Genetics of Dystonia

J Neural Transm (2006) [Suppl] 70: 489–495 # Springer-Verlag 2006

Genetics of dystonia

S. Bressman Beth Israel Medical Center, Department of Neurology, Albert Einstein College of Medicine, Phillips Ambulatory Care Center, New York, NY, USA

Summary. Primary torsion dystonia (PTD) expression. Defining genetic etiologies has has a broad clinical spectrum, with earlier altered the way neurologists diagnose and onset of symptoms associated with more gen- counsel patients, including the important need eralized muscle involvement. The causes for to provide genetic counseling to patients and most dystonia are unknown although several their families. Ultimately, understanding the monogenic subtypes have been identified. genetic causes of dystonia, and the effects One important genetic cause of PTD is of these alterations, holds the promise of DYT1; a three base pair deletion in this gene rational, targeted therapies. is a major cause for early-onset dystonia. Its identification has allowed the development of Primary dystonia cellular and animal models; it has also permit- ted studies that identify both ‘‘manifesting’’ There are many classification schemes to and ‘‘non-manifesting’’ DYT1 mutation car- organize the causes of dystonia. Most create riers. These studies have expanded our under- at least two broad categories: primary torsion standing of clinical expression to include dystonia (previously named idiopathic tor- psychiatric symptoms and also have enabled sion dystonia), and secondary (or non-primary) imaging studies of endophenotypes. These ad- dystonia (see Table 1 and Fig. 1). Primary vances provide a widened platform for future torsion dystonia (PTD) is defined as a syn- research. drome in which dystonia is the only clinical sign (except for tremor) and there is no evidence of neuronal degeneration or an acquired Introduction cause. Secondary (non-Primary) dystonias in- Genetic etiologies have long been suspected clude all other dystonia subtypes and can for many subtypes of dystonia. Recent molec- further divided into inherited, complex, and ular advances have lead to the identification acquired etiologies. of an increasing number of genes for primary and secondary dystonia subtypes (see Table 1). Primary torsion dystonia (PTD) This information has opened the way for stud- Early-onset PTD and DYT1 ies aimed at characterizing basic pathogenic mechanisms, including cellular and animal Early-onset PTD is 3–5 times more common models. It has also allowed for a broader in Ashkenazi Jews compared to other popu- analysis of phenotype and endophenotype lations and is transmitted in an autosomal to further characterize the spectrum of gene dominant fashion with reduced penetrance 490 S. Bressman

Table 1. Causes of dystonia

Primary Autosomal dominant Early limb (DYT1, other genes to be determined) Mixed (DYT6, DYT13, other genes to be determined) Late Focal (DYT7, other genes to be determined) Other genetic causes ?autosomal recessive, complex Secondary Inherited Dystonia plus (non-degenerative) DRD (DYT5-GCH1, DYT14, other biopterin deﬁciencies, tyrosine hydroxylase deﬁciency) Myoclonus – Dystonia (DYT11-epsilon sarcoglycan, 18p locus) Rapid-Onset Dystonia Parkinsonism (DYT12, ATP1A3) Degenerative Autosomal Dominant (e.g., Huntington’s disease, SCA’s especially SCA3) Autosomal Recessive (e.g., Wilson’s, NBIA1, GM1, GM2, Parkin) X-linked (e.g., X-linked Dystonia-Parkinsonism=Lubag, DDP) Mitochondrial Complex=unknown Parkinsonism (e.g., Parkinson’s disease, multisystem atrophy, progressive supranuclear palsy, corticobasal degeneration) Acquired e.g., drug-induced, perinatal injury, head trauma, cervical trauma, peripheral trauma, infectious and post infectious, tumor, AVM, stroke, central pontine myelinolysis, multiple sclerosis

