Role of carbohydrate binding module (CBM3c) of GH9 β-1,4 endoglucanase (Cel9W) from Hungateiclostridium thermocellum ATCC 27405 in catalysis

Carbohydrate Research 484 (2019) 107782

Contents lists available at ScienceDirect

Carbohydrate Research journal homepage: www.elsevier.com/locate/carres

Role of carbohydrate binding module (CBM3c) of GH9 β-1,4 endoglucanase (Cel9W) from Hungateiclostridium thermocellum ATCC 27405 in catalysis

T

Krishan Kumara, Shubham Singalb, Arun Goyala,∗ a

Carbohydrate Enzyme Biotechnology Laboratory, Department of Biosciences and Bioengineering, Indian Institute of Technology Guwahati, Guwahati, 781039, Assam, India b Department of Microbiology, Panjab University, Chandigarh, 160014, India

ARTICLE INFO

ABSTRACT

Keywords: Hungateiclostridium thermocellum β-1,4-Glucanase Family 9 glycoside hydrolase Carbohydrate binding module 3c (CBM3c)

The function of CBM3c in the enzymatic catalysis varies among the members of family 9 Glycoside Hydrolases (GH). A new member of family 9 GH (Cel9W) from thermophilic anaerobic bacterium, Hungateiclostridium thermocellum was explored for elucidation of the role of CBM3c in catalysis by GH9. Cel9W is a multimodular theme B1, cellulase enzyme comprising a catalytic module of family 9 glycoside hydrolase (HtGH9t) at Nterminal, a family 3c carbohydrate binding module (HtCBM3c) and a dockerin domain at the C-terminal. The ORF of Cel9W encoding full length β-1,4-glucanase, HtGH9 (containing both GH9 and CBM3c modules), the truncated GH9 catalytic module (HtGH9t) and module CBM3c (HtCBM3c) were cloned and over-expressed using E. coli BL21 cells. HtGH9 showed maximum activity at pH 6.5 and 90 °C. It displayed highest activity of 64 U/mg against lichenan followed by 44.6 U/mg (β-glucan) and 22.3 U/mg (Carboxymethyl cellulose). HtGH9 showed stability in the pH ranging from 5.0 to 9.0 and thermal stability up to 70 °C for 1.0 h. The presence of EDTA and EGTA decreased the activity of HtGH9 and also shifted the melting curve peak from 93 °C to 88 °C indicating that the enzyme inherently possesses metal ions, which play role in catalysis and structural stability. The TLC analysis of HtGH9 hydrolysed Avicel showed the presence cellotetraose indicating the processive endoglucanase activity in HtGH9. The module, HtGH9t alone exhibited very low level of activity (1.22 U/mg) against lichenan. It partially recovered (51%) the enzyme activity in presence of equimolar concentration of HtCBM3c. Non-denaturing gel electrophoresis revealed a non-covalent binding interaction between the catalytic HtGH9t module with HtCBM3c module and their physical association showed the partial recovery of its endoglucanase activity. This study confirmed that the physical association of the catalytic module HtGH9t with HtCBM3c is necessary for HtGH9 to efficiently hydrolyze the cellulosic substrates.

1. Introduction Cellulose is an unbranched polysaccharide of glucose units linked by β-1,4-glycosidic bond. This carbohydrate is plenty in plant biomass that can be converted into simple sugar (glucose) and subsequently to alcohol and therefore can be used as a renewable energy source [1]. Cellulose and other polysaccharides of plant biomass are degraded or hydrolysed by many bacteria and fungi by releasing different types of cellulases and other specific enzymes either in free form or in a multienzyme complex called cellulosome [2–6]. Cellulases are the catalytic enzymes that hydrolyze the β-1,4-glycosidic bond of the cellulose chain and categorized in to different glycoside hydrolase (GH) families based on the amino acid sequences [7–9]. Cellulose can be converted in to a simple form of sugar (glucose) by the synergistic action of three enzymes namely endo-β-1,4-glucanase, exo-β-1,4-glucanase and β-1,4∗

glucosidase [1]. Endo-β-1,4-glucanase is also called non-processive cellulase, because its active site structure possesses an open groove which can fit any part of the linear chain of the cellulose and produces cellodextrins as breakdown products [10]. The processive endo-β-1,4glucanase makes initially a random internal cut on the cellulose chain followed by sequential (processive) cleavage of either cellobiose or cellotetrose [11,12]. However, the processive exo-β-1,4-glucanase attaches to one end of the substrate (either reducing or non-reducing end) and splits off the cellobiose in a sequential (processive) manner [10,13,14]. Severval cellulases and other carbohydrate degrading enzymes are often associated with carbohydrate binding modules (CBMs), which increase the enzyme-substrate proximity and augment substrate binding and degradation [15]. CBMs have been classified into 85 families in CAZY database (http://www.cazy.org/Carbohydrate-Binding-

Corresponding author. E-mail addresses: [email protected] (K. Kumar), [email protected] (S. Singal), [email protected] (A. Goyal).

https://doi.org/10.1016/j.carres.2019.107782 Received 31 May 2019; Received in revised form 6 August 2019; Accepted 20 August 2019 Available online 20 August 2019 0008-6215/ © 2019 Elsevier Ltd. All rights reserved.

Carbohydrate Research 484 (2019) 107782

K. Kumar, et al.

Modules.html) as of May 2019. Among CBMs, family 3 CBM show cellulose binding function and are sub-grouped in to CBM3a, CBM3b and CBM3c [16–17). CBM3a and CBM3b strongly bind the crystalline cellulose, while CBM3c does not bind the crystalline cellulose, but modulates the glucanase (endo- or exo-) activity of family 9 GH [16–18]. The removal of CBM3c from catalytic GH9 module of Clostridium thermocellum Cel9I [11], Clostridium cellulolyticum Cel9G [19] and Paenibacillus barcinonensis Cel9B [20] significantly reduced the enzyme activity indicating the importance of CBM3c in the catalysis. However, the removal of CBM3c from catalytic GH9 module of Thermomonospora fusca Cel9A did not significantly affect the enzyme activity. Though the removal reduced the soluble to insoluble reducing sugar ratio from 3.5 to 0.6. This indicated that CBM3c removal may

