Peptide Recognition Capabilities of Cellulose in Molecular Dynamics Simulations

Knowing the amino acid discriminating capacity of the surface of cellulose can be valuable information toward the rational design and re-engineering of the cellulosome in order to improve its catalytic properties. This aim can be achieved by the determination of the binding free energies of amino acids in molecular dynamics simulations. Conventional simulations do not always allow for sufficient sampling of the configuration space, especially in a system where large energy barriers occur. A better sampling can be obtained by replica exchange molecular dynamics (REMD). Here, we use REMD combined with umbrella sampling to determine the potential of the mean force for amino acids, analogues of their side chains, dipeptides, and tripeptides for the interactions with the (100) face of the crystalline cellulose I beta. We also use REMD to characterize the adsorption dynamics of a small protein, tryptophan cage. Our results show that all 20 standard amino acids adsorb on the surface of cellulose with binding energies ranging from 3 to 10 kJ/mol with the specificity of approximately 30%. The largest affinity to cellulose is shown to belong to the aromatic residues. On the basis of simulations of selected dipeptides and tripeptides, we determine that cellulose-adsorption energies should be scaled down to perhaps about 20% to correctly describe adsorption of protein residues as opposed to free amino acid molecules. We notice that the rough and dynamic surface of cellulose favors adsorption of flexible parts of the protein, in contrast to the specificity expressed by flat and static surfaces of nonorganic solids. We show that the contact angle for a droplet of water on the (100) face of the cellulose is about 24, reflecting the heterogeneous hydrophilicity of the surface.