An actinobacteria lytic polysaccharide monooxygenase acts on both cellulose and xylan to boost biomass saccharification

Background Lytic polysaccharide monooxygenases (LPMOs) opened a new horizon for biomass deconstruction. They use a redox mechanism not yet fully understood and the range of substrates initially envisaged to be the crystalline polysaccharides is steadily expanding to non-crystalline ones. Results The enzyme KpLPMO10A from the actinomycete Kitasatospora papulosa was cloned and overexpressed in Escherichia coli cells in the functional form with native N-terminal. The enzyme can release oxidized species from chitin (C1-type oxidation) and cellulose (C1/C4-type oxidation) similarly to other AA10 members from clade II (subclade A). Interestingly, KpLPMO10A also cleaves isolated xylan (not complexed with cellulose, C4-type oxidation), a rare activity among LPMOs not described yet for the AA10 family. The synergistic effect of KpLPMO10A with Celluclast® and an endo-β-1,4-xylanase also supports this finding. The crystallographic elucidation of KpLPMO10A at 1.6 Å resolution along with extensive structural analyses did not indicate any evident difference with other characterized AA10 LPMOs at the catalytic interface, tempting us to suggest that these enzymes might also be active on xylan or that the ability to attack both crystalline and non-crystalline substrates involves yet obscure mechanisms of substrate recognition and binding. Conclusions This work expands the spectrum of substrates recognized by AA10 family, opening a new perspective for the understanding of the synergistic effect of these enzymes with canonical glycoside hydrolases to deconstruct ligno(hemi)cellulosic biomass. Electronic supplementary material The online version of this article (10.1186/s13068-019-1449-0) contains supplementary material, which is available to authorized users.


Background
Lytic polysaccharide monooxygenases (LPMOs) employ an oxidative step to cleave polysaccharides not yet fully understood at the atomic level [1,2]. Given their flat surfaces, LPMOs are able to access the crystalline portion of polysaccharides releasing new ends that can elicit the activity of canonical enzymes, promoting a boosting on sugar release [3].
LPMOs on the main component of hemicellulose, xylan. MtLPMO9A, an AA9 isolated from Myceliophthora thermophila C1, releases oxidized xylo-oligosaccharides only when xylan is bound to cellulose [11]. In a similar way, the first component of AA14 family, PcAA14B, from Pycnoporus coccineus, only recognizes xylan when interacting with cellulose chains [12]. The only LPMO reported to act on isolated xylan is LsAA9A, from the Basidiomycete Lentinus similis [13]. Beyond xylan, LsAA9A is also active on cellulose, xyloglucan, mixed-linkage glucan and glucomannan. Structural analysis of LsAA9A bound to xylo-oligosaccharides suggests other mechanisms of interaction during the catalysis than that observed for cello-oligosaccharides; however, it is unclear for other LPMO families.
The activity of LPMOs on a plethora of substrates is desirable especially when added to enzyme cocktails aiming at the deconstruction of ligno(hemi)cellulosic biomass to simple sugars that can be converted into green chemicals and fuels [14].
In this work, we investigated the functional and structural properties of a LPMO from the actinomycete Kitasatospora papulosa, KpLPMO10A. Beyond the already described activities on chitin and cellulose (PASC and Avicel ® ), typical of AA10 from the clade II (subclade A), KpLPMO10A was shown to be active on xylan. This finding expands the range of substrates recognized by this family and depicts a new perspective of the synergistic action of these enzymes with canonical CAZymes to enhance ligno(hemi)cellulosic biomass deconstruction, beyond breaking down glycosidic bonds at crystalline patches in cellulose.

