Twisting mice move the dystonia field forward

A common form of the hyperkinetic movement disorder dystonia is caused by mutations in the gene TOR1A (located within the DYT1 locus), which encodes the ATPase torsinA. The underlying neurobiological mechanisms that result in dystonia are poorly understood, and progress in the field has been hampered by the absence of a dystonia-like phenotype in animal models with genetic modification of Tor1a. In this issue of the JCI, Liang et al. establish the first animal model with a dystonic motor phenotype and link torsinA hypofunction to the development of early neuropathological changes in distinct sensorimotor regions. The findings of this study will likely play an important role in elucidating the neural substrate for dystonia and should stimulate systematic neuropathological and imaging studies in carriers of TOR1A mutations.

For many brain disorders, identification and characterization of the underlying neurobiological mechanisms remains a challenge for clinicians and scientists. Lack of defined neural substrates and an understanding of the pathways responsible for neurological and psychiatric symptoms has limited the development of novel therapies, which are urgently needed to improve the care and quality of life of affected individuals. In this issue, Liang and collaborators present animal models that recapitulate the major clinical symptomatology of dystonia (1). The study by Liang and colleagues represents an important leap forward for the dystonia research field.

Dystonia is characterized by sustained or intermittent muscle contractions that cause abnormal, often repetitive, twisting movements and postures and is now recognized as a heterogenous group of hyperkinetic movement disorders. The term dystonia was coined in 1911 by Herman Oppenheim, who used “dystonia musculorum deformans” to describe a childhood-onset form of generalized dystonia (2). These disorders have traditionally been classified as either primary or secondary dystonias. Primary dystonia is considered to only present with tremor or myoclonus as an additional neurological symptom in the absence of any neuropathological changes, whereas secondary dystonia is considered the consequence of a hereditary neurodegenerative disorder or an insult to the brain, and other neurological symptoms can be manifest. More recently, the dystonia classification system has been called into question. In particular, the division into primary and secondary dystonia has been criticized as neuroimaging and neuropathological studies have begun to indicate the presence of structural changes in primary dystonia (3). The recent international consensus classification has instead proposed to use a two-axes classification based on clinical characteristics and etiology (4).

There is a considerable genetic contribution for many forms of dystonia. The first dystonia locus (DYT1) was identified in 1990 and localized to chromosome 9 (5). The gene, which was identified seven years later and named TOR1A, encodes for the ubiquitously expressed torsinA protein (6). TorsinA is a member of the AAA⁺ family of ATPases present in the endoplasmic reticulum/nuclear envelope space and is thought to play a role in structural integrity and protein trafficking in the cell (reviewed in ref. 7). The DYT1/TOR1A mutation is responsible for around 50% of early onset primary dystonia cases (8). Most often, the genetic mutation in DYT1-related dystonia is an in-frame GAG deletion (termed ΔE mutant), which is thought to result in at least a partial loss of torsinA function. The penetrance of the mutation is around 30% to 40%, and in these patients the symptoms are variable, ranging from a mild focal presentation to disabling generalized dystonia (8). Although a few of the available studies suggest the presence of rather unspecific structural and morphological alterations in the substantia nigra and the cerebellum (9, 10), there is a general paucity of more comprehensive clinical studies examining structural changes in the brain of DYT1-dystonia patients.

