Computational Methods
1. Computational approach
The crystal structures were retrieved from the RCSB database and missing residues were added using Modeller 10.2. (Webb, B, & Sali, A. 2016). Simulations were conducted with Gromacs2020 (http://www.gromacs.org/) using CHARMM27 (CHARM22 plus CMAP for proteins) force field periodic boundary conditions in the ORACLE server (Huang et al, 2017). The structures were solvated in a truncated octahedron box of Simple point charge water model. The solvated system was neutralized with Na+ or Cl− counter ions using the tleap program. Particle Mesh Ewald was employed to calculate the long-range electrostatic interactions. The cut-off distance for the long-range van der Waals energy term was 12.0 Å. The system was then minimized at a maximum force of 1000.0 KJ/mol/nm, by using 50,000 steps. The solvated and energy minimized systems were further equilibrated for 100 ps under NVT and NPT ensemble processes. The dynamics were integrated using the velocity Verlet integrator, with a time step of 2 fs and bonds constrained using the LINCS algorithm. The simulations were performed for 100, 200 ns or 500 ns for different immune complexes, at 3 or 4 different temperatures: 37℃ (310K), 38℃ (311K), 39℃ (312K) and 40℃ (313K). In some instances, duplicate or triplicate runs were also produced, starting with different random seeds. We have deployed two thermostats: an initial Berendsen thermostat, due to its capability to equilibrate rapidly large protein complexes at the temperatures of interest, followed by the use of the V-rescale (Bussi-Donadio-Parrinello) thermostat for the MD production runs (Bussi et al, 2007). All files include PDBs obtained every 50 ns, starting with the initial structure at 0 ns.
2. Determination of Binding Free Energies
To determine the binding free energies, we utilized gmx_MMPBSA, which combines GROMACS and AmberTools functions for performing end-state free energy calculations (Valdés-Tresanco et al, 2021). This tool calculates the average molecular mechanical energy, taking into account both bonded (bond, angle, dihedral) and nonbonded interactions (van der Waals and electrostatic interactions), using molecular mechanics (MM) force-field parameters. This approach allows us to break down the binding energy into contributions from each residue, providing a more comprehensive understanding of the binding mechanisms. In our study, we performed a MM/GBSA analysis on the trajectories obtained from molecular dynamics simulations of antigen-antibody complexes (Valdés-Tresanco et al, 2023), with a focus on the contributions of all amino acids located within 6 Å from the binding interface to the binding energies. Energy calculations were performed using the MMPBSA.py core. Results were analyzed and visualized with gmx_MMPBSA_ana tool.
References:
Bussi, G., Donadio, D., & Parrinello, M. (2007). Canonical sampling through velocity rescaling. The Journal of chemical physics, 126(1), 014101. https://doi.org/10.1063/1.2408420 .
Huang, J., Rauscher, S., Nawrocki, G., Ran, T., Feig, M., de Groot, B. L., Grubmüller, H., & MacKerell, A. D., Jr (2017). CHARMM36m: an improved force field for folded and intrinsically disordered proteins. Nature Methods, 14(1), 71–73. https://doi.org/10.1038/nmeth.4067.
Valdés-Tresanco, M. S., Valdés-Tresanco, M. E., Valiente, P. A., & Moreno, E. (2021). gmx_MMPBSA: a new tool to perform end-state free energy calculations with GROMACS. Journal of chemical theory and computation, 17(10), 6281-6291. https://doi.org/10.1021/acs.jctc.1c00645.
Valdés-Tresanco, M. E., Valdés-Tresanco, M. S., Moreno, E., & Valiente, P. A. (2023). Assessment of Different Parameters on the Accuracy of Computational Alanine Scanning of Protein-Protein Complexes with the Molecular Mechanics/Generalized Born Surface Area Method. Journal of physical chemistry B, 127(4), 944–954. https://doi.org/10.1021/acs.jpcb.2c07079.
Webb, B., & Sali, A. (2016). Comparative protein structure modeling using MODELLER. Current protocols in bioinformatics, 54(1), 5-6. https://doi.org/10.1002/cpbi.3.