How low would researchers have to go before glyphosate stopped producing signs of genetic damage?
In a new study using human intestinal cells, the answer was surprisingly simple.
They didn’t find out.
Researchers tested glyphosate at five concentrations, beginning at just 0.01 mg/L and increasing through 0.1, 1, 10 and 100 mg/L. At every concentration tested, glyphosate produced a significant increase in micronuclei — a recognised marker of genotoxicity, including chromosome damage or disruption during cell division.
Even at the lowest concentration they tested, the effect was there.
That doesn’t tell us that 0.01 mg/L of glyphosate in food or drinking water will damage someone’s DNA. This was an experiment on human cells in a laboratory, not on people, and laboratory concentrations cannot simply be translated into everyday human exposure.
But it does leave us with an interesting question.
What might the researchers have found if they had gone lower still?
And perhaps a bigger one:
How much do we really know about the biological effects of low-level glyphosate exposure?
Researchers Went Looking at Concentrations Others Had Largely Overlooked
The study, published in the peer-reviewed Annals of Global Health in July 2026, was part of the EU-funded SPRINT project investigating pesticide exposure and its potential effects on environmental and human health.
This wasn’t originally an experiment about glyphosate alone.
Researchers selected ten widely used pesticides and tested them individually and in mixtures on Caco-2 human intestinal epithelial cells — a widely used laboratory cell line originally derived from a human colorectal adenocarcinoma.
The choice of cells is interesting.
Rather than using an animal model, the researchers wanted to investigate effects directly in human cells. And because these were intestinal epithelial cells, they could be used as a laboratory model of what the researchers describe as a “site of first contact” following oral ingestion.
They then did something else worth noticing.
They deliberately extended their testing down to comparatively low concentrations — concentrations the researchers said had largely been overlooked in previous published studies.
For each pesticide, they tested 0.01, 0.1, 1, 10 and 100 mg/L.
And when they tested glyphosate, something stood out.
Every Glyphosate Concentration Produced an Effect
The researchers were looking for micronuclei.
These are tiny additional nuclei that can appear when chromosomes, or fragments of chromosomes, fail to be incorporated properly into the main nucleus as a cell divides.
In everyday language, they are a warning sign that something has gone wrong with the cell’s genetic material or the way its chromosomes are being divided.
That is why the micronucleus assay is an internationally recognised test for genotoxicity.
The untreated cells recorded 33.11 micronuclei per 1,000 binucleated cells.
After exposure to the lowest glyphosate concentration — 0.01 mg/L — that increased to 90.91.
At 0.1 mg/L it was 107.44.
At 1 mg/L, 97.78.
At 10 mg/L, 98.58.
And at the highest concentration, 100 mg/L, it rose dramatically to 268.44.
There wasn’t a concentration in the experiment at which the researchers could say: below here, we no longer see a significant effect.
They simply hadn’t gone that low.
And Glyphosate Stood Out From the Other Pesticides
Glyphosate wasn’t the only pesticide that produced concerning results.
Significant dose-related increases in micronuclei were also reported for lambda-cyhalothrin, deltamethrin, tebuconazole, piperonyl butoxide and fluopyram.
Other pesticides produced significant effects without a consistent dose-related pattern.
But among the pesticides tested individually, glyphosate produced the highest level of micronucleus formation in the Caco-2 cells.
At 100 mg/L, they counted 273 micronuclei following glyphosate exposure.
For comparison, the positive control used to make sure the experiment could detect chromosome damage — mitomycin C, a known chromosome-damaging agent — produced 217.
That doesn’t mean glyphosate is “more dangerous” than mitomycin C. These are experimental results under particular laboratory conditions, not a comparison of what the two substances would do to a person.
But within this experiment, the strength of the glyphosate response is difficult to ignore.
Is Glyphosate Genotoxic? Haven’t We Been Told It Isn’t?
This is where the study becomes particularly interesting.
Whether glyphosate is genotoxic has been debated for years.
The International Agency for Research on Cancer [PDF] concluded in 2015 that there was strong evidence that glyphosate causes genotoxicity as part of its assessment that glyphosate was probably carcinogenic to humans.
Other assessments and studies have reached different conclusions.
The authors of this new study acknowledge that controversy.
But they also point to something potentially important: the type of human cell being tested.
Previous experiments using human blood cells generally found effects from glyphosate at substantially higher concentrations.
Here, researchers were looking at human intestinal epithelial cells — cells chosen specifically because the intestine can be a first point of contact following ingestion.
And they found evidence of genotoxicity beginning at 0.01 mg/L.