Table 1 Oligonucleotide primer sequences used for cloning full length HtGH9 and truncated derivatives HtGH9t and HtCBM3c from Hungateiclostridium thermocellum. Construct

Primer sequence

HtGH9

Forward: 5′- CG GCTAGC TTCAACTACGGAGAAGCCC -3′ Reverse: 5′- CC CTCGAG TTACTCAGGTTCTATGCCTGCA -3′ Forward: 5′- CG GCTAGC TTCAACTACGGAGAAGCCC -3′ Reverse: 5′- CC CTCGAG TTACGGTCTTCCGTCATATCTG -3′ Forward: 5′- CG GCTAGC GCTATTGAAGTGCCGGAAG Reverse: 5′- CC CTCGAG TTACTCAGGTTCTATGCCTGCA

HtGH9t HtCBM3c

Fig. 1. Schematic diagram of (A) molecular architecture of Cel9W and its truncated derivatives HtGH9, HtGH9t and HtCBM3c developed by Dog2.0 software and (B) Phylogenetic analysis of family 9 GH enzymes and HtGH9. The tree was generated using neighborjoining method in MEGA 7 software.

2

Carbohydrate Research 484 (2019) 107782

K. Kumar, et al.

Table 2 Substrate specificity of HtGH9 against different substrates. Substrate (1%, w/v)

HtGH9

HtGH9t

HtGH9t:HtCBM3c (1:1)*

Specific Activity (U/mg) Lichenan β-glucan Carboxy methylcellulose (CMC) Avicel Carob galactomannan Beechwood Xylan Laminarin

64.3 44.6 22.3 0.26 – – –

1.22 0.93 0.62 – – – –

34 27 7.2 – – – –

*Molar ratio. -No activity detected.

constructed using the neighbor-joining method that utilizes Mega 7 software to understand the evolutionary relationship between HtGH9 and characterized protein of family 9 GH. 2.3. Cloning of HtGH9, HtGH9t and HtCBM3c The Hungateiclostridium thermocellum ATCC 27405 genomic DNA was purchased from DSMZ, Germany. The ORF regions of Cel9W encoding β-1,4-glucanase full length (HtGH9), truncated GH9 catalytic domain (HtGH9t) and carbohydrate binding module 3 (HtCBM3c) were amplified by using the pair of oligonucleotide primers for each (Table 1). The forward primer and reverse primer contained NheI and XhoI restriction sites. Each gene was amplified by Phusion High-Fidelity DNA Polymerase (ThermoFisher Scientific, USA) using PCR conditions of initial denaturation at 98 °C for 1 min followed by 30 cycles of denaturation at 98 °C for 10s, annealing for 30 s at 48 °C for HtGH9 and HtGH9t and 52 °C for HtCBM3c, extension at 72 °C for HtGH9 and HtGH9t for 90 s and at 72 °C for 30 s for HtCBM3c and final extension of each gene was at 72 °C for 10 min. The PCR amplified products were run on 0.8% (w/v) agarose gel electrophoresis. The DNA fragment matching to the size of gene of interest were excised out and purified by the GelElute gel extraction kit (Sigma-Aldrich USA). Each purified PCR product and expression vector pET28a(+) were digested by NheI and XhoI restriction enzymes (Promega, USA). Each digested gene was ligated to pET28a(+) vector by using T4 DNA ligase (Promega, USA). Then the recombinant plasmids were transformed in to E. coli TOP10 cells. The cloned recombinant plasmids were isolated from TOP10 cells and purified. They were subjected to restriction digestion by NheI and XhoI enzymes to confirm the positive clone. After confirmation the recombinant plasmids were transformed in to E. coli BL21 cells for protein expression.

Fig. 2. Purified recombinant protein on SDS-PAGE gel (12%, w/v). Lane M − Molecular mass marker (14.3–120 kDa, Bio-Bharti), lane 1- HtGH9, lane 2HtGH9t and lane 3- HtCBM3c.

change the processive endo-acting GH9 catalytic module to non-processive endoglucanase that released lower soluble sugars [21,22]. The cellulosome (multienzyme complex) of Hungateiclostridium thermocellum contains a set of cellulase enzymes. One such cellulosomal enzyme (Cel9W) from Hungateiclostridium thermocellum belongs to family 9 GH are cellotetraose producing processive endoglucanase enzyme [12]. In this study, a multimodular enzyme, Cel9W comprising a GH9 catalytic module (HtGH9t) at N-terminal followed by a carbohydrate binding module (CBM3c) and a dockerin domain at its C-terminal were studied. The gene encoding full length β-1,4-glucanase (HtGH9), GH9 catalytic module (HtGH9t) and CBM3c module (HtCBM3c) from thermophilic bacterium Hungateiclostridium thermocellum ATCC 27405 were cloned and expressed in E. coli BL21 cells, purified and characterized. The role of HtCBM3c in catalysis was explored by studying its physical interactions with catalytic module HtGH9t. 2. Material and methods 2.1. Plasmid, bacterial strains and chemicals Cloning and expression hosts used were Escherichia coli strain TOP10 (Novagen) and BL-21 (Novagen), respectively. The amplified genes were cloned in to pET28a(+) vector for cloning and expression. Carboxy methylcellulose (CMC), Lichenan, β-D-Glucan from Barley, Avicel, Beechwood xylan and Laminarin (from Laminaria digitata) were procured from Sigma-Aldrich Chemical Co., USA. Carob galactomannan was purchased from Megazyme Ltd. Ireland. Thin layer chromatography plate (TLC Silica gel 60 F254, 20 × 20 cm) was from Merck, India.

