A Foundation Model Identifies Broad-Spectrum Antimicrobial Peptides against Drug-Resistant Bacterial Infection

Area of Science:

• Biochemistry
• Computational Biology
• Drug Discovery

Background:

• Antimicrobial resistance (AMR) poses a significant global health threat, necessitating the development of novel therapeutic strategies.
• Antimicrobial peptides (AMPs) are a promising class of molecules with broad-spectrum activity, but their discovery and optimization remain challenging.

Purpose of the Study:

• To develop and validate a deep generative framework, deepAMP, for the accelerated discovery of potent and broad-spectrum antimicrobial peptides.
• To identify novel AMP candidates with enhanced membrane-disrupting abilities and efficacy against drug-resistant bacterial pathogens.

Main Methods:

• A peptide language-based deep generative framework (deepAMP) was employed for AMP design.
• 18 Tier 1 (T1) and 11 Tier 2 (T2) AMP candidates were synthesized and experimentally evaluated using a two-round design with cross-optimization and validation.
• In vitro antibacterial assays against Gram-positive (S. aureus) and Gram-negative bacteria (K. pneumoniae, P. aeruginosa) were performed.

Main Results:

• Over 90% of designed AMPs demonstrated superior inhibition compared to penetratin against both Gram-positive and Gram-negative bacteria.
• AMP T2-9 exhibited the strongest antibacterial activity, comparable to FDA-approved antibiotics.
• Three AMPs (T1-2, T1-5, T2-10) significantly reduced S. aureus resistance compared to ciprofloxacin and showed efficacy in a P. aeruginosa infected wound mouse model.

Conclusions:

• The deepAMP framework effectively accelerates the discovery of potent, broad-spectrum antimicrobial peptides.
• The identified AMPs show significant potential for combating drug-resistant bacterial infections, including topical wound infections.