Biochemical and inhibitor analysis of recombinant cellobiohydrolases from Phanerochaete chrysosporium

Abstract

The demand for greener energy sources necessitates the development of more efficient processes. Lignocellulosic biomass holds significant potential for biofuels production, but improvements in its enzymatic degradation are required to mitigate the susceptibility of enzymes by reaction products and pretreatment impurities. In this work, two cellobiohydrolases (CBHs) from the basidiomycete Phanerochaete chrysosporium (PcCel7C and PcCel7D) were heterologously expressed, characterized, and analyzed in the presence of their products (glucose and cellobiose) and harmful compounds commonly found in industrial processes (phenolics), as well as their adsorption to lignin and cellulose. The enzymes exhibited an optimum temperature of 55 °C and displayed a pH profile similar to the model CBHI from Trichoderma reesei (TrCel7A). Activity decreased consistently for all CBHs in the presence of cellobiose, while glucose significantly impacted the basidiomycete CBHs. Phenolic compounds with a higher content of OH groups were found to be more detrimental to the enzymes, with the location of the OH group on the phenolic ring playing a crucial role in enzyme deactivation. Molecular docking simulations predicted that the product-binding site of CBHs has the highest affinity for interaction with phenolics; however, they are unlikely to interact at this site in the presence of substrate. PcCel7C and PcCel7D exhibited poorer adsorption on cellulose compared to the TrCel7A enzyme. These findings provide insights into how the structure of CBHs influences their susceptibility to inhibitors and deactivating compounds present in saccharification reaction medium.

Introduction

Several research studies have been conducted to increase the contribution of lignocellulosic biomass to the global energy matrix [1]. However, converting its polysaccharides into monosaccharides necessitates a pretreatment process followed by enzymatic hydrolysis. To ensure the economic viability of these processes, scientific and technological advancements are essential for improving the industrial production of monosaccharides from lignocellulose [1,2,3].

Beyond the inhibitory effect of soluble sugars on enzymatic hydrolysis efficiency, which is significantly reduced due to product inhibition, other factors also play a crucial role. These include thermal inactivation, unproductive adsorption onto insoluble lignin, and inhibition by soluble lignin-related compounds (LRCs). Most pretreatment methods generate soluble lignin degradation products, with the concentration and type of derivatives depending on the lignocellulosic substrate, the pretreatment strategy (acid, alkaline, steam explosion), the severity of physicochemical conditions, and solid loading [4]. Although earlier studies suggest that soluble lignin compounds are less inhibitory to cellulases compared to insoluble ones [5], LRCs have a more detrimental effect than several low-molecular-weight reaction contaminants [6] due to their acceleration on thermal denaturation, irreversible inhibition, and cellulase precipitation [7].

During the saccharification step, inhibitors and undesired compounds, including LRCs and insoluble lignin, are present and affect cellobiohydrolases (CBHs), which play a pivotal role in cellulose depolymerization. CBHs exhibit processivity, continuously breaking down insoluble cellulose chains, dismantling the fibril structure, and releasing soluble cellobiose into the reaction medium [8]. Modern enzymatic cocktails, rich in CBHs from ascomycetes, primarily produced by fungi of the genus Trichoderma sp [8, 9], have demonstrated efficiency in hydrolyzing pretreated biomass. However, nonproductive adsorption onto lignin and enzyme inhibition by low-molecular-weight lignin degradation products (i.e., LRCs) remain challenging. These issues necessitate the use of high enzyme loadings, increasing costs to achieve efficient digestion outcomes [8, 10,11,12].

In natural environments, ascomycetes and basidiomycetes colonize different substrates. Basidiomycetes are renowned for their extensive array of oxidative enzymes, which enable them to thrive on highly lignified materials. In contrast, ascomycetes generally colonize substrates that have been pre-degraded by basidiomycetes or are inherently less lignified [8, 13]. This suggests that CBHs derived from basidiomycetes may exhibit distinct characteristics regarding unproductive lignin adsorption and susceptibility to phenolic compounds from lignin due to their harsh native reaction environment, sequences, and structural features. For instance, PcCBHs have a higher dissociation rate constant from cellulose compared to TrCel7A, which confers higher hydrolytic velocity and increased endo-initiation activity to PcCBHs. This enables them to efficiently attach to free end chains of cellulose, resulting in a higher cleavage of amorphous cellulose [14]. However, as far as we know, little attention has been given to their performance in cellulose catalysis in the presence of low-molecular-weight lignin degradation products or their susceptibility to insoluble lignin compared to ascomycete CBHs. Investigating enzymes from various sources present the opportunity to develop more efficient enzymatic cocktails for industrial applications.

In this study, we aimed to clone and express all seven CBH-coding genes from P. chrysosporium. The large-scale production of two CBHs from this white-rot basidiomycete was successfully achieved through heterologous expression in Aspergillus nidulans, followed by comprehensive biochemical evaluation. The recombinant enzymes were systematically compared based on their susceptibility to glucose, cellobiose, and phenolic compounds. Molecular docking analysis was conducted to elucidate the preferred interaction sites of phenolics during the enzymatic reaction. Additionally, their carbohydrate-binding modules (CBM) were experimentally assessed for the adsorption onto lignin and cellulose. The objective was to identify characteristics that could enhance enzyme robustness and facilitate the development of more efficient enzymatic cocktails for the saccharification process of pretreated biomass, specifically focusing on sugarcane bagasse pretreated under mild conditions, which still contains significant amounts of inhibitors, insoluble lignin, and its soluble monomers.

Methodology

Microbial strains and chemicals

The A. nidulans strain A773 (genotype pyrG89; wA3; pyroA4) was obtained from the Fungal Genetic Stock Center (FGSC, Manhattan, USA). A. nidulans A773 was cultured in Petri dishes with minimal medium (1 L), composed of 50 mL of 20 × Clutterbuck salt solution (6.0 g/L NaNO₃, 1.5 g/L KH₂PO₄, 0.5 g/L KCl, and 0.5 g/L MgSO₄), 1 mL of 1000 × trace elements (2.2 g of ZnSO₄·7H₂O, 1.1 g of H₃BO₃, 0.5 g of MnCl₂·4H₂O, 0.5 g of FeSO₄·7H₂O, 0.16 g of CoCl₂·5H₂O, 0.16 g of CuSO₄·5H₂O, 0.11 g of Na₂MoO₄·4H₂O, and 5.0 g of Na₂EDTA in 100 ml), and 1% (w/v) glucose at pH 6.5 and 37 °C, supplemented with pyridoxine (1 mg/L), uracil and uridine (2.5 mg/L each) added by agar 2% (w/v), as described before [15]. The P. chrysosporium strain RP-78 (ATCC MYA-4764) was a sample maintained at the Department of Biotechnology at Lorena School of Engineering, University of São Paulo (EEL-USP, Brazil). P. chrysosporium was cultured in Petri dishes with Norkrans medium [16] and incubated at 37 °C for 7 days.

