MONTE-CARLO STUDIES OF THE THERMODYNAMICS AND KINETICS OF REDUCED PROTEIN MODELS - APPLICATION TO SMALL HELICAL, BETA-PROTEIN, AND ALPHA BETA-PROTEIN/

Citation
A. Kolinski et al., MONTE-CARLO STUDIES OF THE THERMODYNAMICS AND KINETICS OF REDUCED PROTEIN MODELS - APPLICATION TO SMALL HELICAL, BETA-PROTEIN, AND ALPHA BETA-PROTEIN/, The Journal of chemical physics, 108(6), 1998, pp. 2608-2617
Citations number
75
Categorie Soggetti
Physics, Atomic, Molecular & Chemical
ISSN journal
00219606
Volume
108
Issue
6
Year of publication
1998
Pages
2608 - 2617
Database
ISI
SICI code
0021-9606(1998)108:6<2608:MSOTTA>2.0.ZU;2-M
Abstract
Employing a high coordination lattice model and conformational samplin g based on dynamic and entropy sampling Monte Carlo protocols, compute r experiments were performed on three small globular proteins, each re presenting one of the three secondary structure classes. The goal was to explore the thermodynamic character of the conformational transitio n and possible mechanisms of topology assembly. Depending on the stabi lity of isolated elements of secondary structure, topology assembly ca n proceed by various mechanisms. For the three-helix bundle, protein A , which exhibits substantial helix content in the denatured state, a d iffusion-collision mechanism of topology assembly dominates, and here, the conformational transition is predicted to be continuous. In contr ast, a model beta protein, which possesses little intrinsic denatured state secondary structure, exhibits a sequential ''on-site'' assembly mechanism and a conformational transition that is well described by a two-state model. Augmenting the cooperativity of tertiary interactions led to a slight shift toward the diffusion-collision model of assembl y. Finally, simulations of the folding of the alpha/beta protein G, wh ile only partially successful, suggest that the C-terminal beta hairpi n should be an early folding conformation and that the N-terminal beta hairpin is considerably less stable in isolation. Implications of the se results for our general understanding of the process of protein fol ding and their utility for de novo structure prediction are briefly di scussed. (C) 1998 American Institute of Physics.