UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of the mannan-degrading system of cellulomonas fimi Stoll, Dominik 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1998-346293.pdf [ 19.71MB ]
JSON: 831-1.0099353.json
JSON-LD: 831-1.0099353-ld.json
RDF/XML (Pretty): 831-1.0099353-rdf.xml
RDF/JSON: 831-1.0099353-rdf.json
Turtle: 831-1.0099353-turtle.txt
N-Triples: 831-1.0099353-rdf-ntriples.txt
Original Record: 831-1.0099353-source.json
Full Text

Full Text

Characterization of the Mannan-Degrading System of Cellulomonas fimi by Dominik Stoll Diplom in Biologie II, Universitat Basel, 1992 A THESIS SUBMITTED I N P A R T I A L F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Microbiology and Immunology / Biotechnology Laboratory) We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH C O L U M B I A August 1998 © Dominik Stoll, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, 1 agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^ W o V o V a . ^ The University of British Columbia Vancouver, Canada Date ft fS/^S DE-6 (2/88) Abstract In this study the mannan-degrading system of the Gram positive bacterium Cellulomonas fimi was characterised. C. fimi can degrade different forms of mannan and can use the degradation products as carbon and energy source. This study focuses on the galactomannan-degrading system which was found to be composed of one secreted endo-1,4-P-mannanase (Man26A), one intracellular l,4-[3-mannosidase (Man2A) and one intracellular 1,6-cc-galactosidase. The genes encoding Man26A and Man2A have been isolated and sequenced, and the enzyme activities were investigated. The endo-l,4-p-mannanase (Man26A) has a multidomain structure and comprises a family 26 catalytic domain, a mannan-binding domain (MBD), a S-layer homology domain (SLH domain) and a domain of yet unknown function. Mannanase activity was detected on the cell surface and in the culture supernatant. It is believed that the SLH domain mediates transient binding of Man26A to the cell surface and the MBD mediates binding to the substrate. Strong binding of the MBD to soluble mannan was detected and its potential as an affinity tag for protein purification in aqueous 2-phase systems was tested. The 1,4-p-mannosidase (Man2A), cleaves (3-1,4 mannosidic linkages with net retention of the anomeric configuration. Man2A was transformed into the glycosynthase Man2A E519A by mutating the catalytic nucleophile, E519 to alanine. Glycosynthases are retaining glycosidases without hydrolytic but with synthetic activity. Using a-mannosyl fluoride as donor and p-nitrophenyl sugars as acceptors, the glycosynthase Man2A E519A catalyzed the synthesis of [3-1,4 and (3-1,3 mannosidic linkages. In this study the biology of mannan-degradation by C. fimi was investigated and the biotechnological potential of its components was explored. i i Table of Contents Abstract i i Table of Contents i i i List of Tables x List of Figures xi List of Abreviations xiv Acknowledgments xvii Chapter 1: Introduction 1 1.1 Gycosidases 1 1.2 Catalytic mechanisms of glycosidases 2 1.3 Classification of Glycosidases 7 1.4 Strategies for the enzymatic degradation of plant cell wall polysaccharides 9 1.4.1 Complexed systems 10 1.4.2 Non-complexed systems 11 1.5 The cellulolytic-hemicellulolytic enzyme system from Cellulomonas fimi 12 1.6 Overall objectives 15 Chapter 2: Materials and Methods 16 2.1. Buffers, enzymes and chemicals 16 2.2 Bacterial strains, plasmids and phages 16 2.3 Media and growth conditions 19 2.4 Oligodeoxyribonucleotide primers 20 2.5 Recombinant D N A techniques 20 iii 2.5.1 Polymerase chain reaction (PCR) 22 2.5.2.Primer synthesis and D N A sequencing 22 2.6 Detection of enzyme activity 23 2.6.1 Plate assays 23 2.6.2 Zymograms 23 2.7 Library sceening 24 2.8 Production and purification of recombinant proteins 25 2.9 Partial purification of Man2A from C. fimi 26 2.10 N-terminal amino acid sequencing 26 2.11 Determination of protein concentration 27 2.12 Size exclusion chromatography of Man2A 27 2.13 Proteolysis of Man26A 27 2.14 Matrix-assisted laser desorption ionisation time-of-flight mass spectrometry 28 2.15 Localization of Man26A in C. fimi cultures 28 2.16 Affinity gel electrophoresis (AFGE) and Western blotting 29 2.17 Genomic Mapping 30 2.17.1 Preparation of genomic D N A 30 2.17.2 Pulsed field gel electrophoresis 31 2.17.3 Southern blotting of pulsed field gels 31 2.18 Enzymology 32 2.18.1 p H and temperature optima for Man2A 32 2.18.2 Steady state kinetic parameters for P N P M hydrolysis by Man2A 33 2.18.3 Inactivation kinetics of Man2A 33 2.18.4 Reactivation of 2FManpF inactivated Man2A 34 iv 2.18.5 Man26A p H and temperature optima 35 2.18.6 Steady state kinetic parameters for LBG hydrolysis by Man26A 35 2.18.7 Mannan and mannoolicosaccharide hydrolysis by Man26A and Man2A 35 2.19 Mannan hydrolysis of intact and proteolytically cleaved Man26A 36 2.20 Proteolysis of Man2A 36 2.21 Electrospray mass spectrometry (ESMS) 37 2.22 Thin layer chroamtography (TLC) 38 2.23 Transglycosylation product analysis 38 Chapter 3: The Cellulomonas fimi Mannan-degrading System 40 3.1 Introduction 40 3.1.1 Mannan 40 3.1.2 Mannan hydrolyzing enzymes 42 (3-mannanase 42 P-mannosidase 43 a-galactosidases 44 3.1.3 Objectives 44 3.2 Results 45 3.2.1 Identification of galactomannan degrading enzymes in Cellulomonas fimi 45 3.2.2 Carbon source-dependent mannanase synthesis 46 3.2.3 Screening of the C.fimi genomic D N A library for (3-mannanases 48 3.2.4 Nucleotide- and deduced amino acid sequence of the C.fimi mannanase 51 3.2.5 Sub-cloning and production of C. fimi Mannanase, Man26A 59 3.2.6 Analysis of the modular organization of C. fimi Man26A 59 3.2.7 Localization of mannanase activity 65 3.2.8 Is Man26A the only mannanase produced by C. fimi 66 3.2.9 pH- and temperature optimum, and kinetic parameters of Man26A 68 3.2.10 Screening of the C. fimi genomic D N A library for (3-mannosidase 68 3.2.11 Nucleotide- and deduced amino acid sequence from the C. fimi p-marmosidase 71 3.2.12 Sub-cloning and expression of mcmlA 80 3.2.13 p H optimum of C.fimi Man2A WT 80 3.2.14 Temperature optimum and thermostability of Man2A WT 80 3.2.15 Steady-state kinetic parameters for P N P M hydrolysis by Man2A 84 3.2.16 Size exclusion chromatography of Man2A 88 3.2.17 Mannan and mannooligosaccharide hydrolysis by Man26A and Man2A .. .88 3.3 Discussion 93 3.3.1 Molecular architecture of Man26A 93 3.3.2 SLH domain mediated binding of Man26A to C. fimi cells 94 3.3.3 C.fimi p-mannosidase, Man2A 95 3.3.4 Degradation of mannooligosaccharides by Man26A and Man2A 95 3.3.5 Hydrolysis of galactomannan and mannan by Man26A and Man2A 98 Chapter 4: Mannan-Binding Domain 100 4.1 Introduction.. 100 4.1.1 Carbohydrate-binding domains 100 4.1.2 C B D c e x 101 4.1.3 CBDN1 104 4.1.4 Objectives 104 4.2 Results 105 vi 4.2.1 Characterization of a novel domain from C. fimi mannanase, Man26A 105 4.2.2 Sub-clorring of mbdnn 107 4.2.3 Characteristics of the mannan binding domain (MBDim) 109 4.2.4 Mannan-binding domain MBD1112 is not a lectin 113 4.2.5 The role of MBD in mannan degradation 115 4.3 Discussion 117 4.3.1 Mannan-binding domain in C. fimi mannanase, Man26A 117 4.3.2 Comparison of MBDim to C B D N I 118 4.3.3 Lectins 119 4.3.4 The role of MBDim in Man26A catalyzed mannan degradation 119 4.3.5 Protein purification by aqueous two-phase systems 121 Chapter 5: Model of Galactomannan Degradation by C.fimi 169 Chapter 6: Conversion of Man2A into a Glycosynthase 125 6.1 Introduction 125 6.1.1 Enzymatic synthesis of oligosaccharides 125 6.1.1 Enzymatic synthesis of oligosaccharides 125 Glycosyltransferases 125 Glycosidases 126 Glycosynthases 128 6.1.2 Identification of the catalytic nucleophile 130 6.1.3 Objectives 132 6.2 Results 133 6.2.1 Identification of the catalytic nucleophile in C. fimi mannosidase, Man2A 133 vii Prediction of the position of catalytic residues 133 Inactivation of Man2A with 2-deoxy-2-fluoro-(3-mannosyl fluoride.134 Reactivation of inactivated Man2A 139 Analysis of the enzyme-inactivator complex by mass spectrometry.141 Identification of the labeled active site nucleophile by ESMS 141 6.2.2 Conversion of Man2A into a Glycosynthase: Mutation E519A 145 6.2.3 Transglycosylation reaction by Man2A WT 146 6.2.4 Transglycosylation by glycosynthase Man2A E519A 150 6.2.5 Screening for good transglycosylation acceptor molecules 152 6.3 Discussion 155 6.3.1 Man2A inactivation 155 6.3.2 Protection of pepsin cleavage site by 2FMan glycosylation 156 6.3.3 Prediction of catalytic residues by hydrophobic cluster analysis (HCA) 157 6.3.4 Transglycosylation by glycosynthase Man2A E519A 157 Stereospecificity of transglycosylation 157 p H effect on Man2A E519A transglycosylation 158 Acceptor preference of Man2A E519A 159 Chapter 7: Genomic M a p of Cellulomonas fimi 161 7.1 Introduction 161 5.1.1 Genetic organization of cellulase and hemicellulase systems 161 5.1.2 Gene cluster in Cellulomonas fimi 162 5.1.3 Objectives 162 7.2 Results 163 5.2.1 Mapping of man26A and manlA on the C.fimi genome 163 viii 7.3 Discussion 170 7.3.1 Genome size and geometry 170 7.3.2 Genetic and physical mapping of the C. fimi genome 170 Chapter 8: References 172 ix List of Tables Table 2.1 Bacterial strains 17 Table 2.2 Bacteriophages 17 Table 2.3 Plasmids used for cloning and protein expression 17 Table 2.4 Plasmids used for genome mapping 18 Table 2.5 Plasmids used for sequencing of tnan26A 18 Table 2.6 Plasmids used for sequencing of manlA 19 Table 2.7 Primers used for man26A sequencing and PCR 20 Table 2.8 Primers used for mbduu PCR 20 Table 2.9 Primers used for manlA sequencing and PCR 21 Table 4.1 Mannan hydrolysis catalyzed by Man26A and Man26A catalytic domain 116 x List of Figures Figure 1.1 Catalytic mechanism of a retaining (3-mannosidase Figure 1.2 Catalytic mechanism of an mverting (3-mannosidase Figure 1.3 Schematic representation of C. fimi cellulases and xylanases Figure 3.1 Schemtatic structure of galactomannan Figure 3.2 Mannanase secretion by C. fimi Figure 3.3 Restriction mapping of pCMan2 and pCMan4 Figure 3.4 6.8 kbp genomic D N A region containing man26A Figure 3.5 Nucleotide- and deduced amino acid sequence of man26A Figure 3.6 Blast search results for Man26A Figure 3.7 Sequence alignment of Cf Man26A and Pf Man A Figure 3.8 Alignment of SLH repeats of Man26A Figure 3.9 Generation of plasmid pETMan26A Figure 3.10 Protease digestion of Man26A Figure 3.11 Summary of protease digestion of Man26A Figure 3.12 Localization of mannanase activity Figure 3.13 Zymogram of proteolytic Man26A fragments Figure 3.14 pH- and temperature optimum of Man26A Figure 3.15 Steady-state kinetics of LBG hydrolysis Figure 3.16 Nucleotide- and deduced amino acid sequence of man2A Figure 3.17 Blast search results for Man2A Figure 3.18 Multiple sequence alignment of family 2 (3-mannosidases Figure 3.19 H C A plots of bovine and C. fimi (3-mannosidase xi Figure 3.20 Generation of plasmid pETMad 82 Figure 3.21 SDS-PAG of Man2A WT and Man2A E519A 83 Figure 3.22 p H optimum of Man2A 85 Figure 3.23 Temperature of Man2A 86 Figure 3.24 Steady-state kinetics of P N P M hydrolysis 87 Figure 3.25 Size exclusion chromatography 89 Figure 3.26 Mannan hydrolysis products 91 Figure 3.27 Mannooligosaccharide hydrolysis products 92 Figure 3.28 Schematic mannooligosaccharide hydrolysis catalyzed by Man26A 97 Figure 4.1 CBDCex structure 102 Figure 4.2 C B D N I structure 103 Figure 4.3 AFGE of Man2A and Man26A 106 Figure 4.4 AFGE of intact and processed Man26A 108 Figure 4.5 AFGE Western blot of MBDim I l l Figure 4.6 Double reciprocal plot of AFGE data 112 Figure 4.7 Competition AFGE 114 Figure 5.1 Schematic mannan degradation by C. fimi 124 Figure 6.1 Proposed Man2A WT transglycosylation mechanism 127 Figure 6.2 Proposed Man2A E519A transglycosylation mechanism 129 Figure 6.3 Multiple sequence alignment of family 2 catalytic core regions 135 Figure 6.4 Inactivation mechanism of Man2A 137 xii Figure 6.5 Man2A inactivation kinetics "138 Figure 6.6 2Fman-Man2A reactivation kinetics 140 Figure 6.7 LCMS and comparison mapping of Man2A 142 Figure 6.8 ESMS/MS spectrum of unlabeled peptide (MW 1036) 143 Figure 6.9 ESMS/MS spectrum of labeled peptide (MW 1520) 144 Figure 6.10 Generation of plasmid pETMadE519A 148 Figure 6.11 p H optimization of Man2A E519A-catalyzed transglycosylation 151 Figure 6.12 Acceptor preference of Man2A E519A 154 Figure 7.1 Pulsed field gel and Southern blot 164 Figure 7.2 Schematic digestion of C. fimi genomic D N A with Hpa I and Mun 1 165 Figure 7.3 Schematic digestion of C. fimi genomic D N A with Hind I and Nsi 1 166 Figure 7.4 Physical and genetic map of the C. fimi genome 169 xiii List of Abbreviations 2FManpF 2-deoxy-2-fluoro-mannosyl-(3-fluoride A Absorbance aa amino acid AFGE Affinity gel electrophoresis Amp Ampicillin Amp R Ampicillin resistance Amu Atomic mass unit bp Base pair BSA Bovine serum albumin CBD Cellulose-binding domain CD Catalytic domain CIF Collision induced fragmentation CM-cellulose Carboxymethylcellulose Da Dalton DP Degree of polymerization EDTA Ethylenediaminetetraacetic acid ESMS Electrospray mass spectrometry EtOH Ethanol FACE Fluorophore assisted carbohydrate electroph H C A Hydrophobic cluster analysis His6 tag Histidine tag H P L C High pressure liquid chromatography xiv I Inhibitor I N M Ivory nut mannan IPTG Isopropyl- p-D-thiogalactoside ISV Ion source voltage Kan* Kanamycin resistance kcat enzyme turnover constant K d Dissociation constant kDa Kilodalton ki Inactivation rate consant Ki Inhibitor equilibrium binding constant K m Michaelis constant kreact Spontaneous reactivation rate constant Ktrans Dissociation constant k trans Transglycosylation rate constant LB Luria-Bertani medium LC Liquid chromatography m / z Mass to charge ratio Man26A 1,4-B-mannananse Man2A E519A Man2A derived glycosynthase Man2A l,4-(3-mannosidase MBD Mannan-binding domain M C A C Metal chelating affinity chromatography M r relative molecular mass MuaGal Methylumbelliferyl a-galactoside MUpMan Methylumbelliferyl (3-mannoside M W molecular weight N M R Nuclear magnetic resonance PCR Polymerase chain reaction PFGE Pulsed field gel electrophoresis PNPC paranitrophenyl (3-cellobiose PNPG paranitrophenyl (3-gentiobiose PNPGal paranitrophenyl (3-galactose P N P M paranitrophenyl B-mannoside P N P M 2 paranitrophenyl (3-mannobioside rpm Revolutions per minute RT Room temperature SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electroph SLH S-layer homology TFA Trifluoroacetic acid TIC Total ion chromatogram TLC Thin layer chromatography TYP Tryptone, yeast extract, phosphate U V Ultraviolet Vmax Maximal velocity of enzyme-catalyzed reaction X-Glu 5-Bromo-4-chloro 3-indolyl-(3-D-glucoside xvi Acknowledgments I wish to extend my thanks to my supervisor Dr. Tony Warren for giving me the opportunity to pursue my Ph.D. at UBC and for providing continual advice and encouragement. Special thanks go to Dr. Steve Withers for his advice and enthusiasm. I thank Drs. Tom Beatty, Brett Finlay, Neil Gilkes, Doug Kilburn and George Spiegelman for their guidance. M y appreciation also goes to the members of the cellulase lab for their team spirit and friendship. In particular I would like to thank Dr. Andreas Meinke for his mentor- and friendship, Dr. Henrik Stalbrand for helping me to get started on my new project, A l Boraston for many fruitful discussions and for being a very good listener, Brad McLean for his computer tutorials and Dr. Laurent Gal for the transglycosylation discussions. Thanks also go to Lloyd Mackenzie for helping me with the transglycosylation product analysis, Shoumin He for doing all the ESMS work, Renee Mosi, Steve Howard and Dave Vocadlo for substrates and advice. My appreciation also goes to John Smit for his electron microscopy work. I am grateful for the help provided by Emily Kwan and Emily Akow. Many thanks go to Helen Smith for all her work to keep the lab organized and for the many opportunities for cellulase family bonding either at BBQs in her backyard or on the skislopes in Whistler. I appreciate the financial support received from the Protein Engineering Network of Centres of Excellence (PENCE). Last but not least, I am especially grateful to my wife, family and friends for their encouragement and support throughout these many years. xvii 1 Introduction 1.1 Gycosidases Carbohydrates make up most of the organic matter on earth, originating mostly from CO2 assimilation by oxygenic photosynthesizers, such as plants, algae and cyanobacteria (Raven, 1996). Carbohydrates are found in all forms of life, displaying a wide variety of biological functions, such as energy storage, structural components, fuel and metabolic intermediates (Stryer, 1988). Most of the carbohydrates participate in glycosidic linkages, such as N-glycosidic linkages, which are found in the base-deoxyribose linkage of D N A nucleotides, or O-glycosidic linkages, which are most commonly found in carbohydrate-carbohydrate linkages. These linkages are the substrates for glycosidases, i.e. N-glycosidases and O-glycosidases, respectively. The abundance of O-glycosidically linked carbohydrates, their biological and medical importance and their biotechnological potential are the driving force for the extensive studies of O-glycosidases (Davies et al, 1998). One class of O-glycosidases is polysaccharidases, which are key components in the recycling of carbohydrates in nature. Polysaccharidases are involved in the microbial degradation of plant cell wall material or in the degradation of nutritional reservoirs, such as starch, glycogen or certain forms of mannan (Warren, 1996). Glycosidases are also present in lysosomes, where they play an important role in the catabolism of the sugars in sphingolipids, glycoproteins or glycosaminoglycans. The malfunctioning or absence of these enzymes can cause severe diseases, known as lysosomal storage diseases. The loss of p^hexosaminidase activity, for example, results in the accumulation of GM2 ganglioside primarily in neuronal lysosomes, causing the 1 neurodegenerative Tay-Sachs disease (Mahuran, 1995). Other lysosomal storage diseases are Sly syndrome, which is caused by a (3-glucuronidase deficiency, and (3-mannosidosis, which is caused by insufficient (3-mannosidase activity (Neufeld, 1991). Extensive research on glycosidases not only helps to understand the biological process of glycoside hydrolysis, but has and wil l continue to reveal many useful applications. Glycosidases, for example, have become useful tools in the field of glycobiology. Structural information about the sugar moiety of glycoproteins can be obtained by the sequential removal of terrrrinal saccharides using exo-acting glycosidases (Edge et al, 1992). Glycosidases are also valuable for crystallography by improving crystallisation by in-vitro deglycosylation of glycosylated proteins (Grueninger-Leitch et al, 1996). A novel and very promising application in the development of new therapeutics is the use of glycosidase mutants, i.e. glycosynthases, for the enzymatic in-vitro synthesis of glycosidic linkages (Mackenzie et al, 1998; Withers et al, 1998). The biotechnological potential for plant cell wall degrading polysaccharidases of microbial origin is well established. It includes applications such as the conversion of cellulose to fuel, biobleaching of paperpulp, and 'stone-washing' of denim fabric. 1.2 Catalytic mechanisms of glycosidases Glycosidases cleave their substrates between a glycon and an aglycon part. They usually have a very high specificity for the atomic nature of the glycon, but fairly low specificity for the aglycon. Aglycons, in nature mostly sugars, can therefore be replaced by chromogenic or fluorogenic substrates such as dinitrophenyl or methylumbelliferyl groups, which make very useful analytic substrates. Two major mechanisms are known for the 2 cleavage of the glycosidic linkage by glycosidases; cleavage with net retention of the anomeric configuration (retaining enzymes) or with net inversion of the anomeric configuration (inverting enzymes) (Sinnott, 1990). As was proposed by Koshland 45 years ago, the mverting mechanism works via a single chemical step, and the retaining mechanism via a double displacement step (Koshland, 1953). The double-displacement mechanism of retairiing glycosidases is facilitated by acid/base catalysis. The reaction proceeds in two reaction steps, a glycosylation and a deglycosylafion step (Figure 1.1). Generally two carboxylic side chains are involved. One acts as an acid and base catalyst, whereas the second carboxylic side group acts as nucleophile. The glycosidic oxygen is protonated by the acid catalyst, and the nucleophile attacks the anomeric center of the glycon, thereby forming a covalent sugar-enzyme intermediate with an inverted anomeric configuration. In the second step, the deglycosylation, water is deprotonated by the acid/base catalyst, increasing its nucleophilicity. Its attack on the anomeric center of the glycosyl-enzyme intermediate, reverses the anomeric configuration again, resulting in an overall retention. The glycosylation and the deglycosylation steps proceed via oxocarbenium ion-like transition states (Davies et al, 1998). A well studied retaining enzyme is the C. fimi exoglucanase/xylanase Cex, a member of family 10 (Notenboom et. al, 1998). Kinetic studies have shown that Glul27 acts as the acid/base catalyst and Glu233 as the catalytic nucleophile (MacLeod et al, 1994; Tull et al, 1991). The covalent ester linkage between the enzyme nucleophile and the sugar was demonstrated with mechanism based inhibitors. 3D structure analysis of the retaining enzymes Cex (White et al, 1994), E. coli (lacZ) [3-galactosidase (Jacobson, 1994) and the human ^glucuronidase (Sanjeev et al, 1996) indicated that the carboxylate oxygens from the two catalytic residues are about 5.5 A apart 3 4 5 (McCarter and Withers, 1994). The catalytic nucleophile does not have to be part of the enzyme; it can be part of the substrate. The 2-acetamido side chains, in chitinase or hexosaminidase substrates can attack the anomeric center, forming an oxazoline intermediate. This intermediate corresponds to the glycosyl-enzyme complex. Such a pathway could explain the lack of a catalytic nucleophile in the chitobiase from Serratia marcescens (Tews et al, 1996). In inverting glycosidases, the protonation of the glycosidic oxygen and the deprotonation of the water molecule is catalyzed by carboxylic amino acid side groups, from either aspartates or glutamates. The thereby deprotonated, nucleophilic water can then attack the anomeric center from "the opposite side", resulting in a net inversion of the anomeric configuration (Figure 1.2). As was shown by kinetic studies for the family 6 endoglucanase CenA from Cellulomonas fimi, two aspartic acid residues, D252 and D392, act as catalytic acid and catalytic base, respectively. These catalytic residues are conserved in all members of family 6 (Damude et al, 1995). From 3-dimensional structures, e.g. of cellobiohydrolase Cbhll from Trichoderma reseei (Rouvinen et al. 1990) and endo-glucanase E2 form Thermomonospora fusca (Spezio et al, 1993), it was found that the two carboxylate oxygens from the catalytic residues are about 9.5 A apart. This distance is believed to be important to enable simultaneous attack of the glycosidic oxygen by the acid catalyst and the nucleophilic attack at the anomeric center by the deprotonated water (McCarter and Withers, 1994). 6 1.3 Classification of Glycosidases A great variety of O-glycosyl hydrolases (EC 3.2.1.-) have been characterized. They are classified into 62 families based on amino acid sequence similarities (Henrissat, 1991; Henrissat, 1993; Warren, personal communication). Since there is a direct relationship between sequence similarities and folding similarities, this classification allows better structural, evolutionary and mechanistic predictions. Not all members of a family have the same substrate specificity: family 1 comprises P-glucosidases, 6-phospho-p-galactosidases, 6-phospho-p-glucosidases, myrosinases, lactase-phlorizin hydrolases and one p-mannosidase (Bauer et al, 1996). This discrepancy between sequence similarity and different substrate specificity is indicative of evolutionary divergence. The stereochemical outcome of hydrolysis however, is the same for all members within a family (e.g. retaining in family 1). Members within a family also adopt a similar fold. A commonly used tool for predicting the secondary structure of a protein, i.e. the folding, is hydrophobic cluster analysis (HCA) (Gaboriaud et al, 1987). H C A was used to compare the folding of enzymes with known 3D structures from families 1, 2, 5, 10, and 17 with other enzymes from families 1, 2, 5, 10, 17, 26, 30, 35, 39, 42 and 53. Even though they share low levels of sequence identities, H C A analyses revealed that the catalytic domains from members of these families have similar (p/oc) 8 folds. The acid/base catalyst and the nucleophile were found to be located at the C-terminal ends of strands P4 and P7, respectively. These families were placed, based on their structural similarities, into the clan GH-A (Henrissat et al, 1996). Four more of these structural superfamilies were defined; clans GH-B, GH-C, GH-D and GH-E. Based on these classification premises, and the use of H C A , the location of the eight P strands and functional groups were predicted for various 7 lysosomal enzymes of the clan GH-A, such as the human P-glucosidase (family 30), human B-galactosidase (family 35) and human P-marvnosidase (family 2) (Durand et al, 1997). A further characteristic of glycosidases is their overall topology. 3D structures from more than 22 families have shown, that regardless of protein fold and reaction mechanism variety, the glycosidases have only three different overall topologies. The pocket or the crater topology is optimal for the recognition of saccharide extremities, and is encountered in monosaccharidases, such as p-galactosidases and p-glucosidases. The cleft or groove is an open structure, which allows random binding of several sugar units. This form is commonly found in endo-acting polysaccharidases, such as endo-cellulases and chitinases. The third topology found is the tunnel, which is basically a cleft covered by long loops. This form is commonly found for exo-acting enzymes such as cellobiohydrolases (Davies et al, 1995). 1.4 Strategies for the enzymatic degradation of plant cell wall polysaccharides The growing plant cell is protected and structurally supported by its primary cell wall. As the plant cell ceases to grow, the secondary cell wall is formed. The secondary cell wall is composed of several lamellae, within which cellulose, a p-l,4-glucose polymer, forms microfibrils that are organized in parallel. Each lamella has a different microfibril orientation. The microfibrils are usually embedded in a matrix of hemicellulose and lignin, forming intimate interactions between the polymers (Tomme et al, 1995). In secondary cell walls, the hemicellulose mainly consists of xylans (P-l,4,-xylose backbone) substituted with a variety of side chains, mannans (P-l,4,-mannose backbone), and glucomannans (mannan with glucose substitutions). Lignin is a highly branched polymer generated by 8 condensation of aromatic alcohols, coupled to hemicellulose by ester bonds. Only a few organisms are able to degrade lignin by a process involving radical oxidation by peroxidases (Beguin et al., 1994). The complexity of the natural substrate, i.e. the plant cell wall, led to enzymatic degradation studies using isolated substrates, such as celluloses, xylans and mannans. Since cellulose is the most abundant polysaccharide in the plant cell wall, cellulose degradation has been studied extensively in many microorganisms. These taxonomically and ecologically very diverse cellulolytic microorganisms generally secrete a variety of cellulases and hemicellulases, which can be divided into two broad categories: non-complexed and complexed systems (Gilbert and Hazelwood, 1993). The non-complexed system, also called a non-aggregating system, is characteristic of aerobic fungi and bacteria: cellulases and related polysaccharidases are secreted into the culture medium without forming aggregates. The secreted enzyme in complexed systems, typically produced by anaerobic microorganisms, are aggregated into multienzyme complexes, which can be either cell-bound or released into the culture medium (Beguin and Lemaire, 1996). This distinction between cellulase systems of aerobes and anaerobes applies generally, but is not absolute. 1.4.1 Complexed systems Secreted cellulases and hemicellulases of some anaerobic microorganisms can associate into large, multienzyme complexes that are very efficient at hydrolyzing crystalline cellulose. These multienzyme complexes are called cellulosomes. One of the best studied organisms producing such cellulosomes is the anaerobic thermophile, Clostridium 9 thermocellum. Up to several hundred cellulosomes form protuberances on the outermost layer of the cell envelope. The cellulosomes have both cellulolytic and xylanolytic activities. C. thermocellum, however, is unable to grow on the degradation products of xylanolysis (Bayer et al, 1996; Beguin and Lemaire, 1996). Cellulosomes are initally cell-bound, but can then detach from the cells without significant loss in enzymatic activity. Their additional ability to bind to cellulose results in adhesion of the cells to the substrate. The binding to cellulose is mediated by a cellulose-binding domain (CBD; type 3), which is part of the scaffoldin protein, CipA. CipA has nine cohesin domains, which are binding sites for the dockerin domains present in most cellulolytic C. thermocellum enzymes. Dockerin domains are found in at least nine endo-glucanases, one xylanase, one lichenase and one cellobiohydrolase. The scaffoldin protein CipA itself possesses a C-terminal dockerin domain (type II), which fails to interact with its own cohesin domains (Bayer et al, 1994). It binds to a specific set of cohesin domains (type II), which are found in several cell-bound polypeptides (e.g. OlpB and SdbA). These anchoring proteins have, furthermore, a triply repeated sequence, called the SLH domain. This domain has homology to domains found in S-layer proteins, which mediate non-covalent binding of the proteins to the peptidoglycan, or to other cell envelop components (Lemaire et al, 1998; Egelseer et al, 1998). This binding relay results ultimately in binding of the cellulolytic enzymes to the substrate and to the C. thermocellum cell wall. The fungal cellulolytic system from Pyromyces sp. is also aggregated in a cellulose binding complex, which is comparable to the C. thermocellum complex. It comprises at least 10 different polypeptides, including cellulase, xylanase and mannanase activities (Ali et al, 1995). 