Characterization of Paenibacillus sp. GKG Endo-β-1, 3-Glucanase, a Member of Family 81 Glycoside Hydrolases

Paenibacillus sp. GKG was isolated based on its ability to produce hydrolysis zones on agar plates containing yeast cell wall substrate as the single carbon source. The extracellular enzymes secreted into the culture medium were identified by LC-MS/MS proteomics. Endo-β-1,3-glucanase PsLam81A containing GH81 catalytic and the CBM56 carbohydrate-binding modules was selected for heterologous expression in Escherichia coli. The identity of the recombinant PsLam81A was confirmed by LC-MS/MS proteomics. The PsLam81A showed the highest activity at 60 °C, and the optimal pH range was between 6.5 and 8.0. The analysis of the full-length PsLam81A and truncated PsLam81AΔCBM56 enzymes showed that the CBM56 module improved the hydrolytic activity towards linear β-1,3-glucans—curdlan and pachyman but had no effect on hydrolysis of β-1,3/β1,6-branched glucans—laminarin and yeast β-glucan. The characterization of PsLam81A enzyme broadens current knowledge on the biochemical properties and substrate specificity of family 81 glycoside hydrolases and allows prediction of the necessity of CBM56 module in the process of designing new truncated or chimeric glycosidases.


Substrate Preparation for Screening of Microorganisms
Compressed baker's yeast (Lallemand, Lublin, Poland) suspension was prepared in reverse osmosis treated water at a concentration of 7.6 g (dry weight)/100 g of suspension. The suspension pH was 7.6 and it contained 0.7% NaHSO 3 and 0.7% Na 2 CO 3 . Yeast cells were disrupted using high shear fluid processor microfluidizer M7250-10 (Microfluidics, Newton, MA, USA). The operating pressure of 690 bar was used and 8 passages were performed. The temperature of the effluent did not exceed 18 • C. The resulting suspension was centrifuged at 12,200× g for 10 min. The supernatant was discarded, the yeast cell debris (further in the text referred as cell walls) layer was removed and transferred to a new centrifuge bottle, remaining undisrupted yeast cell layer was discarded. The resulting cell wall fraction was washed with water: 2 parts of water (by weight) was added to a yeast cell wall fraction. The mixture was homogenized using a kitchen blender SHB 4460WH-EUE3 (Sencor, Ricany, Czech Republic). The resulting suspension was centrifuged at 12,200× g for 10 min, the supernatant was discarded, yeast cell wall layer was removed and transferred to a new centrifuge bottle. The procedure was repeated two more times. After washing with water, the yeast cell wall fraction was frozen at −80 • C and freeze-dried for 48 h at −48 • C using a freeze-dryer Modulyo D-115 (Thermo Savant, Waltham, MA, USA). Freeze-dried yeast cell wall fraction was ground using laboratory mill M20 (IKA, Staufen, Germany).

Isolation of Bacteria
Soil sample was collected from the pond located in the city of Vilnius, Lithuania (54.709437, 25.271094). The sample was enriched for yeast cell wall degrading microorganisms by adding yeast cell walls to a final concentration of 10 mg/mL and incubated at 22 • C for 120 h. Enriched samples were diluted 25-fold using 0.9% NaCl, then 50 µL aliquots were plated on a solid mineral medium containing a water-insoluble yeast cell wall substrate as a carbon source. The solid mineral medium contained 5 g/L NaCl, 1 g/L NH 4 H 2 PO 4 , 1 g/L K 2 HPO 4 , 2 g/L MgSO 4 × 7H 2 O, 15 g/L agar, 3 g/L lyophilized yeast cell walls and 6 mg/L clotrimazole, pH 7.2. The plates were incubated at 30 • C for 72 h. Then the plates were visually inspected and colonies that were able to produce hydrolysis zones were isolated by repeated streaking.

Identification of Isolates
Chromosomal DNA from isolated colonies was extracted as described previously [17]. The 16S rRNA gene was amplified by PCR using chromosomal DNA as a template, Phusion Plus PCR Master Mix (Thermo Fisher Scientific, Vilnius, Lithuania) and primers W001 and W002 (Table S1) [18]. The PCR products were purified using GeneJET Gel Extraction Kit (Thermo Fisher Scientific, Vilnius, Lithuania). Purified PCR products were cloned into pJET1.2/blunt cloning vector using CloneJET PCR Cloning Kit (Thermo Fisher Scientific, Vilnius, Lithuania). E. coli strain DH5α (Fermentas, Vilnius, Lithuania) cells were transformed with ligation mixtures and spread on LB agar plates supplemented with 50 µg/mL ampicillin. pDNA was extracted and purified using the ZR Plasmid Miniprep kit (Zymo Research, Irvine, CA, USA) and sequenced using pJET1.2 sequencing primers (Table S1) by the Sanger method (Macrogen Europe, Amsterdam, The Netherlands).

