A Thermostable Multifunctional GH5 Endoglucanase From Bacillus sp. for Lignocellulosic Biomass Saccharification

ABSTRACT

The enzymatic conversion of lignocellulosic biomass to glucose remains a bottleneck for cost‑effective biofuel production, largely due to the complexity and expense of enzymes. Here we describe BsEG2, a multifunctional GH5‑like endoglucanase cloned from Bacillus sp. strain BS and heterologously expressed. BsEG2 displays optimal activity at pH 6.0 and 55°C and efficiently hydrolyzes both amorphous and crystalline cellulose. Kinetic analysis on carboxymethyl cellulose (CMC) indicates high substrate affinity and catalytic efficiency. Unlike typical endoglucanases, BsEG2 combines processive endoglucanase activity with broad substrate specificity. It degrades long cellulose chains and oligosaccharides and also shows β‑glucosidase‑like activity on cellobiose, yielding glucose across a wide range of chain lengths. BsEG2 is highly thermostable (a half‐life of ~15 days at 55°C), tolerant of high salt and ionic liquids, and notably resistant to cellobiose inhibition up to 200 mM. It remains active in seawater, suggesting potential to reduce freshwater demand in biomass processing. These properties, namely, multifunctionality, stability, inhibitor tolerance, and halotolerance, make BsEG2 a strong candidate for simplified, cost‑effective enzyme formulations for glucose production from lignocellulosic feedstocks.

Graphical Abstract and Lay Summary

A multifunctional cellulase, BsEG2, converts biomass (left) into the cello‐oligosaccharides, cellobiose, and glucose (right). The enzyme is catalytically competent in seawater, is tolerant to ionic liquids and cellobiose, shows a half‐life of more than 15 days at ∼55°C, and produces cellobiose and glucose.

1. Introduction

Lignocellulosic biomass is an abundant feedstock for biofuels, biomaterials, and chemicals; however, efficient cellulose‐to‐glucose conversion requires a synergistic action of a cellulase cocktail [1]. Cellulose depolymerization typically involves three enzyme types: endoglucanases (EC 3.2.1.4), which randomly cleave β‐1,4 glycosidic bonds to produce shorter oligosaccharides and expose new chain ends; cellobiohydrolases (EC 3.2.1.91), which act on the newly exposed ends to release cellobiose; and β‐glucosidases, which hydrolyze cellobiose to glucose. Producing such cocktails is complex and costly, accounting for nearly 40% of total bioethanol production costs [2, 3]. Therefore, the development of robust, cost‐effective cellulases with high specific activity, tolerance to hydrolysis products, metal ions, salts, and ionic liquids, and long half‐lives is critical.

An attractive alternative is multifunctional, processive cellulases that combine several activities in a single enzyme. The efficient conversion of lignocellulosic biomass is heavily dependent on multienzyme cocktails, which often suffer from high production costs, instability, and suboptimal synergy among individual components. Additionally, the need to balance multiple enzymes' activity at different optimal conditions further complicates their industrial application. Multifunctional endoglucanases that can act across different stages of cellulose depolymerization offer an alternative by reducing enzyme complexity and improving overall efficiency. For example, bifunctional cellulases may combine the activities of endoglucanases and exoglucanases, while trifunctional cellulases may also include β‐glucosidase activity. Several bifunctional cellulases have been identified, such as CelB from Caldocellum saccharolyticum, CelA from Anaerocellum thermophilum, and CtCel5E, a bifunctional cellulase/xylanase from Clostridium thermocellum [4, 5, 6, 7]. However, to date, no naturally occurring trifunctional cellulase that yields glucose as a primary product has been reported. Processivity, the enzyme's ability to continuously thread a cellulose chain through its active site to perform successive cleavages, enhances hydrolytic efficiency by generating oligosaccharides that are amenable to further conversion into glucose [8].

Bacterial cellulases, especially those from Bacillus species, are attractive because these organisms tolerate diverse environmental conditions suitable for harsh industrial conditions. They possess efficient secretion systems that enable direct extracellular enzyme recovery, reducing downstream processing costs. Additionally, their enzymes often exhibit broad substrate specificity, making them strong candidates for multifunctional biomass hydrolysis [8]. Bacillus subtilis is particularly noteworthy for producing a variety of industrially relevant enzymes, including proteases, amidases, and cellulases, making it a promising candidate for pharmaceutical applications [9]. Certain strains of B. subtilis, such as KB‐1111 and KB‐1122, have been investigated as biocontrol agents and are known to secrete endo‐1,4‐β‐glucanase [10].

This study identifies and biochemically characterizes a novel GH5 endoglucanase, BsEG2, for biomass saccharification. The gene encoding BsEG2 was cloned into pET21b+ as two constructs, BsEG2, containing the signal peptide and catalytic domain, and BsEG2+CBM, containing the signal peptide, catalytic domain, and a short carbohydrate‐binding module. Both constructs were heterologously expressed in E. coli and purified. BsEG2 exhibits multifunctional and processive behavior that challenges its classification as a conventional endoglucanase. Molecular dynamics simulations of cellooligosaccharide (G6) binding to BsEG2, with and without the CBM, reveal altered substrate orientation and interaction energetics, suggesting that CBM influences substrate positioning at the catalytic site.

