Simulating Protein Adsorption for Experimental Comparison

Abstract

Many biological processes and technological applications involve proteins coming into contact with a solid surface. Generally, we know that proteins experience some degree of conformational change at the solid/liquid interface, and can measure these changes in the lab. However, while many experimental techniques exist for characterizing surface-bound proteins, none have been able to resolve high-precision structures. Computer simulation offers a unique route to determining how proteins adsorb. Herein, we apply a popular statistical sampling technique - Parallel Tempering Metadynamics - to all-atom molecular dynamics simulations of ex- plicitly solvated proteins interacting with solid surfaces. We show that by biasing specific degrees of freedom - or collective variables - a protein can be influenced to exhaustively explore conformational space both on and off a surface. The results from these simulations can be post-processed to reveal details such as: surface- bound conformations, orientations, and finer structural details like interatomic distances and Ramachandran angles - which, in turn, can be compared to, and validated by, experimental measurements. Ultimately, this work should convey the descriptive power that can arise from a mutually beneficial partnership between surface science and computer simulation in the context of biomolecular adsorption.