Page 1


METHODS

IN

MOLECULAR BIOLOGYâ„¢

Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK

For further volumes: http://www.springer.com/series/7651


Biomass Conversion Methods and Protocols Edited by

Michael E. Himmel Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA


Editor Michael E. Himmel Biosciences Center National Renewable Energy Laboratory Golden, CO, USA

ISSN 1064-3745 ISSN 1940-6029 (electronic) ISBN 978-1-61779-955-6 ISBN 978-1-61779-956-3 (eBook) DOI 10.1007/978-1-61779-956-3 Springer New York Heidelberg Dordrecht London Library of Congress Control Number: 2012942183 Š Springer Science+Business Media, LLC 2012 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution under the respective Copyright Law. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Humana Press is a brand of Springer Springer is part of Springer Science+Business Media (www.springer.com)


Dedication To my wife of 40 years, Rita Jeanette, who passed away during the compilation of this book


Preface This book highlights methodological challenges encountered in pursuit of research critical for developing cost-effective technologies for conversion of biomass to transportation fuels and chemicals. The book begins with an overview chapter discussing the special needs and pitfalls encountered in conducting original biomass conversion research. In brief, biomass conversion research is a combination of basic science, applied science, and engineering testing and analysis. To conduct this work effectively, it is important to understand the essential characteristics of the feedstock considered for conversion. Conversion science includes the initial treatment (called pretreatment) of the feedstock to render it more amenable to enzyme action as well as the saccharification and chemical/biochemical conversion steps that follow. Based on this view, the book is divided into three high-level topics which include Biomass Feedstocks and Cellulose, Plant Cell Wall-Degrading Enzymes and Microorganisms, and Lignin and Hemicelluloses. Each chapter within these general themes stresses the state of the problem as well as methodological challenges encountered in conducting research and ways to improve existing methods or the introduction of new methods. It is our intention that this book will benefit both scientists working to understand the fundamental problems associated with biomass conversion research and chemical and mechanical engineers working to design new conversion processes. The conceptualization and construction of this book was funded by the BioEnergy Science Center (BESC), which is a U.S. Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science as well as the U.S. Department of Energy Office of the Biomass Program (OBP). Golden, CO, USA

Michael E. Himmel

vii


Contents Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Challenges for Assessing the Performance of Biomass Degrading Biocatalysts . . . . . Michael E. Himmel, Stephen R. Decker, and David K. Johnson

PART I 2 3 4

5 6

7

8

1

BIOMASS FEEDSTOCK AND CELLULOSE

Overview of Computer Modeling of Cellulose . . . . . . . . . . . . . . . . . . . . . . . . . . . . Malin BergenstrĂĽhle-Wohlert and John W. Brady Imaging Cellulose Using Atomic Force Microscopy . . . . . . . . . . . . . . . . . . . . . . . . Shi-You Ding and Yu-San Liu Preservation and Preparation of Lignocellulosic Biomass Samples for Multi-scale Microscopy Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bryon S. Donohoe, Peter N. Ciesielski, and Todd B. Vinzant Coherent Raman Microscopy Analysis of Plant Cell Walls . . . . . . . . . . . . . . . . . . . . Yining Zeng, Michael E. Himmel, and Shi-You Ding Immunological Approaches to Plant Cell Wall and Biomass Characterization: Glycome Profiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sivakumar Pattathil, Utku Avci, Jeffrey S. Miller, and Michael G. Hahn Immunological Approaches to Plant Cell Wall and Biomass Characterization: Immunolocalization of Glycan Epitopes. . . . . . . . . . . . . . . . . . . . Utku Avci, Sivakumar Pattathil, and Michael G. Hahn A Method to Evaluate Biomass Accessibility in Wet State Based on Thermoporometry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bon-Wook Koo and Sunkyu Park

PART II

vii xi

11 23

31 49

61

73

83

PLANT CELL WALL

9

Cellulase Processivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . David B. Wilson and Maxim Kostylev 10 A Simple Method for Determining Specificity of Carbohydrate-Binding Modules for Purified and Crude Insoluble Polysaccharide Substrates. . . . . . . . . . . . Oren Yaniv, Sadanari Jindou, Felix Frolow, Raphael Lamed, and Edward A. Bayer 11 Bacterial Cadherin Domains as Carbohydrate Binding Modules: Determination of Affinity Constants to Insoluble Complex Polysaccharides. . . . . . . Milana Fraiberg, Ilya Borovok, Ronald M. Weiner, Raphael Lamed, and Edward A. Bayer

ix

93

101

109


x

12

13

14

15

16

17

18

Contents

Affinity Electrophoresis as a Method for Determining Substrate-Binding Specificity of Carbohydrate-Active Enzymes for Soluble Polysaccharides . . . . . . . . . 119 Sarah Moraïs, Raphael Lamed, and Edward A. Bayer Single-Molecule Tracking of Carbohydrate-Binding Modules on Cellulose Using Fluorescence Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 Yu-San Liu, Shi-You Ding, and Michael E. Himmel Bioprospecting Metagenomics for New Glycoside Hydrolases. . . . . . . . . . . . . . . . . 141 Jack Gilbert, Luen-Luen Li, Safiyh Taghavi, Sean M. McCorkle, Susannah Tringe, and Daniel van der Lelie Anaerobic High-Throughput Cultivation Method for Isolation of Thermophiles Using Biomass-Derived Substrates . . . . . . . . . . . . . . . . . . . . . . . . 153 Scott D. Hamilton-Brehm, Tatiana A. Vishnivetskaya, Steve L. Allman, Jonathan R. Mielenz, and James G. Elkins Assessing the Protein Concentration in Commercial Enzyme Preparations . . . . . . . 169 William S. Adney, Nancy Dowe, Edward W. Jennings, Ali Mohagheghi, John Yarbrough, and James D. McMillan Reducing the Effect of Variable Starch Levels in Biomass Recalcitrance Screening. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Stephen R. Decker, Melissa Carlile, Michael J. Selig, Crissa Doeppke, Mark Davis, Robert Sykes, Geoffrey Turner, and Angela Ziebell Analysis of Transgenic Glycoside Hydrolases Expressed in Plants: T. reesei CBH I and A. cellulolyticus EI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Roman Brunecky, John O. Baker, Hui Wei, Larry E. Taylor, Michael E. Himmel, and Stephen R. Decker

PART III

LIGNINS AND HEMICELLULOSES

19

Structural Characterization of the Heteroxylans from Poplar and Switchgrass . . . . . 215 Koushik Mazumder, Maria J. Peña, Malcolm A. O’Neill, and William S. York 20 Laser Microdissection and Genetic Manipulation Technologies to Probe Lignin Heterogeneity and Configuration in Plant Cell Walls . . . . . . . . . . . 229 Oliver R.A. Corea, Chanyoung Ki, Claudia L. Cardenas, Laurence B. Davin, and Norman G. Lewis 21 Lignin-Degrading Enzyme Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 Yi-ru Chen, Simo Sarkanen, and Yun-Yan Wang Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

269


Contributors WILLIAM S. ADNEY • Research Triangle Institute, International Research Triangle Park, NC, USA STEVE L. ALLMAN • BioEnergy Science Center, Oak Ridge National Laboratory, Oak Ridge, TN, USA UTKU AVCI • BioEnergy Science Center Complex Carbohydrate Research Center, The University of Georgia, Athens, GA, USA JOHN O. BAKER • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA EDWARD A. BAYER • Department of Biological Chemistry, The Weizmann Institute of Science, Rehovot, Israel MALIN BERGENSTRÅHLE-WOHLERT • Department of Food Science, Stocking Hall, Cornell University, Ithaca, NY, USA ILYA BOROVOK • Department of Molecular Microbiology and Biotechnology, The Daniella Rich Institute for Structural Biology, Tel Aviv University, Ramat Aviv, Israel JOHN W. BRADY • Department of Food Science, Cornell University, Ithaca, NY, USA ROMAN BRUNECKY • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA CLAUDIA L. CARDENAS • Institute of Biological Chemistry, Washington State University, Pullman, WA, USA MELISSA CARLILE • Bioenergy Center, National Renewable Energy Laboratory, Golden, CO, USA YI-RU CHEN • Department of Bioproducts and Biosystems Engineering, University of Minnesota, St. Paul, MN, USA PETER N. CIESIELSKI • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA OLIVER R.A. COREA • Institute of Biological Chemistry, Washington State University, Pullman, WA, USA LAURENCE B. DAVIN • Institute of Biological Chemistry, Washington State University, Pullman, WA, USA MARK DAVIS • Bioenergy Center, National Renewable Energy Laboratory, Golden, CO, USA STEPHEN R. DECKER • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA SHI-YOU DING • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA CRISSA DOEPPKE • Bioenergy Center, National Renewable Energy Laboratory, Golden, CO, USA

xi


xii

Contributors

BRYON S. DONOHOE • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA NANCY DOWE • National Bioenergy Center, National Renewable Energy Laboratory, Golden, CO, USA JAMES G. ELKINS • BioEnergy Science Center, Oak Ridge National Laboratory, Oak Ridge, TN, USA MILANA FRAIBERG • Department of Molecular Microbiology and Biotechnology, The Daniella Rich Institute for Structural Biology, Tel Aviv University, Ramat Aviv, Israel FELIX FROLOW • Department of Molecular Microbiology and Biotechnology, The Daniella Rich Institute for Structural Biology, Tel Aviv University, Ramat Aviv, Israel JACK GILBERT • Department of Ecology & Evolution, University of Chicago, Chicago, IL, USA MICHAEL G. HAHN • BioEnergy Science Center, Complex Carbohydrate Research Center, The University of Georgia, Athens, GA, USA; Department of Plant Biology, University of Georgia, Athens, GA, USA SCOTT D. HAMILTON-BREHM • BioEnergy Science Center, Oak Ridge National Laboratory, Oak Ridge, TN, USA MICHAEL E. HIMMEL • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA EDWARD W. JENNINGS • National Bioenergy Center, National Renewable Energy Laboratory, Golden, CO, USA SADANARI JINDOU • Faculty of Science and Technology , Meijo University, Shiogamaguchi, Tempaku, Nagoya, Japan DAVID K. JOHNSON • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA CHANYOUNG KI • Institute of Biological Chemistry, Washington State University, Pullman, WA, USA BON-WOOK KOO • Department of Forest Biomaterials, North Carolina State University, Raleigh, NC, USA MAXIM KOSTYLEV • Department of Molecular Biology & Genetics, Cornell University, Ithaca, NY, USA RAPHAEL LAMED • Department of Molecular Microbiology and Biotechnology, The Daniella Rich Institute for Structural Biology, Tel Aviv University, Ramat Aviv, Israel NORMAN G. LEWIS • Institute of Biological Chemistry, Washington State University, Pullman, WA, USA LUEN-LUEN LI • Biology Department, Brookhaven National Laboratory, Upton, NY, USA YU-SAN LIU • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA KOUSHIK MAZUMDER • Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA SEAN M. MCCORKLE • Biology Department, Brookhaven National Laboratory, Upton, NY, USA


Contributors

xiii

JAMES D. MCMILLAN • National Bioenergy Center, National Renewable Energy Laboratory, Golden, CO, USA JONATHAN R. MIELENZ • BioEnergy Science Center Oak Ridge National Laboratory, Oak Ridge, TN, USA JEFFREY S. MILLER • BioEnergy Science Center, Complex Carbohydrate Research Center, The University of Georgia, Athens, GA, USA ALI MOHAGHEGHI • National Bioenergy Center, National Renewable Energy Laboratory, Golden, CO, USA SARAH MORAÏS • Department of Biological Chemistry, The Weizmann Institute of Science, Rehovot, Israel MALCOLM A. O’NEILL • Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA SUNKYU PARK • Department of Forest Biomaterials, North Carolina State University, Raleigh, NC, USA SIVAKUMAR PATTATHIL • BioEnergy Science Center, Complex Carbohydrate Research Center, The University of Georgia, Athens, GA, USA MARIA J. PEÑA • Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA SIMO SARKANEN • Department of Bioproducts and Biosystems Engineering, University of Minnesota, St. Paul, MN, USA MICHAEL J. SELIG • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA ROBERT SYKES • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA SAFIYH TAGHAVI • Center for Agriculture and Environmental Biotechnology, RTI International, Research Triangle Park, NC, USA LARRY E. TAYLOR • Biosciences Center National Renewable Energy Laboratory, Golden, CO, USA SUSANNAH TRINGE • DOE Joint Genome Institute, Walnut Creek, CA, USA GEOFFREY TURNER • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA DANIEL vAN DER LELIE • Center for Agriculture and Environmental Biotechnology, RTI International, Research Triangle Park, NC, USA TODD B. VINZANT • Biosciences Center National Renewable Energy Laboratory, Golden, CO, USA TATIANA A. VISHNIVETSKAYA • BioEnergy Science Center, Oak Ridge National Laboratory, Oak Ridge, TN, USA YUN-YAN WANG • Department of Bioproducts and Biosystems Engineering, University of Minnesota, St. Paul, MN, USA HUI WEI • Biosciences Center, National Renewable Energy Laboratory, Golden, CO, USA RONALD M. WEINER • Department of Cell Biology and Molecular Genetics, University of Maryland, College Park, MD, USA DAVID B. WILSON • Department of Molecular Biology & Genetics, Cornell University, Ithaca, NY, USA


xiv

Contributors

OREN YANIV • Department of Molecular Microbiology and Biotechnology, The Daniella Rich Institute for Structural Biology, Tel Aviv University, Ramat Aviv, Israel JOHN YARBROUGH • Biosciences Center National Renewable Energy Laboratory, Golden, CO, USA WILLIAM S. YORK • Complex Carbohydrate Research Center, University of Georgia, Athens, GA, USA YINING ZENG • BioEnergy Science Center, Biosciences Center National Renewable Energy Laboratory, Golden, CO, USA ANGELA ZIEBELL • Bioenergy Center, National Renewable Energy Laboratory, Golden, CO, USA


Chapter 1 Challenges for Assessing the Performance of Biomass Degrading Biocatalysts Michael E. Himmel, Stephen R. Decker, and David K. Johnson Abstract Common analytical challenges impact current work to estimate the cost of converting plant biomass to fermentable sugars. The most noteworthy are measuring cellulase and hemicellulase activities, cellulase and hemicellulase protein, biomass compositions (before and after pretreatment), and the products formed. The use of high-throughput (HTP) methods has shown considerable promise for improving both analytical precision and technician efficiency, but can also present pitfalls regarding experimental accuracy and relevance. Recent work demonstrates that HTP systems which include biomass composition analysis, thermal chemical pretreatment, and biomass saccharification can be realized. Key words: Biomass, Cellulase activity, Protein determination, High-throughput screening, Cellulose crystallinity

1. Introduction Plant cell walls are known to be recalcitrant to natural processes of deconstruction, both enzymatic and microbial (1, 2). Cost-effective conversions are critical to enable the emerging biomass to renewable fuels and chemical industries. Because measuring performance of biocatalysts acting on biomass in a consistent and relevant manner is an obvious metric for producing competent process models, these challenges are examined in detail.

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_1, Š Springer Science+Business Media, LLC 2012

1


2

M.E. Himmel et al.

2. Problems Most Often Encountered 2.1. Cellulase Enzyme Performance

There are several aspects of measuring cellulase action that routinely emerge as being problematic. These are measuring protein concentration, use of relevant substrates, choice of saccharification conditions, enzyme inhibition and loss, and conducting hydrolysis to a meaningful level of conversion. Various attempts have been made to determine the protein measurement assay best suited to the direct comparison of diverse cellulase preparations (3, 4). Because these enzyme preparations are often contaminated with mono and polysaccharides, the dye binding assays (i.e., Bradley assay) do not function adequately (5). The bicinchoninic acid assay (BCA) appears to function well for routine enzyme measurement and does not require the more severe color forming steps required by the more traditional assay for proteins, such as the biuret assay. However, if a single recombinant enzyme is used, then using calculated or experimentally determined molar extinction coefficients works well. For problematic samples, protein estimation by amino acid analysis can be used to standardize a classical assay (5). It is important to use substrates for glycosyl hydrolases (GH) that are appropriate for the desired outcome. Enzymes that act upon polymers, especially those applied to plant cell walls, must ultimately be assessed on relevant substrates. Although surrogate, soluble substrates can be useful for comparing activities within a closed set of similar enzymes; to be process relevant, testing for performance on polysaccharides must eventually be done. For cellulases, this consideration is especially important. Useful surrogate substrates include various methylumbelliferyl (MU)- and paranitrophenyl (PNP)-glycosides which generate fluorophoric and colorimetric species, respectively, upon hydrolysis. For example, MU-lactoside and PNP-glucoside are often used for quick measurements of the performance of cellobiohydrolases (GH families 6 and 7) and β-D-glucosidases (GH families 1–3), respectively. Cellulases must be evaluated for performance on cellulose of known history and allomorphic composition. For example, Avicel, a commonly used cellulose substrate for testing enzymes, is known to be approximately 20–40% amorphous cellulose (6, 7). Because amorphous cellulose is hydrolyzed more readily by cellulases than crystalline cellulose, conducting conversion to a level greater than the amorphous content is critical if translation of this performance to an industrial plant cell wall application is anticipated. Another source of error in evaluating cellulase performance can come from the conditions selected for performing the digestions. Ideally a cellulase would be evaluated using an enzyme loading study. A much clearer picture of cellulase activity can be obtained by performing digestions at several loading levels of


1

Challenges for Assessing the Performance of Biomass Degrading Biocatalysts

3

enzyme to cellulose. In addition, digestion conditions can have a significant impact on enzyme performance, particularly cellulose concentration and temperature. Quite different performance metrics can be obtained at cellulose loadings of 1–4% compared to more commercially relevant loadings of 15–20%. The standard temperature for performing digestions is 50°C; however, it may be easier to distinguish differences in activity between enzymes if digestions are slowed somewhat by operating at slightly lower temperatures, such as 45°C. Conversely, with the development of more thermally tolerant enzymes, it may be better to explore higher temperatures to determine if higher conversion rates are possible. The conclusion we reach is that for an assay of cellulase performance to be commercially relevant (i.e., to inform process design), it must be tested under conditions that mimic the actual use condition as closely as possible. Surrogate tests are not normally useful in this regard. It is also necessary to be aware that enzymes can be inhibited by various components that can be produced during an activity assay. Enzymes can often be inhibited by their products, glucose and cellobiose, that are formed during the digestion of cellulose. In addition, if digestions are performed on lignocellulosic substrates there are other components that can be present in the substrate or that are released during the digestion that may inhibit enzymes. There is recent evidence that the extent of inhibition varies by enzyme and the inhibitory component (8). It is also well known that enzymes will bind to cellulose and other components of lignocellulose, especially lignin, nonhydrolytically. When digesting lignocellulosic substrates it is likely that a fraction of the loaded enzymes will be lost through what is termed “nonspecific binding” with the substrate, and it is possible that some enzymes may be affected to a greater degree than other enzymes. 2.2. Product Measurement

For the case of measuring the hydrolysis of natural substrates, detecting and quantifying the release of mono-, di-, and oligosaccharides is needed. This goal can be accomplished using HPLC, TLC, or enzyme linked assays (ELA). Examples of the latter include the Glucose Oxidase/Peroxidase (GOPOD) assay for glucose and the xylose dehydrogenase (XDH) assay for xylose. ELAs are also amenable to high-throughput (robotics) methodologies which are useful for the interrogation of large sample sets; however, the precision is not comparable with HPLC techniques and can be affected by the presence of oligomers. Recent advances in columns and detectors have increased the options available for analysis of mono-, di- and oligosaccharide products from cellulase digestions. Improved HPLC detectors relative to the refractive index detector, the workhorse of carbohydrate analysis, include the pulsed amperometric detector (9), evaporative light scattering detector (10), and the charged aerosol detector (11). Advantages of these detectors include much lower


4

M.E. Himmel et al.

detection limits (into the ng injected range) and the possibility of using nonisocratic elution. Use of anion chromatography also allows the baseline separation of monosaccharides (12) and the selective analysis of oligosaccharides by varying the alkalinity of the mobile phase as well as the choice of column (13). Silica based amino columns can also used for separating monosaccharides and oligosaccharides by selection of the column and mobile phase (14). New ultrahigh pressure LC columns and systems can also be used to obtain chromatographic separations in minutes, instead of hours as was the historical performance of traditional systems. 2.3. Substrate Particle Size and History

Insoluble substrates must be milled to sizes amenable to laboratory scale assays. Most work in biomass conversion science utilizes biomass processed through knife type mills fitted with 1–2 mm rejection screens which generates material in the 30–60 mesh size range. It is clear that enzymatic and microbial conversion of biomass particles proceeds more rapidly as the particle size is reduced (15). This is assumed to be a consequence of the basic process itself, which is a surface active (or ablative) phenomenon. This means that to be meaningful, digestions of biomass must be compared only in cases where the particle sizes used were similar. Of course, eventual translation to full-scale industrial process application must be evaluated for each particular case. Substrate history is another important consideration. It is now known that the preparation of cellulose can affect its convertibility, especially if heating, drying, or oxidative (bleaching) procedures are applied. It is normally more informative to test biocatalysts against both processed cellulose, such as Avicel, SigmaCel, Solka Floc; as well as thermal chemically treated, never dried plant cell wall fractions to generate the most informative digestion results. If microbial cellulose is used, it should be prepared with washing and extraction procedures that come from published or standardized methods. In this case, it is important to note that bacterial cellulose is primarily of the Iι allomorph and of average crystal sizes that are much greater than the microfibrils found in plant cell wall cellulose. We now suspect that these differences are important when considering comparing enzyme performance on different celluloses. Commercial celluloses are prepared so that they are very stable and have a long shelf life. When selecting the substrate upon which a cellulase will be tested, it is important to be aware of the cellulose history. Microcrystalline celluloses are prepared by acid hydrolysis which removes much of the amorphous cellulose and decreases the molecular weight of the remaining cellulose. The saccharification of Avicel slows dramatically after 60% conversion compared to other celluloses, which is probably due to the method of its preparation. Cellulose isolated from corn stover was found to be completely digested in 72 h under the same saccharification conditions (16).


1

Challenges for Assessing the Performance of Biomass Degrading Biocatalysts

3. HighThroughput Solutions

3.1. Benefits

5

High-throughput (HTP) methods can be used to assess the performance of libraries of enzymes acting on single substrates or defined sets of enzymes acting on libraries of substrates with equivalent effectiveness. The overarching truth of any HTP method is that you get what you screen for. While many enzymes involved in biomass hydrolysis have activity on simple, fluorometric or colorimetric substrate analogues, the intrinsic activities are not the same between these artificial compounds and natural (or preprocessed) biomass and care must be take in extending kinetic or structural data from these assays to activity on real-world biomass substrates. The advantage of the analogues lies in their ease of use and assaying. For applications requiring more biomass-specific results, the best substrate to use is obviously one as close as possible to the biomass in question, whether that is native or pretreated. While automated handling of enzymes is simple and straightforward using modern liquid-handling robots and analytical instruments designed to work with standardized microtiter plates, handling the substrate (i.e., biomass) using similar platforms is fraught with obstacles and only recently have researchers overcome these using a variety of unique approaches and modifications to standard laboratory automation. There are numerous obstacles involved in obtaining meaningful results when carrying out high-throughput campaigns using biomass and these have been addressed in several recently (17–22). Briefly, these include the heterogeneity of the biomass, accurate dispensing of solids (whether dry or in slurries), static control (in dry material), extractives and starch removal (which can skew sugar assay results), even heating and pressure containment (for thermochemical pretreatment), post-pretreatment hydrolysate neutralization (for dilute acid pretreatment), evaporation control during multiday enzyme digestion, compositional analysis of large sample numbers (if percent yield is desired), and sugar analysis. While any of these barriers are easy to address for small sample numbers at the bench scale, addressing these issues for thousands of samples is far from trivial. The advantages of high-throughout screening are mostly obvious, in that large numbers of samples, as high as 1,000 per day, can be compared directly to each other for any given assay (19, 20). Multiple enzyme components can easily be mixed and matched at different ratios to determine the most effective combinations of minimum loadings required for a given substrate. Temperature and pH optima and ranges can be quickly determined for given enzymes, as well as substrate specificity or preference. Whereas classical biochemical techniques have been used for these kinds of


6

M.E. Himmel et al.

studies for decades, HTP methods have made these assays much faster and easier to carry out with decreased potential for operator error; however, there are drawbacks and limitations to these techniques that cannot be ignored. 3.2. Drawbacks

The largest and most obvious drawback to HTP methodology is the need for laboratory automation platforms. In general, lab robots are large, expensive, unfamiliar to most researchers, and often high-maintenance ones. Their limitations must be carefully considered before their applicability to any specific assay can be evaluated. In general, separate solid and liquid handlers are far more common than integrated systems. Most research laboratories that do not have HTP as a core business simply do not have the expertise or resources to integrate multiple stations into a unified onebutton-pusher platform. In most cases, solids dispensing, liquid handling, plate sealing, incubation, and microtiter plate reading are handled separately, requiring manual transfer steps between instruments. In addition, data handling for these multiple systems requires extensive LIMS capability which is not present in many research labs. Many manufacturers offer completely integrated systems; however, these tend to be either very specialized or extremely complex and are outside the range of most research facilities. This is primarily because these platforms are driven by the pharmaceutical industry, which is highly regimented and resource rich. In the past decade or so, many of the automation manufacturers have offered more flexible, smaller systems that are of great use to general research labs, specifically those involved in molecular biology; however, these systems have not been designed to address the needs of biomass conversion research. Thus, issues such as the handling of biomass, evaporation control for extended enzyme digestions, high temperature pretreatments, handling of acids, and dealing with solutions containing high levels of particulates have not be solved. Other drawbacks impact the ways data are interpreted. In HTP screening, the results obtained are usually comparative, rather than quantitative. While many methods exist which control and standardize experiments, it is difficult to maintain absolute control over every aspect of such complicated screens. Dealing with very small quantities of liquids and solids means that inaccuracies of only a few micrograms or microliters can translate to a much larger cumulative errors than those common to bench-scale techniques. This is further compounded by the use of highly concentrated enzyme and other stock solutions in order to minimize the volumes of the plate based reactions. Another critical challenge is the need to use finely ground samples to test in microtiter plates. As described above, this need necessitates considerable work to extrapolate automated results to the eventual application of enzymes and substrates to industrial processes where the substrates are tree chips or fragments of grass stems.


1

Challenges for Assessing the Performance of Biomass Degrading Biocatalysts

7

Acknowledgments This work was supported by the US Department of Energy Office of the Biomass Program; the Center for direct Catalytic Conversion of Biomass to Biofuels (C3Bio), an Energy Frontier Research Center funded by the DOE Office of Science, Office of Basic Energy Sciences; and the BioEnergy Science Center (BESC) Oak Ridge National Laboratory, which is a US DOE Bioenergy Research Center supported by the Office of Science, Office of Biological and Environmental Research. References 1. Himmel ME, Ding S-Y, Johnson DK, Adney WS, Nimlos MR, Brady JW, Foust TD (2007) Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science 315(5813):804–807 2. Lynd LR, Laser MS, Bransby D, Dale BE, Davison B, Hamilton R, Himmel ME, Keller M, McMillan JD, Sheehan J, Wyman CE (2008) How biotech can transform biofuels. Nat Biotechnol 26:169–172 3. Nieves RA, Ehrman CI, Adney WS, Thomas SR, Himmel ME (1998) Survey and analysis of commercial cellulase preparations suitable for biomass conversion to ethanol. World J Microbiol Biotechnol 14:301–304 4. Adney WS, Ehrman CI, Baker JO, Thomas SR, Himmel ME (1994) Cellulase assays: methods from empirical mathematical models. In: Himmel ME, Baker JO, Overend RP (eds) Enzymatic conversion of biomass for fuels production, ACS Book Series 566. American Chemical Society, Washington, DC, pp 218–235 5. Adney WS, Mohagheghi A, Thomas SR, Himmel ME (1995) Comparison of protein contents of cellulase preparations in a worldwide round-robin assay. In: Saddler JN, Penner MH (eds) Enzymatic degradation of insoluble polysaccharides, ACS Book Series 618. American Chemical Society, Washington, DC, pp 256–271 6. Park S, Johnson DK, Ishizawa CI, Parilla PA, Davis MF (2009) Measuring the crystallinity index of cellulose by solid state C-13 nuclear magnetic resonance. Cellulose 16:641–647 7. Park S, Baker JO, Himmel ME, Parilla PA, Johnson DK (2010) Cellulose crystallinity index: measurement techniques and their impact on interpreting cellulase performance. Biotechnol Biofuels 3:10

8. Qin Q, Yang B, Wyman CE (2010) Xylooligomers are strong inhibitors of cellulose hydrolysis by enzymes. Bioresour Technol 101:9624–9630 9. Steinbach A, Wille A (2009) Determining carbohydrates in essential and non-essential foodstuffs using ion chromatography. LC-GC Europe 3(7):15–18 10. Filson PB, Dawson-Andoh BE (2009) Characterization of sugars from model and enzyme-mediated pulp hydrolyzates using high-performance liquid chromatography coupled to evaporative light scattering detection. Bioresour Technol 100:6661–6664 11. Asa D (2006) Carbohydrate and oligosaccharide analysis with a universal HPLC detector. Am Lab 38:16–18 12. Widmer W (2011) Analysis of biomass sugars and galacturonic acid by gradient anion exchange chromatography and pulsed amperometric detection without post-column addition. Biotechnol Lett 33:365–368 13. Grey C, Edebrink P, Krook M, Jacobsson SP (2009) Development of a high performance anion exchange chromatography analysis for mapping of oligosaccharides. J Chromatogr B: Anal Technol Biomed Life Sci 877: 1827–1832 14. Agblevor FA, Hames BR, Schell D, Chum HL (2007) Analysis of biomass sugars using a novel HPLC method. Appl Biochem Biotechnol 136: 309–326 15. Ishizawa CI, Jeoh T, Adney WS, Himmel ME, Johnson DK, Davis MF (2009) Can delignification decrease cellulose digestibility in acid pretreated corn stover? Cellulose 16: 677–686 16. Mittal A, Katahira R, Himmel ME, Johnson DK (2011) Effects of alkaline or liquid-ammonia treatment on crystalline cellulose: changes in


8

M.E. Himmel et al.

crystalline structure and effects on enzymatic digestibility. Biotechnol Biofuels 4:41 17. Chundawat SP, Balan V, Dale BE (2008) Highthroughput microplate technique for enzymatic hydrolysis of lignocellulosic biomass. Biotechnol Bioeng 99:1281–1294 18. Decker SR, Brunecky R, Tucker MP, Himmel ME, Selig MJ (2009) High-throughput screening techniques for biomass conversion. Bioenergy Res 2:179–192 19. Selig MJ, Tucker MP, Sykes RW, Reichel KL, Brunecky R et al (2010) Biomass recalcitrance screening by integrated high throughput hydrothermal pretreatment and enzymatic saccharification. Indus Biotechnol 6:104–111

20. Selig MJ, Tucker MP, Law C, Doeppke C, Himmel ME et al (2011) High throughput determination of glucan and xylan fractions in lignocelluloses. Biotechnol Lett 33: 961–967 21. Studer MH, Brethauer S, Demartini JD, McKenzie HL, Wyman CE (2011) Co-hydrolysis of hydrothermal and dilute acid pretreated populus slurries to support development of a high-throughput pretreatment system. Biotechnol Biofuels 4:19 22. DeMartini JD, Studer MH, Wyman CE (2011) Small-scale and automatable high-throughput compositional analysis of biomass. Biotechnol Bioeng 108:306–12


Part I Biomass Feedstock and Cellulose


Chapter 2 Overview of Computer Modeling of Cellulose Malin BergenstrĂĽhle-Wohlert and John W. Brady Abstract Although it has a deceptively simple primary structure, the collective organization of bulk cellulose, particularly as it exists in cellulose fibers in the cell walls of living plants and other organisms, is quite diverse and complex. While some experimental techniques, such as vibrational spectroscopy and diffraction from partially crystalline samples, are able to provide insights into the organization of bulk cellulose, its intrinsic complexity has left many questions still unanswered. For this reason, additional probes of cellulose structure would be highly desirable. With the continuing advances in computer power through massive parallelization, and the steady progress in computer codes and force fields for modeling carbohydrate systems, molecular mechanics simulations have become an attractive means of studying cellulosic systems at the atomic and molecular level. The coming decade will almost certainly see remarkable advances in the understanding of cellulose using such simulations. Key words: Cellulose, Computer modeling, Molecular dynamics

1. Introduction Cellulose has long been one of the most studied of all biopolymer molecules, due to its singular importance. As the principal structural component of plant cell walls and also produced by algae, tunicates, and some bacteria, it is the most abundant biological molecule in the biosphere, and forms the basis for several major industries, including paper, wood products, and cotton fiber textiles. It also represents the most important feedstock for the industrial production of liquid fuels through biomass conversion, a topic of considerable current interest (1–3). However, in spite of extensive attention and its simple, regular homopolymeric sequence, the aggregate structure of cellulose is still incompletely known. This situation is somewhat analogous to that for water. Although the water molecule is quite simple in structure, its collective behavior is so rich and complex that many aspects of its behavior and Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_2, Š Springer Science+Business Media, LLC 2012

11


12

M. Bergenstråhle-Wohlert and J.W. Brady

biological functions are still not fully understood, in spite of the decades of intensive study that its crucial importance in chemistry and biology has warranted.

2. Cellulose Structure The uncertainties about the structure of cellulose are due in part to the fact that the problem of its spatial organization is actually much more complex than might at first be expected. Native cellulose is generally produced in partially crystalline fibers, by rosettes of six cellulose synthetase molecules arranged hexagonally (or linearly in some algae). In higher plants, six such rosettes are further arranged in a hexagonal ring, so that the overall complex of synthetases produces 36 individual parallel cellulose chains, constituting the primary fibril structural element of cellulose. Each of these bundles of chains contains both regular, crystalline regions and less regular, amorphous regions. Cellulose has several ways to organize into ordered or crystalline structure. Native cellulose is found in two different allomorphs called Iα and Iβ (4, 5). If the cellulosic material is processed in some way, its crystal organization often changes. For instance, regenerated or mercerized cellulose forms cellulose II, and cellulose heated and treated with liquid ammonia forms cellulose III. Native cellulose I differs from these two latter crystal forms in its arrangement of the hydrogen bonding network and, as a consequence, in the conformation of the primary alcohol groups. In cellulose I, the primary alcohol group has been found to be, in principle, all TG (6) (trans with respect to O5 and gauche with respect to C4) (7, 8), and the crystal structure is a layered structure with no hydrogen bonds in between the layers (Fig. 1). Cellulose II and III are both three-dimensional hydrogen bonding networks with alternating conformations of the primary alcohol group. In living cell walls, the cellulose fibrils have a high tendency to associate into larger fibril aggregates. The fibril aggregates intimately associate with other plant constituents such as hemicelluloses (a family of polysaccharides), lignin (aromatic polymers), and water to form the macroscopic plant fibers (9). Thus, the structure of cellulose is highly dependent on its source of origin, its processing history, and environmental conditions. As a consequence, investigating the structure of cellulose is a difficult task, as compared for instance to the crystal structures of proteins, which are routinely determined from single crystals. First of all, as mentioned, even individual native cellulose fibrils are structurally heterogeneous, with at least two different crystal forms (Iα and Iβ) (7, 8, 10) alternating in an unknown spatial distribution with amorphous regions whose character and spatial distribution are also unknown. Second, its source of origin is


2

Overview of Computer Modeling of Cellulose

13

Fig. 1. Front and side view of a 36 chain cellulose Iβ crystal with DP 20.

important to the structure, since fibrils from different sources may have different sizes, different fractions of Iι and Iβ allomorphs and different crystallinity. The sample may also be more or less pure, and the processing history might have affected the original structure. In addition to this, the complete insolubility of cellulose fibrils in water under normal conditions is adds extra complexity to the task (11). As a result, the reported crystal structures for cellulose are actually fiber diffraction studies, with the determination of the molecular coordinates dependent in part on modeling and other assumptions, and are deduced from a surprisingly small number of diffraction spots in the experimental data. While the most recent reported crystal structures (8, 10) seem plausible, some uncertainty remains. The diversity of cellulose structure is for instance highlighted by the discussions around the possible existence of a fibril twist. Experimental evidence suggests that cellulose fibrils may be twisted (12), possibly violating the assumption of a small crystal unit cell, with cellobiose as the repeat unit. The twist is mainly seen for bacterial cellulose (13) and it is not clear if the origin is biological or physical. On the other hand, there are also examples of observations where the fibrils of Valonia algae for instance show no twist at all (14). In addition, it has been suggested that the drying of experimental samples, necessary in the preparation of the crystalline fibrils for diffraction experiments, might result in a change in structure, producing an artifactual lattice different from that of never-dried cellulose. There is also a


14

M. Bergenstråhle-Wohlert and J.W. Brady

general tendency to be seduced by the appearance of precision in crystal structures that might distract from a more critical evaluation of general assumptions. For example, a crystal structure for the regenerated crystal form cellulose II has been reported (10, 15, 16), based on the assumption that in this form of cellulose the chains are antiparallel. However, a study in which the reducing ends of cellulose II chains are labeled has found that the overwhelming majority of the microfibers are labeled at only one end (17), indicating parallel packing of the chains. Another recent labeling study found that as much as 15% of the chains are apparently antiparallel (labeled at both ends), but also found that the larger portion are either unlabeled, or are labeled at only one end, indicating either parallel packing or incomplete conversion (18). Resolution of such an issue is a necessary precondition for a “crystallographic” structure determination based on a minimal set of diffraction spots. Given the many difficulties involved in the experimental characterization of cellulose structure, particularly under noncrystalline conditions, it is reasonable to turn to computational models for cellulose fibrils and crystals, in the hope that such models might further illuminate the experimental data and possibly even be able to resolve some of the uncertainties.

3. Molecular Computer Models and Simulations

Molecular mechanics computer studies now have a very long history of successful application to all manner of biological systems, including solids, liquids and liquid solutions, proteins, phospholipid bilayers, and carbohydrates, including polysaccharides like cellulose (19–30). Molecular mechanics (MM) studies, such as molecular dynamics (MD) simulations or Monte Carlo calculations, directly model the properties of molecular systems using an assumed knowledge of the way in which the energy of the system varies with the atomic coordinates. These simulations have been very useful in studying both biological molecules and liquid solutions. For example, MD simulations have been used to successfully model the folding of a small protein, the villin headpiece subdomain, in aqueous solution, which folded to approximately the experimentally known conformation over a simulation time of 1 μs (at the time requiring four CPU months on the fastest supercomputer available) (31), demonstrating that the principal limitations in the application of such methods to many biological questions are system size and simulation time scale. Modeling of this type has become an essential and integral part of pharmaceutical design and is routinely used in many technological applications.


2

Overview of Computer Modeling of Cellulose

15

One possible approach to the modeling of cellulose crystal structures might be to use straightforward energy minimization calculations (32). This approach adjusts the atomic coordinates of a system in the directions which reduce the forces acting on each atom until the sum of all these forces is zero, representing a local minimum-energy state. However, the potential energy hypersurfaces for cellulose crystalline packing contain an enormous number of local energy minima, many of similar energies, all separated from one another by energy barriers of varying heights. Unfortunately, there are no general procedures for finding the lowest of these local energy minima. The conformation predicted by energy minimization calculations will be a strong function of the conformation chosen as the starting point for the calculation (such as the reported crystal structures). This so-called multiple-minimum problem (33) makes successful structural predictions by energy minimization alone very unlikely to succeed. Perhaps the most common and successful type of computer modeling of biomolecules is molecular dynamics calculations. MD simulations use the forces calculated from MM potential energy functions to numerically integrate Newton’s equations of motion for all of the atoms in a system as they move in response to the forces acting on them, in order to provide a description of the evolution of a system with time. These forces are computed directly from the derivatives of an empirical potential energy function of the atomic coordinates. In such calculations, physical observables (properties) are calculated as time averages over the various states that arise during the course of the simulation, since in principle the frequency of occurrence of each state and the time spent in each state will, for simulations of sufficient length, converge to the value determined by the Boltzmann distribution. Unlike Monte Carlo calculations, MD trajectories allow a picture of how a system might evolve with time if it is not at equilibrium, and at least in principle allow the possibility of calculating rates. Also, unlike the situation with minimization studies, the momentum carried by the system in MD trajectories can allow it to cross energy barriers, allowing it to escape from local energy wells and more completely explore its full configuration space. Since entropy arises naturally from the application of Newtonian dynamics, this approach has the added advantage of sampling the free energy surface, since it is the free energy, including the entropy, which determines experimental properties. The ease with which water and cosolutes can be incorporated into MD simulations is another advantage that these calculations have over energy minimization approaches. Solvent not only serves as a heat bath, further promoting transitions over barriers, but may also be an essential component of the free energy of chain association in crystals (34). The atomic and molecular detail available in such simulations allows unprecedented insight into the physical properties unavailable from any type of experiment. Molecular mechanics simulations


16

M. Bergenstråhle-Wohlert and J.W. Brady

also offer other advantages over experiments. For example, in computer simulations, systems and properties can be modeled which are currently inaccessible to experiment. Furthermore, questions can be asked which could never be answered experimentally (for example, how do results depend on mass, charge, atomic radius, etc., which can be changed in a computer model but of course cannot be adjusted in reality). In some cases (however, probably not most) the simulations may be easier than the experiments and can thus increase the efficiency, for example, of selecting protein mutants to be produced by recombinant techniques, or ligands to be screened as possible drug candidates (or pretreatments, for example, in the case of cellulose modeling). Using molecular graphics, the results of such simulations can be displayed pictorially for a dramatically improved understanding of the system, which facilitates the development of conceptual models, although they should be looked upon with caution and not taken as a verity just because of their attractive appearance. It should also be remembered, however, that molecular mechanics calculations are only useful in understanding physical systems to the extent that the force fields and other approximations employed realistically model the way in which the system energy changes with atomic coordinates. It is essential to validate the results of simulations through comparison with experiment whenever possible, to increase the confidence with which those computational results which cannot be verified experimentally can be accepted. The fundamental requirement of molecular mechanics studies of any type is a complete description of the variation of the total potential energy of the system as a function of the atomic coordinates. For macromolecules and condensed phases the accurate calculation of this quantum mechanical energy is not possible, and it is thus common to employ analytic, empirical energy functions which have theoretically reasonable functional forms and which have been parameterized to the results of experiment and simple calculations. These semiempirical potential energy functions may vary somewhat in their details; most, however, represent the intramolecular potential energy as a sum of electrostatic and van der Waals interactions between nonbonded atoms and terms for hindered rotation about molecular bonds, with bond stretching and bond bending forces derived from quadratic restoring potentials, V (q) = ∑ kb (b − b0 )2 + ∑ kθ (θ − θ0 )2 + ∑ kφ [1 + cos(n φ − δ)] + ∑ Aij / rij 12 − Bij / rij 6 + qiq j / rij .

(1)

The parameters in these potential energy surfaces, such as atomic partial charges, bond lengths and angles, and force constants, are usually obtained from quantum mechanical calculations for small molecular models that contain functional groups in analogous covalent environments. The results are then validated by extensive


2

Overview of Computer Modeling of Cellulose

17

comparison with available experimental data, such as crystal structures, vibrational spectra, or solvation free energies. Several MM software packages have been developed and widely employed in modeling biomolecules, including CHARMM (35, 36), AMBER (37), GROMOS/GROMACS (38, 39), NAMD (40, 41), and TINKER (42). It should be noted that for straightforward calculations like MD simulations, the results should not depend significantly on the actual program employed, barring the accidental introduction of a programming “bug,” which is remotely possible since all of these programs undergo periodic updating and development. However, the results of MD simulations might depend strongly on the energy function (force field) used to model the system, including the water model used. Enormous effort has gone into the development of ever more realistic MM energy parameters over the past few decades. Several force fields have been developed specifically to model carbohydrates and the specific covalent situations encountered in such molecules. The most widely used current carbohydrate force fields are the newly developed “Team Sugar” CHARMM parameters (43), the most recent version of the GLYCAM (44) (AMBER) parameters, and the recent GROMOS carbohydrate parameters (45). It should be reemphasized that while these sets were developed in association with specific software packages, they can in principle be used in any of the MM programs with the same results (with greater of lesser degrees of difficulty in implementation) (46). 3.1. Molecular Simulations of Cellulose

As mentioned, the most crucial limitations of MD simulations are system size and the timescale of the simulations. The equations of motion are integrated by discretizing them into a series of finite steps during which the forces can be treated as approximately constant. The forces must be calculated at each of these discrete steps, which is an expensive operation for a polymer like cellulose since the number of contributing interactions grows approximately as N2, where N is the number of atoms. Because of the strengths of the chemical bonds involved, bond vibrations are rapid, and the size of the individual time steps between these force calculations must be quite small, on the order of 1 fs (i.e., 10−15 s), to ensure that the changes across the step are small. Thus, a very large number of these CPU-intensive force calculations are necessary to amount to even relatively short simulation times (hundreds of picoseconds, for example). Using current conventional computers, typical MD simulations of solvated proteins model times in the nanosecond range, although the microsecond range is within reach when using well-parallelized simulation routines on hundreds of CPUs (47). Modeling cellulose fibrils is much harder than simulating the typical protein; however, since a typical fibril chain in wood cellulose has a degree of polymerization of around 15,000. Even with only 36 such chains, and the explicit water molecules necessary


18

M. Bergenstråhle-Wohlert and J.W. Brady

to solvate such a large aggregate, the number of atoms N is vastly larger than for even large proteins, further limiting the timescales that can be simulated. Unfortunately, however, cellulose crystals change their structures on timescales ranging from milliseconds to far longer. Thus, there is a difference of at least four orders of magnitude between the timescale of a typical MD simulation and the timescale for cellulose structural changes. Keeping in mind these limitations, the types of questions about cellulose that might be addressed using modeling could include: what is the native fibril structure, including both the ordered or crystalline portion, as well as the amorphous fraction; what is the nature of the interactions with the hemicelluloses and lignins and the role of water in the native state; what are the effects of temperature, solvent, and cosolutes (as in pretreatments); how do cellulose fibrils interact with each other in nanostructured cellulose-based materials and how do cellulose substrate fibrils interact with cellulases? With the widespread availability of highly parallel MM software packages and the speed of parallel arrays of CPUs, all of these types of questions are becoming tractable, at least in principle, and several such preliminary studies have already been attempted. The first MM simulations of cellulose date back to only a few years after such calculations were applied to proteins (48–51), but it is only quite recently that realistic simulations of large unrestrained crystals in an aqueous environment have been possible (24, 26– 30). Earlier simulations necessarily were confined to relatively small model crystals, for short times, and employing severe approximations that significantly limited any changes that might have taken place (19, 20, 52, 53). Simulations have also begun to be applied to the problem of cellulose interacting with cellulase enzymes (54–56). An overview of the in silico modeling of cellulose during the last decade is found within a recent review of the subject (57). Unfortunately, at the time of this writing, it is still too early for the MM simulations of cellulose crystals that have been reported to date to definitively settle any of the remaining questions concerning cellulose structure (57). However, several general statements can be made. Perhaps the most important is that the reported cellulose Iβ crystal structure (10), when simulated for sufficient time periods, and without artificial constraints, is not stable with any of the MM force fields currently available (43–45). Apparently the heights of the barriers that separate the reported crystal structure from lower energy structures vary among the different force fields, so that the time for significant structural reorganization varies among the different models, but with enough time, all will undergo changes from the proposed crystal structure (30). Of course, it is possible that all of these force fields are wrong in some critical way, but it is worth noting that all three of the most commonly used carbohydrate force fields were developed independently, and that similar force fields for other systems, including


2

Overview of Computer Modeling of Cellulose

19

Fig. 2. A DP 20 cellulose chain represented as an all-atom model and a coarse-grained model with three interacting sites per glucose unit.

carbohydrates, have successfully reproduced known experimental properties. In addition, some of the differences observed in the MM simulations with different force fields are common across the various force fields, such as the observation of chain tilts within layers (24, 25, 30, 52), the occurrences of non-TG primary alcohol conformations (30), and the development of hydrogen bonds between layers (24, 30). Chain and fibril twists are a common feature of the MM simulations (24, 25, 30), but apparently such twisting is frozen out at longer simulation times, regardless of the force field used, as a three-dimensional, interlayer hydrogen bond network develops as the result of the reorganization of the crystal structure. Such troubling observations suggest that more work is probably needed, both in the crystallographic analysis and in the MD simulations. Other possibilities for the profitable application of MM simulations to cellulose systems in the future would include studies of the interactions of cellulose with lignin (58) and hemicellulose, and of cellulase enzymes interacting with different types of cellulosic substrates (55). Since one of the most important limitations on MM simulations of cellulosic systems is that of size and timescale, there is the potential for extending the scope of such simulations through the use of so-called coarse grain calculations, in which certain presumably unimportant details are represented only approximately rather than in specific detail. In MM calculations, this often involves models in which molecules such as cellulose are not treated in atomic detail. Rather, groups of atoms which are presumed to exhibit a similar character and which move approximately as a unit are represented usually as a single entity, and new force fields are parameterized so as to give approximately the same results in test simulations as MD simulations at the full atomiclevel detail. Figure 2 illustrates an example of such a coarse-grained representation of a cellulose chain, compared to an all-atom model. Often implicit solvation models are employed that allow the very expensive treatment of many thousands of water molecules to be avoided (59–62). The principal advantage of coarse grain calculations is that they allow the simulation times to be substantially extended, often by orders of magnitude, which is of critical importance since, as


20

M. BergenstrĂĽhle-Wohlert and J.W. Brady

already observed, the most important limitation of the utility of MM simulations is the timescale that can be modeled. Already coarse grain models have been developed for cellulose (63) and used successfully to understand the interaction of a cellulose binding domain from the cellulase CBH I of Trichoderma reesei with a crystalline cellulose surface (64). Such methods are not yet feasible for studying intrafibrillar structure, but hold great promise for understanding such large-scale structures as cell wall fibers or cellulose fibril network materials like cellulose nanopaper or foams (65) in the future. In addition to coarse grain simulations, a new generation of polarizable force fields is under development for all-atom simulations which account for the changing electronic structure of molecules as their local environment changes (A.D. MacKerell, personal communication). Since the changes in aggregation that occur in crystalline cellulose in MM simulations largely involve interchain and interlayer hydrogen bonding, such polarization effects may prove to be important in an ultimate understanding of the structures of cellulose. The steady increase in computer size and speed through parallelization will make long simulations of large systems progressively more practical, even with the greater computational expense of including electronic polarization. In conclusion, advances in computer power, software design, and force field development have set the stage for molecular mechanics simulations of cellulosic systems at an unprecedented level of realism. Future calculations will potentially involve large scale models of the full complexity of cell walls, including cellulose fibers interacting with lignin, hemicelluloses, water, as well as potentially pectins and phospholipids at the interfaces. Simulations of extended realistic cellulose microfibrils, in tandem with diffraction, NMR, and spectroscopic experiments, may ultimately resolve any lingering uncertainties about the crystalline structure of cellulose, and provide insight into the nature of amorphous cellulose as well. Finally, models of the interactions of cellulase proteins with cellulose may help in the development of more efficient process for cellulose degradation in biomass conversion. The coming decade should be an exciting time indeed in the study of cellulose using MM techniques.

Acknowledgments This work was supported under the SciDAC Program by the US DOE Office of Biological and Environmental Research and the Office of Advanced Scientific Computing Research. The authors also thank James Matthews, Michael Crowley, and Jakob Wohlert for helpful discussions.


2

Overview of Computer Modeling of Cellulose

21

References 1. Atalla RH, Brady JW, Matthews JF, Ding S-Y, Himmel ME (2008) In: Himmel ME (ed) Biomass recalcitrance: deconstructing the plant cell wall for bioenergy. Oxford, Blackwell 2. Himmel ME, Ding S-Y, Johnson DK, Adney WS, Nimlos MR, Brady JW, Foust TD (2007) Science 315:804–807 3. Atalla RH (1999) In: Pinto BM (ed) Comprehensive natural products chemistry, vol 3. Carbohydrates. Elsevier, Cambridge, pp 529–598 4. Atalla RH, VanderHart DL (1984) Science 223:283–285 5. Atalla RH, VanderHart DL (1999) Solid State Nucl Magn Reson 15:1–19 6. Marchessault RH, Peréz S (1979) Biopolymers 18:2369–2374 7. Nishiyama Y, Langan P, Chanzy H (2002) J Am Chem Soc 124:9074–9082 8. Nishiyama Y, Sugiyama J, Chanzy H, Langan P (2003) J Am Chem Soc 125:14300–14306 9. Carpita NC, Gibeaut DM (1993) Plant J 13:1–30 10. Langan P, Sukumar N, Nishiyama Y, Chanzy H (2005) Cellulose, preprint 11. Bergenstråhle M, Wohlert J, Himmel ME, Brady JW (2010) Carbohydr Res 345: 2060–2066 12. Hanley SJ, Revol J-F, Godbout L, Gray DG (1997) Cellulose 4:209–220 13. van Daele Y, Revol JF, Gaill F, Goffinet G (1992) Biol Cell 76:87–96 14. Elazzouzi-Hafraoui S, Nishiyama Y, Putaux J-L, Heux L, Dubreuil F, Rochas C (2008) Biomacromolecules 9:57–65 15. Langan P, Nishiyama Y, Chanzy H (1999) J Am Chem Soc 121:9940–9946 16. Langan P, Nishiyama Y, Chanzy H (2001) Biomacromolecules 2:410–416 17. Maurer A, Fengel D (1992) Holz als Roh- und Werkstoff 50:493 18. Kim N-H, Imai T, Wada M, Sugiyama J (2006) Biomacromolecules 7:274–280 19. Heiner AP, Sugiyama J, Teleman O (1995) Carbohydr Res 273:207–223 20. Heiner AP, Kuutti L, Teleman O (1998) Carbohydr Res 306:205–220 21. Tanaka F, Fukui N (2002) Cellulose Research & Development, 82–83 22. Tanaka F, Fukui N (2004) Cellulose 11:33–38 23. Zhang Q, Jaroniec J, Lee G, Marszalek PE (2005) Angew Chem Int Ed 44:2723–2727

24. Matthews JF, Skopec CE, Mason PE, Zuccato P, Torget RW, Sugiyama J, Himmel ME, Brady JW (2006) Carbohydr Res 341:138–152 25. Yui T, Nishimura S, Akiba S, Hayashi S (2006) Carbohydr Res 341:2521–2530 26. Yui T, Hayashi S (2007) Biomacromolecules 8:817–824 27. Bergenstråhle M, Berglund LA, Mazeau K (2007) J Phys Chem B 111:9138–9145 28. Bergenstråhle M (2008) Polymer technology. KTH Chemical Science and Engineering, Stockholm, Sweden 29. Bergenstråhle M, Wohlert J, Larsson PT, Mazeau K, Berglund LA (2008) J Phys Chem B 112:2590–2595 30. Matthews JF, Bergenstråhle M, Beckham GT, Himmel ME, Nimlos MR, Brady JW, Crowley MF (2011) J Phys Chem B 115:2155–2166 31. Duan Y, Kollman PA (1998) Science 282: 740–744 32. Scheraga HA (1968) Adv Phys Org Chem 6: 103–184 33. Scheraga HA (1983) Biopolymers 22:1–14 34. Wohlert J, Schnupf U, Brady JW (2010) J Chem Phys 133:155103 35. Brooks BR, Bruccoleri RE, Olafson BD, Swaminathan S, Karplus M (1983) J Comput Chem 4:187–217 36. Brooks BR, Brooks CL, MacKerell AD, Nilsson L, Petrella RJ, Roux B, Won Y, Archontis G, Bartels C, Boresch S, Caflisch A, Caves L, Cui Q, Dinner AR, Feig M, Fischer S, Gao J, Hodoscek M, Im W, Kuczera K, Lazaridis T, Ma J, Ovchinnikov V, Paci E, Pastor RW, Post CB, Pu JZ, Schaefer M, Tidor B, Venable RM, Woodcock HL, Wu X, Yang W, York DM, Karplus M (2009) J Comput Chem 30:1545–1614 37. Pearlman DA, Case DA, Caldwell JC, Seibel GL, Singh UC, Weiner P, Kollman PA (1991) 4th edn. University of California, San Francisco, San Francisco 38. Berendsen HJC, van der Spoel D, van Drunen R (1995) Comput Phys Commun 91:43–56 39. Hess B, Kutzner C, van der Spoel D, Lindahl E (2008) J Chem Theor Comput 4:435–447 40. Kale L, Skeel R, Bhandarkar M, Brunner R, Gursoy A, Krawetz N, Phillips J, Shinozaki A, Varadarajan K, Schulten K (1999) J Comp Phys 151:283–312 41. Phillips JC, Braun R, Wang W, Gumbart J, Tajkhorshid E, Villa E, Chipot C, Skeel RD, Kale L, Schulten K (2005) J Comput Chem 26:1781–1802


22

M. Bergenstråhle-Wohlert and J.W. Brady

42. Ponder JW (1999) 3.7 edn. Washington University, St. Louis, MO 43. Guvench O, Greene SN, Kamath G, Brady JW, Venable RM, Pastor RW, Mackerell AD (2008) J Comput Chem 29:2543–2564 44. Kirschner KN, Yongye AB, Tschampel SM, Gonzáles-Outeiriño J, Daniels CR, Foley BL, Woods RJ (2008) J Comput Chem 29:622 45. Lins RD, Hünenberger PH (2005) J Comput Chem 26:1400–1412 46. Bjelkmar P, Larsson P, Cuendet MA, Hess B, Lindahl E (2010) J Chem Theor Comput 6:459–466 47. Gruber CC, Pleiss J (2011) J Comput Chem 32:600–606 48. Yathindra N, Rao VSR (1970) Biopolymers 9:783–790 49. Sathyanarayana BK, Rao VSR (1971) Biopolymers 10:1605–1615 50. Simon I, Scheraga HA, Manley RSJ (1988) Macromolecules 21:983–990 51. Simon I, Glasser L, Scheraga HA, Manley RSJ (1988) Macromolecules 21:990–998 52. Heiner AP, Teleman O (1997) Langmuir 13: 511–518 53. Mazeau K, Heux L (2003) J Phys Chem B 107: 2394–2403 54. Nimlos MR, Matthews JF, Crowley MF, Walker RC, Chukkapalli G, Brady JW, Adney WS, Cleary JM, Zhong L, Himmel ME (2007) Protein Eng Des Select 20:179–187 55. Zhong L, Matthews JF, Hansen PI, Crowley MF, Cleary JM, Walker RC, Nimlos MR, Adney WS, Himmel ME, Brady JW (2009) Carbohydr Res 344:1984–1992

56. Tavagnacco L, Mason PE, Schnupf U, Pitici F, Zhong L, Himmel ME, Crowley M, Cesàro A, Brady JW (2011) Carbohydr Res 346: 839–846 57. Bellesia G, Asztalos A, Shen T, Langan P, Redondo A, Gnanakaran S (2010) Acta Crystallogr Sect D-Biol Crystallogr D66 58. Petridis L, Smith JC (2009) J Comput Chem 30:457–467 59. Still WC, Tempczyk A, Hawley RC, Hendrickson T (1990) J Am Chem Soc 112:6127–6129 60. Hummer G, Pratt LR, García AE (1997) J Am Chem Soc 119:8523–8527 61. Qiu D, Shenkin PS, Hollinger FP, Still WC (1997) J Phys Chem A 101:3005–3014 62. Beckham GT, Bomble YJ, Matthews JF, Taylor CB, Resch MG, Yarbrough JM, Decker SR, Bu L, Zhao X, McCabe C, Wohlert J, Bergenstråhle M, Brady JW, Adney WS, Himmel ME, Crowley MF (2010) Biophys J 99:3773–3781 63. Wohlert J, Berglund LA (2011) J Chem Theor Comput 7:753–760 64. Bu LT, Beckham GT, Crowley MF, Chang CH, Matthews JF, Bomble YJ, Adney WS, Himmel ME, Nimlos MR (2009) J Phys Chem B 113:10994–11002 65. Eichhorn SJ, Dufresne A, Aranguren M, Marcovich NE, Capadona JR, Rowan SJ, Weder C, Thielemans W, Roman M, Renneckar S, Gindl W, Veigel S, Keckes J, Yano H, Abe K, Nogi M, Nakagaito AN, Mangalam A, Simonsen J, Benight AS, Bismarck A, Berglund LA (2010) J Mater Sci 45:1–33


Chapter 3 Imaging Cellulose Using Atomic Force Microscopy Shi-You Ding and Yu-San Liu Abstract Cellulose is an important biopolymer primarily stored as plant cell wall material. Plant-synthesized cellulose forms elementary fibrils that are micrometers in length and 3–5 nm in diameter. Cellulose is a dynamic structure, and its size and property vary in different cellulose-containing materials. Atomic force microscopy offers the capability of imaging surface structure at the subnanometer resolution and under nearly physiological conditions, therefore providing an ideal tool for cellulose characterization. Key words: Cellulose, Plant cell walls, Atomic force microscopy (AFM)

1. Introduction Natural cellulose forms nanometer scale bundles of linear β-1, 4-linked glucan chains. A network of inter- and intrachain hydrogen bonds facilitates cellulose structure. Cellulose in higher plant cell walls is the major source of glucose that can be fermented to produce biofuels by microorganisms in the proposed biomass conversion process. However, natural cellulose is embedded in a complicated polymeric matrices composed of hemicelluloses, lignin, and pectin in the plant cell walls, which protect cellulose from hydrolysis by biocatalysts. In order to enhance the efficiency of cellulose conversion, a thermochemical pretreatment process is commonly used to remove or relocate matrixing polymers, thus in turn exposing the cellulose to enzymes used for saccharification (1). However, the pretreatment step may alter the native structure of cellulose. Although some mild pretreatment approaches, such as dilute acid, may have minimum impact of the cellulose. Indeed, cellulose is a dynamic structure, the physical and

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_3, Š Springer Science+Business Media, LLC 2012

23


24

S.-Y. Ding and Y.-S. Liu

structural characteristics may change in different plant cell walls (i.e., living, dead, or different types of cells), or in different conditions (i.e., fresh, dry, isolated, or pretreated). Changes may include the degree of polymerization (DP), crystallinity, crystalline allomorphs (e.g., cellulose Iα, Iβ, II, III, and IV), size, and shape of cellulose microfibrils, which have great impact of cellulase accessibility and performance. To further optimize biomass conversion processes, it is therefore important to understand the correlation between cellulose structure and enzyme function. Hydrolysis of cellulose to glucose requires three different enzymatic activities. The endoglucanase cleaves the surface cellulose chains creating ends, and then exoglucanase hydrolyzes these chains processively producing cellobiose that is finally broken down to glucose by the β-D-glucosidase. In nature, it has been found that cellulase enzymes are composed of functional modules acting synergistically. In different cellulolytic organisms, cellulase can be “free” containing one catalytic module with or without the carbohydrate-binding module (CBM), “multidomain” with more than one catalytic module, or “complex” composed by many enzymes through domain–domain interaction, i.e., the cellulosomes. Besides enzymatic pathway, cellulose can also be depolymerized by non enzymatic catalysts (i.e., hydrogen peroxide, iron, and oxalic acid) in some so-called brown rot fungi. Nevertheless, the molecular mechanism of cellulose hydrolysis has not been fully elucidated. Cellulose is the most abundant biopolymer on earth. Because of its importance to human life, many analytical techniques have been used to characterize cellulose structure. However, due to the small size of cellulose microfibrils, 3–5 nm in diameter and as long as millimeter in length, and their dynamic and intertwining arrangement with other matrix biopolymers in the plant cell walls, most of methods used to analyze cellulose require extensive sample preparation that involves strong acid or alkaline reagents and even elevated temperature—resulting nonnatural cellulose (2, 3). Atomic force microscopy (AFM) offers an ideal approach to image cellulose surface at the sub-nanometer resolution with minimum sample preparation, and most importantly the sample can be imaged under nearly physiological conditions, representing nearly native structure. It is worth noting that the surface structure of the plant cell walls (cellulose and other polymers) is where the cellulolytic enzyme functions, so that AFM imaging provides invaluable information about the molecular basis and mechanisms of enzyme– plant cell wall interaction. The basic principles and apparatus of AFM for biological applications have been reviewed extensively elsewhere (4). In this chapter, a practical protocol is given for imaging plant cell wall microfibrils in dry and fresh samples.


3

Imaging Cellulose Using Atomic Force Microscopy

25

2. Materials 2.1. Plant Cell Walls

Despite many different cell types, plant cell walls can be basically categorized into two types. The primary cell wall is the wall of growing cell, and the secondary cell wall is the thick layer deposited after cell growth ceases in some cell types. The inner surface of the secondary cell walls is commonly covered by the warty layer, and cellulose structure cannot be imaged (5). All plant cell walls used here are the primary cell walls (Fig. 1).

2.1.1. Maize Fresh Cell Walls

In a petri dish, corn grains are germinated on filter paper soaked with water. After about a week, 2–3 cm seedlings can be collected and hand-cut into small pieces by razor blade (see Note 1). These cell wall pieces can then be spun down in an Eppendorf tube and wash with water for at least three times. Fresh cut mica pre-coated with poly-L-lysine is used as substrate (see Note 2). 5–10 μL of cell wall suspension is dropped onto mica, and after a 5–10 min incubation to allow the cell wall preparation to settle down on the mica surface, a filter paper strip is used to remove extra water. Cell walls should be imaged immediately.

2.1.2. Maize Dry Cell Walls

Pieces of primary cell walls can be obtained from the pith of corn stover stem by hand cutting using a razor blade. These dry powder-like cell wall pieces can be subjected onto fresh-cut mica surface precoated with poly-L-lysine and imaged with AFM directly (see Note 1).

Fig. 1. The primary cell wall from maize.


26

S.-Y. Ding and Y.-S. Liu

3. AFM Operation It is our intention to give a practical protocol of AFM operation. Although there are many different manufacturers that produce the atomic force microscope, the general principles of AFM are the same and it has described in many other sources. The operation procedure presented here can also be used with other AFM systems. Generally, AFM images the surface structure of almost any type of materials. However, because high resolution is expected, the actual size of sample is usually limited in millimeters to allow fitting into the sample holder. In addition, relative flat surface is necessary for high resolution imaging. 3.1. The AFM System

A Multi-Mode™ scanning probe microscope with NanoScope V controller (Veeco, Santa Barbara, California) was used for all imaging under dry and aqueous conditions. To ensure absolute stability, the AFM was located in a specially designed laboratory with acoustic and vibration isolation. A customized Nikon optical microscope with deep focus (800× magnification) was used to aid in positioning of the AFM tip to the desired location (Fig. 1). For most experiments, the standard 15-μm scanner was used with the Tapping Mode. For imaging dry samples, etched silicon probes (TESP, Veeco NanoProbe) or the aluminum-coated probes (HI’RES, MikroMasch, Portland, Oregon). For imaging sample in aqueous condition, silicon nitride probes (DNP) or sharp nitride lever (SNL) probes (SNL, Veeco NanoProbe) were used (see Note 3).

3.2. Imaging in Dry Conditions

Imaging in dry conditions (6).

3.3. Imaging in Water or Buffer (See Note 4)

Locate the probe on the sample.

Tune the adequate resonance frequency using the auto tuning process, usually in the range of 250–300 kHz.

Begin with a scan rate of 1 Hz.

Start with small scan size, for example, 100 × 100 nm, then increase scan size slowly to 200 × 200 nm, 500 × 500 nm, until a maximum 15 × 15 μm.

Adjust drive amplitude, set point, and scan rate during measuring to minimize tip artifacts.

Imaging in aqueous conditions is absolutely nontrivial and timeconsuming process. Success largely relies on the stability of the sample. ●

Locate the probe on sample when it is still wet (Fig. 2).

Manually adjust the tip to close approaching to the sample surface.


3

Imaging Cellulose Using Atomic Force Microscopy

27

Fig. 2. Locate AFM probe onto plant cell wall.

Take out the tip holder carefully without changing the sample position, add 100–150 μL distilled water or buffer onto the sample.

Put back the tip holder and auto tune the resonance frequency.

3.4. Image Analysis 3.4.1. Height Image

Begin scan rate of 1 Hz and scan size of 1 × 1 μm. Adjust drive amplitude, set point, and scan rate during measuring to minimize tip artifacts.

Height images are common in all scan modes, presenting the topographical mapping of the sample surface (Fig. 3). These images are usually used to measure the lateral (xy) and height (z). The xy measurement is normally not accurate due to tip broadening, so that the picture appears too wide, especially when the features are at the nanometer scale. The height measurement is more accurate, and can be used to render a “3-D” image (Fig. 4).

3.4.2. Phase Image

The differences of physical or chemical properties of sample surfaces can be attributed to the phase change of cantilever oscillation when the tip interacts with the surface in tapping mode. The phase images usually show high contrast when imaging plant cell walls in dry condition. The heterogeneous structure of cellulose and other polymers in plant cell walls can be imaged by AFM tapping mode yielding fine details (Fig. 2).

3.4.3. Amplitude Image

When imaging in aqueous conditions, the amplitude image is most useful. This is true particularly for the biological samples where the surface is usually rough, the amplitude images display more contrast of the feature shape, whereas the intensity (z) of the amplitude image is usually meaningless (Fig. 3).


28

S.-Y. Ding and Y.-S. Liu

Fig. 3. Height image of maize primary cell wall imaged in dry condition, scan size 1 Ă— 1 Îźm.

Fig. 4. 3-D rendering of Fig. 3.

4. Notes 1. In dry corn stover, parenchyma cells that have primary cell walls can be easily found in the pith of stem. In the fresh maize seedling, the coleoptiles that contain mostly growing cells


3

Imaging Cellulose Using Atomic Force Microscopy

29

Fig. 5. Phase image of the same area of Fig. 3.

Fig. 6. Amplitude image of fresh plant cell wall from maize seedling imaged in water, scan size 1 × 1 μm.

can be used. The size of plant cells is in the micrometer range. When cell walls are prepared using razor blades and hand cutting, the pieces should be smaller than a cell to break the “box” structure that is a problem for AFM tips to access the cell wall surface (Figs. 5 and 6). 2. Fresh mica surface can be easily obtained using scotch tape to peel off one layer. 5 μL of poly-L-lysine solution (0.1 %, w/v, in water, Sigma-Aldrich, St. Louis, MO) was added to freshly cleaved mica. After 5 min incubation, the excess poly-L-lysine solution was removed using a spin coater (Model KW-4A,


30

S.-Y. Ding and Y.-S. Liu

Chemat Technology, Northridge, CA) operated at 500 rpm for 30 s followed by 4,000 rpm for 30 s. 3. There are many different types of probes commercially available for tapping mode AFM. Usually it is recommended to use more soft and low force tip for imaging in aqueous conditions. 4. We found that application of a 150-μL liquid droplet (water or buffer) provided the sample with an aqueous environment that was sufficient for imaging for at least 10 h without significant artifacts caused by water evaporation (7).

Acknowledgments This work was supported by the BioEnergy Science Center (BESC), which is a US Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science, Oak Ridge National Laboratory. References 1. Himmel ME, Ding S-Y, Johnson DK, Adney WS, Nimlos MR, Brady JW, Foust TD (2007) Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science 315: 804–807 2. Yarbrough JM, Himmel ME, Ding S-Y (2009) Plant cell wall characterization using scanning probe microscopy techniques. Biotechnol Biofuels 2:17 3. Ding S-Y, Xu Q, Crowley M, Zeng Y, Nimlos M, Lamed R, Bayer EA, Himmel ME (2008) A biophysical perspective on the cellulosome: new opportunities for biomass conversion. Curr Opin Biotechnol 19:218–227

4. Morris VJ, Kirby AR, Gunning AP (1999) Atomic force microscopy for biologists. Imperial College Press, London 5. Ding S-Y, Himmel ME (2008) Anatomy and ultrastructure of maize cell walls: an example of energy plants. In: Himmel ME (ed) Biomass recalcitrance. Blackwell, Oxford, pp 38–60 6. Ding S-Y, Himmel ME (2006) The maize primary cell wall microfibril: a new model derived from direct visualization. J Agric Food Chem 54:597–606 7. Liu YS, Baker JO, Zeng YN, Himmel ME, Haas T, Ding S-Y (2011) Cellobiohydrolase hydrolyzes crystalline cellulose on hydrophobic faces. J Biol Chem 286:11195–11201


Chapter 4 Preservation and Preparation of Lignocellulosic Biomass Samples for Multi-scale Microscopy Analysis Bryon S. Donohoe, Peter N. Ciesielski, and Todd B. Vinzant Abstract Biomass exhibits structural and chemical complexity over multiple size scales, presenting many challenges to the effective characterization of these materials. The macroscopic nature of plants requires that some form of size reduction, such as dissection and microtomy, be performed to prepare samples and reveal features of interest for any microscopic and nanoscopic analyses. These size reduction techniques, particularly sectioning and microtomy, are complicated by the inherent porosity of plant tissue that often necessitates fixation and embedding in a supporting matrix to preserve structural integrity. The chemical structure of plant cell walls is vastly different from that of the membrane bound organelles and protein macromolecular complexes within the cytosol, which are the focus of many traditional transmission electron microscopy (TEM) investigations in structural biology; thus, staining procedures developed for the latter are not optimized for biomass. While the moisture content of biomass is dramatically reduced compared to the living plant tissue, the residual water is still problematic for microscopic techniques conducted under vacuum such as scanning electron microscopy (SEM). This requires that samples must be carefully dehydrated or that the instrument must be operated in an environmental mode to accommodate the presence of water. In this chapter we highlight tools and techniques that have been successfully used to address these challenges and present procedural details regarding the preparation of biomass samples that enable effective and accurate multi-scale microscopic analysis. Key words: Biomass surface characterization, Plant cell walls, FE-SEM, TEM, Electron tomography, Cryo-preservation

1. Introduction The material properties of biomass are inherently heterogeneous and vary significantly across plant species and even among cell types of the same plant, presenting a challenge for sample preparation and characterization methods. In order to address these issues, NREL has assembled a set of versatile imaging tools, specialized specimen preparation facilities, and expert investigators into a facility

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_4, Š Springer Science+Business Media, LLC 2012

31


32

B.S. Donohoe et al.

