Deletion of AA9 Lytic Polysaccharide Monooxygenases Impacts A. nidulans Secretome and Growth on Lignocellulose

To understand better the A. nidulans response to lignocellulose in the environment, secretomes (set of extracellular proteins) induced on crystalline cellulose (Avicel) or agroindustry-derived lignocellulose, such as steam-exploded sugarcane bagasse (SCB) and sugarcane straw (SCS), were investigated by mass spectrometry (MS). Out of the 253 proteins identified in all secretomes, 104 proteins presented CAZymes domains, corresponding to 64 glycoside hydrolases (GHs), 3 pectin lyases (PLs), 15 carbohydrate esterases (CEs), and 22 AAs (Table S1). Most of CAZymes were commonly secreted in all inducing conditions, while only a few members were substrate-specific (Fig. 1A). The highest level of CAZymes secretion, estimated by the sum of normalized peptide counts, was observed in the presence of Avicel followed by SCB and SCS. Concerning the cultivation period, the trend was increasing secretion overtime in all tested conditions (Fig. 1B).

A total of 22 members were assigned to different AA families, numerically corresponding to the second most secreted class of CAZymes. The identified AAs corresponded to members from families AA2 (1), AA3 (5), AA7 (9), AA9 (6) and AA13 (1). Apart from AA9s (described below), most of the identified AAs were concomitantly secreted on the different substrates, excepting the AA2 catalase/peroxidase (AN7388) exclusively found on glucose, the AA13 LPMO (AN5463 appended to CBM20) secreted only on Avicel, and the AA7 dehydrogenase (AN2648) absent only in the glucose secretome. Among AA3s, one AA3_2 glucose oxidase (AN4006) was secreted on Avicel and SCB, one AA3_3 alcohol oxidase (AN0567) was exclusively found on SCB, and the AA3_1 CDH (AN7230 appended to CBM1) was present in all secretomes, excepting glucose (Table S1). Indeed, this CDH (AnCDH1) was abundantly secreted, corresponding to 57, 34 and 30% of the total AA spectra on Avicel, SCB and SCS, respectively. While its secretion decreased overtime on Avicel, it was found at variable levels on SCB and SCS (Fig. 1C).

The set of LPMO9s, named according to Nekiunaite et al. (48) consists of AnLPMO9A (AN2388), AnLPMO9B (AN1602, appended to CBM1), AnLPMO9C (AN6428), AnLPMO9D (AN3046), AnLPMO9F (AN3860) and AnLPMO9G (AN10419). In contrast to the increasing profile of CAZymes secretion, the overall secretion of LPMO9s was generally more abundant in the early phase of cultivation, decreasing in a time-dependent manner (Fig. 1D). The secretion profiles on Avicel and SCS were quite similar both quantitatively and qualitatively, being mainly composed of AnLPMO9F complemented by minor and constant amounts of AnLPMO9G. Additionally, the SCS secretome included small amounts of AnLPMO9D and AnLPMO9C which were found at 24 and 72 h, respectively. In turn, SCB induced a more diversified secretome, which included the concomitant secretion of all the six LPMO9s at variable levels overtime. AnLPMO9D was highly secreted at 24 h and decrease at late periods, whereas AnLPMO9F was detected at a steady and intermediate level. The SCB repertoire also included the secretion of AnLPMO9C that decreased overtime, AnLPMO9G at intermediate-to-small amounts, and small amounts of AnLPMO9A and AnLPMO9B (exclusively found in the SCB secretome). Remarkably, small amounts of AnLPMO9A were also identified on glucose at 24 and 72 h (Fig. 1D). Because of the AnLPMO9F and -G production in all inducing conditions, and the exclusive secretion of AnLPMO9C on lignocellulosic substrates, these three enzymes were selected for further functional studies.

