Side Chain Entropy in Enzymes and its Role in Catalysis

Abstract

Directed Evolution is a popular biochemical engineering technique used to improve, and in some cases alter the activity of enzymes. Although many empirical strategies have been developed to carry out directed evolution, it is still an opaque process that could benefit from more rational design. In this work, we try to rationalize directed evolution based on the physical principles of thermodynamics. Our model system is the Kemp Eliminase KE07, which was designed computationally to catalyze the conversion of 5-nitrobenzisoxazole to cyanophenol, but showed poor activity in solution (kcat/KM ∼12 M-1s-1). Seven rounds of directed evolution lead to a 2 order of magnitude improvement in kcat/KM but structural analysis could not explain why many of the mutations were made. Using a sequential Monte Carlo technique we have been able to calculate and identify systematic changes in the side chain entropy through the 7 rounds of directed evolution. Our technique uses a realistic energy function coupled with a rotamer library to estimate the important rotameric states. These results suggest that the functional partitioning of structure (enthalpy) and statistical fluctuations (entropy) that occurs at the side chain level could give us important leads to predict further mutations. This work also provides further evidence to start looking beyond the traditional structure-function paradigm and incorporate entropic contributions for designing/improving enzymes.