For an animal to survive in a new environment, it has to adapt its feeding habits, its behaviour, to the food available in that environment. If it doesn’t, it will not be able to survive. The rule also applies to fruit flies. For instance, Drosophila melanogaster and Drosophila simulans are two “generalist” species, which feed on a diversified diet. On the contrary, Drosophila sechellia, which is native to the tropical Seychelles islands, feeds only on a fruit called noni (Morinda citrifolia), which is particularly acidic and bitter – a fruit that the other two species actively avoid.
How did D. sechellia’s peculiar food preference evolve? Fruit flies taste food with their whole body, from their mouth to their legs to their wings. So genetic changes in taste receptors on the body surface, altering their number and sensitivity, were deemed sufficient to explain the changes in the flies’ food choices. However, a new study, published today (26/11/2025) in the journal Nature by researchers in Switzerland, Portugal and Germany, calls this view into question.
The researchers started out with the hypothesis – as senior co-author Thomas Auer, from the University of Fribourg, explained – that if Sechellia doesn't like the sweet taste as much as Melanogaster or Simulans do, maybe it is because it changed neuron sensitivity or just has less neurons that respond to sweet substances. Or alternatively, because it no longer detects the bitter substances in noni, allowing it to feed on the fruit.
Using grape and noni juice as stimulants, Auer and his colleagues in Switzerland tested the first possibility (on the sweet receptors). The results surprised them: “We didn't see that Sechellia sweet receptors responded more to noni than the other species”, says Auer. The sweet taste sensing neurons, which are usually important for attraction, presented the same activity across all three species and also neuron numbers were the same.
“This was surprising”, adds Auer, “because we were thinking that there might be a specific tuning towards noni, so that Sechellia liked it more – and what we saw was that it was not the case.”
They then looked at the bitter-sensing neurons. “In this case”, explains Auer, “our initial hypothesis was that as a specialist species, Sechellia might lose aversion to noni because it does not sense it as bitter any longer.” Sechellia has in fact lost some bitter receptors in its genome, so the researchers thought that could have led to a loss of noni aversion. But this was also not the case. On the contrary, what they found was that Sechellia responded more strongly to noni than the other species. “It was actually the opposite of what we were expecting”, Auer points out.
Overall, these results indicated that the responses of peripheral taste neurons could not explain the change in feeding behaviour seen in Sechellia. The Swiss team then became interested in looking more closely at the flies’ processing of taste information in the central brain.
That is where Daniel Münch and Carlos Ribeiro, respectively second co-author and senior co-author of the study, came into the picture. Münch, currently at the Justus-Liebig University in Giessen (Germany), was at the time a researcher in Ribeiro’s Behaviour and Metabolism lab, at the Champalimaud Foundation in Lisbon. “In the lab, we developed a technique to image the activity of all the neurons located in the brain taste processing center of the fly”, Münch explained.
The Champalimaud team imaged the taste-elicited responses, that is, the patterns of neuron activity, via calcium imaging in the brain of genetically modified flies of Melanogaster, Simulans and Sechellia provided by Auer’s lab. More specifically, they looked across a part of the fly’s central nervous system called the subesophageal zone, a cluster of nerve cells located below the esophagus that is crucial for eating.
“We created an imaging dataset from the three different species and then compared their activity patterns”, Münch explains.
What they found was that the feeding behaviour differences between the species were not due to changes in the sensory input, but in the processing of these inputs and how they were treated in the brain. “Basically, some regions in the subesophageal zone of Sechellia responded in a stronger way to noni as compared to grape juice, whereas this was inverted in Melanogaster flies”, says Münch.
Also, in the associated motor neuron regions (responsible for the flies’ actual ingestion of the food), they found almost no response to sweet taste in Sechellia flies. In other words, their results corroborated the hypotheses that food choices were regulated by information processing in the brain.
"When we developed this imaging method with Daniel, one of the ideas was that it could be used to look at brain activity across different insects and species, because you don't need many of the genetic techniques normally required to do classic imaging," says Ribeiro. "It's incredibly satisfying to see this approach applied here and yielding such insightful pictures of how evolution alters taste and feeding habits through changes in brain function."
This opens up future research in this direction, says Auer. “I think what's also important for the field is that there has always been a bit of an oversimplistic view: that whatever activates bitter sensory neurons is aversive, and whatever activates sweet sensory neurons is attractive. And here's a case where one sees that it's more complicated, more fine-grained than that.”
One of the potential applications of these new results could be for insect control strategies. “We usually think we can manipulate the feeding behaviour of insects by blocking the peripheral responses”, Auer notes. But maybe that's actually not the only way to do it.
Original article here
Image by Benjamin Fabian.
Text by Ana Gerschenfeld, Health&Science Writer of the Champalimaud Foundation.
