Advances in the Genetics of Parkinson's Disease: A Guide for the Clinician

Authors

  • Una-Marie Sheerin MRCP,

    1. Department of Molecular Neuroscience, UCL Institute of Neurology, University College London, London, United Kingdom
    Search for more papers by this author
  • Henry Houlden PhD, FRCP,

    1. Department of Molecular Neuroscience, UCL Institute of Neurology, University College London, London, United Kingdom
    Search for more papers by this author
  • Nicholas W. Wood PhD ,FRCP

    Corresponding author
    1. UCL Department of Molecular Neuroscience and UCL Genetics Institute, University College London, London, United Kingdom
    • Correspondence to: Prof. Nicholas W. Wood, Department of Molecular Neuroscience (Box 12), UCL Institute of Neurology, Queen Square, London WC1N 3BG, United Kingdom; E-mail: n.wood@ucl.ac.uk

    Search for more papers by this author

Abstract

Over the last 16 years, insights in clinical and genetic characteristics of Parkinson's disease (PD) have increased substantially. We summarize the clinical, genetic, and pathological findings of autosomal dominant PD linked to mutations in SNCA, leucine-rich repeat kinase 2, vacuolar protein sorting-35, and eukaryotic translation initiation factor 4 gamma 1 and autosomal recessive PD linked to parkin, PINK1, and DJ-1, as well as autosomal recessive complicated parkinsonian syndromes caused by mutations in ATP13A2, FBXO7, PLA2G6, SYNJ1, and DNAJC6. We also review the advances in high- and low-risk genetic susceptibility factors and present multisystem disorders that may present with parkinsonism as the major clinical feature and provide recommendations for prioritization of genetic testing. Finally, we consider the challenges of future genetic research in PD.

Over the last 16 years, our understanding of Parkinson's disease (PD) has undergone a set change mainly driven by genetic research, from a disease with little or no genetic predisposition to a disease with a subset of patients with Mendelian forms of the disease. More recently, it has become clear that genetic predisposition also plays a role in patients with the “sporadic” form of the disease.

The term parkinsonism defines the combination of two or more of four cardinal motor signs: bradykinesia; resting tremor; muscular rigidity; and postural instability.[1-3] PD is the most common cause of parkinsonism. The modified Queen Square Brain Bank criteria are the most frequently used diagnostic criteria, which rely on the presence of three cardinal signs (bradykinesia, rigidity, and rest tremor), responsiveness to dopaminergic therapy, and the absence of exclusion criteria, which, if present, usually mean the diagnosis is that of Parkinson-plus syndrome or parkinsonism.[4] The pathological hallmark of classical PD comprises of loss of nigrostriatal dopaminergic neurons in the substantia nigra pars compacta SNc and alpha-synuclein (α-Syn) containing Lewy bodies (LBs) in the surviving neurons,[5] although we will see, in the course of this review, that some of the monogenic forms of PD lack this typical LB pathology. PD is the second-most common neurodegenerative brain disorder demonstrating age-related prevalence, with 1% of the population affected over the age of 60 years and ~4% over the age of 85.[6] Epidemiological studies have revealed that smoking and caffeinated-coffee consumption are associated with a reduced risk of developing PD, whereas pesticide exposure increases the risk.[7]

PD occurs as a sporadic disorder the vast majority of patients. However, in approximately 5% to 10% of cases, the disease occurs as a Mendelian disorder. As genes were identified to cause monogenic forms of PD, they were assigned PARK loci and numbered in chronological order of their identification. However, the PARK loci not only contain genes that cause monogenic forms of PD, but also loci identified from genome-wide linkage screens, which may not have been replicated in subsequent studies (Table 1).

Table 1. PARK-designated PD-related loci
LocusChromosomeGeneInheritancePhenotypeAdditional Information
  1. a

    PARK4 is an erroneous locus; the family was subsequently proven to have an SNCA triplication (i.e., PARK1).

