- Top of page
Early onset torsion dystonia, the most common form of hereditary primary dystonia, is caused by a mutation in the TOR1A gene, which codes for the protein torsinA. This form of dystonia is referred to as DYT1. We have used a transgenic mouse model of DYT1 dystonia [human mutant-type (hMT)1 mice] to examine the effect of the mutant human torsinA protein on striatal dopaminergic function. Analysis of striatal tissue dopamine (DA) and metabolites using HPLC revealed no difference between hMT1 mice and their non-transgenic littermates. Pre-synaptic DA transporters were studied using in vitro autoradiography with [3H]mazindol, a ligand for the membrane DA transporter, and [3H]dihydrotetrabenazine, a ligand for the vesicular monoamine transporter. No difference in the density of striatal DA transporter or vesicular monoamine transporter binding sites was observed. Post-synaptic receptors were studied using [3H]SCH-23390, a ligand for D1 class receptors, [3H]YM-09151-2 and a ligand for D2 class receptors. There were again no differences in the density of striatal binding sites for these ligands. Using in vivo microdialysis in awake animals, we studied basal as well as amphetamine-stimulated striatal extracellular DA levels. Basal extracellular DA levels were similar, but the response to amphetamine was markedly attenuated in the hMT1 mice compared with their non-transgenic littermates (253 ± 71% vs. 561 ± 132%, p < 0.05, two-way anova). These observations suggest that the mutation in the torsinA protein responsible for DYT1 dystonia may interfere with transport or release of DA, but does not alter pre-synaptic transporters or post-synaptic DA receptors. The defect in DA release as observed may contribute to the abnormalities in motor learning as previously documented in this transgenic mouse model, and may contribute to the clinical symptoms of the human disorder.
Dystonia is a disorder characterized by sustained muscle contractions, frequently causing twisting and repetitive movements, or abnormal postures (Jankovic and Fahn 1993). The most common type of early onset generalized dystonia is an autosomal dominant disorder caused by a 3-bp deletion (ΔGAG) in the TOR1A gene that encodes torsinA (Ozelius et al. 1997). Symptoms of this disorder typically develop before the age of 21 years and include involuntary sustained muscle contractions that cause abnormal posturing of a foot, leg or arm, which frequently generalize to other body regions (Bressman et al. 2002). The pathophysiologic mechanisms, by which the ΔGAG mutation causes dystonia is unclear. There is no evidence for neurodegeneration in the limited number of postmortem examinations of DYT1 patients that have been conducted, although enlargement of dopamine (DA) neurons (Rostasy et al. 2003) and ubiquitin inclusions (McNaught et al. 2004) have been reported. The affected protein, torsinA, shares homology with members of the AAA + ATPase gene family (Lupas et al. 1997; Neuwald et al. 1999). Members of this family participate in numerous cellular functions, including protein folding and degradation, membrane trafficking and vesicle fusion (Ogura and Wilkinson 2001).
Genetic and pharmacological studies have led to the hypothesis that at least some forms of dystonia arise from disturbances in striatal dopaminergic signaling. Mutations within the tyrosine hydroxylase gene (Knappskog et al. 1995) and the GTP cyclohydrolase gene (Ichinose et al. 1994), both of which limit DA synthesis, result in a dystonic phenotype in humans. In addition, mechanical or ischemic lesions of the striatum and administration of pharmacological agents that block DA D2 receptors in vivo can result in a dystonic phenotype (Rupniak et al. 1986,Skidmore and Reich 2005). Several types of hereditary dystonia have Parkinsonian features and early onset Parkinson’s disease can present as dystonia (Leung et al. 2001; Klien et al. 2007). In human brain, the mRNA for torsinA is strongly expressed in dopaminergic neurons of the substantia nigra, although it is also found in other regions of the brain including the hippocampus and cerebellum (Augood et al. 1999). Positron emission tomography has demonstrated reduced binding of D2 ligands in human carriers of the DYT1 mutation (Asanuma et al. 2005). Our laboratory conducted a study, which we directly examined DA metabolism in postmortem human brain from three individuals with the DYT1 mutation and found no change in striatal DA content, but there was an alteration in the ratio of dihydroxyphenyl acetic acid (DOPAC) to DA (Augood et al. 2004). This alteration suggests a change in DA turnover, but the number of cases available for study was small and thus the strength of this conclusion is limited.
A transgenic mouse model which reproduces some features of the human disorder has been described previously (Sharma et al. 2005). These mice express either human wild-type or human mutant-type (hMT) torsinA under the control of a constitutive cytomegalovirus promoter, as well as endogenous wild-type mouse torsinA under its endogenous promoter. They do not have overt dystonia, but do demonstrate a defect in motor learning comparable with that seen in human non-manifesting carriers of the DYT1 mutation (Sharma et al. 2005). We have used one of these mouse models (hMT1 mice) to examine the effect of the hMT torsinA protein on tissue levels of DA and its metabolites in the striatum using HPLC with electrochemical detection. We also examined pre-synaptic DA transporters [both DA transporter (DAT) and vesicular monoamine transporter (VMAT2)] and post-synaptic DA receptors (D1 and D2 class) using in vitro ligand binding methods. Moreover, we examined basal and amphetamine-stimulated striatal extracellular DA levels in these animals using in vivo microdialysis. Our data show that presence of the hMT torsinA does not alter the tissue content of DA or total binding densities of pre-synaptic transport sites or post-synaptic striatal DA receptors, but does result in impaired release of DA after amphetamine treatment. These observations suggest that the mutation of torsinA protein may interfere with stimulated striatal DA signaling, and that this defect may contribute both to the motor learning defect previously documented in this mouse model, as well as to the clinical symptoms of the disorder in humans.