To confirm their oxidative activities, the genes encoding AnLPMO9C, -F and -G were cloned with their native signal peptides and produced in A. nidulans by homologous expression. The purified recombinant enzymes showed higher MW (30.8, 29.5 and 33.4 kDa for AnLPMO9C, -F and -G, respectively) than the predicted from amino acid sequence (22.5, 24.1 and 26.5 for AnLPMO9C, -F and -G, respectively). The higher MW observed may be attributed to glycosylation, considering the presence of both N- and O-glycosylation sites predicted by NetNGlyc 1.0 Server and NetOGlyc 4.0 Server, respectively. According to the GLYCAM-Web tool, the AnLPMO9C is the only target harboring accessible sites for N-glycosylation, and indeed the MW decreased by Endo H and PNGase F treatment (25.8 kDa - data not shown), confirming the presence of N-glycosylated residues. Correct processing of the purified enzymes was confirmed by determining the N-terminal sequence, which yielded sequences identical to those predicted from the corresponding nucleotide sequence (data not shown).

Differential scanning fluorimetry (DSF) assays revealed higher stability (Tm of 60, 62, 59°C for AnLPMO9C, -F and -G, respectively) of the holoenzymes (copper loaded) than the Cu-deprived enzymes obtained after incubation with EDTA (Tm of 44, 59, 55°C for AnLPMO9C, -F and -G, respectively). The change in thermal stability confirmed the presence of copper coordinated in the active site of the enzymes providing stabilization to their structures (Fig. 2A). Enzymatic activity of the recombinant AnLPMO9s was initially verified by Amplex Red assays, in which hydrogen peroxide formation indicates that the enzymes were active, therefore responding to the presence of different reductants. The highest activity for the three enzymes was observed using ascorbic acid (Fig. 2B).

Enzymatic assays with the AnLPMO9s were then performed using phosphoric acid-swollen cellulose (PASC) as the substrate and ascorbic acid as external reductant. Analysis of reaction products from the three AnLPMO9s by HPAEC-PAD and MALDI-TOF showed the presence of native and oxidized cellooligosaccharides. Taken together, these assays confirmed the activity and regioselectivity of each enzyme on cellulose, i.e., the C1-oxidation performed by AnLPMO9C, and the mixed C1/C4-oxidation by AnLPMO9F and -G (Fig. 2C and D). When similar enzymatic reactions were performed using Avicel the enzymes maintained their regioselectivity, but weaker signals were observed especially for AnLPMO9C (Fig. S1A) in comparison to the reactions using PASC as the substrate (Fig. 2C). In turn, upon incubation with cellohexaose, beechwood xylan, or PASC/xylan mixtures, no oxidized products from cellohexaose or xylan were identified (data not shown).

To further investigate the catalytic performance of the AnLPMO9s on cellulose, we used a β-glucosidase-assisted assay which allowed degradation of the soluble gluco-oligosaccharides released from reactions applying AnLPMO9F and -G on PASC, filter paper (FP) and Avicel. The results correspond to an estimate of mainly native and minor C1 oxidized species, as oxidation at the C4-position possibly hinders β-glucosidase activity. Higher glucose amount was released from PASC using AnLPMO9F, whereas using AnLPMO9G more glucose was obtained from reactions with cellulosic substrates containing crystalline regions such as FP and Avicel (Fig. 3). In addition, activity was also evaluated on cellulose microcrystals extracted from the algae Valonia ventricosa, and only AnLPMO9G was active (Fig. S1B).

Based on the LPMO9s secretion profiles and the verified oxidative activity on cellulose, we hypothesized that AnLPMO9F and -G, as well as the cosecreted AnCDH1, would be important oxidative players for cellulose degradation by A. nidulans. To investigate the biological role of these enzymes in A. nidulans, knockout strains carrying single deletions, the respective double mutants as well as the triple mutant were built using CRISPR-Cas9 approach, and gene knockout was confirmed by PCR and Southern blot (Fig. S2).