2.4. Expression and purification of HtGH9, HtGH9t and HtCBM3c The E. coli BL-21 cells harbouring the recombinant plasmid were grown in 400 ml Luria broth medium having kanamycin (50 μg/ml) at 37 °C till the absorbance (A550) reached 0.5. Then 0.4 ml of 1 mM isopropyl 1-thio-ß-D-galactopyranoside (IPTG) was added to the culture for enzyme induction and incubated at 24 °C with shaking at 180 rpm for 16 h. The induced cells were centrifuged at 5,000 g at 4 °C for 10 min. The cell pellet was resuspended in 8 ml 50 mM Tris-HCl buffer (pH 7.4) containing 60 mM Imidazole and 300 mM NaCl. The resuspended cells were sonicated on ice with 10s on and 10s off pulse at 32% amplitude. The sonicated cells were centrifuged at 16,000 g, 4 °C for 45 min. The cell free supernatant was filtered through 0.45 μm membrane and loaded onto through 5 ml Ni+2 ion charged sepharose column (HiTrap, GE Healthcare, USA). The column was washed with 100 ml washing buffer (50 mM TrisHCl, pH 7.5, 300 mM NaCl and 60 mM imidazole) and the protein was eluted in 1 ml fractions by using 16 ml elution buffer (50 mM Tris-HCl buffer, pH 7.5, 300 mM NaCl and 300 mM Imidazole). The eluted fractions

2.2. Molecular architecture of Cel9W and phylogenetic analysis The protein sequence of family 9 GH, β-1,4-glucanase (Cel9W) with a GenBank accession number ABN51980.1 and UniProt ID A3DDF1 was retrieved from NCBI database. The conserved domain in the sequence was analysed by submitting the amino acid sequence of Cel9W to the conserved domain database (http://www.ncbi.nlm.nih.gov/cdd/) followed by BLAST analysis. The location of signal peptide was determined by using the SignalP 3.0 server. The phylogenetic tree was 3

Carbohydrate Research 484 (2019) 107782

K. Kumar, et al.

Fig. 3. pH and Temperature profile of HtGH9 and its stability using 1.0% (w/v) carboxymethyl cellulose and 0.02 mg/ml of enzyme HtGH9. A) HtGH9 was incubated at 90 °C at different pH ranging from pH 3.0–9.0 using 50 mM sodium citrate-phosphate (3.0–7.0), sodium phosphate (6.0–8.0) and Tris-HCl (7.5–9.0) buffers for 2 min. (B) pH stability of HtGH9 was determined by pre-incubating the enzyme at different pH ranging from 3.0 to 9.0 using 50 mM sodium citrate-phosphate (3.0–7.0), 50 mM sodium phosphate (6.0–8.0) and Tris-HCl (7.5–9.0) buffers at 25 °C. (C) 100 μl reaction mixture containing HtGH9 (0.02 mg/ml) and 1% (w/v) CMC in 50 mM sodium citrate-phosphate (pH 6.5) incubated at different temperature ranging from 30 °C to 100 °C for 2 min. (D) Enzyme thermostability was determined by pre-incubating the HtGH9 at pH 6.5 different temperatures ranging between 4 and 100 °C. The highest activity observed at 90 °C and pH 6.5 in sodium citrate-phosphate was defined as the 100% activity to calculate the relative activity. The reaction was carried out in duplicate sets.

the reaction mixture (100 μl) containing the enzyme (0.02 mg/ml) and 1% (w/v) CMC in 50 mM citrate-phosphate (pH 6.5) at varied temperature, ranging from 30 °C to 100 °C for 2 min. The amount of reducing sugar and enzyme activity was examined as mentioned above. HtGH9 stability at different pH was examined by incubating the enzyme in different 50 mM buffers (citrate-phosphate (pH 3.0–7.0), phosphate (pH 5.8–8.0) and Tris (pH 7.5–9.0) for 60 min at 25 °C. The specific activity was measured at optimized conditions of pH 6.5 and temperature 90 °C. The temperature stability of HtGH9 was determined by incubating the HtGH9 enzyme in 50 mM citrate-phosphate (pH 6.5) at varied temperature range (30 °C–100 °C) for 60 min and specific activity was determined at optimum conditions of pH 6.5 and 90 °C. The kinetic parameters of HtGH9 were determined by incubating the enzyme HtGH9 with different concentrations of CMC, β-D-glucan and Lichenan at pH 6.5 and 90 °C. Michaelis-Menten plots and Lineweaver-Burk plots were generated using GraphPad Prism 6 to determine the Km and Vmax. The activity of HtGH9 in the presence of various metal ions (10 mM MgCl2, CaCl2, MnCl2, CuSO4, FeCl3, CoCl2, ZnSO4 and NiSO4) and chelating agent (10 mM EDTA and EGTA) was also determined by incubating 10 mM of each in 100 μl reaction mixture containing 1% (w/v) CMC. The presence of inherent Ca2+ ion(s) in the HtGH9 protein was explored with the help of flame photometer (Systronics, India), attached with a calcium filter. A known concentration of HtGH9 (0.4 mg/ml, 5.6 μM) was introduced to flame photometer and concentration of Ca+2 ions was determined.

containing protein were pooled and dialysed against 50 mM Tris-HCl buffer (pH 7.5) to remove the salt and imidazole. The size and purity of each purified protein was examined by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) using 14% (w/v) gel [23]. The polyacrylamide gels were stained with Coomassie Brilliant Blue R-250 using conventional procedures [24]. 2.5. Substrate specificity of HtGH9 and HtGH9t The substrate specificity of HtGH9 and HtGH9t was investigated by using 1% (w/v) of Carboxy methylcellulose (CMC), Lichenan, β-Dglucan from Barley, Avicel, Beechwood xylan, Laminarin (from Laminaria digitate) and Carob galactomannan. The 100 μl reaction mixtures contained 1% (w/v) substrate dissolved in 50 mM citratephosphate buffer, pH 6.5 and 10 μl of HtGH9 (0.067–0.02 mg/ml) or HtGH9t (1.4 mg/ml), respectively. The reaction mixtures of HtGH9 and HtGH9t were incubated at 90 °C for 2 min and 60 °C for 5 min, respectively. The reducing sugar produced in the reaction mixture was quantified as described earlier [25,26] by using glucose as a standard and specific activity calculated. 2.6. Biochemical characterization of HtGH9 The effect of pH on the activity of HtGH9 was analysed by using 50 mM buffers (sodium citrate-phosphate (pH 3.0–7.0), sodium phosphate (pH 5.8–8.0) and Tris HCl (pH 7.5–9.0). The 100 μl reaction mixture contained 0.02 mg/ml enzyme and 1% (w/v) CMC dissolved in different buffers and incubated at 90 °C for 2 min and the reducing sugar was estimated and enzyme activity calculated. The effect of different temperatures on HtGH9 activity was determined by incubating