Production of Hydrolytic Enzymes
The liquid mineral medium used for the production of extracellular enzymes contained 5 g/L NaCl, 1 g/L NH 4 H 2 PO 4 , 1 g/L K 2 HPO 4 , 2 g/L MgSO 4 × 7H 2 O and 3 g/L lyophilized yeast cell walls, pH 7.2. A single colony from a solid mineral medium was inoculated in 20 mL of liquid mineral medium for 24 h at 30 • C 180 rpm. 10 mL of inoculum was transferred to 200 mL of liquid mineral medium and incubated for 48 h at 30 • C 180 rpm. The culture medium was centrifuged at 7200× g for 20 min at 4 • C. The supernatant was collected and pellet was discarded. The supernatant was concentrated 20-fold using Amicon Stirred Ultrafiltration Cell 8200 (Millipore, Burlington, MA, USA) equipped with Omega 76 mm 30K membrane discs (Pall Corporation, New York, NY, USA). Concentrated supernatant was dialyzed in a 14K dialysis membrane tubing (Carl-Roth, Karlsruhe, Germany) at 4 • C against 10 mM Tris-HCl, pH 7.0, for 18 h. Afterwards, the dialyzed samples were subjected to a tryptic digest.

Filter Aided Protein Sample Preparation (FASP) for Mass Spectrometry Analysis
FASP of protein samples was performed using Nanosep ® , MWCO 10 kDa ultrafiltration centrifugal units (Pall Corporation, New York, NY, USA) at 13,000× g. Sample preparation and tryptic digestion were performed according to a modified FASP method described earlier [19]. The protein sample was mixed with denaturing buffer (8 M urea 100 mM ABC pH 8.0) in a ratio 1:7 (sample:denaturing buffer, v/v). Denaturing buffer was exchanged for 100 µL reducing buffer (10 mM DTT in denaturing buffer). The sample in reducing buffer was incubated in a Thermomixer C (Eppendorf, Hamburg, Germany) at 37 • C 600 rpm for 60 min. After incubation, 200 µL of denaturing buffer were added to centrifugal units, after centrifugation, flowthrough was discarded. 100 µL of alkylation buffer (50 mM iodoacetamide in denaturing buffer) were added and centrifugal units were incubated at 37 • C 600 rpm for 60 min. DTT was added to yield a final concentration of 50 mM to deactivate residual iodoacetamide. Alkylation reagents were removed by centrifugation, flowthrough was discarded. Centrifugal units were washed once with denaturing buffer, flowthrough was discarded. Three more buffer exchanges were made to FASP digestion buffer (50 mM ABC), 200 µL of FASP digestion buffer each time. Proteins were digested overnight in 100 µL FASP digestion buffer with a 1:50 enzyme-to-sample (w/w) ratio using Pierce, trypsin protease MS-grade (Thermo Fisher Scientific, Vilnius, Lithuania) or Pierce, chymotrypsin protease (TLCK treated), MS-grade (Thermo Fisher Scientific, Vilnius, Lithuania). Centrifugal units with digestion reaction mixture were incubated in a 37 • C water bath Sub36 (Grant Instruments, Shepreth, UK) for 16 h. After digestion, the flowthrough was collected into new collection tubes, the centrifugal filter units were rinsed twice with 50 µL FASP digestion buffer, the flowthrough was collected. The solvent from the recovered peptide fraction was evaporated using a vacuum dryer Speed Vac SC110 (Thermo Savant, Waltham, MA, USA). Dried peptides were resuspended in 100 µL 50% methanol and vacuum dried two more times to remove volatile salts completely. Dried peptides were redissolved in 0.1% formic acid and Hi3 PhosB standard peptide mixture (Waters Corporation, Milford, MA, USA) was added to yield a final concentration of 0.2 pmol/µL. Prepared samples were analyzed by LC-MS/MS.

