MODELING PROTEIN STABILITY - A THEORETICAL-ANALYSIS OF THE STABILITY OF T4 LYSOZYME MUTANTS

Citation
Dl. Veenstra et Pa. Kollman, MODELING PROTEIN STABILITY - A THEORETICAL-ANALYSIS OF THE STABILITY OF T4 LYSOZYME MUTANTS, Protein engineering, 10(7), 1997, pp. 789-807
Citations number
71
Categorie Soggetti
Biology
Journal title
ISSN journal
02692139
Volume
10
Issue
7
Year of publication
1997
Pages
789 - 807
Database
ISI
SICI code
0269-2139(1997)10:7<789:MPS-AT>2.0.ZU;2-7
Abstract
Free energy calculations were conducted to determine the relative stab ility of the unnatural amino acid mutants of T4 lysozyme norvaline (Nv l) and O-methyl-serine (Mse) and of alanine at residue 133, which is l eucine in the native sequence. These calculations were performed both to assess the validity of the methodology and to gain a better underst anding of the forces which contribute to protein stability, Peptides o f different length were used to model the denatured state. Restraints were employed to force sampling of the side chain;chi(1) dihedral of t he perturbed side chain, and the effect of protein repacking in respon se to mutation was studied through the use of different constraint set s. In addition, the convergence behavior and hysteresis of the simulat ions in the folded and unfolded states were determined. The calculated results agree well with experiment, +1.84 versus +1.56 kcal/mol for M se-->Nvl and -3.48 versus -2.2 to -3.6 kcal/mol for Nvl-->Ala. We find that free energy calculations can provide useful insights to protein stability when conducted carefully on a well chosen system. Our result s suggest that loss of packing interactions in the native state is a m ajor source of destabilization for mutants which decrease the amount o f buried nonpolar surface area and that subtle responses of the backbo ne affect the magnitude of the loss of stability. We show that the con formational freedom of the chi(1) dihedral has a noticeable effect on protein stability and that the solvation of amino acid side chains is strongly influenced by interactions with the peptide backbone.