Microcalorimetric study of cellulose degradation by Cellulomonas uda ATCC 21399

A newly designed batch calorimeter was used to investigate the degradability of some celluloses having varying degrees of crystallinity. The PTC of an aerobic culture of Cellulomonas uda ATCC 21399 obtained revealed a diauxic growth which is attributed to the presence of hemicellulose contaminating Avicel and MN300 cellulose. The microcrystalline celluloses used were not completely utilized, whereas amorphous cellulose was easily metabolized, indicating that under the growth conditions used here, the physical structure of cellulose strongly influenced its microbial degradability. An equivalent growth yield of ca. 0.44 g/g was found with all the substrates used. The heat evolved by metabolism of 1 g cellulose was -5.86 kJ/g, a value similar to that obtained with glucose culture. The growth rate was the only variable parameter. The data obtained showed as expected that the hydrolysis product of cellulose was consumed in the same way as that of glucose and that the only limiting factor to the biodegradability of cellulose was the breakdown of the polymeric substrate. It is concluded that data obtained with glucose metabolism can be used to evaluate the extent of cellulose degradation.     

Comparative study of samples of powdered and microcrystalline celluloses of different natural origins: Supermolecular structure and the chemical composition of powdered samples

Methods of wide-angle X-ray scattering (WAXS), high resolution solid-state ¹³C NMR, and Fourier transform IR-spectroscopy are applied to study supermolecular structures and functional compositions of lignocellulose samples of wood and grass origins and powdered celluloses (PC) obtained from them under identical hydrolysis conditions. It was shown by WAXS that the structure of cellulose I is preserved in samples of powdered celluloses, however, an increased degree of crystallinity and cross-section sizes of crystallites are observed in PC samples. Specific features of changes in the supermolecular structure of cellulose occurred after the hydrolysis, i.e., an increase in the content of cellulose I_β in PC compared to the initial samples, are established by ¹³C NMR method. It was shown by means of ¹³C NMR and Fourier transform IR-spectroscopy that the functional chemical composition of lignocelluloses is weakly affected by the hydrolysis. The presence of residual lignin functional groups in the samples is confirmed.     

Factors influencing the regeneration of cellulose I from phosphoric acid

The regeneration of cellulose I from phosphoric acid solution was studied, with emphasis both on the conditions required for the regeneration of cellulose I and characterization of the resulting cellulose by X-ray diffraction. Raman spectroscopy and SS¹³C n.m.r. As the conditions of regeneration were varied with respect to temperature and time a variety of polymorphs were produced. As previously reported, the cellulose I polymorph dominated at high temperature and long regeneration time. The question of the authenticity of the regenerated cellulose I was addressed, with several tests confirming that it was not an insoluble residue. Further analysis of the various regenerated celluloses revealed that they all had a small cellulose I component that could be isolated by acid hydrolysis. It is suggested that during regeneration, nuclei of different cellulose polymorphs are formed simultaneously, the proportion of each dependent upon the relative rates of nucleation. Under degradative regeneration conditions the polymorphs more susceptible to hydrolysis are attacked preferentially, leaving behind resistant cellulose I     

Detrimental effect of cellulose oxidation on cellulose hydrolysis by cellulase

To effectively convert complex and recalcitrant biomass carbohydrates to simple platform sugars useful for fuel and chemicals production, mechanical or chemical pre-treatments are often required to make the carbohydrates more accessible for enzymatic hydrolysis. Due to their harsh conditions, some pre-treatments might negatively affect enzymatic hydrolysis because of events such as cellulose oxidation. To study how oxidative modification may impact cellulose's reactivity toward hydrolysis by cellulases, we prepared three cellulose substrates by cupric ion and hypochlorite oxidations, and subjected the derived celluloses to hydrolysis by various cellobiohydrolases from glycoside hydrolase families 6 and 7, and one cellulolytic Hypocrea jecorina extracellular enzyme mixture. We observed a profound decrease of enzymatic hydrolysis on the oxidized celluloses. The effect was attributed to the interference, from oxidized functional groups in cellulose, on its binding/activation in the active pocket/tunnel of cellobiohydrolases. Potential implication of the observed effect from cellulose oxidation on pre-treatment optimization and cellulase improvement was discussed.     

