Logo Logo
Switch language to English
Förstner, Friedrich (2011): The morphological identity of insect dendrites. Dissertation, LMU München: Fakultät für Biologie



Dendrite morphology is the most prominent feature of nerve cells, investigated since the origins of modern neuroscience. The last century of neuroanatomical research has revealed an overwhelming diversity of different dendritic shapes and complexities. Its great variability, however, largely interferes with understanding the underlying principles of neuronal wiring and its functional implications. This work addresses this issue by studying a morphological and functional exception- ally conserved network of neurons located in the visual system of flies. Lobula Plate Tangential Cells (LPTCs) have been shown to compute motion vision and contribute to the impressive flight capabilities of flies. Cells of this system exhibit a high degree of constancy in topographic location, morphology and function over all individuals of one species. This constancy allows investigation of functionally identical cells over a large population of flies, and therefore potentially to truly understand the underlying principles of their morphologies. Supported by a large database of in vivo cell reconstructions and a computational quantification framework, it was possible to uncover some of those principles of LPTC anatomy. We show that the key to the cells’ morphological identity lies in the size and shape of the area they span into. Their detailed branching structure and topology is then merely a result of a common growth program shared by all analyzed cells. Application of a previously published branching theory confirmed this finding. When grown into the spanning fields obtained from the in vivo cell reconstruction, artificial cells could be synthesized that resembled all anatomical properties that characterize their natural counterparts. Furthermore, the morphological comparison of the same identified cells in Calliphora and Drosophila allowed to study a functionally conserved system under the influence of extensive down-scaling. The huge size reduction did not affect the underlying branching principles: Drosophila LPTCs followed the very same rules as their Calliphora coun- terparts. On the other hand, we observed significant differences in complexity and relative diameter scaling. An electrotonic analysis revealed that these differences can be explained by a common functional architecture implemented in the LPTCs of both species. Finally, we could modify the LPTC neuronal interaction behavior thanks to the genetical accessibility of Drosophila’s wiring program. The transmembrane protein family Dscam has been shown to mediate the process of adhesion and repulsion of neurites. By manipulating the molecular Dscam profile in Drosophila LPTCs it was possible to change their morphological expansion. The low variability of the LPTCs spanning field in wild type flies and their two-dimensional extension allowed to thoroughly map these morphological alterations in flies with Dscam modifications. In line with the LPTCs retinotopic input arrangement, electrophysiological experiments yielded an inherent linear relationship of their locally reduced dendritic coverage and their locally reduced stimulus sensitivity. With this work I hope to contribute to the general understanding of neuronal morphology of LPTCs and to present a valuable workflow for the analysis of neuronal structure.