Skip to main content
NCBI home page
UKPMC Funders Author Manuscripts logoLink to UKPMC Funders Author Manuscripts
. Author manuscript; available in PMC: 2023 May 1.
Published in final edited form as: Enzyme Microb Technol. 2022 Feb 1;156:110000. doi: 10.1016/j.enzmictec.2022.110000

Expression and Characterization of a Family 45 Glycosylhydrolase from Fomitopsis pinicola and comparison to Phanerochaete chrysosporium Cel45A

Neus Gacias Amengual 1, Florian Csarman 1, Lena Wohlschlager 1, Roland Ludwig 1,*
PMCID: PMC7613719  EMSID: EMS155487  PMID: 35123123

Abstract

To efficiently decompose biomass, fungi have developed various enzymatic and non-enzymatic strategies and are a source of versatile biocatalysts. The endoglucanases in glycosyl hydrolase CAZy family 45 (GH45) are known for their small size, a high thermostability and a broad substrate specificity that has been employed in textile and detergent industries. Here we report the heterologous expression and characterisation of an GH45 endoglucanase from the brown rot Fomitopsis pinicola and its direct comparison to an already characterised GH45 from the white rot Phanerochaete chrysosporium. Both enzymes were recombinantly expressed in Pichia pastoris and purified by two chromatographic steps. The biochemical characterisation highlighted the acidophilic character, with an optimal pH of 4, and a preference for amorphous substrates as carboxymethyl cellulose (CMC) and substrates containing β-1,4-glucans rather than the previously reported β-1,3/1,4-glucans lichenan and β-glucan. The dominating products from β-1,4-glucans were C3–C6 oligosaccharides, whereas from β-1,3/1,4-glucans glucose was the main reaction product. From the characterisation no differences between the brown rot and the white rot GH45 was evident.

Keywords: characterisation, Fomitopsis pinicola, glycosyl hydrolase family 45, Phanerochaete chrysosporium, recombinant expression

Introduction

Wood decaying fungi are the most efficient and prevalent plant biomass decomposers in nature and they have developed diverse strategies to decompose and assimilate the biopolymers forming the lignocellulose matrix [1]. White rot species such as P. chrysosporium are able to mineralise the lignin fraction together with cellulose and hemicelluloses to CO2 and water by a set of hydrolytic and oxidative enzymes as well as non-enzymatic reactions [2]. In contrast, brown rot fungi such as F. pinicola preferentially depolymerise cellulose and hemicelluloses leaving behind a highly modified lignin. Brown rot evolution was accompanied by a reduction of key cellulolytic and ligninolytic enzymes [3]. Nevertheless, they have acquired alternative mechanisms to break down cellulose and extensively also modify the lignin via a chelator-mediated Fenton system (CMF) which has been hypothesized to confer an advantage in wood colonization [4–6].

Due to their efficient enzymatic portfolio which is listed in the carbohydrate-active enzyme database (CAZymes, http://www.cazy.org/) [7], white rot and brown rot decay mechanisms have been extensively explored as potential source of biocatalysts to valorise lignocellulosic biomass for the sustainable production of biofuels, bio-based materials and building blocks. One of the first enzymes applied in industrial processes as textile finishing, paper and pulp drainage or detergent preparations were endoglucanases (EC 3.2.1.4). These enzymes randomly cleaves within the amorphous regions of carbohydrate polysaccharides opening the structures for subsequent hydrolysis by exo-acting enzymes and [8]. In the CAZy database, enzymes with endoglucanase activity have been so far classified into 14 different families (GH5, GH7, GH8, GH9, GH10, GH12, GH26, GH44, GH45, GH48, GH61, GH74, GH124 and GH148) based on the protein sequence similarity and structure of the catalytic domain [7]. Glycosyl hydrolases belonging to family 45 (GH45) are industrially relevant enzymes used in textile and detergent industries for colour clarification, fuzz removal and fabric biopolishing [9]. GH45 endoglucanases are generally small in size (20-45 kDa) by comparison with other GH families except than GH12, which is suggested to provide an advantage for penetrating into smaller pores and cavities and exhibit a high thermostability [5]. GH45 is a widespread family of endoglucanases scattered among the tree of life with representatives found among fungi [10], bacteria [11] or insects [12]. Structural and phylogenetic analysis proposes a division of GH45 into three subfamilies A, B and C [13]. A few GH45 endoglucanases also feature a family 1 carbohydrate binding module. They are closely related expansins, which are extracellular cell-wall loosening proteins secreted by plants and involved in plant development [14]. They are also related with fungal swollenins (expansin-like proteins), acting on the crystalline regions of cellulose to promote swelling and a slight depolymerization [15].

The first enzyme classified into GH45 subfamily A was isolated from the supernatant of the thermophilic fungus Humicola insolens (HiCel45A). In 1993 Davies and coworkers reported the protein structure consisting of a catalytic groove containing an aspartic pair: Asp121 acting as catalytic acid andAsp10 as catalytic base in an anti-parallel position to Asp121. Upon substrate binding, a conformational change brings another aspartate residue (Asp114) in proximity to Asp121 (5.6 Å) increasing the hydrophobic environment and promoting the correct protonation state of the catalytic residues. Mutational studies carried by the same research group demonstrated that while the activity is abolished for the mutants D10N and D121N, only Kcat was reduced for D114N mutant [10]. Since then, 12 subfamily A members have been characterized. In contrast, only limited information is available for subfamilies B and C with only few members characterised to date. The best characterised member of subfamily C is the PcCel45A from the white rot fungus P. chrysosporium [13]. This fungal endoglucanase contains an aspartate residue (Asp114) mapping onto the strand β5 and forming part of the conserved motif termed HFD (histidine-phenylalanine, aspartic acid) present in GH45 family members and expansins [14]. For PcCel45A this residue is located at similar position to Asp121, the general acid residue of HiCel45A, but lacks the counterpart to the aspartic acid designated as proton acceptor (Asp10) or assisting residue (Asp114) present in subfamily A [16]. Instead, PcCel45A possesses an Asp85 within the β4 strand forming the substrate binding groove. The distance between Asp85 and Asp114 (8.0 Å) is comparable to that observed between Asp121-Asp10 (11.3 Å [17]) for HiEGV. However, the parallel localisation of Asp85 and Asp114 does not support the inverting mechanism as for subfamily A. Alternatively, it has been proposed that PcCel45A utilises an asparagine at position 92 (Asn92) as the general base in a “Newton’s cradle” proton relay mechanism. However, this residue is not conserved in all subfamily C members suggesting that the catalytic mechanism lacks some details [16].

GH45 enzymes possess wide substrate specificity [5]. In general, they exhibit low or negligible activity against crystalline cellulose and higher activity against amorphous cellulose preparations including non-natural substrates as carboxymethyl cellulose (CMC) and phosphoric swollen cellulose (PASC). Furthermore, they are also active on mixed β-1,3/1,4-glucans as barley β-glucan, lichenan, and glucomannan and inactive on xylan or xyloglucan [13,18].