DRD dopa-responsive dystonia; GCH1 GTP cyclohydrolase1; SCA spinocerebellar ataxia; NBIA1 neurode- generation with brain iron accumulation; DDP deafness dystonia peptide of 30–40% in both Ashkenazi Jews and non- PTD. The DYT1 gene encodes a novel pro- Ashkenazim (Bressman, 1989; Pauls, 1990). tein, torsinA, that is 332 amino acids long The difference in disease frequency is thought (38 kD), with potential sites for glycosyla- to be the result of a founder mutation in tion and phosphorylation, as well as an DYT1 (see below) that was introduced into amino terminal hydrophobic leader sequence the Ashkenazi population at the time of consistent with membrane translocation= a ‘‘bottleneck’’ in the 1600s, followed by a targeting (Ozelius, 1997). The GAG deletion period of tremendous population growth results in the loss of one of a pair of glutamic (Risch, 1995). acid residues near the carboxy terminus of the The gene at locus DYT1 was identiﬁed in protein. TorsinA shares functional regions with 1997 (Ozelius, 1997) and is responsible for a the AAAþ superfamily. These proteins are large proportion of early limb onset PTD characterized by Mgþþ-dependent ATPase (also known as dystonia musculorum defor- activity and typically form six-membered, ho- mans or Oppenheim’s disease) across many momeric ring structures. Many serve as chap- different populations (Valente, 1998; Leube, erones that mediate conformational changes 1999; Bressman, 2000). Only one recurring in target proteins. They are associated with a mutation in DYT1, an in – frame GAG dele- number of functions including protein fold- tion, has been associated unequivocally with ing and degradation, cytoskeletal dynamics, Genetics of dystonia 491

Fig. 1. Dystonia (DYT) genetic loci membrane trafficking and vesicle fusion and phic factor withdrawal (Shashidharan, 2004; response to stress. Mclean, 2002; Kuner, 2003); in contrast, TorsinA is widely distributed in normal mutant torsinA fails to protect cells from these adult brain and brain tissue from patients toxic insults and leads to the formation of with a variety of movement disorders includ- perinuclear inclusion bodies (Shashidharan, ing Parkinson’s disease and DYT1 dystonia 2004; Hewett, 2000; Gonzalez-Alegre, 2004). (Augood, 1998, 2002; Shashidharan, 2000; Despite these findings, suggesting torsinA Walker, 2002). There is intense expression in protects against various stressors, examina- substantia nigra compacta dopamine neurons, tion of DYT1 brains have been relatively cerebellar dentate nucleus, Purkinje cells, unrevealing (Walker, 2002; Furukawa, 2000; basis pontis, locus ceruleus, numerous thal- Rostasy, 2003), except for one recent study amic nuclei, the pedunculopontine nucleus, of four clinically affected DYT1 brains the oculomotor nuclei, hippocampal forma- (McNaught, 2004). This study utilized highly tion and frontal cortex. TorsinA immunolo- sensitive antibodies to ubiquinated proteins calization in the human CNS appears to be and found perinuclear inclusion bodies in the restricted to the cytoplasm of neurons midbrain reticular formation and periaque- (Konakova, 2001), and when transfected into ductal grey, specifically involving cholinergic mammalian cells behaves as lumenally ori- neurons of the pedunculopontine nucleus. In ented glycoproteins (Kustedjo, 2000; Hewett, addition tau=ubiquitin immunoreactive aggre- 2003). Overexpression of wild-type torsinA gates were observed in pigmented neurons of in cultured cells prevents protein aggregation the substantia nigra compacta and locus co- and protects against cytotoxicity after pro- eruleus. These pathological findings are con- teasomal inhibition, oxidative stress, or tro- sistent with the notion that mutated torsinA 492 S. Bressman may impair protein handling; they also point cases; this compares with 16–53% in early- to involvement of the pedunculopontine nu- onset non-Jewish populations. cleus and other brainstem structures, which With identification of the DYT1 gene it have extensive connections to the basal gan- has become possible to more fully investigate glia. Recent transgenic mouse model results the clinical features and expression of the support these findings as pathological for gene using a variety of approaches, including DYT1. Shashidharan and colleagues reported imaging, psychological, and neurophysio- that 40% of four overexpressing transgenic logical testing. By comparing DYT1 gene lines developed hyperkinesias; those with carriers, including non-manifesting carriers, hyperkinesias were found to have decreased to non – carriers, gene associated features striatal dopamine compared to increased dopa- can be distinguished. Using this paradigm, mine levels in transgenic mice without hy- 18F-fluorodeoxyglucose positron emission perkinesias. Immunohistochemistry revealed tomography (PET) and network analysis, perinuclear brainstem inclusions similar Eidelberg and colleagues demonstrated an to those identified in DYT1 human brains abnormal pattern of glucose utilization char- (Shashidharan, 2005). acterized by covarying metabolic increases Although all DYT1 PTD has a single in the basal ganglia, cerebellum, and sup- molecular basis, clinical expression is extra- plementary motor cortex (SMA) that was ordinarily broad, even within families; 70% present in both ‘‘manifesting’’ and ‘‘non- of gene carriers have no definite signs of dys- manifesting’’ gene carriers (Eidelberg, 1998). tonia and among the remaining 30% dystonia Glucose PET studies have also been utilized ranges from focal to severe generalized for psychomotor testing in ‘‘non-manifesting’’ (Bressman, 2000). There are however com- gene carriers and show subtle abnormalities mon DYT1 clinical characteristics: the great in sequence learning both in the motor per- majority of people with dystonia due to DYT1 formance and recruitment of brain networks have early onset (before 26 years) that first (Ghilardi, 2002). Other imaging studies of affects an arm or leg. About 65% progress DYT1 gene carriers have found decreased to a generalized or multifocal distribution, striatal D2 receptor binding (Asanuma, 2005), the rest having segmental (10%) or only focal and microstructural changes involving the (25%) involvement. When viewed in terms of subgyral white matter of the sensorimotor body regions ultimately involved, one or more cortex (Carbon, 2004). Non-imaging studies limbs are almost always affected (over 95% using this paradigm include an assessment have an affected arm). The trunk and neck of possible psychiatric manifestations; an may also be affected (about 25–35%) and equal and increased risk for major recurrent they may be the regions producing the great- depression in DYT1 gene carriers manifest- est disability (31); the cranial muscles are less ing and not manifesting dystonia was found likely to be involved (<15–20%). Rarely, (Heiman, 2004). Electrophysiological anal- affected family members have late-onset (up yses have also identified abnormalities, to age 64 years) (Opal, 2003). Also, although namely reduced intracortical inhibition and the arm is the body region most commonly a shortened cortical silent period (Edwards, affected in those with focal disease, the neck 2003). These studies strongly support the or cranial muscles have been reported as iso- presence of wider clinical gene expression, lated affected sites (Tuffery-Giraud, 2001). abnormal brain processing and associated Because of the founder effect, the DYT1 structural brain changes in gene carriers GAG deletion is more important in the regardless of overt motor signs of dystonia, Ashkenazi population, where it accounts for expanding the notion of penetrance and about 80% of early (less than 26 years) onset phenotype. Genetics of dystonia 493

