Apigenin, Luteolin, and the Celery Phthalides

Table of Contents

1. Three Families of Compounds
2. Apigenin
3. Luteolin
4. Where the Flavones Actually Sit in the Plant
5. The Phthalides and That Smell
6. 3-n-Butylphthalide: From Celery Seed to a Stroke Drug
7. The Dose Gap: Micromolar Versus a Stalk
8. What Cooking, Storage and Juicing Do
9. Celery Seed, Celery Salt and Celery Seed Extract
10. What It Is Reasonable to Conclude
11. Practical Ways to Get More of Them
12. Key Research Papers
13. Connections
14. Featured Videos

Three Families of Compounds

Celery’s pharmacologically interesting chemistry falls into three groups, and they behave completely differently. Keeping them apart is the single most useful thing you can do when reading anything about celery research.

1. Flavones — apigenin and luteolin. Water-soluble pigment-family compounds, present in the plant mostly bound to sugars. Studied for anti-inflammatory, antioxidant and anticancer activity, almost entirely in cells and animals.
2. Phthalides — sedanolide, sedanenolide, 3-n-butylphthalide and relatives. Oily, volatile, strongly aromatic. These are the smell and taste of celery, they concentrate in the seed, and they carry the plant’s best claim to real pharmacology.
3. Furanocoumarins — psoralen, bergapten, xanthotoxin. The plant’s chemical defence system, and the reason handling large amounts of celery can cause a sunlight-triggered skin burn. These are covered on the Allergy and Photosensitivity page and are not a benefit.

Alongside these sits the entirely separate nitrate story, which is about a simple inorganic ion rather than a plant compound at all. Four distinct arguments, one vegetable.

Back to Table of Contents

Apigenin

Apigenin is a flavone — a subclass of flavonoid — found in celery, parsley, chamomile, artichoke and a scattering of other plants. Parsley and dried chamomile are far richer sources than celery; celery matters because people eat it in bulk while nobody eats a bowl of parsley.

What the laboratory literature shows, and it is a very large literature, is that apigenin does several things to cells:

• Dampens inflammatory signalling. In cultured cells and animal models it inhibits NF-κB, a master switch that turns on inflammatory gene programmes, and reduces production of inflammatory messengers.
• Induces apoptosis in cancer cell lines. Apigenin has been tested against a long list of tumour cell types and repeatedly triggers programmed cell death and cell-cycle arrest, with relative selectivity for malignant over normal cells in several models. Sanjeev Shukla and Sanjay Gupta’s 2010 review is the standard entry point.
• Interacts with benzodiazepine binding sites. This is the pharmacological basis usually offered for chamomile’s calming reputation. In animal work apigenin shows mild anxiety-reducing effects. There is no good human trial establishing this for dietary apigenin.
• Acts as an antioxidant in chemical assays and reduces markers of oxidative stress in animal models.

The important caveat, which the reviews themselves state clearly, is that essentially all of this is preclinical. There are no large randomised human trials showing that apigenin from food prevents or treats any disease. It also has poor water solubility and limited oral bioavailability, which is why so much of the recent literature is about formulation and delivery rather than about outcomes.

Back to Table of Contents

Luteolin

Luteolin is apigenin’s close structural relative — the same flavone skeleton with one extra hydroxyl group — and it travels with it. Celery, parsley, thyme, peppers and chamomile all carry both. Miguel López-Lázaro’s widely cited review lays out the distribution and the biology.

Luteolin’s laboratory profile overlaps apigenin’s and adds a few distinctive threads. It is a fairly potent inhibitor of inflammatory enzymes and cytokine release; it stabilises mast cells, which is why it appears in the literature on allergic inflammation; and it has been studied for neuroinflammation specifically, with animal work suggesting it crosses into brain tissue and reduces microglial activation. That neuro-inflammatory angle is the reason luteolin turns up in supplement formulations aimed at brain fog and mast cell activation.

