Metadynamics Study of a β-Hairpin Stability in Mixed Solvents

• Published in: Journal of the American Chemical Society 2011-03, Vol.133 (9), p.2897-2903
• Main Authors: Saladino, Giorgio, Pieraccini, Stefano, Rendine, Stefano, Recca, Teresa, Francescato, Pierangelo, Speranza, Giovanna, Sironi, Maurizio
• Format: Article
• Language: English
• Subjects: Betaine - chemistry; Molecular Dynamics Simulation; Peptides - chemistry; Protein Denaturation; Protein Folding; Protein Stability; Protein Structure, Secondary; Solvents - chemistry; Thermodynamics; Urea - chemistry
• ISSN: 0002-7863; 1520-5126
• DOI: 10.1021/ja105030m

Understanding the molecular mechanisms that allow some organisms to survive in extremely harsh conditions is an important achievement that might disclose a wide range of applications and that is constantly drawing the attention of many research fields. The high adaptability of these living creatures is related to the presence in their tissues of a high concentration of osmoprotectants, small organic, highly soluble molecules. Despite osmoprotectants having been known for a long time, a full disclosure of the machinery behind their activity is still lacking. Here we describe a computational approach that, taking advantage of the recently developed metadynamics technique, allows one to fully describe the free energy surface of a small β-hairpin peptide and how it is affected by an osmoprotectant, glycine betaine (GB) and for comparison by urea, a common denaturant. Simulations led to relevant thermodynamic information, including how the free energy difference of denaturation is affected by the two cosolvents; unlike urea, GB caused a considerable increase of the folded basin stability, which transposes into a higher melting temperature. NMR experiments confirmed the picture derived from the theoretical study. Further molecular dynamics simulations of selected conformations allowed investigation into deeper detail the role of GB in folded state protection. Simulations of the protein in GB solutions clearly showed an excess of osmoprotectant in the solvent bulk, rather than in the protein domain, confirming the exclusion from the protein surface, but also highlighted interesting features on its interactions, opening to new scenarios besides the classic “indirect mechanism” hypothesis.