Antimicrobial peptide developed with machine learning sequence optimization targets drug resistant Staphylococcus...

Area of Science:

• Antimicrobial drug discovery
• Peptide therapeutics
• Computational biology

Background:

• Rising antimicrobial resistance necessitates novel antibacterial agents.
• Antimicrobial peptides (AMPs) are a promising class of therapeutics.
• Citropin 1.1 shows limited efficacy against Staphylococcus aureus.

Purpose of the Study:

• To engineer a potent antimicrobial peptide with enhanced activity against Staphylococcus aureus.
• To optimize citropin 1.1 using sequence space information and machine learning.

Main Methods:

• Utilized sequence space data from 14,743 functional antimicrobial peptides (AMPs).
• Employed traditional peptide design (truncation, substitution, structure-guided alteration).
• Integrated machine learning (ML)-backed sequence optimization.
• Tested CIT-8 efficacy against methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant S. aureus (VRSA) persister and biofilm cells.
• Assessed membrane depolarization and permeation.
• Evaluated CIT-8 in a mouse model of MRSA skin infection.

Main Results:

• Developed CIT-8, a 13-residue peptide with potent anti-staphylococcal activity.
• CIT-8 eradicated MRSA and VRSA persister cells (1 × 10^8) within 30 minutes at 40 μg/ml.
• CIT-8 reduced MRSA and VRSA biofilm viability by 3 and 4 log10, respectively.
• CIT-8 (32 μg/ml) induced membrane depolarization and permeation in S. aureus MW2.
• CIT-8 (2% w/w) significantly reduced bacterial burden in a mouse MRSA skin infection model (2.3 log10 reduction).

Conclusions:

• CIT-8 is a highly effective antimicrobial peptide against drug-resistant Staphylococcus aureus.
• The developed methodology accelerates antimicrobial peptide design through ML integration.
• CIT-8 demonstrates therapeutic potential for treating staphylococcal infections.