Toxic compounds in microgreens: what recent studies measured in borage, sorghum and buckwheat

This is the third part of the series on microgreens that can be toxic, after the overview and the compound-by-compound article. Here we look at three species where recent analytical work has produced numbers worth knowing about — and where the gap between “measured in a laboratory” and “harmed someone” needs stating plainly.

One thing to be clear about up front: no human poisoning case has been reported from eating borage, sorghum or buckwheat as microgreens. What the studies below report are measured concentrations of particular compounds, and — for sorghum and buckwheat — documented poisonings in grazing livestock. The concern is what has been measured, not what has been observed in people. Where no measurement exists for the microgreen stage, we say so.

Borage: the one with actual microgreen measurements

Borage (Borago officinalis) is known for its edible flowers and its traditional use in teas, and it has become popular as a microgreen for its cucumber-like flavour. It is also the only species in this article for which someone has directly measured the compound of concern at the microgreen stage.

A 2024 analysis in Food Control profiled pyrrolizidine alkaloids — hepatotoxic compounds — across borage leaves, flowers and microgreens. What it found:

• the microgreens carried higher pyrrolizidine alkaloid concentrations than the mature leaves
• in the samples analysed, levels were high relative to the maximum levels the EU sets for comparable herb products under Regulation (EU) 2023/915
• the alkaloid profile — which specific compounds are present — differs between plant parts, which matters because they are not equally toxic

What pyrrolizidine alkaloids do

They are liver toxins. At sustained exposure they are associated with liver lesions, fibrosis and chronic inflammation, and with sinusoidal obstruction syndrome; EFSA also treats several of them as genotoxic carcinogens, which is why its assessment is framed as a margin of exposure rather than a simple safe dose.

That framing matters for how you read this. It does not mean a borage garnish will injure you — it means there is no threshold below which regulators are willing to call repeated exposure risk-free, and the sensible response to a food that concentrates them is to keep intake occasional rather than habitual.

What this means for growers

Borage had been treated as a distinctive, strongly flavoured microgreen with no particular caveats. On the current evidence the reasonable position is to avoid growing it for regular consumption until there is more work on safe levels and on whether cultivation practice can reduce the alkaloid load. This is a change of advice driven by new measurements, not by a poisoning.

Sorghum: cyanogenic glycosides, and a gap in the evidence

Sorghum, widely grown as forage, produces a cyanogenic glycoside called dhurrin. When tissue is damaged — cut, crushed, or stressed by drought — dhurrin meets the enzymes that break it down and hydrogen cyanide is released. Dhurrin content is at its highest in young seedlings and declines as the plant matures, which is exactly the wrong shape for a crop you harvest at seedling stage and then cut up.

Where the evidence actually comes from

Cyanide poisoning produces headache, dizziness, breathing difficulty, weakness, confusion and — in severe cases — respiratory arrest. Those effects are documented in grazing livestock that eat young sorghum in quantity, and that is where the evidence base sits.

No study has measured dhurrin in sorghum grown as a culinary microgreen, and no human case has been reported. We are not going to convert livestock data into a warning about a human portion size, because that is not what the data supports. What it does support is this: the compound is present, it peaks at exactly the stage in question, nobody has measured the culinary product, and therefore no safe portion can be stated. Leaving sorghum off the microgreen list until someone measures it is a decision about uncertainty, not about a known dose.

Buckwheat: fagopyrin, and a gradient that is usually reported backwards

Buckwheat contains fagopyrin, a photosensitising compound. Accumulated in the body it causes fagopyrism: severe sunburn-like reactions, rashes, pain and inflammation on sun-exposed skin, and marked UV hypersensitivity. It is rarely fatal, but the skin effects can be significant.

What is known about fagopyrism

Two things are commonly stated about buckwheat that the measurements do not support. The first is “younger means more” — in fact fagopyrin is highest in the flowers, with the seedling stage lower; within a sprout it concentrates in the cotyledons. The second is the implication that eating buckwheat microgreens is a documented cause of fagopyrism in people. The clear cases are in animals fed buckwheat in quantity; human reports involve concentrated extracts or sustained heavy consumption of leaves.