2.7. Protein melting curve of HtGH9 The melting temperature of HtGH9 enzyme was analysed by incubating HtGH9 (50 μg/ml) in 50 mM Tris-HCl buffer (pH 6.5) at 4

Carbohydrate Research 484 (2019) 107782

K. Kumar, et al.

varying temperatures, ranging between (25–100 °C) in a UV–Visible spectrophotometer (Varian, Cary 100 Bio) connected to Peltier temperature controller. A change in absorbance at 280 nm with respect to the temperature was measured and a curve between absorbance vs temperature was generated. The experiments were also carried out separately by adding 10 mM CaCl2, EDTA or EGTA to examine their effect on the melting of HtGH9 enzyme. 2.8. Catalytic mechanism of HtGH9 The mode of action of HtGH9 was analysed by running the hydrolysed products of CMC and avicel on Thin layer chromatography (TLC) (TLC Silica gel 60 F254, 20 × 20 cm, Merck). 100 μl of reaction mixture containing 10 μl of HtGH9 (0.2 mg/ml) and 90 μl of 1.1% (w/v) CMC and the reaction mixture containing 10 μl of HtGH9 (3 mg/ml) and 90 μl of 1.1% (w/v) avicel dissolved in 50 mM citrate-phosphate buffer (pH 6.5) were incubated at 70 °C for different time intervals. Enzymatic hydrolysis reaction was stopped by adding 400 μl of absolute ethanol and centrifuged at 12,000 g for 15 min. The resulting supernatant containing the released hydrolysed product was transferred to a clean 1.5 ml micro-centrifuge tube and concentrated to 10 μl by evaporating at 70 °C in the hot-air oven. One microliter of 10 μl concentrated reaction mixtures and 1 μl of standards viz. glucose, cellobiose, cellotriose and cellotetraose (1.0 mg/ml each) were loaded on the TLC plate. The mobile phase consisted of n-butanol/acetic acid/water in the ratio 2:1:1 [27]. The TLC plate was dried and resulting oligosaccharides were visualized by pouring the visualizing solution (sulphuric acid: methanol 5:95, v/v and α-naphthol 0.5%, w/v) and heating at 100 °C. 10 μl aliquot from 100 μl reaction mixture containing CMC was taken for reducing sugar estimation. The amount of reducing sugar released by CMC hydrolysis was quantified by the method as described earlier [25,26]. 2.9. Interaction study of HtCBM3c with catalytic HtGH9t The interaction of HtCBM3c with catalytic HtGH9t was carried out by loading the mixture containing 10 μl of HtGH9t (19 μmol) with 10 μl of HtCBM3c in different concentrations, 9.8, 19 and 38 μmol making 1:0.5, 1:1 and 1:2 M ratios, respectively, of HtGH9t: HtCBM3c on native-PAGE using 7.5% (w/v) gel The mobility pattern of each mixture was compared with HtGH9t (19 μmol) and HtCBM3c (38 μmol) separately loaded on to the native gel. The polyacrylamide gel was stained with Coomassie Brilliant Blue R-250 using conventional procedures [24].

Fig. 4. Michaelis-Menten and Lineweaver-Burk plots (inset) for determining the kinetic parameters of HtGH9 for (A) lichenan (B) β-glucan and (C) CMC. The assay was performed in 100 μl reaction mixture containing different concentrations of substrates at pH 6.5 (sodium citrate-phosphate) and incubated at 90 °C for 2 min. The reaction was carried out in duplicate sets.

2.10. Activity of catalytic module HtGH9t in absence and presence of HtCBM3c The activity of catalytic module HtGH9t was analysed in the absence and in presence of HtCBM3c to study the role of HtCBM3c in the enzyme catalysis. The effect of different temperature was analysed by incubating the reaction mixture (100 μl) containing 1% (w/v) carboxymethyl cellulose in 50 mM citrate-phosphate buffer (pH 6.5) with 0.9 μM HtGH9t or mixture of 0.9 μM HtGH9t and 0.9 μM HtCBM3c at different temperature ranging from 30 to 100 °C for 2 min. The reducing sugar produced in the reaction mixture was quantified as described earlier [25,26] and specific activity calculated as mentioned earlier. The effect of different molar concentration of HtCBM3c on the activity of catalytic module HtGH9t was determined by incubating the fixed amount of HtGH9t (0.9 μM) with different concentrations of HtCBM3c (0.23 μM–3.6 μM) in 100 μl reaction mixture containing 1% (w/v) carboxymethyl cellulose in ctirate-phosphate buffer (pH 6.5) at 80 °C for 2 min. The amount of reducing sugar and specific activity was calculated as mentioned above.

Table 3 Effect of metal ions on the activity of HtGH9. Metal ions/reagents (10 mM)

Relative activity (%)

Control Ca2+ Mg2+ Mn2+ Ni2+ Co2+ Zn2+ Cu2+ Fe2+ EDTA EGTA

100.0 97.7 ± 1.5 92.3 ± 1.9 95.3 ± 1.9 15.2 ± 1.5 6.2 ± 0.6 4.1 ± 0.6 3.7 ± 1.1 2.8 ± 0.2 63.5 ± 2.1 62.9 ± 2.5

5

Carbohydrate Research 484 (2019) 107782

K. Kumar, et al.

Fig. 5. Temperature melting curve of (A) HtGH9, (B) HtGH9 + 10 mM Ca2+ ions, (C) HtGH9 + 10 mM EDTA and (D) HtGH9 + 10 mM EGTA. 50 μg of HtGH9 protein was mixed in 1 ml 50 mM Tris-HCl buffer (pH 7.5) and exposed to different temperatures ranging 25–100 °C on a UV–Visible spectrophotometer.