Results
KpLPMO10A is a putative chitin/cellulose-oxidizing enzyme from AA10 family KpLPMO10A, a putative LPMO-coding gene (558 base pairs subtracting those spanning the signal peptide) was isolated from the total DNA of K. papulosa (Additional file 1: Table S1) and cloned into pET-22b(+) vector downstream of the pelB signal peptide to permit the correct processing of the catalytic N-terminal histidine. KpLPMO10A is a non-modular enzyme with 186 amino acids and a predicted molecular mass and isoelectric point (pI) of 20.8 kDa and 4.1, respectively. KpLPMO10A was overexpressed in Escherichia coli cells and purified to homogeneity by Ni-NTA metal affinity and size-exclusion chromatography, with a yield of 0.75 mg per liter of culture.
The phylogenetic analysis of all characterized auxiliary activity 10 (AA10) proteins available on CAZy database (access October 2018), except the fusolin from Anomala cuprea entomopoxvirus, classified the sequences in two clades as predicted by [15] (Fig. 1). Clade I comprises those enzymes exclusively active on chitin, whereas the clade II (subclade A) encompasses LPMOs with the ability to recognize both chitin and cellulose.

KpLPMO10A oxidizes cellulose (C1/C4) and chitin (C1)
KpLPMO10A was predicted as a cellulose/chitin-oxidizing protein by our phylogenetic analysis. Preliminary substrate-binding assays monitored by SDS-PAGE (Additional file 2: Figure S1) showed that KpLPMO10A binds preferably to cellulosic substrates (Avicel ® and PASC, phosphoric acid-swollen cellulose) than α-chitin considering the proportion of enzyme that remained bound to the substrate (insoluble fraction).
HPAEC-PAD chromatograms show the release of native (Glc 2 to Glc 6 ) and C1-oxidized cello-oligosaccharides (17-23 min) from PASC after treatment with KpLPMO10A during 16 h/37 °C (Fig. 2a). Native cello-oligosaccharides appearing in the HPAEC-PAD chromatogram are partly due to on-column degradation of C4-oxidized species, in addition to the native ones naturally formed in the reaction [18]. Evidences of C4-oxidized products were observed in elution times from 23 min (Fig. 2a). In the absence of a reducing agent (ascorbic acid, AscAc), KpLPMO10A is not active against PASC as observed upon its addition (Fig. 2a). Reactions with Avicel ® resulted in the same product profile, but peaks were less intense than that observed for PASC, which is typical for LPMOs (Additional file 3: Figure  S2A).
Unmodified cello-oligosaccharides with degree of polymerization (DP) 5-8 and their C1 and/or C4-oxidized counterparts from PASC were detected by MALDI-TOF MS (Fig. 2b). The inset shows the mono-sodiated unoxidized form of celloheptaose (m/z 1175.314), monosodiated lactone or ketoaldose (m/z 1173.297), and mono-and di-sodiated adducts of C1-aldonic acids (m/z 1191.329 and 1213.263, respectively). In addition, the m/z 1189.243 and 1211.263 correspond to the  double-oxidized (C1 and C4) products and their sodium adducts, respectively. This profile unequivocally indicates that KpLPMO10A releases C1 and/or C4-oxidized products. A similar profile was found for ScLPMO10B [7]. The conversion of PASC into cellobionic acid (GlcGlc1A) was quantified over 24 h (Fig. 2c). KpLP-MO10A remains active for about 2-4 h, releasing 80% of the total GlcGlc1A within the first 30 min. Around 0.4% of PASC was converted into GlcGlc1A in the conditions assayed.
Activity of KpLPMO10A was not observed when the cello-oligosaccharides Glc 5 and Glc 6 were used as substrates.
Experiments with colloidal chitin released chitooligosaccharides up to DP9 (Additional file 4: Figure  S3B). Overall, the intensity of oxidized peaks was higher for this substrate than that observed for non-treated α-chitin. In addition, DP6 was the most intense peak, as observed for PASC treatment in contrast with the ladder pattern observed for α-chitin. Additional file 4: Figure  S3B (inset) also shows the pentamer released from colloidal chitin.

