Cellulose hydrolysis – the role of monocomponent cellulases in crystalline cellulose degradation

Changes in the molecular structure of cellulose during hydrolysis with four recombinant -1,4-glycanases from the cellulolytic bacterium Cellulomonas fimi were assessed and compared in an attempt to elucidate the mechanism of crystalline cellulose degradation. It was apparent that the two endoglucanases, Cel6A and Cel5A, degraded sigmacell cellulose differently; Cel5A liberated more soluble sugars (cellobiose and cellotriose) and significantly altered the molecular weight distribution, while Cel6A had a limited effect on the polymer size and liberated primarily cellobiose and glucose. Additionally, both endoglucanases slightly increased the crystallinity of cellulose. In contrast, the cellobiohydrolases, Cel6B and Cel48A, had no effect on cellulose molecular weight and liberated only cellobiose and cellotriose. However, Cel48A was shown to be effective at reducing the crystallinity of the cellulosic substrate, while Cel6B increased the crystallinity index. Synergistic hydrolysis using combinations of the different enzymes showed that, although the cellulose was extensively hydrolysed, the molecular structure of the substrate was similar to the original material. This phenomenon suggests that the actions of individual monocomponent enzymes are offset by the concurrent modification by the complementing enzymes during synergistic hydrolysis.     

Integration of computer modeling and initial studies of site-directed mutagenesis to improve cellulase activity on Cel9A from Thermobifida fusca

Cellulases are a complex group of enzymes that are fundamental for the degradation of amorphous and crystalline cellulose in lignocellulosic material. Unfortunately, cellulases have a low catalytic efficiency on their substrates when compared to similar enzymes such as amylases, which has led to a strong interest in improving their activities. Thermobifida fusca secretes six cellulose degrading enzymes: two exo- and three endocellulases and an endo/exocellulase Cel9A (formerly called E4). Cel9A shows unique properties because of its endo- and exocellulase characteristics, strong activity on crystalline cellulose, and good synergistic properties. Therefore, it is an excellent target for mutagenesis techniques to improve crystalline cellulose degradation. In this article, we describe research conducted to improve Cel9A catalytic efficiency using a rational design and computer modeling. A computer model of Cel9A was created using the program CHARMM plus its PDB structure and a cellohexose molecule attached to the catalytic site as a starting model. Initially molecular graphics and energy minimization were used to extend the cellulose chain to 18 glucose residues spanning the catalytic domain and cellulose-binding domain (CBD). The interaction between this cellulose chain and conserved CBD residues was determined in the model, and mutations likely to improve the binding properties of the CBD were selected. Site-directed mutations were carried out using the pET vector pET26b, Escherichia coli DH5-α, and the QuickChange mutagenesis method. E. coli BL21-DE3 was used for protein production and expression. The purified proteins were assayed for enzymatic activity on filter paper, swollen cellulose, bacterial microcrystalline cellulose, and carboxymethylcellulose (CMC). Mutation of the conserved residue F476 to Y476 gave a 40% improved activity in assays with soluble and amorphous cellulose such as CMC and swollen cellulose.     

Effect of sodium hydroxide treatment of bacterial cellulose on cellulase activity

The synergistic action between Thermobifida fusca exocellulase Cel6B and endocellulase Cel5A on sodium hydroxide pretreated bacterial cellulose (BC) was determined. The activities of Cel6B and Cel5A were tested singly and both activities were dramatically increased on pretreated BC, especially in the early stage of hydrolysis. Cel5A, which attacks the cellulose chain randomly, showed a larger increase on NaOH treated BC than Cel6B. Mixtures of the two enzymes were also able to degrade NaOH treated BC faster than BC and the kinetics of the mixture differed from that of the individual enzymes. The degree of synergistic effect (DSE) on BC decreased dramatically with time of hydrolysis. However, the DSE on NaOH treated BC was almost constant throughout the incubation, with a smaller effect at higher NaOH concentrations. The change caused by NaOH did not increase the DSE, although each individual cellulase activity increased. This showed that synergistic activity was more effective on recalcitrant cellulose, which requires effective cooperation between the cellulase components for hydrolysis.     