Liquid Chromatography and Mass Spectrometry
Liquid chromatography (LC) separation of trypsin cleaved peptides was performed with ACQUITY UPLC I-Class System (Waters Corporation, Milford, MA, USA). Peptides were separated on a reversed-phase analytical column ACQUITY UPLC Peptide BEH C18 Column 300 Å, 1.7 µm, 2.1 mm × 150 mm (Waters Corporation, Milford, MA, USA) at a flow rate of 40 µL/min The isocratic gradient 5% solvent B was set until 2.5 min, then a linear gradient from 5% to 35% solvent B was set until 50 min, followed by a linear gradient from 35 to 85% solvent B until 51 min (solvent A: 0.1% formic acid, solvent B: 100% acetonitrile and 0.1% formic acid). The analytical column temperature was set to 40 • C.
The LC was coupled online through an ESI ionization source with Synapt G2 mass spectrometer (Waters Corporation, Milford, MA, USA). Data was acquired using MassLynx version 4.2 software (Waters Corporation, Milford, MA, USA) in positive ion mode. LC-MS data was collected using data-independent acquisition (DIA) mode MSE. The source/TOF conditions were set as follows: resolution mode, capillary voltage 2.6 kV, sampling cone voltage 40 V, extraction cone voltage 4 V, source temperature 120 • C, desolvation gas flow 800 L/h at 450 • C. During spectral data acquisition, the trap collision energy of the mass spectrometer was ramped from 18 to 35 eV for high-energy scans in MSE mode, for low-energy scans collision energy was disabled. The mass range was set to 50-2000 Da with a scan time set to 0.5 s. A reference compound (2 ng/µL in 50% acetonitrile, 0.1% formic acid) leucine enkephalin (Waters Corporation, Milford, MA, USA) was co-infused continuously at a 7 µL/min flow rate and scanned every 45 s as a reference for accurate mass measurements (reference mass: m/z 556.2771).

Data Processing and Protein Identification
For peptide and protein identification raw data files were processed using ProteinLynx Global SERVER (PLGS) version 3.0.3 (Waters Corporation, Milford, MA, USA). The following parameters were used to generate peak lists: low and elevated energy thresholds, 135 and 20 counts, respectively; reference mass correction window, 0.25 Da at 556.2771 Da/e. Processed data was analyzed using the following parameters: trypsin or chymotrypsin was selected as a primary digest reagent, one missed cleavage was permitted, carbamidomethylation of cysteines was set as a fixed modification, deamidation of asparagine and glutamine, oxidation of methionine were set as variable modifications. Minimal identification criteria included 2 fragment ions per peptide, 5 fragment ions per protein and a minimum of 1 peptide per protein. The false discovery rate (FDR) was set to 2%.

Cloning of GH81 Gene
Gene sequence variants coding PsLam81A (native enzyme sequence with C-terminal 6×His tag), PsLam81A∆CBM56 (truncated enzyme sequence without CBM56 domain with C-terminal 6×His tag) were cloned from genomic DNA of Paenibacillus sp. GKG using Phusion Plus PCR Master Mix (Thermo Fisher Scientific) and designed primers PsLam81A and PsLam81A∆CBM56 (Table S1). The PCR products were stained with ROTI Load DNA stain 1 SYBR Green (Carl-Roth, Karlsruhe, Germany) and ran in a 1% agarose gel electrophoresis. Bands corresponding to target gene sizes were excised from gel and plasmid DNA was extracted from gel slices using GeneJET Gel Extraction Kit (Thermo Fisher Scientific). Purified PCR products were cloned into a pLATE31 cloning vector using aLICator LIC Cloning and Expression Kit 3 (C-terminal His-tag) (Thermo Fisher Scientific, Vilnius, Lithuania). E. coli strain DH5α (Fermentas, Vilnius, Lithuania) cells were transformed with reaction mixtures and spread on LB agar plates supplemented with 75 µg/mL ampicillin. pDNA was extracted and purified using the ZR Plasmid Miniprep kit (Zymo Research, Irvine, CA, USA). pDNA was sequenced using LIC sequencing primers (Table S1) by the Sanger method (Macrogen Europe, Amsterdam, The Netherlands). Two additional sequencing primers PsLam81A-mid were designed for native enzyme sequence coding gene to obtain whole gene sequence coverage.

Expression and Purification of Recombinant Proteins
The recombinant pLATE31-PsLam81A and pLATE31-PsLam81A∆CBM56 plasmids were transformed into E. coli strain BL21(DE3) (Novagen, Darmstadt, Germany) cells. Transformed bacterium were cultured in 20 mL LB medium supplemented with 75 µg/mL ampicillin until it reached OD 0.6-0.8 at 600 nm. Expression of recombinant protein was induced with 0.5 mM IPTG by incubating at 20 • C for 24 h. Bacterial cells were harvested by centrifugation at 7200× g for 30 min at 4 • C. Collected cells were resuspended in 5 mL 1 × PBS pH 7.0, sonicated and centrifuged at 30,100× g for 30 min at 4 • C. The supernatant was loaded onto a 1 mL HiTrap Chelating HP column (GE Healthcare, Uppsala, Sweden) that had been pre-equilibrated with 10 column volumes of 1 × PBS pH 7.0. Column was washed with 5 column volumes of 28 mM imidazole in 1 × PBS pH 7.0. The recombinant protein was eluted with 400 mM imidazole in 1 × PBS pH 7.0. The collected recombinant protein fraction was desalted on a 5 mL Sephadex G25 column (GE Healthcare, Uppsala, Sweden) using 1 × PBS pH 7.0. The molecular weight of the recombinant protein was evaluated using 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using Prestained Protein Ladder (10-170 kDa) (Thermo Fisher Scientific, Vilnius, Lithuania). Protein bands were visualized using Coomassie Brillant Blue G-250 (Fluka, Buchs, Switzerland) staining. The protein concentration was estimated using Pierce Coomassie (Bradford) Protein Assay Kit (Thermo Fisher, Rockford, IL, USA).