Enzymatic hydrolysis and recrystallization behavior of initially amorphous cellulose

Cellulose samples from cotton and wood pulps with varying low degrees of crystallinity (mechanically decrystallized) were studied. The influence of initial cellulose crystallinity on sugar yield after enzymatic hydrolysis was determined by two different methods. As expected, samples with low crystallinity were much more accessible to enzymatic attack and glucose yields were higher than were samples of high initial crystallinity. Hydrolysis of cellulose seems more dependent on cellulose crystallinity than on the source of cellulose. It is known that decrystallized or amorphous cellulose can recrystallize under proper conditions, e.g., during acid hydrolysis. The data reported here also reveal some recrystallization during enzymatic hydrolysis which probably occurs simulataneously with a selective enzymatic attack on the amorphous regions of cellulose. In all cases, the amorphous celluloses recrystallized in the original lattice form, that of native cellulose.     

Determination of the number-average degree of polymerization of cellodextrins and cellulose with application to enzymatic hydrolysis

A rapid and accurate method for determining the number-average degree of polymerization (DP(n)) was established for insoluble cellulose and soluble cellodextrins as the ratio of glucosyl monomer concentration determined by the phenol-sulfuric acid method divided by the reducing-end concentration determined by a modified 2,2'-bicinchoninate (BCA) method. The modified BCA method, featuring incubation at 75 degrees C for 30 min, did not result in beta-glucosidic bond cleavage, whereas substantial cleavage was observed at higher temperature. Solubilization of insoluble cellulose in cold phosphoric acid prior to measurement of the reducing-end concentration by the BCA method was found not to be necessary for several model celluloses such as microcrystalline cellulose, but such solubilization was required for large fibers of cellulose such as Whatman No. 1 filter paper. The phenol-sulfuric acid method can be used for measuring the glucosyl monomer concentration of soluble cellodextrins, and also for insoluble cellulose if preceded by a liquefaction step. Standard deviations of < or =2% were obtained for both reducing and glucosyl monomer determination and of < or =3% for overall determination of DP. By use of the reported method, hydrolysis of phosphoric acid-swollen cellulose (PASC) by the Trichoderma reesei cellulase system was shown to result in a rapid decrease in DP as hydrolysis proceeded. By contrast, the DP of Avicel remained nearly constant during hydrolysis. The specific enzymatic cellulose hydrolysis rate is 100-fold higher for PASC as compared to Avicel.     

Micropatterned thin film honeycomb materials from regiospecifically modified cellulose

Thin film honeycomb materials were prepared from regioselectively modified celluloses. The method uses water condensation at the surface of a cellulosic solution as an ordered template to form honeycomb structures. Pore size and distribution is controlled by several factors, one of which is the hydrophilicity of the cellulosic used. The amphiphilic nature of the celluloses was modified with varying lengths of ethylene glycol side chains using 2,6-thexyldimethylsilyl cellulose. It was found that the side chains do affect the honeycomb formation, with longer ethylene glycol chains leading to increased pore uniformity but having little influence on the pore size.     

