Dynamic profile analysis to characterize dynamics-driven allosteric sites in enzymes.

Area of Science:

• Biophysics
• Structural Biology
• Computational Biology

Background:

• Allosteric binding sites are crucial for drug discovery but identifying them remains challenging.
• Previous identification of allosteric sites relied on protein-drug co-crystal structures.
• Understanding the dynamic nature of these sites is key to predicting their function.

Purpose of the Study:

• To investigate the dynamic characteristics of known allosteric binding sites.
• To identify reliable indicators of dynamics-driven allostery for drug target identification.
• To develop a computational method for predicting potential allosteric drug binding sites.

Main Methods:

• Comprehensive search of the Protein Data Bank for protein-ligand complex structures.
• Generation of computational models to simulate ligand binding to potential pockets.
• Application of Normal Mode Analysis (NMA) of the elastic network model (ENM) to assess protein dynamics.
• Evaluation of nine profiles to characterize pocket dynamics and positional features.

Main Results:

• Identified 10 protein-ligand complex structures with similar functional forms.
• The change in Mean Square Fluctuation (ΔMSF) around the ligand emerged as the most effective profile.
• ΔMSF successfully distinguished allosteric sites from other sites in 8 out of 10 structures.
• Highlighted cases exemplify dynamics-driven allostery with significant protein dynamic changes.

Conclusions:

• Dynamics-driven allostery is characterized by substantial alterations in protein dynamics.
• ΔMSF is a promising metric for identifying potential dynamics-driven allosteric sites.
• This approach can aid in the discovery of novel allosteric drug targets for therapeutic intervention.