Fig. 1. Bright-field toluidine blue staining (a), SEM (b), and TEM (c) images of liquid hot water pretreated switchgrass samples. These images highlight how multi-scale and multimode imaging can be used to visualize the physical changes to cell walls caused by biomass processing steps.

referred to as the Biomass Surface Characterization Laboratory (BSCL). Imaging techniques central to the BSCL include stereomicroscopy, confocal scanning laser microscopy, confocal Raman microscopy, environmental and high-resolution scanning electron microcopy (ESEM, FE-SEM), transmission electron microcopy (TEM), electron tomography and atomic force microscopy (AFM). These characterization methods are supported by sample preparation techniques such as chemical and cryo-preservation, resin embedding, ultramicrotomy, cytochemical staining, immunolabeling, and others that are presented in this chapter. By employing these techniques in a multi-scale imaging approach, researchers are investigating the physical and chemical underpinnings of the phenomena of biomass recalcitrance to deconstruction (Fig. 1). This information is then used to refine biomass pretreatment strategies that optimize the susceptibility of the biomass to subsequent enzymatic digestion or thermochemical depolymerization. As an example, high-resolution microscopic analysis has been effectively used by NREL and other groups in recent studies to reveal the impact of hot water pretreatment (1), dilute acid pretreatment (2, 3), treatment in ionic liquids (4), and AFEX pretreatment on cell wall ultrastructure (5). Many of the challenges that are encountered in detailed biomass characterization arise from the fact that plants are macroscopic organisms that exhibit multi-scale complexity. For example, vascular bundles and pith tissues are evident within the same grass plant stem at a scale of 10−5 m, differing cell types may be identified at a scale of 10−4 m, the primary and secondary walls of individual cells may be differentiated at the scale of ~10−6 m, and the individual cellulose microfibers that comprise these walls have dimensions on the order of ~10−9 m. It has become clear that the physical and chemical underpinnings of the phenomena of biomass recalcitrance to deconstruction exist at all of these scales. In addition, the chemical properties of biomass become increasingly heterogeneous as the


4

Preservation and Preparation of Lignocellulosic Biomass Samples…

33

analytical size scale increases: cellulose microfibrils of relatively uniform chemistry form bundles which are closely associated with hemicelluloses, pectin, and lignin which are present in varying quantities and compositions across different cell types, locations within the stem, and plant species (6). This variation in chemical composition complicates the process of staining biomass samples for light microscopy and TEM because the individual components of plant cell walls display varying affinities for staining reagents that may be changing with changes in accessible reactive surface area during bioconversion processing steps. Furthermore, staining protocols developed for investigations of living plant cells and their cytoplasmic contents often display poor efficacy for highlighting cell walls. The moisture content of biomass can also be problematic, since some microscopic techniques are performed on dry samples or even under vacuum and require further dehydration of the samples prior to analysis. If not done carefully, changes in surface tension during the drying process can introduce structural artifacts such as wrinkles, compaction, or shrinking in the cell walls. Confronting these challenges necessitates the application of robust sample preservation and preparation methods in tandem with characterization techniques that accommodate diverse chemical structure and span many orders of magnitude in spatial resolution.

2. Materials We have listed here the tools and materials we are currently using in the BSCL. Often there are equivalent products available from other sources, but we have found these materials to be reliable and produce quality results in our hands. 2.1. Fixation and Embedding by Microwave Processing

2.5% Gluteraldehyde in buffer (EMS, Hatfield, PA).

0.1 M Sodium cacodylate buffer or 1× PBS buffer (EMS, Hatfield, PA).

1% Buffered osmium tetroxide (EMS, Hatfield, PA).

Dry acetone or ethanol (Sigma-Aldrich, St. Louis, MO).

Calibrated PELCO microwave with thermocouple and PELCO ColdSpot® (Ted Pella, Redding, CA).

EMbed 812 or LR White resin (EMS, Hatfield, PA).

Beem ® size 00 pyramid embedding capsules, Easy Molds size 00, or PELCO flat embedding molds (Ted Pella, Redding, CA).

Vacuum oven (VWR, Radnor, PA).

Beem® capsule press (Ted Pella, Redding, CA).


34

B.S. Donohoe et al.

2.2. Fixation and Embedding by High-pressure Freezing and Freeze Substitution

2.3. Drying and Coating for SEM Analysis

0.15 M sucrose solution used as cryoprotectant for stem and shoot samples (Sigma-Aldrich, St. Louis, MO).

1-Hexadecene used as a cryoprotectant for leaf and cotyledon samples (Sigma-Aldrich, St. Louis, MO).

0.2 mm deep planchets or other planchets appropriate to the specific model of high-pressure freezer (Leica Microsystems, Wetzlar, Germany).

Leica EMPact2 high-pressure freezer (Leica Microsystems, Wetzlar, Germany).

Nunc® cryo-tube vials prefilled with freeze substitution cocktails and stored in a liquid nitrogen dewar (VWR, Radnor, PA).

Leica AFS2 automated freeze substitution unit (Leica Microsystems, Wetzlar, Germany).

1% Osmium tetroxide/0.1% uranyl acetate in acetone for freeze-substitution of samples for better structural preservation (EMS, Hatfield, PA).

0.25% Gluteraldehyde/0.1% uranyl acetate in acetone for freeze-substitution of samples for better antigen preservation (EMS, Hatfield, PA).

Millrock bench-top freeze dryer (Kingston, NY).

Labconco Co. freeze dying glassware (Kansas City, MO).

Liquid nitrogen or dry ice in ethanol.

SPI-DRY critical point dryer (CPD) jumbo (SPI supply, West Chester, PA).

CPD platform and controlling loop (BSCL NREL, Golden, CO).

Recirculating chiller model 13270-120 (VWR, Bristol, CT).

Heating recirculator model 1104 (VWR, Bristol, CT).

Liquid carbon dioxide.

Cressington high resolution sputter coater 208HR (Ted Pella, Inc.).

Thickness controller MTM-20 (Ted Pella).

25 mm Aluminum Studs (Ted Pella, Inc.).

Double stick carbon tape (SPI Supplies, West Chester, PA).

Pelco colloidal silver liquid no. 16034 (Ted Pella, Inc.).

Electrodag 502 (Ted Pella, Inc.).

Conductive coating targets (i.e., Au, Pt/Pd, Ir) (Ted Pella, Inc.).

Argon.

2.3.1. Freeze Drying 2.3.2. Critical Point Drying

2.3.3. Sputter Coating


4

2.4. Sectioning and Staining for Light and Electron Microscopy

Preservation and Preparation of Lignocellulosic Biomass Samples… ●

Diatome ultra 45° diamond knife (Diatome, Hatfield, PA).

Leica EM UTC ultramicrotome (Leica, Wetzlar, Germany).

0.5% Formvar in ethylene dichloride (EMS, Hatfield, PA).

3 mm diameter 2 × 1 mm copper slot grids (SPI Supplies, West Chester, PA).

Diatome cryo 45° diamond knife (Diatome, Hatfield, PA).

Leica EM UTC ultramicrotome with the Leica EM FCS low temperature sectioning system (Leica, Wetzlar, Germany).

0.5% Formvar coated 100 mesh copper grids (SPI Supplies, West Chester, PA).

Dumont non-magnetic reverse action tweezers (Ted Pella, Redding, CA).

2.5% Nonfat dry milk.

0.1% Tween-20 in 1× PBS (PBST) (Sigma-Aldrich, St. Louis, MO).

Primary and secondary antibodies or other labeling probes diluted in 1% milk PBST.

10 and 15 nm gold conjugated secondary antibodies (BBI, Cardiff, UK).

PELCO 22510 grid-staining matrix system for multi-grid staining (Ted Pella, Redding, CA).

Parafilm lined petri dishes for single grid staining.

2% Uranyl acetate in milliQ H2O (EMS, Hatfield, PA).

Reynolds lead citrate stain. See Note 8 for tips on making this stain.

1% Potassium permanganate (KMnO4) in milliQ H2O (SigmaAldrich, St. Louis, MO).

Nikon EZ-C1 acquisition.

Digital Micrograph for TEM image acquisition (http://www. gatan.com/imaging/).

Serial EM software for TEM tilt-series acquisition (http:// bio3d.colorado.edu/SerialEM/).

xT microscope for SEM image acquision (FEI Co.).

xT Docu for SEM image processing (FEI Co.).

IMOD software for TEM tomogram reconstruction (http:// bio3d.colorado.edu/imod/).

OMERO platform for image database (http://www.openmi croscopy.org/site).

2.4.1. Ultramicrotomy 2.4.2. Cryo-sectioning

2.4.3. Immuno-labeling

2.4.4. Grid Staining

2.5. Image Capture, Curating, Processing, and Analysis

35

and

NIS-Elements

for

CSLM

image


36

B.S. Donohoe et al.

Fig. 2. (a) Laboratory microwave with vacuum chamber and water convection platform to minimize temperature gradients within the microwave interior. (b) Commercial freeze-drying apparatus. (c) Custom built critical-point dryer system. Samples previously dehydrated with a transitional fluid such as ethanol are exposed to liquid CO2 that displaces the dehydrant. The chamber is heated until the CO2 reaches its critical point at which the density of the liquid and vapor phases is equivalent. ●

ImageJ software for image processing and analysis (http://rsb. info.nih.gov/ij/).

Adobe Photoshop for montage assembly and image processing (Adobe Systems Inc., SanJose, CA).

Osirix software for 3D visualization and analysis (http://www. osirix-viewer.com/).

Chimera software for 3D visualization and analysis (http:// www.cgl.ucsf.edu/chimera/).

3. Methods 3.1. Fixation and Embedding by Microwave Processing

1. This protocol has been modified from those developed and used by Elaine Humphrey, and Lacey Samuels at the University of British Columbia and others (7–9). See Note 1 for how to choose between chemical and cryo-fixation methods. 2. All steps are carried out in the PELCO laboratory microwave unless stated otherwise (Fig. 2a). 3. Samples should be dissected to less that 2 mm in their longest dimension and placed in 1.8 mL Eppendorf® or other suitable micro tube. 4. Three to four duplicate samples of the same treatment can be processed in the same tube. 5. Primary fixation is carried out in 2.5% gluteraldehyde in 0.1 M sodium cacodylate buffer or 1× PBS buffer. 6. Fixation is done under vacuum, on power level 1, at <30ºC for 6 min total, with the microwave running 2 min on, 2 min off, 2 min on.


4

Preservation and Preparation of Lignocellulosic Biomass Samples…

37

7. Rinse 3× in buffer on power level 2 for 45 s. 8. A secondary fixation step using 1% buffered osmium tetroxide should be used if microwave processing is being used for fresh plant samples but is not necessary for senesced, field-dried, or otherwise pretreated materials. 9. Secondary fixation is carried out in 1% OsO4 in 0.1 M sodium cacodylate buffer or 1× PBS buffer. 10. Fixation is done under vacuum, on power level 1, at <30°C for 6 min total, with the microwave running 2 min on, 2 min off, 2 min on. 11. Replace with fresh fixative and repeat. 12. Rinse 3× in buffer on power level 2 for 45 s. 13. Rinse 3× in water on power level 2 for 45 s. 14. Dehydrate samples in a graded series (30, 60, 90, 3× 100%) of either acetone (for EMbed 812 embedding) or ethanol (for LR White embedding) by adding solvent to achieve the desired concentration for each step. See Note 2 for dehydration issues. 15. For EMbed 812, we use the medium hardness recipe and mix 141 g EMbed 812 resin, 66 g DDSA, and 81 g NMA to make ~250 mL of resin. 16. Aliquot resin into 60 mL syringes, cap, and store at −20°C for future use. 17. Warm resin to RT before use and add DMP-30 accelerator (180 μL per 10 mL resin) to the 3× 100% resin exchanges. 18. Each step is processed on power level 3 for 45 s. 19. Exchange 3× in 100% acetone or ethanol on power level 2 for 45 s. 20. Infiltrate the samples in a graded series (7.5, 15, 30, 60, 90, 3× 100%) of either EMbed 812 (for better structural preservation) or LR White (for better antigen preservation). See Note 3 for resin selection and infiltration tips. 21. Each step is processed under vacuum, on power level 3, for 3 min. 22. Exchange 3× in 100% resin on power level 3 for 3 min. 23. Transfer and orient samples into capsules or molds for polymerization. See Note 4 for sample labeling and capsule/mold selection. 24. Polymerize resin for 48 h in a vacuum oven at 70°C (for EMbed 812) or 60°C (for LR White). 25. LR White polymerizes best if the oven is purged with nitrogen. 26. Remove embedded samples from capsules or molds for sectioning.


38

B.S. Donohoe et al.

3.2. Fixation and Embedding by High-Pressure Freezing and Freeze Substitution

1. This protocol for high-pressure freezing and freeze substitution is based on those developed and used by Mary Morphew, Tom Giddings, and Byung-Ho Kang, at the University of Colorado at Boulder and others (10–13). 2. Samples should be dissected to be less than 1.5 × 0.2 mm under 0.15 M sucrose solutions and placed into the No. 707899 type Leica planchets. 3. Any remaining volume in the planchets is filled with either 0.15 M sucrose in H2O (for stems, shoots) or 1-hexadecene (for leaves, cotyledons). 4. Once the planchets are loaded, the mechanics of high-pressure freezing depends entirely on which model of freezer you have access to. We use the Leica EMPact2 and have found it to be a user-friendly and reliable unit that provides consistent highquality preservation across a range of samples from bacteria, fungi, algae, and plants. See Note 5 for EMPact2 tips. 5. Cryo-preserved samples may be placed in cryo-tube vials and stored in a liquid nitrogen dewar or transferred directly into vials containing 1 mL aliquots of freeze substitution cocktails in cryo vials for further processing. 6. Our freeze substitution is carried out in a Leica AFS2 unit and proceeds through the following temperature regime: See Note 6 for tips on freeze-substitution. (a) Hold at −90°C for 72 h. (b) Ramp to −30°C over 3 h. (c) Hold at −30°C for 21 h. (d) Ramp to 3°C over 3 h. (e) Hold at 3°C for 21 h. (f) Ramp to 24°C over 3 h. (g) Hold at 24°C for ~1 h. (h) Rinse 3× in dry acetone at RT. 7. Samples should be removed from the planchets using finetipped forceps and minimal agitation before proceeding with infiltration. 8. At this point, the samples are ready for infiltration starting at step 20 from Subheading 3.1 above. 9. An exception for freeze-substituted samples is that we do not always use microwave-assisted infiltration, but instead place samples in a rotary mixer between increases in resin concentration over several days. 10. Also, because of the size and shape of samples from the planchets, the Easy Molds usually provide the best sample orientation.


4

3.3. Controlled Drying and Coating for SEM analysis 3.3.1. Freeze Drying

Preservation and Preparation of Lignocellulosic Biomass Samples…

39

1. Freeze drying is utilized when the effects of surface tension from air drying wet samples would otherwise create the deformation and collapse of the inherent sample structure. Water is removed from the frozen sample by sublimation under vacuum and collected on a condensing coil as water vapor, bypassing the transitional liquid phase and surface tension associated with air drying (Fig. 2b). 2. Ready the freeze dryer for operation (the procedure may differ for specific makes and models) but in general ensure the condensing temperature (−55oC to −85oC) and vacuum state (>20 mT) are suitable for sublimation. 3. Quick freezing samples in liquid nitrogen (preferred) or dry ice in ethanol falls into two general categories: (a) Solids that are wet (above ~8 to 10% solids) are frozen in a compatible cryo-container by simply submerging the container and sample into a dewar filled with liquid nitrogen at a slight angle until the violent bubbling ceases. At this point the sample is completely submerged into the liquid nitrogen for several minutes to ensure the sample is completely frozen, allow the excess liquid nitrogen to boil off before attaching to the freeze dryer manifold. (b) Aqueous suspensions of solids are shell frozen by rotating the contained liquid sample in a liquid nitrogen bath slowly allowing a thin layer of frozen sample to develop on the inside surface of the sample container. Continue rotating the sample until the entire liquid sample has frozen in consecutive layers on the inside surface of the container; submerge the container into the bath for several minutes to ensure the entire sample is frozen solid. 4. It is critical that the sample is completely frozen before adding it to the freeze drier manifold. Once the sample has been frozen, quickly add it to the freezer manifold and open that particular manifold vacuum valve and ensure the overall vacuum state recovers to its set point. 5. Samples are considered dry after the sample has completely transitioned to room temperature.

3.3.2. Critical Point Drying

1. Critical point drying (CPD) is utilized to reduce the surface tension of drying to zero, preventing deformation and collapse of specimen structure. Recommended for samples that are sensitive to ice crystal formation and subsequent damage (Fig. 2c). 2. Wash fixed or non-fixed samples in water (5 min; 2×). 3. Dehydrate using a series of ethanol washes (a) 50% ETOH and 50% water (5 min). (b) 75% ETOH and 25% water (5 min). (c) 95% ETOH and 5% water (5 min). (d) 100% ETOH (5min; 3×).


40

B.S. Donohoe et al.

4. Cool the CDP chamber to ~10°C by positioning the chilled water circuit to flow freely. (a) Place the wet ethanol saturated sample into the CPD chamber contained within an appropriate sized CPD specimen holder and secure the threaded chamber end cap. (b) Flood the chamber with liquid carbon dioxide until the entire sample is submerged and allow the sample to equilibrate for 5 min. Then, exhaust the liquid carbon dioxide using the bottom porting valve and exchange the liquid carbon dioxide at least three times. (c) Begin heating the chamber to 31.1°C/1,072 psi by switching the cooling valve circuit over to the heating loop configuration and monitor the temperature and pressure until the critical for the critical fluid has been achieved. Vent the chamber slowly using the upper port valve until the chamber pressure has reached atmospheric pressure. (d) Remove the specimen from the chamber and complete the appropriate sample preparation. 3.3.3. Sputter coating

1. Sputter coating of dried biological specimens with a conductive metal (i.e., Au, Au/Pd, Pt/Pd, or Ir) is a general requirement under high vacuum SEM imaging. Several methods have been developed and should be considered based on the design of the particular plasma sputter coater and choice of conductive coating to be utilized (14, 15). Actual procedures are somewhat dependent on the specific make and model of sputter coater utilized and type a vacuum system employed; refer to the operations manual. 2. Conductive target material should have good electrical conductivity, high secondary electron emission, considered electron dense, non-oxide forming and maintain a nongranular structure as a thin layer. Good choices of conductive coatings include gold (Au) and Pt/Pd as they inherently possess four of the criteria listed; however, at high resolution FE-SEM imaging some granularity may be detected. 3. Samples should be secured to aluminum stubs with doublesided carbon tape in a manner that maximizes contact with the holder. If this proves difficult it will be necessary to paint an electrical circuit between the sample and the grounding stub with colloidal silver paint or electrodag to ensure good electrical conductivity. 4. In general deposit 5–10 nm of conductive metal with the planetary stage tilted roughly at 20–30° from the horizontal plane of rotation.


4 3.3.4. Coating for High-Resolution FE-SEM

Preservation and Preparation of Lignocellulosic Biomass Samples…

41

1. Coating for high resolution may be required if a regular or repeating surface granularity is detected at extremely high resolution imaging régimes. 2. This type of coating requires a pumping speed of at least 300 L/ min at 0.1 mb. Using a turbo pump in series with the standard roughing pump may be necessary to develop a fine conductive metal granularity with the ionizing argon stream. 3. Iridium has proven to be the best conductive coating for FE-SEM work above roughly 100,000× or 250 nm resolution. 4. Ensure sample is quite dry and attached to the carbon tape in a manner that allows for good sample conductivity. 5. Deposit 1–3 nm of iridium and repeat at least once. Several coating iterations may be required to reduce charging effects. 6. The rotation speed of the planetary mixer should be relatively fast and the stage tilted 20–30° from the plane of rotation.

3.4. Sectioning and Staining for Light and Electron Microscopy 3.4.1. Ultramicrotomy

1. These protocols are modified from texts on biological electron microscopy including Electon Microscopy by Bozzola and Russell, and Plant Microtechnique and Microscopy by Ruzin. 2. Embedded samples are cut with glass knives at 60 μm for confocal scanning laser microscopy or SEM analysis, or with a Diatome diamond knife at ~60 nm for standard TEM and immuno-EM analysis, or at ~250 nm for electron tomography all on a Leica EM UTC ultramicrotome. See Note 7 for ultramicrotomy tips. 3. For light microscopy, sections are collected on glass microscope slides. 4. For SEM, sections are collected on aluminum sample stubs with double stick carbon tape. 5. For TEM, sections are usually collected on 0.5% formvarcoated copper slot grids, although formvar-coated 100 mesh grids are occasionally used. 6. Sections may be treated to remove embedding resin prior to immuno-labeling or staining.

3.4.2. Cryo-microtomy

1. As an alternative to resin embedding, we have also employed cryo-microtomy to generate sections for immuno-labeling or chemical microscopy such as confocal Raman. 2. These methods have been adapted from Mark Ladinsky and others (16). 3. In this case, high-pressure frozen samples are kept under liquid nitrogen and taken directly to the microtome to be mounted


42

B.S. Donohoe et al.

Fig. 3. (a) Individual grid staining. Grids are exposed to a drop of the staining solution for the desired time and then rinsed by advancing the grid onto drops of deionized water for ~1 min per drop. (b) A multigrid staining setup allows for a larger number of grids to be stained simultaneously. After exposure to the staining solution bath, grids are rinsed by soaking the holder in 3–4 baths of deionized water for ~1 min per bath.

and sectioned using the Leica EM FCS low-temperature sectioning system. 4. Samples in vitreous ice are cut with a Diatome cryo 45° diamond knife at ~100 nm for immuno-EM or chemical microscopy. 5. Sections are collected on 0.5% formvar-coated copper grids or glass coverslips. 3.4.3. Immuno-labeling

1. Place grid section side down on ~10 μL drops of the following solutions in a Parafilm-lined petri dish (Fig. 3a). 2. 2.5% Nonfat dried milk in PBST (0.1%) for 30 min. 3. 1° Antibody diluted at least 1:10 in1% milk PBST (0.1%) for 1.5 h at RT or overnight at 4°C. 4. Rinse 3 times w/ PBST (0.1%) over 3 min. 5. 2° Antibody diluted at least 1:100 in 1% milk PBST (0.1%) for 1.5 h at RT or overnight at 4°C. 6. Rinse 3 times w/ PBST (0.1%) over 3 min. 7. Rinse 3 times w/ H2O over 3 min.

3.4.4. Grid Staining

1. Multigrid vs. single grid staining (Fig. 3). See Note 8 for staining issues. 2. We have found that 0.5–2 wt% solutions of potassium permanganate (KMnO4) to be effective for staining biomass cell walls. 3. Exposure times of 2–4 min are used for sections 50–100 nm thick, while sections of 200 nm and thicker require 10–12 min of exposure to the solution in order to achieve adequate staining throughout the sample.


4

Preservation and Preparation of Lignocellulosic Biomass Samplesâ&#x20AC;Ś

43

4. Grids are then rinsed in three consecutive water baths for at least 1 min in each bath. 5. Additional rinsing is performed by holding the grid in reverseaction forceps while a gentle stream of water is applied to the forceps several centimeters above the tip of the forceps. 3.5. Image Capture, Curating, Processing, and Analysis

1. Bright-field, epi-fluorescence, and CSLM images are taken with a SPOT camera and software or the Nikon EZ-C1 software on a Nikon E800 microscope equipped with the Nikon C1 confocal system or a Nikon C1Si spectral confocal system. 2. SEM images are captured with a Everhart-Thornley secondary electron detector on an FEI Quanta 400 FE-SEM. 3. TEM images and tilt series for tomographic reconstruction are taken with a 4 mega-pixel Gatan UltraScan 1000 camera (Gatan, Pleasanton, CA) on an FEI Tecnai G2 20 Twin 200 kV LaB6 TEM driven by Digital Micrograph standard imaging or Serial EM for tilt series acquisition. 4. We curate our images in an image database using The Open Microcopy Environmentâ&#x20AC;&#x2122;s OMERO platform. 5. Image processing and analysis is carried out using a variety of software and analysis tools. The most commonly used are ImageJ, Adobe Photoshop, IMOD, and Chimera. 6. Electron tomogram reconstruction and analysis is done with the IMOD software package and the eTomo GUI. 7. EDS overlay. 8. See Note 9 for issues with image analysis and interpretation.

3.6. Notes

1. Microwave processing is chosen for biomass samples that have senesced, been field dried, milled, and/or pretreated in such a way that only the structure of the intact cell walls is to be analyzed. One of the concerns with extensively pretreated samples is that there may not be much protein left to cross-link during fixation, however in our hands, including the fixation steps yields better structural preservation. High-pressure freezing and freeze substitution is a superior preservation method and is the technique of choice for live or fresh plant materials, or samples that include living microbes or active enzyme complexes that need to be visualized in their native state (17). 2. There is much discussion in the field about the impact of drying on cell wall structure and indeed water loss can cause a collapse of ultrastructure and a decrease in porosity. However, any biorefinery relevant biomass material will be dried, usually on the field, prior to harvesting. Techno-economics prohibit the transportation of wet biomass. Also, there are indications that after the initial drying event, most walls are not rehydrated to


44

B.S. Donohoe et al.

Fig. 4. TEM micrographs of corn stover cell walls from well infiltrated (a) and poorly infiltrated (b) samples. Poor infiltration (b) results in the embedding resin pulling away from biomass cell walls leaving white spaces along the surface (arrow ). Also, insufficient infiltration leads to more wrinkling distortion artifacts within cell walls (arrowhead ). These problems can be solved by slow, gradual resin infiltration over several days.

a native state. Therefore, the solvent-based water replacement used in sample preparation for microscopy does not alter the structure of biomass cell walls beyond what has already taken place during cell development or senescence and represents an authentic condition for biomass. However, for microscopy, dry biomass samples should be soaked in water overnight and even under vacuum to remove trapped air that will inhibit fixation and infiltration. 3. Biomass samples must be embedded prior to sectioning if the structure is going to be preserved. Sectioning unsupported samples, especially if they are pretreated or digested, creates mechanical stress that can be misinterpreted as an effect of the pretreatment or digestion. In some samples, the damage may be isolated to the surface that was cut, but in the case of plant tissues, where cells can be several hundred micrometers long, the damage is usually propagated well into the sample (Fig. 4). 4. Samples can be labeled by printing the sample codes in very small ~6 point fonts on a printer, cutting the labels out, and rolling them into the capsules. Also, using a font like Helvetica Neue Ultra Light will use a minimum amount of toner that can otherwise sometimes dislodge from the paper and settle onto the bottom of the block. This method creates a permanently affixed label. Similar, but smaller labels can also be placed into flat embedding molds. 5. We usually freeze 3â&#x20AC;&#x201C;4 planchets of each sample during a highpressure freezing session. The sample pods and bayonet holders are rinsed in methanol and thoroughly dried between freeze runs. Be sure to precool all forceps or other tools before handling cryopreserved samples.


4

Preservation and Preparation of Lignocellulosic Biomass Samplesâ&#x20AC;Ś

45

Fig. 5. TEM micrographs of dilute acid pretreated corn stover cell walls. The highest contrast imaging of biomass samples has been achieved by staining grids with potassium permanganate in addition to the traditional contrasting agents (OsO4, uranyl acetate) used during freeze substitution. This stain has been especially effective for tracking lignin relocalization within cell walls. The large globules seen within the cell lumen in image (a) are the lignin globules.

6. We usually place the cryo-tube vials containing freeze-substitution cocktails into the AFS unit and then transfer the HPF planchets by pouring them directly into the new cryo-tubes. While the freeze-substitution protocol outlined here may seem long compared to protocols for other types of specimen, a shorter protocol will yield unsatisfactory results for most plant samples. 7. High quality fixation and embedding with complete infiltration generates samples that are relatively easy to section on the ultramicrotome. Smaller, more perfectly trimmed trapezoid faces are easier to section. LR White-embedded samples are more challenging to section and benefit from a lower water level in the diamond knife boat to minimize water adhering to the block face. If we have a problem with section compression, we use chloroform vapors to flatten sections. 8. The main challenge in staining biomass samples is that the major component, cellulose, does not stain well if at all. This means that the staining of the hemicellulose sheath surrounding the microfibrils usually negatively stains the cellulose. On the other hand, lignin stains very well, especially with KMnO4 (Fig. 5). Our routine has been to stain dried, microwave processed biomass sections with uranyl acetate and KMnO4 and live, high-pressure frozen samples with uranyl acetate and lead citrate. We use a combination staining in a multigrid stainer for survey imaging of larger sample sets and single grid staining on drops for blocks that are already known to be high quality. 9. One of the biggest challenges in interpreting biomass cell wall ultrastructure and surface analysis in bioconversion is realizing


46

B.S. Donohoe et al.

that entire classes of tissue and cell types can be removed by processing. The untrained eye can easily misinterpret a different cell type for a changed cell. Another issue is distinguishing any disruption caused by the mechanical milling of samples from the chemical disruption caused by pretreatment and digestion. Similarly, in fresh samples, it can be difficult to distinguish plant cellular debris from deconstruction debris. The only way this can be overcome is by having the right controls and spending enough time on the microscope to become familiar and confident in the inter- vs. intra-sample variability. Direct observation and microscopic analysis does not always lend itself to statistical analysis and often relies on considerable experience and expertise of the operator.

Acknowledgements This work was supported by the US Department of Energy Office of the Biomass Program (BSD and TBV) and the Center for direct Catalytic Conversion of Biomass to Biofuels (C3Bio), an Energy Frontier Research Center funded by the DOE, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC0000997 (BSD and PNC). NREL is a national laboratory of the US Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC. References 1. Zeng MJ, Mosier NS et al (2007) Microscopic examination of changes of plant cell structure in corn stover due to hot water pretreatment and enzymatic hydrolysis. Biotechnol Bioeng 97:265–278 2. Donohoe BS, Decker SR et al (2008) Visualizing lignin coalescence and migration through maize cell walls following thermochemical pretreatment. Biotechnol Bioeng 101:913–925 3. Donohoe BS, Selig MJ et al (2009) Detecting cellulase penetration into corn stover cell walls by immuno-electron microscopy. Biotechnol Bioeng 103:480–489 4. Singh S, Simmons BS et al (2009) Visualization of biomass solubilization and cellulose regeneration during ionic liquid pretreatment of switchgrass. Biotechnol Bioeng 104:68–75 5. Chundawat SPS, Donohoe BS et al (2011) Multi-scale visualization and characterization of lignocellulosic plant cell wall deconstruction during thermochemical pretreatment. Energy Environ Sci 4:973–984

6. McCann MC, Carpita NC (2008) Designing the deconstruction of plant cell walls. Curr Opin Plant Biol 11:314–320 7. Chebli Y, Daher F et al. (2008) Microwave assisted processing of plant cells for optical and electron microscopy. Bull Micro Soc 36(3):15–19 8. Humphrey E (2005) Microwave processing in a modern microscopy facility. Scanning 27: 74–74 9. Kaneda M, Rensing KH et al (2008) Tracking monolignols during wood development in lodgepole pine. Plant Physiol 147:1750–1760 10. Giddings TH (2003) Freeze-substitution protocols for improved visualization of membranes in high-pressure frozen samples. J Microsc Oxf 212:53–61 11. Hess MW (2007) Cryopreparation methodology for plant cell biology. Cell Electron Microsc 79:57–100 12. Kang BH (2010) Electron microscopy and high-pressure freezing of Arabidopsis. Electron Microsc Model Syst 96:259–283


4

Preservation and Preparation of Lignocellulosic Biomass Samples…

13. McDonald K, Morphew MK (1993) Improved preservation of ultrastructure in difficult-to-fix organisms by high-pressure freezing and freeze substitution 1. Drosophila melanogaster and Stronglyocentrotus purpuratus Embryos. Microsc Res Techn 24:465–473 14. Echlin P (1978) Coating techniques for scanning electron microscopy and X-ray microanalysis. Scanning Electron Microscopy I, 109–132

47

15. Echlin P (1981) Recent advances in specimen coating techniques. Scanning Electron Microsc I:79–90 16. Ladinsky MS, Pierson JM et al (2006) Vitreous cryo-sectioning of cells facilitated by a micromanipulator. J Microsc Oxf 224:129–134 17. Gilkey JC, Staehelin LA (1986) Advances in ultra-rapid freezing for the preservation of cellular ultrastructure. J Electron Microsc Tech 3: 177–210


Chapter 5 Coherent Raman Microscopy Analysis of Plant Cell Walls Yining Zeng, Michael E. Himmel, and Shi-You Ding Abstract Coherent Raman scattering (CRS) microscopy is a label-free method for chemical imaging, as it offers chemical specificity with orders of magnitude better sensitivity than the state-of-the-art confocal Raman scattering microscopy. Currently CRS technique includes coherent anti-Stokes Raman scattering (CARS), and stimulated Raman scattering (SRS). This chapter describes the methods of using CRS microscopy to image major polymers in plant cell wall (i.e., lignin and cellulose). This method can also be used to realtime monitor the chemical processes involved in biomass pretreatment. These together demonstrate CRS as an effective method for imaging complex chemistry in biological systems. Key words: Plant cell wall, Biofuels, Raman spectroscopy, Coherent Raman scattering, Coherent anti-Stokes Raman scattering miscroscopy, Stimulated Raman scattering microscopy, Lignin, Cellulose

1. Introduction Biomass is receiving more and more attention as a sustainable source for energy, fuels, and chemicals (1). Biomass materials are typically plant cell walls mainly composed of lignin and various polysaccharides that form a complex three-dimensional matrix (2). In the current biomass conversion process, cell wall polysaccharides chains are broken down into sugars and sugars will be later fermented to produce ethanol as a fuel source, or used as a source for chemicals. Lignins are branched polymers that are composed of hydroxyphenylpropanoid units which polymerized from three primary precursors, coumaryl alcohol, coniferyl alcohol, and sinapyl alcohol. In the plant, lignin provides the necessary mechanical support for stem strength and general cell integrity, and acts as a water barrier in lignified tissues (3, 4). In the biomass conversion process, lignins contribute for biomass recalcitrance (5). It is believed that lignin contributes to the inherit biomass recalcitrance partly

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_5, Š Springer Science+Business Media, LLC 2012

49


50

Y. Zeng et al.

by blocking the access of enzymes to cell wall polysaccharides and therefore limiting the saccharification efficiency (6). This biomass recalcitrance contributes mostly to the relatively high cost of the chemical/biological processes that convert the insoluble polymeric plant cell wall materials into soluble sugars (7). To lower the cost, deeper understanding of the cell wall chemical structure and its role in the process is needed in order to develop more effective methods to digest the biomass. In this sense, chemical imaging techniques that can specifically locate polysaccharides (as the energy carrier) and lignin (as the contributor to recalcitrance) will be helpful for tracing the kinetics of the biomass conversion process. For example, removal of lignin is believed to improve the efficiency biomass conversion. Although it is still not clear how the amount of lignin and its composition affect the hydrolysis of biomass in general, it has been indeed demonstrated by the pulp and wood industry that the cellulose yield can be improved from the down-regulating genes encoding enzymes of monolignol biosynthesis (8). Chen (9) has demonstrated that down-regulated lignin modification in alfalfa (Medicago sativa L.) plants show improves sugar yields during the saccharification processes for bioethanol production. In their study, the lignin biosynthesis pathway is down-regulated independently at six different enzymatic steps in alfalfa. It has also been shown that the lignin content is negatively correlated to the levels of sugars released by enzymatic hydrolysis. Therefore, it is important to use effective chemical imaging techniques that can determine how the lignin distribution in the cell wall is affected by the down-regulation of lignin biosynthesis. Besides lignin, the other major chemical species of interest in the conversion process such as cellulose and hemicelluloses can also be studied. Because of the recalcitrance, usually a thermochemical pretreatment process is necessary in current biomass conversion technology. This process uses harsh chemical conditions and relatively high temperatures to remove or modify lignins and hemicelluloses structure, therefore improve the accessibility for the digestive enzymes used in the saccharification process (5, 10, 11). Interestingly, the lignin amount is not always the less the better. Complete removal of lignins can actually reduce saccharification yield because the enzyme digestibility decreases due to cellulose aggregation (12). Therefore, careful control of the process is critical. To optimize the overall conversion efficiency, improved understanding in the change of polysaccharides and lignin during the hydrolysis kinetics is critical. The above challenges request imaging techniques that have high sensitivity, high chemical specificity, and high spatial resolution. In addition, few or no modification of the substrate is usually preferable. In order to study the conversion process in situ, the sensitivity should be higher enough to achieve the speed for realtime monitoring. Infrared spectroscopy have very good chemical


5

Coherent Raman Microscopy Analysis of Plant Cell Walls

51

resolution, but the microscopy based on infrared absorption has limited spatial resolution because the long infrared wavelengths and penetration depth into aqueous plant samples is limited (13). Raman microspectroscopy also offers rich chemical information with high spatial resolution (14, 15). However, the Raman scattering effect in nature is weak. In order to achieve decent image quality, long pixel dwell times (on the order of 0.1–1 s) are often required. This means that the imaging acquisition rate would be too slow to carry out any real-time imaging measurements. Consequently, with those traditional chemical imaging microscopic tools the dynamic processes involved in the conversion cannot be followed at high spatiotemporal resolution. Coherent anti-Stokes Raman scattering (CARS), a nonlinear optical microscopy based on two pulsed lasers, has been demonstrated to be a powerful tool for chemical imaging. Like traditional Raman spectroscopy, CARS microscopy provides a contrast mechanism also based on molecular vibrations. Therefore, it does not need any additional labeling to the sample. Due to its nonlinear optical nature, the signal is supposed from only the focus. CARS therefore also provides high spatiotemporal resolution, and is free of background from one-photon-excited fluorescence (16). CARS technique uses two picoseconds laser pulses, pump (wp), and Stokes (ws) beams that are temporally and spatially overlapped to generate anti-Stokes signals at frequency was = 2wp − ws. When the frequency difference (wp − ws) is tuned to match a particular Raman-active vibration frequency, the resonant anti-Stokes signal at frequency was = 2wp − ws is orders of magnitude greater than that from spontaneous Raman scattering. Compared to traditional Raman, CARS offers much greater sensitivity and faster acquisition time (17). Although CARS microscopy offer label-free chemical imaging with high sensitivity and high spatial resolution, its signal also suffers from a nonresonant electronic background component that can to certain degree distort the chemical information. Recently stimulated Raman scattering (SRS) has been developed as an effective tool for chemical imaging. SRS microscopy offers chemical contrast still based on the intrinsic Raman vibrational responses of molecules in the sample, similar to ordinary spontaneous Raman scattering. However, because of the stimulated excitation in combination with a high frequency modulation and lock-in detection scheme, it offers orders of magnitude stronger signal than spontaneous Raman. This means that high quality images with high sensitivity and intrinsic chemical contrast can be acquired at high speed. Compared to the CARS technique that uses almost the same laser excitation scheme, SRS offers the major advantages that it is free of the nonresonant background and its signal intensity is linearly dependent on the analyte concentration only. These merits make background-free, quantitative imaging possible. SRS has identical spectral as spontaneous Raman scattering, and this will


52

Y. Zeng et al.

allow for straightforward comparisons to the available Raman spectra for identifying chemical species of interests. Like CARS, SRS microscopy uses two near infrared laser pulse trains (wp and ws) that are overlapped in time and space and focused onto the sample of interest by a laser scanning microscope with a high numerical aperture objective lens. When the difference frequency between the two beams (wp − ws) matches a vibrational resonance intrinsic to the sample, the vibrational transition rate is enhanced due to stimulated excitation. This process is accompanied by energy transfer from the pump beam (wp) at higher frequency to the Stokes beam (ws) (18). Importantly, this energy transfer only occurs when the frequency difference (wp − ws) matches an intrinsic Raman-active vibrational mode in the sample. SRS signal is measured by this amount of energy transfer. The energy transfer is relatively small (less than 10−3 compared to the laser beam) and is buried in the noise of the laser system. To detect the small signal level, an amplitude modulation to the laser beam combined with a lock-in detection is used (19). Raman spectra of the cell walls from different plant species have been extensively studied, and major Raman bands have been assigned to the major plant cell wall polymers. Lignin structure contains high concentration of phenyl units (20) that show a strong Raman band at 1,600 cm−1 due to its symmetric aromatic ring C=C stretch (12, 21). This strong Raman band is traditionally used as a characteristic Raman response for lignin. In this chapter, we demonstrate using CARS to quantify the relative lignin content in the alfalfa lines wild-type (wt) and two previously generated transgenic alfalfa lines, in which hydroxycinnamoyl CoA: shikimate hydroxycinnamoyl transferase (HCT) or coumaroyl shikimate 3-hydroxylase (C3H). We also demonstrate using two-color SRS to simultaneously monitor the lignin and cellulose content during the acidic pretreatment of corn stover chips.

2. Materials 2.1. Plant Cell Walls Materials

The wild-type and the transgenic down-regulated alfalfa lines in HCT and C3H are kindly provided by Rick Dixon’s group at Noble Foundation. The plants are grown in the greenhouse. For details of the transgenic HCT and C3H lines and growth condition, please refer to previous publication from Chen and Dixon (9). The fresh plants are harvested and the stems are cut off and stored at −80°C. The corn stover sample used to study the pretreatment kinetics by SRS is from the field-dried Mo17 maize harvested from Madison, WI in the late fall of 2003.


5

Coherent Raman Microscopy Analysis of Plant Cell Walls

53

2.2. Reagents and Disposals

Deionized (DI) water is used in all experiments. #1 coverslip (25 × 50 mm) are from VWR (vwr.com). HCl (aq) and NaClO2 solution in the lignin blenching experiments are 0.1N and 10%, respectively, in DI water.

2.3. CRS Microscope

The CRS imaging system is a custom-built microscope (Zeiss LSM5-MP/DuoScan or Olympus IX71/Fluoview). The laser source is a mode-locked Nd:YVO4 laser (High Q Laser US, Inc.). The two optical parametric oscillators (OPOs) used to generate the pump beams are from Levante Emerald, APE-Berlin.

3. Methods 3.1. Preparation of Plant Cell Wall Cross-Section

Plant sample was cut into slices 25–50 μm thick. These crosssections were gently washed by distilled water prior to CARS measurement. For CARS measurement, the stem slices were soaked in distilled water and placed between two #1 glass coverslips. For real-time observation, plant internode sections are taken from the upper one third of the stem and transversely cut into 150 mm slices on rotary microtome. For static imaging, the 150 mm thick sample slice is placed in 1× PBS solution between two #1 coverslips. The coverslips are attached by double-sided tape to seal the sample. This structure also forms a flowing chamber as the channel between the two double-side tapes as functions a flow channel to allow solution pass through. Two holes are drilled on the top of the coverslip as solution in/outlet and each hole glued with a Teflon tube to deliver solution. Initially, the sample slice is imaged in phosphate buffer solution in the flow channel between the two coverslips. To initiate the delignification reaction, a solution of HCl and NaClO2 was mixed immediately before the inlet and delivered into the channel.

3.2. CRS Microscopy

A mode-locked Nd:YVO4 laser was used to generate a 7 ps, 76 MHz pulse train of both 1,064 nm (1 W average power) and 532 nm (5 W average power) laser beams. The 1,064 nm beam was used as the CARS Stokes beam. The 532 nm beam was used to pump an OPO in order to generate the CARS pump beam, which was a 6 ps pulse train. The OPO provides a tunable pulse train that is locked in time to the pump laser pulse train. The tuning range is from 680 to 1,010 nm, and ~1 W average power can be obtained from each OPO. In order to measure the CARS signal from lignin’s aromatic ring vibration frequency at 1,600 cm−1, the wavelength of the pump beam is set to 910 nm. The pump and Stokes pulse trains were overlapped in time using a delay stage and sent collinearly into the microscope. Conventional beam scanning confocal microscopes can be modified as a CARS microscope. The Zeiss


54

Y. Zeng et al.

LSM5-MP/DuoScan microscope and the Olympus IX71/ Fluoview microscope are the common ones for this purpose. In the Zeiss LSM5-MP/DuoScan microscope for example, the two overlapped beams were sent to a Zeiss C-Apochromat 40 × 1.1 UV-VIS-IR water immersion objective lens via a pair of scanning mirrors and then focused onto the sample through the objective lens. The excitation power after the objective was usually kept at ~100 mW for the pump and the Stokes beam. When the laser power is too high, the samples slice might be damaged by the strong light intensity at the focus spot. The anti-Stokes light or the CARS signal can be collected in the Epi direction via the objective lens or/and the forward direction via the microscope air condenser. The anti-Stokes light was separated from the pump and the Stokes beams by a bandpass filter (for example, to separate the CARS signal for lignin by using the HQ750/210, Chroma Technology) in the forward CARS direction and by a long-wave pass dichroic mirror (880DXXR, Chroma Technology, which reflects wavelengths shorter than 870 nm and transmits wavelengths longer than 890 nm) in combination with a bandpass filter (HQ750/210, Chroma Technology) for the Epi-CARS channel. The residual two-photonexcited fluorescence signal in the epi direction was removed by a narrowband clean-up filter (HQ800/40m, Chroma Technology). The two-color SRS imaging instrument uses the same laser and OPO except that two OPOs are used to provide two pump beams at two colors. About 2 W the 1,064 nm beam from the Nd:YVO4 laser is used directly as the Stokes beam. The remaining laser power, typically 9–10 W, is allocated to the 532 nm beam, which is split evenly by a non-polarizing beam splitter into two beams each to independently pump the synchronously pumped OPO. In the following example we show using two-color SRS to independently imaging lignin and cellulose, and the pump beam from one OPO is tuned to 910 nm in order to match the lignin resonance at 1,600 cm−1 when combined with the 1,064 nm Stokes beam. The signal wave from the other OPO is tuned to 953 nm so that when combined with 1,064 nm beam it will be in match of the cellulose vibration frequency at 1,100 cm−1. Then the two pump beams are combined in space via a dichroic mirror (q930lp, Chroma Technology) that reflects wavelengths shorter than 925 nm and transmits wavelengths longer than 940 nm. The 953 nm beam also passes a delay stage which varies its pulse delay to overlap the pulse of the 910 nm. The intensity of the 1,064 nm beam is modulated by an acoustic optic modulator (3080-122, Crystal Technology) by using a 10 MHz square wave from a function generator (DS345, Stanford Research Systems). The 1,064 nm beam is also combined in time (by using another delay stage) with the combined pump beams (910 and 953 nm beams) and in space on a dichroic mirror (1064dcrb, Chroma Technology). All three beams are directed into an Olympus laser scanning microscope scanning unit


5

Coherent Raman Microscopy Analysis of Plant Cell Walls

55

(BX62WI/FV300, Olympus). After that, the three beams are sent to a high numerical aperture water-immersion objective (UPlSApo 60X 1.20 NA W, Olympus) and focused down to a diffractionlimited spot typically 350–400 nm. The light transmitted through the sample is collected by an oil-immersion condenser (1.45 NA O, Nikon) with an even higher numerical aperture to minimize the cross phase modulation effect. An optical filter (CARS980/220, Chroma Technology) is used to block the 1,064 nm beam completely. To separate the 910 and 953 nm beams, a dichroic mirror identical to the beam combiner (q930lp, Chroma Technology) is used. The two pump beams are focus independently to a large area silicon PIN photodiode (FDS1010, Thorlabs) back-biased at 70 V. A narrowband filter centered at each of the two pump beam wavelengths was placed directly in front of each photodiode. The signal output from each photodiode is filtered by a 9.2–10.5 MHz frequency bandpass filter (BBP-10.7+ MiniCircuits) and sent into a lock-in amplifier (SR844, Stanford Research Systems) referenced to the amplitude modulation frequency. The DC charge build-up is eliminated by a 50 Ω terminator. The output demodulated SRS signal (as R from the lock-in amplifier) is sent to the A/D converter of the microscope. Depending on the scan rate and image pixel resolution, the sensitivity on the lock-in amplifier is set in the range of 100 μV to 1 mV. The reference phase of the lock-in amplifier can be slightly adjusted to yield maximum contrast.

4. Results Proteins such as phenylalanine and tyrosine residues in cell wall proteins also have symmetric phenol ring structures which would show Raman signal similar to lignin. However, because they are in much lower concentration as compared to lignin, the CARS signal at 1,600 cm−1 contributed by proteins should be negligible. This is indeed confirmed by the fact that in some protein-rich tissues such as phloem no significant CARS signal was observed. The specificity of the CARS signal was proven by alternately blocking the pump beam and the Stokes beam. This procedure proves that the multiphoton fluorescence intensity generated by the pump beam or the 1064 nm Stokes beam are much smaller as compared to the CARS signal. The CARS signal is supposed to be generated from the tightly focused focal spot. Practically, the spatial resolution is usually ~300–400 nm. The signal is generated in the small focal volume and hence is dependent on the local molecular density, or concentration of molecular oscillators within the focal volume. The total sample thickness would not matter. The CARS signals measured in this type of measurement, therefore, could be considered relatively quantitative. Photobleach, which is a common issue in fluorescence


56

Y. Zeng et al.

imaging, is not a problem for CARS imaging. Plant cell wall materials have only small absorption at the IR wavelengths which will minimize photodamage. In fact, there was no significant photo damage observed after hundreds of scans when both pump and Stokes beams are around 100 mW. The CARS signal was consistent after 10 min of continuous image scanning. By taking CARS images at different cell types across the stem, from epidermal tissue to the pith, one can see that the CARS signal was primarily detected from cell walls that are heavily lignified such as the outer wall of the epidermis, or the vascular tissues with thick walled elements or those fibers at vascular and interfascicular sclerenchyma region. In general, for the wild-type alfalfa plant, the CARS intensity was found to be in the following order: fiber (highest) > xylem > epidermis > phloem > parenchyma; and at the cellular level in the order: cell corner (CC, highest) > compound middle lamella (CML, including middle lamellae and primary cell walls from adjacent cells) > secondary wall (SW). Within the secondary wall region, the CARS signal appeared homogenously distributed on both sides of the CML, which appeared as two flat shoulder peaks. In the pore areas, the pit membranes appeared to contain less lignin than that in the CML (Fig. 1). This provides direct visualization to support the Donaldsonâ&#x20AC;&#x2122;s argument that in Pinus radiata lignification is initiated in the central zone of the cell corner and CML areas (22). Compared to the wild-type control, the lignin down-regulated lines also exhibit similar patterns of lignin distribution across the stem. The signal intensity distribution in the cell corner, CML, and secondary all regions of the interfascicular fiber cells are particularly interesting. Figure 1 shows the line profiles of representative fiber cell walls. In the two down-regulated lines HCT and C3H, the signal intensity distribution followed the same pattern as in the wildtype control. There is no significant change in the wall thickness of the lignified cell wall types. In the wide type plant sample, the intercellular spaces at adjacent cells were mostly filled with lignified materials, whereas the triangular shape of the intercellular space was clearly visible in the down-regulated lines. In the cell corner area, the lignin CARS signal intensity was reduced more than that in the secondary wall of both the HCT and C3H down-regulated lines. For example, compared to wild-type, the lignin CARS signal intensity in both HCT and C3H lines are reduced mostly in the cell corner area; whereas in the CML and SW areas, CARS signal was also reduced. In the C3H line, a slightly greater reduction in CARS signal intensity occurs in the CML and secondary wall than that in the HCT line. Clearly, the lignin CARS signal in the down-regulated lines was not reduced equally in different areas of the cell walls. Previous studies have shown that HCT and C3H downregulated alfalfa lines have more accessibility to acids and enzymatic hydrolysis (9). However, there are also studies that show


5

Coherent Raman Microscopy Analysis of Plant Cell Walls

57

Fig. 1. (Top ) CARS micrographs of lignin distribution in cross-sections of wild-type (wt) and lignin-down-regulated alfalfa lines (HCT and C3H). (Bottom) Corresponding line profiles of CARS intensities show different lignin contents across the cell wall layers, in the direction shown by the blue arrows. The reduction of the CARS signal occurs specifically in CC and CML areas. CML = compound middle lamella (middle lamella + primary walls) and SW = secondary wall. The CARS images were taken in 130 × 130 μm scans and cropped to show the representative cells. From Zeng et al. (24), Copyright © Springer for BioEnergy Res 3: 272–277 (2010). DOI 10.1007/s12155-010-9079-1. Reproduced with permission.

down-regulation of lignin negatively affects plant performance and decreases biomass yields. The mechanism between the hydrolysis process and lignin content or lignin distribution is still remained unknown. It is believed that during the biomass conversion process, the transport of chemicals/enzymatic within plant tissue is the rate-limiting step that significantly affects the efficiency of hydrolysis. It has been reported that liquid transport in corn stover primarily follows the channels between the intercellular space (cell corners) and fissures that are formed during natural senescence and material treatment (23). One hypothesis might be that chemicals and enzymes accumulate and initiate action at the intercellular space, such as in the cell corner. After that, hydrolysis happens from middle lamella to the primary cell wall, and finally to the secondary cell wall. Therefore, it is possible that the reduction of lignin in the cell corner area could enhance the accessibility of catalysts, thus improving the hydrolysis. In this regard, imaging of lignin by CARS could provide a direct visualization approach for biomass evaluation.


58

Y. Zeng et al.

Fig. 2. Real-time SRS imaging of a delignification reaction in corn stover. All images were taken in the same vascular bundle. (a) Lignin signal at 1,600 cm−1 before beginning the reaction. (b) The cellulose signal at 1,100 cm−1 before beginning the reaction. (c) Lignin signal after a 53 min time course of acid chlorite treatment showing significant reduction (more than eightfold) compared to (a). (d) Cellulose signal after treatment remains roughly the same as in (c). (e) False color heat map of the reaction rate constant, obtained by fitting the time series of the lignin decay in the reaction (the initial and final points are shown in images (a) and (c)) to a single exponential. Faster reaction rates are observed in the phloem and on the edges of the cells compared to the centers of the cell walls. The rate constant (in s−1) color scale is shown in the bottom left corner. Representative time traces (red dots) and single-exponential fits (blue lines) from four locations labeled as green spots in (e) representing a phloem element (f), vessel (g), fiber (h), and background with no plant cell wall (i) in the corn stover sample. The image in part (e) consists of 256 × 256 pixels, each of which has an associated single exponential decay fit to obtain the rate constant (in s−1). We observed no decay in background, rapid decay in the phloem, and slower decay in the fiber and vessel cells. The time to acquire this data is about 8 s/frame and the spatial resolution is 900 nm (limited by the sampling of the images). This data demonstrates the capability of SRS to perform real-time monitoring of the acid chlorite pretreatment reaction in fresh, unstained plant materials. Scale bars, 40 μm. From Saar et al. (25), Copyright © Wiley-VCH Verlag GmbH & Co. KGaA for Angew Chem Int Ed 49: 5476–5479 (2010). Reproduced with permission.

In principle cellulose can also be imaged by CARS when the pump beam is tuned to 953 nm. But the problem of this is that the anti-Stokes signal shifts more to the longer wavelength region, where the PMT detector has significantly lower efficiency. SRS solves this problem as it uses a photodiode with pretty flat spectral response. Figure 2 shows the SRS images of lignin and cellulose simultaneously acquired with a total pixel dwell time of 50 s. In comparison, a Raman imaging of lignin and cellulose in poplar wood by traditional confocal Raman employed pixel dwell times of 1 s (21), which is more than four orders of magnitude slower. This indicates that dynamic studies with high time resolution and spatial resolution can be performed with SRS microscopies that are not possible in traditional confocal Raman studies. In Fig. 2, the parenchyma cells show very weak lignin signal. Phloem cells also have pretty low lignin content. The vessel, tracheid, and fiber cells are observed to be highly lignified. The annular rings show intermediate lignin content. In the large vessel cell, the lignin distribution appears to vary by more than a factor of three on different sides of the wall: the side adjacent to the tracheids appears to have less


5

Coherent Raman Microscopy Analysis of Plant Cell Walls

59

lignin. A possible explanation could be that these tracheids help control the lateral distribution of water from vessels across the stem. Figure 2a–d shows images of the lignin and cellulose channels before and after the delignification reaction. It is clear that the lignin signal decreases dramatically, while the cellulose signal remains. The time trace of the reaction at each point in the sample is extracted from the sequence of images with submicron resolution. This time trace could then be fitted with a single exponential to determine an effective reaction rate at each location, which is plotted as a heat map in Fig. 2e. Representative traces from different locations in the sample are shown in Fig. 2f–i, which clearly demonstrates that the reaction rate is heterogeneous even on the micron scale. The heat map (Fig. 2e) shows that the phloem region has the most rapid bleaching kinetics of lignin. The bleaching rate is higher on the edges of the plant tissue than at the center of the thicker cell wall, possibly because the edges are more exposed to the surrounding solvent. In general all the tissue types show quite similar bleaching kinetics, suggesting that the accessibility of the lignin to the bleaching reagents is similar.

5. Notes To minimize possible impact to sample’s chemical composition, no chemical bleach (usually to remove cell pigments) is applied to the sample. For optical microscopic observation, plant cell wall samples need to be sectioned into slices and place on glass slides. In order to avoid the subsequent solvent extraction which might also introduce any potential impact to the cell wall composition, all the samples are cut into slices without any embedment. Plant stem samples are transversely cut into fairly thick 50–100 μm slices by a rotary microtome (RM2255, Leica Microsystems Inc.). The preparation of the plant cell wall cross-section is important as improper sectioning will damage the cell wall structure. Since the samples are not embedded prior to slicing on microtome, sometimes rotary microtome is not good option at making thin slices. In this case, one can also try hand cut the samples.

Acknowledgements This work was supported by the US Department of Energy, the Office of Science, Office of Biological and Environmental Research through the BioEnergy Science Center (BESC), a DOE Bioenergy Research Center.


60

Y. Zeng et al.

References 1. Perlack RD, Wright LL, Turhollow AF, Graham RL, Stokes BJ, Erbach DC (2005) Biomass as feedstock for a bioenergy and bioproducts industry: the technical feasibility of a billionton annual supply. DOE/GO-102005-2135, ORNL/TM-2005/66 2. Ding S-Y, Himmel ME (2006) The maize primary cell wall microfibril: a new model derived from direct visualization. J Agric Food Chem 54:597–606 3. Chabannes M, Barakate A, Lapierre C, Marita JM, Ralph J, Pean M, Danoun S, Halpin C, Grima-Pettenati J, Boudet AM (2001) Strong decrease in lignin content without significant alteration of plant development is induced by simultaneous down-regulation of cinnamoyl CoA reductase (CCR) and cinnamyl alcohol dehydrogenase (CAD) in tobacco plants. Plant J 28:257–270 4. Jones L, Ennos AR, Turner SR (2001) Cloning and characterization of irregular xylem4 (irx4): a severely lignin-deficient mutant of Arabidopsis. Plant J 26:205–216 5. Jamin N, Dumas P, Moncuit J, Fridman W-H, Teillaud J-L, Carr GL, Williams GP (1998) Highly resolved chemical imaging of living cells by using synchrotron infrared microspectrometry. Proc Natl Acad Sci U S A 95:4837–4840 6. Himmel ME, Ding S-Y, Johnson DK, Adney WS, Nimlos MR, Brady JW, Foust TD (2007) Biomass recalcitrance: engeering plants and enzymes for biofuels production. Science 315:804–807 7. Foust TD, Ibsen KN, Dayton DC, Hess JR, Kenny KE (2008) In: Himmel ME (ed) Biomass recalcitrance: deconstructing the plant cell wall for bioenergy. Blackwell Publishing, Oxford, pp 7–37 8. Xu L, Jing-Ke W, Clint C (2008) Improvement of biomass through lignin modification. Plant J 54:569–581 9. Chen F, Dixon RA (2007) Lignin modification improves fermentable sugar yields for biofuel production. Nat Biotechnol 25:759–761 10. Kumar P, Barrett DM, Delwiche MJ, Stroeve P (2009) Methods for pretreatment of lignocellulosic biomass for efficient hydrolysis and biofuel production. Ind Eng Chem Res 48:3713–3729 11. Mousdale DM (2008) Biofuels: biotechnology, chemistry, and sustainable development. CRC, Boca Raton, FL 12. Agarwal UP (2006) Raman imaging to investigate ultrastructure and composition of plant cell walls: distribution of lignin and cellulose in black spruce wood (Picea mariana). Planta 224:1141–1153

13. Ahlgren PA, Goring DAI (1971) Removal of wood components during chlorite delignification of black spruce. Can J Chem 49:1272–1275 14. Andresen ER, Nielsen CK, Thøgersen J, Keiding SR (2007) Fiber laser-based light source for coherent anti-Stokes Raman scattering microspectroscopy. Opt Exp 15: 4848–4856 15. Kieu K, Saar BG, Holtom GR, Xie XS, Wise FW (2009) High-power picosecond fiber source for coherent Raman microscopy. Opt Lett 34:2051–2053 16. Cheng J-X, Xie XS (2004) Coherent antiStokes Raman scattering microscopy: instrumentation, theory, and applications. J Phys Chem B 108:827–840 17. Cheng J-X, Volkmer A, Book LD, Xie XS (2001) An epi-detected coherent anti-Stokes Raman scattering (E-CARS) microscope with high spectral resolution and high sensitivity. J Phys Chem B 105:1277–1280 18. Levenson MD, Kano SS (1988) Introduction to nonlinear laser spectroscopy. Academic, San Diego 19. Freudiger CW, Min W, Saar BG, Lu S, Holtom GR, He C, Tsai JC, Kang JX, Xie XS (2008) Label-free biomedical imaging with high sensitivity by stimulated raman scattering microscopy. Science 322:1857–1861 20. Boudet AM, Lapierre C, Pettenati JG (1995) Tansley review no-80 - biochemistry and molecular-biology of lignification. New Phytol 129:203–236 21. Gierlinger N, Schwanninger M (2006) Chemical imaging of popolar wood cell walls by confocal Raman microscopy. Plant Physiol 140:1146–1154 22. Donaldson LA (1994) Mechanical constraints on lignin deposition during lignification. Wood Sci Technol 28:111–118 23. Viamajala S, Selig M, Vinant T, Tucker M, Himmel ME, McMillan J, Decker S (2006) Catalyst transport in corn stover internodes. Appl Biochem Biotechnol 130:509–527 24. Zeng Y, Saar B, Friedrich M, Chen F, Liu Y-S, Dixon R, Himmel ME, Xie X, Ding S-Y (2010) Imaging lignin-downregulated alfalfa using coherent anti-Stokes Raman scattering microscopy. BioEnergy Res 3:272–277 25. Saar B, Zeng Y, Freudiger C, Liu YS, Himmel ME, Xie X, Ding S-Y (2010) Label-free, realtime monitoring of biomass processing with stimulated raman scattering microscopy. Angew Chem Int Ed 49:5476–5479


Chapter 6 Immunological Approaches to Plant Cell Wall and Biomass Characterization: Glycome Profiling Sivakumar Pattathil, Utku Avci, Jeffrey S. Miller, and Michael G. Hahn Abstract The native complexity of plant cell walls makes research on them challenging. Hence, it is advantageous to have a diversity of tools that can be used to analyze and characterize plant cell walls. In this chapter, we describe one of two immunological approaches that can be employed for screening of plant cell wall/biomass materials from diverse plants and tissues. This approach, Glycome Profiling, lends itself well to moderate to high-throughput screening of plant cell wall/biomass samples. Glycome Profiling is being further optimized to reduce the amount of sample required for the analysis, and to improve the sensitivity and throughput of the assay. We are optimistic that Glycome Profiling will prove to be a broadly applicable experimental approach that will find increasing application to a wide variety of studies on plant cell wall/biomass samples. Key words: Biomass, Cell wall glycans, ELISA, Glycome profiling, Heatmap, Monoclonal antibody

1. Introduction Cell walls provide essential functions that are important for structure, intercellular communication, defense responses, growth, and development in plants (1). Cell walls are also dynamic in that they change in their structure and composition depending on the tissue and cell type being examined, its developmental stage, and biotic and abiotic factors. Plant cell walls are the major component of plant biomass and its natural resistance to biodegradation, otherwise termed â&#x20AC;&#x153;cell wall recalcitrance,â&#x20AC;? is a major hurdle that must be overcome if lignocellulosic biomass is to be used effectively for biofuel production (2, 3). Thus, understanding the features of plant cell walls that underlie recalcitrance is of prime interest to cell wall researchers. Such an understanding will facilitate fine-tuning the properties of plant biomass and optimizing biomass processing in order to more efficiently and cost-effectively break down the biomass to its component sugars for subsequent conversion to biofuel and other valuable products. Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_6, Š Springer Science+Business Media, LLC 2012

61


62

S. Pattathil et al.

The composition and structural features of plant cell walls are complex. Plant cell walls are largely composed of diverse types of polysaccharides. Additionally, the structural make up of cell walls varies in different plants, tissue and cell types, and even at subcellular levels (1, 4, 5). Many of the polysaccharides found in plant cell walls are themselves highly heterogeneous in their composition. Multiple structural forms have been demonstrated for most of the major cell wall hemicellulosic polysaccharides (6, 7), pectic polysaccharides (8), and arabinogalactan glycoproteins (9). One of the ways to study the diverse complex structures that are present in plant cell walls is by in situ visualization and/or in vitro detection of specific substructures (epitopes) present on them using suitably designed diverse molecular probes (10). Plant cell wall glycan-directed monoclonal antibodies (mAbs) are highly specific tools for structural and compositional analyses of plant cell walls. The current worldwide collection of about 200 mAbs is sufficiently large and diverse in its binding specificities (Fig. 1) to be an effective toolkit for the study of most major classes of plant cell wall glycans (11). The size and diversity of the mAb collection is particularly important to its utility, as there are multiple antibodies available to detect diverse structural features of a given plant cell wall polysaccharide class. For example, studies of xyloglucan are now not just limited to the use of a single monoclonal antibody (e.g., CCRC-M1 or LM15), but can take advantage of a comprehensive suite of more than 25 mAbs that recognize different substructures present on xyloglucans in diverse plants and plant tissues. Equally large and diverse sets of antibodies are also available for other plant cell wall glycans such as xylans, arabinogalactans, and pectins. In our laboratory, these antibodies are used extensively for plant cell wall/biomass characterization in two different, but complementary approaches: Glycome Profiling and immunohistochemistry. This chapter focuses on an ELISA-based method, Glycome Profiling, which is a moderate to high-throughput method for obtaining a comprehensive picture of the glycan epitope composition of a cell wall/biomass sample and some information about how tightly those epitopes are bound into the wall. In Glycome Profiling, the wall is first fractionated by sequential extraction with increasingly harsh chemical or enzymatic extractants. Each of these fractions is subjected to ELISA against the toolkit of antibodies in order to identify the nature of the polysaccharides present in them. The ELISA responses are depicted as heat maps for analysis. The wall fractionation carried out prior to the ELISAs has the advantage that possible masking of epitopes by the presence of other wall components (12â&#x20AC;&#x201C;14) is precluded and all epitopes in the solubilized materials are freely accessible for detection. This wall fractionation also allows inferences to be made about how tightly the various glycan epitopes (and hence the polysaccharides that contain those epitopes) are bound to the wall. However, all information about the distribution and cellular localization of epitopes


6

Immunological Approaches to Plant Cell Wall and Biomass Characterizationâ&#x20AC;Ś

63

Fig. 1. Current worldwide tool kit of ~200 plant cell wall glycan-directed monoclonal antibodies. The antibodies were all tested by ELISA against a diverse panel of plant cell wall polysaccharide preparations, and the resulting data subjected to hierarchical clustering to group the antibodies based on commonalities in their binding specificities (11).

is lost. Furthermore, epitopes may be lost or modified by the harsh chemicals used to prepare some of the wall extracts. It is hence advisable to also utilize the antibodies for in situ localization of epitopes (see following chapter) in order to maximize the extent of information about wall composition and structure that can be obtained using antibody probes. In this chapter, we describe our protocols for cell wall isolation, extraction, and Glycome Profiling. Our application of Glycome Profiling utilizes a substantial subset (155) of the worldwide collection of plant glycan-directed mAbs to probe plant cell wall/biomass extracts. This subset of antibodies includes antibodies that recognize diverse epitopes present on each of the major classes of plant polysaccharides, except cellulose and rhamnogalacturonan II. We have successfully applied Glycome Profiling to comparative glycomics studies both between different tissues from a single plant and to walls from different


64

S. Pattathil et al.

plants, to the functional characterization of wall biosynthetic genes by analysis of mutant walls (15, 16), the characterization of microbial action on biomass (17), and the study of biomass structural properties under various pretreatment regimes (18).

2. Materials Prepare all solutions using deionized water at room temperature, unless otherwise instructed. Some reagents used are potentially hazardous and appropriate personal protection (e.g., gloves, fume hood) should be used to avoid breathing hazardous fumes or contact with skin. All reagents should be reagent grade and stored in an appropriate storage area; this is particularly true for the bases. Proper disposal of the waste chemicals should be done according to waste disposal regulations of your institute. 2.1. Cell Wall Extraction

1. 50 mM ammonium oxalate (pH ~5.0): To about 800 mL of water in a 1 L glass beaker, add 7.1 g of ammonium oxalate monohydrate. Stir until solid is completely dissolved, then make up to 1 L with water and stir again. This solution can be stored at room temperature until needed. 2. 50 mM sodium carbonate with 0.5% (w/v) sodium borohydride (pH ~10.0): To about 800 mL of water in a 1 L glass beaker, add 5.3 g of anhydrous sodium carbonate and 5.0 g of sodium borohydride. Stir until reagents are completely dissolved, then make up to 1 L with water and stir again. This solution must be freshly made for each experiment. 3. 1 M potassium hydroxide (KOH) with 1.0% (w/v) sodium borohydride: To about 800 mL of water in a 1 L glass beaker, add 56.11 g of KOH pellets and 10.0 g of sodium borohydride powder. Stir until reagents are completely dissolved, then make up to 1 L with water and stir again. This solution must be freshly made for each experiment. 4. 4 M potassium hydroxide with 1.0% (w/v) sodium borohydride: To about 800 mL of water in a 1 L glass beaker, add 224.44 g of KOH pellets and 10.0 g of sodium borohydride powder. Stir until reagents are completely dissolved, then make up to 1 L with water and stir again. This solution must be freshly made for each experiment. 5. Sodium chlorite, glacial acetic acid, and 2-octanol for AIR fractionation.

2.2. Total Sugar Estimation

1. 5% (v/v) phenol: Take 5 mL of phenol (89%) and dilute to 100 mL with deionized water. 2. Concentrated sulfuric acid (96%).


6

Immunological Approaches to Plant Cell Wall and Biomass Characterization…

2.3. Enzyme-Linked Immunosorbent Assay

65

1. Enzyme-linked immunosorbent assay (ELISA) Plate: Costar 3598 acid-resistant flat-bottom 96-well plate (Corning Life Sciences, Acton, MA). 2. 0.1 M Tris-buffered saline (TBS), pH 7.6: Fill about 3500 mL of ultrapure (~18.2 MΩ cm) water into a 4 L container, add 23.38 g of sodium chloride to the water and stir until dissolved. Add 1.11 g of Tris-Base and 4.85 g of Tris–HCl, respectively, to the container and stir until dissolved. Make up the volume to 4 L with ultrapure water and stir completely until uniformly mixed. Buffer stock can be stored at room temperature until needed. 3. Blocking Buffer: 1.0% (w/v) Milk in 0.1 M TBS: To about 500 mL of 0.1 M TBS and stir in 10.0 g of Nonfat Dry Milk (Instant Non-fat Dry Milk, Publix). After 3 h of stirring, make up the solution to 1 L with 0.1 M TBS, then mix uniformly and store at 4°C. 4. Wash Buffer: 0.1% (w/v) Milk in 0.1 M TBS. Dilute the Blocking Buffer solution tenfold using 0.1 M TBS. 5. Primary antibodies: CCRC series of antibodies are generated in mouse; JIM, MAC, and LM series of antibodies are generated in rat. A Web-accessible database listing most of the available plant cell wall glycan-directed mAbs and providing information about their characteristics and suppliers can be found at WallMabDB (http://www.wallmabdb.net). The three main suppliers of plant glycan-directed antibodies are: CarboSource (http://www.carbosource.net), PlantProbes (http://www.plantprobes.net) and BioSupplies (http://www. biosupplies.com.au/.) 6. Secondary antibody: Anti-Mouse (Sigma Catalogue A4416) or Anti-Rat (Sigma Catalogue A9037) IgG whole molecule Goat antibody, conjugated with horseradish peroxidase. The secondary antibody should be diluted according to the manufacturer’s instructions; in our case, the secondary antibody is diluted 1:5,000 in Wash Buffer to the required volume just prior to use. Secondary antibody stocks should be stored at −20°C when not in use. 7. Substrate: TMB Peroxidase Substrate Kit SK-4400 (Vector Laboratories, Inc., Burlingame, CA). Make fresh without using the stabilizer (that is included in the kit), mix two drops of the pH 5.3 Buffer, three drops of the TMB and two drops of hydrogen peroxide in 15 mL of water. Store kit at 4°C when not in use. 8. Stop solution: 0.5 N Sulfuric acid (for 100 mL of 0.5N Sulfuric Acid, mix 1 mL of 18 M Sulfuric Acid with 71 mL of deionized water).


