Engineering potassium activation into biosynthetic thiolase

Activation of enzymes by monovalent cations (M⁺) is a widespread phenomenon in biology. Despite this, there are few structure-based studies describing the underlying molecular details. Thiolases are a ubiquitous and highly conserved family of enzymes containing both K⁺-activated and K⁺-independent members. Guided by structures of naturally occurring K⁺-activated thiolases, we have used a structure-based approach to engineer K⁺-activation into a K⁺-independent thiolase. To our knowledge, this is the first demonstration of engineering K⁺-activation into an enzyme, showing the malleability of proteins to accommodate M⁺ ions as allosteric regulators. We show that a few protein structural features encode K⁺-activation in this class of enzyme. Specifically, two residues near the substrate-binding site are sufficient for K⁺-activation: A tyrosine residue is required to complete the K⁺ coordination sphere, and a glutamate residue provides a compensating charge for the bound K⁺ ion. Further to these, a distal residue is important for positioning a K⁺-coordinating water molecule that forms a direct hydrogen bond to the substrate. The stability of a cation-π interaction between a positively charged residue and the substrate is determined by the conformation of the loop surrounding the substrate-binding site. Our results suggest that this cation-π interaction effectively overrides K⁺-activation, and is, therefore, destabilised in K⁺-activated thiolases. Evolutionary conservation of these amino acids provides a promising signature sequence for predicting K⁺-activation in thiolases. Together, our structural, biochemical and bioinformatic work provide important mechanistic insights into how enzymes can be allosterically activated by M⁺ ions.