In this study, we produced and characterised two endoglucanases of the GH45 family for a direct comparison to reduce experimental errors: the uncharacterised FpiGH45 from the brown rot F. pinicola and the previously characterised PcCel45A from P. chrysosporium. Both genes were successfully cloned and heterologously expressed in Pichia pastoris strain X-33. The proteins were biochemically characterised to verify if FpiGH45 is a member of subfamily C and to investigate its thermostability. Moreover, the substrate specificity and the hydrolysis patterns for both endoglucanases were determined to identify potential divergences between subfamily C GH45s from fungi with different lifestyles.

Materials and methods

Chemicals

Pichia pastoris X-33 strain, the EasySelect™ Pichia expression Kit and pPICZA vector were purchased from Invitrogen (Carlsbad, CA). Restriction enzymes were purchased from New England Biolabs (Ipswich, Massachusetts, USA). All chemicals were of the highest purity grade available and were purchased from Sigma-Aldrich (St. Louis, MO, USA) unless stated differently. Methanol and microcrystalline cellulose (MCC) were purchased from Merck (Kenilworth NJ, EEUU), carboxymethylcellulose sodium salt (CMC) with a degree of substitution of 0.60–0.95 was purchased from Fluka (Buchs, Switzerland), xylan (Carl Roth, Karlsruhe, Germany), glucomannan, lichenan, barley β-glucan, cellotriose, cellotetraose, cellopentaose and cellohexaose were purchased from Megazyme (Wicklow, Ireland). Kobe Agar-agar and yeast extract were purchased from Carl Roth and yeast nitrogenous base without amino acids (YNB) was obtained from VWR chemicals (Radnor, Pensylvania, USA). Phosphoric acid swollen cellulose (PASC) was prepared from MCC as described previously [19].

Bioinformatic analysis

Sequence homology and similarity of the amino sequence of FpiGH45 were performed using BLASTP algorithm against the NCBI protein database (https://blast.ncbi.nlm.gov/) [20]. For phylogenetic analysis, fungal GH45 enzymes previously classified into three subfamilies [5,13] were evaluated using NGPhylogeny server (https://ngphylogeny.fr) [21]. The sequence alignment was performed using Clustal Omega, trimmed by trimAI (0.5 gap threshold) (Figure S1). Prior to the alignment, the signal peptide sequences were predicted using SignalP5-0 (http://www.cbs.dtu.dk/services/SignalP/) [22] and removed. The phylogenetic tree was constructed using PhyML+SMS (statistical criteria using the Akaike Information criterion (AIC), and a subtree pruning and regrafting (SPR) tree topology employing a maximum likelihood method, with 500 bootstrap replicates and 5 random starting trees). The alignment and the resulting phylogenetic tree were visualized using ESPript 3.0 (https://espript.ibcp.fr) [23] and MEGA X respectively [24].

FpiGH45 structural model was predicted using AlphaFold artificial intelligence deep learning networks from the protein sequence (https://alphafold.ebi.ac.uk/) [25].

Identification of GH45 genes and construction of expression vector

For the heterologous expression, the genes EPT00417.1 from F. pinicola and BAG68300.1 from P. chrysosporium genomes were used [13]. Both genes were synthesised by BioCat (Heidelberg, Germany) integrated into the pPICZA expression vector flanked by a BclI restriction site located at the 5’-end of the gene and a NotI restriction site at the 3’-end with the native signal peptide and without codon optimization. For the selection of fungal transformants, the vector contains BleoR gene conferring resistance to Zeocin.

Transformation and selection of clones

Plasmids were linearized with SacI to target its integration into the aox1 gene locus of P. pastoris. The integration of the linearized plasmids was verified by colony PCR and agarose gel electrophoresis using a 0.8% agarose gel, Gel Loading Dye (6X NEB) and 2-Log DNA Ladder (NEB). The transformation of the expression cassette into P. pastoris X-33 was performed according to EasySelect™ Pichia expression guidelines. Briefly, transformations were performed by electroporation using 50 μL of competent cells in 0.1 cm electroporation cuvettes with 3.5 μg of linearized plasmid. Electroporation was achieved with a voltage of 1.5kV, 125 ohms of resistance and a pulse length of 3 ms using a Micropulser system (BioRad, Hercules, CA, USA). Subsequently, cells were transferred in 800 μL YPD supplemented with sorbitol (20% w/v peptone from casein, 10% w/v yeast extract, 4% w/v glucose, 2% w/v agar and 1M sorbitol) at 30°C for 2 h.

Expression and purification of the recombinant enzyme

The selection of the recombinant clones was performed on YPD agar plates (20 % w/v peptone from casein, 10 % w/v yeast extract, 4 % w/v glucose, 2 % w/v agar) containing zeocin as selective agent with a concentration varying from 100 to 1000 μg mL-1 (Invitrogen, Carlsbad, CA, USA). After two days at 30°C, zeocin-resistant P. pastoris transformants were screened for protein expression. Single colonies were used for the inoculation of 100 mL baffled flasks containing 25 mL YPD growth medium and incubated on a rotary shaker (140 rpm) for 18 h at 30°C. Yeast cells were harvested by centrifugation at 1000 × g for 10 min in sterile 50 ml tubes. The supernatants were removed, and the pellets resuspended in 1-L baffled flasks containing 250 mL BMGY medium (phosphate buffer 100 mM pH 6.0, 10 % w/v yeast extract, 20 % w/v peptone from casein, 13.4 % w/v yeast nitrogenous base without amino acids (YNB), 1 % v/v glycerol, 4 × 10-5 w/v D-biotin) and incubated at 110 rpm and 30°C. To induce protein expression, every 24 h cultures were supplemented with 50 % methanol to a final concentration of 1 % (v/v) using aseptic conditions. To select the best transformant, samples of 1 mL were collected. The protein expression was measured by the Bradford assay, the endoglucanase activity was determined by the slightly modified Nelson–Somogyi method described in [26] and supernatant was further analysed by SDS-PAGE.

Heterologous protein production and purification

The transformant that showed the highest extracellular endoglucanase activity was selected to produce the recombinant protein in a bioreactor. The process was performed using the Eppendorf BioFlo120 5-L fermentation system (Eppendorf, Hamburg, Germany) according to Pichia fermentation process guidelines (Invitrogen). The pH value was set to 5.0 and controlled by ammonia (25 %) and the temperature was kept at 30°C throughout the entire fermentation at a constant aeration of 1 vvm. To induce protein expression, a methanol feed was started after 26 h. To follow the process, 10 mL of the supernatant was collected at the end of each fermentation phase to determine: wet biomass (mg mL-1), protein concentration (g L-1) using Bradford assay, endoglucanase specific activity (U mg-1) and volumetric activity (U mL-1). Three days after induction with methanol, the culture medium was harvested by centrifugation 2000 × g for 30 min at 4°C to remove cells clarified by filtration. The supernatant was concentrated using a MiniKros Plus TFF filtration system (Repligen, Waltham, MA, USA) with a modified polyethersulfone membrane (mPES) and a molecular weight cut off (MWCO) of 10 kDa.

