Multidrug-resistant (MDR) pathogens due to overuse of antibiotics have rendered many frontline treatments ineffective, necessitating novel antimicrobial agents from unexplored biological sources. Antimicrobial peptides (AMPs), exhibiting broad-spectrum activity and multi-target modes of action, are promising therapeutics with a reduced likelihood of resistance development. This study presents a computational workflow to identify and mechanistically characterize NCR-like peptides from Trigonella foenum-graecum . Three cationic peptides-TfgNCR7, TfgNCR102, and TfgNCR276 were prioritized for molecular dynamics simulations on Escherichia coli and Staphylococcus aureus membrane models. MM/PBSA analyses revealed that membrane adsorption is energetically favorable and predominantly driven by electrostatic interactions. However, TfgNCR102 exhibited surface-restricted membrane interactions and limited lipid disordering, suggesting functional divergence. Conversely, lipid order parameter ( S C D ) showed that TfgNCR7 induces pronounced disorder in S. aureus sn-1 acyl chains, whereas TfgNCR276 promotes transient water-wire formation and ion defects in E. coli membranes, suggesting potential pore-forming mechanism. Structural clustering positioned TfgNCR7 and TfgNCR276 alongside validated NCRs like CaNCR13 and MtNCR335. Both demonstrated strong interfacial binding, with residue-level energy decomposition highlighting dominant contributions from Lys, Arg, Tyr, and Trp residues. These findings position TfgNCR7 and TfgNCR276 as computationally promising candidates for experimental validation and illustrate a scalable pipeline for mining antimicrobial peptide diversity in non-model legume genomes.
Bidvi et al. (Thu,) studied this question.