Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Identification of genetic modifiers in LRRK2 parkinsonism Trinh, Joanne 2016

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


24-ubc_2017_february_trinh_joanne.pdf [ 2.27MB ]
JSON: 24-1.0340339.json
JSON-LD: 24-1.0340339-ld.json
RDF/XML (Pretty): 24-1.0340339-rdf.xml
RDF/JSON: 24-1.0340339-rdf.json
Turtle: 24-1.0340339-turtle.txt
N-Triples: 24-1.0340339-rdf-ntriples.txt
Original Record: 24-1.0340339-source.json
Full Text

Full Text

IDENTIFICATION OF GENETIC MODIFIERS IN LRRK2 PARKINSONISM by Joanne Trinh BSc, The University of British Columbia, 2012   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in The Faculty of Graduate and Postdoctoral Studies (Medical Genetics)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)    December 2016  © Joanne Trinh, 2016   ii  Abstract Genetic studies have been extremely informative to the pathophysiology of PD. The most common pathogenic mutation discovered is LRRK2 p.G2019S which accounts for 30-40% of Parkinson disease (PD) in North African Arab Berbers, 18-30% in Ashkenazi Jews and 1-3% in Caucasians. Although LRRK2 p.G2019S parkinsonism is considered a monogenic form of disease, disease penetrance of motor symptoms is variable. We hypothesize that genetic factors can modulate the phenoconversion of LRRK2 p.G2019S which could lead to treatments that prevent onset or delay disease progression. Clinical characterization of LRRK2 p.G2019S carriers from Tunisia was performed by analysis of motor and non-motor features. Genetic analysis of age of onset as a genetic trait was performed in a cohort of Tunisian Arab Berbers with LRRK2 p.G2019S. Short-tandem repeat genotyping (4cM resolution) and non-parametric and model-based genome-wide linkage was evaluated in 41 multi-incident LRRK2 p.G2019S families. High-density locus-specific genotyping and association analyses were also performed in 232 unrelated LRRK2 p.G2019S carriers. Genome sequencing in a subset of 25 subjects informed imputation and haplotype analyses. Validation analysis used Sanger sequencing and Taqman genotyping on additional LRRK2 p.G2019S carriers originating from Algeria, France and Norway. Whole transcriptome, candidate gene and protein expression was assessed in striatum from 60 human brains. Significant linkage was identified on chromosome 1q23.3-24.3 (model-based LOD=4.99, D1S2768). In the chromosome 1q23.3-24. interval higher-resolution SNP genotyping, association and haplotype mapping nominated genetic variability within DNM3 as an age of onset modifier of disease penetrance (rs2421947 nominal p<10-5; haplotype p=1.67  x 10-7). In terms of age of onset the penetrance of parkinsonism in LRRK2 p.G2019S carriers varies as a iii  function of DNM3 genotype; rs2421947 is a haplotype-tag for which median onset in GG carriers is 13 years younger than CC carriers (HR 1.63 CI=1.05-2.63, p=0.03). DNM3 rs2421947 variability is also directly correlated with dynamin 3 mRNA and protein expression in human brain striatum (p<0.05).    Dynamin 3, shown to complex with endophilin A, LRRK2 and vacuolar protein sorting 35, localizes to the endocytic machinery of dendritic spines to modulate receptor recycling and excitatory synaptic transmission, now suggests novel targets for therapeutic development in Parkinson’s disease.   iv  Preface  All of the work presented was conducted at the Centre for Applied Neurogenetics (CAN), part of the Djavad Mowafaghian Centre for Brain Health (CBH) at the University of British Columbia. CAN was established by Dr. Matthew Farrer (Principal Investigator). Study and experimental approaches were approved by the University of British Columbia Ethics Board. UBC Research Ethics (H10-02191) and ethics certificate (#5885 – 13) for Disease penetrance of LRRK2 Gly2019Ser parkinsonism, LRRK2 G2019S disease penetrance modifiers and Clinicogenetic studies of LRRK2 G2019S in Tunisia was obtained.  All manuscripts published or in preparation have been written under the guidance of Dr. Matthew Farrer. I collected all genetic data and performed clinical and genetic analysis with contributions from collaborators (Dr. Jan Aasly, Dr. Faycel Hentati, Dr. Suzanne Lesage, Dr. Alexis Brice, Dr. Tatiana Foroud, Dr. Rick Myers), graduate student (Emil Gustavsson), and committee members (Dr. Angie Brooks-Wilson, Dr. Denise Daley, Dr. Carolyn Brown).   Chapter 1: Parts of this chapter has been published as a review: Trinh et al (2013) Advances in the genetics of Parkinson disease, Nature Neurology Reviews. All tables and figures have been adapted and added on from Trinh et al 2013.   Chapter 2: Parts of chapter contains published data from Trinh et al (2014) Disease penetrance of late-onset Parkinson disease, JAMA Neurology and Trinh et al (2014)  and LRRK2 parkinsonism in Tunisia and Norway: A comparative analysis of disease penetrance (2014) Neurology. Published work was done through collaborations with Dr. Jan Aasly from Trondheim University and Dr. Faycel Hentati from Tunis Neurologie institute.  v   Chapter 3: Parts of chapter contains published data Trinh et al (2013) A comparative study on LRRK2 parkinsonism. Neurobiology of Aging. The collection of data and questionnaires was funded by the Michael J Fox Foundation. Conference abstract: A comparative study of LRRK2 G2019S parkinsonism and idiopathic Parkinson’s disease in Tunisia. 3rd World Parkinson Congress. October 1-4, 2013. Montreal, Canada.  Conference abstract:  Identification of LRRK2 p.G2019S disease modifiers. 62nd Annual Meeting of The American Society of Human Genetics, November 6-10, 2012 in San Francisco, California.  Chapter 4: Written as a manuscript: DNM3 modifies age of onset in LRRK2 parkinsonism. Lancet Neurology. 2015 (accepted). Conference abstract: DNM3; a genetic modifier of LRRK2 parkinsonism. 64th  Annual Meeting of The American Society of Human Genetics . October 18-22 2014, San Diego, California, USA.  Chapter 5: Section 1 has been used in a DFG grant, project title “Reduced penetrance in hereditary movement disorders: elucidating mechanisms of endogenous disease protection”.    For the use of article, figures and tables, all copyright permissions have been obtained through journal publishers.   vi  Table of contents Abstract ........................................................................................................................................... ii Preface............................................................................................................................................ iv Table of contents ............................................................................................................................ vi List of tables ................................................................................................................................... ix List of figures ................................................................................................................................. xi List of abbreviations .................................................................................................................... xiii Glossary of terms ........................................................................................................................ xvii Acknowledgments........................................................................................................................ xix 1. Chapter 1: Introduction ....................................................................................................... 1 1.1. General features of Parkinson disease ........................................................................ 1 1.1.1. Motor features ..................................................................................................... 1 1.1.2. Non-motor features ............................................................................................. 2 1.1.3. Pathology ............................................................................................................ 2 1.2. Identification of genetic mutations in PD ................................................................... 3 1.2.1. Linkage analysis.................................................................................................. 3 1.2.2. Next generation sequencing ................................................................................ 4 1.2.3. Genome-wide case-control association............................................................... 5 1.3. Genes implicated in late-onset autosomal dominant PD ............................................ 6 1.3.1. SNCA .................................................................................................................. 6 1.3.2. LRRK2 ................................................................................................................ 7 1.3.3. MAPT ................................................................................................................. 8 1.3.4. EIF4G1 ................................................................................................................ 9 1.3.5. VPS35 and DNAJC13....................................................................................... 10 1.3.6. CHCHD2........................................................................................................... 11 1.3.7. Recessively inherited gene mutations ............................................................... 12 1.4. GWAS in PD............................................................................................................. 13 1.5. Neurobiological interactions: is there one pathway for PD? .................................... 20 1.6. Reduced penetrance .................................................................................................. 23 2. Chapter 2: Disease penetrance estimates of mutations in late-onset PD .......................... 25 2.1. Introduction: penetrance estimates ........................................................................... 25 vii  2.2. Methods..................................................................................................................... 28 2.3. Results ....................................................................................................................... 29 2.3.1. SNCA: description of duplications, triplication and point mutations ............... 29 2.3.2. LRRK2 penetrance findings between populations ........................................... 36 2.3.3. Other autosomal dominantly-inherited mutations in familial PD ..................... 37 2.4. Discussion ................................................................................................................. 40 3. Chapter 3: A clinical comparison between LRRK2 parkinsonism and idiopathic PD ..... 53 3.1. General clinical features of LRRK2 parkinsonism ................................................... 53 3.2. Methods..................................................................................................................... 54 3.2.1. Motor symptom assessment .............................................................................. 54 3.2.2. Non-motor symptom assessment ...................................................................... 55 3.2.3. Genetic assessment and statistical analysis....................................................... 56 3.2.4. Michael J Fox Foundation (MJFF) database storage ........................................ 56 3.3. Results ....................................................................................................................... 60 3.3.1. Motor features ................................................................................................... 60 3.3.2. Non-motor features ........................................................................................... 68 3.3.3. Disease progression .......................................................................................... 74 3.4. Discussion ................................................................................................................. 75 4. Chapter 4: Dynamin 3 modifies age at onset in LRRK2 parkinsonism ............................ 79 4.1. Introduction ............................................................................................................... 79 4.2. Methods..................................................................................................................... 79 4.2.1. Discovery cohort and replication series ............................................................ 79 4.2.2. Linkage analysis and STR genotyping ............................................................. 80 4.2.3. Genome-wide SNP genotyping and association ............................................... 81 4.2.4. Whole genome sequencing and imputation ...................................................... 82 4.2.5. Sequencing and genotyping .............................................................................. 83 4.2.6. Brains, RNA, ampliseq transcriptome, antibodies ............................................ 83 4.3. Results ....................................................................................................................... 85 4.3.1. Linkage and association of LRRK2 p.G2019S families ................................... 85 4.3.2. Higher resolution mapping ............................................................................... 86 4.3.3. DNM3 expression in brain ................................................................................ 87 viii  4.3.4. Replication cohorts ........................................................................................... 88 4.4. Discussion ................................................................................................................. 88 5. Chapter 5: Elucidating mechanisms of reduced penetrance in Mendelian disease ........ 113 5.1. The importance of reduced penetrance ................................................................... 113 5.2. Factors that influence penetrance............................................................................ 114 5.3. Methods and approaches to identify genetic modifiers .......................................... 116 5.4. Dynamin 3 as potential therapeutic target of LRRK2 parkinsonism ...................... 118 5.5. Conclusion .............................................................................................................. 119 References ................................................................................................................................... 121    ix  List of tables Table 1. Phenotypes associated with genes implicated in late-onset Lewy body PD ................... 15 Table 2. Selected genome-wide association studies in Parkinson disease.................................... 19 Table 3. Estimates of LRRK2 p.G2019S age-associated cumulative incidence ........................... 27 Table 4. Summary of patients included for each mutation into penetrance estimates .................. 31 Table 5. Demographics of unrelated patients and control subjects .............................................. 58 Table 6. Demographics of patients with a family history of parkinsonism within 1o .................. 59 Table 7. Clinical summary of patients .......................................................................................... 61 Table 8.  Parkinsonism in LRRK2 p.G2019S carriers by gender ................................................. 62 Table 9. UPDRS Part IA Mentation, Behaviour and Mood ......................................................... 63 Table 10. UPDRS Part IB Mentation, Behaviour and Mood ........................................................ 64 Table 11. UPDRS Part II Activities of Daily Living .................................................................... 65 Table 12. UPDRS Part III ............................................................................................................. 66 Table 13. UPDRS Part IV Complications of Therapy .................................................................. 67 Table 14. Autonomic dysfunction (SCOPA-Aut) individual scores ............................................ 69 Table 15.  Summary of autonomic assessments compared between LRRK2 parkinsonism and iPD....................................................................................................................................................... 71 Table 16. Summary of cognitive assessment compared between iPD, LRRK2 parkinsonism and control subjects ............................................................................................................................. 72 Table 17. Comparison of sleep scales among LRRK2 parkinsonism and iPD.............................. 73 Table 18. Rate of disease progression associated with age at onset in patients ............................ 74 Table 19. Demographics of discovery cohorts: Tunisian Arab-Berber LRRK2 p.G2019S carriers....................................................................................................................................................... 96 x  Table 20 . Demographics of LRRK2 p.G2019S carriers: replication series ................................. 97 Table 21. Demographics of healthy control brains for expression analysis ................................. 98 Table 22. Primer pairs and custom TaqMan probe design for different DNM3 transcript isoforms in human striatum ......................................................................................................................... 99 Table 23. PLINK association underneath linkage regions.......................................................... 100 Table 24. DNM3 haplotypes associated with AAO.................................................................... 101 Table 25. DNM3 transcript levels correlate with LRRK2, VPS35 and SYNJ1 expression in striatal tissue transcriptome data from normal controls (n=17). ............................................................. 102 Table 26. Sensitivity analysis for different age cut-offs on chromosome 1q23.3-24.3 using non-parametric linkage ....................................................................................................................... 103    xi  List of figures Figure 1. Neurobiological Interactions between implicated genes for PD ................................... 22 Figure 2. Kaplan-Meier survival curves for SNCA mutations. .................................................... 33 Figure 3 Kaplan-Meier survival curves for SNCA ....................................................................... 35 Figure 4. Population-specific penetrance estimates of LRRK2 p.G2019S mutations. ................. 38 Figure 5. Kaplan-Meier survival curves for LRRK2 mutations. .................................................. 38 Figure 6. Kaplan-Meier survival curves for VPS35, EIF4G1 and DNAJC13 mutations. ............ 39 Figure 7. Comparison of SNCA and LRRK2 mutations. ............................................................. 45 Figure 8. Cumulative Incidence of SNCA triplication carriers. ................................................... 45 Figure 9. Cumulative Incidence of SNCA duplication carriers. ................................................... 46 Figure 10. Cumulative Incidence of LRRK2 p.N1437H carriers. ................................................ 46 Figure 11. Cumulative Incidence of LRRK2 p.R1441C carriers. ................................................. 47 Figure 12. Cumulative Incidence of LRRK2 p.R1441G carriers. ................................................ 47 Figure 13.  Cumulative Incidence of LRRK2 p.Y1699C carriers. ............................................... 48 Figure 14. Cumulative Incidence of LRRK2 p.G2019S carriers. ................................................. 48 Figure 15. Cumulative Incidence of Ashkenazi Jewish LRRK2 p.G2019S carriers. ................... 49 Figure 16. Cumulative Incidence of Tunisian Arab-Berber LRRK2 p.G2019S carriers. ............. 49 Figure 17. Cumulative Incidence of Norwegian LRRK2 p.G2019S carriers. .............................. 50 Figure 18. Cumulative Incidence of EIF4G1 p.R1205H carriers. ................................................ 50 Figure 19. Cumulative Incidence of VPS35 p.D620N carriers. ................................................... 51 Figure 20. Cumulative Incidence of DNAJC13 p.N855S carriers................................................ 51 Figure 21. World map with LRRK2 mutations ............................................................................ 52 Figure 22. Chromosome 1 linkage peak ....................................................................................... 94 xii  Figure 23. Age-associated cumulative incidence of LRRK2 p.G2019S carriers. ........................ 95 Figure 24. Whole genome sequencing and imputation workflow .............................................. 104 Figure 25. A schematic of the thirteen dynamin isoforms. ......................................................... 105 Figure 26. Multipoint model-based and non-parametric linkage analysis of Tunisian Arab-Berber LRRK2 p.G2019S families. ........................................................................................................ 107 Figure 27 Chromosome 1 Q-Q plot values ................................................................................. 108 Figure 28. DNM3 transcript levels normalized by geometric mean of housekeeping genes ...... 109 Figure 29. Dynamin 3 protein levels normalized by GAPDH .................................................... 110 Figure 30. Dynamin 3 staining in cortical neurons ..................................................................... 111 Figure 31. Flow diagram of discovery and replication cohorts .................................................. 112    xiii  List of abbreviations AAO Age at onset ABCA7 ATP-Binding Cassette, Sub-Family A (ABC1), Member 7 AD Alzheimer’s disease ALS Amyotrophic lateral sclerosis AOO Age of onset APOC3 Apolipoprotein C-III APOE Apolipoprotein class E APP Amyloid Beta (A4) Precursor ATP13A2 ATPase Type 13A2 BACE1 Beta-site amyloid precursor protein cleaving enzyme 1 BLBD Brainstem Lewy body disease C9orf72 Chromosome 9 open reading frame 72 CF Cystic fibrosis CFTR Cystic fibrosis transmembrane conductance regulator CHCHD2 Coiled-coil-helix-coiled-coil-helix domain containing 2 CI Confidence interval CRF Clinical research forms CNV Copy number variation DaT  Dopamine transporters DJ1 Protein deglycase peptidase C56 family DLB Dementia with lewy bodies DLBD Diffuse Lewy body disease DNAJC13 DnaJ (Hsp40) Homolog, Subfamily C, Member 13 DNA Deoxyribonucleic acid DNM3 Dynamin 3 DYT1 Torsion dystonia-1 EIF4G1 Eukaryotic Translation Initiation Factor 4 Gamma, 1 ESP Exome sequencing project ET Essential tremor xiv  ExAC Exome aggregation consortium F-DOPA Fluorodopa FTD Frontotemporal dementia GAK-DGKQ Cyclin G-associated kinase/ Diacylglycerol Kinase loci in GWAS GAPDH Glyceraldehyde-3-phosphate dehydrogenase GCH1 GTP cyclohydrolase 1 GDS Geriatric depression scale GRS Genetic risk score GSK GlaxoSmithKline GTP Guanosine-5’-triphosphate GWAS Genome-wide association studies HD Huntington disease hLOD heterogeneity LOD HPRT Hypoxanthine phosphoribosyl-transferase HTT Huntingtin  HWE Hardy-Weinberg equilibrium IBD Identity by descent IBS Identity by state IPD Idiopathic PD IPSC Induced human pluripotent stem cells LD Linkage disequilibrium LDL Low-density lipoprotein L-dopa Levodopa LOD Logarithm of odds LRRK2 Leucine-rich repeat kinase 2 MAF Minor allele frequency MAPT Microtubule-Associated Protein Tau MDS-UPDRS Movement disorders society unified Parkinson disease rating scale MJFF Michael J Fox Foundation MMSE Mini-Mental State Examination MOCA Montreal Cognitive Assessment xv  MSA Multiple system atrophy NBIA Neurodegeneration with brain iron accumulation NPC1L1 Niemann-Pick C1-Like 1 NPL Nonparametric linkage NUCKS1 Nuclear Casein Kinase And Cyclin-Dependent Kinase Substrate 1 OR Odds Ratio PCSK9 Proprotein convertase subtilisin/kexin type 9 PD  Parkinson disease PINK1 PTEN-induced putative kinase 1 PLA2G6 Phospholipase A2, Group VI PRIMA Preferred Reporting Items for Systematic Reviews and Meta-analyses PRKN Parkin; official name PARK2 RAB7L1 Member of the RAS Oncogenefamily; also known as RAB29 REM Rapid eye movement REP1 Dinucleotide repeat sequence in promoter of SNCA RIN RNA integrity number RNA Ribonucleic acid RME-8 Receptor mediated endocytosis 8 SCOPA-AUT Scales for Outcomes in Parkinson’s disease – Autonomic SGCE Sarcoglycan, Epsilon SLC41A1 Solute carrier family 41 magnesium transporter, member 1 SLC45A3 Solute carrier family 45, member 3 SNCA  alpha-synuclein SNP Single nucleotide polymorphism STR Short tandem repeats SYNJ1 Synaptojanin 1 SYP Synaptophysin TDP TAR DNA-binding protein THAP1 THAP Domain Containing, Apoptosis Associated Protein 1 TLBD Transitional Lewy body disease TMEM175 Transmembrane protein 175 xvi  TOR1A Torsin family 1, member A (torsin A) UPDRS Unified parkinson disease rating scale VCF Variant calling file VPS26 Vacuolar protein sorting 26 VPS29 Vacuolar protein sorting 29 VPS35 Vacuolar protein sorting 35 WES Whole exome sequencing WGS Whole genome sequencing    xvii  Glossary of terms 1000 Genomes First project to sequence the genomes of a large number of people to provide a comprehensive resource on human genetic variation ( ) Apoptosis Process of programmed cell death Bradykinesia Slowness of movement ClinVar A freely accessible, public archive of reports of the relationships among human variations and phenotype with supporting evidence (  dbSNP Database of single nucleotide polymorphisms (  DYT1 Dystonia caused by mutations in TOR1A ENCODE The Encyclopedia of DNA Elements (  Exome Part of the genome formed by exons, sequences which when transcribed remain within the mature RNA after introns are removed by RNA splicing Familial  With family history Idiopathic Disease with unknown pathogenesis Imprinting Epigenetic phenomenon by which certain genes are expressed in a parent-of-origin-specific manner Levodopa Drug treatment for symptoms of Parkinson disease Low frequency variant Changes in the genome that deviates from the reference that is 1-5% minor allele frequency in the general population Modifiers Potential factors that influence the disease phenotype Monogenic Inheritance of a phenotype controlled by one gene Mutation Change in the genome that is rare <1%.  