Monoclonal Antibodies Recognize Distinct Conformational Epitopes Formed by Polyglutamine in a Mutant Huntingtin Fragment

Huntington disease (HD) is a neurodegenerative disorder caused by an expansion of a polyglutamine (polyQ) domain in the N-terminal region of huntingtin (htt). PolyQ expansion above 35–40 results in disease associated with htt aggregation into inclusion bodies. It has been hypothesized that expanded polyQ domains adopt multiple potentially toxic conformations that belong to different aggregation pathways. Here, we used atomic force microscopy to analyze the effect of a panel of anti-htt antibodies (MW1–MW5, MW7, MW8, and 3B5H10) on aggregate formation and the stability of a mutant htt-exon1 fragment. Two antibodies, MW7 (polyproline-specific) and 3B5H10 (polyQ-specific), completely inhibited fibril formation and disaggregated preformed fibrils, whereas other polyQ-specific antibodies had widely varying effects on aggregation. These results suggest that expanded polyQ domains adopt multiple conformations in solution that can be readily distinguished by monoclonal antibodies, which has important implications for understanding the structural basis for polyQ toxicity and the development of intrabody-based therapeutics for HD.Huntington disease (HD)⁵ is a fatal neurodegenerative disorder that is caused by an expansion of a polyglutamine (polyQ) domain in the protein huntingtin (htt), which leads to its aggregation into fibrils (1). HD is part of a growing group of diseases that are classified as “conformational diseases,” which include Alzheimer disease (AD), Parkinson disease (PD), the prion encephalopathies, and many more (2–4). The length of polyQ expansion in HD is tightly correlated with disease onset, and a critical threshold of 35–40 glutamine residues is required for disease manifestation (5). Biochemical and electron microscopic studies with htt fragments demonstrated that expanded polyQ repeats (>39) form detergent-insoluble aggregates that share characteristics with amyloid fibrils (6–8), and the formation of amyloid-like fibrils by polyQ was confirmed by studies with synthetic polyQ peptides (9). Collectively, these studies demonstrated a correlation between polyQ length and the kinetics of aggregation. This phenomenon has been recapitulated in cell-culture models that express htt fragments (10–12). Although it is clear that proteins with expanded polyQ repeats assemble into fibrils in vitro, recent studies have reported that htt fragments can also assemble into spherical and annular oligomeric structures (13–16) similar to those formed by Aβ and α-synuclein, which are implicated in AD and PD, respectively.While the major hallmark of HD is the formation of intranuclear and cytoplasmic inclusion bodies of aggregated htt (17), the role of these structures in the etiology of HD remains controversial. For instance, the onset of symptoms in a transgenic mouse model of HD follows the appearance of inclusion bodies (18), while other studies indicate that inclusion body formation may protect against toxicity by sequestering diffuse, soluble forms of htt (10, 19, 20). Based on the direct correlation between polyQ length, htt aggregation propensity, and toxicity (6), it has been hypothesized that the aggregation of htt may mediate neurodegeneration in HD. However, there is no clear consensus on the aggregate form(s) that underlie toxicity, and there likely exist bioactive, oligomeric aggregates undetectable by traditional biochemical and electron microscopic approaches whose formation precedes disease symptoms. Although identification of the one or more toxic species of htt that trigger neurodegeneration in HD remains elusive, such species might exist in a diffuse, mobile fraction rather than in inclusion bodies (19). A thioredoxin-polyQ fusion protein was recently reported to exhibit toxicity in a meta-stable, β-sheet-rich, monomeric conformation (21), suggesting that polyQ can adopt multiple monomeric conformations, only some of which may be toxic. Consistent with such a scenario, molecular dynamic simulations and fluorescence correlation spectroscopy experiments with synthetic polyQ peptides indicate that polyQ domains can adopt a heterogeneous collection of collapsed conformations that are in equilibrium before aggregation (22–25).Although biochemical, biophysical, and computational approaches have yielded insight into the structures formed by polyQ in vitro, whether such structures