Probing the role of N-linked glycans in the stability and activity of fungal cellobiohydrolases by mutational analysis

The filamentous fungi Trichoderma reesei and Penicillium funiculosum produce highly effective enzyme mixtures that degrade the cellulose and hemicellulose components of plant cell walls. Many fungal species produce a glycoside hydrolase family 7 (Cel7A) cellobiohydrolase, a class of enzymes that catalytically process from the reducing end of cellulose. A direct amino acid comparison of these two enzymes shows that they not only have high amino acid homology, but also contain analogous N-linked glycosylation sites on the catalytic domain. We have previously shown (Jeoh et al. in Biotechnol Biofuels, 1:10, 2008) that expression of T. reesei cellobiohydrolase I in a commonly used industrial expression host, Aspergillus niger var. awamori, results in an increase in the amount of N-linked glycosylation of the enzyme, which negatively affects crystalline cellulose degradation activity as well as thermal stability. This complementary study examines the significance of individual N-linked glycans on the surface of the catalytic domain of Cel7A cellobiohydrolases from T. reesei and P. funiculosum by genetically adding or removing N-linked glycosylation motifs using site directed mutagenesis. Modified enzymes, expressed in A. niger var. awamori, were tested for activity and thermal stability. It was concluded that N-linked glycans in peptide loops that form part of the active site tunnel have the greatest impact on both thermal stability and enzymatic activity on crystalline cellulose for both the T. reesei and P. funiculosum Cel7A enzymes. Specifically, for the Cel7A T. reesei enzyme expressed in A. niger var. awamori, removal of the N384 glycosylation site yields a mutant with 70% greater activity after 120 h compared to the heterologously expressed wild type T. reesei enzyme. In addition, similar activity improvements were found to be associated with the addition of a new glycosylation motif at N194 in P. funiculosum. This mutant also exhibits 70% greater activity after 120 h compared to the wild type P. funiculosum enzyme expressed in A. niger var. awamori. Overall, this study demonstrates that “tuning” enzyme glycosylation for expression from heterologous expression hosts is essential for generating engineered enzymes with optimal stability and activity.     

Enzymatic kinetic of cellulose hydrolysis

The ethanol effect on the Trichoderma reesei cellulases was studied to quantify and clarify this inhibition type. To determine inhibition parameters of crude cellulase and purified exoglucanase Cel7A, integrated Michaelis-Menten equations were used assuming the presence of two inhibitors: cellobiose as the reaction product and ethanol as a possible bioproduct of cellulose fermentation. It was found that hydrolysis of cellulose by crude enzyme follows a model that considers noncompetitive inhibition by ethanol, whereas Cel7A is very slightly competitively inhibited. Crude cellulase is much more inhibited (K _iul=K _icl=151.9 mM) than exoglucanase Cel7A (K _icl=1.6 × 10¹⁵ mM). Also, calculated inhibition constants showed that cellobiose inhibition is more potent than ethanol inhibition both for the crude enzyme as well as exoglucanase Cel7A.     

Exopolysaccharide production by a marine cyanobacterium Cyanothece sp.

The adsorption and the hydrolytic action of purified cellulases of Trichoderma reesei, namely, cellobiohydrolase I (CBH I), endoglucanase II (EG II), and their core proteins, on steam-pretreated willow were compared. The two enzymes differed clearly in their adsorption and hydrolytic behavior. CBH I required the cellulose-binding domain (CBD) for efficient adsorption and hydrolysis, whereas EG II was able to adsorb to steam pretreated willow without its CBD. Absence of the CBD decreased the hydrolysis of cellulose by EG II, but the decrease was less pronounced than with CBH I. A linear relationship was observed between the amount of enzyme adsorbed and the degree of hydrolysis of cellulose only for CBHI. EG II and EG II core appeared to be able to hydrolyze only 1 to 2% of the substrate regardless of the amount of protein adsorbed.     