PLINK Free, open-source whole genome association analysis toolset, designed to perform a range of basic, large-scale analyses in computationally efficient manner xviii  (  Polymorphism Common change in the genome (minor allele frequency >5% Postural instability Disabling sign of Parkinson disease influenced by balance  Rare variant Changes in the genome that deviates from the reference that is <1% minor allele frequency in the general population Rigidity Stiffness of the body Sporadic Disease with unknown pathogenesis and no family history Transcriptomics Study of the transcriptome: the complete set of RNA transcripts that are produced by the genome under specific circumstances or in a specific cell, using high throughput technologies such as microarray analysis  Tremor  Involuntary quivering movement Variant Changes in the genome that deviates from the reference    xix  Acknowledgments I would like to thank the many patients and their families who volunteered, and the longitudinal efforts of the clinical teams involved. Initial studies in Tunisia on familial parkinsonism were in collaboration with Lefkos Middleton, Rachel Gibson and the GlaxoSmithKline PD Programme Team (2002-2005). Subsequent clinical and molecular genetic analysis was supported through Mayo Foundation, GlaxoSmithKline and National Institutes of Health. The Michael J Fox Foundation generously supported clinical studies of LRRK2 p.G2019S in Tunisia and subsequent whole genome sequencing (2008-2011), Canada Excellence Research Chairs program, CIHR/IRSC 275675 (2010-2017) and the Don Rix BC Leadership Chair in Genetic Medicine. Replication series were made possible through the support of the France-Parkinson Association, the Roger de Spoelberch Foundation (R12123DD), the French program “Investissements d’avenir” (ANR-10-IAIHU-06), the Research Council of Norway, Reberg’s legacy, the Norwegian Parkinson Foundation, Parkinson’s Study Group (PSG) PROGENI Investigators. I would like to thank members of my thesis committee, Dr. Angie Brooks-Wilson, Dr. Denise Daley, Dr. Carolyn Brown, for their mentorship and a positive research environment. The critical questions have allowed me to refocus and enhance this dissertation. Most importantly, they have helped me mature into a more independent scientist.  A special thanks to Dr. Matthew Farrer for his knowledge, intellect and supervision. Thank you for believing in this project and my abilities. The opportunities you have given me have been helpful for my academic career. I would like to thank my closest friend, Dr. Carles Vilarino-Guell for his ongoing intellectual and emotional support. I wouldn’t have been able to do this without you. Matt and Carles, I will always look up to you and your scientific values. I xx  would also like to thank all the lab members at the Centre for Applied Neurogenetics, especially Emil Gustavsson and Jas Khinda for their love and humor. I am deeply grateful for the graduate scholarships/financial support I received from the Canadian institutes of health reesearch (CIHR), UBC Faculty of Medicine (FoM), UBC Four year fellowship (4YF), Michael Smith Foreign Exchange Supplement (MSFSS), James Miller committee, Genome BC (LEEF), and the Simons Foundation.  Lastly, I would like to thank my mom, Linda Ninh, my dad, David Trinh, my uncle Thien (Tommy) Ha Trinh and my sister, Angel Trinh for their unconditional love and encouragement. I dedicate my thesis dissertation to my sister, Angel Trinh, who has been through a difficult battle with neurological complications. You have motivated me in every way throughout my graduate career. I owe an immense debt to my family who has sacrificed so much to give me every opportunity to pursue my passions and fulfill my ambitions.     1  1. Chapter 1: Introduction  1.1. General features of Parkinson disease 1.1.1. Motor features Parkinson disease (PD) is the most common neurodegenerative movement disorder with age-related prevalence (Bower, Maraganore, McDonnell, & Rocca, 1999). The mean age of onset is 70 years although 4% of patients develop early-onset disease before the age of 50 (Schrag & Schott, 2006). Approximately 1% of the population is affected at 65 years, increasing to 4–5% in 85-year-olds (de Lau & Breteler, 2006). The burden to patients, families, caregivers and society is increasing steadily with population aging and the increased proportion of ‘baby boomers’ aging.  Parkinsonism is characterized clinically by motor dysfunction; a triad of resting tremor, bradykinesia, rigidity and postural instability (Fahn, 2003; L. W. Ferguson, Rajput, Muhajarine, Shah, & Rajput, 2008). Initially, the symptoms are insidious and typically asymmetric, and most patients suffer an inexorable decline. In diagnosed subjects ‘tremor-dominant’, ‘akinetic-rigid’, or ‘mixed’ subtypes may dominate. However, an individual’s age-at-onset, disease course or subsequent co-morbidities are difficult to predict. (Burn et al., 2006; L. W. Ferguson, et al., 2008) A beneficial response to levodopa drug therapy (which treats the clinical motor features) may remain late into the disease. However, with disease progression, optimizing the treatment to patients is challenging and generally requires increased dosing. Side effects include troubling ‘on/off’ motor fluctuations and peak-dose dyskinesias (uncontrolled hyperkinetic movements). (Fahn, 2000) The progressive loss of dopaminergic innervation to the striatum may be confirmed using several imaging modalities, including DaTscan,18 F-DOPA positron emission tomography, and via metabolic changes in brain glucose utilization and blood flow.  2   1.1.2. Non-motor features Non-motor features of PD include autonomic (constipation, cardiac denervation, impotence, orthostatic hypotension and seborrhea), cognitive (bradyphrenia, cognitive decline and dementia), psychiatric (depression, apathy, hallucinations and delusions), sensory problems (hyposmia, anosmia and pain) and sleep disorders (REM sleep behavior disorder and excessive daytime somnolence). (Chaudhuri, Healy, Schapira, & National Institute for Clinical, 2006) These symptoms may be problematic long before the onset of movement disorder and are difficult to treat. For a neuropathologic diagnosis of PD there must be evidence of neuronal loss in the substantia nigra pars compacta accompanied by Lewy body pathology (alpha-synucleinopathy; brainstem (BLBD) or more transitional Lewy body disease (TLBD)) in surviving neurons. (Braak et al., 2003; Goedert, Spillantini, Del Tredici, & Braak, 2013a, 2013b; Spillantini et al., 1997)  With revised clinical criteria, the majority of patients with probable PD that come to autopsy are now confirmed pathologically (Goedert, et al., 2013a). Fatigue is difficult to treat and an important problem by patients. Similar to sleep disturbance, fatigue is also almost universal in patients with PD (Alves, Wentzel-Larsen, & Larsen, 2004; Brown, Dittner, Findley, & Wessely, 2005). 1.1.3. Pathology A clinical diagnosis of dementia with Lewy bodies (DLB) is associated with much more extensive cortical and limbic Lewy pathology and pathologically defined as diffuse Lewy body disease (DLBD), that is often but not invariably associated with parkinsonism (Goedert, et al., 2013a). Similarly, multiple system atrophy (MSA) has parkinsonism and prominent dysautonomia but is characterized pathologically by glial cytoplasmic alpha-synuclein inclusions 3  (Fellner, Wenning, & Stefanova, 2015; Koga et al., 2015). More sparse or ‘incidental’ Lewy body pathology is often found in the healthy aged. Conversely, and more rarely, patients with parkinsonism and clinically atypical ‘Parkinson-plus’ syndromes may have tauopathy (such as post-encephalytic parkinsonism, cortico-basal degeneration, progressive supranuclear palsy (Golbe, 1999; Papapetropoulos et al., 2005) and the parkinsonism-dementia complex of Guam (lytico-bodig)), (Forman et al., 2002) ubiquitin or TDP-43 (TAR DNA-binding protein) proteinopathy (such as Perry syndrome) (Tsuboi et al., 2008) or have non-specific findings such as nigral neuronal cell loss with gliosis (including rapid-onset dystonia-parkinsonism, X-linked dystonia-parkinsonism (Lubag) (Waters et al., 1993).  In PD, genetic mutations can now inform a diagnosis, disease-modeling and basic/pre-clinical research. However, as the rare may inform the general, the text is also punctuated with references to molecular findings from Parkinson-plus syndromes. Past discoveries, especially within monogenic families, appear to coalesce about three interconnected processes: 1) synaptic transmission (exo-, endocytosis) and endosomal receptor sorting and recycling; 2) lysosomal-autophagy, and; 3) mitochondrial quality control and stress response. The emerging synthesis may provide a unified molecular foundation for hypothesis-testing, pharmaceutical development and future trials aimed at disease modification, not only symptomatic relief.    1.2. Identification of genetic mutations in PD 1.2.1. Linkage analysis Linkage analysis methods were theoretically developed in the 1950-70s, but applied in thel late 1990s and early 2000  to analyze rare traits influenced by a major variant or strong genetic effects in families. Linkage analysis is a study of genetic markers and recombination in 4  families with disease. Traditionally, microsatellites were used as genetic markers for linkage. These include polymorphic sequences of DNA that are characterized by repeated sequences. They can be repeats of 2(dinucleotide), 3(trinucleotide), 4(tetranucleotide), which makes them highly heterozygous. The probability of genetic markers segregating with disease within families is calculated and represented as a logarithm of odds (LOD) score (Dawn Teare & Barrett, 2005). Two genetic loci are linked if transmitted from parent to offspring. This concept is used to identify variants that segregate with a disease phenotype in a family. The LOD score is the function of the recombination fraction, the higher the LOD score the higher the evidence of cosegregation (of disease marker and phenotype).  When a significant linkage region is identified (LOD > 3.0) fine-mapping and sequencing underneath the linkage peak is often pursued to elucidate the causal variant (Dawn Teare & Barrett, 2005). There are now many linkage marker sets publicly available (DeCode, Marshfield resources). There are multiple methods for linkage. One is the ‘parametric’ model which requires estimation of disease penetrance, mode of inheritance (i.e. is it dominant or recessive), disease marker allele frequencies. Often times, these allele frequencies are taken from population studies and inheritance is estimated from the pedigree information.  Parametric linkage uses identity by descent within a family. However, a combined score across families is possible. Another is the non-parametric or ‘model-free’ linkage which does not require an input of inheritance and penetrance estimates. In PD, linked regions to disease were often given a “PARK” locus designation. For example, before the gene was identified, the region containing LRRK2 was named “PARK8”.   1.2.2. Next generation sequencing Whole exome or genome sequencing is another approach to identify causal variants segregating with disease. This involves massive parallel sequencing of ‘short-sequences’ that are 5  aligned to a reference genome through computational methods. Variants that deviate from the reference genome are identified and called. However, this leads to many variations which may or may not have an impact on disease. Some more challenges involve the inability to detect differences in repetitive elements, other large repeat expansions. In the case of exome sequencing, only coding variations are captured and the initial hypothesis includes only protein-coding variation (non-synonymous, deletions, frameshifts and loss of function mutations). The advantages of looking at exome sequencing include the ability to determine pathogenic impact through annotation and available protein crystal structures.  Large publicly-available datasets can be used as references. The ‘ExAC’ database (exome aggregation consortium) is one such database to determine potential pathogenicity of a variant of interest. Over 60,000 individuals have been exome sequenced in this consortium, which allows a reliable estimation of frequency in the general population. Many prediction tools have allowed us to estimate how amino-acid changes influence the folding. On the other hand, non-coding regulatory regions may also contribute to disease pathogenesis. These regions are not covered in exome sequencing. Although whole-genome sequencing databases are soon to be available, the annotation in these regions is not as informative as coding regions.  1.2.3. Genome-wide case-control association Genome-wide association studies test common variants’ association with common disease. Linkage detects segments of inheritance within pedigrees, and association detects alleles whose presence is correlated with a trait. Association makes use of linkage disequilibrium (the likelihood of two markers traveling together in a population), thus association takes into account the identity-by-state status rather than the identity-by-descent. Linkage analysis can localize 5-10cM but association extends less than 1cM. There are two main types of genetic association: 6  one is the population-based and the other family-based. Population-based association compares genetic polymorphisms across case and controls. Genetic polymorphisms are variations in the genome which can be common (frequency>5%), low (frequency<5% and >1%) or rare (frequency<1%). Most recently, testing for association in GWAS is at an upwards of 500,000-5 million markers. Family-based association investigates transmission disequilibrium of alleles through pedigrees. This method is not as commonly used, since it is easier to obtain independent cases rather than families. Also case-control design was demonstrated to have greater power than family based designs, provided the disease allele is common (Risch & Merikangas, 1996) . This analysis has been a seminal driver for case-control designs (Risch & Merikangas, 1996) .  Allelic and genotypic frequencies between cases and controls are compared and the expected contribution of these genetic polymorphisms are low (effect sizes or odds ratios = 1-1.5).    1.3. Genes implicated in late-onset autosomal dominant PD A summary of pathogenic mutations and genes implicated in late-onset autosomal dominant PD is described in Table 1.  1.3.1. SNCA  SNCA encodes for protein a-synuclein. Single nucleotide polymorphisms (SNPs)  in SNCA are found to be associated with PD across multiple ethnic populations: Caucasian, Japanese, Tunisian Arab Berbers. A mega-meta GWAS of PD has replicated and shown SNCA to have the most robust effect (Nalls, Pankratz, Lill, Do, Hernandez, Saad, DeStefano, Kara, Bras, Sharma, Schulte, Keller, Arepalli, Letson, Edsall, Stefansson, Liu, Pliner, Lee, Cheng, International Parkinson's Disease Genomics, et al., 2014). There is association of the promoter (REP1) in the 5’ end of SNCA. REP1 associated SNPs also influence transcription of SNCA. 7  However, the functional consequence of other associations within intron 4 and the 3’ end of SNCA has yet to be discovered.  Many pathogenic mutations within SNCA have been found, traditionally through linkage analysis and fine mapping. The first mutation identified was SNCA p.A53T (Polymeropoulos et al., 1997). Since then, the field has identified more pathogenic point mutations such as p.A30P, p.E46K, p.H50Q,  p.G51D and copy number variations such as duplications and triplications (Lesage, Anheim, Letournel, Bousset, Honore, Rozas, Pieri, Madiona, Durr, Melki, Verny, Brice, et al., 2013; Trinh & Farrer, 2013).  Alpha-synuclein  is a key component of Lewy body inclusion. Patients with these missense variations have predominantly DLBD pathology. In addition to DLBD, duplication and triplication carriers have prominent nigral and hippocampal neuronal loss. SNCA copy numbers lead to earlier onset and more fulminant LBD and dementia is a prominent clinical feature. SNCA mutations are overall rare and p.A53T seems to be the most frequent one.   1.3.2. LRRK2 LRRK2 (Leucine-rich repeat kinase 2) mutations confer the highest genotypic risk for PD. Thus far, there are six pathogenic mutations identified in LRRK2: p.N1437H, p.R1441C/G/H, p.Y1699C, p.G2019S, p.I2020T. LRRK2 p.G2019S is especially frequent in PD patients of Ashkenazi Jewish or North African Arab-Berber origin, accounting for 13% and 30% of cases in these populations respectively. Common risk factors include p.R1628P and G2385R. Genome-wide association studies have highlighted and replicated LRRK2 as an associated gene for PD. The effect of LRRK2 in GWAS studies has been smaller than that of SNCA. Large-scale genotyping and gene sequencing of LRRK2 have identified risk factors associated with PD. One 8  example is LRRK2 p.G2385R as a risk factor in Asian populations. There’s also evidence of a haplotype in LRRK2 that is inversely associated with PD and other Parkinson-plus syndromes, suggesting that LRRK2 variants in cis or in trans can have different influences on PD risk  (Heckman, Elbaz, et al., 2014; Heckman, Schottlaender, et al., 2014; Heckman et al., 2013; Trinh & Farrer, 2013; Trinh, Farrer, Ross, & Guella, 1993) LRRK2 parkinsonism has pleomorphic pathology. At autopsy, patients with LRRK2 parkinsonism typically have Lewy body or neurofibrillary tangle pathology, with nigral neuronal loss and gliosis in some cases, TDP-43 proteinopathies have also been observed. Pleomorphic pathology can be evident even within families with the same mutation. Intracellular Lewy bodies and Lewy neurities, by definition the pathological hallmark of PD, are largely comprised of aggregated a-synuclein. Clinically, patients with mutations in LRRK2 closely resemble patients with idiopathic PD.   1.3.3. MAPT There are two major haplotypes for MAPT (Tau): H1 and H2. The ancestral haplotype for MAPT involves a paracentric inversion spanning 1.5 Mb (Zody et al., 2008). The H1 allele is overrepresented in patients(Skipper et al., 2004; Spillantini & Goedert, 2001).  Importantly, the most significant associations of MAPT H1 and an H1-subtype (H1c; defined by the major allele of rs242557) in neurodegeneration are with progressive supranuclear palsy, corticobasal degeneration and Parkinson–dementia complex of Guam. These disorders are rare forms of parkinsonism defined by their primary neurofibrillary tangle pathologies consisting of hyperphosphorylated 4R tau—a tau protein isoform with four microtubule-binding domains that results from alternative gene splicing and inclusion of MAPT exon 10. Similar to SNCA and 9  LRRK2, MAPT has been consistently associated with PD in GWAS studies. (Simon-Sanchez & Gasser, 2015)  Postmortem studies of Lewy body disease in which patients had a longitudinal clinical diagnosis of PD have also observed a MAPT H1 association. The H1 association in Alzheimer’s disease is not as compelling  but there are mutations in MAPT segregating in frontal-temporal dementia (FTD)(Rademakers, Cruts, & van Broeckhoven, 2004; Rademakers et al., 2003; Roks et al., 1999).   1.3.4. EIF4G1 EIF4G1 encodes for eukaryotic translation initiation factor 4 gamma 1. A dominantly inherited p.R1205H is linked to late-onset PD. Although seen in multiple families in the initial paper, there is incomplete penetrance of the mutation. Support for EIF4G1 in PD remains equivocal. Some studies have shown that the p.R1205H mutation is present in more controls than patients in Iceland (N. Nichols, Bras, Hernandez, Jansen, Lesage, Lubbe, Singleton, et al., 2015; Siitonen et al., 2013). The study assessed the relevance of EIF4G1 in a large cohort by imputing the p.R1205H mutation. They found 76 icelandic subjects older than 65 years of age that carried the mutation. Another study with the NeuroX chip assayed over 12,000 patients and controls and found 5 control subjects carrying the p.R1205H mutation. The control subjects range between 68-75 years of age. This led the studies to conclude that p.R1205H is a benign variant. However, imputation accuracy needs to be taken into account and the evidence for p.R1205H is still arguable. Reduced penetrance of the mutation could be an explanation for the observed asymptomatic carriers. Exome sequencing was performed for the original French families affected with Parkinson disease. Importantly, EIF4G1 p.R1205H remains the only mutation identified in 10  chromosome 3q26 linked to parkinsonism .  It has not been identified by the Exome Aggregation Consortium (ExAC; Jan 13th 2015 release) of 60,706 subjects, that includes contributions for the 1000 Genomes and NHLBI-GO Exome Sequencing Project (ESP). Of course this does not rule out non-coding variation that is in close proximity to p.R1205H and could be the causal mutation.   1.3.5. VPS35 and DNAJC13 VPS35 p.D620N causes autosomal dominantly inherited parkinsonism. The mutation segregated in a Mendelian fashion. (Vilarino-Guell et al., 2011; Zimprich et al., 2011) The mutation was found in large multi-incident families through exome sequencing. The families do not share haplotypes and the mutation seems to have arisen de novo. Thus far, VPS35 p.D620N is extremely rare and other pathogenic mutations in VPS35 have yet to be found. VPS26 and VPS29 bind to VPS35 to form a functional retromer. However, variants identified thus far in these genes do not seem to segregate with disease (Gustavsson, Guella, et al., 2015). Nonetheless, VPS35 p.D620N has been independently replicated in large Dutch, French, Japanese families and other sporadic patients has made a convincing case as a gene for PD (Ando et al., 2012; Kumar et al., 2012; Lesage et al., 2012; Sharma et al., 2012) . VPS35 is also implicated in other neurodegenerative diseases. Haploinsufficiency in VPS35 increases the neuropathology with AD and the protein conforms to the spatiotemporal model of AD(Small et al., 2005; Wen et al., 2011) . VPS35 could regulate Abeta peptide levels(Small, et al., 2005). The retromer sorts cargo from endosomes in all cell types. VPS35 is expressed in axons and dendrites of neurons, involved in retrograde tracking of APP, BACE1 and is important in plasma membrane trafficking (Bhalla et al., 2012; Steinberg et al., 2013) . VPS35 may have neuron-11  specific functions: there is evidence that overexpressing VPS35 is neuroprotective to dopaminergic neurons (Bi, Li, Huang, & Zhou, 2013). DNAJC13 p.N855S has been found in one large Mennonite kindred by exome sequencing. However, there is a phenocopy and unaffected carriers in the family. The mutation needs to be independently replicated in PD. In the meantime, DNAJC13 p.N855S mutations have also been found to be implicated in Essential Tremor (ET). Other variants in DNAJC13 besides p.N855S are extremely rare and there is lack of segregation analysis done in families thus far (Gustavsson, Trinh, et al., 2015; Rajput et al., 2015; Vilarino-Guell et al., 2014)  Both DNAJC13 and VPS35 are the first genes found to be implicated in PD through next generation sequencing and open new methods for the application of exome and whole genome sequencing in mapping genes for disease. Since the discovery of VPS35, genes in early-onset parkinsonism have also been  identified by exome sequencing. For example, SYNJ1 mutations seem to segregate with early-onset disease in some families (Olgiati et al., 2014; Quadri et al., 2013) . SYNJ1 was first discovered through homozygosity mapping and exome sequencing in an Italian consanguineous family with parkinsonism and dystonia (Quadri, et al., 2013). However, this thesis will focus on autosomal dominant Parkinson disease rather than early-onset Parkinson disease/ Parkinson-plus syndrome.  1.3.6. CHCHD2 CHCHD2 (full name is coiled-coil-helix-coiled-coil-helix domain containing 2) has been implicated in late-onset autosomal PD. CHCHD2 p.T61I was described in two large Japanese families with autosomal dominant PD (Funayama et al., 2015).  Furthermore, CHCHD2 p.R145Q, and 300+5G>A were also identified in other smaller Japanese families with autosomal 12  dominant PD. There is no evidence of common variants in CHCHD2 being associated with PD. Further sequencing revealed rare exonic mutations with unknown significance in LBD patients: the majority of these rare variants were located within the gene’s mitochondrial targeting sequence (Ogaki et al., 2015).    CHCHD2 contains cysteine-x9-cysteine motifs that are important for regulating enzymes in the mitochondrial respiratory chain.  Mutations in Parkin and PINK1 (PTEN-induced putative kinase 1) in juvenile/early-onset parkinsonism are important in the mitochondria respiratory chain. PINK1 knock-outs show reduced mitochondrial ATP synthesis (Grunewald et al., 2009; Pilsl & Winklhofer, 2012; Rakovic et al., 2011; Vos, Verstreken, & Klein, 2015).  Thus, CHCHD2 is functionally compelling and replication of these three mutations in other families is warranted.   1.3.7. Recessively inherited gene mutations  Recessively inherited mutations (homozygous or compound heterozygous loss of function) have also been identified by linkage analysis in parkin (PARK2;PRKN) (Cookson et al., 2003; Mata et al., 2005; Tan et al., 2003; West, Lockhart, O'Farell, & Farrer, 2003), PTEN-induced putative kinase 1 (PINK1) (Ishihara-Paul et al., 2008; Lee et al., 2009; Toft et al., 2007) and DJ-1 (PARK7) (Lockhart, Bounds, et al., 2004; Lockhart, Lincoln, et al., 2004; Maraganore et al., 2004)  , albeit clinical syndromes with juvenile (≤20 years at diagnosis) or early-onset disease (≤45 years at diagnosis). While the majority of cases that have come to autopsy suffer neuronal loss without Lewy body pathology there are noteworthy exceptions in compound heterozygotes (Farrer et al., 2001; Samaranch et al., 2010) .  PRKN loss-of-function may explain 13  ~15% of early-onset cases and the majority (~50%) in which there is a family history of parkinsonism and/or parental consanguinity, albeit without Lewy pathology. Early-onset parkinsonism accounts for <4% of PD in the community although it is more frequently encountered in movement disorders neurology clinics. Recessively inherited mutations have been also implicated in rare, rather atypical Parkinson-plus disorders including Kufor-Rakeb syndrome due to mutations in ATP13A2, neuroaxonal dystrophy due to loss of PLA2G6, (Morgan et al., 2006; Paisan-Ruiz, Washecka, Nath, Singleton, & Corder, 2009) and neurodegeneration with brain iron accumulation (NBIA) due to mutations PANK2, C2orf37, C19orf12, FA2H and WDR45 (Gregory & Hayflick, 2011; Haack, Hogarth, Gregory, Prokisch, & Hayflick, 2013; Haack et al., 2012)    1.4. GWAS in PD Genetic association study (GWAS) looks to find alleles that are observed more often than expected by chance in individuals with a trait of interest than those without.  