Transcript
A claim circulating widely right now goes something like this: a gut bacterium called Lawsonibacter asaccharolyticus essentially "drinks" your morning coffee, and in doing so it lowers your blood pressure and blood sugar. Coffee's real benefit, the claim continues, comes not from caffeine but from polyphenols — the plant compounds that survive roasting. Parts of that story are supported by solid science. Other parts are not. And the distinction matters, because the gap between "we found a striking association" and "we know this bacterium does something good for your health" is exactly where nutritional science tends to go wrong.
What does the human evidence actually show? What is lab-only? What is association-only? And what, if anything, can be said with reasonable confidence?
Start with Lawsonibacter asaccharolyticus itself. This is a relatively newly characterised species — first formally described in 2018 as a butyrate-producing bacterium isolated from human faeces. Butyrate is a short-chain fatty acid that gut cells use as fuel, and species that produce it tend to attract scientific interest. But as of 2018, this organism was obscure. Most microbiome researchers had not encountered it. Then in 2024, a study published in Nature Microbiology by Manghi and colleagues changed that.
The Manghi study is the largest investigation ever conducted into the relationship between coffee drinking and the gut microbiome. The primary analysis drew on 22,867 participants — nearly twenty-three thousand people — who provided stool samples for DNA sequencing and detailed reports of their long-term coffee habits. The researchers then integrated 211 additional cohorts, bringing the total to 54,198 individuals. Those are not numbers that appear often in microbiome research, where studies of a few hundred participants have historically been the norm. When a finding replicates across nearly twenty-three thousand people and then holds again across tens of thousands more drawn from entirely different cohorts, the association is not a statistical accident. That level of consistency is hard to dismiss.
The finding itself is striking. The researchers could predict whether someone was a coffee drinker simply by looking at the composition of their gut microbiome, with an accuracy of 0.89 out of a maximum of 1.0. That means a coffee drinker's gut microbiome is statistically quite distinguishable from a non-drinker's.
The pattern across intake levels makes the association harder to wave away. People who drank more than three cups of coffee per day had levels of L. asaccharolyticus that were between 4.5 and 8 times higher than people who rarely or never drank it. Moderate drinkers still had levels 3.4 to 6.4 times higher. But the dose effect was limited: the step from moderate to high intake was at most about 1.4-fold, and not statistically significant in three of five cohorts.
So the association is real, it is large, and it replicates. That is where the confidence should sit for now — on the association. The researchers themselves are explicit about what comes next. As Scientific American reported on this study, the scientists described their next step as "trying to link these bacteria and the related metabolites to health outcomes." They are saying the causal chain from bacterium to human health benefit has not yet been established. What the bacterium actually does for the person carrying it remains an open question.
To recap what we have so far: the link between coffee drinking and L. asaccharolyticus abundance is real, consistent across tens of thousands of people, with only a limited dose effect. What it means for health is, at this point, unknown.
Now, what is driving the association? This is where the polyphenol story becomes relevant, and where some of the viral claims are closer to the evidence.
The Manghi study included plasma metabolomics — chemical analysis of blood — on 438 samples. Among all the compounds measured, quinic acid emerged as the metabolite that tracked most closely with L. asaccharolyticus abundance. Quinic acid is a breakdown product of chlorogenic acid, one of the primary polyphenols in coffee. When gut bacteria metabolise chlorogenic acid, quinic acid is among the compounds they produce. Its co-elevation with the bacterium in blood suggests a mechanistic relationship — the bacterium is not just sitting alongside coffee consumption, it may be actively processing coffee's plant compounds.
The lab experiments support this interpretation. When L. asaccharolyticus was grown in culture and coffee was added to its medium, the bacterium grew faster. Crucially, the same growth stimulation occurred with decaffeinated coffee. If caffeine were the active ingredient feeding this bacterium, decaf would not produce the same effect. The fact that it does points toward the polyphenol fraction — chlorogenic acid, caffeic acid, quinic acid, and related compounds — as the relevant inputs.
This is, to be clear, in vitro evidence — a petri dish, not a person. Bacteria behave differently in culture than in the complex, competitive environment of the human gut. Growing faster on a lab plate does not automatically translate to growing faster inside you, or to producing more of any particular metabolite. But the convergence between the metabolomics finding in blood and the culture experiments in the lab is at least consistent. It gives the polyphenol hypothesis some real footing.
The broader polyphenol picture also fits. A review of habitual coffee intake published in 2025 notes that both caffeinated and decaffeinated coffee appear to reduce inflammation markers, and that the microbiome changes associated with coffee overlap between the two types. Caffeinated coffee showed some distinct effects — reductions in anxiety, psychological distress, and blood pressure in an intervention study — while decaffeinated coffee was associated with improved sleep and memory. Both types lowered stress and depression markers. Two beverages sharing their polyphenol content but differing dramatically in caffeine producing overlapping benefits is consistent with polyphenols doing meaningful work.
To recap the mechanism: evidence from both lab culture and blood metabolomics points toward coffee's plant compounds rather than its caffeine as the driver of the L. asaccharolyticus association. That is a reasonable, evidence-grounded inference. It does not yet explain what, if anything, the bacterium then does.
Before moving into the broader health claims — the diabetes and cardiovascular associations that make coffee one of the most studied dietary exposures in epidemiology — it is worth being explicit about where we stand. Everything discussed so far is about the microbiome link. We have a robust association between coffee and one bacterium, plausible lab evidence for why polyphenols might drive it, and metabolomics data pointing to quinic acid as a connecting molecule. None of that tells us whether the bacterium mediates any of coffee's known health associations. That step has not been taken yet.
