nagamani et al. signalling impairs skeletal muscle growth and performance: in- 2012b; Erez 2013) or multiorgan-targeted gene replacement. volvement of mitochondrial dysregulation. Skelet Muscle 4:22 Deignan JL, De Deyn PP, Cederbaum SD et al (2010) Guanidino com- pound levels in blood, cerebrospinal fluid, and post-mortem brain Acknowledgements Study funded by Action Medical Research for material of patients with argininemia. Mol Genet Metab 100(Suppl Children Charity (grant GN2137). 1):S31–S36 PBM is in receipt of a Great Ormond Street Hospital (GOSH) Demehri S, Homayoun H, Honar H et al (2003) Agmatine exerts anticon- Children’s Charity Leadership award (V2516) and is supported by the vulsant effect in mice: modulation by alpha 2-adrenoceptors and National Institute for Health Research Biomedical Research Centre at nitric oxide. Neuropharmacology 45:534–542 GOSH for Children NHS Foundation Trust and University College D’Hooge R, Pei YQ, Marescau B, De Deyn PP (1992) Convulsive action London. Miss Emma Reid and Mr Matthew Wilson provided technical and toxicity of uremic guanidino compounds: behavioral assessment assistance in ASL sequencing. The authors are indebted to the patients and and relation to brain concentration in adult mice. J Neurol Sci 112: families for their participation in this study. The authors thank the meta- 96–105 bolic teams involved in the care of the patients, especially dieticians. Diesen DL, Kuo PC (2011) Nitric oxide and redox regulation in the liver: part II. Redox biology in pathologic hepatocytes and implications for intervention. J Surg Res 167:96–112 Compliance with ethical standards Erez A (2013) Argininosuccinic aciduria: from a monogenic to a complex disorder. Genet Med 15:251–257 Conflict of interest None. Erez A, Nagamani SC, Shchelochkov OA et al (2011) Requirement of argininosuccinate lyase for systemic nitric oxide production. Nat Open Access This article is distributed under the terms of the Creative Med 17:1619–1626 Commons Attribution 4.0 International License (http:// Ficicioglu C, Mandell R, Shih VE (2009) Argininosuccinate lyase defi- creativecommons.org/licenses/by/4.0/), which permits unrestricted use, ciency: longterm outcome of 13 patients detected by newborn distribution, and reproduction in any medium, provided you give screening. Mol Genet Metab 98:273–277 appropriate credit to the original author(s) and the source, provide a link Garthwaite J (2008) Concepts of neural nitric oxide-mediated transmis- to the Creative Commons license, and indicate if changes were made. sion. Eur J Neurosci 27:2783–2802 Haberle J, Boddaert N, Burlina A et al (2012) Suggested guidelines for the diagnosis and management of urea cycle disorders. Orphanet J Rare Dis 7:32 References Hu L, Pandey AV,Balmer C et al (2015) Unstable argininosuccinate lyase in variant forms of the urea cycle disorder argininosuccinic aciduria. Ah Mew N, Krivitzky L, McCarter R, Batshaw M, Tuchman M, Urea J Inherit Metab Dis 38:815–827 Cycle Disorders Consortium of the Rare Diseases Clinical Research Jaffrey SR, Erdjument-Bromage H, Ferris CD, Tempst P, Snyder SH Network (2013) Clinical outcomes of neonatal onset proximal versus (2001) Protein S-nitrosylation: a physiological signal for neuronal distal urea cycle disorders do not differ. J Pediatr 162:324–329 e321 nitric oxide. Nat Cell Biol 3:193–197
463 Chapter 10 - Appendices
JInheritMetabDis
Kan MJ, Lee JE, Wilson JG et al (2015) Arginine deprivation and im- Premkumar MH, Sule G, Nagamani SC et al (2014) Argininosuccinate mune suppression in a mouse model of Alzheimer’sdisease.J lyase in enterocytes protects from development of necrotizing entero- Neurosci 35:5969–5982 colitis. Am J Physiol Gastrointest Liver Physiol 307:G347–G354 Keskinen P, Siitonen A, Salo M (2008) Hereditary urea cycle diseases in Riccio A (2010) Dynamic epigenetic regulation in neurons: enzymes, Finland. Acta Paediatr 97:1412–1419 stimuli and signaling pathways. Nat Neurosci 13:1330–1337 Kleijer WJ, Garritsen VH, Linnebank M et al (2002) Clinical, enzymatic, Ruegger CM, Lindner M, Ballhausen D et al (2014) Cross-sectional and molecular genetic characterization of a biochemical variant type observational study of 208 patients with non-classical urea cycle of argininosuccinic aciduria: prenatal and postnatal diagnosis in five disorders. J Inherit Metab Dis 37:21–30 unrelated families. J Inherit Metab Dis 25:399–410 Sampaleanu LM, Yu B, Howell PL (2002) Mutational analysis of duck Kolker S, Valayannopoulos V, Burlina AB et al (2015) The pheno- delta 2 crystallin and the structure of an inactive mutant with bound typic spectrum of