Processing of odorant timing and quality in the olfactory system of Drosophila melanogaster

Abstract

How does a brain process the information about its surrounding world that it gathers through its sensory systems? How is the external world encoded in neural signals? How are these codes reformatted, suitable for efficient decoding of the relevant features embedded in the vast amount of activity? To study such an external world representation and its transformations by an internal set of neurons, this work focused of brain of Drosophila melanogaster. This insect’s nervous system has been subjected to detailed genetical, morphological, functional and behavioral studies, allowing to address detailed questions about its inner functions. Using the well described olfactory processing pathways and the influence of learning on them, I studied several aspects of information processing, learning and perceptual segregation and in this insect’s brain. Inspired by the finding that Drosophila can bridge perceptual gaps during associative learning, this thesis contains work on determining the likeliest neural candidates for maintaining the required stimulus trace for this cognitive capacity (Chapter 4). We identified the Kenyon cells (KCs) of the mushroom body, a central brain area for associative learning in insects, as the only cell type suitable for maintaining a stimulus trace predictive of past stimulus presentations. Additionally, this thesis contains work that revealed a novel plastic synapse in the neural circuitry mediating associative learning in a comparable paradigm: We could show that the mushroom body associated dopaminergic neurons of the fly modify their response to reinforced odorants (Chapter 5). Previous studies have shown, that insects and other invertebrates perform better at an odor-background segregation task when the olfactory stimulus contains a temporal structure, although the overall chemical composition remains the same. Such temporal stimulus cues therefore contain information about the external world that the animals are able to both de- and encode within their neural activity. This thesis contains work at elucidating the involved mechanisms of such temporally aided smelling capabilities on both behavioral and physiological levels (Chapter 6). Behaviorally, we employed a behavioral paradigm involving upwind plume-tracking to confirm temporally aided odor-background segregation in Drosophila. Physiological experiments include a series of both electrophysiological and functional imaging studies on olfactory sensory neurons, projection neurons, Kenyon cells and output neurons of the mushroom body, to search for neural correlates of temporally aided odorant identity processing. This study also contained both hardware and software development. A novel type of olfactory stimulator that is capable of precisely controlling the relative timing of 2 odorants in a binary mixture is presented (Chapter 3). Additionally, several software tools for the analysis of both electrophysiological and microscopical recordings of neural activity were developed over the time course of this study (Chapter 7).