Bioinformatics-Structural Approach to the Search for New D-Amino Acid Oxidases

Abstract

D-amino acid oxidase (DAAO, EC 1.2.1.2) plays an important role in the functioning of prokaryotes as well as of lower (yeast and fungi) and higher eukaryotes (mammals). DAAO genes have not yet been found in archaean genomes. D-amino acid oxidase is increasingly used in various fields, which requires the development of new variants of the enzyme with specific properties. However, even within one related group (bacteria, yeasts and fungi, mammals), DAAOs show very low homology between amino acid sequences. In particular, this fact is clearly observed in the case of DAAO from bacteria. The high variability in the primary structures of DAAO severely limits the search for new enzymes in known genomes. As a result, many (if not most) DAAO genes remain either unannotated or incorrectly annotated. We propose an approach that uses bioinformatic methods in combination with general 3D structure and active center structure analysis to confirm that the gene found encodes D-amino acid oxidase and to predict the possible type of its substrate specificity. Using a homology search, we obtained a set of candidate sequences, modelled the tertiary structure of the selected enzymes, and compared them with experimental and model structures of known DAAOs. The effectiveness of the proposed approach for discrimination of DAAOs and glycine oxidases is shown. Using this approach, new DAAO genes were found in the genomes of six strains of extremophilic bacteria, and for the first time in the world, one gene was identified in the genome of halophilic archaea. Preliminary experiments confirmed the predicted specificity of DAAO from Natronosporangium hydrolyticum ACPA39 with D-Leu and D-Phe.