Protein allostery at the solid-liquid interface: endoglucanase attachment to cellulose affects glucan clenching in the binding cleft

At phase boundaries, physical activities of enzymes such as substrate complexation play critical roles in driving biocatalysis. A prominent example is the cellulase cocktails secreted by fungi and bacteria for deconstructing crystalline cellulose in biomass into soluble sugars. At interfaces, molecular mechanisms of the physical steps in biocatalysis remain elusive due to the difficulties of characterizing protein action with high temporal and spatial resolution. Here, we focus on endoglucanase I (Cel7B) from the fungus Trichoderma reesei that hydrolyzes glycosidic bonds on cellulose randomly. We employ all-atom molecular dynamics (MD) simulations to elucidate the interactions of the catalytic domain (CD) of Cel7B with a cellulose microfibril before and after complexing a glucan chain in the binding cleft. The calculated mechanical coupling networks in Cel7B-glucan and Cel7B-microfibril complexes reveal a previously unresolved allosteric coupling at the solid-liquid interface: attachment of the Cel7B CD to the cellulose surface affects glucan chain clenching in the binding cleft. Alternative loop segments of the Cel7B CD were found to affix to intact or defective surface structures on the microfibril, depending on the complexation state. From a multiple sequence alignment, residues in surface-affixing segments show strong conservation, highlighting the functional importance of the physical activities that they facilitate. Surface-affixing residues also demonstrate significant sequence correlation with active-site residues, revealing the functional connection between complexation and hydrolysis. Analysis of the Cel7B CD exemplifies that the mechanical coupling networks calculated from atomistic MD simulations can be used to capture the conservation and correlation in sequence alignment.     

Synergism in binary mixtures of Thermobifida fusca cellulases Cel6B,Cel9A,and Cel5A on BMCC and Avicel

In an earlier binding study conducted in our laboratory using Thermobifida fusca cellulases Cel6B, Cel9A, and Cel5A (formally Thermomonospora fusca E₃, E₄, and E₅), it was observed that binding capacities for these three cellulases were 18–30 times higher on BMCC than on Avicel. These results stimulated an interest in how the difference in accessibility between the two cellulosic substrates would affect synergism observed with cellulase mixtures. To explore the impact of substrate, accessibility on the extent of conversion and synergism, three binary T. fusca cellulase mixtures were tested over a range of cellulase ratios and total molar cellulase concentrations on Avicel and BMCC. Higher extents of conversion were observed for BMCC due to the higher enzyme to substrate ratio resulting from the higher binding The processive endoglucanase, Cel9A, had four times the extent of conversion of the end endocellulase Cel5A, while the exocellulase Cel6B had three times the extent of conversion of Cel5A. Approximately 500 nmol/g of the cel9A+Cel6B mixture was needed to obtain 80% conversion, while the Cel6B+Cel5A and Cel9A+Cel5A mixtures required 1500 and 1250 nmol/g, respectively, to obtain 80% conversion. Thus, it appears that the more accessible structure of BMCC, as reflected by its binding capacity, results in relative higher processive activity.     

Mixtures of Thermostable Enzymes Show High Performance in Biomass Saccharification

Optimal enzyme mixtures of six Trichoderma reesei enzymes and five thermostable enzyme components were developed for the hydrolysis of hydrothermally pretreated wheat straw, alkaline oxidised sugar cane bagasse and steam-exploded bagasse by statistically designed experiments. Preliminary studies to narrow down the optimization parameters showed that a cellobiohydrolase/endoglucanase (CBH/EG) ratio of 4:1 or higher of thermostable enzymes gave the maximal CBH-EG synergy in the hydrolysis of hydrothermally pretreated wheat straw. The composition of optimal enzyme mixtures depended clearly on the substrate and on the enzyme system studied. The optimal enzyme mixture of thermostable enzymes was dominated by Cel7A and required a relatively high amount of xylanase, whereas with T. reesei enzymes, the high proportion of Cel7B appeared to provide the required xylanase activity. The main effect of the pretreatment method was that the required proportion of xylanase was higher and the proportion of Cel7A lower in the optimized mixture for hydrolysis of alkaline oxidised bagasse than steam-exploded bagasse. In prolonged hydrolyses, less Cel7A was generally required in the optimal mixture. Five-component mixtures of thermostable enzymes showed comparable hydrolysis yields to those of commercial enzyme mixtures.     

