Molecular-Scale Investigations of Cellulose Microstructure during

Jun 29, 2010 - Kevin Conley , Louis Godbout , M.A. (Tony) Whitehead , Theo G.M. van ... Jeremy S. Luterbacher , Jose M. Moran-Mirabal , Eric W. Burkho...
0 downloads 0 Views 3MB Size
Download PDF
2000

Biomacromolecules 2010, 11, 2000–2007

Molecular-Scale Investigations of Cellulose Microstructure during Enzymatic Hydrolysis Monica Santa-Maria and Tina Jeoh* Biological and Agricultural Engineering, University of California, One Shields Avenue, Davis, California 95616 Received April 5, 2010; Revised Manuscript Received May 28, 2010

Changes in cellulose microstructure have been proposed to occur throughout hydrolysis that impact enzyme access and hydrolysis rates. However, there are very few direct observations of such changes in ongoing reactions. In this study, changes in the microstructure of cellulose are measured by simultaneous confocal and atomic force microscopy and are correlated to hydrolysis extents and quantities of bound enzyme in the reaction. Minimally processed and never-dried cellulose I was hydrolyzed by a purified cellobiohydrolase, Trichoderma reesei Cel7A. Early in the reaction (∼30% hydrolysis), at high hydrolysis rates and high bound cellulase quantities, untwisting of cellulose microfibrils was observed. As the hydrolysis reaction neared completion (>80% hydrolysis), extensively thinned microfibrils (diameters of 3-5 nm) and channels (0.3-0.6 nm deep) along the lengths of the microfibrils were observed. The prominent microstructural changes in cellulose due to cellobiohydrolase action are discussed in the context of the overall hydrolysis reaction.

Introduction Due to agronomic and environmental attributes, lignocellulosic biomass is viewed as a renewable feedstock for sustainable fuel production.1,2 However, the cost-intensive conversion of lignocellulose to fermentable sugars still limits its widespread utilization.1,2 One of the most costly steps in the biomass conversion process is the breakdown of structural polysaccharides of plant cell walls, such as cellulose, to sugars.2 To overcome this limitation, a fundamental understanding of the biomass deconstruction process at the molecular and chemical levels is necessary.3 One of the main limitations in cellulose depolymerization by cellulases is the accessibility of the β-1,4 glycosidic linkage. It is repeatedly observed in cellulose-cellulase reactions that hydrolysis proceeds at an initial fast rate followed by a rapid decrease in conversion rates at longer times.4,5 The mechanisms leading to this decrease are not fully understood and in some cases controversial.6,7 For instance, increased cellulose crystallinity and changes in the microstructure of cellulose during hydrolysis have been proposed to impact enzyme accessibility and cellulose digestibility.4,5,8 Furthermore, nonproductive adsorption and enzyme inactivation or inhibition, fractal and jamming effects,9 and diffusion constraints due to water structuring at the surface of the cellulose crystallite10 have also been proposed as limiting factors to cellulase action. In any case, the major factors that limit cellulase access to reactive sites in cellulose along with the mechanisms of processivity and enzyme synergism still need to be elucidated.3,6,7,11 High resolution techniques such as Raman spectroscopy, X-ray diffraction, electron microscopy (EM), and atomic force microscopy (AFM) have been used to study cell wall architecture and chemical composition,3,12 to elucidate the crystal structure and organization of cellulose microfibrils13,14 and to determine the effect of pretreatments on physicochemical properties of cellulosic substrates and enzyme accessibility.15,16 * To whom correspondence should be addressed. E-mail: tjeoh@ ucdavis.edu.

