Cellulose surface degradation by a lytic polysaccharide monooxygenase and its effect on cellulase hydrolytic efficiency

Abstract

Lytic polysaccharide monooxygenase (LPMO) represents a unique principle of oxidative degradation of recalcitrant insoluble polysaccharides. Used in combination with hydrolytic enzymes, LPMO appears to constitute a significant factor of the efficiency of enzymatic biomass depolymerization. LPMO activity on different cellulose substrates has been shown from the slow release of oxidized oligosaccharides into solution, but an immediate and direct demonstration of the enzyme action on the cellulose surface is lacking. Specificity of LPMO for degrading ordered crystalline and unordered amorphous cellulose material of the substrate surface is also unknown. We show by fluorescence dye adsorption analyzed with confocal laser scanning microscopy that a LPMO (from Neurospora crassa) introduces carboxyl groups primarily in surface-exposed crystalline areas of the cellulosic substrate. Using time-resolved in situ atomic force microscopy we further demonstrate that cellulose nano-fibrils exposed on the surface are degraded into shorter and thinner insoluble fragments. Also using atomic force microscopy, we show that prior action of LPMO enables cellulases to attack otherwise highly resistant crystalline substrate areas and that it promotes an overall faster and more complete surface degradation. Overall, this study reveals key characteristics of LPMO action on the cellulose surface and suggests the effects of substrate morphology on the synergy between LPMO and hydrolytic enzymes in cellulose depolymerization.

Keywords: Atomic Force Microscopy (AFM); Biofuel; Cellulase; Cellulose; Copper Monooxygenase; GH61-AA9; Lytic Polysaccharide Monooxygenase (LPMO); Oxidative Cellulose Surface Degradation; Synergy.

FIGURE 1.

a, hydrolysis of crystalline-ordered and amorphous-unordered (brown) cellulose is catalyzed by a typical set of fungal cellulases, including chain end-cleaving cellobiohydrolases (CBH I, red; CBH II, green) and internally chain-cleaving endoglucanase (brown). Linear cellulose chains are represented by their cellobiosyl-dimer units, with the reducing end shown in red. Cellobiose is the main soluble sugar produced by the cellulases. b, oxidative O₂-dependent attack of LPMO on the cellulose surface, resulting in internal chain cleavages and release of C1′ or C4 oxidized oligosaccharides harboring one oxygen atom from the O₂ oxidant (indicated in blue) (18–20). C1′ oxidation produces a 1,5-lactone product that hydrolyzes rapidly to give a d-gluconic acid moiety, which is shown. LPMO contains surface-exposed catalytic copper (brown sphere) that is required for activity (12, 14, 16). The structure of GH61D (PDB code 4B5Q) from Phanerochaete chrysosporium is used for depiction (21). Electrons required in the LPMO reaction in vitro can be delivered from the flavo-heme protein cellobiose dehydrogenase or from a small-molecule reductant such as l-ascorbic acid (12, 14).

FIGURE 2.

LPMO is specific for oxidizing crystalline cellulose surfaces and does so by introducing carboxylic acid groups. a, visual appearance of the mixed amorphous-crystalline cellulose substrate prepared from Avicel PH-101 using partial dissolution in and regeneration from ionic liquid. b, AFM height image of the substrate surface in a ultramicrotome-cut cellulose sample. c, height distribution analysis reveals the nano-flat character of the substrate surface. d, maximum intensity projection of the recorded transmission microscopic images of a substrate area where a large cellulose crystallite is embedded in an amorphous matrix. The crystallite is a remnant from the original Avicel PH-101 not dissolved during ionic liquid treatment. It is composed of cellulose allomorph I, as shown by XRD (33). e, maximum intensity projection of the recorded fluorescence signal from the same substrate area after incubation with LPMO (9 mg g⁻¹; supplemented with 7.5 μm l-ascorbic acid; 25 °C; 12 h) and subsequent staining for 12 h with 5 μm SYTO-62, a small molecule fluorescent probe for carboxyl groups (38). The figure shows that crystalline surfaces are preferentially labeled with the fluorescent dye after substrate incubation in the presence of LPMO. This suggests preferred attack of LPMO on these surfaces. f, transmission microscope images (left panel) with the corresponding fluorescence images (right panel) from two focal planes separated by a vertical distance of 4 μm. Focused specimen parts, which are recognized clearly by detailed structures in the transmission images (indicated with green arrows), correspond to the surface of the cellulose crystallite. CLSM images reveal that the SYTO-62 fluorescence signal is confined exclusively to crystalline parts in focus over different planes (yellow arrow in image in plane 15) and that the fluorescence signal is not present at interior parts of the crystallite (yellow arrow in image in plane 10). Therefore, these images show that LPMO action is restricted to the surface of the cellulose crystallite. Suitable controls showed that untreated cellulose crystals or cellulose crystals treated with LPMO in the absence of l-ascorbic acid did not become fluorescent when stained with SYTO-62. Fluorescence images in (e and f) are background-subtracted and contrast-enriched. Scale bars, 10 μm.

FIGURE 3.

