The Isoflavones in Red Clover: Formononetin, Biochanin A and the Rest

Almost every article about red clover stops at the same sentence: “rich in isoflavones.” It is true, and it is nearly useless. Isoflavones are a family, not a single substance, and which members of that family a given batch of Trifolium pratense carries — in what form, in what ratio, at what concentration — is the difference between an extract worth formulating with and an extract worth nothing in particular.

This article is the chemistry underneath the marketing. What an isoflavone is, why these particular molecules are called phytoestrogens, which four dominate red clover and how they relate to one another structurally, what form they take inside the living plant, how much of them is actually there, what makes that number move, and how any of it gets measured. It is written for formulators, pharmacists and readers who want the primary literature rather than a summary of a summary.

What an isoflavone actually is

Isoflavones belong to the flavonoids, the enormous class of plant polyphenols responsible for a great deal of colour, bitterness and chemical defence in the plant kingdom. The flavonoid skeleton is three rings: two aromatic rings (A and B) joined by an oxygen-containing middle ring (C).

What separates an isoflavone from an ordinary flavone is simply where the B-ring is attached. In a flavone it hangs off position 2 of the C-ring; in an isoflavone it has migrated to position 3. That single positional shift is not a detail — it is the whole story. It rearranges the molecule into a shape that no longer resembles a typical plant pigment and starts to resemble something else entirely.

Isoflavones are also unevenly distributed across the plant kingdom. They are concentrated almost entirely in the Fabaceae — the legume family — which is why the conversation about dietary isoflavones is always a conversation about soy, chickpeas, lupin, alfalfa and clover, and almost never about anything else.

Why these count as phytoestrogens

Move the B-ring to position 3 and the isoflavone acquires roughly the same distance between its two terminal hydroxyl groups as 17β-estradiol has between its 3-hydroxyl and its 17-hydroxyl. Both molecules are also broadly planar and similarly sized. To a protein binding pocket that recognises shape and hydrogen-bonding geometry rather than reading a formula, an isoflavone is a passable imitation of the body’s principal estrogen.

“Passable” is the operative word, and the numbers make the point better than adjectives do. In the analysis of a clinical red clover extract by Booth and colleagues, competitive binding assays against recombinant human estrogen receptors gave IC₅₀ values for genistein of roughly 0.30 μM at ERα and 0.020 μM at ERβ, and for daidzein roughly 17 μM and 1.2 μM respectively. Biochanin A landed at approximately 35 μM and 4.1 μM. Formononetin was weakest of all — around 104 μM at ERα and 60 μM at ERβ — and was not estrogenic in the cell-based assay the same team ran alongside it.

Two things follow from that table, and both are frequently missed.

First, the preference for ERβ is consistent and often an order of magnitude or more. That receptor selectivity is why isoflavones are usually discussed as tissue-selective compounds rather than as blanket estrogen substitutes. Our primer Phytoestrogens 101 covers what that distinction does and does not imply.

Second — and this matters enormously for red clover specifically — the plant’s two dominant isoflavones are among its weakest direct binders. Formononetin barely engages the receptor at all in isolation. Its relevance to the research literature comes almost entirely from what happens to it after ingestion, which is the subject of our companion article, Red Clover vs Soy Isoflavones: What Actually Differs.

Red clover’s main four — and how they relate

Red clover contains a dozen or more isoflavones, but four account for the overwhelming majority: formononetin, biochanin A, daidzein and genistein.

They are not four unrelated molecules. They are two pairs:

• Formononetin is the 4′-O-methyl ether of daidzein. Same skeleton, with a methyl group capping the hydroxyl at the 4′ position of the B-ring.
• Biochanin A is the 4′-O-methyl ether of genistein. Same relationship, one ring-position apart in the parent compounds (genistein carries an extra 5-hydroxyl that daidzein lacks).

Strip the methyl groups and red clover’s chemistry collapses onto soy’s. Leave them on and you have a different plant with different behaviour. In red clover, the methylated forms are decisively dominant: at flowering, formononetin and biochanin A together made up around 91% of total isoflavones in the Lithuanian genotype survey below — roughly 51% and 40% respectively — leaving genistein and daidzein as minor components. Soy has the opposite profile, dominated by the unmethylated genistein and daidzein.

That inversion is the single most important fact about red clover phytochemistry, and it is the reason soy findings cannot simply be copied across.

