Electrostatic Engineering of Phosphoketolase Enhances Activity on Small Non-phosphorylated Sugars and Improves Cell-Free ATP Regeneration from Inexpensive C 2 -Substrates

Abstract Phosphoketolases can be used to convert non-phosphorylated sugars to the high energy compound acetyl phosphate and the versatile metabolic precursor acetyl-CoA. The performance of these pathways is limited by low catalytic activity of natural phosphoketolases towards these sugars. Here, we report the rational engineering of the phosphoketolase from Bifidobacterium adolescentis (Bad.F6Pkt) to enhance its activity and affinity towards glycoaldehyde (GA) and D-erythrulose (ERU) through re-organisation of the protein electric field to reproduce the role of terminal phosphate groups in cognate substrates. Guided by predicted induced side-chain pKa shifts, visualisation of electrostatic potential difference maps alongside molecular modelling and sequence variation analyses, we identified mutations that could promote in situ ring opening of the pre-dominant cyclic GA dimer form in solution. This approach to the electrostatic inverse design problem yielded the GA-specific double mutant H142N:E153D, exhibiting a ten-fold improved affinity and slightly enhanced catalytic efficiency (KM = 4.4 mM, kcat/ KM = 26.3 s-1 M-1) compared to the previously reported H142N variant (KM = 42.3 mM, kcat/ KM = 20.6 s-1 M-1). We additionally constructed a H256Y:H260Y:H548Y variant comprising long-range electrostatic mutations with a 3.8-fold increased catalytic efficiency (kcat/ KM = 49.6 s-1 M-1) on the acylic four-carbon ERU ketose compared to the wild-type enzyme. The engineered enzymes were evaluated in cell-free enzyme cascades for ATP regeneration via acetyl phosphate formation. The H142N variant enabled efficient ATP regeneration from GA and ethylene glycol, whereas the H142N:E153D mutant exhibited reduced stability under synthesis conditions. Furthermore, coupling of a highly GA-specific D-threose aldolase and a D-threose isomerase with the PKT triple mutant enabled rapid conversion of GA into C4 sugar intermediates and significantly improved ATP regeneration from GA. Competing Interest Statement The authors have declared no competing interest.