pipette
ESEspañol

CHARMM-Compatible Molecular Mechanics Force Fields for Multiscale Modeling of HIV-1 Replication Inhibitors

Juan S. Rey, Craig A. Peeples, Amanda C. Macke, Juan R. Perilla

Peer-reviewed journalReal-world use

In the authors' words

Abstract Molecular dynamics simulations provide insight into inhibitor-protein interactions and support rational drug design; however, their accuracy depends on reliable molecular mechanics force field parameters and coarse-grained representations that retain relevant molecular features. Although force field parameter sets for biomacromolecules are well established, the chemical diversity of small-molecule inhibitors often requires the development of system-specific parameters. Here, we present a parametrization framework integrating quantum mechanical optimization, experimental benchmarking, and small-molecule coarse-graining, demonstrated using HIV-1 inhibitors targeting multiple stages of the viral replication cycle. We provide a curated data set of CHARMM-compatible force field parameters for 12 HIV-1 inhibitors and the HIV assembly cofactor inositol hexakisphosphate (IP6), together with geometry-optimized atomic structures and ligand restraint files. The resulting parameters are validated against experimental structural and thermodynamic benchmarks, including lenacapavir binding free energy estimates, through inhibitor-bound complex simulations. We demonstrate applications of the parameter data set by quantifying inhibitor-protein interactions, including the free-energy profile of an integrase strand transfer inhibitor interacting with the viral capsid lattice, and the binding energetics of an allosteric integrase inhibitor. Furthermore, we extend the all-atom inhibitor parameter data set through a novel coarse-graining strategy for small molecules that preserves key electrostatic features. The HIV-1 inhibitor force field parameter data set is a publicly available resource that establishes a unified framework for multiscale modeling of HIV-1 inhibitors and facilitates computational and structural biology studies of drug resistance and inhibitor design. More generally, the framework presented here is broadly applicable to small-molecule parametrization and multiscale simulations in other therapeutic and biomolecular contexts.

Main resultThe abstract does not state a limitation.

Appeared: Tuesday, November 10. Journal of the American Chemical Society. Peer-reviewed journal.

DOI: 10.1021/jacs.6c11578