Beyond the four, the same clinical extract analysed by Booth’s group contained measurable irilone (about 3.2% of the preformulated extract), plus pratensein, pseudobaptigenin, prunetin and calycosin at roughly 0.4–1.1% each. Irilone in particular is a red clover marker with no soy counterpart.

Glycosides and aglycones: ononin and sissotrin

A living red clover plant does not store its isoflavones in the free form described above. It stores them as glycosides — sugar-conjugated versions, typically 7-O-glucosides, frequently with a malonyl ester attached to the sugar. Glycosylation makes the molecules more water-soluble, less reactive and easier to park in the vacuole.

The two you need to know by name:

• Ononin = formononetin 7-O-glucoside
• Sissotrin = biochanin A 7-O-glucoside

The unconjugated forms are called aglycones. The distinction is not academic bookkeeping. Glycosides and aglycones differ in solubility, in extraction behaviour, in chromatographic retention and in how they are absorbed. And the ratio between them in any given sample is not fixed by the plant alone: it is heavily influenced by what happened after harvest. Tsao and colleagues, profiling thirteen cultivars, noted that samples held at −5 °C for a few days before freeze-drying arrived at the analytical laboratory containing mainly aglycones — plant enzymes had done the hydrolysis during handling.

The practical consequence: two certificates of analysis reporting “total isoflavones” may be measuring genuinely different things, depending on whether the laboratory hydrolysed the sample first and reported aglycone equivalents, or quantified the native glycosides as they were.

How much is actually in the plant

The cleanest published figures for whole-plant material at flowering come from Lemežienė and colleagues, who profiled red clover genotypes at the flowering stage and reported concentrations on a dry-matter basis.

Read the spread rather than the midpoint. Total isoflavone content varies by roughly 1.5-fold across genotypes grown in the same trial, in the same soil, harvested at the same stage. Genetics alone accounts for that much. Formononetin averaged around 3.4 mg/g across the genotypes surveyed and biochanin A around 2.7 mg/g, with daidzein an order of magnitude lower than either.

Note also how lopsided the ratios are. There is roughly thirty times more formononetin than daidzein in the plant. Anyone reasoning about red clover as though it were a modest source of daidzein and genistein is reasoning about the wrong molecules.

What moves the number

Four variables dominate.

Plant part

Isoflavones are not distributed evenly through the plant. In the thirteen-cultivar survey by Tsao and colleagues, leaves carried the highest overall concentration, with stem, petiole and flower following. Other surveys have ranked flowers above stems. The honest summary is that leaf tissue is consistently the richest fraction, the ranking of the remaining aerial parts differs between studies and cultivars, and roots and woody stem contribute least. Which is why “aerial parts” on a label tells you less than it appears to.

Harvest timing

Isoflavone composition differs measurably between the early bud stage and late flowering, in both the total and the balance between individual compounds. Harvest date is a formulation decision, not a farming detail.

Cultivar and genetics

See the table. Same field, same day, different genotype, up to 1.5× difference.

Drying, storage and processing

Post-harvest handling determines the glycoside-to-aglycone ratio, as described above, and aggressive drying can degrade the compounds outright. This is the stage where the most value is quietly lost, and the stage least often disclosed.

Preparation changes the profile

Extraction is not a neutral act of retrieval. It selects.

Malca-Garcia and colleagues compared traditional red clover preparations directly — aqueous infusions, decoctions and 45% ethanolic tinctures — using quantitative NMR, LC-MS/MS and UHPLC-UV, with eight isoflavone aglycones plus ononin and sissotrin as markers. The three preparation types produced clearly different chemical profiles. Both the infusion and the decoction contained higher concentrations of the glucosides ononin and sissotrin than the 45% ethanolic tinctures did — water pulls the sugar-conjugated forms preferentially, as solubility would predict.

The same study found something less expected. The tincture was not chemically static. Followed over a month, biochanin A and formononetin concentrations shifted, peaking at around six days — the authors describe this as “dynamic residual complexity.” A tincture, in other words, has a composition that depends on its age.

For anyone specifying a raw material: solvent system, solvent ratio and time in contact are not process trivia. They determine which molecules you end up with.

What else is in there: clovamide and the rest

Isoflavones dominate the literature, but they are not the only thing in the plant, and in leaf tissue they are not even the only major phenolic.

