The challenge of rationally predicting distal activity-enhancing mutations in computational enzyme design

Abstract

Enzymes exist as an ensemble of conformational states, whose populations can be shifted by substrate binding, allosteric interactions, but also by introducing mutations to their sequence. Tuning the populations of the enzyme conformational states through mutation enables evolution towards novel activity.[1] A common feature observed in many laboratory-evolved enzymes, is the introduction of remote mutations from the catalytic center, which often have a profound effect in the enzyme catalytic activity. [2] As it happens in allosterically regulated enzymes, distal mutations regulate the enzyme activity by stabilizing pre-existing catalytically important conformational states. In this talk, our new computational tools based on inter-residue correlations from microsecond time-scale Molecular Dynamics (MD) simulations and enhanced sampling techniques are applied in Tryptophan synthase (TrpS) complex. TrpS is composed of TrpA and TrpB subunits, which allosterically activate each other and have no activity when isolated. [3,4] We show how distal mutations introduced in TrpS resuscitate the allosterically-driven conformational regulation and alter the populations and rates of exchange between multiple conformational states, which are essential for the multistep reaction pathway of the enzyme.[3] The exploration of the conformational landscape of TrpS is key for identifying conformationally-relevant amino acid residues of TrpB and TrpA distal from the active site.[4] We predict positions crucial for shifting the inefficient conformational ensemble of the isolated TrpB and TrpA to a productive ensemble through intra-subunit allosteric effects. The experimental validation of the new conformationally-driven TrpB and TrpA design demonstrates their superior stand-alone activity in the absence of binding partner, comparable to those enhancements obtained after multiple rounds of experimental laboratory evolution. Our work evidence that the current challenge of distal active site prediction for enhanced function in computational enzyme design can be ultimately addressed.