  2. AD, autosomal dominant; AR, autosomal recessive

PARK1a4q21 SNCA [8, 9] ADEOPDMissense mutations, genomic multiplications
PARK26q25.2-q27 Parkin [10] AREOPD 
PARK32p13Unknown[11]ADClassical PDUnconfirmed
PARK54p13 UCLH1 [12] ?Classical PDUnreplicated mutations in a single sibling pair
PARK61p36 PINK1 [13] AREOPD 
PARK71p36 DJ-1 [14] AREOPD 
PARK812q12 LRRK2 [15, 16] ADClassical PD 
PARK91p36 ATP13A2 [17] ARJuvenile parkinsonism, pyramidal signs, dementia 
PARK101p32Unknown[18]Risk FactorLate-onset parkinsonismUnconfirmed
PARK112q37Unknown[19]Risk FactorLate-onset parkinsonism 
PARK12Xq21-25Unknown[20]X-linkedLate-onset parkinsonismUnconfirmed
PARK132p12 HTRA2 [21] AD or risk factorLate-onset parkinsonismUnconfirmed
PARK1422q12-q13 PLA2G6 [22] AREarly-onset parkinsonism-dystonia 
PARK1522q12-q13 FBXO7 [23] ARJuvenile parkinsonism and pyramidal signs 
PARK161q32Unknown[24, 25]Risk factorLate-onset parkinsonismUnconfirmed
PARK1716q11.2 VPS35 [26, 27] ADClassical PD 
PARK183q27.1 EIF4G1 [28] ADClassical PDUnconfirmed

Autosomal Dominant Genes

α-Syn, SNCA (PARK1)

Identification of point mutations in α-Syn (SNCA) as a cause of familial PD was the first evidence for a genetic etiology in a subset of PD cases.[8] Later, whole-locus multiplications of SNCA were identified.[9] Both are rare: Duplications are detected in approximately 1% to 2% of patients with autosomal dominant PD.[29] Point mutations and triplications are very rare; the A53T mutation has only been found in a few families, predominately of Greek ancestry.[30-37] The A30P and E46K mutations have been detected in a single family of German and Spanish ancestry, respectively.[38, 39]

Patients with SNCA missense mutations usually have early-onset PD (EOPD; age of onset <45 years), although presentation with diffuse LB disease is recognized in some cases.[39, 40] Most patients have a good initial response to levodopa. However, the disease progresses rapidly, often with dementia as a clinical feature. Atypical features, such as central hypoventilation and myoclonus, are occasionally reported.[37] Penetrance of the missense mutations appears to be high and estimated at 85% for the A53T mutation.[8] Intriguingly, a dose relationship between α-Syn levels and severity of the clinical phenotype is observed in the SNCA multiplications families. Patients with SNCA duplications (three copies of SNCA) bear a closer clinical resemblance to idiopathic PD (iPD) patients (onset 50–60 years) with reduced penetrance (as low as 33% in one family),[41] whereas triplications (four copies of SNCA) are fully penetrant, with onset a decade earlier than duplication carriers (40–50 years) and a rapidly progressive clinical course associated with dementia, autonomic dysfunction, and psychiatric features.

Recently, two additional SNCA point mutations (H50Q and G51D) have been identified. The H50Q mutation was reported in 2 British patients; both had l-dopa-responsive, late-onset PD with cognitive impairment. In 1 case, there was no family history of the disease.[42, 43] The G51D SNCA mutation has been reported in association with a parkinsonian-pyramidal phenotype, with early-onset, moderate response to l-dopa and rapid progression leading to death within a few years of onset.[44]

Brain pathology in patients with SNCA mutations is characterized by abundant α-Syn-positive neuronal inclusions (LBs and Lewy neurites). Soon after the identification of SNCA mutations in familial PD cases came the discovery that α-Syn is the major component of LBs, the pathological hallmark of PD, therefore eloquently linking familial and sporadic forms of the disease. The A53T, A30P, and E46K SNCA mutations have been shown to cause a toxic gain of function, increasing the propensity of α-Syn to form β-pleated sheets, which, in turn, are thought to increase its tendency to form the aggregates that are segregated in LBs[45] G51D α-Syn oligomerizes more slowly and its fibrils are more toxic than those of wild-type (WT) protein,[44] whereas triplications and duplications result in a 100% and 50% overexpression of WT α-Syn.[46]