In contrast to other techniques, AFM imaging can be done under physiologically benign conditions in fluid with minimal alteration of the sample native state.17 In AFM, a sharp tip at the end of a flexible cantilever scans the sample in the X-Y direction, while interacting forces between the cantilever tip and the sample are recorded in the (vertical) Z-direction to give topographic details with subnanometer resolution.18 In this way, continuous monitoring of samples with minimal disruption of their native structure can be conducted to obtain temporal resolution of molecular-scale events. AFM has been used to study cellulose crystal structure and microfibril assembly,13,19,20 the mechanisms of cellulase action on cellulosic substrates,8,21-23 physicochemical and mechanical properties of cellulose surfaces and nanocomposites,24,25 and cell wall architecture and structure of pretreated lignocellulosic substrates.26,27 While relevant information has been obtained by AFM methods, sample preparation is still a major concern. Most AFM studies have relied on drying cellulosic samples onto the AFM substrate to promote attachment and resist lateral forces during scanning.19-21 However, drying cellulose results in irreversible aggregation of microfibrils, an effect known as hornification,28 which renders a substrate more recalcitrant to enzymatic depolymerization due to decreased enzyme accessibility.11,29,30 Interaction between cellulases and never dried cellulose microfibrils by AFM has been recently reported.22,23 For instance, Igarashi et al.22 investigated the mechanism of processivity of Trichoderma reesei Cel7A by high-speed AFM (HS-AFM) on Cladophora cellulose crystals. The authors measured a linear processivity rate of T. reesei Cel7A of about 3.5 nm/sec and demonstrated that both the hydrolysis of the glycosidic bond and loading of the cellulose chain into the active site tunnel of Cel7A are essential for the sliding movement. Similarly, Quirk et al.23 investigated the action of CenA, an endoglucanase from Cellulomonas fimi, on bacterial cellulose (BC). BC microfibrils were mechanically dispersed and attached to a thio-D-glucose functionalized gold-coated glass and changes in cellulose surface topography were monitored after adding enzyme.

10.1021/bm100366h  2010 American Chemical Society Published on Web 06/29/2010

Molecular-Scale Investigation of Cellulose Hydrolysis

In this work, we present a detailed study of the mechanisms of cellulose hydrolysis by cellulases by integrating AFM, confocal fluorescence microscopy and biochemical assays. Minimally processed and never-dried cellulose microfibrils are monitored for changes in cellulose micro- and nanostructure at various extents of hydrolysis during the reaction. Detailed descriptions of the method developed in this work along with observations of changes in cellulose microstructure at different stages in the hydrolysis reaction are presented herein.

Experimental Section Cellulose Sample. A single colony of Gluconacetobacter xylinum sbsp sucrofermentans (ATCC 700178) was grown in 3 mL of HestrinSchramm (HS) medium with 1% ethanol in static culture at 30 °C for 4 days.31,32 A total of 35 µL of grown broth were used to inoculate 500 µL of HS medium in 24-well plates. Static G. xylinum cultures were grown for 24 h at 30 °C to obtain thin and homogeneous cellulose films of 25 mm diameter cellulose films were washed overnight in 1% sodium hydroxide (NaOH) at 4 °C with agitation (350 rpm) followed by extensive rinsing with ultrapure water. Grafting of cellulose films with dichlorotriazinylaminofluorescein (DTAF; Sigma-Aldrich) was done as described previously.33 Films were placed in 1.5 mL of DTAF solution (0.6% DTAF in 0.2 N NaOH) and incubated overnight at room temperature with agitation (350 rpm). DTAF-grafted films were rinsed thoroughly with ultrapure water and put in 5 mM sodium acetate (pH 5) with 0.02% sodium azide. Cellulose microfibrils were dispersed by ultrasonication (Sonicator Model W-375, Heat Systems Ultrasonic, Inc.) in an ice-water bath. Cellulose concentration in sonicated samples was determined by the anthrone assay using a glucose standard curve.34 Aqueous samples (500 µL) and standards were cooled in an ice-water bath, mixed with 1 mL anthrone reagent (0.2% anthrone in concentrated sulfuric acid), and incubated for 10 min at 100 °C. Reactions were cooled down and absorbance was read at 630 nm in microtiter plates using a Synergy 4 microplate reader (BioTek). The difference in the reactivities of DTAF-grafted and ungrafted bacterial cellulose to T. reesei Cel7A were shown to be minimal (data are presented in Supporting Information). Cellulose from green algae, Cladophora sp, was also used in this study. Cladophora cellulose samples were kindly provided by Dr. Paul Langan from Los Alamos National Laboratory, prepared as previously described.35 DTAF grafting and dispersed suspensions of Cladophora cellulose were obtained as described above. T. reesei Cel7A Purification and Labeling. T. reesei Cel7A was purified from Celluclast (Novozymes) following a variation of previously described methods.11,36 The enzyme preparation, buffered in 20 mM Tris-HCl, pH 8.0, was fractionated by anion exchange chromatography on a ResourceQ column (GE Healthcare) by eluting with a linear gradient of 0-0.5 M sodium chloride in 20 mM Tris-HCl pH 8.0. The fractions with activity on p-nitrophenyl-β-D-lactoside (pNPL) were pooled, adjusted to 1.0 M ammonium sulfate, and loaded onto a phenyl sepharose column (HIC SPEC, GE Healthcare). Bound enzymes were eluted with a linear gradient of 0.8-0 M ammonium sulfate in 20 mM Tris-HCl, pH 8.0. The fractions with activity on pNPL were pooled and loaded onto a sizing column (Superdex 200, GE Healthcare) and eluted with 5 mM sodium acetate and 100 mM sodium chloride, pH 5. SDS-PAGE of the Cel7A fractions showed minor lower molecular weight bands, indicating the presence of impurities. Thus, the fractions were pooled, adjusted to 1 mM gluconolactone, and loaded onto a p-aminophenyl-β-D-cellobioside (pAPC) affinity column37 and eluted in 5 mM sodium acetate, 1 mM gluconolactone, and 0.01 M cellobiose, pH 5.0. The eluted fractions were pooled, concentrated by ultrafiltration with a 10000 MWCO spin concentrator (Amicon) and loaded again on the sizing column. The Cel7A fractions eluted in 5 mM sodium acetate, 100 mM sodium chloride, pH 5.0, were collected, concentrated, and stored for further use. A single band in the SDS-PAGE analysis at ∼62 kDa indicated purity of the final Cel7A preparation. The identity