Enzymatic hydrolysis and physical characterization of commercial celluloses and cellulose-based ion-exchange powdered mixed resins

Commercial celluloses (BH20, Epicote, FC+) and their cellulose-containing powdered mixed resins (PMR) were evaluated using enzymatic and physical methods. Samples were hydrolyzed with purified Trichoderma viride cellulase extract and measured for released reducing sugar using the dinitrosalicylic acid method. Physical characterization was performed with gross specific surface areas (GSSA) and relative crystalline indices (RCI). In addition, FC+ was exposed to physical and chemical processing commonly encountered in spent PMR processing to determine potential effects on reducing sugar release in high intensity containers. Reducing sugar released from the celluloses by T. viride cellulase ranged from 135.37 to 244.48 mg day(-1); the celluloses were highly crystalline, ranging from 82.47 to 84.57%; and the GSSA medians for the celluloses ranged from 1,298.60 cm(2) g(-1) to 2,493.20 cm(2) g(-1). Most processing treatments on the FC+ reduced the amount of reducing sugar released and increased RCI. Cellulose hydrolysis rates did not show a strong correlation with the physical characterization. These results suggest that (1) celluloses and PMR can serve as abundant sources of bioavailable carbon in water treatment systems, and (2) the use of correlative physical characteristics to evaluate a cellulose-based commercial product may not accurately predict microbial activity; a complementary microbial test such as cellulose hydrolysis with cellulase may prove useful.     

Regenerating cellulose from ionic liquids for an accelerated enzymatic hydrolysis

The efficient conversion of lignocellulosic materials into fuel ethanol has become a research priority in producing affordable and renewable energy. The pretreatment of lignocelluloses is known to be key to the fast enzymatic hydrolysis of cellulose. Recently, certain ionic liquids (ILs) were found capable of dissolving more than 10wt% cellulose. Preliminary investigations [Dadi, A.P., Varanasi, S., Schall, C.A., 2006. Enhancement of cellulose saccharification kinetics using an ionic liquid pretreatment step. Biotechnol. Bioeng. 95, 904-910; Liu, L., Chen, H., 2006. Enzymatic hydrolysis of cellulose materials treated with ionic liquid [BMIM]Cl. Chin. Sci. Bull. 51, 2432-2436; Dadi, A.P., Schall, C.A., Varanasi, S., 2007. Mitigation of cellulose recalcitrance to enzymatic hydrolysis by ionic liquid pretreatment. Appl. Biochem. Biotechnol. 137-140, 407-421] suggest that celluloses regenerated from IL solutions are subject to faster saccharification than untreated substrates. These encouraging results offer the possibility of using ILs as alternative and non-volatile solvents for cellulose pretreatment. However, these studies are limited to two chloride-based ILs: (a) 1-butyl-3-methylimidazolium chloride ([BMIM]Cl), which is a corrosive, toxic and extremely hygroscopic solid (m.p. approximately 70 degrees C), and (b) 1-allyl-3-methylimidazolium chloride ([AMIM]Cl), which is viscous and has a reactive side-chain. Therefore, more in-depth research involving other ILs is much needed to explore this promising pretreatment route. For this reason, we studied a number of chloride- and acetate-based ILs for cellulose regeneration, including several ILs newly developed in our laboratory. This will enable us to select inexpensive, efficient and environmentally benign solvents for processing cellulosic biomass. Our data confirm that all regenerated celluloses are less crystalline (58-75% lower) and more accessible to cellulase (>2 times) than untreated substrates. As a result, regenerated Avicel((R)) cellulose, filter paper and cotton were hydrolyzed 2-10 times faster than the respective untreated celluloses. A complete hydrolysis of Avicel((R)) cellulose could be achieved in 6h given the Trichoderma reesei cellulase/substrate ratio (w/w) of 3:20 at 50 degrees C. In addition, we observed that cellulase is more thermally stable (up to 60 degrees C) in the presence of regenerated cellulose. Furthermore, our systematic studies suggest that the presence of various ILs during the hydrolysis induced different degrees of cellulase inactivation. Therefore, a thorough removal of IL residues after cellulose regeneration is highly recommended, and a systematic investigation on this subject is much needed.     