66

S. Pattathil et al.

3. Methods 3.1. General Protocol for Plant Biomass/Cell Wall Extractions

All procedures are performed at room temperature unless otherwise specified. 1. Preparation of alcohol insoluble residue (AIR) from plant biomass. Plant biomass is ground by mortar and pestle to a fine powder in liquid nitrogen (for vegetative tissues) or in a Wiley mill to 20–80 mesh (for woody tissues). The ground biomass is then resuspended in 80% (v/v) ethanol (100 mL/g) and stirred for at least an hour (can be longer; up to overnight) and then centrifuged at 3,000 × g for 15 min. The supernatant is discarded and the pellet is washed sequentially with absolute ethanol once and acetone twice at room temperature. The resulting cell wall residue (AIR) is vacuum dried for 24 h at room temperature. 2. AIR fractionation. The cell walls (AIR) are subjected to sequential extractions with increasingly harsh reagents in order to isolate fractions enriched in various cell wall components (see Note 1). Based on the starting weight of the AIR used, all of the following extractions are carried out in suspension at 10 mg/mL in a 50 mL polypropylene centrifuge tube. (a) Suspend the AIR in 50 mM ammonium oxalate (pH 5.0) and incubate for 24 h with constant shaking (100 rpm). After incubation, the suspension is centrifuged at 4,000 × g for 15 min. Decant the supernatant from the pellet and store at 4°C in a new, 50 mL centrifuge tube labeled as “Ammonium Oxalate Extract.” (b) Resuspend the residual pellet from the previous step in the same volume of deionized water. After mixing, re-pellet by centrifugation as before and discard the supernatant. (c) Resuspend the washed pellet from the previous step in 50 mM sodium carbonate, containing 0.5% (w/v) sodium borohydride, and incubate for 24 h with constant shaking (100 rpm). After incubation, re-pellet the sample by centrifugation as above. Decant the supernatant and store at 4°C in a new, 50 mL tube labeled as “Carbonate Extract.” (d) Wash the pellet from the previous step with deionized water as before and discard the water wash. (e) Resuspend the washed pellet in 1 M KOH, containing 1.0% (w/v) sodium borohydride, and incubate for 24 h with constant shaking (100 rpm). After incubation, repellet the sample by centrifugation as above. Decant the supernatant and store at 4°C in a new, 50 mL tube labeled


6

Immunological Approaches to Plant Cell Wall and Biomass Characterization…

67

as “1 M KOH Extract” (see Note 2 for further action on this fraction). (f) Resuspend the pellet from the previous step directly in 4 M KOH containing 1.0% (w/v) sodium borohydride, and incubate for 24 h with constant shaking (100 rpm). After incubation, re-pellet the sample by centrifugation as above. Decant the supernatant and store at 4°C in a new, 50 mL tube labeled as “4 M KOH Extract” (see Note 2 for further action on this fraction). (g) After the 4 M KOH treatment, wash the pellet three times with deionized water (as described above) to remove all KOH prior to the next extraction. (h) Treat the residual pellet with sodium chlorite (ACS Reagent, MW: 90.44) and glacial acetic acid (19) to break down the lignin, as follows. The pellet is suspended in 20 mL of deionized water and placed in a 70°C water bath. Each sample is then subjected to three additions of 0.125 g of sodium chlorite and 50 μL of glacial acetic acid; each addition is separated by a 1-h incubation. The final product contains dissolved chlorine gas, as indicated by the yellow color of the solution. This chlorine gas must be removed by slowly bubbling dry air or nitrogen gas through the sample until the yellow color disappears (carry out in a fume hood). The final sample is pelleted by centrifugation as above, and the supernatant is decanted and stored at 4°C in a new, 50 mL tube labeled as “Chlorite Extract.” (See Note 1 on this extraction step.) (i) Wash and pellet the chlorite residue as above with deionized water and discard the water wash. (j) Finally, resuspend the washed pellet from the previous step in 4 M KOH containing 1.0% (w/v) sodium borohydride, and incubate for 24 h with constant shaking (100 rpm). After incubation, re-pellet the sample by centrifugation as above. Decant the supernatant and store at 4°C in a new, 50 mL tube labeled as “4 M KOHPC fraction” (see Note 2 for further action on this fraction). The residual pellet after this step is washed with deionized water and stored at 4°C. (See Note 1 on this extraction step.) (k) All of the extracts are dialyzed (using 3,500 Da molecular weight cut-off tubing; #S632724, Spectrum Laboratories Inc., CA, USA) against four changes of deionized water (sample:water ~1:60) at room temperature for a total of 48 h and then lyophilized and weighed. The extracts can then be stored at room temperature in a dessicator over drying agent until needed.


68

S. Pattathil et al.

3.2. Total Sugar Estimation

All cell wall extracts are dissolved in deionized water at a concentration of 0.2 mg/mL in a 15 mL centrifuge tube. Total sugar estimation in samples is done using the phenol-sulfuric acid micro plate assay (20, 21), as follows: 1. Assays are done in duplicate in disposable 13 × 100 mm glass test tubes. First, add 100 μL of each sample into respective duplicate test tubes with samples being added directly to the bottom of the tubes. 2. Add 100 μL of 5.0% (v/v) phenol to each culture tube with direct addition to the sample. 3. Add 500 μL of concentrated sulfuric acid (18 M) to each reaction tube. Acid should be added in such a way that the entire 500 μL drops directly into the sample mixture without contacting the walls of the test tube. Vortex each tube of sample solutions gently to get uniform mixing. 4. Incubate the reaction tubes for 20 min in a fume hood. 5. After incubation, transfer 250 μL of each reaction mixture into a Costar 3598 ELISA plate. 6. The color development in the reactions is measured using an ELISA plate reader by reading the OD at 490 nm. [Comment: A standard curve prepared using solutions with varying amounts of D-glucose (5 μg to 50 μg) is used to calculate the exact glucose equivalents of sugars in each of the wall extract samples.]

3.3. Enzyme-Linked Immunosorbent Assay

1. Coating of the ELISA plates: All wall extract samples are diluted to a sugar concentration of 20 μg/mL. Add 50 μL per well of each wall extract dilution onto the ELISA plates (with the number of wells coated equaling the number of mAbs to be tested plus controls) and evaporate to dryness overnight in a ventilated 37°C incubator (see Note 3). 2. Blocking: Nonspecific sites in the coated ELISA plates are blocked by adding 200 μL per well of Blocking Buffer and incubating the plates for an hour at room temperature. 3. Addition of primary antibodies (mAbs): Aspirate the blocking buffer from each well. Dispense 50 μL of primary mAb into each well. Incubate plates with the primary antibodies for an hour at room temperature. 4. Washing the plates: Aspirate the primary antibodies from each well. Wash each plate with 300 μL of wash buffer, wait 5 s, and completely aspirate the buffer. Repeat two more times for a total of three washes. 5. Secondary antibodies: After washing, 50 μL per well of secondary antibody is added to the ELISA plates. Anti-mouse or anti-rat secondary antibodies (mixed at a 1:5,000 dilution in


6

Immunological Approaches to Plant Cell Wall and Biomass Characterization…

69

Wash Buffer) are dispensed into the respective mouse (for example, CCRC series) and rat (for example, JIM series) primary antibody-bound wells and incubated at room temperature for 1 h. 6. Washing secondary antibodies: Aspirate the secondary antibodies from each well after the incubation. Wash each plate with 300 μL of Wash Buffer, wait 5 s, and completely aspirate the buffer. Repeat four more times for a total of five washes. 7. Adding substrate and termination: Dispense 50 μL per well of TMB substrate solution into each well, allow each plate to incubate for precisely 20 min, and then stop the reaction using 50 μL per well of 0.5 N sulfuric acid. 8. Quantitation: Immediately after termination, measure the net OD values of the color formation in the wells of ELISA plates using an ELISA plate reader at 450 nm and subtracting a background reading at 655 nm. 3.4. Representation of ELISA Data in Heat Maps

Sequential extracts of plant cell wall/biomass materials are probed against 155 cell wall glycan-directed antibodies belonging to 26 reactivity groups (Fig. 1). The ELISA responses of these antibodies to each extract are presented as a heat map using a modified version of R-Console software (22). The R software assists in converting the raw absorbance values (that are provided to the software in an Excel™ “csv file” format in which raw ELISA response values are organized into any pattern of users’ interest) into color gradients using a set of color keys that are chosen by the user. Thus, easier visual inspection and interpretation of the entire data set is possible, irrespective of its size. An example of Glycome Profiling data representation is shown in Fig. 2 and depicts the glycome profile of Arabidopsis thaliana leaves (see Notes 3 and 4).

4. Notes 1. For biomass samples that are not lignified or do not contain significant levels of cell wall phenolics, the last two extractions (chlorite and post-chlorite 4 M KOH) are optional. In this case, the respective steps for these extractions can be omitted. 2. The 1 M KOH, 4 M KOH, and 4 M KOHPC fractions obtained during cell wall extraction require neutralization prior to dialysis and/or storage. Place these samples, in their new 50 mL tube, upright and uncapped in ice, add ~3–5 drops of 2-octanol to each, and slowly add glacial acetic acid until a pH of 7 is reached (use pH indicator paper to monitor). Be careful


Fig. 2. Glycome Profile of Arabidopsis thaliana leaf cell walls. Sequential extracts of cell walls from leaves of 10 week old Arabidopsis thaliana plants were prepared and were probed by ELISA with an array of cell wall glycan-directed mAbs shown in right hand panel. The strengths of the ELISA signals are represented in a yellow to black scale with bright yellow depicting strongest binding (representing an A450â&#x20AC;&#x201C;655 of 0.8 in this experiment) and black no binding. The reagents used for extraction are labeled at the bottom of the heat map. The bar graph at the top of the glycome profile shows the total amount of material extracted from the cell wall/biomass sample in each extraction step.


6

Immunological Approaches to Plant Cell Wall and Biomass Characterizationâ&#x20AC;Ś

71

not to allow the samples to bubble over the rim while adding the acid to avoid sample losses. 3. The cell wall extracts are immobilized onto the ELISA plates by drying down the extract solutions. Subsequently, these dried down carbohydrates that are stuck to the wells are probed with mAbs. However, experiments in our laboratory have shown that polysaccharides with molecular sizes of 20 kDa or less do not efficiently stick to the ELISA plates. Thus, Glycome Profiling yields an assessment of the epitope composition of the larger MW polysaccharides present in each extract, since the small polysaccharides and oligosaccharides wash off of the plate and antibodies that recognize epitopes on these smaller polysaccharides will give no signal in the ELISAs. Therefore, one can fail to observe appreciable antibody binding to an extract in Glycome Profiling if it contains a significant proportion of smaller polysaccharides, even though carbohydrate was loaded onto the plate. In order to analyze the epitope composition of the low MW fraction, it is necessary to use alternative means to ensure immobilization to the ELISA plate, for example by incorporating a biotin group at the reducing ends of the carbohydrates and then immobilizing the modified carbohydrates on streptavidin-coated ELISA plates. 4. The various cell wall/biomass extracts are loaded onto the ELISA plates in equal sugar amounts for Glycome Profiling. Thus, the method does not take into account the quantities of material released at each step. Glycome Profiling carried out in this way does not, therefore, give an accurate representation of the total amount of an epitope released in a given extraction step. So one cannot make quantitative comparisons of the abundance of an epitope in a biomass sample from the glycome profile alone. We typically add a bar graph to the top of the glycome profile (see Fig. 2) to show the total amount of material extracted from the cell wall/biomass sample in each extraction step to allow for more meaningful comparisons of epitope contents to be made.

Acknowledgements Research in our laboratory on immunological approaches to biomass characterization is supported by the Office of Biological and Environmental Research in the DOE Office of Science through the BioEnergy Science Center (BESC) funded by grant DE-AC0500OR22725. Generation of the CCRC series of plant glycandirected mAbs used in this work was supported by the NSF Plant Genome Program (DBI-0421683).


72

S. Pattathil et al.

References 1. Keegstra K (2010) Plant cell walls. Plant Physiol 154:483–486 2. Wyman CE (2007) What is (and is not) vital to advancing cellulosic ethanol. Trends Biotechnol 25:153–157 3. Lynd LR, Laser MS, Brandsby D, Dale BE, Davison B, Hamilton R, Himmel ME, Keller M, McMillan JD, Sheehan J, Wyman CE (2008) How biotech can transform biofuels. Nat Biotechnol 26:169–172 4. Pauly M, Keegstra K (2010) Plant cell wall polymers as precursors for biofuels. Curr Opin Plant Biol 13:305–312 5. McCann MC, Knox JP (2011) Plant cell wall biology: polysaccharides in architectural and developmental contexts. In: Ulvskov P (ed) Plant Polysaccharides: biosynthesis and bioengineering. Annu Plant Rev 41: 343–366 6. York WS, O’Neill MA (2008) Biochemical control of xylan biosynthesis - which end is up? Curr Opin Plant Biol 11:258–265 7. Scheller HV, Ulvskov P (2010) Hemicelluloses. Annu Rev Plant Biol 61:263–289 8. Mohnen D (2008) Pectin structure and biosynthesis. Curr Opin Plant Biol 11:266–277 9. Showalter AM (2001) Arabinogalactanproteins: structure, expression and function. Cell Mol Life Sci 58:1399–1417 10. Lee KJ, Marcus SE, Knox JP (2011) Cell wall biology: perspectives from cell wall imaging. Mol Plant 4:212–219 11. Pattathil S, Avci U, Baldwin D, Swennes AG, McGill JA, Popper Z, Bootten T, Albert A, Davis RH, Chennareddy C, Dong R, O’Shea B, Rossi R, Leoff C, Freshour G, Narra R, O’Neil M, York WS, Hahn MG (2010) A comprehensive toolkit of plant cell wall glycan-directed monoclonal antibodies. Plant Physiol 153:514–525 12. Puhlmann J, Bucheli E, Swain MJ, Dunning N, Albersheim P, Darvill AG, Hahn MG (1994) Generation of monoclonal antibodies against plant cell wall polysaccharides. I. Characterization of a monoclonal antibody to a terminal α(1 → 2)-linked fucosyl-containing epitope. Plant Physiol 104:699–710 13. Marcus SE, Blake AW, Benians TA, Lee KJ, Poyser C, Donaldson L, Leroux O, Rogowski

14.

15.

16.

17.

18.

19.

20.

21.

22.

A, Petersen HL, Boraston A, Gilbert HJ, Willats WG, Knox JP (2010) Restricted access of proteins to mannan polysaccharides in intact plant cell walls. Plant J 64:191–203 Marcus SE, Verhertbruggen Y, Hervé C, Ordaz-Ortiz JJ, Farkas V, Pedersen HL, Willats WG, Knox JP (2008) Pectic homogalacturonan masks abundant sets of xyloglucan epitopes in plant cell walls. BMC Plant Biol 8:60 Freshour G, Bonin CP, Reiter WD, Albersheim P, Darvill AG, Hahn MG (2003) Distribution of fucose-containing xyloglucans in cell walls of the mur1 mutant of Arabidopsis. Plant Physiol 131:1602–1612 Kong Y, Zhou G, Yin Y, Xu Y, Pattathil S, Hahn MG (2011) Molecular analysis of a family of Arabidopsis genes related to galacturonosyltransferases. Plant Physiol 155:1791–1805 Lee SJ, Warnick TA, Pattathil S, Alvelo-Maurosa JG, Serapiglia MJ, McCormick H, Brown V, Young NF, Schnell DJ, Smart LB, Hahn MG, Pedersen JF, Leschine SB, Hazen SP (2012) Biological conversion assay using Clostridium phytofermentans to estimate plant feedstock quality. Biotechnol Biofuels 5:5 DeMartini JD, Pattathil S, Avci U, Szekalski K, Mazumder K, Hahn MG, Wyman CE (2011) Application of monoclonal antibodies to investigate plant cell wall deconstruction for biofuels production. Energy Environ Sci 4: 4332–4339 Ahlgren PA, Goring DA (1971) Removal of wood components during chlorite delignification of black spruce. Can J Chem 49: 1272–1275 Dubois M, Gilles DA, Hamilton JK, Rebers PA, Smith F (1956) Colorimetric method for the determination of sugars and related substances. Anal Chem 28:350–356 Masuko T, Minami A, Iwasaki N, Majima T, Nishimura SI, Lee YC (2005) Carbohydrate analysis by a phenol-sulphuric acid method in microplate format. Anal Biochem 339:69–72 R Development Core Team (2006) R: A language and environment for statistical computing. R Foundation for Statistical Computing. http://www.R-project.org


Chapter 7 Immunological Approaches to Plant Cell Wall and Biomass Characterization: Immunolocalization of Glycan Epitopes Utku Avci, Sivakumar Pattathil, and Michael G. Hahn Abstract Plant cell walls are dynamic structures that show changes in composition and configuration depending on the developmental stage, biotic, and abiotic factors. Therefore, it is necessary to have tools for visualizing the components of the cell wall in situ at any stage. Here, we describe how specific monoclonal antibodies can be used to examine the distribution of plant cell wall glycan epitopes at the whole plant, tissue, cell, and subcellular levels. Understanding the basic cell wall structure is essential for devising efficient strategies to convert cell walls to fermentable sugars for ethanol production. Key words: Biomass, Cell wall glycans, Epitope, Fluorescence, Immunolocalization, Microscopy, Monoclonal antibody

1. Introduction Plant cell walls are composed of diverse and structurally complex polysaccharides, whose presence, absence, and/or structural variation in different plants, tissues, cell types, and even cell walls complicate studies of these walls (1â&#x20AC;&#x201C;4). Therefore, it is vital to have diverse tools that can be used to analyze and characterize plant cell walls. Plant cell wall glycan-directed monoclonal antibodies (mAbs) are highly specific tools for structural and compositional analyses of plant cell walls. The current worldwide collection of about 200 mAbs is sufficiently large and diverse in its binding specificities (Fig. 1) to be an effective toolkit for the study of most major classes of plant cell wall glycans (5). The usefulness of this toolkit continues to be improved, primarily through studies to define in greater detail the specific carbohydrate epitopes recognized by more of the antibodies in the toolkit.

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_7, Š Springer Science+Business Media, LLC 2012

73


74

U. Avci et al.

Fig. 1. Current worldwide toolkit of ~200 plant cell wall glycan-directed monoclonal antibodies. The antibodies were all tested by ELISA against a diverse panel of plant cell wall polysaccharide preparations, and the resulting data subjected to hierarchical clustering to group the antibodies based on commonalities in their binding specificities (5).

These immunological tools provide valuable information and insight into cell wall structure and composition. However, they are most powerful when combined with other analytical/chemical tools. The main advantage of the immunological tools is that, for the most part, they appear to be applicable to all plant species, are relatively rapid, and can guide scientists in the appropriate direction(s) for more time consuming chemical analyses. Immunohistochemistry (also known as immunolocalization) is a technique to observe the cellular and subcellular locations of antigens of interest. It is a powerful approach that is frequently overlooked in cell wall research projects due to the need for special training and/or specialized equipment. However, the basic equipment necessary for immunohistochemistry is now fairly broadly available and the specialized training needed for carrying out the


7

Immunological Approaches to Plant Cell Wall and Biomass Characterizationâ&#x20AC;Ś

75

Fig. 2. Schematic representation of the two main approaches for immunological characterization of plant biomass.

procedures can be obtained in microscopy centers present on many university campuses. Thus, there is less reason to avoid this powerful approach, particularly given the wealth of information about wall structure and complexity that can be obtained from immunohistochemical analyses. Immunohistochemistry shows the distribution of wall glycan epitopes in plant cell walls at the whole plant, tissue, cell, and subcellular levels. Furthermore, one can take advantage of intrinsic developmental gradients that exist in plant tissues and use immunohistochemistry to monitor changes in wall composition and structure as a function of development and differentiation. Immunohistochemical techniques also result in minimal if any modification of the glycans, and thus such studies provide a fairly complete picture of cell wall composition, limited only by how completely the structural diversity of the walls is covered by the available probe collection. Since immunohistochemistry is often laborious, it is sometimes useful to narrow the selection of antibodies to be used for such localization studies using information obtained from other experimental approaches. For example, Glycome Profiling, another immunological approach described in the previous chapter in this volume (6), provides a quick overview of cell wall glycan epitopes in a given sample. However, there are some fundamental differences in these two immunological methods (Fig. 2), which must be taken into account when comparing data generated using these approaches. First, the harsh reagents used to prepare some wall extracts for Glycome Profiling can alter the extracted polysaccharides and hence alter the spectrum of antibody reactivity compared with immunohistochemistry, which does not typically necessitate the use of such harsh reagents. Second, other wall components may interfere with antibody access to epitopes in situ (7â&#x20AC;&#x201C;9), whereas in Glycome Profiling, the wall is fractionated and all epitopes are freely accessible


76

U. Avci et al.

for detection. Third, glycan epitopes may be localized to specific cells or subcellular structures that are not in the plane of section taken for immunohistochemistry and therefore remain undetected, whereas these same epitopes may be detectable in Glycome Profiling, which is carried out on walls prepared from bulk tissue. In this chapter, we describe how cell wall glycan-directed monoclonal antibodies can be used to examine the distribution of plant cell wall glycan epitopes at the whole plant, tissue, cell, and subcellular levels using immunohistochemistry. Other molecular probes, such as carbohydrate-binding modules (3, 10) or glycosylhydrolases (11–13), have also been used for localization of cell wall glycan structures, but methods for their use are not discussed here. The following protocols have been successfully used in the authors’ laboratory to localize cell wall glycan epitopes in diverse plants, including Arabidopsis thaliana, switchgrass (Panicum virgatum), and poplar (Populus deltoides). Depending on the tissue of interest, modifications and optimization of protocols provided here might be required to meet researcher’s needs or to take into account the characteristics of specific tissues of interest.

2. Materials Some of the chemicals used below are toxic. Therefore, proper lab safety practices should be followed (gloves, lab coat, hood, etc.). Proper disposal of the waste chemicals should be done according to waste disposal regulations of your institute. The vendors for specialized equipment listed here are the ones in use in our laboratory. Equivalent equipment from other vendors exists. 2.1. Fixation

1. Phosphate buffers: Stock sodium phosphate buffers; 0.2 M monobasic and dibasic: working sodium phosphate buffer; 50 mM pH 7.1: washing sodium phosphate buffer; 25 mM pH 7.1. 2. Formaldehyde/glutaraldehyde fixative [to make 100 mL of 1.6% (v/v) paraformaldehyde/0.2% (v/v) glutaraldehyde in 25 mM sodium phosphate buffer]: 10 mL 16% (v/v) paraformaldehyde (Electron Microscopy Sciences, 15710), 2 mL 10% (v/v) glutaraldehyde (Electron Microscopy Sciences, 16120), 2 mL 1% (v/v) Triton X-100, 50 mL 50 mM stock phosphate buffer, 36 mL H2O. 3. Microcentrifuge tubes or glass vials. 4. Ethanol dehydration: Prepare graded ethanol series diluting with deionized water except 100% [35% (v/v), 50% (v/v), 70% (v/v), 95% (v/v), 100%].


7

Immunological Approaches to Plant Cell Wall and Biomass Characterization…

77

5. Infiltration: LR White Resin (Medium grade, Ted Pella, 18181). 6. Embedding: Gelatin capsules (Ted Pella, 130–14). 7. Polymerization: Cryo chamber (Ted Pella, 6202) or UV light source at 4°C. 2.2. Sectioning

1. Sharp razor blades for trimming (Ted Pella, 121–3). 2. A loop (Ted Pella) to pick up sections. 3. Microscope glass slides (Fisherbrand Plus Microscope slides, 12–550) that have special coating to facilitate adherence of sections to the glass slide. 4. Heater (VWR 12365–454). 5. Ultramicrotome (Leica EM UC6).

2.3. Immunolabeling

1. Primary antibodies: CCRC series of antibodies are generated in mouse; JIM, MAC and LM series of antibodies are generated in rat. A web-accessible database listing most of the available plant cell wall glycan-directed monoclonal antibodies and providing information about their characteristics and suppliers can be found at WallMabDB (http://www.wallmabdb.net). The three main suppliers of plant glycan-directed antibodies are CarboSource (http://www.carbosource.net), PlantProbes (http://www.plantprobes.net), and BioSupplies (http://www. biosupplies.com.au/). 2. Secondary antibody (Alexa Fluor 488 anti-mouse IgG (H + L), Invitrogen, A11001 for CCRC series of antibodies; Alexa Fluor 488 anti-rat IgG (H + L), Invitrogen, A-11006 for LM, JIM, and MAC series of antibodies). 3. Petri dishes (glass or plastic) for creating a humid chamber with soaked tissue-wipes. 4. Potassium phosphate buffered saline (KPBS); 10 mM pH 7.1 with 500 mM NaCl. 5. 3% (w/v) dry milk (instant nonfat dry milk, Publix) in 10 mM KPBS for blocking. 6. Citifluor antifadent mounting media (AF1, 17970, Electron Microscopy Sciences). 7. Cover slides (Corning 2935–246).

3. Methods 3.1. Fixation

1. Cut the tissue of interest into small pieces (1–5 mm, thinner is better) with a razor blade and immediately put into freshly prepared fixative solution sufficient to cover the tissue in


78

U. Avci et al.

microcentrifuge tubes or glass vials. Apply vacuum for 30 min and further fix for 2–3 h (see Note 1). 2. Wash samples for 15 min with washing buffer sufficient to cover the tissue (three times). Repeat washing steps with water (two times). 3. Dehydrate the samples by washing the tissue pieces in the graded ethanol series for 25–30 min at each step (one time). The 100% ethanol step should be repeated two times to remove as much water from the tissue as possible. 4. Infiltrate the samples by immersion sequentially with LR White Resin as follows: 1:1 (LR White:100% ethanol) for 12–24 h (one time); LR White (100%, no ethanol) for 12–24 h (three times) (see Note 2). 5. Insert the samples into gelatin capsules and fill the capsules with fresh LR White. Tightly cap them. 6. Polymerize the closed capsules under UV light for 48 h at 4°C. 3.2. Sectioning

Trimming the blocks and sectioning require experience and special equipment. Therefore, consult experienced personnel at your microscopy facility to get trained. 1. Trim the blocks with a sharp razor blade under a dissecting microscope to create a sectioning surface with desired sample orientation. 2. Take 250 nm ribbon sections (serial sections) with a histodiamond knife on an ultramicrotome. 3. Pick the sections up one by one with a loop and carefully transfer them onto the glass slide. Make sure that the sections are not folded after transferring. Also make sure that the sections are spaced sufficiently far apart to preclude mixing of droplets later during the immunolabeling stages. We generally put six sections on a slide. 4. Dry the sections on a heater set to a low setting or around 50°C.

3.3. Immunolabeling

1. Make circles around the sections with a Sharpie marker (use the side without sections). All of the following steps should be conducted in a petri dish with a soaked piece of tissue paper (we use laboratory wipes) to create a humid chamber, which will prevent solutions from drying out. 2. Block the sections with 3% (w/v) nonfat dry milk in 10 mM KPBS (about 10–15 μL) for at least 30 min.


7

Immunological Approaches to Plant Cell Wall and Biomass Characterization…

79

Fig. 3. Cross section of a whole switchgrass leaf blade close to the leaf tip labeled with CCRC-M149 (a xylan-directed antibody). Note: Labeling of almost all cell types within the leaf by this antibody, with the outer wall of epidermal cells showing particularly strong labeling. Bar = 100 μm.

3. Incubate the sections with the choice of primary antibodies (10–15 μL) for at least 1 h at room temperature. We typically use hybridoma supernatants as primary antibody and dilute them 5 to 10-fold as a good starting point (see Note 3). 4. Wash the sections with 10 mM KPBS (three times) for at least 5 min each. 5. Incubate the sections with anti-mouse Alexa fluor 488 (or alternatively anti-rat Alexa fluor 488, depending on the animal source of the primary antibody). A 100-fold dilution in 10 mM KPBS works well for us. 6. Wash the sections with 10 mM KPBS (two times) and with water for at least 5 min each. 7. Apply Citifluor antifadent mounting media and cover with a cover slide. Examine under a fluorescent microscope. Keep the slides at 4°C for future observations for up to 1 week. 8. We visualize labeling using an Eclipse 80i light microscope (Nikon, Melville, NY) equipped with epifluorescence optics and Nikon B-2E/C filter. Images are captured with a Nikon DS-Ri1 camera head (Nikon, Melville, NY) using NISElements Basic Research software. Images are assembled without further processing using Adobe Photoshop (Adobe Systems, San Jose, CA). Figure 3 shows an example of immunolabeling of a cross section of switchgrass leaf blade with one of our xylan-directed antibodies (CCRC-M149). This antibody labels almost all cell types in a cross section of switchgrass leaf blade. The labeling intensity varies, however, in different parts of the tissue, reflecting the availability of the particular xylan epitope recognized by this antibody (see Notes 4 and 5).


80

U. Avci et al.

4. Notes 1. Some plant tissues (especially leaves) contain air spaces, so they float on the fixative. Therefore, initial application of vacuum is essential to help the tissue to sink and thereby facilitate complete fixation of the tissue. In fixation, it is very important to use freshly made fixative. For denser tissue samples, like poplar (P. deltoides), fixation, dehydration, and embedding times should be increased, depending on the age and nature of the sample. Researchers should optimize the protocols according to the specific attributes of their tissue samples. We have not observed any nonspecific background with the glycan-directed monoclonal antibodies that we have used, though such nonspecific binding may be encountered when using polyclonal antibodies. 2. Some caution is called for in the choice of embedding medium used for tissue preparation in advance of sectioning. We use LR White, a hydrophilic acrylic resin for all of our immunohistochemistry. The use of LR White (14) and other plastic/ resin-based embedding media yield sections that are impenetrable to the antibodies. Labeling of tissue is therefore limited to those cellular structures that are exposed on the cut surface of the section (15). In the case of wax-embedded tissues, the wax embedding medium is removed prior to immunolabeling, leading to exposure of additional tissue surfaces that can subsequently be accessed by antibodies during labeling of the sections. Such differences in the nature of the sections must be taken into account when interpreting the results of immunohistochemical studies. For example, changes in labeling patterns using probes against xylan or mannan epitopes have been observed after treatment of wax-embedded sections with pectic-degrading enzymes (8, 9), but were not observed when LR White is used as the embedding medium ((16, 17); and unpublished results of the authors). 3. It is important to run immunological controls along-side of the antibody incubations on each slide. One important set of controls is to include sections incubated with buffer in place of either the primary or secondary antibody. This control should show no labeling of the section. Another type of immunological control is to incubate a section with an antibody generated in the same animal as the test antibody, but one that is not known to bind to anything in plant tissues and should therefore show no labeling. An additional control that can be included is to preincubate the primary antibody (e.g., CCRC-M1) with a ligand known to be recognized by this antibody (e.g., fucose, which is an effective competitor for the CCRC-M1 binding site (18)) before applying the antibody to


7

Immunological Approaches to Plant Cell Wall and Biomass Characterization…

81

the section. In this case, antibody labeling of the section should be abolished. 4. It should be noted that the absence of labeling in immunohistochemistry can never be unambiguously interpreted as an indication of the absence of the epitope. 5. It must also be kept in mind that glycan-directed antibodies should be primarily considered to be epitope-specific, rather than polymer-specific (5). This is because some epitopes might be present in more than one context among wall polymers [for example, arabinogalactan epitopes present on both polysaccharide (e.g., rhamnogalacturonan I) and peptide (e.g., arabinogalactan protein) backbones]. Furthermore, the absence of labeling by an antibody against one epitope on a class of cell wall glycans does not always mean that that class of cell wall glycan is not present in the tissue analyzed. For example, the absence of labeling using an antibody specific for methylated homogalacturonan (e.g., JIM7) does not mean that homogalacturonan is absent from the tissue, since the homogalacturonan might be demethylated, which would not be recognized by JIM7, but would be detected by an antibody like CCRC-M38, which binds to demethylated homogalacturonan. Likewise, an antibody that binds in vitro to a polysaccharide that was isolated using harsh chemical reagents that can alter its structure might not bind to the polysaccharide in its native form in the tissue section. For example, xylans are typically isolated from cell walls using strong base, which cleaves acetyl groups frequently present on this polysaccharide. If an antibody recognizes the deacetylated form of xylan (e.g., CCRC-M146) in ELISAs, it might not label the native acetylated xylan present in the tissue section.

Acknowledgment Research in our laboratory on immunohistochemistry of plant cell walls is supported by a grant from the NSF Plant Genome Program (IOS-0923992). Generation of the CCRC series of plant glycandirected monoclonal antibodies used in this work was supported by the NSF Plant Genome Program (DBI-0421683). References 1. Keegstra K (2010) Plant cell walls. Plant Physiol 154:483–486 2. Pauly M, Keegstra K (2010) Plant cell wall polymers as precursors for biofuels. Curr Opin Plant Biol 13:305–312

3. McCann MC, Knox JP (2011) Plant cell wall biology: polysaccharides in architectural and developmental contexts. In: Ulvskov P (ed) Plant polysaccharides: biosynthesis and bioengineering. Annu Plant Rev 41:343–366


82

U. Avci et al.

4. York WS, O’Neill MA (2008) Biochemical control of xylan biosynthesis—which end is up? Curr Opin Plant Biol 11:258–265 5. Pattathil S, Avci U, Baldwin D, Swennes AG, McGill JA, Popper Z, Bootten T, Albert A, Davis RH, Chennareddy C, Dong R, O’Shea B, Rossi R, Leoff C, Freshour G, Narra R, O’Neil M, York WS, Hahn MG (2010) A comprehensive toolkit of plant cell wall glycandirected monoclonal antibodies. Plant Physiol 153:514–525 6. Pattathil S, Avci U, Miller JS, Hahn MG (2012) Immunological approaches to plant cell wall and biomass characterization: Glycome Profiling. In: Himmel M (ed) Biomass conversion: Meth Mol Biol 908:63–74 7. Puhlmann J, Bucheli E, Swain MJ, Dunning N, Albersheim P, Darvill AG, Hahn MG (1994) Generation of monoclonal antibodies against plant cell wall polysaccharides. I. Characterization of a monoclonal antibody to a terminal α(1 → 2)-linked fucosyl-containing epitope. Plant Physiol 104:699–710 8. Marcus SE, Verhertbruggen Y, Hervé C, Ordaz-Ortiz JJ, Farkas V, Pedersen HL, Willats WG, Knox JP (2008) Pectic homogalacturonan masks abundant sets of xyloglucan epitopes in plant cell walls. BMC Plant Biol 8:60 9. Marcus SE, Blake AW, Benians TA, Lee KJ, Poyser C, Donaldson L, Leroux O, Rogowski A, Petersen HL, Boraston A, Gilbert HJ, Willats WG, Knox JP (2010) Restricted access of proteins to mannan polysaccharides in intact plant cell walls. Plant J 64:191–203 10. Knox JP (2008) Revealing the structural and functional diversity of plant cell walls. Curr Opin Plant Biol 11:308–313

11. Blanchette RA, Abad AR, Cease KR, Lovrien RE, Leathers TD (1989) Colloidal gold cytochemistry of endo-1,4-β-glucanase, 1,4-β-D-glucan cellobiohydrolase, and endo1,4-β-xylanase: ultrastructure of sound and decayed birch wood. Appl Environ Microbiol 55:2293–2301 12. Blanchette RA, Abad AR, Farrell RL, Leathers TD (1989) Detection of lignin peroxidase and xylanase by immunocytochemical labeling in wood decayed by basidiomycetes. Appl Environ Microbiol 55:1457–1465 13. Benhamou N (1989) Cytochemical localization of β-[1–4]-D-glucans in plant and fungal cells using an exoglucanase-gold complex. Electron Microsc Rev 2:123–138 14. Brorson SH, Roos N, Skjorten F (1994) Antibody penetration into LR-White sections. Micron 25:453–460 15. Newman GR, Hobot JA (2001) Resin microscopy and on-section immunocytochemistry, 2nd edn. Chapter 10. Resin embedding and immunolabeling, pp. 215-228 Springer, Berlin 16. Donaldson LA, Knox JP (2012) Localization of cell wall polysaccharides in normal and compression wood of Radiata pine: relationships with lignification and microfibril orientation. Plant Physiol 158:642–653 17. Brennan M, Harris PJ (2011) Distribution of fucosylated xyloglucans among the walls of different cell types in monocotyledons determined by immunofluorescence microscopy. Mol Plant 4:144–156 18. Freshour G, Bonin CP, Reiter WD, Albersheim P, Darvill AG, Hahn MG (2003) Distribution of fucose-containing xyloglucans in cell walls of the mur1 mutant of Arabidopsis. Plant Physiol 131:1602–1612


Chapter 8 A Method to Evaluate Biomass Accessibility in Wet State Based on Thermoporometry Bon-Wook Koo and Sunkyu Park Abstract The substrate accessibility to enzyme has been considered as one of the most important factors for biomass conversion. To avoid the irreversible collapse of pore structure during the drying of sample, the measurement needs to be performed in a wet state. In this report, a thermoporometry method based on DSC isothermal step procedure is explained in details. This detects the amount of nonfreezing bound water in a wet sample and the value is expressed into pore size distribution with the Gibbsâ&#x20AC;&#x201C;Thomson equation. Information on pore size distribution and pore volume can be used to evaluate biomass accessibility. Key words: Biomass accessibility, Pore size distribution, Pore volume, Thermoporometry

1. Introduction Lignocellulosic biomass has been a research and development focus as a feedstock for biomass conversion into transportation fuels and chemicals. It is inherently difficult to deconstruct the biomass using biocatalyst, e.g., cellulolytic enzymes, due to its heterogeneity. To enhance biomass conversion efficiency, pretreatment is a necessary step by making substrates more accessible to enzymes. The substrate accessibility has been considered one of the most important factors for biomass reactivity and can be measured by several analytical methods both in dry and wet states. However, the irreversible physical collapse of cell wall structure through the aggregation of cellulose chains is caused during the drying process (1), and thus, the measurement with dried sample could not provide comparable accessibility information as in a wet state (2). Several methods have been applied to evaluate the accessibility in a wet state based on the adsorption characteristics of probe molecules such as dye, cellulase, dextran, and water.

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_8, Š Springer Science+Business Media, LLC 2012

83


84

B.-W. Koo and S. Park

The Simonsâ&#x20AC;&#x2122; staining method measures accessible surface area through the adsorption of two dyes which have different diameters and affinity for the hydroxyl groups in biomass (1). The accessible surface area can be expressed based on the Langmuir isotherm with the amount of dye adsorption on cellulose (3). Cellulase itself has been used to evaluate the accessibility of biomass and the amount of protein adsorbed on substrates is determined by nitrogen analyzer (4). A method to differentiate cellulose and noncellulose accessibility has been developed based on the assumption that BSA is preferentially bounded to lignin (5). Fluorescencelabeled cellulase binding modules have been used to determine the accessibility of substrates (6). The measurement of pore properties such as pore size distribution and pore volume can provide information on the accessibility of substrate and solute exclusion technique has been used to assess pore size distribution of biomass in a swollen state using a series of dextran molecules with different sizes (7). A thermoporometry method using differential scanning calorimetry (DSC) or solid-state NMR can determine pore size distribution by measuring the amount of water that has its melting temperature depressed at each isothermal step procedure (2). Bound water absorbed in hydrophilic substrates is categorized as nonfreezing and freezing bound water (8, 9). Nonfreezing bound water is the first 1â&#x20AC;&#x201C;3 layers of water adjacent to a surface and does not freeze because the motion of water structures is severely limited by the association with the surfaces (10). Freezing bound water is water that has its melting temperature depressed called freezing point depression. The freezing bound water has thermodynamically different behavior than unbound water, and the quantity could be determined by the integration of either an exotherm (crystallization of water) or an endotherm (melting of water) (11). However, the freezing bound water is generally characterized by the integration of the endotherm, since the exothermic curve may not be accurately detected during cooling in some cases (12). Water held in the capillaries of porous materials has a depressed melting temperature because of the lower pressure at a curved interface in cavities. The melting temperature depression has a reciprocal relationship with the pore diameter and thus the pore size distribution can be determined. This is the principle of thermoporometry based on the phenomenon called freezing point depression (13, 14). This method has been applied to cellulose fibers with an isothermal step melting procedure (15). In this chapter, a practical protocol for determining pore size distribution using the DSC thermoporometry method is described in details, which can be applied to evaluate the accessibility of pretreated lignocellulosic substrate to enzymes.


8

2. Apparatus and Sample Preparation 2.1. Apparatus

A Method to Evaluate Biomass Accessibility in Wet State…

85

1. A differential scanning calorimetry equipped with a refrigerated cooling system to measure the heat flow during the experiment. 2. Aluminum hermetic pan and lid, which has enough inside volume to hold wet sample (Fig. 1a). 3. Temperature calibration of the DSC is performed using a small amount (a few mg) of deionized water. A pan including the water is prepared and cooled to −30°C. Then the temperature is maintained for 5 min to ensure that all of the water is frozen. The pan is then heated to 5°C at a heating rate of 1°C/min. It is important to use the same heating rate throughout the experiments; otherwise, the results are influenced by the heating rate. The temperature at which the ice starts to melt should be calibrated to 0.00°C (onset of the melting peak). Repeat the experiment to ensure the reproducibility of the melting value about ±0.02°C.

2.2. Sample Preparation

1. Place a small amount of sample to a humidity chamber to adjust the amount of moisture in sample. Optimal moisture ratio is 0.8–1.2 g/g (gram of water per gram of oven dry

Fig. 1. The Hermetic pan and lid and sample pan preparation.


86

B.-W. Koo and S. Park

sample) (see Note 1). Make sure that samples are not dried, as drying will irreversibly change the pore structure of biomass. 2. Predry the aluminum hermetic pan and lid by placing them into an oven and cool those in a desiccator to remove any moisture (see Note 2). 3. Measure the weight of the pan and lid. 4. Take approximately 10 mg of sample into the pan (Fig. 1b). If the moisture ratio is 1.0 g/g, it would be 5 mg dry weight. 5. Put sample on the bottom side of the pan. Make sure of a good contact between sample and the pan to avoid any unnecessary heat flow resistance. 6. Put the hermetic lid and seal the pan using a press (Fig. 1c). 7. Measure the total weight of the sample pan. 8. Prepare an empty pan sealed with a lid for control.

3. DSC Operation and Analysis 3.1. Isothermal Step Melting Procedure

The sample and the control pan are placed into the DSC and cooled to −40°C. The cooling is maintained for 5 min and the temperature is then raised to −30°C at 1°C/min. Subsequent heating steps to slightly higher temperatures (−20, −15, −10, −6, −4, −2, −1.5, −1.1, −0.8, −0.5, −0.2, and −0.1°C) are then performed in succession (see Note 3). In each step, the temperature is raised at 1°C/min to the target temperature and then the sample is maintained isothermally until the heat flow returns to the baseline value. An example of the isothermal step melting program and the resulting heat flow is shown in Fig. 2. Total heat (Ht, J/g) for each segment is obtained by the integration of each endothermic reaction. After running of the whole step procedure, the moisture ratio in the initial sample is determined.

3.2. Data Analysis

Pore size distribution is determined by measuring the amount of water that has its melting temperature depressed at each isothermal step procedure. The heat absorbed during the heating and isothermal time period is calculated by integrating the endotherm. Thus, a melting enthalpy (Hm) is calculated by subtracting a sensible heat (Cp·ΔT) from the total heat (Ht) for each segment as shown in Eq. (1). The sensible heat is determined using the first segment (−40°C to −30°C), assuming that there is no water melting. H m = H t − C p · ΔT .

(1)

The relationship between a pore diameter (D) and the depressed melting temperature (Tm) is described by Eq. 2, which reduces to the Gibbs–Thomson equation when the contact angle is assumed


8

A Method to Evaluate Biomass Accessibility in Wet State…

87

Fig. 2. The melting water in substrate during an isothermal step melting program.

Table 1 The melting temperature in DSC analysis and its corresponding pore diameter Tm (°C)

−30

−20

−15

−10

−6

−4

D (nm)

1.3

2.0

2.6

4.0

6.6

9.9

Tm (°C)

−1.5

−1.1

−0.8

−0.5

−0.2

−0.1

D (nm)

26.4

36.0

49.5

79.2

198.0

396.0

−2 19.8

to be 180° (16). The use of Eq. 2 is based on the assumptions that the substrates are not soluble in the water and its pore shape is cylindrical. Each melting temperature depression (DT) represents a specific pore diameter. ΔT = T0 − Tm = (−4·T0 · γ 1 s · cos q ) / (D · r · H f ),

(2)

where T0 is the melting temperature of unbound water (273.15 K), g1s is the surface energy at the ice–water interface (12.1 mJ/m2) (17), r and Hf are the density and the specific heat of fusion of freezing bound water, respectively, assumed to be the same as that of unbound water (1,000 kg/m3, 334 J/g) (2), D is the diameter of the pore, and DT is the melting temperature depression (K). Thus, water held in a smaller pore has a larger melting temperature depression. The relationship between the melting temperature and the pore diameter is shown in Table 1. Pore size distribution is plotted by the amount of freezing bound water based on the melting enthalpy (Hm) versus its specific pore size and the sum of freezing bound water represents the total pore volume of the sample.


88

B.-W. Koo and S. Park

4. Pitfalls and Major Assumptions ●

The surface chemistry of sample can affect the amount of bound water measured due to the different affinity with water molecules. For example, hemicellulose and lignin have more and less available hydroxyl group, respectively, per gram basis than cellulose. Therefore the interpretation should be performed carefully.

For the analysis of DSC step melting isotherms, followings have been assumed. ●

Pore shape is considered to be cylindrical although pores exist in a variety of shapes in biomass.

Several equation parameters have been assumed constant such as surface energy at the ice–water interface, contact angle, and the density and specific heat of fusion of freezing bound water.

There is no melting during the first segment to determine the sensible heat of wet samples.

5. Notes 1. Moisture ratio can be adjusted by gently squeezing water from the wet sample using wipes. It is noted that too high moisture ratio of sample may mask the melting behavior at each step, which decrease the detection sensitivity. Too low moisture ratio also causes inaccurate detection of freezing bound water. 2. Use latex gloves or tweezers to handle the pan. Heating flow detection is sensitive to any contaminants. 3. The temperature scales used in this procedure are optimized for one type of biomass. The scale can be adjusted to make each melting enthalpy having a similar value. References 1. Chandra R, Ewanick S, Hsieh C, Saddler J (2008) The characterization of pretreated lignocellulosic substrates prior to enzymatic hydrolysis, part 1: a modified Simons’ staining technique. Biotechnol Prog 24:1178–1185 2. Park S, Venditti R, Jameel H, Pawlak J (2006) Changes in pore size distribution during the drying of cellulose fibers as measured by differential scanning calorimetry. Carbohydr Polym 66:97–103

3. Inglesby M, Zeronian S (1996) The accessibility of cellulose as determined by dye adsorption. Cellulose 3:165–181 4. Kumar R, Wyman CE (2009) Access of cellulase to cellulose and lignin for poplar solids produced by leading pretreatment technologies. Biotechnol Prog 25:807–819 5. Rollin JA, Zhu Z, Sathitsuksanoh N, Zhang YHP (2011) Increasing cellulose accessibility is more important than removing lignin:


8

6.

7.

8.

9.

10.

11.

A Method to Evaluate Biomass Accessibility in Wet State…

a comparison of cellulose solvent based lignocellulose fractionation and soaking in aqueous ammonia, Biotechnol Bioeng 108:22–30 Jeoh T, Wilson DB, Walker LP (2002) Cooperative and competitive binding in synergistic mixtures of Thermobifida fusca cellulases Cel5A, Cel6B, and Cel9A. Biotechnol Prog 18:760–769 Mooney CA, Mansfield SD, Touhy MG, Saddler JN (1998) The effect of initial pore volume and lignin content on the enzymatic hydrolysis of softwoods. Bioresour Technol 64:113–119 Liu WG, Yao KD (2001) What causes the unfrozen water in polymers: hydrogen bonds between water and polymer chains? Polymer 42:3943–3947 Ping Z, Nguyen Q, Chen S, Zhou J, Ding Y (2001) States of water in different hydrophilic polymers–DSC and FTIR studies. Polymer 42:8461–8467 Berlin E, Kliman P, Pallansch M (1970) Changes in state of water in proteinaceous systems. J Colloid Interface Sci 34:488–494 Nakamura K, Hatakeyama T, Hatakeyama H (1981) Studies on bound water of cellulose by

89

differential scanning calorimetry. Textile Res J 51:607 12. Hori T, Zhang HS, Shimizu T, Zollinger H (1988) Change of water states in acrylic fibers and their glass transition temperatures by DSC measurements. Textile Res J 58:227–232 13. Brun M, Lallemand A, Quinson JF, Eyraud C (1977) A new method for the simultaneous determination of the size and shape of pores: the thermoporometry. Thermochim Acta 21:59–88 14. Burghoff HG, Pusch W (1979) Characterization of water structure in cellulose acetate membranes by calorimetric measurements. J Appl Polym Sci 23:473–484 15. Maloney T, Paulapuro H, Stenius P (1998) Hydration and swelling of pulp fibers measured with differential scanning calorimetry. Nordic Pulp Paper Res J 13:31–36 16. Skapski A, Billups R, Rooney A (1957) Capillary cone method for determination of surface tension of solids. J Chem Phys 26:1350–1351 17. Ishikiriyama K, Todoki M (1995) Pore size distribution measurements of silica gels by means of differential scanning calorimetry II. Thermoporosimetry. J Colloid Interface Sci 171:103–111


Part II Plant Cell Wall


Chapter 9 Cellulase Processivity David B. Wilson and Maxim Kostylev Abstract There are two types of processive cellulases, exocellulases and processive endoglucanases. There are also two classes of exocellulases, ones that attack the reducing ends of cellulose chains and ones that attack the nonreducing ends. There are a number of ways of assaying processivity but none of them are ideal. It appears that exocellulases, all of which have their active sites in a tunnel, couple movement along a cellulose chain with cleavage of cellobiose from the end of the cellulose molecule. There are two sets of structures that suggest how an exocellulase might move along a cellulose chain. For family 48 exocellulases there are two different ways that a chain can be bound in the active site while for family 6 exocellulases there are several different ligand-bound structures. Site-directed mutagenesis of Thermobifida fusca exocellulases Cel48A and Cel6B and the processive endoglucanase Cel9A have identified some mutations that increase processivity and some that decrease processivity. In addition a mutation in Cel6B was identified that appears to allow the mutant enzyme to move along a cellulose chain in the absence of cleavage. Key words: Cellulose, Cellulase processivity, Exocellulases, Processive endoglucanases, Thermobifida fusca

1. Introduction The first processive cellulases to be identified were exocellulases (also called cellobiohydrolases), which attack the end of a cellulase chain and cleave off cellobiose residues sequentially from one end of a cellulose chain until they dissociate or stall (1). There are two classes of exocellulases; one class which attacks the reducing ends of cellulose chains is found in either family GH-7 or GH-48 (2), while the other class of exocellulases attacks the nonreducing ends of cellulose chains and it is found in family GH-6 (2). All known exocellulases have their active sites in a tunnel, which is consistent with their processive activity (3â&#x20AC;&#x201C;5). More recently, a new type of cellulase, processive endoglucanase, was discovered that contains a GH-9 catalytic domain with a family 3c carbohydrate binding

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_9, Š Springer Science+Business Media, LLC 2012

93


94

D.B. Wilson and M. Kostylev

module (CBM) bound at its C-terminus (6). The family 3c CBM was shown to be essential for processivity (7). Another type of processive endoglucanase was found recently that contains a family GH-5 catalytic domain (8). It appears that their processivity results from unusual subsite binding, as has been seen for some processive chitinases (9).

2. Processivity Assays There is no perfect assay for determining cellulase processivity, which is defined as the average number of cleavages that an enzyme carries out on a cellulose chain before it dissociates from the chain. One assay that has been used, measures both the amount of soluble reducing sugars that an enzyme produces from filter paper or other insoluble substrate during an appropriate incubation, often overnight, and the amount of insoluble reducing ends it produces in the filter paper during the same incubation. This can be done by removing the filter paper at the end of the incubation and measuring the reducing sugars in the solution and in the rinsed filter paper disc using the DNS assay (10). Exocellulases produce more than 93% of the total reducing sugar in the solution while endocellulases produce 30â&#x20AC;&#x201C;40% of the total reducing ends in the filter paper (insoluble). Thermobifida fusca Cel48A and Trichoderma reesei Cel7A showed the highest processivity in this assay with only 4% insoluble reducing ends while T. reesei Cel6A and T. fusca Cel6B gave 7% insoluble reducing ends. This may reflect the fact that family GH-6 cellulases contain only eight glucose-binding subsites while families GH-48 and GH-7 have more than ten such subsites (10). At this time it is not clear how the small number of insoluble ends are produced by an exocellulase, but it is unlikely that they result from endocellulolytic cleavages by the exocellulase. This is due to the fact that the buried surface area present in the loops that form the active site tunnel in exocellulases is large enough to prevent the tunnel from opening, especially in family GH-7 and GH-48 enzymes. The soluble-to-insoluble reducing sugar ratio assay is useful for distinguishing nonprocessive cellulases (most endoglucanases) from processive cellulases. A better assay for measuring the processivity of exocellulases is to determine the ratio of cellobiose to cellotriose that is produced by the exocellulase using HPLC (11). This assay is based on the assumption that during the first cleavage by an exocellulase there is an equal chance that it will produce either cellobiose or cellotriose depending on the stereochemistry of the chain end, as the glucose-binding subsites in an enzyme alternate in their binding specificity and cellulose chains are believed to have an equal number of each type of end. After the first cleavage, cellobiose will be the only product, as all the enzyme-bound ends have the same


9

Cellulase Processivity

95

stereochemistry after the first cleavage. Thus, more processive enzymes will produce a higher ratio of cellobiose to cellotriose. This assay requires that the hydrolysis of cellotriose by the enzyme is slow and that cellobiose is not hydrolyzed by the enzyme, which is true for most processive cellulases Cel6B slowly hydrolizes cellotriose so the formula: G3-G1/G2+G1. This assay gives a value of 12â&#x20AC;&#x201C;14 for the processivity of T. fusca WT Cel48A or Cel48cd on both bacterial cellulose and amorphous cellulose, (Kostylev M, Wilson DB, unpublished). Using electron microscopy, it was shown that T. reesei Cel7A acts processively from the reducing end towards the nonreducing end of crystals of Valonia cellulose (12). A direct assay for processivity is single-molecule studies using either fluorescently labeled cellulases or atomic force microscopy (AFM). An AFM study of T. reesei Cel7A and Cel6A, which was published in 2009, provided clear evidence for processivity in T. reesei Cel7A (CBHI) but the movement of Cel6A (CBH II) on a cellulose chain was very limited (13). The Cel7A molecules moved a distance covering from 35 to 50 CB units on the cellulose showing a processivity of near 50. Cel7Acd did not bind unless its concentration was tenfold higher than the native enzyme, but the cd moved at the same rate as the intact enzyme showing that the CBM was important for binding the catalytic domain to cellulose but was not needed to allow cleavage or movement along the cellulose molecule. An inactive Cel7A mutant did not show any movement, supporting the idea that for the wild-type enzyme, cleavage is essential for movement along a cellulose chain. There are two studies of the cleavage of oligosaccharides by T. reesei Cel6A, which show that it acts processively, since this enzyme hydrolyzes cellohexose to cellobiose without releasing cellotetraose (14, 15). Another assay of processivity is to label the reducing ends of cellulose by reacting them with a fluorescent group such as anthranilic acid or diaminopyridine. Then the release of labeled cellobiose under conditions that allow only one cycle of cellulase binding can be compared with the release of unlabelled cellobiose, which will be produced by all subsequent cleavages (16, 17). A similar assay is to reduce the reducing end to an alcohol, react the reduced cellulose with an exocellulase and measure the cellobiose produced along with the number of insoluble reducing groups that are produced when the alcohol group is cleaved off the end of a cellulose chain (16). By use of this assay, it was shown that the processivity of Cel7A from two different fungi was about three times higher on bacterial cellulose than on amorphous cellulose. The processivity values that have been determined seem fairly low relative to the length of the cellulose chains suggesting that release of the enzyme occurs easily, which is surprising given the large number of subsites in the active site tunnel of Cel7A. This may indicate that these assays have some undiscovered flaws.


96

D.B. Wilson and M. Kostylev

A kinetic model of the initial burst phase of T. reesei Cel7A acting prosessively on cellulose was proposed and tested by a calorimetric assay using amorphous cellulose. It was proposed that the initial binding and processive cleavage is fast but that the enzyme gets stalled and dissociation of the stalled enzyme is slow (18).

3. Mechanistic Studies of Processivity

There are two structural studies that provide possible mechanisms for the processive movement of exocellulases. In one, two different binding modes for long oligosaccharides were seen in structures of two different mutants of CelF (Cel48A) from Clostridium cellulolyticum. One mode was identical to that seen in the WT enzyme, which is believed to be the catalytic site, while the other was in a site above the catalytic site and it was suggested that this site might be used during processive movement of the cellulose chain (19). In the other study a set of structures of Cel6A from Humicola insolens bound to different ligands suggests that the processive movement is possible due to the flexibility of the hydrophobic residues that bind the cellulose in the active site as well as the extensive hydration of the bound cellulose (20). We have used the ratio of cellobiose to cellotriose assay to study the processivity of various site-directed mutants in the exocellulase: T. fusca Cel6B and the soluble/insoluble assay to study the processivity of mutants in T. fusca Cel9A. These experiments showed that certain mutations increase processivity while others decrease processivity and that both types of mutations were found in both enzymes. For Cel9A, it appeared that processivity depends on the balance between the binding affinity of the −4 to −1 subsites to the affinity of regions upstream of the cleavage site especially the family 3 CBM (19). Most mutations in potential substrate-binding residues in subsites −1 to −4 have decreased processivity and this seems reasonable, as the weaker binding would make dissociation of the chain from the CBM more likely than binding of the chain into the empty subsites after cleavage. It is less clear why most mutations in the family 3 CBM increase processivity since weaker binding to the CBM should increase both dissociation and movement of the chain into the subsites to about the same extent. A surprising finding is that a double mutant enzyme containing a cd mutation that increases productivity by itself and a CBM mutation that also increases processivity by itself, produces an enzyme with much lower than WT processivity and activity (21). The studies of T. fusca Cel6B mutants showed that there was not a strong correlation between the activity of a mutant enzyme and its processivity (11). In addition, mutation of Asp226 to Ala appeared to allow the mutant enzyme to move along the substrate


9

Cellulase Processivity

97

Fig. 1. Structural comparison of three GH family 48 exocellulases.

without cleavage. This is different from the WT enzyme and most mutant enzymes where movement along the substrate is coupled to cleavage (22). The evidence for this change was that the mutant enzyme had greatly reduced activity on swollen cellulose and bacterial cellulose but WT activity on carboxymethylcellulose [CMC]. Furthermore, it did not produce cellobiose from CMC but it did produce cellotriose, cellotetraose, cellopentose, and cellohexose suggesting that it made random cleavages along the CMC molecule. Further study is needed to explain how this mutation uncouples movement of the enzyme along the substrate from cleavage. For Cel48A, we have mutated surface residues that are close to the entrance of the active site tunnel (see Fig. 1). All such residues in the three GH-48 cellulases that we examined are potential cellulose-binding residues, and aromatic residues, which have the highest affinity for sugars, are about four times enriched relative to all the surface residues (19, 22). It is interesting that only 3 of the 13 tunnel entrance surface residues are conserved in all three enzymes. So far the three conserved residues have been mutated to Ala and the mutant enzymes have been characterized. Mutation of a highly conserved Trp residue(313) to Ala caused a decrease in activity on both amorphous and bacterial cellulose as does mutation of conserved Tyr213 while mutation of conserved Ser311 did not change the activity (Kostylev M, Wilson DB, unpublished). This seems reasonable, as the Tyr and Trp residues would be expected to bind cellulose more tightly than Ser. The Trp mutation decreased processivity while the other two mutations did not. A Trp residue is present at the entrance to the active site tunnel


98

D.B. Wilson and M. Kostylev

in all three exocellulase families, and it has been shown to be specifically required for crystalline cellulose hydrolysis in both family GH-6 and GH-48 exocellulases (Kostylev M, Wilson DB, unpublished) (24).

4. Conclusion There is clearly a need to develop better assays to measure processivity. In addition, more research is needed to understand why the measured processivity of most exocellulases is quite low even though the processivity predicted based on the ratio of the on and off rates for Cel7A on cellulose is much higher (17).

Acknowledgments This work was supported by the BioEnergy Science Center (BESC), which is a part of the U.S. Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science, Oak Ridge National Laboratory. We thank Mo Chen for preparing the figure. References 1. Teeri TT, Koivula A, Linder M, Wohlfahrt G, Divne C, Jones TA (1998) Trichoderma reesei cellobiohydrolases: why so efficient on crystalline cellulose? Biochem Soc Trans 26:173–178 2. Barr BK, Hsieh YL, Ganem B, Wilson DB (1996) Identification of two functionally different classes of exocellulases. Biochemistry 35:586–592 3. Rouvinen J, Bergfors T, Teeri T, Knowles JK, Jones TA (1990) Three-dimensional structure of cellobiohydrolase II from Trichoderma reesei. Science 249:380–386 4. Divne C, Stahlberg J, Reinikainen T, Ruohonen L, Pettersson G, Knowles JKC, Teeri TT, Jones A (1994) The three-dimensional structure of the catalytic core of cellobiohydrolase I from Trichoderma reesei. Science 265:524–528 5. Parsiegla G, Juy M, Reverbel-Leroy C, Tardif C, Belaich JP, Driguez H, Haser R (1998) The crystal structure of the processive endocellulase CelF of Clostridium cellulolyticum in complex

6.

7.

8.

9.

with a thiooligosaccharide inhibitor at 2.0 Å resolution. EMBO J 17:5551–5562 Sakon J, Irwin D, Wilson DB, Karplus PA (1997) Structure and mechanism of endo/ exocellulase E4 from Thermomonospora fusca. Nat Struct Biol 4:810–818 Irwin D, Shin D-H, Zhang S, Barr BK, Sakon J, Karplus PA, Wilson DB (1998) Roles of the catalytic domain and two cellulose binding domains of Thermomonospora fusca E4 in cellulose hydrolysis. J Bacteriol 180:1709–1714 Watson BJ, Zhang H, Longmire AG, Moon YH, Hutcheson SW (2009) Processive endoglucanases mediate degradation of cellulose by Saccharophagus degradans. J Bacteriol 191: 5697–5705 Zakariassen H, Aam BB, Horn SJ, Vårum KM, Sørlie M, Eijsink VG (2009) Aromatic residues in the catalytic center of chitinase A from Serratia marcescens affect processivity, enzyme activity, and biomass converting ef fi ciency. J Biol Chem 284:10610–10617


9 10. Irwin DC, Spezio M, Walker LP, Wilson DB (1993) Activity studies of eight purified cellulases: specificity, synergism, and binding domain effects. Biotechnol Bioeng 42:1002–1013 11. Vuong TV, Wilson DB (2009) Processivity, synergism, and substrate specificity of Thermobifida fusca Cel6B. Appl Environ Microbiol 75:6655–6661 12. Imai T, Boisset C, Samejima M, Igarashi K, Sugiyama J (1998) Unidirectional processive action of cellobiohydrolase Cel7A on Valonia cellulose microcrystals. FEBS Lett 432:113–116 13. Igarashi K, Koivula A, Wada M, Kimura S, Penttilä M, Samejima M (2009) High speed atomic force microscopy visualizes processive movement of Trichoderma reesei cellobiohydrolase I on crystalline cellulose. J Biol Chem 284:36186–36190 14. Harjunpää V, Teleman A, Koivula A, Ruohonen L, Teeri TT, Teleman O, Drakenberg T (1996) Cello-oligosaccharide hydrolysis by cellobiohydrolase II from Trichoderma reesei. Association and rate constants derived from an analysis of progress curves. Eur J Biochem 240:591 15. Nidetsky B, Zachariae W, Gercken G, Hayn M, Steiner W (1994) Hydrolysis of cello-oligosaccharides by Trichoderma reesei cellobiohydrolases; experimental data and kinetic modeling. Enzyme Microb Technol 16:43–52 16. Kurasin M, Väljamäe P (2011) Processivity of cellobiohydrolases is limited by the substrate. J Biol Chem 286:169–177 17. Kipper K, Väljamäe P, Johansson G (2005) Processive action of cellobiohydrolase Cel7A from Trichoderma reesei is revealed as ‘burst’ kinetics on fluorescent polymeric model substrates. Biochem J 385:527–535

Cellulase Processivity

99

18. Praestgaard E, Elmerdahl J, Murphy L, Nymand S, McFarland KC, Borch K, Westh P (2011) A kinetic model for the burst phase of processive cellulases. FEBS J 10.1111/j.1742-4658 19. Parsiegla G, Reverbel C, Tardif C, Driguez H, Haser R (2008) Structures of mutants of cellulase Cel48F of Clostridium cellulolyticum in complex with long hemithiocello oligosaccharides give rise to a new view of the substrate pathway during processive action. J Mol Biol 375:499–510 20. Varrot A, Frandsen TP, von Ossowski I, Boyer V, Cottaz S, Driguez H, Schülein M, Davies GJ (2003) Structural basis for ligand binding and processivity in cellobiohydrolase Cel6A from Humicola insolens. Structure 11:855–864 21. Li Y, Irwin DC, Wilson DB (2007) Processivity, substrate binding, and mechanism of cellulose hydrolysis by Thermobifida fusca Cel9A. Appl Environ Microbiol 73:3165–3172 22. Guimarães BG, Souchon H, Lytle BL, Wu D, Alzari PM (2002) The crystal structure and catalytic mechanism of cellobiohydrolase CelS, major enzymatic component of the Clostridium thermocellum cellulosome. J Mol Biol 320:587–596 23. Vuong TV, Wilson DB (2009) The absence of a single identifiable catalytic base residue in Thermobifida fusca exocellulase Cel6B. FEBS J 276:3837–3845 24. Koivula A, Kinnari T, Harjunpää V, Ruohonen L, Teleman A, Drakenberg T, Rouvinen J, Jones TA, Teeri TT (1998) Tryptophan 272: an essential determinant of crystalline cellulose degradation by Trichoderma reesei cellobiohydrolase Cel6A. FEBS Lett 429:341–346


Chapter 10 A Simple Method for Determining Specificity of Carbohydrate-Binding Modules for Purified and Crude Insoluble Polysaccharide Substrates Oren Yaniv, Sadanari Jindou, Felix Frolow, Raphael Lamed, and Edward A. Bayer Abstract Experimental identification of carbohydrate-binding modules (CBM) and determination of ligand specificity of each CBM are complementary and compulsory steps for their characterization. Some CBMs are very specific for their primary substrate (e.g., cellulose), whereas others are relatively promiscuous or nonspecific in their substrate preference. Here we describe a simple procedure based on in-tube adsorption of a CBM to various insoluble polysaccharides, followed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS PAGE) for determining the distribution of the CBM between the bound and unbound fractions. This technique enables qualitative assessment of the binding strength and ligand specificity for each CBM. Key words: Carbohydrate-binding module (CBM), Binding assay, Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS PAGE)

1. Introduction The noncatalytic polysaccharide-recognizing modules of glycoside hydrolases were originally defined as CBDs (cellulose-binding domains), because the first examples of these protein domains bound crystalline cellulose as their primary ligand (1â&#x20AC;&#x201C;3). Subsequently, the more inclusive term CBM (carbohydrate-binding module) evolved to reflect the diverse ligand specificity of these modules (4). Many CBMs have now been identified experimentally, and several hundred putative CBMs can be further identified on the basis of amino acid similarity. In some CBM families, ligand specificity is invariant, while other families contain proteins that bind to a range of different carbohydrates (5). Polysaccharide binding assay, presented in this Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_10, Š Springer Science+Business Media, LLC 2012

101


102

O. Yaniv et al.

chapter, is a simple and elegant procedure aiming to identify CBM experimentally and to determine qualitatively the binding strength and the ligand specificity of each CBM. The procedure is suitable for most common insoluble polysaccharides, and it composed of two steps, in tube polysaccharide adsorption step in which each polysaccharide is incubated with purified sample of CBM, followed by protein separation on sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (SDS PAGE) (6) for visualization of the results. The binding of family 3 CBMs (CBM3) to insoluble cellulosic polymers is used as an example for this approach. Quantitative estimation of affinity constants can be achieved by an extension of this approach as outlined in an accompanying chapter (7).

2. Materials Prepare all solutions using analytical grade reagents and ultrapure water (prepared by purifying deionized water to attain a sensitivity of 18 MW cm at 25°C). Prepare and store all reagents at room temperature (unless indicated otherwise). Follow all waste disposal regulations carefully when disposing waste materials. 2.1. Carbohydrate Binding Assay

1. Binding buffer: 50 mM phosphate buffer, 300 mM NaCl, pH 8.0 or 50 mM Tris–HCl buffer, 300 mM NaCl, pH 7.5. 2. Polysaccharides (most common insoluble polysaccharides are suitable for this assay): Avicel (microcrystalline cellulose, Merck AG, Darmstadt, Germany), amorphous cellulose (8) (see Note 1), xylan (from birch wood/oat spelt, Sigma Chem. Co., St. Louis, MO), chitin (from crab shells, Sigma), lichenan (Sigma), polygalacturonic acid (Sigma) (see Note 2), pectin (Sigma) (see Note 2), neutral detergent fiber of alfalfa cell walls, banana fruit stem, and wheat straw (9) (see Note 3). 3. Purified CBMs (20–40 mg) (see Note 4).

2.2. SDS Polyacrylamide Gel

1. Resolving gel buffer: 1.5 M Tris–HCl, pH 8.8, 0.4% SDS. Store at 4°C. 2. Stacking gel buffer: 0.5 M Tris–HCl, pH 6.8, 0.4% SDS. Store at 4°C. 3. 30% Acrylamide/bis solution (29:1). Store at 4°C. 4. Ammonium persulfate: 10% Solution in water. 5. N,N,N,N¢-tetramethyl-etyhlenediamine (TEMED). Store at 4°C. 6. Running buffer: 192 mM glycine, 25 mM Tris–HCl, pH 8.9, 0.1% SDS.


10

A Simple Method for Determining Specificity of Carbohydrate-Binding Modules…

103

7. Sample buffer: 62.5 mM Tris–HCl pH 6.8, 10% glycerol, 5% b-mercaptoethanol, 3% SDS, 0.05% bromophenol blue. Store at −20°C. 8. 50% Glycerol. 9. Staining buffer: 25% (v/v) isopropanol, 10% (v/v) acetic acid, 0.05% (w/v) Coomassie brilliant Blue R-250. 10. Destaining buffer: 20% (v/v) methanol, 7% (v/v) acetic acid.

3. Methods Carry out all procedures at room temperature unless otherwise specified. 3.1. Carbohydrate Binding Assay

1. Preliminary washing of polysaccharides: Mix 5 mg of each polysaccharide with 200 mL binding buffer in 1.5 mL Eppendorf tube, centrifuge at 12,000 × g for 5 min to sediment the polysaccharides. Discard the supernatant fluids. 2. Binding step: Mix 20–40 mg of the purified CBM with binding buffer to a final volume of 200 mL. Transfer the mixture to the polysaccharide-containing Eppendorf tube (see Note 5). 3. Maintain the assay tubes at room temperature for 1 h with gentle rotation. 4. Centrifuge the assay tubes at 12,000 × g for 10 min to sediment the polysaccharides. Transfer the supernatant fluids (containing the unbound fraction) to 1.5 mL Eppendorf tubes. 5. Determination of unbound fraction concentration: Measure the absorbance at a wavelength of 280 nm, and calculate the concentration of the sample according to its extinction coefficient (see Note 6). 6. Add 100 mL SDS gel sample buffer to each of the Eppendorf tubes containing the unbound fraction. Heat at 95–100°C for 10 min. 7. Determination of adsorbed CBM concentration: Subtract the unbound fraction from the total protein added to the assay tube. 8. Washing: Add 1 mL of binding buffer to each polysaccharide containing eppendorf tube and centrifuge at 12,000 × g for 5 min. Discard the supernatant fluids. Repeat washing four times to remove traces of unbound protein. 9. Elution of bound fraction: Add 135 mL of binding buffer and 65 mL of SDS gel sample buffer and heat at 95–100°C for 10 min. 10. Centrifuge the bound fraction tubes at 12,000 × g for 10 min to sediment the polysaccharides.


104

O. Yaniv et al.

3.2. 12.5% SDS Polyacrylamide Gel Electrophoresis

1. Prepare the resolving gel (12.5%) by mixing 4.1 mL of acrylamide mixture, 3.2 mL of water, 2.5 mL of separating gel buffer, 200 mL of 50% glycerol, 40 mL of 10% APS, and 10 mL TEMED (see Note 7). 2. Cast gel within a 7.25 cm × 10 cm × 1.5 mm gel cassette. Gently overlay with water (see Note 8). Allow polymerization time of about 30 min. 3. Prepare the stacking gel 3.25 mL of water, 1.25 mL of stacking gel buffer, 0.5 mL of acrylamide mixture, 30 mL of 10% APS, and 5 mL TEMED. 4. Insert a 15-well gel comb immediately without introducing air bubbles. Allow polymerization time of about 15–20 min. 5. Load protein standard (5–10 mL), and then load 15–20 mL of each sample (see Note 9). 6. Apply a voltage of 8 V/cm and run gel until the dye front have entered into the resolving gel, increase to 15 V/cm and run the gel until the dye front has reached the bottom of the resolving gel. 7. Turn off the power supply, remove the glass plate from the electrophoresis apparatus and carefully pry the plates apart. 8. Immerse the gel in staining solution and place on slowly rotating platform for at least 2 h. 9. Transfer the stained gel to distaining buffer (see Note 10), place on slowly rotating platform for 4–8 h. Replace the distaining buffer 3–4 times (see Fig. 1).

