Characterization of protein-folding pathways by reduced-space modeling

Area of Science:

• Computational Biology
• Biophysics
• Protein Folding Dynamics

Background:

• Simulating protein folding pathways for large proteins is challenging due to computational limitations.
• Experimental and simulation studies of protein folding and unfolding are complex.
• Chymotrypsin inhibitor 2 (CI2) and barnase are well-characterized model systems for folding research.

Purpose of the Study:

• To simulate the initial folding stages of larger proteins using a reduced-space model.
• To identify initiation and nucleation sites crucial for protein folding.
• To validate the accuracy of knowledge-based potentials in reproducing folding pathways.

Main Methods:

• Utilized the CA-CB-side chain (CABS) reduced-space protein-modeling tool.
• Employed isothermal Monte Carlo (MC) dynamics with knowledge-based potentials.
• Simulated folding initiation sites and tertiary interaction networks.

Main Results:

• Identified initiation sites with residual structure and weak tertiary interactions essential for folding.
• Located nucleation sites that initiate the tertiary interaction network.
• MC simulations revealed an unambiguous folding sequence and cooperative substructures for CI2, consistent with experimental data.

Conclusions:

• Knowledge-based potentials in reduced models are effective for predicting protein structure and reproducing folding pathways.
• The CABS model and MC dynamics successfully simulate protein folding mechanisms, extending the applicability of reduced models.
• This approach provides insights into protein folding initiation and overcomes the Levinthal paradox.