Gut bacteria and brain chemistry: the first human study, read honestly
For fifteen years the gut-brain field has rested on two kinds of evidence. Animal work, where you can cut a nerve or raise a germ-free mouse and watch what happens. And human association studies, where you compare the bacteria of people with a diagnosis against people without one. What has been missing is the piece in the middle: a look inside a living human head to see whether what the gut bacteria are equipped to do has anything to do with the chemistry actually present in the brain.
In 2026 a team from the University of Surrey and the University of Roehampton published that study.
What they did
Sixty-one healthy young women aged 17 to 25 gave a stool sample and had their brains scanned. The scan was proton magnetic resonance spectroscopy, which measures the concentration of specific chemicals in a defined piece of brain tissue without any dye or tracer. They measured two: GABA, the brain's main calming messenger, and glutamate, its main excitatory one. The balance between the two is what researchers call excitatory-inhibitory balance, and it matters for learning, attention and emotion.
They measured those two chemicals in three places: the dorsolateral prefrontal cortex, involved in cognitive control and emotion regulation; the anterior cingulate cortex, involved in attention and emotion regulation; and the inferior occipital gyrus, which handles visual processing and was included partly as a comparison region.
The stool samples went through shotgun metagenomic sequencing. That does not just list which bacteria are present. It reads the genes they carry, which tells you what those bacteria are equipped to make and break down. The team looked at pathways for glutamate degradation, GABA metabolism, short-chain fatty acids, inositol, p-cresol, tryptophan metabolism and propionate production.
What they found
Different microbial pathways were associated with GABA, glutamate and excitatory-inhibitory balance, and which pathway mattered depended on which brain region was being measured. The associations were region-specific rather than global.
They also compared those pathways against how the participants said they were feeling. A GABA-related microbial pathway was associated with trait anxiety and social anxiety. Pathways involved in tryptophan metabolism, the route the body uses to make serotonin, were associated with depressive symptoms and social anxiety. A pathway for producing propionate, one of the short-chain fatty acids that gut bacteria make from fibre, was associated with poorer sleep quality.
Why this is a first
The single most cited experiment in this field is still from 2011, when mice fed a Lactobacillus showed changes in GABA receptors across brain regions, and the effect vanished when the vagus nerve was cut. That established the wiring, in mice. Separately, a 2012 laboratory study screened 91 human gut bacteria and found that a strain of Lactobacillus brevis was the most efficient GABA producer of the lot, converting up to 90 per cent of its raw material. That established the capability, in a dish.
Neither told you whether any of it registers in a living human brain. This study is the first to put the two measurements side by side in people.
Four things it cannot tell you
This matters more than the finding, and the authors say so themselves.
It is an association, not a cause. The study is cross-sectional: everything was measured once, at the same time. Dr Nicola Johnstone, who led it, put it plainly: “We are not saying that a particular gut bacterium causes anxiety or changes a particular brain chemical.” The gut could be influencing the brain, the brain could be influencing the gut, or something else could be driving both.
It measured potential, not production. Sequencing tells you which genes are present. It does not tell you those genes were switched on, or that the compound was actually made, or that it reached the brain. The team is explicit that they measured genetic capacity rather than metabolite output.
It was 61 healthy young women. Not children, not teenagers below 17, not men, not anyone with a diagnosis. Whether the same associations hold in a nine-year-old, or a forty-year-old, or someone who is unwell, is unknown.
Nobody was given anything. No probiotic, no supplement, no diet change. Nothing was tested to see if it moved. Professor Kathrin Cohen Kadosh, the senior author, named the next step directly: “The next question is whether changing these pathways can actually alter brain chemistry. That will require larger longitudinal and intervention studies.”
Where this leaves the field
Honestly, roughly where it was, but with a floor under it. The gut-brain axis has never been short of mechanism. It has been short of human measurement. A study showing that the neurochemical capacity of someone's gut bacteria tracks with the GABA and glutamate in specific parts of their brain makes the mechanism less hypothetical, without making any product or any diet proven.
The honest position is the one the researchers themselves take: this is a signpost for the intervention studies that have not been done yet, not a result you can act on.
The paper
Johnstone N., Cohen Kadosh K. and colleagues. “Empirical evidence for gut microbial influence on human brain neurochemistry via the gut-brain axis.” Molecular Psychiatry, 2026. DOI 10.1038/s41380-026-03813-y. PMID 42642466. A preprint appeared on bioRxiv in May 2025.