Growth analysis on solid minimal medium (MM) supplemented with different substrates showed the mutant strains growing at the same rate on glucose, whereas reduced growth was verified for the single mutants ΔAnLPMO9F and ΔAnLPMO9G growing on cellulose and xylan, in comparison to the reference strain (Fig. 4A). The strains were then cultivated in liquid MM with Avicel, and the extracellular enzymatic activities profiles, as well as fungal growth, were investigated. Initial analysis revealed no changes in the secretory system since, overall, the secretomes presented similar levels of protein secretion relative to the reference strain (Fig. 4B). Deletion of the AnLPMO9F had a major impact on the activity on PASC (PASCase), which was reduced by more than 60%. Additional deletions, however, had no further impact, i.e., the secretomes of the double knockout strains ΔAnLPMO9F/ΔAnLPMO9G and ΔAnLPMO9F/ΔAnCDH1, as well as the triple mutant, presented similar activity levels than the ΔAnLPMO9F single mutant. Additionally, single and double deletions of AnLPMO9G and AnCDH1 had no impact on the activity on this substrate (Fig. 4C). Filter paper activity (FPase) was also strongly affected by AnLPMO9F deletion, whereas single deletions of AnLPMO9G and AnCDH1 also reduced this activity at moderate levels. Double deletions of the AnLPMO9s reduced FPase to the lowest level (Fig. 4D). Deletions were less influential for the degradation of hydroxyethyl cellulose (HEC) as the activities were partially reduced to the same level in all mutants; and the activity on β-glucan underwent reductions similar to that observed on PASC. Moreover, some mutants showed reduced activity on 4-nitrophenyl β-d-glucopyranoside (pNPG), xylan and 4-nitrophenyl β-d-xylopyranoside (pNPX), whereas most of the mutants displayed higher activity on 4-nitrophenyl β-d-cellobioside (pNPC) (Fig. S3). Despite the remarkable reductions in the extracellular cellulolytic activity of the knockout strains, no significant changes in fungal growth were verified when the strains were cultivated on Avicel (Fig. 5A). However, when cultivation was performed on SCS, deletions caused about 25% reduction in A. nidulans growth, of A. nidulans growth, as observed for the single and the double ΔAnLPMO9F/ΔAnLPMO9G mutants (Fig. 5B).

To further explore modifications in the microbial metabolism, the secretomes of the single deletion mutants induced on Avicel were analyzed by MS (Table S2). Overall, no changes in protein secretion were verified considering normalized peptide counts, confirming the data from the protein quantification assay. However, quantitatively a differential secretion was verified by categorizing enzymes in the following groups: cellulases, hemicellulases, proteases, and miscellaneous enzymes and proteins with unknown functions (named as “others”). Peptide counts of cellulases were reduced by around 20% in the LPMO9s knockout strains, whereas hemicellulases were approximately 30% more abundant (Fig. 6A). Particularly, changes in the secretion of β-glucosidase (bglL, AN2828) and cellobiohydrolase (cbhA, AN5176) can be related to the changes in the cellulolytic profile of the mutant strains, while secretion of xylanase (xlnA, AN3613), absent in the reference strain secretome, is a remarkable change in the hemicellulases group (Fig. 6B). Notably, in the ΔAnCDH1 strain, peptide counts of cellulases were reduced by more than 50% and a striking increase occurred in the secretion of the category “others” (Fig. 6A), which included fungal cell wall-associated proteins (AN7269, AN7657, AN8969 and AN4055) (Fig. 6B).

