Mechanism-Driven Screening of Membrane-Targeting and Pore-Forming Antimicrobial Peptides

Area of Science:

• Biochemistry
• Computational Biology
• Microbiology

Background:

• Antibiotic resistance necessitates novel antimicrobial agents.
• Current screening methods often neglect antimicrobial peptide (AMP) mechanisms of action.
• Targeting bacterial membranes is a key strategy for AMP development.

Purpose of the Study:

• To introduce a mechanism-driven screening approach for identifying AMPs.
• To utilize machine learning (ML) to predict peptide sequences targeting bacterial membranes.
• To discover novel AMPs with pore-forming capabilities and antimicrobial activity.

Main Methods:

• Employed ML-based computational models for mechanism-driven screening.
• Screened metaproteomes from frog and human sources.
• Validated identified peptides using liposome leakage assays, hemolysis tests, and cytotoxicity assays.
• Confirmed membrane interaction and pore formation via single-molecule experiments and electrophysiology.

Main Results:

• Identified seven novel peptides with antimicrobial activity and low cytotoxicity.
• Demonstrated membrane disruption and pore formation in bacterial models.
• Three peptides exhibited broad-spectrum activity against Gram-positive and Gram-negative bacteria.
• Confirmed peptide oligomerization and pore formation correlating with broad-spectrum activity.

Conclusions:

• The ML-driven, mechanism-focused approach is effective for discovering novel AMPs.
• Identified AMPs show potential as therapeutic agents against antibiotic-resistant bacteria.
• Pore-forming ability is a key characteristic of broad-spectrum antimicrobial activity in these AMPs.