From virtual screening to functional validation: An MM/GBSA and molecular dynamics driven strategy for targeting phytopathogenic fungal ferrochelatase

This is the authors' abstract. We don't add key points for this paper.

Fusarium graminearum is one of the primary causal agents of wheat Fusarium head blight, a devastating disease that poses a significant threat to global grain security through yield reduction and contamination with the mycotoxin deoxynivalenol (DON). Our previous study showed that fungal ferrochelatase (FgFc), the terminal enzyme in the fungal-specific heme biosynthesis pathway in F. graminearum , serves as a promising target for novel fungicide development. A high-confidence structural model of FgFc was generated using AlphaFold3 , enabling the virtual screening of 1.6 million compounds from the ChemDiv library. From this screen, compound 13324 was identified as a high-affinity candidate, exhibiting a binding energy of -9.8 kcal mol -1 , an MM/GBSA score of -50.58 kcal mol -1 , and a stable RMSD of 2.35 Å. In bioassays, compound 13324 exhibited broad-spectrum antifungal activity, showing effective mycelial inhibition against Rhizoctonia solani (EC 5 0 = 5.25 µg/mL) and Stagonosporopsis cucurbitacearum (EC 5 0 = 8.31 µg/mL). Against F. graminearum , the compound inhibited more than 98% of conidial germination at a concentration of 8 µg/mL and reduced DON production by 2.52-fold at 100 µg/mL. Mechanistic analysis revealed a marked upregulation (1.26-fold) of the FgFc gene, which was consistent with a compensatory cellular response to target inhibition. These results demonstrate that compound 13324 serves as a promising chemical scaffold targeting the enzyme FgFc for fungicide development. This integrated strategy—combining AI-based structure prediction with functional validation—offers a robust framework for accelerating the discovery of pesticides with novel modes of action.