MMGBSA as a tool to understand the binding affinities of filamin-peptide interactions.

Filamins (FLN) are large dimeric proteins that cross-link actin and work as important scaffolds in human cells. FLNs consist of an N-terminal actin-binding domain followed by 24 immunoglobulin-like domains (FLN1-24). FLN domains are divided into four subgroups based on their amino acid sequences. One of these subgroups, including domains 4, 9, 12, 17, 19, 21, and 23, shares a similar ligand-binding site between the β strands C and D. Several proteins, such as integrins β2 and β7, glycoprotein Ibα (GPIbα), and migfilin, have been shown to bind to this site. Here, we computationally estimated the binding free energies of filamin A (FLNa) subunits with bound peptides using the molecular mechanics-generalized Born surface area (MMGBSA) method. The obtained computational results correlated well with the experimental data, and they ranked efficiently both the binding of one ligand to all used FLNa-domains and the binding of all used ligands to FLNa21. Furthermore, the steered molecular dynamics (SMD) simulations pinpointed the binding hot spots for these complexes. These results demonstrate that molecular dynamics combined with free energy calculations are applicable to estimating the energetics of protein-protein interactions and can be used to direct the development of novel FLN function modulators.