Plasmid replication was carried out using Escherichia coli strain NEB Turbo-competent (New England BioLabs, Ipswich, USA), cultured in lysogeny broth (LB) medium with ampicillin (100 µg/mL) at 37 °C for 16 h and 180 rpm. Recombinant A. nidulans strains were cultured in expression medium composed by minimal medium supplemented with 3% (w/v) maltose and pyridoxine (1 mg/L) for 48 h at 37 °C without agitation [17]. The CBHI enzyme from T. reesei (TrCel7A) was kindly provided by Matti Siika-aho from Technical Research Centre of Finland (VTT, Espoo, Finland).

The model LRCs used in this study included ferulic acid (FER), p-coumaric acid (COU), 4-hydroxybenzoic acid (HBA), 4-hydroxybenzaldehyde (HBL), syringic acid (SGA), syringaldehyde (SGL), vanillin (VNL), vanillic acid (VNA) and isomeric forms of dihydroxybenzoic acids (2,3-, 2,4-, 2,5-, and 2,6- dihydroxybenzoic acids), as well as glucose and cellobiose were obtained from Sigma-Aldrich. A stock solution of LRCs (100 mM) was prepared by dissolving them in water adjusted to pH 8.0 with 10 M NaOH. Sodium acetate buffer (pH 5.0) was added at a final concentration of 50 mM. When necessary, the pH of the solution was adjusted to 5.0 with 4 M HCl and heated to 50 °C for 5–10 min to facilitate the dissolution of the compounds.

The 4-nitrophenyl (pNP), 4-nitrophenyl β-D-cellobioside (pNPC), 4-nitrophenyl β-D-glucopyranoside (pNPG), 4-nitrophenyl β-D-xylopyranoside (pNPX), carboxymethyl cellulose (CMC) and Avicel (Avicel PH101) were purchased from Merck.

Cloning, heterologous expression, purification, and mass spectrometry analysis

Genomic DNA from P. chrysosporium was extracted using the Wizard Genomic DNA Purification kit (Promega, Madison, USA). This DNA served as the template for amplifying the full-length PcCel7C (accession number Z22528/CAA80253) and PcCel7D (accession number Z29653/CAA82761) genes, including introns and regions corresponding to the secretion signal peptides, through polymerase chain reaction (PCR). The amplified fragments were generated using the oligonucleotides (Exxtend Biotechnology Ltd., Campinas, Brazil) as follows: PcCel7C forward (5′–CATTACACCTCAGCAATGTTCCGCACTGCTACTTTGCTCGCATTC–3′), PcCel7C reverse (5′–GTCCCGTGCCGGTTATTAGTAGCACTGGGAGTAGTCTGCCATGGA–3′), PcCel7D forward (5′–CATTACACCTCAGCAATGTTCCGCGCCGCCGCACTC–3′) and PcCel7D reverse (5′–GTCCCGTGCCGGTTATTAGCTAAAGACACTCCGGGCGAGCAGACT–3′). Regions in bold facilitate gene assembly into the pEXPYR expression vector using the Gibson Assembly (GA) method, as previously described [15].

The GA reactions were used to transform NEB Turbo-competent E. coli through heat-shock. Ampicillin-resistant colonies were screened by colony PCR. Positive colony for PcCel7C as well for PcCel7D was selected for recombinant plasmid propagation by culturing in a 50 mL conical tube containing 10 mL of LB medium supplemented with ampicillin (100 µg/mL) for 18 h at 37 °C and 180 rpm. Plasmid DNA was recovered using the Wizard Plus SV Minipreps DNA Purification System kit (Promega).

A. nidulans A773 protoplasts were prepared [15] and transformed using polyethylene glycol-mediated integration with 15–20 µg of recombinant plasmid containing PcCel7C or PcCel7D encoding genes [18]. Positive recombinant A. nidulans strains, which had reverted uracil and uridine auxotrophic marker, were selected and analyzed for their ability to secrete PcCel7C and PcCel7D in the expression medium (see Sect. "Microbial strains and chemicals"). These strains were subsequently used to produce the recombinant proteins.

For the larger-scale production of recombinant proteins, spores were obtained by inoculating the recombinant strain for PcCel7C and PcCel7D onto a Petri dish containing minimal medium supplemented with pyridoxine (1 mg/L). The Petri dishes were then incubated at 37 °C for 48 h, as previously described. The pre-inoculum was cultured by harvesting spores to achieve a concentration of 10⁸ spores/mL in a 50-mL centrifuge tube. The spore solution was transferred to a 250-mL Erlenmeyer flask containing 100 mL of minimal medium (see Sect. "Microbial strains and chemicals"), followed by incubation at 37 °C with agitation at 150 rpm for 16 h.

Recombinant protein production was carried out utilizing the Minifors 2 bioreactor (Bench-Top Bioreactor, Infors, Switzerland), containing 3 L of minimal medium (see Sect. "Microbial strains and chemicals") supplemented with 3% (w/v) maltose (WE Consultoria, Porto Alegre, Brazil) and pyridoxine (1 mg/L). The pre-sterilized equipment, including the pH and dissolved O₂ probes previously calibrated, received a 1% (v/v) pre-inoculum into the culture medium. The pH of culture medium was maintained at pH 6.5 by acid (8 M HCl) and base (6 M NaOH) addition. The incubation period ranged from zero to 112 h for PcCel7D and up to 140 h for PcCel7C.

Samples were withdrawn periodically to analyze protein concentration using the DC™ Protein Assay (Bio-Rad Laboratories, CA, USA), reducing sugar concentration using the 3,5-dinitrosalicylic acid (DNS) method [19], enzyme activity on pNPC, and protein integrity analysis using 15% SDS-PAGE [20].

The recombinant PcCel7C and PcCel7D were concentrated through ammonium sulfate precipitation (at 70% saturation) and subsequently purified using two chromatographic steps, including anion exchange and molecular weight exclusion. The first step was performed with a DEAE-Sepharose® CL6B resin (GE Healthcare Life Sciences, MA, USA) packed in an Econo-Pac® Chromatography Column (Bio-Rad, CA, USA) and the molecular weight exclusion step was performed in the system Äkta Pure 25 M using a HiPrep™ 26/60 Sephacryl® S-200 HR (GE Healthcare Life Science, MA, USA) column, as detailed previously [17]. The purity of the samples was evaluated using SDS-PAGE, and their activity on pNPG and pNPC was assessed (see Sect. "Biochemical characterization"). To confirm the identity of the recombinant proteins, the bands corresponding to PcCel7C and PcCel7D on the SDS-PAGE were excised and subjected to mass spectrometry analysis at the multiuser facility LaCTAD (Central Laboratory for High-Performance Technologies in Life Sciences, Campinas, Brazil).