form in vivo remains largely unknown. Indeed, determining the conformational state of any misfolded/aggregated protein in situ and/or in vivo remains a major technical challenge.Toward this goal, antibodies have been explored as a potentially powerful tool for detecting specific conformations or multimeric states of aggregated proteins in situ. Antibodies specific for amyloid fibrils often do not react with natively folded globular proteins from which they are derived, suggesting that such antibodies recognize a conformational epitope (26, 27). Several antibodies display conformation-dependent interactions with amyloids, aggregation intermediates, or natively folded precursor proteins. For example, there are antibodies specific for paired helical filaments of Tau (28–31), of aggregated forms of Aβ ranging from dimers to fibrils (32–34), and of native (35) or disease-related (36) forms of the prion protein. Antibodies have also been developed that are specific for common structural motifs associated with amyloid diseases, such as oligomers (37) and fibrils (38), independent of the peptide sequence of the amyloid forming protein from which they are derived, suggesting the potential for a common mechanism of aggregation and toxicity for these diseases.With regard to htt, several antibodies (MW1, MW2, MW3, MW4, MW5, IC2, and IF8), which are specific for polyQ repeats, stain Western blots of htt with expanded polyQ repeats much more strongly than htt with normal polyQ length (39, 40), suggesting that these antibodies may recognize abnormal polyQ conformations. Furthermore, these polyQ-specific antibodies have distinct staining patterns in immunohistochemical studies of brain tissue sections (39). In one study, the affinity and stoichiometry of MW1 binding to htt increased with polyQ length, suggesting a “linear lattice” model for polyQ (41). This model is supported by a crystal structure of polyQ bound to MW1, which showed that polyQ can adopt an extended, coil-like structure (42). However, an independent structural study showed that the anti-polyQ antibody 3B5H10 binds to a compact β-sheet-like structure of polyQ in a monomeric htt fragment.⁶ These results clearly indicate that polyQ domains can fold into at least two unique, stable, monomeric conformations and suggest that the “linear lattice” model is not generally applicable to all polyQ structures.Not only are antibodies useful for understanding what polyQ structures exist in situ, especially in the diffuse htt fraction of neurons, but antibodies and/or intrabodies may also have potential as therapeutic agents. For example, several studies showed that intrabodies reduce htt toxicity in cellular models (44–49). Moreover, one intrabody (C4) slows htt aggregation and prolongs lifespan in a Drosophila model of HD (50, 51), while another (mEM48) ameliorates neurological symptoms in a mouse model of HD (48).Three of the antibodies examined in this study (MW1, MW2, and MW7) modulate htt-induced cell death when co-transfected as single-chain variable region fragment antibodies (scFvs) in 293 cells with htt exon 1 containing an expanded polyQ domain (46). In these studies MW1 and MW2, which bind to the polyQ repeat in htt, increased htt-induced toxicity and aggregation (46). Conversely, MW7, which binds to the polyproline (polyP) regions adjacent to the polyQ repeat in htt, decreased its aggregation and toxicity (46). Interestingly, MW7 has also been shown to increase the turnover of mutant htt in cultured cells and reduce its toxicity in corticostriatal brain slice explants (49).Given the difficulty in understanding which specie(s) of htt exist and mediate pathogenesis in the putative toxic diffuse fraction of neurons, we sought to rigorously characterize the conformational specificity of a panel of anti-htt antibodies, the best in situ probes currently available for distinguishing specie(s) of htt. We reasoned that if htt can adopt multiple conformations that mediate different aggregation pathways, then anti-htt antibodies should differentially alter htt aggregation pathways by stabilizing or sequestering the specific conformers or aggregates they recognize. We therefore examined the effects of various antibodies on mutant htt fragment fibril formation and stability by atomic force microscopy (AFM). Our results are consistent with the hypothesis that monoclonal antibodies recognize distinct conformational epitopes formed by polyQ in a mutant htt fragment.     

