Generation of a novel rodent model for DYT1 dystonia

Abstract

A mutation in the coding region of the Tor1A gene, resulting in a deletion of a glutamic acid residue in the torsinA protein (∆ETorA), is the major cause of the inherited autosomal-dominant early onset torsion dystonia (DYT1). The pathophysiological consequences of this amino acid loss are still not understood.

Currently available animal models for DYT1 dystonia provided important insights into the disease; however, they differ with respect to key features of torsinA associated pathology.

We developed transgenic rat models harboring the full length human mutant and wildtype Tor1A gene. A complex phenotyping approach including classical behavioral tests, electrophysiology and neuropathology revealed a progressive neurological phenotype in ∆ETorA expressing rats. Furthermore, we were able to replicate key pathological features of torsinA associated pathology in a second species, such as nuclear envelope pathology, behavioral abnormalities and plasticity changes. We therefore suggest that this rat model represents an appropriate new model suitable to further investigate the pathophysiology of ∆ETorA and to test for therapeutic approaches.

Introduction

DYT1 dystonia is an autosomal-dominantly inherited movement disorder. Most cases are caused by a 3 base pair (GAG) deletion in the coding region of the Tor1A gene (DYT1/TOR1A) (Ozelius et al., 1997). DYT1 dystonia manifests predominantly in childhood as twisting movements and abnormal postures induced by involuntary muscle contractions. The protein torsinA belongs to the AAA + (ATPase associated with different cellular activities) protein family. Studies in recent years have implicated torsinA in cellular response to stress (Hewett et al., 2003, Kuner et al., 2004), in neurite outgrowth (Ferrari-Toninelli et al., 2004, Hewett et al., 2006), synaptic plasticity (Martella et al., 2009) and dopaminergic transmission (Augood et al., 2002, Augood et al., 2003, Marsden et al., 1985, Torres et al., 2004). Despite the increasing knowledge about the normal function of the protein, the relationship between the impaired function of mutant torsinA protein and the development of DYT1 dystonia is still unclear.

Findings from cell biological studies, neurochemical analyses, functional imaging and electrophysiological studies in mouse models and in humans indicate that dystonic movements are the result of dysfunction of CNS motor systems involving selective regions implicated in movement control (Martella et al., 2009, Niethammer et al., 2011, Tanabe et al., 2009). These disease characteristics require the investigation of the pathophysiology of the torsinA mutation in an in vivo setting. Accordingly, a variety of animal models for DYT1 dystonia have been generated including transgenic and gene targeted mouse models (Dang et al., 2005, Goodchild et al., 2005, Grundmann et al., 2007, Sharma et al., 2005, Shashidharan et al., 2005). These mouse models display important characteristics of the human disease. However, the phenotype of the existing mouse models appears to be inconsistent most likely due to differences in genetic backgrounds, the promoters chosen to drive the expression of the transgene but also to the varying degree of characterization.

Collectively, it is difficult to dissect at this stage which phenotypic features are related to torsinA dysfunction and which are side effects of the models (Grundmann et al., 2007, Sharma et al., 2005, Shashidharan et al., 2005). One limitation of the existing mouse models overexpressing the wildtype and mutant torsinA is the use of foreign promoters to drive the expression of the torsinA protein (Grundmann et al., 2007, Sharma et al., 2005, Shashidharan et al., 2005). Although the mouse models for DYT1 dystonia provided important insights into disease mechanisms there is still need for an animal model belonging to a different species. Replication of pathological features observed in preceding mouse models might provide further evidence that these features are direct results of an impact of the mutant protein and not related to the model system expressing the mutant protein.

This motivated us to generate a transgenic rat model overexpressing the human mutant and wildtype torsinA protein. We used a 16.25 kbps human genomic fragment containing the full length human Tor1A gene including promoter and regulatory regions with the intention to express the human torsinA protein under the control of the endogenous human promoter and its regulatory elements. This model enabled us to investigate which of the so far reported pathological characteristics caused by the mutant torsinA protein in cell culture or in mouse models can be replicated in vivo in a second rodent species.