There are many strengths in this approach and GWAS have made important contributions in the scientific field. Most notably, in neurodegeneration, the APOE association was identified for Alzheimer’s disease (AD) (Harold et al., 2009; Lambert et al., 2013) . The APOE allele had a large and robust effect on AD. The SNCA signal in PD is the most robust across populations in GWAS. Furthermore, many genes that have been linked (identified through families) have also been nominated in GWAS. A basic summary of associated genes found from GWAS is present in Table 2. SNCA, LRRK2, GCH1 are a few genes that harbor genetic risk factors and pathogenic mutations segregating in families (Nalls, Pankratz, Lill, Do, Hernandez, Saad, DeStefano, Kara, Bras, Sharma, Schulte, Keller, Arepalli, Letson, Edsall, Stefansson, Liu, Pliner, Lee, Cheng, 14  International Parkinson's Disease Genomics, et al., 2014; Simon-Sanchez et al., 2011). GWAS findings can be followed up with additional sequencing to identify genetic variants of pathogenicity. For example, loss-of function (LOF) mutations in ABCA7 in patients with AD have been identified. In fact, an ABCA7 LOF mutation segregated in late-onset autosomal dominant AD families (Cuyvers et al., 2015; Hollingworth et al., 2011) .  Like many studies, there are also caveats to GWAS. One main problem is difficulty in pin-pointing the real associated genes and/or functional variants.The PARK16 locus contains five genes (SLC45A3, NUCKS1, RAB7L1, SLC41A1) (Trinh, Vilarino-Guell, & Ross, 2015; Vilarino-Guell et al., 2010).  RAB7L1 seems to be the most studied and most compelling candidate. RAB7L1 has been shown to co-immunoprecipitate with VPS35 and LRRK2 (D. A. MacLeod et al., 2013). RAB7L1 seems to interact with common variants in LRRK2 to modify risk. However, the effect of RAB7L1 is different across populations. Thus, elucidating the real functional variant for RAB7L1 is challenging. Another example is the GAK-DGKQ locus has also been nominated by GWAS and the genomic region consists of three genes. GAK is the most interesting, due to its involvement in clathrin-mediated endocytosis. Even if the locus points to one gene of interest, there may be multiple risk variants to consider. There are multiple variants associated in SNCA and the effect between two variants very close together is difficult to distinguish because they co-segregate during inheritance. However, each variant can alter expression levels of SNCA differently. A novel strategy to identify such functional variants is with human pluripotent stem cells (IPSC) and genome editing techniques. Through CRISPR/Cas9, a common PD-associated risk variant in a non-coding distal enhancer element (located in intron 4) was found to regulate the expression of α-synuclein (Soldner et al., 2016).    15  Table 1. Phenotypes associated with genes implicated in late-onset Lewy body PD Disease OMIM identifier  Gene  Mutations  Age at onset (range)  Synopsis of clinical features  Predominant pathology  References Dominantly inherited late-onset PD  168601  SNCA  Missense: Ala30Pro, Glu46Lys, His50Gln, Gly51Asp, Ala53Thr  60 years (30–80)  Levodopa-responsive parkinsonism  Diffuse LBD  (Kruger et al., 1998; Lesage, Anheim, Letournel, Bousset, Honore, Rozas, Pieri, Madiona, Durr, Melki, Verny, & Brice, 2013; Polymeropoulos, et al., 1997; Proukakis et al., 2013; Zarranz et al., 2004)  605543  SNCA  Locus duplication (and triplication)  31–71 years  (24–48)  Levodopa-responsive parkinsonism, cognitive decline, autonomic dysfunction and dementia; progression more rapid in SNCA triplication cases   Diffuse LBD, with prominent  nigral and hippocampal  (CA2–3) neuronal loss  (Chartier-Harlin et al., 2004; J. Fuchs et al., 2008; Ibanez et al., 2004; Nishioka, Wider, et al., 2010; Singleton et al., 2003) 607060  LRRK2  Missense: Asn1437His, Arg1441Cys/Gly/His, Tyr1699Cys, Gly2019Ser, Ile2020Thr  60 years  (32–79)  Levodopa-responsive parkinsonism consistent with sporadic PD; Brainstem LBD, neurofibrillary tangle or TDP-43 pathology and/or nigral (Paisan-Ruiz, Lang, et al., 2005; Ross et al., 2011; Zimprich et al., 2004)  16  Disease OMIM identifier  Gene  Mutations  Age at onset (range)  Synopsis of clinical features  Predominant pathology  References Common polymorphisms: Ala419Val, Arg1628Pro, Gly2385Arg (Asia)  Protective haplotype: Asn551Lys– Arg1398His–Lys1423Lys  occasionally dystonia, amyotrophy, gaze palsy and dementia  neuronal loss  614203  VPS35  Missense: Asp620Asn  53 years  (40–68)  Tremor-dominant levodopa-responsive parkinsonism, dyskinesia and dystonia, occasionally dementia   Inconclusive, possibly without LBD  (Ando, et al., 2012; Kumar, et al., 2012; Nuytemans et al., 2013; Sheerin et al., 2012; Vilarino-Guell, et al., 2011; Zimprich, et al., 2011)  616361 DNAJC13 Missense: Arg855Ser 67 years  (57.5-76.5) Slowly progressive, late-onset asymmetric parkinsonism, good response to L-dopa.  Lewy body inclusions in carriers and also DNAJC13 staining within these inclusions (Vilarino-Guell, et al., 2014) 616244 CHCHD2 Missense: Thr61Ile 55.5 years (48-61) Typical parkinsonsonian features (bradykinesia, rigidity, gait). L-dopa responsive  NA: Post-mortem yet to be tested for Lewy body inclusions (Funayama, et al., 2015) 17  Disease OMIM identifier  Gene  Mutations  Age at onset (range)  Synopsis of clinical features  Predominant pathology  References  Juvenile and early-onset recessively inherited parkinsonism   600116  PARK2  Numerous missense, exon deletion and duplication mutations  <45 years  (12–58)  Levodopa-responsive parkinsonism, often juvenile and typically slowly progressive  Predominantly nigral neuronal loss, occasionally with synuclein or tau pathology   (Kitada et al., 1998) 605909  PINK1  Missense: Gln129X, Gln129fsX157, Pro196Leu, Gly309Asp Trp437X, Gly440Glu, Gln456X  Rare: locus and exon deletion  Typically <45 years (18–56)  Levodopa-responsive parkinsonism, often akinetic with postural instability/gait disturbance with slow progression; sleep benefit   One case with LBD  (Ishihara-Paul, et al., 2008; Samaranch, et al., 2010; Valente, Abou-Sleiman, et al., 2004; Valente, Salvi, et al., 2004) 606324  DJ-1  Missense: Glu163Lys, Leu166Pro  Exon 1–5 deletion, g.168–185dup  <40 years (24–39)  Levodopa-responsive parkinsonism, psychological and behavioural disturbances, amyotrophy and cognitive impairment     Unknown  (Annesi et al., 2005; Bonifati et al., 2003) 18  Disease OMIM identifier  Gene  Mutations  Age at onset (range)  Synopsis of clinical features  Predominant pathology  References 606693  ATP13A2  Missense: Phe182Leu, Gly504Arg, Gly877Arg, 1019GfsX1021  Exon 13 1306+5G>A Exon 16 22-bp deletion  <20 years  (10–33)  Levodopa-responsive atypical parkinsonism associated with supranuclear gaze palsy, spasticity and dementia   Neuroradiological atrophy with iron accumulation in basal ganglia  (Di Fonzo et al., 2007; Ramirez et al., 2006)  Abbreviations: fs, frameshift; LBD, Lewy body disease; OMIM, Online Mendelian Inheritance in Man; PD, Parkinson disease; X, stop codon.  19  Table 2. Selected genome-wide association studies in Parkinson disease Gene  Chromosome Population SNCA 4q21 USA, UK, France, Japan MAPT 17q21.1 USA, UK, France LRRK2 12q12 USA, Japan HLA-DRA 6q21.3 USA, UK  GAK–DGKQ  4p16 USA, UK PARK16 1q32 USA, UK, Japan BST1 4p15 France, USA    20   1.5. Neurobiological interactions: is there one pathway for PD? Many genes implicated in PD are expressed in the endosomes, synaptic vesicle sorting and recycling and membrane curvature (Figure 1). Alpha-synuclein is important at the presynaptic terminals and promotes exocytosis. Alpha-synuclein has roles in membrane curvature and is expressed in endosomes, multi-vesicular bodies and lysosomes. There is evidence that a-synuclein is also involved in endocytosis with dynamins during clathrin-mediated endocytosis (Vargas et al., 2014). At the post-synapse (medium spiny neurons), LRRK2 is involved with endocytosis by phosphorylating endophilin A at S75 (Matta et al., 2012).  Activation of LRRK2 and PINK1 (recessive mutations cause early-onset parkinsonism) phosphorylate Rab family GTPases(Lai et al., 2015; Steger et al., 2016). An unbiased phospho-proteomics approach identified Rabs with a pThr73 autophosphorylation site (Rab3 , Rab8 and Rab 10) as LRRK2 substrates in vitro. Furthermore, Rab8A, Rab8B and Rab13  are indirectly phosphorylated by PINK1 (Lai, et al., 2015) . Loss of Rab39B causes early-onset parkinsonism(Wilson et al., 2014). Rabs are important for vesicular trafficking and cellular compartmentalization(Clague & Rochin, 2016).  There is also evidence that suggests LRRK2 regulates chaperone-mediated autophagy, microtubule stabilization, mitochondria and Golgi pathways. LRRK2 co-immunoprecipitates with VPS35 and RME-8 (DNAJC13), and is involved in actin polymerization(Munsie et al., 2015). VPS35 is part of the retromer, formed with VPS26 and VPS29. The retromer complex mediates cargo recognition of early endosomes and membrane recruitment. VPS35 mediates recycling from endosomes to the Golgi apparatus. LRRK2, VPS35 and RME-8 directly mediate endosomal protein sorting and recycling, including the delivery of synaptic neurotransmitter receptors and lysosomal proteins to either degradation or endosome to membrane recycling. 21  VPS35 has been shown to interact with eIF4G1 in yeast to modulate alpha-synuclein toxicity(Dhungel et al., 2015) .  Perhaps impairment in synaptic vesicle trafficking and recycling is central to the pathophysiology of PD. When this process is perturbed, cargo retention in the endosome lead to the formation of multivesicular bodies that are destined to fuse with lysosomes for exosomal release. Cell-to-cell transmission of alpha-synuclein proteinopathy has been a highlight in recent research (Luk, Kehm, Carroll, et al., 2012; Luk, Kehm, Zhang, et al., 2012; Wang et al., 2012)  and this may be a consequence of cargo retention and synaptic vesicle trafficking/recycling.    22   Figure 1. Neurobiological Interactions between implicated genes for PD  Key molecular processes in neurons for important genes implicated in PD. Dopaminergic neuron is in green, glutamatergic cortical neuron in blue and medium spiny neuron is in yellow.   23   1.6. Reduced penetrance  It has been almost 20 years since the discovery of the first SNCA mutation in familial PD. Before the discovery of genes implicated in PD, many scientists had thought that PD was caused by environmental factors. Genetics have been extremely informative in the biology of PD, which could lead to new therapies that could help all those with the disease. However, treatment to prevent symptom onset or delay progression has yet to be developed.  Interestingly, large numbers of putative pathogenic mutation carriers are free of disease: there is evidence of reduced penetrance in patients carrying known pathogenic mutations. Penetrance is formally known as conditional probability of being affected with a disease given a genotype. Penetrance can be age-dependent and may even border ‘variable expressivity’ in very subtle disease manifestations (expressivity describes the extent a certain phenotype manifests).  Within genomic research, reduced penetrance has been neglected and is now emerging as a new field of research. The identification of genetic, environmental, lifestyle and biological factors influencing the phenomena of reduced penetrance is of great interest in neurodegeneration. The idea behind discoveries of ‘protective’ genetic factors can help developing relevant therapeutic targets to halt the development of disease. In large-scale 1000 Genomes and ExAC projects, many pathogenic mutation carriers have been identified. For example, there are 47 LRRK2 p.G2019S mutation carriers in the ExAC database that are potentially asymptomatic, although these individuals are not well phenotyped ( In fact, an average genome has 150 sites of protein truncating variants, 10-12,000 sites with protein-altering variants and even up to 30 mutations implicated in rare disease (Auton et al., 2015) . Thus, penetrance estimates may be more reduced than what is estimated in literature. Recognizing the potential of disease modifiers or protective factors has already sparked research initiatives such as the 24  ‘resilience project’ ( at Mount Sinai, ‘wellderly’ ( at Scripps amongst others. These projects focus on healthy, elderly individuals. Studying modifier genes in Parkinson disease and other neurodegenerative disorders are more difficult as this requires much more stringent phenotyping from neurologists with clinical expertise. LRRK2 p.G2019S is the most common gene mutation in familial PD and accounts for the highest attributable risk in PD. The high frequency of LRRK2 p.G2019S in North African Arab Berbers and Ashkenazi Jewish populations give a larger sample size to discover genetic modifiers that can influence penetrance. Although LRRK 2 p.G2019S parkinsonism is considered a monogenic form of disease, the mutation is not fully penetrant. We hypothesize that genetic factors can modulate phenoconversion of LRRK2 p.G2019S.We postulate that the novel modifier genes and DNA variants that are identified will advance our understanding of the biological mechanisms of LRRK2. Second, these genetic factors may prove to be useful therapeutic targets that could be used to delay the onset of PD among those with LRRK2 mutations. Third, screening of these genetic variants could be included as part of LRRK2 genetic testing and results provided as part of genetic counseling to yield better estimates of the likely onset of PD for a particular at-risk individual.  There are three main studies in this thesis 1) comparative analysis of disease penetrance of mutations implicated in late-onset autosomal dominant PD 2) detailed clinical analysis of LRRK2 p.G2019S carriers compared to idiopathic PD 3) Identification of a potential age-at-onset modifier in LRRK2 parkinsonism.    25  2. Chapter 2: Disease penetrance estimates of mutations in late-onset PD   2.1. Introduction: penetrance estimates  There have been many pathogenic mutations identified for PD. However, these mutations are not 100% penetrant. Penetrance is defined as the probability of individuals with a given genotype who exhibit a certain phenotype. There are many methods to assess penetrance in age-associated diseases. In late-onset autosomal dominant PD, the disease onset, progression, pathology and clinical features of mutation carriers can be vastly distinct. For example, SNCA has been associated with LOPD in every population tested (Nalls, Pankratz, Lill, Do, Hernandez, Saad, DeStefano, Kara, Bras, Sharma, Schulte, Keller, Arepalli, Letson, Edsall, Stefansson, Liu, Pliner, Lee, Cheng, Ikram, et al., 2014; Simon-Sanchez & Gasser, 2015). However, a penetrance comparison of point mutations and copy number variations in SNCA has not been assessed in detail. On the other hand, the penetrance of LRRK2 p.G2019S has been explored through various case sampling and statistical analyses (Table 1) (Healy et al., 2008; Marder et al., 2015; Trinh, Amouri, et al., 2014; Trinh, Guella, & Farrer, 2014) . The two main methods that have been used to assess the penetrance of LRRK2 p.G2019S are cumulative incidence plots and kin-cohort. Some studies have inferred genotypes within families (Marder, et al., 2015). The penetrance varies between 10-50% at age 60 (Table 3). The statistical analyses used are also quite variable. The first study on LRRK2 p.G2019S published an age-dependent penetrance within families and derived an estimation by a simple equation (proportion of affected/total carriers) (Kachergus et al., 2005) . Kachergus et al report at age 50 the LRRK2 p.G2019S mutation is 17% penetrant and at age 70 it is 85%. A world-wide consortium of LRRK2 carriers in Europe have reported a similar estimation age 59-79 (28-74%). This was further replicated in 26  Tunisian Arab Berbers (Hulihan et al., 2008) . Interestingly, the Ashkenazi Jewish LRRK2 carriers and Italian LRRK2 carriers had a more reduced estimation ranging from 15-32%.   27  Table 3. Estimates of LRRK2 p.G2019S age-associated cumulative incidence Ethnicity Sample  Statistical Analysis Age range (penetrance) Reference Norwegian, American (United States), Irish and Polish 13 LRRK2 families  22 familial affected carriers Proportion of affected/total carriers 50-70 (17-85%)  (Kachergus, et al., 2005) French and North African families 2 LRRK2 families 6 familial affected carriers Not reported 55-76 (33-100%)   (Lesage et al., 2005) Ashkenazi Jews 2975 familial relatives of 459 probands Kin-cohort  (Wacholder et al., 1998)  Relatives were not genotyped for mutation: probability of carrying mutation was estimated 60-80 (12-24%)  (Clark et al., 2006) Ashkenazi Jews 22 affected carriers   Penetrance calculated from odds ratio Lifetime risk = 35% (Ozelius et al., 2006) Italian (UK Parkinson’s Disease Brain Bank) 36 familial affected carriers Kaplan Meier   60-80 (15-32%)  (Goldwurm et al., 2011; Goldwurm et al., 2007) World-wide (mostly European) 133 LRRK2 families 327 affected members  Kaplan Meier  59-79 (28-74%)  (Healy, et al., 2008)  28     2.2. Methods In this study, we have created a large meta-analysis of published literature on disease onset and clinical phenotypes of late-onset autosomal dominant Parkinson disease. Published literature was included if there was information on ethnicity, mutation, age and age-at-onset/first motor symptom and confirmation of mutation (not inferred). If age-at-onset or age at last contact were not available in the published literature, we requested information from corresponding authors. We also excluded autosomal recessive genes. Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRIMA) guidelines were followed (Vrabel, 2015) .   Literature search involved keywords: SNCA point mutation, SNCA duplication, VPS35, EIF4G1, LRRK2, DNAJC13, Parkinson disease, autosomal dominant, late-onset parkinsonism. Published studies were included if they have the following information: 1) ethnicity of patient or unaffected Ethnicity Sample  Statistical Analysis Age range (penetrance) Reference Arab-Berber 72 affected carriers Kaplan Meier  60-80 (50-100%)  (Hulihan, et al., 2008) European countries, mainly Italy 154 first degree relatives and 190 second degree relatives of 10  probands with p.G2019S   Kin-cohort  (Wacholder, et al., 1998)  *No relatives were genotyped for mutation: probability of carrying mutation was estimated 1st degree  60-80 (12-33%)  2nd degree  60-80 (10-30%) (Goldwurm, et al., 2011)  Northern Spain (Cantabria) 32 carriers Kaplan Meier  60-80 (12-47%)  (Sierra et al., 2011) 29  individual; 2) confirmation of mutation (not inferred); 3) age of patient or unaffected individual; 4) age of onset of patient; 5) gender of the patient or unaffected individual; 6) first motor symptom of patient, and;7) non-motor symptom of patient .  When age of onset, age, gender and age-at-last contact data was not available in the published article, information was requested through the corresponding authors. If this information was not obtainable, the study subjects were excluded. We also excluded articles:1) about autosomal recessive parkinsonism; 2) that reported duplicate data; 3) that were not written in English, and; 4) genes for which significant genetic linkage was not reported.  The age-associated cumulative incidence (disease penetrance) was estimated using a Kaplan-Meier method with age-at-onset as the time variable; asymptomatic carriers were right censored at the age-at-last contact or age-at-death (JMP software, SAS Institute Inc., Cary, NC). Statistical comparisons between survival curves were done with log-rank tests unless otherwise stated.   2.3. Results 2.3.1. SNCA: description of duplications, triplication and point mutations  SNCA harbors both copy number variation and point mutations in patients with PD. Clinically, SNCA triplication carriers have an earlier onset, faster progression and more fulminant disease compared with duplication carriers. These findings from duplication carriers are more closely comparable to typical late-onset PD. SNCA triplication and duplication carriers seldom have dementia. However, the frequency of SNCA mutation carriers are extremely rare and thus, clinical comparisons are difficult to interpret. We assessed the cumulative incidence of  SNCA copy number and point mutation carriers. Penetrance of triplications was higher than duplications and point mutations (log rank p<0.01) 30  Penetrance of triplications had a lower quartile of 31 years, median of 39 years and upper quartile of 46 years (Figure 2). Point mutations had a lower quartile of 42 years, median of 49 years and upper quartile of 60 (n=59). Point mutations were comparable to duplications (log rank p=0.97) which had a lower quartile of 40 years, median of 48 years and upper quartile of 61 years (n=41).  We observed penetrance differences in point mutations. However, the sample sizes were too small for a meaningful interpretation. SNCA p.A53T (n=35) had a mean age-at-onset of 45.9 years; p.A30P (n=5) had a mean age-at-onset 59.8 years; p.E46K had a mean age at onset of 62.3 years; p.H50Q (n=3) had a mean age-at-onset 64.7 years and p.G51D (n=3) had a mean age-at-onset of 32.7 years (Figures 3a-f). A summary of patients included for each mutation is described in Table 4. Cumulatively, SNCA mutations (triplications, duplications, point mutations) had an earlier onset age (mean age at onset 38.5-49.5 years) compared to LRRK2 mutations (mean age at onset 46.8-68.8 years). Unlike SNCA, two copies of the mutant allele (i.e. homozygous G2019S mutations) do not confer significantly higher risk or higher penetrance .     31  Table 4. Summary of patients included for each mutation into penetrance estimates Mutation Carriers included (n) Affected carriers (n) Unaffected carriers (n) Ethnic backgrounds References SNCA point mutation 47 43 4 Greek, British, Korean, Polish, Swedish, English (Golbe, 1999)  SNCA duplications 41 39 2 French, Korean, Italian (J. Fuchs et al., 2007; J. Fuchs, et al., 2008; Nishioka et al., 2009) SNCA triplications 15 15 0 Swedish- American, Japanese (Farrer et al., 2004; Nishioka, et al., 2009) LRRK2 N1437H 10 9 1 Norwegian (Johansen, White, Farrer, & Aasly, 2011) LRRK2 R1441C 27 17 10 Norwegian (Haugarvoll et al., 2008) LRRK2 R1441G 104 62 42 Basque (Haugarvoll & Wszolek, 2009; Marti-Masso et al., 2009; Pchelina, Ivanova, Emel'ianov, & Iakimovskii, 2011; Ruiz-Martinez et al., 2010) LRRK2 Y1699C 7 7 0 Norwegian (Khan et al., 2005; Zimprich, et al., 2004) LRRK2 G2019S 330 291 39 Norwegian, Tunisian, Ashkenazi Jewish (Healy, et al., 2008; Trinh, Amouri, et al., 2014)  32  Mutation Carriers included (n) Affected carriers (n) Unaffected carriers (n) Ethnic backgrounds References LRRK2 I2020T 29 23 6 Japanese (Tomiyama et al., 2006) VPS35 D620N 61 54 7 German, Tunisian, Yemen Jews, Japanese, French (Ando, et al., 2012; Lesage, et al., 2012; Sharma, et al., 2012; Sheerin, et al., 2012; Vilarino-Guell, et al., 2011; Zimprich, et al., 2011) EIF4G1 R1205H 20 20 0 French, Tunisian, Yemen Jews (Chartier-Harlin et al., 2011; N. Nichols, Bras, Hernandez, Jansen, Lesage, Lubbe, & Singleton, 2015; Nuytemans, et al., 2013; Siitonen, et al., 2013) DNAJC13 N855S 18 12 6 Mennonites-Canadian (Vilarino-Guell, et al., 2014) Total Combined 709 592 117      33      Figure 2. Kaplan-Meier survival curves for SNCA mutations.  The probability of being affected at median age 56 is 0.90 for SNCA triplication carriers, 0.70 for SNCA duplication and missense carriers.    34       A B C 35      Figure 3 Kaplan-Meier survival curves for SNCA a) cumulative incidence for all SNCA point mutation carriers, b) p.A30P, c) p.A53T, d) p.E46K, e) p.G51D, f) p.H50Q. The dotted lines represent confidence intervals.  D E F 36  2.3.2. LRRK2 penetrance findings between populations The LRRK2 p.G2019S mutations account for up to 15% in Ashkenazi Jewish populations, 30% in Arab Berber populations and 1% in Caucasian populations. We assessed the penetrance of LRRK2 in the Arab Berber population with an expanding a Tunisian cohort since 2005 (Figure 4) (Hulihan, et al., 2008). The Arab Berber population  cumulative incidence and mean age at onset estimates were consistent with previous studies: mean age at onset 57.1 years; 95%CI, 45.5-68.7 years n=220 (Healy, et al., 2008; Hulihan, et al., 2008). However, the Norwegian penetrance estimates (mean age at onset 63 years; 95% CI 51.4-74.6 years) were reduced compared to Tunisia (p<0.0001). Lastly, the Ashkenazi Jewish population from Israel were comparable to Tunisia (mean age at onset 57.9 years, 95% CI, 54-63 years) (Figure 4).  Furthermore, there are six pathogenic mutations in LRRK2 (p.N1437H, p.R1441C/G/H, Y1699C,  p.G2019S, p.I2020T).  Higher penetrance for p.N1437H and p.Y1699C mutations, may reflect the small sample size (n=10). The cumulative incidence of LRRK2 mutations are significantly different from each other.  Penetrance within the kinase domain (p.G2019S and p.I2020T) are similar and significantly higher than the Roc domain mutations (p.R1441C/G/H) (Figure 5). The p.N1437H is also in the Roc domain but hampered by small sample size. The cumulative incidence of LRRK2 p.I2020T had a lower quartile of 51 years or younger, a median of 55 years of age, and an upper quartile of 60 years or older (n = 29). The estimation was similar to LRRK2 p.G2019S, which had a lower quartile of 49 years or younger, a median of 57 years, and an upper quartile of 67 years or older (n = 330). Lastly, the cumulative incidence of LRRK2 p.R1441C and p.R1441G were the least penetrant. LRRK2 p.R1441C had a lower quartile of 65 years or younger, a median of 71 years, and an upper quartile of 77 years or older (n = 27). p.R1441G had a lower quartile of 60 years or younger, a median of 65 years of age, and an upper 37  quartile of 72 years or older (n = 104).The LRRK2 p.I2020T mutation is largely Japanese, R1441G is largely Basque, R1441C is mostly Belgian. These estimates could also reflect diagnostic or referral differences across regions. Perhaps there is better clinical care for neurodegenerative diseases in different parts of the wold which could reflect higher reporting of disease and earlier age-at-onset estimates.   2.3.3. Other autosomal dominantly-inherited mutations in familial PD VPS35 p.D620N (lower quartile ≤ 45years, median 49 years and upper quartile ≥ 59 years; n= 61) was significantly more penetrant than EIF4G1 p.R1205H (lower quartile ≤ 56 years, median 62 years and upper quartile ≥ 69.5 years; n=20) and DNAJC13 p.N855S (lower quartile ≤ 61 years, median 68 years and upper quartile ≥ 76 years; n=18) (Figure 6).  The age-dependent cumulative incidence was significantly different across mutations (p<0.0001) . Overall SNCA triplication carriers (n=15) had the highest cumulative incidence (penetrance) and LRRK2 p.G2019S carriers in Norway (n=84) had the lowest (Figure 7).  38   Figure 4. Population-specific penetrance estimates of LRRK2 p.G2019S mutations.   Figure 5. Kaplan-Meier survival curves for LRRK2 mutations.  39   Figure 6. Kaplan-Meier survival curves for VPS35, EIF4G1 and DNAJC13 mutations.    40   2.4. Discussion This study summarizes and systematically compares the age-dependent cumulative incidence of  all known  mutations leading to late-onset parkinsonism. Fifteen rare pathogenic variations in five genes (SNCA, LRRK2, VPS35, EIF4G1, and DNAJC13) were assessed. All mutation carriers were combined, whether from the literature or contributed by corresponding authors, providing the most accurate penetrance estimates to date.  