Leucine-Rich Repeat Kinase 2 (PARK8)

Mutations in the leucine-rich repeat kinase 2 (LRRK2) gene are the most frequent known cause of autosomal dominant PD. More than 100 LRRK2 variants have been reported in association with PD (http://www.molgen.vib-ua.be/PDMutDB)[47, 48]; however, of these, only seven are unequivocally pathogenic based on cosegregation with disease in families and absence in controls: R1441G,[15] R1441C,[16] R1441H,[49] Y1699C,[15, 16] G2019S,[50, 51] I2020T,[16] and N1437H.[52] The clinical presentation of most patients with LRRK2 mutations is that of late-onset asymmetric l-dopa-responsive tremor-predominant PD, which is clinically indistinguishable from iPD.[53, 54] However, early onset and atypical features, such as autonomic dysfunction, motor neuron disease, and choreoathetosis, and pyramidal tract signs are reported.[16, 55-57] LRRK2 mutations are found in ~10% of autosomal dominant PD kindreds.[53] The most frequent LRRK2 mutation is G2019S, accounting for ~2% of sporadic and ~5% of familial PD cases in Northern European and North American populations.[51, 58] Prevalence of the G2019S mutation is enriched in certain populations, accounting for ~20% of PD patients of Ashkenazi Jewish ancestry[59, 60] and ~40% of North African Berber Arab PD patients[61, 62] it is very rare in Asia[63] and high in Southern European countries.[55, 64] An incomplete, age-related penetrance (ranging from approximately 30% to 70% by the age of 80 years in different studies) has been estimated for the G2019S mutation,[54, 65-67] explaining those patients without a family history of the disease. Of the remaining pathogenic LRRK2 mutations, R1441C is the second-most common and reported in diverse populations.[68-71] An increased prevalence of the R1441G mutation is observed in PD patients with Basque and Spanish ancestry, but is rare elsewhere.[15, 72] Patients with LRRK2 mutations show pleiomorphic pathologies that overlap with other neurodegenerative diseases. The most common pathology is LBs with dopaminergic cell loss and gliosis in the SN. However, a minority of cases demonstrate nigral loss without LBs,[16, 73-76] whereas other cases have been reported with neurofibrillary tangle pathology,[77] glial cytoplasmic inclusions similar to those observed in MSA,[78] and some with anterior horn cell degeneration associated with axonal spheroids.[16]

Several pathogenic LRRK2 mutations cluster in functionally important domains and have been shown to alter their function: R1441C, R1441G, and R1441H reduce protein GTPase activity, whereas the G2019S and I2020T mutations in the kinase domain have been shown to increase its activity (as reviewed previously[79]). LRRK2 kinase inhibitors are in preclinical development for potential use in treating patients with mutations in the kinase domain. A variety of functions and mechanisms have been ascribed to LRRK2, including neurite outgrowth,[80] protein translation through regulation of microRNA processing,[81] and vesicle storage and mobilization within the recycling pool.[82] The physiological and pathological functions of LRRK2 have not yet been fully characterized.

Vacuolar Protein Sorting-35 (PARK17)

Mutations in vacuolar protein sorting-35 (VPS35) as a cause of familial PD were identified by exome sequencing, demonstrating the utility of this technique.[26, 27] To date, only one pathogenic mutation has been identified (D620N) and it accounts for ~1% of familial PD cases, across several populations.[83-86] The clinical phenotype associated with VPS35 mutations closely resembles that of iPD. Reduced penetrance has been observed.[85, 86] Dementia, although reported, is not a prominent feature of presentation; age of onset varies, with some patients presenting with EOPD.[83, 85, 86] Thus far, very little is known about the specific role of VPS35 in PD, except that it is a highly conserved component of the retromer, a heteromeric complex that mediates retrograde transport of the transmembrane cargo from endosomes back to the trans-Golgi network.[87, 88] The pathology associated with mutations in VPS35 has not been reported on.

