A Drosophila model of early onset torsion dystonia suggests impairment in TGF-beta signaling.

To investigate the cellular and molecular etiology of early onset torsion dystonia, we have established a Drosophila model of this disorder. Expression of mutant human torsinA deleted for a single glutamic acid residue (DeltaE HtorA), but not normal HtorA, elicits locomotor defects in Drosophila. As in mammalian systems, DeltaE HtorA in flies forms protein accumulations that localize to synaptic membranes, nuclei and endosomes. Various morphological defects at the neuromuscular junction in larvae expressing DeltaE HtorA were observed at the EM level, some of which resemble those recently reported for mutants with defects in TGF-beta signaling. These results together with the distribution patterns and localizations of DeltaE HtorA accumulations suggested that DeltaE HtorA could interfere with some aspect of TGF-beta signaling from synapses to endosomes or nuclei. Consistent with this possibility, neuronal overexpression of Drosophila or human Smad2, a downstream effector of the TGF-beta pathway, suppressed the behavioral and ultrastructural defects of DeltaE HtorA flies. These results raise the possibility that a defect in TGF-beta signaling might also underlie early onset torsion dystonia in humans.