Processivity and enzymatic mechanism of a multifunctional family 5 endoglucanase from Bacillus subtilis BS-5 with potential applications in the saccharification of cellulosic substrates

Abstract

Background: Presently, enzymes still constitute a major part of the cost of biofuel production from lignocellulosic biomass. Processive endoglucanases, which possess both endoglucanase and exoglucanase activity, have the potential to reduce the costs of biomass saccharification when used together with commercial cellulases. Therefore, the exploration of new processive endoglucanases has attracted much attention with a view to accelerating the industrialization of biofuels and biochemicals.

Results: The endoglucanase EG5C and its truncated form EG5C-1 from Bacillus subtilis BS-5 were expressed and characterized. EG5C was a typical endoglucanase, comprised of a family 5 catalytic domain and family 3 carbohydrate-binding domain, and which had high activity toward soluble cellulosic substrates, but low activity toward insoluble cellulosic substrates. Importantly, the truncated form EG5C-1 was a processive endoglucanase that hydrolyzed not only carboxymethyl cellulose (CMC), but also insoluble cellulosic substrates. The hydrolytic activities of EG5C-1 towards CMC, phosphoric acid-swollen cellulose (PASC), p-nitrophenyl-β-d-cellobioside, filter paper and Avicel are 4170, 700, 2550, 405 and 320 U/μmol, respectively. These data demonstrated that EG5C-1 had higher activity ratio of exoglucanase to endoglucanase than other known processive endoglucanases. When PASC was degraded by EG5C-1, the ratio of soluble to insoluble reducing sugars was about 3.7 after 3 h of incubation with cellobiose and cellotriose as the main products. Importantly, EG5C-1 alone was able to hydrolyze filter paper and PASC. At 5% substrate concentration and 10 FPU/g PASC enzyme loading, the saccharification yield was 76.5% after 60 h of incubation. Replacement of a phenylalanine residue (F238) by an alanine at the entrance/exit of the substrate binding cleft significantly reduces the ability of EG5C-1 to degrade filter paper and Avicel, but this mutation has little impact on CMCase activity. The processivity of this mutant was also greatly reduced while its cellulose binding ability was markedly enhanced.

Conclusions: The processive endoglucanase EG5C-1 from B. subtilis BS-5 exhibits excellent properties that render it a suitable candidate for use in biofuel and biochemical production from lignocellulosic biomass. In addition, our studies also provide useful information for research on enzyme processivity at the molecular level.

Keywords: Bacillus subtilis; Family 5 glycoside hydrolase; Multifunctional; Processive endoglucanase; Saccharification.

Fig. 1

a SDS-PAGE analysis of cellulose-binding proteins secreted by B. subtilis BS-5. Lane M marker, Lane 1 fermentation broth of B. subtilis BS-5, Lane 2 unbound fractions in the fermentation broth after Avicel adsorption, Lane 3 cellulose-binding proteins bound to Avicel, eluted with SDS-PAGE buffer. b SDS-PAGE analysis of recombinant EG5C, EG5C-1 and EG5C-2 produced by E. coli. Lane M-marker, Lane 1-crude extracts of BL21-pET28a, Lanes 2, 4 and 6-crude extracts of E-pET-eg5c-1, E-pET-eg5c, E-pET-eg5c-2, Lines 3, 5 and 7-purified EG5C-1, EG5C and EG5C-2 with a HisTrap FF column

Fig. 2

Effect of temperature and pH on the activity and stability of EG5C and its derivatives. a Activity of EG5C (○), EG5C-1 (□) and EG5C-2 (∆) was measured at different pH values at 45 °C. b pH stability of EG5C (○), EG5C-1 (□) and EG5C-2 (∆) evaluated following incubation at 45 °C for 2 h at different pH values. The buffer systems were as follows: 50 mmol/L citric acid/sodium citrate (pH 3.0–6.0), 50 mmol/L Na₂HPO₄/KH₂PO₄ (pH 6.0–8.5), 50 mmol/L Gly/NaOH (pH 8.5–10.5). c The effect of temperature on activity of EG5C (○), EG5C-1 (□) and EG5C-2 (∆) was studied by changing the temperature from 30 to 80 °C at pH 6.5. d Thermosatbility of EG5C (○), EG5C-1 (□) and EG5C-2 (∆) measured by incubating the enzyme at various temperatures for 2 h. Error bars are standard deviation of 3 independent experiments

