Protein adsorption is a ubiquitous phenomenon that plays a critical role in many processes and thus in the design and development of new systems and applications. However, gaining insight into the complexities of protein adsorption, such as binding affinity and surface-induced orientation and conformational change, are difficult from both an experimental and computational perspective. For example, there are currently few, if any, experimental techniques that can completely resolve the structure of an adsorbed protein. In the same fashion, computational prediction of adsorbed structures has typically been impossible without the inclusion of some amount of experimental data to guide simulations in the right direction. Here we show that in a bottom-up fashion enhanced sampling computational methods based on the metadynamics family can be used to probe the equilibrium structure and preferred orientation of peptides on surfaces. This talk will highlight simulations of peptide/protein adsorption with explicitly solvated molecular dynamics in three different contexts: 1) adsorption of a 14-residue peptide fragment inspired by the carbohydrate binding module (CBM) of a cellulase to cellulose, with relevance to biomass processing, 2) adsorption of the 15-residue binding domain of the salivary protein statherin (SN15) to hydroxyapatite, with relevance to the biomineralization process, and 3) adsorption to silica of a 12-residue peptide linker (Car9) alone and attached to green fluorescent protein (GFP), for the design of bioinspired or hierarchical nanostructures. In the first study, simulations confirm recent experimental work that shows CBM aligns to cellulose fibers in a specific orientation and predict the bound equilibrium structure to resemble a loose β-hairpin with strong surface-hydrogen bonding capabilities. In the second case, calculation of 22 structural observables from molecular simulations of SN15 adsorbed to hydroxyapatite show strong agreement to experimental measurements. Lastly, binding free energies calculated from simulations of Car9 on silica agree within 2 kcal/mol. Simulations also show the number and location of Car9 peptides attached to GFP influences the silica-adsorbed structures of Car9. The talk will conclude with broader perspectives on the combined use of experiments and simulations to control biomolecular structure for new and existing applications.

Predicting the structure and orientation of adsorbed peptides with molecular simulations

253rd National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.