Biomacromolecules, Vol. 11, No. 8, 2010

2001

of Cel7A was verified by LC/MS/MS at the proteomics facility at University of California, Davis (http://proteomics.ucdavis.edu/). Purified T. reesei Cel7A was labeled with Alexa Fluor 488 (AF488TrCel7A) and 594 (AF594-TrCel7A) fluorophores (Invitrogen, Inc.) according to procedures specified by the manufacturer. Both labels are functionalized with N-hydroxysuccinimidyl ester that reacts with the primary amines of lysine residues in the enzyme. Similarly labeled cellulases have previously been shown to retain its original activity on cellulose.38 This was reverified with each labeled batch. AFM Substrate and Sample Immobilization. Cover glass No. 1 (25 mm dia.), modified by self-assembly of aminopropyltriethoxysilane (APTES; Sigma-Aldrich) monolayers via chemical vapor deposition (CVD),39 was used as the AFM substrate. Glass discs were cleaned by successive immersions in acetone, isopropanol, and water, blow-drying with compressed nitrogen in-between washes.40 Further cleaning was done by the RCA method;41 glass discs were immersed in a H2O/H2O2 (30%)/NH4OH (29%) solution (5:1:1) at 70 °C for 15 min and then rinsed thoroughly with ultrapure water. CVD was done in a glass desiccator (150 mm i.d.) with moisture removed by oven drying. A total of 100 µL of APTES and 10 µL of N,N-diisopropylethyleneamine were placed at the bottom of the desiccator, and a clean cover glass was placed on a plastic tray 3 cm above. Silanization was done under an APTES atmosphere overnight. Silanized glass was allowed to stand under a nitrogen atmosphere for at least 24 h prior to use. For cellulose immobilization, silanized glass was assembled into a cover glass adapter (Asylum Research, part nos. 111.789 and 111.790) of the BioHeater closed fluid cell (Asylum Research) and 50 µL of sonicated cellulose suspension was spun down onto the silanized surface by centrifugation at a low speed (121 × g, 1 min). The excess volume was collected to estimate the mass of the nonadhered cellulose, and the amount of cellulose on the AFM substrate was estimated by subtraction. Samples were imaged immediately afterward under buffer. AFM Observations of Batch Reactions. Cellulose (5 µg/mL) was incubated with AF594-TrCel7A in 5 mM sodium acetate (pH 5) in 1.5 mL total volume at room temperature. The enzyme loading was 18.5 µmol per gram of cellulose. Reactions were stopped at different time intervals by filtration in 0.45 µm GHP multiwell filter plates (Pall Corp.). The flow-through was used to estimate the amount of cellobiose released in the reaction. Cellulose fibrils with bound AF594-TrCel7A retained in the filter were resuspended in 100 µL of buffer and used for measuring the amount of bound enzyme according to what was previously described.42 Fluorescence intensity of bound AF594TrCel7A was recorded and its concentration obtained from an AF594TrCel7A standard curve. Reactions were carried out in replicates (n ) 5). Cellulose fibrils from two replicates at each reaction time were pooled and adhered to silanized glass for AFM observations. Time-Lapse AFM. Changes in cellulose structure during an ongoing hydrolysis reaction were monitored by continuous “time-lapse” AFM scanning. A total of 250 µmol of purified T. reesei Cel7A were added per gram of cellulose in a 2 mL reaction. Incubation and imaging was done in 5 mM sodium acetate (pH5) in a closed cell assembly. In every experiment, cellulose samples were imaged several times before enzyme addition. After adding enzyme, the same cellulose fibrils were monitored for up to 16 h at room temperature. For each AFM experiment, two cellulose samples were similarly prepared, one of which was monitored by AFM and the other served as a replicate to measure the extent of hydrolysis. To stop the reaction, the reaction volume was collected and filtered in 0.45 µm GHP multiwell filter plates (Pall Corp.). The flow-through was used to estimate the amount of cellobiose released in the reaction. Cellobiose Quantification. Cellobiose concentration in the reaction supernatants was estimated by high performance anion exchange chromatography with pulsed amperometric detection (HPAE-PAD) in a Dionex BioLC system with a CarboPac PA1 analytical column (Dionex).43 Samples and cellobiose standards prepared in 5 mM sodium acetate (pH5) were diluted to 0.2-fold with ultrapure water. A total of 200 µL of sample was loaded onto the column and sugars were eluted