AFM imaging of LPMO action on the cellulose surface and its effect on CBH I activity. a, three-dimensional surface representation overlaid with phase information to show the attack of LPMO (9 mg g⁻¹ substrate; 20 °C; 12 h) on a cellulose fibril (blue outline) in the middle and from the top, thus causing a large amount of internal degradation. b, phase images showing fibril degradation by LPMO through major thinning from the sides, as indicated by green arrows. AFM images show substrate before and after 12 h of incubation with LPMO supplemented with 7.5 μm l-ascorbic acid as reductant. Amorphous material is visualized by a dark color in phase or phase-overlaid images. It was resistant against degradation by LPMO. Scale bar, 30 nm. c, time-resolved AFM sequences (height images on top, phase images on bottom) showing surface degradation by CBH I on a substrate preincubated with LPMO. Gradual degradation of a crystalline surface feature is observed as a result of thinning from the top and from the sides, as indicated by green arrows, eventually leading to complete dissolution of a large set of fibril bundles. Fiber orientation on the crystalline surface is demonstrated by 2D Fast Fourier Transformation analysis (embedded in Gwyddion 2.31) of the highlighted rectangle. Deviation from a circular symmetric to an ellipsoid spectrum (white dotted envelope) shows the fiber orientation by a 90° rotation (white bold arrows). Scale bar, 300 nm.

FIGURE 4.

Synergy between LPMO and CBH I during degradation of different cellulosic substrates determined in sequential (a) and simultaneous (b) assays. The substrates used were MACS, nanocrystalline cellulose (NCC), and Avicel PH-101. For the sequential assay (panel a), LPMO (9 μg ml⁻¹) was incubated with either cellulose preparation (1 mg ml⁻¹) in 50 mm sodium phosphate buffer, pH 6.0, for 5 h at 25 °C. l-Ascorbic acid (7.5 μm) was present as the reductant. The total volume was 500 μl, and incubations were done in static Eppendorf tubes. CBH I (100 μg ml⁻¹) supplemented with β-glucosidase (5 μg ml⁻¹) was added to the reaction mixture, and incubation was continued at 50 °C for 1 h. The simultaneous assay (panel b) used identical conditions except that all reactants were already present at reaction start. Temperature was 50 °C, and reaction time was 96 h. The d-glucose concentration was measured with an enzymatic assay. It was confirmed using analysis with high performance anion exchange chromatography with pulsed amperometric detection that d-glucose accounted for ≥99% of the total soluble products present in the supernatant under these conditions. The measured data are plotted as ratio of sugar formed in reactions containing or lacking LPMO. A value exceeding unity, therefore, indicates a synergistic effect. A value below unity indicates that preincubation with LPMO (panel a) or the presence of LPMO (panel b) results in apparent inhibition of the d-glucose release by CBH I. Results are shown in bars with S.D. from the three independent experiments indicated.

FIGURE 5.

Effect of substrate pretreatment with LPMO on activity of the complete T. reesei cellulase for degrading mixed amorphous-crystalline cellulose analyzed in supernatant (a) and directly on the cellulose surface (b and c). a, LPMO (9 mg g⁻¹) was incubated with either cellulose preparation (1 mg ml⁻¹) in 50 mm sodium phosphate buffer, pH 6.0, for 12 h at 25 °C. l-Ascorbic acid (7.5 μm) was present as the reductant. Cellulase (25 mg g⁻¹) supplemented with β-glucosidase (5 mg g⁻¹) was added to the reaction mixture, and incubation was continued at 50 °C for 1 h. The d-glucose concentration was measured enzymatically. It was confirmed using analysis with HPAEC-PAD that d-glucose accounted for ≥99% of the total soluble products present in the supernatant under these conditions. Percent conversion is calculated from the anhydroglucose release. Symbols show measured data, and error bars show S.D. from three independent experiments. Pretreatment of substrate by LPMO boost the cellulase activity by a factor of ∼2. b and c, experiments in the AFM liquid cell were performed in exactly the same way as described above, except that the temperature of the hydrolysis reaction was 20 °C. Panel c shows AFM height images that depict a representative surface area of the substrate after LPMO treatment immediately before the addition of cellulase (left image) and after 5 h of hydrolysis reaction (right image). A large cellulose crystallite (bright color) is seen embedded in amorphous material (darker color). A blue circle is used to indicate part of the cellulose surface where structural disruptions occur within highly crystalline material. Comparison of the left and right image reveals that cellulase activity results in strong volume degradation in regions of amorphous cellulose, causing the overall surface to become completely rugged with time. Additionally, there is clear activity in the highlighted crystalline region of the substrate where fissures become larger and generally more distinct with time. It is important to emphasize that activity in the highly crystalline areas has only been observed when substrate was pretreated with LPMO. Panel b shows a quantitative analysis of time-resolved AFM sequences that were used to measure vertical surface degradation by the cellulases. Slowly degraded crystallites such as the one seen in panel c were used as reference points. Pretreatment of substrate with LPMO enhanced surface degradation by the cellulases by a factor of ∼2. Despite different reaction temperatures used in hydrolysis reactions shown in panel a and panel b, it is worth noting that synergy factors determined from the measurement of soluble product release and measurement of surface degradation were identical within error limit.