Eukaryotic Translation Initiation Factor 4 Gamma 1

Mutations in eukaryotic translation initiation factor 4 gamma 1 (EIF4G1) were recently proposed as the cause of autosomal dominant PD in a kindred with LB pathology.[28] However, subsequent studies have not been able to convincingly replicate this finding and have identified reportedly pathogenic variants in controls.[89-93]

Dynactin 1

Mutations in dynactin 1 (DCTN1) cause Perry syndrome, an autosomal dominant, rapidly progressive form of parkinsonism associated with depression, weight loss, and hypoventilation. Dynactin is a multisubunit protein complex that is required for most long-distance trafficking of protein complexes and membrane organelles. All reported mutations cluster in exon 2, the P150glued highly conserved N-terminal cytoskeleton-associated protein, glycine-rich domain, which serves as a parking brake of the dynein motor. It is a very rare cause of familial parkinsonism, with only five missense mutations reported (P.G71A/E/R, pT72P, and p.Q74P) in a handful of families worldwide.[94-98] Histology shows severe neuronal loss in the SN without LBs and a unique distribution of TDP-43-positive pathology in neurons and glial cells in a pallidonigral distribution.[99]

Autosomal Recessive, Early-Onset (≤45 Years) Typical Parkinsonism

Parkin, PARK2

Mutations in parkin are the most common cause of autosomal recessive, early-onset typical parkinsonism, accounting for up to ~50% of cases with familial PD compatible with recessive inheritance and disease onset before the age of 45 years, as well as ~10% of sporadic cases with onset before the age of 45. No specific features differentiate patients with parkin mutations from other EOPD forms; however, symmetrical onset, dystonia at onset and hyperreflexia, slower progression of the disease, a tendency toward a better response to l-dopa, and preservation of olfaction might be more frequent among patients with parkin-related PD.[100, 101] Patients with parkin-related PD may present with isolated l-dopa-responsive dystonia mimicking dopa-responsive dystonia; therefore, this differential should be kept in mind, because the morbidity and recurrence risks differ significantly.[102] Few patients with pathogenic parkin mutations have come to autopsy. Many have not shown LB pathology, suggesting pathogenetic differences between the autosomal recessive and dominant forms of PD.[103] Many pathogenic mutations in parkin have been reported on (http://www.molgen.vib-ua.be/PDMutDB).[47, 48] More than 50% of parkin mutations are copy number mutations (deletions, duplications, and rarely triplications); consequently, dosage assay of all exons is required, in addition to sequencing, to achieve a high sensitivity of mutation detection.[104] The Parkin protein functions as an E3 ubiquitin ligase in the process of ubiquitination, a form of post-translational modification that conjugates ubiquitin protein(s) to lysine residue(s) of target proteins, which, in turn, determines their cellular fate. Most parkin mutations lead to loss of its E3 ligase function; therefore, loss of the parkin function might lead to accumulation of nonubiquitylated substrates.[105]

Phosphatase and Tensin Homolog-Induced Putative Kinase 1, PARK6

Mutations in phosphatase and tensin homolog-induced putative kinase 1 (PINK1) are the second-most common cause of autosomal recessive EOPD, accounting for ~1% to 8% of familial PD with recessive inheritance and ~1% of early-onset sporadic cases.[106] Depression and anxiety may be overrepresented in patients with PINK1-related PD,[107-110] whereas cognitive involvement is rare. More than 90% of the reported mutations in PINK1 are missense mutations, whereas the remaining are copy number mutations (http://www.molgen.vib-ua.be/PDMutDB). The pathology of PINK1-linked PD has been reported in a single patient and was characterized by neuronal loss in the SNc, LBs, and sparing of the locus ceruleus.[111] PINK1 and Parkin function in a common pathway for sensing and selectively eliminating damaged mitochondria from the mitochondrial network. PINK1 is stabilized on mitochondria with lower membrane potential and recruits Parkin from the cytosol. Once recruited to mitochondria, Parkin becomes enzymatically active and initiates the autophagic clearance of mitochondria by lysosomes (i.e., mitophagy).[112]