The concentrated supernatant was fractionated on a Phenyl-Sepharose FF column (250 mL, GE Healthcare, Chicago, IL, USA) equilibrated with a 200 mM sodium acetate buffer containing 1 M ammonium sulphate (pH 5.0). The concentrated supernatant was supplemented with ammonium sulphate to a final concentration of 1 M and loaded to the column at room temperature. The elution was performed with a gradient to 200 mM sodium acetate buffer (pH 5.0) [13]. Active fractions were pooled and concentrated using Vivaspin sample concentrators with 10 kDa MWCO and rebuffered to 20 mM potassium phosphate buffer (pH 7.0). An anion exchange column containing Source 15Q (19 mL, GE Healthcare) was used to elute the protein of interest by a linear gradient from 0 to 1 M NaCl in the same buffer in four column volumes. Finally, the enzymatic activity of the purified enzyme was tested by the Nelson–Somogyi method, and the enzyme purity verified by SDS-PAGE.

Electrophoretic analysis

SDS-PAGE analysis was carried out using Mini-PROTEAN TGX Stain-Free precast gels (Bio-Rad) according to the manufacturer guidelines. To determine the molecular mass, the Precision Plus Protein Dual Color standard marker (Bio-Rad) was used. For the deglycosylation of the expressed FpiGH45 and PcCel45A Endo Hf was used according to NEB protocol.

Enzymatic activity assay

The endoglucanase activity was measured by the Nelson-Somogyi method slight modifications [28]. The enzymatic activities towards different cellulosic substrates (CMC, α-cellulose and PASC) and complex polysaccharides (barley β-glucan and lichenan) were evaluated. In brief, 100 μL of suitably diluted enzyme (1 μM) was mixed with 150 μL of 1 mg mL-1 substrate in 50 mM sodium acetate buffer at pH 4.0. The reaction mixtures were incubated at 30°C for 30 min at 1200 rpm, unless otherwise indicated. The reaction was stopped by heating the samples to 99°C for 5 min followed by centrifugation (10000 × g for 5 min) to completely separate the supernatant from the rest of remaining solid substrate. Subsequently, 200 μL of Somogyi’s copper reagent was added and samples were incubated in a boiling water bath for 20 min. The reaction mixtures were cooled at room temperature, and 200 μL of Nelson’s arsenomolybdate reagent was added. The solution was carefully mixed, and 2.4 mL of water were added. After centrifugation at 10 000 × g for 5 min, the reducing sugars were quantified by measuring the absorbance at 540 nm. The absorbance values for the substrate and the enzyme banks were subtracted from the analysed sample and a glucose standard curve (0–1 mg mL-1) was used to calculate the sugar release. One unit of enzyme was defined as the amount of protein releasing 1 μmol of glucose per min.

Effect of pH and temperature

The apparent optimal pH for FpiGH45 and PcCel45A, was evaluated using an enzyme concentration of 1 μM and 1 mg mL-1 CMC at 30°C. The buffer solutions used were 50 mM sodium citrate (pH 3.5–5.5), 50 mM sodium acetate (pH 4.5–6.0), 50 mM sodium phosphate (pH 6.0–7.5), 50 mM Tris citrate (7.0–8.0) and 50 mM sodium glycine (8.0–9.0) The temperature stability of the enzymes was determined by measuring the residual activity after pre-incubation of the purified enzyme in absence of substrate at 60°C and 80°C during different incubation times. The enzyme residual activity towards CMC was assayed as described above and expressed in seconds.

Differential Scanning Calorimetry

DSC measurements were performed on a MicroCal PEAQ-DSC System (Malvern Panalytical, Malvern, Worcestershire, UK) equipped with a 96-well plate autosampler cooled to 4°C. The experiments were performed under increased pressure (~4.2 bar). At least five buffer runs (for the buffer in the sample as for the reference cell) were performed at the beginning of the experiment to set the thermal history of the cells. FpiGH45 and PcCel45A (22 μM in 50 mM sodium acetate buffer, pH 5.0) were heated from 20 to 90°C with a temperature ramp of 1°C min-1 in high feedback mode. The reverse process was applied for the downscans. Subsequently, the rescans were measured up to 90°C with the same temperature ramp of 1°C min-1. The data analysis was performed using the MicroCal PEAQ-DSC Software version 1.4 (Malvern Panalytical, Malvern). In all analyses, the baseline correction was performed using the spline function and a non-two-state thermal unfolding model was fitted to the data points after subtraction of the buffer baselines and normalization for the protein concentration.

HPLC analysis of reaction products

The hydrolysis reactions of different substrates including CMC, α-cellulose, PASC, barley β-glucan, lichenan, cellobiose (1 mg mL-1) were analysed by HPLC. The hydrolysis reactions were terminated after 60 min following the procedure described previously. Samples were applied to an Aminex HPX-42A high-performance liquid chromatography (HPLC) column (7.8 × 300 mm), equilibrated in 60°C deionized water at 0.2 mL min-1 and prefitted with an Aminex guard cartridge (125–0507). Identification and quantification of the hydrolysis products (DP) was performed relative to cellulose oligosaccharide standards of various concentrations (1–20 ppm). The detected products are expressed in μM and the enzyme conversion in seconds.

Results

Sequence analysis of proteins from GH45 family

The gene of the previously characterised PcCel45A [13] was used as template to identify homologous genes in the F. pinicola genome (FP-58527 SS1) [3]. Using Basic Local Alignment Search Tool (BLAST) analysis [20], the query resulted in one hit with a sequence identity of 69%, a query coverage of 99 % for the gene EPT00417.1 from F. pinicola. The selected sequence showed the highest sequence identity (87.86%) for the endoglucanase from Fomitopsis palustris (GenBank accession number AVV62522.1) [27] classified in subfamily C. The identified sequence EPT00417.1 from F. pinicola was aligned with other members of the GH45 family and the expansin sequence from the tomato plant (AAC63088.1) [28] using the multiple sequence alignment program Clustal Omega trimmed by trimAI (Figure S1) [29]. Prior to the alignment, the signal peptide sequences were identified and removed [22]. The phylogenetic analysis coincides with previous reports showing that GH45 endoglucanases are divided into three subfamilies A, B and C and are related to plant expansins (α-expansin family) [5,13]. The analysis showed that FpiGH45 clusters in subfamily C together with PcCel45A and FpCel45 as reported in earlier studies and it is certainly distant form members of subfamily A. (Figure 1).

Figure 1.

Figure 1

Open in a new tab

Phylogenetic tree of fungal GH45 family members. The sequences were obtained from NCBI database and were previously classified into subfamily A, B, and C. For comparison, an expansin from Solanum lycopersium (AAC63088.1) was also included. The tree was constructed by PhyML+SMS employing the maximum likelihood method with 500 bootstrap replicates. The sequence of FpiGH45 obtained in this study (red arrow) was included in subfamily C.

Structural analysis of catalytic residues of FpiGH45

FpiGH45 structural model was predicted by AlphaFold, an artificial intelligence program relying on deep neural networks trained to predict protein properties from its genetic sequence [25]. FpiGH45 structure possess an anchor shape that constitutes of a six-stranded β-barrel and four α-helix domains forming a catalytic cleft running across the surface (Figure 2).

Figure 2.

