Characterization of distinct contrast- and luminance-sensitive pathways in the Drosophila visual system

Abstract

Visual systems are capable of extracting relevant information from dynamically changing environments. This allows animals to locate objects, escape predators and collisions or estimate self-motion, among others. During the day, illumination of the environment undergoes big intensity changes. Visual systems of many animals have evolved to cope with these changing light conditions by adapting to the input, which allows animals to perform well in bright daylight as well as at dusk and dawn, or when the world is fast changing e.g. due to self motion. This is true because visual systems ignore background illumination by computing contrast. Downstream of photoreceptors lamina monopolar cells (LMCs) were thought to discard luminance information and amplify photoreceptor contrast signals (Laughlin, 1989; Laughlin et al., 1987). Here we show that in the visual system of Drosophila, luminance information is retained past photoreceptors. We show that two distinct OFF-pathway inputs, the two lamina neurons L2 and L3, located downstream of photoreceptor cells, are contrast- and luminancesensitive, respectively. To understand what is shaping these early differences in visual processing, we tested the contribution of different photoreceptors inputs, the effect of lateral circuit inputs, as well as cell-autonomous differences. We show that L2 as well as L3 receive predominant input from R1-R6 and that two different types of photoreceptor-to-lamina transformations occur between the outer photoreceptors and L2 or L3. Contrast sensitivity of L2 is achieved by a circuit-dependent elimination of baseline, whereas luminance sensitivity of L3 depends on a cell-autonomous process that requires the L3-specific transcription factor, dFezf. We furthermore show that contrast- and luminance-sensitive information are combined in visual circuitry postsynaptic to lamina neurons. This suggests the importance of retaining the peripheral visual feature, luminance, in image processing.