Effect of solvent exchange on the supramolecular structure, the molecular mobility and the dissolution behavior of cellulose in LiCl/DMAc

We investigated the effect of solvent exchange on the supramolecular structure and the molecular mobility of the cellulose molecule to clarify the mechanism of the dissolution of cellulose in lithium chloride/N,N-dimethylacetamide (LiCl/DMAc). Among the celluloses that were solvent exchanged in different ways, the DMAc-treated celluloses dissolved most rapidly. Dissolution of the acetone-treated celluloses was much slower than the DMAc-treated ones, but considerably faster than the untreated one. Such differences in the dissolution behavior were well explained by the differences in the surface fractal dimension calculated from the small-angle X-ray scattering profiles and in the (1)H spin-lattice and spin-spin relaxation times estimated from the solid-state NMR spectroscopic measurements. Furthermore, it was suggested from the IR spectra and the (13)C spin-lattice relaxation times of cellulose that DMAc is adsorbed on the surface of cellulose even after vacuum-drying and affects the molecular mobility and hydrogen-bonding state of cellulose.     

Hyperthermostable Endoglucanase from Pyrococcus horikoshii

An endoglucanase homolog from the hyperthermophilic archaeon Pyrococcus horikoshii was expressed in Escherichia coli, and its enzymatic characteristics were examined. The expressed protein was a hyperthermostable endoglucanase which hydrolyzes celluloses, including Avicel and carboxymethyl cellulose, as well as β-glucose oligomers. This enzyme is the first endoglucanase belonging to glycosidase family 5 found from Pyrococcus species and is also the first hyperthermostable endoglucanase to which celluloses are the best substrates. This enzyme is expected to be useful for industrial hydrolysis of cellulose at high temperatures, particularly in biopolishing of cotton products.     

Distribution of carboxylate groups introduced into cotton linters by the TEMPO-mediated oxidation

The 2,2,6,6-tetramethylpiperidine-1-oxy radial (TEMPO)-mediated oxidation was applied to aqueous slurries of cotton linters. The water-insoluble fibrous fractions thus obtained in the yields of more than 78% were characterized by solid-state ¹³C-NMR, X-ray diffraction and scanning electron microscopic analyses for evaluation of distribution of carboxylate groups formed in the TEMPO-oxidized celluloses. The patterns of solid-state ¹³C-NMR spectra revealed that the oxidation occurred at the C6 primary hydroxyl groups of cellulose. X-ray diffraction and scanning electron microscopic analyses showed that such C6 oxidation took place at the surfaces of cellulose I crystallites without any oxidation at the C6 of inside cellulose I crystallites. Thus, carboxylate and aldehyde groups introduced into the TEMPO-oxidized celluloses are densely present on the surfaces of cellulose I crystallites. In addition, the obtained results revealed that the shoulder signal due to non-crystalline C6 carbons at about 63 ppm in solid-state ¹³C-NMR spectra of native celluloses is ascribed to those of surfaces of cellulose I crystallites or those of cellulose microfibrils.     

Correlation between X-ray diffraction measurements of cellulose crystalline structure and the susceptibility to microbial cellulase

Chemical and physical treatments of cotton cellulose have been studied in order to elucidate the relationship between the degree of crystallinity of cellulose and the susceptibility of cellulose to cellulase. Cotton cellulose powder was treated with the following solvents: 60% H₂SO₄, Cadoxen, and DMSO-p -formaldehyde. The dissolved celluloses were recovered at high yield of over 97% by addition of nine volumes of cold acetone. X-ray diffraction for measurements of relative crystallinity showed that the crystalline structure of cellulose declined in quantity and perfection by the dissolving treatment and changed to an amorphous form that is highly susceptible to enzymatic hydrolysis. These reprecipitated celluloses were hydrolyzed almost completely within 48 hr by Aspergillus niger cellulase containing mainly 1,4-β-glucan glucanohydrolase (EC 3.2.1.4), without action of 1,4-β-glucan cellobiohydrolase (EC 3.2.1. 91). On the other hand, cryo-milled cellulose (below 250 mesh) still had a crystalline structure, was resistant to cellulase, and gave a low percentage of saccharification. These results indicate that in pure cellulose there are good correlations between x-ray diffractograms and susceptibility to microbial cellulase.     