2002

Biomacromolecules, Vol. 11, No. 8, 2010

Santa-Maria and Jeoh

Figure 1. Schematic representation of a cellulose microfibril adhered onto the imaging surface by electrostatic interactions between negative charges in the DTAF-label of cellulose and positive charges on the APTES-modified glass surface.

with 20% NaOH 1 M isocratic flow at 1 mL/min. The cellobiose concentration in sample hydrolyzates was estimated from a cellobiose standard curve (n ) 3). The fraction of original cellulose that was converted to cellobiose was estimated. Mass gain due to the addition of water per hydrolyzed glycosidic bond was accounted for by multiplying cellobiose mass by a factor of 0.9. Simultaneous AFM and Confocal Microscopy. AFM and confocal microscopy of a sample preparation was done in a MFP3D BIO AFM system (Asylum Research) coupled to an inverted laser scanning confocal microscope (FV1000 system; Olympus America). AFM imaging was done in tapping mode with an AC240TS cantilever (force constant 2 N/m, tip radius < 10 nm; Asylum Research) in 5 mM sodium acetate (pH 5) using a closed cell assembly. Set point and integral gain were optimized for each scan. Scan rate was 1 Hz with 512 scan points and 512 scan lines. Amplitude and height AFM data are presented. Amplitude AFM data corresponds to changes in the amplitude of oscillation of the cantilever due to changes in surface topography. Height AFM data corresponds to the calibrated position of the cantilever with respect to a set point, which provides an accurate measure of sample dimensions in the vertical Z-direction. Images were not corrected for broadening effects, thus, lateral dimensions were not considered in the analysis. Images were analyzed using IGOR Pro 6.11 (Wavemetrics). Microfibril dimensions (height and length) of over 100 microfibrils from several independent scans were measured. 3D maps of surface topography were generated using IGOR Pro 6.11 software. Confocal imaging was done with a 60× objective. Excitation/emission spectra for the DTAF-, AF488-, and AF594-labels were 488/513, 495/519, and 590/617 nm, respectively.

Figure 2. DTAF-grafted bacterial cellulose microfibrils immobilized onto silanized glass are visualized by confocal and AFM. (A) Confocal image obtained with 488/513 nm excitation/emission spectra. (B) Same frame showing the colocation of the AFM cantilever and microfibrils in the fluorescence and transmitted-light channel (bar ) 5 µm). (C) AFM height data of the same cellulose sample (scan area ) 10 × 10 µm). (D) Higher resolution image of the same microfibril (area in box; scan area ) 1 × 1 µm).

