A fungus taught to make egg-white protein of its own

An egg protein written into a fungal chromosome

On 25 July 2026 the Journal of Agricultural and Food Chemistry published online a study in which Wenwen Pei and ten co-authors expressed chicken and duck ovalbumins in Fusarium venenatum, the filamentous fungus grown commercially as mycoprotein. The abstract calls the technique multisite chromosomal integration: copies of the inserted genes are placed at several positions in the fungus's own chromosomes and are copied along with them from then on. The print issue is dated 5 August 2026, volume 74, issue 30, pages 23662–23673. The work was done at the Tianjin Institute of Industrial Biotechnology of the Chinese Academy of Sciences and at Tianjin University of Science and Technology, and the free PubMed record 42503790 carries the abstract in full.

That last point is not incidental. The article itself is not open access, and its publisher page returns a 403 to automated clients. Only the abstract can be read publicly, every figure quoted below comes from it, and the Crossref record reproduces the same abstract word for word along with both publication dates.

Six numbers and what they are measured against

The engineered mycoprotein had 12.8% more total protein and 39.5% more essential amino acids than the wild-type strain. Sulphur-containing amino acids — the abstract names methionine and cysteine — increased threefold. Water and oil holding, emulsifying and gelling abilities were all described as enhanced, with hardness 3.7 to 5.1 times that of the wild type. Precision is worth keeping here: the abstract says hardness, and does not specify gel hardness. The modified strain reached 62.6% digestibility in 240 minutes, and its initial gastric digestion rate doubled.

Every one of those figures shares a single reference point, the wild-type Fusarium venenatum control run in the same experiments. None is a comparison with meat, with egg, or with anything sold in a shop. It is also worth noticing where the authors begin. Their own abstract says that low essential amino acid levels and poor functional properties limit this fungus as a food. The percentages are therefore an improvement on an acknowledged deficit, not a verdict that a complete protein has been created.

The awkward part, stated plainly

Ovalbumin is the main protein of hen egg white and one of the best-characterised food allergens there is. Putting it into mycoprotein adds that allergen rather than removing it. The abstract does not mention any allergenicity testing — and that is as far as anyone outside the paywall can go, because the full text cannot be read, so whether such tests appear in the article is simply unknown here.

Which leads to the misreading most likely to happen. Mycoprotein built its market position as protein without meat, so it is read reflexively as a vegan food. This strain cannot be called vegan, vegetarian or animal-free: the inserted sequences come from a chicken and a duck, and the organism is genetically modified. Nor is this the first allergy conversation about mycoprotein, which has a literature of its own that has nothing to do with the Tianjin strain. Ovalbumin here is a second, deliberately added and well-described allergen, not allergenicity arriving in a food that previously had none.

What the result does not cover

The 62.6% figure was measured in a four-hour gastric digestion assay. The abstract never uses the phrase in vitro, but a measurement of that kind is a bench result, not an observation in people. There is nothing on taste, aroma or consumer acceptance either; the reported gains are compositional, functional and textural. No production-scale yield, cost or environmental figure is given. The strain is not approved or sold in any jurisdiction — it is a genetically modified laboratory organism, and no mycoprotein manufacturer uses it.

Something to look at without a laboratory

The most accessible observation is an ingredients list in a supermarket. Mycoprotein rarely reaches a shelf on its own: egg white is a common binder mixed into the fungal mass after fermentation, while vegan versions of the same foods usually reach for potato protein instead, and the pack says which is which. The Tianjin paper describes something different: not an ingredient added at the end, but a fungus producing egg protein itself as it grows. The distance between growing a fungus and rebuilding one is the whole story here, and it is that distinction, rather than any percentage, that will decide how food regulators eventually treat a product of this kind.