Researchers Reveal a New Twist in Torsion Dystonia

Courtesy of Guy Caldwell CLEARING THE DRAIN: (Left) The worm expresses polyglutamine GFP in both the presence and absence of torsin. (Right) A Caenorhabditis elegans torsin protein, TOR-2 (red), localizes to sites of polyglutamine green fluorescent protein (GFP) aggregation (green) in a worm cell. A movement disorder can start as a twinge. A child's leg turns in while walking. Writing becomes difficult, painful. For many, these types of diseases--broadly termed dystonias--progress no further than persistent muscle cramps. Yet in many children affected by rare, heritable, early-onset dystonia, a generalized movement disorder called torsion dystonia develops as well. The disorder can affect the entire body: Opposing muscles work against each other, twisting the posture, causing repetitive movements, or contorting arms and legs into unnatural positions. Oftentimes, the earlier in life symptoms appear, the worse they get. To uncover the roots of this dominant trait, which has only 30% to 40% penetrance, researchers spent more than 15 years studying afflicted, diverse families and a population of Ashkenazi Jews to zero in on a responsible mutation: a three-basepair deletion in DYT1 that appears exactly the same across ethnic groups.1 Recent efforts to elucidate the function for torsinA--the protein this gene encodes--reveal that torsins may act as molecular chaperones, altering protein folding and clearing aggregates from living cells. In worms, torsins alleviate engineered poly-glutamine clumps,2 and in cell culture, torsinA suppresses a-synuclein aggregation.3 These results and others reveal ties to troublesome neurodegenerative disorders such as Parkinson, Huntington, and Alzheimer disease. Protein aggregates are hallmarks of these diseases, but their role remains controversial. Likewise, torsinA's role in clearing the aggregates is murky, and such clumps are not even found in patients with dystonia. Yet, some draw parallels between torsion dystonia and Parkinson disease: Uncontrollable movements are common to both, and some treatments, including deep-brain stimulation, seem to ameliorate symptoms, even though dystonia is not a neurodegenerative disease. Others researching dystonia reason that, at least, the connection adds political clout to their work on an orphan disease. "I see [patients] suffer, and just because they don't die from it, just because nobody famous has it ... doesn't mean that it's less worthy a problem," says Laurie Ozelius, who cloned DYT1 in 1997.1 The first clues to torsinA's function came from its sequence. Torsins look similar to chaperone proteins, and they've been placed in the family called AAA+, for ATPases Associated with a variety of cellular Activities. "That very obtuse name tells you that these things have tons of different functions, and they're all lumped together based on some limited sequence homology," says Ozelius, now at Albert Einstein College of Medicine in New York. And indeed, researchers invoke torsins in cell polarity, membrane trafficking, and neurotransmission, in addition to protein folding and aggregate degradation. Yet no one is certain which, if any, of these functions might play a role in dystonia symptoms. MOLECULAR DRANO? Guy Caldwell, an assistant professor at the University of Alabama, recently claimed a function for torsins. He engineered Caenorhabditis elegans to produce high amounts of polyglutamine repeats fused to green fluorescent protein; GFPs with an 82-glutamine stretch clump together in the cytoplasm of muscle cells. Concurrent overexpression of the gene tor-2, which encodes a worm torsin, alleviates the clumps in vivo, but a mutant of the gene does not. Additionally, human torsinA disrupts the aggregates.2 Previously, a research group at Massachusetts General Hospital demonstrated torsinA chewing through a-synuclein aggregates in cell culture.3 Xandra Breakefield, who is also at Harvard University, explains, "If you overexpress the wild-type torsin, you get fewer cells with aggregates. If you express the mutant torsin, you don't affect the a-synuclein aggregates and you get separate mutant torsin aggregates." Ozelius, who worked with Breakefield while cloning DYT1, says this is not wholly unexpected because heat-shock proteins, which torsins resemble, have previously been shown to alleviate protein aggregates. This chaperone function, which Caldwell likens to a kind of "molecular Drano," could be important. In his nematode model, he's screening for other interacting partners for tor-2. Furthermore, he and other researchers are drawing links to neuro-degenerative disorders. Torsin proteins have been found in the brains of patients with Parkinson disease. The proteins were incorporated into Lewy bodies, protein aggregates consisting largely of a-synuclein.4 More recently, torsins were isolated in the polyglutamine inclusions found in Huntington disease and two related disorders, according to Ruth Walker at the Mount Sinai School of Medicine in New York. "Maybe torsinA is bound to the expanded polyglutamines as some way of helping degrade them," Walker postulates. Courtesy of Jeffrey W. Hewett LIFE IN THE ER: Immunofluorescence with antibody to torsinA (green), and an endoplasmic reticulum marker (red) in a catecholaminergic cell line (CAD); overexpressed