Predicting Toxicity of Insect Venom-Derived Antimicrobial Peptides Using MD Simulations: A Comparative Study of...

Area of Science:

• Biochemistry
• Computational Biology
• Pharmacology

Background:

• Insect venom-derived antimicrobial peptides (AMPs) show therapeutic potential but face limitations due to mammalian cell toxicity.
• Current toxicity assays are resource-intensive, requiring peptide synthesis, purification, and testing for large libraries.

Purpose of the Study:

• To evaluate molecular dynamics (MD) simulations using mammalian membrane models for predicting AMP toxicity.
• To benchmark different mammalian membrane models for their efficacy in toxicity prediction.

Main Methods:

• Conducted 25 µs of MD simulations on 30 AMP analogs (16 toxic, 14 non-toxic) from five insect families using two distinct mammalian membrane models.
• Analyzed MD trajectories (500 ns each) focusing on structural stability and membrane permeability.
• Utilized Root Mean Square Deviation (RMSD) over the final 100 ns of simulations for toxicity prediction.

Main Results:

• MD simulations revealed significant differences in structural stability and membrane permeability between toxic and non-toxic AMPs, correlating with experimental data.
• Realistic mammalian membrane models accurately distinguished toxic from non-toxic AMPs with 90% accuracy based on RMSD.
• A commonly used multicomponent mammalian membrane model showed poor predictive performance for mammalian toxicity.

Conclusions:

• MD simulations provide an efficient and robust method for preliminary toxicity prediction of venom-derived AMPs.
• Realistic membrane models are crucial for accurate toxicity assessment using MD simulations.
• This approach facilitates the accelerated development of safer antimicrobial therapies.