An animal’s feeding state changes its behavioral priorities and thus influences even nonfeeding-related decisions. How the feeding state information is transmitted to nonfeeding-related circuits and what circuit mechanisms are involved in biasing nonfeeding-related decisions remain open questions. By combining calcium imaging, neuronal manipulations, behavioral analysis and computational modeling, we determined that the competition between different aversive responses to mechanical cues is biased by changes in the feeding state. We found that this effect is achieved by the differential modulation of two different types of reciprocally connected inhibitory neurons promoting opposing actions. This modulation results in a more frequent active type of response and, less frequently, a protective type of response if larvae are fed sugar than when they are fed a balanced diet. Information about the internal state is conveyed to inhibitory neurons through homologs of the vertebrate neuropeptide Y, which is known to be involved in regulating feeding behavior.
Understanding the genetic basis of natural phenotypic variation remains challenging, particularly for complex traits. However, causal genetic variants and their effects may be more accessible through endophenotypes, intermediate traits along the trajectory from genotype to complex organismal phenotype. Here, we investigate natural variation in R8 photoreceptors of the Drosophila visual system, using Rhodopsin 5 (Rh5) and Rhodopsin 6 (Rh6) expression as a readout of their subtype identities. Simple rules governing this expression allowed us to predict eight qualitatively distinct phenotypes. We observed deviations from the typical R8 subtype pattern in individual flies extracted from the field and in wild-derived inbred lines. Remarkably, we found all eight predicted phenotypes, including shifts in the proportions of Rh5- and Rh6-expressing R8s, Rh5/Rh6 co-expression, and loss of detectable Rh5 or Rh6 expression. We identified causal coding and regulatory variants in sevenless , Rh5 , Rh6 , and melted , affecting different levels of the well-characterised developmental program. Our findings demonstrate that standing genetic variation in natural populations spans the full qualitative phenotypic space of a stereotyped neuronal system, suggesting that substantial deviations, including those associated with likely loss-of-function alleles, can be tolerated in nature.