Predicting internal protein dynamics from structures using coupled networks of hindered rotators

Area of Science:

• Biophysics
• Structural Biology
• Nuclear Magnetic Resonance (NMR) Spectroscopy

Background:

• Internal protein motions, like amide bond oscillations, are crucial for protein function.
• These motions can be quantified using a local order parameter.
• Experimental determination involves measuring relaxation rates of nitrogen-15 (15N) nuclei via NMR.

Purpose of the Study:

• To demonstrate that local order parameters (S(ii)^2) can be predicted directly from protein structure.
• To develop a computational method for predicting protein dynamics from static structural data.

Main Methods:

• Describing the diffusive motion of amide bond internuclear vectors within a potential energy landscape.
• Incorporating angular deviations between neighboring internuclear vectors and heavy atom vectors.
• Defining atomic vicinity using a 7.5 Angstrom cutoff distance.

Main Results:

• Local order parameters (S(ii)^2) were successfully predicted from protein structures.
• Prediction accuracy was achieved using a limited set of heavy atom coordinates.
• Including more heavy atoms did not significantly improve prediction accuracy.

Conclusions:

• Protein structure alone is sufficient for predicting local dynamic parameters.
• The developed method provides a valuable tool for understanding protein flexibility.
• Successful applications to calmodulin, calbindin, and interleukin-4 highlight the method's utility and limitations.