Designing allosteric control into enzymes by chemical rescue of structure

Abstract

In recent years chemical biology has been used to engineer ligand-dependent activity into enzymes to control a variety of signaling pathways, such as modulating protein production, degradation, and localization. These successes have generally incorporated a naturally allosteric domain to an enzyme of interest. Here, we demonstrate a novel approach for designing de novo allosteric effector sites directly into an enzyme. Our approach diverges from traditional chemical rescue in that it does not rely on the disruption and restoration of active site chemistry as a means to control enzyme function, but rather on the disruption and restoration of structure. We present two examples, W33G in a beta-glycosidase enzyme (beta-gly) and W492G in a beta-glucuronidase enzyme (beta-gluc), in which we engineer indole-dependent activity into enzymes by removing a buried tryptophan sidechain that serves as a buttress for the active site architecture, which results in a loss of activity for both enzymes. In both cases we demonstrate that subsequent addition of exogenous indole restores catalytic function. We demonstrate through analysis of enzyme kinetics that the rescued beta-gly W33G enzyme is fully functionally equivalent to that of the wild type enzyme. We establish the structural basis for inactivation and rescue by presenting the apo and indole-bound crystal structures of beta-gly W33G. Finally, we use this rationally designed switch to control beta-glycosidase activity in living cells with the use of exogenous indole. Chemical rescue of protein structure may represent a general approach for designing allosteric control into enzymes, and thus may serve as a starting point for building a variety of bioswitches and sensors.