Early-onset but not DYT1 dystonias but the role of DRD5 and DYT1 as a susceptibility genes remain to be eluci- A large group of early-onset PTD, especially dated (Misbahuddin, 2002; Clarimon, 2005). among non-Jewish populations, is not due to the DYT1 GAG deletion. Two loci, DYT6 (Almasy, 1997) and DYT13 (Valente, Secondary dystonia and dystonia 2001), have been mapped in families having plus syndromes an average age-onset in adolescence. However, neither locus has been confirmed in other Etiological subgroups for secondary dystonias families and they are suspected to account include: 1) inherited causes; 2) a group of for only a minority of non-DYT1 early-onset primarily parkinsonian disorders including cases. Further, overall clinical features in these Parkinson’s disease that are thought to have two families differ from DYT1 (although fea- complex etiologies and; 3) environmental or tures in any single family member may over- acquired causes. In addition most classifica- lap with DYT1). The family phenotypes for tions also include other movement disorders DYT6 and 13 are marked by prominent in- that may display dystonic phenomenology volvement of cranial and cervical muscles such as tics and the paroxysmal dyskinesias with variable spread; also, compared to DYT1 and the pseudo-dystonias. The latter are not a greater proportion of family members have considered true dystonia but muscle contrac- later adolescent and adult onset. To distin- tions mimicking dystonia such as seen in guish this phenotype from the typical early- Sandifer’s syndrome, orthopedic conditions onset phenotype associated with DYT1 and and psychogenic dystonia. typical late-onset focal phenotypes the term Among the inherited forms of secondary ‘‘mixed’’ has been applied. dystonia is a relatively newly defined category of dystonia-plus syndromes consisting of Late-onset PTD three clinically defined entities: dopa – responsive dystonia (DRD), myoclonus – dystonia Like early-onset PTD, late-onset PTD also (M-D), and rapid – onset dystonia-parkinson- appears to have autosomal dominant in- ism (RDP). The dystonia-plus category was heritance (Defazio, 1993). However, unlike distinguished from both primary dystonia and early-onset dystonia, most studies show that other inherited secondary dystonias because penetrance is even more reduced (about it shares some but not all features of both 12%–15% compared to 30% for early-onset); groups. That is, like primary dystonia these alternatively, penetrance may be higher in a three syndromes do not appear to be degener- subset with the remainder sporadic. Consis- ative. Although pathology is limited, evidence tent with the notion of increased penetrance to date supports genetic defects that result in in a subset of late-onset PTD, are descrip- functional brain changes not associated with tions of large families with more highly progressive neuronal death. Further, unlike penetrant autosomal dominant disease. One primary dystonia, but similar to the other such family with adult-onset torticollis was degenerative secondary dystonias, the dysto- studied resulting in the mapping of DYT7 nia plus group has as characteristic clinical (Leube, 1996). Other clinically similar fam- features signs other than dystonia including ilies have been excluded from DYT7 (47) parkinsonism for DRD and RDP and myoclo- suggesting yet other loci for adult onset focal nus for M-D. PTD. Most recently a polymorphism in the As our understanding of these syndromes D5 dopamine receptor (DRD5) gene and a is expanding the complexity of their genetic DYT1 haplotype (but not the GAG deletion) and clinical heterogeneity is being detailed. have been associated with adult-onset focal For example, for DRD there are currently 494 S. Bressman several known genetic biochemical etiologies Clarimon J, Asgeirsson H, Singleton A, Jakobsson F, each with protean clinical manifestations, and Hjaltason H, Hardy J, Sveinbjornsdottir S (2005) Torsin A haplotype predisposes to idiopathic dys- for myoclonus – dystonia there appear to be at tonia. Ann Neurol 57(5): 765–767 least two genetic etiologies (see Table 1). Defazio G, Livrea P, et