organic acidurias and urea cycle disorders. substrate provide insight into the enzymatic mechanism of Part 2: the evolving clinical phenotype. J Inherit Metab Dis argininosuccinate lyase. J Biol Chem 277:4166–4175 38:1059–1074 Satriano J, Cunard R, Peterson OW, Dousa T, Gabbai FB, Blantz RC Lagas PA, Ruokonen A (1991) Late onset argininosuccinic aciduria in a (2008) Effects on kidney filtration rate by agmatine requires activa- paranoid retardate. Biol Psychiatry 30:1229–1232 tion of ryanodine channels for nitric oxide generation. Am J Physiol Lameu C, Trujillo CA, Schwindt TT et al (2012) Interactions between the Renal Physiol 294:F795–F800 NO-citrulline cycle and brain-derived neurotrophic factor in differ- Shu X, Keller TC, Begandt D et al (2015) Endothelial nitric oxide syn- entiation of neural stem cells. J Biol Chem 287:29690–29701 thase in the microcirculation. Cell Mol Life Sci 72:4561–4575 Leitch B, Shevtsova O, Reusch K, Bergin DH, Liu P (2011) Spatial Sijens PE, Reijngoud DJ, Soorani-Lunsing RJ, Oudkerk M, van Spronsen learning-induced increase in agmatine levels at hippocampal CA1 FJ (2006) Cerebral 1H MR spectroscopy showing elevation of brain synapses. Synapse 65:146–153 guanidinoacetate in argininosuccinate lyase deficiency. Mol Genet Metab 88:100–102 Marble M, McGoey RR, Mannick E et al (2008) Living related liver Stockler S, Holzbach U, Hanefeld F et al (1994) Creatine deficiency in the transplant in a patient with argininosuccinic aciduria and cirrhosis: brain: a new, treatable inborn error of metabolism. Pediatr Res 36: metabolic follow-up. J Pediatr Gastroenterol Nutr 46:453–456 409–413 Mercimek-Mahmutoglu S, Moeslinger D, Haberle J et al (2010) Long- Stockler-Ipsiroglu S, van Karnebeek C, Longo N et al (2014) term outcome of patients with argininosuccinate lyase deficiency Guanidinoacetate methyltransferase (GAMT) deficiency: outcomes diagnosed by newborn screening in Austria. Mol Genet Metab in 48 individuals and recommendations for diagnosis, treatment and 100:24–28 monitoring. Mol Genet Metab 111:16–25 Molderings GJ, Haenisch B (2012) Agmatine (decarboxylated L-argi- Tuchman M, Lee B, Lichter-Konecki U et al (2008) Cross-sectional mul- nine): physiological role and therapeutic potential. Pharmacol Ther ticenter study of patients with urea cycle disorders in the United 133:351–365 States. Mol Genet Metab 94:397–402 Nagamani SC, Erez A, Lee B (2012a) Argininosuccinate lyase deficien- Vallance P, Charles I (1998) Nitric oxide as an antimicrobial agent: does cy. Genet Med 14:501–507 NO always mean NO? Gut 42:313–314 Nagamani SC, Campeau PM, Shchelochkov OA et al (2012b) Nitric- van Spronsen FJ, Reijngoud DJ, Verhoeven NM, Soorani-Lunsing RJ, oxide supplementation for treatment of long-term complications in Jakobs C, Sijens PE (2006) High cerebral guanidinoacetate and argininosuccinic aciduria. Am J Hum Genet 90:836–846 variable creatine concentrations in argininosuccinate synthetase Newnham T, Hardikar W, Allen K et al (2008) Liver transplantation for and lyase deficiency: implications for treatment? Mol Genet Metab argininosuccinic aciduria: clinical, biochemical, and metabolic out- 89:274–276 come. Liver Transplant Off Publ Am Assoc Study Liver Dis Int von Wendt L, Simila S, Ruokonen A, Puukka M (1982) Argininosuccinic Liver Transplant Soc 14:41–45 aciduria in a Finnish woman presenting with psychosis and mental Nott A, Riccio A (2009) Nitric oxide-mediated epigenetic mechanisms in retardation. Ann Clin Res 14:145–147 developing neurons. Cell Cycle 8:725–730 Waisbren SE, Gropman AL, Members of the Urea Cycle Disorders Nott A, Nitarska J, Veenvliet JV et al (2013) S-nitrosylation of HDAC2 Cosortium, Batshaw ML (2016) Improving long term outcomes in regulates the expression of the chromatin-remodeling factor Brm urea cycle disorders—report from the Urea Cycle Disorders during radial neuron migration. Proc Natl Acad Sci U S A 110: Consortium. J Inherit Metab Dis 39:573–584 3113–3118 Widhalm K, Koch S, Scheibenreiter S et al (1992) Long-term follow- Odent S, Roussey M, Journel H, Betremieux P, David V, Le Marec B up of 12 patients with the late-onset variant of argininosuccinic (1989) Argininosuccinic aciduria. A new case revealed by psychi- acid lyase deficiency: no impairment of intellectual and psycho- atric disorders. J Hum Genet 37:39–42 motor development during therapy. Pediatrics 89:1182–1184 Peunova N, Enikolopov G (1995) Nitric oxide triggers a switch to growth Wishart DS, Tzur D, Knox C et al (2007) HMDB: the Human arrest during differentiation of neuronal cells. Nature 375:68–73 Metabolome Database. Nucleic Acids Res 35:D521–D526
464