Oncogene, DJ-1, PARK7

Mutations in DJ-1 are rare and found in ~1% of EOPD patients.[113] Both missense mutations and copy number mutations are reported (http://www.molgen.vib-ua.be/PDMutDB). The clinical phenotype seems to be comparable to Parkin- and PINK1-related PD; however, dystonic and psychiatric features may be overrepresented.[14, 113, 114] Atypical phenotypes have been reported, including one family with early-onset parkinsonism, dementia, and amyotrophic lateral sclerosis.[115] The pathology of DJ-1-related PD has not been published.

Role of Heterozygous Mutations in Recessive PD Genes

A pathogenic effect of certain heterozygous mutations in parkin and PINK1 has been postulated,[116-119] complicating the genetic counseling of these individuals. A number of theories have been put forward as to why a heterozygous mutation in these genes might cause PD: A second pathogenic mutation might escape detection by the standard screening methods (e.g., deep intronic mutations); second, it is possible that some mutations are dominant and cause disease in the heterozygous state, or a heterozygous mutation may contribute to disease development as a genetic risk factor. However, a recent study did not find a significant difference in heterozygous parkin mutations in cases and controls.[120] Furthermore, another study found that 10% of cases with a heterozygous parkin mutation had pathogenic homozygous mutations in other recessive PD genes, emphasizing the importance of comprehensive screening of these individuals.[121]

Autosomal Recessive, Juvenile (≤21 Years) Atypical Parkinsonism

ATPase 13A2, ATP13A2, PARK9 (Also Known as Kufor-Rakeb Syndrome)

Mutations in ATP13A2 cause Kufor-Rakeb syndrome,[17] a juvenile-onset parkinsonism syndrome with a variety of associated features, such as supranuclear palsy, oculogyric dystonic spasms, facial-faucial-finger minimyoclonus, and spasticity. Dementia and visual hallucinations are recognized cognitive features. Brain imaging often reveals moderate cerebral and cerebellar atrophy, sometimes with iron deposition in the basal ganglia.[122] Copy number mutations have not been reported to date in ATP13A2. Mutations in ATP13A2 are likely to have a role in lysosomal degradation based on putative gene function, sural nerve pathology, identification of a homozygous ATP13A2 mutation in a kindred with a clinical and pathological diagnosis of neuronal ceroid lipofuscinosis (NCL),[123] and that ATP13A2 mutations cause a form of canine NCL.[124, 125]

F-Box Only Protein, PARK15

Patients with mutations in F-box only protein (FBXO7) present with juvenile-onset, severe l-dopa-responsive parkinsonism and pyramidal signs. The pathological features of this condition are not currently known. Recently, FBXO7 was shown to act in a common pathway with Parkin and PINK1 to induce mitophagy in response to mitochondrial depolarization and that disease-associated mutations in FBXO7 interfere with this pathway.[126]

Phospholipase A2, Group VI (Cytosolic, Calcium Independent), PARK14

Mutations in phospholipase A2, group VI (PLA2G6) usually cause an early-onset recessive degenerative disorder with spasticity, ataxia, and dystonia. However, later adult-onset forms of the disease can present with a dystonia-predominant parkinsonism often accompanied by cognitive decline, eyelid opening apraxia, and supranuclear gaze palsy.[22] Brain MRI may show iron accumulation in the globus pallidus, SN, and/or striatum. Pathologically, both early-infantile and late-onset PLA2G6 is characterized by LBs and synuclein-positive dystrophic neuritis in the SN and cortex as well as tau immunoreactive cortical neurofibrillary tangles.

