Evolution of taste processing shifts dietary preference
Article Meta
Article Date: 26 November 2025
Article URL: https://www.nature.com/articles/s41586-025-09766-6
Article Image: (not provided)
Summary
This Nature paper by Bertolini et al. investigates how changes in taste processing drive dietary specialisation in fruit flies. Focusing on Drosophila species with different host preferences (notably the noni-specialist D. sechellia), the authors combine behavioural assays (72-well, petri-dish, flyPAD), genetic manipulations (CRISPR knock-ins, mutants, transgenics), volumetric calcium imaging and connectomic analyses to show that peripheral taste receptor populations and central taste-circuit responses have diverged between species. These sensory and circuit-level changes alter how sweet, bitter and fatty-acid signals are represented and integrated, shifting feeding and oviposition choices toward distinct natural substrates such as noni fruit.
The work links changes in receptor expression and neuron numbers (for example Gr64f and Gr66a-expressing populations), altered physiological responses in taste organs and the subesophageal zone (SEZ), and behavioural preference shifts. Functional experiments (neuronal silencing, receptor rescue and mutant analysis) show causal roles for specific taste neurons and pathways in the evolved dietary preference. Data and code are available via the authors’ GitHub.
Key Points
- Behavioural assays across species reveal distinct, reproducible preferences: D. sechellia prefers noni while other species favour grape/sucrose.
- Evolutionary changes include shifts in peripheral taste neuron populations and receptor expression (notably sweet and bitter receptor-expressing cells).
- Calcium imaging shows species-specific differences in taste-evoked activity patterns in peripheral organs and in SEZ circuits, indicating central processing changes as well as peripheral ones.
- Genetic manipulations (CRISPR knock-ins, mutants, neuronal silencing and rescue experiments) provide causal evidence that altered taste circuits drive changed feeding behaviour.
- Analytical workflows (flyPAD metrics, volumetric imaging, connectomic segmentation) demonstrate how sensory encoding and integration explain ecological specialisation.
- All data, code and newly generated fly lines are available on the authors’ GitHub or from corresponding authors, supporting reproducibility.
Context and relevance
This study sits at the intersection of neurobiology, evolution and ecology. It builds on prior work showing rapid evolution of chemoreceptors and host specialisation in Drosophila, and it goes further by directly linking cell-type changes and circuit-level reorganisation to altered dietary choices. The findings matter for anyone interested in how nervous-system changes enable ecological adaptation — from basic researchers studying sensory coding and decision-making to applied groups considering pest behaviour and host preference.
More broadly, the paper illustrates a general principle: evolutionary shifts in both peripheral receptors and central processing can reassign value to the same chemical cues, producing new feeding and reproductive behaviours. This helps explain rapid dietary specialisation in insects and provides experimental routes to dissect similar processes in other animals.
Author style
Punchy: This is a tightly executed, multi-level demonstration that taste-system evolution — at the level of receptors, neurons and circuits — can rewire preference. If you work on sensory coding, evolution or insect behaviour, the methods and causal tests here are worth studying closely: they show how to go from gene to neuron to circuit to behaviour in an evolutionary context.
Why should I read this?
Because it neatly explains how tiny changes in taste neurons and brain wiring let a fly switch its favourite food — and the paper actually proves it with genetics, imaging and behaviour. Quick skim or deep dive: skim to get the core result (circuit + receptor changes = new diet), dive in if you want the experimental tricks and datasets — the authors have made code and lines available so you can reuse their approach.