Atomic-level characterization of the structural dynamics of proteins

Area of Science:

• Biophysics
• Computational Biology
• Structural Biology

Background:

• Molecular dynamics (MD) simulations are crucial for understanding atomic-level protein motions.
• Current MD simulations are limited to shorter timescales than many biologically significant conformational changes.

Purpose of the Study:

• To investigate protein folding and conformational changes using extended all-atom MD simulations.
• To overcome timescale limitations in traditional MD simulations for studying protein dynamics.

Main Methods:

• Conducted extremely long all-atom MD simulations on a specialized machine.
• Performed equilibrium simulations of a WW protein domain to observe folding/unfolding events.
• Simulated substructures of the WW protein to identify pathway determinants.
• Executed a 1-millisecond simulation of the BPTI protein.

Main Results:

• Observed consistent folding and unfolding pathways for the WW protein domain.
• Identified key determinants influencing the protein folding pathway.
• Revealed a small set of distinct conformational states in BPTI.
• Demonstrated that interconversion between BPTI states is over 1000 times slower than local relaxations.

Conclusions:

• Extended MD simulations can capture biologically relevant protein dynamics, including folding and conformational transitions.
• A well-defined pathway governs the folding of the WW protein domain.
• Proteins like BPTI exhibit slow conformational dynamics involving distinct states.