According to the researchers, to the best of their knowledge, this is the first study using human epithelial cells to report genotoxicity from a single glyphosate exposure at concentrations of 0.01–1 mg/L.
So perhaps the question isn’t simply:
Is glyphosate genotoxic — yes or no?
Perhaps we should also be asking:
In which cells? At what concentrations? Under what conditions? And have we looked carefully enough at the lower end of the exposure range?
What About Glyphosate’s Acceptable Daily Intake?
Here we need to tread carefully.
The researchers themselves discuss their findings alongside regulatory exposure limits, including the Acceptable Daily Intake — the amount regulators consider can be consumed each day over a lifetime without appreciable health risk.
And understandably, seeing an effect at 0.01 mg/L alongside discussions about “acceptable” exposure immediately raises questions.
But the numbers are not directly comparable.
The cells in this experiment were exposed to concentrations measured in milligrams per litre (mg/L).
An Acceptable Daily Intake is measured in milligrams per kilogram of bodyweight per day (mg/kg/day).
One describes the concentration placed directly around cells in a laboratory dish. The other describes an amount a person consumes relative to their bodyweight.
They are not the same thing.
So we cannot take this study and say that it proves current regulatory limits for glyphosate are unsafe.
But neither should that end the conversation.
If researchers are finding evidence of chromosome damage in human intestinal cells at the lowest concentration they chose to test, surely it is reasonable to ask:
Should we be looking more closely at what happens at still lower concentrations — and at exposures that more closely resemble those experienced by people?
That seems a considerably more useful question than pretending this one experiment can settle the matter either way.
Then There Is the Question of Mixtures
There is another wrinkle.
Most pesticide safety discussions concern individual chemicals.
Real life isn’t necessarily so tidy.
Earlier work from the SPRINT project found mixtures of pesticide residues across agricultural environments — in soil, water, sediment, crops and outdoor air — and even in dust inside farmers’ homes.
So the researchers didn’t just test their ten substances individually.
They mixed some of them together.
One mixture contained three pesticides:
glyphosate + acetamiprid + tebuconazole.
It produced a significant genotoxic effect at the lowest concentration tested.
A second mixture containing eight pesticides, including glyphosate, produced a significant effect at the highest concentration.
There is an important qualification here.
The researchers did not find evidence that these chemicals acted synergistically to increase micronucleus formation. In other words, this experiment doesn’t support claiming that mixing these pesticides somehow magnified their genotoxicity beyond their individual contributions.
But the underlying question remains relevant.
How well does assessing pesticides one at a time reflect the way people are actually exposed to them?
So What Does This Study Really Tell Us?
It would be easy to take a result like this too far.
This experiment does not prove that glyphosate causes cancer in people.
It doesn’t show that eating food containing a particular glyphosate residue will damage your DNA.
It doesn’t reproduce the way the human body absorbs, distributes, metabolises and eliminates glyphosate.
And it doesn’t tell us what concentration actually reaches intestinal cells following everyday exposure.
But equally, those limitations don’t make the result disappear.
Researchers used human intestinal cells.
They used an internationally recognised test for genotoxicity.
They deliberately investigated concentrations lower than many previous studies had examined.
And every concentration of glyphosate they tested produced a significant increase in micronuclei — a recognised marker of genotoxicity and chromosome damage.
The lowest was 0.01 mg/L.
We don’t know what would have happened at 0.001 mg/L.
Or 0.0001 mg/L.
They weren’t tested.
And perhaps that is one of the most important things this study leaves us with.
Not an answer.
A reason to keep asking the question.
At No More Glyphosate NZ, our purpose has always been to raise awareness about the potential risks associated with glyphosate use and exposure.
That doesn’t mean treating every new study as proof of harm.
It means asking what the evidence is telling us — and sometimes, what we still haven’t bothered to find out.
When researchers find evidence of chromosome damage in human intestinal cells at every glyphosate concentration they test, the appropriate response surely isn’t to declare the debate over.
It’s to ask:
How low do we need to go before the effect disappears?
And perhaps more importantly:
Shouldn’t we want to know?
Read the Study
Truzzi F, Tibaldi E, Noferini R, et al. Multiple Pesticides and their Mixtures Tested for Genotoxicity in the Micronucleus Assays on Intestinal Caco-2 Cells.
Annals of Global Health. 2026;92(1):69.
Read the study: DOI: 10.5334/aogh.5345
Image Source & Attribution
The feature image on this page was created using AI-assisted image generation from an original concept developed by No More Glyphosate NZ and refined for publication in Canva.