2. Results and Discussion

2.1. Identification of a Novel Endoglucanase‐Like Gene

PCR and Sanger sequencing of Bacillus sp. strain BS yielded a full‐length 1263 bp gene encoding the protein designated BsEG2+CBM. The gene encodes a protein that contains a signal peptide, a catalytic domain, and a carbohydrate‐binding module (CBM). Pairwise comparison to the endoglucanase Q6YK34 showed 88.7% nucleotide identity and 89.8% amino‐acid identity, indicating a putative novel endoglucanase‐like gene hereafter referred to as BsEG2.

2.2. Cloning, Expression, and Purification of BsEG2

To dissect the contribution of the CBM, two coding sequences, the catalytic domain alone (BsEG2) and the catalytic domain plus CBM (BsEG2+CBM), were cloned into pET‐21b+ under a T7 promoter with an N‐terminal His6 tag, to enable high‐level T7 RNA polymerase‐driven expression of the proteins in E. coli BL21(DE3). The constructs were verified by Sanger sequencing (primers in Table S1). Both constructs were overexpressed in E. coli BL21(DE3) and recovered predominantly in the soluble fraction following IPTG induction and purified from soluble extracts by HisTrap affinity chromatography. SDS–PAGE (12%) showed each purified protein as a single predominant band (BsEG2, Figure S1a; BsEG2+CBM, Figure S1b). The purity and integrity of the heterologously expressed proteins were sufficient for all downstream biochemical and processivity analyses.

Domain and family assignments were performed using InterPro [11] and PANTHER [12], which placed the enzyme firmly within the GH5 family. A conserved motif (VIYEIANEPN) maps to Glycoside Hydrolase family 5 (PTHR34142), InterPro annotations identify endoglucanase features and a CBM3a domain (Table S2), and multiple‐sequence alignment with GH5 members revealed conserved catalytic and structural sequence stretches (Figures S2 and S3). The amino acid sequence of the signal peptide in BsEG2 is MSNMKRSNSIFITCVLIAVLTIGSLLGSTASA as identified by SignalP 6.0 [13].

2.3. Optimum Temperature, pH, and Specific Activity of BsEG2

Having confirmed the identity of both constructs, we next characterized their pH and temperature dependence on the soluble substrate carboxymethyl cellulose (CMC). BsEG2+CBM exhibited the highest activity at 60°C within the pH range of 5.0–6.0 (Figures S4 and S5), while BsEG2 displayed an optimal temperature (T _opt) of 55°C, retaining over 90% activity between 50°C–65°C (Figure 1a). At its T _opt, BsEG2 retained more than 80% of its specific activity over the pH range of 5.0–7.5, with peak activity at pH 6.0–6.5 in McIlvaine buffer (Figure 1b). These optimal conditions are consistent with previously reported endoglucanases from Bacillus sp. [14, 15, 16, 17].

FIGURE 1.

Determination of temperature and pH optima of BsEG2: (a) The effect of temperature on BsEG2 activity was measured between 40°C and 80°C using 1% CMC as the substrate in McIlvaine buffer, pH 6.0. Specific activity was measured using the DNS assay and reported as % specific activity. (b) The effect of pH on BsEG2 activity was determined by measuring specific activity in McIlvaine buffer across a pH range of 4–7 at the T_opt (55°C) on 1% CMC. 100% specific activity denotes the BsEG2 specific activity (340.57 ± 10.61 µmoles min⁻¹ mg⁻¹) recorded at the optimum temperature and pH of BsEG2. Error bars represent standard deviations from independent reactions.

Under these optimal conditions (pH_opt 6.0 and T _opt 55°C), BsEG2 exhibited a specific activity of 340.6 ± 10.6 U mg⁻¹ (where 1 U = 1 µmol of reducing sugars generated per minute of reaction per mg of enzyme), comparable to the 327 ± 24.8 U mg⁻¹ measured for BsEG2+CBM under its respective optimal conditions, indicating that the short CBM did not significantly alter activity on this soluble substrate. This pattern held on insoluble substrate as well. When activity was tested on both untreated and pre‐treated sugarcane bagasse, no significant difference was observed after CBM removal (Figure 2), in contrast to previous studies that highlight the importance of CBMs in binding insoluble substrates [18]. While these results suggest that the CBM does not enhance substrate catalysis in BsEG2, further investigations are required to assess its potential role in thermostability or other functional properties. Enzymes may exhibit high optimal temperatures but shorter half‐lives, or vice versa, as observed in examples such as I7DLX2 from Caldicellulosiruptor sp. F32, P50388 from Saccharolobus shibatae, P22498 from Saccharolobus solfataricus, and D9PZ08 from Acidilobus saccharovorans [19, 20, 21, 22]. In this report, however, most experiments were carried out only with the catalytic domain (BsEG2).