3. Results and discussion

3.2. Substrate specificity of HtGH9 and HtGH9t

3.1. Cloning, expression and purification of HtGH9, HtGH9t and HtCBM3c

The cellulosic polysaccharides hydrolysed by HtGH9 and HtGH9t are shown in Table 2. HtGH9 exhibited maximum activity against lichenan (64.3 U/mg) followed by β-D-glucan (44.6 U/mg), carboxy methylcellulose (22.3 U/mg) and avicel (0.26 U/mg), while HtGH9t showed only 1.22 U/mg, 0.93 U/mg and 0.62 U/mg activity against lichenan, β-D-glucan and carboxy methylcellulose, respectively. Both HtGH9 and HtGH9t were unable to hydrolyze carob galactomannan, beechwood xylan and laminarin. These results showed that both HtGH9 and HtGH9t hydrolysed mixed β-1,3; 1,4-linked glucan (β-D-glucan and lichenan) and β-1,4glucan (CMC) indicating their β-1,4-glucanase activity. The activity of HtGH9 against insoluble substrates (avicel) was very less as compared with the soluble substrates, similar to the activities of endoglucanases from Bacillus pumilus and Paenibacillus sp. X4 [29,30]. In contrast endoglucanases of Caldicelllulosiruptor bescii and Clostridium thermocellum showed high activity against crystalline celluloses [31,32]. The activity of catalytic module (HtGH9t) in the absence of carbohydrate binding module was significantly low as compared to the full length module (HtGH9), indicating the necessity of carbohydrate binding module (HtCBM3c) in the enzymatic catalysis. Similar, results were reported for GH9 module of Clostridium thermocellum Cel9I [11], Clostridium cellulolyticum Cel9G [19] and Paenibacillus barcinonensis Cel9B [20].

Sequence analysis of Cel9W by using conserved domain database and BLAST tool revealed the presence of an N-terminal signal peptide followed by catalytic domain family 9 GH (HtGH9t), carbohydrate binding module (HtCBM3c) and type-1 dockerin domain at the Cterminal as shown in Fig. 1A. The amino acid sequences of HtGH9 was compared with previously characterized family 9 GH enzymes using phylogenetic tree analysis (Fig. 1B). The enzymes are separated into two distinct clusters (branch I and branch II, Fig. 1B). The nine enzymes of branch II are more distantly related to HtGH9. The branch I of phylogenetic tree contains 16 enzymes closely related to HtGH9. In branch I Cel9N of Clostridium thermocellum is the closest enzyme in family 9 GH to HtGH9 [28]. This enzyme also has CBM3c to the downstream of the catalytic domain but it was found as a non-processive endoglucanase [28] while, HtGH9 (Cel9W without dockerin) is prossessive endoglucanase [12]. The gene encoding HtGH9, HtGH9t and HtCBM3c were amplified by PCR and cloned in to pET28a(+) vector. The recombinant plasmid for each gene was confirmed by NheIXhoI restriction digestion and sequencing. The recombinant proteins HtGH9, HtGH9t and HtCBM3c were purified by immobilized metal ion affinity chromatography (IMAC) and examined by SDS-PAGE electrophoresis (Fig. 2). The purified proteins HtGH9, HtGH9t and HtCBM3c showed molecular sizes of approximately, 71.7, 52.9 and 20.7 kDa, respectively which were of the expected size as calculated from their amino acid sequences.

3.3. Biochemical properties of HtGH9 The effect of pH and temperature on the activity and stability of HtGH9 was determined by using carboxymethyl cellulose as substrate. 6

Carbohydrate Research 484 (2019) 107782

K. Kumar, et al.

Fig. 7. Native-PAGE (7.5%, w/v gel) displaying the interaction of HtGH9t with HtCBM3c. 10 μl of HtGH9t (19 μmol) was mixed with increasing amount of HtCBM3c (9.5 μmol–38 μmol) and loaded onto the gel. Lanes: (a) BSA, (b) HtCBM3c, (c) HtGH9t and (d, e and f) HtGH9t mixed with increasing ratios of HtCBM3c.

83% activity at pH 5.0 (Fig. 3B). However, below pHs 4.0, HtGH9 rapidly lost its activity. HtGH9 exhibited maximum activity at 90 °C (Fig. 3C) and retained around 84% activity even at 100 °C. The optimum temperature of HtGH9 was much higher than 50 °C of endoglucanase, Cel9K from Paenibacillus sp. X4 [30], 60 °C of BlCel9 from Bacillus licheniformis [34], 53 °C of CelB from Paenibacillus sp. BP-23 [33] and 70 °C of processive endoglucanase, CenC from Clostridium thermocellum [32]. HtGH9 displayed thermal stability up to 70 °C and retained 67% of its activity at 75 °C after 1 h incubation (Fig. 3D). These results showed the thermophilic nature of the enzyme. The stability of HtGH9 over a wide range of pH and temperature makes it suitable for industrial applications. Several industrial processes require thermostable enzymes for better substrate solubility, easy mixing, high mass transfer rate and reduced risk of contamination [35]. The kinetic parameters of HtGH9 against lichenan, β-glucan and carboxy methylcellulose were analysed with the help of Michaelis-Menten plot and Lineweaver-Burk plot. HtGH9 displayed Vmax of 70.2 ± 2.0, 48.86 ± 2.6 and 26.19 ± 1.0 U/mg and Km of 1.1 ± 0.1, 0.72 ± 0.1 and 2.7 ± 0.2 mg/ml against lichenan, β-glucan and carboxy methylcellulose, respectively (Fig. 4). The activity of HtGH9 in the presence of 10 mM divalent metal ions, EDTA and EGTA was also explored by using carboxymethyl cellulose as substrate. The presence of Ca2+, Mg2+ or Mn2+ ions did not affect the activity of HtGH9 (Table 3). HtGH9 activity was adversely affected by Zn2+, Cu2+, Co2+, Ni2+ and Fe2+ ions, while it gave 63% residual activity in the presence of EDTA or EGTA (Table 3). The Ca2+ ions did not alter the enzyme activity of HtGH9. This result was similar to endoglucanase, Blcel9 of Bacillus licheniformis which also showed no effect of Ca2+ ions [34], whereas, the activity of endoglcuanase, Cel9k of Paenibacillus sp. X4 [30] and processive endoglucanase, CenC from Clostridium thermocellum was increased [32]. The decrease in the activity of HtGH9 in the presence of EDTA and EGTA indicated that the native enzyme, HtGH9 may contains inherently bound Ca2+ ion(s), which play a role in the enzyme catalysis. The presence of Ca2+ ion(s) in the protein was determined by flame photometer. The result showed that 5.6 μM of HtGH9 contained 2.5 ppm (62.5 μM) of Ca2+ ions. This is equivalent to, one molecule of HtGH9 protein contains approximately, 11 number of Ca2+ ions. This result confirmed that HtGH9 protein contains inherent Ca2+ ions which play a role in enzyme catalysis.

