Why mountain ash berries should be cooked: parasorbic acid explained

In October, the orange clusters of mountain ash trees are a feast for birds, and they catch foragers' eyes too. In Quebec, you will come across American mountain ash (Sorbus americana), showy mountain ash (Sorbus decora) and European mountain ash, or rowan (Sorbus aucuparia), a European tree introduced to North America as an ornamental. Everywhere, the same advice comes up: these berries are eaten cooked, never raw, because of parasorbic acid.

To identify these trees, pick at the right time and prepare your harvest, start with our guide to mountain ash and highbush cranberry, the wild fruit to forage after the frost and always eat cooked. Here, we focus on the molecule itself: what it is, what it does, what cooking and frost change, and above all what is proven and what is not. Same approach as for tomatine in green tomatoes: no panic, no promises.

Parasorbic acid: a defence molecule hidden in a sugar-bound form

Its chemical name, according to the International Agency for Research on Cancer (IARC), is (S)-5,6-dihydro-6-methyl-2H-pyran-2-one. What matters is that it is a lactone, a small molecule closed in a ring, and a cousin of sorbic acid. You know sorbic acid without realizing it: it is food additive E 200, a preservative that the European Food Safety Authority (EFSA) re-evaluated in 2015 without finding any genotoxic activity. According to a 2020 scientific review on Sorbus species, heat converts parasorbic acid into sorbic acid. We will come back to this below, because it is what cooking is all about.

In the fruit, the molecule hides mostly in a form bound to a sugar, called parasorboside. According to 1971 work cited in the reference study on the subject, published in 1987 by Oster's team, parasorbic acid forms during extraction, when the glucose breaks off, and parasorboside itself has little or no antibiotic activity. This sugar-bound precursor has another feature: according to a monograph on European mountain ash published in 2000 in the Journal of Ecology, it is this compound, not the polyphenols, that makes the berries bitter. The dry, rough feeling in your mouth comes instead from tannins, the 2020 review adds.

For the tree, the molecule has at least one known effect: it slows germination. In the lab, berry extracts stop other seeds from germinating, and parasorbic acid is the most important of the three inhibitors Oster's team found in them.

How much is in the berries? A single documented measurement, from 1987

The figure repeated in recent reviews comes from that 1987 study: 4 to 7 mg of parasorbic acid per gram of fresh fruit, or 0.4 to 0.7% of its weight, and 0.08 to 0.12 mg per gram in the seeds, dozens of times less. The 2000 monograph and the 2020 review cite it as is.

But you have to read the method. The researchers froze ripe berries at −18°C right after picking, ground them, then boiled them in water for 24 hours in a reflux setup, which sends condensed steam back into the vessel, before analyzing the liquid. Since parasorbic acid forms precisely during extraction, this figure mostly measures what a long boil can release, not necessarily what a raw berry picked from the tree contains. A 1998 Bavarian publication gives a much lower figure, 0.02 to 0.2% of the pulp (0.04% on average), without saying where it comes from or how it was obtained. We found no documented measurement of the free acid in fresh berries.

Two more limits. These data concern European mountain ash: we found no measurements for American or showy mountain ash. And none of the studies consulted compares the content of a jelly, syrup or compote with that of the starting fruit.

Eating raw berries: what poison control centres say

Non-toxic or slightly toxic, depending on the source

Good news: none of the organizations consulted ranks mountain ash berries among dangerous plants. In its 2017 brochure, the Quebec Poison Control Centre places mountain ash (Sorbus americana) on its list of non-toxic plants, while pointing out that a non-toxic plant is not necessarily edible. In Germany, the Federal Institute for Risk Assessment (BfR) cites European mountain ash berries as an example of a non-toxic plant: only a large quantity can cause digestive upset, usually mild.