KpLPMO10A boosts sugar release from Avicel ® , Filter Paper and pretreated sugarcane bagasse
Given the activity of KpLPMO10A on cellulosic substrates, we investigated its synergy with Celluclast ® to deconstruct Avicel ® , Filter Paper and alkaline-treated sugarcane bagasse (Fig. 3). The experiments were carried out at 50 °C and pH 5.0 to promote the activity of Celluclast ® . The boosting on glucose release reached 76 and 61% for Avicel ® and Filter Paper (Fig. 3a, b), respectively. When alkaline-treated sugarcane bagasse was used as substrate, a total of 25 and 38% more glucose and cellobiose, respectively, were measured upon the addition of KpLPMO10A. Interestingly, the concentration of xylose was also higher in this condition (52%), suggesting synergy with enzymes other than cellulases eventually present in the cocktail. Assays with sugarcane bagasse were not added with ascorbic acid given the potential role of lignin components as electron donors.

Activity of KpLPMO10A on xylan expands the substrate spectrum recognized by bacterial LPMOs
Given the boost on xylose release observed in the synergy experiments, we evaluated whether KpLPMO10A is able to oxidatively cleave hemicellulosic compounds. Enzymatic reactions with 2 mg/mL xyloglucan, guar galactomannan and konjac glucomannan resulted in HPAEC-PAD profiles identical to the control ones carried out in the absence of KpLPMO10A (Additional file 3: Figure S2B-D). These reactions were also evaluated by MALDI-TOF MS and no products were observed (data not shown).
Xylan is a polymer of β-(1→4)-d-xylosyl residues that resembles the β-(1→4)-d-glucan found in cellulose [19]. Incubation of KpLPMO10A with xylan in the same conditions adopted for cellulose and chitin gave rise to several products and the corresponding peaks were eluted from 10 to 23 min according to HPAEC-PAD profile (Fig. 4a). Peaks from 10 to 14 min were assigned as Xyl 3 to Xyl 6 .
The occurrence of oxidized products was evaluated by MALDI-TOF MS (Fig. 4b). Xylo-oligosaccharides from DP3 to DP7 were produced by KpLPMO10A. Peaks related to the mono-sodiated lactone/ketone (m/z 699.209) and mono-sodiated gemdiol/aldonic acid (m/z 717.200) species are highlighted for the pentamer. Additionally, the sodium adduct of the sodium salt of aldonic acid (m/z 739.213), determinant of C1 oxidation, is not visible.
The absence of C1-oxidized products was further confirmed by comparing the HPAEC-PAD profile of xylan/KpLPMO10A reactions with C1-oxidized xylooligosaccharides produced upon the enzymatic treatment of Xyl 2 -Xyl 5 with cellobiose dehydrogenase (CDH) (Additional file 6: Figure S5). No peaks corresponding to C1-oxidized species were found, suggesting that the oxidative cleavage of xylan by KpLPMO10A occurs exclusively at C4.
KpLPMO10A activity on isolated xylan was also demonstrated by synergy assays with the endo-β-1,4-xylanase xynB (GH11, [20]). Experiments were performed using xylan from beechwood as substrate and the release of xylo-oligosaccharides was measured after 24 h reaction. Reactions with KpLPMO10A (ascorbic acid added) released 25% more oligosaccharides compared to the control experiment (Fig. 4c). Analyzing the released xylo-oligosaccharides separately, it was observed that the increase in Xyl 2 and Xyl 3 reached 23 and 24%, respectively, Xyl 2 being the main product. Results regarding Xyl, Xyl 4 , Xyl 5 and Xyl 6 did not show statistically significant differences (data not shown).
The addition of the strictly cellulose-oxidizing LPMO from fungal origin "LPMO9" (Corrêa et al. to be submitted) to xynB did not increase the conversion of xylan into xylo-oligosaccharides, thus, validating the xylan-oxidizing activity of KpLPMO10A (Fig. 4c).
Xyl 5 and Xyl 6 were also evaluated as substrates for KpLPMO10A, but no products were recorded.