Fig. 6. The amount of reducing sugar released by (A) HtGH9 hydrolysis of CMC (inset containing reducing sugar released up to 1 h. TLC analysis (B) of CMC hydrolysed products at different time interval and (C) avicel at different time interval. The standards used in TLC are G1 (Glucose), G2 (Cellobiose), G3 (Cellotriose) and G4 (Cellotetraose).

HtGH9 displayed maximum activity at pH 6.5 of both, citrate-phosphate and phosphate buffers as shown in Fig. 3A. The optimum pH of HtGH9 was higher than pH 5.5 of endoglucanase, Cel9K from Paenibacillus sp. X4 [30], pH 6.0 of processive endoglucanase, CenC from Clostridium thermocellum [32] and pH 5.5. of Celb from Paenibacillus sp. BP-23 CelB [33] and lower than pH 7.0 of BlCel9 from Bacillus licheniformis [34]. The β-1,4-glucanase activity of HtGH9 was stable over a wide range of pH (5.0–9.0) as it retained 87% of activity at pH 9.0 and 7

Carbohydrate Research 484 (2019) 107782

K. Kumar, et al.

showed that the hydrolysis reached saturation at 12 h. The catalytic mechanism of HtGH9 was examined by analyzing the hydrolysed products of CMC and avicel at different time intervals by TLC (Fig. 6 B&C). The chromatogram of hydrolysed CMC displayed the predominant presence of cellotetraose and higher cello-oligosaccharides at all the time intervals indicating the endo-β-1,4-glucanase activity of the HtGH9. However, the chromatogram of hydrolysed avicel displayed the presence of cellotetraose as a major product released at all the time intervals (Fig. 6C), however, no larger cello-oligosaccharides were detected. This indicated that HtGH9 is cellotetraose producing processive endoglucanase as also described by Leis et al. (2017) [12]. The cellotetraose producing processive endoglucanase activity of HtGH9 is different from other family GH9 proteins like Cel9B from Paenibacillus barcinonensis which is cellobiose producing processive endoglucanase, while Cel9N from Clostridium thermocellum showed the non-processive endoglucanase activity [20,28]. 3.6. Interaction study of HtCBM3c with catalytic HtGH9t The interaction of HtCBM3c with HtGH9t was studied by performing the native PAGE. The catalytic module HtGH9t was mixed with increasing molar ratios of HtCBM3c, ranging from 1:0.5 to 1:2. The mobility of each protein mixture was compared with the individual protein on the native-PAGE. The mixture of HtGH9t and HtCBM3c showed the formation of new band and its intensities increased with increase in HtCBM3c molar concentrations. The presence of new band indicated the formation of complex by non-covalent interaction of HtCBM3c with HtGH9t (Fig. 7). These results indicated that HtCBM3c may play a role in the HtGH9 catalytic activity. A similar study was also reported for processive endoglucanase, Cel9I from Clostridium thermocellum displaying the physical interaction between CBM3c module and catalytic GH9 module essential for catalysis [11].

Fig. 8. Effect of HtCBM3c on the activity of HtGH9t. (A) 0.9 μM HtGH9t or mixture of 0.9 μM HtGH9t and 0.9 μM HtCBM3c were preincubated at 25 °C for 15 min in 50 mM sodium ctirate-phosphate buffer, pH 6.5, then their activity was calculated using 100 μl reaction mixture containing 1% (w/v) carboxymethyl cellulose in 50 mM sodium citrate-phosphate buffer, pH 6.5 at different temperature ranging from 30 to 100 °C for 2 min. (B) The effect of different molar concentrations of HtCBM3c on the activity of HtGH9t was determined by preincubating 0.9 μM of HtGH9t with varying concentration of HtCBM3c (from 0.23 μM to 3.6 μM) in 50 mM sodium ctirate-phosphate buffer, pH 6.5 at 25 °C for 15 min and then the enzyme activity was determined by using 100 μl reaction mixture containing 1% (w/v) CMC in 50 mM sodium ctirate-phosphate buffer, pH 6.5 at 80 °C for 2 min.

3.7. Activity of catalytic module HtGH9t in absence and presence of HtCBM3c The effects of physical association of HtGH9t and HtCBM3c on the catalysis at varying temperatures at equimolar concentration and also at different molar ratios of (HtGH9t: HtCBM3c) were determined. The equimolar mixture of HtGH9t and HtCBM3c showed maximum activity of 7.2 U/mg with CMC at 80 °C, while HtGH9t alone showed maximum activity (0.62 U/mg) at 60 °C (Fig. 8A). Whereas, the full length HtGH9 gave maximum activity of 22.3 U/mg at 90 °C (Table 2). These results also showed that the associated CBM3c module imparts thermostability as well as plays a role in catalysis. The enzyme activity of mixture increased from 0.62 U/mg to 12.5 U/mg against CMC at 80 °C with increase in the molar ratio of HtGH9t: HtCBM3c from 1:0 to 1:3 (Fig. 8B). The enzyme activity of mixture steadily increased beyond the equimolar ratio of HtGH9t: HtCBM3c and saturated at 1:3 M ratio (Fig. 8B) at 12.5 U/mg. The truncated catalytic module, HtGH9t showed very low enzyme activity against lichenan, β-glucan and CMC (Table 2). However, it partially recovered activity against soluble substrates, lichenan (51%), β-glucan (58%) and CMC (30%) in the presence of equimolar ratio with HtCBM3c (Table 2). These results confirmed that the physical association of HtCBM3c with HtGH9t is essential for the efficient catalysis. The activity of HtGH9t was not recovered against insoluble avicel in presence of HtCBM3c. The enzyme activity of full length HtGH9 itself is very low (0.26 U/mg) against avicel. Moreover, the truncated derivative, HtGH9t does not show any activity against avicel. It indicated that physical association of HtCBM3c with HtGH9t is not sufficient for hydrolysis of insoluble substrates, they also require covalent bond for hydrolysis.

