We have shown that OxdC requires manganese and dioxygen for activity. Oxalate decarboxylase and oxalate oxidase share not only the same substrate and manganese cofactor, but also the key manganese-binding residues of cupin sequence motifs. Dioxygen is a substrate for the oxidase. By contrast, it is a unique cofactor for the decarboxylase, despite the overall reaction involving no net redox change. We were the first to propose a divergent free radical mechanisms to account for these observations (see below left for the current proposal). We have developed the first real-time assay for this enzyme that detects both the substrate and the products simultaneously allowing mechanistic studies. Crystal structures of the decarboxylase and the plant oxidase have been available for some time, but it was not completely clear what differences in the active sites of these enzymes determined the chemical fate of the cleavage of the carbon-carbon bond of oxalate. It had been suggested that the C-terminal domain manganese-binding site was the catalytically active site. However, we solved an alternative oxalate decarboxylase structure with David Lawson, which has helped identify the N-terminal domain manganese-binding site to be the main catalytic site and Glu-162 to be the crucial general acid that is required for the decarboxylase reaction. This residue forms part of a lid that closes over the active site (see below right). Mutagenesis and additional structural studies have supported these findings and highlighted the importance of lid flexibility in catalysis. Furthermore, guided by sequence comparisons with a novel fungal oxalate oxidase, we have demonstrated the remarkable interconversion of decarboxylase and oxidase activities. Thus, reaction specificity is controlled by the presence/absence of the general acid Glu-162 together with the saturation behaviour towards dioxygen, which is dictated by the peptide sequence of the lid. Moreover, we have interconverted these activities with a single amino acid change, by-passing the need for a promiscuous evolutionary intermediate enzyme (see below middle).

Fungal Oxalate Oxidase and Lignin Degradation

Ceriporiopsis subvermispora is a lignin-degrading fungus with applications in the paper pulping industry. It utilises the hydrogen peroxide-requiring enzyme manganeses peroxidase to degrade lignin. The most likely source of hydrogen peroxide appears to be oxalate oxidase which converts oxalate and dioxygen to hydrogen peroxide and carbon dioxide. We have cloned and sequenced two allelic isoforms of oxalate oxidase from this fungus in collaboration with Raphael Vicuña (Santiago, Chile). Surprisingly, their sequences resemble those of the bacterial bicupin oxalate decarboxylases rather than the plant monocupin oxalate oxidases (also known as germins). The fungal oxidases therefore form a new class of oxalate oxidases. Sequence alignments between the bicupin fungal oxidases and the bacterial decarboxylases together with molecular modelling have provided independent evidence for the identity of the active sites of these enzymes and the importance of a crucial Glu residue in the decarboxylase reaction. These observations have guided the interconversion of decarboxylase and oxidase activities as described above.

Plant Oxalate Oxidase and Disease Resistance

Oxalate oxidase (also known as germin) is expressed in cereal crops in response to a number of biotic and abiotic stresses including the powdery mildew infection of barley. Like oxalate decarboxylase, this enzyme is used in the clinical detection of oxalate and its expression in crop plants confers fungal disease resistance. Our new found understanding of the catalytic mechanism will allow the design of new chemical catalysts for the production of hydrogen peroxide.

We have shown that the barley enzyme does not contain copper, flavin or iron, which are normally utilised as cofactors by oxidases that convert dioxygen directly to hydrogen peroxide. We were the first to identify the presence of mononuclear manganese(II). This may explain the link between the severity of take-all fungal infection of wheat and low manganese content of seed and soil.

Adjacent is a model of the barley oxalate oxidase active site, derived from a homology model. The manganese(II) ion is ligated by the highly conserved histidines and glutamate of the cupin sequence motifs. The two water molecules indicate where the two substrates could bind. This published prediction was later shown to be correct by RW Pickersgill et al. We have also cloned a similar but distinct fungal oxalate oxidase as described above.