Biochemical characterization

To evaluate the enzymatic activity of purified PcCel7C, PcCel7D, and TrCel7A on various substrates, assays were conducted employing 1 µM of enzyme in combination with 1 mM pNPC, pNPG and pNPX in 50 mM sodium acetate buffer (pH 5.0) and with a total reaction volume of 100 µL. The reactions were incubated at 50 °C for 20 min under static conditions, after which 100 µL of 1 M Na₂CO₃ was added to terminate the enzyme activity. The absorbance of the released pNP was subsequently measured at a wavelength of 405 nm. Enzymatic activity unit (U) was defined as the amount of enzyme necessary to release 1 μmol of pNP per minute, determined using a standard curve established with pNP.

For cellulosic substrates such as Avicel, phosphoric acid swollen cellulose (PASC) [21], and CMC, a concentration of 10 g/L was used in a total reaction volume of 500 µL for Avicel and 200 µL for PASC and CMC. The reactions using these substrates were then incubated with 1 µM of enzyme (and 0.5 µM for PASC) at 50 °C for 1 h (for Avicel and PASC) and 10 min (for CMC) under agitation at 800 rpm in the ThermoMixer C coupled with a ThermoTop lid (Eppendorf, Hamburg, Germany). To monitor the release of reducing sugar, the reaction mixtures were centrifuged, and DNS was added to the supernatant. The resulting mixture was boiled for 5 min, and absorbance was measured at a wavelength of 545 nm [19]. Enzymatic activity unit (U) was defined as the amount of enzyme required to release 1 µmol of reducing sugar per minute, determined using a standard curve established with cellobiose.

The kinetics parameters were determined using pNPC as a substrate at concentrations ranging from 0.1 to 5 mM (final volume of 100 µL) and 0.5 µM of PcCel7C and PcCel7D at 50 °C in 50 mM sodium acetate buffer (pH 5.0). The reactions were stopped by adding 100 µL of 1 M sodium carbonate after 10 min of incubation for concentrations of 0.1 to 0.3 mM, and after 20 min for concentrations of 0.5 to 5 mM. The kinetic parameters K_M and k_cat were calculated using GraphPad Prism 5 software (GraphPad Software) through non-linear regression based on the Michaelis–Menten equation.

To determine the optimal pH, enzymatic activity was evaluated utilizing four distinct buffers: 50 mM sodium citrate (pH 2–3), 50 mM sodium acetate (pH 4–5), 50 mM sodium phosphate (pH 6–7), and 50 mM Tris–HCl (pH 8–9) at a constant temperature of 50 °C. For pH stability assessment, the enzymes were incubated in 50 mM sodium acetate (pH 5.0) and sodium phosphate (pH 6.0) over periods of 0, 8, 24, 48, and 72 h at 30 °C. The optimal temperature was determined in a range of 30 to 70 °C in 50 mM sodium acetate buffer (pH 5.0), while thermal stability was investigated at 50 and 60 °C in 50 mM sodium acetate buffer (pH 5.0) without the substrate for a period of up to 6 h. The residual activity was measured adding 1 mM of pNPC, incubating for 20 min at 55 °C, terminating the reaction by adding 100 µL of 1 M Na₂CO₃, and measuring the absorbance at 405 nm. Throughout these experiments, pNPC was used as the substrate with an enzyme loading of 1 µM under static conditions and a total reaction volume of 100 µL. All the experiments were performed in triplicates.

Decrease in activity by glucose and cellobiose

The reactions to investigate the impact of glucose and cellobiose concentrations on the activity of CBHs (1.45 µM) were conducted using 1 mM pNPC in a 50 mM sodium acetate buffer (pH 5.0). The procedure entailed varying concentrations of glucose and cellobiose (ranging from 0 to 1.5 M and 0 to 20 mM, respectively) at a temperature of 50 °C for 20 min under static conditions and were terminated as previously described.

Activity of CBHs in presence of lignin-related compounds

The endo-activity of PcCel7C and PcCel7D (0.2 µM each) was assessed in the presence of LRCs at a concentration of 30 mM, utilizing PASC (10 g/L) as the substrate. The incubation period lasted 60 min at 50 °C and 800 rpm in a 50 mM sodium acetate (pH 5.0) and total reaction volume of 500 µL. The selection of 30 mM concentration (approximately 5 mg/mL) for analyzing the effects of phenolic compounds was based on prior research [10, 12]. TrCel7A (1.45 µM) was assayed utilizing Avicel as a substrate (4 g/L), and the reaction proceeded for 48 h at 50 °C under 1000 rpm in 50 mM sodium acetate (pH 5.0) and total reaction volume of 500 µL, followed by termination through centrifugation at 10,000 rpm for 10 min. Protein concentration was determined using the DC™ Protein Assay (Bio-Rad Laboratories, CA, USA). The release of reducing sugars was monitored employing DNS method [19]. Controls containing Avicel and PASC in the buffer with LRCs were also integrated into the analysis. The enzymatic hydrolysis assay results were expressed in terms of increase or decrease in the enzyme activity, as defined by Eq. 1:

$$Increase\, or\, decrease\, in\, the\, enzyme\, activity \left(\,\%\right)=\frac{{A}_{phenolic\, compound}}{A}-1,$$

(1)

where "A_{phenolic compound}" refers to the enzymatic activity in the presence of the phenolic compound, and "A" represents the enzymatic activity without the phenolic compound.

Hydrophobic patch and docking analysis

The hydrophobic patches (hpatch) [22] were calculated utilizing Rosetta software version 3.12, employing the Fixed Backbone Design (fixbb) application with adherence to recommended constraints [22]. The three-dimensional (3D) structures of CBMs from PcCel7C and PcCel7D were modelled using AlphaFold 2 [23]. Molecular docking procedures were exclusively conducted by Maestro (Schrödinger, LLC, New York, NY, 2022-4). The 3D structures of PcCel7D and cellobiose (PDB ID 1z3t), LRCs (obtained from PubChem), and cellononaose (extracted from PDB ID 4c4c) were minimized under pH 5.0 conditions utilizing the Protein Preparation Workflow and Refinement, and LigPrep modules, respectively. Molecular docking was performed using Glide software, integrated into Maestro (version 2022-4). A grid of 36 Å was generated through the Receptor Grid Generation of Glide, enabling docking of the entire protein structure with ligands using default parameter. Free energies were computed using MM-GBSA within the PRIME software, also integrated into Maestro, with all atoms considered flexible and default parameters.