Synaptojanin 1

A homozygous mutation in synaptojanin 1 (SYNJ1; R258Q) was recently identified by exome sequencing, in two kindreds, as the putative cause of a complex early-onset autosomal recessive parkinsonian syndrome,[127, 128] with rapidly progression and severe l-dopa-induced dystonia. Additional features may be present, including generalized seizures, cognitive decline, dysarthria, eyelid apraxia, and supranuclear gaze palsy. Brain MRI demonstrated diffuse cortical atrophy with T2 hyperintensities. SYNJ1 is a phosphoinositide phosphatase protein involved in clathrin-mediated endocytosis in the adult brain.[129]

Auxilin, DNAJC6

Recently, two recessive mutations (splicing mutation c.801-2 A>G and Q734X) have been identified in DNAJC6 as the probable cause of a juvenile-onset parkinsonism syndrome with rapid progression to dependence 10 to 15 years after onset. In one family, the parkinsonism was not l-dopa responsive. Associated features, which may be present, include dysarthria, generalized seizures, and brain atrophy on MRI.[130, 131] DNAJC6 encodes auxilin, which is a clathrin-associated protein enriched in nerve terminals, and may play a role in synaptic vesicle recycling.[132]

Identification of further multiplex kindreds with DNAJC6 and SYNJ1 mutations will confirm these as PD genes.

High-Risk Susceptibility Loci for PD

Glucocerebrosidase

The astute clinical observation that patients and relatives of patients with Gaucher's disease seem to present with PD more often than expected led to studies that confirmed that mutations in glucocerebrosidase (GBA) are a risk factor for PD.[133-137] The odds ratio (OR) for developing PD with a single GBA mutation is estimated to be 5.4 (largely reflecting the common N370S mutation), with mutations in 15% of Ashkenazi Jewish patients and 3% of non-Ashkenazi Jewish patients, compared to 3% and <1% of matched controls.[138] Recently, mutations in GBA have also been confirmed as a significant risk factor for dementia with LBs (LBD; estimated OR: 8.28) and PD with dementia (PDD; estimated OR: 6.48), confirming that PD, PDD, and LBD form a continuum within the spectrum of LB diseases.[139] Evidence suggests that impairment of the aging lysosome, enhanced by deficient or mutant GBA, can affect α-Syn degradation.[140, 141] Functional loss of GBA causes accumulation and aggregation of α-Syn, which, in turn, inhibits the trafficking and lysosomal activity of GBA, leading to a self-propagating disease.[142]

Genome-Wide Association Study Identified PD Genes

A large number of genome-wide association studies have been performed in PD,[24, 25, 143-152] resulting in a growing number of loci. A recent meta-analysis provided evidence for 16 independent risk loci.[145, 153] Common variations in genes that cause monogenic PD (SNCA and LRRK2) have been found to modify risk of developing the “sporadic” form of the disease. Two common variants in LRRK2 (G2385R and R1628P) appear to increase the risk of PD by approximately 2-fold, particularly in Asian populations.[154-156] An eloquent link between sporadic and monogenic forms was found in the case of SNCA, where risk alleles are associated with increased expression, mirroring the dosage effects observed in multiplications of SNCA, with age of onset in patients with triplications a decade earlier than duplications. However, despite the inclusion of these risk loci in commercially available genetic profiling, genetic susceptibility testing is not currently advised because of the numerous challenges of estimating disease risk, integrating genetic information with other factors that influence disease expression, such as environmental exposure and health behaviors, and the lack of preventative therapies. Extrapolating risk from aggregate data to make inferences at a personalized level can be problematic. In order to interpret results, we will need to understand better gene-gene and gene-environment interactions implicated in human diseases. Currently known PD risk alleles each confer only a ~1.3-fold increase in risk. Even when considering cumulative risk burden for those in the highest quintile of disease risk versus the lowest quintile, the OR is ~2.5.[146] Genetic susceptibility testing raises important ethical and practice issues related to test access, informed consent, risk estimation and communication, return of results, and policies to prevent genetic discrimination.[157] However, as larger meta-analyses identify additional low-risk genetic susceptibility alleles and possibly lower frequency variants of large effect size using next-generation sequencing (NGS), in the future, we may be able to develop better genetic risk prediction models that are clinically useful. Nonetheless, the risk factors identified by these studies provide clues to the underlying molecular mechanisms involved and offer potential therapeutic targets.

