Oxygen reactivity of pyranose oxidoreductases

Reactions of flavoenzyme oxidoreductases consist of a reductive half-reaction, in which the substrate reduces the flavin, and an oxidative half-reaction, in which flavin is re-oxidized by an electron acceptor. Flavin-dependent oxidases use dioxygen as electron acceptor, producing hydrogen peroxide, whereas dehydrogenases and electron transferases react either very slowly or not at all with oxygen, and utilize alternative electron acceptors such as various (substituted) quinones or redox-proteins. Despite intensive research, the chemical and structural basis of oxygen reactivity remains largely undetermined, and specific binding sites for oxygen or alternative electron acceptors could be found. Several amino acid residues were shown to influence oxygen reactivity substantially in various proteins of different structural families, but no structural determinants governing whether an enzyme reacts as an oxidase or a dehydrogenase could be estabished. Pyranose 2-oxidase and pyranose dehydrogenase are two flavin-dependent sugar oxidoreductases from mainly white-rot and litter-decomposing basidiomycetes, respectively, which catalyze the regioselective oxidation of various aldopyranoses at the C-2 position (and sometimes at C-3). P2Ox is a homotetrameric protein and acts as an oxidase, PDH is a monomeric glycoprotein and is dependant on alternative acceptors such as 1,4-benzoquinone, the ferricenium ion and ferricyanide. Both belong to the GMC-family of flavin-dependant enzymes, and their active sites share a high degree of similarity. Genes encoding both enzymes were isolated in our laboratory, and heterologous expression was established. We propose to use an approach combining methods of directed evolution as well as semi-rational protein engineering, in order to change, and to identify amino acids influencing, the oxygen reactivity. Mini-libraries of enzyme variants will be produced by saturation mutagenesis of amino acids surrounding the active site, as well as libraries of randomly mutated p2ox- and pdh-genes, and screened in a parallel assay for changes in catalytic activity with oxygen and alternative electron acceptors. Mutations that show an effect will be biochemically characterized in detail. The similarity of the two enzymes active sites will allow comparisons of effects of certain amino acid exchanges, as well as the confirmation of these effects by introduction of identical and complementary alterations in both enzymes. Besides increasing basic knowledge about the reaction chemistry of enzymes from the GMC family of flavin-dependent enzymes, this research has some applicational relevance as well, as it would, if successful, allow to, e.g., eliminate the detrimental hydrogen peroxide production by P2Ox, or utilize the more flexible PDH in biocatalysis without redox mediator regenerations.

The project resulted in the identification of several amino acid residues in positions around and near the active site and the flavin cofactor of the carbohydrate oxidoreductase pyranose 2-oxidase (POx) that, when exchanged to different amino acids, significantly reduced the ability of the enzyme to activate molecular oxygen and act as an oxidase. The conversion of a very similar and structurally related enzyme, pyranose dehydrogenase (PDH), that cannot utilize oxygen as electron acceptor following the oxidation of a carbohydrate, to an oxidase was not successful, despite the exchange of amino acids in similar or identical positions to POx. A more substantial alteration of the enzyme structure, that was supposed to change the entire architecture of the enzymes active center, only resulted in unstable and non-functional enzyme variants. One amino acid was identified that allowed electron transfer to oxygen to a very minor degree. This amino acid connects the flavin cofactor covalently to the peptide chain of the protein. The cofactor continues to be linked to the enzyme, but in a non-covalent fashion, and the activity of the enzyme is reduced, but otherwise intact. Similar amino acid exchanges in other flavin-dependent oxidoreductases lead to both similar as well as divergent results, depending on the enzyme. The change in cofactor-binding reduced the redox potential of the cofactor and presumably is responsible for the ability of the enzyme to transfer electrons to oxygen. A detailed investigation revealed that the first half-reaction (oxidation of the sugar substrate) is slowed down to about a third in the variant, whereas the second one (oxidation of the cofactor by transferring the electrons to an electron acceptor) is sped up approximately two-fold. How this influences oxygen activation remains open for now. The example of pyranose 2-oxidase showed that dehydrogenase activity (ability to transfer electrons to acceptors other than oxygen) is a much more robust and resilient process that is not negatively affected by various amino acid exchanges, but that oxygen reactivity is based on a delicate balance of numerous factors, and can be negatively affected by amino acid exchanges comparably easily. A detailed investigation of the effects of the discovered amino acid exchanges on the two half-reactions of the enzyme as well as the potential of combinations of such exchanges will be the subject of further research. The conversion of PDH to an oxidase, which would be beneficial for applications in carbohydrate biocatalysis, remains an unsolved question. Applications of POx for biosensors or enzymatic biofuel cells, where electron transfer to oxygen and production of hydrogen peroxide is detrimental, may benefit from changes that abolish this property.