Adsorption on microcrystalline cellulose and lignin

Avicel was utilized as microcrystalline cellulose to generate adsorption isotherms. Prior the assay, Avicel underwent a through purification process, involving seven washes with distilled water followed by two washes with 50 mM sodium acetate buffer at pH 5.0 to eliminate any residual soluble sugars [24]. The washed Avicel was recovered by allowing it to settle by gravity. The lignin utilized in the adsorption experiments was obtained from sugarcane bagasse previously pretreated by steam explosion [25]. Shortly, pretreatment was performed in a pilot-plant reactor loaded with 10 kg sugarcane bagasse treated at 190 °C for 15 min. Pretreated material was extracted using 200 L of 1% (w/v) NaOH at 100 °C for 1 h. Dissolved lignin was further precipitated by addition of concentrated H₂SO₄ up to the liquor reached pH 2.0. Precipitated lignin was recovered using a press filter and dried at 60 °C. For the currently developed experiments, this lignin underwent seven washes with 50 mM sodium acetate buffer at pH 5.0. These washings involved incubation at 50 °C for 3 h under agitation of 800 rpm to remove impurities and low-molecular-weight lignin compounds. Following each washing cycle, the lignin was separated by centrifugation at 10,000 rpm for 10 min. The absorbance of the supernatant was continuously monitored until a consistent absorption profile was achieved, across wavelengths from 270 to 750 nm. In the assays, Avicel and lignin were loaded at concentrations of 10 and 20 g/L, respectively, while the enzyme concentration ranged from 0.05 to 0.4 mg/mL [11].

The binding of protein to Avicel was assessed at 25 °C for 1 h under 800 rpm, while the evaluation with lignin was conducted at 50 °C for 3 h. The protein initially present in the solution was recovered through centrifugation, and its initial and final concentrations were determined using the DC™ Protein Assay (Bio-Rad Laboratories, CA, USA). To mitigate potential interference from low-molecular-weight lignin released during the assay, free proteins in the supernatant were precipitated before the colorimetric assay. The precipitation underwent by adding 1:4 (v/v) sample to acetone (at −20 °C) and incubating at −20 °C for 1 h. The samples were centrifuged at 10,000 rpm at 4 °C for 10 min. The pellet was air-dried for 5 min and resuspended in 250 µL of 50 mM sodium acetate (pH 5.0). The amount of adsorbed enzyme was calculated by the disparity between the initial and final protein concentration in solution. The adsorption isotherms of CBHs were fitted to the Langmuir curve, which characterizes the adsorption phenomenon in heterogeneous systems (Eq. 2) [26, 27]:

$$P=\frac{{P}_{max}.k.{E}_{L}}{1+k.{E}_{L}},$$

(2)

where P represents the adsorbed protein content on the substrate (mg/g lignin), P_max represents the maximum adsorption capacity (mg/g lignin), k represents the Langmuir constant for affinity equilibrium (mL/mg), and E_L represents the free enzyme in the supernatant after the assay (g/L). The acquired adsorption data were fitted to the Langmuir non-linear regression using Origin 2020 software (OriginLab Corporation, Northampton, MA, U.S.A.). The assays were performed in triplicate.

Results

Gene cloning, heterologous expression, and identification of PcCel7C and PcCel7D

Seven genes encoding for CBHs were identified in the genome of P. chrysosporium, denoted as cel7A to cel7F and cel6A, based on data from the Joint Genome Institute [28] and previous literature (Supplementary Fig. 1A) [29,30,31]. Among these, six genes belong to the CBHs within the glycosyl hydrolases family 7 (GH7), while one is classified in the family 6 (GH6). In the current effort to heterologous express PcCBH genes in A. nidulans, cel7A, cel7B, cel7C, cel7D and cel6 were cloned but failed in terms of expression and secretion, while it was not possible to amplify genes cel7E and cel7F. In contrast, cel7C and cel7D genes were successfully amplified by PCR from the genomic DNA of P. chrysosporium (Supplementary Fig. 1B), assembled into the pEXPYR vector, transformed, and successfully expressed and secreted in A. nidulans.

The recombinant PcCel7C and PcCel7D enzymes displayed a molecular weight (MW) of approximately 66.0 kDa, as determined by SDS-PAGE analysis (Supplementary Fig. 2). In silico analysis predicted a MW between 53 and 55 kDa for these enzymes, suggesting that the observed increase in MW of the proteins likely resulted from post-translational modifications (Supplementary Table 2) [32]. LC–MS/MS analysis of the excised bands on SDS-PAGE gel of recombinant enzymes identified 33 peptides for PcCel7C with a sequence coverage of 38% and 26 peptides for PcCel7D with a sequence coverage of 44%, confirming the heterologous expression of recombinant PcCBHs in A. nidulans (Supplementary Fig. 2).

On average, 0.5 g/L of the PcCel7C and PcCel7D recombinant proteins was achieved in 3-L cultured bioreactors. Enzyme activities were discernible between 72 and 96 h of culturing, with continuous accumulation observed up to 140 h for PcCel7C (Supplementary Fig. 3A) and 112 h for PcCel7D (Supplementary Fig. 3B). At the end, the total accumulated protein in both cultures reached approximately 0.5 g/L, while the carbon source (glucose) and inducer (maltose) had depleted to minimal levels (around 10 g/L). The pH was carefully maintained at a constant level throughout all culturing periods to inhibit excessive secretion and activity of proteases [33]. The recombinant proteins in the cultured broth were concentrated by precipitation in ammonium sulfate, followed by dialysis, and purified through ion exchange and size-exclusion chromatography steps (Supplementary Fig. 4).

Analysis of substrate specificity, glucose and cellobiose inhibition, and the impact of pH and temperature on the enzymatic activity of PcCel7C and PcCel7D

The biochemical characteristics of the purified recombinant PcCel7C and PcCel7D were compared with those of TrCel7A (CBHI from T. reesei) (Supplementary Fig. 4), a well-studied enzyme as reported in the literature [8, 34,35,36].

All enzymes exhibited low activity on pNPG as expected, while TrCel7A demonstrated the ability to hydrolyze pNPX (Fig. 1A). Substrate specificity evaluation revealed that PcCel7C exhibits higher activity on pNPC compared to PcCel7D and TrCel7A (Fig. 1A). Consistently, PcCel7C showed a predominance of endo-initiation activity due to its higher activity on CMC (Fig. 1B) and PASC (Fig. 1C) compared to TrCel7A. PcCel7D also displayed higher activity on CMC (Fig. 1B) and PASC (Fig. 1C) compared to TrCel7A, albeit to a lesser extent than observed for PcCel7C. The kinetic parameters were determined using pNPC for PcCBHs (Supplementary Fig. 5). PcCel7C demonstrated a higher affinity for pNPC than PcCel7D and similar k_cat values, resulting in a higher catalytic efficiency for pNPC hydrolysis (Supplementary Fig. 5).

