Correlation of Experimental and Calculated Inhibition Constants of Protease Inhibitor Complexes

Area of Science:

• Computational chemistry
• Biophysics
• Drug discovery

Background:

• Predicting inhibitor potency is crucial for in silico screening of potential therapeutic compounds.
• Accurate prediction of binding affinity aids in identifying effective synthetic and natural drug candidates.

Purpose of the Study:

• To develop and validate a predictive workflow for calculating inhibitor potency (K_i and K_D values).
• To assess the workflow's applicability across different protease families and other protein interaction systems.

Main Methods:

• Utilized YASARA's FoldX plugin to calculate free interaction energy (ΔG) from PDB structures of protease-inhibitor complexes.
• Employed the PRODIGY server to obtain corresponding dissociation constants (K_D).
• Correlated calculated values with experimental data for various protease types.

Main Results:

• Calculated inhibitory values showed good agreement with experimental data, especially for serine proteases.
• PRODIGY server data demonstrated higher consistency across cysteine, aspartic, and metalloproteases.
• Identified a more rigid Pro14 variant of sunflower trypsin inhibitor (SFTI-1) with potentially higher potency.
• Suggested a hirudin variant as a basis for potent thrombin inhibitors.

Conclusions:

• The developed computational workflow accurately predicts inhibitor potency for protease-inhibitor complexes.
• The approach is versatile and applicable to diverse protein-protein interactions, including antibody-antigen and effector-receptor systems.
• This method facilitates efficient in silico screening for drug discovery and understanding protein interactions.