In contrast to gene deletion, we used A. nidulans recombinant strains overexpressing AnLPMO9C, -D, -F and -G, aiming to increase the titer and/or change the LPMO9s profile relative to the originally verified on cellulose. The strategy adopted to obtain LPMO9s-enriched secretomes was initial cultivation on Avicel, a condition which naturally induces the secretion of AnLPMO9F, AnLPMO9G and AnCDH1 along with several other cellulases (Fig. 1C and D, Table S1), followed by recombinant AnLPMO9s induction by maltose. We evaluated a second strategy in which recombinant LPMO9s maltose-dependent induction was concomitant to the cultivation on Avicel (data not shown); however, the two-step strategy resulted in secretomes displaying higher levels of the target enzymes (Fig. 7A). The recombinant strains secreted lower levels of proteins (Fig. 7B) in comparison to the reference strain (transformed with the empty vector), and the enzymatic profiles using various substrates were distinct among them. In this regard, the AnLPMO9F-enriched secretome was more active on PASC (Fig. 7C), whereas the AnLPMO9G-enriched secretome presented increased FPAse activity (Fig. 7D). Variations on xylanase activity were also observed in the AnLPMO9C, -F and -G-enriched secretomes. Notably, the AnLPMO9D enriched secretome retained the same activity levels or underwent severe activity reductions (Fig. S4). When applied as supplementing agents for lignocellulose saccharification, the secretomes enriched with AnLPMO9C and -G demonstrated to increase both glucan and xylan conversions. Remarkably, supplementation with the AnLPMO9G-enriched secretome resulted in glucan conversion yields superior to that of using total enzyme load made up of the commercial cocktail (Fig. 7E and F).

A. nidulans A773 (pyrG89; wA3; pyroA4; veA1) was obtained from the Fungal Genetic Stock Center (FGSC, University of Missouri, Kansas City, MO, USA) and used to generate A. nidulans Δku (nkuA/AN7753). The A. nidulans strains constructed in this project (Table S3) corresponded to those homologously expressing AnLPMO9C, AnLPMO9D, AnLPMO9F and AnLPMO9G and the single, double, and triple knockouts for AnLPMO9F, AnLPMO9G and AnCDH1. These strains were routinely maintained on minimal medium (MM) composed of 20× Clutterbuck salts (72), 1000× pyridoxine, and 1000× trace elements, pH 6.5 supplemented with 1% (wt/vol) glucose and cultivated at 37°C. Parental strains were cultivated in the same medium, including 5 mM uracil and uridine. Escherichia coli DH5α was used to propagate all plasmids.

A. nidulans A773 conidia suspension were inoculated (5 × 10⁶ conidia/ml final concentration) for precultivation on Erlenmeyer flasks (250 mL) containing 50 mL liquid MM 1% glucose (wt/vol) pH 6.5 for 24 h at 37°C. Mycelia were removed by filtration, washed abundantly and transferred to Erlenmeyer flasks (250 mL) containing 50 mL liquid MM pH 6.5 supplemented with 1% (wt/vol) Avicel, glucose, steam-exploded sugarcane bagasse (SCB) and steam-exploded sugarcane straw (SCS), cultivation then proceeded for 24, 72 and 120 h at 37°C. Cultivation was performed using three replicates. Mycelia were removed by filtration and the crude filtrates were centrifuged (10.000g, 40 min, 4°C) and concentrated in 10 kDa cutoff Amicon (Millipore, USA). The concentrated secretomes (20 μg) were partially resolved on 12% SDS-PAGE, excised, reduced, and digested with 20 mg/ml trypsin (Promega) (73). After extraction, samples were dried under vacuum, peptide mixtures were analyzed by LC-MS/MS -LTQ Orbitrap Velos (Thermo Fisher Scientific), and data were analyzed as previously described (5).

The full-length ORFs encoding the targets AnLPMO9C (AN6428), AnLPMO9D (AN3046), AnLPMO9F (AN3860) and AnLPMO9G (AN10419) were PCR amplified from A. nidulans A773 gDNA using Phusion DNA polymerase, according to standard protocols and primers listed in Table S4. Amplicons were ligated into the pEXPYR expression vector (59) using BbvCI and XbaI restriction sites and transformed into E. coli DH5α. After cloning and confirmation by sequencing, the vector was transformed into A. nidulans A773 protoplasts (74). Transformants growing in the absence of uracil and uridine were screened for target enzyme secretion by cultivation in liquid MM supplemented with 1% (wt/vol) glucose and 2% (wt/vol) maltose, under static conditions for 36 h at 37°C. Cultivation supernatants were concentrated using 10 kDa cutoff Amicon (Millipore, USA) and protein profiles were analyzed by SDS-PAGE.