The same caution applies, in the same words. This is real biology, it is measured in real experiments, and it has not been translated into demonstrated human benefit from eating luteolin-containing food. Anyone selling you luteolin capsules on the strength of the cell-culture literature is skipping the step that matters. See Luteolin and Apigenin for the compound-level pages.

Back to Table of Contents

Where the Flavones Actually Sit in the Plant

Two practical facts follow from how celery stores its flavones, and both are useful in the kitchen.

First: they are mostly bound to sugars. In fresh celery, apigenin and luteolin exist largely as glycosides — the flavone with sugar molecules attached — rather than in the free form used in laboratory experiments. Celery’s characteristic version is apiin, apigenin carrying a glucose and an unusual branched sugar called apiose. That sugar was first characterised from parsley and celery and is named after the genus Apium itself. Analytical chemists working on celery routinely have to hydrolyse the glycosides — strip the sugars off enzymatically or with acid — before they can even measure the free flavones, which is exactly what the extraction papers cited below are doing. Your gut does some of that hydrolysis too, with help from gut bacteria, but the conversion is partial and variable between people.

Second: the leaves beat the stalks. Flavone concentrations are substantially higher in celery’s leafy tops than in the pale blanched stalks, and higher again in the dark green outer stalks than in the pale inner heart. This is the direct consequence of the blanching described on the History page: growing the stalks in darkness suppresses exactly the light-driven secondary chemistry that produces these compounds. The vegetable was deliberately bred and grown to be less bitter, and bitterness in the Apiaceae is largely these compounds.

Which produces a small, genuinely actionable conclusion. Most people buy celery, use the pale stalks, and bin the leaves — throwing away the most flavone-rich part of the plant. The leaves are edible, taste like concentrated celery, and belong in soups, stocks, salads and stuffings.

Back to Table of Contents

The Phthalides and That Smell

Phthalides are what make celery smell like celery. They are small, oily, volatile molecules built on a benzofuranone skeleton, and celery’s aroma is a blend of several — sedanolide and sedanenolide contribute most of the characteristic scent, with 3-n-butylphthalide and its relatives alongside. Related compounds give lovage and fennel their own smells; phthalides are an Apiaceae family signature.

Because they are volatile and fat-soluble rather than water-soluble, phthalides behave in the opposite way to the flavones:

• They are concentrated in the seed and the essential oil. Celery seed oil is many times richer in phthalides than the stalk.
• They are lost to heat and air. Long boiling drives them off — which is why a soup smells intensely of celery while it cooks and less so by the time it is served, and why celery’s flavour mellows so much with cooking.
• They dissolve into fat and alcohol better than into water, which is why celery seed extract preparations are usually made with an organic solvent rather than by brewing.

Traditional herbal use of celery for the circulation, gout and blood pressure has always centred on the seed rather than the stalk, and the phthalides are the most likely reason. Traditional practice reached the concentrated fraction long before anyone could name what was in it.

Back to Table of Contents

3-n-Butylphthalide: From Celery Seed to a Stroke Drug

This is the strongest pharmacological story celery has, and it is worth telling properly because it is usually either ignored or wildly overstated.

3-n-butylphthalide — NBP, or in its synthetic racemic form dl-3-n-butylphthalide — was originally isolated from celery seed. Chinese pharmacological research pursued it from the 1970s onwards, and the synthetic version was developed into a licensed medicine for acute ischaemic stroke, approved for use in China in the early 2000s. It is a genuine, regulated, prescription drug, given by capsule and by infusion, and it has been through randomised controlled trials — including a ninety-day double-blind trial reported by Cui Liying and colleagues in the Chinese Medical Journal in 2013.

The mechanistic work suggests NBP acts on several fronts at once in ischaemic tissue: improving microcirculation and collateral blood flow, protecting mitochondria, reducing oxidative injury and limiting inflammatory damage. Ibrahim Abdoulaye and Yi-Jing Guo’s review surveys that literature, and the drug has since been studied in vascular cognitive impairment and other neurological settings.