Enzyme Assay
The activity of the recombinant PsLam81A and PsLam81A∆CBM56 enzymes was determined using the DNS method by determining the reduced sugars released during the hydrolysis of polysaccharides as described previously [20], with modifications. DNS reagent contained 10 g/L 3,5-dinitrosalicylic acid, 300 g/L sodium potassium tartrate and 16 g/L NaOH. The hydrolysis reaction was terminated by immediately transferring of tubes to a boiling water bath for 5 min. After the hydrolysis reaction the sample was centrifuged at 16,000× g for 2 min, then a sample was mixed with DNS reagent at a sample-to-DNS reagent ratio (1:5, v/v). The mixture was heated for 5 min in a boiling water bath, then cooled down for 5 min in an ice bath, then left for 5 min at 22 • C for the sample temperature to equilibrate. 100 µL of the reaction mixture were transferred to a round bottom microplate and absorbance at 540 nm was measured. The one unit of endo-1,3-β-glucanase activity is determined as the enzyme amount required to produce the amount of reducing sugar equal to 1 µmol of glucose per min of reaction time. Measurements were performed at least in triplicate and expressed as mean ± standard deviation (SD).
The effect of temperature on PsLam81A and PsLam81A∆CBM56 activity was determined in a reaction mixture (100 µL) containing 5 µg/mL enzyme and 0.9% (w/v) laminarin in 50 mM sodium phosphate buffer pH 7.0 at different temperatures ranging from 10 • C to 80 • C for 15 min.
The effect of temperature on the stability of PsLam81A was determined by incubating the enzyme in 1 × PBS pH 7.0 at different temperatures ranging from 4 • C to 60 • C for 15 or 60 min. After incubation activity was determined in a reaction mixture (100 µL) containing 10 µg/mL enzyme and 0.9% (w/v) laminarin in 50 mM sodium phosphate buffer pH 7.0 incubated at 40 • C for 15 min.

Determination of Substrate Specificity
The mode of action of PsLam81A and PsLam81A∆CBM56 was determined using laminari-oligosaccharides (laminaribiose, laminaritriose, laminaritetraose, laminaripentaose, laminarihexaose (Table 1) as substrates. Reaction mixtures (50 µL) contained 25 µg/mL enzyme and 0.9% (w/v) laminari-oligosaccharide substrate in 50 mM sodium phosphate buffer pH 7.0 and were incubated at 40 • C for 2 h, during control reactions the enzymes were absent. The substrate specificity of PsLam81A and PsLam81A∆CBM56 was determined using various polysaccharides: laminarin, yeast β-glucan, curdlan, pachyman, barley β-glucan (Table 1). Reaction mixtures (100 µL) contained 25 µg/mL enzyme and 0.9% (w/v) polysaccharide substrate in 50 mM sodium phosphate buffer pH 7.0 and were incubated at 40 • C for 18 h, during control reactions the enzymes were absent. No temperature or chemical pre-treatment was applied to laminari-oligosaccharides or polysaccharides prior to the hydrolysis reaction. The substrates were dissolved or suspended in appropriate buffers by mixing at 1000 rpm at 22 • C for 30 min.

Analysis of Hydrolysis Products
Samples collected at different times of hydrolysis reaction were immediately boiled for 5 min, and then analyzed by thin layer chromatography (TLC). TLC was performed as described [21], with slight modifications. Briefly, samples were spotted on TLC Aluminum Sheets coated with silica gel matrix and developed in butan-1-ol/acetic acid/water (2:1:1, v/v) solvent. TLC results were visualized by p-anisaldehyde. Developed silica plates were dipped into dye solution (p-anisaldehyde/99.8% acetic acid/96% ethanol/96% sulfuric acid 0.54:0.54:50:1.9 v/v/v/v) for a few seconds and heated above heating plate for a few minutes until spots corresponding to hydrolysis products appeared.

Nucleotide Sequence Accession Number
The Paenibacillus sp. GKG family GH81 endo-β-1,3-glucanase gene coding sequence was deposited in GenBank under accession no. OP093985. The Paenibacillus sp. GKG 16S rRNA gene sequence was deposited in GenBank under accession no. OP081028.

