Fungal Killer’s Achilles Heel Found in RNA Splicing Machinery

A single RNA-binding protein sits at the heart of one of agriculture’s most destructive pathogens, and scientists have now shown that removing it collapses nearly every weapon the fungus possesses. Fusarium graminearum, the mold behind Fusarium head blight, devastates wheat and maize crops worldwide and laces infected grain with deoxynivalenol, a mycotoxin that threatens food and feed safety. A new study published in Stress Biology by Yiyi Ren, Yun Chen and colleagues at Zhejiang University reveals that a deeply conserved protein called Lsm8 acts as a master regulator of RNA splicing fidelity, and that its loss cripples the fungus’s growth, stress tolerance, reproduction, virulence and toxin production all at once. The finding exposes a previously hidden post-transcriptional control layer in fungal biology and points to RNA-processing machinery as a promising target for broad-spectrum antifungal strategies.

Lsm8 is the defining subunit of the nuclear Lsm2-8 complex, a heptameric ring of Like-Sm proteins that binds the 3-prime end of U6 small nuclear RNA, the catalytic core of the spliceosome. The spliceosome is the macromolecular machine that removes introns from pre-messenger RNA, and without properly assembled Lsm2-8, U6 is left unprotected and splicing cycles falter. Sequence analysis showed that the Fusarium Lsm8 protein shares roughly half its identity with its yeast counterpart and about 65 percent with the human version, underscoring how ancient this module is. Although Lsm8 lacks a predicted nuclear localization signal, a fluorescent Lsm8-GFP fusion co-localized precisely with a histone marker in fungal nuclei, confirming where the protein operates.

To map the complex’s architecture, the team deployed a battery of interaction assays. Yeast two-hybrid tests revealed direct contacts between Lsm8 and Lsm2 and Lsm3, bimolecular fluorescence complementation captured a nuclear interaction between Lsm8 and Lsm4, and co-immunoprecipitation validated all of these associations in living fungal cells. Most decisively, affinity purification followed by mass spectrometry, using Lsm8-GFP as bait, co-purified all six remaining Lsm subunits. The picture that emerged is of Lsm8 as a central architectural keystone. When the researchers deleted the LSM8 gene, GFP-tagged Lsm2, Lsm3 and Lsm4 abandoned the nucleus entirely and diffused throughout the cytoplasm, demonstrating that Lsm8 is indispensable for both assembling the ring and escorting it to its workplace.

The consequences of losing this keystone were dramatic. Deletion mutants grew slowly on every medium tested, from rich potato dextrose agar to minimal medium. Transcriptome sequencing revealed massive reprogramming: 1,710 genes were significantly upregulated and 1,796 downregulated. The upregulated set was enriched in ribosome biogenesis, spliceosome function and RNA degradation, which the authors interpret as a compensatory feedback response to the missing splicing machinery. The downregulated genes clustered in carbon and nitrogen metabolism, including glycolysis and the pentose phosphate pathway, providing a molecular explanation for the growth collapse. Quantitative PCR confirmed the sequencing results for splicing factors such as PRP4 and PRP5, degradation factors such as RRP6 and XRN2, and glycolytic genes such as FBA1 and ENO1.

The most striking defect, however, lay in splicing itself. Across 13,982 detected introns, the median intron retention rate jumped from 0.011 in the wild type to 0.089 in the mutant, a highly significant eight-fold increase. In total, 2,344 introns from 1,821 genes showed significantly elevated retention, while only 45 introns became better spliced. Crucially, most of the affected genes, 1,270 of them, showed no change in overall transcript abundance, meaning the damage was purely post-transcriptional: transcripts were made normally but never spliced correctly. Retained introns disrupt reading frames, producing truncated or non-functional proteins, or triggering nonsense-mediated decay. The affected genes were enriched in chromatin remodeling, RNA polymerase activity, nucleocytoplasmic transport and DNA repair, processes essential to cellular integrity.

The splicing failures mapped directly onto the fungus’s environmental vulnerabilities. When challenged with a panel of eight stress agents, the mutant proved hypersensitive to all of them, with the strongest effects under osmotic stress from salt and the fungicide fludioxonil, which works by hyperactivating the osmotic stress pathway. The team traced this to the high osmolarity glycerol, or HOG, MAPK cascade. HOG1 carries eight introns, and nearly all of them were inefficiently spliced in the mutant, with defects worsening under stress. Total HOG1 mRNA levels were unchanged, but Hog1 protein levels dropped significantly, showing that the defect lies in protein production rather than transcription. Intriguingly, the residual Hog1 that remained was hyperphosphorylated, and intracellular glycerol accumulated to roughly three times wild-type levels, suggesting distorted pathway feedback. The downstream transcription factor ATF1 also showed splicing abnormalities.

