FV 4 Analyzing for central network changes in the basal ganglia of a DYT-TOR1A mouse model with a dystonia-like phenotype

Abstract

Introduction: DYT-TOR1A is the most common inherited form of dystonia with still unclear pathophysiology and a reduced penetrance of 30–40%. To analyse whether environmental factors in genetically predisposed DYT-TOR1A patients can act as a potential trigger for the development of a dystonic phenotype, we aimed to induce dystonia-like symptoms in hΔGAG3 mice by a peripheral nerve crush. Furthermore, our objective was to identify mechanisms of dystonia development in the basal ganglia. Materials and methods: Potential dystonia-like movements (DLM) of the right hindlimb in hΔGAG3 and wildtype (wt) mice after peripheral nerve crush were analysed during a tail suspension test over an observational period of 12 weeks using an observer-based scoring system from 0-5 points and an unbiased deep learning pipeline. Nerve conduction studies were performed 10 weeks after the nerve crush in order to study the functional recovery. The number and volume of striatal parvalbumin+, nNOS+, calretinin+, ChAT+ interneurons and of dopaminergic neurons of the substantia nigra were estimated by unbiased stereology. Striatal medium spiny neurons were visualized via a Golgi Cox staining followed by a Sholl analysis. Results: The observer-based scoring system showed that a peripheral nerve crush leads to significantly more DLM in hΔGAG3 animals compared to wt animals (week 12; nerve-crushed hΔGAG3 mice: 2.10 ± 0.30 points vs nerve-crushed wt mice: 0.40 ± 0.20 points, p < 0.0001). Using a trained DeepLabCut network for the tail suspension test, we were able to identify significantly more dystonia-like retraction and clenching of the right hind limb in the hΔGAG3 animals compared to wt mice (week 12; 5.28 ± 1.13 vs 1.09 ± 0.38 cramps/minute, p < 0.01). Nerve conduction studies showed an equal recovery of the sciatic nerve in hΔGAG3 and wt mice (nerve conduction velocity: 23.97 ± 1.80 m/s vs 25.30 ± 2.80 m/s). The dopaminergic neurons of the substantia nigra were found to be unchanged in number across all groups, however, the cell volume was found to be significantly increased in the nerve-crushed hΔGAG3 mice compared to the wt littermates (wt naïve: 3116.00 ± 221.50 µm 3 , nerve-crushed wt: 3039.00 ± 110.90 µm 3 , hΔGAG3 naïve: 3681 ± 221.40 µm 3 , nerve-crushed hΔGAG3: 4787.00 ± 197.90 µm 3 , p < 0.0001 comparing nerve-crushed hΔGAG3 to nerve-crushed wt and naïve wt, p < 0.01 comparing nerve-crushed hΔGAG3 to naïve hΔGAG3). The striatal interneurons showed no significant differences in number and volume. The analysis of medium spiny neurons revealed significantly reduced dendrite length, dendrite thickness and spine numbers in hΔGAG3 mice compared to wt mice. Conclusion: The induction of a dystonia-like phenotype in genetically predisposed DYT-TOR1A mice highlights the importance of environmental factors in the symptomatogenesis of dystonia. Morphological analyses of the basal ganglia revealed endophenotypic changes as well as changes related to the dystonia-like phenotype.