Predicting Optimal DEER Label Positions to Study Protein Conformational Heterogeneity

Area of Science:

• Biophysics
• Computational Biology
• Structural Biology

Background:

• Double Electron-Electron Resonance (DEER) spectroscopy is vital for probing protein conformational heterogeneity by measuring distances between spin labels.
• Selecting optimal residue pairs for spin labeling is challenging due to the vast number of possibilities and the need to capture relevant protein dynamics.

Purpose of the Study:

• To develop and validate a computational methodology for predicting optimal residue pairs for DEER spectroscopy in proteins.
• To identify spin-labeling strategies that maximize information gain about protein conformational dynamics.

Main Methods:

• Utilized all-atom molecular dynamics simulations to model protein dynamics.
• Employed Markov state models (MSMs) to identify slow conformational processes.
• Applied a genetic algorithm to efficiently search for optimal residue-pair combinations based on dynamic information.

Main Results:

• The developed method successfully ranks residue-pair sets for their ability to represent protein dynamics.
• Predicted optimal residue pairs outperformed experimentally used pairs in capturing conformational changes for selected proteins (β2 adrenergic receptor, calmodulin C-terminal domain, PepT_So).

Conclusions:

• The computational approach provides a rational strategy for selecting DEER residue pairs, enhancing insights into protein conformational heterogeneity.
• This method optimizes information extraction by focusing on slow and orthogonal conformational processes, minimizing the number of required labels.