4. Results Family 3 CBMs have been classified into three subtypes according to sequence homology. Family 3a and 3b are very similar in primary sequence, except that the very few known examples of CBM3a are components of cellulosomal scaffoldings and contain an extra loop that includes an additional cellulose-binding residue. Subfamily 3b can be further divided into 3b and 3b¢, where CBM3b¢ modules lack a conserved cellulose-binding Trp residue, which characterizes the 3a and 3b modules. Other sequence discrepancies indicate that the CBM3b subfamily exhibits more divergence than originally considered. In the described experiment, the binding ability of selected CBM3 modules to purified and crude cellulosic polymers was examined. The CBM3s that were investigated were CBM3b¢ from Acetivibrio cellulolyticus Cel9B, CBM3b¢ from Clostridium thermocellum Cel9U and Cel9V, and the scaffoldin-borne CBM3b from A. cellulolyticus ScaA. Furthermore, mutations in subtypes 3a and 3b¢ were also examined in order to check the contribution of the additional


10

A Simple Method for Determining Specificity of Carbohydrate-Binding Modules…

105

Fig. 1. Interaction of family-3 CBMs with cellulosic substrates. Recombinant forms (all highly expressed in Escherichia coli) of the designated CBM3s (20 mg) were mixed in predetermined amounts with samples of the given substrate (10 mg Avicel, 2.5 mg amorphous cellulose and 5 mg of the other substrates). The suspension was centrifuged, the washed pellet (bound CBM) and supernatant fluids (unbound CBM) were collected, mixed with sample buffer, and aliquots corresponding to 10% of the bound and unbound fractions were subjected to SDS-PAGE. Each experiment was repeated at least three times. The partition of the CBM3 bands between the bound (+) and unbound states (−) is shown. From Jindou et al. (10). Copyright © Wiley-Blackwell Publishing Ltd., FEMS Microbiol Lett 254, 308–316. doi:10.1111/j.1574-6968.2005.00040.x.

aromatic residues to cellulose binding. The additional mutated CBM3s that were examined were CBM3a (Y67A and W118A) from C. thermocellum CipA and CBM3b¢ (A112W) from A. cellulolyticus Cel9B. CBM3a from C. thermocellum CipA and CBM3b from C. thermocellum Cel9I were used as positive controls. As shown in Fig. 1, the positive controls bound strongly to the various polymers. In contrast, all three CBM3b¢ modules failed to bind clearly to any of the cellulose polymers. In the scaffolding-borne CBM3a from C. thermocellum CipA, the mutation W118A had a deleterious effect on the binding of the mutant to cellulosic polymers, thus indicating the contribution of this residue to the cellulosebinding function of CBM3a,b. The mutation A112W in CBM3b¢ from A. cellulolyticus Cel9B, however, failed to restore binding to cellulose, indicating that in this case other factors are either required for, or interfere with, the interaction to the cellulosic polymers (10).


106

O. Yaniv et al.

5. Notes 1. Preparation of amorphous cellulose: Dissolve 10 g Avicel (Merck) in 500 mL concentrated phosphoric acid at 25°C. After the cellulose has dissolved (2 h), add 5 volumes of distilled water. Centrifuge the precipitated amorphous cellulose at 12,000 × g for 10 min. Resuspend in distilled water and recentrifuge. Repeat washing step six times. Adjust the pH in the final suspension (~12 mg mL) to 6.5 with 1 M NaOH (8). 2. Immerse pectin and polygalacturonic acid in buffer containing 7 mM CaCl2 in order to precipitate the polysaccharides (both are soluble in the absence of calcium) (11). 3. Preparation of neutral detergent fibers: Air-dry samples of alfalfa cell walls, wheat straw, and banana fruit stem. Ground the samples to pass 1 mm screen. Add 0.5 g of ground sample to 100 mL of neutral detergent solution containing 50 mg of heat stable a-amylase, 0.5 g of sodium sulfite (Na2SO3) and some drops of octyl-alcohol (n-octanol). Boil the sample in the solution for 1 h, and filter on Whatman 54 paper (Whatman, Piscataway, NJ). Wash each sample three times with boiling water, followed with two washes with cold acetone. Dry samples for 8 h at 105°C, and cool in desiccator (9). 4. Overexpression of recombinant CBM can be easily done in Escherichia coli (12). Purification can be done by using His-tag fusion protein (13) or by using the inherent affinity of some of the CBMs to cellulose (8). 5. To ensure homogeneity of CBM concentration. Prepare the CBM-binding buffer mixture for all polysaccharides in a single 2 mL Eppendorf tube, then dispense 200 mL from this 2 mL tube to each 1.5 mL tubes. 6. The extinction coefficient is based on the absorbance of the aromatic side chains (i.e., mainly tryptophan, with lesser but measurable contribution of tyrosine and even less by phenylalanine). The absorbance of 1 g protein per 1 L can be obtained using the Expasy ProtParam tool: http://expasy.org/tools/ protparam.html (14). 7. Polymerization of the resolving and stacking gel begins as soon as the TEMED has been added. 8. Water overlay prevents contact with atmospheric oxygen (which inhibits acrylamide polymerization) and helps leveling the separating gel solution. 9. For every polysaccharide: Load bound fraction into one well and unbound fraction into the adjacent well. 10. Staining buffer can be saved for future use.


10

A Simple Method for Determining Specificity of Carbohydrate-Binding Modules…

107

References 1. Vantilbeurgh H, Tomme P, Claeyssens M, Bhikhabhai R, Pettersson G (1986) Limited proteolysis of the cellobiohydrolase I from Trichoderma reesei—separation of functional domains. Febs Lett 204:223–227 2. Tomme P, Van Tilbeurgh H, Pettersson G, Van Damme J, Vandekerckhove J, Knowles J, Teeri T, Claeyssens M (1988) Studies of the cellulolytic system of Trichoderma reesei QM 9414. Analysis of domain function in two cellobiohydrolases by limited proteolysis. Eur J Biochem 170:575–581 3. Gilkes NR, Warren RAJ, Miller RC, Kilburn DG (1988) Precise excision of the cellulose binding domains from 2 Cellulomonas fimi cellulases by a homologous protease and the effect on catalysis. J Biol Chem 263:10401–10407 4. Boraston AB, McLean BW, Kormos JM, Alam M, Gilkes NR, Haynes CA, Tomme P, Killburn DG, Warren RAJ (1999) Carbohydrate-binding modules: diversity of structure and function. Royal Society of Chemistry, Cambridge 5. Boraston AB, Bolam DN, Gilbert HJ, Davies GJ (2004) Carbohydrate-binding modules: finetuning polysaccharide recognition. Biochem J 382:769–781 6. Shapiro AL, Vinuela E, Maizel JV (1967) Molecular weight estimation of polypeptide chains by electrophoresis in SDS-polyacrylamide gels. Biochem Biophys Res Commun 28:815 7. Fraiberg M, Borovok I, Lamed R, Bayer EA (2011) Bacterial cadherin domains as carbohydrate binding modules: Determination of affinity constants to insoluble complex polysaccharides. Methods Mol. Biol. This volume, Chapter 11

8. Lamed R, Kenig R, Setter E, Bayer EA (1985) Major characteristics of the cellulolytic system of Clostridium thermocellum coincide with those of the purified cellulosome. Enzyme Microb Technol 7:37–41 9. Vansoest PJ, Robertson JB, Lewis BA (1991) Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J Dairy Sci 74: 3583–3597 10. Jindou S, Xu Q, Kenig R, Shulman M, Shoham Y, Bayer EA, Lamed R (2006) Novel architecture of family-9 glycoside hydrolases identified in cellulosomal enzymes of Acetivibrio cellulolyticus and Clostridium thermocellum. FEMS Microbiol Lett 254:308–316 11. Kahel-Raifer H, Jindou S, Bahari L, Nataf Y, Shoham Y, Bayer EA, Borovok I, Lamed R (2010) The unique set of putative membraneassociated anti-sigma factors in Clostridium thermocellum suggests a novel extracellular carbohydrate-sensing mechanism involved in gene regulation. FEMS Microbiol Lett 308:84–93 12. Baneyx F (1999) Recombinant protein expression in Escherichia coli. Curr Opin Biotechnol 10:411–421 13. Hengen PN (1995) Methods and reagents— purification of His-Tag fusion proteins from Escherichia coli. Trends Biochem Sci 20:285–286 14. Gasteiger E, Gattiker A, Duvaud S, Wilkins MR, Appel RD, Bairoch A (2005) Protein identification and analysis tools on the ExOASy server. In: Walker JM (ed) The proteomics protocols handbook. Humana, Tolowa, NJ, pp 571–607


Chapter 11 Bacterial Cadherin Domains as Carbohydrate Binding Modules: Determination of Affinity Constants to Insoluble Complex Polysaccharides Milana Fraiberg, Ilya Borovok, Ronald M. Weiner, Raphael Lamed, and Edward A. Bayer Abstract Cadherin (CA) and cadherin-like (CADG) doublet domains from the complex polysaccharide-degrading marine bacterium, Saccharophagus degradans 2â&#x20AC;&#x201C;40, demonstrated reversible calcium-dependent binding to different complex polysaccharides, which serve as growth substrates for the bacterium. Here we describe a procedure based on adsorption of CA and CADG doublet domains to different insoluble complex polysaccharides, followed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) for visualizing and quantifying the distribution of cadherins between the bound and unbound fractions. Scatchard plots were employed to determine the kinetics of interactions of CA and CADG with several complex carbohydrates. On the basis of these binding studies, the CA and CADG doublet domains are proposed to form a new family of carbohydrate-binding module (CBM). Key words: Cadherin-like, Binding affinity, CBM, Carbohydrate binding assay, SDS-PAGE, Scatchard plot

1. Introduction The cadherins constitute a large family of calcium-dependent cell-adhesion proteins, which, in higher organisms, play a major role in development and tissue morphogenesis (1, 2). The homotypic adhesion between cells arises from homophilic interactions between extracellular tandemly repeated cadherin domains (3). In the past, the cadherins were broadly studied in the metazoan lineage, but in the prokaryotic world their biological role has been poorly explored.

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_11, Š Springer Science+Business Media, LLC 2012

109


110

M. Fraiberg et al.

The genome of Saccharophagus degradans contains at least 52 CA and CADG domains in five large secreted proteins (4). In the CabC and CabD proteins (3477 and 3474 amino acid residues, respectively) the CA and CADG domains (residues 2288–2496 and 2261–2500, respectively) each occurs as a doublet (single repeat). Both displayed carbohydrate-binding features for a wide range of different insoluble complex polysaccharides, in a calciumdependent manner (5). In this chapter we present a qualitative carbohydrate-binding assay of bacterial cadherins to several complex carbohydrates and quantitative determination of binding parameters for these interactions by Scatchard analysis. The procedure is composed of an in-tube polysaccharide adsorption step, in which each polysaccharide is incubated with a purified sample of CA or CADG, followed by protein separation by SDS-PAGE (6) for visualization of the results. The kinetics of the interactions between the CA/CADG doublet domains and the complex carbohydrates exhibiting the highest binding capacities were further investigated by determining the distribution of bound and free fractions and analyzing the data by Scatchard plots (7).

2. Materials Prepare all solutions using analytical grade reagents and ultrapure water (prepared by purifying deionized water to attain a sensitivity of 18 MW cm at 25 °C). Prepare and store all reagents at room temperature (unless otherwise indicated). Follow all waste disposal regulations carefully, when disposing waste materials. 2.1. Cloning

1. DNA-spinTM—Total DNA extraction kit (iNtRON Biotechnology, Gyeonggi-do, Korea). 2. i-MAXTM DNA Polymerase (iNtRON Biotechnology). 3. MEGA quick-spinTM PCR & Agarose Gel DNA .Extraction kit (iNtRON Biotechnology). 4. pET28a Vector (Novagen, Darmstadt, Germany). 5. dNTPs (New England Biolabs, Ipswich, MA). 6. Restriction enzymes: NcoI, NotI (New England Biolabs). 7. ReadyMix for PCR (New England Biolabs or Sigma). 8. DNA-spinTM—plasmid Biotechnology).

DNA

purification

kit

(iNtRON

9. Ethidium bromide (Sigma Chem. Co., St. Louis, MO): 0.5 mg/mL aqueous stock solution. 10. Ligase (New England Biolabs): 400,000 U/mL. 11. Kanamycin (Sigma): 50 mg/mL aqueous stock solution.


11

Bacterial Cadherin Domains as Carbohydrate Binding Modules…

111

12. Luria Bertani (LB) medium: 10 g Bacto-Trypton, 5 g yeast extract, 10 g NaCl per 1 L H2O. Add 2 % agar for LB agar plates. 13. TAE × 50 buffer: 2 M Tris, 1 M glacial acetic acid, 50 mM EDTA, pH 8.0. 2.2. Protein Expression

Isopropyl b-D-thio-galactopyranoside (IPTG) (New England Biolabs): 1 M aqueous stock solution.

2.3. Protein Purification

1. Phenylmethylsulphonylfluoride (PMSF) (Sigma): 0.1 M ethanolic stock solution. 2. His-tag capture resin (Zephyr ProteomiX, Kiryat-Shmona Israel). 3. Sonication buffer: 50 mM Na2HPO4, 50 mM NaH2PO4, pH 8.0, 300 mM NaCl. 4. Washing buffer: 50 mM Na2HPO4, 50 mM NaH2PO4, pH 6.0, 300 mM NaCl, 10 % glycerol. 5. Elution buffer: 50 mM Na2HPO4, 50 mM NaH2PO4, pH 6.0, 300 mM NaCl, 10 % glycerol, 300 mM imidazole. 6. Superdex®75 XK 16 gel filtration column (GE Healthcare, Piscataway, NJ), 80 mL column volume, 1.6 cm diameter. 7. Gel filtration buffer: 50 mM Tris–HCl, pH 7.5, 0.15 M NaCl, 0.05 % NaN3. 8. Äkta-prime system (GE Healthcare). 9. Centriprep Centrifugal Filter Unit with Ultracel-3 membrane (Millipore, Billerica, MA).

2.4. Binding Assay

1. Binding buffer: 50 mM Tris–HCl buffer, 300 mM NaCl, 7 mM CaCl2 (or 10 mM EDTA), pH 7.5. 2. Complex carbohydrates: Avicel (microcrystalline cellulose, Merck AG, Darmstadt, Germany), agar (Acumedia Manufacturers, Inc., Lansing, MI), agarose (Hispanagar, Burgos, Spain), starch (Sigma), pectin (Sigma), lichenan (Sigma), chitin (from crab shells, Sigma), xylan (from birch wood/oat spelts, Sigma). 3. Bradford reagent (Sigma).

3. Methods Carry out all procedures at room temperature unless otherwise specified. 3.1. Cloning of CA and CADG Doublet Domains into pET28a

1. Isolate S. degradans 2–40 chromosomal DNA using DNAspinTM-Total DNA extraction kit according to the manufacturer’s manual.


112

M. Fraiberg et al.

Table 1 Constructs, expression vectors, restriction enzymes, and primers used in this study Construct name CAHis6 (CabC/ Sde_3323)

Expression Restriction vectors enzymes Primer sequences (5¢–3¢) pET28a

NcoI, NotI

F¢: ACCATGGCCGCAGTAGAAGTAAATGTATTG R¢: GCGGCCGCGCCATCGATATCGTCATCATTAAC

CADGHis6 (CabD/ pET28a Sde_0798)

NcoI, NotI

F¢: AACCATGGCTGTAAATGAAGATTCGGTG R¢: TTGCGGCCGCTATACTTTCCGATCCGTCGTGC

2. Prepare PCR reaction of CA and CADG doublet domains using appropriate primers (Table 1) as follows: 100 ng genomic DNA of S. degradans 2–40, 50 pmol of each tailed primer, 1 mL of i-Max Taq DNA polymerase, 5 mL of 10× i-Max Taq buffer, 2.5 mM dNTPs mix, brought to 50 mL with sterile double distilled water. 3. Perform PCR amplification as follows: Pre-denaturation step of 4 min at 94 °C, 29 cycles comprise 20 s denaturation steps at 94 °C, a 20-s annealing step at 60 °C, an extension step at 72 °C for 1 min, and a final extension step at 72 °C for 10 min. Store at 4 °C. 4. Purify the PCR products using a PCR product purification kit according to the manufacturer’s manual. 5. Separate PCR products on 1.5 % agarose gel, containing 0.5 mg/mL ethidium bromide (8). 6. Cut appropriate DNA bands from the gel, and purify DNA using a MEGA agarose gel purification kit according to the manufacturer’s manual. 7. Restrict PCR-amplified products and pET28a by NcoI and NotI as follows: 1 mg of purified DNA (insert or plasmid), 10 units of each restriction enzyme, suitable buffer diluted tenfold, brought to 50 mL with double distilled water. Incubate for 1 h at 37 °C. 8. Ligate DNA encoding CA and CADG into pET28a as follows: 5 mL plasmid, 12 mL PCR product, 2 mL ligase buffer × 10, 1 mL ligase. Incubate overnight at 16 °C. 9. Introduce the ligated DNA into 100 mL of competent Escherichia coli XL 1-Blue host cells as follows: Add total volume (20 mL) of the DNA into 100 mL of the competent cells, place on ice for 30 min, heat 2 min at 42 °C, add 1 mL of LB medium, and incubate for 1 h at 37 °C. Grow single colonies using LB agar plate containing added kanamycin (50 mg/mL). 10. Check for the presence of the desired sequence in the colonies obtained after transformation by PCR using ReadyMix, 2 ml of


11

Bacterial Cadherin Domains as Carbohydrate Binding Modules…

113

each primer, brought to 10 ml with sterile double distilled water. Perform PCR amplification as was described above, and separate PCR products on 1.5 % agarose gel contained 0.5 mg/mL ethidium bromide (see Note 1). 11. Purify pET28a with cloned DNA for CA or CADG domains using plasmid purification kit according to the manufacturer’s instructions. 12. Introduce the resulting recombinant plasmid into competent E. coli BL21(lde3) pLysS cells (see Note 2). 3.2. Expression of CA and CADG Doublet Domains

1. Dilute an overnight culture of E. coli BL21(lde3) pLysS cells to a cell density of 0.1 (A600) in LB containing kanamycin (50 mg/mL), and shake vigorously at 37 °C. 2. When the culture reaches a cell density of 0.6 (A600), add IPTG to a final concentration of 0.5 mM. 3. Incubate the culture for an additional 3 h at 37 °C (see Note 3). 4. Harvest the cells by centrifugation at 4,000 × g for 20 min at 4 °C; store the pellet at −20 °C.

3.3. Purification of CA and CADG Doublet Domains

1. Suspend the pellet in sonication buffer containing 1 mM of PMSF. 2. Sonicate the mixture until clear in a discontinuous mode (10 s pulse on and 15 s pulse off) (see Note 4). 3. Centrifuge the sonicate at 15,000 × g for 15 min at 4 °C. 4. Mix the supernatant with His-tag capture resin (with ratio of 1 mL resin per 70 mL of supernatant fluids) equilibrated with sonication buffer. 5. Incubate this suspension for 1 h at 4 °C with gentle rocking. 6. Centrifuge the suspension for 5 min at 2,000 × g and discard the supernatant. 7. Wash the pellet with 40 mL washing buffer and centrifuge for 5 min at 2,000 × g, repeat three times. 8. Add 2 mL of elution buffer, incubate for 5 min, centrifuge for 5 min at 2,000 × g, and collect the resolved supernatant (desired protein). Repeat the elution four more times. 9. Inject 2 mL of His-tagged protein into a Superdex®75 XK 16 gel filtration column, using gel filtration buffer in the effluent. 10. Collect the purified peak fractions using an Äkta system (GE Healthcare), flow rate 1.5 mL/min, and concentrate the proteins using a Centriprep centrifugal filter device with a molecular weight cut off of 3 kDa. Store at −20 °C. 11. Visualize the purified proteins by SDS-PAGE 12.5 % (9).


114

M. Fraiberg et al.

3.4. CA Binding Assay to Different Complex Carbohydrates

1. Preliminary washing of polysaccharides: Mix 15 mg of each insoluble polysaccharide preparation with 200 mL binding buffer in 2 mL Eppendorf tube, centrifuge at 12,000 × g for 2 min to sediment the polysaccharides. Discard the supernatant fluids. Repeat five times. 2. Binding: Mix 0.5 mg/mL of the purified CA/CADG with binding buffer to a final volume of 200 mL. Transfer the mixture into the prewashed carbohydrate-containing Eppendorf tube (see Note 5). 3. Maintain the assay tubes at room temperature for 20 min with gentle rotation. 4. Centrifuge the assay tubes at 12,000 × g for 20 min to sediment the carbohydrates. Transfer the supernatant fluids (containing the unbound fraction) into a 1.5 mL Eppendorf tube. 5. Determine unbound fraction concentration: quantify the protein concentration by Bradford reagent using bovine serum albumin as a standard. 6. Determine protein concentration in bound fractions: subtract the unbound fraction from the total protein added to the assay tube (see Note 6). 7. Add 10 mL sample buffer to 40 mL of each unbound fraction. Heat at 95–100 °C for 10 min. 8. Wash the pellet fractions that contain the carbohydrates and bound protein with 1 mL binding buffer, centrifuge at 12,000 × g for 5 min. Discard the supernatant fluids. Repeat washing four times to remove traces of unbound protein. 9. Dilute the bound fraction to 80–200 mL with washing buffer, add 50 mL sample buffer to each bound fraction. Heat at 95–100 °C for 10 min. 10. Subject the bound and unbound fractions to SDS-PAGE (9) (see Note 7). Both CA and CADG doublet domains bound to most of complex carbohydrates in a calcium-dependent manner. The proteins failed to bind these complex carbohydrates in the presence of 10 mM EDTA. Moreover, the binding reaction was found to be reversible, and all bound proteins could be removed from the complex carbohydrates by incubation of the bound fraction with 10 mM EDTA (see Fig. 1).

3.5. Estimation of Binding Capacities of CA and CADG to Complex Carbohydrates

1. Multiply the bound CA/CADG concentration by total reaction volume (200 mL) and divide by the amount of insoluble complex polysaccharides (15 mg). 2. Choose the reactions which exhibit the highest binding capacities for further investigation and estimation of the kinetics of interaction between CA/CADG and insoluble complex polysaccharides (see Fig. 2).


11

Bacterial Cadherin Domains as Carbohydrate Binding Modules…

115

Fig. 1. Interactions of the CA and CADG doublet domains with different complex carbohydrates. The CA and CADG doublet domains were mixed with the indicated insoluble complex polysaccharides in the presence of 7 mM CaCl2 or 10 mM EDTA as indicated. Samples were separated into supernatant (−) and pellet (+) fractions, which were subjected to SDS-PAGE. From Fraiberg et al. (2011). Copyright © American Society for Microbiology, J Bacteriol 193:283–285. doi:10.1128/ JB.00842-10.

Fig. 2. Binding capacity of CA and CADG doublet domains to different complex carbohydrates. Protein concentration in the unbound fraction was determined, and binding capacities for the different insoluble complex polysaccharides were calculated for CA and CADG, separately. From Fraiberg et al. (2011). Copyright © American Society for Microbiology, J Bacteriol 193:283–285. doi:10.1128/JB.00842-10.

3.6. Estimation of Binding Parameters

1. Incubate incremental concentrations of the CA or CADG (e.g., 0.1, 0.15, 0.2, 0.25, 0.3, 0.35 mg/mL) with 15 mg of the chosen complex polysaccharides, which gave the highest binding capacities. Follow the binding protocol as mentioned above (see Fig. 3). 2. Estimate CA/CADG concentrations in unbound fractions by Bradford reagent.


116

M. Fraiberg et al.

Fig. 3. Binding of CA and CADG doublet domains to insoluble carbohydrates as a function of incremental substrate concentrations. Concentrations between 0.1 and 0.35 mg/mL were tested. CA binding was performed with chitin, birchwood xylan and lichenan, and CADG binding was performed with pectin and chitin.

Fig. 4. Scatchard plot analysis of CA and CADG binding to selected complex carbohydrates. Incremental concentrations of CA and CADG doublet domains were mixed with the indicated insoluble complex polysaccharides. Samples were separated into supernatant (unbound) and pellet (bound) fractions, and protein was determined using the Bradford method. From Fraiberg et al. (2011). Copyright © American Society for Microbiology, J Bacteriol 193:283–285. doi:10.1128/JB.00842-10.

3. Calculate CA/CADG concentration in bound fractions as mentioned above. 4. Represent the results in Scatchard plots (see Fig. 4), where the Y-axis is the bound/free protein concentration and the X-axis is the bound protein concentration. 5. Determine the Kd and Bmax values by linear regression, where Bmax is the intercept at X-axis, and the slope of the linear curve is −1/Kd (Table 2).


11

Bacterial Cadherin Domains as Carbohydrate Binding Modules…

117

Table 2 Binding parameters for the interaction of CA and CADG with selected ICPs CA

Chitin

Xylan (from birch wood)

Lichenan

Kd (mM)

1.74 ± 0.1

1.37 ± 0.15

1.45 ± 0.12

Bmax (mM)

16.7 ± 2.3

14.6 ± 1.8

17.6 ± 2.5

CADG

Chitin

Pectin

Kd (mM)

1.08 ± 0.1

0.57 ± 0.11

Bmax (mM)

15 ± 1.5

15 ± 1.5

4. Notes 1. Extend the pre-denaturation step to 10 min, in order to break the cell wall and expose DNA. 2. Use E. coli BL21(lde3) pLysS cells for better protein expression. 3. Alternatively, incubate the culture at 16 °C overnight. 4. Add DNAse powder or 1 mM PMSF. Always keep the suspension on ice to prevent protein degradation. 5. To ensure homogeneity of CA/CADG concentration, prepare the CA/CADG containing binding buffer mixture for all complex carbohydrates in a single 2-mL Eppendorf tube, then distribute 200 mL from the tube into a new 2-mL tube for each reaction. 6. Use washing buffer with 10 mM EDTA in order to remove CA/CADG from bound fraction. 7. For each polysaccharide: Load bound fraction into one well and unbound fraction into the adjacent well. References 1. Koch AW, Manzur KL, Shan W (2004) Structure-based models of cadherin-mediated cell adhesion: the evolution continues. Cell Mol Life Sci 61:1884–1895 2. Leckband D, Prakasam A (2006) Mechanism and dynamics of cadherin adhesion. Annu Rev Biomed Eng 8:259–287 3. Tsuiji H, Xu L, Schwartz K, Gambiner BM (2007) Cadherin conformations associated with dimerization and adhesion. J Biol Chem 282:12871–12882

4. Fraiberg M, Borovok I, Weiner RM, Lamed R (2010) Discovery and characterization of cadherin domains in Saccharophagus degradans 2–40. J Bacteriol 192:1066–1074 5. Fraiberg M, Borovok I, Bayer EA, Weiner RM, Lamed R (2011) Cadherin domains in the polysaccharide-degrading marine bacterium Saccharophagus degradans 2–40 are carbohydrate-binding modules. J Bacteriol 193: 283–285


118

M. Fraiberg et al.

6. Shapiro AL, Vinuela E, Maizel JV (1967) Molecular weight estimation of polypeptide chains by electrophoresis in SDSpolyacrylamide gels. Biochem Biophys Res Commun 28:815 7. Scatchard G (1949) The attractions of proteins for small molecules and ions. Ann N Y Acad Sci 51:660â&#x20AC;&#x201C;672

8. Sambrook J, Russell D (2001) Molecular cloning: a laboratory manual, 3rd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY 9. Yaniv O, Jindou S, Frowlow F, Lamed R, Bayer EA. A simple method for determining specificity of carbohydrate-binding modules for insoluble purified and crude polysaccharide substrates. Methods Mol Biol This volume, Chapter 10


Chapter 12 Affinity Electrophoresis as a Method for Determining Substrate-Binding Specificity of Carbohydrate-Active Enzymes for Soluble Polysaccharides Sarah MoraĂŻs, Raphael Lamed, and Edward A. Bayer Abstract Affinity electrophoresis is a simple and rapid tool for the analysis of protein-binding affinities to soluble polysaccharides. This approach is particularly suitable for the characterization of the carbohydrate-active enzymes that contain a carbohydrate-binding module and for their mutants and chimeras. Knowledge of the binding characteristics of these enzymes can be the first step to elucidate the enzymatic activity of a putative enzyme; moreover in some cases, enzymes are able to bind polysaccharides targets other than their specified substrate, and this knowledge can be essential to understand the basics of the intrinsic mechanism of these enzymes in their natural environment. Key words: Carbohydrate-active enzymes, Carbohydrate-binding module (CBM), Substrate binding, Non-denaturating affinity electrophoresis

1. Introduction Carbohydrate-active enzymes, and glycoside hydrolases in particular, are modular proteins involved in the hydrolysis of soluble or insoluble polysaccharides (1). In addition to their catalytic module(s) that perform the actual cleavage of the polymeric substrate, a large number of these enzymes contain one or several carbohydrate-binding module (CBM) in their modular organization (2). One particularity of these enzymes is to have the ability to bind to their substrate or other polysaccharides either via their CBM or by the direct interaction of the catalytic module towards the polysaccharide or by the combined affinity of several modules for the sugar (3).

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_12, Š Springer Science+Business Media, LLC 2012

119


120

S. MoraĂŻs et al.

In this chapter, we present a rapid, simple, cheap, and efficient method based on non-denaturing affinity electrophoresis, which allows the identification of specific affinities for a protein to a soluble polysaccharide and insight into the strength of this interaction. The procedure is suitable for most common soluble polysaccharides and is composed of a single step, which involves direct protein application in a non-denaturing gel containing a soluble polysaccharide. Binding is easily detected by a reduced mobility of the protein in the gel containing the polysaccharide relative to the mobility of the protein in a gel lacking the polysaccharide. In addition to its value in identifying glycoside hydrolase ligands, affinity electrophoresis also provides a rapid screening method for comparing the binding characteristics of mutants or chimeras. In the present chapter, the binding of modular wild-type and chimeric xylanases to soluble xylan and amorphous cellulose is used as an example for this approach.

2. Materials Prepare all solutions using analytical grade reagents and ultrapure water (prepared by purifying deionized water to attain a sensitivity of 18 MW cm at 25 °C). Prepare and store all reagents at room temperature (unless indicated otherwise). Follow all waste disposal regulations carefully when disposing waste materials. 2.1. Affinity Electrophoresis Assay

1. Polysaccharides (most common soluble polysaccharides are suitable for this assay): amorphous cellulose (4), xylan (from beech wood, birch wood and oat spelts, Sigma Chem. Co., St. Louis, MO), chitin (from crab shells, Sigma), arabinan (Megazyme International, Ltd. Wicklow, Ireland), pectin (Sigma). 2. Purified carbohydrate-active enzymes (with or without resident CBMs) (5 mg minimum per well in the gel) (see Note 1). 3. Bovine serum albumin (BSA) as a negative non-binding control (see Note 2).

2.2. Preparation of Amorphous Cellulose

1. Dissolve 12 g Avicel (Merck) in 600 mL concentrated phosphoric acid. 2. Incubate for 2 h while stirring in the chemical hood at room temperature. 3. Add 2,400 mL of water to the homogeneous suspension, and centrifuge 12,000 Ă&#x2014; g for 10 min. 4. Repeat the washing step three times. 5. Adjust the pH of the final suspension (~12 mg/mL) to 6.5 with 1 M NaOH (4).


12

Affinity Electrophoresis as a Method for Determining Substrate-Binding…

2.3. Non-denaturating Polyacrylamide Gel

121

1. Native gel upper buffer (7×): 5.7 g Tris base, adjust pH to 6.7 with H3PO4, bring to 100 mL with water. Store at 4 °C. 2. Native gel lower buffer (4×): 18.2 g Tris base, adjust pH to 8.9 with HCl, bring to 100 mL with water. Store at 4 °C. 3. Running buffer (10×): 7.5 g Tris base, 36 g glycine, bring to 100 mL with water. 4. Sample buffer (3×): Mix 3 mL glycerol, 0.6 mL running buffer, 6.4 mL water, and 0.05 % bromophenol blue. 5. 30 % Acrylamide/Bis solution (29:1). Store at 4 °C. 6. N,N,N,N¢-tetramethyl-ethylenediamine (TEMED). Store at 4 °C. 7. Ammonium persulfate (APS) (×10): Dissolve 1 g in 10 mL water, filter, and store at −20 °C. 8. Staining buffer: Dissolve 2 g Coomassie Brilliant Blue R-250 (0.2 % (w/v)) in 500 mL methanol (50 % (v/v)). Filter through 3 mm Whatmann filter paper, and add 100 mL acetic acid (10 % (v/v)) and 400 mL water. 9. Destaining buffer: Mix 400 mL methanol (20 % (v/v)), 140 mL acetic acid (7 % (v/v)), and 1,460 mL water.

3. Methods Carry out all procedures at room temperature unless otherwise specified. 3.1. Non-denaturating Gel Electrophoresis

1. Prepare the separating gel as a function of the molecular weight of the proteins in the assay (the higher the molecular weight of the protein the less concentrated the gel should be). As indicated in Table 1, a 12 % gel can be prepared as follows, by mixing 3.5 mL of water to 4 mL of acrylamide and 2.5 mL of lower gel buffer. At least two gels have to be prepared, one without polysaccharides (the reference gel) and one containing 0.1 % (w/v) soluble polysaccharide (see Note 3 and 4). Add 100 mL of 10 % APS and 5 mL TEMED to each preparation (see Note 5). 2. Cast gel within a 7.25 cm × 10 cm × 1.5 mm gel cassette. Gently overlay with water (see Note 6) and allow polymerization time of about 20 min. 3. Prepare the stacking gel, for two 4.3 % gels: 5 mL of water, 1 mL of upper gel buffer, 1 mL of acrylamide, 50 mL of 10 % APS and 8 mL TEMED. 4. Insert a 10-well gel comb immediately, taking care not to introduce air bubbles. Allow polymerization time of about 10 min (see Notes 7 and 8).


122

S. Moraïs et al.

Table 1 Preparation of separating and stacking gels with various concentrations of acrylamide Lower/separating gela 15 %

12 %

9%

7.5 %

6%

Water

5 mL

7 mL

9 mL

10 mL

11 mL

Lower buffer

5 mL

5 mL

5 mL

5 mL

5 mL

Acrylamideb

10 mL

8 mL

6 mL

5 mL

4 mL

TEMED

10 mL

10 mL

10 mL

10 mL

10 mL

10 % APS

200 mL

200 mL

200 mL

200 mL

200 mL

4.3 %

3.5 %

Water

5 mL

5 mL

Upper buffer

1 mL

1 mL

Acrylamideb

1 mL

0.8 mL

TEMED

8 mL

8 mL

10 % APS

50 mL

50 mL

Upper/stacking gela

Quantities for two gels

a

b

Acrylamide = 30 % Acrylamide/Bis

5. Load protein standard (4 mL) (see Note 9) and then load 15–20 mL of each protein containing the sample buffer (see Note 10). 6. Apply a voltage of 100 V maximum until the dye front (from the bromophenol blue dye in the samples) has reached the bottom of the gel (see Note 11). 7. Immerse the gels in staining solution and place on slowly rotating platform for 30 min. 8. Transfer the stained gel into destaining buffer (see Note 12), place on slowly rotating platform for 2 h to overnight (see Note 13). 3.2. Comments

Affinity electrophoresis was used in this chapter to test the binding properties of two xylanases, Xyn10A and Xyn11A, and their chimeras from the thermophilic soil bacterium Thermobifida fusca towards cellulose and xylan. These two modular enzymes contain a CBM at their C-terminal with different substrate-binding specificities, either cellulose for Xyn10A or both cellulose and xylan for Xyn11A (5, 6). The enzymes of T. fusca (7, 8) have been developed as a model to explore the conversion from a free enzyme system to the cellulosomal mode (9–13), in which the enzymes are in close


12

Affinity Electrophoresis as a Method for Determining Substrate-Bindingâ&#x20AC;Ś

123

Fig. 1. Schematic representation of the recombinant proteins used in this chapter. The source of the representative module (see key) is indicated as follows: light gray (Thermobifida fusca), dark gray (Ruminoccoccus flavefaciens), and black (Acetivibrio cellulolyticus). In the shorthand notation for the engineered enzymes, the numbers 10 and 11 refer to the corresponding GH family (GH10 and GH11) of the catalytic module; lower case characters (a and f) indicate the source of the dockerin module: A. cellulolyticus and R. flavefaciens, respectively.

proximity via their integration into a complex by the cohesinâ&#x20AC;&#x201C; dockerin interaction of an appropriate scaffoldin subunit (14). In order to convert these two T. fusca xylanases into the cellulosomal mode, each was joined to a dockerin of divergent specificity for future integration into a designer cellulosome, as presented in Fig. 1 (15â&#x20AC;&#x201C;25). The Xyn11A xylanase has been the topic of a study in which the intrinsic cellulose/xylan-binding module (XBM) has been tested for its contribution to the overall enzymatic activity of the designer cellulosomes (23). For that purpose, two recombinant forms of Xyn11A were designed: one, 11A-XBM-a, in which the ScaB dockerin from Acetivibrio cellulolyticus was appended at the C terminus of the original Xyn11A, thus retaining the original catalytic module and XBM, and a second, 11A-a, in which XBM was replaced by the same A. cellulolyticus dockerin. The resultant fusion protein was identical to11A-XBM-a but lacked the XBM. The Xyn10A xylanase was converted to the cellulosomal mode by replacing its native CBM by a dockerin from Ruminoccoccus flavefaciens (26) (chimera 10A-f ). Affinity electrophoresis was used to determine whether or not the substrate-binding affinities of the xylanases were conserved in the chimeras (Fig. 2). Relative mobilities for the proteins were calculated (normalized) as the migration distance of the protein in mm divided by the migration of the initial relative mobility in the absence of polysaccharide in mm (see Note 14). Thus, the relative mobility of the reference protein (not affected by the presence of any polysaccharides) should be 1.0 (Table 2).


124

S. MoraĂŻs et al.

Fig. 2. Affinity electrophoresis of family 10A and 11A xylanases. Affinity electrophoresis was performed in the absence of ligand (a) or in the presence of 0.1 % (w/v) oat spelt xylan (b), or 0.1 % (w/v) amorphous cellulose (c). Samples included: lane 1, BSA; lane 2, wild-type Xyn10A; lane 3, 10A-f, lane 4, wild-type Xyn11A, lane 5, 11A-a, lane 6, 11A-XBM-a. See Fig. 1 for modular content of the xylanases.

Table 2 Thermobifida fusca xylanases binding to insoluble polysaccharides. Values represent normalized mobilities (see Text) BSA

Xyn10A

10A-f

Xyn11A

11A-a

11A-XBM-a

Xylan

1

0.92

0.93

0.01

0.78

0.02

Amorphous cellulose

0.98

0.2

0.87

0.65

0.78

0.75

Cellulose-binding affinity was demonstrated for Xyn10A and was essentially abolished in 10A-f, which has a dockerin module in place of the CBM, thus demonstrating that the C-terminal CBM of the protein is responsible for this binding property. Interestingly, neither the wild-type enzyme nor the chimera appeared to bind to xylan. Xyn11A and its chimera 11A-XBM-a exhibited a strong ability to bind both cellulose and xylan, whereas 11A-a, which lacks the XBM, demonstrated weakened abilities for both of these substrates. This was also demonstrated with polysaccharide-binding assay method in a previous publication (23). As expected, the BSA negative control was not able to bind either xylan or cellulose (see Note 15).

4. Notes 1. Over-expression of recombinant carbohydrate-active enzymes can be performed in Escherichia coli (27). Purification can be achieved either by using His-tag fusion protein (28) or by using the inherent affinity of the CBM, for example, if the CBM binds to cellulose (4).


12

Affinity Electrophoresis as a Method for Determining Substrate-Binding…

125

2. Relative mobilities can either be calculated using the migration distance of the non-binding reference protein (BSA) or the bromophenol blue dye front. 3. Dissolving some polysaccharides (such as arabinan) may require heating-stirring of the separating gel preparation. 4. It must also be noted that neutral polysaccharides should be used. Charged polysaccharides may migrate in the gels resulting in ambiguous results. 5. Polymerization of the separating and stacking gel begins as soon as the TEMED has been added. 6. Water overlay prevents contact with atmospheric oxygen (which inhibits acrylamide polymerization) helps leveling the separating gel solution and avoids dehydration of the gel. 7. It is recommended to use large wells to get nicer bands and allow better interpretation of the results. 8. Gel can be enveloped in wet tissue and conserved for a week at 4 °C in a plastic bag containing water. 9. In non-denaturing gel electrophoresis, proteins do not separate according to their molecular weight. The mobility depends on both the protein’s charge and its hydrodynamic size. Protein standards will serve as an indicator of the position of each band and not their molecular weight. 10. Each protein has to be loaded in the gel at a minimum amount of 5 mg for proper visibility of the band after gel staining. 11. A maximum voltage of 100 V avoids smearing of the bands. 12. Staining buffer can be saved for future use. 13. For quicker destaining, a sponge can be added to the solution and destaining buffer can be replaced 1 to 2 times. 14. Alternatively, the mobility can be calculated as the mobility of the protein in mm divided by the mobility distance of the reference in millimeter. 15. Interactions with soluble polysaccharides can be quantified: after the identification of polysaccharide ligands, affinity constants can be determined by varying the ligand concentration in the affinity gels (29). References 1. Gilbert HJ (2010) The biochemistry and structural biology of plant cell wall deconstruction. Plant Physiol 153:444–455 2. Shoseyov O, Shani Z, Levy I (2006) Carbohydrate binding modules: biochemical properties and novel applications. Microbiol Mol Biol Rev 70:283–295

3. Boraston AB, Bolam DN, Gilbert HJ, Davies GJ (2004) Carbohydrate-binding modules: finetuning polysaccharide recognition. Biochem J 382:769–781 4. Lamed R, Kenig R, Setter E, Bayer EA (1985) Major characteristics of the cellulolytic system of Clostridium thermocellum coincide with


126

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

S. Moraïs et al. those of the purified cellulosome. Enzyme Microb Technol 7:37–41 Irwin D, Jung ED, Wilson DB (1994) Characterization and sequence of a Thermomonospora fusca xylanase. Appl Environ Microbiol 60:763–770 Kim JH, Irwin D, Wilson DB (2004) Purification and characterization of Thermobifida fusca xylanase 10B. Can J Microbiol 50:835–843 Lykidis A, Mavromatis K, Ivanova N, Anderson I, Land M, DiBartolo G, Martinez M, Lapidus A, Lucas S, Copeland A, Richardson P, Wilson DB, Kyrpides N (2007) Genome sequence and analysis of the soil cellulolytic actinomycete Thermobifida fusca YX. J Bacteriol 189: 2477–2486 Wilson DB (2004) Studies of Thermobifida fusca plant cell wall degrading enzymes. Chem Rec 4:72–82 Lamed R, Setter E, Bayer EA (1983) Characterization of a cellulose-binding, cellulase-containing complex in Clostridium thermocellum. J Bacteriol 156:828–836 Lamed R, Setter E, Kenig R, Bayer EA (1983) The cellulosome—A discrete cell surface organelle of Clostridium thermocellum which exhibits separate antigenic, cellulose-binding and various cellulolytic activities. Biotechnol Bioeng Symp 13:163–181 Bayer EA, Chanzy H, Lamed R, Shoham Y (1998) Cellulose, cellulases and cellulosomes. Curr Opin Struct Biol 8:548–557 Shoham Y, Lamed R, Bayer EA (1999) The cellulosome concept as an efficient microbial strategy for the degradation of insoluble polysaccharides. Trends Microbiol 7:275–281 Fontes CM, Gilbert HJ (2010) Cellulosomes: highly efficient nanomachines designed to deconstruct plant cell wall complex carbohydrates. Annu Rev Biochem 79:655–81 Yaron S, Morag E, Bayer EA, Lamed R, Shoham Y (1995) Expression, purification and subunit-binding properties of cohesins 2 and 3 of the Clostridium thermocellum cellulosome. FEBS Lett 360:121–124 Bayer EA, Morag E, Lamed R (1994) The cellulosome—A treasure-trove for biotechnology. Trends Biotechnol 12:378–386 Fierobe H-P, Mechaly A, Tardif C, Belaich A, Lamed R, Shoham Y, Belaich J-P, Bayer EA (2001) Design and production of active cellulosome chimeras: selective incorporation of dockerin-containing enzymes into defined functional complexes. J Biol Chem 276: 21257–21261

17. Fierobe H-P, Bayer EA, Tardif C, Czjzek M, Mechaly A, Belaich A, Lamed R, Shoham Y, Belaich J-P (2002) Degradation of cellulose substrates by cellulosome chimeras: substrate targeting versus proximity of enzyme components. J Biol Chem 277:49621–49630 18. Fierobe H-P, Mingardon F, Mechaly A, Belaich A, Rincon MT, Lamed R, Tardif C, Belaich J-P, Bayer EA (2005) Action of designer cellulosomes on homogeneous versus complex substrates: controlled incorporation of three distinct enzymes into a defined tri-functional scaffoldin. J Biol Chem 280:16325–16334 19. Caspi J, Irwin D, Lamed R, Shoham Y, Fierobe H-P, Wilson DB, Bayer EA (2006) Thermobifida fusca family-6 cellulases as potential designer cellulosome components. Biocatal Biotransformation 24:3–12 20. Caspi J, Irwin D, Lamed R, Fierobe H-P, Wilson DB, Bayer EA (2008) Conversion of noncellulosomal Thermobifida fusca free exoglucanases into cellulosomal components: comparative impact on cellulose-degrading activity. J Biotechnol 135:351–357 21. Caspi J, Barak Y, Haimovitz R, Irwin D, Lamed R, Wilson DB, Bayer EA (2009) Effect of linker length and dockerin position on conversion of a Thermobifida fusca endoglucanase to the cellulosomal mode. Appl Environ Microbiol 75: 7335–7342 22. Caspi J, Barak Y, Haimovitz R, Gilary H, Irwin D, Lamed R, Wilson DB, Bayer EA (2010) Thermobifida fusca exoglucanase Cel6B is incompatible with the cellulosomal mode in contrast to endoglucanase Cel6A. Syst Synth Biol 4:193–201 23. Moraïs S, Barak Y, Caspi J, Hadar Y, Lamed R, Shoham Y, Wilson DB, Bayer EA (2010) Contribution of a xylan-binding module to the degradation of a complex cellulosic substrate by designer cellulosomes. Appl Environ Microbiol 76:3787–3796 24. Moraïs S, Barak Y, Caspi J, Hadar Y, Lamed R, Shoham Y, Wilson DB, Bayer EA (2010) Cellulase-xylanase synergy in designer cellulosomes for enhanced degradation of a complex cellulosic substrate. MBio 1:e00285–00210 25. Vazana Y, Moraïs S, Barak Y, Lamed R, Bayer EA (2010) Interplay between Clostridium thermocellum family-48 and family-9 cellulases in the cellulosomal versus non-cellulosomal states. Appl Environ Microbiol 76:3236–3243 26. Ding S-Y, Rincon MT, Lamed R, Martin JC, McCrae SI, Aurilia V, Shoham Y, Bayer EA, Flint HJ (2001) Cellulosomal scaffoldin-like proteins from Ruminococcus flavefaciens. J Bacteriol 183:1945–1953


12

Affinity Electrophoresis as a Method for Determining Substrate-Binding…

27. Baneyx F (1999) Recombinant protein expression in Escherichia coli. Curr Opin Biotechnol 10:411–421 28. Hengen P (1995) Purification of His-Tag fusion proteins from Escherichia coli. Trends Biochem Sci 20:285–286

127

29. Tomme P, Boraston A, Kormos JM, Warren RA, Kilburn DG (2000) Affinity electrophoresis for the identification and characterization of soluble sugar binding by carbohydrate-binding modules. Enzyme Microb Technol 27: 453–458


Chapter 13 Single-Molecule Tracking of Carbohydrate-Binding Modules on Cellulose Using Fluorescence Microscopy Yu-San Liu, Shi-You Ding, and Michael E. Himmel Abstract Single-molecule fluorescence detection is an invaluable technique for the study of molecular behavior in biological systems, both in vitro and in vivo. In this chapter, we focus on detailed protocols that utilize Total Internal Reflection Fluorescence Microscopy (TIRF-M) to visualize single molecules of carbohydratebinding module (CBM) labeled with green fluorescent protein (GFP). The content describes step-by-step sample preparation and data acquisition, processing, and analysis. These methods can also be further used to study interactions between domains of cellulase molecules and between cellulases and cellulose. Key words: Single-molecule tracking, Total internal reflection fluorescence microscopy, Fluorophore, Carbohydrate-binding module, Cellulose

1. Introduction Proteins are dynamic structures. The interactions between proteins, between proteins and ligands, and between proteins and substrate (i.e., enzymes) are essential to their biological function. However, these protein dynamic behaviors are usually hidden in ensemble measurement, which averages out changes occurring in large numbers of protein molecules. Single-molecule measurements, in contrast, record changes in the locations and physical states of individual molecules, which can lead to in-depth quantitative understanding of individual steps in complex biological phenomena. Single-molecule detection techniques include fluorescence-based techniques such as centroid tracking, fluorescence resonance energy transfer (FRET), and fluorescence correlation spectroscopy (FCS), as well as force-based techniques such as atomic force microscopy, optical traps, and magnetic

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_13, Š Springer Science+Business Media, LLC 2012

129


130

Y.-S. Liu et al.

tweezers (1). Among fluorescence-based methods, centroid tracking can study molecule translocation and rotational motion, FRET allows the study of molecular conformation changes, and FCS is useful for studying freely diffusing molecules (2). Single-molecule tracking by fluorescence is a direct and straightforward method, and has been widely used to study proteins under in vitro, in vivo, and in situ conditions (3â&#x20AC;&#x201C;5). In biomass conversion, one of the key steps is saccharification of cellulose, that is, the extraction of cellulose from its plant cell wall source by converting (depolymerizing) it to produce simple soluble sugars that can be fermented to fuels. This process usually involves many types of enzymes including not only cellulases but also hemicellulases and lignin-modifying enzymes. Evidence obtained by traditional biochemistry and analytical chemistry methods has suggested that synergistic interactions between these enzymes are critical for efficient biomass conversion. The cellulosedegrading enzymes usually contain two globular domains, a carbohydrate-binding module (CBM) and a catalytic domain (CD), connected by a flexible polypeptide-chain linker. The CBM functions as a binding unit to allow the enzyme to associate with the cellulose surface; the CD includes either an active-site tunnel or an active-site cleft in which the cellulose chain is precisely positioned, permitting hydrolysis of the glycosidic bond and release of soluble sugars such as cellobiose. To understand the molecular basis of the mechanisms of these enzymes, we have developed a practical fluorescence-based protocol for measurement of single-molecule binding to cellulose (6, 7). It includes sample preparation, microscope setup, and data acquisition and analysis.

2. Sample Preparation A practical requirement for fluorescence-based single-molecule techniques is the availability of a fluorophore or fluorophores with suitable optical/fluorescence properties, and in such a form that it can be attached to the protein of interest by chemical methods that are compatible with retention of the function of the protein. Ideal fluorophores for single-molecule detection should be (a) photostable, to ensure sufficient observation time before photobleaching occurs, (b) bright enough (high quantum yield and large extinction coefficient) to deliver sufficient photons to the detector, and (c) have absorption and emission spectra in the visible wavelength range to accommodate commercial excitation sources and filter cubes. Common fluorophores are organic dyes, inorganic nanocrystals, and fluorescent proteins (FPs). The structures are shown in Fig. 1. Each of them has pros and cons, so choice of fluorophores is mainly based on each specific application.


13

Single-Molecule Tracking of Carbohydrate-Binding Modules…

131

Fig. 1. Structures of fluorophores commonly used for single-molecule fluorescence detection. (a) Alexa Fluor 488 carboxylic acid, tetrafluorophenyl (TFP) ester, bis (triethylammonium salt) (MW ~ 885). (b) Ribbon representation of GFP structure. Eleven beta-strands make up the beta-barrel and an alpha-helix runs through the center. The chromophore is located in the middle of the beta-barrel. PDB: 1EMB. (c) Quantum dots with CdSe/ZnS core/shell structure capped with carboxyl group molecules.

2.1. Organic Dyes

One of the popular types of dyes for labeling proteins is the Alexa FluorTM (Molecular Probes, Inc., Eugene, OR) (AF) family. The absorption spectra of the AF family cover the entire visible spectrum (AF488, AF514, AF532, AF555, AF546, AF568, AF594, and AF610) and match the principal output wavelengths of common excitation sources. AF labeling compounds have an ester moiety that reacts efficiently with primary amines of proteins to form stable dye–protein conjugates. Though commercially available AF reagents serve as convenient labeling agents for enzymes/proteins, the labeling is not specific. For macromolecules, such as an enzyme/ protein having numerous amine groups on the surface, random labeling with AF on the protein surface increases the difficulty of identifying the precise position and number of AF on the protein, and thus increases measuring uncertainty in single-molecule detection.

2.2. Fluorescent Protein

Similar to the excitation and emission spectra of the fluorescent dyes, the excitation and emission spectra of fluorescent proteins (FPs) cover the region of 350–750 nm (BFP, CFP, GFP, YFP, RFP) and are suitable for use with common lasers and detectors. FPs are both photostable and bright. FPs can be used to label proteins by constructing a recombinant protein through genetic engineering. Therefore, the numbers of FPs on the host protein are well controlled and their positions on the protein of interest can be well defined.

2.3. Inorganic Nanocrystals

The photoluminescence properties of colloidal luminescent semiconductor nanocrystals, or quantum dots (QDs), have narrow, size-tunable emission spectra, broad absorption spectra (from UV extended to visible), limited photobleaching, and high brightness


132

Y.-S. Liu et al.

(quantum yield) (8–10). The capping molecules of QDs can be modified to make QDs water-soluble and usually to have available carboxyl groups (–COOH) allowing them to be linked to amino groups (–NH2) on the protein surface by use of the chemical coupling reagent 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC). The functional groups of the capping molecules on the QD surface can be varied to fit different applications. Unlike single fluorescent dyes and fluorescent proteins having simple on-off emission patterns, single QDs can exhibit multiple levels of emission. Though QDs have excellent photo properties enabling observation for longer time than fluorescent dyes and fluorescent proteins, their complicated fluorescence intermittency (blinking) behaviors reduce the usefulness of QDs as indicators for identifying single molecules (11). Blinking of QDs can be suppressed by antiblinking agents, such as dithiothreitol and b-mercaptoethanol (12).

3. Microscope Setup One of the common techniques used for single-molecule tracking is total internal reflection fluorescence microscopy (TIRF-M) (13– 15), which limits fluorescence imaging to an extremely thin area at the surface of a transparent substrate and therefore results in an enhanced signal-to noise ratio (S/N) and higher imaging contrast. The technique applies a laser beam at an angle through a cover glass to excite chromophores in the sample. The laser incident angle can be varied and the beam is refracted according to Snell’s law (n1 sin q1 = n2 sin q2), where the index of refraction n1 is typically 1.50 for glass and n2 is 1.33 for water. When the laser light enters the interface (cover glass–sample boundary) at incident angles greater than the critical angle (qc = sin−1(n2/n1)), the result is “total internal reflection” (Fig. 2), meaning that all of the “real” light delivered by the beam is reflected, and none crosses the interface into the sample. On the other side of the interface, however,

Fig. 2. Two things happen as laser incident angle q1 increases. First, the refraction angle q2 approaches 90 °. Second, the fraction of the light energy transmitted decreases while the fraction reflected increases. The refracted light vanishes at the critical angle and the reflection becomes 100 % for any angle q1 ³ qc.


13

Single-Molecule Tracking of Carbohydrate-Binding Modules…

133

Fig. 3. Schematic diagram of objective-type TIRF-M. The laser beam enters the interface (cover glass–sample boundary) at incident angles greater than the critical angle resulting in total internal reflection and forming an evanescent wave. TIRF-M utilizes this evanescent wave to excite single molecules in the thin section in contact with the cover glass.

there is set up a much weaker “evanescent wave” that decays exponentially and very rapidly as distance from the interface increases, illuminating only a thin region of the sample, typically 100 nm in thickness (Fig. 3).

4. Data Acquisition and Analysis Fluorescence signals are recorded in real time by an electronmultiplying charged coupled device (EMCCD) camera. With the presence of a solid state electron multiplying (EM) register at the end of the normal serial register, EMCCD allows weak signals to be multiplied before any readout noise is added by the output amplifier, thereby rendering the read noise negligible. The EM register has several hundred stages that use higher-than-typical CCD clock voltages (up to 50 V). As charge is transferred through each stage, impact ionization produces secondary electrons, resulting in EM gain. Over several hundred stages, the resultant gain can be software-controlled from unity to hundreds or even thousands of times. EMCCD can nowadays combine >90 % quantum efficiency with high speed (30 ms per 512 × 512 pixel frame) to obtain diffraction-limited images of thousands of single molecules simultaneously, and thus boost the popularity of TIRF-M (16–18). The recorded fluorescence signals (individual point spread functions (PSF)) in an image sequence are fitted to Gaussian distributions to determine the peak of the fitting Gaussian function. Successive extractions of the peak of the fitting Gaussian function yield singlemolecule trajectories.


134

Y.-S. Liu et al.

In the example protocol described here, carbohydrate-binding modules (CBMs), the noncatalytic protein domains found in the substrate-recognition portions of cellulases (19), are combined with green fluorescent protein and bind the fluorescent label to the surface of Valonia crystalline cellulose. The motions of single, fluorescently labeled CBMs on the crystalline cellulose is then are then tracked by means of TIRF-M. Individual steps discussed include choice of protein-labeling fluorophore, optimal concentrations for labeling of proteins with fluorophores, assembly of CBM–cellulose complexes, image capture setting and skills, and trajectory analysis.

5. Materials 5.1. Microscopy and Accessories

1. Olympus IX 71 inverted microscope. 2. QuantEM: 512SC camera (Roper Scientific, Trenton, NJ, USA) with resolution 512 × 512, and pixel size 16 mm × 16 mm. 3. Linear-polarized 488 nm Albuquerque, NM, USA).

argon

laser

(Melles

Griot,

4. Notch filter (488-S3D, Omega, Stamford, CT, USA) (see Note 1). 5. Filter cube (SPEC Z488RDC) with excitation filter Ex = 488/10×, emission filter 525/50 M nm, and dichroic filter Z488RDC. 6. 150× objective lens (UApo, oil immersion and N.A. = 1.45) (see Note 2). 7. No. 1 coverslips with 0.13–0.17 mm thickness, (12-542A, 18 x 18 mm, Fisherbrand; 12-544-12, Fisherfinest, 40 × 24 mm). 8. Neutral density filters 0.3, 0.6, 0.9 (see Note 3). 9. Vacuum pen. 5.2. Reagents, Equipment and Software

1. CBM-GFP fusion protein (see Note 4) in Tris buffer: 50 mM Tris, pH 8.0 (see Note 5). 2. Centrifuge (5417R, Eppendorf). 3. NanoDrop 1000 Spectrophotometer (Thermo Scientific, Wilmington, DE, USA). 4. Image-Pro 6.2 software (MediaCybernetics). 5. DIATRACK software (Semasopht, North Epping, Australia).


13

Single-Molecule Tracking of Carbohydrate-Binding Modules…

135

6. Methods 6.1. Assembly of Complexes of Labeled CBMs and Cellulose

1. 1 mg of CBM-GFP protein (see Note 6) in solution is incubated with 25 mg of Valonia cellulose crystals in 200 mL of 50 mM Tris, pH 8.0 with gentle mixing for 10 min. 2. The above mixture is centrifuged (10,000 × g for 5 min at 4 °C (see Note 7)) to obtain the protein-bound cellulose pellets. 3. The protein-bound cellulose pellets are washed three times by re-suspension in Tris buffer followed by centrifugation (10,000 × g for 5 min at 4 °C) to remove unbound CBM-GFP present in the solution. 4. The resulting CBM-GFP bound cellulose complexes are suspended in 50 mL of Tris buffer for use in single-molecule experiments.

6.2. Microscopic Observation

1. Freshly prepared samples (2-mL volume) are placed between two glass coverslips. The vacuum pen is used to delicately place one coverslip onto the other to ensure a bubble free sample environment. 2. After the sample spreads out by capillary effect, the coverslips are pressed together hard to minimize optical depth of the sample and to obtain a thin and even distribution of cellulose substrate fibers, maintained in an aqueous environment. 3. A 200-frame image sequence is captured (see Note 8), with exposure time 500 ms for each frame and on-chip multiplication set at gain factor 200 (see Note 9) (Fig. 4).

Fig. 4. A typical image set for single-molecule tracking (see Note 13). (a) White light image of cellulose crystals. (b) TIRF image of GFP-labeled CBMs bound to cellulose. GFPs appear as white spots. Scale bar is 2 mm.


136

Y.-S. Liu et al.

6.3. Imaging Processing

1. The image sequence is exported from Image-Pro 6.2 software as a TIF file. 2. The image sequence is imported to DIATRACK 3.03 software. 3. The intensity threshold is set to exclude spots on an image that are too bright (i.e., which obviously contain multiple fluorophores). All other spots should represent single molecules (see Note 10) and are selected to be analyzed (see Note 11) through the image sequence. 4. Trajectories are plotted as single-molecule movement; with errors estimated for centroid locations (see Note 12).

6.4. Notes

1. A 488 nm notch filter is placed in front of the camera to cut off the excitation background. To further suppress noise, an additional emission filter 525/50 M nm can be used to get a high-purity signal. 2. A high N.A. (numerical aperture) objective lens (N.A. = 1.45) makes it easier to achieve TIRF conditions due to the increase in alignment working area for the laser angle adjustment, and also decreases the width of the diffraction limited spot s = 0.61Îť / NA . The spatial resolution of an optical system is limited by Rayleigh criterion (20), where l is the wavelength of the collected photons. In our case, the emission peak of GFP is at 510 nm which will in turn give a diffraction limit of 215 nm. 3. A neural-density filter can be used to adjust the excitation power, which is usually 3 mW after the objective. 4. The structures and ligand specificities of many CBMs have been determined experimentally (21), and several hundred other CBMs have been putatively identified and grouped into 62 families according to their amino acid sequence similarity (http://www.cazy.org/fam/acc_CBM.html). CBMs have also been classified into three types based on their function: Type A CBMs, referred to as surface-binding CBMs, bind specifically to insoluble crystalline cellulose. Type B and C bind to soluble polysaccharides, and are described as either chain-binding or end-binding CBMs. We have used genetic engineering methods to produce labeled CBMs. For example TrCBM1-GFP is a fungal family-1 CBM cloned from the cellobiohydrolase (CBH) I gene of Trichoderma reesei and fused with GFP; AcCBM2-GFP is a bacterial family-2 CBM cloned from a free cellulase enzyme of Acidothermus cellulolyticus and fused with GFP; and CtCBM3-GFP is a bacterial family-3 CBM cloned from cellulosome scaffoldin of Clostridium thermocellum and fused with GFP. These three CBMs represent the three major natural cellulase paradigms (22). In our group, we have successfully


13

Single-Molecule Tracking of Carbohydrate-Binding Modules…

137

labeled CBM with fluorescent proteins (FPs) (7, 23, 24) and quantum dots (QDs) (25, 26) through genetic engineering. 5. CBMs-GFP are stable to storage in Tris buffer with NaCl 300 mM; to get high image quality, however, there should be low (20 mM) or no salt in the Tris buffer in actual experimental samples. 6. The concentration of CBMs-GFP applied to the cellulose is important for single-molecule detection (6, 27–29). A serial dilution test is needed. Usually the appropriate concentration for single-molecule detection falls within the picomolar (pM, 10−12 M) range. The protein concentration can be measured by NanoDrop 1000, and subsequently diluted with (low-salt, see Note 5) Tris buffer. 7. High speed centrifugation can increase the temperature of solutions thus has potential to denature protein. Centrifugation at 4 °C in refrigerated centrifuge can ensure the protein activity. 8. Capturing images in a dark laboratory environment not only prevents photobleaching of GFP molecules photobleaching, but will also prevent light other than the desired fluorescence from contaminating the signals. Placing a black curtain around the microscope can reduce the light noise from the environment, such as light coming from the computer monitor. 9. On-chip multiplication gain multiplies the number of electrons generated in each pixel. It controls the amplification of the signal coming from the CCD chip. This amplifies the whole signal, including any associated background noise. 10. The major identifying single-molecule characteristic is single step photobleaching (29, 30) (Fig. 5). More than one step of photobleaching implies multiple fluorophores, which can be either multiple labeled protein molecules, or in the case of chemically linked fluorophores, more than one label on a single protein molecule (31). By plotting fluorescence intensity versus time, we can verify that the observed CBM-GFP molecule is single. 11. Each spot was fitted with a two-dimensional Gaussian function (32). The centroid of a spot in one image was determined as the peak of the fitting Gaussian function. The centroid of the same spot throughout the sequence of images was determined from one frame to the next and finally reconstructed as a spatial trajectory recording the movement of the spot. 12. The accuracy in the centroid determination is calculated by (32) s≈

1 N

⎛ 2 a2 ⎞ ⎜⎝ s + 12 ⎟⎠


138

Y.-S. Liu et al.

Fig. 5. A plot of fluorescence intensity vs. time shows single-step photobleaching.

where s is the width of diffraction limited spot (see Note 2), a is the pixel size (The QuantEm 512SC camera has 16 mm pixel size. If we use a 150Ă&#x2014; objective, each pixel is 107 nm), and N is the number of photons detected. With higher numbers of photons detected, the error in the centroid determination is lower. To decrease the uncertainty in determining the centroid position, one can either increase the exposure time (which will increase the number of photons counted, N, and ultimately improve the signal-to-noise ratio) or, with this EMCCD camera, increase the on-chip multiplication gain setting, which will also improves signal-to-noise ratio. However, one needs to compromise between accuracy and the temporal resolution. 13. A typical individual image set contains one white-light image and one TIRF image. TIRF image sequences are used to track movement of labeled molecules, while the corresponding white-light images are used to identify the cellulose location for each image-set. The cellulose crystals in the white-light image do not reflect their real size due to the diffraction limit, but do reliably reflect the locations and alignment of the axes of the linear crystals.

Acknowledgments This work was supported by the BioEnergy Science Center (BESC), which is a US Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science, Oak Ridge National Laboratory.


13

Single-Molecule Tracking of Carbohydrate-Binding Modules…

139

References 1. Greenleaf W J, Woodside MT, Block SM (2007) High-resolution, single-molecule measurements of biomolecular motion. Annu Rev Biophys Biomol Struct 36:171–190 2. Joo C, Balci H, Ishitsuka Y, Buranachai C, Ha T (2008) Advances in single-molecule fluorescence methods for molecular biology. Annu Rev Biochem 77:51–76 3. Goulian M, Simon SM (2000) Tracking single proteins within cells. Biophys J 79:2188–2198 4. Yildiz A, Forkey JN, McKinney SA, Ha T, Goldman YE, Selvin PR (2003) Myosin V walks hand-over-hand: Single fluorophore imaging with 1.5-nm localization. Science 300:2061–2065 5. Xie XS, Choi PJ, Li GW, Lee NK, Lia G (2008) Single-molecule approach to molecular biology in living bacterial cells. Annu Rev Biophys 37:417–444 6. Liu YS, Zeng YN, Luo YH, Xu Q, Himmel ME, Smith SJ, Ding S-Y (2009) Does the cellulosebinding module move on the cellulose surface? Cellulose 16:587–597 7. Liu YS, Luo YH, Baker JO, Zeng YN, Himmel ME, Smith SJ, Ding S-Y (2010) A single molecule study of cellulase hydrolysis of crystalline cellulose. Proc SPIE 7571:757103 8. Alivisatos AP (1996) Semiconductor clusters, nanocrystals, and quantum dots. Science 271:933–937 9. Chan WCW, Nie SM (1998) Quantum dot bioconjugates for ultrasensitive nonisotopic detection. Science 281:2016–2018 10. Michalet X, Pinaud FF, Bentolila LA, Tsay JM, Doose S, Li JJ, Sundaresan G, Wu AM, Gambhir SS, Weiss S (2005) Quantum dots for live cells, in vivo imaging, and diagnostics. Science 307:538–544 11. Kai Z, Hauyee C, Aihua F, Alivisatos AP, Haw Y (2006) Continuous distribution of emission states from single CdSe/ZnS quantum dots. Nano Lett 6:843–847 12. Hohng S, Ha T (2004) Near-complete suppression of quantum dot blinking in ambient conditions. J Am Chem Soc 126:1324–1325 13. Selvin P, Ha T (2008) Single molecule techniques a laboratory manual. Cold Springs Harbor Laboratory Press, Cold Springs Harbor, NY 14. Axelrod D (2008) Total internal reflection fluorescence microscopy. Method Cell Biol 89:169–221

15. Zander C, Enderlein J, Keller RA (2002) Single-molecule detection in solution: methods and applications. Wiley-VCH, Berlin 16. Axelrod D (2001) Total internal reflection fluorescence microscopy in cell biology. Traffic 2:764–774 17. Moerner WE, Fromm DP (2003) Methods of single-molecule fluorescence spectroscopy and microscopy. Rev Sci Instrum 74:3597–3619 18. Walter NG, Huang CY, Manzo AJ, Sobhy MA (2008) Do-it-yourself guide: how to use the modern single-molecule toolkit. Nat Methods 5:475–489 19. Levy I, Shoseyov O (2002) Cellulose-binding domains biotechnological applications. Biotechnol Adv 20:191–213 20. Born M, Wolf E (1997) Principles of optics. Cambridge Univ, Press, Cambridge 21. Boraston AB, Bolam DN, Gilbert HJ, Davies GJ (2004) Carbohydrate-binding modules: finetuning polysaccharide recognition. Biochem J 382:769–781 22. Wei H, Xu Q, Taylor LE, Baker JO, Tucker MP, Ding S-Y (2009) Natural paradigms of plant cell wall degradation. Curr Opin Biotechnol 20:330–338 23. Dagel DJ, Liu YS, Zhong LL, Luo YH, Himmel ME, Xu Q, Zeng YN, Ding S-Y, Smith SJ (2011) In situ imaging of single carbohydratebinding modules on cellulose microfibrils. J Phys Chem B 115:635–641 24. Porter SE, Donohoe BS, Beery KE, Xu Q, Ding SY, Vinzant TB, Abbas CA, Himmel ME (2007) Microscopic analysis of corn fiber using starch- and cellulose-specific molecular probes. Biotechnol Bioeng 98:123–131 25. Ding S-Y, Xu Q, Ali MK, Baker JO, Bayer EA, Barak Y, Lamed R, Sugiyama J, Rumbles G, Himmel ME (2006) Versatile derivatives of carbohydratebinding modules for imaging of complex carbohydrates approaching the molecular level of resolution. Biotechniques 41:435–443 26. Xu Q, Tucker MP, Arenkiel P, Ai X, Rumbles G, Sugiyama J, Himmel ME, Ding S-Y (2009) Labeling the planar face of crystalline cellulose using quantum dots directed by type-I carbohydrate-binding modules. Cellulose 16: 19–26 27. Steinmeyer R, Noskov A, Krasel C, Weber I, Dees C, Harms GS (2005) Improved


140

Y.-S. Liu et al.

fluorescent proteins for single-molecule research in molecular tracking and co-localization. J Fluoresc 15:707–721 28. Gopich IV (2008) Concentration effects in “Single-Molecule” spectroscopy. J Phys Chem B 112:6214–6220 29. Jung G, Wiehler J, Gohde W, Tittel J, Basche T, Steipe B, Brauchle C (1998) Confocal microscopy of single molecules of the green fluorescent protein. Bioimaging 6:54–61

30. Pierce DW, HomBooher N, Vale RD (1997) Imaging individual green fluorescent proteins. Nature 388:338–338 31. Gordon MP, Ha T, Selvin PR (2004) Singlemolecule high-resolution imaging with photobleaching. Proc Natl Acad Sci U S A 101: 6462–6465 32. Thompson RE, Larson DR, Webb WW (2002) Precise nanometer localization analysis for individual fluorescent probes. Biophys J 82: 2775–2783


Chapter 14 Bioprospecting Metagenomics for New Glycoside Hydrolases Jack Gilbert, Luen-Luen Li, Safiyh Taghavi, Sean M. McCorkle, Susannah Tringe, and Daniel van der Lelie Abstract To efficiently deconstruct recalcitrant plant biomass to fermentable sugars in industrial processes, biocatalysts of higher performance and lower cost are required. The genetic diversity found in the metagenomes of natural microbial biomass decay communities may harbor such enzymes. The aim of this chapter is to describe strategies, based on metagenomic approaches, for the discovery of glycoside hydrolases (GHases) from microbial biomass decay communities, especially those from unknown or never-been-cultivated microorganisms. Key words: Metagenomics, Glycoside hydrolase, Cellulase, Biomass decay community, 16S rRNA

1. Introduction In recent years, and in the face of fossil fuels depletion and a growing global environmental awareness, biofuels have attracted more interest as an alternative, renewable source of energy. Plant biomass has long been recognized as a potential sustainable source of mixed sugars for biofuels production via fermentation. However, in order to develop cost-effective processes for converting biomass to fuels and chemicals several technical barriers related to biomass recalcitrance, such as attainment of minimal biomass pretreatments matched to super active enzymes, still need to be overcome (1). In nature, cellulosic biomass is decomposed by a complex and efficient microbial process. Various microorganisms produce cellulolytic enzymes that function synergistically to decompose plant biomass (2–4). These environments contain microbial communities that can efficiently decompose natural plant biomass; they include the animal rumen (5–8), digestive tracks of termites (9–11) and wood boring insects (12), and decomposed biomass (13–15).