Identification of PsLam81A Coding Gene
The bacterium (latter named, Paenibacillus sp. GKG) was isolated from the water pond sample based on its ability to produce hydrolysis zones on an agar plate containing insoluble yeast cell wall substrate ( Figure S1). The isolate was identified using 16S rRNA gene sequence analysis, which showed the highest similarity (99%) with Paenibacillus sp. B01 (accession no. CP045802). To identify hydrolytic enzymes, Paenibacillus sp. GKG was Microorganisms 2022, 10, 1930 8 of 16 cultivated in a liquid mineral medium supplemented with yeast cell walls. The supernatant was collected, concentrated, dialyzed and analyzed by LC-MS/MS proteomics in order to identify the extracellular hydrolytic enzymes. The protein sequences (1,671,348 entries) retrieved from UniProt using the Paenibacillus genus keyword were used as a reference database. The proteomic analysis revealed that 710 proteins from the database had one to 35 peptides matched to their amino acid sequences. The clustering of these proteins using a 90% identity allowed the identification of 163 proteins in the cell-free culture medium of Paenibacillus sp. GKG. The majority of identified proteins were not related to carbohydrate degradation. A manually curated list of 24 proteins selected as CAZymes can be found in Supplementary materials (Table S2). The selection was made based on available assigned EC numbers and protein descriptions. According to the published data, a GH81 family includes a low number of the characterized bacterial enzymes, therefore, a putative glycoside hydrolase belonging to family 81 had 29% sequence coverage and was selected for further analysis to broaden our knowledge about this group of enzymes.

Cloning, Expression and Purification of Recombinant PsLam81A and PsLam81A∆CBM56 Enzymes
The two gene variants of selected glycoside hydrolase from family 81 (coding full PsLam81A and without carbohydrate-binding module PsLam81A∆CBM56 proteins) were cloned into a pLATE31 plasmid, expressed in E. coli strain BL21(DE3) cells. Both enzymes were soluble independently of induction at either 20 • C or 30 • C temperatures ( Figure S2). Small amounts of target proteins were detected in the induction medium suggesting a lowefficiency secretion. The presence of N-terminal signal peptide sequence (1-23 aa) in the PsLam81A protein was predicted by SignalP 6.0 server [30]. Both enzymes with C-terminal 6 × His tags were purified successfully using immobilized metal ion affinity chromatography and desalted by gel permeation chromatography. The obtained recombinant enzymes were analyzed by SDS-PAGE ( Figure 2). The molecular weights of both enzymes PsLam81A and PsLam81A∆CBM56 were comparable to the ones deduced from theoretical amino acid sequences (see Supplementary Materials). However, a resolution was too low to precisely determine the presence or absence of the signal peptides.  For that reason the purified recombinant PsLam81A enzyme was analyzed by LC-MS/MS. The PsLam81A protein was digested with two proteases separately-trypsin and chymotrypsin. The sequence coverage obtained after tryptic and chymotryptic digestion was 86% and 78%, respectively ( Figures S3 and S4). Overlayed sequences of both digestions yielded in 96% coverage of the recombinant protein sequence. The sequence had one uncovered region (1-27 amino acid (aa)), which included a predicted signal peptide sequence (1-23 aa), determined using SignalP 6.0 server [30]. The absence of peptides from this region in both digests indicated that the signal peptide was cleaved off in the E. coli cells. Only a small amount of recombinant enzymes was detected in the cellfree culture medium ( Figure S2). As the major part of induced proteins remained in the For that reason the purified recombinant PsLam81A enzyme was analyzed by LC-MS/MS. The PsLam81A protein was digested with two proteases separately-trypsin and chymotrypsin. The sequence coverage obtained after tryptic and chymotryptic digestion was 86% and 78%, respectively (Figures S3 and S4). Overlayed sequences of both digestions yielded in 96% coverage of the recombinant protein sequence. The sequence had one uncovered region (1-27 amino acid (aa)), which included a predicted signal peptide sequence (1-23 aa), determined using SignalP 6.0 server [30]. The absence of peptides from this region in both digests indicated that the signal peptide was cleaved off in the E. coli cells. Only a small amount of recombinant enzymes was detected in the cell-free culture medium ( Figure S2). As the major part of induced proteins remained in the cell and had its signal sequence removed it is probable that proteins remained located in a periplasmic space of E. coli cells. Earlier findings support the prediction that the presence of native Paenibacillus sp. GKG signal peptide results in periplasmic secretion of PsLam81A and PsLam81A∆CBM56 in E. coli BL21(DE3) cells. Hence, the endoglucanase EG5B (a full-length enzyme with a native signal peptide) from Paenibacillus sp. IHB B 3084 was transported to the periplasm of E. coli BL21(DE3) cells [31]. Moreover, the xylanase XynA from Paenibacillus sp. DG-22 heterologously produced in E. coli M15 (pREP4) cells was also located in the periplasm when the host cells were carrying the plasmid pXA8 with a gene encoding a full-length protein including a native signal peptide, or in the cytoplasm when host cells were carrying the plasmid pQE60-XynA with a truncated gene (the sequence encoding a native signal peptide was removed) [32].