Studies of animals with genetic modification of the Dyt1/Tor1a gene have suffered from the inability of these models to display overt dystonia; therefore, these preclinical models lack validity for studying overt clinical symptoms. Previous models have been engineered with a heterozygous knockin of the Tor1a ΔE mutation or with brain region–specific knockout of the Dyt1 locus; however, these mutations do not result in a clear dystonic motor phenotype (recently reviewed in ref. 11). On the other hand, homozygous knockout of Tor1A or homozygous knockin of the Tor1a ΔE mutation in mice results in neonatal lethality, and therefore, these models are not useful (12). Based on previous attempts to develop a murine dystonia model, Liang and collaborators hypothesized that a dystonia mouse would require torsinA deficiency in critical brain areas during postnatal CNS development in order to develop a clinically relevant phenotype. Toward this goal, Liang and colleagues generated mice that lacked the Tor1a gene in the entire CNS and found that these mice developed striking abnormal twisting movements that are indicative of dystonia. The severe early-onset phenotype observed in mice lacking torsinA in the CNS was accompanied by premature death at postnatal day 16. Liang et al. took their dystonia mouse model one step further and created mice that are heterozygous for the Tor1a ΔE mutation and a floxed WT Tor1a allele, which can be deleted in select tissues, allowing expression of torsinA ΔE in the absence of the WT protein in the CNS. Liang et al. elegantly demonstrate that these animals develop a dystonia-like phenotype as early as the second postnatal week, but survive to adulthood. Interestingly, the CNS-specific Tor1a ΔE mutant mice showed an age-related improvement in behavioral phenotypes, suggesting that the residual function of torsinA may partly compensate for the early deficits in the maturing CNS (1).

Importantly, the behavioral abnormalities in the mouse models developed by Liang and colleagues were accompanied by perinuclear accumulation of ubiquitin as well as neuropathological changes, including gliosis, caspase-3 activation, and ER stress in selective sensorimotor regions. Neuropathology was observed in deep layers of sensorimotor cortex, ventral posterior thalamus, globus pallidus, deep cerebellar nuclei, red nucleus, and the facial nerve nuclei. Thus, a second and equally important aspect of the study by Liang and colleagues is the linkage of neuropathological changes in the brain to a model of “primary” dystonia, which provides further support in the debate on the presence of structural brain changes in this disease. Although Liang and collaborators detected a selective vulnerability of certain sensorimotor cells to torsinA deficiency, further neuropathological analyses will be important to determine the full extent of structural and morphological changes in these models. Nevertheless, the findings by Liang et al. call for further systematic analyses of postmortem tissue from individuals with DYT1-related dystonia in order to elucidate the level of neuropathology in the clinical setting.

The development of animal models for the study of brain disorders in general has been important for the progress of neuroscience research. Initially, nonspecific lesions in different brain areas of mice helped lead to knowledge about their function. More specific neurotoxic lesions mimicked neuropathological changes observed in clinical conditions and provided a basis both for testing hypotheses related to mechanisms of disease and evaluating restorative therapies. With the identification of the genetic causes of neurodegenerative disorders, the generation of animal models with better construct validity has been possible; however, studies in mice that replicate the genetic mutations identified in patients have seldom resulted in an exact phenocopy of clinical symptoms. In many instances, it appears that the genetic burden in an animal model needs to be exaggerated in order to produce a robust and clinically relevant phenotype. In the case of dystonia, Liang and collaborators engineered mice with alterations of the Dyt1/Tor1a gene that are more severe than what is present in patients and therefore compromised construct validity in favor of the face validity (1). A similar approach to murine model development and analysis has also been beneficial for Huntington disease (HD), a neurodegenerative movement disorder caused by an expanded CAG repeat in the gene coding for the huntingtin protein (13). In the case of HD, the animal models with severe phenotypes were obtained only when mice were engineered to express either a mutant HD protein with a much longer polyglutamine stretch than seen in any patients or a truncated and more toxic form of the protein (14). Nevertheless, the similarities found between clinical manifestations in murine models and patients, such as in the HD field, support the value of these models for examining potential disease mechanisms as well as for testing therapeutic strategies (15). Novel and sometimes unexpected discoveries made in animal models often become the foundation for studies using patient material, which can confirm the clinical relevance of the findings. Availability of a wide range of animal models with different advantages and varying degrees of validity in different domains directly affects the pace of advancement in translational and clinical research for diseases of the brain. The mouse models produced by Liang and collaborators are likely to play an important role in these endeavors in the dystonia research field.

D. Kirik is supported by the Swedish Research Council (2008-3092, 2012-2586, 2012-5854) and a European Research Council ERC Starting Grant (TreatPD, 242932). Å. Petersén is supported by the Swedish Research Council (2013-3537).