Results Cellulose Sample Preparation and Imaging by AFM. An important requirement for AFM scanning is that samples are well attached to the imaging surface. The use of DTAF-grafted cellulose promoted effective adhesion of microfibrils onto silanized glass without a drying step such that the microfibrils were not removed while continuous scanning for up to 16 h under buffer. This was not the case when using nongrafted cellulose and silanized glass, or DTAF-cellulose on plain glass or mica surfaces. The effective adhesion of cellulose microfibrils is most likely due to electrostatic interactions between positive charges in the APTES-modified surface with negative charges in the DTAF label (Figure 1). On average, 8.6 ( 1.0 µg (n ) 9; ( standard error) corresponding to ∼67% of the cellulose mass originally added to the silanized glass was retained on the surface, with microfibrils generally aligned in the direction of the centripetal force applied as part of the adhesion protocol (Figure 2). In testing an alternate means of conducting the experiments, ungrafted bacterial cellulose microfibrils with bound AF488Cel7A were adhered to the APTES-modified surface of the imaging substrate (Figure 3). Here, adhesion of cellulose to the silanized surface was most likely due to interactions between negative charges in the bound enzyme and the silanized surface. This means of capturing the microfibrils was not as effective

Figure 3. Bacterial cellulose microfibrils with bound AF488-TrCel7A are observed by confocal and AFM. (A) Confocal image of cellulose microfibrils with bound enzyme obtained with 495/519 nm excitation/ emission spectra (bar ) 5 µm). Arrows indicate “hot spots” of enzyme binding identified as areas of higher fluorescence intensity. (B) 3D rendition of height AFM data of the same microfibrils showing irregular topography with protuberances of up to 50 nm (scan area ) 10 × 10 µm). Arrow indicates shadow of microfibril removed by interactions with the AFM tip.

as that between DTAF-grafted cellulose onto silanized glass as some microfibrils were removed during AFM scanning (Figure 3B). Some observations could be made of the microfibrils, notably, “hot spots” of enzyme binding were identified toward the microfibril ends as evidenced by higher fluorescence intensities (Figure 3A), and the surface topography of cellulose microfibrils with bound Cel7A appeared highly irregular, with pronounced protuberances more prominently than in the scans

Molecular-Scale Investigation of Cellulose Hydrolysis

Figure 4. 3D renditions of height AFM data showing bacterial cellulose microfibrils in a time-lapse experiment (scan area ) 1 × 1 µm). (A) Microfibril before adding enzyme (T. reesei Cel7A). (B) Same sample after enzyme addition. Enzymes are seen at the imaging surface as 3.2-3.8 nm height particles. (C) Enzymes are removed by lateral forces after a higher contact scan.

where no enzyme was added (Figure 3B). In general, cellulose microfibrils with bound Cel7A were around 25 nm in diameter with protuberances of up to 50 nm, possibly a result of overlapping microfibrils and the presence of bound enzyme. Ultimately, DTAF-grafted cellulose provided more reliable adhesion to the imaging substrate, thus, was used for all batch and time-lapse AFM experiments. An atomically flat surface is preferred for AFM imaging. Although glass is not as flat as mica or graphite, surface roughness was significantly reduced by the RCA1 treatment, minimizing topographic noise to within 1 nm (not shown). The RCA1 cleaning also seemed to improve silanization, however, the degree of silanization was most impacted by silanization time. Glass discs that were silanized for extended periods (>40 h) appeared to have a higher charge density as evidenced by a high proportion of enzyme bound to the surface (Figure 4B). This was not the case with shorter silanization times (80% hydrolysis), the particles do not appear to physically change. Although the particles were shown to be composed of

Santa-Maria and Jeoh

cellulose, the mechanism by which they are formed is not clear and we speculate that these are aggregates of cellulose microfibrils. Production of cellulose particles of macroscopic dimensions was first noted by Din et al.58 when treating cotton fibers with an endoglucanase (CenA). The authors hypothesized that the cellulose particles were generated as a consequence of the enzymatic hydrolysis of the fibers. In our study, cellulose particles were also observed in undigested controls, thus they are most likely a result of physical reaggregation of dispersed fibers after sonication. Irreversible aggregation of cellulose microfibrils has been shown to occur in the cell wall when noncellulose constituents such as hemicelluloses are removed from between microfibrils.59 Such aggregation results in samples with reduced accessible surface area and average pore diameter, increased cellulose crystallinity, and decreased digestibility by cellulases.30,60,61 Cellulose spheres of 40-60 nm diameter have also been observed in aerated cultures of G. xylinum growing on rice bark in contrast to static cultures of the same strain.62 The cellulose particles observed in this study could also be a result of degradation during extended sonication.63 For instance, the high energy imparted during ultrasonication treatment have been shown to reduce microfibril lengths in cellulose samples presumably due to scission of the β-D-(1f4) glycosidic linkages.64

