Optimization of a Neurotoxin to Investigate the Contribution of Excitatory Interneurons to Speed Modulation In Vivo.

Precise control of speed during locomotion is essential for adaptation of behavior in different environmental contexts [1-4]. A central question in locomotion lies in understanding which neural populations set locomotor frequency during slow and fast regimes. Tackling this question in vivo requires additional non-invasive tools to silence large populations of neurons during active locomotion. Here we generated a stable transgenic line encoding a zebrafish-optimized botulinum neurotoxin light chain fused to GFP (BoTxBLC-GFP) to silence synaptic output over large populations of motor neurons or interneurons while monitoring active locomotion. By combining calcium imaging, electrophysiology, optogenetics, and behavior, we show that expression of BoTxBLC-GFP abolished synaptic release while maintaining characterized activity patterns and without triggering off-target effects. As chx10(+) V2a interneurons (V2as) are well characterized as the main population driving the frequency-dependent recruitment of motor neurons during fictive locomotion [5-14], we validated our silencing method by testing the effect of silencing chx10(+) V2as during active and fictive locomotion. Silencing of V2as selectively abolished fast locomotor frequencies during escape responses. In addition, spontaneous slow locomotion occurred less often and at frequencies lower than in controls. Overall, this silencing approach confirms that V2a excitation is critical for the production of fast stimulus-evoked swimming and also reveals a role for V2a excitation in the production of slower spontaneous locomotor behavior. Altogether, these results establish BoTxBLC-GFP as an ideal tool for in vivo silencing for probing the development and function of neural circuits from the synaptic to the behavioral level.