Effect of pH and Temperature on the Activity of PsLam81A and PsLam81A∆CBM56
The purified recombinant protein PsLam81A was active towards yeast cell walls. Therefore, the effect of pH on the activity and the effect of temperature on enzyme activity and stability were tested. The enzyme showed a broad optimum pH range exhibiting 97% of the maximum activity within the pH range of 6.5 and 8.0 and 80% of the maximum activity within the pH range of 6.0 and 9.0 (Figure 3a). Similar optimum pH ranges with 80% of the maximum activity were reported for two previously characterized bacterial GH81 hydrolases-pH range from 5.5 to 8.0 for CtLam81A from Clostridium thermocellum (Acetivibrio thermocellus) [3] and pH range from 5.5 to 10.0 for TfLam81A from Thermobifida fusca [4].
The highest activity of the enzyme (PsLam81A) was recorded at 60 • C (Figure 3b). In comparison, the reported optimal temperature for CtLam81A was 75 • C [3] and 50 • C for TfLam81A [4]. However, the temperature stability of the enzyme (PsLam81A) was low (Figure 3c). After 15 min of incubation at temperatures below 40 • C enzyme retained 97% of maximum activity, whereas at 45 • C only 59% of maximum activity was retained. Moreover, after 60 min incubation the enzyme retained 77% of the maximum activity at 40 • C temperature and showed no activity after incubation at temperatures above 50 • C. Generally, the previously characterized members of the GH81 family have been more thermostable. In comparison, CtLam81A retained 93% of maximum activity after incubation for 60 min at 75 • C [3] and TfLam81A retained approximately 90% of its maximum activity after incubation for 18 h at 50 • C [4].
Literature research revealed that the removal of CBM modules can result in either improved [33,34] or decreased [35,36] thermal stability of an enzyme. Moreover, improved and decreased thermal stability can be observed for different truncation variants for the same enzyme. For example, truncated endo-β-1,4-xylanase rXTMA∆NC with all four (2 × CBM22, 2 × CBM9) CBM modules removed shows lower thermal stability, whereas rXTMA∆C with only two out of four CBM modules (2 × CBM22) removed shows higher thermal stability compared to native enzyme rXTMA where all four CBM modules are present [37]. In this study, we had observed a slight improvement of thermal stability of the PsLam81A enzyme when the CBM56 module was removed (the enzyme PsLam81A∆CBM56). After 60 min of incubation at 45 • C, PsLam81A retained 24.0 ± 2.7% of maximum activity, PsLam81A∆CBM56 retained 35.6 ± 1.0% of maximum activity and after 60 min incubation at 50 • C PsLam81A retained no detectable activity, when PsLam81A∆CBM56 showed a trace activity. The fusion of the homologous CBM56 module with chitosanases improved the thermal stability of fusion enzymes compared to native ones [38]. However, the thermal stability enhancing effect was different for the two chitosanases, hence, CBM56-GsCsn46A had improved thermal stability only at 35 • C temperature, while CBM56-BaCsn46A was more stable in the 40-55 • C temperature range. These observations on the CBM56 module effect on temperature stability supports the pattern that every enzyme and CBM combination is case-dependent and there is no general rule to conclude whether CBM module presence will improve or decrease the thermal stability of a truncated or fused enzyme.
We have selected the temperature of 40 • C for the substrate specificity reactions (Section 3.4). This is a compromise between an increasing reaction rate (Figure 3b) and the decreasing thermal stability of the enzyme (Figure 3c) due to increasing reaction temperature.