KpLPMO10A catalytic interface is indistinguishable from other clade II (subclade A) AA10 LPMOs
To understand the structural basis of KpLPMO10A activity on both crystalline and non-crystalline substrates, we have solved its crystal structure at 1.6 Å resolution ( Table 1). As expected, KpLPMO10A displays the fibronectin/immunoglobulin-like β-sandwich fold typical of LPMOs [21] with two β-sheets, the first one comprising three antiparallel strands (S1, S4 and S7) and the later with four parallel strands (S5, S6, S8 and S9). The 85-residue-long motif consisting of short α-helices and loops between the strands S1 and S4 of the antiparallel β-sheet is known as loop 2 (Fig. 5a). This motif forms a significant patch of the substrate binding (Fig. 5b) that is conserved among the characterized chitin and C1/C4-cellulose-oxidizing AA10 s (Additional file 7: Figure S6). No copper occupancy was observed in KpLPMO10A structure, but the superimposition with ScLPMO10B (PDB code: 4OY6) (Fig. 5c) showed that all residues relevant for copper coordination are fully conserved, including the histidines 1 and 108 (brace of histidines, H1 and H108) and the tyrosine 177 (Y177). This aromatic residue is conserved among AA10s and occupies an axial position for copper coordination. Conserved residues neighboring the copper ion center also include Asp104, Ala106, His172 and Gln175. Studies carried out by [17] showed that mutations on residues near the copper center led to a decrease in activity, but did not affect the regioselectivity displayed by C1/C4-oxidizing bacterial LPMOs. Although the copper was not found in both molecules in the asymmetric unit, thermal shift assays revealed its key role in thermal stability of KpLPMO10A with an increase in melting temperature (T m ) of ~ 12 °C and ~ 10 °C at pH 5.0 and 6.0, respectively (Additional file 8: Figure S7). The process was shown to be reversed following the addition of EDTA to copper-saturated KpLPMO10A as also assayed by circular dichroism spectroscopy (data not shown).
The catalytic interface of KpLPMO10A was compared in detail with other structurally characterized AA10 members from the clade II (subclade A) including ScLP-MO10B, Micau_1230 (PDB code: 5OPF) and E7 (PDB code: 5UIZ), which did not reveal any evidence of structural and conformational difference that could be correlated with the ability to oxidize xylan chains. This analysis suggests that closely related members may also be active on xylan, being the clade II (subclade A), a polyspecific subfamily of the AA10 family or that the substrate recognition and binding involve yet elusive mechanisms.

