by Dr Alicja Witwicka
For a while now researchers have been warning governments, farmers and the public about the harm that pesticides cause to pollinating insects. And many have listened, some of the most toxic insecticides, called neonicotinoids, have been under strict restrictions within the EU and other parts of the world. But this raises an uncomfortable question: why were pesticides licensed for use in the field later found to be so toxic? Part of the answer may lie in how these chemicals are tested. Most insecticide safety testing is conducted on western honey bees. It’s understandable, this species is well-studied and economically important. However, when a pesticide is tested and passes as relatively safe to honey bees, it is assumed that the chemical will be safe for all other pollinators too. A new study from Queen Mary University of London suggests that this ‘assumption’ deserves a second look.
The problem with testing on just one species
Honey bees sit on one tiny branch on an enormous evolutionary tree that encompasses other pollinating insects that are all crucial for food production and ecosystem health. Bees, flies, and butterflies split apart hundreds of millions of years ago and adapted to a multitude of different habitats and stressors that made their jobs within ecosystems (including pollinating flowers) very effective. At the same time, this long divergence time created vast differences in their metabolisms. When a chemical passes a honey bee safety test, can it tell us anything about its effect on a butterfly?
We exposed four pollinators, bumblebees, solitary mason bees, blowflies, and painted lady butterflies to two real-world insecticides: clothianidin, a restricted neonicotinoid, and sulfoxaflor, a newer chemical marketed as a safer alternative. We chose a dose that was deliberately small, the kind of trace amount these insects might realistically encounter while foraging in a field. Rather than just watching to determine which species survived, we looked inside each one’s brain and tried to detect which genes were switched on or off in response to each chemical.
Bees, butterflies, and flies vary greatly in their behaviour, limiting our ability to compare exposure effects through behavioural changes alone. Genes, however, are the building blocks of all living organisms, and some of them play similar functions across species. These similarities allowed us to zoom into the processes happening in the insects’ brains and disentangle which ones are disrupted in each species under exposure.
Both insecticides caused similar disruptions within species
Within each single species, the two insecticides affected the same processes. More than 70% of the genes disrupted by one compound were also disrupted by the other. This makes biological sense, both chemicals disrupted the same type of receptor in insect neurons, so a shared downstream signature isn’t surprising once we are looking at one species in isolation.
But what about between species?
Here’s where it gets interesting. When we compared those same two pesticides between species, the picture fell apart entirely. Out of nearly 3,000 genes shared across all four species, not even one showed a consistent response to insecticide exposure in every species. Even between the two bee species, which are relatively closer to one another in evolutionary terms, the overlaps in effects were minimal. Interestingly, while sulfoxaflor barely touched the bees’ brain chemistry at the tested dose, in the fly and the butterfly it triggered far more disruption than the supposedly more dangerous clothianidin, the exact opposite of what their safety labels would predict.
Why does this matter?
If regulatory testing keeps focusing on just one or two model species, it risks missing the very real possibility that a chemical deemed “safe” is doing something very different to a fly, a butterfly, or any of the thousands of other insects keeping our food systems alive.
The fix is not more testing for testing’s sake. Testing should reflect the actual diversity of pollinators present in our fields, rather than assuming one species can speak for them all. But better testing alone is not enough. Insecticides are designed to kill insects and will always pose risks to the ecosystems they enter, regardless of how carefully they are designed or assessed. The deeper challenge is rethinking the agricultural model itself, moving away from chemical dependency and toward farming practices that work with nature rather than against it. Let’s give the insects that support our food system a fighting chance.
Alicja Witwicka is a postdoctoral research fellow at the Wellcome Sanger Institute in Cambridge. Alicja completed her PhD at Queen Mary University of London, where she studied how pesticide exposure alters gene regulation in pollinating insects. Her work bridges molecular biology and applied ecology, providing actionable data for policymakers, land managers and industry to protect insect biodiversity.
