Decoupling enzyme catalysis from thermal denaturation

Abstract

The equilibrium model (EM) (Daniel et al., 2001) postulates two forms of a folded enzyme, one catalytically active (Eact) and the other inactive (Einact), which interconvert via a fast thermal equilibrium (Keq) (Figure A). This model for enzyme catalysis accounts for experimentally observed time and temperature profiles of enzyme/substrate systems more accurately than the classically derived, single folded-species, model (Figure B). In both models, the denatured species (X) is formed via the kinact process, which is temperature and time-dependent. (FIgure A) (Figure B) Comparison between the equilibrium model and classical model for enzyme catalysis. For both, the vertical axis is catalytic rate (M s-1), the left-right axis is increasing temperature (K) and the back-front axis is assay duration (s). The physical basis for the Eact/Einact equilibrium is unknown. To study the equilibrium, the temperature midpoint of the Eact/Einact transition (Teq) has to be separated from the thermostability of the enzyme (Tm) to allow the Einact species to exist in measurable concentrations without exhibiting denaturation. Mutations were made in a well-studied and NMR-accessible ribonuclease, barnase, to alter the thermostability and/or the Teq of the enzyme activity. The stability properties of each mutant were measured and the activity against two substrates assayed. New models were derived and fitted against wild-type barnase, and an ideal data set, to give insight into alternative irreversible and reversible denaturation pathways. Simulations of these models were developed to benchmark potential dynamics work and explain the movements of species within each model's framework. Assay data fits to the EM and alternative models show a preference for irreversible denaturation pathways via the Einact species. A mathematically simplified model was also found that accounts for data and could provide an alternative method for determining EM parameters. Although fits of barnase to the EM were statistically good, the denaturation properties could not be reconciled with the literature or experimentally determined values for stability and unfolding. Simulations illustrating how the Eact, Einact and denatured (X) species interact also corroborate this finding. Despite this discrepancy (in fitted parameters to the EM), it is hypothesised that the Teq and Tm of a disulphide-bridged mutant of barnase have been successfully decoupled. This mutant has been 15N- labelled for future NMR dynamics measurements. New approaches to the EM model are proposed where the separate determination of enzyme thermodynamic properties (e.g., rate and free energy of denaturation) would allow other EM parameters to be fitted independently to each data set.