Hydrocelluloses with low degree of polymerisation from liquid ammonia treated cellulose

The heterogeneous hydrolytic degradation of cellulose after treatment with liquid ammonia has been studied. The level off degree of polymerisation (LODP) of liquid ammonia treated (LAT) linters is reached after 3 h when hydrolysed in hydrochloric acid (6.5 mol/l) at 60 °C. The hydrocelluloses were characterized as trimethylsilyl derivatives and as tricarbanilates. LODPs of non-activated celluloses were in the range from 55 to 77, while LAT celluloses had LODPs between 27 and 39. Trimethylsilyl derivatives and tricarbanilates gave almost identical elution curves in size exclusion chromatography indicating comparable hydrodynamic volumes. Glass transition temperatures of trimethylsilyl celluloses with DPs from 27 to 39 were found to be lower than those of the derivatives of the parent celluloses (Avicel, cotton linters) and showed a dependence on molar mass indicating that oligomeric celluloses are obtained by the method reported. Treatment of cellulose in aqueous ammonia was less efficient than liquid ammonia treatment.     

Hyperthermostable endoglucanase from Pyrococcus horikoshii.

An endoglucanase homolog from the hyperthermophilic archaeon Pyrococcus horikoshii was expressed in Escherichia coli, and its enzymatic characteristics were examined. The expressed protein was a hyperthermostable endoglucanase which hydrolyzes celluloses, including Avicel and carboxymethyl cellulose, as well as beta-glucose oligomers. This enzyme is the first endoglucanase belonging to glycosidase family 5 found from Pyrococcus species and is also the first hyperthermostable endoglucanase to which celluloses are the best substrates. This enzyme is expected to be useful for industrial hydrolysis of cellulose at high temperatures, particularly in biopolishing of cotton products.     

Effect of cellulose fine structure on kinetics of its digestion by mixed ruminal microorganisms in vitro

The digestion kinetics of a variety of pure celluloses were examined by using an in vitro assay employing mixed ruminal microflora and a modified detergent extraction procedure to recover residual cellulose. Digestion of all of the celluloses was described by a discontinuous first-order rate equation to yield digestion rate constants and discrete lag times. These kinetic parameters were compared with the relative crystallinity indices and estimated accessible surface areas of the celluloses. For type I celluloses having similar crystallinities and simple nonaggregating particle morphologies, the fermentation rate constants displayed a strong positive correlation (r2 = 0.978) with gross specific surface area; lag time exhibited a weaker, negative correlation (r2 = 0.930) with gross specific surface area. Crystallinity was shown to have a relatively minor effect on the digestion rate and lag time. Swelling of microcrystalline cellulose with 72 to 77% phosphoric acid yielded substrates which were fermented slightly more rapidly than the original material. However, treatment with higher concentrations of phosphoric acid resulted in a more slowly fermented substrate, despite a decrease in crystallinity and an increase in pore volume. This reduced fermentation rate was apparently due to the partial conversion of the cellulose from the type I to the type II allomorph, since mercerized (type II) cellulose was also fermented more slowly, and only after a much longer lag period. The results are consistent with earlier evidence for the cell-associated nature of cellulolytic enzymes of ruminal bacteria and suggest that ruminal microflora do not rapidly adapt to utilization of celluloses with altered unit cell structures.     

Effect of cellulose fine structure on kinetics of its digestion by mixed ruminal microorganisms in vitro.

