Apigenin

The verdict

Is Apigenin worth taking?

Apigenin is a flavone found in parsley, celery and chamomile, sold at 50 mg as a nightly sleep aid on the strength of a mechanism that the specialist literature does not support.

No randomized controlled trial of apigenin for sleep has ever been run. Not one, at any dose, in any population. The 2024 review most often cited in its favor says so in its own abstract: the human clinical support comes from studies of chamomile extract, and apigenin's own sleep and sedative findings are in mice, rats, flies and worms.

The chamomile trials are anxiety trials, not sleep trials, and apigenin was 1.2 percent of what participants swallowed. The extract used in those trials was standardized to 1.2 percent apigenin at a maximum of 1,100 mg a day — about 13 mg of apigenin, alongside the other 98.8 percent of the plant. Chamomile has been tested for insomnia exactly once, and that trial found no significant change in insomnia severity.

And the mechanism runs the wrong way. Apigenin does bind the benzodiazepine site of the GABA-A receptor. But the electrophysiology study that measured what happens next found it reduced GABA-activated chloride currents — the opposite direction from a benzodiazepine — and two separate rat studies found its locomotor-slowing effect was not blocked by flumazenil, meaning it is not working through that receptor at all. Three of the four animal studies that tested apigenin for anxiolytic effect found none.

The bottom line. Chamomile tea at night is a pleasant and harmless habit with modest evidence behind it as a whole plant. Isolated apigenin at 50 mg is a different proposition: a dose with no study behind it, for an indication with no trial behind it, on a mechanism its own specialist literature describes as weak, contradictory, and probably not the one being claimed. It appears safe. That is the strongest thing the evidence supports.

What it does

What Apigenin does in the body

The story on every label is that apigenin binds the same receptor site as benzodiazepines, so it calms you down the way a very mild sedative would. The first half is true. The second half is where it falls apart, and the contradictions are in the primary literature rather than hidden anywhere.

1995, Planta Medica — the paper the claim comes from. Apigenin isolated from chamomile competitively inhibited flunitrazepam binding at the central benzodiazepine receptor with a Ki of 4 µM. In mice it was anxiolytic on the elevated plus maze — and the authors were explicit that this happened without sedation and without muscle relaxation. Only at ten times that dose did they see mild sedation, a 26 percent reduction in locomotor activity. So the foundational paper describes an anxiolytic that does not sedate, which is not what a sleep supplement needs.

2000, Biochemical Pharmacology — what the receptor actually does. The same compound, tested on cultured cerebellar granule cells, reduced GABA-activated chloride currents in a dose-dependent fashion. A benzodiazepine increases them. Apigenin also reduced the latency to chemically induced convulsions, which is the pro-convulsant direction. In rats it lowered locomotor activity but showed no anxiolytic, muscle-relaxant or anticonvulsant activity, and the authors concluded the locomotor effect “cannot be ascribed to an interaction with GABA-A benzodiazepine receptor” because flumazenil did not block it.

2000, Fitoterapia — the same result independently. Apigenin reduced locomotor activity in rats at 25 mg/kg and above, but failed to produce an anxiolytic effect, and again the effect was not reversed by flumazenil. In the same experiment chrysin, a closely related flavone, was anxiolytic and was blocked by flumazenil — so the method could detect a genuine benzodiazepine-site effect. It just did not find one for apigenin.

A 1997 paper testing apigenin on learning tasks reported, in passing, that it “showed no anxiolytic action in an elevated plus maze.” That is three independent groups failing to find the effect the supplement is sold for.

How the specialists describe it. A 2017 reference chapter written by GABA receptor pharmacologists calls apigenin “a weak inhibitor of flumazenil binding,” notes that in cultured cerebellar granule cells it “is described as a negative modulator of GABA action,” and adds that it is unlikely to work by enhancing diazepam binding “as it is known to inhibit such binding.” No affinity constant for apigenin at that site appears in the chapter at all.

None of which means apigenin does nothing. It plainly does things — it slows locomotor activity in rodents reliably, it inhibits the enzyme CD38, and it has a long list of effects in cell culture. It means the specific mechanism printed on the bottle is the one piece of the story the literature least supports.

The evidence

What the research actually shows

Emerging on the site. For apigenin as an isolated compound taken for sleep, the honest count of human trials is zero, and everything below is about the plant it is extracted from.