Figure 2

Open in a new tab

Predicted structure for FpiGH45 depicted as a surface representation (A). The model was obtained by AlphaFold. Prior to modelling, the signal peptide of FpiGH45 was predicted using SignalP 5.0 and removed. The aromatic residues Tyr71 and Trp157 are highlighted in blue. The catalytic residues (Asn95, Asp117) and the residues involved in the proton relay pathway (Asp88, Tyr22, Trp98 and His117) are showed as sticks. (B) Sequence alignment of FpiGH45 with FpCel45 and PcCel45A. Prior to the alignment the signal peptides were predicted via Signal P5.0 and removed. The strictly conserved residues are shown in red blocks and similar residues in red text. The conserved amino acid residues are boxed in blue. PcCel45A secondary structure is depicted above the sequence alignment. The proposed catalytic residues are shown with orange stars, the residues involved in the proton relay network are indicated in cyan triangles and the aromatic residues are indicated by a blue circle.

The FpiGH45 structural model reveals the absence of Asp10 that has been suggested as a proton acceptor for members of subfamily A enzymes. Instead, an aspartate residue is found at position 88 (Asp88), which is homologous to Asp85 in PcCel45A with similar distances between Asp117–Asp88 (9 Å) and Asp114–Asp85 (8Å). The residues forming the HFD motif, present in GH45 members and plant expansins [14], are also present in FpiGH45 (His115 and Asp117). The aromatic residues Tyr67 and Trp154 identified as key residues for holding the substrate inside the catalytic cleft and stabilizing the product of enzymatic reaction in PcCel45A were present in FpiGH45 (Tyr71 and Trp157, Figure 2) [16].

Heterologous protein production and purification of FpiGH45 and PcCel45A

To investigate the enzyme properties, the cDNA encoding the genes EPT00417.1 (FpiGH45) and BAG68300.1 (PcCel45A) were cloned into the pPICZα vector for heterologous expression in Pichia pastoris X33 as already described for PcCel45A [13]. Transformants were plated onto selective PDA plated supplemented with Zeocin and resistant clones were selected after 48 h of incubation at 30°C. A screening for endoglucanase activity was performed using shake flasks containing BMGY medium supplemented with methanol. The transformant showing the highest extracellular endoglucanase activity was selected for fed-batch fermentations using glycerol as a carbon source and methanol for inducing the AOX1 promoter.

The time course of the heterologous expression of both enzymes is shown in Figure 3. For FpiGH45 the first endoglucanase activity could be detected after 47 h of fermentation (21 h after induction). The highest specific enzymatic activity (1.19 U mg-1) and protein concentration (0.26 g L-1) was achieved after 96 h of fermentation. During PcCel45A fermentation, endoglucanase activity was first detected after 49 h (20 h after induction) and the highest enzymatic activity (1.10 U mg-1) and protein concentration 0.33 g L-1 were obtained after 101 h of fermentation. After harvesting, the enzymes were filtered and concentrated using a hollow-fibre cross-flow filter module. Subsequently, ammonium precipitation, hydrophobic interaction chromatography and anion exchange chromatography was used to purify the enzymes which resulted in 65 mg of homogeneously purified FpiGH45 with a specific activity of 5.88 U mg-1 and 72.9 mg of PcCel45A with a specific activity of 7.10 U mg-1 determined in 50 mM sodium acetate buffer at pH 5.0 (Table 1).

Figure 3.

Figure 3

Open in a new tab

Recombinant expression of FpiGH45 (top) and PcCel45A (bottom) in 5-L fermentations monitoring wet biomass (mg mL-1) (squares), protein concentration (g L-1) determined by Bradford (slash line), enzymatic activity (U mg-1) measured with the Nelson-Somogyi method (diamonds) and volumetric activity (U mL-1, dotted line, filled triangles). Error bars indicate the standard deviation from triplicates.

Table 1.

Purification strategy for FpiGH45 (top) and PcCel45A (bottom). The protein concentration was determined using the Bradford assay. PF…purification factor.

Purification step Volume [mL] Protein conc [mg mL-1] Total protein [mg] Total activity [U] Specific activity [U mg-1] PF Yield [%]
FpiGH45 Supernatant 3020 0.26 785 938 1.19 1 100
(NH4)2SO4 precipitation 900 0.56 702 905 1.28 1.09 96
Phe-Sepharose (HIC) 310 0.93 288 736 2.55 2.12 78
Q-Source 15 (AEX) 48 4.51 216 910 4.21 3.55 97
Vivaspin (10 kDaMWCO) 10 6.5 65 382 5.88 5.0 40
PcCeI45A Supernatant 3300 0.33 1089 1198 1.10 1 100
(NH4)2SO4 precipitation 1000 0.69 690 998 1.45 1.31 83
Phe-Sepharose (HIC) 350 0.82 287 918 3.20 2.90 76
Q-Source 15 (AEX) 120 0.91 110 548 4.99 4.53 45
Vivaspin (10 kDaMWCO) 17 4.29 72.9 517 7.10 6.45 43
Open in a new tab

The purity of both enzymes was verified by SDS-PAGE. The molecular weight of FpiGH45 and PcCel45A estimated from the gel elution was approximately 18 kDa for both proteins (Figure 4). To investigate if P. pastoris glycosylated the expressed enzymes, which both show no N-glycosylation motifs, the enzymes were treated with Endo Hf for 1 h at 37°C and analysed by SDS-PAGE. The determined molecular mass is in good agreement with the theoretical values calculated from the protein sequences without signal peptides (FpiGH45 18.591 kDa, PcCel45A 18.984 kDa). The measured molecular mass of FpiGH45 was found to be 18 kDa and that of PcCel45A 18.2 kDa. None of the two expressed endoglucanases showed any sign of glycosylation. The occurrence of O-glycosylation, for which some potential sites are available on both enzymes, also seems not supported by the measured molecular masses that are below the calculated ones.

Figure 4.

Figure 4

Open in a new tab

SDS-PAGE of the purified and deglycosylated FpiGH45 and PcCel45A FpiGH45 (Lane 2), Endo Hf FpiGH45 (Lane 3) and PcCel45A (Lane 6), Endo Hf PcCel45A (Lane 7). The Precision Plus Protein Dual Color Standard (Bio–f treated samples (Lane 3 and Lane 7) show a band at 70 kDa corresponding to its molecular mass.

Biochemical characterization

The pH profiles of FpiGH45 and PcCel45A were determined using 1 μM of purified enzyme and carboxymethyl cellulose (CMC) as substrate at 30°C using four different buffer solutions over the pH range 3.5–9.0. The pH-optimum for both enzymes was 4.0 in 50 mM sodium acetate buffer. The activity at pH 5 is 92 % of the maximum and both enzymes retained approximately 75 % of their activity between 3.5 and 5.5. Above pH 5.0 the activity drops monotonously. At pH 8.0 about 10 % residual activity was retained for FpiGH45 and 5 % for PcCel45A, at pH 9.0 no activity could be detected. (Figure 5).

Figure 5.