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_14, © Springer Science+Business Media, LLC 2012

141


142

J. Gilbert et al.

Many of these systems have proven to be attractive sources for exploring novel plant biomass degrading organisms and enzymes. Estimates suggest that approximately 4–6 × 1030 prokaryotes inhabit the earth (16) and constitute the world’s major reserve of genetic diversity. However, about 95–99.9 % of microorganisms have not been cultured by standard laboratory techniques (17). In order to bypass the limitation of cultivation-based methodologies, metagenomic approaches provide a powerful tool to directly study the diversity of genes within microbial communities, analyze their biochemical activities, and prospect novel biocatalysts from environmental samples (18–21). Here we describe metagenomics and data mining as a tool to discover glycoside hydrolases for various microbiomes. Advances in high-throughput sequencing technologies have provided tools with lower cost and facilitated the progression of metagenome projects. 1.1. Determine Community Composition Using Cultivation Independent Approaches

Before performing in depth metagenome sequencing it is important to determine the complexity of the community, as this will provide valuable information on the depth of sequencing required to obtain sufficient coverage of the community for its functional analysis. In addition, information on community composition can be exploited to address sample heterogenicity by comparing samples taken from different locations within the same microbiome, allowing for the selection of representative samples. With the availability of new high throughput sequencing technologies it is impossible to advise on the number of samples that should be analyzed; this should be driven by the presumed heterogenicity of the microbiome. Unlike in the recent past, where sequencing was the driving cost behind metagenomics, data storage and analysis have become the limiting factors. Cultivation-independent approaches, based on 16S rRNA gene sequencing for bacteria and archaea and 18S rRNA gene sequencing for eukaryotes including protozoa, yeast, fungi, and nematodes, are routinely used to determine the compositions of the microbial communities found in various microbiomes. In the past, many groups, including us, have successfully used phylogenetic sequencing to determine microbial community composition and how this composition was affected by environmental factors, such as the effects of elevated CO2 on the microbial communities associated with trembling aspen (22). At that time, full length 16S and 18S rRNA genes were PCR amplified and cloned, before being sequenced using Sanger chemistry. Nowadays, pyrotag sequencing of small DNA regions covering a hypervariable region of the rRNA gene has become the new standard (23). The advantage of this new approach, in addition to its incredible through put, is that it eliminates problems of chimeric sequences that were often created during the PCR amplification step. However, a disadvantage of the smaller pyrotags is that they are less discriminatory, as they contain less phylogenetic information.


14

Bioprospecting Metagenomics for New Glycoside Hydrolases

143

One of the critical factors for metagenome sequencing is the quality of the DNA. The MoBio Powersoil kit with an initial heating step has been successfully applied for processing environmental samples (including difficult types containing a high humic-acid content such as compost, sediment, and manure) (24), and is one of the kits recommended for total DNA extraction. This DNA can be subsequently used to amplify the 16S rRNA V3-V4 region, which will result in a 305 bp amplicon. Subsequently, Illumina sequencing is performed by generating a 125 bp single read through the reverse primer capturing the V4 region (23). This will produce approximately 30 million reads per lane minimum. Alternatively, pyrotag sequencing using the 454 technology can be used. To fully exploit the high-throughput sequencing capabilities of these methods, samples can be barcoded. In the case of Illumina sequencing, between 250 and 450 samples can be multiplexed per lane of Illumina, resulting in approximately 50,000 reads per sample. In the near future it is expected that a protocol for 16S/18S/ ITS rRNA analysis will become available for the Illumina HiSeq2000 platform, This platform should allow to multiplex up to 500 samples per lane, generating up to 40 million sequences per sample. To perform rRNA gene-based species identification and use it to describe the community composition for each sample, the Quantitative Insights into Microbial Ecology (QIIME) analysis pipeline available through MG-RAST version 3 is one of the most advanced tools (for more information, see http://press.mcs.anl. gov/mg-rast/about). Using standard nonparametric statistical analysis (e.g., Bray-Curtis matrices, UniFrac distances, nonmetric multidimensional scaling [NMDS], Principle Component Analysis [PCA], clustering, SimProf), the species richness, diversity, and dominance across every sample can be explored. The analyses of community composition can subsequently be used to examine overlap in species distributions, to determine a subset of the samples within a specific microbiome to provide replicate representatives of each community type (defined as a representing a similar community profile at a cutoff determined by the data). These representative subsets can subsequently be applied to more in-depth phylogenetic sequencing to determine the abundance of rare species and the potential for every species to be present in every sample in variant abundances (25), for instance by applying 100 million reads per sample and performing UniFrac analysis of phylogenetic novelty added by new sequences. Additionally, these subset samples can be further analyzed using a shotgun metagenomic sequencing approach to address community functionality. As part of metagenome exploration it is very important to collect a very detailed suite of environmental contextual data, which should be used to determine the fundamental drivers of microbial community composition. Within the context of metagenomics for the discovery of enzymes important for biomass breakdown,


144

J. Gilbert et al.

information on the type of biomass, biological or physicochemical pretreatment (e.g., grinding of biomass by wood feeding insects), redox conditions, pH and temperature are important parameters to record. To determine the fundamental drivers of microbial community composition, canonical correlation analysis and BEST correlation statistics (26) can be employed. To further interrogate metagenome community composition in function of sample composition, we also propose the use of discriminant function analysis, regression modeling, and autocorrelations studies to determine the linkages between taxa and environment and taxa and functionality through contemporary and time-lag correlations. 1.2. Metagenomic Sequencing to Determine Community Functionality

Complementary to the microbial community composition, metagenomic sequencing will provide valuable insights into how resilient the functional potential of biomass decomposing communities are, e.g., to various forms of biomass input or pretreatment. This can subsequently be complemented by predictive metabolomics. For biomass decomposing communities, the emphasis should be on the role of the microbial communities in the cycling of essential nutrients, especially recalcitrant plant cell wall polymers (as the major source for carbon). Comparison between taxonomic and functional data can subsequently provide valuable insights on the resilience of the communities at different levels defining biodiversity, including composition and functionality. To address the functionality of various biomass decomposing communities it is advised to initially perform shallow metagenomic sequencing on a subset of the samples to provide replicates representatives of each community type. This is especially useful when dealing with a large number of samples that are closely related, e.g., taken within a similar environmental context. Using Illumina sequencing, per metagenome, five million to ten million, 200 bp long reads are sequenced, resulting in approximately 1â&#x20AC;&#x201C;2 Gbp of sequences generated per sample. To place this in perspective, 401 Mbp of sequence was performed for functional analysis of the leaf-cutter ant fungal garden microbiome (27). The final annotation pipeline applied to each sample should be determined by the most appropriate technology available at the time. This field is developing rapidly, but will resemble the annotation pipeline of MG-RAST, which uses BLAT (instead of BLAST) to provide sequence homology annotation against KOG, COG, KEGG, SEED, NR, and IMG. These data should be used to link environmental conditions with specific metabolic activities inferred from metagenomic data (28), and to cluster samples based on their apparent functional potential using nonparametric statistical techniques (as previously described for the rRNA data). At this stage, it is advisable to carry out a comparison between the clustering of the communities, based on their phylogenetic composition, versus their clustering, based on the distribution of functional groups.


14

Bioprospecting Metagenomics for New Glycoside Hydrolases

145

Fig. 1. Comparison of the metagenomes from communities that are characterized by a high biomass turnover rates using the distribution and relative abundance of glycoside hydrolase families as variables for metagenomic comparison on the level of functionality related to biomass breakdown (29).

We successfully used this approach to compare the metagenomes from communities that are characterized by high biomass turnover rates. Rather than comparing various metagenomes at the species level, we used the distribution and relative abundance of glycoside hydrolase families as variables for metagenomic comparison on the level of functionality related to biomass breakdown (29). This information was obtained for the various metagenomic sequence sets by performing Blastx against the CAZy database and was subsequently used to calculate the correlation distances between the various metagenomes via PCA based on Spearmanâ&#x20AC;&#x2122;s rank correlation distance (Fig. 1). In this example, the gut communities form a distinct group from the free-living biomass decay communities, even though genes involved in hostâ&#x20AC;&#x201C;symbiont interactions were not part of the analysis. In addition, the microbial communities found in compost and decaying poplar woodchips show, at least on the level of glycoside hydrolase functionality, a close affiliation. This close correlation is also observed for the glycoside hydrolase distributions among the maize, miscanthus, and switchgrass associated rhizosphere communities, as well as the microbiomes of the top and bottom parts of the fungal garden of the leaf-cutter ant (Atta colombica). Within the gut communities, the distribution of GHases in the termite hindgut microbial community is unique from that in mammals.


146

J. Gilbert et al.

1.3. Predicted Relative Metabolic Turnover for Comparative Analyses of Environmental Metabolomes Predicted from Metagenomic Data

Predicted Relative Metabolic Turnover (PRMT), which enables comparative analyses of environmental metabolomes predicted from metagenomic data, can be used as a way to better predict community functionality. PRMT was successfully validated (personal communication, Larsen P, 2011) to a metagenomic data set from the time-series study of the bacterial environmental metabolome in the Western English Channel (30). PRMT generates information on the relative potential for consumption or synthesis of a metabolite between two metagenomic data sets. For microbial communities actively degrading plant biomass, PRMT can be used to predict the microbial communities’ potential to synthesize metabolites that directly influence the synthesis of secondary metabolites that have the potential to be used as biofuels or commodity chemicals. Essentially, although the standard metagenomic annotation will help to understand the putative change in the relative abundance of certain functions, PRMT will identify the metabolites that will be synthesized and consumed and a vector will be used to describe the rate of change. Therefore the application of PRMT enables to explore the ecosystem services (such as secondary metabolite synthesis) provided by the microbial community and how various disturbances or changes will impact these service provisions. By using a Bayesian inference network, with the environmental metadata as parent nodes of a linear transformation matrix, it will be possible to develop a predictive model describing the metabolite-derived ecosystem services provided by the community, and how these will change.

1.4. Mining for Enzymes of Interest: Glycoside Hydrolases

Sequence-based screening methods rely on known conserved sequences, and cannot uncover nonhomologous enzymes. Therefore, the drawback of this “closed approach” is its failure to detect fundamentally different “new” genes. However, unlike function-based methods, it can disclose target genes, regardless of gene expression and protein folding in the host, and irrespective of the completeness of the target gene’s sequence. The success of this approach rests on meeting several conditions: (A) The more complex the community is, the larger must be the sequencing effort in order to get representative coverage and subsequent sequencing assembly. Here, the development of new sequencing technology, such as the latest-generation Illumina and 454-pyrosequencing, have changed the outcome. (B) While the metagenomic approach captures representative DNA samples from diverse organisms, many sequence reads remain unassembled due to the variety of sizes of the environmental genomes, and their abundance. Therefore, a shift in focus emerged, from complete metagenome sequencing to bulk sequencing of as many possible genes/functions. In this latter approach, where there is less need to assemble sequences into contigs, the limiting factor becomes the lengths of the fragments that can be obtained for high-throughput screening


14

Bioprospecting Metagenomics for New Glycoside Hydrolases

147

and cloning. As previously discussed by Allgaier (15) and Li (21), full-length genes are desirable for enzyme characterization, but difficult to obtain from highly fragmented metagenome sequence data. Gene-finding tools, such as MetaGene, were demonstrated to predict 90 % of shotgun sequences (31). (C) New bioinformatics tools are needed for data mining, based not only on primary sequence homology but also able to predict protein structures, putative catalytic sites, and activities. With the betterment of protein classification tools, models might be designed to correlate enzyme mechanisms and protein folding. Based on this folding and the creation putative active sites, gene function can be predicted (32–37). We anticipate that soon sequence-based metagenome databases searches combined with bioinformatics tools will have a greater influence on mining novel biocatalyst genes than will function-based methods. To prospect for genes encoding glycoside hydrolases in microbial communities, tiled blastx searches can be performed against the CAZy database (http://www.cazy.org/) (filtered with E-value of 1−10 or better). From the homologues, candidate genes should be selected for further investigation using the following criteria: (A) Enzyme functions of interests. GHase families that represent key enzymes for the most efficient decomposition of plant cell wall recalcitrants: cellulase (GH5, 6, 8, 9, 48); hemicellulase (GH 8, 10, 11, 12, 26, 28, 53, 74); debranch enzyme (GH51, 54, 62, 67, 78, 74). (B) The quality of sequences, including gene length and homology, and exclude genes with potential gene rearrangements, disruptions, and deletions. A scheme of the cloning strategy is showed in Fig. 2. 1.5. Cloning the Full-Length Open Reading Frame of Glycoside Hydrolase Gene Directly from Metagenomic DNA

To obtain flanking sequences of candidate gene fragments (in order to reconstruct the full-length open reading frame of each candidate), inverse PCR and DNA walking should be performed where necessary. We successfully applied the following protocol (38): for inverse PCR, purified metagenome DNA was partially digested with restriction endonuclease BamHI or EcoRI and subsequently diluted and treated with T4 DNA ligase. Two sets of primers that are specific to each candidate gene were used successively to amplify flanking regions from self-ligated metagenome DNA. DNA walking was successfully performed by using the DNA Walking SpeedUp kit (Seegene, Seoul, Korea) according to manufacturer’s protocol. PCR products from both inverse PCR and DNA walking were inserted into a vector using the TOPO TA Cloning Kit (Invitrogen, Carlsbad, CA), and plasmids were isolated for sequencing analysis. Coding sequences are subsequently PCR amplified from metagenome DNA using primers that are designed according to each candidate’s sequence, and subsequently cloned into the pET28a vector (Novagen, Gibbstown, NJ). Each plasmid was confirmed by DNA sequencing and introduced into E. coli host ER2566 (New England BioLabs, Ipswich, MA) for protein expression and characterization.


148

J. Gilbert et al. Metagenome sequence data (From the microbial community decaying poplar biomass)

Collection of partial glycoside hydrolase gene sequences (Genes identified based on translated sequence homology )(blastx)

Select candidates for further analysis 1. Emphasis on most interesting GHase families: Family 5, 9, 48, 51 2. Emphasis on predicted specificity: Endo-β-1, 4-glucanase, Exo-β-1, 4glucanase, Exo-β-1, 4-glucosidase, cellulase, hemicellulase, β-1, 4-xylanase 3. Check for potential problems: gene rearrangement, deletion, mutation….

Using inverse PCR to identify flanking sequences of selected GHase fragment in order to obtain the complete gene.

Evaluate the sequencing resu lt: Complete GHase? Gene rearrangement, deletion, mutation? Upstream or downstream gene (if sequences available) functionally related? No further analysis

Yes No Cloning of the potential GHase gene and examining for GHase activity

No further analysis

Yes No Expression, purification, and characterization of the cloned GHase

Fig. 2. Proposed cloning strategy and decision scheme.

1.6. Protein Expression and Purification

For batch culture of E. coli bearing plasmid, cells were incubated in LB medium with 50 μg/mL kanamycin at 37 °C until OD600nm = 0.5– 0.6. The culture was induced with isopropyl-beta-d-thiogalactopyranoside (IPTG, 0.4 mM final concentration) at 18 °C for 4 h. The cells were harvested by centrifugation, resuspended in 1/20 culture volume of LEW buffer (50 mM sodium phosphate, pH 8.0, 300 mM NaCl, 10 mM 2-mercaptoethanol, 10 % Triton X-100), and disrupted by sonication. The cell lysates were centrifuged at 15,000 × g for 15 min and both the supernatant and the pellet were examined with SDS-PAGE. PrepEase His-Tagged High Yield Purification Resin (USB, Cleveland, OH) was added into the supernatant and


14

Bioprospecting Metagenomics for New Glycoside Hydrolases

149

gently mixed at 4 °C for 1 h. After binding, the resin was pelleted and washed twice with ten resin volumes of LEW buffer, and was subsequently eluted with elution buffer (LEW buffer plus 250 mM imidazole). Eluted proteins were examined with SDS-PAGE, dialyzed, and then concentrated for further enzymatic assays. 1.7. Enzyme Activity Assays

For initial testing, E. coli strains bearing candidate genes on plasmids were cultured as described in previous section. Cells were harvested and resuspended in 50 mM sodium phosphate (pH 8.0) buffer with 100 mM NaCl. After sonication, the whole cell lysate was tested for substrate specificity. In a repeat experiment, CelLytic B lysis reagent (Sigma-Aldrich, St. Louis, MO) was also used for cell lysis and supernatant was used for enzyme activity examination. Candidate clones and proteins were tested for enzyme activities using following substrates: p-nitrophenyl β-d-cellobioside, p-nitrophenyl β-d-glucopyranoside, p-nitrophenyl β-d-lactopyranoside, p-nitrophenyl β-d-galactopyranoside, p-nitrophenyl β-d-xylopyranoside, and p-nitrophenyl α-l-arabinofuranoside (all purchased from Sigma-Aldrich, St. Louis, MO). Cell lysates or proteins were tested at 37 °C, in 50 mM sodium phosphate buffer (pH 8.0) containing 100 mM NaCl and 0.5 mM substrate. After incubation for appropriate amount of time, the reactions were stopped by adding a quarter volume of 1 M Na2CO3 solution, and the hydrolysis product p-nitrophenol was measured by absorbance at 405 nm. A pure p-nitrophenol (Sigma-Aldrich, St. Louis, MO) was used for producing a standard curve. The assay was performed with biological duplicates for each clone on every substrate. To determine the pH optimum of candidate proteins we added 1 μL cell lysate, or purified proteins (final concentration 6.5 μg/ mL), to solutions of p-nitrophenyl substrates (final concentration 0.625 mM) in the buffer range pH 4 to 8.5. The total volume for the reaction was 200 μL. Reactions were conducted for 30 min at 30 °C and quenched by the addition of 50 μL 1 M NaCO3. Absorbance was read at 405 nm to determine the extent of conversion. Data was normalized for 100 % response at the maximum conversion. To determine the temperature optimum of candidate proteins we added 1 μL cell lysate, or purified proteins (final concentration 6.5 μg/mL), to solution of p-nitrophenyl substrates (final concentration 0.625 mM) in a buffer that has the optimal pH range for the candidate. The total volume for the reaction was 200 μL. The temperature range used was from 25 °C to 55 °C in 5 °C increments. Both the enzyme stock solution and the reaction mix were preequilibrated for 5 min at each tested temperature prior to mixing. Reactions were conducted for 10 min and quenched by the addition of 50 μL 1 M NaCO3. Absorbance was read at 405 nm to determine the extent of conversion. Data was normalized for 100 % response at the maximum conversion.


150

J. Gilbert et al.

2. Conclusions Metagenome sequencing and mining can be a good resource to explore and prospect new functional enzymes for biomass deconstruction and biofuels production.

Acknowledgments The authors wish to acknowledge a subcontract from the BioEnergy Science Center, which is a US Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science. References 1. Himmel ME (ed) (2008) Biomass recalcitrance deconstructing the plant cell wall for bioenergy. Blackwell, London 2. Himmel ME (ed) et al (1997) Fuels and chemicals from biomass. American Chemical Society Symposium Series, American Chemical Society, Washington, DC 3. Bayer EA, Shimon LJ et al (1998) Cellulosomesstructure and ultrastructure. J Struct Biol 124:221–234 4. Bayer EA, Belaich J-P et al (2004) The cellulosomes: multienzyme machines for degradation of plant cell wall polysaccharides. Annu Rev Microbiol 58:521–554 5. Ferrer M, Golyshina OV et al (2005) Novel hydrolase diversity retrieved from a metagenome library of bovine rumen microflora. Environ Microbiol 7:1966–2010 6. Feng Y, Duan C et al (2007) Cloning and identification of novel cellulase genes from uncultured microorganisms in rabbit cecum and characterization of the expressed cellulases. Appl Microbiol Biotechnol 75:319–328 7. Flint HJ, Bayer EA et al (2008) Polysaccharide utilization by gut bacteria: potential for new insights from genomic analysis. Nat Rev Microbiol 6:121–131 8. Singh B, Gautam SK et al (2008) Metagenomics in animal gastrointestinal ecosystem: potential biotechnological prospects. Anaerobe 14: 138–144 9. Warnecke F, Luginbuhl P et al (2007) Metagenomic and functional analysis of hindgut microbiota of a wood-feeding higher termite. Nature 450:560–565

10. Ohkuma M (2008) Symbioses of flagellates and prokaryotes in the gut of lower termites. Trends Microbiol 16:345–352 11. Scharf ME, Tartar A (2008) Termite digestomes as sources for novel lignocellulases. Biofuels, Bioprod Bioref 2:540–552 12. Zachary A, Colwell RR (1979) Gut-associated microflora of Limnoria tripunctata in marine creosote-treated wood pilings. Nature 282: 716–717 13. Yu H, Zeng G et al (2007) Microbial community succession and lignocellulose degradation during agricultural waste composting. Biodegradation 18:793–802 14. Schluter A, Bekel T et al (2008) The metagenome of a biogas-producing microbial community of a production-scale biogas plant fermenter analysed by the 454-pyrosequencing technology. J Biotechnol 136:77–90 15. Allgaier M, Reddy A et al (2010) Targeted discovery of glycoside hydrolases from a switchgrass-adapted compost community. PLoS ONE 5:e8812 16. Whitman WB, Coleman DC et al (1998) Prokaryotes: the unseen majority. Proc Natl Acad Sci U S A 95:6578–6583 17. Amann RJ, Binder BL et al (1990) Combination of 16S rRNA targeted oligonucleotide probes with flow-cemetry for analysing mixed microbial populations. Appl Environ Microbiol 56:1910–1925 18. Tyson GW, Chapman J et al (2004) Community structure and metabolism through reconstruction of microbial genomes from the environment. Nature 428:37–43


14

Bioprospecting Metagenomics for New Glycoside Hydrolases

19. Daniel R (2005) The metagenomics of soil. Nat Rev Microbiol 3:470–478 20. Lorenz P, Eck J (2005) Metagenomics and industrial applications. Nature 3:510–516 21. Li LL, McCorkle S et al (2009) Bioprospecting metagenomes: glycosyl hydrolases for converting biomass. Biotechnol Biofuels 2:10 22. Lesaulnier C, Papamichail D et al (2008) Elevated atmospheric CO2 affects soil microbial diversity associated with trembling aspen. Environ Microbiol 10:926–941 23. Costello EK, Lauber CL et al (2009) Bacterial community variation in human body habitats across space and time. Science 326:1694–1697 24. Foster JS, Green SJ et al (2009) Molecular and morphological characterization of cyanobacterial diversity in the stromatolites of Highborne Cay, Bahamas. ISME J 3:573–587 25. Caporaso JG, Field D et al. Everything is everywhere: a 19th century hypothesis confirmed with 21st century science. ISME J (in Press) 26. Gilbert JA, Field D et al (2009) The seasonal structure of microbial communities in the Western English Channel. Environ Microbiol 11:3132–3139 27. Suen G, Scott JJ et al (2010) An insect herbivore microbiome with high plant biomassdegrading capacity. PLoS Genet 6(9) 28. Gianoulis TA, Raes J et al (2009) Quantifying environmental adaptation of metabolic pathways in metagenomics. Proc Natl Acad Sci U S A 106:1374–1379 29. van der Lelie D, Taghavi S, McCorkle SM, Li L-L, Malfatti SA, Monteleono D, Donohoe BS, Ding S-Y, Adney WS, Himmel ME, Tringe SG (2012) The metagenome of an anaerobic microbial community decomposing poplar wood chips, PLoS One 7(5) 30. Gilbert JA, Field D et al (2010) The taxonomic and functional diversity of microbes at a

31.

32.

33.

34.

35.

36.

37.

38.

151

temperate coastal site: a ‘multi-omic’ study of seasonal and diel temporal variation. PLoS One 5:e15545 Noguchi H, Park J et al (2006) MetaGene: prokaryotic gene finding from environmental genome shotgun sequences. Nucleic Acids Res 34:5623–5630 Henrissat B (1991) A classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem J 280:309–316 Claudel-Renard C, Chevalet C, Faraut T, Kahn D (2003) Enzyme-specific profiles for genome annotation: PRIAM. Nucleic Acids Res 31:6633–6639 Bateman A, Coin L, Durbin R, Finn RD, Hollich V, Griffiths-Jones S, Khanna A, Marshall M, Moxon S, Sonnhammer EL, Studholme DJ, Yeats C, Eddy SR (2004) The Pfam protein families database. Nucleic Acids Res 32(Database issue):D138–D141 Rost B, Yachdav G, Liu J (2004) The PredictProtein Server. Nucleic Acids Res 32(Web Server issue):W321–W326 Selengut JD, Haft DH, Davidsen T, Ganapathy A, Gwinn-Giglio M, Nelson WC, Richter AR, White O (2007) TIGRFAMs and Genome Properties: tools for the assignment of molecular function and biological process in prokaryotic genomes. Nucleic Acids Res 35(Database issue):D260–D264 Cantarel BL, Coutinho PM et al (2009) The Carbohydrate-Active EnZymes database (CAZy): an expert resource for Glycogenomics. Nucleic Acids Res 37(suppl 1):D233–238 Li L-L, Taghavi S, McCorkle SM, Zhang Y-B, Blewitt MG, Brunecky R, Adney WS, Himmel ME, Brumm P, Drinkwater C, Mead DA, Tringe SG, van der Lelie D (2011) Bioprospecting metagenomics of decaying wood: mining for new glycoside hydrolases. Biotech Biofuels 4(23)


Chapter 15 Anaerobic High-Throughput Cultivation Method for Isolation of Thermophiles Using Biomass-Derived Substrates Scott D. Hamilton-Brehm, Tatiana A. Vishnivetskaya, Steve L. Allman, Jonathan R. Mielenz, and James G. Elkins Abstract Flow cytometry (FCM) techniques have been developed for sorting mesophilic organisms, but the difficulty increases if the target microbes are thermophilic anaerobes. We demonstrate a reliable, high-throughput method of screening thermophilic anaerobic organisms using FCM and 96-well plates for growth on biomass-relevant substrates. The method was tested using the cellulolytic thermophiles Clostridium thermocellum (Topt = 55 °C), Caldicellulosiruptor obsidiansis (Topt = 78 °C) and the fermentative hyperthermophiles, Pyrococcus furiosus (Topt = 100 °C) and Thermotoga maritima (Topt = 80 °C). Multi-well plates were incubated at various temperatures for approximately 72–120 h and then tested for growth. Positive growth resulting from single cells sorted into individual wells containing an anaerobic medium was verified by OD600. Depending on the growth substrate, up to 80 % of the wells contained viable cultures, which could be transferred to fresh media. This method was used to isolate thermophilic microbes from Rabbit Creek, Yellowstone National Park (YNP), Wyoming. Substrates for enrichment cultures including crystalline cellulose (Avicel), xylan (from Birchwood), pretreated switchgrass and Populus were used to cultivate organisms that may be of interest to lignocellulosic biofuel production. Key words: Thermophile, Avicel, Cellulose, Anaerobic, Flow cytometry

1. Introduction Flow cytometry (FCM) enables the rapid analysis of single cells and has become an increasingly important tool for microbiologists. Mixed sets of cells or spores can be analyzed, enumerated, and sorted using physical measurements of morphology (1, 2), physiology (3), or fluorescence (4, 5). These diagnostics are important when considering organism enrichment, isolation, substrate utilization, and growth conditions of an individual cell from a population of cells or diverse environmental sample (6, 7). Connon and Giovannoni developed techniques based on FCM to recover novel marine isolates grown under oligotrophic conditions (8). Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_15, © Springer Science+Business Media, LLC 2012

153


154

S.D. Hamilton-Brehm et al.

Gel microdroplets have also been used to grow microcolonies of marine or soil microbes, which can be sorted into multiwall plates via FCM for isolation (9). The utility of isolation and recovery of microorganisms by FCM is obvious though this capability becomes problematic if the organism is a strict anaerobe. With some bacteria, strict anaerobic conditions can be avoided if the organism is capable of forming spores allowing study by FCM, but not always (1). Lignocellulosic biomass will likely be one of the main renewable energy sources in the future. However, overcoming the recalcitrance of lignocellulosic materials to hydrolysis will require significant research and development before commercial processes to liquid fuels are feasible (10, 11). Cellulolytic microorganisms have been of scientific interest for many years, with more recent focus on Clostridium sp. and Caldicellulosiruptor sp. (12â&#x20AC;&#x201C;18). There is also an increased interest in the discovery of novel cellulolytic thermophiles. Yellowstone National Park (YNP), WY, USA, encompasses the worldâ&#x20AC;&#x2122;s richest assortment of hot springs and other geothermal features. Cellulosic detritus may enter a spring and provide carbon and nutrients for a diverse consortium of uncharacterized microorganisms. Standard plating and serial dilution techniques for isolation are not always successful with anaerobic thermophiles. Successful isolation and characterization of novel microorganisms is typically the rate limiting factor in any discovery pipeline. One advantage of the FCM method described herein is its ability to isolate and recover strictly anaerobic and thermophilic organisms at moderate to high-throughput. Without the ability to rapidly screen and generate putative novel cellulolytic isolates, the process would take significantly longer, requiring more tedious techniques such as serial dilution, roll tubes, and plating. Additionally, some of these techniques are ineffective with thermophilic anaerobes. Isolation and analysis by FCM of anaerobic thermophiles should advance the speed of acquiring new isolates for study. By combining cellulosic substrate enrichments with YNP samples to screen for cellulose hydrolyzing microbes and FCM to sort single cells from the enrichments, we developed a reliable high-throughput method of generating putative isolates from thermal environments with bioenergy-relevant phenotypes such as cellulose utilization or ethanol production.

2. Materials and Equipment Unless otherwise specified, all solutions were prepared using ultrapure water (prepared by purifying deionized water). Equipment and solutions were sterilized by autoclaving for 30 min at 121 °C or filtration through a 0.2 mm filter where appropriate. All anaerobic liquids were boiled then sparged with oxygen-free gas mixtures


15

Anaerobic High-Throughput Cultivation Method…

155

Table 1 Characteristics of model organisms selected for FCM and cultivation in 96-well plates Characteristic

Clostridium thermocellum

Caldicellulosiruptor obsidiansis

Thermotoga maritima

Pyrococcus furiosus

Cell size (mm)

0.5 × 4.0

0.2 × 2.0

0.6 × 5.0

0.8–2.5

Cell shape

Rod

Rod

Rod

Cocci

Flagellation

No

No

Yes

Yes

Temperature optimum (°C)

60

78

80

100

pH optimum

6.8

7.0

6.5

7.0

Doubling time (min)

150

96

75

30

Metabolism

Fermentative, utilizes C6 polysaccharides including cellulose

Fermentative, utilizes C5 and C6 sugars including xylan and cellulose

Fermentative, So-reducer, utilizes mono-, disaccharides and peptides

Fermentative, So-reducer, utilizes mono-, disaccharides and peptides

Reference

(23)

(20)

(24)

(25)

consisting of 80:20 of N2/CO2 for at least 30 min. Source of xylan is from Birchwood. All chemicals were purchased from SigmaAldrich, Difco, or Fisher Scientific. 2.1. List of Control Organisms

2.2. Media Composition for Positive Control Organisms

The following control organisms were acquired from the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ) and American Type Culture Collection (ATCC) (see Table 1): ●

Clostridium thermocellum (ATCC 27405).

Caldicellulosiruptor obsidiansis (ATCC 2073).

Thermotoga maritima (DSMZ 3109).

Pyrococcus furiosus (DSMZ 3638). 1. P. furiosus (media adapted from ref. 19): 4.4 mM KCl, 4.6 mM NH4Cl, 14.2 mM MgSO4·7H2O, 480 mM NaCl, 1.0 mM CaCl2·2H2O, 1.0 mM Na2WO4·2H2O, 0.25 mg/mL Resazurin, 2.8 mM cysteine-HCl, 2.1 mM Na2S, 6.0 mM NaHCO3, 1 mM phosphate buffer, 10 mM 3-(N-morpholino)propanesulfonic acid (MOPS) pH 6.8, 1× ATCC trace minerals (Manassas, VA),


156

S.D. Hamilton-Brehm et al.

1× ATCC vitamin supplement (Manassas, VA), 0.05 % weight/ volume (w/v) yeast extract, and 0.5 % (w/v) maltose. 2. T. maritima (adapted from ref. 19): 4.4 mM KCl, 4.6 mM NH4Cl, 14.2 mM MgSO4·7H2O, 480 mM NaCl, 1.0 mM CaCl2·2H2O, 1.0 mM Na2WO4·2H2O, 0.25 mg/mL Resazurin, 2.8 mM cysteine-HCl, 2.1 mM Na2S, 6.0 mM NaHCO3, 1 mM phosphate buffer, 10 mM MOPS pH 6.8, 1× ATCC trace minerals (Manassas, VA), 1× ATCC vitamin supplement (Manassas, VA), 0.05 % (w/v) yeast extract, and 0.5 % (w/v) glucose. 3. C. obsidiansis (adapted from ref. 20): 4.5 mM KCl, 4.7 mM NH4Cl, 2.5 mM MgSO4·7H2O, 1.0 mM NaCl, 0.7 mM CaCl2·2H2O, 0.25 mg/ml Resazurin, 2.8 mM Cysteine-HCl, 2.1 mM Na2S, 6.0 mM NaHCO3, 1 mM phosphate buffer, 10 mM MOPS pH 6.8, 1× ATCC trace minerals (Manassas, VA), 1 ATCC vitamin supplement (Manassas, VA), 0.1 % (w/v) yeast extract, and 0.4 % (w/v) cellobiose. 4. C. thermocellum (adapted from ref. 21): 4.5 mM KCl, 4.7 mM NH4Cl, 4.0 mM MgSO4·7H2O, 1.0 mM CaCl2 × 2H2O, 6 mM Na3Citrate × 2H2O, 10 mM Urea, 0.25 mg/ml Resazurin, 5.6 mM Cysteine-HCl, 6.0 mM NaHCO3, 10 mM phosphate buffer, 23 mM MOPS pH 6.8, 1× ATCC trace minerals (Manassas, VA), 1× ATCC vitamin supplement (Manassas, VA), 0.1 % (w/v) yeast extract, and 0.5 % (w/v) cellobiose. 2.3. Composition of Flow Cytometer Sheath Fluid

2.4. LB Broth and Agar Composition

2.5. List of Kits and Equipment Needed for this Method

Phosphate buffered saline (PBS) used for sheath fluid and sample preparation came from a 10× stock solution (G Biosciences, St. Louis, MO). ●

1× PBS at pH 7.2 final concentrations: 8 mM Na2HPO4, 150 mM NaCl, 2 mM KH2PO4, and 3 mM KCl.

LB Broth: 10 g/L tryptone, 5 g/L yeast extract, and 5 g/L NaCl.

LB Agar: 10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl, and 15 g/L agar.

Sterile polystyrene nonpyrogenic 96-well plates with low evaporation lids (Becton Dickenson Labware, NJ).

The AnaeroPack System (Mitsubishi Gas Chemical CO., Inc., Japan).

MACS Anaerobic Workstation (Don Whitley Scientific Limited, England).

Cytopeia Influx flow cytometer (Cytopeia Inc., Seattle, WA).

Synergy 2 multi-well plate reader, (BioTek, VT).


15

Anaerobic High-Throughput Cultivation Method…

157

Waters Breeze 2 HPLC system (Milford, MA), RI detector (Waters 2414).

Aminex HPX-87H column (BioRad Laboratories).

QIAquick Gel Extraction kit (Quagen Inc, Valencia, CA).

PowerSoil™ DNA Isolation Kit (Mo Bio Labs, Inc., Carlsbad, CA).

GoTaq DNA polymerase (Promega, Madison, WI).

UltraPureTM Agarose (Invitrogen, Carlsbad, CA).

pCR 2.1-TOPO vectors (Invitrogen).

BigDye Terminator v3.1 Cycle Sequencing kit.

3. Methods 3.1. Media Preparation for Control Organisms

1. Media was prepared in batches of 500 mL containing vitamins, trace minerals, salts, and soluble carbon sources (see Subheading 2.2). The media was thoroughly mixed and then microwaved until boiling, after which it was sparged with N2/ CO2 for 30 min, reduced with 2.8 mM cysteine-HCl and 2.1 mM sodium sulfide. Resazurin (0.5 mg/L) was included to indicate removal of oxygen. 2. 120 mL serum bottles were flushed with N2/CO2, then filled with 50 mL of anaerobic media, plugged with a butyl rubber stopper, sealed with an aluminum cap, and then autoclaved for 30 min. 3. Carbon sources for each organism used in 50 mL culture volumes contained 0.5 % weight/volume (w/v): glucose (T. maritima), maltose (P. furiosus), and cellobiose (C. obsidiansis and C. thermocellum) and each was supplemented with 0.1 % (w/v) yeast extract. 4. Stock cultures were grown for 10–16 h or to mid log phase at each organism’s optimal growth temperature and stored in the dark at RT until inoculated.

3.2. Media for Environmental Samples

1. Carbon and energy sources for enrichment cultures were dilute acid-pretreated switchgrass and Populus, Avicel, and xylan with yeast extract. 2. Media was prepared in batches of 500 mL, with 500 mg dry weight of appropriate substrate (final concentration 0.1 % w/v). (a) To determine dry weight of dilute sulfuric acid-pretreated switchgrass and Populus (22), a wet sample was weighed out and dried at 85 °C for 16 h then weighed again in triplicate to determine the ratio of dry to wet weight.


158

S.D. Hamilton-Brehm et al.

Fig. 1. Flow diagram of environmental sampling, enrichment cultures, cell sorting, and cultivation of putative isolates. (a) Sampling pictures of Rabbit Creek. (b) Anaerobic flow cytometry system with regulated nitrogen source for charging the sheath fluid and purging a plate storage chamber.

3. Media was microwaved until boiling, sparged with N2/CO2 (80:20) for 30 min, and then reduced with cysteine-HCl and sodium sulfide. 4. While vigorously stirred to ensure homogeneous distribution of insoluble substrate, 50 mL of media was transferred by pipette to a 120 mL serum bottle flushed with N2/CO2 (see Note 1). 3.3. Environmental Sample Collection

1. A water sample (collection number RC0708-11) containing grass and visible biofilms was collected from a runoff channel of a small hot spring located ~5 m from Rabbit Creek Source Pool (N44°31¢279², W110°48¢690²) within the thermal area Rabbit Creek Hot Springs Group, which represent Midway Geyser Basin thermal region of YNP, WY (see Fig. 1a). 2. The sample, with ambient temperature of 67.5 °C and pH 7, was collected in a 250 mL sterile anaerobic Nalgene bottle in July 2008. The sample was immediately reduced with a few crystals of sodium dithionite and 1 mL of a 2.5 % (w/v) cysteine-HCl solution (pH 7.2) and tightly sealed with a butyl


15

Anaerobic High-Throughput Cultivation Method…

159

rubber stopper. The sample was kept at room temperature during transportation to Oak Ridge National Laboratory (Oak Ridge, TN) by air and then stored at 4 °C. 3.4. Flow Cytometer Parameters and Settings

1. The Cytopeia Influx flow cytometer (see Note 2) is an open stream, high-speed cell sorting flow cytometer which sorts electrically charged droplets using high voltage deflection plates. 2. It is capable of processing and sorting at up to 200,000 events per second. For our bacterial cell sorting, we used a 70 mm nozzle, with PBS sheath fluid typically pressurized at 18–25 psi above atmospheric pressure. 3. The sheath fluid and cell sample tube were pressurized using compressed high purity nitrogen for anoxic sorting. 4. Three liters of PBS sheath fluid was heated to the boiling point before being transferred into the Influx sheath reservoir to be sparged with oxygen-free N2. 5. Following sparging, the N2 pressurized sheath fluid reservoir was allowed to cool to room temperature before being used for sorting. 6. Droplet formation in the stream was stabilized by an oscillating piezo-electric drive, and under typical operating parameters, produced 48,000 droplets per second. 7. A cell dilution was sampled and continuously mixed/injected into the sheath fluid stream to achieve a detection rate of a few hundred forward scatter events per second or less. 8. Sorting speed was limited by the motorized movement of the microtiter multi-well plate. 9. A 488 nm, 200 mW solid state laser was used for forward and side scatter measurements. The laser was operated at 50 % of its maximum output during bacterial cell sorting (The laser is a Coherent SapphireTM 488-200 CDRH Optically Pumped Semiconductor Laser). Forward and side-scatter detectors were adjustable gain photomultipliers capable of single photon sensitivity.

3.5. Cell Sorting: Proof of Design Using Control Organisms

1. In an MACS Anaerobic Workstation, each well of a 96-well plate (see Note 3) was filled with 200 mL of appropriate reduced media (see Subheading 2.2) containing the appropriate soluble sugar for each organism and yeast extract. 2. Each plate with sterile growth medium was allowed to equilibrate for 16 h in the anaerobic chamber before sorting. 3. Control organisms were grown to early mid-log phase in a serum bottle or Balch tube (see Subheading 2.2), and then chilled on ice for 30 min before being sorted.


160

S.D. Hamilton-Brehm et al.

4. Cells were diluted 1:100 in anaerobic, ice cold 1× PBS solution (see Note 4). 5. The sheath fluid of 1× PBS for the flow cytometer was kept anaerobic by boiling and sparging with N2 (see Subheading 3.4). 6. Cell sorting is monitored by sorting several thousand cells onto a microscope slide for observation via phase contrast microscopy. 7. Control organisms were sorted using an InFlux Flow Cytometer via forward/side scatter into the 96-well plates, starting with one cell per well in the “A” column and increasing to eight cells per well in the “H” column (see Fig. 1b). 8. Once the 96-well plates were inoculated, they were transferred immediately to an anaerobic chamber until sorting was complete. 9. The 96-well plates were placed in anaerobic, gastight plastic boxes (see Fig. 1, section IV) with an anaerobic gas pack (see Note 5). 10. Gastight plastic boxes were incubated at the optimal growth temperatures of 60, 75, 80, and 90 °C, respectively, for the control organisms C. thermocellum, C. obsidiansis, T. maritima, and P. furiosus. A pressure release valve was installed to prevent bulging at higher temperatures. 11. Organisms were incubated until turbidity was observed within the wells (anaerobic boxes were made of clear polycarbonate), which generally required 3–5 days at optimal temperature (see Subheading 3.12). 3.6. Environmental Enrichments

1. A volume of 1 mL was taken from the environmental sample, and injected into four serum bottles containing one of each of the carbon sources Avicel, xylan, pretreated switchgrass or pretreated Populus, at a concentration of 0.1 % (w/v) (see Fig. 1, section II). 2. A set of four carbon sources was incubated at each of the temperatures: 55, 60, 65, 70, 75, 80, and 85 °C. Serum bottles were not shaken (see Note 6). 3. After 24–30 h of incubation, an inoculum of 1 mL was transferred from each serum bottle to fresh bottles containing the same enrichment medium and incubated at the same temperature. 4. Transfers were performed three times before cell sorting with the flow cytometer.

3.7. Cell Sorting: Environmental Enrichments

1. Enrichment cultures inoculated from Rabbit Creek had been transferred three times at each of the following temperatures: 55, 60, 65, 70, 75, 80, and 85 °C.


15

Anaerobic High-Throughput Cultivation Method…

161

2. Cultures that had reached densities of ³108 cells/ml and tested by HPLC for the presence of >1 mM ethanol, were selected for FCM sorting. Cultures that had been cultivated at 60 °C possessed both of these conditions. 3. In an anaerobic chamber, each well of a 96-well plate (see Note 3) was filled with 200 mL of 0.4 % (w/v) glucose, 0.4 % (w/v) xylose, and 0.1 % (w/v) yeast extract (see Note 7). 4. Each plate was allowed to degas for 16 h in the anaerobic chamber before sorting. 5. The targeted 60 °C cultures were grown to early mid-log phase in a serum bottle or Balch tube (using appropriate reduced insoluble substrate medium), and then chilled on ice for 30 min before being sorted. 6. Cells were diluted 1:100 in anaerobic, ice cold 1× PBS solution (see Note 5). 7. The sheath fluid of 1× PBS for the flow cytometer was kept anaerobic by boiling and sparging with N2 as described above (see Fig. 1b). 8. Cell sorting was performed as described above and microscopy is used to find sorting gates that yield primarily single cells. 9. Once the 96-well plates were inoculated with one cell/particle per well, the plates were transferred to an anaerobic chamber for 12 h to ensure anaerobic conditions. 10. The 96-well plates were then transferred to an anaerobic gastight plastic box (see Fig. 1, section IV) and then transferred to an incubator set for 60 °C for 5 days (see Note 8). 3.8. Cell Transfer and Cultivation

1. Anaerobic boxes were removed from the incubator and transferred into an anaerobic chamber. 2. The 96-well plates were removed from the boxes and a 100 mL volume was removed from visibly turbid wells and was transferred to a Balch tubes containing 0.1 % (w/v) yeast extract with either 0.1 % (w/v) pretreated switchgrass, 0.1 % (w/v) pretreated Populus, 0.1 % (w/v) Avicel, 0.1 % (w/v) xylan, or 0.4 % (w/v) cellobiose (see Note 10). 3. Tubes were then transferred to a 60 °C oven for incubation (see Fig. 1, section V). 4. The 96-well plates were then scanned using a plate reader at 600 nm.

3.9. Analysis: Plate Reading

1. Once plates achieved sufficient growth, they could be measured by optical density at 600 nm with a scanning plate reader, or wells could be sampled and viewed by microscopy (see Note 9).


162

S.D. Hamilton-Brehm et al.

2. Cell proliferation by optical density at 600 nm was determined by an absorbance reading above an uninoculated blank (see Subheading 3.12). 3.10. Analysis: HPLC

1. All standards were prepared in uninoculated culture media to account for interference of salts in the refractive index (RI) detector. 2. HPLC analysis was used to determine the concentration of acetate, lactate, and ethanol in Rabbit Creek sorted cells transferred from 96-well plate into Balch tubes. 3. A sample of one mL was filtered (0.2 mm) and acidified with 200 mM sulfuric acid (final concentration of 5 mM) before it was injected into an HPLC. 4. Metabolites were separated on an Aminex HPX-87 H column under isocratic temperature (60 °C) and 5 mM sulfuric acid. Each sample was run for 30 min on the column with a flow rate of 0.6 mL/min and passed through an RI detector. 5. Samples were run in technical triplicate. Identification of metabolites was performed by comparison of retention times with known standards. Quantification of metabolites was calculated against linear standard curves.

3.11. DNA Extraction

1. The total community genomic DNA (cgDNA) was isolated from 3 mL of each enrichment culture using the PowerSoil™ DNA Isolation Kit. 2. The cgDNA was amplified using GoTaq DNA polymerase and bacteria-specific primers 8F (5¢-AGA GTT TGA TCC TGG CTC AG-3¢) and 1492R (5¢-GGT TAC CTT TTA CGA CTT-3¢). 3. The PCR products (~1.5 kb) were purified from UltraPureTM Agarose using QIAquick Gel Extraction kit. 4. PCR products were ligated in pCR 2.1-TOPO vectors, transformed into One Shot TOP10 chemically competent E. coli, and plated onto LB agar containing 50 mg/mL kanamycin and 40 mg/mL X-gal. 5. Transformants were incubated overnight at 37 °C and white colonies were selected and regrown separately in LB broth with 50 mg/mL kanamycin at 37 °C with shaking at 200 rpm. 6. 16S rRNA genes were then sequenced using the BigDye Terminator v3.1 Cycle Sequencing kit and a set of universal bacterial primers to get overlapping fragments.

3.12. Example Results and Discussion

This chapter describes the application of FCM for cell sorting and recovery of strict anaerobic thermophilic microorganisms. To test the procedure, we sorted strictly anaerobic, thermophilic and hyperthermophilic organisms C. thermocellum, C. obsidiansis,


15

Anaerobic High-Throughput Cultivation Method…

163

Table 2 Success rates for cultivating control strains in 96-well plates Percentage of turbid wells after sorting and incubation # of cells sorted per well Organism sorted

1

2

3

4

C. thermocellum

66.7

83.3

96.7

100

C. obsidiansis

73.3

96.7

98.3

100

T. maritima

95.0

100

100

100

P. furiosus

86.7

98.3

98.3

100

T. maritima, and P. furiosus (see Table 1). Using FCM methods, no less than 67 % growth recovery from a single cell inoculated into a well was achieved (see Table 2). Using the same method to isolate organisms from enrichment could accelerate thermophilic microbial discovery. Rabbit Creek, YNP is a thermal site where an abundant plant material is constantly falling into the thermal springs, providing carbon and energy for degradation by cellulolytic and xylanolytic microorganisms. The general work-flow is outlined in Fig. 1, where samples from Rabbit Creek were enriched for growth on model substrates (xylan, Avicel, dilute acid-pretreated switchgrass, and Populus) and incubated at a range of temperatures (55, 60, 65, 70, 75, 80, and 85 °C). Many organisms were observed to grow in the enrichments, but only organisms growing at 60 °C produced measurable ethanol concentrations and attained a cell density of >108 cells/mL. This result agreed with expectations as the measured temperature of the Rabbit Creek sample site was 67.5 °C. All substrate enrichments from the 60 °C incubation were sorted using FCM, yielding an average of 17 % growth recovery, with the highest recovery observed at 27 % from the pretreated switchgrass enrichment (see Fig. 2). We believe the variability between successfully cultivated cells from Rabbit Creek enrichments (17–27 %) versus the control organisms (>66 %) was due primarily to the gate setting on the FCM not being able to distinguish between cells and insoluble particles suspended in the sample. Fluorescent dyes could potentially be used to distinguish between cells and debris, but DNAbinding fluorophores may diminish the cells’ viability, and/or nonspecific binding of dyes to abiotic particles may occur. Limiting the gating parameters could potentially also limit the diversity (based on morphology) of the enriched microbial population.


164

S.D. Hamilton-Brehm et al.

Fig. 2. Success rates for obtaining putative isolates from Rabbit Creek enrichment cultures using FCM. Primary enrichments (SWG, switchgrass) were established on insoluble substrates (x-axis) and vortexed before sorting into 96-well plates. The mean percentage of turbid wells (OD600 from 0.1 to 0.6) after 5 days of incubation at 60 °C was determined for three replicate 96-well plates. Error bars represent ± standard deviation.

Thus, a decrease in recoverable cell viability is unavoidable if the enrichments are conducted on insoluble substrates including plant biomass. Following enrichment of the Rabbit Creek sample and isolation via FCM, 16 wells displayed turbid growth. These wells were selected to be transferred into new media tubes for end product and phylogenetic analysis. Fermentation end products were determined via HPLC as described previously (20). The average amount of acetate detected was 6.5 mM and the average ethanol concentration was 3.7 mM from 14 cultures, with two putative isolates, SWG E03 and SWG F04, producing 16 mM acetate and 30–31 mM ethanol (see Fig. 3). Phylogenetic analysis of these two putative isolates showed that the organisms belong to the genus Caloramator (see Fig. 4). Sequences have been deposited for several Caloramator sp. isolates under GenBank Accession numbers HQ342684HQ342689. A total of 24 clones from 0.1 % Avicel enrichment RC0708-11 culture medium at 60 °C were sequenced. Twelve of the clones were related to Caldicellulosiruptor with 96 % identity to Caldicellulosiruptor obsidiansis (ATCC 2073), and 12 clones were related to Caloramator (see Fig. 4 and Note 10). FCM methods have been used as an inoculation tool for applications in microbiology and biotechnology (6), this study expands its potential to include isolation of anaerobic microorganisms inhabiting thermal environments. Reducing the metabolic activity of the inoculum (4 °C treatment of thermophiles) and minimizing


15

Anaerobic High-Throughput Cultivation Methodâ&#x20AC;Ś

165

Fig. 3. Several putative isolates were screened for ethanol production resulting from fermentative growth on cellobiose. Gray bars, acetate; white bars, ethanol. Lactate was not detected. Error bars represent Âą the standard deviation of the mean for triplicate samples.

Fig. 4. Phylogenetic analysis of small subunit rRNA genes for single-cell isolates (indicated in bold ) obtained from enrichment cultures using anaerobic FCM. The control organisms used in this study are underlined. The tree was produced by a neighbor-joining method. Bootstrap values are based on 1,000 replicates. The scale bar represents 0.05 changes per nucleotide position. Swg, switchgrass; Pop, Populus; Xyl, xylan. Sequences have been deposited under GenBank Accession numbers HQ342684-HQ342689.


166

S.D. Hamilton-Brehm et al.

exposure to air is required for best results. One potential application could be screening for desirable mutant phenotypes from a wildtype background without relying on tedious plating techniques (e.g., roll tubes). Hence, we have successfully shown cell sorting of thermophilic anaerobic microbes for both validly described model organisms and novel fermentative thermophiles from biomassdegrading environments in YNP. These results confirm FCM methods effectively assist in isolation and screening of novel microbes from extreme environments. 3.13. Notes

1. Vigorous stirring is necessary to make sure insoluble substrates do not settle, causing uneven distribution of substrate between bottles. Too much stirring, however, can also allow air to enter the media bottle while gassing. 2. In 2008, Cytopeia was acquired by Becton Dickenson. BD currently markets the flow cytometer as the BD Influx. 3. Sterile polystyrene nonpyrogenic 96-well plates with low evaporation lids (Becton Dickenson Labware, NJ) are necessary for two reasons. First, the polystyrene nonpyrogenic plastic is nontoxic to cells being incubated at elevated temperatures. Second, the lid keeps evaporation at the elevated temperature under control for the maximum of 5-day incubations (when inside a sealed incubation box). 4. Thermophiles are typically in a low metabolic state at ambient or colder temperatures. 5. The 96-well plates can be briefly removed from the anaerobic chamber and quickly placed into the plastic boxes while flushing with nitrogen or the plastic boxes can be brought into the anaerobic chamber and the plates placed inside them at that moment. In either case, the gastight anaerobic plastic boxes must be securely sealed with anaerobic gas packs inside. 6. Shaking was shown to decrease growth of some organisms. Characterization of the effects of agitation on an organism can be performed upon further characterization of a bone fide isolate. 7. Soluble glucose and xylose were chosen since they comprise the monomers for cellulose and xylan, respectively. Any insoluble debris would interfere with optical density readings. 8. Sufficient incubation time is organism dependant; but, five days was sufficient to allow organisms with slower growth rates to be detected. Anaerobic boxes have been incubated in ovens up to 95 °C with only minor modifications to the box to adjust for gas pressure. 9. When viable cells in individual wells are transferred to new media, culture plates should be kept under anaerobic conditions. 10. Additional tests should be used to confirm a pure culture has been obtained as two cells can be adhered to each other and


15

Anaerobic High-Throughput Cultivation Method…

167

rarely the FCM can place two cells in one well. Secondary steps to verify that a pure culture has been obtained include application of classical microbiology methods for cell isolation combined with rRNA gene sequence analysis. These approaches have been used to verify purity of organisms that were characterized in more detail including C. obsidiansis and Caloramator sp. isolates.

Acknowledgments We thank Sue Carroll and Marilyn Kerley who provided excellent technical assistance for many of the procedures described in this chapter. Jennifer J. Moser provided additional metabolite analysis. Martin Keller and Anthony V. Palumbo participated in helpful discussions. We also thank the National Park Service for coordinating and allowing sampling under permit YELL-2008-SCI-5714. This work was supported by the BioEnergy Science Center (BESC), which is a US Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science, Oak Ridge National Laboratory. Oak Ridge National Laboratory is managed by UT-Battelle, LLC, for the US Department of Energy under contract DE-AC05-00OR22725. References 1. Tracy BP, Gaida SM, Papoutsakis ET (2008) Development and application of flow-cytometric techniques for analyzing and sorting endosporeforming clostridia. Appl Environ Microbiol 74:7497–7506 2. Papadimitriou K, Pratsinis H, Nebe-vonCaron G, Kletsas D, Tsakalidou E (2007) Acid tolerance of Streptococcus macedonicus as assessed by flow cytometry and single-cell sorting. Appl Environ Microbiol 73:465–476 3. Forster S, Snape JR, Lappin-Scott HM, Porter J (2002) Simultaneous fluorescent gram staining and activity assessment of activated sludge bacteria. Appl Environ Microbiol 68: 4772–4779 4. Bergquist PL, Hardiman EM, Ferrari BC, Winsley T (2009) Applications of flow cytometry in environmental microbiology and biotechnology. Extremophiles 13:389–401 5. Prigione V, Lingua G, Marchisio VF (2004) Development and use of flow cytometry for detection of airborne fungi. Appl Environ Microbiol 70:1360–1365

6. Zengler K (2009) Central role of the cell in microbial ecology. Microbiol Mol Biol Rev 73:712–729 7. Tracy BP, Gaida SM, Papoutsakis ET (2010) Flow cytometry for bacteria: enabling metabolic engineering, synthetic biology and the elucidation of complex phenotypes. Curr Opin Biotechnol 21:85–99 8. Connon SA, Giovannoni SJ (2002) Highthroughput methods for culturing microorganisms in very-low-nutrient media yield diverse new marine isolates. Appl Environ Microbiol 68:3878–3885 9. Zengler K, Toledo G, Rappe M, Elkins J, Mathur EJ, Short JM, Keller M (2002) Cultivating the uncultured. Proc Natl Acad Sci U S A 99:15681–15686 10. Wyman CE (2007) What is (and is not) vital to advancing cellulosic ethanol. Trends Biotechnol 25:153–157 11. Shaw AJ, Podkaminer KK, Desai SG, Bardsley JS, Rogers SR, Thorne PG, Hogsett DA, Lynd LR (2008) Metabolic engineering of a thermophilic


168

12.

13.

14.

15.

16.

17.

18.

S.D. Hamilton-Brehm et al. bacterium to produce ethanol at high yield. Proc Natl Acad Sci U S A 105:13769–13774 Yang SJ, Kataeva I, Hamilton-Brehm SD, Engle NL, Tschaplinski TJ, Doeppke C, Davis M, Westpheling J, Adams MW (2009) Efficient degradation of lignocellulosic plant biomass, without pretreatment, by the thermophilic anaerobe “Anaerocellum thermophilum” DSM 6725. Appl Environ Microbiol 75:4762–4769 Zhang YH, Lynd LR (2005) Cellulose utilization by Clostridium thermocellum: bioenergetics and hydrolysis product assimilation. Proc Natl Acad Sci U S A 102:7321–7325 Lynd LR, Weimer PJ, van Zyl WH, Pretorius IS (2002) Microbial cellulose utilization: fundamentals and biotechnology. Microbiol Mol Biol Rev 66:506–577 Lynd LR, Grethlein HE, Wolkin RH (1989) Fermentation of cellulosic substrates in batch and continuous culture by Clostridium thermocellum. Appl Environ Microbiol 55:3131–3139 Blumer-Schuette SE, Kataeva I, Westpheling J, Adams MW, Kelly RM (2008) Extremely thermophilic microorganisms for biomass conversion: status and prospects. Curr Opin Biotechnol 19:210–217 Roberts SB, Gowen CM, Brooks JP, Fong SS (2010) Genome-scale metabolic analysis of Clostridium thermocellum for bioethanol production. BMC Syst Biol 4:31 Miroshnichenko ML, Kublanov IV, Kostrikina NA, Tourova TP, Kolganova TV, Birkeland NK, Bonch-Osmolovskaya EA (2008) Caldicellulosiruptor kronotskyensis sp. nov. and Caldicellulosiruptor hydrothermalis sp. nov., two extremely thermophilic, cellulolytic, anaer-

obic bacteria from Kamchatka thermal springs. Int J Syst Evol Microbiol 58:1492–1496 19. Verhagen MF, Menon AL, Schut GJ, Adams MW (2001) Pyrococcus furiosus: large-scale cultivation and enzyme purification. Methods Enzymol 330:25–30 20. Hamilton-Brehm SD, Mosher JJ, Vishnivetskaya T, Podar M, Carroll S, Allman S, Phelps TJ, Keller M, Elkins JG (2009) Caldicellulosiruptor obsidiansis sp. nov., an anaerobic, extremely thermophilic, cellulolytic bacterium isolated from Obsidian Pool, Yellowstone National Park. Appl Environ Microbiol 76:1014–1020 21. Zhang YH, Lynd LR (2005) Regulation of cellulase synthesis in batch and continuous cultures of Clostridium thermocellum. J Bacteriol 187:99–106 22. Schell DJ, Farmer J, Newman M, McMillan JD (2003) Dilute-sulfuric acid pretreatment of corn stover in pilot-scale reactor: investigation of yields, kinetics, and enzymatic digestibilities of solids. Appl Biochem Biotechnol 105–108: 69–85 23. Freier D, Mothershed CP, Wiegel J (1988) Characterization of Clostridium thermocellum JW20. Appl Environ Microbiol 54:204–211 24. Huber R, Langworthy TA, Konig H, Thomm M, Woese CR, Sleytr UB, Stetter KO (1986) Thermotoga maritima sp-nov represents a new genus of unique extremely thermophilic eubacteria growing up to 90 degrees C. Arch Microbiol 144:324–333 25. Fiala G, Stetter KO (1986) Pyrococcus furiosus sp-nov represents a novel genus of marine heterotrophic archaebacteria growing optimally at 100-degrees C. Arch Microbiol 145:56–61


Chapter 16 Assessing the Protein Concentration in Commercial Enzyme Preparations William S. Adney, Nancy Dowe, Edward W. Jennings, Ali Mohagheghi, John Yarbrough, and James D. McMillan Abstract Although a poor indicator of how a cellulase preparation will perform on biomass, the filter paper unit (FPU) still finds wide use in the literature as an apparent measure of performance efficacy. In actuality, the assessment of commercial enzyme preparation performance in terms of biomass conversion or solubilization of insoluble polysaccharides is largely dependent on the substrate composition, which cannot be easily standardized. Commercial cellulase preparations are evaluated based upon their performance or specific activity. The ability to compare commercial enzyme preparation efficacy across a wide variety of different preparations requires defining the amount of enzyme protein required in milligrams per gram of cellulose to achieve a targeted level of cellulose hydrolysis in a specified timeframe. Since biomass substrates are highly variable, reproducible and accurate protein determination is as important as performance testing to be able to rank order the effectiveness of diverse preparations. This chapter describes a protocol that overcomes many of the difficulties encountered with determining the protein concentration in commercial cellulase preparations. Key words: Cellulase, Cellulose, Protein assay, Enzymatic hydrolysis

1. Introduction The use of cellulosic biomass for the production of fuels and chemicals is gaining momentum due to mounting geopolitical issues that impact petroleum supply and costs, coupled with recent scientific advancements in biochemical conversion technologies. Decreasing the cost of the enzymesâ&#x20AC;&#x201D;generally referred to as cellulasesâ&#x20AC;&#x201D;used to convert the complex, highly structured, polymeric carbohydrates cellulose and hemicellulose found in lignocellulosic biomass is an area of active research. Funding by the United States Department of Energy has led to substantial

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_16, Š Springer Science+Business Media, LLC 2012

169


170

W.S. Adney et al.

reductions in the cost of cellulase enzymes for biomass utilization (1–4), albeit the cost of these enzymes remains one of the barriers to economically producing ethanol and other fuels and chemicals from cellulosic biomass (5–7). The term “cellulase” as used here refers to multienzyme mixtures exhibiting a wide variety of activities and specificities towards the different structural carbohydrates in biomass. Cellulases are produced by a large number of diverse fungi and bacteria, with the most important commercial cellulase preparations currently produced by filamentous fungi using submerged aerobic cultivation (8–10) or solid-state fermentation (10). Although cellulases are beginning to be developed and marketed for the biomass conversion industry, they are already in longstanding commercial use in many applications including laundry detergents, animal feed, fruit juice, and beverage processing and bio-polishing of denim fabrics and other textiles (8). In a typical enzyme-based biomass conversion processes, a thermochemical pretreatment is used to reduce the amount and complexity of enzyme types required to efficiently hydrolyze biomass polysaccharides to fermentable sugars (11). A variety of different pretreatment chemistries are being considered; however, the type of pretreatment can substantially alter the type and quantity of commercial enzymes required to achieve high enzymatic hydrolysis yields. This means that a “true” assessment of the performance of a cellulase enzyme preparation has as much if not more to do with the level of the reactivity of the substrate than it does with the characteristics of the enzyme preparation. The cost to produce industrial enzymes, including cellulases, is typically proportional to the quantity and titer of enzyme protein produced and the amount of formulation applied, and favorable production economics are achieved by minimizing the cost per unit volume of final product. Commercial cellulase preparations used for commodity applications like cellulose conversion to sugars and biofuels are generally marketed and priced according to their efficacy (value provided) on a volumetric activity basis rather than on their protein concentration per se. However, poor surrogates of volumetric effectiveness like activity on filter paper, expressed in filter paper units (FPU), are often interpreted as approximate indicators of preparation potency (12, 13). Unfortunately, the measurement of FPU, which is based upon the amount of enzyme protein needed to solubilize 4 % of a 50 mg Whatman #1 filter paper strip in 1 h, gives relatively little insight about how a given commercial enzyme preparation will perform on an actual sample of pretreated biomass. Because the protein concentrations in commercial cellulase preparations vary widely (14), preparations are usually assessed and compared on a protein concentration-normalized basis. Accurate determination of protein concentration thus becomes a critical aspect of cellulase performance evaluations.


16

Assessing the Protein Concentration in Commercial Enzyme Preparations

171

Commercial enzyme preparations are highly complex mixtures of many disparate components. In addition to containing many different enzyme proteins, a variety of additives used as stabilizers and preservatives as well as many constituents carried over from the fermentation production broth are also often present. Thus, there are several features of most commercial cellulase preparations that must be carefully considered before beginning an analysis. These include the high protein concentration, preparation viscosity, pH, and pH behavior of preparation diluents including residual metabolites produced by the production microorganism or present in the nutrient medium. Most preparations have very high protein concentrations (³100 mg/mL), which tend to increase protein– protein interactions, contribute to high viscosity, and promote greater protein self-association and aggregation. Commercial cellulases generally contain polyol additives such as glycerol and sorbitol, which are intended to stabilize and preserve the preparation but also can significantly affect the polarity and osmolarity of the solution. Examples of commercial enzyme preparation compositions published by Nieves et al. showed concentrations of sorbitol and glycerol as high as and 19 mg/mL, respectively (14). While the presence of such co-solvents in commercial cellulase preparations helps to stabilize the concentrated protein solutions, it can also have substantial consequences on the measurement of total protein content. It is typically assumed, sometimes incorrectly, that all components of a commercial cellulase preparation remain soluble when the preparation is diluted and used for an application such as simultaneous saccharification and fermentation (SSF). However, the maximum protein solubility depends on a number of factors including salt concentrations and the presence of modifying agents that may have been added to the preparation as well as on the characteristics of the enzyme proteins themselves. The salt concentration and pH of diluents introduced when applying or measuring preparation performance can influence protein–protein interactions and cause proteins which were considered to be “salted in” to precipitate out as salt concentrations change. Proteins in general are less stable near their isoelectric point where precipitation can occur as the net charge of the protein approaches zero. This issue becomes more important as enzyme formulations are produced that contain heterologously expressed proteins that are more or less hydrophobic and have different net charges. Changes in the surface characteristics of a protein can also have large effects on protein interactions and thus on solubility. Depending on a protein’s surface chemistry, molecular size, and its tertiary structure flexibility, the addition of low molecular weight polyol stabilizers may alter protein solution behavior in a number of ways. The addition of glycerol, sorbitol, or other polyols are thought to stabilize proteins in solution by preferentially hydrating the proteins to


172

W.S. Adney et al.

minimize surface contact between the proteins and primary solvent (14). The mechanism of protein stabilization by glycerol is thought to be a result of electrostatic interactions that induce orientations of glycerol molecules at the protein surface such that glycerol is further excluded (15). Since the assessment of a given cellulase preparation is primarily dependant on the substrate composition, which is difficult to standardize, the critical measure when comparing the performance of different commercial preparations on a specific substrate is the protein dosage or loading needed to achieve a desired target conversion level. There are a number of protein concentration measurement methods and commercial assay kits available that can be used to determine the concentrations of proteins in solution. Ideally, to be useful and widely adopted, any method used for a commercial cellulase preparation needs to be sensitive, accurate, and easily applied. Selection of measurement method or test kit must be considered carefully based upon the compatibility of the detection method with the samples being assayed. The use of kits produced and sold by companies such as Thermo Scientific and reported on here give the advantages of being broadly available and backed up by substantial quality control and standardization provided by the manufacturer.

2. Materials 2.1. Sample Preparation

1. Sample equilibrated at room temperature (see Notes 1 and 2).

2.2. Desalting Chromatography

1. Enzyme, at room temperature.

2. Millipore 0.22 Mm filter or equivalent low protein binding filter.

2. HiPrep 26/10 desalting column, GE Healthcare Life Sciences, Piscataway, NJ (see Note 3). 3. FPLC with 0.5 and 2.0 mL calibrated loop, UV (280 and 254 nm), conductivity detector, and fraction collector. 4. Buffer, 20 mM acetate, 100 mM NaCl, pH 5.0 filtered and degassed at least 1 L (see Note 4).

2.3. Protein Determination Assay

1. BCA Protein Assay Kit, Thermo Scientific, Fremont, CA.


16

Assessing the Protein Concentration in Commercial Enzyme Preparations

173

3. Methods 3.1. Desalting Chromatography

1. Exchange buffers in FPLC and equilibrate column with starting buffer. 2. Select a calibrated loop and attach to the FPLC. Use the 0.5 mL loop for high concentration samples. 3. Filter sample through a 0.22 Mm filter. 4. Overfill the calibrated loop with at least half the volume of the loop. When using the 0.5 mL loop, load the column with 1.0 mL and when using the 2 mL loop use at least 3.0 mL of sample. After loading the loop do not remove the syringe from the filling port. 5. Establish a flow rate of 5 mL/min (see Note 5). 6. Inject the sample onto the column using an injection volume of twice the volume of the calibrated loop. 7. The sample is isocratically eluted at a flow rate of 5 mL/min. Follow the elution of the protein at UV280 and the elution of the salt peak using conductivity. 8. Collect 5 mL fractions using pre-weighed 16 Ă&#x2014; 100 mm glass tubes. Accurate and reproducible flow rate are needed for resolution and reproducible separations. The use of preweighed glass tubes helps to confirm accurate volumes. 9. Pool the fractions containing the peak containing the high molecular weight fraction. This generally represents a tenfold dilution of the original sample.

3.2. Protein Determination

1. Dilute the BSA standards according to the product instructions. 2. Dilute the desalted sample to a concentration within the range of your BSA standard. Depending on the dilution factor of the desalting chromatography and the concentration of the starting material several dilutions may need to be tried. 3. Set up the standards and diluted sample according to the BCA product instructions. This will depend primarily on using a microtiter plate or test tubes (see Note 6). 4. Measure the absorbance of the standards and the samples at 562 nm. 5. Prepare a standard curve using absorbance valued obtained for the BSA standards according to the manufacturerâ&#x20AC;&#x2122;s instruction. It is important to note that the standard curve may not be linear, but instead must be fixed with a quadratic equation.


174

W.S. Adney et al.

6. Using only measurements of your sample that fall within the standard curve calculate the concentration of protein in the diluted samples. 7. Calculate the protein concentration of the original sample by taking into account the product of all the dilutions made, including the dilution from the desalting chromatography step.

4. Discussion The method describes here includes an initial â&#x20AC;&#x153;desaltingâ&#x20AC;? chromatography protocol that serves to remove low molecular weight compounds and put the macromolecular, enzyme proteincontaining components in a commercial cellulase preparation into a standardized buffer prior to measuring protein concentration. The desalting chromatography step separates the sample into two fractions, a high molecular weight fraction containing constituents of molecular size greater than 5,000 Da and a low molecular weight fraction containing small molecules less than 5,000 Da such as salts, additives, and various residual medium components. The purpose of this step is to remove compounds that could otherwise suppress, enhance, or elevate apparent protein concentrations by interfering with the protein assay chemistry. Figure 1 shows an example of a desalting chromatography run.

Fig. 1. Example of a desalting chromatography run illustrating the separation of the macromolecular component (peak A) greater than 5 kDa, from the low molecular components that can interfere with assaying total protein (peak B). In this example the protein peak was collected from a total of four fractions or 5 mL, or a total of 20 mL. The dilution factor from the desalting step in this case is 10Ă&#x2014;.


16

Assessing the Protein Concentration in Commercial Enzyme Preparations

175

Fig. 2. To illustrate the potential for salting out a commercial cellulase was diluted in two different buffers; one at pH 5.0 and the other at pH 6.5. Aggregation of a component from the cellulase complex can occur depending upon the pH, or upon dilution. The effect is generally immediate and easily visible.

Before the protein solution sample can be analyzed, it must be equilibrated to room temperature and all protein must remain solubilized. Tests should be performed beforehand to ensure compatibility with the desalting chromatography elution buffer. This requires verifying that proteins will not precipitate or â&#x20AC;&#x153;salt outâ&#x20AC;? of solution during any phase of the chromatography run. When making dilutions of commercial enzyme preparations or exchanging buffers during desalting chromatography, it is important to avoid extreme changes in solution pH that may cause protein precipitation and column blockage. Figure 2 shows an example of the effect solution pH changes can have on protein solubility, in this case caused by diluting a commercial cellulase preparation into a different pH buffer. The picture on the left shows a commercial enzyme diluted tenfold into 50 mM acetate buffer at pH 5.0 and on the right similarly diluted tenfold but into 50 mM Bisâ&#x20AC;&#x201C;Tris at pH 6.5. The solution is considerably more turbid, indicative of increased protein precipitation, when it is diluted tenfold at pH 5.0 compared to pH 6.5. The rationale for the selection of the BCA assay for protein determination over other rapid dye-based chemistries has its basis in a protein assay comparison previously published by Adney et al. (16).