Analysis of Free Energy Signals Arising from Nucleotide Hybridization Between rRNA and mRNA Sequences during Translation in Eubacteria

A decoding algorithm is tested that mechanistically models the progressive alignments that arise as the mRNA moves past the rRNA tail during translation elongation. Each of these alignments provides an opportunity for hybridization between the single-stranded, -terminal nucleotides of the 16S rRNA and the spatially accessible window of mRNA sequence, from which a free energy value can be calculated. Using this algorithm we show that a periodic, energetic pattern of frequency 1/3 is revealed. This periodic signal exists in the majority of coding regions of eubacterial genes, but not in the non-coding regions encoding the 16S and 23S rRNAs. Signal analysis reveals that the population of coding regions of each bacterial species has a mean phase that is correlated in a statistically significant way with species () content. These results suggest that the periodic signal could function as a synchronization signal for the maintenance of reading frame and that codon usage provides a mechanism for manipulation of signal phase.[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32]     

Algorithms for Finding Small Attractors in Boolean Networks

A Boolean network is a model used to study the interactions between different genes in genetic regulatory networks. In this paper, we present several algorithms using gene ordering and feedback vertex sets to identify singleton attractors and small attractors in Boolean networks. We analyze the average case time complexities of some of the proposed algorithms. For instance, it is shown that the outdegree-based ordering algorithm for finding singleton attractors works in time for , which is much faster than the naive time algorithm, where is the number of genes and is the maximum indegree. We performed extensive computational experiments on these algorithms, which resulted in good agreement with theoretical results. In contrast, we give a simple and complete proof for showing that finding an attractor with the shortest period is NP-hard.[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32]     

Arachidonic Acid Stimulates Cell Adhesion through a Novel p38 MAPK-RhoA Signaling Pathway That Involves Heat Shock Protein 27

Rho GTPases are critical components of cellular signal transduction pathways. Both hyperactivity and overexpression of these proteins have been observed in human cancers and have been implicated as important factors in metastasis. We previously showed that dietary n-6 fatty acids increase cancer cell adhesion to extracellular matrix proteins, such as type IV collagen. Here we report that in MDA-MB-435 human melanoma cells, arachidonic acid activates RhoA, and inhibition of RhoA signaling with either C3 exoenzyme or dominant negative Rho blocked arachidonic acid-induced cell adhesion. Inhibition of the Rho kinase (ROCK) with either small molecule inhibitors or ROCK II-specific small interfering RNA (siRNA) blocked the fatty acid-induced adhesion. However, unlike other systems, inhibition of ROCK did not block the activation of p38 mitogen-activated protein kinase (MAPK); instead, Rho activation depended on p38 MAPK activity and the presence of heat shock protein 27 (HSP27), which is phosphorylated downstream of p38 after arachidonic acid treatment. HSP27 associated with p115RhoGEF in fatty acid-treated cells, and this association was blocked when p38 was inhibited. Furthermore, siRNA knockdown of HSP27 blocked the fatty acid-stimulated Rho activity. Expression of dominant negative p115-RhoGEF or p115RhoGEF-specific siRNA inhibited both RhoA activation and adhesion on type IV collagen, whereas a constitutively active p115RhoGEF restored the arachidonic acid stimulation in cells in which the p38 MAPK had been inhibited. These data suggest that n-6 dietary fatty acids stimulate a set of interactions that regulates cell adhesion through RhoA and ROCK II