Figure 5

Open in a new tab

Determination of pH optimum for FpiGH45 (left) and PcCel45A (right). The turnover number (TN) was measured by incubating 1 μM of purified enzyme with 1 mg mL-1 CMC for 30 min at 30°C in a buffer system ranging from 3.5 to 9.0. The hydrolysis of the substrate was analysed by the Nelson–Somogyi method.

Substrate specificity

The two endoglucanases were evaluated for their apparent ability to hydrolyse a range of cellulosic and hemicelullosic substrates (Table 2). FpiGH45 converted the same substrates as PcCel45A and both enzymes characterised in parallel also showed similar activities towards these substrates. The activity, given as turnover number (TN) for both endoglucanases in 50 mM sodium acetate buffer pH 4.0, was detected towards the soluble substrate CMC followed by the dispersed PASC, both model substrates for endo-acting cellulases containing a high percentage of amorphous cellulose strands. Both enzymes were able to hydrolyse substrates containing β-1,3/1,4-glucans but in minor proportion. The lowest hydrolytic activities were measured for solid α-cellulose, with a few unordered regions, and glucomannan. No activity was detected for microcrystalline cellulose or xylan.

Table 2.

Substrate specificity of FpiGH45 and PcCel45A evaluated with cellulosic and hemicellulosic substrates. The enzymes (1.0 μM) were incubated with 1 mg mL-1 substrate in 50 mM sodium acetate buffer (pH 4.0) at 30°C for 30 min. The products were detected by the Nelson-Somogyi method. n.d. …not detected

Substrates FpiGH45 TN [s-1] PcCel45A TN [s-1] Main linkage type and solubility
CMC 0.18 ± 0.01 0.18 ± 0.01 β-(1,4)-D-Glc, amorphous, soluble
PASC 0.14 ± 0.008 0.13 ± 0.01 β-(1,4)-D-Glc, amorphous, dispersed
β-glucan from barley 0.12 ± 0.005 0.10 ± 0.008 β-(1,3)/(1,4)-D-Glc, soluble
Lichenan 0.09 ± 0.006 0.073 ± 0.007 β-(1,3)/(1,4)-D-Glu, soluble
α-cellulose 0.042 ± 0.008 0.042 ± 0.002 β-(1,4)-D-Glc, crystalline, insoluble
Glucomannan 0.007 ± 0.003 0.009 ± 0.001 β-(1,4)-D-Man, D-Glu, soluble
MCC n.d. n.d. β-(1,4)-D-Glc, crystalline, insoluble
Xylan n.d. n.d. β-(1,4)-D-Xylp, insoluble
Open in a new tab

Reaction products

The activity of both endoglucanases was additionally determined by HPLC detection of the hydrolysis products of the complex cellulosic and hemicellulosic substrates. The sum of the concentration of all detected products of a GH45 hydrolysate were used to calculate an observed turnover number (TNobs), which is a conservative estimate of the enzymes activity since it does not contain non-soluble reaction products such as C7 and above (Table 3). The obtained TNobs correlate very well with the soluble CMC and the disperse PASC, but are lower for β-1,3/1,4-glucans substrates such as β-glucan, lichenan or insoluble substrates as α-cellulose. The products formed by FpiGH45 upon hydrolysis of soluble CMC were C3–C6 oligosaccharides with C5 and C4 being the most prominent reaction products, but no C2 and very minor traces of C1 were detected. The same product distribution was also found for PASC and α-cellulose, however at a much lower concentration. Substrates containing β-1,3/1,4-glucans (barley β-glucan and lichenan) resulted in a much lower product concentration, about 10 to 20-fold less than for β-1,4 substrates as CMC or PASC. From the β-1,3/1,4-glucans the most prominent oligosaccharides are the C4, C5 and C3 products. Interestingly, FpiGH45 produces a high amount of glucose from both substrates but no C2 product. PcCel45A showed a very similar hydrolysis pattern than the brown rot enzyme from F. pinicola. However, it was about 5 times more efficient in the conversion of α-cellulose and produced much less glucose from barley β-glucan. Neither of the enzymes showed a detectable hydrolysis towards glucomannan, microcrystalline cellulose or xylan. The incubation of both enzymes with cellobiose did not show any glycosidic activity towards the disaccharide. The incubation with cellobiose or glucose also showed no transglycosylation activity.

Table 3.

End products (μM) detected by HPLC upon substrate hydrolysis of FpiGH45 and PcCel45A. The substrates (1 mg mL-1) were incubated with 1.0 μM of purified enzymes during 60 min at 30°C. n.d. …not detected

Substrates Gluc [μM] C2 [μM] C3 [μM] C4 [μM] C5 [μM] C6 [μM] Total [μM] TNobs [s-1]
FpiGH45 CMC 5.94 n.d 121.73 196.07 184.26 34.41 542.42 0.151
PASC 0.00 n.d 15.86 67.51 99.07 32.29 214.73 0.060
β-glucan 56.06 n.d 6.15 15.68 8.33 n.d 86.21 0.024
Lichenan 25.53 n.d 1.19 10.05 4.71 n.d 41.48 0.012
α-cellulose n.d n.d 1.25 5.06 5.07 n.d 11.37 0.003
Glucomannan 3.08 n.d n.d n.d n.d n.d 3.08 0.001
MCC n.d n.d n.d n.d n.d n.d n.d n.d
Xylan n.d n.d n.d n.d n.d n.d n.d n.d
PcCeI45A CMC 2.78 n.d 90.80 164.12 159.29 33.11 450.09 0.125
PASC n.d n.d 14.67 62.26 72.05 25.44 174.42 0.048
β-glucan 3.89 n.d 4.96 18.30 7.60 0.00 34.75 0.010
Lichenan 25.09 n.d n.d 5.51 2.20 n.d 32.79 0.009
α-cellulose n.d n.d 1.25 5.06 5.07 n.d 11.37 0.017
Glucomannan 3.07 n.d n.d n.d n.d n.d 3.07 0.001
MCC n.d n.d n.d n.d n.d n.d n.d n.d
Xylan n.d n.d n.d n.d n.d n.d n.d n.d
Open in a new tab

Thermal stability

The thermal denaturation was measured by two methods. The residual activity of preincubated enzymes at 60°C and 80°C was measured with the Nelson-Somogyi assay, given as turnover number (TN), and was found to be very similar for both EG (Figure 6). After 2 h of incubation at 60 and 80°C the enzymatic activity decreased by 48.8 and 92.1 % (FpiGH45) and 50.8 and 80.1 % (PcCel45A), respectively. In a second approach differential scanning calorimetry (DSC) was used to determine the thermal stability of the proteins (Figure 7). Both endoglucanases showed a sharp transition and a similar transition midpoint temperature (Tm) of 69.9°C for FpiGH45 and 69.6°C for PcCel45A demonstrating a high thermostability. To investigate protein unfolding and refolding, sample downscans and rescans were performed by DSC. Results revealed that refolding was happening, but not completely reversible. The ratio of ΔHcal,rescan over ΔHcal,scan showed a value of 0.76 for FpiGH45 while for PcCel45A the ratio was 0.48. The measured Tm-values suggest a very robust structural fold and the ratios of observed enthalpies a big fraction of unfolded enzyme molecules that refold at lower temperatures. It is however interesting, that despite the similar protein composition a twofold difference between the heat capacities (Cp) were observed with FpiGH45 (Cp 2750 kJ mol-1 K-1) having a higher heat capacity than PcCel45A (1389 kJ mol-1 K-1).

