Functional characterization of neurons in the Drosophila medulla, which is the first processing center for color vision, using a behavioural assay
Final Report Abstract
The Drosophila eye consists of about 800 unit eyes, each of which contains eight photoreceptors. The six outer photoreceptors are involved in motion detection, while the two inner photoreceptors are believed to mediate color vision. Each photoreceptor expresses a single type of Rhodopsin. Four Rhodopsins with different absorbance spectra are present in different inner photoreceptors. They define two subtypes of unit eyes specialized in the detection of short or longer wavelengths of light, allowing the visual range of Drosophila to cover the spectrum from UV to green light. Color vision not only depends on the presence of different photoreceptors expressing individual photopigments, but also requires an underlying network of neurons that is capable of comparing and processing the photoreceptor output. In order to distinguish lights of different wavelengths, color-opponent mechanisms are often employed. This means that color-opponent neurons receive excitatory input from certain photoreceptors and inhibitory input from other photoreceptors. This is in contrast to pure luminance mechanisms that only rely on additive inputs. Using Drosophila as a model system, it is possible to specifically and very precisely manipulate photoreceptors and processing neurons to analyze their function, which is not feasible for color vision in other model organisms. While many genetic tools are already available to study the functions of neurons in the optic lobe, I had to adapt and improve them to use them in a behavioral assay. For instance, more specific GAL4 lines, which allow the expression of inhibitors or activators of neuronal function, are required. I have generated 35 GAL4 lines that will be useful in studying color vision in Drosophila. Even though some of them have similar expression patterns, it is nevertheless possible, together with previously available GAL4 lines, to address about thirty of the eighty different cell types in the medulla, which is the first processing center for color information. However, most of these GAL4 lines have expression domains outside of the cells of interest. Interfering with the function of these neurons might cause problems in a behavioral assay that relies on intact motor and cognitive skills of the flies. Therefore, it is important to restrict the manipulations to the neurons of interest. This can be achieved by adding a second level of control, such as the Flp-FRT system. Then, the inhibitors of neuronal function can only be expressed when the GAL4 line and the Flp line have overlapping expression patterns. I have already identified six GAL4/Flp pairs that restrict expression to one or a few cell types in the medulla, while there is no expression at all in other regions of the brain. Further studies are under way. In order to understand the exact function of neurons involved in color vision, I have chosen to establish a behavioral assay using a flight simulator that consists of a circular arena of blue and green LED panels. Single flies are tethered to small metal pins, which are then suspended between two magnets inside the arena. The flies can then freely turn and orient towards visual stimuli displayed in the arena. The flies are trained to prefer one color over the other by associating a pleasant smell such as vinegar odor with this color. I have preliminary data that demonstrate the feasibility of this approach. In the future, I will address whether the flies indeed memorize the color, or whether they use intensity differences for the task. Then, I will inhibit different classes of neurons in the medulla and study the effect of these manipulations on the ability of the flies to distinguish the two colors.
Publications
- (2007). The first steps in Drosophila motion detection. Neuron 56, 5-7
Vogt, N., and Desplan, C.