Call data point the same way. From 2010 to 2019, the joint poison information centre in Erfurt, which serves four states in eastern Germany and covers five others at night, received 248 calls about European mountain ash. In 199 cases there were no symptoms; 30 people had mild symptoms and 8 had moderate symptoms. No severe cases or deaths were reported, and the outcome was unknown or could not be assessed for the other 11. The study places mountain ash in its category 1, "slightly toxic," and names the seeds and raw berries as the parts involved. Across all the fruiting plants studied, nearly 70% of calls involved children aged 1 to 5. So the sources do not fully agree, between "non-toxic" and "slightly toxic," but none of them describes a serious danger.

As for the mechanism, the 2020 review describes parasorbic acid as irritating to the stomach lining and able to cause indigestion in large amounts. It adds that it can damage the kidneys, but without citing any study, and we found no documented case. Above all, none of the sources consulted gives a safe quantity of raw berries, or a threshold at which symptoms start.

If a child has eaten mountain ash berries, or if you are not sure about the tree, call the Quebec Poison Control Centre at 1 800 463-5060, open 24 hours a day, even if there are no symptoms. Keep a branch with its leaves and berries, or take a photo, to help identify it. And watch out for look-alikes: several small red berries that grow in the same places are truly toxic. Our article on the red fall berries not to mix up goes over them.

What about cancer in rats?

You sometimes read that parasorbic acid causes cancer. Here is what IARC says, in a 1976 evaluation: injected under the skin of rats, it caused sarcomas, a type of tumour, at the injection site. Feeding studies in rats did not allow any conclusion: in those where it was given together with sorbic acid, the dose was too low and there was no control group followed in parallel; in another, too few animals survived. IARC had no human data. In 1987, IARC placed it in Group 3: not classifiable as to its carcinogenicity to humans.

Injections under the skin do not reproduce what happens when you eat a spoonful of jelly, but these old data do not prove safety either. It is a reason for caution, not a cause for alarm.

Cooking: the right precaution, even if its effect has never been measured

The usual reasoning is simple: heat turns parasorbic acid into harmless sorbic acid. That is what the 2020 review says, without citing any source or measurement.

The only available measurement actually calls for nuance. In the 1987 study, 24 hours of boiling did not make parasorbic acid disappear, since that is precisely the treatment used to measure it. The conditions were not those of a kitchen, with a setup that held in the steam and a full day of boiling, but we found no study measuring what remains in a homemade jelly, syrup or compote. A 1998 Bavarian collection on mountain ash even notes that in jams, picking after frost and adding plenty of sugar soften the effect of parasorbic acid, but a clear loss of quality remains. The known chemistry does not look like a kitchen either. In the lab, parasorbic acid can be turned into sorbic acid: in the 19th century by heating it with potash or hydrochloric acid, and today with an acid catalyst, in an organic solvent, at about 80 to 140°C (175 to 285°F) and under pressure. In those experiments, water slows the reaction down. The molecule can also leave with the steam: it is in fact obtained by steam-distilling the acidified berry juice. What happens in your pot, conversion or evaporation, no one has measured, as far as we know.

So why cook? For three good reasons, which hold up even without that measurement.

• It is the precaution the sources point to. The Erfurt study targets raw berries and seeds, not cooked preparations, and the 2020 review presents heat as what makes parasorbic acid harmless.
• Straining removes the seeds. The Erfurt study lists them among the parts involved, and the 2020 review notes that they contain prunasin, a compound that can release cyanide. A well-strained jelly or syrup leaves them in the bag.
• The flavour improves enormously. Raw, these berries are bitter and astringent. Cooked with crabapples, then strained and sweetened, as in our mountain ash and crabapple jelly, they make a tangy jelly that goes well with game. The principle is the same as for our rose hip and crabapple jelly.

In practice: cook the berries until they are very tender, strain, and do not serve them raw, whether as a snack, in a smoothie or as cold-pressed juice. Pick away from roads, with the owner's permission, and leave some clusters for the birds. And when in doubt about the tree's identity, don't pick.

Frost and freezing: a belief without data

Another widespread tip: wait for the first frosts, or put the berries in the freezer, to "neutralize" parasorbic acid. The 2020 review says so too: freezing would convert it into sorbic acid, and berries picked after the first frost would be rid of it. But it reports no measurements before and after frost. The only reference it cites on this point, a 2014 Czech study of hybrid mountain ash, does not mention frost or freezing, and it did not measure parasorbic acid. We found no measurements either.

