Functional characterization of fungal lytic polysaccharide monooxygenases for cellulose surface oxidation

Abstract

Background: Microbial lytic polysaccharide monooxygenases (LPMOs) cleave diverse biomass polysaccharides, including cellulose and hemicelluloses, by initial oxidation at C1 or C4 of glycan chains. Within the Carbohydrate-Active Enzymes (CAZy) classification, Auxiliary Activity Family 9 (AA9) comprises the first and largest group of fungal LPMOs, which are often also found in tandem with non-catalytic carbohydrate-binding modules (CBMs). LPMOs originally attracted attention for their ability to potentiate complete biomass deconstruction to monosaccharides. More recently, LPMOs have been applied for selective surface modification of insoluble cellulose and chitin.

Results: To further explore the catalytic diversity of AA9 LPMOs, over 17,000 sequences were extracted from public databases, filtered, and used to construct a sequence similarity network (SSN) comprising 33 phylogenetically supported clusters. From these, 32 targets were produced successfully in the industrial filamentous fungus Aspergillus niger, 25 of which produced detectable LPMO activity. Detailed biochemical characterization of the eight most highly produced targets revealed individual C1, C4, and mixed C1/C4 regiospecificities of cellulose surface oxidation, different redox co-substrate preferences, and CBM targeting effects. Specifically, the presence of a CBM correlated with increased formation of soluble oxidized products and a more localized pattern of surface oxidation, as indicated by carbonyl-specific fluorescent labeling. On the other hand, LPMOs without native CBMs were associated with minimal release of soluble products and comparatively dispersed oxidation pattern.

Conclusions: This work provides insight into the structural and functional diversity of LPMOs, and highlights the need for further detailed characterization of individual enzymes to identify those best suited for cellulose saccharification versus surface functionalization toward biomaterials applications.

Fig. 1

Sequence relationship of 5,328 AA9 catalytic modules. A Sequence similarity network (SSN) created with SSNpipe [58] and displayed in Cytoscape with yFiles Organic layout (Shannon et al. 2003a). Each node corresponds to one of the curated 5328 catalytic modules used as an input to build the SSN. Edges represent an alignment bit-score threshold of 250 that clusters the sequences into subgroups. AA9 members whose regioselectivity is available are colored in blue, yellow and red for C1, C4 and C1/C4 oxidizing enzymes, respectively (functional annotations were obtained from the CAZy database [20]). Sequences expressed in Aspergillus niger are colored in black. B Maximum-likelihood phylogenetic tree. 10 representative sequences for each subgroup defined by the sequence similarity network in A. Bootstrap values based on 100 replicates are shown

Fig. 2

HPAEC-PAD analysis of LPMO regioselectivity towards three cellulosic substrates. Chromatograms over the retention times of 7.5–35 min. Native and C1-oxidized cello-oligosaccharide standard peaks are annotated. All activity assays were conducted using 5 µM of enzyme, 0.1% PASC (A), 1% Avicel (B) or 1% SA-Avicel (C) and 1 mM ascorbic acid for 16 h at 50 °C. The negative controls (substrate + electron donor) contained all assay components except enzyme

Fig. 3

Activity of 5 µM LPMOs after 16 h on PASC (0.1%), Avicel (1%), and SA-Avicel (1%) with 1 mM ascorbic acid as an electron donor. Semi-quantitative analysis of C1 and C4 oxidized products was completed by summing the peak areas of C1-oxidized and C4-oxidized oligosaccharides. Each bar is the average of three independent assays measured singly by HPAEC-PAD, with error bars indicating the standard error of the mean

Fig. 4

Time-course release of soluble oxidized products for A MfLPMO9A, B MsLPMO9B, C DcLPMO9A, and D MytLPMO9A. For each panel, 5 µM LPMOs were incubated at 50 °C for 24 h with 1 mM of either ascorbic acid (black lines), gallic acid (red lines) or cysteine (blue lines) with PASC (squares), Avicel (circles) or SA-Avicel (triangles). For each time point, T. reesei cellulase cocktail was used to convert all C1-oxidized products into cellobionic acid and quantified as the total C1-oxidized ends generated (µM). Total oxidized ends were obtained by quantifying cellobionic acid by HPAEC-PAD against a standard curve. Every point is the average of three independent assays measured individually by HPAEC-PAD, with error bars indicating the standard error of the mean

Fig. 5

Assessing LPMO (5 µM) activity on SA-Avicel (1%) fiber after 16 h with 1 mM ascorbic acid as an electron donor. T. reesei cellulase cocktail was used to convert all oxidized insoluble products into cellobionic acid and quantify the total C1-oxidized ends introduced to the fiber (nanomoles per mg of starting fiber). Total oxidized ends were obtained by quantifying cellobionic acid by HPAEC-PAD against a standard curve. Each bar is the average of three independent assays measured singly by HPAEC-PAD with error bars indicating the standard error of the mean

Fig. 6

Brightfield and confocal images of LPMO-treated SA-Avicel labeled using cyanine aminooxy dye (blue) for C4 products and rhodamine chloride (green) for C1 products. For each panel, 5 µM LPMOs were incubated at 50 °C for 24 h with 1% SA-Avicel and 1 mM of gallic acid. Insoluble products were separated, labeled with fluorescent dye, and visualized using a confocal microscope. Untreated substrates did not show fluorescence signals (Additional file 1: Figure S2)