The practical upshot is moderation rather than exclusion — and rather more caution than average if you are already photosensitive or taking a medication that makes you so.

Why young plants often carry more of this chemistry

A seedling has no bark, little lignin and no robust physical defences. What it has instead is chemistry, and it invests in it heavily at exactly the stage when losing the growing point would be fatal. Across the species covered in this series that means alkaloids (nightshades, borage), cyanogenic glycosides (sorghum), lectins (Phaseolus beans), oxalates (rhubarb) and photosensitisers (buckwheat).

In the mature plant many of these fall away or relocate to parts nobody eats. In a microgreen they sit in the young leaves and stems — which is the entire harvest. That is the structural reason the microgreen stage deserves its own safety assessment rather than inheriting the mature plant’s reputation.

It is worth adding the other half of that sentence, though: it is not a general rule that seedlings are more toxic. For most cultivated microgreen species the same defensive chemistry is either absent or harmless, which is why the list of exclusions is short and the list of good options is long.

In conclusion

Not all microgreens are safe, even when the mature plant is, and the toxicity is not accidental — it is what the seedling is doing to stay alive. On current evidence the species to leave alone or restrict are:

• Borage — measured, and the measurements are the reason
• Sorghum — unmeasured at this stage, which is itself the reason
• Buckwheat — fine in normal culinary quantities; moderation rather than avoidance
• Nightshades, including pepper
• Phaseolus beans eaten raw
• Rhubarb
• Ornamental species, always

Where we have written “no case has been reported”, that is the finding — not a gap we filled with a warning to be safe. Being straight about the difference between a measurement, an animal study and a human case is what makes the rest of the advice on this site worth following.

References

• Sattler, M., Huch, M., Bunzel, D., Soukup, S.T., & Kulling, S.E. (2024). “Pyrrolizidine alkaloid contents and profiles in Borago officinalis leaves, flowers and microgreens: Implications for safety”. Food Control, 168, 110930. Open the study ↗
• EFSA Panel on Contaminants in the Food Chain (CONTAM) (2017). “Risks for human health related to the presence of pyrrolizidine alkaloids in honey, tea, herbal infusions and food supplements”. EFSA Journal, 15(7), 4908. Open the study ↗
• European Commission (2023). “Commission Regulation (EU) 2023/915 of 25 April 2023 on maximum levels for certain contaminants in food, Section 2.4 Pyrrolizidine alkaloids, entry 2.4.1”. Official Journal of the European Union (EUR-Lex). Open the regulation ↗
• Wang, B., Xiong, W., & Guo, Y. (2024). “Dhurrin in Sorghum: Biosynthesis, Regulation, Biological Function and Challenges for Animal Production”. Plants, 13, 2291. Framed around livestock forage, not human food. Open the study ↗
• EFSA Panel on Contaminants in the Food Chain (CONTAM) (2019). “Evaluation of the health risks related to the presence of cyanogenic glycosides in foods other than raw apricot kernels”. EFSA Journal, 17(4), 5662. Open the study ↗
• Kreft, S., Janeš, D., & Kreft, I. (2013). “The content of fagopyrin and polyphenols in common and tartary buckwheat sprouts”. Acta Pharmaceutica, 63(4), 553–560. Open the study ↗
• Tavčar Benković, E., & Kreft, S. (2015). “Fagopyrins and Protofagopyrins: Detection, Analysis, and Potential Phototoxicity in Buckwheat”. Journal of Agricultural and Food Chemistry, 63(24), 5715–5724. Paywalled; abstract free. Open the study ↗
• EFSA Panel on Contaminants in the Food Chain (CONTAM) (2020). “Risk assessment of glycoalkaloids in feed and food, in particular in potatoes and potato-derived products”. EFSA Journal, 18(8), e06222. Open the study ↗