Discussion
LPMOs are enzymes that have revolutionized the polysaccharide deconstruction field. Recent discoveries such as the activity on hemicellulose [9] and starch [22], alternative electron donors [23][24][25] and co-substrates [2] point out that our knowledge about these biocatalysts is in its infancy. Further studies are required to shed light on the catalytic cycle steps, substrate selectivity, synergy with other CAZymes and the biological roles on carbon cycle. The expansion on the understanding of these aspects is instrumental for industrial applications.
The prediction of KpLPMO10A as a chitin and C1/ C4-cellulose-oxidizing enzyme was confirmed by experimental assays. KpLPMO10A showed a product profile similar to ScLPMO10B, Micau_1230 and E7 acting on PASC [7,17,32]. PASC probably adopts the same orientation in the similar binding patch and catalytic center shared by these enzymes. Mutation studies from Micau_1230 showed that the regioselectivity of AA10 relies on the positioning of substrate on the copper site [17], while for AA9 it sounds to be determined by structural features of loop 2, aromatic residues on substratebinding patch and exposed N-glycans [33,34].
KpLPMO10A releases GlcGlc1A after up to 2 hours of reaction, which resembles the production of C4 oxidized species by MtLPMO9J [35]. For PaLPMOH, the release of GlcGlc1A is time dependent [36].
Treatment of α-chitin with phosphoric acid ("colloidal chitin") gives rise to a substrate with low crystallinity, molecular mass, surface area and particle size [40]. KpLPMO10A was also able to release oxidized products from this substrate, with preference for DP6. Beyond this, the DP of products released was extended, in this particular case to 9, which is in accordance with data from [41]. In addition, peaks regarding the oxidized products were higher than those observed for α-chitin.
Synergy assays of KpLPMO10A with Celluclast ® , an enzyme cocktail containing mainly cellobiohydrolase (CBH) and endo-1,4-β-glucanase (EG) activities [42], promoted an increase in glucose and cellobiose release from Avicel ® and Filter Paper, highlighting the potential application of this enzyme as additive in enzyme cocktails. Some examples of sugar release boosting by LPMOs are highlighted in [14]. We also observed higher titers of xylose, in addition to glucose and cellobiose, when sugarcane bagasse was employed as substrate. Beyond CBHs and EGs, Celluclast ® contains a range of other activities, for example, xylanases [43]. Then, we hypothesized that KpLP-MO10A could act also in hemicellulosic components.
Activity on isolated xylan (not complexed to cellulose), as shown by KpLPMO10A, is a rare event for LPMOs.
The only record is found for LsAA9A, from L. similis, which oxidizes xylan and xylo-oligosaccharides (Xyl 6 ) at C1/C4 and C4, respectively [13]. Other two works show that the release of oxidized products from xylan by MtLPMO9A (C1/C4 oxidation) and PcAA14B (C1 oxidation) is only possible when this substrate is complexed with cellulose [11,12]. PcAA14B is the first member of AA14 family, comprising exclusively lytic xylan oxidases. In contrast to that observed for PcAA14B and MtLP-MO9A, which oxidize xylan at C1 and C1/C4, respectively, KpLPMO10A releases C4-oxidized species, which was confirmed by HPAEC-PAD analysis.
Comparison of KpLPMO10A, LsAA9A (5NLO) and PcAA14B (5NO7) structures did not show any clue about the activity/regioselectivity on xylan (Additional file 9: Figure S8). LsAA9A is the only LPMO structure complexed with substrates (cello and xylo-oligosaccharides) available in the Protein Data Bank. LsAA9A presents a less pronounced binding surface and a protuberance near the subsites + 1 and + 2 [46], while PcAA14B has a clamp formed by two surface loops [12]. These data suggest different structural determinants for xylan specificity in LPMOs. Unfortunately, the structure of MtLPMO9A, the other LPMO that cleaves xylan, is not available.
This work is the first to describe the activity of an AA10 member on xylan. Structural comparisons with other AA10, such as ScLPMO10B, Micau_1230 and E7 indicate no significant differences that would explain the ability of KpLPMO10A to act on xylan, which tempted us to suggest that these homologues might also exhibit this similar feature or yet obscure mechanisms of substrate interaction are involved.
The activity on xylan is also corroborated by synergy assays with Celluclast ® and the endoxylanase xynB. KpLPMO10A promoted a higher overall sugar release from biomass when compared to an exclusively cellulose-oxidizing AA9 (LPMO9) also studied by our group and the activity on hemicellulose probably facilitates the accessibility to cellulose chains.

Conclusions
In this work, we describe the cloning, overexpression, functional and structural characterization of a new bacterial LPMO (named as KpLPMO10A), which is active on xylan, beyond the classical activity on chitin and cellulose (PASC/Avicel ® ). KpLPMO10A is the first bacterial LPMO to exhibit such activity on xylan and only very few fungal LPMOs are able to attack this polysaccharide. Moreover, KpLPMO10A shows a regioselectivity for xylan (C4) distinct to that observed for cellulosic substrates (C1/C4), which could serve as an excellent model to address mechanistic questions regarding regioselectivity. The discovery of new activities in already known families opens new horizons regarding the mechanistic of LPMOs, its role on carbon cycle and its suitability for new industrial applications.