Figure 6.

Figure 6

Open in a new tab

Thermostability of FpiGH45 (left) and PcCel45A (right). Residual activity given as turnover numbers (TN) was analysed at 60°C and 80°C during 240 min. The assays were conducted in 50 mM sodium acetate (pH 4.0) using 1 mg mL-1 CMC as a substrate.

Figure 7.

Figure 7

Open in a new tab

DSC thermograms for FpiGH45 (left) and PcCel45 (right) upscan and rescan are displayed as overlay graphs (upscan, dark blue line) and (rescan, light blue line). The molar heat capacity (Cp) after buffer baseline subtraction and baseline fitting was plotted assuming a non-two-state model with ΔHvH/ΔHcal < 1.0 against temperature. The protein samples (22 μM) were dialyzed against 50 mM sodium acetate pH 4.0 and reheated at a rate of 1°C min-1.

Discussion

The genome of Fomitopsis pinicola encodes a hypothetical protein of 206 amino acids (GenBank accession number EPT00417.1) that has been annotated as a GH45 endoglucanase [3]. The sequence alignment revealed a high similarity with previously identified GH45 endoglucanases from F. palustris (87 %) and P. chrysosporium (69 %) with most conserved residues located along the catalytic groove. The highest diversity in these sequences is observed in the region of the substrate cleft, which could indicate a different substrate specificity. The AlphaFold prediction of FpiGH45 structure revealed the absence of Asp10, which has been appointed as the proton acceptor for subfamily A enzymes. Instead, FpiGH45 contains Asn95 which is at the same position as Asn92 in PcCel45A, which has been proposed to act as a general base in the Newton’s cradle proton relay catalytic mechanism [16] and for which a later study showed that if mutated, the activity decreases [30]. Major divergences are located between positions 105 to 112 in close vicinity to the HFD motif highly conserved among GH45 enzymes, also present in expansins, containing Asp117 (Asp114 in PcCel45A and Asp121 in HiCel45A) which has been shown to act as a catalytic proton donor and whose mutation to alanine resulted in total loss of enzymatic activity [30]. In comparison to the PcCel45A sequence, the major position divergence is found at position 106 where both FpCel45 and FpiGH45 display a small and flexible glycine residue while PcCel45A contains a proline which may confer a higher rigidity to the structure.

Recombinant FpiGH45 and PcCel45A displayed remarkable biochemical characteristics that might be crucial for industrial and biotechnological purposes. The apparent pH optimum for both endoglucanases was found to be 4.0 in 50 mM sodium acetate buffer. Both endoglucanases have an acidophilic character, retaining about 75 % of their activity between pH 3.5 and 5.5 and very little activity above pH 7.5 (Figure 5). Our results confirm these previously reported data for PcCel45 [13]. In contrast to FpiGH45, FpGH45 has its pH optimum at pH 5.0 [27]. Our data support former studies reporting that the optimal pH of basidiomycetous endoglucanases are usually acidophilic [4], which is also in good agreement with natural pH found in fungus-colonized wood [6]. Thermostable cellulases have gained attention due to their ability to withstand the often relatively harsh conditions used in industrial processes. Several thermostable endoglucanases have been reported among family GH45 [31]. In this study, preincubation experiments revealed a lower thermostability of FpiGH45 compared to GH45 enzymes from F. palustris (t1/2 ~ 1.5 h at 80°C) [27], Pichia pastoris GS11 (t1/2 of 6 h at 80°C) [32] or from thermophilic organisms as Gloeophyllum trabeum (24 h at 70°C), Myceliophthora thermophila (24 h at 70°C) [18]. DSC analysis showed that FpiGH45 fully retained its secondary structure until 60°C where unfolding started. The Tm of 69.9°C for FpiGH45 supports the results from the preincubation experiments. Very similar results were obtained for PcCel45A (Tm = 69.59°C) except that the heat capacity of FpiGH45 (2750 kJ mol-1 K-1) is about twofold higher than for PcCel45A (1389 kJ mol-1 K-1) (Figure 7). Changes in heat capacity for unfolded proteins are attributed to the hydration effect. In the folded state, many of the non-polar or hydrophobic amino acids are buried in the interior of the protein, thus a smaller surface area is accessible to the solvent [33]. When the protein unfolds, the solvent accessible area increases. However, when comparing both amino acid sequences, PcCel45A has only a slightly higher proportion of hydrophobic residues in the protein core buried from the interaction with water molecules [34,35] which cannot explain the observed difference.

Previous studies showed that enzymes from the GH45 family are active on a relatively wide range of substrates containing β-1,4 or β-1,3/1,4-glucans [10,13,18]. FpiGH45 and PcCel45A displayed activity towards β-1,4 and β-1,3/1,4-glucans although with different efficiencies (Table 2). Contrary to earlier reports stating that PcCel45A had higher specific activity towards β-1,3/1,4-glucans than for β-1,4-glucans [13] our results indicate that PcCel45A has a 1.3-2 fold higher hydrolytic activity towards β-1,4-glucans. The highest hydrolytic activity of FpiGH45 was found towards CMC, an amorphous model substrate for endo-acting cellulases consisting of β-1,4-bound glucose units. FpiGH45 showed a similar hydrolysis pattern as PcCel45A as well as the closely related FpCel45 [27]. Fungal enzymes from the GH45 family are reported to preferentially degrade amorphous cellulose rather than crystalline cellulose. In our study, both FpiGH45 and PcCel45A displayed a moderate activity towards α-cellulose but no activity against microcrystalline cellulose. Commercial α-cellulose has amorphous regions that are susceptible to endoglucanases [36]. The major soluble products quantified by HPLC were oligosaccharides with high a degree of polymerisation between 3 and 5 (Table 3). No cellobiose was detected and only minor amounts of glucose could be quantified. This result contrasts with the hydrolysis patterns of endoglucanases belonging to subfamilies A and B were the major end products obtained are cellobiose and glucose [5,10]. However, when the enzyme models/structures are compared, FpiGH45 as well as PcCel45A exhibit a much longer catalytic groove and lack the end loops present in the canonical members of subfamilies A and B. Instead of large loops to properly lock the substrate, the PcCel45A structure involves two aromatic residues at the ends of the binding cleft (Tyr67 and Tyr154), which are also present in FpiGH45 (Tyr71 and Trp157). The Trp154 residue is also conserved along the domain I of expansins [14] and is reported to be involved in the polysaccharide recognition and binding through a ring stacking mechanism, which might be an alternative for the lack of the carbohydrate-binding module (CBM) [3]. Moreover, mutational investigations with PcCel45A showed that both residues are essential for a productive substrate binding into the cleft [30]. However, in near vicinity to Trp157 at position 158, FpiGH45 contains a threonine while the closely related FpCel45 has an aspartic acid and PcCel45A has an asparagine. Interestingly, all three residues display different properties that might affect substrate binding.

