Accurate Protein Dynamic Conformational Ensembles: Combining AlphaFold, MD, and Amide ¹⁵N(¹H) NMR Relaxation

Area of Science:

• Structural biology
• Biophysics
• Computational biology

Background:

• Protein conformational heterogeneity is vital for function.
• Validating molecular dynamics (MD) simulations against experimental data is challenging.
• Accurate protein structure prediction and dynamics are key to understanding biological mechanisms.

Purpose of the Study:

• To develop and validate a method for generating time-resolved 4D conformational ensembles.
• To integrate free MD simulations with experimental NMR relaxation data.
• To identify biologically relevant protein conformations and flexible regions.

Main Methods:

• Utilized AlphaFold for initial protein structure generation.
• Performed extensive molecular dynamics (MD) simulations.
• Integrated refined experimental Nuclear Magnetic Resonance (NMR) relaxation data.
• Selected specific MD trajectory segments (RMSD plateaus) matching experimental observables.

Main Results:

• Developed an approach integrating MD simulations and NMR data for conformational ensemble analysis.
• Identified specific segments of MD trajectories consistent with experimental NMR relaxation data for Streptococcus pneumoniae PsrSp.
• Revealed two distinct regions of increased flexibility in the extracellular region of PsrSp.
• Demonstrated the functional importance of these flexible regions.

Conclusions:

• The integrated approach successfully validates theoretical MD ensembles with experimental data.
• The identified flexible regions in PsrSp are functionally significant.
• This method provides a robust framework for studying protein dynamics and conformational heterogeneity.