A. nidulans transformants expressing AnLPMO9C, AnLPMO9F and AnLPMO9G were grown in Erlenmeyer flasks (2 L) containing 200 mL MM pH 6.5 supplemented with 1% (wt/vol) glucose and 2% (wt/vol) maltose. Cultivation was performed for 36 h under static conditions at 37°C. Cultivation supernatants were centrifuged (5.000g, 40 min, 4°C), filtered (0.45 um) and loaded onto a HiPrep DEAE FF 16/10 column (GE Healthcare), equilibrated in 50 mM Tris–HCl buffer pH 7.0. Proteins were eluted using a linear gradient up to 1 M NaCl in the same buffer. Fractions containing a protein band with the expected theoretical MW were pooled and concentrated. The concentrated sample was applied to a HiTrap Phenyl HP 1 mL column (GE Healthcare) and eluted in the same buffer with a linear gradient from 1.5 – 0.0 M ammonium sulfate. Fractions containing a protein band with the expected theoretical MW were pooled. Partially purified enzymes were incubated with Cu(II)SO₄ (3:1 molar ratio of copper:enzyme) overnight at 4°C. Then, samples were concentrated, washed in 50 mM MES buffer pH 6.0 and applied to a Superdex 75 column (GE Healthcare) equilibrated in the same buffer. Eluted fractions were analyzed in SDS-PAGE and those presenting the purified enzyme were pooled. Protein concentration was determined by reading A₂₈₀ and considering the theoretical MW and molar extinction coefficient of each enzyme.

Differential scanning fluorimetry (DSF) with SYPRO Orange fluorescent dye (Sigma-Aldrich) was used to monitor protein unfolding (75). Reactions (30 μL) were composed of 25 mM Tris HCl buffer pH 7.0, 30 μg of purified AnLPMO9C, AnLPMO9F and AnLPMO9G and 5-fold diluted SYPRO Orange. Stabilization provided by copper-binding to the LPMOs was determined by comparing their temperature of melting (T_m) to the controls containing the enzymes previously incubated with 5 mM EDTA. Assays were performed as triplicate independent experiments. The T_m was determined by plotting fluorescence against temperature and fitting the data to a sigmoid function (76). The influence of reductants on the activity of AnLPMO9C, AnLPMO9F and AnLPMO9G was verified with Amplex Red assays by measuring peroxide formation (77). The evaluated reductants were 50 μM ascorbate, pyrogallol, hydroquinone, ferulic acid, 2,6-dimethoxyphenol (2,6-DMP) and gallic acid. Assays were performed as triplicate independent experiments and controls correspond to reactions with the copper-deprived AnLPMO9s. Enzymatic reactions with substrates were carried out in 50 mM sodium acetate buffer pH 5.0 containing 0.1% (wt/vol) PASC, 0.2% (wt/vol) beechwood xylan (Sigma-Aldrich), 0.2% (wt/vol) PASC/xylan mixture, 0.1% (wt/vol) Valonia cellulose, and 50 μM cellohexaose, 1.0 μM AnLPMO9C, AnLPMO9F and AnLPMO9G and 1 mM ascorbate as reductant. PASC was prepared from Avicel PH-101 (Sigma-Aldrich) (78). Valonia cellulose microcrystals were prepared from vesicles of the algae Valonia ventricosa (64). Reaction mixtures were shaken in a ThermoMixer (Eppendorf) at 800 rpm and 50°C for 24 h. Enzymatic activity was stopped by heating the samples at 100°C for 5 min, soluble fractions were separated by centrifugation (13,000g, 5 min, 4°C) and stored at 4°C. All assays were performed as triplicate independent experiments and controls correspond to reactions without ascorbic acid and a reaction without any enzyme.

