Impact of mutations on the allosteric conformational equilibrium

Area of Science:

• Biochemistry
• Computational Biology
• Structural Biology

Background:

• Allostery is a key regulatory mechanism in proteins, involving effector binding that alters distant functional sites.
• Allosteric regulation relies on a conformational equilibrium between different protein substates.
• Understanding and predicting allosteric modulation is crucial for protein engineering and drug discovery.

Purpose of the Study:

• To develop a computational method for predicting mutation-induced shifts in protein allosteric conformational equilibrium.
• To identify key thermodynamic properties governing allosteric regulation.
• To provide a tool for analyzing ligand-induced conformational transitions.

Main Methods:

• Utilized molecular dynamics simulations on simplified energy landscapes to model ligand-induced conformational changes.
• Developed a feature set based on molecular mechanics, stereochemistry, and site coupling to quantify allosteric properties.
• Applied a machine-learning algorithm trained on data from 10 proteins and 179 mutations.

Main Results:

• Successfully predicted the magnitude and sign of allosteric conformational equilibrium shifts caused by mutations.
• Achieved an average unsigned error of 1k(B)T for a significant fraction of mutations.
• Demonstrated accurate prediction of mutation effects on an independent 11th protein not used in training.

Conclusions:

• The developed machine-learning model accurately predicts the impact of mutations on protein allosteric conformational equilibrium.
• The study identifies critical thermodynamic features that drive allosteric modulation.
• This approach offers a valuable tool for understanding and engineering protein allostery.