Enzyme-substrate matching in biocatalysis: in silico studies to predict substrate preference of ten putative ene-reductases from Mucor circinelloides MUT44

Abstract

Ene-reductases are flavoproteins able to catalyse the reduction of carbon-carbon double bonds with many potential applications in biocatalysis.The fungus Mucor circinelloides MUT44 has high ene-reductase activity when grown in the presence of substrates carrying different electron-withdrawing groups. Genome sequencing revealed the presence of ten putative genes coding for ene-reductases that can be potentially exploited for biocatalytic purposes. To this end, the availability of a method able to predict which isoform binds and turns over a specific substrate would help to choose the best catalyst for the desired bioconversion.Here, homology models of the ten putative enzymes are first generated, validated and show that the proteins share the typical TIM barrel fold with a conserved β-hairpin cap on one side of the barrel and a non-conserved subdomain capping the other side, where the FMN cofactor is accommodated. The active site of the ten enzymes is different in terms of both volume and charge distribution whereas the residues responsible for substrate recognition and catalysis are generally conserved.Docking of cyclohexenone into the active site of the ten enzymes shows a binding almost superimposable to that found in pentaerythritol tetranitrate reductase in complex with this substrate (PDB ID 1GVQ) in isoforms 1, 2, 6 and 10.The data demonstrate that in silico predictions can be used for new putative fungal ene-reductases to predict the best substrate-enzyme matching for the selection of the most suitable catalyst for the desired biotransformation.