The eventful journey of a ligand in and out of the binding site
- 1Atomistic Simulations
- 2Italian Institute of Technology
Abstract
The route a ligand follows to reach and leave a protein binding site typically involves complex molecular mechanisms that shape its efficacy as a biologically active compound. These mechanisms are reflected in the binding and unbinding kinetics, which often predict biological activity better than binding affinity alone, yet their calculation remains a bottleneck in ligand characterization. Here, we propose an approach that combines committor-based enhanced sampling with spectral analysis of the infinitesimal generator to characterize in detail how a ligand reaches the binding site of a protein and is eventually expelled. Applied to the paradigmatic benzamidine (BEN)–trypsin complex and the congeneric para-amino benzamidine (PAB), this framework reveals distinct and multifaceted binding pathways for the two ligands, uncovering a previously uncharacterized BEN binding site that competes with the crystallographic site, while identifying a common rotational bottleneck for unbinding. The computed unbinding free energies closely agree with experimental measurements. The spectral decomposition of the infinitesimal generator provides access to the slow dynamical modes of the system. From the slowest mode, we derive residence times and kinetic rate constants that agree with available experimental data and with independent OPES flooding simulations, while requiring substantially less computational effort. This approach offers a privileged perspective to resolve the complexity of protein–ligand binding-unbinding process, with potential applications to ligand screening and the characterization of multiple binding sites.
Related
- Ligand binding remodels the oligomeric state and active site of Klebsiella pneumoniae azoreductase
- Identification and Ligand-Binding Mechanism Analysis of a C4-Dicarboxylate Chemoreceptor from Agrobacterium tumefaciens
- From Ligand Binding to Pore Opening in the Mosquito Odorant Receptor AaegOR10
- Frustration Quenching and Network Topology of the Energy Landscape as Primary Determinants of Protein-Ligand Binding Pose Prediction by Deep Learning Models
- Structural basis of phospholipid sensing nuclear receptor LRH-1 recognition by carboxylic acid bioisosteres reveals an alternative ligand interaction site
BibTeX
@misc{berselli2026eventful,
title = {{The eventful journey of a ligand in and out of the binding site}},
author = {Alessandro Berselli and Michele Parrinello},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15010461/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010461/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - The eventful journey of a ligand in and out of the binding site
AU - Alessandro Berselli
AU - Michele Parrinello
PY - 2026
DA - 2026/10/14
PB - ChemRxiv
DO - 10.26434/chemrxiv.15010461/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010461/v1
AB - The route a ligand follows to reach and leave a protein binding site typically involves complex molecular mechanisms that shape its efficacy as a biologically active compound. These mechanisms are reflected in the binding and unbinding kinetics, which often predict biological activity better than binding affinity alone, yet their calculation remains a bottleneck in ligand characterization. Here, we propose an approach that combines committor-based enhanced sampling with spectral analysis of the infinitesimal generator to characterize in detail how a ligand reaches the binding site of a protein and is eventually expelled. Applied to the paradigmatic benzamidine (BEN)–trypsin complex and the congeneric para-amino benzamidine (PAB), this framework reveals distinct and multifaceted binding pathways for the two ligands, uncovering a previously uncharacterized BEN binding site that competes with the crystallographic site, while identifying a common rotational bottleneck for unbinding. The computed unbinding free energies closely agree with experimental measurements. The spectral decomposition of the infinitesimal generator provides access to the slow dynamical modes of the system. From the slowest mode, we derive residence times and kinetic rate constants that agree with available experimental data and with independent OPES flooding simulations, while requiring substantially less computational effort. This approach offers a privileged perspective to resolve the complexity of protein–ligand binding-unbinding process, with potential applications to ligand screening and the characterization of multiple binding sites.
ER -