The only concrete clue points the other way. In the 1987 study, the berries had been frozen at −18°C and kept that way until analysis, and the extraction still drew 4 to 7 mg per gram from them. Freezing had therefore not removed what the analysis measures. Whether frost softens the taste is another question: in any case, nothing supports using it as a detox method.

The same caution applies to mountain ash trees selected for sweeter fruit, notably from a breeding program started in Russia in the early 20th century. The 2020 review claims that parasorbic acid is almost absent from them, again without figures to back it up. For a less bitter European variety of rowan, a 1998 Bavarian publication gives at most a quarter of the usual content, or 0.01%, without saying where the figure comes from. For the Russian hybrids, we found no measurements: cook them too.

Waiting for frost remains a forager's habit, and our guide to mountain ash and highbush cranberry explains what the sources say about it. But freezing your berries does not replace cooking them. To know when to look for the clusters, also check our foraging calendar.

In short: proven, likely or unknown?

So the advice to always cook mountain ash berries holds up, for more modest reasons than people think: caution with a poorly studied molecule, seeds you would rather strain out, and a flavour that only comes through with cooking. The exact dose and the precise effect of heat and cold still remain to be measured.

Sources

• Des plantes toxiques chez nous? C'est le bouquet! (brochure with a list of non-toxic plants; in French), Quebec Poison Control Centre (Centre antipoison du Québec), 2017.
• Risiko Pflanze – Einschätzung und Hinweise (plant risk: assessment and advice; in German), German Federal Institute for Risk Assessment (BfR), 2017.
• Exposures to fruit plants in Germany from 2010–2019 (article in German), S. Wendt, D. Prasa, C. Lübbert, K. Begemann and H. Franke, Bundesgesundheitsblatt, 2023.
• Parasorbic acid (IARC Summary & Evaluation, Volume 10), International Agency for Research on Cancer (IARC), 1976, classified in Group 3 in 1987.
• Natural Inhibitors of Germination and Growth, IV. Compounds from Fruits and Seeds of Mountain Ash (Sorbus aucuparia), U. Oster, I. Blos and W. Rüdiger, Zeitschrift für Naturforschung C, 1987.
• Pharmazie und Medizin, Volksmedizin (pharmacy, medicine and folk medicine; in German), Klaus Storm, Beiträge zur Vogelbeere (LWF report no. 17), Bavarian State Institute of Forestry (LWF), 1998.
• Die Vogelbeere – ein Obstbaum (mountain ash as a fruit tree; in German), Reinhard Eder, Beiträge zur Vogelbeere (LWF report no. 17), Bavarian State Institute of Forestry (LWF), 1998.
• Sorbus aucuparia L. (Biological Flora of the British Isles), O. Raspé, C. Findlay and A.-L. Jacquemart, Journal of Ecology, 2000.
• The Sorbus spp. – Underutilised Plants for Foods and Nutraceuticals: Review on Polyphenolic Phytochemicals and Antioxidant Potential, V. Sarv, P. R. Venskutonis and R. Bhat, Antioxidants, 2020.
• Bioactive compounds in sweet rowanberry fruits of interspecific Rowan crosses, J. Mlček, O. Rop, T. Juríková, J. Sochor, M. Fišera, Š. Balla, M. Baroň and J. Hrabě, Central European Journal of Biology, 2014.
• Scientific Opinion on the re-evaluation of sorbic acid (E 200), potassium sorbate (E 202) and calcium sorbate (E 203) as food additives, EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS), EFSA Journal, 2015.
• Production of 2,4-hexadienoic acid and 1,3-pentadiene from 6-methyl-5,6-dihydro-2-pyrone (US patent 8,404,890 B2), J. A. Dumesic and M. Chia, Wisconsin Alumni Research Foundation, 2013.
• Parasorbic acid (DB14121), DrugBank Online.