10 1.4.2 Non-complexed systems: A well studied, non-complexed system is the plant cell wall-degrading system from Trichoderma reesei. This soft-rot, aerobic fungus secretes two major cellobiohydrolases, CBHI and CBHII (Teeri et al, 1987), at least four endoglucanases, EGI, EGII, EGIII and EGV (Saloheimo et al, 1994), two xylanases, Xynl and Xynll (Torronen et al, 1995), at least two mannanases (Stalbrand et al, 1995), one B-glucosidase and one B-mannosidase (Stalbrand personal communication). Synergistic interactions of isolated components from non-complexed systems have been demonstrated in several independent studies. Commonly, the combined activity on crystalline, but not soluble, cellulose was greater than the sum of their individual activities (Tomme et al, 1995). Synergism was found not only between endo- and exo-acting glucanases (endo-exo synergism) but was also reported to occur between two exo-acting cellobiohydrolases, i.e. CBHI and CBHII from T. reesei (Nidetzki et al, 1994). Cellulomonas fimi, a gram positive, mesophilic soil bacterium is another producer of a well studied, non-complexed cellulolytic-hemicellulolytic enzyme system. Since this organism is being investigated in the present study, its system wil l be described in detail (Section 1.5). 1.5 The cellulolytic-hemicellulolytic enzyme system from Cellulomonas fimi Cellulomonas fimi, a member of the actinomycete family Cellulomonadaceae (Stackerbrandt and Prauser, 1991) can use a variety of polysaccharides as carbon sources. C. fimi secretes several glucanases and xylanases into the culture medium (Figure 1.3). These 11 enzymes all have a modular architecture, composed of two or more domains. The domains are usually separated by linker sequences, such as the proline-threonine rich linker regions found in CenA and Cex (Wong et al, 1986; O'Neill et al, 1986; Gilkes et al, 1991). Four endoglucanases, CenA, CenB (Owolabi et al, 1988), CenC (Coutinho et al, 1991) and CenD (Meinke et al, 1993), two cellobiohydrolases, CbhA (Meinke et al, 1994) and CbhB (Shen et al, 1995), one mixed function p-l,4-xylanase/p-l,4,-glucanase (Cex) and two xylanases, XYLC (Clarke et al, 1996) and XYLD (Millward-Sadler et al, 1994) have been analyzed to date. Both the cellulases and the xylanases from C. fimi have one or more cellulose binding domains (CBD) that bind to soluble and/or insoluble cellulose. Fibronectin type 3 repeats (Fn3) are domains found in CenB, CenD and in the two cellobiohydrolases, CbhA and CbhB. Their function is as yet unknown but it is speculated that they might be involved in protein-protein interactions or might just be used as spatial separation between the catalytic domain and the CBD (Tomme et al, 1995). NodB is a domain which is homologous to the N-acetylglucosamine deacetylating enzyme NodB from Rhizobium. The NodB homologues are prevalent among xylanases from different soil bacteria; their function, however, is as yet unknown (Millward-Saddler et al, 1995). The unusual high temperature optimum of the xylanase XYLC (60° C), might be explained by the presence of a thermostabilizing domain, adjacent to the catalytic domain (Clarke et al, 1996). Endo-glucanases and exo-glucanases interact synergistically to optimize cellulose degradation (endo-exo synergy). The endo-glucanases attack the cellulose at disordered, or amorphous regions and thereby provide new sites for the attack of exo-acting enzymes. Differences in initial attack were seen for the four C. fimi endo-glucanases (Kleyman-Leyer et al, 1994; Tomme et al, 1996). Like the T.reesei cellobiohydrolase CBHI, C. fimi 13 cellobiohydrolase B (CbhB) attacks the p-l,4-glucan chain from the reducing end, whereas CbhA and the corresponding T.reesei CBHII attacked the substrate from the non-reducing end (Gilkes et al, 1997; Nidetzky et al, 1994). The major end product from the concerted cellulose degradation by secreted C. fimi endo- and exo-cellulases is cellobiose, which once imported into the cells can be further processed into glucose (Wakarchuk et al, 1984, Kim and Pack, 1989). The biological significance of the secretion of at least four endo-glucanases by Cellulomonas fimi is currently unclear. It is possible that the enzymes are more versatile than assumed and that the classification into exo- and endo- acting enzymes is too simplistic, as was shown for CenC. CenC, which was classified as an endo-glucanase, has not only endo-but also exo-activity on CM-cellulose. It is able to hydrolyze linkages within the P-1,4-glucan chains, and then to move processively along the chain, releasing cellobiose. This process was classified as a semiprocessive mechanism (Tomme et al, 1996). One way to study the role of each cellulase in the biological process of cellulose, or even plant cell wall, degradation would be a genetic approach, i.e. regulation and knockout studies. This approach, however, was obstructed by the inability to manipulate Cellulomonas fimi genetically (results not shown). 14 1.6 Overall objectives The overall objective of this study was to analyze the ability of Cellulomonas fimi to degrade mannan, an important hemicellulose component. It was of scientific interest to analyze the mannan degrading components in order to compare the mannanolytic system to the cellulolytic system and to find out whether the C. fimi cellulolytic system is a blueprint for polysaccharide degradation in this organism. The project was also driven by application-based objectives. The finding of novel carbohydrate binding domains in secreted enzymes might be a very valuable contribution for the study of protein-carbohydrate interactions and the finding of an enzyme with transglycosylation activity might provide a good candidate for the glycosynthase approach to enzymatic synthesis of glycosidic linkages. 15 2 Materials and Methods 2.1 Buffers, enzymes and chemicals Buffers and solutions used in this study were generally prepared as described by Sambrook et al. (1989). A l l buffers and solutions were sterilized by autoclaving or filtration. Locust bean gum (LBG) and carboxymethylcellulose were purchased from Sigma, ivory nut mannan, azo-carob galactomannan and mannooligosaccharides were obtained from Megazyme (Australia) and birchwood-xylan from Roth (Germany). Zeocin™ was purchased from Invitrogen (USA), Anti His6 tag mouse antibodies from Dianova, and anti-mouse IgG-horse radish peroxidase conjugate from Dako Diagnostics (Canada). Potassium was generally used as counterion in phosphate buffer. 2-deoxy-2-fluoro-B-D-mannosyl fluoride (2FManBF) and a-mannosyl fluoride were gifts from Lloyd Mackenzie, Dept. of Chemistry, UBC. C B D c e n D was a gift from A l Boraston (Cellulase lab, UBC), and C.fimi protease a gift from Emily Kwan (Cellulase lab, UBC). 2.2 Bacterial strains, plasmids and phages The bacterial strains, plasmids and phages used in this study are described in Tables 2.1-2.5. Bacterial stocks were maintained at -70° C in LB medium containing 15 % glycerol. Plasmid D N A was stored in water at -20° C. Phages were stored in TYP medium at 4° C or -70" C. 16 Table 2.1.: Bacterial strains Bacterial strain Genotype Reference or source Cellulomonas fimi wild type ATCC 484 Escherichia coli strains: BL21(DE3) F- ompT hsdSB(rB-niB-)gal dcm (DE3) Grodberg et al. 1988 DH5a F- endAl, hsdR17 (rk, mk+), supE44, thi-1, Hanahan, 1983 recAl, (argE-laczya)U169, <f>lacZM15 JM101 supE thi A(lac-proAB)[F traA36 Yanisch-Perrom proABlacIqZAM15] etal, 1985 XLl-Blue endAl, hsdR17, supE44, thi-1, recAl, Jerseth et dl, gyrA96, relAl, lac, [¥', proAB, ladiZAM15, 1992 TniO(tetR) XLOLR/SOLR el4-(mcrA)A(mcrCB-fcsdSMR-mn-)171 Stratagene sbcC recB red] «rawC::Tn5(KanR), uvrC lac gyrA96 relAlthil endAl XR[F proAB lach Aml5] Su-Table 2.2: Bacteriophages Bacteriophages: description: Source: ExAssist A. ZapII-C. fimi genomic library f l derived helper phage used for Stratagene pBluescript excision from X ZapII lambda C. fimi expression library Stratagene Table 2.3: Plasmids used for cloning and protein expression Plasmid name: Description: Source: pZErO™ cloning vector; ZeocinR Invitrogen pBluescript SK and KS cloning vector; Amp R Stratagene pET27b and pET28a expression vector; Kan R Novagen 17 Table 2.4: Plasmids used for genome mapping plasmid name: description reference: p T A L l pDAMl-5 p A L M l . l pTCln pTZ18R-cex pTZ18R-1.6cenA PTugSH3 pET28aMad pET27Man26A pTZ18R; cenB; Amp R pTZ18R; cenD; Amp R pTZ18R; cbhA; Amp R pTug07K3; cenC; Kan R pTZ18R; cex; Amp R pTZ18R; cenA; Amp R PTugE07K3; cbhB; Kan R pET28a; manlA; Kan R pET27b; man26A; Kan R Meinke etal, 1991 Meinke et al, 1993 Meinke et al, 1994 Tomme ef al, 1996 MacLeod'eraZ.,1994 Damude al, 1996 Shenef fl/.,1995 this study this study Table 2.5: Plasmids used for sequencing of man26A plasmid name: description: pCMan2 pCMan4 pCMan30 pBSManPst0.9 pBSManPstl pBSManPstl.9 pBSManXho0.9 pCMan30AXho pCManBam/XhoO. 9 pCManABam pCManAKpn pBluescript (SK); 4.3 kbp man26A genomic C. fimi D N A insert (at EcoR I); Amp R . pBluescript (SK); 6.3 kbp man26A genomic C. fimi D N A insert (at EcoR I); Amp R . pBluescript (SK); 3.0 kbp man26A genomic C. fimi D N A insert (at EcoR I); Amp R . pCMan2 0.9 kbp Pst I fragment cloned into pBluescript (KS); Amp R . pCMan2 1 kbp Pst I fragment cloned into pBluescript (KS); Amp R . pCMan2 1.9 kbp Pst I fragment cloned into pBluescript (KS); Amp R . pCMan2 0.9 kbp Xho I fragment cloned into pBluescript (KS); Amp R . pCMan30 with deletion of all Xho I fragments; Amp R pCMan2 0.9 kbp BamH l-Xho I fragment cloned into pBluescript (KS); Amp R . pCMan2 with deletion of all BamH I fragments; Amp R pCMan2 with deletion of all Kpn I fragments; Amp R 18 Table 2.6: Plasmids used for sequencing of manlA plasmid name: description: pCMadl pBluescript (SK); 4.3 kbp manlA genomic C. fimi D N A insert (at EcoR I); Amp R pCMadll pBluescript (SK); 4.3 kbp manlA genomic C. fimi D N A insert (at EcoR I); Amp R pBSMadPst/Bam pCMadl 1.1 kbp Pstl-Xhol fragment cloned into pBluescript (KS); Amp R . pCMadABam pCMadl with deletion of BamH I fragment; Amp R pCMadAPst pCMan2 with deletion of all Pst I fragments; Amp R pBSMadXho/Pst pCMadl 1.1 kbp Pst l-Xho I fragment cloned into pBluescript (KS); Amp R . pCMadABamAXho pCMadABam with deletion of Xho I fragment 2.3 Media and growth conditions E. coli strains were grown in LB medium (10 g tryptone, 5 g yeast extract, 10 g NaCl per liter) at 37° C, or for gene expression at RT in TYP (16 g tryptone, 16 g yeast extract, 5 g NaCl and 2.5 g K2HPO4 per liter; p H 7.0) supplemented with either 50 ug kanamycin or 100 (ig ampicillin mL-1. pZErO™ clones and subclones thereof were grown in LB low salt (LB with only 0.5% NaCl) and 50 ug/mL Zeocin. C. fimi was grown in basal salt medium (1 g NaNC>3, 1 g K2HPO4, 0.5 g KC1, 0.5 g MgS04.7H20, 0.5 g yeast extract per liter; p H 7.2) supplemented with 0.2 % (w/v) carbon source (Stewart et ah, 1976). Kanamycin was added at 50 ug/mL. C. fimi cultures were grown at 30° C with agitation at 220 rpm. Solid media contained 1.5 % agar. 19 2.4 Oligodeoxyribonucleotide primers Primers used for sequencing and PCR reactions are summarized in Tables 2.6-2.8. The primer sequences are indicated in the 5' end to 3' end orientation. Table 2.7: Primers used for man26A sequencing and PCR Primer name: Primer Sequence: Manl GTC TTC GGC TGG G A C A C G Man2 GTGTCGAGGCCGAACACG Man3 GAT C A A GGC C G A CCC CGT Man 31 GGA GCT CTA CCG GTT C A C Man 32 CGT TCT TGA CGT CGG AGT Man 33 G A C CTG TTC CGT CCT G A C Man 34 C A A CGC TAC ATG G A G A C G Man 35 GTG CTG TCG TAG GTG TCG Man 37 CTC GTA GGC G A C A C C A C C Man 41 C A A CGT CTA CGT C A A CG Man 42 GGT A C C TGT A C C GGT TCG A C Man 43 A C C TCT A C C TCA A C G C A G GC Man 44 CGC G A G GTG GTA GTC GGA CC Man 45 AGT A C G CCG A G A CGT GGA T Man 46 CGC GGT CCG ACT A C G A C C T Man 47 TAC A A C GTC GTG A G G TCG A G Man 48 TGA G C A CCC C G A A C G TCG A C Man 50 (PCR) GTC C A G GCT A G C GCG GCG C A G CTC G A C GTC GTC (Nhe I-site) Man 51 (PCR) G G A GGT CAT ATG A C G A A C CGC A G C A G C CGT CCG (Nde I-site) T3 ATT A A C CCT C A C T A A A G T7 A A T A C G ACT C A C TAT A G Table 2.8: Primers used for MBDim PCR Primer name: Primer Sequence: MBD 11 A G C GCG C A G CTC G A C A A C A C C ATG GGC A C C GTC A C C GCG A C G GCG (Nco I-site) MBD12 GCC GGG GTG GGC GGC GAT GGC GGC CGC G A C G A G C G A GCC C G A CGC (Not I-site) 20 Table 2.9: Primers used for manlA sequencing and PCR Primer name: Primer Sequence: M a d l G G A CCG TCA CGT TCG A C G Mad2 TCG GTG CGC A C G C A C GTC Mad3 TCG TTG A C G A C C GCG A G C Mad4 (PCR) GCG TCC ATG GTC A C C C A G G A C ATC TAC (Ncol-site) Mad5 CTC GGC GTG C A C GAT CGC GG MadlO G A G GTC G A C A C G A C G M a d l l CCC ACT CGT GGT GCG Madl2 CCT TGA C G A A G A CGG Madl3 TCG C C A CCG CTC G A C Mad21 G A A TGA TCC CGG GTA CCT Mad22 ATC TTG CGG A C C ATG TTG Mad23 A A C ATG GTC CGC A A G ATG Mad24 GTT G A C CGG A A C GTG A A C Mad25 ATG A A G CCC C A G A G G TTC Mad26 GTC TGG C A G G A C TTC CTC Mad27 C C A G C A GTC GTT TGA GCT Mad28 CGG TTC TGC TCC G A G TTC Mad29 C G A CCG GTA GTG CTC GAT Mad51 (PCR) GCG A A C CGT TCA GCG GCC GCG CGG G A C TGC TGG (Noil-site) MadE519A (used CGG C A C GCG TGA GGG TCG A C C A C G TCG TCG GCG GGC CCT to be GGA A G C C G A A G G CGG A G C A G A A C C MadE622A) (PCR) T7 and T3 see table 2.6 2.5 Recombinant D N A techniques Recombinant D N A work was generally carried out as described by Sambrook et al, 1989. D N A fragments were isolated from agarose gels and purified using the Qiaex II Gel Extraction Kit (Qiagen). Restriction endonucleases were used as recommended by the suppliers (New England Biolabs, GIBCO BRL or Bohringer Mannheim). Ligations were performed with 100-500 ng of total D N A at insert to vector molar ratios of approximately 10:1 in 20 uL reaction volume. 1 U T4 D N A ligase (GIBCO BRL) was 21 used per reaction. Ligations were incubated at 23° C for approximately 2 h. Reaction mixtures were desalted by butanol precipitation (Thomas et ah, 1994). E. coli cells were transformed by either electroporation (GenePulser, BioRad) or by heat shock of cells prepared by the quick chemical competent cell protocol in TSS buffer (10 g bacto tryptone, 5 g yeast extract, 5 g NaCl, 100 g PEG 4000, 50 mL DMSO, 50 mL I M MgCl 2 per liter; p H 6.5) (Dower, 1987). 2.5.1 Polymerase chain reaction (PCR) PCR conditions using C. fimi D N A templates were as follows: 100 p.L reaction volume with 10-100 ng D N A template, 40 pmol of each primer, recommended polymerase buffer, 10 % DMSO, 0.2 m M 2'-deoxynucleoside 5'triphosphates and 1 unit of either Vent polymerase (New England Biolabs), PWO or HiFi polymerase (both Bohringer Mannheim). D N A was amplified by repeating cycles of denaturation ( 94° C, 30 s), annealing (64-67° C, 30 s) and elongation steps (72° C, 45-90 s) 30 times. 2.5.2 Primer synthesis and D N A sequencing Oligodeoxyribonucleotide primers were synthesized by the UBC Nucleic acid and Protein synthesis unit (NAPS) with an Applied Biosystem D N A synthesizer and purified by precipitation with n-butanol (Sawadogo and VanDyke, 1991). D N A sequencing was performed by NAPS using an Applied Biosystem D N A sequencer Model 377 (Perkin-Elmer). Sequencing reactions were performed according to the AmpliTaq dye termination cycle sequencing protocol with the addition of 5 % DMSO. 22 2.6 Detection of enzyme acitvity 2.6.1 Plate assays Mannanase activity was assayed on 2 % agar plates containing 0.5 % LBG in 50 m M buffer (pH 7.0) (Stalbrand et al, 1993). Ten to 20 uL of sample were deposited onto the plate, which then was incubated for 12 to 16 h at 37° C. Hydrolysis of the polymer was visualized by flooding the plate with Congo red solution (2 mg/mL) followed by two washing steps with 1 M NaCl (Teather and Wood, 1982). E. coli clones producing mannanase activity were detected on LB plates (Section 2.2) contairiing 0.2 % azo-carob galactomannan. Azo-carob galactomannan is a galactomannan dyed with Remazol Brilliant Blue (McCleary, 1978). After 12 to 16 h at 37° C zones of mannan hydrolysis (halos) were detectable (Braithwaite et al, 1995). To detect (3-mannosidase or a-galactosidase activity, 0.1 m M to 0.5 m M 4-methylumbelliferyl (3-mannoside (MUpMan) or 4-methylumbelliferyl oc-galactoside (MUccGal), respectively were added to LB plates. The release of fluorescent methylumbelliferol was detected under UV light (Wakarchuk et al, 1984). 2.6.2 Zymograms Zymograms for the detection of polypeptides with mannanase activity were prepared as 10 % SDS-polyacrylamide gels (SDS- PAGE) (Laemmli, 1970) supplemented with 0.5 % azo-carob galactomannan (adapted from Braithwaite et al, 1995). Equal volumes of 2 x non-reducing loading buffer were added to samples: 125 m M Tris (pH 6.8), 5 % SDS, 0.3 % 23 bromophenol blue, 50 % glycerol. Samples (18 uX/lane) were separated at constant voltage (200 V) using a Bio-Rad Mini-PROTEAN™ apparatus. To detect enzyme activity, gels were rinsed and then incubated in 200 mL 50 m M phosphate buffer (pH 7.0) at 37° C until halos were visible (2 to 16 h). To detect B-mannosidase or a-galactosidase activity, non-reducing SDS-PAGE gels were rinsed in 50 m M phosphate buffer p H 7.0 and incubated in 1 m M MUBMan and 1 m M MUaGal , respectively for approximately 5 min. Hydrolysis was observed under UV (Wakarchuk et al, 1984). To detect B-glucosidase activity, gels were incubated in 0.5 m M 5-bromo-4-chloro-indolyl B-D-glucoside (X-Glu) (Kohchi et al, 1996). 2.7 Library sceening The C. fimi lambda ZapII library prepared by Stratagene was screened as recommended by the supplier with a few changes: 150 mm diameter petri dishes were filled with 30 ml N Z Y medium (5 g NaCl, 2 g MgS0 4 .7H 2 0, 5 g yeast extract, 15 g agar and 10 g casamino acids per liter). On top of the solidified NZY medium 10 mL of 0.4 % Azo-carob galactomannan containing 1.5 % agar solution was added. The plates were dried for several hours before the overlay of phage and host cells mixed in 6.5 mL NZY top agar (0.7 % agar, 0.1 to 0.5 m M IPTG) was poured on. Mannanase activity was detected as halos, which were visible after incubation periods of 16 to 24 h at 37° C (adapted from Braithwaite et al, 1995). After secondary and tertiary screeriings, mannanase positive phagemids (pBluescript SK) containing genomic D N A inserts were excised in vivo from lambda DNA, recircularized and transduced into E. coli XLOLR host cells (Shen et al, 1995). 24 The excised form of the entire genomic library was used to screen for (3-mannosidase producing E. coli clones. Only replica plates containing 0.5 m M MUBMan were induced with 0.3 m M IPTG. 2.8 Production and purification of recombinant proteins Genes were expressed from pET vectors (Novagen) in E. coli BL21(DE3) host cells. Cells were grown at RT and 150 rpm in TYP supplemented with 50 ug/mL kanamycin to mid exponential phase. Protein synthesis was induced with 0.2 to 0.4 m M IPTG at RT for 24 to 36 h. Cultures were centrifuged at 5000 rpm for 10 min. Either the supernatant, as for the purification of secreted Man26A, or the cells for the purification of Man2A WT, Man2A E519A, M B D 1 1 1 2 , or Man26A, were further processed. Cells from 1 or 2 liter cultures were resuspended in l x Binding Buffer (5 m M imidazole, 500 m M NaCl, 20 m M Tris-HCI p H 7.9) and ruptured by passing them 3 times through the French pressure cell. Insoluble cell debris was sedimented for 30 min at 15 k rpm and the cell-free extract was loaded onto a 5-10 mL His-bind® metal chelating affinity column (MCAC) at a flow rate of 0.5 mL/min. Column chromatography was performed as recommended by the supplier (Novagen). Culture supernatant was concentrated to 50 mL and buffer exchanged with Binding Buffer using an Ultrasette™ tagential flow concentrator with a 1 kDa cut off (Filtron) . The concentrate was loaded onto a 15 mL His-bind column (Novagen) at a flow rate of 0.5 mL/min. Columns were washed with 3 to 6 column volumes of Binding Buffer at 1 mL/min. Proteins were eluted by step wise increases in the concentration of immidazole from 0 to 0.5 M imidazole elution buffer (0.5 M NaCl, 20 m M Tris-HCI p H 7.9) at a flow rate of 1 25 mL/min. The desired proteins were eluted with 50 m M to 120 m M imrnidazole, with the exception of MBDim which eluted at 20 mM. The appropriate fractions were pooled, buffer exchanged and concentrated by diafiltration through an Amicon PM10 membrane. The M B D i m fraction was concentrated with a Microsep™ Microconcentrator (3 kDa cutoff; Filtron). Protein purity was estimated by Coomassie-stained SDS-PAGE gels. 2.9 Partial purification of Man2A from C. fimi C. fimi cells from a 2 L LBG culture were collected by centrifugation (10 min at 5 k rpm). The resuspended cells were ruptured by passing them 4 times through the French pressure cell. Debris and unbroken cells were removed by centrifugation at 40 k rpm for 1 h. The supernatant was buffer exchanged with 20 m M Bis-Tris HC1 p H 5.8. The sample was fractionated by anion-exchange chromatography (EconoQ, Bio-Rad) with a linear gradient of NaCl (0 m M to 700 mM) in 20 m M Bis-Tris HC1 p H 5.8 (20 column volumes). Fractions were tested for B-mannosidase activity with 200 uM MUpMan. The fractions with the highest activity were pooled, concentrated and prepared for N-terminal sequence analysis (Section 2.11). 2.10 N-terrrvinal amino acid sequencing Protein samples were subjected to SDS-PAGE (Laemmli, 1970) then electroblotted onto polyvinylidene difluoride (PVDF) membranes (Millipore). Protein bands were visualized by Coomassie staining (0.025 % Coomassie in 40 % MeOH) and the bands of interest were excised and submitted for N-terminal sequence analysis by the NAPS unit at University of British Columbia or by the Protein Microchemistry Facility, University of 26 Victoria. The N-terminal sequence was determined by automated Edman degradation, using a Perkin Elmer Applied Biosystems 476A gas-phase sequenator (Matsudaira, 1990). 2.11 Determination of protein concentration Protein concentrations were determined from A280 using the Beer-Lambert law (Absorption (A)= extinction coefficient (s) x path length (1) x molar concentration (c)). The molar extinction coefficient (e) was calculated according to equation E 2.1 (Pace et al, 1995). E 2.1: s(280)(M-i.cm-i)= (#Trp)(5,500) + (# Tyr)(l,490) + (#Cys)(125) 2.12 Size exclusion chromatography of Man2A Fractogel TSK HW-55S from EM-Science (Merck) was used as matrix for size exclusion chromatography of Man2A (100 ug) with a 400 x 10 mm column. A buffer of 50 m M phosphate p H 7.0 with 100 m M NaCl was used (flow rate of 0.5 mL/ min) as mobile phase. The void volume was determined with Blue Dextran, and the column was equilibrated with ferritin (440 kDa), catalase (233 kDa), aldolase (158 kDa) and ovalbumin (50 kDa) (Gehrmann et al, 1994). The sample volume was 50 \\L. 2.13 Proteolysis of Man26A Man26A was treated with C. fimi protease at 37° C (1 U protease/0.55 nmol Man26A in 20 m M Tr i s /HCl p H 7.5 (Meinke et al, 1992)). Samples were removed and immediately 27 added to SDS loading buffer and heated at 100° C for 2 min. Samples were analyzed by 12 % SDS-PAGE (Laernmli, 1970) with approximately 6 ug protein loaded per lane. 2.14 Matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOFMS) Protein samples (0.5 to 3 mg/ mL) were desalted by drop dialysis for 12 to 16 h by depositing 5 to 10 pL drops on dialysis membrane disks (MF, pore size 0.025 um, Millipore) floating on water (500 mL). One uL of dialyzed sample was loaded onto the sample holder and dried for 5 min. The sample was overlaid with 1 ul matrix (supersaturated sinapiruc acid solution in 70 % acetonitrile, 0.1 % trifluoroacetate) and dried for 5 min (Kallweit et al, 1996). Mass spectra were recorded on a Mass Phoresis instrument (Ciphergen Inc.). 2.15 Localization of Man26A in C.fimi cultures One hour before analyzing samples from a 6 day old C. fimi culture, grown on minimal medium plus 0.2 % LBG, chloramphenicol was added (40 ug/mL). Mannanase activity in the culture, supernatant, and in cells (washed 3 times with 50 m M phosphate buffer p H 7.0) was tested on azo-carob galactomannan (McCleary, 1978). Reactions with 75 uL azo-carob galactomannan (2 %), 5 uL azide(10 %), 6 uL chloramphenicol (20 mg/mL) and 210 uL sample were incubated at 37° C for 16 h. Reactions were stopped by adding 750 uL EtOH (95 %) and the precipitated mannan was removed by centifugation for 2 min at 13 k rpm. The release of EtOH-soluble RBB labeled mannooligosaccharides in the supernatant was determined from A 5 9 0 . Datapoints were collected in triplicates. 28 2.16 Affinity gel electrophoresis (AFGE) and Western blotting The binding of proteins to soluble substrates was analyzed by affinity gel electrophoresis (Takeo et al, 1972; Takeo, 1984). 7.5 % separating polyacrylamide gels (Mini-Protean II system (Bio-Rad); 0.75 mm spacers) were supplemented with soluble galactomannan, either locust bean gum or azo-carob galactomannan, at final concentrations from 0 to 1 %. Protein samples in native loading buffer (20 % glycerol, 0.1 % Bromopheol blue, 125 m M Tr i s /HCl p H 8.8) were loaded on the gels. Five ug/lane were used for detection by Coomassie staining and 70 ng/lane for Western blotting. Samples were electrophoresed at 150 V for 2 h at 4° C. Protein bands were visualized either by staining with Coomassie blue, by zymograms (for azo-carob galactomannan gels only) or by Western blotting (Towbin et al, 1979). For Western blotting, the protein bands were transferred electrophoretically to PVDF membranes (Millipore) and probed with mouse anti-His6 antibodies (Dianova; 1/250 dilution) using anti-mouse IgG-horse radish peroxidase conjugate (Dako Diagnostics, Canada) as secondary antibody (1/5000 dilution). ECL™ detection reagents were used as recommended by the supplier (Amersham). The relative mobility, which is the ratio of migration distance of the protein to that of the reference protein, was measured at various substrate concentrations and the values used to calculate dissociation constants according to equation E 4.1 (Section 4.2.3). Reference 29 proteins were either Man2A or acetylated BSA (New England Biolabs). C B D c e n D was used in Western blots to mark the top of the separating gels. Binding to monomeric sugars was tested by competition affinity gel electrophoresis (Takeo, 1984) through 1.25 x l O 2 % LBG affinity gels (vide supra) containing either mannose or galactose (1.8 %). Gels were electrophoresed as described and proteins with a His6 tag were detected by Western blotting (vide supra). The concentration of LBG was calculated by measuring total sugar concentration using the phenol-sulfuric acid method (Chaplin, 1986) and reducing sugar concentration with the the p-hydroxybenzoic acid hydrazide method (Lever, 1973). Both concentrations were calculated as D-galactose and D-mannose equivalents assuming the presence of 5 mannose residues for each galactose residue, because these sugars differ in their responses in the assay (Lever, 1977). 2.17 Genomic Mapping 2.17.1 Preparation of genomic D N A The protocol for preparation of C. fimi genomic D N A was adapted from Birren and Lai (1993). C. fimi cultures were grown in LB lowsalt (0.5 g NaCl per liter) supplemented with 0.5 % glucose and 50 ug kanamycin mL-1 to mid log phase (A6oo= 0.8 to 1.5). Two hours before harvesting, 3 (ig Penicillin G mL^was added. Cells were harvested by centrifugation (5 k rpm, 10 min) and washed in Wal (10 m M Tris /HCl, p H 7.2, 100 m M EDTA, 200 m M NaCl). The washed cells were resuspended in Wal (1/20 of the culture volume) and warmed to 37° C. The suspension was mixed with an equal volume of molten 1.5 % low melting preparative grade agarose (Bio-Rad) at 40° C. The cell suspension was transferred to 30 1 cc syringes. After the agarose solidified, the tips of the syringes were cut off and little agarose-embedded cell blocks (2 to 5 mm) were cut with sterile razor blades. Twenty to thirty blocks were incubated in sterile LSI (10 m M Tri s /HCl p H 7.5, 50 m M NaCl, 100 m M EDTA, 0.2 % Na-Deoxycholate, 0.5 % NaSarcosyl, 3 mg lysozyme mL-1) for 16 h at 37° C with gentle agitation. The agarose blocks were transfered to 30 mL sterile DB1 (0.5 M EDTA, 1 % NaLaurosylsarkosyl, 0.5 mg/mL Proteinase K) and incubated for 48 h at 53° C. Blocks were extensively dialysed with restriction enzyme buffer prior to D N A digestions with restriction endonucleases. Each block was digested in 100 uL of the appropriate restriction buffer containing 20 to 50 U restriction endonuclease for 16 h at the temperature recommended by the supplier. 2.17.2 Pulsed field gel electrophoresis Genomic D N A blocks were embedded in 1.2 % agarose gels (high strength analytical ultra pure D N A grade agarose (Bio-Rad), in TAE buffer). D N A size standards were Saccharomyces cerevisiae chromosmes (NEB) and Hind III digested lambda D N A (BRL). Pulsed field gel electrophoresis was performed on a 2015 Pulsaphor apparatus (LKB; Pharmacia). The gel was submerged in sterile 1 x TAE buffer and equilibrated to 12° C. Switch intervals of 60 s for 9 h, 80 s for 9 h, 100 s for 9 h and 120 s for 9 h were applied with a constant voltage of 160 V. D N A bands were visualized under UV light after ethidium bromide staining (50 ug/mL in H 2 O ) . 2.17.3 Southern blotting of pulsed field gels 31 D N A bands from PFGE gels were transfered onto Hybond™ N+ membranes (Amersham) by capillary alkaline blotting (Sambrook et al, 1989). Genes were detected by using the Fluorescein Gene Images™ system (Amersham) according to the recommendation of the supplier. D N A probes were labeled either by the random primer method, or by PCR (Feinberg, et al, 1983). PCR conditions were as described in Section 2.5.1, whereby the deoxynucleotides were supplemented with fluorescein-labeled dUTP. The probes were: man26A, 1.2 kbp Sac ll fragment; man!A, 650 bp PCR fragment (Mad26, Mad29: PCR primers); cex, 511 bp Nrw I fragment; cenB, 811 bp Nru I fragment; cenD, 711 bp Pvu I fragment, cbhB: 750 Sst I fragment; xynD, PCR product 101; the cenA and cenC specific probes were the entire plasmids (Table 2.4). The hybridization temperature applied ranged from 63° C to 72° C and the wash stringency was 1 x SSC, 0.1 % (w/v) SDS and 0.5 x SSC, 0.1 % SDS (20 x SSC: 3 M NaCl, 0.3 M Na3citrate, p H 7.0), at the same temperatures as the hybridization. 2.18 Enzymology 2.18.1 p H and temperature optima for Man2A The p H optimum for Man2A was determined by measuring P N P M hydrolysis at A 4 0 0 . 