Building a living laboratory
The team currently maintains three insect species in the laboratory: the mustard beetle, which serves as the project's model species, and two agricultural pests: the African Cotton Leafworm Spodoptera littoralis (which is a moth) and the aphid Myzus persicae. By the end of the project, they hope to work with more than nine different species, allowing them to investigate how different insects respond to RNA-based plant protection products.
Maintaining the colonies is a team effort. Together with research assistant CamilaZanini and student assistant Nora Iziki, Martina spends much of her week feeding insects, cleaning cages, collecting eggs, establishing new generations and carefully recording observations.
Some species are easier to manage than others.
"Ladybugs look sweet and pretty, but once the larvae grow older, they start eating each other," Martina laughs. "The same happens with Spodoptera. At that point, we have to separate them so every larva has its own compartment."
Even maintaining the colonies can mean individually housing dozens of larvae every few days to ensure there are enough healthy insects for future experiments.
Knowing what "normal" looks like
One obvious question is why researchers don't simply buy insects whenever they need them.
The answer is that maintaining colonies gives the team something they cannot order: experience.
When running bioassays, the researchers do not only measure whether insects survive. They also monitor growth, development and behaviour over time.
"If you don't know what normal looks like, you can't recognise when something has changed," Martina explains.
Some insects change colour when stressed. Aphids may develop wings. Developmental stages can vary slightly depending on conditions. Without understanding these natural patterns, researchers cannot confidently interpret experimental results.
Every observation is carefully documented in laboratory notebooks, making it possible to trace unexpected findings back to the condition of the colony if necessary.
Every detail matters
Maintaining healthy insects also means paying close attention to what they eat.
For some species, such as Spodoptera, the team prepares an artificial diet.
Many laboratories include antibiotics to prevent infections in densely populated colonies. Martina's team deliberately chose another approach.
"We wanted to avoid antibiotics because they could potentially make the insects weaker, affect their microbiome and thereby influence our experiments," she says.
Instead, the researchers optimised the diet itself and developed procedures that allowed them to maintain healthy colonies through careful hygiene.
During experiments, standardisation becomes even more important. Every insect must receive exactly the same amount and quality of food.
For leaf-feeding species, for example, researchers cut leaf discs of identical size so that every larva receives the same meal. This allows the team to estimate how much of the RNA-based product each insect actually consumes, making the results comparable between individuals and experiments.
When experiments become detective work
Even with careful preparation, biology can still be full of surprises.
Early in the project, the team encountered a puzzling problem.
The mustard beetle—the very pest the dsRNA was designed to kill—didn't die.
"We kept asking ourselves, 'Why aren't the beetles dying?'"
The researchers systematically investigated every possible explanation. They compared different sources of leaves, tested different washing procedures and discussed the problem with colleagues at other research institutions.
Eventually they realised that seemingly small details could matter. The way the leaves are washed could affect the integrity of the dsRNA applied to the leaf surface.
The team decided to grow their own kale in the greenhouse, produced a fresh batch of double-stranded RNA, and finally achieved the expected results.
The experience highlighted something important: when working with RNA-based plant protection products, not only the active ingredient but also its surroundings and its "packaging" can influence performance. This is one of the reasons why other ENSAFE researchers are investigating different formulations and delivery systems that may help stabilise RNA under real-world conditions.
Building confidence in the data
Every experiment also includes carefully designed controls.
In each experiment, some insects receive only water to ensure that the experimental conditions themselves are not causing unexpected effects. Others receive a conventional chemical with a well-established response, providing a benchmark against which the dsRNA treatment can be compared.
If insects in the water control begin dying unexpectedly, the researchers immediately know something is wrong with the experiment, and will not use the data.
These quality checks are essential for producing robust and trustworthy data.
Much of this work happens quietly behind the scenes and rarely appears in scientific publications. Yet without healthy insect cultures, detailed observations and rigorous quality control, it would be impossible to generate the reliable evidence needed for environmental risk assessment.
Behind every data point lies months of patient care, method optimisations, careful observation and teamwork. It may not be the most visible part of ENSAFE's research, but it is one of the foundations on which good science is built.