FIGURE 2.

Hydrolytic activity of BsEG2 and BsEG2 + CBM on various biomass substrates, including SCB, sugarcane bagasse; HFB, hemicellulose‐free biomass; MCC, microcrystalline cellulose; LF, lignin‐free biomass; AC, and α‐cellulose. The hydrolysis reactions were performed for 2 h using 5 µg enzyme under optimal temperature and pH. The resulting hydrolysis products were quantified using the DNS assay.

2.4. Substrate Specificity and Kinetics of BsEG2

The kinetic parameters of BsEG2 on CMC were determined under optimal assay conditions (pH 6.0, 55°C) in McIlvaine buffer. Data were analyzed using the non‐linear regression fit in GraphPad PRISM (version 8.0). Complete substrate saturation could not be achieved within the usable concentration limit of 24 mg mL⁻¹, necessitating the extrapolation of the curve. Similar cases of unsaturated Michaelis‐Menten kinetics have also been reported for other enzymes, such as RsEG_m and EGL1 [23, 24]. On CMC, BsEG2 has an apparent V _max of 148.9 µM min⁻¹, an apparent K _m of 24.9 mg mL⁻¹, and a k _cat of 517 s⁻¹ (Figure S6). Beyond CMC, substrate specificity was assessed using both natural and commercially available substrates (Table S3). The enzyme displays higher activity on β‐glucan than on CMC, a trend observed in many endoglucanases [25, 26, 27]. This preference is likely due to β‐glucan's greater solubility in water and its mixed β‐1,3 and β‐1,4 glycosidic linkages [28]. BsEG2 also hydrolyzes insoluble crystalline substrates, such as Avicel PH‐101, filter paper, lichenan, and natural biomass, albeit at lower rates, likely due to reduced accessibility and higher crystallinity [29]. In addition, BsEG2 hydrolyzes chromogenic substrates, including azo‐CMC, pNPClb, and pNPLac (Table S3).

Among the natural cellulosic substrates tested were both pre‐treated and untreated sugarcane bagasse, a fibrous residue after sugarcane stalks are crushed to extract juice, which is an abundant biomass in many parts of the world, including India. Sugarcane bagasse (SCB) obtained from a local sugar industry (M/s Gayatri Sugars Ltd., Hyderabad, India) was cleaned to remove impurities, sun‐dried, oven‐dried, and ground to a uniform particle size. Different pretreatment approaches were applied to selectively isolate biomass components. Hemicellulose was removed by acid treatment (5% H₂SO₄, 1:10 solid–liquid ratio, autoclaving), yielding hemicellulose‐free bagasse (HFB) [30]. Lignin removal was achieved using an alkaline peroxide method (NaOH and H₂O₂ under controlled pH), producing lignin‐free bagasse (LFB). For α‐cellulose preparation, HFB underwent further alkali treatment followed by bleaching with acetic acid and sodium hypochlorite. Microcrystalline cellulose (MCC) was then generated by oxidative depolymerization of α‐cellulose using hydrogen peroxide. All pretreated fractions were washed to neutrality, dried, and subsequently used for enzymatic saccharification. Among various substrates, BsEG2 exhibits the highest activity on microcrystalline cellulose (MCC), followed by lignin‐free bagasse (LF). The BsEG2 activity on real biomass remains similar with or without CBM, as BsEG2 and BsEG2+CBM show comparable activity under identical reaction conditions (Figure 2).

To visualize this hydrolytic action directly, the effect of BsEG2 on filter paper and sugarcane bagasse was further examined using scanning electron microscopy (Figure S7). The SEM image shows that the untreated filter paper (control) has a smooth, uniform, and dense mesh of cellulose fibers. Upon BsEG2 treatment, this network became undulated, more open and non‐uniform, and thinner, with interstitial fibers becoming more susceptible to enzymatic digestion. These results suggest that BsEG2 can act on diverse cellulosic substrates and may have the potential to convert agricultural residues, such as sugarcane bagasse, into value‐added products.

A comparative analysis of the kinetic parameters of BsEG2 was performed not only against other GH5 endoglucanases (Table S4), but also against endoglucanases reported from other Bacillus species (Table S5). These comparisons enable direct evaluation of substrate affinity, catalytic turnover, and overall hydrolytic efficiency of BsEG2 relative to previously characterized endoglucanases. Collectively, these parameters provide a comprehensive view of the enzyme's substrate binding and catalytic performance, thereby positioning BsEG2 within the broader functional landscape of related endoglucanases.