Nevertheless, the study has many limitations. These include cultural and environmental differences between populations, access to health care and ascertainment bias. Various diagnostic criteria have to be considered, and the movement disorders neurology expertise at different centres .  Moreover, age at onset is retrospective and subjective, dependent upon a variety of symptoms and signs, although well correlated with age at diagnosis (Reider et al., 2003) .  All comparisons utilized the same statistical measure to estimate cumulative incidence which simplifies comparisons between mutations. The Kaplan-Meier method is a reverse survival curve analysis, ideally suited for sporadic patients and unrelated probands that censors for asymptomatic carriers. In contrast, the kin-cohort method excludes probands, employing just relatives with inferred genotypes to specifically avoid inflating penetrance estimates. However, a disadvantage is that the phenotypic and genotypic information of the relatives may be inaccurate. While analyses have been adapted to compensate for a variety of study designs, Kaplan-Meier and kin-cohort are the major methods employed in penetrance estimates of PD. Bias from the inclusion of probands and family members has been assessed using a variety of statistical methods and sensitivity analyses for LRRK2 p.R1441G and p.G2019S show comparable results (Trinh, Amouri, et al., 2014; Trinh, Guella, et al., 2014) . The sensitivity analysis compared penetrance estimates derived from different methods and sample groups (kin-cohort methods 41  versus Kaplan meier as well as families versus unrelated patients). Herein we are limited by published data, by the relatedness of subjects and the total number of carriers/families with each gene. With these caveats acknowledged, confidence intervals are provided for genetic counseling (Figure 7-20).   Penetrance estimates for monogenic parkinsonism vary by gene, by mutation and by ethnicity. SNCA triplications are more penetrant than duplications for which genomic dosage has been directly correlated to mRNA and protein expression (Farrer, et al., 2004; Nishioka, et al., 2009) . Clinically, SNCA triplication carriers have an earlier onset, faster progression and more fulminant disease compared to duplication carriers which more closely resembles late-onset idiopathic PD (Muenter et al., 1998; Nishioka, Kefi, et al., 2010; Nishioka, et al., 2009) . Seldom do SNCA triplications or duplication carriers have dementia as a first symptom; typically cognitive decline is noted several years after the onset of parkinsonism. Nevertheless, many have a clinical diagnosis of dementia with Lewy bodies (DLB), with diffuse Lewy body disease on autopsy (DLBD).  Overall SNCA point mutations and SNCA duplications are quite similar in penetrance. While the majority of missense carriers of SNCA p.A53T have been described with young onset parkinsonism, with an aggressive course (Golbe, 1999) , and most duplication carriers are described as DLB, they are most comparable. The frequency of SNCA multiplications and point mutations is extremely rare (less than 1% in different populations), thus meaningful comparisons of  clinical features is problematic, although a global study of SNCA multiplication and missense carriers has recently been initiated  (The Parkinson Progression Markers Initiative by The Michael J Fox Foundation for Parkinson’s Research).   LRRK2 mutations confer the highest population-attributable risk to PD but the function of the encoded protein still remains unclear. The majority of pathogenic variants are within three 42  contiguous domains: kinase, ROC and COR (Mills, Mulhern, Liu, Culvenor, & Cheng, 2014) . Penetrance of mutations within the kinase domain (LRRK2 p.G2019S and p.I2020T) are similar, and significantly higher than ROC domain mutation (p.R1441C and p.R1441G). LRRK2 p.R1441C and p.R1441G mutations have similar penetrance estimates (p=0.31). The sample size was too small to compare p.R1441H. However, we observe higher penetrance of LRRK2 p.N1437H, which could be hampered by the rarity of this variant (n=10). COR domain mutations are highly penetrant, but could also be due to a smaller sample size (n=7).   SNCA mutations (triplications, duplications and point mutations) had a larger effect, with an earlier onset (AAO means were from 38.5-49.5 years) compared to LRRK2 mutations (AAO means were from 46.8 to 68.8 years) . SNCA mutation carriers have a more aggressive phenotype whereas  LRRK2 carriers have a more benign clinical course compared to idiopathic PD. In LRRK2 parkinsonism, there is less REM sleep behavior disorder and gastrointestinal dysfunction (Trinh, Amouri, et al., 2014)  which are two main clinical features affected by Braak staging. SNCA mutation carriers primarily have Lewy-body-like inclusions of α-synuclein aggregates (Conway et al 2000 oligomerization SNCA, Wood et al alpha-synuclein 1999). In contrast, LRRK2 carriers (albeit p.N1437H, p.R1441C/G/H, p.G2019S, or p.I2020T) have pleiomorphic pathologies including α-synuclein, 4-repeat-tau, or tar-dna binding-43 proteinopathies on a background of neuronal loss and gliosis. The clinical course probably reflects the burden and type of end-stage pathology.  Age at onset is only one measure of variability across these pathogenic mutations. The pathology in LRRK2 mutation carriers are extremely pleiomorphic. Furthermore, clinical features such as cognitive decline/dementia, autonomic dysfunction can vary between mutation carriers. The LRRK2 p.Y1699C mutation is most unusual with amotrophy, dementia and not just 43  parkinsonism, compared to LRRK2 p.I2020T which has typical parkinsonism which is comparable to idiopathic PD and tauopathies (Hasegawa et al., 2009; Ujiie et al., 2012).  This study highlights the role of ethnicity or environmental factors as a major contributor of penetrance. Stratification of LRRK2 p.G2019S parkinsonism by ethnicity was possible because of the large sample size. Israeli Ashkenazi Jews have a significantly higher penetrance compared to Norwegian LRRK2 p.G2019S mutation carriers, and are comparable in penetrance to Tunisian Arab-Berbers. In New York, the disease in Ashkenazi Jewish carriers is less penetrant (24% penetrance at age 80) (Figure 21); these differences may reflect a sample of 7 carriers, the exclusion of family members (Clark, et al., 2006)  and environmental factors such as orthodox or unorthodox practices. The study by Clark et al has now been further expanded to 90 LRRK2 carriers and include a much larger cohort of Ashkenazi LRRK2 G2019S in New York. The reduced penetrance estimate still stands (26% at age 80 years). However, the statistical method used was different and the sampling included more unaffected LRRK2 p.G2019S carriers with family history (Marder, et al., 2015) . In contrast, similarities in age of onset between Israeli Jews and Tunisian Arab-Berbers carriers may reflect similar genetic and environmental backgrounds (Nebel et al., 2000) . The environment may also play a role in the differences. Furthermore, ascertainment bias in the patients sample sets may influence the data. These cohorts are predominantly tertiary referral clinic-based samples from specialist hospitals, which means age-at-onset can be inflated as patients from the same family may be more aware of symptoms. Nevertheless, ethnic differences are an important consideration in genetic counseling.  Mutations in SNCA, LRRK2, VPS35, EIF4G1 and DNAJC13 have been directly implicated in familial parkinsonism (Trinh & Farrer, 2013) . These proteins are centrally involved in synaptic transmission, early endosomal sorting and recycling, and lysosomal 44  autophagy. Indeed, LRRK2, VPS35, and DNAJC13 directly immunoprecipitate with members of the WASH complex (Helfer et al., 2013) , which regulates actin remodelling and membrane trafficking in these processes. Whether this network is similarly perturbed in idiopathic PD has yet to be established. Differences in the penetrance estimates may reflect the type of substitution, its location and functional consequence. Mutations may affect interactions with binding partners and downstream signaling pathways, thus influencing expression (transcript or protein), and ultimately compensatory mechanisms (genetic and non-genetic).  Age is considered the greatest risk factor for PD and genetic susceptibility is only one influence. The penetrance of mutations in late-onset parkinsonism is also dependent on ethnicity and potentially environmental factors. Thus, heterogeneity between mutation carriers may be an important consideration when identifying modifiers of disease. Prospective, longitudinal evaluation of carriers and further meta-analyses will be required for more precise penetrance estimates, and provide the opportunity to inform therapeutic trials.       45    Figure 7. Comparison of SNCA and LRRK2 mutations.  SNCA mutations are illustrated as red lines and LRRK2 mutations are illustrated as black lines. The cumulative incidence for distinct pathogenic mutations in each gene are shown in figure 2 and figure 4.    Figure 8. Cumulative Incidence of SNCA triplication carriers.  Dotted lines represent confidence intervals. 46   Figure 9. Cumulative Incidence of SNCA duplication carriers. Dotted lines represent confidence intervals.   Figure 10. Cumulative Incidence of LRRK2 p.N1437H carriers. Dotted lines represent confidence intervals.  47   Figure 11. Cumulative Incidence of LRRK2 p.R1441C carriers. Dotted lines represent confidence intervals.   Figure 12. Cumulative Incidence of LRRK2 p.R1441G carriers. Dotted lines represent confidence intervals.  48    Figure 13.  Cumulative Incidence of LRRK2 p.Y1699C carriers. Dotted lines represent confidence intervals.   Figure 14. Cumulative Incidence of LRRK2 p.G2019S carriers. Dotted lines represent confidence intervals.  49   Figure 15. Cumulative Incidence of Ashkenazi Jewish LRRK2 p.G2019S carriers. Dotted lines represent confidence intervals.    Figure 16. Cumulative Incidence of Tunisian Arab-Berber LRRK2 p.G2019S carriers. Dotted lines represent confidence intervals.  50   Figure 17. Cumulative Incidence of Norwegian LRRK2 p.G2019S carriers. Dotted lines represent confidence intervals.   Figure 18. Cumulative Incidence of EIF4G1 p.R1205H carriers. Dotted lines represent confidence intervals.  51   Figure 19. Cumulative Incidence of VPS35 p.D620N carriers. Dotted lines represent confidence intervals.    Figure 20. Cumulative Incidence of DNAJC13 p.N855S carriers. Dotted lines represent confidence intervals.   52    Figure 21. World map with LRRK2 mutations  53  3. Chapter 3: A clinical comparison between LRRK2 parkinsonism and idiopathic PD  3.1. General clinical features of LRRK2 parkinsonism  PD is characterized by four cardinal signs: resting tremor with asymmetry at onset, bradykinesia, rigidity, postural instability and positive response to Levodopa (Postuma et al., 2015) . LRRK2 p.G2019S has the highest genotypic and population attributable risk. The mutation was first shown to segregate with PD in a Norwegian family (Kachergus et al 2005). Overall, the clinical presentation of idiopathic PD (iPD) is similar to LRRK2 parkinsonism (Aasly et al., 2005) . However, there is heterogeneity in the cohorts, sample size and methodology.   Some reports suggest a more severe phenotype in LRRK2 mutation carriers compared to idiopathic PD. LRRK2 p.G2019S carriers were reported to have more severe motor symptoms and dyskinesias (Nishioka, et al., 2009) (Oosterveld et al., 2015) . Depression, hallucinations, sleep issues were reported to be more common in LRRK2 p.G2019S carriers (Pchelina, et al., 2011) . Furthermore, postural instability and gait problems were more common in early-onset LRRK2 carriers (Alcalay et al., 2009) (Alcalay et al., 2015; Marras et al., 2016) . On the other hand, LRRK2 carriers have reported to have slower disease progression, less cognitive impairment, lower depression, less autonomic dysfunction and UPDRS scores (Alcalay, et al., 2009; Healy, et al., 2008) (Marras, et al., 2016) (Tijero et al., 2013) .  A better characterization of LRRK2 carriers is warranted.  Furthermore, collecting and analyzing a large database on clinical features of LRRK2 p.G2019S carriers can lead to a better understanding of the progression of the disease and study of endophenotypes (both motor and non-motor). In the present study, we have expanded the 54  Tunisian Arab-Berber LRRK2 cohort over a period of six years to compare the disease onset, clinical symptoms and disease progression.    3.2. Methods All patients were recruited at the same neurological centre: Mongi Ben Hamida National Institute of Neurology, Tunis. The center provides out-patient and in-patient services for neurological disorders in Tunisia. Local on-site monitoring was independently performed by PRN clinical research ( every 18 months. Clinical examinations were performed and questionnaires were administered by movement disorder specialists Dr. Faycal Hentati, Dr. Samia Ben Sassi, Dr. Fatma Nabli, Dr. Emna Farhat. Diagnoses of PD were made according to the UK PD Society brain bank criteria. Enrollment information including additional family medical history and origin was also collected. Patients and control subjects completed standardized clinical research forms (CRFs), all medical history of patients and families were recorded. Clinical data and blood samples were collected for 778 patients and 580 unaffected subjects (Table 5-6).   3.2.1. Motor symptom assessment Movement disorders society unified Parkinson disease rating scale (MDS-UPDRS) were performed on patients with symptoms of parkinsonism. The MDS-UPDRS consists of four parts: Non-motor experiences of daily living, motor experiences of daily living, motor examination and motor complications (Goetz, Nutt, & Stebbins, 2008; Goetz et al., 2008).  All assessments in the MDS-UPDRS have five responses: 0=normal, 1=slight, 2=mild, 3=moderate, 4=severe. “Slight” refers to symptoms with low frequency or intensity that have no impact on function, “mild” has 55  modest impact on function, “moderate” has considerable impact on function and lastly, “severe” refers to symptoms that prevent function. Medication status of L-dopa “on” and “off” stages were recorded. Hoehn and Yahr staging (1-5), to objectively rate the patient’s disability at a certain time (stage 0 means no signs of disease and stage 5 is wheelchair bound or bedridden), was also recorded.  3.2.2. Non-motor symptom assessment Multiple questionnaires were used for non-motor symptoms. The Schwab and England questionnaire was used for activities of daily living (McRae, Diem, Vo, O'Brien, & Seeberger, 2000) . The rating was performed by the neurologist and ranges from 0% (only vegetative functions are working, bedridden and helpless) to 100% (completely independent, able to do all chores, no slowness or difficulty). Autonomic dysfunction was assessed with Scales for Outcomes in Parkinson’s disease – Autonomic (SCOPA-AUT) (Visser, Marinus, Stiggelbout, & Van Hilten, 2004). There are 25 items that assess gastrointestinal, urinary, cardiovascular, thermoregulatory, and sexual dysfunction. Autonomic problems increase significantly with disease severity (Visser, Marinus, Stiggelbout, et al., 2004; Visser, Marinus, van Hilten, Schipper, & Stiggelbout, 2004) . Geriatric depression scale and Epworth daytime sleepiness scale was used to assess depression and sleep function, respectively (Johns, 1991; Yesavage et al., 1982). REM sleep behavior disorder was noted as 50% of all REM sleep disordered patients by polysonography develop PD. A specially modified “sniffin’ test” was created with the help of neurologist Dr. John Duda for odorant descriptors and distractors for the Arab Berber population. “Sniffin tests” is a validated test consisting  of odor detection, discrimination and sensitivity. It was adopted a trial of 100 control subjects to be ‘culturally’ appropriate. Simplified, culturally appropriate tests were created for largely illiterate population. Cognition was measured using the 56  mini-mental state examination (MMSE), Montreal cognitive assessment (MOCA), six picture test or frontal assessment battery. The latter two were adapted for a largely illiterate population.   3.2.3. Genetic assessment and statistical analysis DNA was extracted by standard procedures (Miller et al 1988) and LRRK2 c.6055G>A (p.G2019S) was genotyped with a TaqMan probe on an ABI7900 analyzer and then verified by sequencing, as previously described (Hulihan, et al., 2008) . Patients with pathogenic mutations in PINK1 or Parkin were excluded from this study (Bonifati, et al., 2003; Valente, Salvi, et al., 2004).  Multivariate regression models were used to investigate and compare different questionnaires and clinical assessments, adjusted for age at onset, disease duration, gender and medical state (on or off levodopa) (JMP software version 10). Cumulative incidence was assessed with Kaplan Meier or kin-cohort analysis and significant differences were detected with either log-rank or Wilcoxon tests.  The log-rank test gives equal weight to all time points in a cumulative incidence plot, whereas the Wilcoxon test gives more weight to earlier time points and requires one group consistently have a higher risk than the other.  3.2.4. Michael J Fox Foundation (MJFF) database storage Each patient and control subjects had a clinical research form ID, linked with a MJFF family ID and individual ID. The data was imported into under six categories: 1) enrollment, 2) UPDRS, 3) medications, 4) non-motor, 5) cognitive testing, 6) environment and lifestyle. The collected data was stored in a database under the LRRK2 cohort consortium ( webpage titled “Parkinson’s disease in Tunisia”.  However, the database is no 57  longer maintained and is warehoused in Tunisia and UBC. It has also been submitted to the MJFF LRRK2 cohort consortium to be made broadly accessible. However, it may be important to make this publicly available.  58  Table 5. Demographics of unrelated patients and control subjects   Non-LRRK2 p.G2019S LRRK2 p.G2019S carriers   Patients Controls Patients Controls N 350 399 220 (38%) 12 (3%) Number of men (%) 187 (53%) 203 (51%) 124 (56%) 6 (50%) Mean age (SD) years 66.6 (12.9) 61.1 (11.1) 67.6 (12.6) 56.7 (10.9) Median age (IQR) 69 (59–76) 59 (53–69) 69 (48–90) 54.5 (38–72) Mean age of onset (SD) 55.3 (14.4) - 57.1 (11.6) - Median age at onset (IQR) 58 (46–66) - 57 (40–74) - Mean disease duration 8.10 (5.2)  9.07 (5.03)  Range disease duration 2-23  3-23    59  Table 6. Demographics of patients with a family history of parkinsonism within 1o   Affected probands Affected family members Unaffected family members N 162 126 169 Number of men (%) 89 (55%) 73 (58%) 74 (44%) Mean age (SD) 67.0 (14.2) 76.6 (14.0) 59.4 (17.9) Median age (IQR) 68 (46–90) 80 (66–95) 57 (26.5–87.5) Mean age at onset (SD) 55.0 (14.0) 59.2 (14.4) - Median age at onset (IQR) 56 (38–74) 60 (42–79) - LRRK2 p.G2019S carriers 80 (49%) 46 (36%) 71 (42%)    60   3.3. Results  3.3.1. Motor features The motor features were largely indistinguishable between iPD or LRRK2 carriers. Age and age-at-onset were similar between LRRK2 p.G2019S homozygotes (n=32), heterozygotes (n=177) and idiopathic PD (n=324) (Table 7). Comparison of first symptoms between idiopathic PD and LRRK2 carriers was similar. Tremor was the most predominant first symptom across all genotypes (range from 71.9% to 81.5%) (Table 7).  When stratifying by gender, the results remain comparable across LRRK2 parkinsonism and iPD (Table 8).  Unified Parkinson disease rating scale (UPDRS) has also been assessed and comparisons were performed with regression modeling. There were no remarkable differences between  LRRK2 parkinsonism and iPD (Table 7-13). However, early-onset LRRK2 carriers tend to suffer more rigidity than late-onset carriers (UPDRS III rigidity score, p=0.05). Although this value is not significant after Bonferroni correction.     61  Table 7. Clinical summary of patients LRRK2 p.G2019S carrier status Homozygous Heterozygous iPD p-values N 32 177 324  Mean age of onset (SD) 54.5 (12.4) 57.3 (11.9) 56.2 (13.8) 0.46 Median age of onset (IQR) 56 (40–72) 58 (41–75) 58 (38–78)  First symptom (%)      Tremor 23 (71.9%) 135 (76.3%) 265 (81.5%) 0.53  Gait or balance deterioration 4 (12.5%) 23 (13.0%) 26 (8.0%)   Muscle cramping or dystonia 2 (6.2%) 5 (2.8%) 7 (2.2%)   Shoulder stiffness 0 5 (2.8%) 8 (2.5%)   Other 3 (9.4%) 7 (4.0%) 10 (3.1%)   NA 0 2 (1.1%) 8 (2.5%)  PD phenotype (%)      Mixed 22 (69%) 101 (57%) 192 (59.1%) 0.29  Akinetic rigid 3 (9.4%) 34 (19.2%) 68 (21.0%)   Tremor dominant 6 (19%) 31 (17.5%) 62 (19.1%)   NA 1 (3.1%) 11 (6.2%) 2 (0.6%)  Hoehn and Yahr (SD)      On (SD, n) 2.5 (0.9, 20) 2.1 (0.8, 70) 1.9 (1.0, 126) 0.21  Off (SD, n) 1.6 (0.9, 8) 2.3 (1.1, 84) 2.3 (0.9, 123)  UPDRS-III score (SD)      On (SD, n) 34.0 (21.1, 20) 36.7 (16.4, 70) 33.3 (19.5, 128) 0.45  Off (SD, n) 45.2 (19.8, 8) 49.0 (21.9, 84) 47.4 (17.2, 123)     62  Table 8.  Parkinsonism in LRRK2 p.G2019S carriers by gender   LRRK2 p.G2019S iPD   Female Male Female Male N 103 106 142 182 Mean age of onset (SD) 60.0 (12.0) 57.8 (12.0) 56.4 (13.4) 56.2 (14.1) Median age of onset (IQR)  56 (40-72) 60 (43.5-76.5) 58 (39.5-76.5) 58 (38-78) First Symptom       Tremor 73 (70.9%) 85 (80.1%) 122 (85.9%) 143 (78.6%)   Gait or balance deterioration 14 (13.6%) 13 (12.3%) 8 (5.6%) 18 (9.9%)   Muscle cramping or dystonia 7 (6.8%) 0 4 (2.8%) 3 (1.6%)   Shoulder stiffness 1 (0.97%) 4 (3.8%) 2 (1.4%) 6 (3.3%)   Other 8 (0.78%) 2 (1.9%) 2 (1.4%) 8 (4.4%)   NA 0 2 (1.9%) 4 (2.8%) 4 (2.2%) PD phenotype       Akinetic rigid 20 (19.4%) 17 (16.0%) 28 (19.7%) 40 (22.0%)   Mixed  58 (56.3%) 65 (61.3%) 84 (59.2%) 108 (59.3%)   Tremor dominant 17 (16.5%) 20 (18.9%) 29 (20.4%) 33 (18.1%)   NA 8 (7.8%) 4 (3.8%) 1 (0.70%) 1 (0.54%) Hoehn &Yahr (SD)       On 2.3 (0.93) (n=48) 2.0 (0.73) (n=42) 1.9 (0.89) (n=58) 2.0 (1.1) (n=68)   Off 2.3 (1.2) (n=40) 2.2 (1.1) (n=52) 2.5 (0.97) (n=51) 2.1 (0.80) (n=72) UPDRS-III score (SD)       On 39.2 (19.4) (n=48) 32.7 (14.4) (n=43) 32.4 (18.2) (n=60) 34.1 (20.7) (n=68)   Off 50.2(21.2) (n=40) 47.5(21.6) (n=52) 51.6(16.9) (n=51) 44.4(16.9) (n=72) NA = not available. Other = change in facial expression, change in speech or voice, decreased dexterity, stooped posture or bradykinesia.   63  Table 9. UPDRS Part IA Mentation, Behaviour and Mood  LRRK2 p.G2019S iPD p-value Cognitive impairment 0.28 (0.67) 0.33 (0.74) 0.66 Hallucinations and psychosis  0.14 (0.45) 0.23 (0.57) 0.35 Depressed mood 1.28 (1.00) 1.40 (1.03) 0.08 Anxious mood 1.06 (1.11) 1.02 (1.07) 0.32 Apathy 1.22 (1.03) 1.19 (1.01) 0.53 Features of dopamine dysregulation syndrome 0.05 (0.25) 0.06 (0.35) 0.95 Quantitative scales are from 0-4: 0 (normal) – 4 (most severe). Mean values (standard deviation) are given.    64   Table 10. UPDRS Part IB Mentation, Behaviour and Mood  LRRK2 p.G2019S iPD p-value Sleep problems 0.96 (1.15) 0.87 (1.17) 0.30 Daytime sleepiness 0.96 (0.96) 0.85 (0.92) 0.14 Pain and other sensations  1.53 (1.21) 1.45 (1.12) 0.75 Urinary problems 1.20 (1.27) 1.19 (1.20) 0.68 Constipation problems 1.15 (1.21) 1.20 (1.14) 0.95 Light headedness on standing 0.92 (1.10) 1.01 (1.06) 0.11 Fatigue 1.86 (1.09) 1.74 (0.97) 0.57 Quantitative scales are from 0-4: 0 (normal) - 4 (most severe). Mean values (standard deviation) are given.   65   Table 11. UPDRS Part II Activities of Daily Living  LRRK2 p.G2019S iPD p-value Speech 1.01 (0.98) 1.00 (0.96) 0.34 Saliva and drooling 0.97 (1.13) 1.19 (1.24) 0.02 Chewing  0.72 (0.90) 0.70 (0.87) 0.95 Eating tasks 1.18 (0.89) 1.04 (0.82) 0.24 Dressing 1.60 (1.08) 1.48 (1.05) 0.79 Hygiene 1.71 (1.13) 1.59 (1.14) 0.34 Handwriting 1.54 (1.19) 1.53 (1.17) 0.76 Doing hobbies and other activities 1.86 (1.30) 1.71 (1.11) 0.99 Turning in bed 1.56 (1.17) 1.44 (1.10) 0.80 Tremor 2.00 (1.00) 1.98 (1.00) 0.74 Getting out of bed, a car, or a deep chair 1.67 (1.15) 1.32 (1.05) 0.01 Walking and balance 1.66 (1.04) 1.37 (0.98) 0.10 Freezing 0.85 (1.13) 0.72 (1.08) 0.92 Quantitative scales are from 0-4: 0 (normal) - 4 (most severe). Mean values (standard deviation) are given.   66   Table 12. UPDRS Part III  Quantitative scales are from 0-4: 0 (normal) - 4 (most severe). Mean values (standard deviation) are given.    LRRK2 p.G2019S iPD P-value Mean (SD)    Part III. Total Rigidity Subscale 1.59 (0.71) 1.53 (0.76) 0.83 Part III. Total Bradykinesia Subscale 1.65 (0.96) 1.49 (0.98) 0.63 Part III. Total Tremor Subscale 0.76 (0.60) 0.65 (0.60) 0.31 Part III. Total Postural Instability and Gait Disorder Subscale 1.33 (0.79) 1.27 (0.85) 0.40 67   Table 13. UPDRS Part IV Complications of Therapy  LRRK2 p.G2019S iPD p-value Time spent with dyskinesias 0.47 (1.00) 0.30 (0.75) 0.83 Functional impact of dyskinesias 0.46 (1.02) 0.27 (0.80) 0.52 Time spent in the off state 1.34 (1.18) 1.14 (1.03) 0.34 Functional impact of fluctuations 1.51 (1.50) 1.34 (1.36) 0.62 Complexity of motor fluctuation 0.95 (0.97) 0.88 (0.91) 0.21 Painful off-state dystonia 0.30 (0.75) 0.23 (0.60) 0.42 Quantitative scales are from 0-4: 0 (normal) - 4 (most severe). Mean values (standard deviation) are given.   68   3.3.2. Non-motor features To assess the validity of the non-motor questionnaires we compared control subjects to iPD for SCOPA-AUT, MOCA, Epworth sleepiness scale, and olfactory assessments (Table 14-17). SCOPA-autonomic assessments can be clearly distinguished between control subjects and iPD (p<0.0001). Interestingly, there is a trend for affected LRRK2 carriers with less gastrointestinal dysfunction (mean score 0.64 in LRRK2 carriers compared to mean 0.74, p=0.04). For example, the score for constipation is 0.64 for LRRK2 carriers compared to 0.74 in idiopathic PD.  Cognitive assessments were done using the Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment (MOCA), six picture test and frontal assessment battery.  However, when looking into control subjects vs iPD these scores were comparable in the Tunisian Arab Berber population. This suggests that cognitive assessments using these scales may not be appropriate, and are not sensitive enough for measuring differences between patients and control subjects (Table 16).  