0.3 pmol Man2A was incubated in 500 uL of 200 uM P N P M , 0.1 % BSA, 150 m M NaCl and 100 m M buffer. Potassium phosphate buffers were used at p H 5.7 to 8.1. After 6 min at RT the reactions were stopped and equilibrated by adding 1 mL of 1 M glycine (pH 10.9). Initial P N P M hydrolysis rates (AA^oo/rnin) were measured to determine the temperature optimum. 0.08 pmol Man2A was added to 1 mL reaction mixture (220 uM P N P M , 0.1 % BSA and 100 m M phosphate buffer p H 7.0). Hydrolysis rates were measured at 32 temperatures between 23° C and 60° C. Reaction mixtures were preincubated at the appropriate temperature for a period of 10 min prior to addition of enzyme. AA4oo/min was measured 3 min after the addition of enzyme. Thermal stability was determined by incubating Man2A at temperatures ranging from 23° C to 55° C. (0.2 nmol Man2A in 500 uL 50 m M phosphate buffer p H 7.0, 0.1 % BSA). The loss of activity over time was measured by removing samples and deterrnining the rate of hydrolysis of P N P M at 23°C (vide supra). 2.18.2 Steady state kinetic parameters for hydrolysis of P N P M by Man2A Steady state kinetic parameters for P N P M hydrolysis were determined by continuous measurement of the release of p-nitrophenol using a Hitachi U 2000 spectrophotometer with a temperature-controlled cell holder set at 37° C. Reaction mixtures (1 mL) contained 50 m M phosphate buffer (pH 7.0), 4.2 n M Man2A and 0.1 % BSA. Cuvettes and solutions were prewarmed at 37° C. The P N P M concentrations ranged from about 1/20 x K m to about 5 x K m . The release of p-nitrophenyl was measured at A 4 0 0 and the molarity calculated according to the Beer-Lambert law, using e = 7280 (M^x cm-1) (Kempton and Withers, 1992). Values for v m a x , K m and k c a t / K m were calculated from a Lineweaver Burk plot (Figure 3.24) for P N P M concentrations of 10 to 150 uM. A curve obeying the Michaelis-Menten equation was fitted to the initial slope using the computer program GraFit 3.0 (Leatherbarrow, 1992). 2.18.3 Kinetics of inactivation of Man2A The inactivation of Man2A by 2-deoxy-2-fluoro-mannosyl-B-fluoride (2FManBF) was monitored by incubating of enzyme (8.3 uM) with various concentrations (19.5 uM to 520 33 uM) of 2FManpF and measuring the release of fluoride using an Orion 96-09 combination fluoride electrode. The reaction mixtures (300 uL) were incubated at 37° C in 50 m M phosphate buffer p H 7.0 containing 0.1 % BSA. Pseudo-first order rate constants at each inactivator concentration (k0bs) were determined by fitting each inactivation curve to a first order rate equation with GraFit 3.0 (Leatherbarrow, 1992). Values for the inactivation rate constant (ki) and the dissociation constant for the inactivator (Ki) were determined by non-linear regression according to equation E 6.1 (Section (Mackenzie et al, 1997). 2.18.4 Reactivation of Man2A after inactivation with 2FMan£F Completely inactivated Man2A was freed from excess inactivator by exchanging the buffer in the sample with 50 m M phosphate buffer, p H 7.0, using Microsep™ Microconcentrators (10 kDa cutoff; Filtron) (Mackenzie et al, 1997). Potential reactivators were screened by incubating 21 pmol of the inactivated enzyme with 2 m M of reactivator in a reaction volume of 200 uL at 23° C. The regain of activity was assayed by removing samples and measuring P N P M hydrolysis at A 4 0 0 (45 uL reactivation mix, 500 uM P N P M , 50 m M phosphate buffer p H 7.0, 0.1 % BSA; 1 mL reactions). The kinetics of reactivation of inactivated Man2A were determined with gentiobiose (glucosyl-B-l,6-glucoside), by mcubating 8.3 uM of the inactivated enzyme with 10 mM, 15 m M , 40 m M and 80 m M gentiobiose in 50 m M phosphate buffer, p H 7.0, 0.1 % BSA in a 500 uL reaction volume at 37° C. Reactivation was monitored by removing samples (7 uL) at appropriate time intervals and assaying the regain of activity with 2 m M P N P M , 50 m M phosphate buffer p H 7.0, 0.1 % BSA, in 600 uL. The observed reactivation rate constant, kQbs, for each gentiobiose concentration was determined from the slope of the plots of ln (full rate 34 minus observed rate) versus time. The rate constant (ktrans) and dissociation constant ( K t r a n s ) for the reactivation were determined from a reciprocal plot of k 0bs versus reactivator concentration (Mackenzie et al, 1997). 2.18.5 p H and temperature optima for Man26A Hydrolysis of P N P M 2 by Man26A was measured at p H 4.7 to 8.0, by incubating 0.4 nmol Man26A with 1 m M P N P M 2 , 150 m M NaCl, 0.1 % BSA and 50 m M buffer in a total volume of 100 uL for 5 h at 32° C. Citrate buffer was used from p H 4.7 to p H 6.3 and phosphate buffer from p H 5.7 to p H 8.0. The reactions were stopped and the p H adjusted by addition of 500 uL I M glycine p H 10.9. A 4 0 0 was recorded. P N P M 2 hydrolysis was analyzed at temperatures between 23° C and 50° C, by incubating 0.4 nmol Man26A with 1 m M P N P M 2 , 150 m M NaCl, 0.1 % BSA and 50 m M citrate buffer p H 5.5 in a total volume of 100 u,L for 5 h. A 4 0 0 was measured after the addition of 500 uL 1 M glycine p H 10.9. A l l determinations were done in duplicate. 2.18.6 Steady state kinetic parameters for the hydrolysis of LBG by Man26A Steady state kinetic parameters for the hydrolysis of LBG by Man26A were determined by monitoring the release of reducing sugars, as D-mannose equivalents, using the p-hydroxybenzoic acid hydrazide method (Lever, 1973). Reaction mixtures contained 22 n M Man26A, 0.01 to 4.5 mg LBG mL-1, 50 m M citrate buffer p H 5.5 and 0.1 % BSA in a total reaction volume of 10 mL. The reactions were incubated at 37° C for 40 min. The 35 concentration of reducing ends on undigested substrate was subtracted from the concentration of reducing ends measured after hydrolysis to yield the enzymatic activity. 2.18.7 Hydrolysis of mannan and mannooligosaccharides by Man26A and Man2A Mannooligosaccharides (1 mM), locust bean gum (0.1 %) and ivory nut mannan (0.1 %) were digested with 0.5 nmol Man2A and/or 0.5 nmol Man26A for 1 h at 37 ° C in a 100 uL reaction volume. The reaction was terminated by separating the sugars from the enzyme with Microsep™ Microconcentrators (3 kDa cut off; Filtron). The mono- and oligosaccharides released by Man26A and Man2A were analyzed by fluorophore assisted carbohydrate electrophoresis (FACE®) as recommended by the supplier (Glyko). The standards were a-1,4 glucooligosaccharides and (3-1,4 mannooligosaccharides. The O-linked oligosaccharide profiling gels were run at 20 mA (constant current) for approximately 60 min and the temperature was controled at 20° C. Fluorescent carbohydrate bands were visualized under U V light (320 nm) and documented with a digital camera (Kodak DC 40). 2.19 Hydrolysis of mannan by intact and proteolytically cleaved Man26A Man26A (390 ug) was incubated in 100 uL phosphate buffer p H 7.0 with either 1.4 U C. fimi protease or without protease for 2 h at RT. A control reaction was set up with protease alone. Six uL samples from these reactions were transferred to 10 mL of 0.2 % locust bean gum or 0.2 % ivory nut mannan in 50 m M citrate, p H 5.5 (0.2 nmol Man26A/10 mL). After incubation for 30 h at 37° C (with agitation) the release of reducing sugars was determined (Section 2.18.6). The concentrations of reducing ends in the control samples 36 were subtracted from the concentrations in the test samples, to yield the enzymatic release of reducing ends. .2.20 Proteolysis of Man2A Man2A was inactivated with 2FManBF as described above (Section 2.18.3). Then 1.2 mg each of Man2A and of inactivated Man2A were treated with 120 ug pepsin at RT in 50 m M phosphate p H 2.0 in 1 mL. Digestions were stopped by freezing (-70° C). The products were analyzed immediately after thawing by liquid chromatography/mass spectrometry (LC/MS) (Mackenzie et al, 1997). 2.21 Electrospray mass spectrometry (ESMS) Mass spectra were recorded on a PE-Sciex API 300 triple quadrupole mass spectrometer (Sciex, Thornhill, Ontario, Canada) equipped with an Ionspray ion source. Errors in the reported masses are 0.1 %. Peptides were separated by reverse phase HPLC on an Ultrafast Microprotein Analyzer (Micchrom BioResources Inc., Pleasanton, CA) directly interfaced with the mass spectrometer. In each of the MS experiments, the proteolytic digest was loaded onto a C18 column (Reliasil, 1 x 150 mm) equilibrated with solvent A (0.05% rrifluoroacetic acid (TFA), 2 % acetonitrile in water). Elution of the peptides was accomplished by using a gradient (0 to 60%) of solvent B (0.045 % TFA, 80 % acetonitrile in water) for 60 min followed by 100 % solvent B for 2 min. Solvents were pumped at a constant flow rate of 50 uL/ min. Spectra were obtained in the single quadrupole scan mode (LC/MS) or in the tandem MS product ion scan mode (MS/MS). 37 In the single quadrupole mode (LC/MS), the quadrupole mass analyzer was scanned over a mass to charge ratio (m/z) range of 300 to 2200 amu with a step size of 0.5 amu and a dwell time of 1.5 ms per step. The ion source voltage (ISV) was set at 5.5 kV and the orifice energy (OR) was 45 V. The labeled and the corresponding unlabeled peptides were identified by comparative mapping, i.e., by comparing the mass spectra of unlabeled versus labeled petide mix. In the tandem MS daughter ion scan mode, the spectrum was obtained by selectively introducing the parent ion (m/z = 1520 for the labeled and m/z = 1036 for the unlabeled sample) from the first quadrupole (ql) into the collision cell (Q2) and observing the product ions in the third quadrupole (Q3). Thus, Q l was locked at either 1036 or 1520; Q3 scan range was 100-1040 and 100-1530, respectively; the step size was 0.5; dwell time was 1 ms; ion source voltage (ISV) was 5 kV; orifice energy was 45 V; collision gas = N2. (Mackenzie et al, 1997) 2.22 Thin layer chromatography (TLC) Thin layer chromatogaphy was performed on 0.2 mm silica gel aluminum plates (#60 F254; Merck). One to two uL of each reaction mixture or standard was applied to the base of the TLC plate. The plate was either air dried for 10 nun or by gentle warming with a hair dryer. The solvent was ethyl actetate : methanol : water (7:2:1 v/v). After air drying, 38 products were detected either under UV and/or by charring; i.e., heating the TLC plates until bands were visible after dipping them in 10 % H2SO4 in methanol. 2.23 Transglycosylation product analysis N M R spectra were recorded on a 200 Hz Bruker AC-200 and are referenced to the solvent peak. Where required, the interpretations were supported by COSY experiments. Mass spectra were obtained by ion-spray mass spectrometry on a PE-Sciex API III triple quadrupole mass spectrometer (Sciex, Thornhill, Ont, Canada). Reactions were monitored by TLC. Separation of the mixtures of regioisomers (B-1,3 and (3-1,4 linked mannobiose and mannotriose) was achieved using both flash chromatography and HPLC. Flash chromatography was performed on short columns (10 to 15 cm) of Merck Kieselgel 60 (230 to 400 mesh). HPLC was performed on a Waters HPLC using a Dynamax column and 70:30 acetonitrile-water as the eluant (flow rate 7.0 mL/min). 39 3 Characterization of the Cellulomonas fimi Mannan-degrading System 3.1 Introduction 3.1.1 Mannan Mannan is a polymer composed of a [3-1,4 linked mannose backbone. It is found in nature in a variety of forms with either backbone substitutions and/or side branches. In soft-woods, mannan is the major hemicellulose component, accounting for up to 25% of the wood dry-weight, whereas in hard-woods heteroxylans are predominant. Soft-wood mannan is present as an acetylated galactoglucomannan (O-acetyl-galactoglucomannan), in which the P-1,4 linked mannose backbone is interspersed with p-1,4 linked glucopyranosides, with a mannose to glucose ratio of 3:1 (Timell 1967; Puis et al, 1993). Galactose monomers are a-1,6 linked side branches linked only to mannose residues (Tenkanen et al, 1997), whereas the acetyl groups are esterified to either the 2 or 3 hydroxyls of the backbone mannoses, and to a much lower extent to glucose (Katz et al, 1964). There are two different forms of galactoglucomannan: water soluble and alkali soluble. They differ in their composition; the water soluble form has a Man:Glu:Gal:Ac ratio of 3:1:1:0.24 whereas the alkali soluble form has a ratio of 3:1:0.1:0.24 (Timell, 1967). The distribution of glucose substitutions is not necessarily uniform: there are runs of glucose residues in pinewood O-acetyl galactoglucomannan, that may be important in the proposed intimate interaction of hemicellulose and cellulose (Tenkanen et al, 1997, Atalla, 1993). In certain green algae mannan replaces cellulose as the crystalline structural component of the wall (Yui et al, 1997). Mannan-based polysaccharides also function as storage carbohydrates in bulbs and the endosperm of carob seeds, ivory nuts beans (Meier, 40 1958; McCleary et al, 1985). Ivory nut mannan, from Phytelephas macrocarpa, and locust bean gum (or carob-) galactomannan, from Ceratonia siliqua, are two commonly used substrates for the study of mannan degrading enzymes (Stalbrand et al, 1993 and 1995; Bolam et al, 1996; McCleary et al, 1983). Ivory nut mannan is an insoluble, crystalline material with linear chains of (3-1,4 linked mannosyl residues (Hendrixson et al, 1993). Its crystal structure strongly resembles that of cellulose, with two mannose residues repeating every 10.3 A (Nieduszynski et al, 1972). Locust bean gum is a heteropolymer, with a (3-1,4 linked mannosyl backbone and a-1,6 linked galactose side chains (Figure 3.1). The ratio of galactose to mannose was reported as 1:5, with the galactosyl groups distributed irregularly along the B-1,4 mannosyl backbone (McCleary et al, 1985). Locust bean gum is an industrially important polysaccharide, being used as a stabilizer and a thickener in the food industry (Rol, 1973). Figure 3.1: Schematic structure of galactomannan. x:y ratio in locust bean gum is 1:4 (McCleary etal, 1983). 41 3.1.2 Mannan hydrolyzing enzymes Hydrolysis of acetylgalactoglucomannan into its monomeric components requires the following enzyme activities: endo-B-mannanase, fi-mannosidase, a-galactosidase, B-glucosidase and acetyl esterase (Puis and Schuseil, 1993). The study of the mannan-degradfng system from C. fimi was simplified by studying the degradation of galactomannan (locust bean gum) and mannan (ivory nut mannan), hydrolysis of which requires only endo-B-mannanase, B-mannosidase and, in the case of galactomannan, oc-galactosidase. B-mannanase B-mannanases (mannan endo-l,4-beta-mannosidase; EC are produced by plants, fungi and bacteria. They catalyse the random hydrolysis of B-1,4 mannosidic linkages within the backbone of substrates such as mannan, galactomannan and glucomannan. Among the best characterized S-mannanases are the Aspergillus niger mannanase (McCleary et al, 1983), Manl from Trichoderma reesei (Stalbrand et al, 1995; Harjunpaa et al, 1995) and ManA from Pseudomonas fluorescens ssp. cellulosa (Bolam et al, 1996). The sequences of several B-mannanases are known (Braithwaite et al, 1995; Stalbrand et al, 1995; Morris et al, 1995; Arcand et al, 1993; Millward-Sadler et al, 1996). Based on amino acid sequence similarities, B-mannanases can be grouped into two different glycosyl hydrolase families: family 5 and family 26, both members of the superfamily GH-A (Bolam et al, 1996). Mechanistic studies of the family 5 enzyme Manl from T. reesei and family 26 Pj ManA have shown that both enzymes cleave their substrate with net retention of the anomeric configuration (Harjunpaa et al, 1995; Bolam et al, 1996). Secreted B-mannanases 42 have a non-modular architecture e.g. Pf ManA (Braithwaite et al, 1995), or a modular architecture e.g. T. reesei Manl (Stahlbrand et al, 1995). B-mannosidase B-mannosidases (B-l,4-mannoside mannohydrolase EC catalyze the hydrolysis from the non-reducing end of terminal B-D-mannose residues. B-mannosidases from animals, plants, fungi and bacteria have been characterized, but very few B -mannosidase-encoding genes have been sequenced (McCleary et al, 1983; Bauer et al, 1996; Alkhayat et al, 1998). Lysosomal B-mannosidase is involved in degradation of the N-linked oligosaccharide moieties of glycoproteins. A deficiency of lysosomal B-mannosidase in animals and humans is responsible for the neurological disorder B-mannosidosis, which is caused by the accumulation of Man B-1,4 GlcNAc and Man B-1,4 GlcNAc B-1,4 GlcNAc di-and trisaccharide, respectively (Alkhayat et al, 1998; Chen et al, 1995). The genes encoding human (Alkhayat et al, 1998), goat (Leipprandt et al, 1996) and bovine (Chen et al, 1995) B -mannosidase have been sequenced and characterized. Based on their sequence similarities, the eukaryotic B-mannosidases are classified into glycosyl hydrolase family 2, a family with mainly B-galactosidase and B-glucuronidase members (Durand et al, 1997). The only bacterial B-mannosidase sequenced is the intracellular B-mannosidase from the archaeon Pyrococcus furiosus, which is a member of glycosyl hydrolase family 1. Several fungal B-mannosidases and their roles in mannan degradation have been studied. The B-mannosidases from Aspergillus niger, Aspergillus awamori and Trichoderma reesei are secreted into the culture medium and play a role in mannan degradation by releasing mannose preferentially from shorter mannooligosaccharides (Elbein et al, 1977; 43 Neustroev et al, 1991; Stalbrand personal communication). Furthermore, B-mannosidases from fungal and bacterial sources are useful tools for the analysis of glycoproteins (McCleary et al, 1983; New England Biolabs, 1998). a-galactosidases a-galactosidases (melibiase, or a-D-galactoside galactohydrolase, EC from plants, fungi and bacteria remove a-1,6 linked galactose groups from galactomannan polymers. The enzymes may be extracellular (Margolles-Clark et al, 1996; McCleary et al, 1983; Talbot et al, 1990) or intracellular (Xiuzhu et al, 1991; Gherardini et al, 1985). a-galactosidases are represented in glycosyl hydrolase families 27 and 36 (Henrissat et al, 1996). 3.1.3 Objectives It was the aim of this project to study the enzymes of Cellulomonas fimi responsible for degrading mannan into its monomers. The genes encoding a B-mannanase and a B-mannosidase were cloned, sequenced and expressed. The products released from mannan and galactomannan by hydrolysis with the recombinant enzymes were identified. 44 3.2 Results 3.2.1 Identification of galactomannan degrading enzymes produced by Cellulomonas fimi At least one endo- 1,4- B-mannanase, one 1,4- (3-mannosidase and one 1,6- a-galactosidase are required to degrade galactomannan into galactose and mannose (Section 3.1.2). To screen for these enzymatic activities, C. fimi cultures were grown for six days on minimal medium, suplemented with 0.2 % (w/v) locust bean gum (LBG) as carbon source. The growth rates of LBG cultures were comparable to those of glucose supplemented cultures, indicating that C. fimi can degrade LBG and use the degradation products as energy and carbon source. Supernatant and cell-free extract from the LBG culture were assayed for activity by the LBG-Congo red plate assay (Section 2.6.1). Most of the mannanase activity was present in the supernatant, with much lower but detectable activity in the cell-free extract. Culture supernatant was further analyzed by azo-carob galactomannan zymograms (Section 2.6.2). At least three different bands with endo-l,4-B-mannanase activity were detected (Section 3.2.2). B-mannosidase and a-galactosidase activities were detected only in the cell-free extract. 4-methylumbelliferyl-B-mannoside (MUpMan) and 4-methylumbelliferyl-a-galactoside (MUaGal) were used as substrates. MUpMan and MUaGal zymograms of cell-free extract separated by non-reducing PAGE (Section 2.6.2) revealed only one band of B-mannosidase and one band of a-galactosidase activity. From the electrophoretic mobilities, the M rs of the a-galactosidase and the B-mannosidase were estimated to be 120 kDa and 60 kDa, respectively. The B-mannosidase comigrated with the C. fimi B-glucosidase (Kim and Pack, 1992; Gene bank accession number: M94865). 45 3.2.2 Carbon source-dependent mannanase synthesis There were at least three bands with B-mannanase activity (Mrs of 100 kDa, 75 kDa and 30 kDa) on azo-carob galactomannan zymograms of the supernatant of a culture of C. fimi grown with LBG. This suggested the presence of more than one B-mannanase encoding gene. Therefore, carbon source-dependent mannanase synthesis was studied. C. fimi was inoculated into basal salt medium supplemented with either LBG, mannose, glycerol, LBG plus glucose, birch wood xylan, CM-cellulose or galactose as carbon source (0.2 % w/v) or without carbon source. Final cell mass was similar in most cultures. Cultures without carbon source or with CM-cellulose as carbon source grew very slowly. Supernatants from all the cultures were analyzed on zymograms after 2, 6, and 11 days (Figure 3.2). Mannanase gene expression was induced only by LBG or CM-cellulose as carbon source. The LBG culture contained two major (Mrs 75 kDa and 30 kDa), and one minor (M r 100 kDa) protein with mannanase activity. In the presence of glucose and LBG the synthesis of mannanase activity was repressed. The CM-cellulose culture contained only one protein with B-mannanase activity (M r 30 kDa). Mannanase activity was not detected in cultures with mannose, galactose, xylan or glycerol as carbon source. These results supported the hypothesis that more than one B-mannananse encoding gene is present in C. fimi, and showed that the B-mannanase genes are differentially regulated (this issue wil l be further addressed in Section 3.7). B-mannosidase synthesis was induced by LBG and mannose but not by CM-cellulose. 46 A M 1 2 3 4 5 6 7 8 105 kDa 82 kDa 49 kDa 33 kDa 28 kDa 20 kDa Figure 3.2: Mannanase secretion by Cellulomonas fimi. Supernatant samples were analyzed by non-reducing SDS-PAGE-zymograms. C. fimi cultures were grown in basal salt medium (Lane 1) or in basal salt medium supplemented with 0.2 % (w/ v) locust bean gum (LBG) (Lane 2), mannose (Lane 3), glycerol (Lane 4), glucose and LBG (Lane 5), xylan (Lane 6), CM-cellulose (Lane 7) and galactose (Lane 8); reduced, prestained molecular size standards (Lane M). Panel A: samples from 2 day old cultures; Panel B: samples from 6 day old cultures and Panel C: samples from 11 day old cultures. Zymograms were incubated for 16 h at 37 °C in phosphate buffer. 47 3.2.3 Screening of the C. fimi genomic D N A library for B-mannanase genes The C. fimi genomic library was prepared by inserting genomic D N A fragments (2 to 5 kbp) into the EcoR I site of the multiple cloning site of lambda ZAPII (prepared by Stratagene). This created translational fusions of the genomic inserts with the first 36 aa of the E. coli (lacZ) B-galactosidase coding sequence transcribed from the lacZ promoter (Stratagene; Meinke et al, 1993). Therefore, the C. fimi lambda ZapII library could be screened for IPTG-inducible B-mannanase activity on azo-carob galactomannan plates (Section 2.7). Eight plaques with mannanase activity were isolated, their phagemids (pBluescript SK+C. fimi D N A insert) excised, and transferred into E. coli XLOLR cells. A l l eight mannanase clones, CManl-8, secreted mannanase activity into the culture supernatant. In the supernatant of each clone at least two bands with mannanase activity could be detected by zymograms. The only clone that produced higher molecular weight bands with significantly more mannanase activity than the other clones was CMan2. From the other seven clones, slightly more mannanase activity could be detected in the supernatant of CMan4. CMan2 and CMan4 were therefore chosen for further analysis. Restriction maps of their plasmids were established (Figure 3.3). pCMan2 had a 4.3 kbp and pCMan4 a 6.3 kbp insert. The similarity of the restriction pattern from pCMan2 and pCMan4 suggested that their inserts had a D N A fragment in common (Figure 3.3 A). By D N A sequencing it was shown that these two plasmids carried inserts with identical 5' ends except for 65 extra bp in pCMan4 (Figure 3.4). This suggested that the 6.3 kbp long insert of pCMan4 contained D N A encoding the same mannanase as pCMan2. The insert in pCMan4, however, was 65 bp longer at the 5' end and 2 kbp longer at the 3'end than 48 c Z L pCMan2: pBluescript II SK C. fimi genomic D N A insert Figure 3.3: Restriction mapping of pCMan2 and pCMan4. Panel A: Restriction digests of pCMan4 (Lane 1) and pCMan2 (Lane 2) separated on a 0.9 % agarose gel. Restriction endo-nucleases used are indicated. X Hind. Ill D N A was used as size standard. Panel B: Plasmid map of pCMan2. The restriction map of the 4.3 kbp C. fimi genomic D N A insert of pCMan2 is shown. Not all the fragments could be mapped. 49 a, x> AH CO t 03 u O H X-0 0 < U U u u < < + 1 CO a u o o u u < o H u O H cN OH X> co O CO " + O H AH o CO o CO fi a u O H OH X> rH - + OH Xi O -+ LO CO CO o CO + •4-1 03 i -2 03 O CU CU S O O H H o cu bo c bo PS l-H 03 03 v 4-1 CU fi o o < Z Q CU -O — tu - M C •J-J o u C H 3J s-^ H 03 co •S s jg 03 CO u3 •H 03 3 & fi o O H o S H 0 C CU bo cu u • H I c fi c to •1 03 03 5 0 g CO ** 03 -f i S o u * Oc ^ JT > fi 03 "» TS CN fi 1 ^ H=H CU U <2 P H . S bo to cu * fi X e.J3 V X. 43 a U « 5 « -5 £ cu CO CU >H O H CU c c c •43 03 42 fi CU CU CO g a « p> 03 P H SH CU " * H ^ bo CU fi 03 ^ (3 03 CU -O O H fi v CU -3 cu a 1 ™ 5 l-H (3 O H +H CU 4_| £ j CO CU CO f H -i a, 1-5 o co o ufiu C O < H CO cu T3 J2 os O CU c 9 O H o v cu X co H cu cu 03 o o b cu « •a cu fi -fi o H CU JO bo £ CU o £ K E-i os 50 pCMan2 (Figure 3.4). Hence, pCMan2 was chosen for D N A sequencing. The subclones and oligodeoxyribonucleotide primers used for sequencing are listed in section 2.2 and 2.3. A putative start codon was found in neither pCMan2 nor in pCMan4. Because the mannanase activities detected in the supernatants of all the other clones, C M a n l , 3, and 5 to 8, were very similar to CMan4 (vide supra) the library was rescreened for more mannanase encoding clones. The extracellular mannanase activities of five clones, CManlO, 20, 30, 40 and 50 were analyzed on zymograms. CMan30 secreted two active polypeptides that were very similar in size to those secreted by C. fimi. The plasmid insert of CMan30, was 3 kbp in size. Comparisons of restriction digests of the plasmids pCMan30, pCMan2 and pCMan4 showed that all three plasmids had a D N A region in common. A n ATG start codon was found in pCMan30, 540 bp downstream from the 5' end of the insert. This start codon was found to be only 6bp and 72 bp upstream from the 5' ends of the inserts in pCMan 4 and pCMan2, respectively (Figure 3.4). 