Clovamide — a caffeic acid–DOPA conjugate — was measured by Tava and colleagues at 15.6 ± 0.6 mg/g dry weight in the leaves of red clover grown in Italy, alongside 24.6 mg/g of isoflavones and 13.2 mg/g of flavonols, within a total phenolic content of 53.7 mg/g. That is a substantial fraction of the leaf’s phenolic load, and it is invisible on any certificate of analysis that reports isoflavones only.

Also present, at varying levels:

• Flavonols — predominantly quercetin derivatives, as glycosides and malonylated glycosides.
• Coumestans — including coumestrol and medicagol. Coumestrol is worth flagging: in the Booth analysis it was the most potent estrogen-receptor binder of any compound identified, with IC₅₀ values around 0.06 μM at ERα and 0.02 μM at ERβ — but it was present at ≤0.01% of the extract. Potency and abundance are different axes.
• Pterocarpans — maackiain and its glucoside trifolirhizin. In the clinical extract profiled by Booth’s group, pterocarpans made up 0.06% of the material.
• Coumarins — present at ≤0.03% in that same extract.
• Saponins — triterpene glycosides, reviewed alongside the phenolics in Kołodziejczyk-Czepas’s survey of the genus.

For orientation, the full breakdown of that phase II clinical extract was 35.54% isoflavones, 1.11% flavonoids, 0.06% pterocarpans, ≤0.03% coumarins and ≤0.03% tyramine. Even a well-characterised standardised extract is roughly two-thirds material that nobody put on the label.

How it is measured

Three analytical approaches account for most of the published numbers.

HPLC-UV remains the workhorse for routine quantification. Isoflavones absorb usefully in the ultraviolet, standards are commercially available, and the instrumentation is in every quality-control laboratory. Lee and colleagues published a validated, optimised HPLC-UV method specifically for quantifying formononetin and biochanin A in Trifolium pratense extract — a sensible starting point for anyone writing a specification.

LC-MS/MS adds mass-selective detection, which is what you need when compounds co-elute, when you are quantifying minor components such as irilone or coumestrol against a large formononetin peak, or when you are working in a complex matrix.

Quantitative NMR (qHNMR) is the outlier and the most useful for reference material. Because the NMR signal is directly proportional to the number of nuclei, qNMR can quantify a compound without a matched reference standard for that exact compound — valuable for minor constituents where no standard is purchasable. Malca-Garcia’s team used it alongside chromatography for precisely this reason.

One caveat that undermines more cross-study comparisons than any other: if the sample was hydrolysed before analysis, the result is aglycone equivalents; if it was not, the result is native glycosides. These are different quantities. Always check which was reported.

Why standardisation matters

All of the above converges on one commercial reality: red clover products vary, substantially, and the label often will not tell you.

Wang and colleagues analysed a set of marketed red clover products and reported that the labelling was ambiguous and the isoflavone content varied significantly between products. They then ran those extracts through a Caco-2 intestinal cell model and found that absorption rates and permeabilities for biochanin A and formononetin differed significantly too — and, unexpectedly, that the surrounding product matrix itself affected the result, not merely the isoflavone concentration. The authors framed their conclusion as scientific support for standardisation, so that buyers can make informed choices. Worth keeping in proportion: that was a cell-culture model of intestinal transport, not a human study, and it says nothing about clinical outcome.

What it does say is that “contains red clover extract” is not a specification. A specification names the marker compounds, the assay, the plant part, the extraction solvent and the acceptance range — and reports whether the figure is aglycone equivalents or native glycosides.

For context on the regulatory side: red clover entries such as Trifolium Pratense Flower Extract appear in the EU’s CosIng cosmetic ingredient database. CosIng is an inventory of ingredients used in cosmetics with their assigned functions — it is a catalogue, not an approval list, and inclusion is not a statement about efficacy.

Minerva108 produces its red clover extract in-house rather than buying a commodity extract, precisely because extraction and post-harvest handling are where an isoflavone profile is preserved or lost. You can read about the plant and our sourcing on our red clover ingredient page, and for what the skin and hair research actually shows, see Red Clover: The Anatolian Flower Science Keeps Coming Back To.

Frequently asked questions

What are the main isoflavones in red clover?

Four dominate: formononetin, biochanin A, daidzein and genistein. Formononetin and biochanin A are far more abundant — in a survey of red clover at flowering they made up roughly 51% and 40% of total isoflavones respectively, with genistein and daidzein as minor components. Red clover also contains irilone, pratensein, pseudobaptigenin, prunetin and calycosin in smaller amounts.

