# The Effects of Heparin Binding and Arg596 Mutations on the Conformation of Thrombin–Antithrombin Michaelis Complex, Revealed by Enhanced Sampling Molecular Dynamics Simulations

## TL;DR

This study uses molecular simulations to explore how heparin and mutations in thrombin affect interactions with antithrombin, revealing structural changes that impact blood clotting.

## Contribution

The study reveals atomic-level conformational changes in thrombin-antithrombin complexes due to heparin binding and Arg596 mutations using enhanced sampling simulations.

## Key findings

- GaMD simulations identified multiple conformational types in thrombin-antithrombin complexes, showing higher flexibility in mutant and heparin-free systems.
- Arg596 mutations impair antithrombin binding, likely due to conformational instability revealed through clustering analysis.
- Heparin enhances protein interactions in ternary complexes by stabilizing specific conformations at the binding exosites.

## Abstract

The inactivation of thrombin by antithrombin is highly enhanced by the presence of heparin chains forming “bridges” between the two proteins. X-ray structures for such ternary complexes have been published, but the molecular background of the lower efficiency of smaller heparinoids on thrombin inhibition remains poorly understood. Antithrombin-resistant prothrombin mutants (mutations affecting Arg596 in prothrombin) have been reported that cause severe thrombophilia. Our aim was to study the interactions in the antithrombin–thrombin Michaelis complex both in the presence and the absence of a heparinoid chain and in the presence of pentasaccharide by using molecular dynamics. We also intended to study the complexes of thrombin mutants as well as a known alternative antithrombin conformation at the “hinge” region built using docking. The binding between the proteins was investigated by Gaussian Accelerated Molecular Dynamics (GaMD). We compared the contribution of several amino acids at the binding “exosites” between AT and the wild type and mutant thrombins and between systems containing or not containing a heparinoid. In the docking-based simulations, several of the analyzed amino acid pairs no longer contributed to the interaction, suggesting that the open “hinge” conformation has limited biological relevance. We could identify multiple conformational types using clustering, revealing high flexibility in mutants and systems without heparinoid, probably indicating lower stability. We were also able to detect the allosteric effects of the ligands on the bound thrombin. In summary, we were able to obtain conformations using GaMD that can explain the better protein–protein interactions in the ternary complexes and the impaired AT binding of the thrombin Arg596 mutants at an atomic level.