A Method to Calculate the Relative Binding Free Energy Differences of α‑Helical Stapled Peptides

Hydrocarbon-stapled
peptides are a class of bioactive α-helical
ligands developed to target protein–protein interactions. Peptide
stapling has benefited from the development of several chemical reactions
to modulate their membrane permeability and binding affinity. However,
in most current programs, choosing the best stapling positions is
usually a trial-and-error process. Here, we develop a protocol to
obtain optimal stapling positions computationally. Our method is based
on molecular dynamics simulations and free energy calculations with
nonequilibrium approaches; here, we predict the binding poses, hot-spot
residues, and binding affinity differences of a set of perfluoroarene
stapled α-helical peptides of the BIM BH3 peptide to the BCLXL receptor. The prediction of the hot-spot residues within
the target peptide through computational alanine scanning anticipates
not only the key residues for the receptor–peptide complex
formation but also which positions should be avoided when applying
the stapling groups. The staple moieties introduce local conformational
changes not only in the replaced positions but also on their neighbor
residues of the template peptide further affecting their binding behavior.
Our approach is successful at rank-ordering the binding affinities
of these stapled peptides with respect to the BIM BH3 peptide.