Conclusions The kinetics of cellulose hydrolysis relies on effective binding of cellulases to cellulose. Understanding the fundamental kinetics of cellulose hydrolysis by cellulases necessitates accurate assessment of the changing nature of the substrate over the course of hydrolysis. The method we describe here integrates AFM, confocal microscopy, and biochemical assays such that measurements of microstructural changes in the cellulose can be directly correlated to the extents of bound enzyme and the extent of hydrolysis during the reaction. By this method, two important observations of cellulose microstructure are reported: (1) early in the reaction at high hydrolysis rates, the untwisting of cellulose microfibrils due to the action of T. reesei Cel7A is observed. Untwisting is not currently incorporated into a generalized hydrolysis mechanism by cellulases and further explorations are needed. (2) Thinning of cellulose microfibrils and the appearance of channels along the microfibril length are observed at late stages of the reaction (>80% hydrolysis), which support the processive action of the cellobiohydrolase on cellulose. These channels were not observed in the initial stages of the reaction and might have become prominent when more organized crystalline microfibrils remained after hydrolysis of less ordered cellulose. The implication of this structural change on enzyme accessibility and synergism will need to be evaluated further. In all, the method proved effective for investigating cellulose depolymerization and could be expanded to investigating more complex cellulosic substrates and reaction conditions. With leading lignocellulose pretreatment technologies relying on removing or displacing hemicellulose and lignin to increase enzyme digestibility, understanding the implications of the microstructure of cellulose that is free from other cell wall components on hydrolysis kinetics is critical. Acknowledgment. This work was supported in part by the Materials Design Institute, funded by the LANL/UC Davis Education Research Collaboration, Los Alamos National Laboratory (LANS Subcontract No. 75782-001-09); and the UC Lab

Molecular-Scale Investigation of Cellulose Hydrolysis

Fees Research ProgramsContingency Funds. This work was also supported by funds from the University of California Academic Senate. The authors thank Dr. John Labavitch for insightful discussions and for the use of his laboratory resources, Dr. Paul Langan (Los Alamos National Laboratory) and Dr. Masahisa Wada (Biomaterials Science Department, University of Tokyo, Japan) for providing samples of cellulose from Cladophora, and Dr. Yoshiharu Nishiyama on helpful insights into the effects of sonication on cellulose microfibrils. Supporting Information Available. Data on the enzymatic conversion of native versus DTAF-grafted bacterial cellulose, AFM data for undigested controls, and larger scans and height AFM data of the figures presented in this manuscript. This material is available free of charge via the Internet at http:// pubs.acs.org.