Fig. 1

Substrate specificity for the activities of PcCel7C, PcCel7D, and TrCel7A using synthetic substrates (pNPC, pNPG, and pNPX) over a 20-min period (A), CMC (B), PASC (C), and Avicel (D) during 1 h at 50 °C

To assess the impact of known released products on the enzymatic activity, the hydrolysis of pNPC by the CBHs was monitored in the presence of glucose and cellobiose. The observed reduction in activity for PcCel7C, PcCel7D and TrCel7A in response to cellobiose was quite similar (Fig. 2A). A significant decrease occurred at 1 mM cellobiose, resulting in approximately 50% reduction in activity, with nearly complete inhibition observed at concentrations exceeding 10 mM. The inhibitory effect of glucose (Fig. 2B) was more pronounced in the P. chrysosporium CBHs, with a 67% loss of activity in the presence of 1.5 M glucose, compared to 30% for TrCel7A.

Fig. 2

Relative activity (%) curves for PcCel7C, PcCel7D and TrCel7A when cellobiose (A) and glucose (B) were added to the reaction medium using pNPC as substrate at 50 °C. Determination of the optimal parameters for the proteins PcCel7C, PcCel7D, and TrCel7A (1 µM) in relation to optimal pH (C), optimal temperature (D), thermal stability at 50 °C (E), and 60 °C (F) using pNPC as substrate

In terms of biochemical properties, the optimal pH for all three CBHs was similar, falling within the range of pH 5.0 to 5.5. However, TrCel7A exhibited better tolerance to acidic pH compared to the basidiomycete CBHs (Fig. 2C). The optimal temperature was found to be 55 °C for both PcCel7C and PcCel7D, while it was 60 °C for TrCel7A (Fig. 2D). Notably, PcCel7C demonstrated greater stability compared to PcCel7D and TrCel7A over a period of 4 h of incubation at 50 and 60 °C, retaining over 80% of its enzymatic activity (Fig. 2E and F). Furthermore, all CBHs maintained their maximum enzymatic activities for 72 h during incubation at pH 5.0 and 6.0 (Supplementary Fig. 6).

Effect of lignin-related compounds on CBH activity

The enzymatic activity of TrCel7A, PcCel7C and PcCel7D on cellulosic substrates was assessed in the presence of 12 model compounds representing low-molecular-weight phenolics. TrCel7A was evaluated by Avicel hydrolysis, as it presented the highest activity on this substrate (Fig. 1D). The activities of PcCel7C and PcCel7D suggested that PASC would be a more appropriate substrate in this context (Fig. 1C). Therefore, PASC was used for the analysis of PcCBHs.

Compounds 4-hydroxybenzoic acid (3), 4-hydroxybenzaldehyde (6) and hydroxycinnamic acids (p-coumaric and ferulic acid, 7 and 8) (Fig. 3) demonstrated the most significant impact on the activity of TrCel7A. This influence led to a reduction in activity ranging from 20 to 30%. Similarly, the activity of PcCBHs exhibited susceptibility to the same LRCs, following a comparable pattern (Fig. 4). Highly substituted LRCs with OCH₃ groups, such as syringic acid (1) and syringaldehyde (4), moderately enhanced the enzymatic activity of PcCel7D by 7 to 15% (Fig. 4). The slightly increase in TrCel7A activity by syringaldehyde (4) and vanillin (5) was not statistically significant compared to the others LRCs (Fig. 3).

Fig. 3

Effect of LRCs on the hydrolytic activity of TrCel7A (1.45 µM). The central line (0%) represents the control (TrCel7A and Avicel as substrate (4 g/L) without LRCs added to the reaction). The addition of 30 mM of LRCs in the reaction containing TrCel7A and Avicel at the same conditions as the control resulted in an increase (positive %) or a decrease (negative %) in TrCel7A activity compared to the control reaction (central line at 0%)

Fig. 4

Effect of LRCs on the hydrolytic activity of PcCel7C and PcCel7D (0.2 µM). The central line (0%) represents the controls (PcCel7C or PcCel7D and PASC as substrate (10 g/L) without LRCs added to the reactions). The addition of 30 mM of LRCs in the reaction containing PcCel7C or PcCel7D and PASC at the same conditions as the controls resulted in an increase (positive %) or a decrease (negative %) in PcCel7C or PcCel7D activity compared to the control reactions (central line at 0%) at a concentration of 30 mM on the enzyme activity (%) of PcCel7C and PcCel7D, each at a concentration of 0.2 µM. The enzymatic activity was measured on 10 g/L PASC for 1 h at 50 °C

Given that LRCs with phenolic OH groups but devoid OCH₃ groups exhibited the most detrimental effects on CBHs performance, the influence of various configurations of OH groups on the aromatic ring were investigated. To this end, assays were conducted using 2,3-, 2,4-, 2,5-, and 2,6-dihydroxybenzoic acids (compounds 9 to 12) (Fig. 4). Remarkably, the ortho-position of the benzene ring, as represented by 2,6-dihydroxybenzoic acid (compound 12), had the most pronounced impact on CBH activity. This configuration led to an 87% reduction in activity for PcCel7C and a 67% decrease for PcCel7D (Fig. 4).

Electrostatic surface charge of PcCBHs and docking analysis of interaction sites

Overall, PcCel7D enzyme demonstrated a slightly reduced susceptibility to inhibition by phenolic compounds in comparison to PcCel7C (Fig. 4). Sequence and structural analysis unveiled notable disparities in the solvent-exposed tunnel region between the two enzymes. Specifically, PcCel7D features a higher abundance of negatively charged amino acids, including Glu55, Asp57, Glu191, Asp381, and Asp403, contrasting with PcCel7C, which exhibited Gln55, Ala191, Asn381, Gln403 (Fig. 5A). This distinction was evident based on the standard pKa values (3.86 and 4.25 for Asp and Glu, respectively) at pH 5.0. Consequently, the solvent-exposed tunnel region of PcCel7D also exhibited a greater concentration of negative electrostatic potential (Fig. 5B).

Fig. 5

Evaluation of pKa for exposed amino acids and LRCs. A Identification of charged residues exposed on the surface of the catalytic domains (CD) of PcCel7C and PcCel7D. Regions in red represent negatively charged residues (Glu, Asp), while those in blue represent positively charged residues (Arg, Lys, and His) at pH 5.0. Figures were generated using PyMOL. B Electrostatic surface potential of catalytic domain of PcCel7C and PcCel7D was calculated using the Poisson–Boltzmann equations in the CHARMM-GUI program (PBEQSolver) [76] and visualized in PyMOL

To further investigate the preferred interaction sites of LRCs with the PcCBHs, preliminary docking simulations were performed. These simulations revealed that the catalytic and product sites play a significant role in the interactions of LRCs, particularly involving residues His223, Glu212, Arg240, and Trp373, which stabilize phenolic compounds through hydrogen bonding and π–π stacking (Supplementary Fig. 7). The calculated free energies of binding ranged from −18 to −36 kcal/mol. Among the LRCs, syringaldehyde demonstrated the highest affinity for the enzyme, exhibiting more hydrogen bonding possibilities as indicated by the free energy of binding, suggesting its potential to interact with PcCel7D (Supplementary Table 1).

