Scientists map fruit-fly taste circuits to feeding

A fruit fly can taste with its legs, wings, mouthparts and throat. A signal on a leg might make it stop walking and investigate; one inside the throat can regulate swallowing. Researchers at the Champalimaud Foundation and their collaborators have now traced those signals through the nervous system to the motor neurons that drive feeding—the first complete map of this taste-to-action wiring in an animal, according to the team.

The map sits inside a much larger reconstruction. In a companion study published in Cell, researchers traced and cataloged 166,700 neurons in the central nervous system of an adult male fruit fly, identifying more than 11,000 distinct cell types. The dataset preserves the link between the brain and the nerve cord, the insect equivalent of a spinal cord, so sensory signals can be followed all the way to movements of the legs, wings and feeding apparatus.

Artificial intelligence (AI) helped trace neurons and their connections through electron-microscopy data, and it predicted whether particular neurons excite or inhibit others. Humans then carried out extensive proofreading and annotation. A connectome—the map of how neurons connect through synapses, the junctions where signals pass—therefore shows more than the wiring’s endpoints: it also offers clues about the kind of message being transmitted.

The organizing pattern is striking but precise. Taste receptors are scattered across the body, yet signals carrying a similar value—whether something is “good” or “bad”—tend to converge onto the same circuits deeper in the brain. The location of the signal is not erased: taste detected on a leg may produce investigation, while taste in the throat can shape swallowing. The same pathways also connect to circuits involved in locomotion, hormone release and courtship.

So what changes, concretely? Researchers now have a route for testing how a sensory judgment becomes a physical act, and for examining pathways that could explain anticipatory insulin release before a single calorie is absorbed. The map could also inform studies of which flowers insects pollinate, which crops pests attack and where disease-carrying mosquitoes feed. Its efficient circuitry may suggest principles for artificial systems and a technical roadmap for larger connectomics projects.

The achievement remains a wiring map of an adult male fruit fly. The team’s broader effort took nearly two decades and involved the Champalimaud Foundation, Howard Hughes Medical Institute’s Janelia Research Campus, the MRC Laboratory of Molecular Biology, the University of Cambridge and Google Research; the next ambitions named by the researchers are connectomes for mice and humans.