Isolation and cloning of KpLPMO10A
The genomic DNA of K. papulosa (DSM41643) was extracted according to [47] with modifications. The gene KpLPMO10A was amplified by polymerase chain reaction (PCR) using the primers KpF and KpR (Additional file 1: Table S1). The reaction contained 0.25 μL of Phusion DNA Polymerase (New England Biolabs), 5 μL of 5X buffer, 0.25 μL of 50 mM MgCl 2 , 1 μL of 10 mM dNTPs, 2 μL of 5 μmol each primer and 12.5 μL of water. PCR conditions were as follows: initial denaturation at 98 °C for 30 s, 35 denaturation cycles at 98 °C for 10 s, annealing at 61 °C for 30 s, extension at 72 °C for 30 s and a final extension at 72 °C for 10 min.
The amplicon (558 base pairs) was purified using QIAquick gel extraction kit (Qiagen) and employed as a template for a second round of PCR using the primers KpFpET22b and KpRpET22b (Additional file 1: Table S1). KpFpET22b and KpRpET22b were designed to add a homology sequence to vector pET-22b(+). The same reaction conditions highlighted above were adopted in this step.
The purified product was inserted into pET-22b(+) immediately after pelB coding sequence to ensure the histidine as the first amino acid and the His 6 -tag at C-terminus by Gibson Assembly methodology [48]. The reaction products were transformed into thermocompetent Escherichia coli DH5α cells. pET-22b(+)-KpLPMO10A was purified using QIAprep Miniprep System (Qiagen) and was sequenced to confirm the correct integration of KpLPMO10A in pET-22b(+).

Overexpression and purification
Thermocompetent cells of E. coli SHuffle ® transformed with pET-22b(+)-KpLPMO10A were grown on Terrific Broth (TB) at 30 °C/250 rpm until it reached an optical density of 0.6. After induction with 0.5 mM isopropyl β-d-1-thiogalactopyranoside (IPTG), cells were maintained overnight at 16 °C. The cells had the periplasmic fraction extracted by osmotic shock protocol [49]. The supernatant was loaded onto a His-Trap column 5 mL (GE Healthcare) equilibrated with buffer A (0.02 M potassium phosphate pH 7.4, 0.5 M NaCl and 0.05 M imidazol). Bound proteins were eluted following a gradient of Buffer B (0.02 M potassium phosphate pH 7.4, 0.5 M NaCl and 0.5 M imidazol). Fractions containing KpLPMO10A were pooled together, concentrated with Vivaspin (cutoff 10 kDa, Sartorius) and applied onto HiLoad Superdex G-75 16/60 (GE Healthcare) previously equilibrated with 0.02 M potassium phosphate pH 7.4 and 0.15 M NaCl. Elution was achieved at a flow rate of 1 mL/min using the same buffer. All fractions were evaluated by SDS-PAGE [50] and those containing the enzyme were pooled together, concentrated and employed for further assays. All purification steps were performed in Äkta Purifier System (GE Healthcare).

Crystallization, data collection, structure determination and refinement
KpLPMO10A was crystallized by the sitting-drop vapordiffusion method at 18 °C using a protein concentration of 6 mg/mL. Crystals were obtained in 0.1 M MMT buffer, pH 4.0 and 25% (v/v) polyethylene glycol (PEG) 1500. Diffraction data were collected on the beamline MX2 at the Synchrotron Light Source Laboratory, LNLS, Campinas, São Paulo, Brazil. A dataset corresponding to 180º was obtained using the fine-slicing strategy (0.1 per image). The wavelength was set to 1.46 Å and a PILATUS2 M detector was used to record the reflections (Dectris, Baden-Dattwil, Switzerland). Data were integrated and scaled using the XDS package [51]. KpLPMO10A was solved by molecular replacement using the crystal structure of the LPMO from Streptomyces coelicolor (PDB code 4OY6) [7]; sequence identity of 86%] using the program phaser [52] as a module of PHENIX package [53].
The final model was obtained after several manual building cycles using Coot [54] along with restrained refinement using REFMAC5 [55]. The atomic coordinates and diffraction data were deposited on Protein Data Bank (https ://www.pdb.org/) under the code 6NDQ. Molecular images were produced with PyMOL software (Schrödinger).