- Top of page
Using a mouse model of human DYT1 dystonia, we observed that expression of human mutant torsinA leads to an impairment of amphetamine-induced DA release in the striatum. This effect cannot be attributed to an impairment in the synthesis or storage of DA, as no reduction in the striatal content of DA or its metabolites was observed; indeed, the data revealed a trend towards an increased content of DA, DOPAC and HVA in the hMT1 mice. In addition, we found no significant differences in the density of pre-synaptic DA uptake sites (DAT or VMAT2), or post-synaptic striatal DA receptors between transgenic mice and their non-transgenic littermates. These observations suggest that the primary defect in DA metabolism produced by the presence of the mutant form of the human torsinA protein is an abnormality of DA release.
As recently reviewed by Sulzer et al. (2005), amphetamine produces DA release by displacing DA from pre-synaptic vesicles. This leads to elevated cytoplasmic DA levels in pre-synaptic terminals, and release into the extracellular space by reversal of the normal action of the DAT. This is of interest because several lines of evidence have pointed to an interaction between torsinA and expression of VMAT and DAT. In HEK293 cells, co-expression of torsinA with DAT reduces [3H]-DA uptake by decreasing the level of DAT present at the plasma membrane (Torres et al. 2004). In Caenorhabditis elegans, over-expression of the torsinA homolog Tor2 leads to down-regulation of DAT expression (Cao et al. 2005), while in SH-SY5Y neuroblastoma cells over-expression of mutant, but not wild-type, torsinA leads to entrapment of VMAT and torsinA in membranous structures (Misbahuddin et al. 2005). In this study, we used binding of [3H]-mazindol to study DAT sites and [3H]-dihydrotetrabenazine to study VMAT2 sites in the striatum of our transgenic mouse model (hMT1 mice). We found that neither [3H]-mazindol sites nor [3H]-dihydrotetrabenazine sites were altered by the expression of human mutant torsinA. However, it is important to note that this assay does not distinguish between surface and intracellular ligand binding sites. Thus, it is possible that the altered release of DA in response to amphetamine in these animals arises from an altered surface expression of DAT or altered subcellular distribution of VMAT2, which were not detected using the ligand binding assays.
In vivo microdialysis is the most direct way to assess extracellular DA in an intact animal, but it is necessarily an invasive technique and there is a possibility that local tissue injury contributes to measurement variability (Westerink 1995; Kuczenski et al. 1997; Carboni et al. 2001). The variability of DA release in response to amphetamine, which we observed is typical of this type of experiment, and the magnitude of inhibition of DA release in the hMT1 mice is similar to that seen with other manipulations of dopaminergic neuron function, such as knockout of GABAB1 receptor function (Vacher et al. 2006).
Several other mouse models of DYT1 dystonia have been described. In a model in which the mutant torsinA protein was over-expressed under control of a neuron specific enolase promoter (Shashidharan et al. 2005), a variable phenotype was observed with some animals exhibiting hyperactivity, while others are behaviorally normal. In these animals, tissue levels of DA were differentially altered, with a small increase in the non-symptomatic animals and decrease in symptomatic mice, while levels of DOPAC and HVA were not altered. In another DYT1 mouse model in which endogenous knock-in ΔGAG mutation was introduced, hyperactivity and impaired beam walking were observed (Dang et al. 2005). In these knock-in animals, tissue levels of DA were not different between groups but there was a slight reduction in HVA. Similar results were obtained in a knock-down mouse model with reduced expression of endogenous torsinA, which did not alter DA content but did slightly reduce DOPAC content (Dang et al. 2006). In the limited studies of human DYT1 dystonia that have been reported, no change in striatal DA content has been observed (Augood et al. 2004;Furukawa et al. 2000). Collectively, these data support the view that in animals as well as humans, neither the presence of mutant torsinA nor a partial deficiency of the normal protein has a substantial effect on the synthesis or storage of DA in striatal pre-synaptic terminals. Our data suggest that the major defect in our animal model is in the dynamic properties of DA release, as demonstrated by the impairment in amphetamine-stimulated release, which we have observed. It will be important to determine whether a similar abnormality of release is present in the other mouse models of DYT1 dystonia which have been described.
The mouse model we have studied does not have overt dystonia, but does demonstrate a reduced ability to learn motor skills in an accelerating rotarod paradigm (Sharma et al. 2005). This defect may be similar to the abnormalities observed in the ‘non-manifesting carrier’ state of the human disorder. Only about 30% of those who carry the DYT1 mutation develop dystonia, but those who carry the gene but lack overt symptoms nevertheless have demonstrable abnormalities, which include a defect in motor learning ability and an altered pattern of cerebral recruitment when learning new motor tasks (Ghilardi et al. 2003). DA release is essential for motor learning, and DA impairment is associated with impaired rotarod performance in rodents (Ogura et al. 2005 and Mizoguchi et al. 2002). Thus, impaired DA release similar to that which we have observed in animals expressing the human mutant torsinA protein may contribute to the motor learning abnormalities of human carriers of the DYT1 mutation.
The identification of dopaminergic abnormalities in animal models of DYT1 dystonia is important, because this defect may contribute to clinical symptoms of the disorder and it suggests alternative therapeutic approaches to this disorder. If in fact presence of the mutant form of torsinA in humans leads to dysregulation of DA release, then strategies to augment release or even therapies directed at activation of post-synaptic DA receptors may prove beneficial.