Fig. 3

Effect of EG5C (○), EG5C-1 (□) and EG5C-2 (∆) on the viscosity of a CMC solution. Error bars are standard deviation of 3 independent experiments

Fig. 4

The ratios of the soluble to insoluble reducing sugars generated by EG5C, EG5C-1, EG5C-2 and F238A mutant of EG5C-1, respectively. Error bars are standard deviation of 3 independent experiments

Fig. 5

The binding ability of EG5C, EG5C-1 and EG5C-2 for Avicel. Lane M-marker, Lines 1–3-initial, unbound and bound proteins of EG5C, Lines 4–6-initial, unbound and bound proteins of EG5C-2, Lines 7–9-initial, unbound and bound proteins of EG5C-1. Bound proteins of EG5C and its derivatives were released from Avicel using the sample buffer of SDS-PAGE

Fig. 6

TLC analysis of the hydrolysis products from various cellulosic substrates by EG5C and its derivatives. a Hydrolysis products from cello-oligosaccharides. Lane M-glucose unit markers, i.e. glucose (G1), cellobiose (G2), cellotriose (G3), cellotetraose (G4) and cellopentaose (G5), Lanes 1–3-products released from cellobiose by EG5C-1, EG5C and EG5C-2, Lanes 4–6-products released from cellotriose by EG5C-1, EG5C and EG5C-2, Lanes 7–9-products released from cellotetraose by EG5C-1, EG5C and EG5C-2, Lanes 10–12-products released from cellopentaose by EG5C-1, EG5C and EG5C-2. b Products obtained from the hydrolysis of CMC, PASC and Avicel. Lane-M glucose unit markers, i.e. glucose (G1), cellobiose (G2), cellotriose (G3), cellotetraose (G4) and cellopentaose (G5), Lanes 2–4-hydrolysis products from CMC by EG5C-1, EG5C and EG5C-2, Lanes 6–8-hydrolysis products from PASC by EG5C-1, EG5C and EG5C-2, Lanes 10–12-hydrolysis products from Avicel by EG5C-1, EG5C and EG5C-2, Lanes 1, 5, 9-supernatant of CMC, PASC and Avicel in the absence of enzyme

Fig. 7

a Hydrolysis yield of PASC by EG5C-1 alone (□), or EG5C-1 in combination with Novozyme 188 (○). This experiment was carried out at pH 6.5 in Na₂HPO₄/KH₂PO₄ buffer at 40 °C with 5% PASC. The EG5C-1 loading was 10 FPU/g PASC. The loading for Novozyme 188 was 20 U/g PASC. Error bars present standard deviations based on three independent experiments. b HPLC analysis of the products from PASC hydrolysis by EGC5-1 alone (lane 2), or by a combination of EG5C-1 and Novozyme 188 (lane 3) after 60 h of incubation. Lane 1-standards of G1 to G3

Fig. 8

Model of the three dimensional structure of EG5C-1. a The protein structure of EG5C-1 shown as cartoon representations with β-strands in yellow, α-helices in red, loops in grey and turn in blue, and the docked substrate cellotetraose styled in green stick representation. b Surface representation of the binding pocket in EG5C-1, highlighting the seven aromatic residues Trp69, Tyr70, Tyr96, Trp207, Tyr231, Phe238 and Trp291 shown as cyan spheres, and the catalytic residues Glu169 and Glu257, shown as violet sticks

Fig. 9

Effect of site-directed mutations of selected aromatic residues of EG5C-1 on the hydrolysis activity towards CMC, PASC, pNPC, filter paper and Avicel. Error bars are standard deviation of 3 independent experiments