3.4. Melting curve of HtGH9 The stability of HtGH9 enzyme was also examined by generating the protein melting curve. HtGH9 displayed melting temperature of 93 °C (Fig. 5A). The melting temperature of HtGH9 was not affected by 10 mM Ca2+ (Fig. 5B). However, in the presence of 10 mM EDTA or EGTA, the melting temperature decreased to 88 °C (Fig. 5C and D) displaying a shift of 5 °C. This observation also indicated that HtGH9 contains inherent Ca2+ ion(s), which provide the thermal stability making it resistant to unfolding and denaturation at the higher temperature. The addition of EDTA or EGTA removes Ca2+ ion(s) from HtGH9 and imparts instability. This result corroborates with the HtGH9 activity data, where the Ca2+ ions do not affect the activity, but EDTA or EGTA decreased the activity of HtGH9 as described in previous section. The results of protein melting curve and enzyme activity elucidated that inherently present Ca2+ ion(s) play role in HtGH9 enzyme structural integrity and its function. 3.5. Catalytic mechanism of HtGH9

4. Conclusion

The hydrolysis of CMC by HtGH9 was monitored by estimating the reducing sugar till the reaction reached saturation (Fig. 6A). The results

A new member of β-1,4-glucanase belonging to family 9 GH (HtGH9) and its catalytic module (HtGH9t) and carbohydrate binding 8

Carbohydrate Research 484 (2019) 107782

K. Kumar, et al.

module (HtCBM3c) from Hungateiclostridium thermocellum were cloned and characterized. The full length HtGH9 showed maximum β-1,4glucanase activity at pH 6.5 and 90 °C. It showed stability in a wide range of pH (5–9) and temperature (10–70 °C) making it suitable for biomass saccharification for bioethanol production. HtGH9 may contain inherent Ca2+ ions which play a role in the enzyme catalysis and stability at high temperature. The TLC analyses of CMC showed the endoglucanase activity of HtGH9 and of avicel showed the presence of cellotetraose displaying the processive endoglucanase activity. HtGH9 displayed 64.3 U/mg against lichenan. The truncated derivative, HtGH9t alone showed very low activity (1.22 U/mg) against lichenan, while in the presence of equimolar ratio of HtCBM3c, it recovered (51%) enzyme activity up to 34 U/mg. The native-PAGE displayed the association of the two modules, HtGH9t and HtCBM3c. This study showed that physical association between catalytic HtGH9t module and HtCBM3c is essential for catalysis by processive endoglucanase, HtGH9.

[13] D.B. Wilson, M. Kostylev, Cellulase processivity, Biomass Conversion, Humana Press, Totowa, NJ, 2012, pp. 93–99. [14] G.T. Beckham, J. Ståhlberg, B.C. Knott, M.E. Himmel, M.F. Crowley, M. Sandgren, C.M. Payne, Towards a molecular-level theory of carbohydrate processivity in glycoside hydrolases, Curr. Opin. Biotechnol. 27 (2014) 96–106. [15] P. Tomme, A. Boraston, B. McLean, J. Kormos, A.L. Creagh, K. Sturch, D.G. Kilburn, Characterization and affinity applications of cellulose-binding domains1, J. Chromatogr. B Biomed. Sci. Appl. 715 (1998) 283–296. [16] O. Yaniv, E. Morag, I. Borovok, E.A. Bayer, R. Lamed, F. Frolow, L.J. Shimon, Structure of a family 3a carbohydrate-binding module from the cellulosomal scaffoldin CipA of Clostridium thermocellum with flanking linkers: implications for cellulosome structure, Acta Crystallogr. F Struct. Biol. Cryst. Commun. 69 (2013) 733–737. [17] S. Petkun, I.R. Grinberg, R. Lamed, S. Jindou, T. Burstein, O. Yaniv, F. Frolow, Reassembly and co-crystallization of a family 9 processive endoglucanase from its component parts: structural and functional significance of the intermodular linker, Peer J 3 (2015) 1126. [18] S. Jindou, Q. Xu, R. Kenig, M. Shulman, Y. Shoham, E.A. Bayer, R. Lamed, Novel architecture of family-9 glycoside hydrolases identified in cellulosomal enzymes of Acetivibrio cellulolyticus and Clostridium thermocellum, FEMS Microbiol. Lett. 254 (2006) 308–316. [19] L. Gal, C. Gaudin, A. Belaich, S. Pages, C. Tardif, J.P. Belaich, CelG from Clostridium cellulolyticum: a multidomain endoglucanase acting efficiently on crystalline cellulose, J. Bacteriol. 179 (1997) 6595–6601. [20] A.I. Chiriac, E.M. Cadena, T. Vidal, A.L. Torres, P. Diaz, F.J. Pastor, Engineering a family 9 processive endoglucanase from Paenibacillus barcinonensis displaying a novel architecture, Appl. Microbiol. Biotechnol. 86 (2010) 1125–1134. [21] D. Irwin, D.H. Shin, S. Zhang, B.K. Barr, J. Sakon, P.A. Karplus, D.B. Wilson, Roles of the catalytic domain and two cellulose binding domains of Thermomonospora fusca E4 in cellulose hydrolysis, J. Bacteriol. 180 (1998) 1709–1714. [22] R. Gilad, L. Rabinovich, S. Yaron, E.A. Bayer, R. Lamed, H.J. Gilbert, Y. Shoham, CelI, a noncellulosomal family 9 enzyme from Clostridium thermocellum, is a processive endoglucanase that degrades crystalline cellulose, J. Bacteriol. 185 (2003) 391–398. [23] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680. [24] A. Makino, T. Mae, K. Ohira, Colorimetric measurement of protein stained with Coomassie Brilliant Blue R on sodium dodecyl sulfate-polyacrylamide gel electrophoresis by eluting with formamide, Agric. Biol. Chem. 50 (1986) 1911–1912. [25] N. Nelson, A photometric adaptation of the Somogyi method for the determination of glucose, J. Biol. Chem. 153 (1944) 375–380. [26] M. Somogyi, Determination of blood sugar, J. Biol. Chem. 160 (1945) 69–73. [27] A. Karnchanatat, A. Petsom, P. Sangvanich, J. Piapukiew, A.J. Whalley, C.D. Reynolds, P. Sihanonth, A novel thermostable endoglucanase from the wooddecaying fungus Daldinia eschscholzii (Ehrenb.: Fr.) Rehm, Enzym. Microb. Technol. 42 (2008) 404–413. [28] V.V. Zverlov, G.A. Velikodvorskaya, W.H. Schwarz, Two new cellulosome components encoded downstream of celI in the genome of Clostridium thermocellum: the non-processive endoglucanase CelN and the possibly structural protein CseP, Microbiology 149 (2003) 515–524. [29] A.O. Lima, M.C. Quecine, M.H. Fungaro, F.D. Andreote, W. Maccheroni, W. Araújo, J.L. Azevedo, Molecular characterization of a β-1, 4-endoglucanase from an endophytic Bacillus pumilus strain, Appl. Microbiol. Biotechnol. 68 (2005) 57–65. [30] J.P. Lee, Y.A. Kim, S.K. Kim, H. Kim, Characterization of a multimodular endo-β-1, 4-Glucanase (Cel9K) from Paenibacillus sp. X4 with a potential additive for saccharification, J. Microbiol. Biotechnol. 28 (2018) 588–596. [31] Z. Yi, X. Su, V. Revindran, R.I. Mackie, I. Cann, Molecular and biochemical analyses of CbCel9A/Cel48A, a highly secreted multi-modular cellulase by Caldicellulosiruptor bescii during growth on crystalline cellulose, PLoS One 8 (2013) e84172. [32] I. ul Haq, F. Akram, M.A. Khan, Z. Hussain, A. Nawaz, K. Iqbal, A.J. Shah, CenC, a multidomain thermostable GH9 processive endoglucanase from Clostridium thermocellum: cloning, characterization and saccharification studies, World J. Microbiol. Biotechnol. 31 (2015) 1699–1710. [33] F.I.J. Pastor, X. Pujol, A. Blanco, T. Vidal, A.L. Torres, P. Diaz, Molecular cloning and characterization of a multidomain endoglucanase from Paenibacillus sp BP-23: evaluation of its performance in pulp refining, Appl. Microbiol. Biotechnol. 55 (2001) 61–68. [34] E.A. de Araújo, M. de Oliveira Neto, I. Polikarpov, Biochemical characterization and low-resolution SAXS structure of two-domain endoglucanase BlCel9 from Bacillus licheniformis, Appl. Microbiol. Biotechnol. 103 (2019) 1275–1287. [35] B.L. Zamost, H.K. Nielsen, R.L. Starnes, Thermostable enzymes for industrial applications, J. Ind. Microbiol. Biotechnol. 8 (1991) 71–81.