What chamomile has actually been tested for is generalized anxiety disorder. The 2009 trial that started the line of work randomized 57 outpatients with mild to moderate GAD to chamomile extract or placebo for eight weeks and found a significantly greater reduction in anxiety scores (p=0.047). A 2016 open-label phase treated 179 people and reported a 58 percent response rate. Then the randomized part of that same 2016 study — the relapse-prevention phase, which is the properly controlled half — missed its primary outcome: time to relapse gave a hazard ratio of 0.52 with a confidence interval of 0.20 to 1.33, p=0.16.

Chamomile has been tested for insomnia once. A 2019 meta-analysis covering twelve randomized trials states it plainly: there is only one trial that evaluated chamomile for insomnia, and it found no significant change in the Insomnia Severity Index. The same meta-analysis found no pooled benefit for anxiety as a symptom (standardized mean difference −0.15, 95 percent CI −0.46 to 0.16, p=0.42), while sleep quality did improve (SMD −0.73).

The sleep meta-analysis, read properly. A 2024 review of ten chamomile sleep studies in 772 people found a significant reduction in the Pittsburgh Sleep Quality Index of 1.88 points — with a confidence interval running from 3.46 to 0.31 and heterogeneity of 88.4 percent, pooled from five studies. Within the same review: sleep efficiency was unchanged in two studies and deteriorated in one, and daytime functioning did not improve in any of the three studies that measured it. The authors also note that of all ten studies, only one checked whether blinding had worked and tested the purity or potency of the product used.

Now the arithmetic that connects the plant to the capsule. The extract in the anxiety trials was standardized to 1.2 percent apigenin, with a maximum daily dose of 1,100 mg. That is roughly 13 mg of apigenin per day — and the other roughly 1,087 mg was everything else chamomile contains, including the flavone chrysin, which in head-to-head rodent work was the one that produced a genuine benzodiazepine-site anxiolytic effect while apigenin did not.

So a 50 mg apigenin capsule is about four times the apigenin the trials delivered, with none of the material the trials actually administered. There is no sense in which it is the same intervention at a cleaner dose.

Where does 50 mg come from? I could not find a published rationale anywhere — no dose-finding study, no pharmacokinetic justification, no human trial at that dose or any other. It is a round number that became a standard because it was recommended in public and then copied across the category.

The concentration problem

Every effect needs a concentration no human has reached

This is the part that decides whether any of the mechanism matters, and it is simple arithmetic once the numbers are in one place.

Apigenin's effects are measured in micromolar concentrations in dishes and membranes. What has been measured in human blood is in nanomolar concentrations — a thousand nanomolar to the micromolar. There is exactly one human pharmacokinetic measurement: eleven adults ate roughly 2 g of parsley per kilogram of body weight, about 17 mg of apigenin, and plasma apigenin came in between 28 and 337 nmol/L six to ten hours later.

The highest concentration ever recorded in a person is about twelvefold below the binding constant for the receptor site the supplement is sold on — and that is plasma, before anything crosses the blood-brain barrier, where levels are lower again.

Two honest qualifications. First, 17 mg of dietary apigenin is not 50 mg of purified apigenin in a capsule on an empty stomach, and nobody has measured what that produces — no human Cmax or Tmax for apigenin has ever been published. The pharmacokinetic figures quoted everywhere, including a bioavailability of around 30 percent and a half-life of about two and a half hours, come from rats. Second, concentrations in the gut wall during absorption are far higher than anything reaching the bloodstream, which is why an intestinal interaction is more plausible than a systemic hormonal one.

A 2021 review that modeled this for a different purpose concluded that reaching the low-micromolar range from food “would require heroic ingestion amounts and is not feasible,” while allowing that capsules might get there. Might. That remains a calculation rather than a measurement, because the measurement has not been taken.

Food first

Getting Apigenin from food

Parsley is the richest source, then celery, then chamomile. And there is a detail here that matters more than the quantities, because it changes what you are actually eating.

In plants, apigenin is almost never free apigenin. It is bound to sugars — in parsley as apiin, in chamomile as apigenin-7-glucoside. A paper validating the analytical method for chamomile preparations states it directly: chamomile flowers present glycosylated apigenin rather than free apigenin. Those sugars have to be cleaved, largely by gut bacteria, before anything is absorbed. The molecule in the capsule is not the molecule in the plant, and the step between them depends on a microbiome that varies from person to person and has never been characterized for this compound.