Monogenic Disorders That May Present With Parkinsonism

Table 2 lists several neurological disorders in which parkinsonism (sometimes with a good l-dopa response) may be the prominent presenting clinical feature. However, for all of these disorders, there is, typically, multisystem neurological involvement.

Table 2. Selected monogenic disorders that can present with clinical features of PD
DisorderGene/OMIMMode of InheritanceClinical Features
Disorders with dystonia as an associated feature
Dopa-responsive dystonia (DRD) (DYT5)GCH1 (128230)Autosomal dominantPresents with a range of phenotypes: juvenile-onset dopa-responsive dystonia, spastic paraplegia mimicking cerebral palsy, and, more recently, “benign” adult-onset PD with good response to low-dose l-dopa and DATSCAN abnormalities[158-161]
Rapid-onset dystonia-parkinsonism (DYT12)ATP1A3 (128235)Autosomal dominantCharacterized by abrupt onset of dystonia-parkinsonism often with bulbar involvement and no response to l-dopa; fever, physiologic stress, and excess alcohol are recognized triggers. Recently, de novo mutations in ATP1A2 have been identified as the cause of alternating hemiplegia of childhood.[162]
Panthotenate-kinase-associated neurodegenerationPANK2 (234200)Autosomal recessiveTypically presents with an early-onset spasticity, ataxia and dystonia; later-onset forms of the disease can present with a dystonia predominant parkinsonism.[22] Brain MRI may show brain iron accumulation.
X-linked dystonia parkinsonism (DYT3)TAF1 (314250)X-linked recessiveParkinsonism is the usual presenting sign, followed by severe dystonia. Only reported in individuals of Filipino ancestry, primarily men and rarely women.
Wilson diseaseATP7B (277900)Autosomal recessive~40% of patients present with neurological symptoms, including tremor, ataxia, chorea, and pseudobulbar involvement through to rigid dystonia and parkinsonism. Investigations show low serum ceruloplasmin and high urinary copper. Slit-lamp examination for Kayser-Fleischer rings will be present in ~90% of those with neurological involvement. Brain MRI may show T2 hyperintensities in the basal ganglia, among other abnormalities.[163]
Disorders with ataxia as an associated feature
SCA2, SCA3, SCA6

ATX2 (183090)

ATX3 (109150)

CACNA1A (183086)

Autosomal dominantSCA2, SCA3, and SCA6 have a wide spectrum of presentation and may include a dopa-responsive parkinsonism, sometimes indistinguishable from PD.[164-167]
Fragile X tremor/ataxia syndrome (FXTAS)FMR1 (300623)X-linked recessivel-dopa-responsive parkinsonism may be a presenting sign of FXTAS. Premutations (55–200 CGG repeats) are found in ~1:800 males and 1:250 females with a penetrance of ~33% and ~10%, respectively.[168, 169] Other clinical features include tremor, ataxia, neuropathy, autonomic dysfunction, cognitive decline, and behavioral changes.
Disorders with dementia as an associated feature
Frontotemporal dementia with parkinsonism-17, GRN and TARDBP-related frontotemporal dementia

MAPT (157140)

GRN (607485)

TARDBP (612069)

C9orf72 (614260)

Autosomal dominantRarely, patients with mutations in these genes will present with parkinsonism. However, most often the clinical phenotype is complicated by frontotemporal dementia, eye movement disorders (MAPT), or amyotrophic lateral sclerosis (TARDBP and C9orf72).[170-175]
Disorders with chorea as an associated feature
Huntington's disease (HD)HTT (143100)Autosomal dominantDopa- and non-dopa-responsive parkinsonism can occur in the later stages of choreic disease and as the predominant feature of juvenile patients (Westphal variant).

Genetic Testing in PD

No guidelines exist for genetic testing in PD, and the question of who to test and what for is not trivial.

