Ab initio folding of helix bundle proteins using molecular dynamics simulations

Area of Science:

• Computational biology
• Biophysics
• Protein folding dynamics

Background:

• Protein folding is crucial for biological function.
• Simulating protein folding accurately requires significant computational resources.
• Understanding folding pathways provides insights into protein structure-stability relationships.

Purpose of the Study:

• To demonstrate the feasibility of ab initio fast folding simulations for small- to medium-sized proteins.
• To investigate the early stages of protein folding, including hydrophobic collapse and helix formation.
• To analyze the thermodynamic behavior and native basin characteristics of studied proteins.

Main Methods:

• Utilized ab initio molecular dynamics (MD) simulations at elevated temperatures (400 K).
• Employed a Generalized Born (GB) implicit solvent model with an all-atom force field.
• Calculated free energy profiles to assess thermodynamic behavior.

Main Results:

• Successfully simulated the spontaneous formation of nativelike structures for villin headpiece and Staphylococcal protein A fragment B.
• Observed initial hydrophobic collapse and rapid helix formation as critical early folding events.
• Protein A fragment B showed preferential early formation and higher stability of its third helix.
• Free energy profiles indicated two-state thermodynamic behavior for both proteins, with protein A exhibiting broader native basins.

Conclusions:

• Implicit solvent models combined with high-temperature MD significantly reduce computational cost for direct folding simulations.
• The study provides a computationally efficient method for predicting protein structures and understanding folding mechanisms.
• Findings highlight the importance of early hydrophobic collapse and helical structure formation in the folding pathways of these proteins.