References and Notes (1) Himmel, M. E.; Ding, S.; Johnson, D. K.; Adney, W. S.; Nimlos, M. R.; Brady, J. W.; Foust, T. D. Science 2007, 5813, 804–807. (2) Lynd, L. R.; Laser, M. S.; Brandsby, D.; Dale, B. E.; Davison, B.; Hamilton, R.; Himmel, M.; Keller, M.; McMillan, J. D.; Sheehan, J.; Wyman, C. E. Nat. Biotechnol. 2008, 2, 169–172. (3) Ding, S. Y.; Himmel, M.; Xie, S. X. Microsc. Microanal. 2008, 14, 1494–1495. (4) Lynd, L. R.; Weimer, P. J.; van Zyl, W. H.; Pretorius, I. S. Microbiol. Mol. Biol. ReV. 2002, 66, 506–577. (5) Wang, L. S.; Zhang, Y. Z.; Gao, P. J.; Shi, D. X.; Liu, H. W.; Gao, H. J. Biotechnol. Bioeng. 2006, 3, 443–456. (6) Yang, B.; Willies, D. M.; Wyman, C. E. Biotechnol. Bioeng. 2006, 6, 1122–1128. (7) Hall, M.; Bansal, P.; Lee, J. H.; Realff, M. J.; Bommarius, A. S. FEBS J. 2010, 6, 1571–1582. (8) Ahola, S.; Turon, X.; Osterberg, M.; Laine, J.; Rojas, O. J. Langmuir 2008, 20, 11592–11599. (9) Xu, F.; Ding, H. Appl. Catal., A 2007, 1, 70–81. (10) Matthews, J. F.; Skopec, C. E.; Mason, P. E.; Zuccato, P.; Torget, R. W.; Sugiyama, J.; Himmel, M. E.; Brady, J. W. Carbohydr. Res. 2006, 1, 138–152. (11) Jeoh, T.; Ishizawa, C. I.; Davis, M. F.; Himmel, M. E.; Adney, W. S.; Johnson, D. K. Biotechnol. Bioeng. 2007, 1, 112–122. (12) Agarwal, U. P. Planta 2006, 5, 1141–1153. (13) Baker, A. A.; Helbert, W.; Sugiyama, J.; Miles, M. J. Biophys. J. 2000, 2, 1139–1145. (14) Nishiyama, Y. J. Wood Sci. 2009, 4, 241–249. (15) Kumar, R.; Mago, G.; Balan, V.; Wyman, C. E. Bioresour. Technol. 2009, 17, 3948–3962. (16) Donohoe, B. S.; Selig, M. J.; Viamajala, S.; Vinzant, T. B.; Adney, W. S.; Himmel, M. E. Biotechnol. Bioeng. 2009, 3, 480–489. (17) Ellis, J. S.; Allen, S.; Chim, Y. T. A.; Roberts, C. J.; Tendler, S. J. B.; Davies, M. C. AdV. Polym. Sci. 2006, 193, 123–172. (18) Seo, Y.; Jhe, W. Rep. Prog. Phys. 2008, 1, 016101. (23 pp.). (19) Yokota, S.; Ueno, T.; Kitaoka, T.; Wariishi, H. Carbohydr. Res. 2007, 17, 2593–2598. (20) Hirai, A.; Tsujii, Y.; Tsuji, M.; Horii, F. Biomacromolecules 2004, 6, 2079–2081. (21) Lee, I.; Evans, B. R.; Woodward, J. Ultramicroscopy 2000, 82, 213– 221. (22) Igarashi, K.; Koivula, A.; Wada, M.; Kimura, S.; Penttila, M.; Samejima, M. J. Biol. Chem. 2009, 52, 36186–36190. (23) Quirk, A.; Lipkowski, J.; Vandenende, C.; Cockburn, D.; Clarke, A. J.; Dutcher, J. R.; Roscoe, S. G. Langmuir 2010, 7, 5007–5013. (24) Bastidas, J. C.; Venditti, R.; Pawlak, J.; Gilbert, R.; Zauscher, S.; Kadla, J. F. Carbohydr. Polym. 2005, 4, 369–378. (25) Stiernstedt, J.; Nordgren, N.; Wagberg, L.; Brumer, H., III; Gray, D. G.; Rutland, M. W. J. Colloid Interface Sci. 2006, 1, 117–123. (26) Ding, S. Y.; Himmel, M. E. J. Agric. Food Chem. 2006, 3, 597–606.