Investigation of endoglucanase selectivity on carboxymethyl cellulose by mass spectrometric techniques

The benefits of applying cellulose selective enzymes as analytical tools for chemical structure characterization of cellulose derivatives have been frequently addressed over the years. In a recent study the high selectivity of cellulase Cel45A from Trichoderma reesei (Tr Cel45A) was utilized for relating the chemical structure to the flow properties of carboxymethyl cellulose (CMC). However, in order to take full advantage of the enzymatic hydrolysis the enzyme selectivity on the cellulose substrate must be further investigated. Therefore, the selectivity of Tr Cel45A on CMC was studied by chemical sample preparation of the enzyme products followed by mass spectrometric chemical structure characterization. The results strongly suggest that, in accordance with recent studies, also this highly selective endoglucanase is able to catalyze hydrolysis of glucosidic bonds adjacent to mono-substituted anhydroglucose units (AGUs). Furthermore, the results also indicate that substituents on the nearby AGUs will affect the hydrolysis.     

Hydrolysis of steam-pretreated lignocellulose

The mechanism of hydrolysis of cellulose is important for improving the enzymatic conversion in bioprocesses based on lignocellulose. Adsorption and hydrolysis experiments were performed with cellobiohydrolase I (CBH I) and endoglucanase II (EG II) from Trichoderma reesei on a realistic lignocellulose substrates: steam-pretreated willow. The enzymes were studied both alone and in equimolar mixtures. Adsorption isotherms were determined at 4 and 40 degrees C during 90-min reaction times. Both CBH I and EG II adsorbed stronger at 40 than at 4 degrees C. The time course of adsorption and hydrolysis, 3 min to 48 h, was studied at 40 degrees C. About 90% of the cellulases were adsorbed within 2 h. The hydrolysis rate was high in the beginning but decreased during the time course. Based on adsorption data, the hydrolysis and synergism were analyzed as function of adsorbed enzyme. CBH I showed a linear correlation between hydrolysis and adsorbed enzyme, whereas for EG II the corresponding curve leveled off at both 4 and 40 degrees C. At low conversion, below 1%, EG II produced as much soluble sugars as CBH I. At higher conversion, CBH I was more efficient than EG II. The synergism as function of adsorbed enzyme increased with bound enzyme before reaching a stable value of about 2. The effect of varying the ratio of CBH I:EG II was studied at fixed total enzyme loading and by changing the ratio between the enzymes. Only a small addition (5%) of EG II to a CBH I solution was shown to be sufficient for nearly maximal synergism. The ratio between EG II and CBH I was not critical. The ratio 40% EG II:60% CBH I showed similar conversion to 5% EG II:95% CBH I. Modifications of the conventional endo-exo synergism model are proposed.     

Thermodynamical studies of designed ligands binding to Cel7A using partial-filling capillary electrophoresis

A convenient experimental method for thermodynamical studies based on partial-filling affinity CE is presented. The advantages of this approach are the possibility to determine binding energies from relatively weak interactions as well as the small amounts of samples consumed. In order to explore the affinity and selectivity of the cellobiohydrolase Cel7A, a number of propranolol analogues were recently designed. The affinities of a selection of these ligands were determined in the temperature interval 15-40 degrees C, and DeltaG degrees , DeltaH degrees and DeltaS degrees were obtained by means of Van't Hoff plots. Through these experiments, the importance of the entropy contribution in the complexation between the ligands and Cel7A has been demonstrated.     

