Uncovering interactions between Descending Neurons as a functional principle of behavioural control

Animals, including humans, exhibit a remarkable variety of complex behaviours. How the nervous system controls all these behaviours ranging from simple, stereotyped movements to flexible, adaptive actions is a central questions of neuroscience. One of the key steps is the transformation from intentions into actions: Movement instructions must pass from the brain to downstream motor circuits through Descending Neurons (DNs). These include small sets of command-like neurons that are sufficient to drive behaviours and specific populations of DNs that are active during different behaviours. Currently it is unclear whether and how different DNs interact with each other to control behaviours.

In this thesis, I present two important, synergistic advances for our understanding of the descending control of behaviour. First, we developed a micro-engineered toolkit to enable long-term recording of the Drosophila ventral nerve cord and cervical connective for up to a month. We demonstrate its utility by visualising sensory neuron degradation after leg injury and by studying the effect of different foods on Ascending and Descending Neuron dynamics. This toolkit will facilitate future studies of DNs across different internal states and throughout motor adaptation.

Second, we studied whether and how DNs interact with each other to control behaviours using a combination of optogenetics, functional population recordings, and connectome analysis. We show that command-like DNs in Drosophila directly recruit networks of additional DNs to orchestrate flexible behaviours. Specifically, we found that optogenetic activation of command-like DNs, previously thought to drive behaviours alone, in fact co-activate larger populations of DNs. Connectome analysis revealed that this functional recruitment can be explained by direct excitatory connections between command-like DNs and networks of interconnected DNs in the brain. The size of downstream DN networks is predictive of whether descending population recruitment is necessary to generate a complete behaviour: DNs with many downstream descending partners require network recruitment to drive flexible behaviours, while neurons with fewer partners can alone drive stereotyped behaviours and simple movements. These results support a mechanism for command-like descending control whereby a continuum of stereotyped to flexible behaviours are generated through the recruitment of increasingly large DN networks which likely construct a complete behaviour by combining multiple motor subroutines.