(427b) Computational Predictions of Mutant Behavior of AraC | AIChE

(427b) Computational Predictions of Mutant Behavior of AraC


Berrondo, M. - Presenter, Johns Hopkins University
Gray, J. J. - Presenter, Johns Hopkins University
Schleif, R. - Presenter, Johns Hopkins University

In the past few years, computational methods have become increasingly successful in designing and redesigning proteins. Such success raises the question of whether the computational methods that have been developed to predict the folded structure of an amino acid sequence can be used for related structure problems, ones where related structural information is available, but where the relevant energy differences may be considerably less than 10 kcal/mol that are typical for the stabilities of folded proteins. We have developed and tested an algorithm to correctly predict the structure of the 17 N-terminal amino acids of the AraC gene regulatory protein when arabinose is bound to the protein and the different structure of this arm when arabinose is absent. Additionally, the transcriptional activity of 43 mutant arm sequences were measured in vivo and compared with predicted folding properties. Seventeen mutant sequences display arabinose response in vivo and can be compared directly against the prediction that computations to fold residues in the presence of arabinose will result in a low energy holo structure. Sixteen out of these seventeen mutants were correctly predicted to fold to the holo structure. The high success ratio shows that ?simple? computational methods like are now capable in some cases of explaining the behavior of mutant proteins with high reliability and have further implications for rational design of new proteins with similar behavior.