Thermoascus aurantiacus CBHI/Cel7A production in Trichoderma reesei on alternative carbon sources

To develop functional enzymes in cellulose hydrolysis at or above 70 degrees C the cellobiohydrolase (CBHI/Cel7A) of Thermoascus aurantiacus was cloned and expressed in Trichoderma reesei Rut-C30 under the strong cbh1 promoter. Cellulase production of the parental strain and the novel strain (RF6026) was examined in submerged fermentation experiments using various carbon sources, which were lactose, Solka Floc 200 cellulose powder, and steam pretreated corn stover. An industrially feasible production medium was used containing only distiller's spent grain, KH(2)PO(4), and (NH(4))(2)SO(4). Enzyme production was followed by measurements of protein concentration, total cellulase enzyme activity (filter paper activity), beta-glucosidase activity, CBHI activity, and endogenase I (EGI) activity. The Thermoascus CBHI/Cel7A activity was taken as an indication of the heterologous gene expression under the cbh1 promoter.     

Catalytically enhanced endocellulase cel5a from Acidothermus cellulolyticus

When Tyr245 in endocellulase Cel5A from Acidothermus cellulolyticus was changed to Gly (Y245G) by designed mutation, the value of K _i for inhibition of the enzyme by the product cellobiose was increased more than 1480%. This reduction in product inhibition enabled the mutant enzyme (used in conjunction with Trichoderma reesei cellobiohydrolase-I) to release soluble sugars from biomass cellulose at a rate as much as 40% greater than that achieved by the wild-type (WT) enzyme. The mutant was designed on the basis of the previously published crystal structure of the WT enzyme/substrate complex (at a resolution of 2.4 Å), which provided insights into the enzyme mechanism at the atomic level and identified Tyr245 as a key residue interacting with a leaving group. To determine the origin of the change in activity, the crystal structure of Y245G was solved at 2.4-Å resolution to an R-factor of 0.19 (R-free=0.25). To obtain additional information on the enzyme-product interactions, density functional calculations were performed on representative fragments of the WT Cel5A and Y245G. The combined results indicate that the loss of the platform (Y245G) and of a hydrogen bond (from a conformational change in Gln247) reduces the binding energy between product and enzyme by several kilo calories per mole. Both kinetic and structural analyses thus relate the increased enzymatic activity to reduced product inhibition.     

Investigation of micro-immobilised enzyme reactors containing endoglucanases for efficient hydrolysis of cellodextrins and cellulose derivatives

Cellulases hydrolysing the interior parts of cellulose, also called endoglucanases, were immobilised in micro-immobilised enzyme reactors (μIMER) made of porous silicon with the purpose of investigating the use of such μIMERs for hydrolysis of cellodextrins and soluble cellulose derivatives. The endoglucanases Trichoderma reesei Cel 12A (TrCel 12A) and Bacillus agaradhaerens Cel 5A (BaCel 5A) were covalently coupled to the surface of a silicon microchip through Schiff base formation. For characterisation cellohexaose was used as substrate for the immobilised enzymes. The characteristics of the μIMER were investigated by studying the product formation when varying the concentration, flow-rate, temperature and pH of the substrate solution. Hydrolysis was performed in the μIMER connected on-line to a chromatographic system, where the products were separated and detected using high-performance anion exchange chromatography (HPAEC) coupled to pulsed amperometric detection (PAD). A comparison of the hydrolytic pattern between BaCel 5A and TrCel 12A was carried out and the results show that the two investigated endoglucanases give specific hydrolytic patterns in the products formed that provide important information about the enzymes. The μIMERs are robust and can be employed continuously over a period of at least several days. Moreover, on appropriate storage, no activity loss is seen after 60 days. The ability of the BaCel 5A containing μIMER to perform hydrolysis of derivatised cellulose was also investigated using carboxymethyl cellulose (CMC) as substrate. Separation and detection were carried out using size exclusion chromatography (SEC) with refractive index detection (RI). The results show that the μIMERs are robust and can be employed for on-line hydrolysis of both cellodextrins and derivatised cellulose of high molecular weight.     

