Predicting protein stability changes upon mutation using database-derived potentials: solvent accessibility determines the importance of local versus non-local interactions along the sequence

Abstract

For 238 mutations of residues totally or partially buried in the protein core, we estimate the folding free energy changes upon mutation using database-derived potentials and correlate them with the experimentally measured ones. Several potentials are tested, representing different kinds of interactions. Local interactions along the chain are described by torsion potentials, based on propensities of amino acids to be associated with backbone torsion angle domains. Non-local interactions along the sequence are represented by distance potentials, derived from propensities of amino acid pairs or triplets to be at a given spatial distance. We ®nd that for the set of totally buried residues, the best performing potential is a combination of a distance potential and a torsion potential weighted by a factor of 0.4; it yields a correlation coef®cient between computed and measured changes in folding free energy of 0.80. For mutations of partially buried residues, the best potential is a combination of a torsion potential and a distance potential weighted by a factor of 0.7, and for the previously analysed mutations of solvent accessible residues, it is a torsion potential taken individually; the respective correlation coef®cients reach 0.82 and 0.87. These results show that distance potentials, dominated by hydrophobic interactions, represent best the main interactions stabilizing the protein core, whereas torsion potentials, describing local interactions along the chain, represent best the interactions at the protein surface. The prediction accuracy reached by the distance potentials is, however, lower than that of the torsion potentials. A possible reason for this is that distance potentials would not describe correctly the effect on protein stability due to cavity formation upon mutating a large into a small amino acid. Last but not least, our results indicate that although local interactions, responsible for secondary structure formation, do not dominate in the protein core, they are not negligible for all that. They have a signi®cant weight in the delicate balance between all the interactions that ensure protein stability.# 1997 Academic Press Limited

Detection of High-Frequency and Novel DNMT3A Mutations in Acute Myeloid Leukemia by High-Resolution Melting Curve Analysis

“…Solvent accessibility also showed considerable changes in 7 of 8 mutants, ranging from 3.57% to 32.85% with no change induced by the C583Y mutant (see Supplemental Figure S1B at http:// jmd.amjpathol.org). Solvent accessibility is an important parameter for determining local versus the nonlocal interactions along the sequence of proteins 31 and the changes measured for the DNMT3A mutants indicate considerable alterations of interresidue interactions. Furthermore, using the Swiss-PDB V 4.0.4 viewer, we could visualize the alterations of local structural aspects and hydrogen bonding induced by mutations, which could explain changes in stability predicted for the mutants (see Supplemental Figure S1B at http://jmd.amjpathol.org).…”

Bioinformatics Approaches to the Functional Profiling of Genetic Variants

“…In traditional molecular dynamics simulation, potential functions from a force field were usually calculated to obtain ∆∆G, which was mainly influenced by interactions between nonlocal amino acids [22]. Although it is generally difficult, if not completely impossible, to infer protein structural architecture accurately based solely on amino acid sequence, pioneering work from [23,24] showed that protein sequence was effective in the prediction of secondary structure and solvent accessibility.…”

Protein folding: Then and now

“…Gilis and Rooman first applied database-derived backbone dihedral potentials to study the change of thermodynamic stability upon point mutations [97][98][99]. They found that torsion-angle potentials predict ΔΔG accurately for mutations of solvent-exposed residues and that distance-dependent statistical potentials are more accurate for predicting the ΔΔG of buried residues.…”