mutant torsinA (left), and overexpressed wild-type torsinA (right) WORKING OUT THE KINKS The connections manifest clinically, as many patients with Parkinson have dystonic movements. Also, dystonia has been misdiagnosed as Parkinson. But one unnerving fact remains: Dystonia is a very different disease. "Whatever's abnormal with the protein torsinA in these conditions, it's not causing proteins to clump [and] it's not causing cells to die," says Walker. Caldwell explains that his worms are neither a model of Huntington, nor of torsion dystonia, which makes it difficult to extrapolate the observed malfunction to diseased states. Another kink is that localization studies place normal torsinA in the endoplasmic reticulum (ER) and out of reach of where these protein aggregates form. "You can have the best chaperone in the world sitting in the middle of the ER lumen, and you can have a [misfolded] protein out in the cytoplasm, they won't ever see each other," says Breakefield. And, Lewy bodies, she says, contain hundreds of proteins that the cell is trying to eliminate. "We can't say that the torsin that's in the aggregate is actually functional torsin." Caldwell says that his group found TOR-2 wherever protein aggregates formed, but adds that overexpression of both tor-2 and the poly-glutamine GFP may be responsible. Phyllis Hanson, Washington University in St. Louis, investigates membrane trafficking. The idea that dystonia might be a neurotransmitter secretion disorder drew her to torsinA. With Breakefield, she recently localized the endogenous protein to the ER. "The question becomes, how are you having an effect on cytoplasmic things? ... You basically have to invoke either indirect mechanisms, or you have to invoke the idea that there is some small pool of torsin that is not going into the ER." In unpublished studies, she says, mutations made to one motif within torsinA's ATP-binding domain caused changes in torsin's localization, which may reflect its function, but she has yet to explain the results. Lesilee Rose, University of California, Davis, previously showed a different function for a closely related worm torsin, OOC-5, in determining cell polarity for asymmetric division during early embryogenesis.5 "The human gene may also be required for a polarity process. Neurons are highly polarized cells," Rose says. Mount Sinai's Pullani Shashidharan demonstrated torsins in Lewy bodies.4 He says that while Caldwell's results confirm the sequence-predicted chaperone function of torsinA, they explain little about dystonia. "We think it has a major role in some way with regards to neurotransmission." He's currently investigating protein-protein interactions and working on a murine model for more clues. Breakefield points to the somewhat obvious conclusion that more research is needed, and she says she's cautious not to rule out numerous hypotheses in favor of one that has tantalizing connections with Parkinson and other funding-rich neuro-degenerative disorders. "It's fine as a way of getting money to do what you want to do, but I think ... it's premature to assign a function to torsinA." The increased attention caused by the neurodegenerative-disease connection to dystonia also benefits patient and research advocacy groups such as the Dystonia Medical Research Foundation, according to the group's president, Rosalie Lewis, who has three sons with the disease. "Dystonia having [a] connection to Parkinson is exciting, because then we get new researchers interested in dystonia as well as hopefully the pharmaceutical and biotech industries." Caldwell says he believes that the protein-aggregate connection exists even though such aggregates have never been found in dystonia. "Less obvious problems with protein misfolding may lead to the disease," he adds. Intermediates between soluble and aggregated proteins have at times been implicated in neuro-pathologies. "Such subtle effects of protein misfolding [are] probably the reason why you only get about 30% to 40% penetrance." A certain stress level, Caldwell says, might override other factors in the cell's ability to cope with a mutant form of torsinA. Now, nearly six years since the discovery of DYT1, some clues, but mostly questions, continue to pop up. Hanson sums up the mixture of excitement and frustration surrounding torsin. "It's an intriguing molecule in a lot of ways, but there's not a lot of answers." Brendan A. Maher can be contacted at bmaher@the-scientist.com. References 1. L.J. Ozelius et al., "The early onset torsion dystonia gene (DYT1) encodes an ATP-binding protein," Nat Genet, 17:40-8, 1997. 2. G.A. Caldwell et al., "Suppression of polyglutamine-induced protein aggregation in Caenorhabditis elegans by torsin proteins," Hum Mol Genet, 12:307-19, February 2003. 3. P.J. McLean et al., "TorsinA and heat-shock proteins act as molecular chaperones: Suppression of a-synuclein aggregation," J Neurochem, 83:846-54, November 2002. 4. P. Shashidharan et al., "TorsinA accumulation in Lewy bodies in sporadic Parkinson's disease," Brain Res, 877:379-81, 2000. 5. S.E. Basham et al., "The Caenorhabditis elegans polarity gene ooc5 encodes a Torsin-related protein of the AAA ATPase superfamily," Development, 128:4645-56, 2001.

Researchers Reveal a New Twist in Torsion Dystonia

Interested in reading more?

Become a Member of