Binding Mechanism of Thrombin–Ligand Systems Investigated by a Polarized Protein-Specific Charge Force Field and Interaction Entropy Method

Abstract

In this study, 2 groups of 10 modified ligand systems with modified P3 and P2 side chains are used to study the binding mechanism with thrombin. Experimental results show that the binding affinity is enhanced by complex ligand side chains. The binding free energy obtained from the polarized protein-specific charge (PPC) force field combined with the newly developed interaction entropy (IE) method is consistent with the experimental values with a high correlation coefficient. On the contrary, poor correlation is obtained using the traditional normal mode (Nmode) method for calculating the entropy change. Furthermore, the binding free energy and hot-spot residue energy are decomposed, and the common hotspot residues in the two groups of systems are Trp50,

Unravelling the mechanism of glucose binding in a protein-based fluorescence probe: molecular dynamics simulation with a tailor-made charge model

“…Similar findings have been repeatedly reported in the literature for a diverse set of systems: there is a large body of evidence that standard force fields with a fixed point charge model fail to describe H-bond strengths with sufficient accuracy for a variety of systems. [20][21][22][23][24][25][26][27] This seems to apply also for the case of GGBP: in pilot simulations, we were unable to find a stable glucose bound state. The H-bonded network broke and set the glucose free, in contrast to the stable bound state found in experiment.…”

“…The importance of polarisation effects on force field parameters, in particular on the force field charges was pointed out in several studies of protein-ligand binding, [20][21][22][23] peptide folding, 24,25 and protein-chromophore complexes. 26,27 Standard force field parameters, in particular the force field charges, can lead to structural instabilities or even a wrong description of the respective systems.…”

Protein–ligand free energies of binding from full-protein DFT calculations: convergence and choice of exchange–correlation functional

“…Based on these findings we are concerned about the use of less than 50 NMA calculations in some applications of MM-PBSA. [89][90][91] 4.3.3 Sampling and statistical error. Based on this study we recommend QM energy sampling at 50 snapshots.…”