This plant compound lowered blood pressure in mice. Wipe out their gut bacteria and it stopped working.

Plenty of plant compounds lower blood pressure in animal studies. What almost none of them show is why — whether the molecule is doing the work itself, or whether your gut bacteria are doing it for you. A new mouse study pulls those apart with a simple test, and the answer is the second one. Strip the animals of their gut microbiome, and a compound that reliably controlled their blood pressure stopped working.

Key Takeaways

• Baicalin, a compound from Chinese skullcap, held hypertensive mice at roughly 135 mmHg systolic instead of the untreated 170 mmHg.
• In mice given broad-spectrum antibiotics, that protection vanished — their blood pressure looked like untreated hypertensive animals.
• Sodium butyrate, a fatty acid made by gut bacteria, reproduced baicalin's blood pressure and immune effects on its own.
• Only seven mice per group were analyzed, there was no standard blood pressure drug for comparison, and this is not a reason to take baicalin.

Baicalin's blood pressure effect disappeared without gut bacteria

Baicalin lowered blood pressure in hypertensive mice only when their gut microbiome was intact, according to a study published in Frontiers in Pharmacology in 2026 by Qiurong Xie, Jun Peng and colleagues at Fujian University of Traditional Chinese Medicine. Untreated mice infused with angiotensin II — a hormone used to drive experimental hypertension — climbed to what the authors describe as a plateau approaching 170 mmHg systolic. Mice given oral baicalin alongside the same infusion stabilized at approximately 135 mmHg.

The decisive arm was the third one. Mice given the identical baicalin dose while receiving a broad-spectrum antibiotic cocktail in their drinking water showed a blood pressure profile that, in the authors' words, "closely resembled that of the untreated Ang II cohort." Same compound, same dose, no benefit — the only thing removed was the bacteria.

That is the structure of a real mechanistic claim rather than a correlation. If a compound works through a pathway, deleting the pathway should delete the effect.

Why this matters for anyone managing high blood pressure

Hypertension is close to universal as health problems go. The CDC reports that nearly half of US adults have high blood pressure — 48.1%, or about 119.9 million people — and that only about 1 in 4 have it under control. That control gap is why alternative approaches attract so much interest, and why the supplement aisle is full of botanical blood pressure claims.

This study runs against the usual grain. It is not evidence that a supplement works. It is evidence that if a botanical compound like this works, the reason may have less to do with the compound reaching your bloodstream than with what your particular gut bacteria do with it — meaning the same pill could behave differently in two people.

It also fits a broader picture. The gut-blood-pressure link is established enough that the American Heart Association has issued a science advisory on hypertension and the gut microbiome, and a review of short-chain fatty acids of gut origin in hypertension lays out multiple routes by which bacterial metabolites influence blood pressure regulation.

Six arms, minipumps, and an antibiotic knockout test

The design is the paper's strongest feature. Male C57BL/6J mice received angiotensin II by implanted osmotic minipump at 500 ng/kg/min for 28 days, a standard and aggressive hypertension model. Animals were randomized into six groups: sham; angiotensin II alone; angiotensin II plus oral baicalin at 50 mg/kg/day; angiotensin II plus injected baicalein; angiotensin II plus baicalin plus antibiotics; and angiotensin II plus sodium butyrate at 200 mg/kg/day.

The antibiotic cocktail — ampicillin, vancomycin, neomycin and metronidazole in drinking water — began a week before the hypertension was induced and continued throughout, which is what makes the negative result interpretable rather than an artifact of timing.

The fourth arm is a clever control most studies skip. Baicalin is poorly absorbed and gut bacteria convert it into baicalein, the form that enters circulation. So the researchers injected baicalein directly, bypassing the gut. If baicalein in the blood were the whole story, injection should have worked at least as well as oral baicalin. It did not — the injected route produced a weaker effect on gut-associated immune cells, the opposite of what a simple absorption model predicts.

Butyrate and regulatory T cells connect the gut to the artery wall

The proposed chain runs from bacteria to a metabolite to the immune system to the blood vessel itself. Angiotensin II infusion collapsed bacterial diversity and depleted cecal butyrate, a short-chain fatty acid produced when gut bacteria ferment fiber. Baicalin restored the butyrate pool and brought back butyrate-producing genera including Faecalibacterium, Roseburia, Coprococcus and Butyricicoccus, while enriching Akkermansia and Lactobacillus.

