Knocking down PxECM33 peeled the outer layer off the fungal cell wall, exposing sugars that melon immunity recognises — and a spray cut powdery mildew severity by about three quarters.

A fungal cell wall is not only armour. It is also the surface a plant inspects, and the fragments that break off it are what plant immune receptors recognise. A successful pathogen therefore has to keep its wall intact for two reasons: to hold itself together, and to avoid being noticed.

A study in Horticulture Research on Podosphaera xanthii, the fungus behind cucurbit powdery mildew, identifies a protein that does both jobs at once — and shows what happens when it is taken away. The answer is that the fungus is not merely weakened, it is exposed.

The work was carried out by researchers at the University of Málaga, at the Instituto de Hortofruticultura Subtropical y Mediterránea “La Mayora”, a joint institute of the University of Málaga and the Spanish National Research Council, and at the University of León and its Institute of Molecular Biology, Genomics and Proteomics.

What ECM33 holds together

ECM33 is a glycosylphosphatidylinositol-anchored protein — tethered to the outer face of the cell membrane by a lipid anchor — and the study characterises it as a structural linker. It binds the wall’s major carbohydrates, connects the inner and outer layers, and limits the release of the fragments that trigger plant immunity.

The wall’s composition was measured first. Glucosamine, mannose and glucose accounted for about 40.9, 30.5 and 26.1 per cent of detected wall sugars respectively, and protein made up 52 per cent of wall dry weight. Purified PxECM33 bound mannans most strongly, followed by chitin and then β-glucan.

That binding preference maps onto the wall’s architecture. Mannoproteins sit in the outer layer while chitin and β-glucan form the inner structural mesh, so a protein that binds all three with a preference for mannans is positioned to staple the outside to the inside.

What happened when it was removed

Double-stranded RNA directed against PxECM33 lowered its transcript level by about half — a partial knockdown rather than a deletion, which makes the resulting phenotype more striking than it would otherwise be.

Fungal growth fell, and electron microscopy showed the outer wall layer separating from, or disappearing above, the inner layer. The staple had been removed and the laminate came apart.

The decisive experiment was the next one. Co-silencing PxECM33 together with CmCERK1 — the melon receptor that perceives chitin fragments — restored fungal growth.

This is the result that establishes mechanism rather than correlation, and the logic is worth spelling out. If losing ECM33 simply made the fungus structurally weaker, disabling the plant’s immune receptor would not help it. That removing the receptor rescues growth means most of the damage was not mechanical: it was the plant detecting fragments that the intact outer layer had been concealing. Strip the mannoprotein coat away, and plant chitinases and β-glucanases reach the inner wall, chew it, and release the oligomers that CmCERK1 is waiting for.

The fungus, in other words, was not killed by the loss of a wall protein. It was turned in to the immune system.

The field-facing number

The applied test used spray-induced gene silencing — double-stranded RNA applied to leaves rather than engineered into either organism — and reduced powdery mildew severity by about 75 per cent in growth-chamber and greenhouse trials.

Spray-induced silencing is an attractive idea for exactly this reason: nothing is genetically modified. The RNA is a topical treatment, the fungus takes it up and silences its own gene, and the plant is unaltered — which places the product in a regulatory category closer to a pesticide than to a transgenic crop in most jurisdictions.

Two properties make ECM33 a better target than most. The authors report it is highly conserved across ascomycete fungi, which include a large share of crop pathogens, and that it has no reported homologues in plants, animals or bacteria. A target present in the pathogens and absent from everything else in the field is the ideal case for off-target safety, and it is rare.

What has not been shown

The authors list what remains untested, and the list is substantial: RNA stability, delivery, cost, field performance and off-target effects.

Those are the five things that have kept spray-induced gene silencing in greenhouses for over a decade. Double-stranded RNA degrades in sunlight and rain within days, which means repeated application; manufacturing RNA at agricultural scale is expensive compared with conventional fungicides; and a 75 per cent reduction achieved in a growth chamber, where humidity, light and inoculum are controlled and nothing washes off, is not a 75 per cent reduction in a field.

The off-target question is also broader than plants and animals. Conservation across ascomycetes is the property that makes ECM33 a promising target and the property that makes it worth checking, since ascomycetes include the mycorrhizal and saprotrophic fungi that soil depends on. A spray that reaches only the leaf surface may never test that concern — and the release does not say whether anyone has looked.

Two gaps in the release itself are worth flagging. It names no individual authors, and it contains no direct quotations from any researcher, so the interpretation above follows the release’s own summary of the paper rather than anything the authors said in their own words.

The paper was published on 13 March 2026. The work was funded by MCIN/AEI and FEDER under grants PID2019-107464RB-C21 and PID2022-136240OB-C21, with a predoctoral contract from MCIN/AEI, and open-access charges covered by the Universidad de Málaga and CBUA. The authors declare no competing interests.

---

Notes

1. “Cracking the wall: the fungal cell wall assembly protein ECM33 is a promising molecular target in powdery mildew fungi.” Horticulture Research, 2026. DOI: 10.1093/hr/uhag101. The release does not name the paper’s authors.
2. Nanjing Agricultural University The Academy of Science. “Silencing a fungal wall protein cuts powdery mildew on melon leaves.” EurekAlert!, 4 August 2026. https://www.eurekalert.org/news-releases/1138768.

Photo by JÉSHOOTS on Pexels.com