al. (1993) Genetic contribution to idiopathic adult-onset blepharospasm and cra- Conclusions nial-cervical dystonia. Eur Neurol 33(5): 345–350 Edwards MJ, Huang YZ, Wood NW, Rothwell JC, There has been tremendous recent growth in Bhatia KP (2003) Different patterns of electrophy- our knowledge of the clinical and genetic siological deficits in manifesting and non-manifesting carriers of the DYT1 gene mutation. Brain 126: complexity of dystonia, including PTD and 2074–2080 the dystonia-plus syndromes. The delineation Eidelberg D, Moeller JR, Antonini A, et al. (1998) of the varied clinical subtypes of dystonia has Functional brain networks in DYT1 dystonia. aided gene identification, e.g., DYT1, GCH1, Ann Neurol 44: 303–312 epsilon sarcoglycan. With the identification of Furukawa Y, Hornykiewicz O, et al. (2000) Striatal do- genes it has been possible to develop cellular pamine in early-onset primary torsion dystonia with the DYT1 mutation. Neurology 54(5): 1193–1195 and animal models; it has also allowed the Ghilardi MF, Carbon M, et al. (2003) Impaired identification of populations harboring muta- sequence learning in carriers of the DYT1 dystonia tions and expanded our understanding of en- mutation. Ann Neurol 54(1): 102–109 dophenotypes. These advances have provides Gonzalez-Alegre P, Paulson HL (2004) Aberrant cel- a widened platform for future research. lular behavior of mutant torsinA implicates nuclear envelope dysfunction in DYT1 dystonia. J Neurosci 24(11): 2593–2601 References Heiman GA, Ottman R, Saunders-Pullman RJ, Ozelius LJ, Risch NJ, Bressman SBB (2004) Increased risk Almasy L, Bressman SB, Kramer PL, et al. (1997) for recurrent major depression in DYT1 dystonia Idiopathic torsion dystonia linked to chromosome mutation carriers. Neurology 63(4): 631–637 8 in two Mennonite families. Ann Neurol 42: Hewett J, Ziefer P, et al. (2003) TorsinA in PC12 cells: 670–673 localization in the endoplasmic reticulum and Asanuma K, Ma Y, Huang C, Carbon-Correll M, response to stress. J Neurosci Res 72(2): 158–168 Edwards C, Raymond D, Bressman SB, Moeller Hewett J, Gonzalez-Agosti C, et al. (2000) Mutant JR, Eidelberg D (2005) Decreased striatal D2 torsinA, responsible for early-onset torsion dysto- receptor binding in non-manifesting carriers of nia, forms membrane inclusions in cultured neural the DYT1 dystonia mutation. Neurology 64(2): cells. Hum Mol Genet 9(9): 1403–1413 347–349 Konakova M, Huynh DP, et al. (2001) Cellular dis- Augood SJ, Penney JB Jr, Friberg IK, Breakefield XO, tribution of torsin A and torsin B in normal human Young AB, Ozelius LJ, Standaert DG (1998) brain. Arch Neurol 58(6): 921–927 Expression of the early-onset torsion dystonia gene Kustedjo K, Bracey MH, et al. (2000) Torsin A and its (DYT1) in human brain. Ann Neurol 43: 669–673 torsion dystonia-associated mutant forms are lu- Augood SJ, Hollingsworth Z, et al. (2002) Dopamine menal glycoproteins that exhibit distinct subcellular transmission in DYT1 dystonia: a biochemical localizations. J Biol Chem 275(36): 27933–27939 and autoradiographical study. Neurology 59(3): Leube B, Rudnicki D, et al. (1996) Idiopathic torsion 445–448 dystonia: assignment of a gene to chromosome 18p Bressman SB, de Leon D, Brin MF, et al. (1989) in a German family with adult onset, autosomal Idiopathic torsion dystonia among Ashkenazi Jews: dominant inheritance and purely focal distribution. Evidence for autosomal dominant inheritance. Ann Hum Mol Genet 5(10): 1673–1677 Neurol 26: 612–620 Leube B, Kessler KR, Ferbert A, et al. (1999) Pheno- Bressman SB, Sabatti C, Raymond D, et al. (2000) The typic variability of the DYT1 mutation in German DYT1 phenotype and guidelines for diagnostic dystonia patients. Acta Neurol Scand 99: 248–251 testing. Neurology 54: 1746–1752 McLean PJ, Kawamata H, et al. (2002) TorsinA and Carbon M, Kingsley PB, Su S, Smith GS, Spetsieris P, heat shock proteins act as molecular chaperones: Bressman S, Eidelberg D (2004) Microstructural suppression of alpha-synuclein aggregation. white matter changes in carriers of the DYT1 gene J Neurochem 83(4): 846–854 mutation. Ann Neurol 56(2): 283–286 Genetics of dystonia 495