Oligosaccharides released from reactions with PASC were initially analyzed by high-performance anion-exchange chromatography (HPAEC) on a Dionex ICS5000 system (Thermo Fisher Scientific) equipped with a CarboPac PA1 14 mm column and pulsed amperometric detection (PAD) (79). Elution and identification of native and oxidized products were based on previously published data (80). Additionally, products from reactions with PASC, Valonia cellulose, cellohexaose, xylan and PASC/xylan mixture were analyzed on Ultraflex MALDI TOF mass spectrometer (Bruker Daltonics) (75). Spectra intensity (V) was expressed relative to the maximum values.

β-glucosidase-assisted assay (81) was performed by incubating 2% (wt/vol) Avicel, FP and PASC in 50 mM ammonium acetate buffer pH 5.0 with 1 μM AnLPMO9F and -G, and 1 mM ascorbic acid. The reactions were carried out under 800 rpm shaking in a ThermoMixer (Eppendorf) at 30°C for 24 h. Samples were centrifuged (13,300g, 15 min, 4°C), filtered (0.22 μm), and 450 μL of the supernatant was mixed with 1 U of commercial β-glucosidase from Aspergillus niger (Megazyme). The reactions were then incubated for 24 h at 40°C. Samples were cooled, diluted, and the products quantified using high-performance anion-exchange chromatography (HPAEC-PAD) (82). The experiment was performed with two replicates and controls corresponded to reactions without enzyme and reactions without ascorbic acid.

CRISPR/Cas9 systems were constructed (83, 84) using primers (Table S4) to amplify noncoding regions flanking the targets AN3860 (AnLPMO9F), AN10419 (AnLPMO9G) and AN7230 (AnCDH1) on the A. nidulans genome. Constructed vectors and synthesized repairing oligonucleotides composed of upstream and downstream sequences of the target genes (Table S4) were used for transformation using A. nidulans Δku protoplasts. Transformants growing in the absence of uracil and uridine were submitted to five rounds of monosporic purification. Double and triple knockout strains were obtained by performing sequential rounds of deletion after having a confirmed mutant. Deletions were confirmed by diagnostic PCR and Southern blot (85) using genomic DNA of the knockout strains.

For growth analysis in solid medium, conidia suspensions of the knockout strains were inoculated on the center of a petri dish containing MM pH 6.5 supplemented with 1% (wt/vol) glucose, carboxymethyl cellulose (CMC) and xylan from beechwood. Cultivation was performed with 5 replicates at 37°C and colony diameters were measured after 96 h. For liquid medium cultivation, conidia of the knockout strains (5 × 10⁶ conidia/ml final concentration) were precultivated in MM with 1% (wt/vol) glucose for 24 h and then mycelia were transferred for cultivation in MM Avicel for additional 48 h, as described before. The mixtures of mycelia and substrate were separated by filtration, manually dried with filter paper and stored at −80°C. The crude filtrates were centrifuged (10.000g, 40 min, 4°C), filtered (0.45 μm), concentrated when necessary (10 kDa cutoff Amicon) and used for protein and activity assays. Cultivation was also performed using 1% SCS (wt/vol) as the substrate at the same conditions. The solid fractions (Avicel and mycelia, and SCS and mycelia) were used for ergosterol extraction and quantification, as described below. Cultivation was performed using six replicates. Enzymatic activities in the secretomes were assayed using different substrates: 5 mM 4-nitrophenyl β-d-cellobioside (pNPC), 4-nitrophenyl β-d-xylopyranoside (pNPX) and 4-nitrophenyl β-d-glucopyranoside (pNPG) and 0.5% (wt/vol), carboxymethyl cellulose (CMC), hydroxyethylcellulose (HEC), xylan from beechwood (Sigma-Aldrich), β-glucan from barley (Megazyme). PASC was used at 1% (wt/vol); and FPAse assays were performed as previously described (86, 87). Reactions were performed with 1 – 20 μg of protein from the concentrated secretomes in 50 mM sodium acetate buffer pH 5.5, 50°C, during 0.5 – 1 h. Reactions with polymeric substrates were interrupted with 100 μL of DNS and the released reducing sugars were measured at 540 nm (88). Reactions with the synthetic substrates were interrupted with 100 μL of 1 M sodium bicarbonate and the released 4-nitrophenolate was measured at 405 nm. All assays were performed in triplicate. One enzyme unit (1 U) corresponds to the amount of enzyme that catalyzes the conversion of one μmol of substrate per minute under the assay conditions. The solid fraction (mycelia and substrate) of cultivations with Avicel and SCS were macerated with a mixture of 0.5 mm and 4 mm glass beads and 20% (wt/vol) KOH in methanol under vigorous vortexing for 2 min. The slurries were sonicated for 30 min and then incubated in a water bath at 65 – 70°C for 1.5 h. After cooling, water was added and ergosterol was extracted twice with hexane. Hexane was evaporated and ergosterol was recovered in methanol under sonication. Diluted samples (50 μL) were applied in an HPLC system equipped with A₂₈₀ detector coupled to a C18 reversed-phase column (Supelco, 25 cm x 4.6 mm, 10 μm) and isocratically eluted with 1 mL/min methanol. Total ergosterol in the samples was quantified based on peak area using standard ergosterol (Sigma).