3.2.4 Nucleotide- and deduced amino acid sequence of the C. fimi mannanase The nucleotide sequence of its gene and the deduced amino acid sequence of the C. fimi mannanase are shown in Figure 3.5. The open reading frame was 3033 bp long, which translated into a 1011 amino acid long protein with a calculated M W of 107,033. The N -terminus of the C. fimi mannanase had an amino acid composition rich in positively charged residues, a characteristic of secretion signal peptides (Nielsen et al, 1997). To determine the N-terminus of the secreted and processed C. fimi mannanase, concentrated C. fimi LBG culture supernatant was separated and analyzed by non-reducing PAGE-51 GCCCGACCTGTGGCTGCGCTGTGCGGCGCCAGGGCCCGGCTGACCTCGGGCGACGCCGGTGAAACGCTTCGCGTGCG AGCCGGGCCCTCGGTAGCGCCTGTCCGACCTGTTCCGTCCTGACTAAGTTGGTAACTTGACGAATCTGTCCGCGTCC GTCCGGAGCCTCCGGTCGGCGCGGCCCCTGGAGGGCTCGGGCGCCGCTGCCCGGCCCGACCGTCACCTGGAGGTTCA 1/1 ' 31/11 ATG ACG AAC CGC AGC CGT CCG CGC GGG CGC ACG GCG GGC CAC GTG CTC GCC GCC ACC GCC M T N R S R P R G R T A G H V L A A T A 61/21 91/31 GCG GCG CTG GCC CTG ACC GGC CTG TCG GCG CTT CCC GCC CAG TCC GCA CCG GCG CCC GCA A A L A L T G L S A L P A Q S A P A P A * 121/41 151/51 GCG CCC GTC GCG GGG GCG CTG CCC ACC GCA CCC GCC GAC GAG ACC ATC GCG ATC GTC GAC A P V A G A L P T 4 A P A D E T I A I V D 181/61 211/71 GCC GAC GCG ACC GCC GAG ACC CGG TCG CTG CTC TCC TAC CTC GAC GGC GTG CGC GGC GAG A D A T A E T R S L ' L S Y L D G V R G E 241 /81 271/91 GGC ATC CTG TTC GGC CAC CAG CAC ACG ACG TCG TTC GGG CTC ACC ACC GGA CCC ACC GAC G I L F G H Q H T T S F G L T T G P T D 301/101 331/111 GGC ACG ACC TCC GAC GTC AAG AAC GTC ACG GGC GAC TTT CCC GCG GTC TTC GGC TGG GAC G T - T S D V K N V T G D F P A V F G W D 361/121 391/131 ACG CTC ATC ATC GAG GGC AAC GAG CGC CCC GGG CTC GCC GAG AAC ACG CGC GAC GAG AAC T L I I E G N E R P G L A E N T R D E N 421/141 451/151 ATC GCG CTG TTC GCC GAC TAC ATC CGC AAG GCC GAC GCG ATC GGC GGC GTC AAC ACC GTG I A L F A D Y I R K A D A I G G V N T V 481/161 511/171 AGC GCG CAC GTC GAG AAC TTC GTC ACC GGC GGC TCG TTC TAC GAC ACC TCG GGC GAC ACG S A H V E N F V T G G S F Y D T S G D T 541/181 571/191 CTG CGC GCC GTG CTG CCG GGC GGC TCG CAC CAC GCC GAG CTC GTC GCC TAC CTC GAC GAC L R A V L P G G S H H A E L V A Y L D D 601/201 631/211 ATC GCG GAG CTC GCC GAC GCG TCG CGC CGC GAC GAC GGC ACG CTC ATC CCG ATC GTC TTC I A E L A D A S R R D D G T L I P I V F 661/221 691/231 CGG CCG TGG CAC GAG AAC GCC GGC TCG TGG TTC TGG TGG GGC GCC GCG TAC GGC TCA CCC R P W H E N A G S W F W W G A A Y G S P 721 /241 751/251 GGC GAG TAC CAG GAG CTC TAC CGG TTC ACC GTG GAG TAC CTG CGC GAC GTC AAG GGC GTC G E Y Q E L Y R F T V E Y L R D V K G V 781/261 811/271 TCC AAC TTC CTC TAC GCG TGG GGT CCG GGC GGC GGC TTC GGC GGC AAC CGC GAC GTC TAC S N F L Y A W G P G G G F G G N R D V Y 841/281 871/291 CTG CGC ACC TAC CCC GGC GAC GCG TTC GTC GAC GTG CTC GGC CTC GAC ACC TAC GAC AGC L R T Y P G D A F V D V L G L D T Y D S 901/301 931/311 ACC GGT TCG G A C . G C G TTC CTC GCC GGG CTC GTC GCC GAC CTG CGG ATG ATC GCC GAG ATC T G S D A F L A G L V A D L R M I A E I 961/321 991/331 GCC GAC GAG AAG GGC AAG GTG TCG GCG TTC ACC GAG TTC GGC GTG AGC GGC GGC GTG GGC A D E K G K V S A F T E F G V S G G V G 1021/341 1051/351 ACG AAC GGC TCG TCG CCC GCG CAG TGG TTC ACC AAG GTG CTC GCC GCG ATC AAG GCC GAC T N G S S P A Q W F T K V L A A I K A D 1081/361 1111/371 CCC GTC GCG AGC CGC AAC GCC TAC ATG GAG ACG TGG GCC AAC TTC GAC GCC GGC CAG CAC P V A S R N A Y M E T W A N F D A G Q H 1141/381 1171/391 TTC GTC CCC GTG CCC GGC GAC GCG CTG CTC GAG GAC TTC CAG GCG TAC GCC GCC GAC CCG F V P V P G D A L L E D F Q A Y A A D P 52 1201/401 TTC ACG CTG TTC GCG TCC GAG GTC ACG GGC F T L F A S E V T G 1261/421 GCG CAG CCG GTC GTG CAC ATC GCC TCT CCG A Q P V V H I A S P 1321/441 ACC ACC GTG CGG GTG CGG GTC GGC GGC ACC T T V R V R V G G T 1381/461 CAG GGC GGC ACC GTC GTC GAC ACT CTG GAC Q G G T V V D T L D 1441/481 GCC CCC TGG TCG CCG ACC AGC GCG CAG CTC A P W S P T S A Q L 1501/501 GCG ACG ACC GCC GCC GGG ACG CTC GAC GTC A T T A A G T L D V 1561/521 ACG TTC CCC GCG GGC GTC GTC GAC GAC TTC T V F P A G V V D D F 1621/541 GCC GAG TAC GTG ACC TAC GGC GCC AAC ACG A E Y V T Y G A N T 1681/561 GGC GCG AAG GCG CTG CGG CTC GAC TAC GAC G A K A L R L D Y D 1741/581 AAG CAG CTG TCC GGC GAC TGG TCC GAC TTC K Q L S G D W S D F 1801/601 GGC TCG AAC AAC AGG ATG GTC CTG CAG CTC G S N N R M V L Q L 1861/621 CCG TCG CTC GCG GGC GAC GAG CCG CAG CTC P S L A G D E P Q L 1921/641 GCA CCG TGG GAC ACC GCG CAC GCC GAC CGC A P W D T A H A D R 1981/661 ACG TCG TTC AAC GTC TAC GTC AAC AGC GCC T S F N V Y A V N S A 2041/681 GTC GAC GAC ATC GCC GCC CAC CCC GGC GTC V D D I A • A H P G V 2101/701 AAG GGC CAC CCC TAC GAG ACG GAG ATC CTG K G H P Y E T E I L 2161/721 TAC GAC GAC GGC ACC TTC CGG CCC GCC AGG Y D D G T F R P A R 2221/741 CTG CAC GCC TAC GAG AAG GCG GTC TTC ACG L H A Y E K A V F T 2281/761 CGC CGG AGC CAC CCC GCC TAC ACC GCG ATC R R S H P A Y T A I 2341/781 GAC GGG CGC GTC TTC CTG CCG AGC GCC CCG D G R V F L P S A P 2401/801 TGG CGG CTC GCC GGG TCC CCC GAG CCC GAG W R L A G S P E P E 2461/821 TGG CAC CGG TAC CGC ACC GCG ATC ACG TGG W H R Y R T A I T W 1231/411 GCG TTC GAC CGG ACC GTC GCC GCA GCG CCC A F D R T V A A A P 1291/431 CC GAC GGC GCG CGC GTC GCG TCC GCC CCG A D G A R V A S A P 1351/451 GAC GTG CAG TCC GTG ACC GTC GAG GTC GCC D V Q S V T V E V A 1411/471 TC GCG TAC GAC GGC GCC CTG TGG TGG ACG L A Y D G A L W W T 1471/491 MBD11 GAC AAC AGC ACC TAC ACC GTC ACC GCG AGG D N S T Y T V T A T 1531/511 ACG AAC GAG GTC GTC CTC GGG CCG AGG CCG T N E V V L G P R P 1591/531 GAG GGC TAC GGC GAC GAC ACC GCG CTG CGT E G Y G D D T A L R 1651/551 ATC TCG CTC GAG ACG GGG TCC GTC GGG GGC I S L E T G S V G G 1711/571 TTC GCG ACG CAG ACC TAC ACC GGC GTC GGC F A T Q T Y T G V G 1771/591 AAC GAG CTC GCG ATC TGG GTC GAC CCC GAC N E L A I W V D P D 1831/611 AAC GCC GGT GGT GTC GCC TAC GAG GCG TAC N A G G V A Y E A Y 1891/631 GTG ACG ATC CCG TTC GTC GAC TGG CGC CCG V T I P F V D W R P 1951/651 CGC ATG TCC GAC GAC GAC CTG CGC GCG CTC R M S D D D L R A L 2011/671 ^ MBD12 GAG GGC GGT GCC GCG TCG GGC TCG CTC GTC E G G A A S G S L V 2071/691 GAG CCG CCG CCG CTG TTC TCC GAC GTC CCG E P P P L F S D V P 2131/711 TGG CTG CAC GCG CAG GGG CTC GAC GAC GGC W L H A Q G L D D G 2191/731 CAG GTC AAG CGG CAG GAC GTC GCG CGG CTG Q V K R Q D V A R L 2251/751 CCG CCG ACG ACA CCG TCG TTC CTC GAC GTC P P T T P S F L D V 2311/771 GAG TGG CTC GTC GCC GAG GGG CTC GTG GAC E W L V A E G L V D 2371/791 CTC GAC CGG GCC ACC GCC GCC GAG CTG CTG L D R A T A A E L L 2431/811 GGG ACG GAG GCG TTC CGC GAC GTC CCC ACG G T E A F R D V P T 2491/831 GCG ACC GAG GTG GGC GTC GTC GAG CCC GTG A T E V G V V E P V 53 2521/841 TCG GCG TCG ACG TTC GGG GTG CTC AAG S A S T F G V L K 2581/861 TAC CGG TTC GAC GCC CTC CCG TCG CCG Y R F D A L P S P 2641/881 GGC GCG CAG GGC TGG GGG CCG CTC GAC G A Q G W G P L D 2701/901 ACG CTC ACG ATC GAG GCC GCC GCG CCC T L T I E A A A P 2761/921 GCC GAC TGG ACC GGG CGG ACG GCG CTC A D W T G R T A L 2821/941 ACC AAG GCG GCC CTG CAG GTC GGC TCG T K A A L Q V G S 2881/961 TGG GTG TCC GGG CCG CGC ACC GGA GAC W V S G P R T G D 2941/981 GCG GGG TGC GGC GCC CAG CTC GCC GAC A G C G A Q L A D 3001/1001 ACC CAC GTG ATC GAC GAC GTC GAG CTG T H V I D D V E L 2551/851 GCC GTG CAG CGG CAG GAG CTC GCG CGG TAC CTG A V Q R Q E L A R Y L 2611/871 CTC GAG CCC GTC GTG CTG TCG GAC TTC GCC GAC L E P V V L S D F A D 2671/891 GCC GGT CCG GGC TCC GCG ACC GCG TCC GGC GGC A G P G S A T A S G G 2731/911 GAC GGC GGG TGG TTC TCG TTC ACC CCG TCC GTC D G G W F S F T P S V 2791/931 GCC CTC GAC GTG GTC TCC ACC ACC GGG TTC GAC A L D V V S T T G F D 2851/951 ACC TGG CAG TGG TGC GAG ACC GCG CAG GCC GGG T W Q W C E T A Q A G 2911/971 GAC GCG CTC GTG CTC GAC CTC ACG ACG TTG TAC D A L V L D L T T L Y 2971/991 GTC AAG CGC GTG AAC CTC TAC CTC AAC GCA GGC V K R V N L Y L N A G 3031/1011 CGC TGA CGCCTGGACCGCACGCGCGGGGGCCGGTCACCC R * ACGTGGTGGCCGGCCCCCGCCACGCGAGGGCGTCGGCTCGCCCGGGGCCCGGGCGGTGCGCCCGCCCGGGGGGCGCT CGGGTCAGGCGGGGGTGAGGAGGCCGTCGGTGACCAGGCCGCGGACCGAGGGCAGGACGTCGGCCGCGACGGCGTCG Figure 3.5: Nucleotide sequence and deduced amino acid sequence of man26A. Putative promoter sequences, identified based on similarities to other C. fimi promoters are underlined. The putative ribosome binding site is shown in bold. A possible transcription termination hairpin is overlined. The PCR primers, MBD11 and MBD12, used for sub-cloning of mbdim are indicated by arrows. * : Predicted signal peptide processing site, • : experimentally identified N-terminus of Man26A, V: predicted C-terminus of Man26A catalytic domain, N-terminus of 29 kDa SLH domain fragment, •: N-terminus of 21 kDa SLH domain fragment (See text for details). 54 zymogram and blotted onto a PVDF membrane for N-terminal sequence analysis. The N -terminal sequence of the processed mannanase, corresponding to the 75 kDa active polypeptide (Section 3.2.2), was determined by Edman degradation as 50APADET55, with the starting methionine being position 1. The cleavage between Thr 49 and Ala 50 was not in agreement with the signal peptide cleavage site as predicted by computer analysis using the program SignalP (Nielsen et al, 1997). Cleavage was predicted to occur between Ala 40 and Ala 41 in the sequence 3 7 P A P A -0- APV43 (-0-: mdicating the cleavage site). This prediction was in agreement with the (-3, -1) rule (Nielsen et al, 1997) and with the consensus cleavage sequence A/VXA -Q- A ( X can be any amino acid) from secreted C. fimi glycanases. Secreted C. fimi protease preferentially cleaves C-terminal to threonines (Gilkes et al, 1989, Sandercock unpublished results), suggesting that the experimentally determined N-terminus APADET (cleaved after Thr 49) of the secreted mannanase might have resulted from signal peptidase processing followed by proteolysis by the secreted C. fimi protease (see also Section 3.2.6). The amino acid sequence of the C. fimi mannanase was compared to those of other proteins. The sequence of the N-terminal half of the C. fimi mannanase was similar to those of the catalytic domains of mannanases in glycosyl hydrolase family 26 (EC The highest identity was with mannanase ManA from Pseudomonas fluorescens ssp. cellulosa (Pf ManA) ( Braithwaite et al, 1995). The two proteins were 42 % identical over a sequence of 328 aa. The C. fimi mannanase was also similar to mannanases ManB from Bacillus subtilis (Mendoza et al, 1995), ManB from Caldocellulosiruytor saccharolyticus (Liithi et al, 1991) and ManA, ManB and ManC from Piromyces sp. (Millward-Sadler et al, 1996), all members of family 26 (Figure 3.6). 55 Color Key for Hlignnent Scores 40-50 >=200 M a n 2 6 A • i i r 0 250 500 750 1000 Pseudomonps fluoresce.™; ManA : 4Jffi/328aa Bacillus anthyjfls^m&^pxot. 24%/174aa 134681 KESfliffB Bacillus sp. ManA: 24%/264aa Bacillus sp. SprB : 31%/116aa Bacil %/264aa I0(><>():>! Bacillus sv. SorB : 31%/116aa IOWi055 Rhodothermus. maritimus Man: 27%/252aa Clostridium thermocellum AncA: 30%/130aa Rhodothermus maritimus ManA: 27%/252aa B . anthracis EA1: 31 %/108aa inicola endo-glucana, l7f>WI9 Pevotella ruminicol >glucanase: 24% /357aa B. licheniformis SLAP: 31%/108aa 1351072 42081 Caldicellulosiruptor saccharolyticum ManB: 26%/256aa Thermotoga maritima OmpA: 39%/56aa Pevotella n g g g ^ e b g ^ 2 7 1 a a Thermus aquaticus SLAP: 30%/112aa Pevotella ruminicola CMCase: 25%/271aa Dictyoglomus thermophilim^an: 25%/256aa 2582053 Pyro7i|||£s sp M a n A^^o /245aa Pyromyces sp ManC: 24%/245aa P y r o 7 ^ g g j j ^ ^ ^ 4 % /247aa Bacillus sp. ManB: 23%/267aa Clostridium thermocellu 121829 H: 31%/85aa Figure 3.6: Results from protein sequence similarity search with Man26A. Each bar re-presents an amino acid sequence which is similar to the sequence of Man26A. The position of each bar is aligned for the homology with Man26A. The alignment score is colour-coded and the identities and alignment lengths are indicated for each protein sequence. The numbers in the or beside the bars are the sequence IDs.(Blast search; Altschul et al, 1997). 56 Cf Man26A 1 • RPRGRTAGHVLAATAAALALTGLSALPAQSAPAPAA:' '. ALPTAPADETIAIVD Pf ManA 1 MKTITTARLPWAAQSFALGICLIALLGCNHAANKSS....ASRADVKPVTVKLVD consensus 1 TA AA AL L A A T VD Cf Man26A 61 ADATAETRSLLSYLDGVRGEGILFGHQHTTSFGLTTGPTDGTTSDVKNVTGDFPAVFGWD Pf ManA 52 SQATMETRSLFAFMQEQRRHSIMFGHQHETTQGLTITRTDGTQSDTFMAVGDFAAVYGWD consensus 61 AT ETRSL R I FGHQH T GLT TDGT SD N GDF AV GWD Cf Man26A 121 TLIIEGNERPGLAENT: . " A " ' : IRKADAIGGVNTVSAHVEN FVTGGS Pf ManA 112 TLSIVAPKAEGDIVAQ VKKAYARGG1ITVSSHFDN VWPVGT consensus 121 TL I G KA A GG TVS H N G Cf Man26A Pf ManA consensus 1 7 3 F Y D T S • D T L R A V L P G G S H H A E L V A Y L D D I A E L A D A S R R D D G T L I P I V F R P W H C I N A G S W F K 1 6 1 S W D Q T . P A W D S L P G G A Y N P V L N G Y L D Q V A E W A N N L K D E Q G R L I P V I F R L Y H E N T G S W F W 1 8 1 D L P G G L Y L D A E A G L I P F R H E N G S W F W Cf Man26A Pf ManA consensus 233 W G A A Y G S P G E Y Q E L Y R F T V E Y L R D V K G V S N F L Y A W G F G G G F G G N R D V Y L R T Y ; A F 220 W G D K Q S T P E Q Y K Q L F R Y S V E Y L R D V K G V R N F L Y A Y S F N N F W D V T E A N Y L E R V ; : E W 241 W G P Y L R V E Y L R D V K G V N F L Y A P Y L Y P G D V D V Cf Man26A Pf ManA consensus 293 L G L D T Y D S T G . S D A F L A G L V A D L R M I A E I A D E K G K V S A F T E F G V S G G V G T N G S S P A Q W F T 2 8 0 L G F D T Y G P V A : , N A D W F R N W A N A A L V A R M A E A R G K I P V I S E I G I R A P D I E A G L Y D N Q W Y R 3 01 L G D T Y V A A A G K E G G Q W Cf Man26A 352 KVLAAIKADPVASRNAYMETWANFDA . HFV . • . AYAADPFTLFASEV'.' Pf ManA 340 KLISGLKADPDAREIAFLLWRNARR. . EFL APMAPRFPIIGCLLT . . consensus 361 K KADP A A W N F A A F T Cf Man2 6A 412 FDRTVAAAPAQPWHIASPADGARVASAPTTVRVRV ' ' - T D V Q S V T V E V A Q G G T W D T L D L Pf ManA 3 84 ARRISTMAPWRTSRPFMPMNSQRSIATSSRSISVRP consensus 421 R AP A VR Figure 3.7: Alignment of the first 460 aa from Cellulomonas fimi mannanase, Cf Man26A with the entire sequence of the Pseudomonas fluorescens mannanase, Pf ManA. Identical residues are highlighted. The two family 26 proteins share 42 % identity. For Pf ManA, Glu212 and Glu320 were experimentally determined to be the catalytic residues acting as acid/base catalyst and nucleophile, respectively. In Man26A these residues are conserved and correspond to Glu225 and Glu332. The catalytic residues are underlined. Alignment was performed with Clustal W (Tompsonef. al, 1994). 57 The best studied enzyme in family 26 is Pf ManA. This enzyme, representative for all family 26 members, cleaves the substrate via a double displacement mechanism, with a net retention of the configuration at the anomeric center. The catalytic residues in Pf ManA were determined by site-directed mutation of conserved family 26 carboxylic residues and kinetic studies of these mutants. Glu212 was identified as the acid-base catalyst, and Glu320 as the catalytic nucleophile (Bolam et al, 1996). Both catalytic residues from Pf ManA are conserved in the C.fimi mannanase and correspond to Glu 225 as acid/base catalyst and to Glu 332 as the catalytic nucleophile (Figure 3.7). In accordance with the nomenclature proposed recently the C. fimi mannanase, the first family 26 enzyme from C. fimi to be described, was named Man26A (or Cf Man26A) (Henrissat et al, 1998). In Man26A, between residues 680 and 860, another region with homology to other proteins was found. A l l the proteins sharing identities with Man26A in this region are either S-layer proteins, or proteins with a S-layer-homology (SLH) domain, e.g. Bacillus anthracis S-layer protein (24 % identical residues over a sequence of 174 aa), Bacillus sp SprB (31 % identity over 116 aa), Clostridium tltermocellum ORF3p, also reported as A N C A (30 % identity over 130 aa), and the endoglucanase from Clostridium josui (23 % identity over 119 aa), to name just a few (Figure 3.6). Typically a SLH domain is composed of three repeated aa sequences (Olabarria et al, 1996). The C.fimi Man26A SLH domain homology region was also composed of three repeats, each of which was about 60 aa long. Twenty percent of the amino acids were conserved in all three repeats (Figure 3.8). Figure 3.8: Alignment of SLH domain repeats (1-3) from Cf Man26A. 696, 757 and 815 are indicate the position of the first amino acid residues of each repeat in the Man26A sequence. The residues that are identical in two or all three repeats are highlighted. 1. 6 9 6 F S D V P K G H P Y E T E I L W L H A Q G L D D G Y D D G T F R P A R Q V K R Q D V A R L L H A Y E K A V F T P P T T P 2 . 7 5 7 F L D V R R S H P A Y T A I E W L V A E G L V D D G R V F L P S A P L D R A T A A E L L W R L A G S P E P E G T E A 3. 9 1 5 F R D V P T W H R Y R T A I T W A T E V G V V E P V S A S T F G V L K A V Q R Q E L A R Y L Y R F D A L P S P L E P V V 58 SLH domains are generally involved in anchoring S-layer proteins to the bacterial cell wall. They also occur in other secreted proteins, such as xylanases, pullulanases and cellulosome anchoring proteins (Ries et al, 1997; Lemaire et al, 1993; Fujino et al, 1993a). Cf Man26A is the first mannanase reported to have a SLH domain (Section 3.2.7). The sequences in Man26A between aa 470 to 680 and 940 to 1011 did not show any significant homologies to other protein sequences. 3.2.5 Sub-cloning of the gene and production of C.fimi mannanase, Man26A A flow sheet illuslrating the sub-cloning of man26A into the expression vector pET27b is shown in Figure 3.9. The E.coli strain BL21(DE3) was the host strain used for production of Man26A fused to a C-terminal His6 tag. Approximately 10 mg of purified protein were obtained from 1 liter of culture supernatant after purification by M C A C (Section 2.8). This suggested that the leader peptide from Man26A was recognized by E.coli cells as a secretion signal. However, this yield was not very good. Better yields were obtained by purifying the enzyme from a cell-free extract: 60 mg of purified and properly folded enzyme were obtained per liter of culture (Section 2.8 and Figure 3.10). 3.2.6 Analysis of the modular organization of C. fimi Man26A The results obtained from aa sequence analysis (Section 3.2.4) suggested the presence of at least two domains. The modular architecture of Man26A was further analyzed. Since folded domains are protected against proteolytic attack by proteases, such as the secreted C. fimi protease, whereas the interdomain regions are very susceptible (Gilkes et al, 1989), 59 Figure 3.9 (following page): Generation of pET27Man26A, encoding the C. fimi P-mannanase, Man26A with a C-terminal His6tag. A Nde I restriction site was introduced at the 5' end and a Nhe I restriction site was introduced at the 3' end of man26A. Only the restriciton sites that were relevant for the cloning steps are shown. • : Man51 (Nde I), Man35, Man41, Man50 (Nhe I) PCR primers : man26A or man26A derived plasmid insert : PCR product : man26A PCR template : D N A fragments used for the sub cloning steps. The D N A fragments were obtained by restriction endonuclease digestions, separation on agarose gels and purification gel extraction. 60 Man51 Nder i 900 bp Man35 Man41 1050 bp Man50 ATG man26A TGA ligate PCR product into pZErO EcoR V site ligate PCR product into pZErO EcoR V site 617 bp Hind III ligate 617 bp PpuM 1/ Hind III fragment into pCMan2 ligate 426 bp Sma l/Not I fragment into pCMan50 ligate 3033 bp Nde I/Nhe I fragment into pET27b T7 promoter pET27Man26A lad (Kan) 61 Man26A produced by E. coli was subjected to proteolysis by the C. fimi protease to probe its modular architecture (Section 2.13; Figure 3.10). The preparation of purified Man26A contained degradation products with M rs of 104, 70, 50, 28 and 21 kDa. Treatment of Man26A with C. fimi protease produced similar fragments (Figure 3.10). The first proteolytic event produced the polypeptide with a M r of 104 kDa, corresponding probably to the loss of the N-terminal 50 aa (Section 3.2.4). This polypetide was further digested into a polypeptide with an estimated M r of 53 kDa (via a 70 kDa intermediate) and into a polypeptide with an estimated M r of 28 kDa. These two polypeptides were further cleaved to produce fragments with M rs of 50 kDa and 21 kDa, respectively, which resisted further proteolysis, even after prolonged (48 h) incubation with protease. Analysis of the two final fragments by mass spectrometry (MALDI-TOFMS) gave masses of 50.4 and 21.2 kDa. N-terminal amino acid sequencing showed that the peptide of M r 50 kDa corresponded to the catalytic domain, the peptide of M r 28 kDa corresponded to the SLH domain, and the peptide of M r 21 kDa arose by cleavage at the N - and C-termini of the 28 kDa peptide (Figure 3.11). The fragmentation pattern was consistent with the hypothesis that Cf Man26A is a modular enzyme, composed of two structural units that resist proteolysis; the catalytic domain (CD 26) and the SLH domain, which are both flanked by exposed protease susceptible sequences. 62 M l 1 2 3 4 5 6 7 8 9 10 11 12 M2 103 kDa 81 kDa 46.9 kDa 34.1 kDa 28.5 kDa 20.2 kDa—> , 116 kDa 97 kDa <— 66 kDa «— 45 kDa <— 31 kDa <— 21.5 kDa <— 14.5 kDa Figure 3.10: SDS-PAGE (12 %) of protease-digested Man26A. Recombinant Cf Man26A was digested with C.fimi protease at 37 °C (1 U protease/60 u-g Man26A). Samples were removed at different times (Lane 1-12) 1: O min, 2:15 min, 3: 30 min, 4: 45 min, 5: 60 min, 6: 90 min, 7: 2 h, 8: 3 h, 9: 5 h, 10: 7 h, 11:16 h, 12: 24 h. Digestion was stopped by boiling samples in SDS loading buffer for 2 min. Molecular weight standards are shown in Lane M l and M2. The M,s are indicated. The N-termini of peptides corresponding to bands labeled with an asterisk $r) were sequenced. 63 CO Id o If* CO p PQ 3J £ •3 c id u c o li C O CM li ft, < < G * • •B-B s ® > , • .SP-o CJ ' S a J w « O ai i C x g a, ^ «s i 0 s IS <-o S 2 S H H g O ^ 8 a -a cu eq .s < TO ui « 5| * I I I 5 o ^ C ^ a TO &, •» g cu y i Si | TS TO VZH & * g g c 'SL -v 9 2 * 7 <2 5b cu a; TO rS Go M g TO cn cu n§ •£ "TO ^ ^ « S 8 0 0) TO 1 1 51 C ai .219 2 u o B 2 -° u H CO CC C 01 S o I • CJ 09 CO DO i : S s 13 £ OJ t. T3 o 2 ' o 1 £ 9 g bos g a 64 3.2.7 Localization of mannanase activity The presence of a SLH domain in Man26A suggested that Man26A might be cell-bound. A C. fimi culture was grown with LBG as carbon source. After six days protein synthesis was inhibited by adding chloramphenicol to the culture before removing samples. Supernatant, washed cells and total culture samples were analyzed for mannanase activity by measuring azo-carob galactomannan hydrolysis (Figure 3.12; Section 2.15). More mannanase activity was found to be cell-bound, than in the supernatant. The highest activity was detected in the total culture samples. These results agreed with the idea that the SLH-domain in Man26A attaches the mannanase to the cell surface. This attachment might be only transient, with subsequent release of some Man26A into the culture supernatant (Lemaire et al, 1995). The activity in the supernatant might also be a consequence of proteolytic cleavage between the catalytic domain and the SLH domain, with release of only the catalytic domain into the culture supernatant (Section 3.2.6). To test in vitro binding of Man26A to C. fimi cell walls, Man26A was added to a peptidoglycan fraction, which was prepared from C. fimi cells as described by Lemaire et al. (1995). However, binding of Man26A to this cell wall preparation could not be detected. 65 120 100 supernatant washed cells culture Figure 3.12: Analysis of mannanase activity from a 6 day old C. fimi culture. Supernatant, washed cells and total culture were analyzed on azo-carob galactomannan. The release of EtOH soluble oligosaccharides was measured at A590nm after incubation at 37° C for 16 h. The activity from total culture was 100 %. 3.2.8 Is Man26A the only mannanase produced by C. fimi ? Since recombinant Man26A was very unstable and susceptible to proteolysis (Figure 3.10), it was possible that all the polypeptides with mannanase activity detected in culture supernatants arose by proteolysis of Man26A. Zymograms of C. fimi culture supernatants and of Man26A proteolytic digests were compared. Supernatants from C. fimi cultures grown with ivory nut mannan (INM), LBG, CM-cellulose, and a mixture of CM-cellulose and LBG were compared to the patterns produced by recombinant Man26A, either untreated, treated with purified C. fimi protease, or treated with supernatant from the C. fimi culture with LBG (Figure 3.13). It was possible to obtain bands of mannanase activity similar to those found in C. fimi cultures, by treating recombinant Man26A with protease. It 66 M 7 103 kDa + 81 kDa-* 46.9 kDa-* 34.1 kDa-* 28.5 kDa-* 20.2 kDa-* Figure 3.13: Non-reducing SDS-PAGE-zymogram to test the origin of multiple mannanase bands in C.fimi culture supernatants. Supernatants from four different C. fimi cultures were tested. The cultures were grown for 11 days at 30 °C in rrunimal medium supplemented with either ivory nut mannan (Lane 1), locust bean gum (Lane 2), carboxymethyl cellulose (Lane 3) or locust bean gum and carboxymethyl cellulose (Lane 4). Lane 5-7 recombinant Man26A (O.OSug), untreated (Lane 5), treated with C.fimi protease(Lane 6), or treated with culture supernatant from C. fimi culture grown on locust bean gum (Lane7). The protease digestions were incubated for lhr at 37 °C. To detect activity the zymogram was incubated in phosphate buffer at 37°C, before staining with Coomassie. 67 appeared, therefore, that Man26A was the only mannanase secreted by C. fimi involved in mannan degradation. The different polypeptides with mannanase activity present in supernatants from CM-cellulose and LBG-grown C. fimi cultures were not encoded by two or more independently regulated genes (Section 3.2); Rather they probably reflected different levels of secreted protease in the different cultures. It is believed that the secretion of protease by C. fimi and the protease susceptibility of Man26A has biological significance (Section 5) 3.2.9 pH- and temperature optima, and kinetic parameters for Man26A The pH- and temperature optima were determined for Man26A with p-nitrophenyl-mannobioside (PNPM2) as substrate. This compound was synthesized by Man2A E519A catalyzed transglycosylation (Section 6.2.4). Optimal hydrolysis of PNPM2 by Man26A occured at p H 5.5 and 42° C (Figure 3.14). Due to very slow hydrolysis rates, the reactions had to be incubated for more than 5 h to detect activity. Therefore these results also suggested that Man26A was stable for at least 2 hrs at 42°C (Section 2.18.5). The kinetics of LBG hydrolysis by Man26A were analyzed for concentrations ranging from 0.01 mg/mL to 4.5 mg/mL. No higher concentations were used in order to avoid viscosity problems. The apparent value of k c a t / K m = 1151 mL/ (mg.rnin) was calculated from the initial slope of the plot of v versus s (Figure 3.15). 3.2.10 Screening of the C. fimi genomic D N A library for (3-mannosidase The C. fimi lambda-ZapII library was screened for expression of (3-mannosidase activity using MUBMan as substrate. Out of 9 x 105 plaques screened, 2 plaques produced (3-68 0.6 0.5 0.4 o o <? 0.3 0.2 0.1 0 + 4 2 ° C 32 ° C • 3 7 ° C 23 ° C 20 25 30 35 40 45 Temperature (°C) 52 ° C * 50 55 Figure 3.14: Panel A : p H optimum of Man26A. Panel B: Temperature optimum of Man26A. The hydrolysis of 1 m M PNPM2 with 0.36 nmol Man26A was determined measuring A 4 0 0 . 100 uL reactions were incubated for 5 h and stopped by the addition of 500 uL 1 M glycine, p H 10.9. 69 0 1 2 3 4 5 [LBG] (mg/mL) Figure 3.15: Steady-state kinetics for hydrolysis of locust bean gum (LBG) by Man26A. Substrate concentrations from 0.01 mg/mL to 4.5 mg/mL LBG were hydrolyzed with 22 n M Man26A in citrate buffer p H 5.5 at 37° C for 40 min. v is the rate of reducing sugars released per minute. The apparent k c a t /K m = 1151.8 mL/mg.min was calculated from the initial slope of the plot v versus [LBG]. To avoid viscosity effects, only concentrations < 4.5 mg/mL were tested. 70 mannosidase activity. These 2 plaques however, did not produce any positive plaques in the secondary screening procedure, possibly a consequence of exposure to methylumbelHferyl. A n alternative approach was used, whereby the C. fimi library was rescreened in its excised form, i.e. as E. coli colonies containing C. fimi genomic D N A fragments inserted into the pBluescript phagemids. The colonies were replica plated on plates containing M U B M to screen for activity. Two clones with B-mannosidase activity were isolated, CMadI and CMadll . The sizes of the genomic D N A inserts were found to be 2.6 kbp and 6.5 kbp for the plasmids pCMadl and pCMadll, respectively. Restriction mapping revealed that both inserts had a D N A region in common, which was confirmed by sequencing the 5' ends of the inserts. The 5' end of the pCMadl insert was found only 150 bp downstream from the begirrning of the pCMadll insert. The shorter insert of pCMadl was sequenced. The subclones and primers used for sequencing are listed in Section 2.2 and 2.4. pCMadl contained the entire C. fimi B-mannosidase open reading frame. 3.2.11 Nucleotide sequence of the C. fimi B-mannosidase gene and its deduced amino acid sequence The nucleotide sequence of its gene and the deduced amino acid sequence of the intracellular C. fimi B-mannosidase are shown in Figure 3.16. The open reading frame is 2526 bp long and encodes a protein of 842 amino acids. The calculated molecular weight of this protein is 94,960, with a calculated pi of 5.1. Similarities of the C. fimi B-mannosidase to other proteins were analyzed (Figure 3.17). It shared significant sequence identity with mammalian B-mannosidases (EC, e.g. 32 % identical residues in a 667 aa long alignment with the caprine B-mannosidase (Figure 3.17; Figure 3.18) (Leipprandt et al, 1996). High alignment scores were also obtained with 71 ACCGGTCAGTGTGCCACTCGATTGACTTACTTTATTCAGTTGTCTAAGAATGATCCCGGGTACCTGGCAGCCGACTCGG CACCTCCCCGGAGGCTCCGAGCCGCCCGCGACGACGCGGGACGGACGCGCCGGACCCGCGGACCCA6GAGCGCGCGTCG 1/1 ATG ATC ACC CAG GAC CTC TAC GAC GGC TGG M I T Q D L Y D G W 61/21 GCC GAG CTG GCC GGC GTG CGC GTG CGG GCC A E L A G V R V R A 121/41 CTC GAC GAG GGC CTG ATC CCC GAC CCG TAC L D E G L I P D P Y 181/61 ATG CGG CGC ACC GAC TGG GCC TAC GAG CGC M R R T D W A Y E R 241/81 GAG CGC GTC GAC CTC GTC TTC GGC GGC ATC E R V D L V F G G I 301/101 CAC GAG CTC GGC CGC ACC GCC AAC CAG CAC H E L G R T A N Q H 361/121 CTG CGC CCC GAC ACC CAG CGC CTG CGC GTG L R P D T Q R C R V 421/141 GCG GAG CGC GAG CGC CTC GGC CAC CGC CCG A E R E R L G H R P 481/161 CGC AAG ATG GCG TGC TCG TTC GGC TGG GAC R K M A C S F G W D 541/181 TGG AAG CCG GTC CGC GTC GAG CGG TGG CGC W K P V R V E R W R 601/201 GTC ACG GTC GAC CCC GAC GGC ACC GGC CGG V T V D P D G T G R 661/221 GGC CTG CCC GGT GGC GAC GCG CCC GTC ACG G L . P G G D A . P V T 721/241 GCC GTC ACC GTG CCC GGC ACG GCC ACG TCG A V T V P G T A T S 781/261 CCC CTG TGG TGG CCG GTC GGC CAC GGC CCG P L W W P V G H G P 841/281 GCG ACG CGC GAC GAC GAG CCG CTC GAC TCC A T R D D E P L D S 901/301 GAG GTC GAC ACG ACG CCC GAC GAG GAC GGG E V D T T P D E D G 31/11 ACC CTC ACG GCC GTC TCC GGA CCC GTC CCG T L T A V S G P V P 91/31 CGC GTC CCC GGG ACG AGC CAC ACC GCG CTG R V P G T S H T A L 151/51 CTC GAC CGC AAC GAG GAC GTC CTC GCG TGG L D R N E D V L A W 211/71 GAG CTC GTC CTC GAC CCG GCC GCC GCC GAC E L V L D P A A A D 271/91 GAC ACC GTC GGG ACC GTC ACG TTC GAC GGC D T V G T V T F D G 331/111 CGG TCC TAC CGG TTC GAC GTG CGC GCC CTG R S Y R F D V R A L 391/131 GAC CTG CGG GCC GCG ATC GTG CAC GCC GAG D L R A A I V H A E 451/151 CTG GCC TAC CCG CAG CCG TTC AAC ATG GTC L A Y P Q P F N M V 511/171 TGG GGC CCC GAC CTG CAG ACC GCG GGC CTG W G P D L Q T A G L 571/191 ACC GCG CGG CTC GCG TCG GTG CGC ACG CAC T A R L A S V R T H 631/211 GTG CGC GTG CTC GTG GAC CTC GAG CGC TCG V R V L V D L E R S 691/231 CTG CGC GCC CGC GTC CTG TCG GCC GAC GTC L R A R V L S A D V 751/251 GCC GTC GTC GAG CTC GAG GTC CCC CGG GCA A V V E L E V P R A 811/271 CAG CCG CTG TCC GAC CTC ACC GTC ACG CTC Q P L S D L T V T L 871/291 TGG TCC CGG CGC ATC GGG TTC CGC ACG GTC W S R R I G F R T V 931/311 ACG CCG TTC ACG TTC CGC GTC AAC GGG CGG T P F T F R V N G R 72 961/321 CCG GTC TTC GTC AAG GGC GCC AAC TGG P V F V K G A N W 1021/341 CGC GAG CGC CTC GCG CAC CGG CTC GAC R E R L A H R L D 1081/361 GTC TGG GGC GGC GGC ATC TAC GAG TCC V W G G G I Y E S 1141/381 CTG CTC GTC TGG CAG GAC TTC CTC CTC L L V W Q D F L L 1201/401 TGG GAC GAG CTC GAG GCC GAG GCG CGC W D E L E A E A R 1261/421 CTC GTG CTG TGG AAC GGC GGC AAC GAG L V L W N G G N E 1321/441 CAG GAG CTC GAG GGC CGC ACG TGG GGC Q E L E G R T W G 1381/461 GTC GCG GAG CTC GAC CCG ACG CGC CCG V A E L D P T R P 1441/481 GCG CTC GAC GAC GTC CAC CCG AAC GAC A L D D V H P N D 1501/501 TGG AAC CGC GTC GAC TAC TCC GCG TAC W N R V D Y S A Y 1561/521 GGC TTC CAG GGC CCG CCG ACG TGG TCG G F Q G P P T W S 1621/541 GGC CCG CTG ACC AAG GAC GAC CCG ACG G P L T K D D P T 1681/561 GGC AAG CTC GAC CGC GGC CTC GCG CCG G K L D R G L A P 1741/581 CAC TGG GCG ACG CAG CTC AAC CAG GCC H W A T Q L N Q A 1801/601 TCG TGG TGG CCC CGG ACC GCG GGC GCG S W W P R T A G A 1861/621 ACG TCG TGG GCC GCG ATC GAC GGC GAC T S W A A I D G D 1921/641 CGC GCC TAC GCC CCG CGC CTG CTC ACC R A Y A P R L L T 1981/661 GTC GTC AAC GAC ACG GGT GGC CTC TGG V V N D T G G L W 991/331 ATC CCC GAC GAC CAC CTG CTC ACC CGC ATC ACG I P D D H L L T R I T 1051/351 CAG GCC GTC GAG GCG AAC CTC AAC CTG CTG CGC Q A V E A N L N L L R 1111/371 GAG GAC TTC TAC GAC CTG TGC GAC GAG CGC GGC E D F Y D L C D E R G 1171/391 GCG TGC GCG GCC TAC CCC GAG GAG CAG CCC ATC A C A A Y P E E Q P I 1231/411 GAG AAC GTC GCC CGG CTC ACG CCG CAC GCC TCG E N V A R L T P H A S 1291/431 AAC CTC TGG GGC TTC ATG GAC TGG GGC TGG CCG N L W G F M D W G W P 1351/451 TAC CGG CTG GCG ACC GAG CTG CTC AAG GGC GTC Y R L A T E L L K G V 1411/471 TAC GCC GAC GGC AGC CCG TAC TCC CCC GGC TTC Y A D G S P Y S P G F 1471/491 CCG GAC CAC GGC ACG CAC CAC GAG TGG GAG GTC P D H G T H H E W E V 1531/511 CGC GAC GAC GTG CCG CGG TTC TGC TCC GAG TTC R D D V P R F C S E F 1591/531 ACC CTC ACG CGT GCC GTC CGC GCC GAC GAC GGC T L T R A V R A D D G 1651/551 TTC CTG CTG CAC CAG AAG GCC GAG GAC GGC AAC F L L H Q K A E D G N 1711/571 CAC CTG GGC GTG CCC GCC GGC TTC GTC GAC TGG H L G V P A G F V D W 1771/591 CGC GCC GTC GCG TTC GCG ATC GAG CAC TAC CGG R A V A F A I E H Y R 1831/611 ATC GTG TGG CAG CTC AAC GAC TGC TGG CCG GTG I V W Q L N D C W P V 1891/631 GAG CGG GTC AAG CCG CTG TGG CAC GCC CTG CGC E R V K P L W H A L R 1951/651 GTG CAG CCG CGC GAC GGC CGC GAC GAG CTC GCG V Q P R D G R D E L A 2011/671 CAG GGC ACG GTC CGG CTC TCG CGG CGC ACG CTC Q G T V R L S R R T L 7 3 2041/681 2071/691 GAC GGT GCG ACG CTC GCC GAG GTC GAG CTC GGG CTG GCC GTC GGC GCG TGG TCG GTC GGC D G A T L A E V E L G L A V G A W S V G 2101/701 2131/711 CTG TTC GCG CTG CCC GAC GAG GTC GCC GCG CCC GAC GAC GCG GCC GGC GAG GTC CTG GTC L F A L P D E V A A P D D A A G E V L V 2161/721 2191/731 GTC GAC CTC GGG GAC GTC CGC ACG GTC CAC ACG TGG GCG CAG GAC GTC GAC CTG CGC CTC V D L G D V R T V H T W A Q D V D L R L 2221/741 2251/751 GAC CCC GAC CCG GTG TCC GCG ACC GTG TCG CCG CTG CAG GAC GGC TAC CGC GTC GAC GTG D P D P V S A T V S P L Q D G Y R V D V 2281/761 2311/771 ACC GCC CGG ACG TTC GCC CGC TCG GTC ACG CTG CAC GTG GAC CGC CTC GAC CCC GAC GCG T A R T F A R S V T L H V D R L D P D A 2341/781 2371/791 ACG GTC GAC GAC GCC CTC GTC GAC GTG CCG GCG GGC GAG ACG TTC TCG TTC CAC GTC CGC T V D D A L V D V P A G E T F S F H V R 2401/801 2431/811 ACC TCC GCA CGG TTC GAC GCC GCC GCC CTC ACG CGC AGC CCC GTG CTG CGC ACG GCG AAC T S A R F D A A A L T R S P V L R T A N 2461/821 2491/831 GAC GTG GTC GTG CCG CGC GGC GCG ACG GCA CCG GCG GGC GCC GAG CGT GAT CTC CAG CAG D V V V P R G A T A P A G A E R D L Q Q 2521/841 > < — TCC CGC TGA ACG GTT CGC GCC CGC CAC TAC GCT GGC GGG CGT GAC CCA CGG GAG GAA TTC S R * GAT ATC AAG CTT ATC GAT ACC GTC GAC CTC GAG GGG GCC CGT ACC AAT Figure 3.16: Nucleotide sequence of manlA and deduced amino acid sequence of the protein. The putative -35 and -10 promoter regions, identified based on similarities to other C.fimi promoters, are underlined. The putative ribosome binding site is shown in bold. The sequences marked with arrows highlight possible trancription termination sequences. 