Genetic Counseling

The primary goal of molecular diagnosis is always to provide help for the patient and/or their family. “Diagnostic testing” refers to when a genetic test is used to confirm or rule out a known or suspected disorder in a symptomatic patient. “Predictive testing” is used to clarify the genetic status of an asymptomatic family member at risk for a genetic disorder. Some considerations regarding genetic testing are listed in Table 3. Genetic testing should always be offered in the framework of genetic counseling based on an informed decision by the patient. Benefits of genetic testing in PD may include prevention of further unnecessary investigations, prediction of the natural history of the disease, and reproductive planning. Considerations against genetic testing may include the lack of neuroprotective drugs in PD (particularly important in predictive testing) and that a molecular diagnosis is currently unlikely to fundamentally change management of the disease. There are established guidelines for predictive testing in Huntington's disease, which should be followed in all cases of predictive testing.[176]

Table 3. Considerations when obtaining consent for genetic testing
Some Considerations for the Clinician and Patient When Considering Genetic Testing in PD
What is the purpose of the test?
Is there a clear explanation of the mode of inheritance, nature, and treatment of the condition?
What are the potential benefits of having the test?
What are the potential disadvantages?
Could the genetic test have implications for the patient's future health, reproductive choices, relatives, family relationships (information about parentage), present or future employment, and insurance prospects?
Likely timescales for availability of tests

Determining Whom to Test

The clinician's first step is to establish the inheritance pattern, because it can exclude disorders that are incompatible in this regard. This is best facilitated with a three-generation pedigree. Interpretation of the mode of inheritance may be complicated by reduced penetrance in autosomal dominant disorders, whereas autosomal recessive PD can be masked in small families where there may not be other affected siblings; in such cases, early age at onset is a better indicator. Parental consanguinity can be considered a red flag for autosomal recessive PD, but this is not always the case.[177] Careful examination of the patient may elicit additional clinical signs, particularly in juvenile PD and multisystem disorders (Table 2), which help guide genetic testing.

We recommend that patients presenting with autosomal dominant PD are tested first for the G2019S mutation in LRRK2 in at-risk populations (European, North African, and Ashkenazi Jewish patients), next for SNCA multiplications, and last for SNCA point mutations and the VPS35 D620N mutation. Genetic testing for juvenile PD (without additional features) and EOPD (<age 45 years), irrespective of family history, should include copy number and point mutations in Parkin, PINK1, and DJ-1. Genetic testing may be appropriate for patients with seemingly “idiopathic” PD who are from at-risk populations, for example, the G2019S LRRK2 mutation in Ashkenazim and North African Arabs and the R1441G mutation in Basque and Spanish patients. The presence of additional features in complex juvenile-onset parkinsonian syndromes will guide genetic testing. Figure 1 gives a suggested algorithm for genetic testing in PD.

Figure 1.

Suggested algorithm for genetic testing in PD.

Looking to the Future

Rare monogenic forms of PD and susceptibility loci have contributed eminently to our understanding of the pathophysiology of “idiopathic” Parkinson's disease by helping to define and understand pathways that lead to neuronal dysfunction and cell death. The increasing availability of NGS is likely to lead to the identification of not only additional Mendelian PD genes, but also high-risk susceptibility loci. Ultimately, it may lead to genetic profiling of patients to determine a personalized therapeutic regime.

Funding Agencies

This work was supported, in part, by the Wellcome Trust/MRC Joint Call in Neurodegeneration award (WT089698) to the UK Parkinson's Disease Consortium (UKPDC), whose members are from the UCL/Institute of Neurology, the University of Sheffield, and the MRC Protein Phosphorylation Unit at the University of Dundee. This work was undertaken at UCLH/UCL, who received a proportion of funding from the Department of Health's NIHR Biomedical Research Center's funding scheme.

Author Roles

(1) Research Project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript: A. Writing of the First Draft, B. Review and Critique.

U.-M.S.: 3A, 3B

H.H.: 3B

N.W.W.: 3B

Financial Disclosures

U.-M.S. is funded by an MRC fellowship grant. N.W.W. holds grants from the Bachmann-Strauss Dystonia Parkinson Foundation, the MRC, and the Wellcome Trust.