The team, led by Hao Qi and Yun Chen of Zhejiang University together with colleagues, began by mapping the complete repertoire of secondary metabolite biosynthetic gene clusters, or BGCs, in the reference strain PH-1. Using the genome mining tool antiSMASH combined with careful re-annotation of the genome, they compiled a catalog of 53 such clusters, each containing a core enzyme such as a polyketide synthase, a non-ribosomal peptide synthase, a terpene synthase, or, in some cases, hybrid combinations of these assembly lines. Non-ribosomal peptide synthase clusters were the most abundant with 23 members, followed by 13 terpene clusters, 11 polyketide clusters, five hybrids, and a single cyclodipeptide synthase cluster. Strikingly, only 15 of the 53 clusters had previously been chemically characterized, meaning most of the fungus’s chemical output remained a complete mystery.

To understand what these silent or poorly understood clusters actually do, the researchers built something unprecedented: a complete knockout library in which the backbone gene of every single cluster was individually deleted through homologous recombination. Each deletion was verified by PCR, with at least three independent transformants per gene, creating a permanent genetic resource for the Fusarium research community. The team then subjected all 53 mutants to an ambitious phenotyping campaign, measuring 24 different traits under a dizzying array of conditions. In total, the study generated 1,272 individual phenotypic measurements, a scale of functional annotation rarely attempted for secondary metabolism in any filamentous fungus.

The results demolished the long-standing assumption that secondary metabolism is a dispensable luxury. When the mutants were grown on six different culture media ranging from nutrient-rich potato dextrose agar to minimal medium, 24 of the 53 mutants showed significantly altered colony growth in at least one condition. The effects were strongly medium-dependent, a textbook example of genotype-by-environment interaction: several mutants that formed conspicuously small colonies on rich media grew perfectly well on minimal medium. This pattern suggests that certain clusters help the fungus exploit specific nutrients rather than simply acting as an energetic burden, directly contradicting the classic energy-drain model of secondary metabolism.

Developmental defects proved even more dramatic. Several clusters turned out to be essential for asexual reproduction, the process that generates the spores by which the fungus spreads through wheat fields. Deleting the backbone gene of cluster BGC52 or the siderophore cluster BGC27 left the fungus nearly or completely unable to produce conidia, with two mutants generating no detectable spores at all under the assay conditions. Other clusters shaped spore architecture itself: one mutant produced spores only half the normal length with fewer cross-walls, while another disrupted germination so severely that fewer than 10 percent of spores sprouted under conditions where wild-type spores germinated at roughly 90 percent. Importantly, these defects far exceeded any effects on vegetative growth, indicating that secondary metabolites actively program development rather than merely supporting it indirectly.

The stress experiments delivered perhaps the most conceptually surprising findings. Across 15 conditions encompassing ionic, osmotic, oxidative, cell wall, pH, temperature, and fungicide challenges, 17 mutants, or 32.1 percent of the library, showed altered sensitivity to at least one stressor. A siderophore-deficient mutant was specifically hypersensitive to vitamin K3-induced oxidative stress, consistent with the known role of siderophores in buffering redox-active iron and preventing Fenton chemistry. Mutants lacking particular polyketide pathways actually tolerated calcium stress better than the wild type, hinting at specialized roles in metal homeostasis. Fungicide treatment itself reprogrammed secondary metabolite gene expression, with carbendazim inducing ten core cluster genes, and several mutants displayed altered sensitivity to tebuconazole or phenamacril. The picture that emerges is of secondary metabolism functioning as a physiological buffer, absorbing environmental shocks rather than merely responding to them.

During actual infection, the gene clusters followed strikingly different timetables in different wheat tissues. In wheat heads, secondary metabolite genes showed their strongest activation three days after inoculation, while coleoptile infection displayed the opposite trajectory: early repression followed by powerful late induction, with some genes upregulated more than hundredfold or even thousandfold as infection progressed. When the researchers tested all 53 mutants for virulence, six showed significantly reduced disease on wheat heads, but the tissue specificity was remarkable. The siderophore cluster mutant, for example, was crippled on wheat heads yet indistinguishable from wild type on coleoptiles, evidence that the fungus deploys distinct chemical strategies depending on the microenvironment it invades.

Two previously uncharacterized clusters emerged as critical virulence factors, and their mechanistic dissection yielded genuine surprises. BGC36, a 26-gene polyketide cluster whose core synthase is homologous to the gibepyrone A biosynthetic gene in Fusarium fujikuroi, proved essential for full virulence on both wheat heads and coleoptiles. Mutants lacking any of several BGC36 genes infected the inoculated spikelet but failed to spread along the rachis, a hallmark of impaired deoxynivalenol production. Consistent with this, the mutants showed reduced expression of the trichothecene biosynthesis genes TRI1, TRI5, and TRI101, failed to form the spherical DON-toxisomes where the toxin is manufactured, and produced significantly less toxin, positioning BGC36 as a positive regulator of the mycotoxin program.

BGC47, a non-ribosomal peptide synthase-type cluster, revealed an entirely different route to virulence. Mutants in this cluster were hypersensitive to the cell wall stressor Congo red and the membrane disruptor SDS, released abundant protoplasts when treated with wall-degrading enzymes, and showed markedly reduced phosphorylation of Mgv1, the map kinase that anchors the cell wall integrity pathway. They also failed to form toxisomes and produced far less deoxynivalenol, suggesting that envelope stress signaling and toxin production are coupled through chromatin-level regulatory programs. The authors propose that BGC-derived metabolites may feed back into stress signaling cascades, reshaping transcriptional and epigenetic states at the trichothecene locus rather than acting through simple linear pathways.

The broader implications reach well beyond basic fungal biology. By demonstrating that nearly half of all predicted secondary metabolite clusters contribute measurably to growth, spore production, stress tolerance, or pathogenicity, the study establishes secondary metabolism as a core pillar of fungal physiology rather than an accessory biochemical layer. The knockout library itself constitutes a permanent community resource for assigning functions to the dozens of clusters whose chemical products remain unidentified. For agriculture, the identification of BGC36 and BGC47 as virulence determinants intertwined with deoxynivalenol biosynthesis offers fresh molecular targets: drugs or biocontrol strategies that disable these clusters could simultaneously blunt the fungus’s ability to spread through wheat heads and reduce mycotoxin contamination of the food supply, a dual benefit in the ongoing battle against one of the world’s most destructive plant diseases.

Subject of Research: Functional analysis of secondary metabolite biosynthetic gene clusters in the wheat pathogen Fusarium graminearum

Article Title: A functional atlas of secondary metabolite biosynthetic gene clusters governing growth, stress adaptation, and pathogenicity in Fusarium graminearum

Article References: Qi, H., Zhao, L., Xu, L., Liu, C., Cheng, H., Han, X., Ren, Y., Xu, C., Yan, J., Jiang, C., Ma, B., Ma, Z., & Chen, Y. (2026). A functional atlas of secondary metabolite biosynthetic gene clusters governing growth, stress adaptation, and pathogenicity in Fusarium graminearum. Crop Health, 4(1), Article 8. https://doi.org/10.1007/s44297-026-00070-x

Image Credits: AI Generated

DOI: 10.1007/s44297-026-00070-x

Keywords: Fusarium graminearum, secondary metabolites, biosynthetic gene clusters, deoxynivalenol, Fusarium head blight, mycotoxins, virulence, cell wall integrity, knockout library, siderophore, wheat infection, fungal physiology