McNaught KS, Kapustin A, et al. (2004) Brainstem Shashidharan P, Paris N, et al. (2004) Overexpression pathology in DYT1 primary torsion dystonia. Ann of torsinA in PC12 cells protects against toxicity. Neurol 56(4): 540–547 J Neurochem 88(4): 1019–1025 Misbahuddin A, Placzek MR, Chaudhuri KR, et al. Shashidharan P, Sandu D, Potla U, Armata IA, Walker (2002) A polymorphism in the dopamine receptor RH, McNaught KS, Weisz D, Sreenath T, Brin MF, DRD5 is associated with blepharospasm. Neuro 58: Olanow CW (2005) Transgenic mouse model of 124–126 early-onset DYT1 dystonia. Hum Mol Genet 14(1): Opal P, Tintner R, Jankovic J, et al. (2002) Intrafamilial 125–133 phenotypic variability of the DYT1 dystonia: from Tuffery-Giraud S, Cavalier L, Roubertie A, Guittard C, asymptomatic TOR1A gene carrier status to dys- Carles S, Calvas P, Echenne B, Coubes P, Claustres tonic storm. Mov Disord 17(2): 339–345 M (2001) No evidence of allelic heterogeneity in Ozelius LJ, Hewett JW, Page C, et al. (1997) The early- DYT1 gene of European patients with early onset onset torsion dystonia gene (DYT1) encodes an torsion dystonia. J Med Genet 38: e35 ATP-binding protein. Nature Genet 17: 40–48 Valente EM, Warner TT, Jarman PR, et al. (1998) The Pauls DL, Korczyn AD (1990) Complex segregation role of primary torsion dystonia in Europe. Brain analysis of dystonia pedigrees suggests autosomal 121: 2335–2339 dominant inheritance. Neurology 40: 1107–1110 Valente EM, Bentivoglio AR, Cassetta E, et al. (2001) Risch N, de Leon D, Ozelius L, et al. (1995) Genetic DYT13, a novel primary torsion dystonia locus, analysis of idiopathic torsion dystonia in Ashkenazi maps to chromosome 1p36.13–36.32 in an Italian Jews and their recent descent from a small founder family with cranial-cervical or upper limb onset. population. Nature Genet 9: 152–159 Ann Neurol 49: 363–364 Rostasy K, Augood SJ, Hewett JW, Leung JC, Walker RH, Brin MF, et al. (2002) TorsinA immuno- Sasaki H, Ozelius LJ, Ramesh V, Standaert DG, reactivity in brains of patients with DYT1 and non- Breakeﬁeld XO, Hedreen JC (2003) Torsin A pro- DYT1 dystonia. Neurology 58(1): 120–124 tein and neuropathology in early onset generalized dystonia with GAG deletion. Neurobiol Dis 12(1): 11–24 Author’s address: Prof. S. Bressman, Beth Israel Shashidharan P, Kramer BC, Walker RH, Olanow CW, Medical Center, Department of Neurology, Albert Brin MF (2000) Immunohistochemical localization Einstein College of Medicine, Phillips Ambulatory and distribution of torsinA in normal human and rat Care Center, 10 Union Square East, New York, NY brain. Brain Res 853: 197–206 10003, USA, e-mail: [email protected]