Of interest, LRRK2 carriers displayed less REM sleep behavior disorder compared to iPD (16% for LRRK2 p.G2019S carriers compared to 29%, p<0.0001) (Table 17). Other sleep disorders: Epworth sleepiness score, sleep apnea and restless legs syndrome were indistinguishable between LRRK2 carriers and iPD.    69  Table 14. Autonomic dysfunction (SCOPA-Aut) individual scores   Control subjects iPD p-value LRRK2 p.G2019S iPD p-value Swallowing/choking  0.20 (0.45) 0.45 (0.65)   <0.0001 0.41 (0.63) 0.45 (0.65) 0.09 Sialorrea  0.13 (0.36) 0.84 (0.93) <0.0001 0.72 (0.90) 0.84 (0.93) 0.10 Dysphagia  0.14 (0.39) 0.47 (0.68) <0.0001 0.47 (0.57) 0.47 (0.68) 0.09 Early abdominal fullness  0.33 (0.74) 0.82 (1.00) <0.0001 0.78 (0.91) 0.82 (1.00) 0.09 Constipation  0.48 (0.73) 1.40 (1.2) <0.0001 1.12 (1.09) 1.40 (1.2) 0.09 Straining for defecation  0.41 (0.78) 1.10 (1.2) <0.0001 0.90 (1.1) 1.10 (1.2) 0.09 Faecal incontinence  0.0078 (0.09) 0.08 (0.32) 0.0035 0.08 (0.88) 0.08 (0.32) 0.09 Straining for urination  0.33 (0.73) 0.90 (1.02) <0.0001 0.75 (0.87) 0.90 (1.02) 0.43 Urinary incontinence  0.20 (0.48) 0.65 (0.85) <0.0001 0.68 (0.92) 0.65 (0.85) 0.47 Incomplete emptying 0.23 (0.62) 0.78 (1.04) <0.0001 0.70 (0.97) 0.78 (1.04) 0.44 weak stream of urine  0.25 (0.65) 0.70 (1.00) <0.0001 0.68 (0.95) 0.70 (1.00) 0.45 frequency of urine passing  0.33 (0.70) 0.98 (1.13) <0.0001 0.86 (1.1) 0.98 (1.13) 0.45 Nocturia  0.63 (0.88) 1.25 (1.13) <0.0001 1.11 (1.14) 1.25 (1.13) 0.45 Light headed when standing up  0.48 (0.75) 0.85 (0.90) 0.0001 0.71 (0.87) 0.85 (0.90) 0.79 Light headed standing some time  0.30 (0.61) 0.63 (0.82) <0.0001 0.61 (0.81) 0.63 (0.82) 0.81 Syncope  0.09 (0.28) 0.14 (0.40) 0.09 0.16 (0.45) 0.14 (0.40) 0.81 Hyperhidrosis during day  0.53 (0.88) 1.06 (1.08) <0.0001 0.98 (1.1) 1.06 (1.08) 0.10 Hyperhidrosis during night  0.46 (0.81) 0.90 (1.03) <0.0001 0.95 (1.1) 0.90 (1.03) 0.18 Oversensitive to bright light  0.24 (0.67) 0.50 (0.85) <0.0001 0.55 (0.90) 0.50 (0.85) 0.56 Cold tolerance  0.49 (0.92) 0.67 (0.96) 0.0302 0.73 (0.92) 0.67 (0.96) 0.82 Heat tolerance   0.58 (0.92) 0.87 (1.00) 0.0013 0.93 (0.99) 0.87 (1.00) 0.34 Men: erection problem  - 0.51 (1.01)  0.53 (1.02) 0.51 (1.01) 0.97 Men: ejaculation problem  - 0.41 (0.91)  0.43 (0.94) 0.41 (0.91) 0.95 Medication for erection disorder (%) - 2 (0.8)  3 (1.8) 2 (0.8) 0.30 Women: vaginal lubrication  - 1.50  (0.53)  1.4 (0.51) 1.50  (0.53) 0.38 Women: orgasm  - 1.58 (0.67)  1.4 (0.51) 1.58 (0.67) 0.03 70   Control subjects iPD p-value LRRK2 p.G2019S iPD p-value Constipation medications (%) - 37 (16%)  23 (14%) 37 (16%) 0.89 Urinary medications (%) - 10 (4.3%)  4 (2.5%) 10 (4.3%) 0.39 PD medications (%) - 40 (17%)  34 (21%) 40 (17%) 0.17 L-Dopa ON State (%) - 119 (51%)  76 (47%) 119 (51%) 0.46 Scale: 0-4. 0=never, 1=sometimes, 2=regularly, 3=often, 4=use catheter. Mean values (standard deviation or %) are given.    71  Table 15.  Summary of autonomic assessments compared between LRRK2 parkinsonism and iPD  LRRK2 p.G2019S iPD p-value     Gastrointestinal (SD) 0.64 (0.51) 0.74 (0.52) 0.04 Urinary (SD) 0.80 (0.80) 0.87 (0.83) 0.10 Cardiovascular (SD) 0.50 (0.59) 0.54 (0.54) 0.35 Thermoregulatory (SD) 0.89 (0.76) 0.87 (0.73) 0.92 SCOPA-Aut subdomain scores were compared between patients. Quantitative scales are from from 0-4: 0 (normal) - 4 (most severe). SD: Standard deviation     72  Table 16. Summary of cognitive assessment compared between iPD, LRRK2 parkinsonism and control subjects   Control subjects iPD p-value LRRK2 p.G2019S iPD p-value MMSE (SD) 27.1 (3.29) 25.4 (3.9) 0.6898 25.7 (3.6) 25.4 (3.9) 0.42 FAB (SD) 12.7 (4.60) 10.6 (4.5) 0.2728 10.8 (4.6) 10.6 (4.5) 0.27 MOCA (SD) 17.10 (10.8) 19.3 (8.5) 0.3550 21.7 (6.8) 19.3 (8.5) 0.03 Mini Mental State Examination (MMSE), Frontal Assessment Battery (FAB), Montreal Cognitive Assessment (MoCA).     73  Table 17. Comparison of sleep scales among LRRK2 parkinsonism and iPD.  Only a subset of patients had levodopa state recorded .     LRRK2 p.G2019S iPD p-value Epworth total score      On state (SD) 5.35 (4.67) 5.25 (4.90) 0.98   Off state (SD) 4.75 (4.48) 4.88 (5.19)  Restless legs      On state (%) 9/75 (12%) 15/118 (13%) 0.75   Off state (%) 6/81 (7.4%) 11/113 (9.7%)  REM sleep disorder      On state (%) 12/75 (16%) 34/118 (29%) 0.001   Off state (%) 14/81 (17%) 40/113 (35%)  Sleep apnea      On state (%) 7/75 (9.3%) 10/118 (8.5%) 0.65   Off state (%) 9/81 (11%) 15/113 (13%)  74  3.3.3. Disease progression   The rate of disease progression was determined by taking motor and autonomic scores over the disease duration. Age at onset was highly correlated with motor and non-motor progression scores in idiopathic PD (R=0.20-0.31, p<0.0001). However, LRRK2 parkinsonism was more uniform in progression (Table 18).   Table 18. Rate of disease progression associated with age at onset in patients   iPD LRRK2 p.G2019S  Correlation to age at onset Correlation to age at onset  R  p-value  R p-value Hoehn and Yahr progression 0.29 <0.0001* 0.00 0.99 GI progression 0.30 <0.0001* 0.09 0.27 Urinary progression 0.22 0.0020* 0.05 0.53 Cardiovascular progression 0.20 0.0054* 0.06 0.51 Thermoregulatory progression 0.24 0.0006* 0.08 0.33 Rigidity progression 0.30 <0.0001* 0.02 0.85 Bradykinesia progression 0.29 <0.0001* 0.07 0.45 Tremor progression 0.22 0.0017* 0.07 0.19 Postural instability and gait disorder progression 0.31 <0.0001* 0.05 0.99 R =pearson’s correlation coefficient to age at onset.  Disease progression is measured by severity score/disease duration. *=significant after Bonferroni    75   3.4. Discussion The main motor feature of LRRK2 p.G2019S parkinsonism in 220 sporadic patients and 126 familial patients was tremor-predominant parkinsonism with bradykinesia and rigidity that responds to dopamine replacement therapy. Patients with LRRK2 p.G2019S parkinsonism are generally indistinguishable from patients with iPD cross-sectionally, however, our data suggests temporal distinction and trending differences in non-motor features. Earlier studies highlighted tremor as the predominant feature of LRRK2 carriers which is supported by more recent meta-analysis (‘dardarin’ the Basque word for tremor remains a colloquial term for LRRK2 protein) (Aasly, et al., 2005; Healy, et al., 2008; Paisan-Ruiz, Lang, et al., 2005; Paisan-Ruiz, Saenz, et al., 2005) . In this study, tremor was observed less in 220 Arab-Berber patients with LRRK2 p.G2019S than in iPD.   Non-motor features occurred at similar frequencies in LRRK2 p.G2019S patients and in iPD. However, affected LRRK2 carriers have less REM sleep disorder and gastrointestinal dysfunction. Less REM sleep disorder and olfactory impairment  has also been seen in Ashkenazi Jewish LRRK2 carriers (Saunders-Pullman et al., 2015; Saunders-Pullman et al., 2014) . Patients with iPD have Lewy body disease in the periphery, most notably the dorsal motor nucleus, vagal nerve, cardiac sympathetic and enteric nervous systems, as well as in the olfactory bulb, brainstem (midbrain, pons, medulla) and cortex (Braak, et al., 2003). Nonmotor features of cognitive impairment and hypotension has been correlated with presence of Lewy bodies (Kalia et al., 2015) . While most LRRK2 p.G2019S patients have similar Lewy body disease some develop alternative 4R-tauopathy or TDP43 proteinopathy (Marras, et al., 2016) . Hence, we speculate marginal differences in REM sleep and gastrointestinal function in LRRK2 p.G2019S carriers may reflect less concomitant alpha-synucleinopathy. Less REM sleep 76  behavioural disorder and peripheral dysfunction can also imply that the background of LRRK2 p.G2019S does not follow Braak staging (Goedert, et al., 2013a) .  The disease penetrance of LRRK2 p.G2019S ranges from 24% to 100% (Marder, et al., 2015; Trinh, Amouri, et al., 2014) (Latourelle et al., 2008)(Latourelle et al., 2008)(Latourelle et al., 2008)(Latourelle, Sun, et al.)(Latourelle et al., 2008)(Latourelle et al., 2008)(Latourelle et al., 2008)(Latourelle et al., 2008)(Latourelle et al., 2008) (Latourelle et al., 2008) , the accuracy of figures is disputed and some of the disparity may reflect differences in population ethnicity, bias in patient recruitment, and differences in statistical analysis. Accurate penetrance estimates of LRRK2 p.G2019S in Arab-Berbers are important given the highest frequency of LRRK2 p.G2019S carriers and the prevalence of parkinsonism in this region of the world (Hulihan, et al., 2008; Lesage, Anheim, Letournel, Bousset, Honore, Rozas, Pieri, Madiona, Durr, Melki, Verny, & Brice, 2013) . The population of Tunisia may also offer greater ethnic, genetic and environmental homogeneity than prior North American, European and Israeli studies. The range of age at onset in LRRK2 p.G2019S carriers is broad spanning 50 years. Ascertainment bias appears an unlikely explanation as the kin-cohort analysis of LRRK2 p.G2019S pedigrees supported the Kaplan Meier findings, which means the familial penetrance estimates are comparable to the unrelated LRRK2 p.G2019S carriers. If there were an ascertainment bias, we would expect higher penetrance estimates in LRRK2 families. Hence, penetrance modifiers that modulate motor symptom onset in LRRK2 p.G2019S carriers appear likely but remain to be defined.  Hoehn & Yahr scores are a composite measure of dysfunction encompassing activities of daily living, motor and cognitive disability. In cross-sectional analysis LRRK2 parkinsonism and iPD cannot be distinguished; the range and distributions of component symptoms and scores 77  overlap. In iPD, mild progressors have an earlier onset age and faster progressors have a later onset age. Patients with LRRK2 parkinsonism  appear to have the same rate of progression regardless of onset age. Age of onset has always been a reasonable predictor of disease progression and morbidity in PD (Diamond, et al., 1989). Interestingly, onset age does not predict progression for LRRK2 parkinsonism. Nevertheless, a more uniform rate of progression in LRRK2 p.G2019S carriers may aid biomarker discovery and clinical trials focused on disease-modification (neuroprotection).  However, this might reflect a sample size and lack of test sensitivity effect in the LRRK2 patient group. Our objective was to compare the clinical features of iPD and LRRK2 parkinsonism and estimate the risk in carriers as an aid for genetic counselling. Kaplan-Meier and kin-cohort methods were used to estimate the risk of parkinsonism in sporadic and familial LRRK2 carriers. Clinic-based and volunteer patient proband series may lead to an overestimate of the penetrance of LRRK2 p.G2019S. However, the kin-cohort method, which does not take the proband into consideration, gave similar results to Kaplan-Meier analyses.  A weakness of our study is that samples were only drawn from Tunisia; while LRRK2 p.G2019S carriers generally inherit the same ancestral haplotype (Kachergus, et al., 2005)  our penetrance findings may not be universally applicable and comparative clinical and genetic studies in different ethnic backgrounds are needed.  Presently, carrier status of this pathogenic mutation does not influence a patient’s choice of treatment, although the discovery of LRRK2 biomarkers and specific molecular interventions are actively sought (Cookson, 2010; R. J. Nichols et al., 2009). The range and severity of motor and non-motor features in idiopathic PD and LRRK2 parkinsonism are comparable which suggests therapies for LRRK2 parkinsonism might be readily generalizable to iPD. Future 78  studies might include comparative whole genome sequencing (WGS) of LRRK2 patients with divergent ages of onset in an attempt to find novel genetic variants that modulate phenoconversion, to symptoms that warrant diagnosis and therapeutic interventions. WGS in large multi-incident pedigrees with the p.G2019S mutation would be a good start in identifying novel genetic modifiers as a rare variant segregating with age at onset in a family tree can be a good indication of  a modifier. Another possibility can be common polymorphisms influencing age at onset, which can be identified in larger case cohorts. Similarly, environmental exposures that influence risk of parkinsonism might be more readily identified in a relatively more homogeneous patient sample.  79  4. Chapter 4: Dynamin 3 modifies age at onset in LRRK2 parkinsonism  4.1. Introduction Genetic variability in leucine-rich kinase 2 (LRRK2) has been linked to familial parkinsonism and associated with idiopathic Parkinson disease (PD): LRRK2 c.6055G>A (p.G2019S) confers the highest genotypic and population attributable risk (Kachergus, et al., 2005; Ross, et al., 2011; Zimprich, et al., 2011) . Penetrance estimates are variable with a wide range in age of onset (AOO) influenced by ethnicity (Healy, et al., 2008; Hentati et al., 2014; Trojano, Moretta, Estraneo, & Santoro, 2010). The relatively homogeneous North-African Arab-Berber population has the highest frequency of LRRK2 p.G2019S carriers, between 30-40% of patients with PD (Lesage, et al., 2005; Trinh, Amouri, et al., 2014) and provides a unique opportunity to identify genetic modifiers of AOO.  LRRK2 is a large multi-domain protein with GTPase (Roc) and kinase activities that appear to modulate cytoskeletal outgrowth and vesicular dynamics, including synaptic transmission, endosomal trafficking and lysosomal autophagy (Orenstein et al., 2013) . Although many binding partners and substrates have been identified, it remains uncertain which are clinically relevant to disease pathophysiology. Herein, a genome-wide approach was used to identify genetic variability that directly influences LRRK2 p.G2019S penetrance.   4.2. Methods 4.2.1. Discovery cohort and replication series Arab-Berber subjects were recruited between 2006 to 2012 by movement disorders neurologists (FH, SBS, FN, EF) at the Mongi Ben Hamida National Institute of Neurology, Tunis. Community-based samples consisted of 41 multi-incident LRRK2 p.G2019S families 80  (150 affected and 103 unaffected LRRK2 carriers), and 232 unrelated LRRK2 p.G2019S carriers (Table 19). All subjects were older than ≥18 years at neurological assessment and provided informed consent prior to their participation. Specific approvals obtained from the local ethics committee at the National Institute and Ministry of Health in Tunis were reviewed by GlaxoSmithKline (GSK), the Institutional Review Board of Mayo Foundation and the Research Ethics Board of the University of British Columbia. Additional replication cohorts included 263 LRRK2 p.G2019S carriers from Algeria (MT), France (AB), Norway (JAA) and North America (PSG–Progeni GenePD Investigators (Latourelle et al., 2011)  (Table 20). Human biological samples were sourced ethically and their research use was in accord with the terms of the informed consents. An overview of the discovery and replication samples is depicted in Figure 31.  4.2.2. Linkage analysis and STR genotyping  Genome wide linkage analysis was performed on 41 LRRK2 p.G2019S families from Tunisia using deCODE’s 4cM density STR (short tandem repeat) marker set, with standard approaches (Abecasis, Cherny, Cookson, & Cardon, 2002) . Allele frequencies derived from Tunisian unrelated, non-carrier, control subjects were used for STRs. Consanguineous loops are noted in ~1/3rd of the families but were split to maximize information content (Abecasis, et al., 2002). Both non-parametric (NPL) and model-based linkage analyses were performed considering early-onset and late-onset groups, dichotomized by median AOO: < or ≥ 56 years. Linkage was performed with merlin. We had two categories: an early PD onset group (all affected carriers with an age at onset <56 years) and a late PD onset group (all affected carriers with age at onset ≥56 years and all unaffected carriers with an age at examination of ≥56 years). Alternatively, AOO in patients and age at recruitment/examination of unaffected carriers were 81  assessed as a quantitative trait. Model-based linkage used an additive model with incomplete penetrance to provide LOD (logarithm of odds) and hLOD (heterogeneity LOD) scores.  4.2.3. Genome-wide SNP genotyping and association  Single nucleotide polymorphisms (SNP) were genotyped for the Tunisian Arab-Berber cohort using Affymetrix 500K NspI and StyI (n=101) and Illumina Multi-Ethnic Genome Arrays (MEGA) (n=131). Affymetrix genotypes were extracted from .cel intensity files using three algorithms, BBRML, JAPL and CHIAMO, and only nominated when there was consensus as previous (Trinh, Gustavsson, et al., 2014) ; GenomeStudio® was used to provide genotypes for Ilumina data. Samples with genotype call rate below 99% were excluded from further analysis. Genotype distributions for all SNPs within control subjects, and all cases combined, satisfied Hardy-Weinberg equilibrium (HWE) expectations (p>0.001) . PLINK was used to assess IBS, IBD and population stratification as quality measures for the MEGA and Affymetrix data (Purcell et al., 2007). Extraction of the DNM3 locus region was performed with PLINK on MEGA and Affymetrix merged datasets. Quality control of MEGA and Affymetrix data was performed. A subset of consistent genotypes/individuals was assessed for population stratification using Eigenstrat, as previously described. Prior to case-control association, genome-wide IBS/IBD (identity by state/identify by descent) estimates were used to identify and exclude sample contamination, duplicates and individuals with unknown relationship (e.g. sibling-pairs in the unrelated case-control series). We assessed IBS and IBD in detail since the unrelated carriers share the same G2019S haplotype. Within regions of linkage, PLINK association analyses were performed (Howey & Cordell, 2014; Purcell, et al., 2007)  . Quantile-quantile plots of p-values were employed to highlight potential confounders (R package qqman).  82   4.2.4. Whole genome sequencing and imputation   Whole genome sequencing (WGS) was accomplished for 14 Tunisian Arab-Berber patients. All are LRRK2 p.G2019S carriers with a family history of parkinsonism, half had early-onset disease (mean onset 34.9 years, SD±7.2, range 22-42) and the remainder are clinically asymptomatic elderly carriers (mean age 77 years, SD±6.9, range 68-90). Sequencing was carried out using Illumina 2x100 nucleotide paired-end reads, with minimum 50-fold mean depth using standard methods for sequence alignment and variant calling (Figure 24).  SNP genotypes in the chromosome 1q23.3-24.3 region of linkage were imputed with Beagle 4.0, (Browning & Browning, 2008, 2009)  employing 14 Tunisian WGS and phased 1000 Genomes data as a reference for MEGA and Affymetrix data (n=232 LRRK2 carriers). Subsequently, haplotype associations were assessed within the linked interval using a variable-length Markov-chain Monte Carlo method (Browning & Browning, 2008, 2009). Affymetrix and MEGA genotype calls were previously merged together. PLINK files were then converted into VCF files with PLINK/SEQ and Beagle 4.0 was used for imputation. The genotype file (gt) was designated as the merged file and 1000 Genomes data was used as the reference VCF file (ref). There were 15075 reference markers, 1595 target markers and 2504 reference samples. Burn-in, phase and imputation iterations were set at 10, to maximize genotype imputation accuracy. The haplotype association was performed using Beagle 3.3 on affected unrelated individuals. Phasing iterations and then haplotype association was performed on allelic, recessive, over-dominant and dominant models. Corrected p-values for haplotype association and multiple-testing were estimated by permutation analyses, randomizing case-control status. The beagle haplotype association p-value was significant after permutation analyses, (p=0.002). 83   4.2.5. Sequencing and genotyping All subjects were screened for LRRK2 p.G2019S by Sanger sequencing or TaqMan SNP assays-on-demand (Life Technologies, Inc, Foster City, CA), and excluded for other pathogenic mutations implicated in PD (Gustavsson, Trinh, et al., 2015; Ishihara-Paul, et al., 2008). Subsequent genotyping was carried out by a combination of Sequenom MassArray iPLEX system (Sequenom, San Diego, CA) and TaqMan genotyping. Cumulative incidence plots (Kaplan Meier) and hazard ratios (Cox proportional hazard regression models) were used to stratify age of initial symptom by genotypes using JMP® software (SAS Institute Inc., Cary, NC). These models were adjusted for family relatedness, gender and population series (Tunisia, Algeria, France, Norway, and North America). Right censoring for asymptomatic carriers was performed at age of examination. Meta-analyses of all populations was performed with R-package ‘metafor’.  4.2.6. Brains, RNA, ampliseq transcriptome, antibodies  Brain tissue from 61 healthy control subjects without any neurological symptoms was obtained from the Oxford Brain Bank, University of Oxford (LP); any with neurodegenerative vascular pathology were excluded (LP). Full ethical approval (REC 07/Q2707/98) and written informed consent are obtained for all participants. Gender, age-at-death and post mortem delay was available for all subjects (Table 21). DNA was prepared from ~20mg of frozen tissue samples of striatum with an Autogen NA1000 and quantified using standard methods. Prism 6.0 (GraphPad Software, Inc) was used for RNA/protein analysis. Total RNA was also extracted (RNeasy Qiagen Minikit) from duplicate samples, and DNase I digested prior to assessing the 84  concentration, quality and integrity (RIN)  with an Agilent 2100 bioanalyzer, RNA 6000 LabChip kit and associated software (Agilent). After extraction, RNA integrity numbers for samples was good quality (mean RIN = 8.9). Out of the 61 human control striatum, 38 have a RIN > 7.0 and were used for TaqMan expression. A subset of the striatum was used for AmpliseqTM whole human transcriptome analysis (n=17) was performed with an Ion Proton (Life Technologies, Inc).  High-quality, total RNA was reverse transcribed and amplified using TaqMan One Step RT-PCR kit following manufacture’s protocol (ABI). Sequencing analysis resulted in an average of over 12 million reads per sample and a read length of 114 bases. AmpliseqRNA was used to map reads and generate absolute/normalized gene expression values (reads per million, RPM).  RNA expression analyses were adjusted by RIN quality.  Expression levels were quantified by dividing 2-Ct by the geometric mean of the expression levels of three commonly used “housekeeping” genes: hypoxanthine phosphoribosyl-transferase (HPRT; Hs02800695_m1), glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Hs02758991_g1) and synaptophysin (SYP; Hs00300531_m1)). DNM3 expression was measured using Taqman probe expression assay ID Hs00399015_m1 (all transcripts) and Hs00927940_m1 (NM_001136127.2 and NM_015569.4 only). Likewise, Ampliseq Transcriptome data was normalized with a variety of housekeeping genes (GAPDH, HPRT, SYP, YWHAZ), including those primarily expressed in neurons (TH, MAP2, ENO2, SV2A, SV2B, SYN1, SYN2) and the expression findings were robust. For protein analysis, 20mg brain tissue (n=17) was lysed with buffer containing 1% NP-40, 20mM HEPES, 125mM NaCl, 50mM NaF and protease inhibitor cocktail (Roche). The lysates were put on ice for 1 hour. Blotting of dynamin-3 was done with a polyclonal rabbit antibody (Synaptic Systems [115 302], 1:1000) and anti-for GAPDH a mouse monoclonal antibody was used (Thermo Scientific [MA5-15738], 1:1000). MAP2 antibodies were used in 85  immunofluorescence (Abcam, 1:1000).LRRK2 p.G2019S mice, primary neuronal littermate cultures, immunostaining and image analysis were as previously described (Beccano-Kelly et al., 2014)  .   4.3. Results 4.3.1. Linkage and association of LRRK2 p.G2019S families Linkage analysis of AOO in 41 Tunisian LRRK2 p.G2019S pedigrees identified chromosome 12q12 using non-parametric (LOD NPL = 3.3,  θ=0 at D12S85) and model-based methods (maximum LOD = 7.6, θ =0 at D12S85 under a dominant model of inheritance), which encompasses the LRRK2 locus. Genome-wide analysis using similar approaches, with allele-dependent penetrances, also identified chromosome 1q23.3-24.3 (LOD NPL =2.90, maximum LOD & hLOD = 4.99, θ =0 at D1S2768 with a recessive model, and LOD = 2.81 and hLOD= 3.81, θ =0 at D1S2768 with a dominant-additive model) (Figure 22). Significant linkage was obtained using AOO as a dichotomous trait and was robust to subsequent ordered subset analyses over a range of divisions (Hauser et al., 2004)  and implications of significant familial heterogeneity.  The highest LOD score across all models was obtained on chromosome 1. However, there was suggestive linkage on chromosome 6, 17 and 21 (Figure 26).  Evidence for association within the chromosome 1q23.3-24.3 linkage region (170.8-172.5Mb, the maximum LOD -1 support interval) was assessed in unrelated LRRK2 p.G2019S carriers (n=232). Only affected individuals were included in this association, the unaffected carriers were excluded. Association with dichotomized AOO revealed three associated SNPs (rs742510, rs2421947 and rs2206543, r2= 0·98; pnominal=2·6 x 10-5,) within the dynamin 3 locus (DNM3)(Table 23). Within the chromosome 1q23.3-24.3 linkage region, 634 SNPs were 86  assessed. A Bonferroni correction was applied to account for multiple testing (corrected p=0.016).  A QQ-plot for association analyses on chromosome 1 deviated from the line of equality but the DNM3 rs2421947 association was confirmed by TaqMan probe genotyping (Figure 27).  Subsequently, all 21 coding exons of DNM3 gene were sequenced in LRRK2 p.G2019S carriers with divergent AOO (n=25) and three rare (MAF<0.01) synonymous variants were identified (p.A81A, p.H128H and p.V609V). Carriers of divergent AOO refer to LRRK2 p.G2019S carriers who have early onset PD (<45 onset year) or were elderly (>75years) without motor signs of PD.  4.3.2. Higher resolution mapping WGS of 14 LRRK2 p.G2019S Tunisian Arab-Berber subjects and 1000 Genomes data provided references for SNP imputation, to improve haplotype analysis and identify specific variability associated with AOO. Within and flanking the DNM3 locus (chr1:171,810,018-172,382,057) a dense framework of informative markers (MAF>0.05) was imputed in all unrelated LRRK2 carriers (n=232) with Affymetrix, MEGA and Sequenom iPLEX genotypes. The shortest, most significant haplotype associated with AOO was subsequently defined between chr1:171,832,491-171,833,094 (rs77565020 to rs2421947, 603bp), using variable-length Markov-chain Monte Carlo methods (p=1.07 x 10-7, Text S1). Within the disease-associated haplotype allelic association with rs2421947 was most significant (p=1.07 x 10-7) (Figure 22, Table 24).  The Kaplan-Meier method was used to calculate median/IQR censoring at age of last examination for unaffected carriers by DNM3 genotype. rs2421947 CC homozygous carriers had 87  a median AOO of 64 years (IQR: 48-67); CG heterozygotes had a median 57 (IQR: 50.5-64 years), and GG homozygotes had a median 51.5 (IQR: 46-61.5 years) (Kaplan Meier log-rank p-value=0.03) (Figure 23). The median onset of LRRK2 parkinsonism in DNM3 rs2421947 GG homozygotes is 12.5 years younger than CC homozygotes. DNM3 rs2421947 has a minor allele frequency (MAF) C= 0·42 in unrelated control subjects from Caucasian populations (HapMap-CEU, n=226), and 0·42 in unrelated control subjects from Tunisia (n=321). In LRRK2 p.G2019S carriers the MAF C=0·39 overall, irrespective of affection status, but increases to C=0·46 with disease onset ≥56 years. To fully estimate the effect of DNM3 rs2421947 in the Tunisian population, we combined the unrelated individuals and families in a Cox proportional hazard model censoring unaffected individuals while adjusting for family relatedness and gender (HR 1.63, CI=1.05-2.63, p=0.03 for alternate homozygous genotypes).  