2.5. Molecular Docking and MD Simulation to Understand the Role of CBM

Because the CBM had little effect on the catalytic activity, we turned to molecular docking and MD simulations to examine how it influences substrate binding, using cellohexaose (G6), as a representative long cellulosic chain bound to BsEG2 in the presence and absence of CBM. We analyzed the conformational changes of cellohexaose (G6) based on the length and radius of gyration of the cellohexaose chain, where the distance between the terminal sugar subunits was considered as the length of the G6 chain. Simulated average structure along with the distribution of length and radius of gyration (Figure S8a,b) of the G6 chain shows that G6 is aligned linearly at the binding site cleft of BsEG2 in the presence of CBM without significant distortion. Besides, the average length and radius of gyration of the G6 chain (Figure S8a,b) are reduced when bound with the catalytic domain of BsEG2 in the absence of CBM. Thus, the G6 chain is aligned deep inside the binding site cleft, adopting a distorted, curved conformation in the catalytic domain of BsEG2 in the absence of the CBM. Along with the altered conformation of G6, the average volume of the catalytic pocket of BsEG2 with G6 in the binding site cleft is found comparatively higher in the presence of CBM (1856 ± 109 Å2) than in the absence of CBM (1713.6 ± 93 Å3) (Figure S8c,d). The average volume of the catalytic pocket of BsEG2 was assessed over the equilibrated simulation trajectories, considering E169 as the center of the catalytic site with the POVME (pocket volume measurer) algorithm [31]. The AlphaFold‐predicted model structure of BsEG2 reveals a structured catalytic domain and a short CBM (Figure S9a). The docked structures of BsEG2 with G6 exhibited the highest docking free energy scores of −7.97 kcal/mol with CBM (Figure 3a) and −6.74 kcal/mol without CBM (Figure 3b). By analyzing the distances between receptor‐ligand interface residue pairs that are below 5.0 Å, we identified the key BsEG2 residues involved in forming bonded and non‐bonded contacts with G6 (Table S6). Notably, the binding site residues remain mostly similar in both scenarios, though the orientation of G6 differs at the binding site pocket depending on the presence of CBM.

FIGURE 3.

(a,b) Initial and (c,d) simulated average G6 docked structures of BsEG2 with and without CBM. The representative G6 binding site residues and G6 are shown as stick representations and marked with cyan and yellow colors, respectively. (e,f) The overlay structures of G6‐bound BsEG2 with and without CBM along the equilibrated MD trajectories, where the conformation of protein and ligand at different simulation times are marked with RWB and RGB color scale.

Furthermore, G6 stabilizes the binding site pocket of BsEG2 in different orientations, relative to the initial docked conformation (Figure S9b,c). The convergence of the root‐mean‐square displacement (RMSD) of the G6‐bound BsEG2 structures (Figure S10a) suggests that G6 can form stable complexes with BsEG2 both in the presence and absence of CBM. Notably, G6 is positioned deeper within the catalytic domain's binding pocket of BsEG2 when CBM is absent. The distance between the mass center of G6 and the catalytic domain of BsEG2 (d _{G6 − catalytic} ) increases in the presence of CBM, as illustrated in Figure S10b. Figure 3c,d shows the simulated average docked structures of G6 along with representative binding site residues of BsEG2 with and without CBM. The structural overlay of G6‐bound BsEG2 along the equilibrated MD trajectories (Figure 3e,f) reveals distinct preferred binding conformations of G6 within the binding pocket, along with the flexible movements of G6 when CBM is present (Figure 3e). In contrast, G6 shows more restricted movements in the absence of CBM (Figure S10c). On average, five favorable polar hydrogen‐bonded contacts can form between G6 and BsEG2 (Figure S10d). Key residues forming stable polar contacts with the oxygen atoms of G6 in the binding pocket of BsEG2 without CBM include (Table S6) Y96, D99, N134, W207, Q209, Y231, and E257. In contrast, in the presence of CBM, residues L133, E169, G233, T234, and D262 from the catalytic domain establish favorable polar contacts with G6. Most residues within the binding pocket of BsEG2 that participate in G6 binding (Table S6) are highly conserved (Figure S11). Based on our molecular modelling and simulation studies, as well as comparison with the crystal structure of BsCel5A, we speculate that E169 is present at the center of the catalytic site nearby the −1 subsite [32]. It was proposed from the functional and structural characterization of BsCel5A that the catalytic machinery involves two acid residues, E257 and E169. Structural superposition of the catalytic domain of modelled BsEG2 with the BsCel5A structure (PDB id: 3PZV), as shown in Figure S12a, represents the unique location of E257 and E169 in the equivalent position at the catalytic pocket. We found that E257 and E169 residues of BsEG2 are highly evolutionarily conserved and exposed on the catalytic site, where E169 is found to form close contacts with the sugar moiety of the G6 chain at the −1 subsite (Figure S12b). Concurrently, W207, Q209, F230, A263, and W291 are found as neighboring residues at +1 subsite.