74 Color Key for Rlignnenb Scores 50-80 40-50 Man2A —| i i i—i—|—i—i—i—i—]—i—i—i—i—|—i—i—i—r—]—i—i—i—r '^F-i—i—i—|—i—r 0 100 200 300 400 500 600 700 800 Arabidopsis thalina ^g^gfljjgggj^zse: 40%/125aa aa taltus norvegicus beta-glucuronidase precursor: 24%/421aa Artkrobactersp. beta-galactosidase: 25%/363aa H^s^'mslreta-glucmortidase precursor: 24%/339aa Homo sapiensbej^glucuronidase^4%/339aa^ • H 114963 Mus musculus beta-glucuronidase: 25%/263aa Mus musculus beta-glucuronidase precursor: 25%/263aa Mus musculu^^^^^^^^^^^^^^^pr. 25%/263aa Mus musculus beta- glucuronidase precursor: 25%/263aa M u s musculus beta- glucuronidase precursor: 25%/263aa Cams /fl?ra7iar^jj^^^uronidase^^^j^6aa Bacillus megaterium beta-galactosidase; 25%/177aa 2764542 Escherichia coli evolved beta-galactosidase alpha subunit: 26%/172aa •HK.114995J Escherichia coli evolved beta-galactosidase alpha chain: 26%/172aa 67497 Escherichia coli phospho-beta^^glactosidase alpha subunit: 26%/172aa Lactobacillus delbruecki beta-galaaoSdase: 22%/452aa Lactobacillus delbruecki beta-galactosidase: 22%/452aa l8 Thermotoga maritima beta-galad idase^6 % / 149aa •173272 Figure 3.17: Results from protein sequence similarity search with Man2A. Each bar re-presents an amino acid sequence which is similar to the sequence of Man2A. The position of each bar is aligned for the homology with Man2A. The alignment score is colour-coded and the identities and alignment lengths are indicated for each protein sequence. The numbers in the or beside the bars are the sequence IDs.(Blast search; Altschul et al, 1997). 75 Goat Mad 1...MLLRLLLLLAPCGAGFATEWSISLRGNWKIHNGNGSLQLPAAVPGCVHSALFNKRI Bovine Mad 1. . .MLLRLLLLLAPCGAGFATKWSISLRGNWKIHSGNGSLQLPATVPGCVHSALFNKRI Human Mad 1...MRLHLLLLLALCGAGTTAAELSYSLRGNWSICNGNGSLELPGAVPGCVHSALFQQGL C.fimi Mad 1 . . .MITQDLYDGWTLTAVSGP.VPAELAG VRVRARVPGTSHTALLDEGL consensus 1 L L G VPG H AL Goat Mad 5 8 IKDPYYRFNNLDYRWIALDNWTYIKKFKLHSDMSEliMKVNLVFEGIDTVAWLLNSVPIG Bovine Mad 58 IKDPYYRFNNLDYRWIALDNWTYIKKFKLHSDMSTWSKVNLVFEGIDTVAWLLNSVPIG Human Mad 58 IQDSYYRFNDLNHRWVSLDNWTYSKEFKIPFEISKWQKVNLILEGVDTVSKILFNEVTIG C.fimi Mad 46 IPDPYLDRNEDVLAWMRRTDWAYERELVL.DPAAADERVDLVFGGIDTVGTVTFDGHELG consensus 6 1 I D Y N W WY V L G D T G Goat Mad 118 KTDNMFRRYSFDITHMVKA.VNIIEVRFQSPVIYANQRS...ERHTAYWVPPNCPPPVQD Bovine Mad 118 KTDNMFRRYSFDITHTVKA.VNIIEVRFQSPVVYANQRS...ERHTAYWVPPNCPPPVQD Human Mad 118 ETDNMFNRYSFDITNVVRD.VNSIELRFQSAVLYAAQQS...KAHTXYQVPPDCPPLVQK C.fimi Mad 105 RTANQHRSYRFDVRALLRPDTQRVRVDLRAAIVHAEAER ER.LGHRPLAYPQP.. consensus 121 T N Y FD A P P Goat Mad 174 GECHVNFIRKMQCSFGWDWGPSFPTQGIWKDVRIEAYNICHLNYFMFTPIYDNYMETVJNL Bovine Mad 174 GECHVNFIRKMQCSFGWDWGPSBTTQGIWKDVRI^ Human Mad 174 GECHVNFVRKEQCSFSWDWGPSFPTQGIWKDVRIEAYNICHLNYFTFSPIYDKSAQEWNL C.fimi Mad 157 ....FNMVRKMACSFGWDWGPDLQTAGLV?KPVRVERWRTARLASVRTHVTVD...PDGTG consensus 181 N RK CSF WDWGP T G WK VR E L D Goat Mad 234 KIESSFDWSSKLVSGEAIVAIPELNIQQRNNIELRHGE....R.TVKLFVKIDKAVIVE Bovine Mad 234 KIESSFDWSSKLVSGEAIVAIPELNIQQTNNIELQHGE....R.TVELFVKIDKAIIVE Human Mad 234 EIESTFDWSSKPVGGQVIXAIPKLQTQQTYSIELQPGK. . . . R. IVELFVNISKNITVE C . f imi Mad 210 RVRVLVDLERSGLPGGDAPVTLRARVLS . ADVAVTVPGT ATSAWELEVPRAP . consensus 241 D G G V Goat Mad 2 89 TWWPHGHGNQTGYDMTVTFEL.DGGLRFEKSAKVYFRTVELVEEPIQ.NSP..GLTFYFK Bovine Mad 289 TWWPHGHGNQTGYNMSVIFEL.DGGLRFEKSAKVYFRTVELVEEPIQ.NSP..GLSFYFK Human Mad 2 89 TWWPHGHGNQTGYNMTVLFEL.DGGLNIEKSAKVYFRTVELIEEPIK.GSP..GLSFYFK C.fimi Mad 2 62 LWWPVGHGPQPLSDLTVTLATRDDEPLDSWSRRIGFRTVEVDTTPDEDGTP FTFR consensus 3 01 WWP G G Q V D S FRTVE P P F F Goat Mad 3 45 INGLPIFLKGSNWIPADSFQ.DRVTSDMLRLLLQSWDANMNALRVWGGGIYEQDEFYEL Bovine Mad 345 INGLPIFLKGSNWIPADSFQ.DRVTSAMLRLLLQSWDANMNALRVWGGGVYEQDEFYEL Human Mad 345 INGFPIFLKGSNWIPADSFQ.DRVTSELLRLLLQSWDANMNTLRVWGGGIYEQDEFYEL C.fimi Mad 317 VNGRPVFVKGANWIPDDHLL.TRITRERLAHRLDQAVEANLNLLRVWGGGIYESEDFYDL consensus 3 61 N P F KG NWIP D R T L L V A N N LRVWGGG YE FYL Goat Mad 404CDELGIMIWQDFMFACALYPTDEDFMDSVREEVTHQVRRLKSHPSIITWSGNNENEAALM Bovine Mad 404CDELGIMIWQDFMFACALYPTDKDFMDSVREEVTHQVRRLKSHPSIITWSGNNENEAALM Human Mad 404CDELGIMVWQDFMFACALYPTDQGFLDSVTAEVAYQIKRLKSHPSIIIWSGNNENEEALM C.fimi Mad 3 7 6CDERGLLVWQDFLLACAAYPEEQPIWDELEAEARENVARLTPHASLVLWNGGNEN. LWGF consensus 421CDE G WQDF ACA YP D E RL H S W G NEN Goat Mad 464 MGWYDTKPGYLHTYIKDYVTLYVKNIRTIVLEGDQTRPFIISSPTNGAKTTAEGWLSPNP Bovine Mad 464 MGWYDTKPGYLQTYIKDYVTLYVKNIRTIVLEGDQTRPFITSSPTNGAKTIAEGWLSPNP Human Mad 464 MNIVYHISFTDRPIYIKDWTLYVKNIRELVLAGDKSRPFITSSPTNGAEWAEAWSQNP C.fimi Mad 43 5 MDWGWPQELEGRTWGYRLATELLK...GWAELDPTRPYADGSP..YSPGFALDDVHPND consensus 481 MW T K V D R P S P A 76 Goat Mad 524 YDLNYGDVHFYD. "fMSDCWNWRTFPKARFVSEYGYQSWPSFSTLEKVSSEED. WSYES . Bovine Mad 524 YDLNYGDVHFYD.YVSDCWNWRTFPKARFVSEYGYQSWPSFSTLEKVSSEED.WSYRS. Human Mad 52 4 NSNYFGDVHFYD.YISDCWNWKVFPKARFASEYGYQSWPSFSTLEKVSSTED.WSFNS. C.fimi Mad 490 PD. .HGTHHEWEVWNRVDYSAYRDDVPRFCSEFGFQGPPTWSTLTRAVRADDGGPLTKD consensus 541 G H RF SE G Q P STL D Goat Mad 580 . SFALHRQHLINGNSEMLQQIELHFKLPNSA DQLRRFKDTLYLTQV Bovine Mad 580 . SFALHRQHLINGNNEMLHQIELHFKLPNST DQLRRFKDTLYLTQV Human Mad 580 . KFSLHRQHHEGGNKQMLYQAGLHFKLPQST DPLRTFKDTIYLTQV C.fimi Mad 548 PTFLLHQKAE . DGNGKLDRGLAPHLGVPAGF VDW.HWATQL consensus 601 F LH GN H P TQ Goat Mad 625 MQAQCVKTETEFYRRSRNEIVDG.KGHTMGALYWQLNDIWQAPSWSSLEYGGKWKMLHYF Bovine Mad 625 MQAQCVKTETEFYRRSRSEIVNG.KGHTMGALYWQLNDIWQAPSWSSLEYGGKWKMLHYF Human Mad 625 MQAQCVKTETEFYRRSRSEIVDQ.QGHTMGALYWQLNDIWQAPSWASLEYGGKWKMLHYF C.fimi Mad 5 87 NQARAVAFAIEHYRSWWP RTAGAIVWQLNDCWPVTSWAAIDGDERVKPLWHA consensus 661 QA V E YR T GA WQLND W SW K L Goat Mad 684 ARRFFAPLLPVGFEDKD..VLFIYGVSDLPSDHQMMLTVRVHTWSSLELVCSELTNPFVM Bovine Mad 684 ARHFFAPLLPVGFEDKD..MLFIYGASHLHSDQQMMLTVRVHTWSSLELVCSESTNPFVI Human Mad 684 AQNFFAPLLPVGFENEN..TFYIYGVSDLHSDYSMTLSVRVHTWSSLEPVCSRVTERFVM C.fimi Mad 63 9 LRRAYAPRLLTVQPRDG..RDELAVVNDTG..GLWQGTVRLSRRTLDGATLAEVELGLAV consensus 72 P L VR Goat Mad 742 KAGESWLYSKPVPELLKGCPGCTRQSCWSFYLSTDGELLSPINYHFLSSLKNAKGLHK Bovine Mad 742 KAGESVLLYTKPVPELLKGCPGCTRQSCWSFYLSTDGELLSPINYHFLSSLKNAKGLHK Human Mad 742 KGGEAVCLYEEPVSELLRRCGNCTRESCWSFYLSADHELLSPTNYHFLSSPKEAVGLCK C.fimi Mad 695 GAWSVG. . .LFALPDEVAAPDDAAGEVLWDLGDVR TVHTWAQDVDLR.LDP consensus 781 W H L Goat Mad 802 ANITATISQQGNTFVFDLKTSAVAPFVWLDVGS.IPGRFSDNGFLMTEKTRTVFFYPWKP Bovine Mad 802 ANITATISQQGDTFVFDLKTSAVAPFVWLDVGS.IPGRFSDNGFLMTEKTRTVFFYPWKP Human Mad 802 AQITAIISQQGDIFVFDLETSAVAPFVWLDVGS.IPGRFSDNGFLMTEKTRTILFYPWEP C.fimi Mad 743 DPVSATVSPLQDGYRVDVTARTFARSVTLHVDRSTPTRRSTTPSSTCRRARRSRSTSAPP consensus 841 A S D A V L V P R S R P Goat Mad 861 TSKSSLEQSFHVTSLADTY. Bovine Mad 861 TSKSELEQSFHVTSLADTY. Human Mad 861 TSKNELEQSFHVTSLTDIY. C.fimi Mad 803 H consensus 901 Figure 3.18: Alignment of family 2 p-mannosidases. Goat Mad: Copra hircus p-mannosidase, Bovine Mad: Bos taurus P-mannosidase, Human Mad: Homo sapiens P-mannosidase, C.fimi Mad: Cellulomonas fimi p-mannosidase. Identities in all five protein sequences are highlighted and indicated in capital letters in the consensus sequence. The putative catalytic residues are underlined. Alignment was performed with Clustal W (Thompson et. al., 1994). 77 78 human and bovine B-mannosidases (Alkhayat et al, 1988; Chen et al, 1995). The C. fimi enzyme also shared sequence identity with B-glucuronidases and B-galactosidases, but the alignment scores were much lower (Figure 3.17). A l l of the enzymes sharing identity with the C. fimi B-mannosidase were members of glycosyl hydrolase family 2 (Henrissat 1991). Therefore, in accordance with the newly proposed nomenclature, the C. fimi B-mannosidase was named Man2A (or Cf Man2A) (Henrissat et al, 1998). A l l the family 2 members cleave sugars from the non-reducing end of the substrate with retention of the anomeric configuration (Gebler et al, 1992). Family 2 is a member of clan GH-A, which represents a group of more than 200 distantly related protein sequences from families 1, 2, 5, 10, 17, 26, 30, 35, 39 and 42. Analysis of known 3D structures of GH-A members revealed that their catalytic domains adopt similar (p/a)s folds, with the acid/base catalyst and the catalytic nucleophile at the C-terminal end of strands (34 and B7, respectively (Henrissat et al, 1996). The same folding motif, the (B/ a)s barrel, and the position of the catalytic residues were predicted by H C A analysis for the bovine B-mannosidase (Durand et al, 1997). The similarities between the H C A plots of the bovine and the C. fimi B-mannosidase were used to predict the positions of the acid/base catalyst as Glu429 and catalytic nucleophile as Glu519 in Man2A (Figure 3.19). For each glycosyl hydrolase family, signature consensus patterns were defined in order to reduce ambiguity in classification of new enzymes. The mannosidase sequences however, do not agree completely with the consensus patterns as defined for family 2 (Henrissat et al, 1996). One of the conserved regions in family 2 is around the general acid/base catalyst (Gebler et al, 1992). The second conserved region, that was defined as a signature pattern, is located about sixty residues upstream of the acid/base catalyst. These 79 two signature patterns are shown below for Cf Man2A, with the residues highlighted that do not agree with the family 2 signature sequences (26 %): consensus sequence 1: N L L R V W G G G I Y E S E D F Y D L C D E R G L L V W consensus sequence 2: L T P H A S L V L W N G G N E This suggested that the B-mannosidases might be a subfamily of family 2. 3. 2.12 Sub-cloning and expression of manlA The gene manlA was sub-cloned into the expression vector pET28a(+) (Novagen) (Figure 3.20). The resulting clone, pET28Mad, was used to produce recombinant Man2A with a C-terminal Fi6 tag in E. coli BL21(DE3) cells (Section 2.8). The protein was purified by M C A C . Up to 300 mg of the purified cytoplasmic protein were obtained with a purity estimated to be >96 % (Figure 3.21). The N-terminal amino acid sequence of the purified recombinant protein was identical to that of the partially purified C. fimi p-mannosidase (Section 2.9). Therefore, the recombinant, purified enzyme was used for further analyses. 3.2.13 p H optimum of C.fimi recombinant Man2A WT P N P M hydrolysis by Man2A was measured at pHs ranging from 5.5 to 8.2. The p H optimum was 7.0 (Section 2.18.1; Figure 3.22 A). 80 Figure 3.20 (following page): Generation of pET28aMad, encoding the C.fimi p-mannosidase, Man2A with a C-terminal His 6 tag. A Nco I restriction site was introduced at the 5' end and a Not I restriction site was introduced at the 3' end of manlA. Only the restriciton sites that were relevant for the sub-cloning steps are shown. • : Mad4 (Nco I) and Mad51 (Not I) PCR primers : manlA or manlA derived plasmid insert : PCR product : manlA PCR template : D N A fragments used for the sub-cloning steps. The D N A fragments were obtained by restriction endonuclease digestion, separation on agarose gels and purification by gel extraction. 81 Mad4 2529 bp Mad51 Nco I -t-A T G manlA I ligate PCR product into pZErO EcoR V site T G A 'Not I replace 1731 bp Pst I fragment in pOMad4/51 2529 bp Nco I Nco I Pst I ligate 2529 bp Nco I/Not I fragment into pET28a promoter pET28aMad lad (KanR) Not I ^His i tag Kan 82 M 116 kDa 97 kDa 66 kDa 45 kDa Figure 3.21: C.fimi p-mannosidase, Man2A WT and Man2A E519A, purified by M C A C , were separated on a 7.5% SDS-PAG (Phast System, Pharmacia) and stained with Coomassie. Lane M : molecular size standards, Lane 1:1 ug Man2A E519A, Lane 2:10 ug Man2A E519A, Lane 3:1 ug Man2A and Lane 4:10 ug Man2A. 83 3.2.14 Temperature optimum and thermostability of recomabinant Man2A WT P N P M hydrolysis was measured at temperatures ranging from 23° C to 65° C. The rates were calculated from the AA4oonm/ min at 190 s after the addition of enzyme. The fastest hydrolysis rate was obtained at 55 °C. This hydrolysis rate was therefore set as 100 %. The activities of Man2A WT at other temperatures were expressed as. values relative to the hydrolysis rate obtained at 55° C. The thermostability of Man2A WT was determined by incubating the enzyme at temperatures ranging from 23° C to 55° C (Section 2.18.5). During the 4 hour incubation samples were removed and P N P M hydrolysis rate was assayed at 23° C. A 3D representation of relative hydrolysis rates as a function of time and temperature is shown in Figure 3.22, with hydrolysis rates at 55° C and t = 190 s corresponding to 100 %. The optimum temperature was defined as the temperature at which the enzyme displayed highest hydrolysis rates and thermostability for at least 2 hours, which was 37° C for Man2A WT. The half-life of Man2A at 37° C was calculated to be 27.3 h. 3.2.15 Steady-state kinetic parameters for hydrolysis of P N P M by Man2A The enzyme (4.2 nM) was inhibited by substrate concentrations >400 uM; therefore values for the apparent k c at = 167 s4, K m = 0.3 m M and k c a t / K m = 501 s^.mM 4 were only estimates (Figure 3.24). Substrate inhibition can result if a second substrate molecule binds to the enzyme-substrate complex, ES, to produce an inactive complex, SES (Cornish-Bowden, 1979): 84 Figure 3.22: p H optimum of Man2A. The release of 4-nitrophenol was measured at A 4 0 0 after stopping the reactions with 1 M glycine p H 10.9. 85 Figure 3.23: 3D plot of Man2A temperature optimum versus temperature stability measured as the relative activity on PNPM. The temperature optimum was determined by P N P M hydrolysis at temperatures from 23 ° C to 60 °C. Temperature stability was monitored by mcubating Man2A for 4 h at temperatures from 23 ° C to 55 °C. Stability was examined by removing aliquots and measuring hydrolysis of P N P M at 23 °C. The activity of Man2A at 55 ° C was set as 100 %. 86 A 4 0 I 1 I 1 I 1 I 1 I 1 I 1 I 1 I 1 0.02 0.04 0.06 0.08 1 1/[PNPM] (1/pM) Figure 3.24: Steady-state kinetics of P N P M hydrolysis by Cf Man2A. Panel A : Kinetic data obtained from incubation of 4.2 n M Man2A with P N P M (10 uM to 1.5 mM). The curve shown was fitted to apparent K m = 0.3 m M and vmax = 42 mmol/min, which were calculated from the double reciprocal replot of hydrolysis initial rates with 10 uM to 150 uM P N P M (Panel B). 87 SES k + i E + S ^ ES 1 ^i> E + P Man2A WT did not hydrolyze p-rutrophenyl-a-mannoside, p-nitrophenyl-B-N-acetylglucoside, p-rutrophenyl-B-glucoside, p-nitrophenyl-(3-xyIoside, p-nitrophenyl-(3-cellobioside or p-rritrophenyl-B-gentiobioside. Low activity was detected on p-nitrophenyl -(3-galactoside. Man2A WT was also able to slowly hydrolyze Man-(3-l,4-GlucNAc into mannose and N-acetyl glucosamine, as was determined by FACE analysis (data not shown). 3.2.16 Size exclusion chromatography of Man2A Size exclusion chromatograpy was used to determine the mass of native Man2A. Ferritin (440 kDa), catalase (233 kDa), aldolase (158 kDa) and ovalbumin (50 kDa) were used as molecular mass standards. The log of M r was plotted against ve/vo (elution volume/void volume) (Figure 3.25). The M r of Man2A was calculated to be 1.03 x 104 Da, indicating that it was a monomer. 3.2.17 Mannan and mannooligosaccharide hydrolysis by Man26A and Man2A Mannan, galactomannan and five manno-oligosaccharides were subjected to hydrolysis by C. fimi (3-mannosidase, Man2A, B-mannanase, Man26A, or by both enzymes (Section 2.18.7). The released products were analyzed by fluorophore assisted carbohydrate electrophoresis (FACE) using glucose (a-1,4 linked) and mannose (B-1,4 linked) oligosaccharides as standards (Figures 3.26 and 3.27). Man26A cleaved randomly within the (3-1,4 linked backbone of galactomannan, releasing oligosaccharides of different sizes. This 88 Figure 3.23: Size exclusion chromatography of Man2A. The column was calibrated with ferritin (440 kDa), catalase (233 kDa), aldolase (158 kDa) and ovalbumin (50 kDa). The void volume (v0) was determined with Dextran Blue. Fractogel® TSK HW-55S was used as matrix. 89 cleavage pattern was characteristic of an endo-acting enzyme. Man2A, however, was unable to hydrolyze galactomannan into oligosaccharides, but released low quantities of mannose. Man26A in combination with Man2A produced a mix of oligosaccharides from galactomannan, very similar to the products released by Man26A alone, the only difference being the complete hydrolysis of the disaccharide (Figure 3.26 A). Unsubstituted mannan (ivory nut mannan) was hydrolyzed by Man26A into mainly mannotriose, mannobiose and mannose, whereas Man2A produced significant amounts of mannose as the only product. Both enzymes together released mannose as the only product (Figure 3.26 B). Mannose was the major product released from mannohexaose, -tetraose, -triose and -biose by Man2A (Figure 3.27), whereas Man26A hydrolyzed mannohexaose, -pentaose and -tetraose into mannobiose and mannose. Hydrolysis of mannotriose into mannobiose and mannose by Man26A was slow, and mannobiose was not hydrolyzed at all (Figure 3.27). 90 A G5 G4 G3 G2 G l B G5 G4 G3 G2 G l Figure 3.26: Analysis of products formed by hydrolysis of locust bean gum (Panel A) and ivory nut mannan (Panel B) with p-mannanase (Man26A) and P-mannosidase (Man2A). The products were analyzed by fluorophore assisted carbohydrate electro-phoresis (FACE) and compared to a-1,4- glucose oligosaccharide standards (Lane S). Lane 1: Man26A, Lane 2: Man2A, Lane 3: Man26A and Man2A. In a 100 u.L reaction volume, 0.1 % (w/v) substrate was digested with 0.5 nmol of either Man26A, Man2A or of both. Incubation time was 1 h at 37° C. 91 Figure 3.27: Analysis of products formed by hydrolysis of mannooligosaccharides with p-mannanase (Man26A) and P-mannosidase (Man2A). The products were analyzed by fluorophore assisted carbohydrate electrophoresis (FACE) and compared to cc-l,4-glucose oligosaccharide standards (Lane S) and P-1,4 mannooligosaccharide standards (Ml-M6)( Lane 1). Lane 2: mannotetraose Man2A digested, Lane 3: manno-hexaose Man2A digested. Lanes 4-9 are mannooligosaccharides digested with Man26A, Lane 4: mannotetraose, Lane 5: mannopentaose; Lane 6: mannohexaose; Lane 7: mannotiose; Lane 8: mannobiose; and Lane 9: mannose.Lanes 10-12 are Man2A digests of Lane 10: mannotriose, Lane 11: mannobiose and Lane 12: mannose. 1 m M oligo-saccharide solution was incubated with 0.5 n M enzyme at 37° C for 1 h (100 uL reaction volume) 92 3.3 Discussion 3.3.1 Molecular architecture of Man26A A modular architecture is common for p-mannanases. Additional domains such as cellulose-binding domains and cohesins may also be present (Stalbrand et ah, 1995; Millward-Sadler et al., 1996; Morris et al., 1995). Some p-mannanases comprise only catalytic domains (Arcand et al., 1993, Braithwaite et al., 1995). The C. fimi P-mannanase, Man26A is comprised of a family 26 catalytic domain, a mannan-binding domain (Chapter 4), a SLH domain and a C-terminal domain of unknown function. This study is the first report of a P-mannanase with a SLH domain and the first such domain described for any C. fimi protein. SLH domains occur in other enzymes, including xylanases, pullulanases, lichenases and for endo-glucanases (reviewed in Beveridge et ah, 1997). In growing C. thermocellum cells, a significant fraction of the cellulosomes are cell-bound. This interaction is mediated by surface proteins that bind to the cell envelope via SLH domains (Section 1.4). SLH domains from S-layer proteins from Bacillus stearothermophilus WT and PV76/p6 strains were shown to non-covalently bind to a secondary cell wall polymer composed of Glc-NAc and Man-NAc, but not to peptidoglycan (Egelseer et ah, 1998; Ries et ah, 1997). It is possible, based on the sequence diversity of SLH domains, that not all the SLH domains recognize the same cell wall components. SLH domains have also been shown to interact with other SLH domains (Lemaire et ah, 1995), which could be an indication that C. fimi could form small aggregation of enzyme complexes on the cell surface. The C-terminal region of Man26A might act as an additional anchoring segment by recognizing other cell wall components or perhaps other proteins, supporting the function of the SLH domain. 93 3.3.2 Localization of mannanase activity in C.fimi cultures Man26A appeared to be transiently cell-bound (Figure 3.12). The activity in the unfractionated culture, however, was lower than the combined activities of the supernatant and washed cells. This discrepancy could be explained by the presence of residual LBG in the culture which interfered with the azo-carob galactomannan activity assay, possibly due to a higher affinity of Man26A for LBG than for the azo-substrate. Man26A has a mannan-binding domain (MBD) (Chapter 4). The centrifugation of the bacterial culture in the preparation of the supernatant fraction removes some of the LBG, bound to the MBD of Man26A and hence to the cells, from the supernatant. Repeated washing of the cells carrying bound LBG could have removed some or all of the LBG. The unfractionated culture sample probably had the highest residual LBG concentration, and therefore showed lower detectable activity on azo-carob galactomannan than expected. Nevertheless, C. fimi appeared to contain cell-bound B-mannanase which, depending on the culture conditions, possibly, could be released into the culture supernatant. More experiments are necessary to study the cell-association of Man26A. One possible experiment would be to synthesize a fluorescently labeled, mechanism-based mannanase inhibitor and detect inactivated enzyme bound to the cell surface by fluorescence microscopy. 3.3.3 The C.fimi p-mannosidase, Man2A The manlA gene is only the second bacterial, and the first eubacterial p-mannosidase gene described (Bauer et al., 1996). The Man2A protein is closely related to the mammalian lysosomal p-mannosidases, on the basis of primary amino acid sequence alignments (Figure 94 3.18). Man2A was inhibited by low concentrations of P N P M . T. reesei B-glucosidase (Estrada et al., 1990) and B-galactosidases from E. coli, Aspergillus and Jack bean (Pocsi et al., 1993) are also inhibited by substrate. Inhibition of Man2A at low concentrations of substrate suggested that the intracellular mannooligosaccharide levels are controlled, in order to maintain optimal activity of the enzyme. Substrate inhibition wil l be discussed in Section The C. fimi p-mannosidase migrated considerably faster during SDS-PAGE under non-reducing conditions than under reducing conditions, an indication that native Man2A forms a tight fold, possibly stabilized by disulfide bonds formed between two or more of the six cysteine residues in Man2A. 3.3.4 Degradation of mannooligosaccharides by Man26A and Man2A Man2A hydrolyzes all of the p-1,4 mannooligosaccharides tested M2 to M6 to mannose as the major product, with some minor products, e.g. mannotetraose cleavage also produced some mannobiose and less mannotriose as minor products. Mannopentaose is hydrolyzed 8 times faster than mannobiose by the exo-P-manannase from guar seeds, whereas the hydrolysis rates from the p-mannosidase from Helix pomatia are identical for mannooligosaccharides with DPs from 2 to 5 (McCleary, 1983). The accumulation of more mannobiose than mannotriose during hydrolysis of mannotetraose and mannohexaose by Man2A indicated that it might prefer oligosaccharides longer than mannobiose as substrates (Figure 3.27). The production of mannotriose and mannobiose also suggested that the degradation of mannooligosaccharides by Man2A was not processive, at least not for mannooligosaccharides with DP from 2 to 6. As the optimum DP of the substrate for 95 Man2A has not been determined, it is unclear if Man2A is an exo-p-mannanase or a p-mannosidase. The major products released from mannohexaose, -pentaose and -tetraose by Man26A were mannobiose and mannose. Mannotriose was hydrolyzed only slowly by Man26A, to mannobiose and mannose, whereas mannobiose was not hydrolyzed at all. Hydrolysis of mannopentaose by the T. reesei p-mannanase, BMANI occurs via a fast initial cleavage into mannotriose and mannobiose, followed by a slower hydrolysis of mannotriose into mannobiose and mannose (Hariupaa et al, 1995). In other systems, hydrolysis of mannohexaose and mannopentaose produces mainly mannobiose and mannotriose, because enzymes such as the Aspergillus niger p-mannanase are unable to cleave mannotriose (McCleary et al, 1983; Arcand et al, 1993). Man26A cleaves mannotriose into mannobiose and mannose. Mannotetraose is also cleaved into mannobiose and mannose. Presumably mannotetraose was first cleaved into mannose and mannotriose, as with the A. niger mannanase (McCleary et al, 1983). The A. niger mannanase, and other mannanases, also transglycosylate (Coulombel et al, 1981; McCleary et al, 1983). It is possible that Man26A hydrolyzes mannotetraose into mannobiose, and that the mannobiosyl-Man26A covalent intermediate transfers mannobiose onto mannotetraose, producing mannohexaose; mannohexaose could then be hydrolyzed to mannobiose and mannose as described. Mannotriose is produced from mannohexaose, -pentaose and -tetraose, but it appeared as a doublet in the FACE gels (Figure 3.27). This doublet, however, was not produced from mannotriose. Since the regiochemistry of glycosidic linkages influences the mobility of oligosaccharides in FACE gels, e.g. a-1,3 mannobiose migrates faster than cc-1,4 mannobiose (Linda Sandercock, personal communication), it is possible that one of the bands seen for mannotriose is P-1,3 linked mannotriose, whereas the other one, comigrating with the 96 standard, is P-1,4 linked mannotriose. Only the faster migrating form of mannotetraose, presumably including a linkage other than p-1,4, was detected. Transglycosylation seems to occur with mannohexaose, -pentaose and -tetraose but not with mannotriose and mannobiose. Man2A also transglycosylates (Chapter 6). The cleavage of mannooligosaccharides by Man26A is illustrated in Figure 3.28. Cleavage of P-1,4 mannohexaose: Cleavage of P-1,4 mannopentaose: Cleavage of P-1,4 mannotetraose: Cleavage of P-1,4 mannotriose: Figure 3.28: Scheme of P-1,4 mannooligosaccharide cleavage by Man26A. Cleavage sites are indicated by arrows and P-1,4 linked mannose molecules by filled circles. The numbers indicate the relative position of the sugars. O--©--©-^-© --© -© 0~©~©-^-©~© O--©--©-*-© O--©-*-© 97 3.3.5 Hydrolysis of galactomannan and mannan by Man26A and Man2A The release of a variety of mannooligosaccharides from LBG, a substituted galactomannan, suggests that the C. fimi p-mannanase was an enzyme with endo activity (Figure 3.26). Mannotriose, mannobiose and mannose were released from unsubstitued mannan (INM) indicating a preference of Man26A for mannooligosaccharides of DP >3. Man2A released only insignificant amounts of mannose from LBG, indicating that Man2A attacked the substrate only from the non-reducing end and that a-l,6-galactose side chains block futher hydrolysis. Significantly more mannose is cleaved from I N M with Man2A, consistent with a high concentration of free ends (Section 4.2.5). The concerted action of Man26A and Man2A was sufficient to degrade I N M into mannose. More mannose was produced in the presence of both enzymes than with Man2A alone. Complete degradation of LBG into monomers could not be obtained with the two enzymes, suggesting a requirement for the intracellular a-galactosidase to remove the galactose side chains (Section 3.2.1). The only detectable difference between products from the hydrolysis of LBG by Man26A and by Man26A and Man2A LBG hydrolysis was the complete cleavage of the disaccharide product in the presence of both enzymes. This disaccharide was therefore mannobiose. The only disaccharide produced from LBG by A. niger p-mannanase is mannobiose, but two trisaccharides, mannotriose and 61-a-galactosyl-p-D-marinobiose are produced (McCleary et al, 1983). The relative amounts of the products with the mobilities of the G3, G4, G5 and G6 standards were very similar in the Man26A and Man26A + Man2A digestions, indicating that most of these oligosaccharides had an a-galactosyl substitution at their non-reducing ends, which blocked further hydrolysis by Man2A. Thus, a semi-processive hydrolysis mechanism, as was proposed for the C. fimi P-1,4 glucanase CenC 98 (Tomme et al, 1996), is proposed for the p-mannanase Man26A. In this proposal Man26A cleaves the backbone of galactomannan internally, holds on to the substrate with the help of the MBD (Section 4.3.4) and moves along the chain from the non-reducing end towards the reducing end, thereby releasing smaller oligosaccharides. The enzyme moves along the chain until the processive action is stopped by an cx-galactosyl side chain. The enzyme has to desorb from the substrate to intiate another endoglycolytic cleavage. This mechanism would explain the production of galactomannanoligosaccharides with galactosyl side chains bound mainly to the mannose at the non-reducing end of the oligomer. Thus the high degree of substitution in LBG, with a mannose to galactose ratio of 5:1 (McCleary et al, 1985), reduces the efficiency of semi-processive cleavage of LBG by Man26A on LBG. 99 4 Mannan-Binding Domain 4.1 Introduction 4.1.1 Carbohydrate-binding domains of polysaccharidases Cellulolytic and hemicellulolytic enzymes of non-complexed degradation systems are very often modular enzymes. The most common non-catalytic domains are cellulose-binding domains (CBDs). CBDs are believed to enhance activity by binding the enzyme to the substrate thereby increasing the enzyme concentration on the substrate (Gilkes et al, 1992 and 1993). In complexed cellulolytic systems, e.g in the cellulosome of C. tltermocellum, the CBD is part of the scaffoldin protein and mediates the anchoring of the entire cellulosome to the substrate. Since cellulosomes are mostly cell bound multienzyme complexes, the scaffoldin CBD is also ultimately responsible for binding the C. thermocellum cells to the substrate, allowing improved uptake of the degradation products (Beguin et al, 1996). Substrate binding domains are a common occurence not only in plant cell wall degrading enzymes, but also enzymes involved in starch and chitin degradation that use starch-binding and chitin-binding domains, respectively, to adsorb to the substrate (Williamson et al, 1997, Tomme et al, 1995). CBDs have been classified into families based on amino acid sequence similarities (Tomme et al, 1995). Currently, more than 170 cellulose-binding domains have been identified and grouped into 17 different families (Creagh et al, 1998; P. Tomme, personal communication). Their sizes range from 33 to over 170 amino acid residues. A l l the cellulases and xylanases secreted by Cellulomonas fimi include at least one CBD, including members from CBD families I, II, III, IV and IX. The CBDs differ in their 100 binding affinities for the different forms of cellulose. The two tandemly repeated CBDs from the C. fimi endoglucanase CenC, C B D N I and C B D N 2 , are unique in their properties. They bind soluble cello-oligosaccharides, and amorphous but not crystalline cellulose (Tomme et al, 1996b, Coutino et al, 1992). The biotechnological potential of CBDs led to detailed structural and functional studies of C B D N I and CBDc e x. The CBD c e x from the C. fimi xylanase/exoglucanase binds, unlike C B D N I , to crystalline cellulose. 