4.3.3. DNM3 expression in brain  DNM3 rs2421947 was genotyped in striatal brain tissue (n=61) to assess any influence on expression. The rs2421947 GG genotype was correlated with higher DNM3 mRNA levels (r=0.25, p=0.006) (Figure 28), 1.25-fold higher for the GG genotype compared to CC. Results were confirmed using Ampliseq whole transcriptome analysis in a subset of samples (n=17; transcriptome data available on request). The findings were robust to normalization with a variety of housekeeping genes. DNM3 total transcript expression was correlated with LRRK2 expression (r2=0.65 p=0.004)(Table 25). Dynamin-3 protein levels in striatum stratified by rs2421947 genotype (n=17) are higher for the GG genotype (1.6 fold higher, p=0.08, Figure 29).   88  4.3.4. Replication cohorts The DNM3 association with AOO was examined in additional LRRK2 p.G2019S carriers including subjects originating from Algeria (n=46), France (n=65), Norway (n=64) and North America (n=88). DNM3 rs2421947 was imputed for the American series using 1000 Genomes as a reference , or was otherwise genotyped. Of note, the MAF for LRRK2 carriers in each population series is different. Cox proportional hazard ratios are provided for each population, censoring unaffected individuals, adjusting for family relatedness and gender as covariates, and combined within a meta-analysis also adjusting for population in the model (HR 1.46 CI=1.04-2.04, p=0.02 for alternate homozygous genotypes)  (Figure 23).   4.4. Discussion Unbiased genome-wide linkage analyses and locus–specific association, with replication of that association in an unrelated series, nominate DNM3 as a genetic modifier of AOO in LRRK2 p.G2019S parkinsonism. The frequency of LRRK2 p.G2019S carriers is higher in North Africa than in any other region reported to date (Kachergus, et al., 2005; Ross, et al., 2011). Hence a strength of our study is the large number of patients and family members with LRRK2 p.G2019S originating from the same population. Clinical exams applied longitudinally by the same team of movement disorder specialists ensure accurate diagnoses and consistent data reporting. Inclusion of unrelated, incident cases at one site also avoids potential selection biases in referrals from multiple centers. The Arab-Berber population of Tunisia provides ethnic, genetic and environmental homogeneity to increase power for discovery. However, there are also many study limitations. In general, AOO is broadly defined and subjective: its variance is large even within LRRK2 families although highly correlated with age of a motor diagnosis. 89  Nevertheless, the variance in AOO in families is less than the variance in unrelated LRRK2 p.G2019S carriers suggestive of penetrance modifiers . AOO is a fixed albeit temporal measure of disease pathophysiology. Hence, in our initial linkage and association analyses a dichotomized approach was used, using AOO about 56 years as a categorical variable. Key findings were assessed using Cox proportional hazards regression models censoring unaffected individuals, adjusting for family relatedness, gender and population series. It would be worthwhile to examine disease onset and progression in other ways. Longitudinal follow up of these families, additional patients and asymptomatic carriers is warranted. In general, AOO is broadly defined and subjective: its variance is large even within LRRK2 families in this study although highly correlated with age of a motor diagnosis. Nevertheless, the variance in AOO in families is less than the variance in unrelated LRRK2 p.G2019S carriers suggestive of penetrance modifiers (Table 19, median interquartile range). AOO is a fixed albeit temporal measure of disease pathophysiology. Hence, it would be worthwhile to assess onset and disease progression in other ways. Longitudinal follow up of these families, additional patients and asymptomatic carriers is warranted. Genome-wide linkage analysis to AOO was performed in large LRRK2 p.G2019S pedigrees employing informative STRs. The highest linkage peak identified is on chromosome 12 and explained by p.G2019S; however, there was no evidence of genetic variability in cis or trans within this region influencing AOO. In Tunisia, several heterozygous ‘married in’ relatives and homozygous carriers are observed in highly consanguineous multi-incident pedigrees, and within the families 150 (59%) of carriers are affected (Table 19). While the incidence of idiopathic PD is generally low (~2% at >65 years), and biologically carrier status and AOO may be independent, in our dataset this does not appear to be the case. Hence, we took a careful look 90  at both the cis/trans effects of the LRRK2 haplotype and association between AOO and common polymorphisms. Nevertheless, no significant effects were identified after a Bonferroni correction (data available on request). However, lack of evidence for association to AOO should not be considered evidence against. The second highest linkage peak is on chromosome 1q23.3-24.3 and remained robust when considering different models and allele frequencies. Other linkage peaks were also present on chromosome 6, 17 and 21; while none showed evidence for association further investigation is warranted in larger datasets. LRRK2 p.G2019S is a relatively rare, pathogenic mutation for disease. Thus our study was limited by the number of LRRK2 p.G2019S carriers available, in families and in population-based series of idiopathic PD. As a continuous trait the distribution of affected carriers was too sparse for AOO analysis; unaffected carriers were not included and there was insufficient information for linkage analysis. However, as a dichotomized trait, we were able to include unaffected carriers’ age greater than or equal to the median AOO. In addition, unaffected carriers younger than the median AOO for the pedigrees were marked as ‘unknown’ status in pedigree analyses and thus contribute their genotype information. Hence, the significance of the DNM3 finding may be driven by the inclusion of unaffected carriers older than the median AOO, not only affected carriers. Overall, rs2421947 appears to have an effect on AAO of LRRK2 p.G2019S parkinsonism. Nevertheless, confidence intervals are wide and span 1.0 for several replication series, albeit relative to sample size, and the effect appears to be in the opposite direction for the French series.  In addition, in replication series, a major caveat is that convenience samples suffer an intrinsic ascertainment bias – as they are from patients with PD of which a subset were found to be p.G2019S carriers.  Worldwide LRRK2 p.G2019S is generally inherited from the same ancestral haplotype (Kachergus, et al., 2005) but the influence of modifiers and their associated allele frequencies 91  may be population specific. There is suggestive linkage (LOD = 2.43) for AOO on chromosome 1q32.1 in predominantly North American LRRK2 p.G2019S families, albeit with no evidence for association in that region in those samples (Latourelle, et al., 2011) .  Nevertheless, genome-wide association analysis of idiopathic PD in Japan robustly implicates PARK16 within 1q32 (Satake et al., 2009), which is reproducibly observed albeit with low effect size (OR ~1.1) in a mega meta-analysis of Caucasian samples (Nalls, Pankratz, Lill, Do, Hernandez, Saad, DeStefano, Kara, Bras, Sharma, Schulte, Keller, Arepalli, Letson, Edsall, Stefansson, Liu, Pliner, Lee, Cheng, Ikram, et al., 2014) . PARK16 includes RAB29 (formerly RAB7L1) investigated as a candidate gene and associated with reduced risk of idiopathic and monogenic parkinsonism (LRRK2 p.G2019S and GBA p.N370S) in Ashkenazi  (Gan-Or et al., 2008) . Functional studies also support an interaction between RAB7L1 and LRRK2 (Beilina et al., 2014; D. A. MacLeod, et al., 2013) .  Nevertheless, chromosome 1 linkage results in this and the previous study appear independent. Patterns of linkage disequilibrium in Tunisian Arab-Berber and Israeli Jewish population samples are also different thus additional tagging SNPs may be required to evaluate DNM3 or other loci as penetrance modifiers. Non-synonymous variability in DNM3 was not observed in LRRK2 carriers which allowed us to focus on polymorphic non-coding eQTLs (expression quantitative trait loci). Variability in DNM3 expression correlates with genotype whether quantified by Ampliseq transcriptome or TaqMan methods. A specific isoform (Dyn3b) co-localizes with clathrin (Cao, Garcia, & McNiven, 1998) and appears more highly expressed in DNM3 rs2421947 GG homozygotes (Figure 28, Figure 29). In striatum DNM3 and LRRK2 expression are correlated suggesting they are involved in the same process. DNM3 rs2421947 does not appear to 92  contribute to risk of idiopathic PD, neither susceptibility nor AOO, but the influence of DNM3 rare variability has yet to be explored. LRRK2 has been implicated in neurite outgrowth (D. MacLeod et al., 2006; Parisiadou et al., 2009) , synaptic vesicle trafficking and neurotransmitter release , and via kinase-dependent mechanisms (Arranz et al., 2015) . Much of the underlying mechanistic biology in these processes remains enigmatic, as does their clinical relevance to PD. However, our genetic study shows DNM3 is an AOO modifier of LRRK2 p.G2019S parkinsonism. LRRK2 co-immunoprecipitates with the dynamin family GTPases that drive membrane fission (DNM1-3 and dynamin-related proteins).  LRRK2 co-immunoprecipitates with the dynamin family GTPases that drive membrane fission (DNM1-3, and dynamin-related proteins) (Stafa et al., 2014) . Amphiphysin recruits dynamin and endophilin A (a LRRK2 kinase substrate) , and recruits synaptojanin (SYNJ1) for endocytic vesicle fission (S. M. Ferguson & De Camilli, 2012) . Recessive mutations in SYNJ1 have been implicated in seizure disorders and early-onset parkinsonism (Krebs et al., 2013; Quadri, et al., 2013)  . In neurons, dynamin 3 localizes to the endocytic machinery of dendritic spines to modulate receptor recycling and excitatory synaptic transmission (Gray, Kruchten, Chen, & McNiven, 2005) . In this process ‘Dyn3b’ isoform expression is also centrally involved in the regulation of actin polymerization, filopodia and spine formation (Cao, et al., 1998; Gray, et al., 2005) . Intriguingly, a significant reduction and redistribution of dendritic dynamin 3 staining is observed in LRRK2 p.G2019S murine cortical cultures (Figure 30), although it may also reflect elevated glutamateric synaptic transmission (Beccano-Kelly, et al., 2014) . We postulated LRRK2 p.G2019S activates kinase activity (Kachergus, et al., 2005), an outcome of which has been the pursuit of competitive LRRK2 inhibitors. Based on similarly 93  unbiased genetic data, we postulate lower levels of DNM3, and perhaps specific dynamin 3 isoforms, will delay the onset of LRRK2 p.G2019S parkinsonism. The crystal structure of the dynamin tetramer has just been elucidated (Reubold et al., 2015)  and might accelerate the development of dynamin GTPase inhibitors (dynasores). These anticonvulsants repress synaptic transmission in seizure disorder (Li et al., 2015)  and delay alpha-synuclein uptake by neuronal and oligodendroglial cells (Konno et al., 2012) . At autopsy, most LRRK2 p.G2019S carriers have alpha-synucleinopathy and Lewy body disease (Ross et al., 2006) .  Thus DNM3 expression represents a target for neuroprotection in LRRK2 p.G2019S carriers, and potentially for disease-modification in LRRK2 parkinsonism.  94   Figure 22. Chromosome 1 linkage peak   a. (LOD score = 4.99). b. Region of association within the LOD -1 linkage interval: Plink SNPs (10-5) and Beagle haplotype (p=1.06 x 10-7) associations     95   Figure 23. Age-associated cumulative incidence of LRRK2 p.G2019S carriers. a. Replication cohorts: Algerian, French, Norwegian and North American, stratified by rs2421947 genotype (log rank p=0.0001). b. All populations combined (Algerian, French, Norwegian, North American and Tunisian Arab Berber) stratified by rs2421947 genotype (log rank p<0.0001).   B A 96  Table 19. Demographics of discovery cohorts: Tunisian Arab-Berber LRRK2 p.G2019S carriers   Unrelated patients Unrelated control subjects Familial patients Unaffected family members N 220  12  150 103 Number of men (%) 124 (56%) 6 (50%) 77 (51.3%) 48 (46.6%) Mean age (SD) years 67.6 (12.6) 56.7 (10.9) 68.6 (15.8) 56.1 (17.5) Median age (IQR) 69 (48-90) 54.5 (38-72) 70.5 (57-81) 53 (43-72.5) Mean age of onset (SD) 57.1 (11.6) - 56.1 (12.8) - Median age of onset (IQR) 57 (40-74) - 56 (47-65) -     97  Table 20 . Demographics of LRRK2 p.G2019S carriers: replication series    Norway France Algeria North America Total  Patient  Unaffected  Patient  Unaffected  Patient  Unaffected  Patient  Unaffected   N 19 45 48 17 45 1 88 - 263 Number of men (%) 8 (42%) 18 (40%) 26 (60%) 7 (39%) 19 (42%) - 41(47%) -  Mean age (SD) years 67.6 (17.5) 63.6 (12.4) 57.7 (13.8) 67.4 (11.8) 55.5 (11.3) 54 NA -  Median age (IQR) 73 (52-82) 62 (54.5-70) 59 (46.8-67.3) 67 (59.5-76.8) 55 (45.3-63) 54 NA -  Mean age of onset (SD) 62.6 (13.0) - 52.1 (13.5) - 49.6 (10.3) - 61.5 (10.1) -  Median age of onset (IQR) 65 (49-74) - 51 (41.3-62) - 50 (43-56) - 63 (56-70) -      98  Table 21. Demographics of healthy control brains for expression analysis   Control subjects N 61 Number of men (%) 30 (49.2%) Mean age at death (SD) years 80.6 (12.1) Median age at death (IQR) 85 (71-89) Tissue type Striatal Average RIN (RNA integrity number) (SD) 8.6 (1.2) Average PMI (Post-mortem interval) (SD) 48.8 (33.2)    99  Table 22. Primer pairs and custom TaqMan probe design for different DNM3 transcript isoforms in human striatum Names 5'->3' Primers Amino acid sequence Region 1   DNM3_Reg1F aaacggaaaggattgttgc  DNM3_Reg1Fprobe tctcttacatcaacaccaacc  DNM3_Reg1BR cccttgcgaatcacaatttg GTNLPPSRQI DNM3_Reg1AR ttgcgaatcacctgatttc  DNM3_Reg1C_F gcaaattgtacgagctaagttc VRAKFCKLYCCFFI DNM3_Reg1_R ttcaggttgtccaagggaag  Region 2   DNM3_Reg2A_F tatcctgacaaatctgtagctg SVAEN DNM3_Reg2_R ggtcctctgaagaatacaac  DNM3_Reg2B_F tctgtagggaacaacaaagc SVGNNKAEN DNM3_Reg2_R ggtcctctgaagaatacaac  Region 3   DNM3_Reg3_F aaaggaggccaacactaag SRRPPPSPTRPTIIRP DNM3_Reg3B_R attatagtgggacgagttgg  DNM3_Reg3_F aaaggaggccaacactaag RFGAMKDEAAEP DNM3_Reg3A_R cagcagcttcatccttcatgg   Probe TaqMan Probe design attggcttcgcaaatgctcagcagag  100  Table 23. PLINK association underneath linkage regions CHR SNP BP (hg19) A1 F_A F_U A2 CHISQ P OR 1 rs742510 171858930 A 0.5 0.1974 G 18.13 2.06 x 10-5** 4.067 1 rs2421947 171833094 C 0.5 0.1974 G 18.13 2.06 x 10-5** 4.067 1 rs2206543 171835493 G 0.5 0.1974 A 18.13 2.06 x 10-5** 4.067 ** Values significant after Bonferroni correction for all SNP association tests within the LOD-1 linkage interval on chr 1   101   Table 24. DNM3 haplotypes associated with AAO  rs77565020 rs75848807 rs192895361 rs74673993 rs142760983 rs559149705 rs185844670 rs541736672 rs563254497 rs530428455 rs190417579 rs74777828 rs192302781 rs146042960 rs566301333 rs376575981 rs183688167 rs114979811 rs56237038 rs72713714 rs2421947 count p-value Major haplotypes: G G G T A G G A A C A T A T G A G T A G G 238 1.07E-07** G G G T A G G A A C A T A T G A G T A G C 138 0.122 Minor haplotypes: G A G T A G G A A C A T A T G A G T A G C 1 NA G G G T A G G A A C A A A T G A G T A G C 5 NA G G G T A G G A A C A T A T G A G T A C C 3 NA G A G T A G G A A C A T A T G A G T A C C 1 NA G A G T A G G A A C A T A T G A G T A G G 2 0.515 G G G T A G G A A C A T A T G A G T A C G 1 0.411 ** Values significant after Bonferroni correction all haplotype associations within the LOD-1 linkage interval on chr 1    102  Table 25. DNM3 transcript levels correlate with LRRK2, VPS35 and SYNJ1 expression in striatal tissue transcriptome data from normal controls (n=17).  Gene DNM3 expression level correlation coefficient p-value Genes implicated in Late-onset autosomal dominant   LRRK2  0.65    0.004** VPS35 0.65 0.008 SNCA 0.65   0.04 DNAJC13 0.21   0.14 Genes implicated in Early-onset recessive    SYNJ1 0.41 0.008 PINK1 0.53   0.02 PARK2 0.21   0.04 FBXO7 0.51   0.12 **Values significant after Bonferroni correction   103  Table 26. Sensitivity analysis for different age cut-offs on chromosome 1q23.3-24.3 using non-parametric linkage Age at onset dichotomization   Chromosome 1q23.3-24.3  NPL LOD score 45 years 2.3 50 years 2.5 55 years 2.9 60 years 1.8 65 years 1.2                     104  Clinical characteristics of subjects with WGS 7 early onset LRRK2 p.G2019S carriers 7 asymptomatic LRRK2 p.G2019S carriers Mean Age of Onset (SD): 34.8 (7.2) Mean Age (SD): 77 (6.9) Mean Sequencing Depth 50X ↓ Align to NCBI hg19 Build38 ↓ Compare SNPs against dbSNP (91% of variants represented in dbSNP) ↓ Average number of SNP variants ~3,000,000 Average number of insertion variants ~325,750 Average number of deletion variants ~349,189 ↓ Imputation of chromosome 1 linkage region (DNM3 locus) ↓ Use imputed data for Beagle haplotype association   .Figure 24. Whole genome sequencing and imputation workflow   105    Figure 25. A schematic of the thirteen dynamin isoforms. Refer to table 22 for primer designs to capture different amino acid sequences. Figure adapted from (Cao, et al., 1998).106      012345670 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22A. Multipoint model-based linkage  LOD (blue) and HLOD (black) dominant model  024680 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22B. Multipoint model-based linkage LOD (blue) and HLOD (black) recessive model  0240 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22C. Multipoint non-parametric linkage (NPL) LOD (cumulative)   LOD score LOD score Chromosome Chromosome LOD score Chromosome 107    Figure 26. Multipoint model-based and non-parametric linkage analysis of Tunisian Arab-Berber LRRK2 p.G2019S families.   A. Parametric linkage with divergent ages at onset (<56 or ≥56 years) , using a dominant model with incomplete penetrance; B Parametric linkage hLOD cumulative scores; C. Non-parametric linkage D. Continuous trait analysis 0120 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24D. Continuous trait linkage analysis (NPL) LOD LOD score 108   Figure 27 Chromosome 1 Q-Q plot values   109  .     Figure 28. DNM3 transcript levels normalized by geometric mean of housekeeping genes   Total DNM3 RNA levelsDNM3 rs2421947DNM3/Geometric MeanCCCGGG01234 ** p=0.006    CC  CG  GG Dyn3A/Dyn3B 1.53  1.46  0.98 Normalized DNM3 110   Figure 29. Dynamin 3 protein levels normalized by GAPDH    DNM3 protein levelsDNM3 rs2421947 genotypesDNM3/GAPDHCCGG0. p=0.08111   Figure 30. Dynamin 3 staining in cortical neurons  A. representative confocal microscopic images of dynamin-3 (red) and MAP2 (blue) staining in wild-type (WT) and GKI (LRRK2 p.G2019S) murine cortical neurons, cultured as previously described (Beccano-Kelly, et al., 2014)Left: 60X 2-times zoom of individual neuron staining. Right: expanded region of interest with and without MAP2; B. Quantification of dynamin-3 intensity in cortical cultures (DIV=21). Scale bars, 50um, n=3 cultures per group; C.  Quantification of dynamin-3 cluster density in cortical cultures (DIV=21) **=p<0.05.    A B C 112    Figure 31. Flow diagram of discovery and replication cohorts   113  5. Chapter 5: Elucidating mechanisms of reduced penetrance in Mendelian disease  5.1. The importance of reduced penetrance  It has been over 10 years since the discovery of LRRK2 mutations in PD. The penetrance of LRRK2 p.G2019S parkinsonism is complex and varies across ethnicities and environments. Other pathogenic mutations in LRRK2 also show variable penetrance estimates. The penetrance estimates for p.G2019S are relevant for genetic counseling, but  treatment and prognosis for these patients are the same as typical PD.  The lack of a definitive cure for PD drives the search for modifier genes that are informative for genetic counseling, disease severity and potential new avenues for therapeutics. However, discovering penetrance modifiers in monogenic forms of disease (albeit genetic or environmental) requires large numbers of mutation carriers (both in families and sporadic unrelated patients) for sufficient power. Here we have a large homogeneous Tunisian population with a high frequency of the LRRK2 p.G2019S mutation. Our study was limited by the rarity of this Mendelian form of parkinsonism, in families and in population-based series of idiopathic PD, and it has taken 10 years of research to build this valuable resource of clinical and genetic data. In PD, homozygous or compound heterozygous mutations in Parkin and PINK1 are highly penetrant genotypes in early onset parkinsonism, heterozygous mutations in LRRK2, VPS35, EIF4G1, have reduced penetrance estimates and heterozygous states of Parkin/PINK1 mutations may be regarded as pathogenic but with very low penetrance, although this is debatable (Klein & Ziegler, 2011) . Thus far, there has not been a genetic modifier identified for genes implicated in monogenic forms of PD. Although few genetic modifiers have been identified and validated, there are many biological candidates. Mutations of GBA are an important risk factor for PD which reduce enzyme activity leading to ER associated degradation. 114  The GBA enzyme, GCase, interacts with alpha synuclein. Reduced GCase in GBA mutation is associated with increased SNCA (Schapira, 2015) .  There are several examples of genetic modifiers in movement disorders and neurodegeneration. Mutations in  SGCE  lead to development of myoclonus-dystonia. However, maternal imprinting of SGCE does not lead to disease in the offspring when transmitted through the mother (Guettard et al., 2008) . The finding is extremely relevant for young female patients with an SGCE mutation, as their children will not suffer from the disease. Likewise, DYT1 dystonia is caused by a TOR1A GAG deletion. However, when p.D216H polymorphism in TOR1A is present in trans, there is reduced penetrance of the TOR1A GAG deletion to 3% (Bruggemann et al., 2009; Kamm et al., 2008; Klein, 2014).   Herein, we have the unique opportunity of a large homogeneous cohort with one identical mutation to study age-at-onset genetic modifiers. The work in this thesis has made use of collected detailed clinical research forms to study and characterize endophenotypes in LRRK2 parkinsonism. The work has demonstrated the usefulness of families in linkage analysis, withsubsequent use of whole-genome sequencing and haplotype analysis. We have identified a potential age-at-onset modifier for the most common mutation in familial PD.    5.2. Factors that influence penetrance The phenotypic manifestation of mutations in neurodegenerative diseases are age-dependent, e.g. c9orf72 in FTD/ALS, LRRK2 mutations in PD, HTT expansion in HD.  The risk of developing the disease increases with age. The mutation type can also influence penetrance. Some mutations are more penetrant than others. For example, we have found that SNCA point 115  mutations, duplications and triplications are more highly penetrant in comparison to LRRK2 point mutations in PD(Trinh, Guella, et al., 2014). This may be due to mutation type (duplication or triplications can severely influence the patient compared to point mutations). It can also be due to gene-specific differences. Perhaps perturbations in SNCA  have a larger effect on disease processes compared to LRRK2. But even within the same gene, the penetrance estimates are vastly different. For example, LRRK2 p.G2019S is more penetrant compared to the R1441G/C/H mutations. Perhaps penetrance correlates with the kinase or GTPase activity in LRRK2. This phenomena is not specific to PD. Cis and trans elements that control gene expression can also influence the penetrance of mutations. If there is unequal expression of the wild-type and the mutant allele, then there would be an influence on expression levels. The ethnic or environmental background can influence penetrance. In our study, we have shown that cumulative incidence estimates are significantly different between Norwegians and Tunisians with the LRRK2 p.G2019S mutation (Hentati, et al., 2014) . Furthermore, new studies with larger sample sizes show that there are differences between Ashkenazi Jews from New York (n=90 LRRK2 p.G2019Scarriers) and Tunisian Arab Berbers (n=220 LRRK2 p.G2019S carriers)with the same LRRK2 p.G2019S mutation (Marder, et al., 2015; Trinh, Guella, et al., 2014). Gender can also influence penetrance, although controversial, females may have higher risk of disease compared to males in LRRK2 p.G2019S (Cilia et al., 2014) (Trinh, Amouri, et al., 2014)  . This is in contrast to the higher risk of idiopathic PD in males compared to females (de Lau & Breteler, 2006) .     116   5.3. Methods and approaches to identify genetic modifiers The most obvious candidates for genetic modifiers may be the top GWAS hits. Genes such as SNCA, MAPT, RAB7L1 that contribute to risk of PD could also modify age-at-onset in idiopathic PD. In fact, there are a few polymorphisms in SNCA and the TMEM175/GAK loci that may influence age-at-onset (Lill et al., 2015; Ritz, Rhodes, Bordelon, & Bronstein, 2012). A combined genetic risk score for age-at-onset of all significantly associated SNPs revealed that the signal was mostly driven by SNCA and the TMEM175/GAK. There was a reduction of the effect when these two top SNPs were removed and thus other PD risk loi besides SNCA and TMEM175/GAK  have a relatively small contribution to AAO variability (Lill, et al., 2015) . Another study has shown that SNCA rs356165 and rs356219 modifies age-at-onset in idiopathic PD (Brockmann et al., 2013) .  However, we have found that SNCA polymorphisms do not have an effect on disease risk or onset age in LRRK2 p.G2019S carriers (Trinh, Gustavsson, et al., 2014), suggesting that PD GWAS risk loci may have a relatively small contribution to modifying endophenotypes in Mendelian forms of PD.  A combination of linkage analyses, genome-wide association studies, meta-analyses and exome sequencing of ‘extreme’ cases have been used to identify modifiers of disease severity and comorbidities in the field of complex genetic disorders (Wright et al., 2011) (Emond et al., 2012) which requires good phenotyping, especially in the context of movement disorders and longitudinal follow up on families and patients. It also requires large Mendelian families or sib-pairs segregating with disease. Many modifiers of endophenotypes in cystic fibrosis have been explored with linkage analysis of phenotypes such as lung disease severity (Corvol et al., 2015; Emond, et al., 2012; Wright, et al., 2011). A genome-wide study on 486 sib-pairs identified linkage on chromosome 20q13.2 that modifies lung disease severity (Wright, et al., 2011) .  