Both electrostatic (E _elec) and van der Waals (E _vdw) interactions favor G6 binding with BsEG2, where the E _vdw is more favorable for BsEG2 without CBM (Table S6). We further evaluated the binding free energy of G6 to BsEG2 and the contributions of individual binding site residues using the MMGBSA approach [33]. The negative total binding free energy (ΔG), as indicated in Table S6, suggests a favorable interaction between G6 and BsEG2, regardless of the presence of the CBM. However, the estimated ΔG values do not accurately represent the actual binding free energy due to the omission of entropic factors related to the release of solvent molecules. Notably, residues such as L133, W207, Q209, H229, F230, Y231, T234, E257, and W291 are highly conserved (Figure S11). Overall, our in silico results characterize the binding site and suggest that G6 binding to BsEG2 and BsEG2+CBM is favorable. Further studies are needed to fully understand the finer implications of the role of the truncated CBM.

2.6. Stability and Tolerance of BsEG2

Beyond catalytic performance, an enzyme's value for biomass processing depends heavily on its robustness under operational conditions, beginning with ionic liquids (ILs), which are excellent non‐aqueous solvents for lignocellulosic biomass pretreatment [34]. However, residual ILs can inhibit cellulase activity, necessitating extensive washing, which increases water consumption and costs [35, 36]. IL‐tolerant enzymes are therefore desirable for compatibility with IL‐pretreated biomass and for probing determinants of enzyme stability. BsEG2 was evaluated for its tolerance to imidazolium‐based ILs, specifically 1‐ethyl‐3‐methylimidazolium [C₂mim] variants (Figure 4a). It maintained full activity in up to 20% [C₂mim][Cl], 15% [C₂mim][OAc], and 10% [C₂mim][Lac], though higher concentrations of [C₂mim][OAc] and [C₂mim][Lac] resulted in 20% and 40% activity loss, respectively. This suggests that anion identity plays a critical role in enzyme stability. Tolerance to ionic liquids (ILs) has also been reported for certain cellulases. NMgh45 retained 92.8% and 43.8% residual activity in the presence of 10% and 20% (v/v) 1‐butyl‐3‐methylimidazolium chloride, respectively [37]. C5‐CEL4 retained over 90% activity in ionic liquids such as BMIM‐Ac, EMIM‐Cl, and BMIM‐BF4 even at concentrations up to 50% [38]. A cellulase system from Stachybotrys microspora showed enhanced activity (115.5% and 114.5%) in the presence of 5% (v/v) [EMIM][DEP] and [AMIM][Cl], and retained 84.4% and 76.7% activity at 30% concentrations, while showing reduced activity in [EMIM][MtSO4] [39]. The effectiveness of ILs for pretreatment followed the trend [EMIM][DEP] > [AMIM][Cl] > [EMIM][MtSO4], with a maximum saccharification yield of 52.4% obtained using 15% IL‐pretreated substrate. ILs typically inhibit cellulase activity via denaturation, aggregation, competitive inhibition, or modification in enzyme dynamics [36, 40]. The surface charge of enzymes has also been proposed to play a role in maintaining the integrity and specific activity in ionic liquids [41, 42]. Notably, BsEG2 contains 24% (92 out of 383) charged residues, among the highest reported for endoglucanases, which may contribute to IL tolerance.

FIGURE 4.

(a) Effect of the [C₂mim] cation‐based ionic liquid (IL) on BsEG2 activity. Specific activity was measured in the presence of 5%–20% IL using 1% β‐glucan as the substrate in McIlvaine buffer, pH 6.0. Residual activity was determined using the DNS assay and expressed as a percentage of the enzyme's activity in the absence of ionic liquid (set at 100%). (b) Salt tolerance of BsEG2. The enzyme was incubated in McIlvaine buffer at pH_opt and 55°C with 0.2–2 M NaCl, and then assayed under standard conditions. The specific activity of BsEG2 in the absence of NaCl was set at 100% specific activity (340.57 ± 10.61 U/mg). Error bars represent the standard deviation from three independent experiments. (c) Comparison of reducing sugar concentrations produced by BsEG2 and BsEG2+CBM in McIlvaine buffer (at pH 6), seawater (pH adjusted to 6), and seawater (pH unadjusted) at 60°C. The reducing sugars generated in each reaction were quantified using the DNS assay.

A relatable form of robustness is salt tolerance, where BsEG2 retained over 90% activity up to 1.5 M NaCl and around 70% activity at 2 M NaCl, though the salt tolerance is not as high as in some previously reported EGs [43] (Figure 4b). Some other reported endoglucanases that exhibit tolerance to high salt concentrations include NMgh45, which retained over 88% relative activity in 4 M NaCl and remained stable for 24 h [37]. Similarly, CelM has been reported as a thermo‐halo‐tolerant GH5 endoglucanase [44]. In another case, a GH5 cellulase from Aspergillus glaucus CCHA showed activity across a wide NaCl range (1.0–4.0 M) [45]. The enzyme C5‐CEL4 maintained 100% activity between 0.5 and 5.0 M NaCl and remained stable even after prolonged exposure (up to 9 months) in 5.0 M NaCl [38]. Since pre‐treated biomass often contains residual salts from acid or base neutralization, such tolerance can reduce the need for extensive washing, lowering overall process costs. Salt‐tolerant cellulases may also allow the substitution of freshwater with non‐potable sources, such as seawater, in bioprocessing operations [46]. BsEG2 functions effectively in seawater at its natural pH (7.8–8.1) and pH 6.0, indicating its suitability for marine or seaweed biomass degradation without pretreatment (Figure 4c).