4.1.2 CBDcex CBDcex is 108 aa long and is at the C-terminal end of Cex (Figure 1.3). Three dimensional structure analysis of this family II CBD revealed a nine-stranded B-barrel fold (Xu et al, 1995). Ln this structure the three exposed tryptophans, W17, W54 and W72, are aligned on one face as shown in Figure 4.1. The role of these aromatic residues in cellulose binding was demonstrated by chemical modifications (Bray et al, 1996). The arrangement of hydrophobic amino acid side chains on one face is also found in other CBDs. NMR structures of T. reesei CBDCBHI (family I) and Envinia chrysanthemi C B D E G Z (family V) revealed either tyrosines or tryptophans on the putative binding faces (Brun et al, 1997). Each of these tyrosines in CBDCBHI was shown by site-directed mutations to be indispensable in cellulose binding (Linder et al, 1995). Binding studies of CBDcex on crystalline cellulose showed that binding is entropically driven, indicating significant dehydration of the protein-ligand contact surfaces. Whereas polar interactions, such as hydrogen bonding and van der Waals interactions, provide the major driving force in the enthalpic binding process of C B D N I to soluble sugars (Tomme et al, 1996b, Creagh et al, 1998). 101 Courtesy of Bradley W. McLean Figure 4.1: Ribbon diagram of the structure of CBDC e x . The amino acid side chains of three of its five tryptophan residues are shown (W17, W52 and W72). These three tryptophans are aligned along one face of the nine-stranded p-barrel fold. These three tryptophans are conserved in all family II cellulose-binding domains. 102 Courtesy of J.M. Kormos Figure 4.2: Ribon diagram of the structure of CBDN 1 . The ten P-strands are organized in two antiparallel P-sheets, which adopt an overall topology of a jelly-roll P-sandwich. Some of the amino acid side chains involved in substrate binding are shown. Hydrophobic side chains are highlighted in blue, polar residues in green and charged residues in red. 103 4.1.3 C B D N I C B D N I is unusual in respect of its binding properties and its structure. It does not bind to crystalline cellulose, but it can bind amorphous cellulose and soluble oligosaccharides (Johnson et al, 1996a; Tomme et al, 1996b). The tertiary fold of this 152 aa long C B D is strikingly different from that of C B D c e x . The jelly-roll B-sandwich lacks a flat binding surface; it forms a binding cleft closely resembling the structure of an endo-1,3-1,4-B-glucanase from Bacillus macerans (Johnson et al, 1996b) (Figure 4.2). The binding cleft can accommodate approximately five glucopyranosyl residues. A strip of hydrophobic amino acid residues runs along the center of the binding cleft, whereas the sides are lined by mostly hydrophilic residues. Two tyrosines, Tyrl9 which is in the cleft and Tyr85, which is on a loop proximal to the cleft, play critical roles in binding the oligosaccharide chain (Kormos, 1998). 4.1.4 Objectives It was the aim of this project to study the substrate-binding of the C. fimi mannanase, Man26A. A putative mannan-binding domain from this enzyme was sub-cloned and expressed in E. coli to study mannan binding. 104 4.2 Results 4.2.1 Characterization of a novel domain from C. fimi mannanase, Man26A The region between the catalytic domain and the SLH domain in Man26A had no significant sequence similarity to any other known protein sequence (Section 3.2.4). A l l of the characterized secreted C. fimi cellulases and hemicellulases are modular enzymes and have at least one cellulose binding domain. It was therefore hypothesized that Man26A might have either a cellulose or hemicellulose-binding domain. To test this hypothesis, binding of Man26A to the insoluble substrates chitin, mannan, cellulose and xylan was analyzed. However, binding was not detected to any of these polysaccharides (data not shown). Binding of Man26A to soluble mannan was studied by affinity gel electrophoresis (AFGE). In this technique native protein samples are electrophoretically separated in polyacrylamide gels containing different concentrations of soluble substrate. Reversible binding of the protein to the soluble substrate would reduce its migration distance (mobility). The change in mobility is proportional to the substrate concentration. The mobility of proteins without binding specificity should not change, even in the presence of high concentrations of the polymeric substrate (Nakamura et al, 1992). For the binding study of Man26A and Man2A, azo-carob galactomannan was added to the AFGE gels. The relative mobility, which is used to calculate the dissociation constant, is expressed as the migration distance of the protein sample versus the migration distance of a non-binding control. BSA was used as a control in the experiments shown in Figure 4.3. The relative mobility of Man2A was not affected by mannan, not even at a concentration of 1 %, 105 106 indicating that Man2A did not bind to soluble mannan. By contrast changes in relative mobility of Man26A, were detected at very low substrate concentrations and the presence of 1 % substrate prevented the protein from migrating into the gel. This suggested a relatively strong interaction of Man26A with azo-carob galactomannan. The AFGE-zymogram method was used to test whether the catalytic domain of Man26A, or the non-catalytic portion of the enzyme was involved in substrate binding (Figure 4.3). The catalytic domain was obtained by protease treatment of Man26A (Section 3.2.6). Changes in relative mobility of undigested Man26A were compared to changes in relative mobility of the Man26A catalytic domain on AF gels with 0 % and 1 % mannan. To confirm that the proteolytic band corresponded to the catalytic domain, the 1 % azo-carob galactomannan gel was incubated in phosphate buffer (1 h at 37° C) prior to Coomassie-staining. The presence of mannan did not affect the relative mobility of the catalytic domain, seen as a zone of clearing on the 1 % substrate gel, whereas the intact enzyme did not even migrate into the gel (Figure 4.4). It appeared that mannan binding was due to a domain other than the catalytic domain. It was hypothesized that a mannan-binding domain might be present between the catalytic domain and the SLH domain. 4.2.2 Sub-Cloning of mbduu The D N A fragment encoding the protein portion between the catalytic and SLH domain, the putative mannan-binding domain (MBD), was cloned into the pET28a vector (Novagen) as described: The primers MBD11 (Nco I) and MBD12 (Not I) were used to amplify the MBD encoding D N A fragment by the polymerase chain reaction (PCR) using pCMan2 as template. The PCR product was cloned into the pZErO™-l.l vector (Invitrogen) 107 A B 1 2 3 4 5 1 2 3 4 5 0 % Mannan 1 % Mannan Figure 4.4: Affinity gel electrophoresis (AFGE) of protease treated Man26A (Lane 1 and 2), BSA (Lane 3) and intact Man26A (Lane 4 and 5) without (Panel A) or with 1 % (Panel B) macromolecular affinity ligand, i.e. azo-carob galactomannan. To detect mannanase activity the gel was incubated in excess potassium-phosphate buffer (pH 7.0) at 37 °C until clearings were visible. Protein bands were then visualized by Coomassie staining. 108 at the EcoR V restriction site. From this plasmid, pOMBDim, the 570 bp mbdllll D N A fragment was excised by Nco I and Not I restriction endonuclease digestion, and ligated into the expression vector pET28a to obtain pET28MBDm2, encoding the putative MBD translationally fused to a C-terminal sequence of six histidines. The mbdllll coding region was expressed in E.coli BL21(DE3) cells, producing a protein, MBDim, with a calculated molecular weight of 20,990. Initial expression levels were low and purification by M C A C was unsuccessful, because of poor binding of MBDim to the affinity column. The experiments presented below were intended to determine the binding characteristics of MBDim and its potential applications, prior to improving protein expression and purification. 4.2.3 Characteristics of the mannan binding domain (MBDim) Binding of MBDim to mannan was analyzed by AFGE. The MBDim protein bands from the partially purified sample were detected after AFGE separation on Western blots, using oligohistidine specific antibodies (Figure 4.5). Man2A was used as the non-mannan-binding control. To mark the top of the AFGE gels, C B D c e n D , a protein that does not migrate into the separating gels under native conditions (Boraston, A., and Tomme, P. unpubUshed results) was applied. The relative mobilities of Man26A and MBDim were compared in gels including 0 %, 5 x 10-4 %, 7.5 x 10-4 %, 1.25 x 10-3 %, 2.5 x 10-3 % and 1.25 x 10~2 % locust bean gum and 1 x 10"3 %, 1.5xl0"3 %, 2.5 x 10"3 %, 5 x 10-3 % and 2.5 x 10"2 % azo-carob galactomannan. The relative mobilities of Man26A and MBDim decreased with increasing substrate concentrations. The decrease was similar for both proteins. However, slightly stronger protein-substrate interactions were detected for Man26A (Figure 4.5). From the 109 double reciprocal plots of l/(R-r) versus 1/c (see Equation E 4.1) the negative reciprocal of the dissociation constant could be determined as the intercept on the abscissa. E4.1: 1 = (1 + Kd/c) R -r R - Rc d: migrating distance of protein (MBDim) D: rmgrating distance of reference (Man2A) r: relative mobility (d/D) of MBDim in the presence of mannan R: relative mobility of unbound MBDim Rc: relative mobility of MBDim -mannan complex, or that value obtained in the presence of an excess amount of mannan with all MBD molecules bound. Ka: dissociation constant of protein for affinity Ligand The dissociation constants for MBDim were determined to be Kd = 4.6 x l O 4 %(+ 4 x lO 5 ) for locust bean gum and Kd = 5.5 x 10 3 % (+1 x 103) for azo-carob galactomannan. The 12 fold weaker binding of MBDim to the azo-substrate could be caused by the Remazol brilliant-Blue R molecules linked to the backbone to an extent of about one dye molecule per 20 sugar residues reducing the accessibility of binding sites. Furthermore, a lower viscosity, as found for the azo-carob galactomannan, could be indicative of molecules with a lower degree of polymerization (DP). Shorter molecules would have fewer potential binding sites, assuming MBDim binds to the mannan backbone (Section 4.2.4). In order to calculate the molar dissociation constant, the molarity of the locust bean gum solution was determined by total and reducing sugar analysis, assuming a galactose to 110 I l l Figure 4.6: Double reciprocal plot of MBDim AFGE data. Relative mobilities with azo-carob galactomannan and locust bean gum as affinity ligands were plotted as 1/(R-r) versus 1/c. Dissociation constants (Kd) were determined from the x-axis intercept (1/-K). Kd for MBD1112 on locust bean gum Kd = 4.6 x IO 4 % was and on azo-carob galactomannan Kd = 5.5 x 10-3 %. 112 mannose ratio of 1:5 (McCleary et al, 1985) (Section 2.16). The soluble galactomannan molecules had an average DP of 130 mannose residues, which meant that a 1 % solution was 0.39 mM. This gave a molar dissociation constant for MBDim on locust bean gum of Kd = 1.8 x IO"7 M . Although only an approximation, it clearly suggested a high affinity of M B D i m for soluble galactomannan. 4.2.4 Mannan-binding domain MBD1112 is not a lectin Mannan-binding proteins (MBD) described previously bind to terminal sugars and are, therefore, classified as lectins and were renamed as mannan-binding lectins (MBL)(Mizuno et al, 1981; van Emmerik et al, 1994). It was the goal of the experiments presented in this section to provide evidence that MBDim, unlike the lectins, binds to the B-1,4-mannose backbone. The first indication of a non-lectin type of binding for M B D i m came from comparisons of the dissociation constants for locust bean gum and azo-carob galactomannan (Section 4.2.3). If the lower viscosity of the latter corresponded to a lower DP value, its molar concentration would be higher. Therefore, the concentration of terminal sugars would also be greater, presenting more binding sites for mannose-binding lectins. Furthermore, if MBDim bound only to terminal residues, the presence of Remazol Brilliant Blue should not significantly influence the binding. Competition affinity gel electrophoresis was used to test the hypothesis that M B D i m is not a lectin-like binding protein. The relative mobility of the mannan-binding domain in a gel with 1.25 x 10-2 % mannan was compared to its relative mobility in 1.25 x 10-2 % mannan gels including 1.8 % mannose or galactose. 113 u 1^1 H M l ro o 3 u OJ c o cC 43 C CJ C O CJ CN H cu a, £ o Q o c H > ro S u <3 1 Q" ro . O H U 5 cu ro a> pa 6 p C o u R3 OH Q 03 O H CK cn • M cn 1 O a, « 2 8 u -rt J cu O) T3 H « cn ro i % Ej ro X d xi •S P 2 £ 43 cn c J eu g i 1 * M3 CU ro cu 8 3 cu cn C cu 01 i f o ro ro X cn _, 0 .S G cu 11 cn M ^ QJ T3 cn CU o ro * ro m bo m r'] $ -cn cu 1 s <D X § I O <J 0 ° I N CU [Si ^3 — (3 >•, OS 2 S ro g tn ro | S PH-H fi i o ro U PH in the presence of an excess of either galactose or mannose. If MBDim preferentially bound monomeric sugars, it would have formed a sugar- MBDim complex and would not have interacted with the mannan. The relative mobility of MBDim was not changed significantly in the presence of excess galactose or mannose (Figure 4.7), indicating binding preference for polymeric mannan and not for monomeric or terminal mannose or galactose residues, as found in galactomannans. Therefore, the mannan-binding domain from the C. fimi mannanase Man26A can be classified as a true mannan-binding domain. This is the first mannan-binding domain to be described. 4.2.5 The role of MBD in mannan degradation The role of MBD in Man26A catalyzed mannan degradation was investigated. Locust bean gum and ivory nut mannan were incubated with either intact Man26A or Man26A catalytic domain (prepared by proteolytic cleavage; Section 2.19). Man26A had a lower specific activity on crystalline ivory nut mannan (36.5 nmol/mL.min: released reducing sugars per nmol Man26A) than on locust bean gum (123.6 nmol/mL.min per nmol Man26A). 10 mL of 0.2 % locust bean gum in 50 m M citrate p H 5.5 and 10 mL of 0.2 % ivorynut in 50 m M citrate p H 5.5 were incubated with 0.2 nmol and 1.2 nmol, respectively, of intact Man26A and Man26A catalytic domain. After a 30 h incubation period at 37° C, the concentration of released reducing sugars was analyzed. The results summarized in Table 4.1 indicated that hydrolysis rates on locust bean gum were similar for intact Man26A and for Man26A catalytic domain. However, the hydrolysis rates for Man26A were 1.4 times higher than the rates for Man26A catalytic 115 domain on crystalline mannan (i.e. ivory nut mannan), indicating a role of MBD in enhancing hydrolysis rates on crystalline substrate. Table 4.1: Reducing sugars produced by enzymatic mannan hydrolysis (30 h at 37° C) Substrate Man26A • atalvtic domain Man26A Locust bean gum (0.2 %) Ivory nut mannan (0.2 %) 2.6 (+0.18) u M o l / m L a 1.69 (+0.23) u M o l / m L b 2.hh (+0.31) L i M o l / m l . ' 2.45 (+0.13) uMol/mL b a: enzyme concentration: 0.02 nmol/ mL b: enzyme concentration: 0.12 nmol/mL 116 4.3 Discuss ion 4.3.1 The mannan-bining domain of C. fimi mannanase, Man26A Unlike all of the C. fimi cellulases and xylanases described previously, Man26A does not have a cellulose-binding domain. Binding of Man26A to insoluble substrates, such as mannan, chitin, xylan or cellulose was not detected. Binding of Man26A to soluble mannan however, could be demonstrated by AFGE. The D N A encoding the mannan-binding domain, located between the catalytic domain and the SLH domain, was sub-cloned and expressed. Expression levels were low and the MBD with a C-terminal His-tag (MBDim) could only be partially purified. In this study, no attempts were undertaken to solve these problems. A possible reason for the purification problems might be the presence of a protease susceptible region in the MBDim C-terminus (Figure 3.1.4). Cleavage at this site would cause the loss of the affinity tag. By using the AFGE-Western blot technique, both of the problems could be circumvented and dissociation constants (Kd) could be calculated for MBDim on soluble mannans. K d = 4.6 x 1CM %(+ 4 x IO5) on locust bean gum and K d = 5.5 x IO"3 % (+1 x IO"3) on azo-carob galactomannan. Plotting the mobility data as 1/r versus c, as is usually done for AFGE data analysis, resulted in convex, instead of linear curves. This non-linear behaviour has been previously reported, e.g. for lower molecular weight dextran-ConA complexes (Takeo, 1984). It is believed that the protein retains some mobility, even when tightly complexed with the substrate. To correct for this behaviour, the data had to be plotted according to Equation 4.1 (Takeo, 1984). The calculations of the molar dissociation contant for M B D i m (Kd = 1.8 x IO 7 M) on locust bean gum were based on an average DP value of 130. This value, however, was obtained by assuming a 1:5 galactose to mannose ratio, as has 117 been reported in the literature (McCleary, et al, 1985). Therefore, the molar dissociation constant for M B D i m can only be seen as an approximation. The production of larger quantity and purer quality of the mannan-binding domain wil l allow to do more sensitive assays (i.e. isothermal titration calorimetry or surface plasmon resonance) to obtain more accurate binding affinties (Holmskov et al, 1996). 4.3.2 Comparison of M B D i m to C B D N I M B D i m and C B D N I are similar in respect to their binding preferences. Both domains bind soluble substrates but do not bind crystalline substrates. Both domains are about 150 amino acid residues long and have relatively low calculated pis (3.5 for C B D N I and 4.3 for M B D i m ) . Unlike Man26A, which has only one binding domain, CenC has two similar tandemly repeated binding domains, C B D N I and C B D N 2 (Coutinho et al, 1992). AFGE experiments were also used to determine a dissociation constant of C B D N I on soluble barley B-glucan (B-1,4 and B-1,3 glucose polymer), Kd = 7.57xl0-6 M (Kormos, 1998), which indicated 42 fold weaker binding of C B D N I to barley B-glucan compared to binding of M B D i m to locust bean gum. Comparison of the putative binding mechanisms for CBDcex, which binds with its hydrophobic binding face to a crystalline substrate (Creagh et al, 1996, Bray et al, 1996), and for C B D N I , which binds flexible, single-stranded soluble substrate in a binding cleft, leads to the speculation that M B D i m adopts a similar overall topology to C B D N I , forming a substrate binding cleft rather than a binding face. For both binding site topologies, hydrophobic amino acid side chains (i.e., tryptophans or tyrosines) were shown to be key components in carbohydrate-protein interaction (Bray et al, 1996; Kormos, 1998; McLean 118 personal communication). With 8 tyrosines and 4 tryptophans present in MBDim there are many putative carbohydrate-binding amino acid side chains. 4.3.3 Lectins Another class of proteins binding to mannan is the collectins. Collectins are C-type (Ca 2 + dependent) carbohydrate-binding proteins that play an important role in the innate immune defence against microorganisms (Lu, 1997). C-type lectins, such as the mannan binding lectin (MBL, previously known as mannan binding protein, MBP) and collectin-43 (CL-43) bind mainly to terminal mannose and glucose in the presence of Ca 2 + . By surface plasmon resonance, the dissociation constant for the monomeric CL-43 lectin was determined on Saccharomyces cerevisiae mannan as Kd = 2.68 x 10-8 (Holmskov et al, 1996). However, the collectins bind only to terminal sugars. The results from competition AFGE indicated, that MBDim preferentially bound to polymeric B-1,4 linked mannose rather than to monomeric mannose or galactose. It was therefore proposed that MBDim did not bind to mannan in a lectin-like fashion but rather as a true B-1,4 mannan-binding protein. This is the first report of such a mannan-binding domain. 4.3.4 The role of MBDim in Man26A catalyzed mannan degradation The role of MBDim in mannan hydrolysis by Man26A was investigated on locust bean gum and on ivory nut mannan. The latter is an unsubstituted, crystalline (3-1,4-mannan, whereas locust bean gum is more amorphous and has (a-1,6) galactose substitutions. Intact Man26A and Man26A catalytic domain released similar quantities of reducing sugars from locust bean gum over a 30 h incubation period. However, the intact 119 Man26A was 1.4 times more efficient than Man26A catalytic domain in producing reducing sugars on ivory nut mannan. Similar results were obtained for the degradation of bacterial micro-crystalline cellulose (BMCC) with CenA. Intact CenA was 3.3 times more efficient in degrading the substrate, than the catalytic domain without the binding domain (Gilkes et al, 1992). For CenA, the results agree with the proposed role of CBDs to increase the local concentration of catalytic domain on the substrate. This, however, can not be the role of MBDim because binding of Man26A to ivory nut mannan could not be detected. It is possible that Man26A is able to degrade ivory-nut mannan semi-processively, but semi-processivity on locust bean gum is impaired by the galactose substitutions. Serni-processivity, endo-cleavage followed by processive exo-cleavage, has been suggested previously for CenC based on C M C viscosimetric analyses (Tomme et al, 1996). CenC and Man26A both have at least one binding domain that bind soluble oligosaccharides. Therefore, it is hypothesized that the binding domain is involved in the semi-processive cleavage of soluble or amorphous substrates, by "feeding" the oligosaccharide chains to the catalytic domain, and that substitutions, such as the galactose side chains in locust bean gum inhibit processive hydrolysis. Semi-processivity was also proposed for the Thermomonospora fusca E4 B-glucanase (Irwin et al, 1998). The crystal structure of E4 (3-glucanase catalytic domain linked to the family IIIc E4 CBD has been solved (Sakon et al., 1997). The binding face of the type IIIc CBD and the shallow catalytic cleft of the catalytic domain are aligned in E4 in such a way that a cellulose strand could bind along both domains and enable semi-processive hydrolysis (Irwin et al, 1998). By analogy it is thought that the catalytic domain of Man26A cleaves crystalline mannan (endo-acting) releasing soluble mannan chains to which the MBD binds. The MBD 120 would feed the strands to the catalytic domain, which would processively degrade the strands (exo-acting) into mannobiose or mannose. This model could be tested by viscosimetric assays using substrates with different degrees of substitutions, or viscosimetric assays with solubilized ivory nut mannan, comparing Man26A to a non-modular, only endo-acting mannanase, i.e., P. fluorescens ManA (Braithwaite et al, 1995). 4.3.5 Protein purification by aqueous two-phase systems Separation of two phases in an aqueous two-phase system has a wide varyiety of applications. One of them is large-scale protein purification. Phase separation with solutions of galactomannans and polyethylene glycol (PEG), dextran or citrate has been reported (Franco et al, 1996). The binding characteristics of M B D i m make this binding domain a very good candidate to use as an affinity tag in the purification of fusion proteins. In preliminary experiments with locust bean gum-PEG two-phase systems, purification of Man26A from 1.3kg of E. coli supernatant was achieved. Adding 12 g of locust bean gum (600 g of 2 % solution) and 200 g of PEG 20,000 to 1.3 kg of supernatant resulted in a two phase system in which Man26A patitioned into the small volume mannan phase (data not shown). However, detailed binding studies with M B D i m are required to find non-denaturing desorption conditions for removal of the fusion protein from the mannan. 121 5 Model of galactomannan degradation by C. fimi The cellulose hydrolyzing system from the Gram positive soil bacterium Cellulomonas fimi does not represent a paradigm for hydrolysis of plant cell wall polysaccharides by this bacterium. The mannan hydrolyzing system in C. fimi is, unlike the cellulolytic system, composed of only one secreted "endo-acting" enzyme, i.e. endo- 1,4-B-mannanase (Man26A), one intracellular 1,4-B-mannosidase (Man2A) and one intracellular 1,6-a-galactosidase. Galactomannan, mannan, mannose and galactose were all good carbon sources for C. fimi. The secreted (3-mannanase can be either cell-bound or released into the culture supernatant as an intact enzyme or as proteolytically processed enzyme. The interaction of Man26A with the cell envelope, presumably with a polysaccharide thereof (Egelseer et al, 1998), is mediated by the SLH domain (S-layer homology domain). This is the first report of a cell-bound C. fimi polysaccharidase. Man26A also comprises a substrate binding domain, i.e., the mannan-binding domain (MBD). This MBD does not bind to crystalline but does bind to soluble mannan, similar to C B D N I from the C. fimi endoglucanase CenC (Tomme et al, 1996b). Possibly, the MBD is involved in the semi-processive hydrolysis mechanism of mannan by Man26A, by "feeding" soluble, single strands of substrate to the catalytic domain as was proposed for Thermomonospora fiisca E4 endoglucanase (Irwin et al, 1998). Man26A hydrolyzes galactomannan into a variety of oligosaccharides, which for further hydrolysis have to be imported into the cells. The cellulase system hydrolyses cellulose to cellobiose outside the cells and presumably imports only cellobiose and glucose. Within the cells the galactomannooligosaccharides are hydrolyzed into monomers by the concerted action of the B-mannosidase, Man2A and the 120 kDa oc-galactosidase. The amount of Man26A (or its catalytic domain) released into the 122 culture medium could be dependent on the concentration of secreted C. fimi protease. The proposed hydrolysis mechanism of galactomannan hydrolysis by C. fimi cultures is summarized in Figure 5.1: The presence of a substrate-binding domain (MBD) and a cell adhesion SLH domain in Man26A promotes binding of the cells to the substrate enabling efficient uptake of the galactomannooligosaccharides produced by Man26A into the cells where they are further hydrolyzed into monomeric mannose and galactose by the a-galactosidase and the (3-mannosidase. As the cell density increases in the culture (Figure 5A) the concentration of Man26A increases as well, with a concomitant increase in galactomannooligosaccharide product concentration. With the increased product concentration in the environment the cells no longer have to be very close to the substrate to guarantee sufficient oligosaccharide uptake. Therefore the cells can afford to release Man26A into the culture supernatant, which possibly is regulated by proteolytic processing of Man26A and is therefore dependent on the concentration of secreted C. fimi protease. At high cell density the concentration of secreted protease is presumably also increased, thereby processing and hence releasing more Man26A into the culture supernatant. In this proposed mannan-degrading system of C. fimi it is possible that the proteolytically processed mannanase takes over the "endo" hydrolysis of the substrate and the cell-bound unprocessed mannanase acts as the "exo"-enzyme. 123 Figure 5.1: Schematic mannan degradation by Cellulomonas fimi. CD: catalytic domain, SLH: SLH domain, MBD, mannan-binding domain; aGal: a-galactosidase. 124 6 Conversion of Man2A into a glycosynthase 6.1 Introduction 6.1.1 Enzymatic synthesis of oligosaccharides Oligosaccharides play important roles in a range of biological processes such as cell recognition, growth and differentiation (Edelman, 1983) and pathogen adherence (Section 4.3.3). The use of oligosaccharides in studies of these processes and their prospective use as therapeutics, is the driving force for the development of methodologies for the efficient synthesis of oligosaccharides. The synthesis of such hetero-oligosaccharides with regio- and stereospecific interglycosidic linkages by means of classical carbohydrate chemistry is a demanding task (Thiem et al, 1995). Therefore different strategies for the enzymatic formation of oligosaccharides using either glycosyl transferases (EC 2.4.-) glycosyl hydrolases (EC 3.2.