Under experimental conditions, cellulose fiber and cellobiose are readily available at the onset of enzymatic activity. CBHs are significantly inhibited by the presence of cellobiose due to its strong affinity. Consistent with this observation, the free energy of binding calculations indicated that cellobiose has a superior affinity for the binding site, with a free energy of −50.70 kcal/mol, surpassing that of LRCs (Supplementary Table 1).

To predict the interaction sites of LRCs with PcCel7D when the tunnel is occupied by cellulose fiber, molecular docking analyses were conducted using cellononaose within the tunnel of PcCel7D (Fig. 6). In this scenario, LRCs predominantly clustered near the charged residues proximate to the tunnel and active site, particularly Asp365, Asp366, His367, and Thr192, establishing hydrogen bonding via the OH groups from the phenolic ring. Among the LCRs, COU and HBA were the compounds that interacted with the enzyme's outermost surface, primarily within hydrophobic or uncharged regions.

Fig. 6

Best scored sites of LRCs (sticks) interaction with PcCel7D (blue surface) with catalytic tunnel occupied by cellononaose (sticks, green). Negatively charged residues are highlighted in red, polar residues in blue, hydrophobic regions in green, and hydrogen bonds are indicated by purple arrows in the schematic representations of the ligand interacting with residues from PcCel7D. The computational analysis was performed by Maestro software

Adsorption on lignin and cellulose

Thus far, research on CBHs has primarily focused on their catalytic activity. However, for comprehensive enzyme characterization, the analysis of the CBM warrants attention, particularly concerning its adsorption characteristics on heterogeneous materials. The adsorption of CBHs onto lignin and cellulose was examined, leading to the development of adsorption isotherms (Fig. 7). Lignin derived from sugarcane bagasse [37], and Avicel as microcrystalline cellulose were employed in this analysis. The results showed that PcCel7C has a lower adsorption capacity onto lignin compared to PcCel7D and TrCel7A, while similar adsorption profiles were presented for PcCel7D and TrCel7A (Fig. 7A). Consequently, PcCel7C exhibited a maximum lignin adsorption capacity (P_max) of approximately one-third of the values observed for TrCel7A and PcCel7D (Table 1).

Fig. 7

Adsorption isotherms to binding on lignin (A) and cellulose (B) by PcCel7C, PcCel7D and TrCel7A

Table 1 Parameters of maximum adsorption capacity (P_max), Langmuir constant K (mL/mg), and the fitting of the non-linear regression curve using the Langmuir equation (R²) for the assays conducted with PcCel7C, PcCel7D, and TrCel7A

Clusters of hydrophobic amino acids on the surface of the proteins, referred as hydrophobic patches (hpatch), serve as interaction sites with insoluble lignin through hydrophobic contact and π–π stacking [22, 38]. To assess the susceptibility of CBM interaction with lignin for various CBHs, the hpatch for the CBM of PcCel7C (PcCel7C_CBM), PcCel7D (PcCel7D_CBM) and TrCel7A (TrCel7A_CBM) (PDB id 1cbh) was calculated. The modeled PcCel7C_CBM (hpatch = 0.8448) exhibited approximately half the hpatch value compared to modeled PcCel7D_CBM (1.5616) and TrCel7A_CBM (1.504), suggesting that PcCel7C_CBM may be less favorable for interacting with lignin, which corroborates the experimental findings. At lower protein concentrations (0.05 and 0.2 mg/mL), PcCel7C demonstrated a higher affinity and likely easier diffusion through the solution boundary on the lignin layer [39]. However, beyond this concentration range, saturation was reached at 3.5 mg of adsorbed protein per g of lignin, leading to decrease adsorption affinity for PcCel7C at higher protein concentrations (0.2 to 0.4 mg/mL).

Adsorption onto cellulose was also assessed. Although PcCel7C exhibited slightly reduced interaction with lignin, it also displayed lower affinity for cellulose (Fig. 7B). Interestingly, in the case of cellulose, the adsorption profile of PcCel7D resembled that of PcCel7C, while TrCel7A demonstrated significantly higher adsorption values (Table 1).

Discussion

The analysis of both coding (exons) and non-coding regions (introns) across all PcCBH genes reveals that cel7F and cel6 harbor a greater number of introns (4 and 6 regions, respectively) compared to the genes cel7A to cel7E (Supplementary Fig. 1A). This variation may potentially impact post-transcriptional modifications and the processing of these genes [40]. In fungi, as in higher eukaryotes, precise gene expression necessitates an accurate splicing process to excise introns and efficiently splice exons. This constitutes a critical step in pre-mRNA splicing, contributing to precise translation, proper folding, and successful protein secretion [41, 42].

Therefore, the cloning of PcCBHs genes was successful for cel7C and cel7D, which exhibited one of the lowest intron contents and are recognized as functional enzymes in the literature. Previous studies have demonstrated that PcCel7D (with minor levels of PcCel7C) can be obtained from the crude extract of P. chrysosporium when using Avicel as the carbon source in the culture medium, albeit requiring an extended growth period (10–14 days) due to its slow metabolism [43]. Additionally, there have been several attempts to heterologous express these genes in E. coli [44] and Saccharomyces cerevisiae [45], yielding low levels of functional enzymes. Hence, the A. nidulans expression system has emerged as an effective strategy for obtaining functional and high titers of PcCBHs, primarily for PcCel7C and PcCel7D.

The substrate specificity reveals findings consistent with previous research, indicating that the native PcCel7C enzyme exhibited a higher velocity of cellobiose production from PASC compared to PcCel7D and TrCel7A (Fig. 2C) [14]. In contrast, TrCel7A demonstrated superior performance on microcrystalline cellulose (Avicel) compared to PcCel7C and PcCel7D (Fig. 2D), which is already reported and attributed to its higher processivity degree on cellulose fibers [14]. Previous research on structural analysis of these enzymes have shown that PcCel7D contains deletions within its loop3 structure, resulting in a conformation characterized by a more flexible cellulose-binding tunnel in comparison to TrCel7A. This enhanced flexibility enables PcCel7D to exhibit significantly augmented endo-initiation activity, thereby showing greater efficacy in targeting the amorphous regions of cellulose [31, 46]. Similarly, it is predicted that PcCel7C possesses a more flexible loop2, due to the replacements of residues G193S and T196A relative to PcCel7D. This contributes to the higher flexibility of this region and alters cellulose affinity, culminating in a higher velocity of hydrolysis and increased rates of product formation on amorphous substrates [14, 30]. These observations are consistent with the present findings, indicating that the heterologous expression of basidiomycete enzymes in A. nidulans does not alter their intrinsic characteristics.