Metal ion tolerance further supports BsEG2's industrial potential (Figure S13). BsEG2 was assayed in the presence of divalent (Ca^{2 +}, Cu^{2 +}, Mg^{2 +}, Ni^{2 +}, Zn^{2 +}) and trivalent (Fe^{3 +}) metal chloride salts as well as monovalent cations (K⁺, Na⁺) at 10 mM concentrations. At 10 mM, divalent and trivalent cations slightly enhanced activity by up to 25%, whereas monovalent cations (K⁺, Na⁺) had no significant effect. The stimulatory effect of divalent cations such as Ca^{2 +} and Mg^{2 +} is consistent with their known roles in stabilizing enzyme–substrate interactions or active‐site geometry in GH5 endoglucanases [47]. The tolerance to Cu^{2 +}, Zn^{2 +}, and Fe^{3 +}, which are common contaminants in process water and lignocellulosic feedstocks, further strengthens BsEG2's suitability for industrial biorefinery contexts. Ca^{2 +} and Mg^{2 +} typically originate from plant biomass ash and process water, whereas Na⁺ and K⁺ are introduced through biomass constituents and alkaline pretreatment chemicals and can reach millimolar to tens‐of‐millimolar levels, even though it should be noted that these ranges of concentration are highly process‐dependent [48]. Therefore, the observed tolerance of BsEG2 to these ions suggests that the enzyme is likely to remain active under the metal ion concentrations typically encountered during industrial biomass saccharification.

BsEG2 thermal stability is outstanding. After 12 days at 55°C, BsEG2 retained 59 ± 2% activity, corresponding to an approximate half‐life of 15 days (Figure 5), and retained 97% ± 3% activity after two months at 4°C. The BsEG2 melting temperature (T _m) is 65.5°C, classifying it as a thermophilic enzyme. Enzymes with high T _m often favor prolonged high‐temperature saccharification and reduced contamination risk, as they resist denaturants better than mesophilic enzymes [49, 50, 51]. In summary, BsEG2 combines IL and salt tolerance, long half‐life, tolerance to metal ions, and thermostability, making it an excellent candidate for biorefinery applications.

FIGURE 5.

Thermostability of BsEG2. The enzyme was incubated in buffer at pH_opt (pH 6.0) and 55⁰C for up to 16 days. At designated time points, aliquots were taken and assayed under standard kinetic assay conditions. The specific activity of BsEG2 at 0 h of incubation (340.57 ± 10.61 U/mg) was set as 100%. Error bars represent the standard deviation from independent reactions.

2.7. Processivity in BsEG2

A further distinctive feature of BsEG2 emerged from its product distribution as hydrolysis yielded 2.6‐fold more reducing sugars in the soluble fraction than in the insoluble fraction (Figure 6), indicating a processive mode of action. This behavior contrasts with typical endoglucanases, which often generate more reducing ends in the insoluble particles than in the soluble fraction [52]. Because processivity is central to efficient cellulose hydrolysis and is quantified by the ratio of soluble to insoluble reducing sugars a higher ratio reflecting greater soluble sugar production, this behavior is notable. While most processive endoglucanases belong to the GH9 family [53], several GH5 members are also processive (e.g., cel5H from Saccharophagus degradans, p4818Cel5_2A from porcine gut microbiota, Cel5A from Gloeophyllum trabeum, CHU_2103 from Cytophaga hutchinsonii, and EG5C‐1 from B. subtilis BS‐5 [14, 27, 54, 55, 56]. Table S7 summarizes the processivity across the GH5 family.

FIGURE 6.

The ratio of soluble to insoluble reducing sugars generated from 30 mg of filter paper by 15 µg of enzyme in 3 h at the enzyme's optimal temperature and pH was determined using the DNS assay. The soluble‐to‐insoluble sugar ratio of BsEG2 was compared to that of another processive endoglucanase, H1AD14, which loses most of its activity in the absence of its CBM. Reactions were performed in triplicate. Error bars represent the standard deviation from independent reactions.

For direct comparison, Figure 6 contrasts the soluble and insoluble fractions of reducing sugars produced by BsEG2 with those generated by the GH9 endoglucanase H1AD14‐CBM (only catalytic domain). Unlike BsEG2, the removal of the CBM in H1AD14 lowered its ratio of soluble to insoluble reducing sugars, whereas the presence of CBM3 enhanced its processivity [54, 57]. The processivity in GH5 endoglucanases does not always depend on the presence of CBMs. For example, deleting the N‐terminal CBM1 in the endoglucanase from Volvariella volvacea affected its processivity, whereas the deletion of CBM6 in GH5 endoglucanases (SdCel5G, SdCel5H, and SdCel5J) from Saccharophagus degradans had no impact [56, 58]. Additionally, endoglucanases from Gloeophyllum trabeum and Cytophaga hutchinsonii have been reported to be processive even without CBMs [55, 56]. CBM3, which is natively present in many GH9 endoglucanases, has been shown to enhance processivity [59].