via a p38 MAPK-dependent association of HSP27 and p115RhoGEF.The ability of tumor cells to metastasize to secondary sites is a hallmark of neoplastic disease. Unfortunately, this propensity to spread is the primary cause of morbidity and death in cancer patients (1). Metastasis is clearly a highly regulated, multistep process that occurs in a spatiotemporal manner (2–4). To escape the restrictive compartment boundaries characteristic of adult tissue, separate intravasation and extravasation steps requiring alterations in co-adhesion, adhesion, invasion, and migration must occur. Execution of these biological processes, involving multiple proteins and cellular organelles, require highly coordinated cell signaling mechanisms.The Rho family of small GTPases regulates many facets of cytoskeletal rearrangements that facilitate cell attachment and migration (5–7). Rho GTPases act as molecular switches by changing from an inactive GDP-bound conformation to an active GTP-bound conformation, thereby regulating a signaling pathway. These proteins are directly regulated by Rho guanine nucleotide exchange factors (GEFs),² Rho GTPase activating proteins, and Rho GDP-dissociation inhibitors (8–12). RhoGEFs bind to the GTPase to catalyze the dissociation of GDP, allowing the binding of GTP and thereby promoting Rho activation (8). The RGS (regulators of G protein signaling) domain-containing RhoGEFs are a recently described family of GEFs. Currently, there are three members of this family, PDZ-RhoGEF, LARG, and p115RhoGEF (13–15), in which the RGS domains function as a heterotrimeric GTPase-activating domain (13, 15, 16). The RGS family of RhoGEFs has been shown to regulate Rho during several processes including cytoskeletal rearrangements, cell adhesion, and cancer progression (17–21).There is significant interplay between the activity of small GTPases and signaling derived from fatty acid metabolism (22–28). Linoleic acid, which is metabolized to arachidonic acid, is an n-6 polyunsaturated fatty acid that is present at high levels in most western diets (29). In animal models, diets high in n-6 polyunsaturated fatty acids have been shown to enhance tumor progression and metastasis (30, 31). Additionally, arachidonic acid is stored in cell membranes and is made available by phospholipases under conditions of increased inflammatory response (32). Arachidonic acid is further metabolized by cyclooxygenases (COX), lipoxygenases (LOX), and cytochrome P450 monooxygenases to yield bioactive products that have myriad effects on cells, and altered metabolism of arachidonic acid by COX, LOX, and P450 has been implicated in cancer progression (31, 33–36).We have studied mechanisms of cell adhesion using the MDA-MB-435 cells as a model of a highly metastatic human cancer cell line (37). These cells have been extensively studied for their ability to recapitulate the metastatic cascade in vivo and in vitro, although recent work indicates that the cells currently in use are most likely a human melanoma line (38). We initially observed that arachidonic acid (AA) enhanced adhesion of MDA-MB-435 cells to type IV collagen through specific integrin-mediated pathways (37). Exogenous AA led to the activation of mitogen-activated protein kinase (MAPK)-activated protein kinase 2 and the phosphorylation of heat shock protein 27 (HSP27) via a p38 MAPK-dependent process (39). Inhibition of p38 MAPK activation blocked cell adhesion as did function-blocking antibodies specific for subunits of the collagen receptor (40). More recently, we identified the key metabolite of AA (15-(S)- hydroxyeicosatetraenoic acid) and the upstream kinases (TAK1 and MKK6) that are responsible for activation of p38 MAPK in this system (41).In this study we investigated the role of Rho activation in the MDA-MB-435 cells after exposure to arachidonic acid. Several aspects of the regulation of Rho signaling in these cells provide insights into the cross-talk between important signaling pathways.     