176

W.S. Adney et al.

This earlier work included a comparison of three different dye-based protein assay methods—the BCA, modified Lowry and Bradford assay methods—for measuring protein concentration in commercial cellulase preparations; results were compared against those determined based on Kjeldahl nitrogen analysis. The Bradford method was found to significantly underestimate protein concentration in this earlier work compared to the modified Lowry and BCA methods. The BCA assay was ultimately adopted because in round robin testing of a standard enzyme preparation it exhibited considerably lower assay-to-assay variability than the Lowry assay. The higher variability observed with the Lowry assay in this earlier protein round robin study may have been due to a number of factors including the stability of the reagents used for the assays themselves, since these were self made. The use of commercial protein determination kits, such as the Pierce® BCA protein assay kit employed in the work reported here, has the benefit of greatly reducing reagent variability and also providing a traceable protein standard for assay calibration (12). Protein determinations must always be referenced by the assay method and protein calibration standard used. The selection of both is critical, particularly when the composition and source of the enzyme preparation to be measured is unknown. Most commercial enzyme preparations are produced in fungi using Trichoderma reesei, Aspergillus niger, or similar filamentous fungal species. A major protein component in all whole broth cellulases produced by such fungi undoubtedly will be a glycoside hydrolase family 7 (Cel7) reducing end-specific cellobiohydrolase. In non-engineered T. reesei preparations, for example, this single enzyme can form 30–50 % of the total enzyme protein present (17). The differential reactivity of individual proteins components to the protein assay chemistry can be quite dramatic and potentially lead to significant errors in estimating total protein content. To illustrate this point, we compared the reactivity of several highly purified mono-component enzymes likely to be present in multicomponent commercial enzyme preparations to the bovine serum albumin (BSA) standard provided in the Pierce® BCA protein assay kit (12). The “true” protein concentrations were calculated by estimating the molar extinction coefficient calculated from the proteins’ primary amino acid sequences using the Web-based ExPASy ProtParam tool (Swiss Institute of Bioinformatics) (18). A comparison of the response of each of the individual proteins is shown in Table 1, with responses normalized as their ratios compared to the BSA protein standard. The response values were determined as the slope calculated by best-fit linear regression of four protein concentrations normalized to a reagent blank, as shown in Fig. 3. Also shown in Table 1 is a comparison of the response of each tested mono-component protein relative to the BSA standard using commonly used protein assay chemistry, the 660 nm Protein


16

Assessing the Protein Concentration in Commercial Enzyme Preparations

177

Table 1 The results from a comparison of the response of purified commercial cellulase components relative to the response of the BSA standard for both the BCA assay and the Pierce 660 assay illustrating the potential to obtain significantly different values depending upon the reagent chemistries Protein

BCA ratio

660 ratio

BSA

1.000

1.000

T. reesei Cel7A

0.823

0.036

T. reesei Cel6A

1.074

0.027

P. chrysosporum Cel7D

2.348

0.019

Fig. 3. The response of different highly purified cellulase components t could be potentially be found in a commercial cellulase compared to the response of the BCA kit standard. The response was obtained over a range of four concentrations of protein plotted after being corrected for a reagent blank. Cel7A and Cel6A were chosen because they contribute to a major fraction of the total protein in most commercial cellulase preparations.


178

W.S. Adney et al.

Table 2 Comparison of protein concentrations before and after desalting chromatography of five commercial cellulase preparations using the standard BCA assay Commercial cellulase

Protein concentration de-salted sample (mg/mL)

Protein concentration “asreceived” sample (mg/mL)

Percent difference due to desalting step

1

127

148

14.19

2

234

255

8.24

a

212

230

7.83

a

122

140

12.86

a

102

114

10.53

3 4 5

The contribution of non-macromolecular components is presented as the percent difference from the sample as it was received a Previously reported in McMillan et al. 2011 (in press)

Assay Kit, also marketed by Thermo Scientific, which is equivalent to the Bradford Coomassie-based dye binding assay (12). As the data in Table 1 show, the individual protein response factors obtained using the BCA assay chemistry are order 1, i.e., of the same magnitude as the response obtained using the BSA standard. In contrast, the responses obtained using the Bradford (“660 nm”) type chemistry are 25-fold to 50-fold lower than for the BSA standard, reflecting the fact that this assay chemistry will significantly underestimate fungal-based cellulase protein concentrations if BSA is being used as the standard. The use of the desalting technique removes small molecular weight compounds that potentially interfere with the detection of the macromolecular component of the protein and adds to the accuracy of the determination. In Table 2 a comparison is made using the BCA assay of five commercial enzyme preparations. In general the desalting chromatography values were from 7.8 to 14 % lower that the protein values obtained when the starting material was assayed. The discrepancy between the two values represents variability of the contribution of small molecular weight compounds found in most commercial enzyme preparations. 4.1. Notes

1. The quality of the results begins with the ability to obtain a representative sample. Frozen samples should be thawed completely, allowed to reach room temperature, and completely mixed before testing. 2. All procedures should be carried out at room temperature unless otherwise specified 3. The HiPrep 26/10 column is packed with Sephadex™ G-25 Fine, has a total volume of 53, and void volume of 15 mL.


16

Assessing the Protein Concentration in Commercial Enzyme Preparations

179

The recommended sample volume is 2.5–15 mL with an elution volume of 7.5–20 mL. The nominal size exclusion limit is 5 kDa for globular proteins. Recommended flow rates are 9–31 mL/min with a maximum of 40 mL/min. A lower flow rate is recommended for highly viscous samples. 4. Always use filtered and degassed buffers; the buffer type and pH must be determined experimentally based on the source of the enzyme preparation. 5. A sample volume must be selected that avoids peak overlap and peak broadening. Lower or reduced flow rates during the desalting chromatography helps to prevent compression of the column packing and allows for better separation with samples containing high protein concentrations. The viscosity of the sample can impact the separation and sample volume is determined by experimentation. 6. There are several formats of the BCA protein assay available. All will work for this method as long as the product instructions are carefully followed. References 1. Aden A, Foust T (2009) Technoeconomic analysis of the dilute sulfuric acid and enzymatic hydrolysis process for the conversion of corn stover to ethanol. Cellulose 16:535–545 2. Merino ST, Cherry J (2007) Progress and challenges in enzyme development for biomass utilization. Adv Biochem Eng Biotechnol 108:95–120 3. Teter SA, Xu F, Nedwin GE, Cherry JR (2006) Enzymes for biorefineries. In: Kamm B, Gruber PR, Kamm M (eds) Biorefineries industrial processes and products, vol 1. WileyVCH, Weinheim, pp 357–383 4. Dean B, Dodge T, Valle F, Chotani G (2006) Development of biorefineries - technical and economic considerations. In: Kamm B, Gruber PR (eds) Biorefineries - industrial processes and products, vol 1. Wiley-VCH, Weinheim, pp 67–83 5. Lynd LR, Laser MS, Brandsby D, Dale BE, Davison B, Hamilton R, Himmel M, Keller M, McMillan JD, Sheehan J, Wyman CE (2008) How biotech can transform biofuels. Nat Biotechnol 26:169–172 6. Himmel ME (2009) Corn stover conversion to biofuels: DOE’s preparation for readiness in 2012. Cellulose 16:531–534 7. Wilson DB (2009) Cellulases and biofuels. Curr Opin Biotechnol 20:295–299

8. Tolan JS, Foody B (1999) Cellulase from submerged fermentation. Adv Biochem Eng/ Biotechnol 65:41–67 9. Tolan JS (2006) Iogen’s demonstration process for producing ethanol from cellulosic biomass. In: Kamm B, Gruber PR (eds) Biorefineries - industrial processes and products, vol 1. WileyVCH, Weinhelm, pp 193–208, 2 vols 10. Cen P (1999) Production of cellulase by solidstate fermentation. Adv Biochem Eng/ Biotechnol 65:69–92 11. Elander RT, Dale BE, Holtzapple M, Ladisch MR, Lee YY, Mitchinson C, Saddler JN, Wyman CE (2009) Summary of findings from the Biomass Refining Consortium for Applied Fundamentals and Innovation (CAFI): corn stover pretreatment. Cellulose 16:649–659 12. Thermo Fisher Scientific I (2010) Thermo Scientific Pierce Protein Assay Technical Handbook, Version 2 13. Ghose GL (1987) Measurement of cellulase activities. Pure Appl Chem 59:257–268 14. Nieves RA, Ehrman CI, Adney WS, Elander RT, Himmel ME (1998) Survey and analysis of commercial cellulase preparations suitable for biomass conversion to ethanol. World J Microbiol Biotechnol 14:301–304 15. Timasheff SN (1998) Control of protein stability and reactions by weakly interacting cosol-


180

W.S. Adney et al.

vents: the simplicity of the complicated. Adv Protein Chem 51:355–432 16. Vagenende V, Yap MGS, Trout BL (2009) Mechanisms of protein stabilization and prevention of protein aggregation by glycerol. Biochemistry 48:11084–11096 17. Adney William S, Mohagheghi A, Thomas Steven R, Himmel Michael E (1996) Comparison of protein contents of cellulase preparations in a worldwide Round-Robin

assay. In: Enzymatic degradation of insoluble carbohydrates, vol 618, pp 256–271. American Chemical Society 18. Adney WS, Taylor LE, Johnson D, Park S, Knoshaug EP, Nimlos MR, Decker SR, Vinzant TB, Selig MJ, Himmel ME (2008) CELL 215-Deconstruction of biomass: Understanding enzyme/substrate interactions. Abstracts of Papers of the American Chemical Society. p 235


Chapter 17 Reducing the Effect of Variable Starch Levels in Biomass Recalcitrance Screening Stephen R. Decker, Melissa Carlile, Michael J. Selig, Crissa Doeppke, Mark Davis, Robert Sykes, Geoffrey Turner, and Angela Ziebell Abstract Cell wall recalcitrance is the largest contributor to the high expense of lignocellulose conversion to biofuels (Himmel ME et al., Science 315:804–807, 2007). In response to this problem, researchers at the BioEnergy Science Center (BESC) are working to determine the contributing factors of biomass recalcitrance. The primary approach to this is screening large sample sets of genetic and environmental variants of model and feedstock plant species for differences in recalcitrance to combined hydrothermal pretreatment and enzymatic hydrolysis (Decker S et al., BioEnergy Res 2:179–192, 2009). To handle these large sample sets (up to several thousand samples per set), the BESC has developed high throughput screening systems to evaluate both cell wall composition and recalcitrance (Selig MJ et al., Biotechnol Lett 33:961–967, 2011; Selig MJ et al., Ind Biotechnol 6, 104–111, 2010). Molecular beam mass spectroscopy and high throughput, 2-stage acid hydrolysis are used to determine amounts and ratios of cell wall components such as lignin, cellulose, and xylan. Recalcitrance is measured by glucose and xylose release after high throughput hydrothermal pretreatment and enzymatic saccharification, screening large numbers (up to 1,000 s per week) of biomass samples (Selig MJ et al., Ind Biotechnol 6, 104–111, 2010; Sykes R et al., Methods Mol Biol 581, 169–183, 2009). Implementation of these high throughput techniques revealed additional concerns when screening biomass samples for recalcitrance, principal among these was the contribution of starch to glucose release quantitation in both compositional analysis and recalcitrance screening. Key words: Starch, High throughput screening, Biomass recalcitrance, Plant cell walls, Composition analysis

1. Introduction Upon implementing the high throughput pretreatment pipeline within the BESC, questions were raised as to whether starch could be significantly contributing to glucose release results in the assay for recalcitrance. The majority of biomass composition performed at NREL is on mature wood or field dried, senesced herbaceous Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_17, © Springer Science+Business Media, LLC 2012

181


182

S.R. Decker et al.

crops, which are both typically very low in starch. When BESC researchers started sending green material for analysis, we noted an increase in variability in our recalcitrance screening results. Subsequent discussions with the plant molecular biologists led to the realization that the sheer number of plant variants being harvested from the greenhouse precluded simultaneous harvest and processing. Since greenhouse space is at a premium and waiting for samples to senesce and dry was not practical, samples were being harvested green, with a full set taking hours or days to harvest, tag, and process. We hypothesized that starch reserves in the plants could be varying by harvest time, i.e., low starch in the morning, high starch later in the day. This phenomenon is not limited to herbaceous material. Assessment of large genetic modification trials and Quantitative Trait Loci (QTL) studies of trees is typically performed before the plants reach maturity. This is necessary due to long rotation ages of trees and allows larger genetic gain in shorter amounts of time. However, this leads to issues with assessing young biomass material and correlating it with mature plants. Immature plants often contain a higher starch level than mature plants. It also can lead to issues with plant morphology due to growing plant material in greenhouses vs. field grown material. Differences between mature and immature tissue must, therefore, be accounted for in experimental methods where possible. Starch is a storage component in most biomass so the part of the plant, the season, and the time of day that the biomass is harvested can affect the amounts of starch present in the sample (3). Unless biomass is always harvested using the same procedures and during the same time of the day, one cannot be sure what proportion of starch is contributing to glucose numbers during the recalcitrance assay. Even when these variables are held as constant as possible, individual plant and environmental variations may contribute to variable starch contents. Though time of harvest data was not available for these large sample sets, we decided to test a set of samples for starch content. After developing a rapid, high throughput starch quantitation assay, testing a set of 250 switchgrass variants revealed a fairly wide range of starch contents. The high throughput (HTP) starch assay was effective in measuring the starch content of large sample numbers; however, it is time and resource intensive and application of the results to recalcitrance data would introduce more uncertainty in the final recalcitrance evaluation. We determined the most straightforward way to address starch variability was to simply eliminate starch from the samples. Therefore, we devised a high throughput method for removing the starch. Our first step was to determine if starch was present in our typical raw materials. Switchgrass, as an herbaceous species, is typically high in starch and may be susceptible to variation of starch content in large sample sets. The BESC standard switchgrass (lowland cultivar Alamo, Ardmore, OK) was used to develop the


17

Reducing the Effect of Variable Starch Levels in Biomass…

183

qualitative and quantitative starch assays. Additionally, we also tested a small sample set of immature poplar in order to evaluate the procedures on woody biomass. A basic iodine starch test was used to qualitatively determine the presence of starch in these materials both before and after amylase treatment to remove starch. Once we validated the presence of starch and the basic conditions needed to de-starch the samples, we developed a high throughput version of a standard quantitative starch measurement assay. The HTP starch assay was used to measure starch in a 250 sample switchgrass experimental dataset in order to assess the variability of starch content in a typical sample set and to measure the effect of starch removal on recalcitrance variability. The presence of starch in the sample set of 250 switchgrass confirmed that starch-derived glucose would influence the sugar release data. The high variability in the starch contents confirmed that starch would negatively influence the conclusions of the recalcitrance assay by unpredictably contributing to the total glucose release. The best way to de-starch the samples was with amylases. The amylase treatment was placed at the start of the sample preparation so that the subsequent ethanol extraction could remove the water from the sample and enable efficient sample drying. This was critical to the recalcitrance pipeline as the robotic biomass dispensing can only be carried out with dry samples.

2. Materials and Methods 2.1. Biomass

The herbaceous standard feedstock used was the BESC standard lowland cultivar Alamo Switchgrass harvested in Ardmore, on Nov. 2, 2007 and baled on Nov. 5, 2007. The BESC switchgrass standard was shipped to NREL in Golden, CO and air dried to <15 % relative humidity. The material was then milled using a Wiley mill to +20/−80 mesh (~1 mm). In addition, 250 switchgrass variants grown at the University of Tennessee-Knoxville were used for evaluation of starch variability and de-starching efficacy. All materials were received dry and milled to pass a 20-mesh screen.

2.2. Sample Preparation

Dry biomass (~95 % dry weight) was either loaded directly into 96-well HTP Ni-Teflon-coated aluminum reactors or wrapped in teabags for extraction and/or de-starching. The 2.5 in. by 2.75 in. teabags were purchased from the Monterey Bay Spice Company (http://www.herbco.com/p-1079-press-n-brew-small-5000case. aspx), and are “made of a special blend of thermoplastic fibers, abaca, and cellulosic fibers” (from vendor’s Web site) (Fig. 1, inset). Approximately 200 mg of dry biomass was loaded by hand onto a teabag, rolled up, and wrapped with 26 gauge copper wire to keep closed. Each teabag was numbered in pencil in order to


184

S.R. Decker et al.

Fig. 1. Processing of biomass in teabags in Soxhlet extractor using 95 % ethanol. Left-most set up shows extractor 2/3 full of ethanol. Inset shows close-up of teabags containing biomass.

track the samples during extraction and/or de-starching. De-starching (see below) was carried out prior to extraction in order to use the extraction process to generate consistently dry biomass. Extraction was carried out in multiple Soxhlet extractors (Fig. 1). Approximately 250 mL of 95 % ethanol was loaded into a round bottom flask nestled in an electric heating mantle. After washing with deionized water, twenty teabags were loaded into each extractor, which was connected to the round-bottom flask as well as 10 °C cooling water. The evaporated ethanol was condensed above the extractor and percolated down through the teabags, which filled with ethanol and drained through an automatic siphon. This cycle was repeated for 24 ± 2 h. After extraction, the teabags were removed from the Soxhlet and spread out in trays to dry in the hood overnight. After drying, samples were transferred into plastic 10 mL dispensing hoppers for loading into custom HTP 96-well Ni-Teflon-coated aluminum reactors (Fig. 2) (1, 2). The biomass plastic biomass hoppers were modified with aluminum funnels and conductive copper tape in order to ground the hoppers and eliminate errant biomass dispensing due to static electricity build up. A Symyx Powderium MTM powder dispensing robot was used to load 5.0 ± 0.3 mg of biomass into the reactor plates (2).


17

Reducing the Effect of Variable Starch Levels in Biomassâ&#x20AC;Ś

185

Fig. 2. Components of high throughput starch and recalcitrance assay. Top left: magnetic base plate. Bottom left: Tefloncoated aluminum HTP reactor plate. Small holes between wells are steam ports, 25 of which hold the magnets. Right: modified powder dispensing hopper. Copper tape and aluminum funnel dissipate static electricity by grounding to the robot while dispensing.

2.3. Amylases and Starch Testing 2.3.1. Starch Iodine Test

Iodine was used as a quick and easy way to detect the presence of starch in biomass. The iodine test solution consisted of Iodine (5 mg/mL) and KI (10 mg/mL) in deionized water. A sensitivity curve was made to test the limit of detection of the iodine starch test. Ten milliliters of water:biomass suspensions with solids loadings of 2.5, 1.5, 1.0, and 0.5 mg of biomass (BESC switchgrass) per 1.0 mL deionized water were boiled in tubes for 10 min, 30 min, 1 h, 2 h, and 3 h. Ten minutes of boiling was found to fully release the starch from the biomass, no gain in sensitivity was seen on boiling for longer. The iodine test was performed by combining 15.0 mg of biomass and 1.0 mL of deionized water in 1.5 mL centrifuge tubes. The tubes were then boiled in water for 10 min. The tubes were removed after boiling and 50.0 ÎźL of iodine was added to each tube. A positive result for starch content was recorded for any tubes presenting a color change from yellow to dark blue. This assay was


186

S.R. Decker et al.

used for rapid determination of starch presence in the initial BESC standard switchgrass material, as well as to evaluate the general efficacy of time and temperature variation of starch removal by amylases. 2.3.2. High Throughput Starch Assay

The HTP starch assay was adapted from a standard commercial starch test from Megazyme, Intl. (http://secure.megazyme.com/ Dynamic.aspx?control=CSViewProduct&categoryName=AssayKit s&productId=K-TSTA) and carried out in our custom HTP reactor plates (Fig. 2). The reactor plates are essentially 96-well aluminum microtiter plates of the standard SBS dimensions with a few variations (1, 2). The wells have a 7.0 mm diameter instead of the standard 6.8 mm and there is a 3.2 mm hole between every four wells as well as the outside pairs of wells. These holes are for steam penetration during thermochemical pretreatment; however, they also have a very useful side benefit. A series of 1/8 in. diameter neodymium cylinder magnets can be inserted into the holes and used to clamp a steel lid down on to the top of the plate, preventing evaporation for extended time. The magnetic clamps consisted of twenty-five 1/8 in. × 5/8 in. neodymium cylinder magnets press-fit into a 1/8 in. aluminum base plate which nests into the bottom of the reactor plate. The magnets fit into the steam ports between the wells of the aluminum reactor plates with the ends recessed just below the surface of the reactor plate. After sealing the plate with a standard microtiter plate adhesive seal, a magnetic stainless steel lid is placed on top of the plate. The pulling force of the magnets on the steel plate keeps pressure on the polypropylene seal, eliminating edge effects, and the resultant differential evaporation from the wells upon long-term incubation. After loading the biomass into the wells of the HTP reactors, each of the samples was wetted with 20 μL of 80 % ethanol and 300 μL of the amylase solution was added (300:1 α-amylase:amyloglucosidase by volume). The plates were sealed with a polypropylene adhesive seal (TiterTops, Diversified Biotech) augmented with custom magnetic clamps. The sealed plates were incubated overnight at 55 °C and 125 rpm in a Hi-Gro microtiter plate incubator. Glucose analysis after amylase digestion was conducted using the glucose oxidase assay described in Selig and coworkers (2). Briefly, the reactor plates are centrifuged to pellet the biomass and 20 μL of supernatant is diluted into 180 μL of deionized water, mixed, and then 20 μL of diluted sample is mixed with 180 μL of the glucose oxidase–peroxidase reagent mix (Megazyme, Intl.). The dilutions are carried out using standard 96-well microtiter plates with mixing for 3 min on a linear plate shaker. The final plate is sealed and incubated at 40 °C for 45 min and then read in a microtiter plate reader at 510 nm. Glucose concentrations are determined using a glucose standard curve included in the plate.


17

Reducing the Effect of Variable Starch Levels in Biomass…

187

2.3.3. Amylases

The initial de-starching of biomass was conducted using mixtures of two amylases; a thermostable α-amylase and an amyloglucosidase for TDF and starch assays, both from Megazyme International Ireland, Wicklow, Ireland. The α-amylase contained 3,000 U/mL (45 U/mg protein). The amyloglucosidase contained 3,260 U/ mL (soluble starch) or 200 U/mL (p-NP β-Maltoside). The initial de-starching was carried out using the same ratio of α-amylase to amyloglucosidase (300:1 by volume) as prescribed in the total starch kit. When scaling up to bulk de-starching of hundreds of samples, we opted to switch to bulk industrial amylases.

2.3.4. De-starching of Biomass

Time conditions for de-starching were determined using standard BESC switchgrass in teabags. Each teabag contained ~200 mg of the biomass and a total of 18 teabags were filled. The switchgrass teabags were placed in triplicate into five separate jars. One group of teabags was set aside for an ethanol extraction only control. An additional jar contained three teabags filled with 200 mg of NIST 8493 standard (Pinus radiata) milled to pass a 20-mesh screen as a negative control. The jars were filled with 20.0 mL of an amylase stock solution (3.0 mL α-amylase, 3.0 mL amyloglucosidase, 10.0 mL of 0.1 M citrate buffer, 90.0 mL de-ionized water). Four switchgrass jars were incubated at 55 °C and the fifth was placed at 30 °C along with the NIST pine samples. All were shaken at 125 rpm. Switchgrass jars were removed at 1, 3, 6, and 24 h. The negative control and switchgrass incubated at 30 °C were removed after 24 h. After removal, the teabags and jars were rinsed ten times with nanopure water. Iodine starch tests were conducted on the sample supernatants before the ethanol extraction. The solids were extracted and then assayed for starch content using the HTP starch assay. Testing of recalcitrance variability reduction by de-starching was carried out on a large (250) switchgrass sample set utilizing the Megazyme amylase enzymes. Samples (~200 mg dry weight) were hand-loaded into teabags (one sample per teabag) and digested 24 h at 55 °C in bulk. After washing with de-ionized water, the samples, still loaded in the teabags, were extracted, dried, and tested for starch content using the HTP starch assay. Each sample was analyzed in triplicate in HTP reactors, 5.0 mg per replicate. Each sample was also analyzed after extraction only (no amylase de-starching).

2.3.5. Commercial Enzyme Optimization

Commercial amylases are easily and inexpensively available from several enzyme manufacturers in standardized bulk packaging. This is important when screening large sample numbers over a long time period. We tested a two-component system; Spirizyme (glucoamylase) and Liquozyme (α-amylase), both from Novozymes A/S, Bagsvaerd, Denmark. As activity levels were not readily comparable between the Megazyme kit enzymes and the bulk amylases from Novozymes, we set up a series of mixing experiments to determine


188

S.R. Decker et al.

Table 1 Optimization of amylase loadings for switchgrass de-starching Enzyme mix

Megazyme mix (mL)

Relative concentration

M-1×

75

M-0.5×

37.5

0.5×

M-2×

150

M-5×

375

Enzyme mix N-1 N-2 N-3 N-4 N-5 N-6 N-7 N-8 N-9 N-10 N-11 N-12 No-E

Spirizyme (μL) 5 5 5 10 10 10 20 20 20 40 40 40 0

Liquozyme (μL) 25 50 100 50 100 200 100 200 400 200 400 800 0

Extracted Raw

Extracted only-no amylase Raw switchgrass

the minimal amount of amylase and the correct ratio of amylase components. Standard BESC switchgrass was loaded into HTP reactor plates and digested with an array of amylase levels and ratios. The mixtures were set up in 50 mM citrate buffer (pH 5.0) with four Spirizyme loadings, each with three Liquozyme loadings (Table 1). The BESC standard switchgrass loaded into teabags was used as substrate with three teabags per assay condition. Samples were incubated overnight at 50 °C. After digestion, the samples were washed with water, extracted, dried, and analyzed for remaining starch using the HTP amylase digestion assay. Three sets of controls were analyzed. One set was extracted but not de-starched, one set was de-starched in buffer only (no amylase), and a third set consisted of raw switchgrass. After 24 h at 55 °C and 125 rpm, glucose release was determined as described previously using the glucose oxidase assay. Though this system was not used in the baseline work described in this paper, we have implemented it as our standard de-starching protocol and have detailed the information here in order to provide the reader with an alternative system and to illustrate the importance of the activity ratio in this type of assay.


17 2.3.6. Effect of De-starching on Recalcitrance Screening

Reducing the Effect of Variable Starch Levels in Biomass…

189

Each of the 250 switchgrass variants was assayed for recalcitrance after pretreatment (180 °C, 40 min, in H2O) both before and after de-starching with the optimized amylase mixture. Both sets were extracted, dried, and robotically dispensed into the HTP reactors. Each sample was analyzed in triplicate, with 5.0 mg of biomass for each replicate. After loading the plates, 200 mL of deionized H2O was added to each well. The plates were sealed and pretreated at 180 °C for 40 min. After pretreatment, the samples were digested with 70 mg Spezyme CP per gram of biomass in 50 mM citrate buffer pH 5.0 for 3 days at 40 °C. Glucose was analyzed by GOPOD and the glucose release reported as glucose released as a percentage of the amount of biomass present.

3. Notes 3.1. Iodine Test Sensitivity

The results for the iodine sensitivity curve for switchgrass show that starch could be detected by the iodine test for a minimum solids loading of 1.5 mg biomass per 1.0 mL of water (Table 2). Boiling time does not appear to affect the sensitivity of the assay; however, a more sensitive, quantitative starch assay may detect lower concentrations and finer differences between samples. The iodine test was useful for quick evaluations on the presence of starch, but was not deemed quantitative.

3.2. De-starching Conditions

After initial treatment of the ethanol extracted BESC switchgrass with the Megazyme amylase mix, the iodine starch test showed negative results for every time period (at 55 °C) except zero hours (extraction only) (Table 3). At least within the limits of the starchiodine test, after one hour of treatment, at 55 °C, the starch in the biomass is digested (Table 3). The iodine test results also show negative for the samples that were treated for 24 h at 30 °C. This indicated several things, primarily that starch is indeed present in our switchgrass standard material, that ethanol extraction does not remove the starch, and that the amylase treatment can be used to remove starch from the switchgrass. One caveat to this is our inability to detect insoluble starch. In order for the iodine to react with starch, the starch is solubilized by boiling in water. Very crystalline or tightly bound starch may not become soluble and react with iodine, potentially resulting in false negatives. We also tested the solids from each sample using the HTP starch assay. After washing in deionized water, the samples were extracted, dried, and assayed for starch content by overnight digestion with amylase followed by glucose detection by GOPOD. Figure 3 shows the quantitative change in starch percentages at different time periods. The iodine test results suggested that after an hour of treatment, the starch was reduced to undetectable levels


190

S.R. Decker et al.

Table 2 Sensitivity of starch detection using the iodine assay mg/ml

Boil time

+/−

2.5

30 min

+

2.5

1h

+

2.5

2h

+

2.5

3h

+

1.5

30 min

+

1.5

1h

+

1.5

2h

+a

1.5

3h

+

1.0

30 min

+a

1.0

1h

1.0

2h

1.0

3h

0.5

30 min

0.5

1h

0.5

2h

0.5

3h

Switchgrass was boiled in water at the indicated times and loadings and assayed for starch by the addition of iodine a Sample presented a very slight blue color with iodine test

Table 3 Without amylase treatment (0 h) we get a positive result from iodine starch test Temperature (°C)

55

55

55

55

55

30

55

Time

0 h-extraction only

1h

3h

6h

24 h

24 h

24 h (NC)

Rep 1

+

Rep 2

+

Rep 3

+

For every other time period at 55 °C, the iodine test is negative. After a 24-h

period at 30 °C, the iodine test results are also negative. The negative control (NC) Pinus radiata was unaffected by treatment


17

Reducing the Effect of Variable Starch Levels in Biomass…

191

3

T=0

3.5

2

NC

30 oC

ON

0.5

T = 6 hr

1

T= 3 hr

1.5

T = 1 hr

% Starch in Biomass

2.5

0 −0.5

Fig. 3. When switchgrass is treated for 3 h or longer the starch percentage drops to under 0.5 %. ON = overnight, 30 °C = overnight incubation at 30 °C, NC = Negative Control (NIST pine).

in the biomass. When compared to the results of the quantitative starch assay, we can see that after an hour’s time there is still about 0.5–1.5 % of starch left in the samples. After 3 h of incubation at 55 °C, the percentage of starch in the samples was reduced to less than 0.5 %. Though the iodine starch test showed negative results for all treated samples, the more sensitive enzyme-based starch assay indicated that amylase treatment time should be six or more hours at a temperature of 55 °C in order to ensure as much starch is removed as is readily possible. All subsequent testing was carried out using overnight treatments. The iodine starch and the HTP starch assay clearly have their limits on starch detection. The HTP starch assay shows clearer results, but with large error bars. The error bars could be a result of there being more variance at a lower end of detection. They may also be due to incomplete mixing on the linear shaker. We have recently converted to mixing by tituration with the robot pipetting head, giving much better mixing with less variation between samples (data not shown). For a better sensitivity range, the dilution during the assay could be changed. As with the iodine test, severely recalcitrant or inaccessible starch would not be quantified by this method. While this was a concern, we did not have an easy or convenient way to test for this type of starch. Also, our primary concern was to remove starch variability in order to generate comparable results within a given


192

S.R. Decker et al. 3.00%

Starch Content (%)

2.50%

2.00%

1.50%

1.00%

0.50%

0.00% Raw

Extracted

No E

N-12

N-11

N-10

N-9

N-8

N-7

N-6

N-5

N-4

N-3

N-2

N-1

M-5X

M-2X

M-0.5X

M-1X

Enzyme Mix Fig. 4. Amylase loadings and ratio optimization for de-starching of switchgrass. See Table 1 for enzyme loadings and ratios.

sample set, not to quantifiably remove 100 % of the starch. To this end, we proceeded to develop the method in order to validate if this was possible. 3.3. Bulk Enzyme Testing

Testing of Spirizyme (glucoamylase) and Liquozyme (α-amylase) ratios and loadings demonstrated a clear effect on starch reduction at a variety of loadings and ratios (Fig. 4). Mixtures 11–14 (Table 1) yielded the highest starch removal. Lower enzyme loadings (especially those with lower Spirizyme, mixtures 5–10) were not as effective, though they were equivalent to the Megazyme mixtures 1–4 at those levels tested. This seems to indicate that the glucoamylase is the limiting activity and once above a certain level, increased enzyme loadings do not work any better (though they may be faster, since we did not do a time-course study). As the Novozymes enzymes are available in bulk, we opted to use them in the “standard” recalcitrance pipeline protocol for future screening, though the Megazyme mix was used to complete this study.

3.4. Starch Variability Reduction

A set of 250 switchgrass variants was analyzed for starch content before and after de-starching with the Megazyme amylase mix. The initial variability in the starch contents ranged from approximately 2 % to approximately 8 % starch after extraction-only, no de-starching (Fig. 5). After de-starching and extraction, the variability was reduced to between approximately 2 and 3 % starch, resulting in up to a 6 % decrease in starch content for some samples and therefore a change in up to 6 % glucose in the subsequent recalcitrance assay. Careful analysis of the data indicated that with


17

Reducing the Effect of Variable Starch Levels in Biomassâ&#x20AC;Ś

193

Fig. 5. Starch variability in biomass before and after amylase de-starching. Blue bars indicate samples that were extracted only. Red bars are the same samples after de-starching with amylase followed by ethanol extraction.

extraction only, the standard deviation of the glucose release was 1.14 % compared to 0.34 % in the sample samples after de-starching and ethanol extraction. This clearly shows the effect de-starching has on the measurable starch content in the biomass (Fig. 6). The 2 % residual starch content in the de-starched sample set indicated that there is some kind of interference in the assay. Further testing indicated that the GOPOD reagent reacted with the amylase mixture used, accounting for about half of the 2 % baseline starch content reading (data not shown). Although the remaining 1 % â&#x20AC;&#x153;starchâ&#x20AC;? is difficult to account for, the results are still valid in showing the effect of starch removal on the reproducibility and accuracy in determining the recalcitrance variation in a large sample set of biomass variants. Some of the residual glucose content may be due to GOPOD reacting with cellulose reducing ends or other compounds in the biomass. Without a well-defined starch-free control, it is difficult to evaluate this possibility. 3.5. Starch Reduction Effect on Recalcitrance Screening

Without removal prior to pretreatment and enzyme hydrolysis, the starch in these samples would contribute to the overall (pretreatment + enzyme hydrolysis) glucose release measurement, rendering a large variability in measured glucan. Since starch is much more easily hydrolyzed than cellulose, samples with large amounts of starch could be interpreted to contain low recalcitrance cellulose if the starch content was not considered. This effect is demonstrated in Fig. 4. For many samples, the amount of glucose released from the extracted-only sample is higher than the same sample after


194

S.R. Decker et al.

Fig. 6. Glucose release from switchgrass variants after hydrothermal pretreatment and cellulase digestion; with and without de-starching. Blue bars indicate Native samples (extracted-only). Red bars are the same samples after de-starching with amylase followed by extraction.

de-starching and extraction. In many cases, 15 % or more of the glucose released could be accounted for from the starch fraction. In an assay system designed to find plant variants that differ in recalcitrance by 10 % or so, this is a significant variable.

4. Conclusions It is clear from the study on the 250 switchgrass variants, that starch can be a major contributor to the measured “recalcitrance” of biomass when glucose release after hydrothermal pretreatment and enzyme digestion is the primary method of comparison. Significant variation in starch levels between switchgrass samples complicates these measurements. Rather than try to measure the starch and subtract from the “recalcitrance” glucose, we have shown that de-starching of biomass with amylases is an effective and easy way to minimize the effect of starch on these kinds of measurements. While highly quantitative starch analysis is difficult, standardizing the starch removal allows for direct comparison of recalcitrance between biomass variants when screening hundreds or thousands of samples.


17

Reducing the Effect of Variable Starch Levels in Biomass…

195

Acknowledgments This work was supported by the BioEnergy Science Center (BESC), which is a US Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science, Oak Ridge National Laboratory. References 1. Decker S, Brunecky R, Tucker MP, Himmel ME, Selig M (2009) High-throughput screening techniques for biomass conversion. BioEnergy Res 2:179–192 2. Selig MJ, Tucker MP, Sykes RW, Reichel KL, Brunecky R, Himmel ME, Davis MF, Decker SR (2010) Lignocellulose recalcitrance screening by integrated high-throughput hydrothermal pretreatment and enzymatic saccharification. Ind Biotechnol 6:104–111

3. Smith SM, Fulton DC, Chia T, Thorneycroft D, Chapple A, Dunstan H, Hylton C, Zeeman SC, Smith AM (2004) Diurnal changes in the transcriptome encoding enzymes of starch metabolism provide evidence for both transcriptional and posttranscriptional regulation of starch metabolism in Arabidopsis leaves. Plant Physiol 136:2687–2699


Chapter 18 Analysis of Transgenic Glycoside Hydrolases Expressed in Plants: T. reesei CBH I and A. cellulolyticus EI Roman Brunecky, John O. Baker, Hui Wei, Larry E. Taylor, Michael E. Himmel, and Stephen R. Decker Abstract Plant cell walls are composed of three basic structural biomolecules: cellulose, hemicellulose, and lignin with cellulose being the most abundant biopolymer on earth. Cellulose is composed of cellodextrins, which are linear polymers of glucose, and considered to be microcrystalline in structure. The conversion of cellulose to free glucose is one of the primary steps in the fermentative conversion of biomass to fuels and chemicals. However, the crystalline nature of this complex, noncovalent structure is highly resistant to enzymatic hydrolysis. Thus, the substantial cost currently associated with biomass saccharification primarily represents the cost of biomass degrading enzymes [1]. Despite the fact that the microbial cellulose hydrolytic “machinery” for the recycling of carbon from plant biomass already exists in nature, the natural enzymatic degradation of plant material is typically a slow and complex process [2]. Thus, if commercial biofuels production is to become a reality, it must be more cost-effective. One method proposed for achieving this objective is to express all or some of the requisite cellulolytic enzymes in planta, thus reducing both enzyme and thermochemical pretreatment costs [3, 4]. Key words: Glycoside hydrolase, In planta expression, Trichoderma reesei, Cellobiohydrolase I, Acidothermus cellulolyticus, Endoglucanase EI

1. Introduction Cellulose is largely degraded by the synergistic action of two general types of cellulases; enzymes that cleave cellulose internally are referred to as endo-1,4-β-D-glucanases (E.C. 3.2.1.4), that serve to provide new reducing and nonreducing chain termini on which more processive enzymes, such as the exo-1,4-β-D-glucanases (cellobiohydrolase, CBH; E.C. 3.2.1.91), can operate. The products of the exoglucanase reactions are typically glucose and/or cellobiose, but other short-chain cello-oligomers are also produced.

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_18, © Springer Science+Business Media, LLC 2012

197


198

R. Brunecky et al.

To date, many cellulase genes have been cloned and sequenced from a wide variety of bacteria, fungi, and plants. Subsequently, these genes have been transferred and expressed in model plants (i.e., Arabidopsis thaliana and Nicotiana tabacum), as well as biomass crops (e.g. Zea mays) (3). Therefore, it is an important prerequisite to be able to assess whether or not the selected plant expressed transgenic enzymes are in fact present in the plant materials and if they are present, whether or not they retain their hydrolytic activities. In this chapter, we describe methods to identify and verify the activity of two important classes of enzymes used for biomass deconstruction: the Acidothermus cellulolyticus endoglucanase I (EI) (Cel5A) and the Trichoderma reesei cellobiohydrolase I (CBH I) (Cel7A). The A. cellulolyticus EI is a well-characterized glycoside hydrolase family 5 endocellulase known to be thermally tolerant (optimal temperature 80 °C) and generally very stable over broad ranges of pH with an optimum of pH 5 (5). The A. cellulolyticus EI has been expressed in both Nicotiana tabacum and Zea mays, where its expression was targeted to the cell wall under a constitutive promoter (6â&#x20AC;&#x201C;8). T. reesei CBH I is a major component of most commercial cellulase preparations. It degrades cellulose processively from nonreducing free chain ends. CBH I shows optimal and substantial activity at temperatures up to 50 °C. Expression of CBH I has been reported in corn leaves and kernels (9).

2. Materials and Methods 2.1. Biomass

Several biomass samples and fractions thereof are discussed in this study. Briefly, they include corn (seeds) transformed with the T. reesei cbh1 gene, corn (stover) transformed with the A. cellulolyticus e1 gene (8), and tobacco transformed with the A. cellulolyticus e1 gene (6). The EI-maize biomass was used as a model lignocellulosic substrate, as corn stover (the nongrain parts of the harvested plant) is one of the leading biofuel feedstock candidates. Ransom and coworkers (10) produced EIcd transformed maize in which EI expression was driven by the 35S CaMV promoter and targeted to the apoplast. EI corn plants were grown to maturity under greenhouse conditions and then allowed to dry in the greenhouse before being harvested (7). Tobacco was transformed successfully according to standard methods using Agrobacterium tumefaciens. After identification of stable lines, the biomass was grown under controlled greenhouse conditions. Individual plants were screened for EI activity. Initial studies focused on two transgenic lines designated EI-1 and EI-7, which showed high and low endoglucanase activity, respectively (7). Wild type (WT) maize kernels or


18

Analysis of Transgenic Glycoside Hydrolases Expressed in Plants…

199

transgenic (TG) maize kernels overexpressing CBH I were created as described by Hood and coworkers, and processed and made into a fine cornmeal by grinding the dried kernels (11). 2.2. Milling

All biomass samples, including both leaves and stalks, were air-dried to a moisture content of approximately 10 %, and milled using a Wiley mill (Thomas Scientific, Swedesboro, NJ, USA) to pass a either an 80- or 20-mesh screen. Corn kernels were milled to 20 mesh using a Wiley mill (Thomas Scientific, Swedesboro, NJ, USA).

2.3. Extraction of Glycoside Hydrolases

Several methods were employed to extract the enzymes from the ground biomass depending on the desired assay. Briefly, aqueous buffers may be a viable option for extracting small amounts of protein from biomass samples for highly sensitive analyses, such as 4-methylumbelliferyl-β-D-cellobiose. Alternatively for detection using gel electrophoresis and/or subsequent western blotting, typically boiling for 5–10 min with SDS or LDS-containing sample buffer is an effective extraction technique, though it denatures the enzyme. Finally, extraction using a combination of ethylene glycol and detergents can be used to remove glycoside hydrolases containing carbohydrate binding modules from biomass more effectively than with simple buffers by disrupting the CBM/biomass interaction; the recipe for this extraction is as follows: 50 % ethylene glycol, 0.5 % SDS, LDS protein loading buffer, 2 % Brij 30, 2 % Triton X-100, 2 % Tween 80, 2 % non-Ident P-40.

2.4. Identification of Glycoside Hydrolases

Western blotting is among the most commonly used techniques to detect the levels of target proteins when specific antibodies are available (12). Thus, we use this technique (with the respective antibodies) to determine the expression of EI and CBH I in the examined sample.

2.4.1. EI Identification Using Western Blot

To identify the transgenic EI construct we used direct extraction of the plant material into LDS loading buffer followed by western blotting. Briefly, 5 mg milled biomass was added directly to 50 μL NuPAGE LDS sample buffer and boiled for 10 min. After centrifuging for 5 min at 14,000 × g, 20 μL of the supernatant was loaded onto a NuPAGE 4–12 % gradient gel run in MOPS (3-(N-morpholino)-propanesulfonic acid) running buffer (Invitrogen, Carlsbad, CA). The gel was then electroblotted onto polyvinylidene fluoride (PVDF) membrane using the Invitrogen XCell II blot module blotting apparatus. The blot was analyzed for the presence of EIcd using mouse anti-EI monoclonal primary antibody and goat anti-mouse secondary antibody with the Invitrogen WesternBreeze immunodetection kit. EIcd purified from expression in Streptomyces lividans was used as a qualitative control for comparison with a dilution series of the EI maize and tobacco extracts.


200

R. Brunecky et al.

2.4.2. CBH I Identification Using Western Blot

A western blot of CBH I enzymes obtained from various sources was performed to illustrate the importance of using more than one antibody with different epitopes to identify CBH I or other multidomain enzymes expressed in plants. Samples (15 μL) of the extracts were loaded onto PAGE/SDS gels then transferred to a PVDF membrane as for EI. Immunodetection was carried out as for EI with either a mouse monoclonal against the CBM/linker domain of CBH I, or a rabbit polyclonal antibody specific to the catalytic domain (CD) portion of CBH I.

2.5. Testing for Transgenic Enzyme Activity

The low EI detection levels indicated by the direct extraction/ western blot detection suggested the need for an enzymatic assay with a high sensitivity such as the MUC assay. Other substrates, such as AZCL-hydroxyethyl cellulose and p-nitrophenyl-β-Dcellobioside did not exhibit the required sensitivity for quantitation. Additionally, the yellow product from the pNPC assay was lost in the high background color generated by tobacco slurries. The 4-methylumbelliferyl-β-D-cellobiose (MUC) assay takes advantage of the ability of endoglucanases to break down the fluorogenic substrate (MUC) to produce the fluorophore, 4-methylumbelliferone (MU) (13). On the basis of western blot estimates of enzyme concentration and the sensitivity of the assay, the MUC assay was chosen to determine the enzymatic activities of EI in the examined samples.

2.5.1. Fluorogenic Assays for EI Activity

The activity of EI in milled corn stover and tobacco can be estimated on the basis of MUC activity. In a 16 × 100-mm test tube, we added 10 mg of 80-mesh biomass, 200 μL of 1.0 M sodium acetate, pH 5.0, and 1,600 μL of H2O. After heating the sample to either 60 °C or 84 °C for 5 min, 200 μL of 2.0 mg/mL MUC substrate was added to the reaction mix. The 60 °C assay condition yielded false positives in the untransformed control stover. Presumably, this is due to a native corn β-glycosidase present in the stover. Heating to 84 °C inhibited this activity in wild-type stover. Each sample was analyzed in triplicate. Samples of 100 μL were taken at 0.0, 0.5, 1.0, and 2.0 h (for stover) or every minute for 15 min (for tobacco), quenched with 200 μL of 1.0 M Na2CO3 and fluorescence was measured at 360-nm excitation/465-nm emission in a Tecan GENios microplate reader (MTX Lab Systems, Vienna, VA, USA). EIcd levels in the stover were estimated by comparison of activity to purified S. lividans-expressed EIcd activity under the same conditions (Fig. 2).

2.5.2. Assay of Transgenic Cornmeal Extracted CBH I Against Microcrystalline Cellulose: Extraction of CBH I from Transgenic Cornmeal

Six aliquots of ground transgenic corn meal, averaging 4.8 g each for a total of 28.8 g of ground corn kernels, were each extracted for 80 min (room temperature with constant mixing by inversion at 20 revolutions per minute in VWR-Brand 50-mL centrifuge tubes) with 37.0 mL each of 6 mM acetate, pH 4.5, containing 0.02 %


18

Analysis of Transgenic Glycoside Hydrolases Expressed in Plants…

201

sodium azide (7.71 mL buffer per g cornmeal). After pellets were centrifuged, supernatants were decanted and filtered using Nalgene filter units (0.2-μm polyethersulfone membrane, Cat # 568-0020) to remove remaining fine particulates. An 8.0 mL aliquot of the total of 182 mL of clarified solution was then exchanged into fresh acetate buffer using two Vivaspin 4 (Vivascience, 10,000 MWCO) centrifugal ultrafiltration units (4.0 mL each). After three concentration steps, the retentate was rinsed out into a total volume of 8.0 mL (returning retained species to their original concentrations) in pH 5.0, 20 mM acetate buffer. The ultrafiltered sample, from which possible inhibitors and other small molecules that might interfere with HPLC analysis have been effectively removed (dilution factor of 1,400), was then assayed for activity against Sigmacell-20. 2.5.3. Sigmacell Enzyme Assay

Assays intended to measure the actual cellulolytic capabilities of the transgenic CBH I extract utilized microcrystalline cellulose as substrate (Sigmacell, Type 20, Sigma-Aldrich Corp, St. Louis, MO). As shown in Table 2, the buffer extract was assayed at two different loadings of extract, acting against Sigmacell-20 (20.83 mg/mL) at 40 °C in 20 mM acetate buffer, pH 5.0, to which sodium azide had been added at 0.02 % (w/v) to retard microbial growth, and which also contained “protease-free” bovine serum albumin (Product No. A3294, Sigma-Aldrich Corp, St. Louis, MO) at 0.16 mg/mL to reduce the possibility of target proteins being adsorbed on glass walls of the reaction vessels. Assay digestions were carried out in a 1.2-mL reaction volume in crimp-sealed HPLC vials (nominally 2.0-mL volume) with constant mixing by inversion at 10 revolutions/min. After a 72-h digestion, all vials were centrifuged at room temperature and 0.5 mL of supernatant from each was diluted with 0.5 mL of acetate buffer in a termination vial, which was sealed, mixed by vortexing, and then immersed in a boiling water bath for 10 min to terminate enzyme activity. After cooling and vortexing, the contents of these termination vials were filtered into HPLC vials for determination of released soluble sugars. Sugar analysis was carried out by chromatography on a BioRad HPX-87H column operated at 65 °C with 0.01N H2SO4 (0.6 mL/min) as eluent in an Agilent model 1100 chromatograph with refractive-index detection.

2.5.4. Assay for Total Cellulase Activity in Cornmeal (Tightly Bound to Biomass Tissues As Well As That Readily Extractable)

For “in situ” assays of cellulase activity from cornmeal, the “enzyme” was un-extracted transgenic cornmeal, weighed directly into HPLC-vial reaction vessels at 100 mg/vial, where it was mixed with 25.0 mg of Sigmacell, type 20 as substrate, the Sigmacell being delivered as a calibrated slurry. Control (enzyme-only) digestion vials contained only the transgenic biomass, with no added microcrystalline cellulose. The mixtures in all vials were adjusted to a total volume of 1.0 mL, in the same pH 5.0, 20 mM acetate buffer used for the assay of the extracts.


202

R. Brunecky et al.

All vials were incubated for 168 h (7 day) at 40 °C, with continuous mixing by inversion at 20 revolutions/min. At the end of the incubation period, all vials were centrifuged at room temperature and 0.5 mL of supernatant from each was diluted with 0.5 mL of acetate buffer in a termination vial, which was sealed, mixed by vortexing, and then immersed in a boiling water bath for 10 min to terminate enzyme activity. After cooling and vortexing, the contents of these termination vials were filtered into HPLC vials for soluble-sugar analysis. Sugar analysis was carried out by chromatography on a BioRad HPX-87H column operated at 65 °C with 0.02N H2SO4 (0.6 mL/min) as eluent in an Agilent model 1100 chromatograph with refractive-index detection.

3. Results 3.1. EI Identification

Western blotting clearly identifies EIcd as being present in both the tobacco and EI-1 corn-stover extracts. The EI-7 sample was very weakly detectable by western (data not visible) but later showed activity against fluorogenic substrate (Fig. 1).

3.2. CBH I Identification

The results of the western blotting (Fig. 2) are highly dependent on the primary antibody used. In the case of a monoclonal antibody that recognizes the CBM/linker portion of CBH I, detection was strong for several positive controls (i.e., CBH I purified from T. reesei broth or commercial cellulase, rCBH I expressed in

Fig. 1. EI expression in planta. Western blot analysis of wild-type (WT) and EI-transformed tobacco and corn stover. (Reproduced with permission ref. 7).


18

Analysis of Transgenic Glycoside Hydrolases Expressed in Plants…

203

Fig. 2. Detection of CBH I using either mouse monoclonal or rabbit polyclonal antibodies.

Aspergillus awamori, or insect cells) but nonexistent for the truncated CBH Icd enzyme. A polyclonal antibody was used in the second western blot experiment. The polyclonal was raised in rabbit against purified CBH I catalytic domain. The polyclonal antiCBH Icd exhibited strong detection of all of the CBH I variants. 3.3. EI MUC Assay

The MUC assay demonstrated that active EI enzyme was present in both transgenic maize and in tobacco samples (Fig. 3). It also showed native plant enzyme activity from corn stover when the samples were incubated at 60 °C. As EI is thermal tolerant, samples were preincubated at 84 °C and then assayed. This higher temperature apparently denatured the native corn enzyme, allowing EI activity only to be measured. The MUC assay also allows us to estimate the EIcd levels based on activity compared to the control EI. In the case of EI-1, this is approximately 0.3 ng/mg biomass. Although the EI-7 sample was not detected by the antibody method, the MUC assay estimated the EI-7 level to be about tenfold less than that of EI-1 approximately 0.03 ng/mg biomass for EI-7 (Fig. 3). Wild-type stover showed no presence of EI by either method. For tobacco, the EIcd expression level was approximately 1,000-fold higher than that in the EI-1 transgenic stover.

3.4. Sigmacell Enzyme Assay Results Activity in Extracts

Tables 1 and 2 describes the release of soluble sugars from Sigmacell-20 by activity found in the acetate extracts of CBHtransgenic cornmeal. The data in Table 1 are those used to construct a standard curve (shown in Fig. 4) for the activity of wild-type T. reesei CBH I against Sigmacell-20 under the same conditions. In assay B1 in Table 2 was loaded with 0.8 mL of the diafiltered buffer-extract and 0.352 mg/mL of total soluble anhydro-sugars were released over the course of a 72-h digestion. This release of soluble


204

R. Brunecky et al.

Fig. 3. EI activity from plant extracts. (a) WT and EI corn stover at 60 °C versus WT and EI corn stover (b) tobacco at 84 °C were extracted and assayed for activity on 4-methylumbelliferyl-β-D-cellobiose (MUC) substrate and (c) WT and E1 tobacco compared at 84 oC (Reproduced with permission ref. 7).

sugars may be the result of action of either intact CBH I, or CBH I catalytic domain, or a mixture of the two on the Sigmacell substrate. We can estimate the amount of apparent CBH I active site present in the sample by a linear interpolation between the “zero CBH I”


18

Analysis of Transgenic Glycoside Hydrolases Expressed in Plants…

205

Table 1 Standard curve of wild-type Trichoderma reesei CBH I activity against Sigmacell CBHI (mg/mL)

AH-CB (mg/mL)

AH-Glc (mg/mL)

Total AH-sugar (mg/mL)

% Cellobiose in product

% Conversion of substrate

A1

0.417

2.217

0.116

2.451

95.04

11.77

A2

0.156

1.271

0.076

1.392

94.35

6.68

A3

0.063

0.715

0.054

0.789

93.01

3.79

A4

0.021

0.390

0.034

0.424

91.97

2.04

A5

0.000

0.000

0.000

0.000

0.00

0.00

Table 2 Activity of acetate-extract of cornmeal against Sigmacell Extract (fractional AH-CB volume in assay) (mg/mL)

AH-Glc (mg/mL)

Total AH-sugar (mg/mL)

% Cellobiose in product

% Conversion of substrate

B1

0.667

0.015

0.337

0.352

4.34

1.69

B2

0.333

0.017

0.246

0.262

6.33

1.26

and the “0.0208-mg/mL CBH I” points. From this, we might conclude that this level of sugar release corresponds to an intact CBH I concentration of 0.017 mg/mL in assay B1 and a concentration of 0.0253 mg/mL in the diafiltered extract used to set up the assay. The perpendicular red lines in Fig. 4 illustrate the method of this estimate. A similar estimation using the sugar-release value of 0.0262 mg/mL for assay B2 yields the higher estimate of 0.0388 mg/mL for the CBH I concentration in the extract. This B1 value, based on sugar release closer to that of the nonzero “standard-curve” CBH I loading of 0.0208 mg/mL, is probably the more meaningful value. Because the response curve is almost certainly nonlinear (significantly concave-downward) in this region, both estimates are likely to be; however, overestimates. Even if all of the sugar-releasing activity seen in assay B1 (Table 2) were ascribable to intact CBH I, the total amount of CBH I available in the extract of 28 g of “CBH I corn” would be relatively low, i.e., 4.4 mg or ~0.015 % by weight.


206

R. Brunecky et al.

Fig. 4. Digestion of Sigmacell 20 (20.83 mg/mL) by purified Trichoderma reesei CBHI at pH 5.0, 40 °C, at 72 h.

Table 3 Release of sugar upon incubation of microcrystalline cellulose (Sigmacell-20) with transgenic (CBH I) cornmeal Sample

Peak area AH-Glc (mg/mL) Average Std. Dev. Difference (mg/mL)

G-CBH I meal + cellulose

12,500

12.30

TG-CBH I meal + cellulose

12,838

12.70

12.50

0.24

2.85 TG-CBH I meal alone

9,925

9.80

TG-CBH I meal alone

9,684

9.50

3.5. Results of the Assay for Cellulase Activity Tightly Bound to Biomass Tissues

9.65

0.17

Table 3 compares the concentrations of soluble sugar detected in the liquid phases of acetate-buffer co-slurries of transgenic (CBH I) corn-kernel material and Sigmacell-20, after a 168-h incubation at 40 °C with constant mixing, with the HPLC-measured solublesugar levels in identically treated slurries of the ground corn-kernel material alone. The soluble sugar concentrations are reported as anhydro-glucose (AH-Glc), in order to facilitate comparison with the starting concentration (25 mg/mL) of Sigmacell-20 in the mixture. All soluble sugar was in fact detected as glucose, with no detectable amounts of cellobiose or other sugars. The Sigmacellonly controls showed no detectable soluble sugars at all and are not included in the table.


18

Analysis of Transgenic Glycoside Hydrolases Expressed in Plants…

207

4. Discussion 4.1. EI Identification and MUC Activity 4.1.1. Identification

4.1.2. Activity

A. cellulolyticus EI was readily extracted from ground biomass samples using LDS sample buffer as described and then subjected to westernblot analysis. This is the simplest method reported so far to extract and identify recombinant proteins, such as EI, in biomass samples. Extraction with buffer or buffer followed by ethylene glycol gave low protein yield (data not shown). This indicates that the EI expressed in maize was either tightly bound to the cell wall (presumably the cellulose fraction) or was localized deeply enough in the cell wall matrix to make extraction difficult. While extraction with detergent (boiling LDS-sample buffer) yielded measurable protein, activity is abolished using this method from general protein denaturation. This technique is useful for qualitative evaluation of expression levels; however, quantitative determination of expressed protein can only be achieved by comparison with known levels of protein measured by band intensity on a western blot (and possibly by ELISA) or direct mass spectral fingerprinting. This estimation assumes that both proteins (purified control and transgenic) are equally antigenic, which may not be true given that they are expressed in different host systems. It also assumes that all of the transgenic protein is released from the biomass and moves into the gel, which may not be the case. To test for active EI enzyme or other highly thermostable recombinant enzymes, we recommend the use of fluorogenic substrates. As EI was not readily extracted from ground biomass in an active form, assays were carried out using milled biomass as the “enzyme.” Heating to 84 °C was required to denature native plant glycosyl hydrolases that interfered with the MUC assay conducted at 60 °C. This same technique could presumably also be used with cellulose as substrate. Incubation of EI-stover at 84 °C with AZCLhydroxyethyl cellulose as a substrate yielded enzymatically released dye (data not shown). Although this hydrolysis result was not quantified, the results suggest that incubation with cellulose would yield new reducing ends. However, because EI is primarily an endo-acting cellulase, the production of detectable, soluble sugars might be low. Inclusion of a β-glucosidase and/or exocellulase would enhance this activity evaluation. In addition to verifying that the plant expressed EI was in fact active, we can also estimate the amount of EI present in the plant material using a known amount of control WT purified EI. It is possible to estimate the amount of EI present in the biomass samples by comparing the MUC digestions to the control during the linear phase of the digestion curve. However, as can be seen in Fig. 3a, native plant glycosyl hydrolases can interfere with the MUC assay at lower incubation temperatures. This possibility must be carefully considered if one is testing for enzymes that do not have high thermostability.


208

R. Brunecky et al.

4.2. CBH I Identification and Activity

Extraction and western blotting of CBH I from various sources illustrates a critical point: that when examining CBH I enzymes, or multidomain enzymes in general, one must consider the possibility of truncation or changes in the antigenicity of CBH I (Fig. 2). Given the possibility that truncation of CBH I can occur when expressed in plants, one must be careful when selecting antibodies to the protein of interest; that is, it must be known what the actual epitope of the antibody is targeted to. In our case, the monoclonal mouse antibody failed to detect certain truncated variants of CBH I, because it binds only to the CBM/linker region of this protein. However, by using the polyclonal rabbit antibody, we ascertained that at least the catalytic domain of CBH I was present in all cases.

4.2.1. Sigmacell Enzyme Assay

One consideration when interpreting the data on digestion of Sigmacell with buffer-extracted rCBH I (Table 3) which in this case was identified using mass spectrometry techniques (data not shown) is that these estimates arise from the low overall extents of conversion shown by the extract assay. It is well known that in nearly all microcrystalline celluloses, a significant portion of the material is present as amorphous content (14). This material is less ordered than the more crystalline material, and is much more readily hydrolyzed by either enzymes or chemical catalysts. And at <2 % conversion, the extract activity is likely to still be working on this more digestible fraction of the material. The shape of the response curve of purified T. reesei CBH I in Fig. 4 would support this view. Below 4 % or so conversion, (approximately 0.087 mg/mL anhydro-sugar released) the steeper slope of the curve indicates that the enzyme is working on much more readily hydrolysable material (i.e., amorphous cellulose) compared to later digestion where the rate is much shallower. Presumably, this later hydrolysis is on more crystalline material. Hydrolysis of this amorphous material does not require an intact CBH I; CBH I catalytic domain alone has significant activity against such material. This suggests that the low extent of conversion of Sigmacell by extracted rCBH I is probably due to the activity of either catalytic domain only CBH I or CBH I with a truncated or modified CBM that does not readily bind to crystalline cellulose. Also present in (Table 3), we see a clear difference between the compositions of the sugar mixtures released from Sigmacell-20 by T. reesei CBH I (chromatographically purified from commercial cellulase) and the mixtures released by the extract of transgenic (CBH I) cornmeal, in that the purified CBH I releases predominantly cellobiose, whereas the extract of maize kernel material releases predominantly glucose. The low percentage of cellobiose as product from this reaction is not typical of CBH I activity. Presumably, the cellobiose produced by the extracted CBH I is being hydrolyzed by a native β-glucosidase-like enzyme in the biomass. This conclusion is supported by results from the MUC assays applied to EI-stover, which clearly showed MUC activity in the wild-type plant material. As the activity of the extracted rCBH


18

Analysis of Transgenic Glycoside Hydrolases Expressed in Plants…

209

I is low, it would not take much glycosidase activity to convert the majority of cellobiose to glucose. The fact that the higher loading of the extract (B1) shows a lower proportion of cellobiose is in agreement with this explanation. For an assay to give strong indication of the presence of the complete, active enzyme, it is necessary for a large enough fraction of the crystalline cellulose be removed. This condition must be met before one can be confident that the enzyme has removed the “easy,” presumably less-ordered or amorphous cellulose, and is working its way through the much more refractory substrate that is presumed to be crystalline. 4.2.2. Assay for CBH I Activity that is Tightly Bound to Biomass

In the previous Sigmacell digestion assay, we found relatively small amounts of cellulose-depolymerizing activity in buffer extracts of several transgenic corn-kernel tissues expected to be expressing T. reesei CBH I. Given the low overall extent of conversion of these digestions, it is possible that much of the activity that we observed in the extracts might be attributable to CBH I molecules having functional catalytic domains, but lacking functional cellulosebinding modules; these CBH Icd molecules were detected by massspectrometry (data not shown). This non-functional CBM phenotype could be explained by truncated protein lacking a complete CBM or by altered glycosylation of the linker peptide and/or CBM which could interfere with optimal binding of the CBM to cellulose. Based on this idea, significant quantities of complete, cellulolytically competent CBH I molecules, possessing not only active catalytic domains, but also linker regions and CBMs properly attached, might be bound through the CBMs to the cellulose component of the corn-kernel cell walls. These putative molecules are contrasted with the catalytic-domain-only molecules actually found, which could be extracted rather quickly using aqueous buffers. Thus, the holoenzymes (those with CBMs) may be extracted at a much slower rate, because of the presumed tighter binding of the intact CBH I. In addition, intact CBH I would be expected to have an equilibrium between being free in solution and bound, possibly very heavily in favor of bound molecules, whereas for CBH Icd, the opposite behavior is expected. Simple buffer extractions, therefore, wash out only a very small fraction of the whole CBH I enzyme in each washing, but should remove most of the truncated CBH Icd. This in situ assay is designed to take advantage of a possible two-step transfer of CBH I. First, corn-kernel-bound CBH I dissociates from the kernel cell-wall cellulose, into the solution. Second, solution CBH I diffuses to, and binds to, microcrystalline cellulose that has been added to the slurry of milled corn-kernel material and is intimately mixed with it. In this scheme, even though production of free (solution) CBH I may be slow, and although there is not much of the free species present at any one time, complete, active CBH I may be funneled through the free population to accumulate on another CBH I-sink such as clean


210

R. Brunecky et al.

cellulose (Sigmacell 20), provided CBH I has adequate affinity for the second sink and the sink is in significant excess to the corn cellulose. This process is seen as an equilibrium process, not a one-way transfer, but if the process is given sufficient time, it can be expected that a significant time-averaged population of CBH I molecules will appear on the coincubated microcrystalline cellulose, and measurable sugar-release from the microcrystalline cellulose will occur as a result. Despite the necessity of subtracting out a large blank of glucose contributed by the (previously unextracted) corn-meal added to the coincubation mixtures; we see that the incubations containing Sigmacell-20 in addition to the cornmeal show sugarrelease greater than that in the cornmeal-alone controls (Table 3). With the reasonably good reproducibility shown by the duplicates, it appears that the difference is statistically significant. Whether the “extra” sugar release can be ascribed to the action of intact CBH I on the Sigmacell-20 is still under study. While the product of the coincubation does not show the characteristic digestion profile of CBH I acting alone on cellulose; in that the detected sugar is entirely glucose, rather than the expected mixture: predominantly cellobiose containing a small fraction of glucose. However, as in the case of the extract assays described above, the absence of cellobiose from the product mixture is not a serious objection to ascription of the observed release to a rCBH I species, in that there is generally even more endogenous cellobiase retained on the cornmeal pellets than is extracted in simple buffer-extractions such as these. The presumed extent of conversion of the Sigmacell in this experiment (approximately 10.1 % based on 2.8 mg/mL anhydro-glucose in the calculated 0.895-mL liquid-phase of the 1.0-g assay mixture) is considerably higher than that seen in the assays of extract. The example results described here for the cellulose-conversion assays unfortunately did not give definitive answers to the question of whether or not the putative rCBH I responsible for the sugarrelease is actually cellulolytically competent (i.e., capable of attacking crystalline cellulose), or if the activity belongs to holoenzyme, isolated catalytic domain, or a mixture of the two. However, demonstrating “cellulolytic competence” would require larger extents of conversion than the higher of the two extents seen here. The assay designs themselves could, in fact, be greatly improved. The “solidto-solid extraction” assay in particular needs a corresponding control utilizing WT control corn-meal, and would benefit from a procedure in which the cornmeal solids were preextracted to remove both the readily extractable cellulase and other endogenous MUCase/MULase activities (with these activities in the extractions/washes carefully accounted for, so that the results of the “pellet” assays would reflect only the more tightly bound activities corresponding to intact rCBH I). This pre-extraction would also avoid the necessity of making a large correction for extractable


18

Analysis of Transgenic Glycoside Hydrolases Expressed in Plants…

211

sugars and other substances potentially by interfering with the HPLC analysis, as was necessary for the results in Table 3. The procedures and results presented here are simply intended to provide a starting point for important considerations of the questions and concerns encountered in such work.

Acknowledgments This work was supported by the DOE Office of Science, Office of the Biological and Environmental Research, through the BioEnergy Science Center (BESC), a DOE Bioenergy Research Center. References 1. Himmel ME, Ding SY, Johnson DK, Adney WS, Nimlos MR, Brady JW, Foust TD (2007) Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science 315:804–807 2. Wei H, Xu Q, Taylor L, Baker J, Tucker M, Ding S-Y (2009) Natural paradigms of plant cell wall degradation, Curr Opin Biotechnol 20:330–338 3. Taylor LE, Dai Z, Decker SR, Brunecky R, Adney WS, Ding S-Y, Himmel ME (2008) Heterologous expression of glycosyl hydrolases in planta: a new departure for biofuels, Trends Biotechnol 26:413–424 4. Sticklen MB (2008) Plant genetic engineering for biofuel production: towards affordable cellulosic ethanol, Nat Rev Genet 9:433–443 5. Baker JO, Adney WS, Nieves RA, Thomas SR, Wilson DB, Himmel ME (1994) A new thermostable endoglucanase, Acidothermus cellulolyticus E1, Appl Biochem Biotechnol 45:245–256 6. Ziegelhoffer T, Raasch JA, Austin-Phillips S (2001) Dramatic effects of truncation and subcellular targeting on the accumulation of recombinant microbial cellulase in tobacco, Mol Breed 8:147–158 7. Brunecky R, Selig MJ, Vinzant TB, Himmel ME, Lee D, Blaylock MJ, Decker SR (2011) In planta expression of A. cellulolyticus Cel5A endocellulase reduces cell wall recalcitrance in tobacco and maize, Biotechnol Biofuels 4:1 8. Ransom C, Balan V, Biswas G, Dale B, Crockett E, Sticklen M (2007) Heterologous Acidothermus cellulolyticus 1, 4- -endoglucanase E1 produced within the corn biomass converts corn stover into

9.

10.

11.

12.

13.

14.

glucose, Appl Biochem Biotechnol 137–140: 207–219 Harrison MD, Geijskes J, Coleman HD, Shand K, Kinkema M, Palupe A, Hassall R, Sainz M, Lloyd R, Miles S, Dale JL (2011) Accumulation of recombinant cellobiohydrolase and endoglucanase in the leaves of mature transgenic sugar cane, Plant Biotechnol J 9(8):884–896 Ransom C, Balan V, Biswas G, Dale B, Crockett E, Sticklen M (2007) Heterologous Acidothermus cellulolyticus 1,4-beta-endoglucanase E1 produced within the corn biomass converts corn stover into glucose, Appl Biochem Biotechnol 137:207–219 Hood EE, Love R, Lane J, Bray J, Clough R, Pappu K, Drees C, Hood KR, Yoon S, Ahmad A, Howard JA (2007) Subcellular targeting is a key condition for high-level accumulation of cellulase protein in transgenic maize seed, Plant Biotechnol J 5:709–719 Burnette WN (1981) “Western blotting”: electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A, Anal Biochem 112:195–203 Heptinstall J, Stewart JC, Seras M (1986) Fluorimetric estimation of exo-cellobiohydrolase and [beta]-d-glucosidase activities in cellulase from Aspergillus fumigatus Fresenius, Enzyme Microb Technol 8:70–74 Park S, Baker J, Himmel ME, Parilla P, Johnson D (2010) Cellulose crystallinity index: measurement techniques and their impact on interpreting cellulase performance, Biotechnol Biofuels 3:10


Part III Lignins and Hemicelluloses


Chapter 19 Structural Characterization of the Heteroxylans from Poplar and Switchgrass Koushik Mazumder, Maria J. Peña, Malcolm A. O’Neill, and William S. York Abstract Heteroxylans are polysaccharides with a backbone composed of 1,4-linked β-D-xylosyl residues. In hardwoods some of these xylosyl residues are substituted at O-2 with 4-O-methyl α-D-glucuronic and occasionally with α-D-glucuronic acid. In grasses, the xylan backbone is predominantly substituted with α-L-arabinofuranosyl residues (most often at O-3, but sometimes at O-2). Grass heteroxylan backbone residues may also have small amounts of α-D-glucuronic acid and/or 4-O-methyl α-D-glucuronic acid at O-2. Heteroxylans have a role in maintaining the structural integrity of the cell walls that comprise the bulk of lignocellulosic biomass. Moreover, differences in the molecular features of these hemicellulosic polysaccharides, including their degree of polymerization, degree of branching and spatial arrangement of side chains along the xylan backbone, have been correlated to altered cell wall properties (Izydorczyk MS, Biliaderis CG, Carbohydr Polym 28:33–48, 1995) and the ease with which biomass can be enzymatically converted to fermentable sugars. Thus, understanding the relationship between heteroxylan structure and biomass properties is required to engineer bioenergy crops with improved processing characteristics. In this chapter we describe some of the analytical methods we routinely use to perform in-depth structural analysis of heteroxylans from poplar and switchgrass biomass. Key words: Lignocelluloseic biomass, Heteroxylan, Compositional analysis, Oligosaccharides, Mass spectrometry, NMR spectroscopy

1. Introduction Lignocellulosic biomass is derived primarily from the cell walls of grasses and woody plants and is composed predominantly of cellulose, hemicellulose, and lignin together with smaller amounts of pectin, protein and minerals. The structure and composition of these cell walls depends on both the type of plant (hardwood, softwood, or grass) and the age of the tissue. Lignocellulosic biomass is typically composed primarily of mature tissues rich in secondary Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_19, © Springer Science+Business Media, LLC 2012

215


216

K. Mazumder et al.

cell walls. Glucuronoxylan is the most abundant hemicellulose present in the secondary walls of hardwoods, with smaller amounts of xyloglucan and glucomannan, whereas glucomannans predominate in softwoods. By contrast, arabinoxylans account for the bulk of the hemicellulose in grasses (1, 2). The heteroxylans of land plants have a backbone composed of (1 → 4)-linked β-D-xylopyranosyl residues, some of which are substituted either at O-2 or O-3 with monosaccharides or short oligosaccharide side chains. In hardwoods, 4-O-methyl α-D-glucuronic acid and (less frequently) α-D-glucuronic acid sidechains are present at O-2, whereas in grasses α-L-arabinofuranosyl side chains are present, predominantly at O-3 but occasionally at O-2. Grass heteroxylans may also contain relatively small amounts of α-D-glucuronic acid and 4-O-methyl α-D-glucuronic acid side chains at O-2 (2, 3). In this chapter we describe some of the methods we have used to isolate and structurally characterize the heteroxylans from poplar and switchgrass lignocellulosic biomass. The methods include the following: 1. Preparation of the alcohol insoluble residue (AIR) from lignocellulosic biomass. 2. Treatment of AIR with aqueous alkali to solubilize the heteroxylans. 3. Endoxylanase treatment of the heteroxylans to generate xylooligosaccharides. 4. Structural characterization of heteroxylans using glycosyl residue, glycosyl-linkage, mass spectrometric, and 1H-NMR spectroscopic analyses of the intact and enzymatically fragmented polysaccharide.