The concentrated secretomes (20 μg) of the mutants carrying single deletions of AnLPMO9F, AnLPMO9G and AnCDH1 (cultivated in Avicel for 24 h) were prepared from in-gel tryptic digestion, as described above. Digested peptides were separated using RP-nanoUPLC (nanoAcquity, Waters) on a C18 column coupled to a Q-TOF Premier mass spectrometer (Waters) with a nanoelectrospray ion source (89) and data analysis was performed as previously described (5).

Conidia suspensions (1 × 10⁶ conidia/ml final concentration) of the A. nidulans recombinant strains expressing AnLPMO9C, AnLPMO9D, AnLPMO9F and AnLPMO9G were precultivated in liquid MM with glucose, as described above. Mycelia was removed by filtration, washed with distilled water and transferred to Erlenmeyer flasks (250 mL) containing 50 mL liquid MM pH 6.5 with 1% (wt/vol) Avicel. After cultivation for 24 h, 180 rpm, 37°C, maltose was added at 2% (wt/vol) final concentration and cultivation proceeded for additional 24 h at the same conditions. Alternatively, induction by maltose was performed concomitantly to the cultivation in Avicel in 24 and 48 h cultivations at the same conditions. Cultivation was performed using four replicates. The crude filtrates were centrifuged (10.000g, 40 min, 4°C), filtered (0.45 μm), concentrated when necessary, (10 kDa cutoff Amicon) and used for protein and activity assays (as described above), and for lignocellulose saccharification.

The AnLPMO9-enriched secretomes were evaluated by replacing 15% of protein load from commercial enzymes Celluclast (Novozymes) + glucosidase from Aspergillus niger (Merck) in the SCS degradation (90), following previously established reaction conditions (91). The products were analyzed by high-performance liquid chromatography (HPLC) in a 1260 Infinity system (Agilent) equipped with a refractive index (RI) detector coupled to an Aminex HPX 87H column (300 mm x 7.8 mm) at 35°C, 0.6 mL/min⁻¹ flow rate and 5 mM sulfuric acid isocratic elution. Glucose, cellobiose and xylose were quantified by external calibration.

Proteins were quantified by the BCA method (Thermo Scientific) and visualized in 12.5% SDS-PAGE (92) after staining with Coomassie blue R-25.

The mass spectrometry proteomics data of A. nidulans A773 growing in different substrates at different periods of cultivation have been deposited to the ProteomeXchange Consortium via the PRIDE (93) partner repository http://proteomecentral.proteomexchange.org/cgi/GetDataset with the data set identifier PXD031886; and data of A. nidulans knockout mutants growing in Avicel with the identifier PXD031881.