Shifting the substrate scope of dimeric pyranose oxidase from monosaccharide to glycoside preference through oligomeric state modification

Pyranose oxidase (POx) and C-glycoside oxidase (CGOx) are FAD-dependent oxidoreductases belonging to the glucose-methanol-choline oxidoreductase superfamily and share the same sequence space. Despite a shared structural fold, these two members possess homologous domains that enable (arm and head domain) or disable (insertion-1 domain and barrel-shaped bottom) oligomerization. POxs with a higher oligomerization state (dimeric or tetrameric) exclusively catalyze the oxidation of monosaccharides (d-glucose, d-xylose). In contrast, the monomeric state of POxs/CGOxs is observed to prefer glycosides (homoorientin, phlorizin) and has low activity with free monosaccharides. We aimed to engineer dimeric POx from Kitasatospora aureofaciens (KaPOx) to form a functional monomer, and monomeric POx/CGOx from Streptomyces canus (ScPOx) to a dimeric structure. Deletion of the head and arm domains of the KaPOx subunit resulted in enzyme variants with a less hydrophobic surface, thus affecting its oligomerization. These monomeric KaPOx variants KaPOx_xal and KaPOx_xalh resembled monomeric wild-type POxs/CGOxs and preferred glycosides as substrates over monosaccharides with catalytic efficiencies for phlorizin being 24 × 10⁶ higher compared to those for d-xylose. The wild-type dimeric KaPOx showed no activity towards glycosides. We hypothesize that KaPOx_xalh is unable to react with monosaccharides because the introduced mutations alter the positions of monosaccharide-binding residues. The inability of KaPOx to react with glycosides is likely caused by steric hindrance and the inaccessibility of the active site to bulky glycosides due to dimerization. The attempt to engineer ScPOx into a dimeric structure failed at the stage of soluble expression, likely due to exposed hydrophobic patches and aggregation.