It also repaired the gut barrier. Angiotensin II more than doubled intestinal leakage of tracer molecules and reduced the tight junction proteins ZO-1 and occludin; baicalin brought leakage back toward normal and suppressed circulating LPS (a bacterial cell wall fragment that provokes inflammation when it escapes the gut).

Downstream, hypertension shifted the immune balance — regulatory T cells (Tregs, which restrain inflammation) fell while pro-inflammatory Th17 cells rose. Baicalin reversed both shifts, and adding antibiotics abolished that reversal. Baicalin-treated mice accumulated more Tregs in the aortic adventitia, the outer layer of the artery wall, and showed reduced arterial wall thickening.

The sodium butyrate arm ties it together: butyrate alone, with no baicalin, blunted the angiotensin II blood pressure rise and reproduced the Treg expansion. In cell culture, Tregs exposed to butyrate suppressed proliferation of the vascular smooth muscle cells driving vessel remodeling — and blocking IL-10 undid that suppression, pinning down how the Tregs acted.

Seven mice per group, and the add-back experiment that was not run

The main limitation is size. Mice were randomized at 8 to 10 per group but analyses were standardized to n = 7 per group, roughly 42 animals across six arms. That is small enough that a single outlier moves a group mean, and it is the reason to treat these effect sizes as provisional even though the logic is sound.

There is also a gap in the causal chain worth being precise about. Butyrate was given as its own arm to hypertensive mice with intact microbiomes — it was not added back to the antibiotic-depleted animals. So the study shows that depleting bacteria abolishes baicalin's effect, and separately that butyrate alone reproduces it. The single cleanest experiment, restoring butyrate to depleted mice and watching the benefit return, was not performed. Nor were fecal transplants or germ-free mice used.

The authors are candid about further gaps: no standard antihypertensive such as losartan or amlodipine was included as a positive control, so there is no benchmark for how meaningful a 35 mmHg difference is against existing drugs; plasma concentrations of baicalin were never measured; blood pressure was recorded by tail-cuff rather than the more reliable radiotelemetry; and exact endpoint values are plotted rather than reported numerically.

What this does not mean for baicalin supplements

This is not a reason to take baicalin, and the study should not be read as supporting any supplement. These were mice on an artificially induced form of hypertension, dosed at 50 mg/kg/day of a purified, 98%-pure compound — not a commercial extract, and not a dose that translates directly to humans. No human trial has shown baicalin lowers blood pressure. Skullcap products have been associated with liver injury in case reports, and baicalin interacts with drug-metabolizing enzymes, which matters enormously if you already take blood pressure or blood thinning medication.

If you have hypertension, the interventions with real human evidence remain the boring ones: sodium reduction, physical activity, weight management, alcohol moderation, and taking prescribed medication consistently. The plausible thread from this research is dietary fiber, since fiber is what gut bacteria ferment into butyrate — though this study tested neither fiber nor any human diet.

What comes next

The obvious next experiments are the ones missing here: butyrate add-back in microbiota-depleted animals, fecal transplants from treated to untreated mice, and a head-to-head against a standard antihypertensive. Human relevance would require a trial measuring blood pressure, gut composition and butyrate together — years away, and most compounds at this stage never make it.

Common Questions

Does baicalin lower blood pressure in people? No human trial has established that. This evidence is entirely from mice with drug-induced hypertension, and the effect depended on mouse gut bacteria that differ substantially from human ones.

Should I take butyrate supplements for blood pressure? The study gave butyrate to mice, not people, and oral butyrate supplements in humans have not been shown to lower blood pressure. Fiber-rich foods are the established way to increase butyrate production, though this study did not test that either.

Why would antibiotics block a plant compound from working? Because baicalin is poorly absorbed on its own and gut bacteria convert it into the absorbable form. Remove the bacteria and much of the conversion stops, along with the downstream bacterial metabolites the effect appears to depend on.

References

Qiurong Xie, Yanyan Yang, Jiekun Lin, et al. (2026). Gut microbiota-derived butyrate contributes to baicalin-induced attenuation of hypertensive vascular remodeling via adventitial immunity. Frontiers in Pharmacology. https://doi.org/10.3389/fphar.2026.1835174