Two honest qualifications belong beside that:

1. The evidence base is geographically narrow. Most trials have been conducted in China, and NBP is not approved in the United States, the European Union or the United Kingdom. Independent replication in other populations is limited.
2. The drug is not the vegetable. NBP therapy uses a synthesised, purified compound at doses on the order of several hundred milligrams a day, given to people in the acute phase of a stroke under medical supervision. Nobody has shown that eating celery delivers a neuroprotective dose of anything, and the arithmetic in the next section explains why that is unlikely.

What the NBP story genuinely demonstrates is something more modest and more interesting: that celery’s traditional reputation for the circulation was pointing at a real compound. Traditional use is often a decent hypothesis generator, and here it generated a drug. That is not the same as the salad being medicine.

Back to Table of Contents

The Dose Gap: Micromolar Versus a Stalk

This section is the reason the page exists, because almost nothing written about celery’s compounds addresses it.

Laboratory experiments on apigenin and luteolin typically expose cells to concentrations in the micromolar range for hours or days. Animal studies typically give purified compound by gavage or injection at doses that, scaled by body weight, correspond to far more than any plausible dietary intake. NBP trials use hundreds of milligrams of a synthesised drug per day.

Now consider what a serving of celery delivers. A cup of chopped celery is around a hundred grams of a vegetable that is roughly ninety-five percent water. The flavones in it are present at the milligram-per-hundred-grams scale, mostly as glycosides that must be hydrolysed before absorption, and free apigenin is absorbed poorly and cleared quickly. Peak blood concentrations after a normal food serving land far below the levels at which the interesting cell-culture effects appear.

None of this means the compounds are useless in food. It means three specific things:

• A cell-culture result is a hypothesis, not a health claim. “Apigenin kills cancer cells” is true in a dish and does not describe what happens when you eat celery.
• The realistic benefit of dietary flavones is cumulative and pattern-level — a lifetime of varied plant food associated with lower disease risk in population studies — not acute and not attributable to any single compound.
• Concentrated supplements are a different question with different risks. A capsule can reach concentrations food cannot, which is precisely why it should be treated as a drug rather than a vegetable.

The gap is not a reason for cynicism. It is a reason to describe celery accurately: a pleasant, hydrating, fibre-carrying, nitrate-rich, flavone-containing vegetable that fits well inside a good diet, and not a delivery vehicle for pharmacology.

Back to Table of Contents

What Cooking, Storage and Juicing Do

Different compounds, different fates. Practically:

• Flavones are reasonably heat-stable but water-soluble. Boiling celery and pouring away the water sends a fair share of the apigenin and luteolin glycosides down the sink. Soups, stews, stocks and braises keep them, because you eat the liquid. This is a general rule for water-soluble plant compounds and it applies to celery’s nitrate and potassium too.
• Phthalides are volatile and are driven off by long cooking. Add celery early for a mellow background note; add chopped leaves or a pinch of celery seed near the end if you want the aroma to survive.
• Juicing extracts the water-soluble fraction and discards the pulp. Flavone glycosides, nitrate, potassium and vitamin C come through into the juice. The fibre does not. Blending keeps everything.
• Storage raises furanocoumarins. Old celery, and especially celery with fungal spoilage, accumulates far more of the defensive furanocoumarins — documented as far back as 1985 and confirmed for fungal infection in the 1990s. Another reason to use celery while it is crisp.
• Freezing is fine for cooking, useless for crunch. Frozen celery collapses on thawing but works perfectly well in stocks and soups. Chopped celery leaves freeze well.

Back to Table of Contents

Celery Seed, Celery Salt and Celery Seed Extract

Three products, three different things, frequently confused on the shelf.

