(348c) Understanding and Predicting Multimodal Ligand-Protein Interactions Using Molecular Simulations

Recent developments in multimodal chromatography made by our group and by others have successfully shown that enhanced selectivities can be achieved for protein separations in multimodal systems as compared to traditional single-mode chromatography by the appropriate choice of ligand chemistry and mobile phase modifiers. The mechanisms underlying these unique binding behaviors are not well understood. While chromatographic retention times provide the overall results of a combination of kinetic and affinity phenomena, individual cases need to be studied in more detail to shed light on the principles that govern mulitmodal interactions. An integrated approach of experimental and theoretical techniques is therefore essential to probe different aspects of protein-ligand binding in these multimodal systems. The work presented here demonstrates how Molecular Dynamics (MD) simulations can be employed to gain fundamental insights into the protein-ligand interactions at an atomistic level. In particular, the interaction of proteins with multimodal ligands was investigated.

Protein-ligand binding in multimodal systems involves a combination of electrostatic and hydrophobic interactions. The accurate measurement of individual contributions of both these modes to the overall binding behavior has been broken down into two steps. The first step requires the characterization of the molecules involved in binding in terms of their electrostatic and hydrophobic potential. While electrostatic interactions between two molecules can be easily calculated, characterizing hydrophobicity at a molecular level is very difficult. The advancements made in the later part of the last decade towards identifying molecular signatures of hydrophobicity and the development of better sampling techniques in simulations have enabled the hydrophobic characterization of biological macromolecules. These methods were used to map regions on proteins which are involved in hydrophobic association to ligands. Multimodal ligand-protein binding is further complicated by the fact that the cooperative effects of the multiple modes present on a ligand, i.e., the synergy present in multimodal interactions, has not yet been characterized. This constitutes the second step in understanding multimodal interactions, where MD simulations were performed with a series of increasingly complex ligands to identify the key contributors to synergistic interactions in multimodal ligand-protein interactions. Simulations were also performed to quantify the effect of fluid phase modifiers in modulating the different types of interactions in multimodal systems.

The results were used to develop a new technique for mapping synergistic regions on protein surface, which have the potential to associate with multimodal ligands, and for understanding avidity effects on resins to make predictions of protein retention behavior in multimodal systems. The change in protein retention times as a response to changes in fluid phase conditions (e.g. changes in temperature, cosolutes like urea) has been explained. The knowledge base created using this study can have a profound impact on the way we think about multimodal systems, especially in the rational design of ligands and choice of appropriate fluid phase conditions for achieving novel bioseparations.