2. Plant Materials 2.1. Alcohol Insoluble Residue Preparation Components

2.2. Preparation of Extracts

Poplar and switchgrass biomass that has been milled to pass through a #20 mesh but is retained by a #80 mesh. ●

Aqueous 80 % ethanol.

Ceramic ball mill jar and zirconium beads (0.25 in.).

Acetone.

50 mM ammonium oxalate prepared by adding 500 mL of deionized water to 3.6 g NH4OCOCOONH4·H2O; 1 M KOH containing 1 % w/v sodium borohydride prepared by adding 500 mL of deionized water to 28.1 g of KOH and 5 g of NaBH4; 4 M KOH containing 1 % w/v sodium borohydride


19

Structural Characterization of the Heteroxylans…

217

prepared by adding 500 mL of deionized water to 112.2 g of KOH and 5 g of NaBH4 (see Note 1). ●

Dialysis bags (3,500 molecular weight cutoff) and clips.

2.3. Neutral Sugar Composition and Linkage Analysis

2 M Trifluoroacetic acid (TFA) prepared by slowly adding 1 mL of concentrated TFA (12.9 M) to 5.5 mL of deionized water; 1 M ammonium hydroxide prepared by adding 6.7 mL of conc. ammonia [14.8 M] to 93.3 mL of deionized water; methanol containing 10 % acetic acid prepared by mixing 90 mL of anhydrous methanol and 10 mL of glacial acetic acid [17.4 M]; 100 mM sodium carbonate prepared by adding 1.1 g of anhydrous Na2CO3 to 100 mL of deionized water.

2.4. NMR Analysis of the Xylooligosaccharides

Deuterium oxide (99.9 %; Cambridge Isotope Laboratories).

3-mm NMR tubes.

2.5. Preparation of Matrix for MALDITOFMS Analysis

To prepare the matrix for MALDI-MS analysis, dissolve 154 mg of 2,5-dihydroxybenzoic acid (0.1 M) and 44 mg of 1-hydroxyisoquinoline (0.03 M) in 10 mL of aqueous 50 % acetonitrile.

3. Methods All procedures are carried out at room temperature unless specified otherwise. 3.1. Preparation of the Alcohol Insoluble Residue

1. Suspend dry biomass (1 g) in 80 % ethanol (60 mL) and transfer suspension to a ceramic ball mill jar (250 mL) half-filled with 0.25 in. zirconium milling beads. Mill the biomass at 100 rpm for 16 h at 4 °C (see Note 2). (Note: Ball-milling typically reduces the particle size of the biomass to <50 μm.) This increases the amounts of polysaccharides that can be solubulized using aqueous buffers and alkali. However, the reduced particle size may result in a decrease in the apparent recalcitrance of the biomass. Thus, the ball-milling step should be omitted for samples also being prepared for recalcitrance analysis. In this case, the biomass is simply stirred in aqueous 80 % ethanol overnight and then centrifuged (step 3 in Subheading 3.1) to remove the 80 % ethanol-soluble extractives. 2. Remove the milling beads by filtration through a kitchen sieve (16 mesh). 3. Transfer the suspension to centrifuge tubes and centrifuge for 15 min (3,600 × g), discard the supernatant. Wash the residue with acetone (50 mL).


218

K. Mazumder et al.

4. Transfer the residue to a Petri dish (8 cm) and allow solvents to evaporate overnight in a fume hood (see Note 3). Finally vacuumdry the residue at room temperature to remove residual acetone. The dry material is stored at room temperature. (Note: Acetone is a health and fire hazard, and should be removed in a fume hood.) 3.2. Solubilization of Hemicellulose

1. Suspend dry AIR (1 g) in 50 mM ammonium oxalate (60 mL) and agitate the suspension for 16 h at 120 rpm on a gyratory shaker. 2. Centrifuge the suspension for 15 min (3,600 × g). Decant the supernatant and suspend the residue in deionized water (10 mL). Repeat centrifugation, combined the supernatants and dialyze against distilled water (5 × 2 L). After dialysis, concentrate the solution to ~10 mL by rotary evaporation and then lyophilize to obtain the ammonium oxalate-soluble material. 3. Suspend the oxalate-insoluble residue in 1 M KOH containing 1 % (w/v) sodium borohydride (60 mL) and stir overnight at 120 rpm (see Note 4). Centrifuge (3,600 × g) the suspension for 15 min. Collect the supernatant and add 3–4 drops of octanol, cool in ice and then adjust to pH 5 by drop-wise addition of glacial acetic acid (see Note 5). Dialyze the neutralized extract against deionized water (5 × 2 L) (see Note 6). If a precipitate forms, remove it by filtration through nylon mesh, concentrate the soluble material by rotary evaporation, and lyophilize to obtain the 1 M KOH-soluble material. 4. Suspend the 1 M KOH-insoluble residue in 4 M KOH containing 1 % sodium borohydride and treat as described in step 3 in Subheading 3.2.

3.3. Glycosyl Residue Composition Analysis

1. Transfer portions (200–500 μg) of the oxalate and alkali solubulized materials to test tubes with teflon-lined screw caps, add 2 M TFA (0.5 mL) and 20 μg of inositol (internal standard, 20 μL of a 1 mg/mL stock solution). Seal the tubes and heat for 2 h at 120 °C in a heating block (4, 5). 2. Evaporate the TFA using a stream of air, add 20 drops of isopropanol and concentrate to dryness. Repeat the addition of isopropanol twice. 3. Add sodium borodeuteride (200–300 μL of a 10 mg/mL solution in 1 M ammonium hydroxide) and incubate for 2 h at room temperature. Destroy the unreacted NaBD4 by adding 3–4 drops of glacial acetic acid and 200 μL of methanol:acetic acid (v/v 9:1). Concentrate to dryness using a stream of air. Add methanol:acetic acid (9:1 v/v, 200 μL) and concentrate to dryness. Repeat this step and then add ~200 μL of anhydrous methanol and concentrate to dryness. Repeat this last step three times.


19

Structural Characterization of the Heteroxylans…

219

Fig. 1. GC-MS chromatogram of the alditol acetate derivatives generated from the 1 M KOH-soluble material from switchgrass AIR.

4. Add acetic anhydride (250 μL) and concentrated TFA (250 μL) to the residue and keep for 15 min at 50 °C. Cool to room temperature and then add 20–30 drops of isopropanol and concentrate to dryness using the stream of air. Repeat this step one more time. 5. Add 0.1 M sodium carbonate (1 mL) followed by methylene chloride (1 mL), vortex, and then remove the (upper) aqueous layer. Wash the organic layer with deionized water (2 × 1 mL). 6. Concentrate the organic layer to dryness and dissolve the residue in acetone (100–200 μL). The alditol acetate derivatives of the monosaccharides are then analyzed by GC-MS using a SP 2330 (30 m × 0.25 mm, 0.25 μm film thickness, Supelco) using helium as carrier gas and the following temperature gradient: 80 °C for 2 min, 80–170 °C at 30 °C/min, 170–240 °C at 4 °C/min, 240 °C held for 20 min using a Agilent 5890 gas chromatograph coupled to a Agilent 5880A mass selective detector. The sample (~1 μL) is injected in the splitless injection mode. The alditol acetate derivatives of the neutral glycoses typically present in poplar and switchgrass biomass are separated in under 30 min (Fig. 1). Alternate methods are required to detect and quantify glucuronic acid and 4-O-methyl glucuronic acid. 3.4. Glycosyl-Linkage Analysis 3.4.1. Preparation of the Solid NaOH/DMSO Reagent

1. Transfer 20–30 drops of aqueous 50 % NaOH to a screw cap vial; add anhydrous methanol (1 mL) and vortex to give a clear solution. Add anhydrous dimethylsulfoxide (DMSO, 1 mL), vortex, and centrifuge for 5 min (3,600 × g). 2. Remove and discard the supernatant; add anhydrous DMSO (1 mL), vortex, and centrifuge to remove the DMSO. 3. Repeat step 2 in Subheading 3.4.1 three times to prepare a transparent NaOH pellet. Suspend the pellet in 200–300 μL of anhydrous DMSO and use immediately for the methylation reaction.


220

K. Mazumder et al.

Fig. 2. GC-MS chromatogram of the per-O-methylated alditol acetates generated from switchgrass arabinoxylan.

3.4.2. Methylation of a Polysaccharide or Oligosaccharide (5, 6)

1. The lyophilized sample (100–200 μg of polysaccharide or oligosaccharide) is transferred to a screw cap vial and dissolved in anhydrous DMSO (0.2–0.5 mL). 2. Freshly prepared NaOH in DMSO (~200 μL from step 3 in Subheading 3.4.1) is added and the mixture kept for 15 min at room temperature with stirring using a small teflon-coated “magnetic flea.” 3. Methyl iodide (300 μL, caution the reagent is toxic) is added and the mixture kept for 15 min at room temperature. Cool the reaction mixture in an ice bath and then add two drops of deionized water. Remove the excess methyl iodide by carefully bubbling air through the reaction mixture (perform in a fume hood). Add methylene chloride (1 mL), vortex, and transfer organic (lower) layer to a clean tube. Wash the organic later three times with deionized water (1 mL). Evaporate the methylene chloride and dissolve the residue containing the methylated materials in methanol and store at −20 °C. 4. The per-O-methylated glycans are converted to their partially methylated aliditol acetate (PMAA) derivatives and analyzed by GC-MS as described in Subheading 3.3 and Fig. 2. The positions of the OMe and OAc substituents in a PMAA are readily determined from the primary fragment ions present in their electron-impact mass spectra (Fig. 3).

3.5. Preparation of Xylooligosaccharides (7, 8)

1. Suspensions of the 1 M KOH and/or 4 M KOH soluble materials (5–20 mg) in 50 mM ammonium formate (1–5 mL, pH 5.0) are vortexed. Endoxylanase (GH family 11 from Trichoderma viride, 1 unit) is then added and the suspension kept for 24 h at 37 °C. (Note: In some instances, the alkali extracts may not be completely dissolved in the buffer). 2. The reaction mixture is kept for 10 min in a boiling water bath to inactivate the enzyme. The suspension is then centrifuged for 15 min (3,600 × g). The soluble material is lyophilized and


19

Structural Characterization of the Heteroxylansâ&#x20AC;Ś

221

Fig. 3. Electron-impact mass spectrum of 1,2,4-tri-O-acetyl-1-deuterio-3,5-di-O-methyl arabinitol, the PMAA derived from a 1,2-linked arabinofuranosyl residue.

then dissolved in 50 mM ammonium formate (1 mL). Portions (200 ÎźL) of the solution are then fractionated by size-exclusion chromatography. Endoglucanase-resistant polymers such as xyloglucan can be separated from the xylo-oligosaccharides using a Supedex-75 HR10/30 column eluted with 50 mM ammonium formate (0.5 mL/min) and a refractive index detector. In this case, the xylo-oligosaccharides are typically eluted just before the totally included volume of the column. Fractions are collected manually. All fractions containing xylooligosaccharides are lyophilized, dissolved in water (1 mL), and lyophilized again to remove the volatile ammonium salt. This lyophilization procedure may have to be repeated up to three times to ensure complete removal of the ammonium formate. Alternatively, xylo-oligosaccharides can be separated from each other and from endoxylanase-resistant polysaccharides using a Bio-Gel P2 column eluted with water. 3.6. NMR Analysis of the Xylooligosaccharides

NMR is a nondestructive technique widely used in the structural analysis of xylans. Over the years, purified xylo-oligosaccharides obtained from diverse tissues and species have been rigorously F by NMR. As a result of these analyses, chemical shifts and coupling constants diagnostic for the most common structural features of xylans are available in the literature (3, 7, 8) or via the Internet (http://ccrc.uga.edu/db/nmr). 1D and 2D 1H-gCOSY experiments are usually sufficient to identify isolated spin systems and to obtain chemical shifts and coupling constants for each type of glycosyl residue in the xylo-oligosaccharides. The NMR data can often be interpreted from first principles, but comparison to chemical shifts


222

K. Mazumder et al.

Fig. 4. Partial 600 MHz 1H-NMR spectra of xylo-oligosaccharides generated by xylanase-treatment of the 1 M KOH-soluble materials from poplar wood (a) and mature switchgrass biomass (b).

previously assigned to similar oligosaccharides is often sufficient to determine the identity, the ring form, and the anomeric configuration of each type of residue. Side chain structures and their linkage positions can also be deduced from this data, especially if ROESY, NOESY, and/or HMBC data are included in the analysis. One advantages of the NMR analysis is the ability to quantify (by signal integration of anomeric protons in the 1D 1H-spectrum) specific xylan structural features, such as the contribution of each type of side chain to the total backbone substitution. The reducing end of hardwood xylan contains the glycosyl sequence 4-β-D-Xylp-(1 → 4)-β-D-Xylp-(1 → 3)-α-L-Rhap-(1 → 2)α-D-GalpA-(1 → 4)-D-Xylp (7, 9). This unique sequence can be used as reference to calculate the average degree of polymerization of the xylan. The 1H-NMR spectra of xylooligosaccharide mixtures (Fig. 4) contains a considerable amount of valuable information, although signal overlap often makes it difficult completely determine the glycosidic sequence from this data. Complementary sequence information can be obtained using electrospray-ionization MS analysis, as described in Subheading 3.8. ●

Oligosaccharides (1–2 mg) are dissolved in deuterium oxide (99.9 %; Cambridge Isotope Laboratories, 0.5 mL), and acetone (1 μL) is added as an internal standard.


19

Structural Characterization of the Heteroxylans…

223

1

H NMR spectra are recorded (e.g., with a Varian Inova NMR spectrometer equipped with a 3-mm cold probe operating at 600 MHz) with a sample temperature of 298 K. Two-dimensional spectra are recorded using standard pulse sequences (typically provided by the instrument manufacturer). The COSY spectra are recorded at high digitization (e.g., 800 × 1,024 complex points) and zero filled to obtain a 2,048 × 2,048 matrix. Data are processed using commercially available software (e.g., MestRe-C, Universidad de Santiago de Compostela, Spain). Chemical shifts are calibrated relative to acetone (δH 2.225 ppm). (Note: The resonance of the H5 of the glucuronic residue is sensitive to pH. It is important that the pH of the sample is adjusted to neutral before recording NMR spectra). 3.7. Preparation of Per-O-Methylated Oligoglycosyl Alditols (8)

3.8. Mass Spectrometric Analysis

The oligoglycosyl alditols are per-O-methylated as described in Subheading 3.4.

Positive ion MALDI-TOF mass spectra are recorded (e.g., using a Bruker Microflex LT mass spectrometer and workstation). An approximately 1 mg/mL solution of oligosaccharides in water or methanol (1 μL) is mixed with an equal volume of matrix solution (0.1 M 2,5-dihydroxybenzoic acid and 0.03 M 1-hydroxyisoquinoline in aqueous 50 % MeCN) on the MALDI target plate and then dried with a stream of warm air (from a hair dryer) (see Note 7).

Spectra from at least 200 laser shots (60–65 % laser power) are collected and averaged to generate a mass spectrum.

3.8.1. MALDI-TOF-MS

3.8.2. ESI Mass Spectrometry

The oligosaccharides (~250–500 μg) are converted to their corresponding oligoglycosyl alditols by treatment for 16 h at room temperature with 1 M sodium borohydride in 1 M ammonium hydroxide (250 μL). The reaction is terminated by the dropwise addition of glacial acetic acid. To remove borate as its methyl ester, acetic acid in methanol (10 % v/v, 0.5 mL) is added and evaporated three times using a stream of nitrogen gas. Anhydrous methanol (0.5 mL) is then added and evaporated three times.

Multiple-step mass spectrometry (ESI-MSn) is a powerful analytical technique that can determine the glycosyl sequences and branching topologies of per-O-methylated oligosaccharides or oligoglycosyl alditols (8, 10). Methylation of the free hydroxyl groups of the oligomers increases the sensitivity and provides mass spectral features that are diagnostic for fragmentation events (8, 10). The introduction of methyl groups labels free hydroxyl groups that are not protected by involvement in a glycosyl linkage. Thus, each free hydroxyl group generated during gas-phase fragmentation of the methylated oligosaccharide can be identified by


224

K. Mazumder et al.

Fig. 5. ESI-MSn spectra showing fragmentation of the quasimolecular precursor ion at m/z 1,045, selected from the ESI-MS spectrum of a per-O-methylated oligoglycosyl alditol containing five pentosyl residues (X and A) and one pentitol (Xol). The oligosaccharide was generated by endoxylanase treatment of the 1 M KOH-soluble material from switchgrass AIR, purified by SEC on Bio-Gel P-2, reduced with NaBH4 and then per-O-methylated. The assignment of pentosyl residues as xylose (X) or arabinose (A) was based on NMR analysis. Glycosidic bonds are indicated by small arrows within each structure. Cleavage of glycosidic bonds exposes hydroxyl groups (scars), which are indicated by dots. The fragment ion at m/z 871 was selected from the MS2 spectrum for fragmentation in MS3. Fragment ions at m/z 697 and 711 were each selected from the MS3 spectrum for fragmentation in MS4.

its characteristic 14 Da mass-difference relative to a methylated site. Such unmethylated “scars” facilitate the interpretation of the resulting MSn data, providing key information required for the identification and location of branch points in the glycosyl sequence. (See Figs. 5–8.) We have previously shown that the most abundant


19

Structural Characterization of the Heteroxylansâ&#x20AC;Ś

225

Fig. 6. ESI-MS spectrum of the per-O-methylated glucuronoxylan oligoglycosyl alditols. Birch glucuronoxylan was fragmented with endoxylanase and the resulting oligosaccharides were isolated by SEC, reduced, per-O-methylated and analyzed as quasimolecular [M+Na]+ ions by ESI-MS in the positive ion mode. The birch glucuronoxylan consists predominantly of xylosyl and 4-O-methyl glucuronosyl residues, but the presence of endogenous methyl groups cannot be established by MS analysis of the per-O-methylated oligoglycosyl alditols prepared using CH3I. Therefore, annotation of ions in this spectrum (and in the spectra shown in Fig. 8) indicates the Xyl and GlcA composition without regard to the presence of endogenous methyl groups.

fragment ions in the ESI-MSn spectra of the per-O-methylated (neutral) oligosaccharides from switchgrass (Fig. 5) are the socalled Y and B fragment ions (8, 10). Due to their high abundance, Y ions are most often selected for further fragmentation to obtain glycosyl sequence information. We have also performed MSn analysis of hardwood xylo-oligosaccharides (Figs. 6â&#x20AC;&#x201C;8), which contain 4-O-methyl glucuronic acid. The carboxylate group of the uronic acid is methyl-esterified during the per-O-methylation reaction. The endogenous methyl group at O-4 cannot be distinguished from methyl groups that are incorporated by chemical derivatization by ESI-MS. Thus, a slight modification of the method is required when using ESI-MS to distinguish glucuronic acid from 4-O-methyl glucuronic acid. That is, trideuteriomethyl derivatives are generated using trideuteriomethyl iodide (11). In this case, the incremental mass of a per-O-deuteriomethyl glucuronic acid is 3 Da more than that of a per-O-deuteriomethyl 4-O-methyl glucuronic acid, allowing these two types of residues to be distinguished. Orthogonal methods (notably NMR) demonstrate that the birch xylan oligoglycosyl alditol whose spectra are shown in


226

K. Mazumder et al.

Fig. 7. MSn fragmentation patterns (sodiated Y-ions) of four possible isomers (I–IV) of per-O-methylated Xyl4GlcAol, which gives rise to quasimolecular ions at m/z 943. Fragmentation pathways that were observed are indicated by large solid arrows and fragmentation pathways that were not observed are indicated by dashed arrows. High abundance of ions at m/z 391 and low abundance of ions at m/z 551, 609 and 217 in the MSn spectra (Fig. 8) indicate that structure II is by far the most abundant pentasaccharide generated by endoxylanase-treatment of the Birch glucuronan.

Figs. 6–8 contains a single endogenous 4-O-methyl glucuronic acid residue. ESI mass spectra are recorded (e.g., with a Thermo Scientific LTQ XL ion trap mass spectrometer). Solutions of per-O-methylated oligosaccharides or oligosaccharide-alditols in methanol are mixed with 50 % acetonitrile–water containing 0.1 % TFA to give sample concentrations of ~1 ng/μL. ●

The sample is introduced in the MS source at flow rate of 3 μL/min through a fused silica capillary (150 μm i.d. × 60 cm, Thermo Finnigan, USA) using a syringe pump.

The electrospray source is operated at a voltage of 5.0 kV and a capillary temperature of 275 °C. To obtain MSn spectra, the collision energy is initially set to 30 V, although the energy may need to be adjusted to obtain optimal fragmentation.


19

Structural Characterization of the Heteroxylansâ&#x20AC;Ś

227

Fig. 8. ESI-MSn spectra showing fragmentation of the quasimolecular precursor ion at m/z 943, selected from the ESI-MS spectrum of the glucuronoxylan oligoglycosyl alditols from birch glucuronoxylan (Fig. 6). Ions at m/z 943 could theoretically arise via sodiation of the four possible isomers of Xyl4GlcA shown in Fig. 7. The fragment ion at m/z 769 was selected from the MS2 spectrum for fragmentation in MS3. The fragment ion at m/z 537 was selected from the MS3 spectrum for fragmentation in MS4.


228

K. Mazumder et al.

4. Notes 1. Use a magnetic stir bar to prepare buffer and alkali solutions to increase the dissolution of solute in solution. 2. During the process of ball milling record the time and speed (8 to 16 h, 90 to 100 rpm), as the milling time and speed affect the final particle size of the biomass. 3. During AIR preparation, Place dishes in rack in the vacuum dessicator and allow acetone to evaporate overnight. Continue drying until the odor of acetone is completely gone (use appropriate “waft technique” to avoid inhaling toxic acetone fumes). 4. Add 1 % sodium borohydride to 1 M and 4 M KOH solution during alkali extraction to avoid the peeling effect on polysaccharides by alkaline solutions. 5. During the neutralization of alkali extracted polysaccharides, add 3–4 drops of octanol required to prevent excessive foaming and sample loss when adjusting the pH of the supernatant (check the neutral pH of the solution using pH paper). 6. Change the distilled water used for dialysis every 6–8 h. 7. Matrix solution used for MALDI-MS analysis can be stored at −20 °C for 6 months. References 1. Izydorczyk MS, Biliaderis CG (1995) Cereal arabinoxylans: advances in structure and physicochemical properties. Carbohydr Polym 28: 33–48 2. Ebringerova A (2006) Structural diversity and application potential of hemicelluloses. Macromol Symp 232:1–12 3. Verbruggen MA, Spronk BA, Schols HA, Beldman G, Voragen AGJ, Thomas JR, Kamerling JP, Vligenthart JFG (1998) Structures of enzymatically derived oligosaccharides from sorghum glucuronoarabinoxylan. Carbohydr Res 306:265–274 4. Fox A, Morgan SL, Gilbart J (1989) Preparation of alditol acetates and their analysis by gas chromatography and mass spectrometry. In: Biermann CJ, McGinnis GD (Eds) Analysis of carbohydrates by GLC and MS. CRC Press: Boca Raton, FL. p 87–117 5. Carpita NC, Shea EM (1989) Linkage structure of carbohydrates by gas chromatographymass spectrometry (GC-MS) of partially methylated alditol acetates. In: Biermann CJ, McGinnis GD (Eds) Analysis of carbohydrates by GLC and MS. CRC Press: Boca Raton, FL. p 157–216

6. Ciucanu I, Kerek F (1984) A simple and rapid method for the permethylation of carbohydrates. Carbohydr Res 131:209–217 7. Pena MJ, Zhong R, Zhou GK, Richardson EA, O’Neill MA, Darvill AG, York WS, Ye ZH (2007) Arabidopsis irregular xylem8 and irregular xylem9: implications for the complexity of glucuronoxylan biosynthesis. Plant Cell 19: 549–563 8. Mazumder K, York WS (2010) Structural analysis of arabinoxylans isolated from ball-milled switchgrass biomass. Carbohydr Res 345:2183–2193 9. Shimizu K, Ishihara M, Ishihara T (1976) Hemicellulases of brown rotting fungus. Tyromyces palustris. II. Mokuzai Gakkaishi 22:618–625 10. Domon B, Costello CE (1988) A systematic nomenclature for carbohydrate fragmentations in FAB-MS/MS spectra of glycoconjugates. Glycoconjugate J 5:397–409 11. Reuhs BL, Glenn J, Stephens SB, Kim JS, Christie DB, Glushka JG, Zablackis E, Albersheim P, Darvill AG, O’Neill MA (2004) L-Galactose replaces l-fucose in the pectic polysaccharide rhamnogalacturonan II synthesized by the l-fucose-deficient mur1 Arabidopsis mutant. Planta 219:147–157


Chapter 20 Laser Microdissection and Genetic Manipulation Technologies to Probe Lignin Heterogeneity and Configuration in Plant Cell Walls Oliver R.A. Corea, Chanyoung Ki, Claudia L. Cardenas, Laurence B. Davin, and Norman G. Lewis Abstract Single and multiple T-DNA knockouts of genes encoding arogenate dehydratases (ADTs) in Arabidopsis were obtained in homozygous form. These were analyzed for potential differences in lignin contents and compositions, as well as for distinct phenotypes over growth and development. Of these different lines, distinct reductions in lignin contents were obtained, with those having different G:S ratios depending upon the combination of ADT genes being knocked out. Results from pyrolysis GC/MS analyses indicated that differential carbon flux occurred into the vascular bundles (vb) and interfascicular fibers (if). These results provide additional new insight into factors controlling lignin heterogeneity and configuration. Key words: Arogenate dehydratase, T-DNA knockouts, Phenylpropanoid metabolism, Metabolic flux, Lignin analyses, Thioacidolysis, Interfascicular fiber and vascular bundle lignins, Laser microscope microdissection, Pyrolysis GC/MS, G:S ratio determining factors

1. Introduction Macromolecular lignins in vascular plant cell walls are complex plant biopolymers, in terms of average monomeric compositions, overall lignin contents, and their compositions in specific cell wall types, such as those in fibers and vascular bundles/vessels. In angiosperms, the lignins are derived from p-coumaryl (p-hydroxycinnamyl, 1), coniferyl (2), and sinapyl (3) alcohol monomeric precursors and are referred to as H, G, and S-lignins (see Structures 1â&#x20AC;&#x201C;3), respectively (1). The resulting lignin biopolymers are, however, heterogeneous from a spatial perspective, as they have different monomeric compositions and contents in distinct cell wall types and subcellular layers (1). More specifically, angiosperm

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_20, Š Springer Science+Business Media, LLC 2012

229


230

O.R.A. Corea et al.

lignins are syringyl-(S) and guaiacyl-(G) rich in fiber and vascular bundles/vessels. Additionally, cell wall types in such tissues have low amounts of H-derived moieties which are considered to be deposited at the early stages of lignification. Taken together, lignins represent Natureâ&#x20AC;&#x2122;s most abundant biopolymers, next to cellulose.

One of the most important issues in sustainable bioenergy considerations is the question of how lignocellulosic recalcitrance can be overcome in order to provide the facile means to generate, for example, either transportation fuels and/or other petrochemical replacements at the scale and cost needed. Accordingly, enhancing our understanding of the lignification process and its potential for manipulations through genetic engineering/plant breeding is considered key to help resolve such questions. The purpose of this contribution is thus to discuss and describe the various technologies that can now be brought to bear upon the still difficult challenge of lignin biogenesis, lignin macromolecular structure and configuration, and lignin deconstruction. In this context, the ability to modulate gene expression of specific biochemical pathway steps has provided scientists with the tools to readily generate genetically modified plant lines, such as those with alterations in lignin contents and compositions. The most common approaches used to date involve generation of either plant knockouts (KOs), through T-DNA insertions affecting genes of interest, and/or downregulation of genes encoding specific biochemical steps. Such approaches are now providing us with the hitherto unprecedented opportunity to systematically dissect various aspects of lignin deposition and formation in specific cell types, and to assess the potential of such plant lines as feedstock for renewable energy/bioproducts.


20

Laser Microdissection and Genetic Manipulation Technologies…

231

In this contribution, for illustration purposes, we describe the potential of modulating different members of the gene family encoding the final biochemical step to phenylalanine, namely, that of the arogenate dehydratases (ADTs). In Arabidopsis, this family has six distinct genes. Interestingly, this biochemical pathways step is upstream of what is considered to be the phenylpropanoid pathway leading to lignins. Nevertheless, as indicated below, using a range of molecular biological, biochemical, chemical, and histochemical procedures and technologies, we are now beginning to gain novel insights into lignin macromolecular configuration and structure, as well as effects on the resulting vasculature in the so-modified plant lines. Of particular interest in this contribution is the differential carbon flux that ultimately occurs into the distinct lignified cell wall types and tissues. As a result of these new technologies and approaches, we are now beginning to gain a true understanding of the lignification process, of lignin primary structures, and of meaningful correlations with biophysical cell-wall composite properties.

2. Materials 2.1. Generation and Verification of Arabidopsis Knockout Lines

2.2. Sampling and Phenotypic Analysis

Search T-DNA insertion lines for each gene of interest at the Salk Institute Genomic Analysis Laboratory (SIGnAL; http:// signal.salk.edu/cgi-bin/tdnaexpress) (2) and order seed stocks from The Arabidopsis Biological Resource Center (ABRC, http://www.arabidopsis.org/abrc/).

3.5 in. pots, soil (Sunshine Mix #1/LC1, Sun Gro Horticulture).

REDExtract-N-AmpTM Plant PCR kit (Sigma-Aldrich), containing REDExtract-N-Amp PCR Ready Mix, Extraction Solution and Dilution Solution.

T-DNA and gene-specific primers.

Agarose (Invitrogen).

RNeasy kit (QIAgen).

Superscript IITM reverse transcriptase (Invitrogen).

Polymerase Chain Reaction (PCR) thermocycler (C1000TM Thermal Cycler, BioRad, Hercules, CA, USA).

Vernier caliper, tape measure, scale.

FalconTM conical tubes (15 mL).


232

O.R.A. Corea et al.

2.3. Histochemical Staining 2.3.1. Common Materials

2.3.2. Phloroglucinol–HCl

2.3.3. Mäule Reagent

2.4. Amino Acid Analysis

2.5. Pyrolysis GC-MS of Vascular Bundles and Interfascicular Fiber Tissues 2.5.1. Laser Microscope Dissection

Double edge stainless steel razor blade (Electron Microscopy Sciences, Hatfield, PA).

Phosphate buffered saline (PBS) solution (1×): To 800 mL of deionized H2O, add NaCl (8 g), KCl (0.2 g), Na2HPO4 (1.44 g), and KH2PO4 (0.24 g). Adjust the pH to 7.4 with HCl and bring volume to 1,000 mL with deionized H2O.

Glutaraldehyde (EM Grade, 10 %; Electron Microscopy Sciences, Hatfield, PA).

Plastic transfer pipette droppers (3 mL).

Reagent reservoirs/staining trays.

Microscopy slides and slide covers.

Olympus BH-2 photomicroscope with a ProgRes C12plus digital microscope camera (JENOPTIK, Jena, Germany).

Ethanol (95 %).

Phloroglucinol (Sigma-Aldrich).

Hydrochloric acid (36.5–38 %, J.T. Baker).

Potassium permanganate (Sigma-Aldrich).

Hydrochloric acid (36.5–38 %, J.T. Baker).

Ammonium hydroxide (30 %, J.T. Baker).

Methanol (Optima® LC/MS, Fisher Scientific).

Chloroform (J.T. Baker).

Norvaline (Sigma-Aldrich, 100 mM).

AccQ•TagTM Ultra Derivatization Kit (Waters, Milford, MA).

AccQ•TagTM Ultra Eluent A, concentrate (Waters, Milford, MA).

AccQ•TagTM Ultra Eluent B (Waters, Milford, MA).

Microcentrifuge (Eppendorf).

Speed-Vac® SC110 (Savant).

Ultra-Performance Liquid Chromatography (UPLC) for amino acid analysis, coupled to a photodiode array detector (Waters, Milford, MA).

AccQ•TagTM Ultra Milford, MA).

Cryocut 1800 microtome (Leica Microsystems, Deerfield, IL).

Accu-Edge® low profile microtome blades (Sakura Finetek USA, Inc.).

Tissue-Tek® optimum cutting temperature (O.C.T.)TM compound (Miles, Inc., Elkhart, IN).

PEN-membrane covered slides (1.0 Microimaging, Munich, Germany).

column

(2.1 × 100

mm)

mm;

(Waters,

Carl

Zeiss


20

2.5.2. Pyrolysis GC-MS of Vascular Bundles and Interfascicular Fiber Regions

2.6. Extraction of Aqueous/Organic Solubles and Estimation of Lignin Constituents

Laser Microdissection and Genetic Manipulation Technologies… ●

Ethanol (200 proof, Decon Laboratories, King of Prussia, PA).

Ethanol (95 %).

Optima® LC/MS-grade H2O (Fisher Chemical).

Microfurnace pyrolyzer (PYR-4A; Shimadzu).

Shimadzu GC-17A gas chromatograph coupled to a Shimadzu QP-5000 mass spectrometer.

SPBTM-5 column [poly (5 % diphenyl/95 % dimethylsiloxane), 30 m × 0.25 mm × 1 mm film, Supelco].

Platinum sample cup.

Micro/Pettor® (Scientific Emeryville, CA).

Freeze-dryer (Labconco®).

Planetary ball mill PM 400 (Retsch, Haan, Germany).

Ethanol (95 %).

Toluene (J.T. Baker).

Centrifuge (RC 3C Plus, Sorvall).

Glass vials (4 mL) with polytetrafluoroethylene (PTFE)-coated silicon septa.

Acetyl bromide (Sigma-Aldrich).

Glacial acetic acid (J.T. Baker).

Perchloric acid (70 %, J.T. Baker).

Sodium hydroxide (J.T. Baker).

Volumetric flasks (50 mL).

Spectrophotometer (Perkin Elmer UV/Vis spectrophotometer Lambda 20).

Boron trifluoride diethyl etherate (³46.5 %, Sigma-Aldrich).

1,4-Dioxane (99.8 %, anhydrous, Sigma-Aldrich).

Ethanethiol (97 %, Sigma-Aldrich).

Tetracosane C24 (99 %, C24H50, Sigma-Aldrich).

Methylene chloride (J.T. Baker).

Sodium bicarbonate (Fisher Scientific).

Anhydrous sodium sulfate (Fisher Scientific).

N,O bis (trimethylsilyl) trifluoroacetamide + trimethylchlorosilane (BSTFA + TMCS, 99:1, Supelco).

Pyridine (99.8 %, anhydrous, Sigma-Aldrich).

Round-bottom glass tubes (20 mL) with Teflon®-lined screwcaps (Pyrex).

2.6.1. Cell Wall Residue Extraction 2.6.2. Acetyl Bromide Reaction

2.6.3. Thioacidolysis

233

Manufacturing

Industries,


234

O.R.A. Corea et al. ●

Rotary evaporator (Büchi Rotavapor-R).

HP 6890 Series GC System (Hewlett Packard) equipped with a Restek-5-SiL-MS (Restek, 30 m × 0.25 mm × 0.25 mm) column and a HP 5973 MS detector.

Authentic silylated standards (3).

3. Methods 3.1. Generation and Verification of Arabidopsis Knockout Lines

1. Utilize the T-DNA Primer Design Tool (http://signal.salk. edu/tdnaprimers.2.html) to design the left primer (LP) and right primer (RP) flanking the T-DNA insertion site. These primers will be designed to amplify a region of ~900 to 1,100 bp long from DNA lacking a T-DNA insert. To verify the presence of a T-DNA insert in that region, use one of the three left border (LB) primers for the T-DNA insert (LBa1, LBb1, or LBb1.3). The T-DNA Primer Design tool can be used to determine the expected length of the T-DNA-specific band. 2. Vernalize seeds at 4 °C for 3–4 days, and then transfer to soil, putting a few seeds at each corner of pots. When germinated, removed seedlings so that one seedling remains in each corner. 3. Allow to grow for 3–4 weeks under greenhouse conditions or in a growth chamber. 4. Harvest one leaf from an individual plant using a pair of clean scissors, then use the lid of a microcentrifuge tube to punch out a disk (simply by closing the lid on the leaf). Repeat for each plant to be screened, as well as a WT control. 5. Use the REDExtract-N-AmpTM Plant PCR kit to extract DNA from the leaf disk (according to the manufacturer’s instructions). Briefly, add 100 mL of Extraction Solution to each tube, and push the leaf disk to the bottom of the tube using the pipette tip, such that it is completely covered by the Extraction Solution. Place in 95 °C water bath for 10 min, remove from heat, and add 100 mL of Dilution Solution (Sigma), vortex. If necessary, store sample for 1–2 days at 4 °C, or at −20 °C for longer storage times. 6. Use the REDExtract-N-AmpTM Plant PCR kit to amplify the DNA region of interest (according to the manufacturer’s instructions). It is recommended to use a 3-primer strategy, whereby the two flanking primers, LP and RP, are used in combination with a third T-DNA-specific primer (LB) (see Note 1). 7. Run PCR products on a 0.8 % agarose gel. Identify lines that have a band corresponding to the expected size for LB + RP


20

Laser Microdissection and Genetic Manipulation Technologies…

235

(T-DNA), and no band corresponding to the expected size of LP + RP (WT). Sequence of each T-DNA specific product was confirmed by DNA sequencing. 8. Keep each individual plant that is confirmed homozygous for the T-DNA insert (see below) and discard all of the other plants. If no homozygous plants were identified, then keep a heterozygous plant and repeat the screening steps (3–8) for the next generation. Grow plant(s) to maturity (~10 weeks) and harvest seed. 9. To confirm that a given homozygous T-DNA insertion line is a knockout, extract the mRNA from an individual plant from each T-DNA insertion line and from a WT plant using a QIAgen RNeasy kit, according to the manufacturer’s instructions. Reverse-transcribe into cDNA using Superscript IITM reverse transcriptase for use as a PCR template for each line. Amplify a unique region of the coding sequence and/or untranslated region using gene-specific primers using the cDNA from either WT or the T-DNA insertion line for a template. Also set up a control reaction, using a “housekeeping gene” such as actin, tubulin, or elongation factor. The WT sample should have both the control and gene-specific PCR products present, while the T-DNA insertion line should have the control, but no gene-specific PCR product. 10. If double KOs are to be generated, plant seeds of confirmed homozygous T-DNA insertion lines, as described above. Shortly after stems begin to bolt (usually 3–4 weeks), the lines will be ready to cross. From one line, select a stem and remove (using needle-nose forceps) all of the siliques and all flowers in which the stamen has grown longer than the petals. From the remaining unopened flowers, remove the sepals, petals and anthers (leaving only the stamen) from six to eight different flowers on a single stem. This allows flowers with a range of developmental stages to be pollinated. From the opposite line, select a single flower and remove the sepals, petals and stamen, leaving only the anthers. Remove the flower from the plant, holding the stem at the base of the flower with the forceps, using it like a paintbrush to gently transfer pollen to the tips of the stamens of the other line. Perform at least three to four reciprocal crosses between the two lines, and tie each of the selected stems to a small wooden stake, labeled with the relevant information for the two lines. Allow plants to grow to maturity, and collect seeds only from the siliques that develop from those selected stems. 11. Plant the seeds resulting from each cross, and repeat steps 1–9 to identify double heterozygous lines (containing one T-DNA insert in each gene of interest). Plant those double heterozygous seeds to screen for double homozygous lines (steps 1–9 once more).


236

O.R.A. Corea et al.

3.2. Sampling and Phenotypic Analysis

1. For amino acid analysis (Subheading 3.4), five flats (18 pots per flat, 4 plants per pot) of Arabidopsis per line are required. Due to their smaller size, a greater number of plants must be harvested for the 3.5 week time point (one flat of plants) in order to obtain enough root and stem tissue for analysis. For each of the following weeks (4–10), half a flat of plants each week will yield a sufficient amount of tissue (see Note 2). 2. Harvest Arabidopsis leaves, roots and stems from each plant. Wash leaves and roots thoroughly in water, and pat dry with a paper towel. For stems, harvest only the main inflorescence stem, remove all leaves, siliques, and flowers. Pool each tissue for each line and transfer to separate 15 mL Falcon tubes. Keep on ice until all individual plants are harvested, then flash-freeze in liquid N2. Store at –80 °C then freeze-dry for 3 days. 3. For lignin analyses (Subheading 3.6), 12 flats (18 pots per flat, 4 plants per pot) of Arabidopsis per line are used. The main Arabidopsis inflorescence stems should be harvested weekly from after initial stem emergence up to maturity (~3.5 to 10 weeks). Again, a greater number of stems must be harvested for the 3.5-, 4-, and 5-week time points due to smaller amounts of material available (3, 2, and 1.5 flats of plants, respectively). For each of the following weeks (6–10), harvest one flat of plants each week. Transfer the whole stems to sealable plastic bags, and store at 4 °C for up to a week. 4. Use the stems collected for lignin analysis to perform the phenotypic analysis; measure the diameter (using Vernier calipers), weights, and lengths of 20 inflorescence stems from each line, for each weekly time point. 5. Cut stems into 0.5–1 cm long pieces, lyophilize, and store at room temperature prior to lignin analyses.

3.3. Histochemical Staining 3.3.1. Phloroglucinol–HCl: Staining G-Derived Lignin Constituents (4–6)

1. Make phloroglucinol stock solution by adding phloroglucinol (10 mg) to 95 % EtOH (50 mL) in a 50 mL Falcon tube. This solution is light sensitive; wrap in aluminum foil to store for up to 1 month. 2. Hand-section freshly-harvested main Arabidopsis inflorescence stems using a double edge stainless steel blade. 3. Place in PBS buffer (1×) to store samples. 4. Remove air bubbles by placing samples under vacuum (30 min to 1 h) (see Note 3). 5. Rinse cross-sections with H2O in a 1.5 mL microcentrifuge tube. 6. Transfer cross-sections to a reagent reservoir using a plastic transfer pipette, remove excess H2O. 7. Make phloroglucinol–HCl solution by mixing phloroglucinol stock solution (10 mL) and concentrated HCl (1.6 mL).


20

Laser Microdissection and Genetic Manipulation Technologies…

237

Add to cross-sections and incubate for 30 min. Reaction is light-sensitive, so cover with aluminum foil (see Note 4). 8. Place stained cross-sections on the glass microscopy slide and carefully apply slide cover. 9. Remove air bubbles, add phloroglucinol–HCl solution if needed. 10. Observe cross-sections under a light microscope using bright field or differential interference contrast. 3.3.2. Mäule Reagent: Staining S-Derived Lignin Constituents (6–8)

1. Hand-section freshly-harvested main Arabidopsis inflorescence stems using a double edge stainless steel razorblade. 2. Place in PBS buffer (1×) to store samples. 3. Remove air bubble by placing samples under vacuum (30 min to 1 h). 4. Fix in glutaraldehyde (4 % in H2O) for 60 min. 5. Rinse with deionized H2O (three times); let samples stand for 10 min (see Note 5). 6. Rinse cross-sections with H2O in a 1.5 mL microcentrifuge tube. 7. Transfer cross-sections to a reagent reservoir using a plastic transfer pipette, remove excess H2O. 8. Immerse cross-sections in KMnO4 (0.5 %) for 10 min (see Note 6). 9. Rinse cross-sections with H2O. 10. Treat with hydrochloric acid (10 %) for 5 min. Rinse crosssections with H2O. 11. Mount cross-sections on glass microscopy slide and add concentrated NaOH. 12. Observe cross-sections under a light microscope using bright field or differential interference contrast.

3.4. Amino Acid Analysis 3.4.1. Amino Acid Extraction

1. Grind each lyophilized tissue sample (see Subheading 3.2) at room temperature using a mortar and pestle until sample becomes a homogeneous powder. 2. Weigh samples in triplicate (3–10 mg each) into 1.5 mL microcentrifuge tubes. 3. Add 500 mL MeOH:CH2Cl2:H2O (12:5:3) containing norvaline (2 mM), vortex, then centrifuge for 10 min at 13,800 × g. Transfer supernatant to a new microcentrifuge tube. 4. To the remaining tissue, add 500 mL MeOH:CH2Cl2:H2O (12:5:3) and repeat the extraction, as before. Combine the supernatant with that of the first extraction.


238

O.R.A. Corea et al.

5. Back wash each sample by addition of 300 mL chloroform, followed by 400 mL H2O. Vortex, then centrifuge for 10 min at 13,800 × g. 6. Transfer the upper aqueous phase to a new 1.5 mL microcentrifuge tube. 7. Dry down in a Speed-Vac. 3.4.2. Derivatization and Analysis of Amino Acids

1. To dry sample, add borate buffer (from the AccQ•Tag kit, 80 mL) and vortex briefly, then centrifuge at top speed for 2 s. 2. Transfer to glass UPLC recovery vials, add freshly prepared AccQ•Tag derivatization reagent (20 mL). 3. Move vials to heating block (55 °C) for 10 min. Allow to cool to room temperature. 4. Inject 1 mL in UPLC and separate amino acids over the following gradient of Eluent A:Eluent B; 0–0.54 min (0.1% B), 0.54– 5.74 min (0.1 to 7.1% B), 5.74–7.74 min (7.1 to 17% B), 7.74–9.04 min (17 to 32% B), 9.04–9.55 min (32 to 90% B), 9.55–10.14 (90% B), with column temperature set to 57 °C (see Note 7). A photodiode array detector was used over a fluorescence detector as it allows detection of Trp under the following conditions: Absorbance at 260 nm.

3.5. Pyrolysis GC-MS Analysis of Vascular Bundles and Interfascicular Regions (9)

1. Harvest ~1 to 2 cm section from the base of an Arabidopsis stem, place in a 1.5 mL microcentrifuge tube and flash-freeze in liquid N2. If using samples immediately, keep in liquid N2; alternatively, they may be transferred to a –80 °C freezer for storage (see Note 8).

3.5.1. Cryotome Cutting of Stem Cross-Sections and Laser Microscope Dissection Thereof

2. Activate a new PALM membrane slide for 45 min under UV light (254 nm). Use treated slides within 10 min (see step 7). If they cannot be used in that time then they must be re-treated under UV light for an additional 45 min. 3. Start the cryotome and set the temperature (–16 to –20 °C). 4. Put O.C.T on the holder (inside the cryotome) and put the frozen sample (Arabidopsis stem) into the O.C.T. If there are bubbles in the O.C.T., move them away from the sample using a pair of forceps. 5. Put the holder into the cryotome and cut (10–16 mm thickness). 6. Lower UV-activated membrane on the PALM slide to touch circa 20 cross-sections; the latter then adhere onto the membrane upon contact. 7. Remove O.C.T. by rinsing with chilled (4 °C) distilled H2O (2 × 30 s). 8. Dry slides under a fume hood for 5 min, then transfer to −20 °C freezer for storage.


20

Laser Microdissection and Genetic Manipulation Technologies…

239

9. Prior to using for microdissection, dip the slide in chilled (4 °C) 70 % EtOH for 1 min. Repeat with chilled 95 % EtOH, and twice more with 100 % EtOH. Allow slide to air dry for 5 min. 10. Start laser microdissection procedure according to the manufacturer’s instructions. 11. Individually dissect either vascular bundle or interfascicular regions and collect microdissected tissues into the caps of 200 mL microcentrifuge tubes containing ~50 mL LC/ MS-grade H2O. 12. Centrifuge the tubes to collect the target tissue in the bottom and place in −80 °C until needed. Tissues collected at different times can be thawed and pooled together in a glass vial prior to freeze-drying. 3.5.2. Pyrolysis GC-MS (9)

1. Extract freeze-dried vascular bundle or interfascicular fiber tissue samples (Subheading 3.5.1) with H2O:acetone (3:7, v/v) twice for 12 h with constant shaking at 25 °C of the glass vial. After centrifugation (1000 × g, 10 min) remove H2O:acetone solution. Add 100 mL LC/MS grade H2O, flash freeze in liquid N2 and freeze dry. 2. Resuspend tissue samples in acetone (100 mL), taking care to rinse all tissue into the bottom of the vial. Transfer the entire sample to the pyrolysis sample cup, 5 mL at a time using a Micro/Pettor®. Allow the acetone to evaporate before adding the next aliquot. 3. Set the pyrolysis gas chromatograph conditions as follows: Pyrolysis microfurnace temperature = 500 °C, pyrolysis:GC interface temperature = 270 °C, GC gradient conditions to 50 °C for 1 min, followed by a 5 °C/min increase from 50 to 270 °C. 4. Load the sample cup into the pyrolyzer, close the lid and allow to equilibrate for 5 min. Initiate pyrolysis by releasing the sample cup into the microfurnace and then open the purge flow after 4 s to allow the vapor phase products to enter the GC-MS. 5. Separate pyrolysis products on the GC/MS, and identify products by comparison to previously reported mass fragmentation data (10, 11) and/or authentic standards.

3.6. Cell Wall Residue Extraction and Lignin Analyses 3.6.1. Cell Wall Residue (CWR) Extraction (12)

1. Individually grind each freeze-dried stem sample (see Subheading 3.2) by ball milling for 2 h at 400 rpm in a Planetary Ball Mill PM 400 (Retsch) using agate bowls and balls. 2. Successively extract each resulting powder with continuous stirring, at room temperature for 12 h each with EtOH:toluene (1:1, v/v, 100 mL/g), EtOH (95 %, 100 mL/g) and then twice with H2O (100 mL/g).


240

O.R.A. Corea et al.

3. After the last H2O extraction, recover the extractive-free CWR of each line by centrifugation (7,200 × g, 15 min), freeze the residue at −80 °C and freeze-dry. 3.6.2. Acetyl Bromide (12, 13)

1. Transfer CWR samples (4–6 mg) to 4 mL glass vials containing 2.5 mL AcBr:glacial acetic acid (5.4:25.7, v/v) and 0.1 mL perchloric acid (70 %) (see Note 9). 2. Seal each vial with a PTFE-coated silicon septa, vortex briefly, and place in an oven at 70 °C for 30 min. Vortex each vial briefly at 10 min intervals during incubation. 3. Transfer solutions to volumetric flasks (50 mL) containing 2 M NaOH (10 mL) and glacial acetic acid (12 mL). 4. Rinse each vial with a minimal amount of acetic acid and transfer to the flask. Make up the solution to 50 mL with glacial acetic acid. 5. Subject each sample to UV analysis using a spectrophotometer, reading at 280 nm. Calibrate against the absorbance of a blank solution (lacking only the sample). 6. Calculate the lignin contents using an absorptivity value for lignin of 20.09/g L/cm for the absorbance at 280 nm. 7. Take average value of duplicate samples to determine the absorptivity of each sample. 8. Percent lignin content may be determined using the following equation: Lignin Content (%) = (100 × absorptivity × 0.05)/(Weight of sample (mg) × 0.001 × 20.09).

3.6.3. Thioacidolysis (3, 14)

1. Weigh individual CWR samples (10–15 mg) into 20 mL glass tubes and add thioacidolysis reagent to each (15 mL, 0.2 M BF3 diethyl etherate in 8.75:1 (v/v) dioxane:ethanethiol). Fit each tube with Teflon-lined screw-cap under nitrogen atmosphere. 2. Perform the thioacidolysis reaction by placing each sample in a heating block (100 °C) for 4 h with constant stirring with magnetic stir bar. 3. After cooling the tubes in ice-water, pour each reaction mixture into CH2Cl2 (20 mL) in a 300 mL beaker. Rinse each reaction tube with H2O (3 × 5 mL) with the aqueous solutions added to the beaker. 4. Add the internal standard (tetracosane C24, 20 mL of 25 mg/ mL CH2Cl2 solution) then adjust each aqueous phase to pH 3–4 with ~6–7 mL of 0.4 M NaHCO3. 5. Next, separate each organic phase by decantation, with the resulting aqueous phases extracted with CH2Cl2 (3 × 20 mL); dry the combined organic phases over anhydrous Na2SO4 and


20

Laser Microdissection and Genetic Manipulation Technologies…

241

then evaporate under reduced pressure at 40 °C using a rotary evaporator. 6. Dissolve the residue of each thioacidolysis product in CH2Cl2 (1.5 mL). Transfer an aliquot (20 mL) to an insert of a GC-MS glass vial and derivatize at room temperature with N,O bis (trimethylsilyl) trifluoroacetamide + trimethylchlorosilane (BSTFA + TMCS, 99:1, 50 mL) in pyridine (20 mL). 7. Inject 1 mL of the resulting derivatives to a HP 6890 Series GC System (Hewlett Packard) equipped with a Rtx® 5-SiL-MS column (30 m × 0.25 mm × 0.25 mm, Restek). Identify products by reference to retention times of authentic silylated standards, and by mass spectrometric analyses using the HP 5973 MS detector (Hewlett Packard, EI mode, 70 eV). The column conditions are as follows: 140 °C for 5 min and then 140– 280 °C at 8 °C/min until the final temperature, this being held at 280 °C for a further 7.5 min. Splitless mode. All samples should be analyzed in duplicate (see Note 10).

4. Results and Discussion For illustrative purposes of our general approach, the role of a biosynthetic step modulating carbon flow into the phenylpropanoid pathway was assessed (15). Specifically, the six-membered gene family, encoding the ADTs which perform the conversion of arogenate to phenylalanine (16), was analyzed for potential differential contributions into these downstream pathways, such as to the monolignols/lignins. Through the generation of single, double, triple, and quadruple KOs in Arabidopsis, it was thus possible to identify specific KO lines that were potentially lignin deficient. By cross-comparison between these various mutant lines, deduction of which ADT isoforms were participating in monolignol metabolic networks as regards differential lignin deposition could be made. Generation of each ADT KO line was performed following the PCR-based screening approach described in Subheading 3.1. Homozygous lines were identified as having an amplified PCR product from the T-DNA-specific primer set, and an absence of an amplified product from the WT-specific primer set. Subsequently, the PCR products were sequenced to verify the presence and location of each T-DNA insert, and finally RT-PCR was used to verify that the corresponding mRNA of each T-DNA insertion line was, indeed, absent. The final RT-PCR screen is of particular importance when the T-DNA insertion lies within a putative promoter, untranslated region (UTR), or intron of the gene, as mRNA may (rarely) be transcribed when the T-DNA insertion is not in an exon. In this analysis, five out of the six homozygous T-DNA


242

O.R.A. Corea et al.

insertion lines had abolished mRNA for the corresponding ADTs, and were considered to be KO lines, whereas ADT2, which had a T-DNA insert in the putative promoter region, still had its corresponding mRNA. Thus, the five verified ADT KO lines were crossed together and screened (see Subheading 3.1) to obtain double ADT KO lines (e.g., adt1×adt3 to obtain adt1/3). This process was repeated again (using double KOs) to obtain triple KO lines (e.g., adt1/3 × adt3/5 to obtain adt1/3/5), and likewise, using triple KOs to obtain one quadruple KO lines (e.g., adt3/4/5 × adt4 /5/6 to obtain adt3/4/5/6). Knockout lines having potential reductions in lignin contents were then identified based on having reduced stem lengths, weights and/or widths over the course of growth and development, as compared to WT. Several lines were identified as having slightly reduced stem size and mass compared to WT, and also had distinctly different overall stature (Fig. 1a, d, g, j). A histochemical approach was used to test if these changes in stem properties were indeed due to a change in the lignin contents. Phloroglucinol–HCl and Mäule reactions were used, respectively, to preferentially qualitatively detect G- and S-derived lignin components in stem crosssections. For WT tissue, reaction with phloroglucinol–HCl (with presumed G-lignin derived constituents) occurred throughout the vascular bundle, vb, and interfascicular fiber, if, regions (Fig. 1b). By contrast, there was a slight reduction of coloration in the if region in one single KO line, adt5 (Fig. 1e), with this being more pronounced in the double KO, adt4/5 (Fig. 1h). In the triple KO, adt3/4/5 (Fig. 1k), coloration from the phloroglucinol reaction was all but absent in the if. Interestingly, the Mäule reaction (with presumed S-lignin constituents) occurred throughout the vb and if of WT (Fig. 1c), as well as those of adt5, adt4/5, and adt3/4/5 (Fig. 1f, i, l, respectively), with no apparent difference in staining intensity between these lines. These data thus qualitatively suggested more pronounced effects on G-lignin deposition. While the qualitative approach described above was useful for identifying potential candidate lines with reduced lignin contents/ compositions, it was necessary to perform quantitative lignin analyses on each line. This was in order to assess the impact on lignin biosynthesis over the course of stem maturation. Thus, for both WT and ADT KO lines, cell wall residues (CWRs) were individually prepared from stem tissue collected weekly over the course of stem growth and development. Amounts and monomeric compositions of lignins within CWRs were estimated using acetyl bromide and thioacidolysis analyses, respectively. For the WT line (Fig. 2a, e, ) both G and S-releasable monomer levels and lignin contents steadily increased during growth and development until reaching maximum levels at maturation/senescence (15) as previously noted in other studies (6, 9, 12, 17, 18). By contrast, the adt5 line (Fig. 2a, e, ) had an overall estimated lignin content reduced to


20

Laser Microdissection and Genetic Manipulation Technologies…

243

Fig. 1. Phenotypes and histochemical staining of WT and ADT knockout lines. (a–c) WT, (d–f) adt5 single knockout line, (g–i) adt4/5 double knockout line, ( j–l) adt3/4/5 triple knockout line. Whole plant phenotypes (a, d, g, j) show a progressively greater prostrate phenotype in the single, double and triple knockouts, compared to WT. Similarly, stem cross-sections treated with phloroglucinol–HCl (b, e, h, k) show reduced intensity of coloration in the interfascicular fiber (if) regions in the double and triple knockouts, while there is no apparent difference in the vascular bundles (vb). By contrast, stem crosssections treated with Mäule reagent (c, f, i, l) do not display an obvious decrease in the intensity of coloration in either the vb or if.

circa 80 % of WT levels at maturation, with concomitant reductions in G:S monomer release. However, interestingly, the reduction was essentially only detectable in G-monomer component release, since S-monomer amounts were either at WT levels and/ or were slightly increased at maturation. Additionally, with double mutant, adt4/5, the reductions in estimated lignin amounts were much more pronounced to circa 61 % of WT levels (Fig. 2b, f, ).


244

O.R.A. Corea et al.

Fig. 2. Comparison of thioacidolysis-determined G- and/or S-lignin-derived monomers contents vs. total “AcBr lignin” contents, for WT, and ADT KO mutant lines, adt5, adt4/5, adt3/4/5, and adt3/4/5/6. G-derived thioacidolysis degradation products compared to total “AcBr-lignin” (a–d) and S-derived thioacidolysis degradation products compared to total “AcBrlignin” (e–h).


20

Laser Microdissection and Genetic Manipulation Technologiesâ&#x20AC;Ś

245

Nevertheless, the amounts of releasable G and S monomers increased linearly with increasing lignin levels during growth/ development; however, in this case, levels of overall S releasable monomers at maturation were also essentially identical to WT. That is, the adt4/5 double mutant again gave a lignin-reduced phenotype, but whose major quantifiable effect was on the cleavable G-monomer content levels. For the adt3/4/5 mutant (Fig. 2c, g, ), the effect was even more dramatic; estimated AcBr lignin contents were reduced to circa 50 % of WT levels, whereas the thioacidolysis releasable G- and S-monomers plateaued at circa 20 and 58 %, respectively. That is, while the adt3/4/5 mutant was still able to produce G- and S-derived monomers/lignins, the overall reduction in carbon flux to the lignin pathway resulted in an accompanying reduction in both G- and S-derived monomers/ lignins. Additionally, somewhat comparable results (Fig. 2d, h, ), were obtained with the quadruple mutant adt3/4/5/6. In that case, the phenotype also had a decreased lignin content at maturation of circa 32 % relative to WT, and reductions in G- and S-releasable monomers to circa 21 and 67 % of WT, respectively. These different analyses and correlations, taken over numerous data time points to identify trends during growth/development for each mutant line, thus also established that quite distinct G:S ratios (from ~4:1 to ~1:1) could be attained by differentially manipulating the genes encoding ADTs; these manipulations also differentially modulated carbon flux into the monolignol/lignin pathways. In this regard, WT lines at maturation contain fiber cells (which are abundant in the if regions) rich in S-lignin, while xylem cells (found in the vb) are mainly G-rich (9). In the vb, the cell wall types mainly contain G-rich vessels together with smaller levels of xylary fibers, xf, which are S-rich. Thus, it was instructive to ascertain whether there was any differential effect on individual cell wall lignin compositions in the mutant lines in these distinct anatomical regions/cell wall types, relative to WT. Accordingly, to assess potential differences in lignin contents/compositions in different tissues/cell wall types, the mutant adt1/4/5 was subjected to laser microdissection of its vb and if regions at 7 weeks growth/development, followed by their analysis by pyrolysis GC/MS, with these results compared to WT. Pyrolysis GC/MS analyses of both tissue types in the triple mutant and WT lines thus demonstrated significant differences: for the G-rich vascular bundles in adt1/4/5, the H-, G-, and S-derived lignin pyrolysis products were clearly evident (Fig. 3b and Fig. 4) except that, relative to the H-derived components, both G and S were significantly reduced in abundance as compared to the WT vb tissue (9). By contrast, for the if regions (Fig. 3a and Fig. 4), relative to H-derived pyrolysis products, the largest reduction was in G-components over that of S-monomer release when compared to if WT pyrolysis products. We currently rationalize these findings to reflect differential carbon


246

O.R.A. Corea et al.

Fig. 3. Pyrolysis GC/MS chromatograms of laser-microdissected interfascicular regions (a) and vascular bundles (b) from adt1/4/5.

allocation into these distinct regions/cell wall types in the adt145 KO line, compared to the if (Fig. 3) and to WT samples from a prior analysis (9). Together, these analyses thus have consistently demonstrated that specific ADT isoenzymes are centrally involved in modulating carbon flux into downstream monolignols/lignins. Amino acid analyses: Given the pivotal position of ADT in modulating carbon flux into phenylalanine/phenylpropanoid metabolism vs. that of other aromatic amino acids [tryptophan (Trp) and tyrosine (Tyr)], and other products, such as proteins,


20

Laser Microdissection and Genetic Manipulation Technologiesâ&#x20AC;Ś

Fig. 4. Pyrolysis GC/MS lignin-derived products.

247


248

O.R.A. Corea et al.

Fig. 5. Amino acid derivatives analyzed by UPLC. (a) Standard mixture (500 pmol), as well as asparagine, glutamine, norvaline, and tryptophan (each at 125 pmol). (b) Amino acids extracted from 6 week old Arabidopsis rosette leaves. Peaks are numbered as follows: 1. 6-Aminoquinoline (AMQ; derivatization by-product), 2. NH3+, 3. Histidine, 4. Asparagine, 5. Serine, 6/7. Glutamine/Arginine, 8. Glycine, 9. Aspartate, 10. Glutamate, 11. Threonine, 12. Alanine, 13. Proline, 14. Derivatization by-product, 15. Cysteine, 16. Lysine, 17. Tyrosine, 18. Methionine, 19. Valine, 20. Norvaline (internal standard), 21. Isoleucine, 22. Leucine, 23. Phenylalanine, 24. Tryptophan.

glucosinolates, etc., it will be important in future to comprehensively establish whether amino acid metabolism is also affected in term of amino acid pools in the tissues and distinct cell wall types. To address this question, a facile high throughput analysis of the 20 protein amino acids has been developed, thereby permitting both the analysis and quantification of all amino acid pools in these tissues in both WT and mutant lines at all stages of growth/development, with the exception of glutamine and arginine which coelute under the conditions described. As one example, Fig. 5a shows the separation of a standard mixture of 20 protein amino acids and norvaline (internal standard), derivatized using the 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) reagent (Waters), and separated using an UPLC (see Subheading 3.4.2). Amino acids extracted from Arabidopsis leaves (Subheading 3.4.1) are also readily detected (Fig. 5b); comparison of WT and adt KO lines will


20

Laser Microdissection and Genetic Manipulation Technologies…

249

provide important insight into how the aromatic amino acid levels are regulated in plants. Taken together, generation of the T-DNA knockouts together with their comprehensive analyses are now providing the path forward to rapidly identify new plant lines more suitable for bioenergy/bioproducts feedstock, as well as enabling a better understanding of their metabolic processes associated with same in planta.

5. Notes 1. If screening a line containing multiple T-DNA inserts in different members of a gene family, it may not be feasible to use the 3-primer PCR strategy. Instead, perform two 2-primer PCR reactions; LP + RP for WT, and LB + RP for the T-DNA insert. 2. After senescence at ~8 to 9 weeks, harvest only the stem and root tissues, as leaves will become desiccated. 3. Skip steps 2–4 if cross-sections will be used the same day. 4. Make phloroglucinol–HCl solution fresh each time. 5. Skip steps 2–4 if cross-sections will be used the same day. 6. Filter KMnO4 with MillexHV 0.45 mm prior to use. 7. Store concentrated Eluent A and Eluent B at 4 °C. Just prior to use, Eluent A should be diluted 10× with LC/MS grade H2O. Both diluted Eluent A and Eluent B should not be used if kept at room temperature for longer than 2 days. 8. Frozen sections are the method of choice for most immunohistochemistry and immunofluorescent applications, since the cross-linking of surface antigens by formalin fixation can cause trouble with specific antigen binding by antibodies. 9. Use glass pipettes for adding all solvents; do not allow any contact with metal or plastic. 10. For quantitative analyses, authentic standards (H-CHSEtCHSEt-CH2SEt, G-CHSEt-CHSEt-CH2SEt, and S-CHSEtCHSEt-CH2SEt) were used to calculate the response factors for monomeric H, G, and S units, respectively.

Acknowledgments This work was supported by the Chemical Sciences, Geosciences, and Biosciences Division, Office of Basic Energy Sciences, Office of Science, US Department of Energy, by the BioEnergy Science Center (DE-FG-0397ER20259), the US Department of Energy


250

O.R.A. Corea et al.

Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science, as well as the G. Thomas and Anita Hargrove Center for Plant Genomic Research. References 1. Davin LB, Jourdes M, Patten AM, Kim K-W, Vassão DG, Lewis NG (2008) Dissection of lignin macromolecular configuration and assembly: comparison to related biochemical processes in allyl/propenyl phenol and lignan biosynthesis. Nat Prod Rep 25:1015–1090 2. Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen H, Shinn P, Stevenson DK, Zimmerman J, Barajas P, Cheuk R, Gadrinab C, Heller C, Jeske A, Koesema E, Meyers CC, Parker H, Prednis L, Ansari Y, Choy N, Deen H, Geralt M, Hazari N, Hom E, Karnes M, Mulholland C, Ndubaku R, Schmidt I, Guzman P, AguilarHenonin L, Schmid M, Weigel D, Carter DE, Marchand T, Risseeuw E, Brogden D, Zeko A, Crosby WL, Berry CC, Ecker JR (2003) Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301:653–657 3. Lapierre C, Monties B (1986) Thioacidoylsis of poplar lignins: identification of monomeric syringyl products and characterization of guaiacyl-syringyl lignin fractions. Holzforschung 40:113–118 4. Wiesner J (1878) Note über das Verhalten des Phloroglucins und einiger verwandter Körper zur verholzten Zellmembran. Sitz Kaiserlichen Akad Wissensch Math-Naturwissensch Classe 77:60–66 5. Ruzin SE (1999) Plant microtechnique and microscopy. Oxford University Press, New York, NY 6. Patten AM, Cardenas CL, Cochrane FC, Laskar DD, Bedgar DL, Davin LB, Lewis NG (2005) Reassessment of effects on lignification and vascular development in the irx4 Arabidopsis mutant. Phytochemistry 66:2092–2107 7. Mäule C (1901) Das Verhalten verholzter Membranen gegen Kalium-permanganat, eine Holzreaktion neuer Art. Beiträge zur wissenschaftlichen Botanik 4:166–185 8. Nakano J, Meshitsuka G (1992) The detection of lignin. In: Lin SY, Dence CW (eds) Methods in lignin chemistry. Springer, Berlin, pp 23–32 9. Patten AM, Jourdes M, Cardenas CL, Laskar DD, Nakazawa Y, Chung B-Y, Franceschi VR, Davin LB, Lewis NG (2010) Probing native lignin macromolecular configuration in Arabidopsis thaliana in specific cell wall types: further insights into limited substrate degeneracy

and assembly of the lignins of ref8, fah 1-2 and C4H::F5H lines. Mol BioSyst 6:499–515 10. Faix O, Meier D, Fortmann I (1990) Thermal degradation products of wood: a collection of electron-impact (EI) mass spectra of monomeric lignin derived products. Holz RohWerkst 48:351–354 11. Faix O, Meier D, Fortmann I (1990) Thermal degradation products of wood. Gas chromatographic separation and mass spectrometric characterization of monomeric lignin-derived products. Holz Roh-Werkst 48:281–285 12. Jourdes M, Cardenas CL, Laskar DD, Moinuddin SGA, Davin LB, Lewis NG (2007) Plant cell walls are enfeebled when attempting to preserve native lignin configuration with poly-p-hydroxycinnamaldehydes: evolutionary implications. Phytochemistry 68:1932–1956 13. Iiyama K, Wallis AFA (1990) Determination of lignin in herbaceous plants by an improved acetyl bromide procedure. J Sci Food Agric 51:145–161 14. Rolando C, Monties B, Lapierre C (1992) Thioacidolysis. In: Lin SY, Dence CW (eds) Methods in lignin chemistry. Springer, Berlin, pp 334–349 15. Corea ORA, Ki C, Cardenas CL, Kim SJ, Brewer SE, Patten AM, Davin LB, Lewis NG (2012) Arogenate dehydratase isoenzymes profoundly and differentially modulate carbon flux into lignins. J Biol Chem 287:11446–11459 16. Cho M-H, Corea ORA, Yang H, Bedgar DL, Laskar DD, Anterola AM, Moog-Anterola FA, Hood RL, Kohalmi SE, Bernards MA, Kang C, Davin LB, Lewis NG (2007) Phenylalanine biosynthesis in Arabidopsis thaliana: identification and characterization of arogenate dehydratases. J Biol Chem 282:30827–30835 17. Kim S-J, Kim M-R, Bedgar DL, Moinuddin SGA, Cardenas CL, Davin LB, Kang C, Lewis NG (2004) Functional reclassification of the putative cinnamyl alcohol dehydrogenase multigene family in Arabidopsis. Proc Natl Acad Sci U S A 101:1455–1460 18. Moinuddin SGA, Jourdes M, Laskar DD, Ki C, Cardenas CL, Zhang D, Davin LB, Lewis NG (2010) Insights into lignin primary structure and deconstruction from Arabidopsis thaliana COMT (caffeic acid O-methyltransferase) mutant Atomt1. Org Biomol Chem 8:3928–3946


Chapter 21 Lignin-Degrading Enzyme Activities Yi-ru Chen, Simo Sarkanen, and Yun-Yan Wang Abstract Over the past three decades, the activities of four kinds of enzyme have been purported to furnish the mechanistic foundations for macromolecular lignin depolymerization in decaying plant cell walls. The pertinent fungal enzymes comprise lignin peroxidase (with a relatively high redox potential), manganese peroxidase, an alkyl aryl etherase, and laccase. The peroxidases and laccase, but not the etherase, are expressed extracellularly by white-rot fungi. A number of these microorganisms exhibit a marked preference toward lignin in their degradation of lignocellulose. Interestingly, some white-rot fungi secrete both kinds of peroxidase but no laccase, while others that are equally effective express extracellular laccase activity but no peroxidases. Actually, none of these enzymes has been reported to possess significant depolymerase activity toward macromolecular lignin substrates that are derived with little chemical modification from the native biopolymer. Here, the assays commonly employed for monitoring the traditional fungal peroxidases, alkyl aryl etherase, and laccase are described in their respective contexts. A soluble native polymeric substrate that can be isolated directly from a conventional milled-wood lignin preparation is characterized in relation to its utility in next-generation lignin-depolymerase assays. Key words: Alkyl aryl etherase, Laccase, Laccase mediator, Light scattering, Lignin, Lignin depolymerase, Lignin peroxidase, Manganese peroxidase, Reactive oxygen species, Versatile peroxidase, White-rot fungi

1. Introduction Lignins are the most recalcitrant of the biopolymers in vascular plant cell walls. They vary quite widely in cell-wall content (15â&#x20AC;&#x201C; 35 % w/w) and macromolecular configuration (i.e., monomer-unit composition and substructure frequency). They are assembled through the dehydrogenative coupling of p-coumaryl alcohol and its 3-methoxy and 3,5-dimethoxy derivatives (coniferyl and sinapyl alcohol) in varying proportions. By these means, coumaryl, guaiacyl, and syringyl units are respectively linked together through six or more different substructures in native lignin macromolecules.

