Exploring the binding mechanism between human profilin (PFN1) and polyproline-10 through binding mode screening.

Area of Science:

• Biochemistry
• Computational Biology
• Structural Biology

Background:

• Protein-protein interactions (PPIs) are vital for cellular functions.
• Predicting PPI complex structures remains a significant challenge due to high degrees of freedom.
• Existing predictive techniques often struggle with accuracy and ranking.

Purpose of the Study:

• To evaluate different computational approaches for predicting and ranking protein-protein interactions.
• To identify reliable criteria for assessing the accuracy of PPI predictions.
• To develop a more effective screening method for PPIs.

Main Methods:

• Umbrella sampling to calculate potential of mean force (PMF) and identify stable binding structures.
• Regular molecular dynamics (MD) simulations to analyze interaction energy and root mean square displacement.
• Steered molecular dynamics (SMD) simulations to calculate average and minimal rupture works.

Main Results:

• PMF confirmed the crystallographic binding structure as the most stable.
• Crucial residues in PFN1 formed hydrogen bonds with P10, indicating a zipping binding mechanism.
• Average and minimal rupture works from SMD simulations showed good correlation (R² = 0.67) with PMF results.
• Conventional MD metrics like interaction energy and RMSD did not correlate well with PMF.

Conclusions:

• Steered molecular dynamics (SMD) simulations provide a promising method for ranking protein-protein interaction predictions.
• Rupture work calculated from SMD simulations serves as a reliable metric for evaluating PPIs.
• This approach advances the development of predictive models for the human interactome.