Phosphorylation Dependence of Hsp27 Multimeric Size and Molecular Chaperone Function

The molecular chaperone Hsp27 exists as a distribution of large oligomers that are disassembled by phosphorylation at Ser-15, -78, and -82. It is controversial whether the unphosphorylated Hsp27 or the widely used triple Ser-to-Asp phospho-mimic mutant is the more active molecular chaperone in vitro. This question was investigated here by correlating chaperone activity, as measured by the aggregation of reduced insulin or α-lactalbumin, with Hsp27 self-association as monitored by analytical ultracentrifugation. Furthermore, because the phospho-mimic is generally assumed to reproduce the phosphorylated molecule, the size and chaperone activity of phosphorylated Hsp27 were compared with that of the phospho-mimic. Hsp27 was triply phosphorylated by MAPKAP-2 kinase, and phosphorylation was tracked by urea-PAGE. An increasing degree of suppression of insulin or α-lactalbumin aggregation correlated with a decreasing Hsp27 self-association, which was the least for phosphorylated Hsp27 followed by the mimic followed by the unphosphorylated protein. It was also found that Hsp27 added to pre-aggregated insulin did not reverse aggregation but did inhibit these aggregates from assembling into even larger aggregates. This chaperone activity appears to be independent of Hsp27 phosphorylation. In conclusion, the most active chaperone of insulin and α-lactalbumin was the Hsp27 (elongated) dimer, the smallest Hsp27 subunit observed under physiological conditions. Next, the Hsp27 phospho-mimic is only a partial mimic of phosphorylated Hsp27, both in self-association and in chaperone function. Finally, the efficient inhibition of insulin aggregation by Hsp27 dimer led to the proposal of two models for this chaperone activity.Oligomeric heat shock protein 27 (Hsp27)² is a ubiquitous mammalian protein with a variety of functions in health and disease (1–8). These functions include ATP-independent chaperone activity in response to environmental stress, e.g. heat shock and oxidative stress, control of apoptosis, and regulation of actin cytoskeleton dynamics. Hsp27 is a member of the α-crystallin small heat shock protein family of which αB-crystallin is the archetype. These proteins are characterized by an α-crystallin domain of 80–90 residues consisting of roughly eight β-strands that form an intermolecular β-sheet interaction interface within a dimer, the basic building subunit of the oligomer (2, 4, 9–11).Hsp27 is in equilibrium between high molecular weight oligomers and much lower molecular weight multimers. It has been reported that unphosphorylated Hsp27 includes predominantly a distribution of high molecular species ranging in size from 12-mer to 35-mer (12–19). Phosphorylation of Hsp27 at serines 15, 78, and 82 by the p38-activated MAPKAP-2 kinase (20–22) or the use of the triple Ser-to-Asp phospho-mimic results in a major shift in the equilibrium toward much smaller multimers (23) and in an alteration of its function (1, 3, 6, 7, 24, 25). The size distribution of the smaller species has been reported to be between monomer and tetramer (12–16, 18, 19).Small heat shock proteins, including Hsp27, behave as ATP-independent molecular chaperones during cellular heat shock. They bind partially unfolded proteins and prevent their aggregation until the proteins can be refolded by larger ATP-dependent chaperones or are digested (7, 8, 26). This function includes the up-regulation and/or phosphorylation of Hsp27.It is not entirely clear what the role of Hsp27 size and phosphorylation state plays in its heat shock function because there are conflicting results in the literature. Some in vitro studies concluded that the unphosphorylated oligomeric Hsp27 (or the murine isoform Hsp25) protects proteins against aggregation better than does the phosphorylation mimic (13, 19, 27), whereas others found no difference (16, 28, 29), and still other studies found that the mimic protects better than does the unphosphorylated wild type (27, 30, 31). In-cell studies found that phosphorylation of Hsp27 was essential for thermo-protection of actin filaments (32), and the Hsp27 phosphorylation mimic decreased inclusion body formation better than did unphosphorylated Hsp27 (33). This study was undertaken to investigate the molecular chaperone function of Hsp27 by correlating chaperone activity with Hsp27 size and by comparing fully phosphorylated Hsp27 with its phospho-mimic.     

The Wavelet-Based Cluster Analysis for Temporal Gene Expression Data

A variety of high-throughput methods have made it possible to generate detailed temporal expression data for a single gene or large numbers of genes. Common methods for analysis of these large data sets can be problematic. One challenge is the comparison of temporal expression data obtained from different growth conditions where the patterns of expression may be shifted in time. We propose the use of wavelet analysis to transform the data obtained under different growth conditions to permit comparison of expression patterns from experiments that have time shifts or delays. We demonstrate this approach using detailed temporal data for a single bacterial gene obtained under 72 different growth conditions. This general strategy can be applied in the analysis of data sets of thousands of genes under different conditions.[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29]     

Multipattern Consensus Regions in Multiple Aligned Protein Sequences and Their Segmentation

Decomposing a biological sequence into its functional regions is an important prerequisite to understand the molecule. Using the multiple alignments of the sequences, we evaluate a segmentation based on the type of statistical variation pattern from each of the aligned sites. To describe such a more general pattern, we introduce multipattern consensus regions as segmented regions based on conserved as well as interdependent patterns. Thus the proposed consensus region considers patterns that are statistically significant and extends a local neighborhood. To show its relevance in protein sequence analysis, a cancer suppressor gene called p53 is examined. The results show significant associations between the detected regions and tendency of mutations, location on the 3D structure, and cancer hereditable factors that can be inferred from human twin studies.[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27]     

Splitting the BLOSUM Score into Numbers of Biological Significance

Mathematical tools developed in the context of Shannon information theory were used to analyze the meaning of the BLOSUM score, which was split into three components termed as the BLOSUM spectrum (or BLOSpectrum). These relate respectively to the sequence convergence (the stochastic similarity of the two protein sequences), to the background frequency divergence (typicality of the amino acid probability distribution in each sequence), and to the target frequency divergence (compliance of the amino acid variations between the two sequences to the protein model implicit in the BLOCKS database). This treatment sharpens the protein sequence comparison, providing a rationale for the biological significance of the obtained score, and helps to identify weakly related sequences. Moreover, the BLOSpectrum can guide the choice of the most appropriate scoring matrix, tailoring it to the evolutionary divergence associated with the two sequences, or indicate if a compositionally adjusted matrix could perform better.[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29]     