Hydrolysis experiments using β-1,3/1,4-glucans revealed similar patterns. Both enzymes displayed slightly higher activity towards barley β-glucan than lichenan. These results agree with the hydrolysis end products obtained on the HPLC. Interestingly, for both substrates, the major soluble product detected was glucose. The glucose concentration was 2–4 times higher than that of the C3–C5 products together with the exception of β-glucan for PcCel45A where glucose was not the major reaction product but C4. The concentration of produced oligosaccharides from β-glucan was twice as high as for lichenan. Our results disagree with previous reports describing that lichenan is the best substrate for PcCel45A [13,34].

Although the substrate could be hydrolyzed, the activity towards CMC was 2 times higher as well as the oligosaccharides quantified by HPLC. In general, FpiGH45 showed higher catalytic activity towards β-1,4-glucans rather than β-1,3/1,4-glucans. This trend coincides with the FpCel45 hydrolysis patterns. Both brown rot fungi at position 98 display a tryptophan while PcCel45A contains a phenylalanine residue at position 95 (Figure S3). Cha et al., supposed that the difference in size between residues could limit the binding space in FpCel45A and as a result the enzyme might possess higher hydrolytic activity towards β-1,4-glucans as CMC and PASC rather than β-1,3/1,4-glucans [27]. However, our results disagree with the previous suggestion as both FpiGH45 and PcCel45A endoglucanases display similar hydrolysis patterns. Residual activity could be detected towards glucomannan, with a slightly higher activity for PcCel45A than FpiGH45. However, the exclusive end product detected for both enzymes was glucose. Those values suggest that the positioning of the mannose residues into the substrate binding cleft might not be optimal leading to a lower hydrolytic efficiently. Finally, our HPLC analysis with xylan showed no soluble end products. This result agrees with the MD simulations carried out by Godoy and coworkers where they showed that xyloheptaose (X7) completely dissociates from the enzyme within the first ~5 ns revealing low substrate affinity attributed to the carbohydrate recognition by the enzyme [30].

In conclusion, phylogenetic and structural analysis suggests that FpiGH45 is a true member of GH45 subfamily C. The characterisation of the brown-rot FpiGH45 in comparison to the white-rot PcCel45A shows that both endoglucanases are active in acidic conditions and quite thermostable which suits industrial purposes. Both endoglucanases showed similar substrate specificities which supports the classification of FpiGH45 as a member of subfamily C. It also implies that both fungi use this endoclucanse for a similar function during the lignocellulose degradation process. Despite the observed structural differences at the binding cleft, no different hydrolysis patterns between the GH45 of a brown rot and white rot fungi was observed.

Supplementary Material

Supplementary Information

Highlights.

  • FpiGH45 is a true member of GH45 subfamily C endoglucanases

  • β-1,4 glucans are hydrolyzed to C3, C4, C5 and C6 products

  • β-1,3/1,4-glucans result in glucose as predominant reaction product

  • FpiGH45 has an acidic pH optimum and is thermostable

Acknowledgements

This project has received funding from the European Union’s Horizon 2020 research and innovation programme (ERC Consolidator Grant OXIDISE) under grant agreement No 726396.

Footnotes

Competing interests

The authors declare to have no competing interests.

Authorship contribution

Neus Gacias Amengual: Investigation, Conceptualisation, Data curation, Writing –original draft, Review & editing. Florian Csarman: Investigation, Data curation, Review & editing. Lena Wohlschlager: Investigation. Roland Ludwig: Supervision, Funding aquisition, Project administration, Writing – review & editing.

Contributor Information

Neus Gacias Amengual, Email: neus.gacias@boku.at.ac.

Florian Csarman, Email: florian.csarman@boku.ac.at.

Lena Wohlschlager, Email: lena.wohlschlager@boku.ac.at.