Address correspondence to: Deniz Kirik, B.R.A.I.N.S. Unit, BMC D11, 221 84 Lund, Sweden. Phone: 46.46.2220564; Fax: 46.46.2223436; E-mail: deniz.kirik@med.lu.se.

Conflict of interest: Deniz Kirik receives financial compensation in the form of retainer fees, milestones, and royalty payments from Genepod Therapeutics AB and receives research grant support from ParkCell AB.

Reference information: J Clin Invest. 2014;124(7):2848–2850. doi:10.1172/JCI76624.

See the related article at TorsinA hypofunction causes abnormal twisting movements and sensorimotor circuit neurodegeneration.

1. Liang C-C, Tanabe LM, Jou S, Chi F, Dauer WT. TorsinA hypofunction causes abnormal twisting movements and sensorimotor circuit neurodegeneration. J Clin Invest. 2014;124(7):3080–3092.
   View this article via: JCI CrossRef Google Scholar
2. Oppenheim H. Ueber eine eigenartige Krampfkrankheit des kindlichen und jugendlichen Alters (dysbasia lordotica progressiva, dystonia musculorum deformans). Neurol Centralbr. 1911;75:323–345.
3. Standaert DG. Update on the pathology of dystonia. Neurobiol Dis. 2011;42(2):148–151.
   View this article via: PubMed CrossRef Google Scholar
4. Albanese A, et al. Phenomenology and classification of dystonia: a consensus update. Mov Disord. 2013;28(7):863–873.
   View this article via: PubMed CrossRef Google Scholar
5. Kramer PL, et al. Dystonia gene in Ashkenazi Jewish population is located on chromosome 9q32-34. Ann Neurol. 1990;27(2):114–120.
   View this article via: PubMed CrossRef Google Scholar
6. Ozelius LJ, et al. The early-onset torsion dystonia gene (DYT1) encodes an ATP-binding protein. Nat Genet. 1997;17(1):40–48.
   View this article via: PubMed CrossRef Google Scholar
7. Granata A, Schiavo G, Warner TT. TorsinA and dystonia: from nuclear envelope to synapse. J Neurochem. 2009;109(6):1596–1609.
   View this article via: PubMed CrossRef Google Scholar
8. Charlesworth G, Bhatia KP, Wood NW. The genetics of dystonia: new twists in an old tale. Brain. 2013;136(pt 7):2017–2037.
   View this article via: PubMed Google Scholar
9. Rostasy K, et al. TorsinA protein and neuropathology in early onset generalized dystonia with GAG deletion. Neurobiol Dis. 2003;12(1):11–24.
   View this article via: PubMed CrossRef Google Scholar
10. Argyelan M, et al. Cerebellothalamocortical connectivity regulates penetrance in dystonia. J Neurosci. 2009;29(31):9740–9747.
    View this article via: PubMed CrossRef Google Scholar
11. Oleas J, Yokoi F, Deandrade MP, Pisani A, Li Y. Engineering animal models of dystonia. Mov Disord. 2013;28(7):990–1000.
    View this article via: PubMed CrossRef Google Scholar
12. Goodchild RE, Kim CE, Dauer WT. Loss of the dystonia-associated protein torsinA selectively disrupts the neuronal nuclear envelope. Neuron. 2005;48(6):923–932.
    View this article via: PubMed CrossRef Google Scholar
13. The Huntington’s Disease Collaborative Research Group. A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell. 1993;72(6):971–983.
    View this article via: PubMed CrossRef Google Scholar
14. Pouladi MA, Morton AJ, Hayden MR. Choosing an animal model for the study of Huntington’s disease. Nat Rev Neurosci. 2013;14(10):708–721.
    View this article via: PubMed CrossRef Google Scholar
15. Crook ZR, Housman D. Huntington’s disease: can mice lead the way to treatment? Neuron. 2011;69(3):423–435.
    View this article via: PubMed CrossRef Google Scholar