Effect of pH and Temperature on the Activity of PsLam81A and PsLam81AΔCBM56
The purified recombinant protein PsLam81A was active towards yeast cell walls. Therefore, the effect of pH on the activity and the effect of temperature on enzyme activity and stability were tested. The enzyme showed a broad optimum pH range exhibiting 97% of the maximum activity within the pH range of 6.5 and 8.0 and 80% of the maximum activity within the pH range of 6.0 and 9.0 (Figure 3a). Similar optimum pH ranges with 80% of the maximum activity were reported for two previously characterized bacterial GH81 hydrolases-pH range from 5.5 to 8.0 for CtLam81A from Clostridium thermocellum (Acetivibrio thermocellus) [3] and pH range from 5.5 to 10.0 for TfLam81A from Thermobifida fusca [4]. The highest activity of the enzyme (PsLam81A) was recorded at 60 °C (Figure 3b). In comparison, the reported optimal temperature for CtLam81A was 75 °C [3] and 50 °C for TfLam81A [4]. However, the temperature stability of the enzyme (PsLam81A) was low (Figure 3c). After 15 min of incubation at temperatures below 40 °C enzyme retained 97% of maximum activity, whereas at 45 °C only 59% of maximum activity was retained. Moreover, after 60 min incubation the enzyme retained 77% of the maximum activity at 40 °C temperature and showed no activity after incubation at temperatures above 50 °C. Generally, the previously characterized members of the GH81 family have been more thermostable. In comparison, CtLam81A retained 93% of maximum activity after incubation for 60 min at 75 °C [3] and TfLam81A retained approximately 90% of its maximum activity after incubation for 18 h at 50 °C [4].

Substate Specificity of PsLam81A and PsLam81A∆CBM56
Various polysaccharides (Table 1) were assayed for hydrolysis by recombinant PsLam81A and PsLam81A∆CBM56. The specific activity of each enzyme for the polysaccharide substrates was calculated by quantifying the amount of reducing sugars released during the hydrolysis reaction. The polysaccharides that were hydrolyzed by both enzyme variants were laminarin, yeast β-glucan, curdlan and pachyman ( Table 2).
The full-length enzyme variant (PsLam81A) showed the highest activity towards laminarin, followed by yeast β-glucan and curdlan with very similar specific activities and was the least active on pachyman. This data is in agreement with previously published data on the specificity of the GH81 hydrolases [3][4][5]22,26,27,39]. PsLam81A exhibited lower specific activity towards pachyman compared to laminarin (∼4.1-fold lower activity) and curdlan (∼1.8-fold lower activity). These ratios are different to previously reported data. A ∼8.5-fold lower specific activity of TfLam81A towards pachyman compared to laminarin [4], and a > 15-fold lower activity of BH026 on pachyman relative to laminarin as well as a > 10-fold lower activity on pachyman versus curdlan [40] have been reported previously. The truncated variant of an enzyme (PsLam81A∆CBM56) hydrolyzed the same substrates as a full-length enzyme. The specific activity of PsLam81A∆CBM56 towards the β-1,6 branches containing polysaccharides such as laminarin and yeast β-glucan was similar or higher in comparison to a full-length counterpart. However, PsLam81A∆CBM56 had lower specific activity in the presence of the β-1,3-linear β-glucans (curdlan and pachyman). The importance of CBM modules on curdlan hydrolysis by the GH81 family enzyme was also demonstrated in a previous study [40]. Moreover, the removal of a CBM1 module from the GH45 hydrolase affected an activity towards polysaccharides [35]. The current understanding of such activity enhancement profile by CBM on insoluble substrates, but not on soluble substrates is observed because of the increased local enzyme concentration on the surface of the polysaccharide particles via CBM binding [40]. However, there is evidence that CBM removal can decrease the enzymatic activity simultaneously on both soluble and insoluble substrates. The activity of endo-β-1,3-glucanase (TnLam16A) containing two CBM4 modules decreased by approximately 70% on laminarin and pachyman, and 60% on curdlan after removal of both CBM modules. Moreover, the enzyme with removed the N-terminal CBM4 module only showed a higher activity loss towards laminarin than curdlan and pachyman [41].
The deletion of the C-terminal module CBM56 from the BH026 enzyme resulted in ∼3-fold lower activity, and the subsequent removal of the second module CBM6 resulted in ∼6-fold lower activity on curdlan relative to a native BH026 enzyme [40]. In this study, the deletion of the CBM56 module from PsLam81A resulted in a ∼2.4-fold lower activity towards curdlan. These results show that the CBM56 module is functional in the full-length enzyme variant PsLam81A and has an effect on a specific activity of the PsLam81A enzyme when insoluble polysaccharides are used as a substrate.
Analysis of hydrolysis products of laminarin, yeast β-glucan, curdlan and pachyman by both PsLam81A and PsLam81A∆CBM56 showed similar profiles (Figure 4a,b) and laminaribiose was the main product. In the presence of linear β-glucans (curdlan, pachyman) the enzymes yielded laminaribiose as the main product and glucose as the second one. The same pattern was observed in the case of β-1,6-branches containing polysaccharides (laminarin and yeast β-glucan). However, some other hydrolysis products corresponding to compounds with a higher degree of polymerization (DP) were detected. Two spots near L6 standard were detected for both branched polysaccharides. Similar differences between laminarin and curdlan hydrolysis products after prolonged hydrolysis reaction were reported earlier [5]. Most likely, these additional spots are linear β-1,3 glucan oligosaccharides with β-1,6 branches appearing due to incomplete hydrolysis limited by β-1,6 bonds. The additional difference (hydrolysis products corresponding to L3 and L4 present in laminarin, but not present in yeast β-glucan) between these two polysaccharides containing β-1,6 branches is possibly visible due to different terminal monosaccharides composition. Yeast β-glucan contains glucose in the β-1,3 glucan backbone, while laminarin can be terminated with mannitol or glucose [8]. Therefore, laminarin may contain longer terminal oligosaccharides that are non-hydrolyzable due to the presence of mannitol in comparison to oligosaccharides with the same DP containing only glucose.
Laminaritetraose, laminaripentaose and laminarihexaose were completely hydrolyzed after 2 h incubation while laminaritriose was not degraded completely. However, the 18 h prolonged incubation of PsLam81A and PsLam81A∆CBM56 with curdlan and pachyman resulted in glucose and laminaribiose without the accumulation of laminaritriose (Figure 4a,b). These observations indicate that tested enzymes have lower activity towards shorter oligosaccharides compared to longer ones. This is in agreement with earlier reported data on the endo-hydrolytic activity of the GH81 hydrolases when a decrease in activity correlated with the decrease of the length of oligosaccharide [22,27,28,39]. Laminaribiose was not degraded by PsLam81A and PsLam81A∆CBM56, indicating that laminaritriose was a minimal substrate for both enzymes. There is one other endo-glucanase belonging to the GH81 family that is reported to use laminaritriose as a minimal substrate [5]. Moreover, the disappearance of laminaritriose during laminarihexaose hydrolysis indicates that laminaritriose can be used as a substrate by β-1,3-glucanase Lam81A from Thermobifida fusca [4]. Also, the decreasing proportion of laminaritriose with increasing curdlan hydrolysis time indicates that laminaritriose is a substrate of the enzyme from Clostridium thermocellum [3]. However, some of the reported data is not straightforward, because a low activity on laminaritriose is not always considered to be sufficient to conclude that laminaritriose is a substrate for the GH81 family glucanases [22,28].  The analysis of hydrolysis products of laminari-oligosaccharides with PsLam81A and PsLam81AΔCBM56 showed that both enzymes were able to degrade laminarihexaose (L6), laminaripentaose (L5), laminaritetraose (L4), laminaritriose (L3) down to laminaribiose (L2) and glucose (Figure 5a,b). The analysis of hydrolysis products of laminari-oligosaccharides with PsLam81A and PsLam81AΔCBM56 showed that both enzymes were able to degrade laminarihexaose (L6), laminaripentaose (L5), laminaritetraose (L4), laminaritriose (L3) down to laminaribiose (L2) and glucose (Figure 5a,b).  All this data showed that PsLam81A as several other hydrolases from the GH81 family can act on laminaritriose as a minimal substrate. In addition, the identical results for both enzymes PsLam81A and PsLam81A∆CBM56 indicated that the CBM56 module does not have an effect on the substrate specificity for oligosaccharides.