Coffee's epidemiological track record is extensive and mostly positive, though it is worth being careful about how strongly any of these findings are stated.
A cross-sectional study drawing on 468,629 individuals linked light-to-moderate coffee intake to lower all-cause mortality, lower cardiovascular mortality, and lower stroke incidence. That is a consistent and large finding. But cross-sectional data has a fundamental limitation: it captures a moment in time and cannot establish that coffee caused those differences. People who drink moderate amounts of coffee may differ from non-drinkers in dozens of ways — exercise habits, socioeconomic status, other dietary patterns — and even careful statistical adjustment cannot account for every potential confound.
The review cited in the habitual coffee research found, across multiple studies, a 27% reduction in Alzheimer's disease incidence among coffee consumers relative to non-consumers. That figure comes from a meta-analysis, which pools data from many studies to produce a more stable estimate — but it remains an observational finding. The consistency across populations and study designs is persuasive as evidence that something is happening. Because these findings are observational, they cannot show that coffee is the cause of those outcomes.
Now for the specific viral claim: that L. asaccharolyticus lowers blood pressure and blood sugar. This is where the evidence falls away most sharply.
No human trial has tested whether changing L. asaccharolyticus abundance — whether by introducing it directly or by altering coffee intake — produces changes in blood pressure or blood glucose. The claim that the bacterium does this is extrapolated from the association between coffee drinking, the bacterium's abundance, and coffee's separate epidemiological associations with cardiovascular and metabolic outcomes. That is three inferential leaps chained together, none of them validated by a trial. The claim is not impossible. It might turn out to be right. But calling it established would be incorrect.
There is also a tension in the data worth naming honestly. The review on habitual coffee intake notes that non-coffee drinkers show more stable blood pressure and potentially healthier neurophysiological rhythms in some analyses, while simultaneously the caffeinated coffee arm of an intervention study showed reduced blood pressure in participants. These two observations — acute measured benefit from caffeinated coffee versus steadier baseline blood pressure in non-drinkers — point in opposite directions and are not yet reconciled. Chronic caffeine exposure and its acute measured effects may operate differently, and the available material does not resolve this.
The NR4A1 pathway, which appears in some of the more detailed viral claims as a molecular mechanism linking caffeic acid to blood pressure and blood sugar regulation, is another area where the research in hand is silent. No human evidence for this molecular pathway appears in the sources underpinning this episode. Cell studies and animal models can generate hypotheses about such mechanisms, but they sit at the far end of the evidence hierarchy — relevant to scientists designing the next experiment, but not to anyone drawing conclusions about what coffee does in humans.
One more honest qualification. The relationship between polyphenol bioavailability and gut microbiome composition runs in both directions. Coffee polyphenols appear to shape which bacteria thrive in your gut. Your existing microbiome composition influences how much quinic acid ends up in your blood after a cup of coffee. This circularity means that the coffee–microbiome relationship may look quite different between individuals, and what is true on average across 54,000 people may not describe what happens in any particular person with their own microbial baseline. The review on habitual coffee intake explicitly flags this variability as unresolved.
There is also a question about timing. One intervention study found that just two days of reintroducing coffee after a period of abstinence was not enough to restore changes in gut microbiome diversity. How quickly the microbiome responds to coffee introduction or withdrawal, and whether changes are durable, has not been worked out. The association we see in large observational cohorts presumably reflects long-term habitual consumption, not what happens after a week of drinking more coffee.
So here is where the evidence actually lands, stated plainly.
The association between coffee drinking and elevated L. asaccharolyticus abundance held across 22,867 primary participants and was extended across 211 additional cohorts. People in the highest-intake group had roughly 4.5 to 8 times more of the bacterium than people who rarely or never drink coffee, but the step from moderate to high intake was small, at most about 1.4-fold, and not statistically significant in three of five cohorts. That association is real, it is large, and it is consistent. It is also, for now, just an association.
The polyphenol mechanism — the hypothesis that coffee's plant compounds, not its caffeine, feed this bacterium — has lab support. Decaf produces the same growth stimulation in culture. Quinic acid, a polyphenol metabolite, tracks with bacterial abundance in blood. The evidence for this mechanism is plausible and coherent, though it comes from in vitro experiments and blood metabolomics rather than from trials that manipulated polyphenol intake and measured bacterial and clinical outcomes.
The leap from there to "this bacterium lowers blood pressure and blood sugar" is not supported by human evidence at this time. Coffee's associations with cardiovascular and metabolic health come from observational studies of coffee drinking, not from any trial that tested the bacterium's effect on clinical outcomes. The causal chain from cup of coffee, to L. asaccharolyticus, to lower blood pressure in humans has not been demonstrated. The researchers who conducted the largest study in this area have themselves identified establishing that link as the work still to be done.
The most important question remains open: does this bacterium mediate any of coffee's known health associations, or is it a co-occurring marker of something else entirely? That cannot be answered yet. The science is moving, the dataset is large enough to support confidence that something real is being observed, and the polyphenol hypothesis gives researchers a plausible mechanism to test. But the viral version of this story — in which the mechanism is settled, the bacterium is proven to regulate blood pressure and glucose, and the lesson is clear — has outrun the evidence considerably. That gap, between what the data shows and what the claim asserts, is precisely where scepticism belongs.