The digestion kinetics of a variety of pure celluloses were examined by using an in vitro assay employing mixed ruminal microflora and a modified detergent extraction procedure to recover residual cellulose. Digestion of all of the celluloses was described by a discontinuous first-order rate equation to yield digestion rate constants and discrete lag times. These kinetic parameters were compared with the relative crystallinity indices and estimated accessible surface areas of the celluloses. For type I celluloses having similar crystallinities and simple nonaggregating particle morphologies, the fermentation rate constants displayed a strong positive correlation (r2 = 0.978) with gross specific surface area; lag time exhibited a weaker, negative correlation (r2 = 0.930) with gross specific surface area. Crystallinity was shown to have a relatively minor effect on the digestion rate and lag time. Swelling of microcrystalline cellulose with 72 to 77% phosphoric acid yielded substrates which were fermented slightly more rapidly than the original material. However, treatment with higher concentrations of phosphoric acid resulted in a more slowly fermented substrate, despite a decrease in crystallinity and an increase in pore volume. This reduced fermentation rate was apparently due to the partial conversion of the cellulose from the type I to the type II allomorph, since mercerized (type II) cellulose was also fermented more slowly, and only after a much longer lag period. The results are consistent with earlier evidence for the cell-associated nature of cellulolytic enzymes of ruminal bacteria and suggest that ruminal microflora do not rapidly adapt to utilization of celluloses with altered unit cell structures.     

Light-scattering analysis of native wood holocelluloses totally dissolved in LiCl-DMI solutions: high probability of branched structures in inherent cellulose

Three holocelluloses (i.e., cellulose and hemicellulose fractions) are prepared from softwood and hardwood by the Wise method. These holocelluloses completely dissolve in 8% lithium chloride/1,3-dimethyl-2-imidazolidinone (LiCl/DMI) after an ethylenediamine (EDA) pretreatment. After diluting the holocellulose solutions to 1% LiCl/DMI, they are subjected to size-exclusion chromatography/multiangle laser-light scattering/photodiode array (SEC-MALLS-PDA) analysis. All holocelluloses exhibit bimodal molecular weight distributions primarily due to high-molecular-weight (HMW) cellulose and low-molecular-weight hemicellulose fractions. Plots of molecular weight vs root-mean-square radius obtained by SEC-MALLS analysis revealed that all the wood celluloses comprise dense conformations in 1% LiCl/DMI. In contrast, bacterial cellulose, which was used as a pure cellulose model, has a random coil conformation as a linear polymer. These results show that both softwood and hardwood HMW celluloses contain branched structures, which are probably present on crystalline cellulose microfibril surfaces. These results are consistent with those obtained by permethylation analysis of wood celluloses.     

The binding of RNA to various celluloses

Several celluloses were tested for their ability to bind homonucleotide oligomers and natural RNAs. Columns of these celluloses were run under conditions favoring hydrogen bond formation (neutral pH and high salt concentration). The materials binding at the high salt were released by eluting the columns with solutions of lower ionic strength. The binding to unesterified cellulose was much less specific than that found with celluloses esterified with oligo(dT) or oligo(dC) residues. The different cellulose preparations were quite different in their ability to bind the oligonucleotides and the RNAs. These variations suggested that an impurity present in the cellulose in varying amounts rather than the cellulose itself, is responsible for the binding properties of these samples. Treatment of the celluloses with sodium bisulfite reduced the amount of poly(A) binding which suggests that the binding is due to a lignin-like contaminant.     

Crystalline index change in cellulose during aerobic and anaerobic Cellulomonas uda growth

Summary Cultures of Cellulomonas uda were monitored under both aerobic and anaerobic conditions using three commercially available celluloses with varying degrees of crystallinity. In all cases, a high level of cellulose was metabolized and the same maximum carboxymethylcellulase activity (2.6 IU/mg of cellular protein) was observed. Measurement of the crystalline index of celluloses during cellulose growth revealed that the amorphous and crystalline regions were solubilized simultaneously. Investigation of the solubilization rate showed that a decline occurred when a considerable amount of cellulose still remained in the medium. Hypotheses were suggested to explain the biphasic pattern of the kinetics obtained.