Printor, a Novel TorsinA-interacting Protein Implicated in Dystonia Pathogenesis

Early onset generalized dystonia (DYT1) is an autosomal dominant neurological disorder caused by deletion of a single glutamate residue (torsinA ΔE) in the C-terminal region of the AAA⁺ (ATPases associated with a variety of cellular activities) protein torsinA. The pathogenic mechanism by which torsinA ΔE mutation leads to dystonia remains unknown. Here we report the identification and characterization of a 628-amino acid novel protein, printor, that interacts with torsinA. Printor co-distributes with torsinA in multiple brain regions and co-localizes with torsinA in the endoplasmic reticulum. Interestingly, printor selectively binds to the ATP-free form but not to the ATP-bound form of torsinA, supporting a role for printor as a cofactor rather than a substrate of torsinA. The interaction of printor with torsinA is completely abolished by the dystonia-associated torsinA ΔE mutation. Our findings suggest that printor is a new component of the DYT1 pathogenic pathway and provide a potential molecular target for therapeutic intervention in dystonia.Early onset generalized torsion dystonia (DYT1) is the most common and severe form of hereditary dystonia, a movement disorder characterized by involuntary movements and sustained muscle spasms (1). This autosomal dominant disease has childhood onset and its dystonic symptoms are thought to result from neuronal dysfunction rather than neurodegeneration (2, 3). Most DYT1 cases are caused by deletion of a single glutamate residue at positions 302 or 303 (torsinA ΔE) of the 332-amino acid protein torsinA (4). In addition, a different torsinA mutation that deletes amino acids Phe³²³–Tyr³²⁸ (torsinA Δ323–328) was identified in a single family with dystonia (5), although the pathogenic significance of this torsinA mutation is unclear because these patients contain a concomitant mutation in another dystonia-related protein, ϵ-sarcoglycan (6). Recently, genetic association studies have implicated polymorphisms in the torsinA gene as a genetic risk factor in the development of adult-onset idiopathic dystonia (7, 8).TorsinA contains an N-terminal endoplasmic reticulum (ER)³ signal sequence and a 20-amino acid hydrophobic region followed by a conserved AAA⁺ (ATPases associated with a variety of cellular activities) domain (9, 10). Because members of the AAA⁺ family are known to facilitate conformational changes in target proteins (11, 12), it has been proposed that torsinA may function as a molecular chaperone (13, 14). TorsinA is widely expressed in brain and multiple other tissues (15) and is primarily associated with the ER and nuclear envelope (NE) compartments in cells (16–20). TorsinA is believed to mainly reside in the lumen of the ER and NE (17–19) and has been shown to bind lamina-associated polypeptide 1 (LAP1) (21), lumenal domain-like LAP1 (LULL1) (21), and nesprins (22). In addition, recent evidence indicates that a significant pool of torsinA exhibits a topology in which the AAA⁺ domain faces the cytoplasm (20). In support of this topology, torsinA is found in the cytoplasm, neuronal processes, and synaptic terminals (2, 3, 15, 23–26) and has been shown to bind cytosolic proteins snapin (27) and kinesin light chain 1 (20). TorsinA has been proposed to play a role in several cellular processes, including dopaminergic neurotransmission (28–31), NE organization and dynamics (17, 22, 32), and protein trafficking (27, 33). However, the precise biological function of torsinA and its regulation remain unknown.To gain insights into torsinA function, we performed yeast two-hybrid screens to search for torsinA-interacting proteins in the brain. We report here the isolation and characterization of a novel protein named printor (protein interactor of torsinA) that interacts selectively with wild-type (WT) torsinA but not the dystonia-associated torsinA ΔE mutant. Our data suggest that printor may serve as a cofactor of torsinA and provide a new molecular target for understanding and treating dystonia.