Conclusions
The endo-β-1,3-glucanase PsLam81A from Paenibacillus sp. GKG was identified, expressed heterologously and characterized as a representative of the GH81 family. PsLam81A hydrolyzed β-1,3 glycosidic bonds containing polysaccharides such as laminarin, yeast βglucan, curdlan, and pachyman down to laminaribiose as the main hydrolysis product. The minimal laminari-oligosaccharide substrate hydrolyzed by PsLam81A was laminaritriose. PsLam81A characterized in this study has a unique domain organization compared to other characterized GHs of the family 81. PsLam81A contains a single CBM56 domain that ensures more efficient hydrolysis of water-insoluble substrates. Collected data encourage further studies focused on the application of PsLam81A for hydrolysis of the yeast wall under industrial conditions. Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/microorganisms10101930/s1, Figure S1: Extracellular hydrolases producing bacteria growing on solid mineral medium containing water-insoluble yeast cell wall substrate as a carbon source; Figure S2: SDS-PAGE analysis of recombinant PsLam81A and PsLam81A∆CBM56 proteins induced at different temperatures; Figure S3: The graphical representation of PsLam81A sequence coverage (86%) obtained after tryptic digestion; Figure S4: The graphical representation of PsLam81A sequence coverage (78%) obtained after chymotryptic digestion; Table S1: Primers used in this study; Table S2

Informed Consent Statement: Not applicable.
Data Availability Statement: The mass spectrometry proteomics data have been deposited to the Pro-teomeXchange Consortium via the PRIDE [42] partner repository with the dataset identifier PXD036565.

Conflicts of Interest:
The authors declare no conflict of interest.