Systematic Characterization of Sensorimotor Transformations in the Drosophila Larva

Abstract

Animal neural systems have evolved to detect and process environmental cues. A central goal in neuroscience is to understand the cellular and molecular mechanisms of sensory systems and how the information they encode is transformed into behavioral decisions. Taking advantage of the optical transparency, simple motor programs, and ample genetic toolbox of the Drosophila melanogaster larva, this study explores chemosensory and thermosensory sensorimotor transformations from a systems-level perspective. Chapter 2 discusses a high-throughput technique developed to quantitatively estimate the transformations from optogenetically-induced neural activity into behavior. Linear-static non-linear (LN) models are used to quantify how different olfactory and gustatory neurons influence Drosophila larvae navigation. This approach succeeds at accurately predicting behavioral responses to new stimuli and captures the differences between complete and partial adaptation to persistent neural activation. Chapter 3 discusses the combination of odor linear gradients, genetic manipulations of odor receptors, and optogenetics to identify the behaviors encoded in the activity of all the genetically accessible larval olfactory receptor neurons (ORNs) (18 of the 21). The early olfactory connectome of larvae is used to analyze correlations between ORN behavior and connectivity, and to build a mathematical model of olfactory processing. Experiments found that ORNs behavioral valence (attraction and repulsion) is strongly correlated with the way they connect with specifc local interneurons. The mathematical model results provide candidate hypotheses for the function of having attraction and repulsion ORNs connected in different patterns. Chapter 4 addresses the problem of controlling odor concentration dynamics in a way that spans both behavioral and neural activity analysis. A novel odor delivery apparatus, and a control method to precisely manipulate odor dynamics for neurophysiology and for high-throughput behavior experiments are presented. I demonstrated how the experimental system can be used to study olfactory sensorimotor transformations by measuring the responses of different components of the olfactory circuit, and by measuring behavioral responses to the same stimulus. This apparatus is also used to study properties of larval ORNs encoding and how they are reflected in behavior. Findings suggest that larval ORNs encode the absolute odor concentration and its rate of change, and the ORNs’ gain is suppressed with increasing odor background (Weber-Fechner law). Behavioral experiments revealed that these two features of ORN encoding are also reflected in navigation dynamics. In Chapter 5, a new set of thermosensory neurons are introduced, the warming cells, whose anatomy, receptive felds (sensitivity), and the combination of ionotropic receptors that convey warming sensing were determined. These new genetic tools allowed for the study of the overall thermotactic strategy of Drosophila larvae using wildtype animals and mutants defective for warming cells, cooling cells, brain temperature sensors(dTRPA1), and their combinations. A cross-inhibition computation between the cooling and warming pathways that determines thermotaxis was discovered. These results suggest that the weight or prevalence of each pathway is determined by dTRPA1, from which a general model for larvae thermotaxis is proposed. Finally, in Chapter6, the implications of these fndings and the new challenges we face in Drosophila systems neuroscience are discussed.