Spatial and network principles behind neural generation of locomotion

Walking is a fundamental action of humans and animals, yet the neural principles underlying how movement is generated remain unclear. In particular, the relationship between neuronal cell types, networks and functions has been difficult to establish. Here, we propose that spatial organization of the spinal cord itself is a key factor governing network-driven locomotor rhythms and patterns. First, an asymmetric "Mexican hat" connectivity -- i.e. an overall local excitation and longer-range inhibition with a longitudinal skew -- can account for the emergence of proper motor dynamics. Second, the role of segregation of cell types in the transversal plane is for descending fibers to find appropriate targets and control the network dynamics. We extract these principles via a model of the mouse spinal cord, where networks are constructed by probabilistic sampling of synaptic connections from cell-specific projection patterns, gleaned from literature. The cell-type distributions are derived from single-cell RNA sequencing combined with spatial transcriptomics. Essential aspects of locomotion are thus readily induced and controlled without requiring extensive parameter optimization, and several experiments can now be explained mechanistically. Besides the described synaptic projections, we predict propagating "bumps" of activity during rhythms. Hence, this work implies universal spatial principles that may underlie motor circuits across species, and provide the link between cell types, connectivity, and behavior. Although this new theory does not incorporate all details, it is an invitation to rethink the neuroscience of spinal cord-driven movement.