-) or glycosynthases are being investigated. Glycosyl transferases Glycosyl transferases are enzymes that are responsible for the biosynthesis of oligo-and polysaccharides. They catalyze the specific transfer of a monosaccharide from a nucleotide donor sugar to the acceptor substrate (Ichikawa et al, 1992). Several disadvantages limit the use of these enzymes in larger scale in vitro processes. A common problem is the cloning, high level expression and stability of specific transferases. Furthermore the high cost of glycosyl nucleotides may be prohibitive, although complex, but effective, recycling schemes have been developed, thereby also reducing the problem of 125 product inhibition. Another problem with glycosyl transferases is their frequently rigid specificities, linuting their use with unnatural substrates (Ichikawa et al, 1992). Glycosidases Retaining glycosidases (Section 1.2) are also useful tools in oligosaccharide synthesis. A wide variety of these enzymes is available (Section 1.3), and, unlike glycosyl transferases, they do not require expensive activated sugar donors for oligosaccharide synthesis. Furthermore, their high glycon- and low aglycon specificities allow the synthesis of unnatural substrates (Murata et al, 1997, Yasukochi et al, 1997). The formation of the glycosyl-enzyme intermediate in the synthetic pathway of retaining glycosidases (Figure 6.1, step 1) corresponds to the glycosylation step of the hydrolytic pathway (Figure 1.1). In the deglycosylation step of the synthetic pathway however, the acceptor sugar competes with water for the nucleophilic attack at the anomeric center of the reactive glycosyl-enzyme complex (Figure 6.1). When an acceptor sugar acts as nucleophile the reaction mechanism is referred to as transglycosylation (Toone et al, 1989). A major disadvantage of this method is that newly formed oligosaccharides can be immediately hydrolyzed by the glycosidase. One way to reduce oligosaccharide hydrolysis is by use of readily synthesized, cheap donor sugars, such as nitrophenyl glycosides or glycosyl fluorides. Preferential cleavage of these donor sugars by the enzyme reduces the overall hydrolysis of the synthesized oligosaccharides (Ichikawa et al, 1992). Other means of improving reaction yields are the selective removal of the newly formed products from the reaction or by lowering the activity of water, using organic solvents (Fan et al, 1995). However, even with these improvements reaction yields are typically low. 126 1^ CO m 73 Ci ca a? CN ^ 5 o cn cu H cu ^ '55 < o CN 01 73 cn O 60 C O PT OI 60 O S H 73 01 (3, c« CJ hi c3 " M _ t>0 <C ^ O) £ rH W PHPH PH cn 3 S3 © 1 & ti 60 2 .S ^ r o 6 ^ CC CJ 01 ) H o PH 01 O CJ ns 01 cn ns en. S3 o o •43 ns cn cj\ 60 Fn cn i^, ro <u 73 PH cu o cn cu PH CJ S3 O cn 0) o •43 cu ro )H PH § § '•S M> ro _ cn C 0 cu X > PH 60 g cn J3 g -a 01 ro 127 Glycosynthases A third and very novel approach to enzymatic oligosaccharide synthesis is the use of glycosynthases. Glycosynthases are mutant forms of glycosidases that are able to synthesize oligosaccharides but can no longer hydrolyze glycosidic linkages (Mackenzie et al, 1998). In the case of retaining glycosidases, the catalytic nucleophile is replaced with a non-nucleophilic amino acid side chain, leaving the rest of the active site, glycon and aglycon site and the acid/base catalyst unaltered. These mutant proteins no longer form covalent enzyme-sugar intermediates, which in wild type glycosidases are formed by the attack of the catalytic nucleophile with inversion of stereochemistry at the anomeric center (Figure 1.1). The catalytic requirement for a glycosyl-enzyme intermediate can be overcome by using activated donor molecules with small aglycon/leaving groups, such as readily synthesized glycosyl fluorides. The anomeric configuration of the activated donor has to be inverted with respect to the natural substrate in order to mimic a glycosyl-enzyme intermediate (Figure 6.2). The binding of an acceptor sugar instead of water in the glycosynthase aglycon site provides the additionally required activation energy to catalyze the release of the leaving group (i.e., fluoride ion) and the formation of the new glycosidic bond. This approach of using glycosynthases in oligosaccharide synthesis includes all the advantages of using glycosidases for oligosaccharide synthesis, but eliminates most of the disadvantages (Section Prior to this study, the concept of glycosynthase catalyzed oligosaccharide synthesis had been demonstrated only with the glycosynthase derived from the Agrobacterium sp. B-glucosidase/galactosidase (Abg) (Mackenzie et al, 1998). This glycosynthase was constructed by mutating the active site nucleophile, glutamate 358, to an 128 129 alanine (Abg E358A). Abg E358A-catalyzed oligosaccharide synthesis was demonstrated with a-galactosyl fluoride or a-glucosyl fluoride as activated donor sugars and a variety of aryl-glycoside acceptors. Product yields between 64 % and 92 % were obtained (Mackenzie et al, 1998). These product yields demonstrate the great potential for the glycosynthase approach for enzymatic oligosaccharide synthesis. However, with Abg E358A being the first and only report of a glycosynthase, the general applicability of this concept had yet to be proved. 6.1.2 Identification of the catalytic nucleophile In the absence of structural information the first step towards the conversion of a retaining glycosidase into a glycosynthase is the identification of its catalytic nucleophile. The covalent glycosyl-enzyme intermediate of retaining glycosidases can be trapped with mechanism-based inactivators if the rate of intermediate formation (i.e., glycosylation), is significantly faster than the rate of deglycosylation. Conduritol epoxides, which incorporate an endocyclic epoxide within a cyclitol ring, form one class of such inactivators (Legler 1990). Protonation of the epoxide and the concomitant attack by the enzymic nucleophile leads to epoxide ring opening and covalently links the inactivator to the nucleophile, forming a stable intermediate. The use of conduritol epoxides however, has led to the misassignment of active site residues in several cases, e.g. E, coli (lacZ) B-galactosidase (Gebler et al, 1992) and human lysosomal B-glucocerebrosidase (Miao et al, 1994). Another class of irreversible mechanism-based inactivators is the activated 2-deoxy-2-fluoro glycosides. The substitution of the hydroxyl group at the glycoside 2 position with a fluorine destabiUzes the transition state, which slows both the glycosylation and 130 deglycosylation steps (McCarter et al, 1992, Namchuk et al, 1995). The glycosylation step however, can be accelerated relative to the deglycosylation step by adding a good leaving group, such as 2,4-dinitrophenolate or fluorine, to the sugar acetal center. This causes not only an accumulation of the glycosyl-enzyme intermediate but also an inactivation of the enzyme. Using a combined approach of reverse phase high pressure liquid chromatography (RP-HPLC) and electrospray ionization (ES) tandem mass spectrometry (MS/MS) the labeled catalytic nucleophile can be identified (Withers and Aebersold, 1995). In these techniques, pepsin digested glycosidase, labeled with inactivator or unlabeled, is analayzed by ESMS for peptides with a mass increase, corresponding to the label, by either comparative mapping or by neutral loss ( The sequence of the labeled peptide can then be determined either by Edman degradation or by MS/MS. The M S / M S sequencing technique is based on collision induced fragmentation of the labeled parent peptide producing daughter ion fragments, whose masses are analyzed. With the deduced glycosidase amino acid sequence available and the mass results from the peptide and the fragments thereof, the sequence of the labeled peptide and the catalytic nucleophile can generally be determined (Withers and Aebersold, 1995). 131 6.1.3 Objectives The aim of this project was to transform the C. fimi mannosidase, Man2A, into a glycosynthase and to test its ability to synthesize oligosaccharides and to prove that the concept of glycosynthases is generally applicable. The glycosynthase was produced by mutating the carboxylic catalytic nucleophile, which was identified by using the mechanism-based inhibitor MS/MS approach, to an alanine. 132 6.2 Results 6.2.1 Identification of the catalytic nucleophile in the C. fimi mannosidase, Man2A Prediction of the position of catalytic residues The conversion of a retaining glycosidase into a glycosynthase by mutating the catalytic nucleophile to an alanine requires knowledge of the exact position of the nucleophile. In many cases the nucleophile can be accurately predicted by alignments of the amino acid sequences of members of the same glycosidase family (Figure 3.7). If the catalytic residues were determined for one member of the family, from either 3D structures, labeling studies or from kinetic studies of catalytic residue mutants, the catalytic residues for new members of the family could be inferred if the sequences were sufficiently similar. This method could be used to predict the catalytic residues for the C. fimi mannanase Man26A (Figure 3.7). In case of insufficient sequence similarity, secondary structure predictions, e.g. hydrophobic cluster analysis (HCA), can be used to look for common secondary structure motives (Section 1.3). This method however, is not always accurate and has lead to misassignments of catalytic nucleophiles (Henrissat et al, 1995 and Vocadlo, personal communication). The C. fimi mannosidase Man2A is a member of family 2, a retaining glycosyl hydrolase family (Section 3.2.9). Three homology groups can be defined within family 2: the (3-mannosidase-, ^glucuronidase- and (3-galactosidase subfamilies. Multiple sequence alignment of members of the (3-mannosidase subfamily revealed at least two conserved carboxylic amino acids that might function as the catalytic nucleophile (Figure 3.18). Multiple alignment of glycosidase sequences from all three subfamilies, however, showed 133 insufficient sequence identity C-terminal of the acid/base catalyst to accurately predict the catalytic nucleophile position in Man2A. In Figure 6.3 an alignment of the sequence around the catalytic residues of five family 2 members is shown. The acid/base catalyst, which is part of the family 2 consensus pattern W-[GS]-x (2,3)-N-E, is conserved in all members shown (Section 3.2.11). In E. coli (lacZ) (3-galactosidase, the glutamate in 536CEY538, was experimentally identified as catalytic nucleophile and, based on kinetic data Glu461 was identified as the acid/base catalyst (Gebler et al, 1992). From comparisons of the 3D structure of E. coli (lacZ) B-galactosidase (Jacobsen et al, 1994) and the 3D structure of human lysosomal B-glucuronidase, the acid/base catalyst and the nucleophile were predicted for the B-glucuronidase to be Glu451 and Glu519, respectively (Jain et al, 1996). The H C A plots of several lysosomal glycosidases, members of the superfamily GH-A, were compared in order to find common motif and to predict the catalytic residues. For the bovine B-mannosidase, Glu457 was predicted to act as acid/base catalyst and Glu554 as nucleophile. These residues are located in the (P/a)s barrel at the C-terminal ends of the (3-4 and B-7 strands, respectively (Figure 3.19; Durand et al, 1997). Although these residues are conserved in Man2A and correspond to Glu429 and Glu519, it was considered to be important to accurately identify the nucleophile in Man2A and to examine its role in catalysis. Inactivation of Man2A using 2-deoxy-2-fluoro-B-mannosyl fluoride Initial studies indicated that 2-deoxy-2-fluoro-P mannosyl fluoride (2FManpF) was a potent inhibitor of the C. fimi B-mannosidase, Man2A (data not shown). Complete inactivation of Man2A, measured as loss of P N P M hydrolytic activity, could be obtained by 134 A/B ChMad CfMad HGus EcLacZ consensus SEETWSGN1 SLVLWNGGJ AVVMWSVAt S V I I W S L G I W NE E ..NEAALM.MGWYD.TKPGY.LHTYIKDYVTLYVKNIRTIVLEGDQTRP. E ..NLWGFMDWGWPQ.ELEG...RTWGYRLATELLKG...WAELDPTRP E . . PASHLE SAG YYLKMVIAHTRS . . LDPSRP Ei >GHGANHDALYRWIKSVDPSRPVQYEGGGADTTATDI ICPMYARVDEDQP D p ChMad CfMad HGus EcLacZ consensus Nu Nu F I I S S P T . N G A K T T A E G W L S P N P Y . D L N Y G D V H F Y D Y M S . . D C W N W R . T F P K A R . . F V ^ E Y A D G S P Y . S P G F A L D D . . V H P N . . . D P D H G T H H E W E V W N R V D Y S A Y R D D V P . . R . . F C ^ E V T F V S . . . N S N Y A A D K G . . A P Y . . . . V T J V I C L N S Y Y S W Y H . D Y G H L E . L I Q L Q L A F P A V P K W S I K K W L S L P G E T R P L ILC^EJIT A H A M G N S L G G F A K . Y W Q A F R . Q Y P R L Q G G F VWD ChMad CfMad HGus EcLacZ consensus Nu Y G Y Q S W P S F S T L E K V S . . S E E D W S Y E S S F A L H R Q H L . . . F G F Q G P P T W S T L T R A V . . R A D D G G P L T K D . D P T F L L H . Q K A . . . T Q F E N W . . Y K K Y Q K P I . . I Q ^ E } Y G A E T W V D Q S L I K Y D E N G N P W S A Y G G D F G D T P N D R Q F C M N G L V F A D R T P H P A L T E A K H Q Q Q F F Q F ChMad . I NGN SEMLQQIE .... LHF KLPNSA DQLRRFKDTLYLTQVMQAQ C f Mad . EDGN GKLDRGLA.... PHL GVPAG FVDWHWATQLNQAR HGus . IAG FHQ DPPLM FTEEYQKSLLEQ . . EcLacZ RLSGQTIEVTSEYLFRHSDNELLHWMVALDGKPLASGEVPLDVAPQGKQLIELPELPQPE consensus --G H P Q--ChMad CVKTETEFYRRSRNEIVDG . . KGHTMGALYWQLNDIWQAPSWSS L C fMad AVAFAIEHYR SW. . WPRTAGAIVWQLNDCWPVTSWAA I HGus . YHLGLDQKR RKYWGELIWNFADFMTEQSPTR V EcLacZ SAGQLWLTVRWQPNATAWSEAGHISAWQQ||RLAENLSVTLPAASHAIPHLTTSEMDFCI consensus R W Figure 6.3: Multiple sequence alignment of glycosidases from family 2. The region around the acid/base catalyst (A/B) and the catalytic nucleophile (Nu) is shown. The roles of the these residues (boxed) were either determined experimentally, or obtained from 3D structures or from H C A predictions. ChMad: Goat beta-mannosidse; CfMad: C. fimi beta-annosidase; HGus: human beta-glucuronidase; EcLacZ: E. coli (lacZ) beta-galactosidase. Conserved amino acids are highlighted. 135 incubating the enzyme with excess inactivator. 2FManPF was known to inhibit B-mannosidase activity in rats (in vivo) and in rat tissue homogenate (in vitro), but not the purified enzyme (McCarter et al, 1994). According to the proposed inactivation mechanism, one fluoride ion is released per inactivated enzyme molecule (Figure 6.4 A and B; McCarter et al, 1992; Street et al, 1992). Therefore, Man2A inactivation was determined by measuring the changes in fluoride concentration with a fluoride electrode. Inactivator (2FManPF) concentrations ranging from 19.5 uM to 520 uM were tested. The rate constant (k0bs) for each inactivator concentration was obtained by non-linear regression analysis of the experimental values. A n example of time-dependent inactivation is shown in Figure 6.5 A. 8.3 uM Man2A were inactivated in the presence of 520 uM 2FManPF. From the plot of kobs versus inhibitor concentration, the inhibition rate constant ki = 0.556 (+ 0.024) min-1 and the dissociation constant Ki = 0.412 (+0.033) m M were obtained by non-linear regression analysis (according to equation E 6.1: k0bs = (ki x [I])/ (Kj + [I]) ). For the purpose of illustration, k0bs versus inhibitor concentration was replotted in a double reciprocal plot (Figure 6.5 B). Complete inactivation of 8.3 uM Man2A, at inactivator concentrations of 260 uM, 390 uM and 520 uM, produced an average final concentration of released fluoride of 8.03 uM. Consistent with the inactivation scheme (Figure 6.4 A and B), inactivation of Man2A by 2FManpF followed a 1:1 stoichiometry, was mechanism-based, most likely active-site directed, and the inactivation rate was dependent on 2FManpF concentration in a saturable manner. 136 f3 PH CN + C3 s PH CN i < n PH < PQ 137 time (sec) B 0 20 40 60 l / [ 2 F M a n p F ] (mM 1 ) Figure 6.5: Man2A inactivation with 2FManpF. Panel A : example of time-dependent inactivation of 8.3 uM Man2A with 520 uM 2FManpF. kobs were calculated for inactivation reactions with 2FManpF concentrations ranging from 19.5 uM to 520 uM. Panel B: Double reciprocal plot of first order rate constants (kobs) versus 2FManpF concentration. 138 Reactivation of inactivated Man2A Catalytic competence of the 2FMan-enzyme intermediate was demonstrated by incubation of inactivated enzyme (freed of excess inactivator) in either buffer only or in buffer plus a reactivator, such as the uncleavable glucose disaccharide gentiobiose (Glu B-1,6 Glu). From the reactivation reactions (incubated at 37° C) aliquots were removed to measure the regain of P N P M hydrolytic activity due to regeneration of free enzyme. The reactivation in buffer followed a first order process with an apparent rate constant of k react = 0.002 min-1, which corresponded to a half-life of the Man2A-2FMan intermediate of 344 min. The reactivation process was significantly accelerated in the presence of gentiobiose. Reactivation with gentiobiose was gentiobiose concentration-dependent and followed pseudo-first order kinetics. Apparent rate constants (kobs) were determined for gentiobiose concentrations of 10 m M , 15 mM, 40 m M and 80 m M (Figure 6.6 A). From the double reciprocal replot of kobs versus gentiobiose concentration, the reactivation rate constant, k t r a ns = 0.043+0.014 min-1 and the dissociation constant Ktrans = 78+25 m M was estimated (Figure 6.6 B). The increased reactivation rate with gentiobiose suggested that this sugar facilitates turnover of the intermediate, probably, as was shown in other systems, via transglycosylation (Street et al, 1992). This sugar-dependent reactivation rate enhancement was used to screen glycoside libraries to find substrates that can act as good acceptor molecules for the glycosynthase catalyzed transglycosylation (Section 6.2.5). 139 A Figure 6 . 6 : 2FMan-Man2A reactivation. Panel A : Time-dependent reactivation of 2FMan-Man2A with 10 mM, 15 mM, 40 m M and 80 m M gentiobiose. Activity versus time is shown on a semilogaritmic plot. Panel B: Pseudo-first order rate constants (kobs) are reploted on a double reciprocal plot versus gentiobiose concentration. 140 Analysis of the enzyme-inactivator complex by mass spectrometry Native and inactivated Man2A were analyzed by ESMS. The native enzyme had a M W of 94,980 (94,960 predicted), whereas the inactivated enzyme, "labeled" with 2FMan(3F had a M W of 95,139 (95124 predicted). The M W difference between the unlabeled and labeled enzyme of 159 was consistent, within error, with the predicted difference of 164, corresponding to the mass of 2FMan. These results confirmed the 1:1 stoichiometry of the inactivation (Section, and further indicated that the inactivator was covalently linked to the enzyme. These findings are in good agreement with the proposed inactivation mechanism (Figure 6.4). Identification of the labeled active site nucleophile by ESMS Peptide mixtures from pepsin digests of native and 2FMan-labeled Man2A were separated by RP-HPLC and analyzed by ESMS in L C / M S mode (Section 6.1.2 and 2.21). The total ion chromatograms (TIC) showed a large number of peaks, each of which corresponded to one or more peptides (Figure 6.7 A and D). One method to identify the labeled peptide in the peptide mix is neutral loss tandem mass spectrometry. Collision-induced fragmentation of the labeled peptide results in the cleavage of only the label-carboxyl ester bond, which generally represents the weakest linkage within the peptide. Peptides losing a predetermined mass (i.e., the mass of the label), can then be selected for by tandem mass spectrometry (MS/MS). A n attempt to identify the 2FMan-labeled peptide from Man2A by M S / M S in neutral loss mode was unsuccessful. Therefore the mass spectra obtained from the labeled and from the unlabeled peptic digests were analyzed by 141 A B ^ioo%-£ 75%-.5 50%-cu | 25%-"a3 c 1115.5 25 30 time (min) 1244.5 1395.0 1542.0 1 1 1 5 2 0 ' 0 I 1110 1200 1290 1380 m/z (amu) 1470 1560 ^ 1 0 0 % H 1036.5 m 5 3 g 66% .S > 33% D 1244.5 1395.0 1542.0 1110 1200 1290 m/z (amu) 1380 1470 1560 time (min) Figure 6.7: Comparison mapping of labeled and unlabeled Man2A pepsin digests. Total ion chromatograms (TIC) from labeled and unlabeled pepsin digests are shown in Panel A and Panel D, respectively. Panel B: Mass spectrum of labeled digest taken at 28.6 min. Panel C: Mass spectrum of unlabeled digest taken at 28.8 min. h: indicates the labeled and the corresponding unlabeled peptides. 142 685.5 557 500.5 403.5 Y" ions B ions Phe--Gly--Ph^Gh^^ 352.5 635 832 m/z (amu) Figure 6.8: ESMS/MS daughter ion spectrum of the unlabeled peptide (MW 1036). The peptide sequence is indicated with the fragment sizes of the B and the Y " ions shown below and above the sequence, respectively. 143 Y" ions 1036.5 889 832.5 685.5 557 500.5 402.5 Cys--Ser--Glu/]^e/^ 799.5 856.5 952.5 1151.5 B ions 100% 90% -80% • 70% -g 60% .g 50% > 40% •»-> £ 30% 20% 10% x4 —H it-sob, 5. 402.5 353.5 i i x6 685/5 557.0 799.5 889.0 $32.5 VIM 1036.5 1151,5 8B6.fc 952.5 i 1356.0 1520.0 200 400 600 800 1000 1200 1400 m / z (amu) Figure 6.9: ESMS/MS daughter ion spectrum of the labeled peptide (MW 1520). The peptide sequence is indicated with the fragment sizes of the unlabeled B and the Y" ions shown below and above the sequence, respectively. 144 comparative mapping (Figure 6.7 B and C). A peptide with a M W of 1520 (peptide 1520) was found in the labeled, but not in the unlabeled sample (Figure 6.7 B), whereas a peptide with a M W of 1036.5 (peptide 1036.5) was found in the unlabeled, but not in the labeled peptic digest (Figure 6.7 C). The difference in M W of the two peptides was 483.5, which did not correspond to the expected difference from the label of 164 (vide infra). Since only small quantities of peptide 1520 were purified, the more abundant peptide, peptide 1036.5 from the unlabeled peptide mix was analyzed. Amino acid sequence information was obtained by collision-induced fragmentation (CIF), and mass data were collected in daughter ion scan mode (Figure 6.8). The sequence for peptide 1036.5 was identified as 520F G F Q G P P T W 5 2 8 . This sequence did not contain any carboxylic amino acid that might act as catalytic nucleophile. However, the N-terminus of this peptide was immediately C-terminal of the predicted nucleophile E519. By purifying more of peptide 1520 its sequence was identified by CIF as 517C S E F G F Q G P P T W 5 2 8 (Figure 6.8). The M W difference between peptide 1520 and peptide 1036.5 corresponded to the M W of the three additional amino acids found in peptide 1520 plus the M W of the label. Since only carboxylic amino acid residues can act as catalytic nucleophiles (Davies et al, 1998), and only one carboxylic amino acid residue was present in the labeled peptide 1520, glutamate E519 was identified as the catalytic nucleophile. 6.2.2 Conversion of Man2A into a Glycosynthase: Mutation E519A To convert the C. fimi mannosidase Man2A into a glycosynthase, the catalytic nucleophile, E519, was mutated to an alanine. This mutation was not only designed to completely abolish hydrolytic activity, but also to create a pocket within the active site, to 145 allow the use of activated a-sugar donors in transglycosylation reactions (Section; Mackenzie etal, 1998). The scheme for in vitro mutagenesis and generation of plasmid pET28aMadE519A, which encodes the Man2A glycosynthase, is illustrated in Figure 6.10. Expression and purification procedures were the same as used for the wild type mannosidase (Section 2.8). The yields (approx. 300 mg/L) and purity (>96 %) of the glycosynthase were comparable to Man2A WT (Figure 3.24). Extreme caution had to be taken not to contaminate the Man2A E519A mutant with Man2A during purification. Purified Man2A E519A did not hydrolyze P N P M , even after a 5 day incubation period with high enzyme concentration (data not shown). The mass of Man2A E519A, with a predicted M W of 94902, was identified by ESMS as 94918. 6.2.3 Transglycosylation reaction by Man2A WT As has been reviewed in Section, retaining glycosidases are not only able to hydrolyze glycosidic linkages, but they are also able to form new glycosidic linkages. Transglycosylation products, hydrolysis substrates themselves, might never accumulate and be detected because hydrolysis is generally much faster than transglycosylation (Harjunpaa et al, 1995). One way to reduce the rate of transglycosylation product hydrolysis is to use a substrate with a good leaving group at high concentration, and low enzyme concentrations (Gusakov, et al, 1991; Namchuk et al, 1995). P N P M was used as the substrate in transglycosylation reactions with Man2A WT. Man2A WT 0.25 nmol/mL was added to a saturated P N P M solution (7.5 mg/mL), which was buffered with 100 m M 146 Figure 6.10 (following page): Generation of pETMadE519A, encoding the glycosynthase Man2A E519A with a C-terminal H 6 tag. PCR mutagenesis of the catalytic nucleophile encoding codon and the three subcloning steps, that were required for the generation of pETMadE519A are illustrated. Only the restriciton sites that were relevant for cloning are shown. > : MadlO and MadE519A PCR primers <• : Mutation introduced by PCR mutagenesis : manlA or manlA derived plasmid insert : PCR product : manlA PCR template . : D N A fragments used for the cloning steps shown. The D N A fragments were obtained by restriction endonuclease digestions, separation on agarose gels and purification by gel extraction. 147 MadlO 700 bp MadE519A ATG manlA Bsiwi Mini ligate PCR product into pZErO EcoRV site TGA • :GAG(Glu) ^GCC(Ala) Mlu I ligate 1123 bp BsiW I/Notl fragment into P ET28aMad pET28aMadE519A lad (KanR) Not l His tag 148 potassium phosphate at p H 7.0,. The reaction was incubated at 33° C and the formation of transglycosylation products was monitored by thin layer chromatography (TLC). Mono-and oligosaccharides linked to the para-nitrophenyl group were readily detected under U V light on TLC plates, after separation in ethyl acetate : methanol : water (7:2:1). The major product formed, as identified by TLC, was PNPmannobioside (PNPM2). PNPmannotrioside (PNPM3) was synthesized as well, but in quantities too small for further analysis. The reaction (20 mL) was stopped after 105 min and the products were purified on a silica gel column and separated by RP-HPLC. Two regioisomers of P N P M 2 were separated and subsequently analyzed. The results from NMR and ESMS analyses are shown (analysis performed by Lloyd Mackenzie; Section2.23): 4-Nitrophenyl 2,3,4,6-tetra-0-a(xtyl-fi-D-mannopyranosyl-(l,4)-2,3,6-tri-0-acetyl-fi-D-manno-pyranoside. * H N M R (200 M H Z , CDC13): d 8.20 (d, 2H, Jy-,2"- 9.3 Hz, H-3'" Ph), 7.05 (d, 2H,J2"',3<" 9.3 Hz, H-2'" Ph), 5.64 (dd, 1 H , J2,3 3.2 Hz, H2), 5.43 (dd, H , fa 3.4 Hz, H-2'), 5.29 (d, 1 H , /i/21.0 Hz, H-l), 5.25 (dd, 1 H , J3,4 9.3 Hz, H-3), 5.22 (dd, 1 H , fa- 9.6 Hz, H-4'), 5.03 (dd, 1 H , fa 9.8 Hz, H-3') 4.74 (d, 1 H , fa 1.0 Hz, H-l') 4.40 (dd, 1 H , fah 12.0 Hz, H-6a), 4.36-4.23 (m, 2 H , H-6a', H-6b'), 4.13 (dd, 1 H , fa 2.7 Hz, H-6b), 4.04 (dd, 1 H , J4,5 9.3 Hz, H -4), 3.88 (m, 1 H , fa 2.7 Hz, H-5), 3.65 (m, 1 H , fa^ 5.6 Hz, H-5'), 2.14-1.97 (7 s, 21 H , Ac); Ms (ionspray) 464 (M + 1) 4-Nitrophenyl 2,3,4,6-tetra-0-acetyl-/3-D-mannopyranosyl-(l/3)-2,3,6-tri-0-acetyl-j3-D-manno-pyranoside. m NMR (200 M H Z , CDCI3): d 8.19 (d, 2H, Jy-,?- 9.3 Hz, H-3'" Ph), 7.06 (d, 2H,j2"</3<" 9.3 Hz, H-2'" Ph), 5.62 (dd, 1 H , /2,3 3.1 Hz, H2), 5.32-5.26 (m, 2 H , H - l , H-2'), (dd, 1 H , fa- 9.8 Hz, H-4'), 5.14 (dd, 1 H , J4/ 5 9.0 Hz, H-4), 5.08 (dd, 1 H , Jy, 2< 3.4 Hz, Jy, 4< 9.9 Hz, H -3'), 4.75 (d, 1 H , fa 1.0 Hz, H-l') , 4.36-4.22 (m, 3 H , H-6a, H-6a', H-6b'), 4.12 (dd, 1 H , fa 2.7 Hz, K e a 12.3 Hz, H-6b) 4.40 (dd, 1 H , fab 12.0 Hz, H-6a), 4.36-4.23 (m, 2 H , H-6a', H -149 6b'), 4.13 (dd, 1 H , J6h,5 2.7 Hz, H-6b), 4.05 (dd, 1 H , J3,4 9.0 Hz, H-3), 3.88 (m, 1 H , H-5), 3.63 (m, 1 H , H-5'), 2.16-1.97 (7 s, 21 H , Ac); Ms (ionspray) 464 (M + 1) A molar product yield of 6.7 % and 3.5 % was obtained for B-1,4 linked PNPM2 and (3-1,3 linked PNPM2, respectively. By using the glycosynthase Man2A E519A, increased yields in transglycosylation were expected, since no product hydrolysis could occur. 6.2.4 Transglycosylation by glycosynthase Man2A E519A The glycosynthase, Man2A E519A, was not able to form a covalent glycosyl-enzyme intermediate at position 519 because the catalytic nucleophile (E519) had been mutated to an alanine. Therefore, a-mannosyl fluoride, which imitates the glycosyl-enzyme intermediate, was used as the donor sugar in transglycosylation reactions (Figure 6.2). The aryl glycoside, P N P M , was used as the acceptor molecule. Product formation in the glycosynthase-catalyzed transglycosylation reaction was followed over time by removing aliquots and analysing them by TLC. Transglycosylation with Man2A E519A was observed; however, the reaction proceeded very slowly. From the different conditions tested, it was found that a donor acceptor ratio of 10:1 and a high enzyme concentration gave the most transglycosylation. The p H had a considerable effect on product formation (Figure 6.11). For p H values ranging from 4.0 to 8.0 the p H 5.4 produced the highest levels of transglycosylation products (PNPM2 and PNPM3); p H 5.1 and lower caused precipitation of the enzyme. To analyze the products formed in the glycosynthase-catalyzed transglycosylation reaction, a 6.3 mL reaction was set up with the following conditions: 41 pM Man2A E519A, 140 m M K-P p H 5.7,42 m M P N P M and 63 m M a-mannosyl fluoride. The reaction was 150 '* II • 'til PNPM, P N P M , P N P M , 1 2 3 Figure 6.11: p H effect on transglycosylation catalyzed by glycosynthase Man2A E519A. Products were analyzed by thin layer chromatography (TLC) and visualized by UV. The reaction conditions were 150 m M a-mannosyl fluoride, 6 m M PNPmannose, 40 u.M Man2A E519A and 150 m M phosphate buffer at Lane 1: p H 5.4, Lane 2: p H 6.0 and Lane 3: pH 7.3. The reactions were incubated for 45 h at RT. 151 incubated at room temperature and product formation was monitored by TLC. After an incubation period of 5 days the reaction, although not gone to completion, was stopped and the products, P N P M 2 and P N P M 3 were purified and separated by HPLC. B-1,4 and B-1,3 linkages were identified by NMR and ESMS (Section 6.2.3). B-1,4 linked PNPmannobiose was the major product formed in this transglycosylation reaction with a molar yield of 6.5 % (8.4 mg). 1.4 % (1.8 mg) of the substrate were found to be converted into B-1,3 linked PNPmannobiose. The PNPmannotriose products (0.7 %) could also be separated into two different products by HPLC. Due to the low yields however (0.8 mg and 0.3 mg), they could not be analyzed by NMR. These results clearly showed that Man2A E519A is a glycosynthase that can be used for enzymatic synthesis of B-1,4 and B-1,3 mannosidic linkages, linkages that are very difficult to synthesize chemically. The optimization of the glycosynthase reaction was addressed in the experiments presented below. 