Convergence in determining enzyme functional descriptors across Kemp eliminase variants

Area of Science:

• Biocatalysis and enzyme engineering
• Computational chemistry and molecular modeling
• Protein dynamics and electrostatics

Background:

• Molecular simulations accelerate biocatalytic discovery by providing enzyme functional descriptors.
• These descriptors guide the search for beneficial enzyme mutants.
• The optimal active-site region size for descriptor computation across enzyme variants is currently unknown.

Purpose of the Study:

• To test the convergence of dynamics-derived and electrostatic descriptors with varying active-site region sizes.
• To establish ideal boundary distances for descriptor calculation in enzyme engineering.
• To provide a reference for selecting appropriate descriptors for predictive modeling.

Main Methods:

• Convergence tests were performed on 18 Kemp eliminase variants using six active-site regions.
• Descriptors evaluated included root-mean-square deviation (RMSD) of the active site, solvent accessible surface area (SASA) ratio, and electric field (EF) projection.
• Methods included molecular mechanics (MM) and quantum mechanics/molecular mechanics (QM/MM).

Main Results:

• Protein dynamics descriptors (RMSD_active_site, SASA_ratio) converged at a 5 Å cutoff from the substrate.
• The electrostatic descriptor (EF_C-H) converged at 6 Å using MM and 4 Å using QM/MM.
• Spearman correlation matrices identified region sizes where descriptor rankings stabilized.

Conclusions:

• Established distance cutoffs for reliable computation of enzyme functional descriptors.
• Provides crucial insights for optimizing active-site region selection in computational enzyme engineering.
• Aims to enhance the accuracy and efficiency of predictive modeling for enzyme discovery.