117  Another method is looking for ‘protective’ alleles. These are alleles that lower risk of getting disease. One example is in the context of cholesterol low-density lipoprotein, loss of function variants in PCSK9 were found in individuals with low levels of LDL cholesterol (Cohen et al., 2005) . Other examples include inactivating mutations in NPC1L1 and APOC3 protecting from coronary heart disease (Crosby et al., 2014; "Inactivating mutations in NPC1L1 and protection from coronary heart disease," 2014; Jorgensen, Frikke-Schmidt, Nordestgaard, & Tybjaerg-Hansen, 2014) . Protein inactivating mutations in NPC1L1 such as p.Arg406X were more associated with lower LDL cholesterol levels and lower risk of coronary heart disease. Protective alleles may also exist in the LRRK2 p.G2019S carriers. The Norwegian population may carry ‘protective’ alleles that the Tunisian Arab-Berber population does not carry. Interestingly, the DNM3 rs2421947 GG genotype is almost absent in the Norwegian LRRK2 p.G2019S carriers.  Penetrance of mutations can be modified by expression levels. Using translational models of human stem cells or other mammalian models to look for transcriptomic differences may be one important step to test potential candidate modifiers or look for novel modifiers. THAP1 mutations can cause early onset primary torsion dystonia, with an autosomal-dominant inheritance and 40% penetrance (T. Fuchs et al., 2009) . THAP1 encodes a transcription factor that regulates expression of TOR1A and also autoregulates its own expression levels (Erogullari et al., 2014) .  Based on this multiplicity of mechanisms and their conceivable interactions, it appears unlikely that a single approach will suffice to arrive at a comprehensive understanding of the molecular mechanisms underlying reduced penetrance of movement disorders. In this respect, a 118  number of DNA and RNA based genetic methods, complemented by functional models may be required for thorough investigations of potential modifier genes.    5.4. Dynamin 3 as potential therapeutic target of LRRK2 parkinsonism LRRK2 has been found to interact with various presynaptic proteins: AP3, clathrin, dynamin-1 (Schreij et al., 2015; Stafa, et al., 2014; Waschbusch et al., 2014)  .These presynaptic proteins are important for maintaining reserve vesicle pools and membrane fusion. LRRK2 binds to purified synaptic vesicles and perhaps regulates exocytosis, modulating vesicle pool mobilization (Piccoli et al., 2014) . We identified a genetic modifier of LRRK2 parkinsonism that is heavily involved in synaptic vesicle fission and release of clathrin (S. M. Ferguson & De Camilli, 2012; Raimondi et al., 2011; Wu et al., 2014) . DNM3 rs2421947 GG is associated with earlier age at onset and  higher gene expression in human control striatum. The dynamin 3b isoform which is involved in regulation of actin polymerization, filopodia and spine formation is also more highly expressed. Lastly, a significant redistribution of dendritic dynamin 3 staining is observed in LRRK2 p.G2019S murine cortical culture. The discovery directs therapeutic development to dynamin 3, as a neuroprotective strategy for LRRK2 parkinsonism or subjects with Parkinson disease. Diagnostics/therapeutics targeting (a) DNM3 nucleic acid , (b) reducing the levels of DNM3 GTPase activity, protein or mRNA may help delay the onset of and prevent symptom progression since the higher gene expression is associated with earlier age of onset. The therapeutic target may even be generalizable to treat other neurodegenerative disorders with similar pathogenesis such as Alzheimer's disease, Huntington disease, immune and inflammatory disorders.  Suppression of dynamin GTPase decreases a-synuclein uptake by neuronal and oligodendroglial cells and amyloid-beta internalization (Konno, et al., 2012; Yu, Nwabuisi-119  Heath, Laxton, & Ladu, 2010) . Drp-1 (dynamin-related protein like 1) inhibitors were shown to protect against ischemic neuronal injury through inhibiting mitochondrial calcium uptake (Tian et al., 2014). Preliminary evidence has shown that small molecule dynamin inhibitors can also be an anticonvulsant drug, acting to control synaptic transmission as a novel target for epilepsy.   A limitation is the therapeutic potential of dynamin 3. Thus far, there has not been a specific drug to readily target dynamin 3, although non-specific inhibitors such as dynasore (which inhibits GTPase activity of dynamin I and dynamin II but not dynamin III) exist.   Other more potent series include: dimeric tyrphostins, long chain amines and ammonium salts (myristyl trimethyl ammonium bromides), dynoles, iminodyns and pthaladyns are other drugs that inhibit dynamin. These drugs have been considered in cancer treatments to induce apoptosis following cytokinesis failure in a concentration-dependent manner (Chircop et al., 2011; Joshi, Braithwaite, Robinson, & Chircop, 2011) . Another disadvantage is that pharmacological inhibition of dynamin in mice has reduced long-term potentiation and resulted in memory loss (Fa, Staniszewski, Saeed, Francis, & Arancio, 2014) . This limitation can be overcome with careful monitoring of dynamin inhibitor levels. Perhaps a moderate to low level of inhibitor will be beneficial and neuroprotective whereas more potent levels lead to apoptosis. Alternatively, allosteric modulators of dynamin GTPase activity might be considered.     5.5. Conclusion The identification of risk loci, genes and mutations in PD has provided new insights into disease aetiology and highlighted new study approaches. Several biological processes involved in PD pathogenesis have been highlighted, and the discovery of novel PD-associated genes in families with Mendelian disease has been particularly informative in this regard. Historically, 120  each major discovery has defined a major theme for translational neuroscience. For example, the discovery of α‑synuclein as a key component of Lewy bodies highlighted protein aggregation and propagation,  and the discovery of parkin highlighted protein ubiquitination and the proteosome. Each discovery generally led to a change and/or replacement of focus. Recently, some pathways have emerged that relate to mitochondrial metabolism (PINK1, PARK2) and lysosomal-autophagy (ATP13A2, GBA, LRRK2). Nevertheless, the ultimate focus must be on late-onset Lewy body PD as, clinically and pathologically, this phenotype describes the vast majority of patients. On the basis of the finding of DNM3 as a penetrance modifier of LRRK2 parkinsonism, we postulate a unifying synthesis whereby deficits in synaptic exocytosis and endocytosis involving DNAJC6, DNAJC13, VPS35, SNCA and LRRK2 are relevant for the clinical phenotype of disease onset.  With the advent of next-generation sequencing, we anticipate genetic advances in PD will continue to flourish, and our understanding of the molecular mechanisms underlying susceptibility, progression and response to treatment will continue to evolve.   121  References Aasly, J. O., Toft, M., Fernandez-Mata, I., Kachergus, J., Hulihan, M., White, L. R., et al. (2005). Clinical features of LRRK2-associated Parkinson's disease in central Norway. Ann Neurol, 57(5), 762-765. Abecasis, G. R., Cherny, S. S., Cookson, W. O., & Cardon, L. R. (2002). Merlin--rapid analysis of dense genetic maps using sparse gene flow trees. Nat Genet, 30(1), 97-101. Alcalay, R. N., Mejia-Santana, H., Mirelman, A., Saunders-Pullman, R., Raymond, D., Palmese, C., et al. (2015). Neuropsychological performance in LRRK2 G2019S carriers with Parkinson's disease. Parkinsonism Relat Disord, 21(2), 106-110. Alcalay, R. N., Mejia-Santana, H., Tang, M. X., Rosado, L., Verbitsky, M., Kisselev, S., et al. (2009). Motor phenotype of LRRK2 G2019S carriers in early-onset Parkinson disease. Arch Neurol, 66(12), 1517-1522. Alves, G., Wentzel-Larsen, T., & Larsen, J. P. (2004). Is fatigue an independent and persistent symptom in patients with Parkinson disease? Neurology, 63(10), 1908-1911. Ando, M., Funayama, M., Li, Y., Kashihara, K., Murakami, Y., Ishizu, N., et al. (2012). VPS35 mutation in Japanese patients with typical Parkinson's disease. Mov Disord, 27(11), 1413-1417. Annesi, G., Savettieri, G., Pugliese, P., D'Amelio, M., Tarantino, P., Ragonese, P., et al. (2005). DJ-1 mutations and parkinsonism-dementia-amyotrophic lateral sclerosis complex. Ann Neurol, 58(5), 803-807. Arranz, A. M., Delbroek, L., Van Kolen, K., Guimaraes, M. R., Mandemakers, W., Daneels, G., et al. (2015). LRRK2 functions in synaptic vesicle endocytosis through a kinase-dependent mechanism. J Cell Sci, 128(3), 541-552. Auton, A., Brooks, L. D., Durbin, R. M., Garrison, E. P., Kang, H. M., Korbel, J. O., et al. (2015). A global reference for human genetic variation. Nature, 526(7571), 68-74. Beccano-Kelly, D. A., Kuhlmann, N., Tatarnikov, I., Volta, M., Munsie, L. N., Chou, P., et al. (2014). Synaptic function is modulated by LRRK2 and glutamate release is increased in cortical neurons of G2019S LRRK2 knock-in mice. Front Cell Neurosci, 8, 301. Beilina, A., Rudenko, I. N., Kaganovich, A., Civiero, L., Chau, H., Kalia, S. K., et al. (2014). Unbiased screen for interactors of leucine-rich repeat kinase 2 supports a common pathway for sporadic and familial Parkinson disease. Proc Natl Acad Sci U S A, 111(7), 2626-2631. Bhalla, A., Vetanovetz, C. P., Morel, E., Chamoun, Z., Di Paolo, G., & Small, S. A. (2012). The location and trafficking routes of the neuronal retromer and its role in amyloid precursor protein transport. Neurobiol Dis, 47(1), 126-134. Bi, F., Li, F., Huang, C., & Zhou, H. (2013). Pathogenic mutation in VPS35 impairs its protection against MPP(+) cytotoxicity. Int J Biol Sci, 9(2), 149-155. Bonifati, V., Rizzu, P., van Baren, M. J., Schaap, O., Breedveld, G. J., Krieger, E., et al. (2003). Mutations in the DJ-1 gene associated with autosomal recessive early-onset parkinsonism. Science, 299(5604), 256-259. Bower, J. H., Maraganore, D. M., McDonnell, S. K., & Rocca, W. A. (1999). Incidence and distribution of parkinsonism in Olmsted County, Minnesota, 1976-1990. Neurology, 52(6), 1214-1220. 122  Braak, H., Del Tredici, K., Rub, U., de Vos, R. A., Jansen Steur, E. N., & Braak, E. (2003). Staging of brain pathology related to sporadic Parkinson's disease. Neurobiol Aging, 24(2), 197-211. Brockmann, K., Schulte, C., Hauser, A. K., Lichtner, P., Huber, H., Maetzler, W., et al. (2013). SNCA: major genetic modifier of age at onset of Parkinson's disease. Mov Disord, 28(9), 1217-1221. Brown, R. G., Dittner, A., Findley, L., & Wessely, S. C. (2005). The Parkinson fatigue scale. Parkinsonism Relat Disord, 11(1), 49-55. Browning, B. L., & Browning, S. R. (2008). Haplotypic analysis of Wellcome Trust Case Control Consortium data. Hum Genet, 123(3), 273-280. Browning, B. L., & Browning, S. R. (2009). A unified approach to genotype imputation and haplotype-phase inference for large data sets of trios and unrelated individuals. Am J Hum Genet, 84(2), 210-223. Bruggemann, N., Kock, N., Lohmann, K., Konig, I. R., Rakovic, A., Hagenah, J., et al. (2009). The D216H variant in the DYT1 gene: a susceptibility factor for dystonia in familial cases? Neurology, 72(16), 1441-1443. Burn, D. J., Rowan, E. N., Allan, L. M., Molloy, S., O'Brien, J. T., & McKeith, I. G. (2006). Motor subtype and cognitive decline in Parkinson's disease, Parkinson's disease with dementia, and dementia with Lewy bodies. J Neurol Neurosurg Psychiatry, 77(5), 585-589. Cao, H., Garcia, F., & McNiven, M. A. (1998). Differential distribution of dynamin isoforms in mammalian cells. Mol Biol Cell, 9(9), 2595-2609. Chartier-Harlin, M. C., Dachsel, J. C., Vilarino-Guell, C., Lincoln, S. J., Lepretre, F., Hulihan, M. M., et al. (2011). Translation initiator EIF4G1 mutations in familial Parkinson disease. Am J Hum Genet, 89(3), 398-406. Chartier-Harlin, M. C., Kachergus, J., Roumier, C., Mouroux, V., Douay, X., Lincoln, S., et al. (2004). Alpha-synuclein locus duplication as a cause of familial Parkinson's disease. Lancet, 364(9440), 1167-1169. Chaudhuri, K. R., Healy, D. G., Schapira, A. H., & National Institute for Clinical, E. (2006). Non-motor symptoms of Parkinson's disease: diagnosis and management. Lancet Neurol, 5(3), 235-245. Chircop, M., Perera, S., Mariana, A., Lau, H., Ma, M. P., Gilbert, J., et al. (2011). Inhibition of dynamin by dynole 34-2 induces cell death following cytokinesis failure in cancer cells. Mol Cancer Ther, 10(9), 1553-1562. Cilia, R., Siri, C., Rusconi, D., Allegra, R., Ghiglietti, A., Sacilotto, G., et al. (2014). LRRK2 mutations in Parkinson's disease: confirmation of a gender effect in the Italian population. Parkinsonism Relat Disord, 20(8), 911-914. Clague, M. J., & Rochin, L. (2016). Parkinson's Disease: A Traffic Jam? Curr Biol, 26(8), R332-334. Clark, L. N., Wang, Y., Karlins, E., Saito, L., Mejia-Santana, H., Harris, J., et al. (2006). Frequency of LRRK2 mutations in early- and late-onset Parkinson disease. Neurology, 67(10), 1786-1791. Cohen, J., Pertsemlidis, A., Kotowski, I. K., Graham, R., Garcia, C. K., & Hobbs, H. H. (2005). Low LDL cholesterol in individuals of African descent resulting from frequent nonsense mutations in PCSK9. Nat Genet, 37(2), 161-165. 123  Cookson, M. R. (2010). The role of leucine-rich repeat kinase 2 (LRRK2) in Parkinson's disease. Nat Rev Neurosci, 11(12), 791-797. Cookson, M. R., Lockhart, P. J., McLendon, C., O'Farrell, C., Schlossmacher, M., & Farrer, M. J. (2003). RING finger 1 mutations in Parkin produce altered localization of the protein. Hum Mol Genet, 12(22), 2957-2965. Corvol, H., Blackman, S. M., Boelle, P. Y., Gallins, P. J., Pace, R. G., Stonebraker, J. R., et al. (2015). Genome-wide association meta-analysis identifies five modifier loci of lung disease severity in cystic fibrosis. Nat Commun, 6, 8382. Crosby, J., Peloso, G. M., Auer, P. L., Crosslin, D. R., Stitziel, N. O., Lange, L. A., et al. (2014). Loss-of-function mutations in APOC3, triglycerides, and coronary disease. N Engl J Med, 371(1), 22-31. Cuyvers, E., De Roeck, A., Van den Bossche, T., Van Cauwenberghe, C., Bettens, K., Vermeulen, S., et al. (2015). Mutations in ABCA7 in a Belgian cohort of Alzheimer's disease patients: a targeted resequencing study. Lancet Neurol, 14(8), 814-822. Dawn Teare, M., & Barrett, J. H. (2005). Genetic linkage studies. Lancet, 366(9490), 1036-1044. de Lau, L. M., & Breteler, M. M. (2006). Epidemiology of Parkinson's disease. Lancet Neurol, 5(6), 525-535. Dhungel, N., Eleuteri, S., Li, L. B., Kramer, N. J., Chartron, J. W., Spencer, B., et al. (2015). Parkinson's disease genes VPS35 and EIF4G1 interact genetically and converge on alpha-synuclein. Neuron, 85(1), 76-87. Di Fonzo, A., Chien, H. F., Socal, M., Giraudo, S., Tassorelli, C., Iliceto, G., et al. (2007). ATP13A2 missense mutations in juvenile parkinsonism and young onset Parkinson disease. Neurology, 68(19), 1557-1562. Emond, M. J., Louie, T., Emerson, J., Zhao, W., Mathias, R. A., Knowles, M. R., et al. (2012). Exome sequencing of extreme phenotypes identifies DCTN4 as a modifier of chronic Pseudomonas aeruginosa infection in cystic fibrosis. Nat Genet, 44(8), 886-889. Erogullari, A., Hollstein, R., Seibler, P., Braunholz, D., Koschmidder, E., Depping, R., et al. (2014). THAP1, the gene mutated in DYT6 dystonia, autoregulates its own expression. Biochim Biophys Acta, 1839(11), 1196-1204. Fa, M., Staniszewski, A., Saeed, F., Francis, Y. I., & Arancio, O. (2014). Dynamin 1 is required for memory formation. PLoS One, 9(3), e91954. Fahn, S. (2000). The spectrum of levodopa-induced dyskinesias. Ann Neurol, 47(4 Suppl 1), S2-9; discussion S9-11. Fahn, S. (2003). Description of Parkinson's disease as a clinical syndrome. Ann N Y Acad Sci, 991, 1-14. Farrer, M., Chan, P., Chen, R., Tan, L., Lincoln, S., Hernandez, D., et al. (2001). Lewy bodies and parkinsonism in families with parkin mutations. Ann Neurol, 50(3), 293-300. Farrer, M., Kachergus, J., Forno, L., Lincoln, S., Wang, D. S., Hulihan, M., et al. (2004). Comparison of kindreds with parkinsonism and alpha-synuclein genomic multiplications. Ann Neurol, 55(2), 174-179. Fellner, L., Wenning, G. K., & Stefanova, N. (2015). Models of multiple system atrophy. Curr Top Behav Neurosci, 22, 369-393. Ferguson, L. W., Rajput, M. L., Muhajarine, N., Shah, S. M., & Rajput, A. (2008). Clinical features at first visit and rapid disease progression in Parkinson's disease. Parkinsonism Relat Disord, 14(5), 431-435. 124  Ferguson, S. M., & De Camilli, P. (2012). Dynamin, a membrane-remodelling GTPase. Nat Rev Mol Cell Biol, 13(2), 75-88. Forman, M. S., Schmidt, M. L., Kasturi, S., Perl, D. P., Lee, V. M., & Trojanowski, J. Q. (2002). Tau and alpha-synuclein pathology in amygdala of Parkinsonism-dementia complex patients of Guam. Am J Pathol, 160(5), 1725-1731. Fuchs, J., Nilsson, C., Kachergus, J., Munz, M., Larsson, E. M., Schule, B., et al. (2007). Phenotypic variation in a large Swedish pedigree due to SNCA duplication and triplication. Neurology, 68(12), 916-922. Fuchs, J., Tichopad, A., Golub, Y., Munz, M., Schweitzer, K. J., Wolf, B., et al. (2008). Genetic variability in the SNCA gene influences alpha-synuclein levels in the blood and brain. FASEB J, 22(5), 1327-1334. Fuchs, T., Gavarini, S., Saunders-Pullman, R., Raymond, D., Ehrlich, M. E., Bressman, S. B., et al. (2009). Mutations in the THAP1 gene are responsible for DYT6 primary torsion dystonia. Nat Genet, 41(3), 286-288. Funayama, M., Ohe, K., Amo, T., Furuya, N., Yamaguchi, J., Saiki, S., et al. (2015). CHCHD2 mutations in autosomal dominant late-onset Parkinson's disease: a genome-wide linkage and sequencing study. Lancet Neurol, 14(3), 274-282. Gan-Or, Z., Giladi, N., Rozovski, U., Shifrin, C., Rosner, S., Gurevich, T., et al. (2008). Genotype-phenotype correlations between GBA mutations and Parkinson disease risk and onset. Neurology, 70(24), 2277-2283. Goedert, M., Spillantini, M. G., Del Tredici, K., & Braak, H. (2013a). 100 years of Lewy pathology. Nat Rev Neurol, 9(1), 13-24. Goedert, M., Spillantini, M. G., Del Tredici, K., & Braak, H. (2013b). 100 years of Lewy pathology. [Research Support, Non-U.S. Gov't]. Nature reviews. Neurology, 9(1), 13-24. Goetz, C. G., Nutt, J. G., & Stebbins, G. T. (2008). The Unified Dyskinesia Rating Scale: presentation and clinimetric profile. Mov Disord, 23(16), 2398-2403. Goetz, C. G., Tilley, B. C., Shaftman, S. R., Stebbins, G. T., Fahn, S., Martinez-Martin, P., et al. (2008). Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS): scale presentation and clinimetric testing results. Mov Disord, 23(15), 2129-2170. Golbe, L. I. (1999). Alpha-synuclein and Parkinson's disease. Mov Disord, 14(1), 6-9. Goldwurm, S., Tunesi, S., Tesei, S., Zini, M., Sironi, F., Primignani, P., et al. (2011). Kin-cohort analysis of LRRK2-G2019S penetrance in Parkinson's disease. Mov Disord, 26(11), 2144-2145. Goldwurm, S., Zini, M., Mariani, L., Tesei, S., Miceli, R., Sironi, F., et al. (2007). Evaluation of LRRK2 G2019S penetrance: relevance for genetic counseling in Parkinson disease. Neurology, 68(14), 1141-1143. Gray, N. W., Kruchten, A. E., Chen, J., & McNiven, M. A. (2005). A dynamin-3 spliced variant modulates the actin/cortactin-dependent morphogenesis of dendritic spines. J Cell Sci, 118(Pt 6), 1279-1290. Gregory, A., & Hayflick, S. J. (2011). Genetics of neurodegeneration with brain iron accumulation. Curr Neurol Neurosci Rep, 11(3), 254-261. Grunewald, A., Gegg, M. E., Taanman, J. W., King, R. H., Kock, N., Klein, C., et al. (2009). Differential effects of PINK1 nonsense and missense mutations on mitochondrial function and morphology. Exp Neurol, 219(1), 266-273. 125  Guettard, E., Portnoi, M. F., Lohmann-Hedrich, K., Keren, B., Rossignol, S., Winkler, S., et al. (2008). Myoclonus-dystonia due to maternal uniparental disomy. Arch Neurol, 65(10), 1380-1385. Gustavsson, E. K., Guella, I., Trinh, J., Szu-Tu, C., Rajput, A., Rajput, A. H., et al. (2015). Genetic variability of the retromer cargo recognition complex in parkinsonism. Mov Disord, 30(4), 580-584. Gustavsson, E. K., Trinh, J., Guella, I., Vilarino-Guell, C., Appel-Cresswell, S., Stoessl, A. J., et al. (2015). DNAJC13 genetic variants in parkinsonism. Mov Disord, 30(2), 273-278. Haack, T. B., Hogarth, P., Gregory, A., Prokisch, H., & Hayflick, S. J. (2013). BPAN: the only X-linked dominant NBIA disorder. Int Rev Neurobiol, 110, 85-90. Haack, T. B., Hogarth, P., Kruer, M. C., Gregory, A., Wieland, T., Schwarzmayr, T., et al. (2012). Exome sequencing reveals de novo WDR45 mutations causing a phenotypically distinct, X-linked dominant form of NBIA. Am J Hum Genet, 91(6), 1144-1149. Harold, D., Abraham, R., Hollingworth, P., Sims, R., Gerrish, A., Hamshere, M. L., et al. (2009). Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer's disease. Nat Genet, 41(10), 1088-1093. Hasegawa, K., Stoessl, A. J., Yokoyama, T., Kowa, H., Wszolek, Z. K., & Yagishita, S. (2009). Familial parkinsonism: study of original Sagamihara PARK8 (I2020T) kindred with variable clinicopathologic outcomes. Parkinsonism Relat Disord, 15(4), 300-306. Haugarvoll, K., Rademakers, R., Kachergus, J. M., Nuytemans, K., Ross, O. A., Gibson, J. M., et al. (2008). Lrrk2 R1441C parkinsonism is clinically similar to sporadic Parkinson disease. Neurology, 70(16 Pt 2), 1456-1460. Haugarvoll, K., & Wszolek, Z. K. (2009). Clinical features of LRRK2 parkinsonism. Parkinsonism Relat Disord, 15 Suppl 3, S205-208. Hauser, E. R., Watanabe, R. M., Duren, W. L., Bass, M. P., Langefeld, C. D., & Boehnke, M. (2004). Ordered subset analysis in genetic linkage mapping of complex traits. Genet Epidemiol, 27(1), 53-63. Healy, D. G., Falchi, M., O'Sullivan, S. S., Bonifati, V., Durr, A., Bressman, S., et al. (2008). Phenotype, genotype, and worldwide genetic penetrance of LRRK2-associated Parkinson's disease: a case-control study. Lancet Neurol, 7(7), 583-590. Heckman, M. G., Elbaz, A., Soto-Ortolaza, A. I., Serie, D. J., Aasly, J. O., Annesi, G., et al. (2014). Protective effect of LRRK2 p.R1398H on risk of Parkinson's disease is independent of MAPT and SNCA variants. Neurobiol Aging, 35(1), 266 e265-214. Heckman, M. G., Schottlaender, L., Soto-Ortolaza, A. I., Diehl, N. N., Rayaprolu, S., Ogaki, K., et al. (2014). LRRK2 exonic variants and risk of multiple system atrophy. Neurology, 83(24), 2256-2261. Heckman, M. G., Soto-Ortolaza, A. I., Aasly, J. O., Abahuni, N., Annesi, G., Bacon, J. A., et al. (2013). Population-specific frequencies for LRRK2 susceptibility variants in the Genetic Epidemiology of Parkinson's Disease (GEO-PD) Consortium. Mov Disord, 28(12), 1740-1744. Helfer, E., Harbour, M. E., Henriot, V., Lakisic, G., Sousa-Blin, C., Volceanov, L., et al. (2013). Endosomal recruitment of the WASH complex: active sequences and mutations impairing interaction with the retromer. Biol Cell, 105(5), 191-207. Hentati, F., Trinh, J., Thompson, C., Nosova, E., Farrer, M. J., & Aasly, J. O. (2014). LRRK2 parkinsonism in Tunisia and Norway: a comparative analysis of disease penetrance. Neurology, 83(6), 568-569. 126  Hollingworth, P., Harold, D., Sims, R., Gerrish, A., Lambert, J. C., Carrasquillo, M. M., et al. (2011). Common variants at ABCA7, MS4A6A/MS4A4E, EPHA1, CD33 and CD2AP are associated with Alzheimer's disease. Nat Genet, 43(5), 429-435. Howey, R., & Cordell, H. J. (2014). Imputation without doing imputation: a new method for the detection of non-genotyped causal variants. Genet Epidemiol, 38(3), 173-190. Hulihan, M. M., Ishihara-Paul, L., Kachergus, J., Warren, L., Amouri, R., Elango, R., et al. (2008). LRRK2 Gly2019Ser penetrance in Arab-Berber patients from Tunisia: a case-control genetic study. Lancet Neurol, 7(7), 591-594. Ibanez, P., Bonnet, A. M., Debarges, B., Lohmann, E., Tison, F., Pollak, P., et al. (2004). Causal relation between alpha-synuclein gene duplication and familial Parkinson's disease. Lancet, 364(9440), 1169-1171. Inactivating mutations in NPC1L1 and protection from coronary heart disease. (2014). N Engl J Med, 371(22), 2072-2082. Ishihara-Paul, L., Hulihan, M. M., Kachergus, J., Upmanyu, R., Warren, L., Amouri, R., et al. (2008). PINK1 mutations and parkinsonism. Neurology, 71(12), 896-902. Johansen, K. K., White, L. R., Farrer, M. J., & Aasly, J. O. (2011). Subclinical signs in LRRK2 mutation carriers. Parkinsonism Relat Disord, 17(7), 528-532. Johns, M. W. (1991). A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep, 14(6), 540-545. Jorgensen, A. B., Frikke-Schmidt, R., Nordestgaard, B. G., & Tybjaerg-Hansen, A. (2014). Loss-of-function mutations in APOC3 and risk of ischemic vascular disease. N Engl J Med, 371(1), 32-41. Joshi, S., Braithwaite, A. W., Robinson, P. J., & Chircop, M. (2011). Dynamin inhibitors induce caspase-mediated apoptosis following cytokinesis failure in human cancer cells and this is blocked by Bcl-2 overexpression. Mol Cancer, 10, 78. Kachergus, J., Mata, I. F., Hulihan, M., Taylor, J. P., Lincoln, S., Aasly, J., et al. (2005). Identification of a novel LRRK2 mutation linked to autosomal dominant parkinsonism: evidence of a common founder across European populations. Am J Hum Genet, 76(4), 672-680. Kalia, L. V., Lang, A. E., Hazrati, L. N., Fujioka, S., Wszolek, Z. K., Dickson, D. W., et al. (2015). Clinical correlations with Lewy body pathology in LRRK2-related Parkinson disease. JAMA Neurol, 72(1), 100-105. Kamm, C., Fischer, H., Garavaglia, B., Kullmann, S., Sharma, M., Schrader, C., et al. (2008). Susceptibility to DYT1 dystonia in European patients is modified by the D216H polymorphism. Neurology, 70(23), 2261-2262. Khan, N. L., Jain, S., Lynch, J. M., Pavese, N., Abou-Sleiman, P., Holton, J. L., et al. (2005). Mutations in the gene LRRK2 encoding dardarin (PARK8) cause familial Parkinson's disease: clinical, pathological, olfactory and functional imaging and genetic data. Brain, 128(Pt 12), 2786-2796. Kitada, T., Asakawa, S., Hattori, N., Matsumine, H., Yamamura, Y., Minoshima, S., et al. (1998). Mutations in the parkin gene cause autosomal recessive juvenile parkinsonism. Nature, 392(6676), 605-608. Klein, C. (2014). Genetics in dystonia. Parkinsonism Relat Disord, 20 Suppl 1, S137-142. Klein, C., & Ziegler, A. (2011). From GWAS to clinical utility in Parkinson's disease. Lancet, 377(9766), 613-614. 