Cerebello-thalamo-cortical network play a key role in motor skill learning in mice

VARANI ANDRÉS P.; MAILHES-HAMON CAROLINE; YOUMIR RIHAM; BARRIER JEAN; PERRIN ELODIE; MENARDY FABIEN; POPA DANIELA; LÉNA CLÉMENT

Congreso; Cerebellum Gordon Research Conference; 2017

Clinical and preclinical studies demonstrate that the cerebello-thalamo-cortical network is involved in motor skill learning. However, how cerebellar outputs could modulate this behavior remains unclear. On one hand, we evaluated the role of cerebello-thalamo-cortical network during motor skill learning (7 trials per day during 7 days) by using in vivo multi-channels recordings. At behavioral level, we were able to distinguish early (acquisition, Day 1) and late (consolidation, Days 2-7) phases in motor learning task. The electrophysiological recordings revealed a decrease in the deep cerebellar nuclei (Dentate, Interpositus and Fastigial nuclei) activity during the late but not early phase. In addition, increased cortical (M1 cortex) activity was observed across trials during early phase. However, the activity within the thalamus (Intralaminar and motor thalamus) increased across trials during early, middle and late phase. On the other hand, we investigated the effects of a specific and transient silencing in the neuronal activity of cerebellar outputs during motor skill learning in mice. We virally expressed inhibitory hM4Di-DREADDs (designed receptors exclusively activated by drugs) in the deep cerebellar nuclei, and we examined the effects of this inhibitory action during motor skill learning using the accelerating rotarod test after clozapine-N-oxide (CNO) injection. Importantly, our results showed that inhibition of deep cerebellar neurons produced an impairment in the beginning of the consolidation phase (Days 2-5) and then a retrieval effect was observed, suggesting that adaptive mechanisms could compensate for the lack of neuronal activity in deep cerebellar nuclei. The motor learning performance was not affected by the CNO treatment in control mice. Open-field, horizontal bar, vertical pole and footprint tests were applied to evaluate locomotor activity and motor coordination. No significant differences were observed between the experimental conditions in any of these behavioral tests. Finally, we combined the use of a CRE-recombinase expressing canine adenovirus-2 (CAV-2) and a floxed hM4Di-DREADD. CAV-2 retrogradely infects projection neurons, which allowed us to specifically express hM4Di-DREADD in neurons that project from dentate nucleus (DN) to intralaminar thalamus (CL). By using this approach we were able to observe that DN-CL projections modulate motor skill learning during middle and late but not early phase. In conclusion, our results indicate that cerebello-thalamo-cortical network participates in the consolidation of motor skill learning.