2.8. Trifunctionality of BsEG2

The processive behavior described above prompted a closer look at the products of BsEG2 across the range of substrates, including oligosaccharides. As noted earlier (vide infra), BsEG2 hydrolyzed chromogenic trisaccharide analogues pNPClb and pNPLac to yield cellobiose, lactose, and pNP as hydrolysis products. To resolve its action on oligosaccharides, we monitored hydrolysis of cellooligosaccharides by HPLC after 10 and 30 min incubations (Figures 7a and S14). The hydrolysis products of cellopentaose (G5) resulted in the formation of cellotetraose (G4), cellotriose (G3), and cellobiose (G2). After 10 min, residual cellotetraose was present, but it became negligible after 30 min, coinciding with an increased accumulation of cellobiose. Similarly, the cellotetraose hydrolysis after 10 min yielded cellotriose, cellobiose, and some unreacted cellotetraose. After 30 min, cellotetraose was nearly depleted, and the accumulation of cellobiose had doubled. A comparable trend was observed in the hydrolysis of cellotriose (Figure S14). The higher accumulation of cellobiose compared to cellotriose at both time points suggests that BsEG2 preferentially produces cellobiose.

FIGURE 7.

(a) HPLC analysis of BsEG2 reaction products after 10 min of incubation with oligosaccharides. The assay was performed as described in the Methods section. The HPLC chromatogram displays the hydrolysis products of cellopentaose (G5), cellotetraose (G4), cellotriose (G3), and cellobiose (G2). (b). Quantification of glucose production from various oligosaccharides (G2, cellobiose; G3, cellotriose; G4, cellotetraose; G5, cellopentaose; G6, cellohexaose) over with time. Within 4 h, nearly 100% of the oligosaccharides were converted to glucose by 10 µM BsEG2 at the optimal temperature and pH, as determined by the GOD‐POD assay. Error bars represent the standard deviation from independent reactions.

Strikingly, this preference was only transient. When the reaction time was extended to four hours, all the oligosaccharide substrates, including cellobiose, cellotriose, cellotetraose, cellopentaose, and cellohexaose, were completely converted to glucose (Figure 7b). The conversion was quantified by GOD POD assay, and TLC analysis confirmed that glucose was the sole product, as indicated by a single spot in each reaction sample (Figure S15). Furthermore, when BsEG2 was incubated with cellobiose for 20 min, HPLC analysis confirmed its hydrolysis to glucose (Figures 7a and S16). Hydrolysis of microcrystalline cellulose showed a substrate‐loading‐dependent increase in glucose release, although the response tended toward saturation at higher substrate concentrations, consistent with limited enzyme accessibility in this heterogeneous system. Under enzyme‐excess conditions, hydrolysis of 8% (w/v) microcrystalline cellulose (16 mg) yielded approximately 0.5 mM glucose. Notably, while some enzymes exhibit higher processivity than BsEG2, they primarily generate cellobiose and cellotriose as end products, with only trace amounts of glucose being produced. Previous studies have detected minimal glucose formation by GH5 endoglucanases, such as in AHK119‐bMs and AHK119‐E5 from Thermobifida alba AHK119, RfGH5_4 from Ruminococcus flavefaciens, and celM from a thermal spring metagenome [44, 60, 61]. In contrast, BsEG2 exhibits a significantly higher glucose yield, suggesting a distinct hydrolytic mechanism that warrants further investigation.

This multifunctionality also has practical appeal, since multifunctional cellulases can reduce enzyme costs by simplifying enzyme cocktail formulations and improving hydrolysis efficiency [62]. Although BsEG2 shares some sequence similarities with other GH5 endoglucanases, an alignment revealed only scattered stretches of conservation and limited overall sequence similarity, underscoring a distinct divergence and unique functional profile (Figures S1 and S2). Earlier attempts to develop single multidomain cellulases involved fusing two distinct cellulases via small linker peptides [63, 64]. However, such engineered multidomain fusions often resulted in misfolding and reduced expression or yield. In contrast, wild‐type BsEG2 naturally combines endoglucanase, cellobiohydrolase, and β‐glucosidase activities, enabling the direct conversion of cellulose to glucose (albeit at a slower rate than dedicated β‐glucosidases) and offering potential for simplifying enzyme cocktails.

To understand the structural basis of its glucose production, we compared BsEG2 with a putative endoglucanase from Bacillus velezensis (B. velezensis) (unpublished data) that shares roughly 92% structural and sequence similarity yet cannot produce glucose. Analysis of the solvent accessible surface areas (SASA) and binding pocket volume of both enzymes using Discovery Studio revealed that their SASA values are nearly identical: 427.31 Ų for the endoglucanase from B. velezensis and 431.28 Ų for BsEG2. However, the volume of the substrate binding pocket is smaller in BsEG2 (3,068.78 ų vs. 14,059.46 ų for B. velezensis). This tighter packing may constrain chain mobility in the cleft, prolonging the residence of small substrates such as cellobiose and promoting their further hydrolysis to glucose. Ongoing dynamic simulations and time‐resolved studies of oligosaccharide binding and catalysis aim to clarify how pocket geometry and residue dynamics enable BsEG2's pronounced glucose production.

