The 40-amino acid-long Fa-AMP1 peptide from buckwheat (Fagopyrum esculentum Moench) seed germ has previously been reported to exhibit potent antifungal and antibacterial effects. The availability of the three-dimensional structure of this peptide could be very useful in investigating its antimicrobial function, but the crystallographic information of this peptide is not currently available. The aim of the present study was to determine the three-dimensional structure of the Fa-AMP1 peptide using a comprehensive approach so that advanced modeling tools such as AlphaFold, SwissModel, and D-I-TASSER could be used to predict and evaluate. The predicted models were compared using validation tools QMEAN, Verify3D, and WHATCHECK. The geometric position of amino acids in all three models was examined by drawing Ramachandran plots. The distance of all three models from each other was compared using the RMSD index, and the number and position of alpha-beta-coil folds, as well as the position of disulfide bonds, were also compared. Overall, the results showed that the models presented by the AlphaFold and SwissModel tools have high convergence with each other, and while the model presented by the D-I-TASSER tool also showed high similarity with the other two models, it did not show the necessary accuracy in the configuration of cysteine amino acids and the formation of expected disulfide bonds. In the second part of the article, the Swiss docking tool was used to investigate the interaction between the model predicted by AlphaFold and chitin, sourced from the fungal cell wall, as a target ligand. The docking results were analyzed based on the position of amino acids involved in binding to the ligand in five complexes run with the peptide model. The docking results showed that the model presented by AlphaFold in the fourth run of the results presented by Swiss docking can be placed at an appropriate distance from each other, with the amino acids related to the binding site in the Fa- AMP1 peptide and the four amino acids forming hydrogen bonds with the NAG1 and two monomers. In future studies, this prediction can be investigated by conducting additional molecular dynamics (MD) simulations.

Keywords