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3_21, Š Springer Science+Business Media, LLC 2012

251


252

Y.-r. Chen et al.

Softwoods (coniferous gymnosperms) possess lignins composed primarily of guaiacyl monomer residues, while hardwoods (deciduous angiosperms) contain lignins with guaiacyl and syringyl units in similar proportions. On the other hand, the lignins of grasses (monocots) are composed of monomer residues formed in varying ratios from all three monolignols (1). Generally speaking, lignin composition and structure can change profoundly during the course of plant growth and development (2, 3). Yet it is possible, by simple physicochemical means alone, to separate the lignins in plant cell walls into different welldefined fractions with contrasting macromolecular configurations (4). Moreover, lignins are covalently linked through about 3 % of their monomer units to hemicellulose chains (5); there is a relationship between the constituent monosaccharide residues and the substructures in the lignin macromolecules to which they are bound (Fig. 1). 1.1. Lignin-Degrading Enzyme Assays

As far as the efficacies of candidate-enzymes for lignin degradation are concerned, there is no consensus about the polymeric lignin substrate that should be used in depolymerase assays. The difficulty appeared already in 1983, when the activity of a putative lignindegrading enzyme was first detected in extracellular culture solutions of the white-rot fungus, Phanerochaete chrysosporium (7, 8). There were two problems with that inaugural study. The methylated aqueous acetone spruce wood extract employed as a substrate was not actually a polymeric lignin preparation. Then, just 3 years later, lignin peroxidase, the enzyme conspicuously present in the extracellular P. chrysosporium culture solution, was found to polymerize alkali-isolated straw lignin and spruce milled-wood lignin (9) in the presence of H2O2 at pH 4.0 (Fig. 2). Such apparent contradictions have confounded the field of lignin biodegradation for more than two decades. The types of enzymes given most attention as plausible agents of lignin degradation are indeed poised between cleavage and polymerization in their effects on the macromolecular substrate. For example, lignin peroxidase has a sufficiently high redox potential to oxidize nonphenolic aromatic rings of the veratryl type in lignin substructures to cation radicals. In nonphenolic alkyl aryl b-O-4 linked lignin model dimers (Fig. 3), the resulting homolytic Caâ&#x20AC;&#x201C;Cb cleavage leads to a Ca-benzaldehyde derivative (as the major initial product) and a phenol formed when the Cbâ&#x20AC;&#x201C;O bond is subsequently broken (10). This phenol is only found in trace amounts, however, because it is much more easily oxidized than nonphenolic aromatic lignin units. Coupling of the ensuing phenoxy radicals in lignin preparations tends to polymerize the substrate and thus counteract the initial degradative effects. Limitingly low substrate concentrations have been employed in an attempt to minimize such phenoxy radical coupling, but the most favorable outcome


21

Lignin-Degrading Enzyme Activities

253

Fig. 1. Structures proposed for ligninâ&#x20AC;&#x201C;carbohydrate complexes isolated by fractionating endoglucanase treated spruce wood meal successively with aqueous 8 M urea, alkaline borate, and/or barium hydroxide (6).

was one where depolymerization and (re)polymerization of a synthetic lignin (a dehydropolymerisate formed from coniferyl alcohol) occurred to comparable extents (11). 1.2. Reactive Oxygen Species

Efficient lignin biodegradation lies within the realm of only one group of microorganisms, namely, the basidiomycetous white-rot fungi. Over the past 30 years, perceptions about the biochemical mechanisms of macromolecular lignin depolymerization have evolved in unexpected ways. In 1980, it appeared that reactive oxygen species were likely to be responsible for cleaving lignin chains


254

Y.-r. Chen et al. 0.6

0.5

A280

0.4

0.3

0.2

0.1

0.0 0.0

0.2

0.4

0.6

0.8

1.0

VR

Fig. 2. Polymerization of alkali-isolated straw lignin (dash-dotted line) by lignin peroxidase (in concentrated extracellular Phanerochaete chrysosporium culture solution) at pH 4.0 with (dotted line) and without (dashed line) veratryl alcohol in the presence of H2O2 added at regular intervals (Sephadex G75/aqueous 0.5 % (w/v) NaOH elution profiles after 24 h (9)).

Fig. 3. Lignin-peroxidase catalyzed homolytic C aâ&#x20AC;&#x201C;C b cleavage of nonphenolic arylglycerol b -aryl ether lignin substructure (10).

in white-rot fungal cultures (12). Before long, it was claimed that ligninolytic activity is correlated with H2O2 production in P. chrysosporium cell extracts (13). Shortly thereafter, it was reported that synthetic lignins (in the form of dehydropolymerisates from 14 C-labeled coniferyl alcohols) could be degraded almost completely by the hydroxyl-radical-generating (Fentonâ&#x20AC;&#x2122;s) reagent, 1 M H2O2/10 mM FeSO4 (14). However, in 1983 the detection of lignin peroxidase activity in extracellular P. chrysosporium culture solutions (7) quickly diverted attention from hydroxyl radicals as likely agents of lignin depolymerization. The redirected flurry of activity quickly embraced


21

Lignin-Degrading Enzyme Activities

255

three more enzymes, manganese peroxidase (15), versatile peroxidase (16) and laccase (the latter acting through an electron-transfer “mediator”) (17) as potentially important components in the arsenal of biological catalysts for lignin biodegradation. Uncertainties about the actual need for lignin peroxidase and/ or manganese peroxidase in lignin biodegradation (18) as well as difficulties in identifying naturally occurring laccase mediators (19) revived an interest in the possible roles of reactive oxygen species as agents for cleaving lignin macromolecules in vivo (20). Lignocellulolytic enzymes are too large to penetrate into undecayed wood cell walls (21, 22); thus, reactive oxygen species could be responsible for zones of incipient decay located far on a molecular scale from the lumen surface. The low molecular weight oxidants in question are thought to include hydroxyl radicals (HO•) as well as less reactive peroxyl (ROO•) and hydroperoxyl (HOO•) radicals. If these strong oxidizing agents are capable of initiating decay within lignocellulosic cell-wall domains, they must be generated at a distance from the hyphae of white-rot (20) or brown-rot (23) basidiomycetes. The only alternative would be electron transfer between closely spaced aromatic rings in interacting lignin chains (24) as a result of enzymatic oxidation of substructures at the lumen surface. For assay purposes, the mineralization of 4,000 molecular weight hydroxyethyl-(14C)poly(ethylene glycol) to 14CO2 has been employed as a general indication that reactive oxygen species are being produced by a fungus (20). The lengths of these poly(ethylene glycol) chains preclude penetration through fungal cell walls. Both hydroxyl and peroxyl radicals are capable of cleaving poly(ethylene glycol), and experimental findings have shown that production of reactive oxygen species is widespread among white- and brown-rot basidiomycetes in the presence of hard- and softwood substrates, respectively (20). Whether or not they play a significant role in degrading lignocellulose depends on where in the cell-wall domains they are generated.

2. Candidates for Lignin-Degrading Enzymes 2.1 Extracellular White-Rot Fungal Peroxidase Structure

The three types of extracellular white-rot fungal peroxidases have molecular weights of about 40 kDa and belong to peroxidase Class II. Lignin peroxidase and versatile peroxidase exhibit considerable substrate versatility, but manganese peroxidase is restricted in its activity to oxidizing Mn2+ through a single-electron transfer process. Lignin peroxidase, a ~50 Å × 40 Å × 40 Å glycoprotein, is partitioned by the placement of its heme prosthetic group into a proximal and distal domain; these are distinguished according to the


256

Y.-r. Chen et al.

relative positions of two axial histidine residues located on either side of the ferric ion in the protoporphyrin of the active site. Access to the heme from bulk solvent is gained through a cleft that is divided by Ile85, Asp183, and heme propionate D into two channels. It is through the smaller of these that peroxide molecules approach the distal side of the heme (25). The peroxide binding pocket in lignin peroxidase is formed by Arg43 (conserved in peroxidases), His47 (the distal axial ligand with respect to the heme ferric ion), and Phe46 (not conserved in peroxidases). The heme iron is pentacoordinated to the four heme tetrapyrrole nitrogens and to the proximal His176 ligand. In the sixth coordination site of the iron, a water molecule is hydrogen bonded to the distal His47 and two other waters. The Mn-binding site on the surface of manganese peroxidase and versatile peroxidase involves a heme propionate and the carboxylate groups of three amino acid residues (Glu35, Glu39, and Asp179 in manganese peroxidase; Glu36, Glu40, and Asp175 in versatile peroxidase) and two water molecules (26, 27). In lignin peroxidase and versatile peroxidase, an invariant tryptophan with neighboring acidic amino acid residues (Trp171 with Glu168, Glu250, and Asp264 in lignin peroxidase isozyme H8; Trp164 with Glu161 and Glu243 in versatile peroxidase isozyme L) is located on the opposite side of the protein from the heme access channel. This tryptophan is not far from the protoporphyrin and thus may participate in long-range electron transfer from aromatic substrates to the heme. In the presence of excess H2O2, Trp171 in lignin peroxidase is stereospecifically b-hydroxylated (25), but there is no Cb-hydroxylation of Trp164 in versatile peroxidase (28). Mutation of Trp171 to a serine eliminates the capacity of lignin peroxidase to cleave a nonphenolic b-O-4 linked tetrameric lignin model compound (29). Similarly, replacing Trp164 in versatile peroxidase with a serine eliminates the ability of the enzyme to oxidize high-redox-potential aromatic compounds such as veratryl alcohol and Reactive Black 5 (28). 2.2. Lignin Peroxidase

The distance between the ferric ion in the protoporphyrin and the Ne2 atom of the proximal histidine ligand (2.15 Å) in lignin peroxidase is longer than the corresponding bond (1.95 Å) in cytochrome c peroxidase (25). Consequently, the lignin peroxidase heme is more electron-deficient than that in most other peroxidases, which may account for its higher redox potential (1.4–1.5 V). Thus, lignin peroxidase is capable of oxidizing a wide range of aromatic compounds not susceptible to the action of other peroxidases (30) such as veratryl alcohol, methoxybenzenes, and nonphenolic b-O-4 linked arylglycerol b-aryl ethers. By reducing H2O2 to water, ferric lignin peroxidase is transformed into a two-electron oxidized intermediate compound I, an enzyme–FeIV–oxo•porphyrin•+ complex. One oxidizing equivalent


21

Lignin-Degrading Enzyme Activities

257

in compound I is present as an oxo-ferryl species (FeIV=O)2+, and the other is probably stored as a porphyrin p-cation radical. Through two one-electron substrate oxidation reactions, compound I is first reduced to oxoferryl compound II and then to the resting state of the enzyme (25). As the sizes of oligomeric lignin-model substrates increase, the catalytic efficiency of lignin peroxidase decreases. For example, kcat/KM for the lignin-peroxidase catalyzed oxidation of a nonphenolic b-O-4-linked lignin model trimer is 25-fold less than that of veratryl alcohol (31). On the other hand, lignin peroxidase oxidizes phenolic lignin model compounds much more readily than the corresponding nonphenolic ones (10). As a result, neither guaiacylglycerol b-guaiacyl ether nor guaiacylglycerol b-syringyl ether is actually degraded by lignin peroxidase in the presence of H2O2 at pH 3.0: both phenolic lignin model dimers presumably undergo phenoxy radical coupling to give 5–5¢ linked biphenyl tetramers (32). Similarly, oxidation of a tetrameric nonphenolic b-O-4 linked lignin model compound by lignin peroxidase results, through Ca– Cb cleavage, in the formation of the expected benzaldehyde-terminal monomer, dimer, and trimer. The respective phenol-terminal trimeric, dimeric, and monomeric counterparts were not detected, however, owing (presumably) to 5–5¢ linked biphenyl oligomer formation (29). The marked preference for phenolic lignin units is responsible for the tendency of lignin peroxidase to polymerize (9) rather than depolymerize lignin samples in vitro (Fig. 2). 2.3. Lignin Peroxidase Assays

Lignin peroxidase activity may be monitored at 310 nm by the oxidation of 2.0 mM veratryl alcohol to veratraldehyde (e310 = 9.3 × 103 M−1 cm−1) in 50 mM sodium tartrate buffer at pH 2.5 or 4.5 in the presence of 0.1 mM H2O2 (33). Phenolic and other aromatic compounds with high absorbance at 310 nm may interfere with this assay. Moreover, nonperoxidizing veratryl alcohol oxidases from some white-rot fungi (such as Pleurotus sajorcaju) also catalyze the same reaction (34). Alternatively, Azure B has higher specificity than veratryl alcohol with respect to lignin peroxidase activity since the dye is not oxidized by P. sajor-caju veratryl alcohol oxidase. Lignin-peroxidase catalyzed oxidation of 32 mM Azure B (e651 = 4.88 × 104 M−1 cm−1) may be monitored by the reduction in absorbance at 651 nm in 50 mM sodium tartrate buffer at pH 2.5 or 4.5 containing 0.1 mM H2O2. The presence of lignin or other aromatic compounds usually does not interfere with this assay which is carried out at the relatively long wavelength of 651 nm (33).

2.4. Manganese Peroxidase

Manganese peroxidase is the only heme peroxidase capable of the one-electron oxidation of Mn2+ in a typical peroxidase reaction cycle. The product Mn3+ is stabilized by a variety of bidentate chelating agents (such as lactate, malonate, or oxalate), hereby averting disproportionation to Mn2+ and MnIV. The chelated


258

Y.-r. Chen et al.

Mn3+ readily oxidizes phenolic compounds such as guaiacol, 2,6-dimethoxyphenol, 4-methoxyphenol, methylhydroquinone (35), phenol red (36), and phenolic residues in lignins, but not veratryl alcohol (35) or nonphenolic lignin substructures (37). When 5–5¢ coupling between the intermediate phenoxy radicals is precluded, phenolic arylglycerol b-aryl ether lignin model dimers may undergo Ca-oxidation and Ca–arene cleavage as a result of manganese peroxidase activity (38). With malonate as the bidentate chelator of Mn3+, 50 % of a syringylglycerol b-(4-hydroxymethyl-2-methoxyphenoxy) ether substrate was transformed to the a-oxo product within 2 min under the conditions employed at pH 4.5. Concomitantly, 15 % Ca–arene cleavage also occurred, and findings from P. chrysosporium cultures indicated that degradation of the phenolic b-O-4 dimer in vivo was proceeding primarily via Ca-oxidation (38). Despite its effectiveness in oxidizing phenolic lignin model compounds, the Mn3+–chelator complex is unable to degrade nonphenolic lignin substructures (37). Thus, manganese peroxidase at pH 4.5 (in the presence of 0.2 mM Mn2+ at 37 °C) has been reported to depolymerize and (re)polymerize synthetic lignin substrates in the form of dehydropolymerisates from coniferyl and/or sinapyl alcohol. After 1 h, (re)polymerization had occurred to a greater extent than depolymerization of each substrate except the dehydropolymerisate produced from sinapyl alcohol alone. In the latter case, further depolymerization had taken place after 7 h, but there was still 20 % more high than low molecular weight material in the transformed substrate taken as a whole (39). 2.5. Manganese Peroxidase Assays

Manganese peroxidase activity can be determined by monitoring the oxidation of 0.01 % phenol red (oxidation product e610 = 2.2 × 104 M−1 cm−1) at 610 nm in 20 mM sodium succinate buffer (pH 4.5) containing 0.10 mM MnSO4 and 0.10 mM H2O2 (36). The direct oxidation of Mn2+ can be observed through formation of a Mn3+–tartrate complex (e238 = 6.5 × 103 M−1 cm−1) using 100 mM sodium tartrate buffer at pH 5.0 in the presence of 0.10 mM MnSO4 with 0.10 mM H2O2 (28).

2.6. Versatile Peroxidase

Versatile peroxidases possess a hybrid molecular architecture with different oxidation sites (corresponding to those in manganese peroxidase and lignin peroxidase) connected to the heme pocket. The manganese-binding site in versatile peroxidase, involving three acidic amino-acid residues (two Glu and one Asp) and a heme propionate, is very similar to that in manganese peroxidase. Like lignin peroxidase, versatile peroxidase also possesses an exposed tryptophan residue (Trp164) that is thought to be involved in long-range electron transfer during oxidation of aromatic substrates. Genes encoding versatile peroxidases are not as widespread as those encoding lignin peroxidase or manganese peroxidase.


21

Lignin-Degrading Enzyme Activities

259

The activity of the hybrid enzyme has been mainly found in the white-rot fungal genera Pleurotus (such as P. eryngii) and Bjerkandera (such as B. adusta) (30). The kinetic parameter kcat/KM of versatile peroxidase isozyme L (from P. eryngii liquid cultures) with respect to Mn2+ oxidation is very similar to that of manganese peroxidase, but versatile peroxidase is tenfold less efficient in oxidizing veratryl alcohol than lignin peroxidase (28, 40). Nevertheless, the versatile peroxidase from P. eryngii liquid culture is capable of degrading both phenolic and nonphenolic b-O-4 lignin model dimers. In a pH 4.5 solution containing 0.1 mM H2O2, 65 % of 0.25 mM guaiacylglycerol b-guaiacyl ether was degraded by ~0.5 U/mL enzyme within 70 min. The process was accelerated 1.5-fold in the presence of 0.1 mM Mn2+. Under similar solution conditions at pH 3.5, 20 % of 0.25 mM veratrylglycerol b-guaiacyl ether was degraded within the same period through a process that was not affected by the presence of veratryl alcohol (41). 2.7. Versatile Peroxidase Assays

Direct versatile-peroxidase catalyzed oxidation of Mn2+ can be monitored at 238 nm in regard to the formation of the Mn3+– tartrate complex (e238 = 6.5 × 103 M−1 cm−1) from 0.10 mM MnSO4 in 100 mM tartrate at pH 5.0 in the presence of 0.10 mM H2O2. Manganese-independent oxidative activity may be determined through the oxidation of Reactive Black 5 (e598 = 3 × 104 M−1 cm−1), an azo dye that is not oxidized by Mn3+ or lignin peroxidase. The assay is carried out with 0.10 mM Reactive Black 5 in 100 mM tartrate at pH 3.5 in the presence of 0.10 mM H2O2. Alternatively, the oxidation of 2.0 mM veratryl alcohol to veratraldehyde may be monitored at 310 nm (e310 = 9.3 × 103 M−1 cm−1) under the same conditions at pH 3.5 (28).

2.8. b-Etherase

An extracellular 65 kDa b-etherase has been isolated from an ascomycete believed to be a member of the genus Chaetomium. This enzyme is capable of cleaving the phenolic b-O-4 lignin model dimer, guaiacylglycerol b-guaiacyl ether, through a mechanism that is thought to proceed through an intermediate quinone methide in which b-aryl ether bond scission occurs. Such a mechanism is consistent with experimentally observed requirements for the presence of a- and 4-hydroxyl groups in the dimeric substrate (42). The fungal b-etherase has also been reported to cleave terminal b-aryl ether bonds in a synthetic lignin prepared by the dehydropolymerization of coniferyl alcohol in the presence of Caoxo-guaiacylglycerol b-O-(4-methyl)umbelliferone. In a final step, the Ca-carbonyl groups in the peripheral dimeric fragments of the synthetic lignin derivative were reduced with sodium borohydride. Cleavage of the ether linkages to the terminal 4-methylumbelliferone residues in the polymeric substrate was monitored by measuring the intensity of fluorescence at 450 nm using 360 nm excitation (42).


260

Y.-r. Chen et al.

2.9. b-Etherase Assay

The activity of the b-etherase is conveniently determined using a fluorometric assay. The reaction is carried out at 28 °C in 10 mM Tris–HCl buffer at pH 7.5 containing 1 mM dithiothreitol and 0.1 mM guaiacylglycerol b-O-(4-methyl)umbelliferone as the substrate (the latter dissolved in DMSO before adding). After 30 or 60 min, the reaction is stopped by adding to the mixture 0.1 mL 500 mM glycine–NaOH buffer at pH 10. Fluorescence of the released 4-methylumbelliferone is measured at 450 nm using an excitation wavelength of 360 nm (43).

2.10. Laccase

Laccases are polyphenol oxidases with four copper ions of three different types in their active sites. A type 1 copper (in the T1 site), having a strong absorption band around 600 nm, gives laccase its characteristic blue color. The coordination of the T1 copper involves the sulfur atom of a cysteine and Nd1 nitrogens of two histidines. One type 2 and two type 3 copper ions form a trinuclear cluster (in the T2/T3 site) coordinated to a highly conserved pattern of four His-X-His motifs. The T2 copper does not show visible absorbance but exhibits superfine splitting in paramagnetic resonance (EPR) spectra characteristic of copper ions in tetragonal complexes. The EPR-silent T3 coppers can be identified by an absorption band at 330 nm. The electron withdrawn from a substrate is transferred from the T1 site through a conserved HisCys-His motif to the T2/T3 site where molecular oxygen is reduced to water (44). Laccases catalyze four single-electron oxidations of substrates. The reaction of the fully reduced enzyme with molecular oxygen occurs through two consecutive two-electron steps giving rise to a peroxy intermediate and then a native intermediate. The completely reduced T2/T3 site interacts with molecular oxygen to form the peroxy intermediate, which is the rate-limiting step. The subsequent formation of the native intermediate is accompanied by cleavage of the peroxide O–O bond. The native intermediate then slowly transforms to the resting state where the T2 copper ion is no longer bound through an intervening O to the two T3 coppers (44). Laccases are usually monomeric with three sequentially arranged domains disposed in a tight 65 Å × 55 Å × 45 Å globule. The T1 site is located in the third of these domains. Substrate binding takes place in a small negatively charged cavity near the T1 site. The T2/T3 site is ligated between the first and third domains. There are two channels, one of which allows molecular oxygen to reach the T3 copper ions while the other allows the release of water from the T2 copper ion. It has been suggested that the coordination distances around the T1 copper affect the redox potential. For example, the Trametes versicolor laccase has an elongated Cu1–Nd2 bond to His458 (a conserved residue) rendering the copper more electrondeficient. This may account for a redox potential at the higher end of the typical range (0.78 V) for T. versicolor laccase (44).


21

Lignin-Degrading Enzyme Activities

261

Due to their relatively low redox potentials (0.42–0.79 V, based on that of the T1 site), laccases can only oxidize phenolic compounds such as polyphenols, methoxy-substituted phenols, and aromatic diamines to the corresponding radicals (44). However, it has been found that laccase can slowly oxidize 2,2¢-azinobis(3-ethylbenzthiazoline-6-sulfonate) (ABTS) to the dication, which is then able to degrade nonphenolic lignin moieties (45); the overall process occurs spontaneously, even though the redox potential of the laccase has not been altered, because the oxidized nonphenolic lignin substructures undergo irreversible transformations once they are formed. A metabolite from the white-rot fungus Pycnoporus cinnabarinus, 3-hydroxyanthranilic acid, has been proposed to act as a natural laccase mediator or electron-transfer agent in lignin biodegradation. However, the ability of P. cinnabarinus to reduce the lignincontent of softwood kraft pulp is apparently not affected when 3-hydroxyanthranilic acid production is blocked while laccase activity is still clearly being expressed by the fungus (19). It has been reported that “yellow” laccases isolated from the white-rot fungi, Panus tigrinus and Phlebia radiata grown on wheat straw, are able to catalyze the oxidation of a nonphenolic b-1 lignin model compound in the absence of a mediator (46, 47). Both the yellow and the more typical blue laccases from the same microorganisms appeared to have the same molecular weight and N-terminal amino acid sequence. It has been proposed that reaction of low molecular weight lignin degradation products with the blue laccase in fungal cultures could give rise to the yellow laccase with a resting state where the copper centers are not oxidized under normal aerobic conditions. Nevertheless, these two laccases exhibit similar specific activities toward ABTS oxidation. One of the most straightforward ways of confirming this hypothesis would have been by documenting the conversion of the blue laccase to the yellow form in vitro. 2.11. Laccase Assays

The fact that laccases have broad substrate specificity makes it difficult to determine their activity unambiguously. Consequently, assays with multiple substrates should be used. Laccase-catalyzed oxidation of 0.5 mM 2, 2¢-azinobis-(3-ethylbenzthiazoline-6sulfonate) (ABTS) to the corresponding cation radical (e420 = 3.6 × 104 M−1 cm−1) in 100 mM sodium acetate buffer at pH 5.0 can be monitored by the increase in absorbance at 420 nm. Alternatively, the oxidation of 0.5 mM syringaldazine (initially dissolved in ethanol) to tetramethoxy-azo-bis-methylenequinone (e525 = 6.5 × 104 M−1 cm−1) in 100 mM sodium phosphate buffer at pH 6.5 may be followed through the absorbance increase at 525 nm. Finally, the oxidation of 1 mM 2,6-dimethoxyphenol to coerulignone (e468 = 4.96 × 104 M−1 cm−1) in 100 mM sodium phosphate buffer at pH 6.0 is conveniently monitored by the increase


262

Y.-r. Chen et al.

in absorbance at 468 nm (16, 48). It is useful to remember that small inorganic anions, such as fluoride, chloride, azide and hydroxide, inhibit laccase activity by binding with the T2/T3 site.

3. LigninDepolymerase Assays 3.1. Selection of LigninDepolymerase Assay Substrates

3.2. Quest for Lignin Depolymerase

The recalcitrance of lignin domains in plant cell walls has its origins in two commanding features of their physicochemical properties. First, most of the linkages between monomer units in lignin macromolecules are rather stable and hence difficult to cleave (both chemically and biochemically). Second, the noncovalent intermolecular interactions between substituted aromatic rings on the one hand, and cyclohexadienone residues on the other, in neighboring lignin chains lead to stabilization energies in the range of 7â&#x20AC;&#x201C;11 kcal/ mol each (24). These energies are so high that cleavage of interunit linkages in lignin macromolecules may not automatically liberate oligomeric degradation products from a lignin domain. Consequently, care should be exercised in selecting substrates for lignin-depolymerase assays. Compounds structurally unrelated to lignin are not likely to be useful for predicting whether an enzyme should be functionally capable of degrading polymeric lignin substrates. Lignin model compounds should also be treated cautiously because the stabilization energies arising from the entirety of the noncovalent interactions between such molecules are much lower than those between lignin macromolecules. Even the configurations of synthetic lignins (produced by dehydropolymerizing monolignols in vitro) will be less well defined than those preserved in polymeric preparations attentively isolated from lignin domains in plant cell walls. The degree of complementarity between interacting lignin chains will influence the stabilization energy created by the noncovalent interactions between them (49). The definitive substrate for an enzyme that exhibits true lignin depolymerase activity would be a polymeric lignin sample derived with minimal chemical modification from a native lignin preparation. Preferably, such a lignin substrate will remain sufficiently soluble during the assay that complications arising from heterogeneous solution conditions can be avoided. Then classical light scattering measurements may be employed to determine how the (weightaverage) molecular weight, Mw, and (z-average) radius of gyration, Rg, of the native lignin substrate change as a result of lignin depolymerase activity. The simplest approach entails the selection of a standard milled-wood lignin from a softwood as the starting material. The milled-wood lignin is isolated by extraction with aqueous 96 % dioxane in the customary way from the appropriate extractive-free


21

Lignin-Degrading Enzyme Activities

263

wood meal. The crude product is purified by dissolving in (9:1:4 v/v/v) pyridine–acetic acid–water and extracting the resulting solution with chloroform. After complete solvent removal, the residue is redissolved in (2:1 v/v) 1,2-dichloroethane–ethanol and precipitated in ether (50). The parent milled-wood lignin is ultrafiltered exhaustively in aqueous 0.10 M NaOH in turn through 30,000 and 10,000 nominal molecular weight cutoff membranes (Amicon YM30 and YM10, respectively; Millipore Corp., Billerica, MA). In each case, the time-course for ultrafiltration should be extended to allow dissociation of the retained macromolecular lignin complexes to occur (51) so that the individual components released may have the opportunity of passing through the membrane being used. When the permeate has become colorless, the sodium hydroxide in the retentate is removed by continuing the ultrafiltration process with distilled water and then triply distilled water until the pH reaches 7.0–7.5. The neutralized retentate solution is centrifuged to remove any precipitate (<10 % of the original lignin in the fraction), and the supernatant with the dissolved milled-wood lignin substrate is stored in a teflon bottle under N2. 3.3. Rayleigh Light Scattering

The determination of Mw and Rg for the native lignin substrate involves extrapolation of the pertinent data to c = 0 g/L in both cases. The substrate solutions in aqueous buffer are each passed through a 1.0 mm porous PTFE membrane (Pall Life Science, TF-1000, 13 mm diameter) prior to being introduced with a syringe pump into the flow-cell of a Wyatt DAWN HELEOS lightscattering photometer equipped with a l0 = 780 nm laser light source. Light scattering detectors (half fitted with interference filters to provide a means of correcting for fluorescence) are placed in a plane at 18 different angles around the position of the scattering volume in the flow-cell. The intensity of the light scattered at each angle q, Iscattered(q), is proportional to the corresponding excess Rayleigh ratio, Rq (52): I scattered (θ ) ∝ Rq = KM wcP (q ) [1 − 2A2 M wcP (q )], where form factor 2

1 ⎛ 4πn0 ⎞ I (q ) ⎛q⎞ = 1− ⎜ sin 2 ⎜ ⎟ Rg2 P (q ) = ⎟ ⎝ 2⎠ 3 ⎝ λ0 ⎠ I (0) and optical constant 2

⎛ dn ⎞ K = 4π 2n0 2 ⎜ ⎟ λ 0 −4N A −1 ⎝ dc ⎠ Here, A2 represents the second virial coefficient of the substrate, n0 denotes the refractive index of the buffer solution without substrate,


264

Y.-r. Chen et al.

NA is Avogadro’s number, and dn/dc represents the refractive index increment (determined with a Wyatt Optilab rEX differential refractometer operating at 780 nm). The effect of solution absorbance at 780 nm upon the incident light intensity (I0) may be taken into account by noting that I true = I 0 (F / F0 )1/ 2 , where F and F0 are the forward laser monitor signal intensities when the substrate solution and buffer alone, respectively, occupy the flow cell. After various incubation times, assay solutions containing the native lignin fraction and enzyme possessing lignin depolymerase activity are diluted to create, in each case, a series of solutions with successively decreasing substrate concentrations. These are consecutively introduced into the DAWN HELEOS light-scattering flow-cell in order of increasing substrate concentrations. As a given set of solutions is being analyzed, the scattered light intensities at angles q change appreciably with time because of the impact of the enzyme on the native lignin substrate. Such effects would compromise a traditional Zimm-plot analysis of the data. The matter is better handled through the Debye formalism, where Rθ /Kc is plotted against sin2 (q/2). As q → 0 , P (θ) → 1 ; a polynomial fit hereby allows the intercept at zero angle, R0 /Kc , to be determined, whereupon: R0 = M w − 2A2cM w 2 . Kc Thus, a plot of R0 /Kc vs. c allows Mw to be determined for the substrate directly from the intercept at c = 0 (Fig. 4). Concomitantly, the apparent Rg can be determined at each concentration, c, from: Rg 2 =

−3m0 λ 2 , 16π 2 M w (1 − 4 A2 M wc)

where m0 ≡ d[Rθ /Kc] / d[sin 2 (q / 2)]θ→0 represents the slope of the corresponding Debye plot at zero angle. The absolute value of Rg may therefore be obtained by extrapolation to c = 0 (Fig. 5). Both m0 and Mw depend on the reciprocal of (dn/dc)2 so that Rg embodies a much weaker dependence on the differential refractive index than Mw. Consequently, if it is suspected that the (dn/dc) of the substrate changes during transformation into product (51), or if it is difficult to ascertain the impact of the enzyme on the overall differential refractive index of the assay solution, the radius of gyration is the preferred parameter for monitoring the effects of enzyme-catalyzed cleavage of lignin macromolecules.


21

Lignin-Degrading Enzyme Activities

265

380

360

R0/Kc x 103

340

320

300

280

260 0.00

Mw

0.02

0.04

0.06

0.08

0.10

0.12

0.14

c g/L

Fig. 4. Determination of weight-average molecular weight for native lignin substrate from light-scattering measurements. Relationship between R0/Kc and substrate concentration, c; R0 denotes excess Rayleigh ratio extrapolated to q = 0 in a Debye plot, and K is an optical constant characteristic of the system (milled-wood lignin fraction incubated at 0.93 g/L with 4 mM lignin depolymerase for 8.5 h in 25 mM phosphate, pH 7.1).

96 94

Rg nm

92 90 88 86 Rg 84 82 0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

c g/L

Fig. 5. Determination of radius of gyration, Rg, for native lignin substrate from lightscattering measurements. Relationship between apparent Rg and substrate concentration, c; values for Rg are deduced from the set of Debye plots used to generate molecular weight data in Fig. 4 (milled-wood lignin fraction incubated at 0.93 g/L with 4 mM lignin depolymerase for 8.5 h in 25 mM phosphate, pH 7.1).


266

Y.-r. Chen et al.

4. Concluding Remark Despite notable experimental efforts over the past 20 years, no enzyme has yet been reported to display significant depolymerase activity toward native polymeric lignin substrates. Indeed, very little has transpired to advance our understanding of what is biochemically necessary for lignin cleavage to succeed without an opposing tendency for (re)polymerization among the individual lignin components and degradation products (53). A resolute focus upon the enzymatic depolymerization of native lignin macromolecules should hasten the discovery of the first enzymes that would deserve nomination to the lignin depolymerase category.

Acknowledgments The authors wish to acknowledge a subcontract from the BioEnergy Science Center, which is a US Department of Energy Bioenergy Research Center supported by the Office of Biological and Environmental Research in the DOE Office of Science. References 1. Wong DWS (2009) Structure and action mechanism of ligninolytic enzymes. Appl Biochem Biotechnol 157:174–209 2. Davin LB, Patten AM, Jourdes M, Lewis NG (2008) Lignins: a twenty-first century challenge. In: Himmel ME (ed) Biomass recalcitrance—deconstructing the plant cell wall for bioenergy. Blackwell, Oxford, pp 213–305 3. Rencoret J, Gutiérrez A, Nieto L, JiménezBarbero J, Faulds CB, Kim H, Ralph J, Martínez ÁT, del Río JC (2011) Lignin composition and structure in young versus adult Eucalyptus globulus plants. Plant Physiol 155: 667–682 4. Lawoko M, Henriksson G, Gellerstedt G (2005) Structural differences between the lignin– carbohydrate complexes present in wood and chemical pulps. Biomacromolecules 6:3467–3473 5. Balakshin MY, Capanema EA, Chang H-M (2007) MWL fraction with a high concentration of lignin–carbohydrate linkages: isolation and 2D NMR spectroscopic analysis. Holzforschung 61:1–7 6. Gellerstedt G (2007) Lignin complexity: fundamental and applied issues. http://rfparois. free.fr/LIG2G/Seminaire%20LIG2G-WEBvs-tout-public.htm.

7. Tien M, Kirk TK (1983) Lignin-degrading enzyme from the hymenomycete Phanerochaete chrysosporium Burds. Science 221:661–663 8. Kern HW, Haider K, Pool W, de Leeuw JW, Ernst L (1989) Comparison of the action of Phanerochaete chrysosporium and its extracellular enzymes (lignin peroxidases) on lignin preparations. Holzforschung 43:375–384 9. Haemmerli SD, Leisola MSA, Fiechter A (1986) Polymerization of lignins by ligninases from Phanerochaete chrysosporium. FEMS Microbiol Lett 35:33–36 10. Lundell T, Schoemaker H, Hatakka A, Brunow G (1993) New mechanism of the Ca–Cb cleavage in non-phenolic arylglycerol b-aryl ether lignin substructures catalyzed by lignin peroxidase. Holzforschung 47:219–224 11. Hammel KE, Jensen KA Jr, Mozuch MD, Landucci LL, Tien M, Pease EA (1993) Ligninolysis by a purified lignin peroxidase. J Biol Chem 268:12274–12281 12. Hall PL (1980) Enzymatic transformations of lignin: 2. Enzyme Microb Technol 2: 170–176 13. Forney LJ, Reddy CA, Tien M, Aust SD (1982) The involvement of hydroxyl radical derived from hydrogen peroxide in lignin degradation by the


21

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

white rot fungus Phanerochaete chrysosporium. J Biol Chem 257:11455–11462 Gold MH, Kutsuki H, Morgan MA (1983) Oxidative degradation of lignin by photochemical and chemical radical generating systems. Photochem Photobiol 38:647–651 Glenn JK, Gold MH (1985) Purification and characterization of an extracellular Mn(II)dependent peroxidase from the lignin-degrading basidiomycete, Phanerochaete chrysosporium. Arch Biochem Biophys 242:329–341 Mester T, Field JA (1998) Characterization of a novel manganese peroxidase–lignin peroxidase hybrid isozyme produced by Bjerkandera species strain BOS55 in the absence of manganese. J Biol Chem 273:15412–15417 Bourbonnais R, Paice MG (1990) Oxidation of non-phenolic substrates—an expanded role for laccase in biodegradation. FEBS Lett 267: 99–102 Nutsubidze NN, Sarkanen S, Schmidt EL, Shashikanth S (1998) Consecutive polymerization and depolymerization of kraft lignin by Trametes cingulata. Phytochemistry 49:1203–1212 Li K, Horanyi PS, Collins R, Phillips RS, Eriksson K-E (2001) Investigation of the role of 3-hydroxyanthranilic acid in the degradation of lignin by white-rot fungus Pycnoporus cinnabarinus. Enzyme Microb Technol 28:301–307 Hammel KE, Kapich AN, Jensen KA Jr, Ryan ZC (2002) Reactive oxygen species as agents of wood decay by fungi. Enzyme Microb Technol 30:445–453 Blanchette RA, Krueger EW, Haight JE, Akhtar M, Akin DE (1996) Cell wall alterations in loblolly pine wood decayed by the white-rot fungus Ceriporiopsis subvermispora. J Biotechnol 53:203–213 Srebotnik E, Messner K, Foisner R (1988) Penetrability of white-rot-degraded pine wood by the lignin peroxidase of Phanerochaete chrysosporium. Appl Environ Microbiol 54:2608–2614 Wei D, Houtman CJ, Kapich AN, Hunt CG, Cullen D, Hammel KE (2010) Laccase and its role in production of extracellular reactive oxygen species during wood decay by the brown rot basidiomycete Postia placenta. Appl Environ Microbiol 76:2091–2097 Chen Y-R, Sarkanen S (2010) Macromolecular replication during lignin biosynthesis. Phytochemistry 71:453–462 Choinowski T, Blodig W, Winterhalter KH, Piontek K (1999) The crystal structure of lignin peroxidase at 1.70 Å resolution reveals a hydroxy group on the Cb of tryptophan 171: a novel radical site formed during the redox cycle. J Mol Biol 286:809–827

Lignin-Degrading Enzyme Activities

267

26. Sundaramoorthy M, Kishi K, Gold MH, Poulos TL (1997) Crystal structures of substrate binding site mutants of manganese peroxidase. J Biol Chem 272:17574–17580 27. Ruiz-Dueñas FJ, Morales M, Pérez-Boada M, Choinowski T, Martínez MJ, Piontek K, Martínez ÁT (2007) Manganese oxidation site in Pleurotus eryngii versatile peroxidase: a sitedirected mutagenesis, kinetic, and crystallographic study. Biochemistry 46:66–77 28. Pérez-Boada M, Ruiz-Dueñas FJ, Pogni R, Basosi R, Choinowski T, Martínez MJ, Piontek K, Martínez AT (2005) Versatile peroxidase oxidation of high redox potential aromatic compounds: site-directed mutagenesis, spectroscopic and crystallographic investigation of three long-range electron transfer pathways. J Mol Biol 354:385–402 29. Mester T, Ambert-Balay K, Ciofi-Baffoni S, Banci L, Jones AD, Tien M (2001) Oxidation of a tetrameric nonphenolic lignin model compound by lignin peroxidase. J Biol Chem 276:22985–22990 30. Hofrichter M, Ullrich R, Pecyna MJ, Liers C, Lundell T (2010) New and classic families of secreted fungal heme peroxidases. Appl Microbiol Biotechnol 87:871–897, and references therein 31. Baciocchi E, Fabbri C, Lanzalunga O (2003) Lignin peroxidase-catalyzed oxidation of nonphenolic trimeric lignin model compounds: fragmentation reactions in the intermediate radical cations. J Org Chem 68: 9061–9069 32. Yokota S, Umezawa T, Higuchi T (1991) Degradation of phenolic b–O–4 lignin model dimers by lignin peroxidase of Phanerochaete chrysosporium. Mokuzai Gakkaishi 37: 535–541 33. Archibald FS (1992) A new assay for lignintype peroxidases employing the dye Azure B. Appl Environ Microbiol 58:3110–3116 34. Bourbonnais R, Paice MG (1988) Veratryl alcohol oxidases from the lignin-degrading basidiomycete Pleurotus sajor-caju. Biochem J 255:445–450 35. de la Rubia T, Linares A, Pérez J, MuñozDorado J, Romera J, Martínez J (2002) Characterization of manganese-dependent peroxidase isoenzymes from the ligninolytic fungus Phanerochaete flavido-alba. Res Microbiol 153:547–554 36. Kuwahara M, Glenn JK, Morgan MA, Gold MH (1984) Separation and characterization of two extracellular H2O2-dependent oxidases from ligninolytic cultures of Phanerochaete chrysosporium. FEBS Lett 169:247–250


268

Y.-r. Chen et al.

37. Hammel KE, Cullen D (2008) Role of fungal peroxidases in biological ligninolysis. Curr Opin Plant Biol 11:349–355 38. Tuor U, Wariishi H, Schoemaker HE, Gold MH (1992) Oxidation of phenolic arylglycerol b-aryl ether lignin model compounds by manganese peroxidase from Phanerochaete chrysosporium: oxidative cleavage of an a-carbonyl model compound. Biochemistry 31:4986–4995 39. Wariishi H, Valli K, Gold MH (1991) In vitro depolymerization of lignin by manganese peroxidase of Phanerochaete chrysosporium. Biochem Biophys Res Commun 176:269–275 40. Tien M, Kirk TK, Bull C, Fee JA (1986) Steadystate and transient-state kinetic studies on the oxidation of 3,4-dimethoxybenzyl alcohol catalyzed by the ligninase of Phanerochaete chrysosporium Burds. J Biol Chem 261:1687–1693 41. Caramelo L, Martínez MJ, Martínez ÁT (1999) A search for ligninolytic peroxidases in the fungus Pleurotus eryngii involving a-keto-g-thiomethylbutyric acid and lignin model dimers. Appl Environ Microbiol 65:916–922 42. Otsuka Y, Sonoki T, Ikeda S, Kajita S, Nakamura M, Katayama Y (2003) Detection and characterization of a novel extracellular fungal enzyme that catalyzes the specific and hydrolytic cleavage of lignin guaiacylglycerol b-aryl ether linkages. Eur J Biochem 270:2353–2362 43. Masai E, Katayama Y, Nishikawa S, Yamasaki M, Morohoshi N, Haraguchi T (1989) Detection and localization of a new enzyme catalyzing the b-aryl ether cleavage in the soil bacterium (Pseudomonas paucimobilis SYK-6). FEBS Lett 249:348–352 44. Giardina P, Faraco V, Pezzella C, Piscitelli A, Vanhulle S, Sannia G (2010) Laccases: a neverending story. Cell Mol Life Sci 67:369–385, and references therein

45. Bourbonnais R, Leech D, Paice MG (1998) Electrochemical analysis of the interactions of laccase mediators with lignin model compounds. Biochim Biophys Acta 1379:381–390 46. Leontievsky A, Myasoedova N, Pozdnyakova N, Golovleva L (1997) ‘Yellow’ laccase of Panus tigrinus oxidizes non-phenolic substrates without electron-transfer mediators. FEBS Lett 413:446–448 47. Leontievsky AA, Vares T, Lankinen P, Shergill JK, Pozdnyakova NN, Myasoedova NM, Kalkkinen N, Golovleva LA, Cammack R, Thurston CF, Hatakka A (1997) Blue and yellow laccases of ligninolytic fungi. FEMS Microbiol Lett 156:9–14 48. Soden DM, O’Callaghan J, Dobson ADW (2002) Molecular cloning of a laccase isozyme gene from Pleurotus sajor-caju and expression in the heterologous Pichia pastoris host. Microbiology 148:4003–4014 49. Guan S-Y, Mlynár J, Sarkanen S (1997) Dehydrogenative polymerization of coniferyl alcohol on macromolecular lignin templates. Phytochemistry 45:911–918 50. Lundquist K, Ohlsson B, Simonson R (1977) Isolation of lignin by means of liquid-liquid extraction. Svensk Papperstidn 80(5): 143–144 51. Contreras S, Gaspar AR, Guerra A, Lucia LA, Argyropoulos DS (2008) Propensity of lignin to associate: light scattering photometry study with native lignins. Biomacromolecules 9:3362–3369 52. Wyatt PJ (1993) Light scattering and the absolute characterization of macromolecules. Anal Chim Acta 272:1–40 53. Sarkanen S (1991) Enzymatic lignin degradation— an extracurricular view. ACS Symp Ser 460: 247–269


INDEX A Acetivibrio cellulolyticus .................................. 100, 101, 119 Acetyl bromide reaction ........................... 227, 234, 236 Acidothermus cellulolyticus endoglucinase I (EI) (Cel5A) ................................................. 194 Adsorption ............................................ 81, 82, 98, 106 Adsorption characteristics........................................... 81 ADTs. See Arogenate dehydratases (ADTs) Affinity electrophoresis non-denaturating........................................ 117–118 AH-Glc. See Anhydro-glucose (AH-Glc) AIR. See Alcohol insoluble residue (AIR) Alcohol insoluble residue (AIR) .................... 62, 64–65, 210–213, 218, 222 Alfalfa (Medicago sativa L.) ............................. 48, 50, 54, 55, 98, 102 transgenic ............................................................. 50 wild-type (wt)........................................... 50, 54, 55 Algae ............................................................... 9–11, 36 Alkyl aryl etherase ............................................ 253, 254 Allomorphs gauche (G) ........................................................... 10 mercerized cellulose.............................................. 10 trans (T) ............................................................... 10 Amino acid analysis .............................. 2, 226, 230–232 Amylase amyloglucosidase ........................................ 182, 183 thermostable α-amylase ...................................... 183 Analysis glycosyl residue composition....................... 212–213 linkage ............................................................... 211 MALDI-TOFMS................................ 211, 212, 222 Anhydro-glucose (AH-Glc)...................... 201, 202, 206 Antigens ................................ 32, 35, 72, 203, 204, 243 Antiparallel chains ...................................................... 12 Arabidopsis thaliana .................................... 67, 68, 74, 194, 225, 228–232, 235, 242 Arabinogalactan glycoproteins.................................... 60 Arogenate dehydratases (ADTs) .............................. 225, 235–238, 240, 242–243 Assays bicinchoninic acid (BCA)............................... 2, 168, 169, 171–175 biuret ..................................................................... 2

cellobiose to cellotriose ratio assay ............................................ 90–92, 205 dye-binding ............................................................ 2 enzyme-linked .......................................... 63, 66–67 fluorogenic assay................................................. 196 glucose oxidase/peroxidase (GOPOD) ................................... 3, 185, 189 high-performance liquid chromatography (HPLC)................................ 3, 90, 153, 157, 158, 160, 197, 198, 202, 207 4-methylumbelliferyl-β-D-cellobiose (MUC) assay ........................ 196, 199, 200, 203–205 protein measurement .............................................. 2 sigmacell enzyme assay ...................... 197, 199–202, 204–205 soluble to insoluble reducing sugar ratio assay .................................................. 90 thin-layer chromatography (TLC)........................... 3 total cellulase activity assay .......................... 197–198 xylose dehydrogenase (XDH) ................................. 3 Atomic force microscopy (AFM) system .............. 21–28, 30, 91, 125–126 nanoscope V controller ......................................... 24 Nikon optical microscope ..................................... 24 probes ............................................................ 24, 28 tapping mode ........................................... 24, 25, 28

B Bacteria .......................... 4, 9, 11, 36, 91, 93, 105–113, 132, 138, 142, 150, 155, 158, 166, 194 Binding affinity .......................................... 92, 119, 120 Binding strength qualitative assessment ............................. 97–98, 106 Biocatalysts .................................. 1–6, 21, 81, 138, 143 Biofuels ...................................................... 21, 59, 137, 142, 146, 166, 194 Bioinformatics tools data mining ........................................................ 143 Biomass characterization .................................. 59–69, 71–79 destarching ................................................ 179, 180, 183, 185, 189, 190 harvesting .................................... 41, 177–179, 194 recalcitrance ..................... 30, 47–48, 137, 177–190

Michael E. Himmel (ed.), Biomass Conversion: Methods and Protocols, Methods in Molecular Biology, vol. 908, DOI 10.1007/978-1-61779-956-3, © Springer Science+Business Media, LLC 2012

269


BIOMASS CONVERSION: METHODS AND PROTOCOLS 270 Index Biomass accessibility accessible surface area ........................................... 82 Biomass conversion process ................................. 4, 6, 9, 18, 21, 22, 47, 48, 55, 81, 126, 166 kinetics ................................................................. 48 recalcitrance ................................................... 30, 47 Biomass surface characterization .......................... 29–30 Biomass surface characterization laboratory (BSCL)................................................ 29–32 Biopolymer molecules .................... 9, 22, 223, 224, 245 Bioprospecting................................................. 137–146 Bound water freezing .................................................... 82, 85, 86 non-freezing......................................................... 82 Bovine serum albumin (BSA) standard ............ 110, 116, 172, 197 Brown rot fungi ......................................................... 22 BSCL. See Biomass surface characterization laboratory (BSCL)

C Cadherin (CA) doublet domains calcium-dependent cell-adhesion proteins ................................................... 105 Cadherin-like (CADG) doublet domains .......... 106–112 Carbohydrate-active enzymes ........................... 115–121 Carbohydrate binding assay cloning ....................................................... 106–107 expression .......................................................... 107 purification ......................................................... 107 Carbohydrate-binding modules (CBMs) family ............................................... 3, 98, 100, 101 non-catalytic polysaccharide-recognizing module glycoside hydrolases ...................... 97 putative ................................................ 97, 132, 205 Carbon flux ............................................. 225, 239, 240 Carboxymethylcellulose (CMC) ................................. 93 Catalysts β-D-glucosidase ................................................ 2, 22 cellobiohydrolase (CBH) ......................... 2, 89, 132, 172, 193, 194 cellulase .......................... 22, 81, 137, 138, 143, 204 glycosyl hydrolase (GH) ................... 2, 74, 203, 204 methylumbelliferyl (MU) glycoside.............................. 2, 195, 196, 200 paranitrophenyl (PNP) glycoside ............................ 2 Catalytic domain (CD) ............................... 89–91, 126, 196, 199, 204–206 Catalytic module ........................................ 22, 115, 119 CAZy database ................................................ 141, 143 CBD. See Cellulose-binding domain (CBD) CBMs. See Carbohydrate-binding modules (CBMs) CD. See Catalytic domain (CD) Cellobiose........................................... 3, 11, 22, 89–93, 126, 152, 153, 157, 161, 201, 202, 204–206

Cellulase cost ............................................................ 165–166 performance efficacy .................................... 2–4, 22, 165–167 processivity ..................................................... 89–94 specific activity............................ 2–3, 197–198, 202 Cellulolytic thermophiles Caldicellulosiruptor obsidiansis ........................... 151–153, 156, 158–160, 163 Clostridium thermocellum ..................................101, 132, 151–153, 156, 158–159 Cellulose amorphous ...................................... 2, 4, 18, 91, 92, 98, 101, 102, 116, 120, 204, 205 bacterial ............................................... 4, 11, 91, 93 bulk ..................................................................... 74 cellodextrin ........................................................ 193 crystalline ............................................ 2, 10, 11, 13, 16, 18, 93–94, 97, 130, 132, 204–206 fibril ............................................... 4, 10–12, 15–18 history of................................................ 2, 4, 10, 11 I-α ................................................................................ 22 I-β ...................................................................11, 16, 22 II ............................................................. 10, 12, 22 III .................................................................. 10, 22 microbial ........................................................ 4, 137 microcrystalline ....................................... 4, 98, 107, 196–197, 202, 204, 206 microfibrils ......................................... 18, 22, 30–31 native .......................................... 10, 16, 21, 22, 43, 91, 116, 117, 196, 203, 245 processed.......................................... 4, 10, 194–195 Valonia crystalline cellulose .............. 11, 91, 130, 131 Cellulose-binding domain (CBD) .............................. 97 Cellulose-binding residue................................... 93, 100 Cellulose synthetase molecules ................................... 10 Cellulosome............................................... 22, 119, 132 Cell walls composition ...................................... 57, 59–61, 69, 71–73, 143, 209, 223, 239, 245 dry ....................................................................... 23 extraction ........................................... 62, 64–65, 67 fresh ..................................................................... 23 primary ........................................ 23, 26–27, 54, 55 recalcitrance ......................................................... 59 residue extraction ............................... 227, 233–235 secondary ................................................. 23, 54, 55 Characterization ..................................... 29–30, 59–69, 71–79, 143, 150, 162, 209–222 Chemical imaging techniques label-free chemical imaging............................. 48, 49 sensitivity ............................................................. 48 spatial resolution ................................ 31, 48–49, 56 specificity.............................................................. 48 Clostridium cellulolyticum .................................................... 92


BIOMASS CONVERSION: METHODS AND PROTOCOLS 271 Index Clostridium thermocellum ........................................101, 132, 151–153, 156, 158–159 CMC. See Carboxymethylcellulose (CMC) Coarse-grain models ............................................ 17–18 Coherent anti-Stokes Raman scattering (CARS) microscopy ............................. 49–56 Coherent Raman scattering (CRS) microscopy .......................................... 47–57 Cohesin–dockerin interaction ........................... 118–119 Columns anion chromatography ............................................ 4 silica-based amino ................................................... 4 ultra-high pressure liquid chromatography .............. 4 Commercial assay kits BCA protein assay kit ................................. 168, 172 Commercial enzyme optimization liquozyme .......................................... 183, 184, 188 spirizyme ............................................ 183, 184, 188 Commercial enzyme preparations..................... 165–175 Complex polysaccharides ........................... 71, 105–113 Compositional analysis ..................................... 5, 60, 71 Computer codes ...................................................................... 9 modeling .......................................................... 9–18 power ....................................................... 18, 34–35 Conditions for cellulose performance assay ........................... 2, 3 for digestion ........................................................... 3 for saccharification .............................................. 2, 4 Containment ............................................................... 5 Corn grains................................................................ 23 Corn stover ........................................ 4, 23, 26–27, 42, 43, 50, 55, 56, 194, 196, 198–200 Costs enzymes ........................................ 1, 47–48, 63, 66, 137, 165–166, 193 thermochemical pretreatment ........................ 21, 48, 166, 193 Cryotome cutting ............................................ 232–233 Crystallinity ................................................... 10–11, 22 Crystallographic analysis ...................................... 12, 17 Cultivation-based methods....................................... 138 Curating ........................................................ 33–34, 41

D Deconstruction enzymatic ............................................... 1, 146, 194 microbial ................................................................ 1 Degree of polymerization (DP) ........................... 11, 15, 17, 22, 216 Desalting chromatography protocol high molecular weight fraction ................... 169, 170 low molecular weight fraction ............................. 170

Detectors charged-aerosol ...................................................... 3 evaporative light-scattering ..................................... 3 pulsed amperometric .............................................. 3 Differential scanning calorimetry (DSC) .............. 82–86 Diffraction ....................... 11, 12, 18, 53, 129, 132–134 Digestion of biomass .............................................................. 4 conditions .............................................................. 3 non-specific binding ............................................... 3 Down-regulated lignin modification ..................... 48, 55 DP. See Degree of polymerization (DP) Drying and coating for SEM analysis critical-point drying (CPD) ...................... 32, 37–38 freeze drying ............................................ 32, 34, 37 liquid carbon dioxide ...................................... 32, 38 sputter coating ............................................... 32, 38 Drying process..................................................... 31, 81 DSC. See Differential scanning calorimetry (DSC)

E ELISA. See Enzyme linked immunosorbent assay (ELISA) ELISA-based method................................................. 60 Endoxylanase treatment ........................... 210, 218, 220 Energy minimization minimum-energy state .......................................... 13 multiple-minimum problem.................................. 13 Environmental enrichment............................... 156–157 Enzymatic hydrolysis.................................... 48, 54, 166 Enzyme inhibition ....................................................... 2 Enzyme linked immunosorbent assay (ELISA) ................... 60, 61, 63, 66–69, 72, 79, 203 Epitope in situ visualization ................................................ 60 in vitro detection................................................... 60 Error .................................................. 2, 5–6, 132, 134, 160, 161, 172, 187 n n ESI-MS . See Multi-step mass spectroscopy (ESI-MS ) Exocellulase (or Cellobiohydrolase)................. 2, 89–94, 132, 172, 193, 194 family 6 (Cel6B) .............................................90, 91 family 48 (Cel48A) ......................................... 90–93 Explicit water molecules ....................................... 15–16 Extraction .......................... 4, 57, 60–62, 64–65, 67–69, 106, 107, 126, 129, 139, 153, 158, 179–180, 183, 185, 186, 188–190, 195–197, 203–207, 222, 225, 227–228, 231–235

F FCM. See Flow cytometry (FCM) Feedstock .............................. 9, 81, 179, 194, 224, 243


BIOMASS CONVERSION: METHODS AND PROTOCOLS 272 Index Fermentative hyperthermophiles Pyrococcus furiosus .................... 151–153, 156, 158–159 Thermotoga maritima .............. 151–153, 156, 158–159 Fibrils ...................................................... 10–12, 15, 16 Filter paper unit (FPU) ............................................ 166 Fixation buffer ........................................... 31, 34–35, 74–76 dehydration ........................................ 31, 35, 74, 78 embedding ........... 31–32, 34–36, 42, 43, 75, 77, 78 fixative................................................ 35, 74–76, 78 infiltration .................................... 35, 36, 42, 43, 75 polymerization ............................................... 35, 75 Fixation and embedding high-pressure freezing and freeze substitution ............................. 32, 36, 41, 42 microwave processing ......................... 31, 34–35, 41 Flow cytometry (FCM) ................................... 149–151, 154, 157–163 Fluorescence-based techniques centroid tracking ........................................ 125, 126 fluorescence correlation spectroscopy (FCS) ............................ 33, 39–40, 125–126 fluorescence resonance energy transfer (FRET) ........................................... 125, 126 Fluorescently labeled cellulose anthranilic acid ..................................................... 91 diaminopyridine ................................................... 91 Fluorophores fluorescent proteins (FPs) .......................... 126–128, 130, 132–133 inorganic nanocrystals ................................ 126–128 organic dyes ............................................... 126, 127 Force-based techniques atomic force microscopy (AFM) ..................................... 22–28, 30, 91 magnetic tweezers ...................................... 125–126 optical trap ................................................. 125–126 Force fields .......................................................... 14–18 FPU. See Filter paper unit (FPU) Functional module ..................................................... 22

H Hemicellulose arabinoxylan ............................................... 210, 214 glucuronoxylan................................... 210, 219, 221 Hemicellulosic polysaccharides ................................... 60 Heteroxylan ..................................................... 209–222 High-throughput (HTP) cultivation method............................................ 149–163 High-throughput (HTP) screening system ............................................. 142–143 High-throughput (HTP) sequencing technologies .................................... 138, 139 High-throughput (HTP) starch assay ................ 178, 179, 182, 183, 185, 187 Histochemical staining ..................... 226, 230–231, 237 Homotypic adhesion ................................................ 105 Humicola insolens ................................................................ 92 Hydrolysis process ...................................... 2–5, 21, 22, 48, 54, 55, 91, 93–94, 115, 126, 145, 150, 166, 189, 203, 204 Hydroxyl group ........................... 82, 86, 217–218, 253

I Image capture ........................................ 33–34, 41, 130 Imaging ............................................ 21–30, 33, 38, 39, 41, 43, 48, 49, 51–56, 128, 132 Imaging tools ...................................................... 29–30 Immunohistochemistry (immunolocalization) ............ 60, 71–79, 243 Immunolabeling .......................... 30, 33, 39, 40, 75–78 Implicit solvation models ........................................... 17 Infrared spectroscopy ........................................... 48–49 In planta .................................................................. 198, 243 Interfascicular fiber lignin ................................. 54, 227, 233, 236, 237, 240 In-tube absorption ................................................... 106 Iodine starch test ..................... 179, 181–183, 185–187 Irreversible cell wall collapse....................................... 81 Isothermal step melting procedure ....................... 82, 84

G

L

Genetic engineering (genetic manipulation) ............ 127, 132–133, 223–243 GFP. See Green fluorescent protein (GFP) Gibb–Thomson equation ..................................... 84–85 Glucose release ................ 177, 179, 184, 185, 188–190 Glycan-directed monoclinal antibodies (mAbs) .................. 60, 61, 63, 66, 68, 69, 71 Glycome profiling .................................... 59–69, 73–74 Glycoside hydrolases ................................. 97, 115, 116, 137–146, 172, 193–207 Green fluorescent protein (GFP) .............. 127, 130–133

Laboratory information management system (LIMS) ............................................. 6 Laccase ............................................ 248–249, 254–256 Laccase mediator ............................................. 249, 255 Laser microscope microdissection..................... 223–243 Ligand-bound structure ............................................. 92 Ligand specificity ................................................. 97–98 Lignin content.............................................. 48, 50, 55, 56, 223, 224, 234, 236–239, 255 depolymerase.............................................. 256–260


BIOMASS CONVERSION: METHODS AND PROTOCOLS 273 Index distribution .............................................. 48, 54–57 heterogeneity ............................................. 223–243 peroxidase .......................................... 246, 248–253 Lignin analysis G:S ratio determining factors .............................. 239 guaiacyl-(G) rich lignins ............. 223–224, 245, 246 H lignins ............................................................ 224 syringyl-(S) rich lignins ............... 223–224, 245, 246 Lignocellulosic biomass switchgrass ................................... 30, 153, 209–210 Lignocellulosic conversion ......................... 81, 210, 224 Lignocellulosic recalcitrance ............................. 150, 224 LIMS. See Laboratory information management system (LIMS) Liquid fuels ......................................................... 9, 150 Loading .............................................. 2–3, 5, 168, 169, 180–182, 184–186, 188, 195, 197, 201, 205

Molecular dynamics (MD) ................................... 12–17 Molecular mechanics (MM) AMBER ............................................................... 15 CHARMM........................................................... 15 GLYCAM ............................................................ 15 GROMOS/GROMACS ....................................... 15 NAMD ................................................................ 15 TINKER .............................................................. 15 Molecular probes ......................................... 60, 74, 127 Monte Carlo calculations ..................................... 12, 13 MS. See Mass spectroscopy (MS) Multi-photon fluorescence intensity ........................... 53 Multi-scale microscopy analysis ............................ 29–44 Multi-step mass spectroscopy n (ESI-MS ) ............................... 217–219, 221 Multi-well plates .............................................. 152, 155

M

Natural cellulose .................................................. 21, 22 Nicotiana tabacum ............................................................. 194 NMR spectroscopy ............................................ 18, 210 Non-isocratic elution ................................................... 4 Non-parametric statistical analysis ............................ 139

Maize ............................................. 23, 26, 27, 50, 141, 194–195, 199, 203–205 Maize kernels transgenic (TG) .......................................... 194–195 wild type (WT) ........................................... 194–195 Manganese peroxidase ..................................... 248–253 Mass spectroscopy (MS).................................. 211, 213, 214, 216–222, 226–228, 232–233, 235, 239–241, 243 MD. See Molecular dynamics (MD) MD trajectories.......................................................... 13 Measurements of cellulose action ........................................... 81, 82 of hydrolysis ........................................................... 3 of molar extinction coefficients ....................... 2, 172 of protein concentration ............................ 111, 133, 168, 170, 172 Membrane-bound organelles Mesophilic organisms............................................... 149 Metabolic flux.......................................................... 223 MetaGene................................................................ 143 Metagenomic methods cultivation-independent approaches .................... 138 Metagenomics ................................................. 137–146 Methylation ..................................... 213, 214, 217, 219 Microbial biomass decay communities ...................... 141 Microorganisms ........................................ 21, 137, 138, 150, 158–160, 167, 247, 255 Milling ...................................... 44, 195, 211, 222, 223 MoBio Powersoil kit ................................................ 139 Modeling atomic level .......................................................... 17 molecular level ..................................................... 17 Moisture content ............................................... 31, 195 Molecular beam mass spectroscopy .......................... 177

N

O Oligosaccharides xylo-oligosaccharides .................. 210, 215–216, 219

P Parallelization ...................................................... 15, 18 Pectic polysaccharides ................................................ 60 Performance of β-D-glucosidase ............................................ 2, 22 of cellobiohydrolase ........................ 2, 132, 172, 192 of cellulase...................................................... 3, 166 Phanerochaete chrysosporium .............................................. 246 Phenotypic analysis .................................. 225, 231–232 Phenylpropanoid metabolism ................................... 240 Photobleaching kinetics ............................................... 126, 127, 133 Plant cell wall...................................... 1, 2, 4, 9, 21–23, 25, 27, 31, 47–57, 59–69, 71–79, 126, 140, 143, 223–243, 245, 246, 256 Polyol additives ........................................................ 167 Polysaccharide binding assay ......................... 2, 97–102, 110, 120, 174 Polysaccharides ................................... 2, 10, 12, 47, 48, 60, 61, 69, 71–73, 79, 97–102, 105–113, 115–121, 151, 168, 210, 211, 214, 222 Poplar (Populus deltoides)................................. 56, 74, 78, 144, 179, 209–222 Pore distribution .............................................. 82, 84, 85 volume ........................................................... 82, 85


BIOMASS CONVERSION: METHODS AND PROTOCOLS 274 Index Porosity of plant tissue ............................................... 41 Potential energy functions .................................... 13, 14 Predicted relative metabolic turnover (PRMT) .................................................. 142 Predictive metabolomics .......................................... 140 Pretreatment AFEX ................................................................... 30 dilute acid ........................................ 5, 21, 153, 159 hot-water ............................................................. 30 hydrothermal...................................................... 190 ionic liquids .......................................................... 30 thermochemical .................... 5, 21, 30, 48, 166, 182 Primary substrate ....................................................... 97 PRMT. See Predicted relative metabolic turnover (PRMT) Probe molecules cellulase .......................................................... 81, 82 dextran ........................................................... 81, 82 dye ................................................................. 81, 82 water .................................................................... 81 Processing and analysis ............................. 33–34, 40–41 Processive endoglucanase Cel9A................................................................... 92 non-reducing end ................................................. 91 reducing end ........................................................ 91 Process model .............................................................. 1 Protein assay ............................................ 168, 170–172 Protein-binding affinities .......................................... 119 Protein concentration ................................. 2, 110–113, 165–175 Protein dosage (or loading).............................. 168, 195 Protein macromolecular complexes .......................... 174 Protein solubility ............................................. 167, 171 Protein stabilization ................................................. 168 Pyrolysis GC/MS ............................ 232–234, 239–241

Q Quantitative insights into microbial ecology (QIIME) analysis ..................................... 140 Quantitative trait loci (QTL) studies ........................ 178

R Raman microspectroscopy .............................. 30, 47–57 Rayleigh light scattering ................................... 257–259 Reactive oxygen species.................................... 247–249 Real-time monitoring..................................... 48, 51, 56 Residual water ..................................................... 64, 65 Residues alcohol insoluble residue (AIR)................ 62, 64, 65, 210, 212, 213, 218, 222 Resolution, sub-nanometer ........................................ 22 Reversible calcium-dependent binding .................................... 105, 106, 110

Robotic methods drawbacks of biomass heterogeneity ................................. 5, 81 extractives removal ............................................ 5 hydrolysate neutralization.................................. 5 starch removal ................................................... 5 static control ..................................................... 5 high-throughput (HTP) ..................................... 5–6 liquid-handling ....................................................... 5 obstacles of meaningful results ................................ 6 screening ............................................................ 185 rRNA gene sequencing 16S sequencing for bacteria and archaea ............................................ 138, 139 18S sequencing for eukaryotes fungi ............................................................. 138 nematodes .................................................... 138 protozoa ....................................................... 138 yeast ............................................................. 138

S Saccharophagus degradans .............................. 2–40, 107, 108 Samples dehydration ................................................ 7, 31, 35 large sets ...................................... 3, 5, 43, 178, 189 preparation ........................................ 11, 22, 29–44, 54, 57, 166, 168, 179–18 Scanning electron microscopy (SEM) .................. 30, 32, 33, 36–39, 41 Scatchard plots................................................. 107, 112 Sectioning and staining for light and electron microscopy ........................ 33, 40, 43, 76, 78 cryo-sectioning ..................................................... 33 grid staining ......................................................... 33 immuno-labeling .................................................. 33 ultramicrotomy..................................................... 33 Sequence-based screening methods illumina sequencing .................................... 139, 140 metagenomic sequencing ............................ 139–141 pyrotag sequencing..................................... 138, 139 Sequence homology ......................... 100, 140, 143, 144 Simons’ staining method ............................................ 82 Simulation ........................................................... 12–18 Simulation timescale ............................................ 15–18 Single-molecule fluorescence detection ......................................... 125–137 Site-directed mutagenesis ........................................... 92 Size reduction techniques .......................................... 29 Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)............. 98, 102, 106, 109, 110, 144, 145, 196 Solid-state nuclear magnetic resonance (NMR) ...................... 81, 210, 211, 215–219


BIOMASS CONVERSION: METHODS AND PROTOCOLS 275 Index Solubilization........................... 166, 167, 171, 210, 212 Solubilization of insoluble polysaccharides................ 115 Soluble polysaccharides ............................ 115–120, 132 Solution osmolarity .......................................................... 167 polarity ............................................................... 167 Solvated proteins model ............................................. 15 Starch .................................................. 5, 107, 177–191 Stimulated Raman Scattering (SRS) microscopy two-color SRS ................................................ 50, 52 Structural artifacts ...................................................... 31 Structural characterization ............................... 209–222 Substrate-binding specificity ............................. 115–121 Substrates Avicel (crystalline cellulose) ................... 4, 101, 153, 156, 157, 159, 160 corn stover ................................................. 194, 198 history of................................................................ 4 insoluble ........................................... 4, 90, 97–102, 154, 157, 160, 162 lignin ................................................. 246, 252–260 lignocelluloses .......................................... 3, 82, 194 plant cell wall ......................................... 2, 4, 23, 63 Populus deltoides ...................... 153, 156, 157, 159, 161 SigmaCel ................................ 4, 197, 199–202, 204 Solka Floc .............................................................. 4 switchgrass (pretreated) ............................... 30, 153, 156, 159–161 xylan (birchwood) ..................................... 151, 153, 156, 157, 159, 161, 162 Surface-active (ablative) phenomenon .......................... 4 Surface chemistry, protein ................................. 85, 127, 128, 167, 168 Surface imaging ......................................................... 21 Surface residues ......................................................... 93 Switchgrass (Panicum virgatum) ...................... 30, 74, 77, 141, 153, 156, 157, 159–161, 178, 179, 181–186, 190, 209–222 System size .......................................................... 12, 15

T T-DNA knockouts, single and multiple .................... 243 TEM. See Transmission electron microscopy (TEM) Thermal tolerance .................................................... 199 Thermobifida fusca ............................................. 90, 118, 119 Thermophilic anaerobes ........................................... 150 Thermoporometry ............................................... 81–86 Thioacidolysis .......................... 227–228, 234–236, 238 Tiled blastx searches................................................. 143 Tobacco........................................... 194–196, 198–200 Total internal reflection fluorescence microscopy (TIRF-M)........................................ 128–130

electron-multiplying charged couple device camera (EMCCD) ........................... 129, 134 evanescent wave.................................................. 129 point-spread functions (PSF) .............................. 129 total internal reflection ............................... 128–129 Total sugar estimation .......................................... 62, 66 Transgenic cornmeal ................................ 196–197, 199 Transgenic enzyme activity ............................... 196–198 Transgenic glycoside hydrolases ....................... 193–207 Transmission electron microscopy (TEM)......................... 30, 31, 33, 39, 41–43 Trichoderma reesei .............................................. 18, 90, 132, 172, 194, 201, 202 cellobiohydrolase I (CBH I) (Cel7A) .................. 194 Tunnel ................................................... 90, 91, 93, 126 Twisted fibrils ............................................................ 11

U UniFrac analysis ....................................................... 139

V Valonia algae ............................................................... 11 Vascular bundle lignin .............................. 223–224, 239 Versatile peroxidase ......................................... 249, 250, 252–253 Vibrational spectroscopy ............................................ 15 Volumetric activity (or effectiveness) ........................ 166

W Western blotting .............................. 195–198, 203, 204 White-rot fungi ........................................ 247, 251, 255

X Xylan ...................................................... 60, 77, 78, 98, 107, 112, 113, 116, 118–120, 150–151, 153, 156, 157, 159, 161, 162, 215, 216, 219–220 soluble ............................................................... 116 Xylanase chimeric ............................................................. 137 wild-type (Xyn10A and Xyn11A) ................ 118–120 Xyloglucan................................................. 60, 210, 215 Xylose release ....................................... 3, 157, 162, 218

Y Yellowstone National Park (YNP).............................150 Rabbit Creek Hot Springs .................................. 154

Z Zea mays ................................................................. 194

1617799556  
1617799556  
Advertisement