Discovery and Prioritization of Plant-Derived Antimicrobial Peptides Through Bioactivity Screening, Proteomics, and In Silico Predictions

Antimicrobial resistance is an escalating global health threat. Antimicrobial peptides (AMPs) are a promising class of therapeutic agents due to their broad presence across all domains of life, ability to target conserved structural features of bacterial membranes, and structural diversity. Despite extensive investigation of small-molecule metabolites, plant-derived AMPs remain comparatively underexplored and represent a source of novel antimicrobial agents. Herein, established AMP discovery workflows integrating bioactivity-guided fractionation, in silico sequence mining, bottom-up proteomics, and PepSAVI-MS were applied to protein extracts from medicinal and culinary plant species including Coffea arabica (arabica coffee), Limnophila aromatica (ngo om), Eryngium foetidum (ngo gai), Perilla frutescens var. frutescens (green shiso), and Perilla frutescens var. crispa (red shiso). Computational tools leveraging signal peptide prediction and cysteine motif analysis were used to generate libraries of putative AMP sequences which were evaluated against antimicrobial activity and LC-MS/MS data. While antimicrobial activity was observed in some chromatographic fractions, direct molecular identification of the active species proved challenging, highlighting key limitations associated with peptide abundance, fractionation pattern, and underlying workflow assumptions. Collectively, this work assesses the feasibility and constraints of existing AMP discovery strategies applied to complex plant systems and provides practical insights to inform future peptide-centric antimicrobial discovery efforts.