Biomacromolecules, Vol. 11, No. 8, 2010

2007

(27) Eronen, P.; Osterberg, M.; Jaaskelainen, A. Cellulose 2009, 2, 167– 178. (28) Weise, U. Pap Ja Puu-Pap Timber 1998, 2, 110–115. (29) Esteghlalian, A. R.; Bilodeau, M.; Mansfield, S. D.; Saddler, J. N. Biotechnol. Prog. 2001, 6, 1049–1054. (30) Wan, J.; Wang, Y.; Xiao, Q. Bioresour. Technol. 2010, 12, 4577– 4583. (31) Schramm, M.; Hestrin, S. Biochem. J. 1954, 163–166. (32) Krystynowicz, A.; Czaja, W.; Wiktorowska-Jezierska, A.; GoncalvesMiskiewicz, M.; Turkiewicz, M.; Bielecki, S. J. Ind. Microbiol. Biotechnol. 2002, 4, 189–195. (33) Helbert, W.; Chanzy, H.; Husum, T. L.; Schulein, M.; Ernst, S. Biomacromolecules 2003, 3, 481–487. (34) Scott, T. A.; Melvin, E. H. Anal. Chem. 1953, 11, 1656–1661. (35) Igarashi, K.; Wada, M.; Hori, R.; Samejima, M. FEBS J. 2006, 13, 2869–2878. (36) Medve, J.; Lee, D.; Tjerneld, F. J. Chromatogr., A 1998, 1-2, 153– 165. (37) Sangseethong, K.; Penner, M. H. Carbohydr. Res. 1998, 3-4, 245– 250. (38) Jeoh, T.; Wilson, D. B.; Walker, L. P. Biotechnol. Prog. 2002, 4, 760– 769. (39) Dong, J.; Wang, A.; Ng, K. Y. S.; Mao, G. Thin Solid Films 2006, 4, 2116–2122. (40) Moran-Mirabal, J. N.; Santhanam, N.; Corgie, S. C.; Craighead, H. G.; Walker, L. P. Biotechnol. Bioeng. 2008, 6, 1129–1141. (41) Kern, W. J. Electrochem. Soc. 1990, 6, 1887–1892. (42) Davis, M.; Wolfrum, E.; Jeoh, T. In Genetic ImproVement of Bioenergy Crops; Vermerris, W., Ed.; Springer-Verlag: Berlin, Germany, 2008; p 143. (43) Jahnel, J. B.; Ilieva, P.; Frimmel, F. H. Fresenius’ J. Anal. Chem. 1998, 360, 827–829. (44) Putman, C. A. J.; Vanderwerf, K. O.; Degrooth, B. G.; Vanhulst, N. F.; Greve, J. Appl. Phys. Lett. 1994, 18, 2454–2456. (45) Haigler, C. H.; Malcolmbrown, R.; Benziman, M. Science 1980, 4472, 903–906. (46) El Kirat, K.; Burton, I.; Dupres, V.; Dufrene, Y. F. J. Microsc. 2005, 218, 199–207. (47) Baker, A. A.; Helbert, W.; Sugiyama, J.; Miles, M. J. J. Struct. Biol. 1997, 2, 129–138. (48) Hanley, S. J.; Revol, J. F.; Godbout, L.; Gray, D. G. Cellulose 1997, 3, 209–220. (49) Josefsson, P.; Henriksson, G.; Wagberg, L. Biomacromolecules 2008, 1, 249–254. (50) OSullivan, A. C. Cellulose 1997, 3, 173–207. (51) Boisset, C.; Fraschini, C.; Schulein, M.; Henrissat, B.; Chanzy, H. Appl. EnViron. Microbiol. 2000, 4, 1444–1452. (52) Hayashi, N.; Sugiyama, J.; Okano, T.; Ishihara, M. Carbohydr. Res. 1997, 1, 109–116. (53) Sugiyama, J.; Okano, T.; Yamamoto, H.; Horii, F. Macromolecules 1990, 12, 3196–3198. (54) Bowling, A. J.; Amano, Y.; Lindstrom, R.; Brown, R. M. Cellulose 2001, 1, 91–97. (55) Samejima, M.; Sugiyama, J.; Igarashi, K.; Eriksson, K. E. L. Carbohydr. Res. 1997, 2, 281–288. (56) Nishiyama, Y.; Johnson, G. P.; French, A. D.; Forsyth, V. T.; Langan, P. Biomacromolecules 2008, 11, 3133–3140. (57) French, A. D.; Johnson, G. P. Cellulose 2009, 6, 959–973. (58) Din, N.; Gilkes, N. R.; Tekant, B.; Miller, R. C.; Warren, A. J.; Kilburn, D. G. Bio/Technology 1991, 11, 1096–1099. (59) Mccann, M. C.; Wells, B.; Roberts, K. J. Cell Sci. 1990, 96, 323– 334. (60) Ishizawa, C. I.; Jeoh, T.; Adney, W. S.; Himmel, M. E.; Johnson, D. K.; Davis, M. F. Cellulose 2009, 4, 677–686. (61) Duchesne, I.; Hult, E.; Molin, U.; Daniel, G.; Iversen, T.; Lennholm, H. Cellulose 2001, 2, 103–111. (62) Goelzer, F. D. E.; Faria-Tischer, P. C. S.; Vitorino, J. C.; Sierakowski, M.; Tischer, C. A. Mater. Sci. Eng., C 2009, 2, 546–551. (63) Nishiyama, Y. personal communication, 2010. (64) Wong, S.; Kasapis, S.; Tan, Y. M. Carbohydr. Polym. 2009, 2, 280–287.

BM100366H