• Celery seed is the whole spice — tiny brown seeds with a concentrated celery aroma, used in pickling, coleslaw, potato salad, rubs and Bloody Marys. It is the phthalide-rich part of the plant, used in the amounts you would use any spice. Perfectly good cooking, negligible as a dose of anything.
• Celery salt is ground celery seed mixed with salt. It is mostly salt. If you are watching sodium, treat it as salt with a flavour, because that is what it is.
• Celery seed extract is a concentrated preparation standardised to an active fraction — usually the phthalides — sold in capsules for blood pressure and gout. This is the form used in the clinical trials, and it is pharmacologically active enough that it can add to the effect of blood-pressure medication. It is also a concentrated source of celery allergen, which matters for anyone with a celery allergy, and it is not appropriate in pregnancy, where celery seed preparations have a traditional reputation as a uterine stimulant.

The rule of thumb is simple: the further you move from the vegetable towards the extract, the more likely the product is to do something — and the more it needs to be treated with the caution you would give a drug.

Back to Table of Contents

What It Is Reasonable to Conclude

Putting the whole picture together:

• Celery genuinely contains apigenin, luteolin and phthalides, and those compounds genuinely have interesting, well-documented biological activity in the laboratory.
• One phthalide became a real medicine, which is a strong vindication of the traditional interest in celery seed — and a story about a purified drug, not about salad.
• The amounts in a normal serving are far below the concentrations that produce the striking laboratory effects, so celery should be described as a good vegetable rather than as a treatment.
• The best case for eating celery is the ordinary one: it is hydrating, filling for very few calories, high in nitrate and potassium, pleasant raw and useful cooked, and it makes vegetables easier to eat in quantity. That case does not need any of the pharmacology to stand up.

Back to Table of Contents

Practical Ways to Get More of Them

• Use the leaves. Do not buy trimmed, leafless celery if you can avoid it. The tops are the richest part and they are usually free.
• Choose the darker outer stalks for cooking and keep the pale heart for snacking. Green means more of the light-driven chemistry.
• Put celery in things you eat the liquid of — soups, stews, braises, stocks, sauces. The classic mirepoix of onion, carrot and celery is not just tradition; it is an efficient way to keep the water-soluble fraction.
• Blend instead of juicing if you want a drink. You keep the fibre and lose nothing else.
• Eat the whole flavone family, not just celery. Parsley, chamomile tea, thyme, oregano, peppers and artichoke all contribute apigenin and luteolin. Variety beats volume of any one food.
• Buy it crisp and use it. Fresh celery has better flavour, more phthalides and fewer furanocoumarins than tired celery.
• Cook it properly at least sometimes. Braised celery hearts, celery in a slow-cooked stew, or roasted celeriac are genuinely good dishes, and cooked celery is far easier to eat in quantity than raw sticks.

Back to Table of Contents

Key Research Papers

Author names, titles and journals are plain text; only the DOI is a link, and each opens in a new tab. Most of the flavone literature below is cell-culture and animal work, and is described that way in the article rather than dressed up as clinical evidence.

