Stability changes upon mutation of solvent-accessible residues in proteins evaluated by database-derived potentials

Area of Science:

• Biophysics
• Protein Engineering
• Computational Biology

Background:

• Amino acid substitutions can alter protein stability, impacting function.
• Experimental measurement of folding free energy changes quantifies these stability alterations.
• Predicting stability changes is vital for protein design and understanding disease mechanisms.

Purpose of the Study:

• To evaluate effective potentials for predicting stability changes caused by single amino acid mutations.
• To assess the accuracy of different potential types for solvent-accessible residues.
• To identify key interactions governing protein stability at the surface.

Main Methods:

• Computed folding free energy changes using database-derived potentials for 106 mutations in barnase, T4 lysozyme, and other proteins.
• Compared computed values with experimentally measured folding free energy changes.
• Analyzed the influence of residue accessibility, backbone structure, and solvent ionic strength on prediction accuracy.

Main Results:

• Distance-dependent residue-residue potentials were inadequate due to dominance of hydrophobic interactions.
• Potentials based on backbone torsion angle propensities showed good correlation (R=0.87) with experimental data for 96 mutations.
• Excluding mutations affecting backbone structure or involving atypical hydrophobic interactions improved correlation.
• Increased ionic strength enhanced correlation by masking electrostatic interactions.
• Overall correlation of R=0.86 for 135 out of 150 mutations across multiple proteins.

Conclusions:

• Backbone torsion angle potentials effectively predict stability changes for solvent-accessible mutations.
• Local interactions along the protein chain are dominant at the protein surface.
• The approximations used in deriving backbone torsion potentials are adequate for this prediction task.