PcCel7C and PcCel7D exhibited kinetic values similar to CBHs from other basidiomycetes and to previous studies using purified PcCel7C (formerly CBH62) and PcCel7D from P. chrysosporium crude extract [47,48,49]. The results also corroborated the greater activity on pNPC (Fig. 1) by PcCel7C, which also showed greater catalytic efficiency compared to PcCel7D on this substrate (Supplementary Fig. 5).

The inhibition of CBHs by glucose and cellobiose was assessed, given that sugars released during the saccharification process exert a significant inhibitory effect on cellulases. Cellobiose, in particular, impacts CBHs significantly, even at millimolar concentrations [50, 51]. The reduction in enzymatic activity in the presence of cellobiose was consistent across all three enzymes. However, more pronounced discrepancies were observed with increasing glucose concentration wherein PcCBHs demonstrated lower tolerance to the monomer. It was observed that β-glucosidases exhibiting lower glucose tolerance are distinguished by a wider tunnel opening at the catalytic site [52, 53]. Conversely, β-glucosidases featuring a narrower and less accessible entrance to the catalytic site, coupled with a deeper cavity, tend to manifest higher glucose tolerance [52, 53]. Considering this, it could be hypothesized that the lower glucose tolerance of PcCBHs may be attributed to the presence of a more opened tunnel compared to TrCel7A [14, 31, 46].

PcCBHs displayed similar characteristics across varying pH and temperature conditions (Fig. 2C and D). Conversely, PcCel7C exhibited superior thermal stability (Fig. 2E and F). The substantial sequence homology observed between PcCel7C and PcCel7D (> 80%) likely underlies their analogous behaviors; nonetheless, it is noteworthy that PcCel7C is predicted to possess a greater abundance of O- and N-glycosylation sites compared to PcCel7D and TrCel7A. This disparity could potentially account for its heightened thermal stability (Supplementary Table 2) [54].

The comparative specificity of CBHs from basidiomycetes and ascomycetes on cellulose has been extensively studied. However, an unresolved issue concerns the susceptibility of CBHs from these fungal classes to inactivation and inhibition by low-molecular-weight lignin derivatives (LRCs), as well as their tendency for unproductive binding to lignin [12, 55, 56]. It is hypothesized that CBHs from basidiomycetes and ascomycetes differ in their susceptibility to these compounds, given that basidiomycetes are known to effectively colonize and degrade lignin-rich substrates in nature [8, 13]. This topic is particularly relevant because several pretreatment methods used in modern biorefineries produce lignin-rich substrates and soluble low-molecular-weight phenolic compounds derived from lignin [12]. These soluble lignin fractions are especially critical in consolidated processes where solids are not separated from the pretreatment liquids before enzymatic saccharification [57].

Therefore, the activity of the enzymes was comparatively analyzed in terms of susceptibility to low-molecular-weight phenolics. The model compounds included acids and aldehydes representing syringyl, guaiacyl and p-hydroxyl coumaryl lignin structures (compounds 1 to 6, Fig. 3). Ferulic acid and p-coumaric acid (compounds 7 and 8, respectively) represented common aromatics easily released into the pretreatment liquor from grass biomass sources [58, 59]. Dihydroxy benzoic acids (compounds 9 to 12) were tested to evaluate the relevance of dihydroxy versus methoxy-hydroxy substitution patterns.

The PcCBHs exhibited a reduction in activity ranging from 40 to 86% (Fig. 4), while TrCel7A showed a smaller decrease in activity, approximately 30% (Fig. 3). This difference in enzyme susceptibility could be attributed to variations in substrate type (amorphous or crystalline), protein content (0.2 and 1.45 µM), and structural characteristics (more open or highly covered tunnel). Further investigations are required to analyze the influence of these factors, but they are beyond the scope of this work. Our primary interest was to understand how the purified CBHs are affected by various LRCs in terms of their chemical structures.

The presence of phenolic OH groups in LRCs without OCH₃ group substitutions on the phenolic ring significantly inhibited/inactivated enzyme activity. This was particularly evident with 4-hydroxybenzoic acid, 4-hydroxybenzaldehyde, p-coumaric and ferulic acid (compounds 3, 6, 7 and 8, respectively) (Figs. 3 and 4), which showed a substantial decrease in enzyme activity. Phenolic OH groups can form hydrogen bonds with proteins [60, 61], leading to enzyme interaction and inactivation. Conversely, LRC with higher OCH₃ group substitutions, such as syringic acid and syringaldehyde (compounds 1 and 4, respectively), displayed a diminished inhibitory effect or even acted as enzyme activators. This trend is consistent with findings in cellulase cocktails and endo-xylanases, where compounds fully substituted with OCH₃ groups showed reduced inactivation potential [62]. The steric hindrance introduced by OCH₃ groups, due to their larger molecular volume, likely impedes the interaction between the phenolic OH groups and the enzyme, thus reducing the inactivating effect [62]. In the food industry, low concentrations of phenolic compounds have been observed to form a hydrophobic monolayer on the protein surface, enhancing hydrophobicity of the proteins [63]. This could potentially increase the activity of cellulases on cellulose by improving their interaction and mobility along the fibers. However, the exact mechanism and preferred interaction sites of phenolic compounds with cellulases remain unclear.

Overall, dihydroxy substitution in the aromatic acids led to a greater decline in the activity of PcCBHs compared to hydroxy-methoxy substitution (Fig. 4). The structure of dihydroxy phenolic acids allows for increased conjugation and polarity, resulting in stronger acids than their hydroxy-methoxy counterparts, and causing more detrimental effects on the enzymes [64]. Among the derivatives, 2,6-dihydroxybenzoic acid (compound 12) (Fig. 4) is capable of forming two intramolecular hydrogen bonds, lowering its internal energy and making it a much stronger acid compared to its isomers [64]. This increased acidity likely enhances its polarity, facilitating interactions via hydrogen bonding with protein residues. Conversely, 2,5-dihydroxybenzoic acid (compound 9), with OH groups in the meta-position, exhibited the least inhibitory effect. The OH groups on opposite sides of the aromatic ring prevents direct conjugation with the carboxylic group, hindering its ionization and resulting in a weaker acid. This arrangement also reduces the polarity of the compound, making it less likely to interact with the enzyme via hydrogen bonding [64]. In 2,4-dihydroxybenzoic acid (compound 11), the para-position of the OH groups create an electron-donating conjugation effect, which also weakens the ionization of the compound [64]. This reduced ionization lessens its ability to inhibit enzyme activity through hydrogen bonding interactions.