2.9. BsEG2 has no Inhibiting Effects From Reaction Products

A practical corollary of efficient saccharification is resistance to product inhibition, which BsEG2 displays. During lignocellulosic biomass hydrolysis at high substrate concentrations, accumulating products such as oligosaccharides, cellobiose, and glucose can inhibit the enzyme and lower saccharification efficiency [65]. Countering this inhibition by raising enzyme loading drives up biofuel production costs. Since cellobiose is reported to inhibit cellulases more strongly than glucose [66], we tested BsEG2's tolerance by measuring its hydrolysis rate against Azo‐CMC in the presence of added cellobiose and other hydrolysis products. BsEG2 retained full specific activity at cellobiose concentrations up to 200 mM (Figure S17). This product tolerance suggests that BsEG2 may help overcome cellulase inhibition, thereby improving the efficiency and cost‐effectiveness of biomass saccharification.

2.10. Synergistic Interactions of BsEG2 With Cellulase

Finally, because efficient conversion of long cellulose chains to glucose relies on cooperation among cellulases, we evaluated the synergy between BsEG2 and the β‐glucosidase B8CYA8 from Halothermothrix orenii on insoluble substrate Avicel (2% w/v), measuring glucose yield from combined enzyme action. The total enzyme load was kept constant to allow for a direct comparison of the impact of individual enzymes and the combined effects. The degree of synergy was calculated as the fold increase in glucose production relative to BsEG2 alone.

BsEG2 and B8CYA8 exhibit complementary activities. BsEG2 primarily generates cellobiose, which B8CYA8 efficiently converts to glucose, enhancing overall cellulose hydrolysis (Figure 8). Glucose production from Avicel is reported as a percentage relative to BsEG2 alone, which is considered 100%. The calculated degree of synergy between BsEG2 and B8CYA8 is 1.6, indicating a cooperative effect on Avicel rather than competition for Avicel within the time periods of the assay, and the product of BsEG2 is utilized as substrate by B8CYA8.

FIGURE 8.

The synergistic effect of endoglucanase BsEG2 with β‐glucosidase (V169C/I246A/E173L B8CYA8) was evaluated on 2% Avicel in McIlvaine buffer, pH 6.0, at 60°C for 2 h. Glucose concentration was quantified using the GOD‐POD assay and expressed as a relative percentage, with the glucose yield from BsEG2 alone set at 100%.

An efficient cellulase cocktail design requires highly active enzymes and balanced functionalities. While BsEG2 can produce glucose on its own over longer time scales, pairing it with B8CYA8 accelerates glucose production via synergistic interaction.

3. Conclusions

BsEG2 represents an unusual GH5 endoglucanase that extends the conventional functional profile of this family. In addition to processive endoglucanase activity, it acts on a broad range of substrates, from cellobiose to longer cellulosic chains and lignocellulosic biomass, with direct glucose release as a major outcome. This distinguishes BsEG2 from many reported GH5 endoglucanases, which are generally associated with endo‐cleavage and oligosaccharide production rather than multifunctional glucose‐generating activity. From an application perspective, BsEG2 combines catalytic efficiency with operationally relevant traits, including broad pH and temperature activity, ionic‐liquid tolerance, seawater compatibility, and synergy with glucose‐tolerant β‐glucosidases. These features suggest potential utility in simplifying enzyme systems for biomass saccharification. However, its industrial applicability will require further validation under process‐relevant conditions, including performance in real biomass hydrolysates, stability over extended operation, and scale‐up studies. Overall, this work expands the known catalytic diversity of GH5 endoglucanases and identifies BsEG2 as a promising candidate for the development of more efficient biomass conversion platforms.

Author Contributions

Debjyoti Ghosh: conceptualization, investigation, formal analysis. Aditi Konar: investigation, formal analysis, writing – review and editing, writing – original draft. Yi Qin Gao: project administration, resources. Manas Mondal: investigation, formal analysis, software, methodology, writing – review and editing. Sushant K. Sinha: conceptualization, investigation, writing – original draft, formal analysis, supervision. Supratim Datta: conceptualization, funding acquisition, writing – review and editing, project administration, supervision, resources. S. Venkata Mohan: resources.

Funding

This work was supported by the Anusandhan National Research Foundation (ANRF), Government of India, CRG/2023/002111 (SD), and the Department of Biotechnology (DBT), Government of India, BT/PR47801/BCE/8/1812/2023, and IISER Kolkata Academic Research Funds.

Conflicts of Interest

The authors declare no conflicts of interest

Supporting information

Supporting File: biot70290‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.