Purposely engineered drug–target mismatches for entropy-based drug optimization

Proteins are dynamic objects that often undergo significant structural change and reduce their conformational possibilities upon ligand binding. Thus, unless dynamic information is incorporated, structure-based drug design becomes of limited applicability. Even within a dynamic approach, a rarely visited scenario arises as proteins increase their entropy content upon ligand binding by locally enhancing conformational exploration in the complex. In this opinion piece, we argue that this binding mode is of primary importance in drug development because it allows for drugs that are not optimized in the conventional way but feature mismatches with the target. Thus, we advocate entropy optimization that exploits dynamic information for drug design.