Cellulase production of Trichoderma reesei rut C 30 using steam-pretreated spruce

Various techniques are available for the conversion of lignocellulosics to fuel ethanol. During the last decade processes based on enzymatic hydrolysis of cellulose have been investigated more extensively, showing good yield on both hardwood and softwood. The cellulase production of a filamentous fungi, Trichoderma reesei Rut C30, was examined on carbon sources obtained after steam pretreatment of spruce. These materials were washed fibrous steam-pretreated spruce (SPS), and hem icellulose hydrolysate. The hemicellulose hydrolysate contained, besides water-soluble carbohydrates, lignin and sugar degradation products, which were formed during the pretreatment and proved to be inhibitory to microorganisms. Experiments were performed in a 4-L laboratory fermentor. The hydrolytic capacity of the produced enzyme solutions was compared with two commercially available enzyme preparations, Celluclast and logen Cellulase, on SPS, washed SPS, and Solka Floc cellulose powder. There was no significant difference among the different enzymes produced by T. reesei Rut C30. However, the conversion of cellulose using these enzymes was higher than that obtained with logen or Celluclast cellulases using steam-pretreated spruce as substrate.     

Effect of digestion by pure cellulases on crystallinity and average chain length for bacterial and microcrystalline celluloses

In this study we employed Size Exclusion Chromatography (SEC) and X-ray diffraction to monitor the molecular weight and crystallinity of bacterial cellulose I and II (BC-I, BC-II) and microcrystalline cellulose (MCC) digested with three “pure” Thermobifida fusca cellulases (Cel6A, Cel6B, and Cel9A ). For each enzyme, cellulose crystallinity was found to increase modestly with treatment time. The digestion rate of BC-II was higher than that of BC-I for Cel6A and Cel9A, both endocellulases. SEC results show that the endocellulases create a very rapid decrease in cellulose molecular weight while a slower molecular weight loss was observed with Cel6B, an exocellulase. This work suggests that conversion of native cellulose I to cellulose II by mercerization may beneficially impact the rate of sugar release by cellulases from biomass. In general, lower conversion rates are observed for MCC compared to BC, possibly due to a higher initial crystallinity for MCC. Surface area effects may also be important.     

Selection of hydrophobic membranes in the lipase-catalyzed hydrolysis of olive oil

A hydrophobic membrane (HVHP, polyvinylidene difluoride) was selected out of HVHP, PTHK and PTGC (polysulfone) membranes to immobilize Candida rugosa lipase by physical adsorption in the hydrolysis of olive oil in a stirred diffusion cell. A previous model that assumed the Michaelis–Menten kinetics and Langmuir adsorption isotherm for the adsorbed lipase was used to interpret the variation of initial hydrolysis rates with enzyme and substrate concentrations. Replacing the aqueous phase by a fresh buffer, with or without containing partially deactivated lipases, during the reaction did not affect the enzyme activity for the adsorbed lipase. Moreover, the same enzyme performance was obtained when a fresh and a regenerated membrane was used as the carrier in the membrane reactor.     

Molecular cloning and biochemical characterization of a family-9 endoglucanase with an unusual structure from the gliding bacteria <Emphasis Type="Italic">Cytophaga hut chinsonii</Emphasis>

Cytophaga hutchinsonii was originally isolated from sugarcane piles. This microorganism therefore probably produces an array of enzymes allowing it to digest cellulosic substrates. C. hutchinsonii thus represents a rich source of potentially effective cellulase enzymes that can be harnessed for conversion of biomass to simple sugars. These sugars can then be used as feedstock for ethanol production or other chemical syntheses. In this study, we report the PCR cloning of an endoglucanase gene (Cel9A) from C. hutchinsonii using degenerated primers directed at the catalytic domain. Alignment of the amino acids sequence revealed that Cel9A has a gene structure totally different from the other known cellulose degraders. The most striking feature of this cloned protein is the absence of a cellulose-binding domain (CBD), which to date was believed to be imperative in cellulose hydrolysis. Consequently, the Cel9A gene, encoding β-1,4 endoglucanase from C. hutchinsonii was over-expressed in Escherichia coli with a His-Tag based expression vector. The resulting polypeptide, with a molecular mass of 105 KDa, was purified from cell extracts by affinity chromatography on cellulose. Mature Cel9A was optimally active at pH 5.0 and 45°C. The enzyme efficiently hydrolyzes carboxymethyl-cellulose (CMC). Analysis of CMC and filter paper hydrolysis suggests that Cel9A is a nonprocessive enzyme with endo-cellulase activities.