1. Shukla S, Gupta S. Apigenin: a promising molecule for cancer prevention. Pharmaceutical Research. 2010;27(6):962-978. — doi:10.1007/s11095-010-0089-7 — the standard survey of apigenin’s preclinical anticancer literature.
2. Salehi B, Venditti A, Sharifi-Rad M, et al. The therapeutic potential of apigenin. International Journal of Molecular Sciences. 2019;20(6):1305. — doi:10.3390/ijms20061305
3. Allemailem KS, Almatroudi A, Alharbi HOA, et al. Apigenin: a bioflavonoid with a promising role in disease prevention and treatment. Biomedicines. 2024;12(6):1353. — doi:10.3390/biomedicines12061353
4. Lopez-Lazaro M. Distribution and biological activities of the flavonoid luteolin. Mini-Reviews in Medicinal Chemistry. 2009;9(1):31-59. — doi:10.2174/138955709787001712
5. Kooti W, Daraei N. A review of the antioxidant activity of celery (Apium graveolens L). Journal of Evidence-Based Complementary & Alternative Medicine. 2017;22(4):1029-1034. — doi:10.1177/2156587217717415
6. Zhang Q, Zhou M, Chen P, et al. Optimization of ultrasonic-assisted enzymatic hydrolysis for the extraction of luteolin and apigenin from celery. Journal of Food Science. 2011;76(5):C680-C685. — doi:10.1111/j.1750-3841.2011.02174.x — illustrates that the flavones must be freed from their sugars before they can even be measured.
7. Han D, Row KH. Determination of luteolin and apigenin in celery using ultrasonic-assisted extraction based on aqueous solution of ionic liquid coupled with HPLC quantification. Journal of the Science of Food and Agriculture. 2011;91(15):2888-2892. — doi:10.1002/jsfa.4553
8. Hang NT, Thi Tu Uyen T, Van Phuong N. Green extraction of apigenin and luteolin from celery seed using deep eutectic solvent. Journal of Pharmaceutical and Biomedical Analysis. 2022;207:114406. — doi:10.1016/j.jpba.2021.114406 — on the flavone content of celery seed specifically.
9. Abdoulaye IA, Guo YJ. A review of recent advances in neuroprotective potential of 3-N-butylphthalide and its derivatives. BioMed Research International. 2016;2016:5012341. — doi:10.1155/2016/5012341
10. Cui LY, Zhu YC, Gao S, et al. Ninety-day administration of dl-3-n-butylphthalide for acute ischemic stroke: a randomized, double-blind trial. Chinese Medical Journal. 2013;126(18):3405-3410. — doi:10.3760/cma.j.issn.0366-6999.20123240
11. Tan TYC, Lim XY, Norahmad NA, et al. Neurological applications of celery (Apium graveolens): a scoping review. Molecules. 2023;28(15):5824. — doi:10.3390/molecules28155824 — maps how much of the celery neuroscience remains preclinical.
12. Kokotkiewicz A, Luczkiewicz M. Celery (Apium graveolens var. dulce (Mill.) Pers.) oils. In: Essential Oils in Food Preservation, Flavor and Safety. 2016:325-338. — doi:10.1016/b978-0-12-416641-7.00037-7 — the phthalide composition of celery essential oil.
13. Tashakori-Sabzevar F, Razavi BM, Imenshahidi M, et al. Evaluation of mechanism for antihypertensive and vasorelaxant effects of hexanic and hydroalcoholic extracts of celery seed in normotensive and hypertensive rats. Revista Brasileira de Farmacognosia. 2016;26(5):619-626. — doi:10.1016/j.bjp.2016.05.012
14. Chaudhary SK, Ceska O, Warrington PJ, Ashwood-Smith MJ. Increased furocoumarin content of celery during storage. Journal of Agricultural and Food Chemistry. 1985;33(6):1153-1157. — doi:10.1021/jf00066a032
15. Shayani Rad M, Moohebati M, Mohajeri SA. Effect of celery (Apium graveolens) seed extract on hypertension: a randomized, triple-blind, placebo-controlled, cross-over clinical trial. Phytotherapy Research. 2022;36(7):2889-2907. — doi:10.1002/ptr.7469

Live PubMed Searches

1. PubMed: apigenin bioavailability in humans
2. PubMed: luteolin and neuroinflammation
3. PubMed: 3-n-butylphthalide in ischaemic stroke
4. PubMed: celery phthalides and aroma composition
5. PubMed: apiin, apiose and celery flavone glycosides

Back to Table of Contents

Connections

• All Food
• Celery
• Celery Benefits Deep Dive
• Celery: History and Origins
• Blood Pressure and Dietary Nitrate
• Celery Allergy and Photosensitivity
• Apigenin
• Luteolin
• Antioxidants
• Parsley — a far richer apigenin source than celery
• Fennel
• Carrots
• Onions
• Broccoli
• Celery Juice

Back to Table of Contents

Featured Videos