What the one human study ate. The single human pharmacokinetic measurement fed eleven adults roughly 2 g of parsley per kilogram of body weight. For an 80 kg adult that is about 160 g of parsley — a couple of large supermarket bunches, leaves and all, at one sitting. That delivered about 17 mg of apigenin and produced plasma levels well below the concentration at which apigenin does anything in a dish. This is what “heroic ingestion amounts” means in practice.

Two things worth knowing in a kitchen. Drying concentrates flavonoids by weight, because what leaves is water — so dried parsley and dried chamomile carry more apigenin per gram than the fresh equivalent, though you also eat far less of them. And when you cover a cup of steeping chamomile, what you are keeping in the cup is the volatile oil, the part that carries the aroma. The apigenin glucoside is water-soluble and not volatile; it comes out with time in the water, not with a lid. Covering the cup is worth doing for the smell and the taste. It is not doing anything for the flavonoid.

The honest summary. Chamomile tea before bed is a good habit — warm, caffeine-free, pleasant, part of a wind-down, and backed by modest evidence as a whole plant. It is not a delivery route for 50 mg of apigenin and it was never trying to be. If you like it, drink it because you like it.

What the guide adds

The part that is not on this page

There is no human trial of this compound. Not for sleep, not for anxiety, not for anything, at any dose. Everything on this page about effects in people is about chamomile extract or about dietary apigenin, and everything about apigenin itself is from animals, cells or membranes. That is the single most important caveat and it is rarely stated.

A 2017 review of its absorption and metabolism concluded that more work was needed “before apigenin can be brought to clinical trials.” That was the state of the science nine years ago, and it is still broadly the state of the science. Meanwhile it is sold for nightly use.

Drug interactions have never been tested in a person, and this is the one caution with a plausible route. Apigenin inhibits CYP3A4 and CYP2C9 in enzyme assays, with a Ki around 6 µM for CYP3A4. Systemic blood levels never approach that, but concentrations in the gut wall during absorption are much higher than in blood, and CYP3A4 in the intestinal wall is where a lot of first-pass drug metabolism happens. CYP3A4 handles something like half of all prescription drugs. Nobody has run the study. If you take a medication on that list, this is a question for the pharmacist rather than for a supplement page.

The hormonal claims cut both ways and neither has human support. Apigenin inhibits aromatase, the enzyme that makes estrogen, with an IC50 of 20 µM in human cells — roughly sixty times the highest level ever measured in human blood, and in the same experiment a pharmaceutical aromatase inhibitor was a thousand times more potent. It has also been shown to act as an estrogen receptor agonist in breast cancer cells. So you will find it marketed as estrogen-lowering and warned about as estrogenic, both from the same in vitro literature, with nothing in humans either way.

Nobody knows what is in the products. The 2024 review of chamomile sleep studies found that of ten trials, only one tested the purity or potency of what it was giving people. I could find no independent analysis of retail apigenin products at all — no third-party testing data, no label-accuracy survey.

Combining it with other sedatives is untested. It is commonly sold and taken stacked with magnesium and theanine. There is no human data on any of those combinations, and the one combination study I found was in mice.

Safety looks genuinely unremarkable. Apigenin is a normal constituent of parsley, celery and chamomile, eaten by everyone in small amounts with no signal of harm, and no serious adverse events have been reported at supplement doses. The problem with this compound is not that it looks dangerous. It is that almost nothing is known, and it is sold as though a great deal is.

About the author

Who graded this

Daniel Edlund is a Le Cordon Bleu-trained private chef in Los Angeles, a certified personal trainer and strength coach, and an internationally licensed massage therapist. Thirty years of strength training and fifteen years cooking in private estate kitchens. He is the author of The Complete Supplement Guide, which grades 147 entries against the clinical literature, and of The EASY Way to Organic Cooking, first published in 2013.

More about Daniel →

This page is information, not medical advice. Supplements interact with prescription medication and with each other, and the right answer depends on your own bloodwork and history. Talk to your doctor before starting anything, particularly if you are pregnant, nursing, managing a health condition, or already taking medication.

Also in Sleep, Stress & Cortisol Management

Related entries

• GABA — Emerging evidence
• Saffron Extract — Moderate evidence
• Reishi Mushroom — Emerging evidence
• CBD — Moderate evidence
• L-Theanine — Strong evidence
• Holy Basil / Tulsi — Moderate evidence
• Melatonin — Strong evidence
• Glycine — Strong evidence
• Magnesium — Strong evidence
• Valerian Root — Moderate evidence

All 147 entries, graded →