Reproduction fared no better. The mutant produced fewer conidia, the asexual spores that spread infection, and those spores were shorter and abnormally septated. Sexual development was abolished outright: on carrot agar under black light, the mutant formed no perithecia at all, the fruiting bodies that generate overwintering ascospores. The molecular culprits were again splicing defects, this time in two developmental regulators. GPA1, encoding a heterotrimeric G protein alpha subunit, retained all three of its introns at elevated rates, and STE12, a transcription factor essential for sexual reproduction, retained both of its introns. As with HOG1, total mRNA levels were largely unaffected, reinforcing the theme that Lsm8’s influence operates after transcription.

Virulence and mycotoxin synthesis collapsed in parallel. Infection assays on wheat heads, wheat coleoptiles and corn silks showed the mutant was nearly nonpathogenic, and deoxynivalenol production fell sharply. The TRI gene cluster that manufactures the toxin was hit twice: key genes including TRI1, TRI4, TRI5, TRI6, TRI10 and TRI101 were transcriptionally downregulated, and the splicing efficiency of TRI1, TRI4 and TRI5 transcripts was significantly reduced. Microscopy of a Tri1-GFP reporter delivered perhaps the most visually compelling result: in the wild type, Tri1 aggregates into the spherical toxisomes, specialized compartments where deoxynivalenol is synthesized, but in the mutant almost no toxisomes formed and Tri1-GFP protein levels plummeted. Lsm8 thus guards the fungus’s chemical arsenal at every regulatory level simultaneously.

Finally, the study answered a long-standing question about what happens to all those defective transcripts. In eukaryotes, aberrant RNAs are normally cleared by nonsense-mediated decay, the RNA exosome, which degrades transcripts in the 3-prime to 5-prime direction, or by Xrn enzymes working in the opposite direction. Mass spectrometry showed that Lsm8 physically interacts with core exosome subunits including Rrp45, Rrp4 and Dis3, and the mutant compensated by upregulating exosome components. Genetic dissection proved decisive: intron-retained HOG1 transcripts accumulated massively in an exosome-subunit deletion mutant and even more in an Lsm8-exosome double mutant, while accumulation in an Xrn1 mutant was comparatively modest. The exosome, the authors conclude, is the predominant degrader of intron-retained transcripts arising from Lsm2-8 dysfunction, with Xrn1 playing a supporting role. Because both the Lsm2-8 complex and the exosome are conserved across eukaryotes, from yeast and plants to worms and humans, this coupled splicing-and-surveillance module appears to be a fundamental feature of eukaryotic biology. In Candida albicans, a human fungal pathogen, Lsm loss already impairs filamentation and pathogenicity, suggesting the Fusarium findings generalize. Targeting this conserved module, the researchers argue, could yield antifungal strategies that strike at core RNA-processing machinery shared by diverse pathogens, simultaneously curbing crop disease and mycotoxin contamination while sparing the host, whose own cells process RNA through homologous but distinct regulatory contexts.

Subject of Research: RNA splicing fidelity and the Lsm8-exosome surveillance module in fungal stress adaptation and virulence

Article Title: A conserved Lsm8–exosome module maintains RNA splicing fidelity to control fungal stress adaptation and virulence

Article References: Ren, Y., Cheng, H., Han, X., Guo, M., Xu, C., Yan, J., Ge, Z., Ma, Z., & Chen, Y. (2026). A conserved Lsm8–exosome module maintains RNA splicing fidelity to control fungal stress adaptation and virulence. Stress Biology, 6(1), Article 14. https://doi.org/10.1007/s44154-026-00285-6

Image Credits: AI Generated

DOI: 10.1007/s44154-026-00285-6

Keywords: Fusarium graminearum, Lsm8, Lsm2-8 complex, RNA splicing, intron retention, RNA exosome, deoxynivalenol, mycotoxin biosynthesis, HOG-MAPK pathway, fungal virulence, post-transcriptional regulation, antifungal targets