Whole genome structural predictions reveal hidden diversity in putative oxidative enzymes of the lignocellulose-degrading ascomycete Parascedosporium putredinis NO1.

IMPORTANCE: An annotated reference genome has revealed P. putredinis NO1 as a useful resource for the identification of new lignocellulose-degrading enzymes for biorefining of woody plant biomass. Utilizing a "structure-omics"-based searching strategy, we identified new potentially lignocellulose-active sequences that would have been missed by traditional sequence searching methods. These new identifications, alongside the discovery of novel enzymatic functions from this underexplored lineage with the recent discovery of a new phenol oxidase that cleaves the main structural ß-O-4 linkage in lignin from P. putredinis NO1, highlight the underexplored and poorly represented family Microascaceae as a particularly interesting candidate worthy of further exploration toward the valorization of high value biorenewable products.

A bioinformatic workflow for in silico secretome prediction with the lignocellulose degrading ascomycete fungus Parascedosporium putredinis NO1.

The increasing availability of microbial genome sequences provides a reservoir of information for the identification of new microbial enzymes. Genes encoding proteins engaged in extracellular processes are of particular interest as these mediate the interactions microbes have with their environments. However, proteomic analysis of secretomes is challenging and often captures intracellular proteins released through cell death and lysis. Secretome prediction workflows from sequence data are commonly used to filter proteins identified through proteomics but are often simplified to a single step and are not evaluated bioinformatically for their effectiveness. Here, a workflow to predict a fungal secretome was designed and applied to the coding regions of the Parascedosporium putredinis NO1 genome. This ascomycete fungus is an exceptional lignocellulose degrader from which a new lignin-degrading enzyme has previously been identified. The 'secretome isolation' workflow is based on two strategies of localisation prediction and secretion prediction each utilising multiple available tools. The workflow produced three final secretomes with increasing levels of stringency. All three secretomes showed increases in functional annotations for extracellular processes and reductions in annotations for intracellular processes. Multiple sequences isolated as part of the secretome lacked any functional annotation and made exciting candidates for novel enzyme discovery.