References

  • [1].Bomble YJ, Lin CY, Amore A, Wei H, Holwerda EK, Ciesielski PN, Donohoe BS, Decker SR, Lynd LR, Himmel ME. Lignocellulose deconstruction in the biosphere. Curr Opin Chem Biol. 2017;41:61–70. doi: 10.1016/j.cbpa.2017.10.013. [DOI] [PubMed] [Google Scholar]
  • [2].Rytioja J, Hildén K, Yuzon J, Hatakka A, de Vries RP, Mäkelä MR. Plant-polysaccharide-degrading enzymes from basidiomycetes. Microbiol Mol Biol Rev. 2014;78:614–649. doi: 10.1128/mmbr.00035-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [3].Floudas D, Binder M, Riley R, Barry K, Blanchette RA, Henrissat B, Martinez AT, Otillar R, Spatafora JW, Yadav JS, Aerts A, et al. The paleozoic origin of enzymatic lignin decomposition reconstructed from 31 fungal genomes. Science. 2012;336(8069):1715–1719. doi: 10.1126/SCIENCE.1221748. [DOI] [PubMed] [Google Scholar]
  • [4].Arantes V, Goodell B. Current understanding of brown-rot fungal biodegradation mechanisms: A review. 2014:3–21. doi: 10.1021/bk-2014-1158.ch001. [DOI] [Google Scholar]
  • [5].Payne CM, Knott BC, Mayes HB, Hansson H, Himmel ME, Sandgren M, Ståhlberg J, Beckham GT. Fungal cellulases. Chem Rev. 2015 doi: 10.1021/cr500351c. [DOI] [PubMed] [Google Scholar]
  • [6].Baldrian P, Valášková V. Degradation of cellulose by basidiomycetous fungi. FEMS Microbiol Rev. 2008;32:501–521. doi: 10.1111/j.1574-6976.2008.00106.x. [DOI] [PubMed] [Google Scholar]
  • [7].Cantarel BI, Coutinho PM, Rancurel C, Bernard T, Lombard V, Henrissat B. The Carbohydrate-Active EnZymes database (CAZy): An expert resource for glycogenomics. Nucleic Acids Res. 2009;37 doi: 10.1093/nar/gkn663. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [8].Henrissat B. Cellulases and their interaction with cellulose. Cellulose. 1994 doi: 10.1007/BF00813506. [DOI] [Google Scholar]
  • [9].Schülein M, Kauppinen MS, Lange L, Lassen SF, Andersen LN, Klysner S, Nielsen JB. Characterization of fungal cellulases for fiber modification. ACS Symp Ser. 1998;687:66–74. doi: 10.1021/BK-1998-0687.CH006. [DOI] [Google Scholar]
  • [10].Davies GJ, Dodson GG, Hubbard RE, Tolley SP, Dauter Z, Wilson KS, Hjort C, Mikkelsen JM, Rasmussen G, Schülein M. Structure and function of endoglucanase V. Nature. 1993 doi: 10.1038/365362a0. [DOI] [PubMed] [Google Scholar]
  • [11].Seon Park J, Russell JB, Wilson DB. Characterization of a family 45 glycosyl hydrolase from Fibrobacter succinogenes S85. Anaerobe. 2007 doi: 10.1016/j.anaerobe.2006.12.003. [DOI] [PubMed] [Google Scholar]
  • [12].Busch A, Danchin EGJ, Pauchet Y. Functional diversification of horizontally acquired glycoside hydrolase family 45 (GH45) proteins in Phytophaga beetles. BMC Evol Biol. 2019;19:100. doi: 10.1186/s12862-019-1429-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Igarashi K, Ishida T, Hori C, Samejima M. Characterization of an endoglucanase belonging to a new subfamily of glycoside hydrolase family 45 of the basidiomycete Phanerochaete chrysosporium. Appl Environ Microbiol. 2008;74:5628. doi: 10.1128/AEM.00812-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Sampedro J, Cosgrove DJ. The expansin superfamily. Genome Biol. 2005;6:1–11. doi: 10.1186/gb-2005-6-12-242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Saloheimo M, Paloheimo M M, Hakola S, Pere J, Swanson B, Nyyssönen E, Penttilä Swollenin, a Trichoderma reesei protein with sequence similarity to the plant expansins, exhibits disruption activity on cellulosic materials. Eur J Biochem. 2002;269:4202–4211. doi: 10.1046/J.1432-1033.2002.03095.X. [DOI] [PubMed] [Google Scholar]
  • [16].Nakamura A, Ishida T, Kusaka K, Yamada T, Fushinobu S, Tanaka I, Kaneko S, Ohta K, Tanaka H, Inaka K, Higuchi Y, et al. “Newton’s cradle”; proton relay with amide-imidic acid tautomerization in inverting cellulase visualized by neutron crystallography. Sci Adv. 2015;1:e1500263. doi: 10.2210/PDB4ZM7/PDB. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Godoy AS, Pereira CS, Ramia MP, Silveira RL, Camilo CM, Kadowaki MA, Lange L, Busk PK, Nascimento AS, Skaf MS, Polikarpov I. Structure, computational and biochemical analysis of PcCel45A endoglucanase from Phanerochaete chrysosporium and catalytic mechanisms of GH45 subfamily C members. Sci Rep. 2018;8:3678. doi: 10.1038/s41598-018-21798-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Berto GL, Velasco J, Tasso Cabos Ribeiro C, Zanphorlin LM, Noronha Domingues M, Tyago Murakami M, Polikarpov I, de Oliveira LC, Ferraz A, Segato F. Functional characterization and comparative analysis of two heterologous endoglucanases from diverging subfamilies of glycosyl hydrolase family 45. Enzyme Microb Technol. 2019;120:23–35. doi: 10.1016/j.enzmictec.2018.09.005. [DOI] [PubMed] [Google Scholar]
  • [19].Kracher D, Scheiblbrandner S, Felice AKG, Breslmayr E, Preims M, Ludwicka K, Haltrich D, Eijsink VGH, Ludwig R. Extracellular electron transfer systems fuel cellulose oxidative degradation. Science (80-) 2016;352:1098–1101. doi: 10.1126/science.aaf3165. [DOI] [PubMed] [Google Scholar]
  • [20].Johnson M, Zaretskaya I, Raytselis Y, Merezhuk Y, McGinnis S, Madden TL. NCBI BLAST: a better web interface. Nucleic Acids Res. 2008 doi: 10.1093/nar/gkn201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Lemoine F, Correia D, Lefort V, Doppelt-Azeroual O, Mareuil F, Cohen-Boulakia S, Gascuel O. NGPhylogeny.fr: new generation phylogenetic services for non-specialists. Web Serv Issue Publ Online. 2019;47 doi: 10.1093/nar/gkz303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Almagro Armenteros J, Tsirigos K, Sønderby C, Petersen T, Winther O, Brunak S, Nielsen H. SignalP 5.0 improves signal peptide predictions using deep neural networks. Nat Biotechnol. 2019;37:420–423. doi: 10.1038/S41587-019-0036-Z. [DOI] [PubMed] [Google Scholar]
  • [23].Robert X, Gouet P. Deciphering key features in protein structures with the new ENDscript server. Nucleic Acids Res. 2014;42:W320–W324. doi: 10.1093/NAR/GKU316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular evolutionary genetics analysis across computing platforms. Mol Biol Evol. 2018 doi: 10.1093/molbev/msy096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, Tunyasuvunakool K, Bates R, Žídek A, Potapenko A, Bridgland A, et al. Highly accurate protein structure prediction with AlphaFold. Nat. 2021;2021:1–11. doi: 10.1038/s41586-021-03819-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Nelson N. A photometric adaptation of the somogyi method for the determination of glucose. J Biol Chem. 1944;153:375–380. [Google Scholar]
  • [27].Cha JH, Yoon JJ, Cha CJ. Functional characterization of a thermostable endoglucanase belonging to glycoside hydrolase family 45 from Fomitopsis palustris. Appl Microbiol Biotechnol. 2018;102:6515–6523. doi: 10.1007/s00253-018-9075-5. [DOI] [PubMed] [Google Scholar]
  • [28].JK R, HH L, AB B. Expression of a divergent expansin gene is fruit-specific and ripening-regulated. Proc Natl Acad Sci U S A. 1997;94:5955–5960. doi: 10.1073/PNAS.94.11.5955. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [29].Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994;22:4673. doi: 10.1093/NAR/22.22.4673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Godoy AS, Pereira CS, Ramia MP, Silveira RL, Camilo CM, Kadowaki MA, Lange L, Busk PK, Nascimento AS, Skaf MS, Polikarpov I. Structure, computational and biochemical analysis of PcCel45A endoglucanase from Phanerochaete chrysosporium and catalytic mechanisms of GH45 subfamily C members. Sci Rep. 2018;8 doi: 10.1038/s41598-018-21798-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Patel AK, Singhania RR, Sim SJ, Pandey A. Thermostable cellulases: Current status and perspectives. Bioresour Technol. 2019;279:385–392. doi: 10.1016/j.biortech.2019.01.049. [DOI] [PubMed] [Google Scholar]
  • [32].Couturier M, Feliu J, Haon M, Navarro D, Lesage-Meessen L, Coutinho PM, Berrin J-G. A thermostable GH45 endoglucanase from yeast: impact of its atypical multimodularity on activity. Microb Cell Fact. 2011;10:103. doi: 10.1186/1475-2859-10-103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Bruylants G, Wouters J, Michaux C. Differential scanning calorimetry in life science: Thermodynamics, stability, molecular recognition and application in drug design. 2005 doi: 10.2174/0929867054546564. [DOI] [PubMed] [Google Scholar]
  • [34].Mazurenko S, Kunka A, Beerens K, Johnson CM, Damborsky J, Prokop Z. Exploration of protein unfolding by modelling calorimetry data from reheating OPEN. doi: 10.1038/s41598-017-16360-y. (n.d.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Panja AS, Bandopadhyay B, Maiti S. Protein thermostability is owing to their preferences to non-polar smaller volume amino acids, variations in residual physico-chemical properties and more salt-bridges. PLoS One. 2015 doi: 10.1371/journal.pone.0131495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].Rojas J, Lopez A, Guisao S, Ortiz C. Evaluation of several microcrystalline celluloses obtained from agricultural by-products. J Adv Pharm Technol Res. 2011 doi: 10.4103/2231-4040.85527. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information
EMS155487-supplement-Supplementary_Information.pdf (945.7KB, pdf)

RESOURCES