Differential scanning fluorimetry (DSF)
The stability of KpLPMO10A in the presence of copper was evaluated by differential scanning fluorimetry (DSF) in a ViiA 7 Real-Time PCR System equipment (Thermo Fischer Scientific). The buffers 0.05 M sodium acetate, pH 5.0, or 0.05 M potassium phosphate, pH 6.0, were mixed with 15 μL (1.3 mg/mL) of copper-treated or -non-treated KpLPMO10A and 15 μL of Sypro Orange (66-fold diluted from a concentrated solution). The fluorescence was measured following the increase in temperature from 30 to 70 °C. The melting temperature (T m ) was obtained by fitting a sigmoidal curve in the Origin Software (Origin Lab).

Binding assays
Binding assays were performed as recommended by [56]. Phosphoric acid-swollen cellulose (PASC) 1%, Avicel ® 5% and α-chitin from shrimp shells 1% were incubated with 80 μg of KpLPMO10A and 0.05 M potassium phosphate buffer, pH 6.0, in a total volume of 200 μL for 1 h on ice and gentle mixing. Additionally, reactions were carried out at 37 °C/850 rpm during 16 h. Ten millimolar of EDTA was added to prevent catalytic activity. After incubation, the samples were centrifuged at 10,000g for 5 min and the supernatants were considered as the soluble fraction. The pellet was washed twice with buffer obeying the same volume of the initial reaction. The remaining pellet was treated with SDS-PAGE without dye to dissociate the bound proteins (insoluble fraction). All reactions were evaluated by SDS-PAGE.

Detection of reaction products by MALDI-TOF MS
The analysis of products was carried out in a MALDI-TOF Autoflex Speed system (Bruker Corporation). One microliter of reaction was added to 1 µL of 2.5-dihydroxybenzoic acid (DHB) in TA30 (30% acetonitrile and 0.1% trifluoroacetic acid). A total of 1 µL was applied on GrounSteel plate and was evaluated in the mass range of 100-2500 Da. Data were acquired on positive polarity reflector mode. All analyses were carried out in comparison with a blank without the addition of KpLPMO10A and only the peaks not found in the blank ones were considered.

Saccharification assays
The synergy of KpLPMO10A (1 mg/g substrate) and Celluclast 1.5 L (0.9 FPU/g substrate, Sigma-Aldrich) was evaluated. The reactions were carried out at 50 °C/200 rpm in 0.05 M sodium acetate buffer, pH 5.0 and 10 mg substrate for 24 h (Avicel ® and Filter Paper) or 72 h [alkaline-pretreated sugarcane bagasse (%); cellulose, 58.6 ± 1.2; hemicellulose, 22.1 ± 1.4 and lignin, 10 ± 0.4] in a total volume of 1 mL. Control reactions were carried out in the absence of KpLPMO10A. The synergy of KpLPMO10A (1 mg/g substrate) and LPMO9 (a fungal strictly cellulose-oxidizing LPMO [Corrêa et al. to be submitted], 1 mg/g substrate) with the xylanase xynB/XAC4254 (0.5 mg/g substrate, [20]) on deconstruction of xylan from beechwood (10 mg) was evaluated in a 1 mL reaction containing 0.05 sodium phosphate buffer, pH 6.0, at 37 °C/1000 rpm. Ascorbic acid (1 mM) was added to some reactions. After the treatment, the samples were boiled for 10 min and centrifuged at 12,000g. The release of glucose/cellobiose/xylose and xylo-oligosaccharides was evaluated by high-performance liquid chromatography (HPLC) and HPAEC-PAD, respectively. All experiments were performed in triplicate.