As PcCel7D exhibited slightly higher tolerance to LRC compared to PcCel7C (Fig. 4), the external surface charge of PcCBHs was investigated. The results demonstrated an increased presence of negative charges near the catalytic and substrate-binding sites (Fig. 5). This likely confers an advantage to PcCel7D by repelling phenolic compounds, some of which carry slight negative charges at pH 5.0 (Supplementary Table 3), thereby reducing their interaction with amino acids near the catalytic tunnel. However, the interaction sites between LRCs and CBHs remain unclear.

In the food industry, studies have shown that phenolic compounds, when oxidized in an alkaline environment, produce quinones. Quinones are electrophilic and can react with nucleophiles in polypeptide chains, such as lysine, methionine, cysteine, and tryptophan [60, 65,66,67,68]. Molecular docking analysis revealed that the product-binding site is likely to accommodate and interact with the LRCs (Supplementary Fig. 7). However, in a realistic scenario where cellulose fiber and cellobiose are present in the reaction medium, the binding affinities of the phenolic compounds are weaker for this site compared to cellulose-related derivatives, such as cellobiose (Supplementary Table 1). Consequently, it can be inferred that in the presence of substrate (and its products), LRCs are unlikely to interact with the active or product sites of CBHs due to their lower affinity (Supplementary Table 1). Instead, in the presence of substrate, LRCs tend to interact with the outermost surface of the protein (Fig. 6), leading to its destabilization and eventual inactivation during prolonged reactions. It was reported that the cleft on the opposite side of the catalytic tunnel of Penicillium oxalicum JU-A10-T CBH was the preferred binding site for lignin due to its hydrophobicity, formed by aromatic amino acids, mainly tyrosine [69]. Given the similar physical properties between lignin and LRCs, such as electrostatics, chemical structure, and aromatic behavior, it can be suggested that, as lignin interacts with hydrophobic patches on the outermost surface of CBHs, LRCs likely perform the same phenomenon, corroborating the previously described results from molecular docking (Fig. 6). These results indicate that enzymes with unique structural properties, such as thermal stability, fewer hydrophobic patches, and negative surface electrostatic characteristics, may be investigated to overcome the deleterious effects of inhibitor compounds in the reaction medium.

The adsorption of enzymes on lignin is a crucial factor, particularly considering that CBHs are the primary driving forces in commercial enzyme cocktails. The unproductive adsorption to lignin is an irreversible process and a frequently reported undesired interaction with lignocellulosic biomass, which drastically decreases enzyme availability to degrade cellulose [39, 70, 71]. This phenomenon primarily occurs through hydrophobic interactions, π–π stacking and hydrogen bonding, depending on temperature, pH, and the sources of both the enzyme and lignin [39, 70, 71]. To observe differences in the CBM among enzymes from different sources, it was essential to experimentally analyze the adsorption of CBHs onto lignin and cellulose. All the enzymes showed maximum adsorptions to lignin similar to that reported in the literature [55]. PcCel7C demonstrated slight less susceptibility to interact with lignin, while PcCBHs exhibited poor adsorption on crystalline cellulose compared to TrCel7A (Fig. 7).

The hydrophobicity of protein surface is reported as major factor influencing unproductive adsorption of cellulases on lignin [72]. The CBM is the primary responsible for interaction with lignin during nonproductive adsorption, in addition to its efficient adsorption on cellulose [36, 73]. A hydrophobic patch (hpatch) index can be computationally calculated to assess the content of hydrophobic patches on the protein surface, indicating its susceptibility to unproductive adsorption onto lignin [22, 72]. The calculated hpatch for PcCel7C_CBM suggests it has fewer hydrophobic amino acid clusters on its surface, which may contribute to its lower adsorption capacity on lignin. Additionally, PcCel7C was predicted to contain a highly glycosylated linker, potentially reducing the interaction of this region with lignin (Supplementary Table 3) by altering its charges and hydrophobicity [36].

TrCel7A_CBM, PcCel7C_CBM and PcCel7D_CBM exhibited differences in their three tyrosine residues, which are crucial for the interaction of these enzymes with cellulose [74]. The highly efficient binding of TrCel7A_CBM to cellulose is attributed to the three tyrosine residues on the planar surface of its CBM. In contrast, PcCel7D_CBM and PcCel7C_CBM have a substitution of Tyr5 to Trp5. Previous reports have indicated a notable negative impact on the interaction of the CBM with the OH groups of cellulose fiber when this residue change occurred [75]. This alteration reduces the potential for hydrogen bonding and decreases the affinity to cellulose.

The findings of this work could aid in investigating the potential of CBHs with increased endo-activity, to elucidate the mechanisms of saccharide and LRC interactions with different cellulases and their unproductive adsorption on insoluble lignin.

Conclusions

The cloning, expression, and large-scale production of both PcCel7C and PcCel7D in A. nidulans were successful, with both enzymes exhibiting a strong preference for hydrolyzing amorphous cellulose. In glucose inhibition assays, P. chrysosporium CBHs were more affected than TrCel7A, likely due to the openness of the tunnel and its accessibility to the glucose monomer. Phenolic compounds with higher OH:OCH₃ ratios in their structures caused more pronounced inhibition/deactivation effects on PcCBHs. The severe negative impact of 2,6-dihydroxybenzoic acid on both enzymes indicates that the arrangement of OH groups in the phenolic ring significantly affects interaction strength with the enzymes. PcCel7D exhibited slightly less susceptibility to LRCs and had a higher concentration of negatively charged residues exposed on the surface near the catalytic tunnel. This characteristic could aid in electrostatic repulsion against LRCs. Additionally, molecular docking simulations suggested that LRCs may interact with the outermost surface of protein when cellulose or cellobiose are present. The P. chrysosporium CBHs displayed weaker interactions with cellulose compared to the TrCel7A. This discrepancy is likely attributed to the substitution of one residue from the common tyrosine triad in the CBM for a tryptophan, which is more hydrophobic and has fewer hydrogen bonding possibilities. These findings highlight how the sequence and structural features of CBHs from different groups of microorganisms can influence their properties during the complex enzymatic saccharification of lignocellulose. Furthermore, this work demonstrated the comparative behavior of purified enzymes in the presence of LRCs, revealing their susceptibility to various phenolics arrangements.

Finally, despite P. chrysosporium, a white-rot basidiomycete, being one of the first microorganism to colonize lignocellulose in nature, its CBHs did not exhibit any special features that would confer greater tolerance to insoluble lignin or its soluble lignin-related monomers. It is likely that the oxidative enzymes arsenal of P. chrysosporium plays a primary role in its remarkable ability to colonize such materials in nature.