Conflicts of interest The authors declare no conflict of interest. Acknowledgements The research work was supported by DBT-Twining project grant number, BT/PR24786/NER/95/853/2017 to AG and in part supported by funding from Indian Institute of Technology Guwahati, Guwahati, Assam, India. References [1] Lee R. Lynd, P.J. Weimer, W.H. Van Zyl, I.S. Pretorius, Microbial cellulose utilization: fundamentals and biotechnology, Microbiol. Mol. Biol. Rev. 66 (2002) 506–577. [2] R.H. Doi, A. Kosugi, Cellulosomes: plant-cell-wall-degrading enzyme complexes, Nat. Rev. Microbiol. 2 (2004) 541. [3] A.L. Demain, M. Newcomb, J.D. Wu, Cellulase, clostridia, and ethanol, Microbiol. Mol. Biol. Rev. 69 (2005) 124–154. [4] E.A. Bayer, E. Morag, R. Lamed, S. Yaron, Y. Shoham, Cellulosome structure: fourpronged attack using biochemistry, molecular biology, crystallography and bioinformatics, R. Soc. Chem. 219 (1998) 39–65. [5] C.M. Fontes, H.J. Gilbert, Cellulosomes: highly efficient nanomachines designed to deconstruct plant cell wall complex carbohydrates, Annu. Rev. Biochem. 79 (2010) 655–681. [6] K. Kumar, M. Correia, V.R. Pires, A. Dhillon, K. Sharma, V. Rajulapati, A. Goyal, Novel insights into the degradation of β-1, 3-glucans by the cellulosome of Clostridium thermocellum revealed by structure and function studies of a family 81 glycoside hydrolase, Int. J. Biol. Macromol. 117 (2018) 890–901. [7] D.B. Wilson, D.C. Irwin, Genetics and properties of cellulases, Recent Progress in Bioconversion of Lignocellulosics, Springer, Berlin, Heidelberg, 1999, pp. 1–2. [8] B. Henrissat, G.J. Davies, Glycoside hydrolases and glycosyltransferases. Families, modules, and implications for genomics, Plant Physiol. 124 (2000) 1515–1519. [9] B.L. Cantarel, P.M. Coutinho, C. Rancurel, T. Bernard, V. Lombard, B. Henrissat, The Carbohydrate-Active EnZymes database (CAZy): an expert resource for glycogenomics, Nucleic Acids Res. 37 (2008) D233–D238. [10] G. Davies, B. Henrissat, Structures and mechanisms of glycosyl hydrolases, Structure 3 (1995) 853–859. [11] T. Burstein, M. Shulman, S. Jindou, S. Petkun, F. Frolow, Y. Shoham, R. Lamed, Physical association of the catalytic and helper modules of a family‐9 glycoside hydrolase is essential for activity, FEBS (Fed. Eur. Biochem. Soc.) Lett. 583 (2009) 879–884. [12] B. Leis, C. Held, F. Bergkemper, K. Dennemarck, R. Steinbauer, A. Reiter, W.H. Schwarz, Comparative characterization of all cellulosomal cellulases from Clostridium thermocellum reveals high diversity in endoglucanase product formation essential for complex activity, Biotechnol. Biofuels 10 (2017) 240.

9