127  Koga, S., Aoki, N., Uitti, R. J., van Gerpen, J. A., Cheshire, W. P., Josephs, K. A., et al. (2015). When DLB, PD, and PSP masquerade as MSA: an autopsy study of 134 patients. Neurology, 85(5), 404-412. Konno, M., Hasegawa, T., Baba, T., Miura, E., Sugeno, N., Kikuchi, A., et al. (2012). Suppression of dynamin GTPase decreases alpha-synuclein uptake by neuronal and oligodendroglial cells: a potent therapeutic target for synucleinopathy. Mol Neurodegener, 7, 38. Krebs, C. E., Karkheiran, S., Powell, J. C., Cao, M., Makarov, V., Darvish, H., et al. (2013). The Sac1 domain of SYNJ1 identified mutated in a family with early-onset progressive Parkinsonism with generalized seizures. Hum Mutat, 34(9), 1200-1207. Kruger, R., Kuhn, W., Muller, T., Woitalla, D., Graeber, M., Kosel, S., et al. (1998). Ala30Pro mutation in the gene encoding alpha-synuclein in Parkinson's disease. Nat Genet, 18(2), 106-108. Kumar, K. R., Weissbach, A., Heldmann, M., Kasten, M., Tunc, S., Sue, C. M., et al. (2012). Frequency of the D620N mutation in VPS35 in Parkinson disease. Arch Neurol, 69(10), 1360-1364. Lai, Y. C., Kondapalli, C., Lehneck, R., Procter, J. B., Dill, B. D., Woodroof, H. I., et al. (2015). Phosphoproteomic screening identifies Rab GTPases as novel downstream targets of PINK1. EMBO J, 34(22), 2840-2861. Lambert, J. C., Grenier-Boley, B., Harold, D., Zelenika, D., Chouraki, V., Kamatani, Y., et al. (2013). Genome-wide haplotype association study identifies the FRMD4A gene as a risk locus for Alzheimer's disease. Mol Psychiatry, 18(4), 461-470. Latourelle, J. C., Hendricks, A. E., Pankratz, N., Wilk, J. B., Halter, C., Nichols, W. C., et al. (2011). Genomewide linkage study of modifiers of LRRK2-related Parkinson's disease. Mov Disord, 26(11), 2039-2044. Latourelle, J. C., Sun, M., Lew, M. F., Suchowersky, O., Klein, C., Golbe, L. I., et al. (2008). The Gly2019Ser mutation in LRRK2 is not fully penetrant in familial Parkinson's disease: the GenePD study. BMC Med, 6, 32. Lee, M. J., Mata, I. F., Lin, C. H., Tzen, K. Y., Lincoln, S. J., Bounds, R., et al. (2009). Genotype-phenotype correlates in Taiwanese patients with early-onset recessive Parkinsonism. Mov Disord, 24(1), 104-108. Lesage, S., Anheim, M., Letournel, F., Bousset, L., Honore, A., Rozas, N., et al. (2013). G51D alpha-synuclein mutation causes a novel parkinsonian-pyramidal syndrome. Ann Neurol, 73(4), 459-471. Lesage, S., Anheim, M., Letournel, F., Bousset, L., Honore, A., Rozas, N., et al. (2013). G51D alpha-synuclein mutation causes a novel parkinsonian-pyramidal syndrome. Ann Neurol, 73(4), 459-471. Lesage, S., Condroyer, C., Klebe, S., Honore, A., Tison, F., Brefel-Courbon, C., et al. (2012). Identification of VPS35 mutations replicated in French families with Parkinson disease. Neurology, 78(18), 1449-1450. Lesage, S., Ibanez, P., Lohmann, E., Pollak, P., Tison, F., Tazir, M., et al. (2005). G2019S LRRK2 mutation in French and North African families with Parkinson's disease. Ann Neurol, 58(5), 784-787. Li, Y. Y., Chen, X. N., Fan, X. X., Zhang, Y. J., Gu, J., Fu, X. W., et al. (2015). Upregulated dynamin 1 in an acute seizure model and in epileptic patients. Synapse, 69(2), 67-77. 128  Lill, C. M., Hansen, J., Olsen, J. H., Binder, H., Ritz, B., & Bertram, L. (2015). Impact of Parkinson's disease risk loci on age at onset. Mov Disord, 30(6), 847-850. Lockhart, P. J., Bounds, R., Hulihan, M., Kachergus, J., Lincoln, S., Lin, C. H., et al. (2004). Lack of mutations in DJ-1 in a cohort of Taiwanese ethnic Chinese with early-onset parkinsonism. Mov Disord, 19(9), 1065-1069. Lockhart, P. J., Lincoln, S., Hulihan, M., Kachergus, J., Wilkes, K., Bisceglio, G., et al. (2004). DJ-1 mutations are a rare cause of recessively inherited early onset parkinsonism mediated by loss of protein function. J Med Genet, 41(3), e22. Luk, K. C., Kehm, V., Carroll, J., Zhang, B., O'Brien, P., Trojanowski, J. Q., et al. (2012). Pathological alpha-synuclein transmission initiates Parkinson-like neurodegeneration in nontransgenic mice. Science, 338(6109), 949-953. Luk, K. C., Kehm, V. M., Zhang, B., O'Brien, P., Trojanowski, J. Q., & Lee, V. M. (2012). Intracerebral inoculation of pathological alpha-synuclein initiates a rapidly progressive neurodegenerative alpha-synucleinopathy in mice. J Exp Med, 209(5), 975-986. MacLeod, D., Dowman, J., Hammond, R., Leete, T., Inoue, K., & Abeliovich, A. (2006). The familial Parkinsonism gene LRRK2 regulates neurite process morphology. Neuron, 52(4), 587-593. MacLeod, D. A., Rhinn, H., Kuwahara, T., Zolin, A., Di Paolo, G., McCabe, B. D., et al. (2013). RAB7L1 interacts with LRRK2 to modify intraneuronal protein sorting and Parkinson's disease risk. Neuron, 77(3), 425-439. Maraganore, D. M., Wilkes, K., Lesnick, T. G., Strain, K. J., de Andrade, M., Rocca, W. A., et al. (2004). A limited role for DJ1 in Parkinson disease susceptibility. Neurology, 63(3), 550-553. Marder, K., Wang, Y., Alcalay, R. N., Mejia-Santana, H., Tang, M. X., Lee, A., et al. (2015). Age-specific penetrance of LRRK2 G2019S in the Michael J. Fox Ashkenazi Jewish LRRK2 Consortium. Neurology, 85(1), 89-95. Marras, C., Alcalay, R. N., Caspell-Garcia, C., Coffey, C., Chan, P., Duda, J. E., et al. (2016). Motor and nonmotor heterogeneity of LRRK2-related and idiopathic Parkinson's disease. Mov Disord. Marti-Masso, J. F., Ruiz-Martinez, J., Bolano, M. J., Ruiz, I., Gorostidi, A., Moreno, F., et al. (2009). Neuropathology of Parkinson's disease with the R1441G mutation in LRRK2. Mov Disord, 24(13), 1998-2001. Mata, I. F., Alvarez, V., Coto, E., Blazquez, M., Guisasola, L. M., Salvador, C., et al. (2005). Homozygous partial genomic triplication of the parkin gene in early-onset parkinsonism. Neurosci Lett, 380(3), 257-259. Matta, S., Van Kolen, K., da Cunha, R., van den Bogaart, G., Mandemakers, W., Miskiewicz, K., et al. (2012). LRRK2 controls an EndoA phosphorylation cycle in synaptic endocytosis. Neuron, 75(6), 1008-1021. McRae, C., Diem, G., Vo, A., O'Brien, C., & Seeberger, L. (2000). Schwab & England: standardization of administration. Mov Disord, 15(2), 335-336. Mills, R. D., Mulhern, T. D., Liu, F., Culvenor, J. G., & Cheng, H. C. (2014). Prediction of the repeat domain structures and impact of parkinsonism-associated variations on structure and function of all functional domains of leucine-rich repeat kinase 2 (LRRK2). Hum Mutat, 35(4), 395-412. 129  Morgan, N. V., Westaway, S. K., Morton, J. E., Gregory, A., Gissen, P., Sonek, S., et al. (2006). PLA2G6, encoding a phospholipase A2, is mutated in neurodegenerative disorders with high brain iron. Nat Genet, 38(7), 752-754. Muenter, M. D., Forno, L. S., Hornykiewicz, O., Kish, S. J., Maraganore, D. M., Caselli, R. J., et al. (1998). Hereditary form of parkinsonism--dementia. Ann Neurol, 43(6), 768-781. Munsie, L. N., Milnerwood, A. J., Seibler, P., Beccano-Kelly, D. A., Tatarnikov, I., Khinda, J., et al. (2015). Retromer-dependent neurotransmitter receptor trafficking to synapses is altered by the Parkinson's disease VPS35 mutation p.D620N. Hum Mol Genet, 24(6), 1691-1703. Nalls, M. A., Pankratz, N., Lill, C. M., Do, C. B., Hernandez, D. G., Saad, M., et al. (2014). Large-scale meta-analysis of genome-wide association data identifies six new risk loci for Parkinson's disease. Nat Genet, 46(9), 989-993. Nalls, M. A., Pankratz, N., Lill, C. M., Do, C. B., Hernandez, D. G., Saad, M., et al. (2014). Large-scale meta-analysis of genome-wide association data identifies six new risk loci for Parkinson's disease. Nat Genet, 46(9), 989-993. Nebel, A., Filon, D., Weiss, D. A., Weale, M., Faerman, M., Oppenheim, A., et al. (2000). High-resolution Y chromosome haplotypes of Israeli and Palestinian Arabs reveal geographic substructure and substantial overlap with haplotypes of Jews. Hum Genet, 107(6), 630-641. Nichols, N., Bras, J. M., Hernandez, D. G., Jansen, I. E., Lesage, S., Lubbe, S., et al. (2015). EIF4G1 mutations do not cause Parkinson's disease. Neurobiol Aging, 36(8), 2444 e2441-2444. Nichols, N., Bras, J. M., Hernandez, D. G., Jansen, I. E., Lesage, S., Lubbe, S., et al. (2015). EIF4G1 mutations do not cause Parkinson's disease. Neurobiol Aging, 36(8), 2444 e2441-2444. Nichols, R. J., Dzamko, N., Hutti, J. E., Cantley, L. C., Deak, M., Moran, J., et al. (2009). Substrate specificity and inhibitors of LRRK2, a protein kinase mutated in Parkinson's disease. Biochem J, 424(1), 47-60. Nishioka, K., Kefi, M., Jasinska-Myga, B., Wider, C., Vilarino-Guell, C., Ross, O. A., et al. (2010). A comparative study of LRRK2, PINK1 and genetically undefined familial Parkinson's disease. J Neurol Neurosurg Psychiatry, 81(4), 391-395. Nishioka, K., Ross, O. A., Ishii, K., Kachergus, J. M., Ishiwata, K., Kitagawa, M., et al. (2009). Expanding the clinical phenotype of SNCA duplication carriers. Mov Disord, 24(12), 1811-1819. Nishioka, K., Wider, C., Vilarino-Guell, C., Soto-Ortolaza, A. I., Lincoln, S. J., Kachergus, J. M., et al. (2010). Association of alpha-, beta-, and gamma-Synuclein with diffuse lewy body disease. Arch Neurol, 67(8), 970-975. Nuytemans, K., Bademci, G., Inchausti, V., Dressen, A., Kinnamon, D. D., Mehta, A., et al. (2013). Whole exome sequencing of rare variants in EIF4G1 and VPS35 in Parkinson disease. Neurology, 80(11), 982-989. Ogaki, K., Koga, S., Heckman, M. G., Fiesel, F. C., Ando, M., Labbe, C., et al. (2015). Mitochondrial targeting sequence variants of the CHCHD2 gene are a risk for Lewy body disorders. Neurology, 85(23), 2016-2025. Olgiati, S., De Rosa, A., Quadri, M., Criscuolo, C., Breedveld, G. J., Picillo, M., et al. (2014). PARK20 caused by SYNJ1 homozygous Arg258Gln mutation in a new Italian family. Neurogenetics, 15(3), 183-188. 130  Oosterveld, L. P., Allen, J. C., Jr., Ng, E. Y., Seah, S. H., Tay, K. Y., Au, W. L., et al. (2015). Greater motor progression in patients with Parkinson disease who carry LRRK2 risk variants. Neurology, 85(12), 1039-1042. Orenstein, S. J., Kuo, S. H., Tasset, I., Arias, E., Koga, H., Fernandez-Carasa, I., et al. (2013). Interplay of LRRK2 with chaperone-mediated autophagy. Nat Neurosci, 16(4), 394-406. Ozelius, L. J., Senthil, G., Saunders-Pullman, R., Ohmann, E., Deligtisch, A., Tagliati, M., et al. (2006). LRRK2 G2019S as a cause of Parkinson's disease in Ashkenazi Jews. N Engl J Med, 354(4), 424-425. Paisan-Ruiz, C., Lang, A. E., Kawarai, T., Sato, C., Salehi-Rad, S., Fisman, G. K., et al. (2005). LRRK2 gene in Parkinson disease: mutation analysis and case control association study. Neurology, 65(5), 696-700. Paisan-Ruiz, C., Saenz, A., Lopez de Munain, A., Marti, I., Martinez Gil, A., Marti-Masso, J. F., et al. (2005). Familial Parkinson's disease: clinical and genetic analysis of four Basque families. Ann Neurol, 57(3), 365-372. Paisan-Ruiz, C., Washecka, N., Nath, P., Singleton, A. B., & Corder, E. H. (2009). Parkinson's disease and low frequency alleles found together throughout LRRK2. Ann Hum Genet, 73(Pt 4), 391-403. Papapetropoulos, S., Scaravilli, T., Morris, H., An, S. F., Henderson, D. C., Quinn, N. P., et al. (2005). Young onset limb spasticity with PSP-like brain and spinal cord NFT-tau pathology. Neurology, 64(4), 731-733. Parisiadou, L., Xie, C., Cho, H. J., Lin, X., Gu, X. L., Long, C. X., et al. (2009). Phosphorylation of ezrin/radixin/moesin proteins by LRRK2 promotes the rearrangement of actin cytoskeleton in neuronal morphogenesis. J Neurosci, 29(44), 13971-13980. Pchelina, S. N., Ivanova, O. N., Emel'ianov, A. K., & Iakimovskii, A. F. (2011). [Clinical features of LRRK2-associated Parkinson's disease]. Zh Nevrol Psikhiatr Im S S Korsakova, 111(12), 56-62. Piccoli, G., Onofri, F., Cirnaru, M. D., Kaiser, C. J., Jagtap, P., Kastenmuller, A., et al. (2014). Leucine-rich repeat kinase 2 binds to neuronal vesicles through protein interactions mediated by its C-terminal WD40 domain. Mol Cell Biol, 34(12), 2147-2161. Pilsl, A., & Winklhofer, K. F. (2012). Parkin, PINK1 and mitochondrial integrity: emerging concepts of mitochondrial dysfunction in Parkinson's disease. Acta Neuropathol, 123(2), 173-188. Polymeropoulos, M. H., Lavedan, C., Leroy, E., Ide, S. E., Dehejia, A., Dutra, A., et al. (1997). Mutation in the alpha-synuclein gene identified in families with Parkinson's disease. Science, 276(5321), 2045-2047. Postuma, R. B., Berg, D., Stern, M., Poewe, W., Olanow, C. W., Oertel, W., et al. (2015). MDS clinical diagnostic criteria for Parkinson's disease. Mov Disord, 30(12), 1591-1601. Proukakis, C., Dudzik, C. G., Brier, T., MacKay, D. S., Cooper, J. M., Millhauser, G. L., et al. (2013). A novel alpha-synuclein missense mutation in Parkinson disease. Neurology, 80(11), 1062-1064. Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M. A., Bender, D., et al. (2007). PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet, 81(3), 559-575. Quadri, M., Fang, M., Picillo, M., Olgiati, S., Breedveld, G. J., Graafland, J., et al. (2013). Mutation in the SYNJ1 gene associated with autosomal recessive, early-onset Parkinsonism. Hum Mutat, 34(9), 1208-1215. 131  Rademakers, R., Cruts, M., & van Broeckhoven, C. (2004). The role of tau (MAPT) in frontotemporal dementia and related tauopathies. Hum Mutat, 24(4), 277-295. Rademakers, R., Dermaut, B., Peeters, K., Cruts, M., Heutink, P., Goate, A., et al. (2003). Tau (MAPT) mutation Arg406Trp presenting clinically with Alzheimer disease does not share a common founder in Western Europe. Hum Mutat, 22(5), 409-411. Raimondi, A., Ferguson, S. M., Lou, X., Armbruster, M., Paradise, S., Giovedi, S., et al. (2011). Overlapping role of dynamin isoforms in synaptic vesicle endocytosis. Neuron, 70(6), 1100-1114. Rajput, A., Ross, J. P., Bernales, C. Q., Rayaprolu, S., Soto-Ortolaza, A. I., Ross, O. A., et al. (2015). VPS35 and DNAJC13 disease-causing variants in essential tremor. Eur J Hum Genet, 23(6), 887-888. Rakovic, A., Grunewald, A., Kottwitz, J., Bruggemann, N., Pramstaller, P. P., Lohmann, K., et al. (2011). Mutations in PINK1 and Parkin impair ubiquitination of Mitofusins in human fibroblasts. PLoS One, 6(3), e16746. Ramirez, A., Heimbach, A., Grundemann, J., Stiller, B., Hampshire, D., Cid, L. P., et al. (2006). Hereditary parkinsonism with dementia is caused by mutations in ATP13A2, encoding a lysosomal type 5 P-type ATPase. Nat Genet, 38(10), 1184-1191. Reider, C. R., Halter, C. A., Castelluccio, P. F., Oakes, D., Nichols, W. C., & Foroud, T. (2003). Reliability of reported age at onset for Parkinson's disease. Mov Disord, 18(3), 275-279. Reubold, T. F., Faelber, K., Plattner, N., Posor, Y., Ketel, K., Curth, U., et al. (2015). Crystal structure of the dynamin tetramer. Nature, 525(7569), 404-408. Risch, N., & Merikangas, K. (1996). The future of genetic studies of complex human diseases. Science, 273(5281), 1516-1517. Ritz, B., Rhodes, S. L., Bordelon, Y., & Bronstein, J. (2012). alpha-Synuclein genetic variants predict faster motor symptom progression in idiopathic Parkinson disease. PLoS One, 7(5), e36199. Roks, G., Dermaut, B., Heutink, P., Julliams, A., Backhovens, H., Van de Broeck, M., et al. (1999). Mutation screening of the tau gene in patients with early-onset Alzheimer's disease. Neurosci Lett, 277(2), 137-139. Ross, O. A., Soto-Ortolaza, A. I., Heckman, M. G., Aasly, J. O., Abahuni, N., Annesi, G., et al. (2011). Association of LRRK2 exonic variants with susceptibility to Parkinson's disease: a case-control study. Lancet Neurol, 10(10), 898-908. Ross, O. A., Toft, M., Whittle, A. J., Johnson, J. L., Papapetropoulos, S., Mash, D. C., et al. (2006). Lrrk2 and Lewy body disease. Ann Neurol, 59(2), 388-393. Ruiz-Martinez, J., Gorostidi, A., Ibanez, B., Alzualde, A., Otaegui, D., Moreno, F., et al. (2010). Penetrance in Parkinson's disease related to the LRRK2 R1441G mutation in the Basque country (Spain). Mov Disord, 25(14), 2340-2345. Samaranch, L., Lorenzo-Betancor, O., Arbelo, J. M., Ferrer, I., Lorenzo, E., Irigoyen, J., et al. (2010). PINK1-linked parkinsonism is associated with Lewy body pathology. Brain, 133(Pt 4), 1128-1142. Satake, W., Nakabayashi, Y., Mizuta, I., Hirota, Y., Ito, C., Kubo, M., et al. (2009). Genome-wide association study identifies common variants at four loci as genetic risk factors for Parkinson's disease. Nat Genet, 41(12), 1303-1307. Saunders-Pullman, R., Alcalay, R. N., Mirelman, A., Wang, C., Luciano, M. S., Ortega, R. A., et al. (2015). REM sleep behavior disorder, as assessed by questionnaire, in G2019S LRRK2 mutation PD and carriers. Mov Disord, 30(13), 1834-1839. 132  Saunders-Pullman, R., Mirelman, A., Wang, C., Alcalay, R. N., San Luciano, M., Ortega, R., et al. (2014). Olfactory identification in LRRK2 G2019S mutation carriers: a relevant marker? Ann Clin Transl Neurol, 1(9), 670-678. Schapira, A. H. (2015). Glucocerebrosidase and Parkinson disease: Recent advances. Mol Cell Neurosci, 66(Pt A), 37-42. Schrag, A., & Schott, J. M. (2006). Epidemiological, clinical, and genetic characteristics of early-onset parkinsonism. Lancet Neurol, 5(4), 355-363. Schreij, A. M., Chaineau, M., Ruan, W., Lin, S., Barker, P. A., Fon, E. A., et al. (2015). LRRK2 localizes to endosomes and interacts with clathrin-light chains to limit Rac1 activation. EMBO Rep, 16(1), 79-86. Sharma, M., Ioannidis, J. P., Aasly, J. O., Annesi, G., Brice, A., Bertram, L., et al. (2012). A multi-centre clinico-genetic analysis of the VPS35 gene in Parkinson disease indicates reduced penetrance for disease-associated variants. J Med Genet, 49(11), 721-726. Sheerin, U. M., Charlesworth, G., Bras, J., Guerreiro, R., Bhatia, K., Foltynie, T., et al. (2012). Screening for VPS35 mutations in Parkinson's disease. Neurobiol Aging, 33(4), 838 e831-835. Sierra, M., Gonzalez-Aramburu, I., Sanchez-Juan, P., Sanchez-Quintana, C., Polo, J. M., Berciano, J., et al. (2011). High frequency and reduced penetrance of LRRK2 G2019S mutation among Parkinson's disease patients in Cantabria (Spain). Mov Disord, 26(13), 2343-2346. Siitonen, A., Majounie, E., Federoff, M., Ding, J., Majamaa, K., & Singleton, A. B. (2013). Mutations in EIF4G1 are not a common cause of Parkinson's disease. Eur J Neurol, 20(4), e59. Simon-Sanchez, J., & Gasser, T. (2015). Parkinson disease GWAS: the question of lumping or splitting is back again. Neurology, 84(10), 966-967. Simon-Sanchez, J., van Hilten, J. J., van de Warrenburg, B., Post, B., Berendse, H. W., Arepalli, S., et al. (2011). Genome-wide association study confirms extant PD risk loci among the Dutch. Eur J Hum Genet, 19(6), 655-661. Singleton, A. B., Farrer, M., Johnson, J., Singleton, A., Hague, S., Kachergus, J., et al. (2003). alpha-Synuclein locus triplication causes Parkinson's disease. Science, 302(5646), 841. Skipper, L., Wilkes, K., Toft, M., Baker, M., Lincoln, S., Hulihan, M., et al. (2004). Linkage disequilibrium and association of MAPT H1 in Parkinson disease. Am J Hum Genet, 75(4), 669-677. Small, S. A., Kent, K., Pierce, A., Leung, C., Kang, M. S., Okada, H., et al. (2005). Model-guided microarray implicates the retromer complex in Alzheimer's disease. Ann Neurol, 58(6), 909-919. Soldner, F., Stelzer, Y., Shivalila, C. S., Abraham, B. J., Latourelle, J. C., Barrasa, M. I., et al. (2016). Parkinson-associated risk variant in distal enhancer of alpha-synuclein modulates target gene expression. Nature. Spillantini, M. G., & Goedert, M. (2001). Tau and Parkinson disease. JAMA, 286(18), 2324-2326. Spillantini, M. G., Schmidt, M. L., Lee, V. M., Trojanowski, J. Q., Jakes, R., & Goedert, M. (1997). Alpha-synuclein in Lewy bodies. Nature, 388(6645), 839-840. Stafa, K., Tsika, E., Moser, R., Musso, A., Glauser, L., Jones, A., et al. (2014). Functional interaction of Parkinson's disease-associated LRRK2 with members of the dynamin GTPase superfamily. Hum Mol Genet, 23(8), 2055-2077. 133  Steger, M., Tonelli, F., Ito, G., Davies, P., Trost, M., Vetter, M., et al. (2016). Phosphoproteomics reveals that Parkinson's disease kinase LRRK2 regulates a subset of Rab GTPases. Elife, 5. Steinberg, F., Gallon, M., Winfield, M., Thomas, E. C., Bell, A. J., Heesom, K. J., et al. (2013). A global analysis of SNX27-retromer assembly and cargo specificity reveals a function in glucose and metal ion transport. Nat Cell Biol, 15(5), 461-471. Tan, L. C., Tanner, C. M., Chen, R., Chan, P., Farrer, M., Hardy, J., et al. (2003). Marked variation in clinical presentation and age of onset in a family with a heterozygous parkin mutation. Mov Disord, 18(7), 758-763. Tian, Y., Li, B., Shi, W. Z., Chang, M. Z., Zhang, G. J., Di, Z. L., et al. (2014). Dynamin-related protein 1 inhibitors protect against ischemic toxicity through attenuating mitochondrial Ca2+ uptake from endoplasmic reticulum store in PC12 cells. Int J Mol Sci, 15(2), 3172-3185. Tijero, B., Gomez Esteban, J. C., Somme, J., Llorens, V., Lezcano, E., Martinez, A., et al. (2013). Autonomic dysfunction in parkinsonian LRRK2 mutation carriers. Parkinsonism Relat Disord, 19(10), 906-909. Toft, M., Myhre, R., Pielsticker, L., White, L. R., Aasly, J. O., & Farrer, M. J. (2007). PINK1 mutation heterozygosity and the risk of Parkinson's disease. J Neurol Neurosurg Psychiatry, 78(1), 82-84. Tomiyama, H., Li, Y., Funayama, M., Hasegawa, K., Yoshino, H., Kubo, S., et al. (2006). Clinicogenetic study of mutations in LRRK2 exon 41 in Parkinson's disease patients from 18 countries. Mov Disord, 21(8), 1102-1108. Trinh, J., Amouri, R., Duda, J. E., Morley, J. F., Read, M., Donald, A., et al. (2014). Comparative study of Parkinson's disease and leucine-rich repeat kinase 2 p.G2019S parkinsonism. Neurobiol Aging, 35(5), 1125-1131. Trinh, J., & Farrer, M. (2013). Advances in the genetics of Parkinson disease. Nat Rev Neurol, 9(8), 445-454. Trinh, J., Farrer, M., Ross, O. A., & Guella, I. (1993). LRRK2-Related Parkinson Disease. In R. A. Pagon, M. P. Adam, H. H. Ardinger, S. E. Wallace, A. Amemiya, L. J. H. Bean, T. D. Bird, C. T. Fong, H. C. Mefford, R. J. H. Smith & K. Stephens (Eds.), GeneReviews(R). Seattle (WA). Trinh, J., Guella, I., & Farrer, M. J. (2014). Disease penetrance of late-onset parkinsonism: a meta-analysis. JAMA Neurol, 71(12), 1535-1539. Trinh, J., Gustavsson, E. K., Guella, I., Vilarino-Guell, C., Evans, D., Encarnacion, M., et al. (2014). The role of SNCA and MAPT in Parkinson disease and LRRK2 parkinsonism in the Tunisian Arab-Berber population. Eur J Neurol, 21(11), e91-92. Trinh, J., Vilarino-Guell, C., & Ross, O. A. (2015). A commentary on fine mapping and resequencing of the PARK16 locus in Parkinson's disease. J Hum Genet, 60(8), 405-406. Trojano, L., Moretta, P., Estraneo, A., & Santoro, L. (2010). Neuropsychologic assessment and cognitive rehabilitation in a patient with locked-in syndrome and left neglect. Arch Phys Med Rehabil, 91(3), 498-502. Tsuboi, Y., Dickson, D. W., Nabeshima, K., Schmeichel, A. M., Wszolek, Z. K., Yamada, T., et al. (2008). Neurodegeneration involving putative respiratory neurons in Perry syndrome. Acta Neuropathol, 115(2), 263-268. Ujiie, S., Hatano, T., Kubo, S., Imai, S., Sato, S., Uchihara, T., et al. (2012). LRRK2 I2020T mutation is associated with tau pathology. Parkinsonism Relat Disord, 18(7), 819-823. 134  Valente, E. M., Abou-Sleiman, P. M., Caputo, V., Muqit, M. M., Harvey, K., Gispert, S., et al. (2004). Hereditary early-onset Parkinson's disease caused by mutations in PINK1. Science, 304(5674), 1158-1160. Valente, E. M., Salvi, S., Ialongo, T., Marongiu, R., Elia, A. E., Caputo, V., et al. (2004). PINK1 mutations are associated with sporadic early-onset parkinsonism. Ann Neurol, 56(3), 336-341. Vargas, K. J., Makani, S., Davis, T., Westphal, C. H., Castillo, P. E., & Chandra, S. S. (2014). Synucleins regulate the kinetics of synaptic vesicle endocytosis. J Neurosci, 34(28), 9364-9376. Vilarino-Guell, C., Rajput, A., Milnerwood, A. J., Shah, B., Szu-Tu, C., Trinh, J., et al. (2014). DNAJC13 mutations in Parkinson disease. Hum Mol Genet, 23(7), 1794-1801. Vilarino-Guell, C., Ross, O. A., Aasly, J. O., White, L. R., Rajput, A., Rajput, A. H., et al. (2010). An independent replication of PARK16 in Asian samples. Neurology, 75(24), 2248-2249. Vilarino-Guell, C., Wider, C., Ross, O. A., Dachsel, J. C., Kachergus, J. M., Lincoln, S. J., et al. (2011). VPS35 mutations in Parkinson disease. Am J Hum Genet, 89(1), 162-167. Visser, M., Marinus, J., Stiggelbout, A. M., & Van Hilten, J. J. (2004). Assessment of autonomic dysfunction in Parkinson's disease: the SCOPA-AUT. Mov Disord, 19(11), 1306-1312. Visser, M., Marinus, J., van Hilten, J. J., Schipper, R. G., & Stiggelbout, A. M. (2004). Assessing comorbidity in patients with Parkinson's disease. Mov Disord, 19(7), 824-828. Vos, M., Verstreken, P., & Klein, C. (2015). Stimulation of electron transport as potential novel therapy in Parkinson's disease with mitochondrial dysfunction. Biochem Soc Trans, 43(2), 275-279. Vrabel, M. (2015). Preferred Reporting Items for Systematic Reviews and Meta-Analyses. Oncol Nurs Forum, 42(5), 552-554. Wacholder, S., Hartge, P., Struewing, J. P., Pee, D., McAdams, M., Brody, L., et al. (1998). The kin-cohort study for estimating penetrance. Am J Epidemiol, 148(7), 623-630. Wang, Y., Shi, M., Chung, K. A., Zabetian, C. P., Leverenz, J. B., Berg, D., et al. (2012). Phosphorylated alpha-synuclein in Parkinson's disease. Sci Transl Med, 4(121), 121ra120. Waschbusch, D., Michels, H., Strassheim, S., Ossendorf, E., Kessler, D., Gloeckner, C. J., et al. (2014). LRRK2 transport is regulated by its novel interacting partner Rab32. PLoS One, 9(10), e111632. Waters, C. H., Faust, P. L., Powers, J., Vinters, H., Moskowitz, C., Nygaard, T., et al. (1993). Neuropathology of lubag (x-linked dystonia parkinsonism). Mov Disord, 8(3), 387-390. Wen, L., Tang, F. L., Hong, Y., Luo, S. W., Wang, C. L., He, W., et al. (2011). VPS35 haploinsufficiency increases Alzheimer's disease neuropathology. J Cell Biol, 195(5), 765-779. West, A. B., Lockhart, P. J., O'Farell, C., & Farrer, M. J. (2003). Identification of a novel gene linked to parkin via a bi-directional promoter. J Mol Biol, 326(1), 11-19. Wilson, G. R., Sim, J. C., McLean, C., Giannandrea, M., Galea, C. A., Riseley, J. R., et al. (2014). Mutations in RAB39B cause X-linked intellectual disability and early-onset Parkinson disease with alpha-synuclein pathology. Am J Hum Genet, 95(6), 729-735. Wright, F. A., Strug, L. J., Doshi, V. K., Commander, C. W., Blackman, S. M., Sun, L., et al. (2011). Genome-wide association and linkage identify modifier loci of lung disease severity in cystic fibrosis at 11p13 and 20q13.2. Nat Genet, 43(6), 539-546. 135  Wu, Y., O'Toole, E. T., Girard, M., Ritter, B., Messa, M., Liu, X., et al. (2014). A dynamin 1-, dynamin 3- and clathrin-independent pathway of synaptic vesicle recycling mediated by bulk endocytosis. Elife, 3, e01621. Yesavage, J. A., Brink, T. L., Rose, T. L., Lum, O., Huang, V., Adey, M., et al. (1982). Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res, 17(1), 37-49. Yu, C., Nwabuisi-Heath, E., Laxton, K., & Ladu, M. J. (2010). Endocytic pathways mediating oligomeric Abeta42 neurotoxicity. Mol Neurodegener, 5, 19. Zarranz, J. J., Alegre, J., Gomez-Esteban, J. C., Lezcano, E., Ros, R., Ampuero, I., et al. (2004). The new mutation, E46K, of alpha-synuclein causes Parkinson and Lewy body dementia. Ann Neurol, 55(2), 164-173. Zimprich, A., Benet-Pages, A., Struhal, W., Graf, E., Eck, S. H., Offman, M. N., et al. (2011). A mutation in VPS35, encoding a subunit of the retromer complex, causes late-onset Parkinson disease. Am J Hum Genet, 89(1), 168-175. Zimprich, A., Biskup, S., Leitner, P., Lichtner, P., Farrer, M., Lincoln, S., et al. (2004). Mutations in LRRK2 cause autosomal-dominant parkinsonism with pleomorphic pathology. Neuron, 44(4), 601-607. Zody, M. C., Jiang, Z., Fung, H. C., Antonacci, F., Hillier, L. W., Cardone, M. F., et al. (2008). Evolutionary toggling of the MAPT 17q21.31 inversion region. Nat Genet, 40(9), 1076-1083.   


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items