Meeting Abstract
Proprioception, the sense of self-movement and body position, is critical for the effective control of behavior. In the absence of proprioceptive feedback, animals are unable to maintain limb posture or coordinate fine-scale movements of the arms and legs. However, despite the importance of proprioception to the control of movement in all animals, little is known about the neural computations that underlie limb proprioception in any animal. We developed new methods to record from proprioceptive neural circuits in the fruit fly, Drosophila. Each fly leg contains ~135 proprioceptive chordotonal neurons positioned at the joint between the femur and the tibia. We used in vivo 2-photon calcium imaging to record calcium sigansl from the axons of this proprioceptor population while manipulating leg position and movement with a magnetic control system. With unsupervised clustering methods, we identified anatomically distinct subpopulations of proprioceptor neurons that encode specific kinematic variables such as leg position, velocity, and direction. Imaging from more specific genetic driver lines, we found that single proprioceptor neurons are sharply tuned for combinations of these variables. We then identified two populations of second-order neurons that process sensory information from leg proprioceptors. Targeted electrophysiological recordings revealed that these two populations are specialized for encoding leg position and directional movement. Overall, our results illustrate how proprioception of a single leg joint is encoded by a diverse population of mechanosensory neurons. Narrowly tuned proprioceptive signals converge onto central pathways that separately represent leg movement and position. This circuit architecture may help to reduce sensory noise while minimizing delays in neural processing. Speed and robustness are critical for feedback control of the limbs during locomotion.