What is the difference between formononetin and daidzein?

One methyl group. Formononetin is the 4′-O-methyl ether of daidzein — the same molecular skeleton with a methyl cap on the hydroxyl at the 4′ position. Biochanin A stands in the same relationship to genistein. In binding assays, the methylated forms are markedly weaker at the estrogen receptor than their unmethylated counterparts.

What are ononin and sissotrin?

They are the glucoside forms in which the plant actually stores its isoflavones. Ononin is formononetin 7-O-glucoside; sissotrin is biochanin A 7-O-glucoside. Aqueous infusions and decoctions have been shown to contain more of these glucosides than 45% ethanolic tinctures do.

How much isoflavone does red clover contain?

In one genotype survey at the flowering stage, total isoflavones ranged from 5.40 to 8.09 mg/g of dry matter, with formononetin at 2.61–4.40 mg/g and biochanin A at 1.79–3.32 mg/g. The figure moves with cultivar, plant part, harvest timing and post-harvest handling, so a single number for “red clover” is always an approximation.

Why do two red clover extracts give different lab results?

Several reasons compound. Different cultivars differ by up to about 1.5-fold at the same growth stage; leaf tissue is richer than other parts; infusions, decoctions and ethanolic tinctures extract different compounds preferentially; and results depend on whether the laboratory hydrolysed the sample and reported aglycone equivalents or quantified native glycosides. Marketed products have been shown to vary significantly in measured isoflavone content.

This article is for general information and education only. It is not medical advice, and nothing here is intended to diagnose, treat, cure or prevent any disease. If you are pregnant, breastfeeding, taking medication or managing a health condition, speak with a healthcare professional before using red clover supplements.

References

1. Lemežienė N, Padarauskas A, Butkutė B, Cesevičienė J, Taujenis L, Norkevičienė E. The concentration of isoflavones in red clover (Trifolium pratense L.) at flowering stage. Zemdirbyste-Agriculture, 2015;102(4):443–448. doi.org/10.13080/z-a.2015.102.057
2. Booth NL, Overk CR, Yao P, et al. The chemical and biologic profile of a red clover (Trifolium pratense L.) phase II clinical extract. Journal of Alternative and Complementary Medicine, 2006;12(2):133–139. doi.org/10.1089/acm.2006.12.133
3. Malca-Garcia GR, Zagal D, Graham J, et al. Dynamics of the isoflavone metabolome of traditional preparations of Trifolium pratense L. Journal of Ethnopharmacology, 2019;238:111865. doi.org/10.1016/j.jep.2019.111865
4. Tsao R, Papadopoulos Y, Yang R, Young JC, McRae K. Isoflavone profiles of red clovers and their distribution in different parts harvested at different growing stages. Journal of Agricultural and Food Chemistry, 2006;54(16):5797–5805. doi.org/10.1021/jf0614589
5. Tava A, Pecio Ł, Stochmal A, Pecetti L. Clovamide and flavonoids from leaves of Trifolium pratense and T. pratense subsp. nivale grown in Italy. Natural Product Communications, 2015;10(6):933–936. pubmed.ncbi.nlm.nih.gov/26197520
6. Lee JH, Paje LA, Kim JY, et al. Validation of an optimized HPLC–UV method for the quantification of formononetin and biochanin A in Trifolium pratense extract. Applied Biological Chemistry, 2021;64:57. doi.org/10.1186/s13765-021-00630-5
7. Wang SW, Chen Y, Joseph T, Hu M. Variable isoflavone content of red clover products affects intestinal disposition of biochanin A, formononetin, genistein, and daidzein. Journal of Alternative and Complementary Medicine, 2008;14(3):287–297. doi.org/10.1089/acm.2007.0617
8. Pfitscher A, Reiter E, Jungbauer A. Receptor binding and transactivation activities of red clover isoflavones and their metabolites. Journal of Steroid Biochemistry and Molecular Biology, 2008;112(1–3):87–94. doi.org/10.1016/j.jsbmb.2008.08.007
9. Kołodziejczyk-Czepas J. Trifolium species – the latest findings on chemical profile, ethnomedicinal use and pharmacological properties. Journal of Pharmacy and Pharmacology, 2016;68(7):845–861. doi.org/10.1111/jphp.12568