6.2.5 Screening for good transglycosylation acceptor molecules One way to optimize a glycosynthase catalyzed transglycosylation is to choose good acceptor molecules. Reactivation of inactivated 2FMan-Man2A WT was accelerated with gentiobiose, presumably by transglycosylation of the inactivator to gentiobiose (Section The more efficient a glycoside reactivates the 2FMan-Man2A WT intermediate, the better its qualities to act as an acceptor molecule in transglycosylation by WT enzyme and presumably by the glycosynthase. Nine PNPglycosyl substrates were tested for their ability to reactivate 2FMan-Man2A WT. Inactivated Man2A WT was incubated either in phosphate buffer at p H 7.0 alone or in the presence of 2 m M PNPglycoside. After 90 min and 150 min 152 incubations at RT, aliquots were removed and the regained activity was assayed on PNPM. The fastest reactivation, from the nine substrates tested, was obtained with PNPgentiobiose. PNPcellobiose was also a good reactivator; however, reactivation was slower than with PNPgentiobiose. Significantly slower reactivation was observed with PNP-B-mannoside (PNPM) and PNP-B-galactoside. PNP-N-acteylglucosamine, PNP-a-mannoside, PNP-B-arabinoside, PNP-(3-glucoside and mannose did not show increased reactivation rates compared to reactivation in buffer alone. Four of these aryl-glycosides were tested as acceptor molecules in transglycosylation experiments with the glycosynthase 2Man E519A (Figure 6.12 A). The conditions used for these reactions were: 90 m M a-mannosyl fluoride, 7.5 m M PNPglycoside, 10 nmol Man2A E519A, 180 m M phosphate buffer at p H 5.7. The reaction volume was 55 uL. Transglycosylation yields with PNPgentiobiose as an acceptor molecule were about 10 times higher, and with PNPcellobioside about 4 times higher than with P N P M (estimated from TLC). PNPgalactose was less efficient than PNPM. These results indicated that Man2A preferentially bound disaccharides, or perhaps even longer oligosaccharides, in its aglycon site. Therefore, PNPgentiobiose and PNPmannobiose were compared in their qualities as acceptor in glycosynthase catalyzed transglycosylation reactions (Figure 6.12 B). The glycosynthase Man2A E519A had a stronger preference for PNPgentiobiose than for mannose or derivatives thereof as acceptor. 153 A B Figure 6.12: Acceptor preference of glycosynthase Man2A E519A in transglycosylation reactions. Comparison of PNP- acceptors. Products were analyzed by thin layer chromatography (TLC). The substrates tested were Lane 1: PNPgentiobiose (PNPG); Lane 2: PNPmannose; (PNPM) Lane 3: PNPcellobiose (PNPC); Lane 4: PNPgalactose (PNPG). Panel B: Lane 5: PNPmannobiose (PNPM2) and Lane 6: PNPgentiobiose(PNPG). Substrates and products are indicated by arrows and labeled. The reactions in Panel A were analysed after 2 days, the reactions in Panel B after 6 days incubation at RT. 154 6.3 Discussion 6.3.1 Man2A inactivation The proposed mechanism for Man2A inactivation by 2-deoxy-2-fluoro-B-D-mannosyl fluoride (2FManpF) is shown in Figure 6.4. In order to trap the 2FMan-enzyme intermediate, the rate of formation (inactivation rate; ki) must be faster than the rate of reactivation (kreact). Under the conditions of k react approaching zero, ki is equal to the glycosylation rate and the dissociation constant K i , is equal to k-i/k+i (McCarter et al, 1992). The fluorine substitution at the C-2 position in 2FManpF, reduces the glycosylation and deglycosylation rate, whereas the good leaving group at the C - l position (fluorine) specifically increases the glycosylation rate (McCarter et al, 1994). Such a situation was clearly obtained in the inactivation of Man2A with 2FManPF. Inactivation, which was time-dependent, followed pseudo-first order kinetics and showed a 1:1 stoichiometry. The detection of a glycosyl-enzyme intermediate by ESMS confirmed not only the 1:1 stoichiometry, but was also a very strong indication of the covalent nature of the glycosyl-enzyme linkage. The catalytic competence of the 2FMan-Man2A intermediate was demonstrated by the analysis of the reactivation rates, which could be accelerated by transglycosylation to acceptor sugars (i.e., gentiobiose). The rate of reactivation of the 2FMan-Man2A intermediate was 21 fold faster upon addition of gentiobiose (ktrans ) compared to the spontaneous reactivation rate (kreact). A 12.6 fold rate increase was observed for the reactivation of the Candida albicans 2FGlc-exo-B-l,3-glucanase intermediate with benzyl-thio-B-D-glucopyranoside (Mackenzie et al, 1997). Glucosyl benzene was used to accelerate 155 the reactivation rate of the Agrobacterium sp 2FGlu-B-glucosidase intermediate 530 fold (Street et al, 1992) and even higher turnover rates were found for this system with p-nitrophenyl B-glucoside. For all three systems the dissociation constants for the reactivator were very similar (78+25 mM, 59+3.1 m M and 56+9.5 mM, respectively). These differences in reactivation versus transglycosylation rates illustrate the affinities of the aglycon site for the reactivator sugars, and the importance of reactivator binding in reducing the activation energy of transglycosylation (Street et al, 1992), which can be expressed as AAG = RTln(kreact/ktrans). Reactivation of 2FGlu-Abg with glucosyl benzene caused a larger change (AAG = - 16.1 kj/mol) than could be obtained from reactivation of 2FMan-Man2A with gentiobiose (AAG = - 7.9 kj/mol). 6.3.2 Protection of pepsin cleavage site by 2FMan glycosylation To identify the catalytic nucleophile in Man2A, labeled and unlabeled enzyme samples were digested with pepsin. The peptide mixes were analyzed by ESMS and by comparison mapping ( Peptide 1520 (MW) was unique for the labeled sample, and peptide 1036.5 (MW) was unique for the unlabeled peptide mix. Peptide 1520 differed from peptide 1036.5 by having three additional amino acids at the N-terminus (517CSE519), with the glutamate at position 519 identified as the catalytic nucleophile. This cleavage pattern indicated that glycosylation of E519 with 2FMan prevented peptic attack between the two amino acid residues E519 and F520, which in the unlabeled sample was cleaved to produce peptide 1036.5. Glycosylation has been shown to protect proteins against proteolysis, for example in secreted C. fimi cellulases (Gilkes et al, 1988). 156 6.3.3 Prediction of catalytic residues by hydrophobic cluster analysis (HCA) The catalytic nucleophile in bovine B-mannosidase was predicted to be E554 by comparison of H C A plots from several lysosomal glycosidases of the clan GH-A (Durand et al, 1997). As seen from multiple sequence alignments, the glutamate corresponding to the bovine catalytic nucleophile is conserved in all family 2 mannosidases and corresponds to residue E519 in the C. fimi mannosidase, Man2A (Figure 3.19). The amino acid E519 was experimentally identified as the catalytic nucleophile (Section 6.2.1), which supports the validity of H C A predictions. 6.3.4 Transglycosylation by glycosynthase Man2A E519A Stereo- and regiospecificity of transglycosylation Detection of transglycosylation by Man2A WT was a good indicatior for a successful approach to transglycosylation by the glycosynthase Man2A E519A. This study is the second report of glycosynthase-catalyzed oligosaccharide synthesis using a-glycosyl fluoride as donor sugars. The Agrobacterium sp. B-glucosidase/galactosidase (Abg) E358A mutant, the first glycosynthase reported, uses a-glucosyl fluoride and a-galactosyl fluoride as donor sugars, whereas a wide range of aryl- and alkyl glycosides can be used as acceptor molecules. The products formed by Abg glycosynthase are almost exclusively (3-1,4 linked, with the exception of glycosyltransfer to B-xylosides, which results in (3-1,3 linkages. The stereo- and regiospecificity of Man2A E519A was analyzed for the P N P M 2 products. Most of the products were (3-1,4 linked although, about 20 % of the products were found to be (3-157 1,3 linked. This ratio of (3-1,4 to (3-1,3 linked product (4.6:1) is considerably higher than the 2:1 ratio found in transglycosylation products formed by Man2A WT. In this study only a-mannosyl fluoride was used as the donor sugar. Since Man2A can hydrolyze PNPgalactoside, a-galactosyl fluoride could possibly be used as an alternative donor sugar in Man2A catalyzed transglycosylation reactions (Section 3.13). The first application of the glycosynthase Man2A E519A was the synthesis of a substrate (PNPM 2) for kinetic analysis of the C. fimi mannanase Man26A (Section 3.2. 9). The effect of p H on Man2A E519A transglycosylation The replacement of the nucleophilic carboxylate with a neutral side chain, or the formation of the glycosyl-enzyme intermediate both reduce the pKa of the acid/base catalyst in a (3-glycosidase, thereby setting it up to function optimally as the general base catalyst for the deglycosylation step (Mcintosh et al, 1996). The p H optimum for the Man2A E519A-catalyzed transglycosylation was p H 5.4, as estimated by product yields from TLC, whereas the optimum for P N P M hydrolysis by the WT was p H 7.0 (Section 3.11). These results were unexpected, and could not be explained by the reduction of the pK a of the acid/base catalyst caused by the mutation of the catalytic nucleophile to alanine (Mcintosh et al, 1996). Other factors resulting from changes in enzyme structure that effect substrate binding might be responsible. Alternatively the lower p H might reduce substrate inhibition by reducing the formation of the inhibitory SES complex (Section 3.2.15). 158 Acceptor preference of Man2A E519A The reactivation of the 2FMan-Man2A intermediate, as demonstrated with gentiobiose (Section, can be used as a fast and easy way to screen for suitable transglycosylation acceptors. In this study, nine glycosides were tested. PNPgentiobiose was the best and PNPcellobiose the second best reactivator. Four of these reactivators were also tested as acceptor molecules for glycosynthase catalyzed-transglycosylation. The glycosynthase results agreed with the reactivation results, demonstrating the usefulness of the screening procedure. The best transglycosylation product yields were obtained with PNPgentiobiose, and somewhat lower yields with PNPcellobiose. Both substrates were significantly better acceptors than PNPmannose. In the case of PNPgentiobiose, the reaction went to almost completion after an incubation period of 6 days (Figure 6.12 A). The comparison of PNPmannobiose and PNPgentiobiose as acceptors in Man2A E519A catalyzed transglycosylation reactions demonstrated that PNPgentiobiose is also a better acceptor than PNPmannobiose, which indicated that PNPgentiobiose is not better solely because it is an aryl-disaccharide (Figure 6.12 B). Gentiobiose has previously been reported to be a good aceptor in tranglycosylation reactions with Fusarium oxysporium B-glucosidase (Christakopoulos et al, 1994). In light of Man2A being an exo-mannanase, preferential binding of mannose or manno-obgosaccharide in its aglycon site would have been expected. A 3D structure of Man2A with a substrate bound in its aglycon site (e.g. gentiobiose) could give some information about the interactions that are involved in binding of the substrate in the aglycon site. For now, the reasons for preferential 159 PNPgentiobiose binding are only speculative. One speculation is that the enzyme has a relatively large aglycon site, which can accommodate non-linear 1,6 linked disaccharides, such as galactose a-1,6 mannose disaccharides, as found in galactomannan, or glucose (3-1,6 glucose as in gentiobiose. Another hypothesis is based on substrate inhibition as was demonstrated for Man2A WT with P N P M (Section 3.2.15). It is assumed that Man2A E519A is also able to form a substrate inhibition complex (SES) with P N P M , but not with PNPgentiobiose and PNPcellobiose. Therefore the latter two substrates appear to be better acceptor molecules. From these transglycosylation studies it was also apparent, that a-mannosyl fluoride did not cause strong substrate inhibition in Man2A E519A. 160 7 Genomic Map of Cellulomonas fimi 7.1 Introduction 7.1.1 Genetic organization of cellulase and hemicellulase systems In many cellulolytic microorganisms, the genes encoding cellulases and hemicellulases are scattered on the genomes with little or no linkage between individual genes (Tomme et al, 1996). Most of the genes encoding components of the Clostridium thermocellum cellulosome are scattered on the genome and are transcribed as monocistronic mRNAs (Beguin et al, 1996; Guglielmi and Beguin, 1998). A notable exception in C. thermocellum is the cluster of structural cellulosome component encoding genes cipA-olpB-ORY2-olpA. In this cluster cipA and olpB are in one operon and ORF2 and olpA in another (Fujino et al, 1993a). Several other gene pairs were found to be closely spaced; e.g., HcA and celC were separated by 4 kbp. In the fungus T. reesei, mapping of the major cellulase and xylanase encoding genes cbhl, cbh2, egll, egl2, bgll, xyll and xyl2 revealed only two genes in proximity to each other. The genes cbh2 and egl2 were located on the same 47 kbp Not I fragment of chromosome I (Carter etal, 1992). Most of the genes encoding enzymes and structural components of the cellulosome from the anaerobe Clostridium cellulolyticum are organized in an approximately 20 kbp long cluster. This is the biggest cluster reported for cellulolytic systems. It is comprised of the genes cipC, celF, celC, celG, celE, cipX, celH and eel] (Gal L., 1997; Belaich et al, 1997). Two genes encoding additional cellulases, CelA and CelD, however, are not closely linked to this cluster. A cluster, almost identical to the C. cellulolyticum cluster, was identified in Clostridium josui, suggesting that this organism might be closely related to C. cellulolyticum 161 (Fujino et al, 1993b). A similar but smaller cluster was detected in the bacterium Caldocellulosiruptor saccarolyticus. This cluster, on a 12 kbp D N A fragment, contains at least the celA, celB, manA and celC genes, all of which encode bifunctional and multidomain enzymes (Borges et al, 1993). 7.1.2 Gene cluster in Cellulomonas fimi In C. fimi, two genes are closely linked: cbhA, the gene encoding cellobiohydrolase A , is upstream of cenD, which encodes endoglucanase CenD. The two genes are separated by only 129 bp. Putative transcriptional termination sequences and putative promoter sequences were found between the two open reading frames, indicating that they are not part of an operon (Meinke et al, 1994). This close linkage suggested a possible linkage of more than only two cellulase/hemicellulase encoding genes in C. fimi. 7.1.3 Objectives The aim of this project was to study the distribution of genes encoding cellulases and hemicellulases in C. fimi, focusing on a possible linkage between man26A and man!A. A physical and genetic map was established for the C. fimi genome using the techniques of pulsed field gel electrophoresis and Southern blotting. 162 7.2 Results 7.2.1 Mapping of the genes man26A and manlA on the C. fimi genome To determine the location of the two genes involved in mannan degradation, man26A and manlA, a physical and genetic map of the C. fimi genome was constructed. Genomic D N A was embedded in agarose blocks and digested with restriction endonucleases (Section 2.17.1), that cleave the genome producing D N A fragments ranging from 100 kbp to 1400 kbp in size. The large D N A fragments were separated in an agarose gel by pulsed field gel electrophoresis (PFGE) (Birren et al, 1993; Section 2.17.2) Restriction endonucleases with AT-rich recognition sequences were tested for cleavage of the GC-rich C. fimi D N A (71.5% G+C (Yamada et al, 1970)). Of the more than twenty enzymes tested, Mun I, Xba I, Nde I, Hpa I, EcoR I, Hind III and Nsil proved to be good candidates. These enzymes produced 7 to 15 D N A fragments, 20 kbp to 1400 kbp in size that were separable on a 1.2 % agarose gel by PFGE. To yield optimal separation of all fragments within the range of 20 kbp to 1400 kbp, the gels were run for 36 h switching the orientation of the electric field every 60 s for the first 9 h, then every 80 s, 100 s and 120 s for 9 h each. The voltage was 160 V and the temperature was controled at 14° C. The D N A bands were visualized after separation with ethidium bromide under UV light (Figure 7.1 A). As molecular size standards, chromosomes from the yeast Saccharomyces cerevisiae and Hind III digested X D N A were used. The sizes of D N A fragments were estimated by comparing their mobilities to the mobilities of the size standards. Fragments released by Hpa I, Mun I, Hind III and Nsi I are shown schematically in Figures 7.2 and 7.3. Addition of fragment lengths from Hpa I, Mun I and Hind III digests resulted in genome sizes of 4120 kbp, 163 A B 1 2 3 4 5 6 7 8 9 10 11 1600kb 1125kb 1020kb 945kb 825kb 785kb 750kb 680kb 610kb 565kb 450kb 365kb 285kb 225kb 23.1kb 2 3 4 5 6 7 8 9 10 11 * i n Figure 7.1: Panel A : Separation of restriction fragments of C. fimi genomic D N A by pulsed field gel electrophoresis (PFGE). The following parameters were used for PFGE: switch intervals of 60 s, 80 s, 100 s and 120 s for 9 h each at 14° C, 1.2 % agarose and 1 x TAE running buffer. The sizes of the size standards are indicated: Saccharomyces cerevisiae cromosomes (Lane 1) and X-Hind III (Lane 2). Panel B: Southern blot of gel shown in Panel A probed with fluorescently labeled, 711 bp Pvu I cenD fragment. The C. fimi genomic D N A was digested with the following restriction endonucleases: Mun I (Lane 3 and Lane 4), Xba I (Lane 5), Nde I (Lane 6), Hpa I (Lane 7), EcoR I (Lane 8), Hind III (Lane 9) and Nsi I (Lane 10). Undigested genomic C.fimi D N A is shown in Lane 11. 164 Hpall cenC/cex/man26A 1,400 kb Hpall Hpa 13 Hpal5 Hpal6 cenD/cbhA cbhB/cenA Hpa 14 cenB/xynD manlA 1,050 kb w T i Mim I I 960 kb Mun 12 Mun 13 Mun 14 Mun 15 Mun 16 MwnI7 150 kb M m " 1 8 530 kb 30 kb Muni 9 ManlA xynD cenD/cbhA/cencC cbhB/cenA cex man26A cenB 1,000 kb 670 kb 610 kb 550 kb 370 kb 350 kb 270 kb 150 kb 30 kb Figure 7.2: Schematic representation of the D N A fragments produced by Hpa I and Mun I restriction endonuclease digestion of C.fimi genomic DNA. The fragments are represented as boxes with their names and sizes indicated. The cellulase and hemi-cellulase encoding genes were mapped to these fragments. The genome size calculated from the sum of Hpa I fragments is 4,120 kbp and 4,000 kbp for the Mun I fragments, respectively. 165 Hind III1 cenB/xynD 920 kb N s i l l l cenD/cbhA 950 kb Hind III 2 Hind III 3 Hind III 4 Hind III 5 Hind III 6 Hind III 7 Hind III 8 cenC/cex cenD/cbhA cbhB cenA 760 kb 660 kb N s i I 2 560 kb 370 kb 290 kb 230 kb 95 kb NsiI3 cex/man26A cenB/xynD 640 kb 400 kb 270 kb 200 kb -20kb Figure 7.3: Schematic representation of D N A fragments produced by Hind III and Nsi I restriction endonuclease digestion of C. fimi genomic DNA. The fragments are re-presented as boxes, with their names and sizes indicated. The cellulase and hemi-cellulase encoding genes were mapped. From the sum of Hind III fragments a genome size of 3,885 kbp was calculated . The D N A fragments produced by Nsi I digestion, 200-20 kbp in size, could not be separated under the conditions used, and are therefore represented as one large box. The relative positions of the genes within this region are indicated. 166 4000 kbp and 3885 kbp, respectively. The average C. fimi genome size was therefore estimated to be approximately 4,000 kbp. C. fimi D N A treated with Xba I resulted only in partial digestion. Therefore, the results obtained from Xba I digests were not considered for further analysis. Undigested C. fimi D N A comigrated as a single band with the 1,125 kbp D N A standard on PFG. This suggested that C. fimi has a single, circular chromosome, with the electrophoretic mobility of supercoiled DNA. The PFGE gel shown in Figure 7.1 A was subjected to Southern blot analysis. Fluorescently labeled probes, i.e., the 711bp Pvu I cenD fragment, were used to locate the cellulase and hemicellulase genes in each set of restriction fragments (Figure 7.1 B) (Section 2.17.3). Good PFGE and Southern blot results were obtained with the C. fimi genes man26A , manlA, cenA, cenB, cenC, cenD, cbhB, cex and xynD as genetic markers from Mun I, Hpa I and Hind III digests. These results are summarized in the schematic restriction patterns shown in Figure 7.2 and 7.3 and were used to create a genetic map of the C. fimi chromosome (Figure 7.4). Results obtained by PFGE and Southern blot analyses of the large D N A fragments from Nsi I and Nde I digests were included in the map as well. Several D N A fragments produced by Nsi I that ranged from 200 kb to 20 kb could not be separated clearly by PFGE. In the schematic representation in Figure 7.3, all of these bands are represented by only one large box. The map of the C. fimi chromosome, shown in Figure 7.4, summarizes all the results obtained by PFGE and Southern blot. Each circle on the map represents the full length of the chromosome and on each circle the results from one restriction endonucleolytic digestion are summarized. Each box represents one restriction fragment and is drawn in its 167 relative size, and gaps between boxes indicate boundaries. For Nsi I and Nde I digests only the fragments that contained one of the used a genetic markers could be mapped. The physical linkages of the three largest Mun I fragments were analyzed. After separation of Mun I digested C. fimi D N A by PFGE, the three D N A fragments Mun II, Mun 12 and Mun 13 (Figure 7.2) were each excised from the gel and purified (QiaexII). After fluorescent labeling they were used as probes for Southern blots. The Mun 11 fragment hybridized to the Hpa I 2, Hpa 13 and Hpa I 5 fragments and to the Nde I 1 fragment. Hybridization to fragments from other digests resulted in ambigous signals. The Mun 12 fragment could only be shown to hybridize to the Hind III1 fragment and the Mun I 3 fragment was shown to hybridize to Hind III 2, Hind III 4, Hpa 11, Hpa I 2 and to Nde I 2 fragments. These results were in good agreement with the genomic map as outlined in Figure 7.4. 168 Figure 7.4: Physical and genetic map of the Cellulomonas fimi genome. Each circle represents D N A fragments produced by one restriction endonuclease. One D N A fragment is represented by one box, with the length of the box being pro-portional to the length of the D N A fragment. The majority of the boxes are labeled with the name and the length of the fragments. Only the fragments that could be physically or genetically mapped are included. The relative positions of the C. fimi cellulase and hemicellulase encoding genes are indicated by arrows. 169 7.3 Discussion 7.3.1 Genome size and geometry The C. fimi genome was determined to be approximately 4,000 kbp in size, comparable to the E. coli genome, which is 4,639 kbp (Blattner et al, 1997). Even though sizes of bacterial genomes vary from 600 kbp for Mycoplasma genitalium to 12,800 kbp for Calothrix strains, the C. fimi genome is similar in size to the average Gram positive genome. The average size of Gram positive genomes was calculated to be 3,115 kbp (Trevor, 1996). Undigested genomic C. fimi D N A comigrated with the 1,125 kbp size standard. This suggested that C. fimi has only one circular chromosome. The majority of the bacterial genomes studied were found to be circular. However, linear chromosomes have been described, e.g. for Borelia burgdorferi, which has a linear chromosome of approximately 1,000 kbp and a number of linear and circular plasmids (Davidson et ah, 1992), and for Streptomyces lividans and Streptomyces griseus both of which have one linear chromosome about 8,000 kbp in size (Lezhava et al, 1995). Plasmids, which might have been useful tools for genetic manipulations, were not detected in C. fimi D N A preparations. 7.3.2 Genetic and physical mapping of the C. fimi genome PFGE and Southern blots were used to establish a genetic and physical map of the C. fimi genome. The genome map clearly shows that neither the genes encoding the mannan degrading system, man26A and manlA, nor the genes encoding the cellulose or xylan degrading systems are organized in clusters. The only exception is the close linkage of cbhA and cenD (Meinke et al, 1994). As in C. fimi, most of the genes known to encode cellulases 170 and xylanases are scattered on the genome of P. fluorescens subsp. cellulosa. However, the xynB and xynC genes, encoding a xylanase and an arabinofuranosidase, respectively, are in the same orientation and only 148 base pairs apart (Kellet et al, 1990). These two examples of small clusters might have arisen from gene duplications. Even though the C. fimi genes encoding hemicellulases and cellulases are scattered all over the chromosome some of them might still be part of one or more regulons, and controlled by the same regulatory element(s). 171 8 References AH, B.R.S., Zhou, L., Graves, F.M., Freedman, R.B., and Black, G.W. (1995). Cellulases and hemicellulases of the anaerobic fungus Pyromyces constitute a multiprotein cellulose-binding complex and are encoded by multigene families. FEMS Microbiol. Lett. 125:15-22. Alkhayat A H , Kraemer SA, Leipprandt JR, Macek M , Kleijer WJ, and Friderici K H (1998). Human beta-mannosidase cDNA characterization and first identification of a mutation associated with human beta-mannosidosis. Hum. Mol. Genet. 7:75-83. Altschul, S.F., Madden, T.L., Schaffer, A.A. , Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402. Arcand, N . , Kluepfl, D., Paradis, F.W., Morosoli, R., and Sharek, F. (1993). B-mannanase of Streptomyces lividans 66: cloning and D N A sequenceing of the manA gene and charaterization of the enzyme. Biochem. J. 290:857-863. Atalla, R.H., Hackney, J.M., Uhlin, I., and Thompson, N.S. (1993) Hemicelluloses as structure regulators in the aggregation of native cellulose. Int J Biol Macromol 15(2):109-112 Atkinson, T. and Smith, M . (1984). Solid phase synthesis of oligodeoxyribonucleotides by the phosphite triester method. In: Oligonucleotide synthesis: a practical approach. N.J. Gait (ed.) IRL Press, Oxford, pp. 35-81. Bauer, M.W., Bylina, E.J., Swanson, R.V., and Kelly, R.M. (1996). Comparison of a B-glucosidase and a B-mannosidase from the hyperthermophilic archaeon Pyrococcus furiosus. J. Biol. Chem. 271(39):23749-23755. Bayer, E.A., Morag, E., and Lamed, R. (1994). The cellulosome - A treasure trove for biotechnology. Trends Biotechnol. 12:378-386. Bayer, E.A., Morag, E., Shoham, Y., Tormo, J., and Lamed, R. (1996). In: Bacterial Adhesion: Molecular and Ecological Diversity, pp 155-182. Wiley-Liss, Inc. Beguin, P., and Aubert, J.-P. (1994). The bioloical degradation of cellulose. FEMS Microbiol. Rev. 13:25-58. Beguin, P., and Lemaire, M . (1996). The cellulosome: A n exocellular, multiprotein complex specialized in cellulose degradation. Crit. Rev. Biochem. Mol. Biol. 31(3):201-236. Belaich, J.-P., Tardif, C , Belaich, A., and Gaudin, C. (1997). The cellulolytic system of Clostridium cellulolyticum. J. Biotech. 57:3-14. Beveridge, T.J., Pouwels, P.H., Sara, M . , Kotiranta, A. , Lounatma, K., Kari, K., Keruso, E., Haapasolo, M . , Egelseer, E.M., Schocher, I., Sleytr, U.B., Morelli, L., Callegari, M.-L., Nomelli, J.F., Bingle, W.H. , Smit, J., Leibovitz, E., Lemaire, M . , Miras, I., Salamitou, S., Beguin, P., Ohayon, H. , Gounon, P., Matuschek, M . , Sahm, K., Bahl, H. , Grogono-Thomas, 172 R., Dworkin, J., Blaser, M.J., Woodland, R.M., Newell, D.G., Kessel, M . , and Koval, S.F. (1997). Function of S-layers. FEMS Microbiol. Rev. 20:99-149. Birren, B., and Lai, E. (1993). Pulsed field gel electrophoresis. A practical guide.Academic Press, Inc., San Diego. Black, G.W., Hazlewood, G.P., Millward-Saddler, S.J., Laurie, J.L, and Gilbert, H.J. (1995). A modular xylanase containing a novel non-catalytic xylan-specific binding domain. Biochem. J. 307:191-195. Blattner, F.R., Plunkett, G., Bloch C.A., Perna, N.T., Burland, V., Riley, M . , Collado-Vides, J., Glasner, J.D., Rode, C.K., Mayhew, G.F., Gregor, J., Davis, N.W., Kirkpatrick, H.A., Goeden M.A. , Rose, D.J., Mau, B., and Shao, Y . (1997) The complete genome sequence of Escherichia col i K-12. Science. 277(5331) :1453-74. Bolam, D.N., Hughes, N . , Virden, R., Lakey, J.H., Hazlewood, G.P., Henrissat, B., Braithwaite, K.L., and Gilbert, H.J. (1996). Mannanase A from Pseudomonas fluorescens ssp. cellulosa is a retaining glycosyl hydrolase in which E212 and E320 are the putative catalytic residues. Biochemistry 35:16195-16204. Borges, K .M. , and Berquist, P.L. (1993). Pulsed-Field gel electrophoresis study of the genome of Caldocellum saccharolyticum. Curr. Microbiol. 27:15-19. Braithwaite, K.L., Black, G.W., Hazlewood, G.P., A l i , B.R., and Gilbert, H.J. (1995). A non-modular endo-beta-l,4-mannanase from Pseudomonas fluorescens subspecies cellulosa. BiochemJ. 305:1005-1010. Bray, M.R., Johnson, P.E., Gilkes, N.R., Mcintosh, L.P., Kilburn, D.G., and Warren, R.A.J (1996). Probing the role of tryptophan residues in a cellulose-binding domain by chemical modification. Prot. Sci. 5:2311-2318. Brun, E., Moriaud, F., Gnas, P., Blackledge, M.J., Barras, F., and Marion, D. (1997). Solution structure of the cellulose-binding domain of the endoglucanase Z secreted by Enuinia chrysanthemi. Biochemistry 36:16074-16086. Carter, G.L., Allison, D., Rey, M.W., and Dunn-Coleman, N.S. (1992). Chromosomal and genetic analysis of the electrophoretic karyotype of Trichoderma reesei: mapping of the cellulase and xylanase genes. Mol. Microbiol. 6(15):2167-2174. Chaplin, M.F. (1986). Chaplin, M.F., Kennedy, J.F. (eds), In: Carbohydrate analysis: a practical approach. IRL Press, Oxford, pp 1-36. Chen, H , Leipprandt, J.R., Traviss, C.E., Sopher, B.L., Jones, M.Z., Cavanagh, K.T., and Friderici, K .H. (1995). Molecular cloning and characterization of bovine beta-mannosidase. J.BioLChem. 270:3841-3848. Clarke, J.H., Davidson, K., Gilbert, H.J., Fontes, C.M.G.A., and Hazlewood, G.P. (1996). A modular xylanase from mesophilic Cellulomomas fimi contains the same cellulose-bining and 173 thermostabilizing domains as xylanases from thermophilic bacteria. FEMS Micobiol. Lett. 139:27-35. Cornish-Bowden, A. (1979). Fundamentals of enzyme kinetics. Butterworth & Co Ltd. Coulombel, C , Clermont, S., Foglietti, M.-J., Percheron, F. (1981). Transglycosylation reactions catalyzed by two B-mannanases. Biochem. J. 195:333-335. Coutino, J.B., Gilkes, N.R., Warren, R.A.J., Kilburn, D.G., and Miller, R.C. (1992). The binding of Cellulomonas fimi endoglucanase C (CenC) to cellulose and Sephadex is mediated by the N-terminal repeats. Mol. Microbiol. 6(9):1243-1252. Creagh, A.L., Koska, J., Johnson, P.E., Tomme, P., Joshi, M.D., Mcintosh, L.P., Kilburn, D.G., Haynes, C A . (1998). Stability and oligosaccharide binding of the N l cellulose-binding domain of Cellulomonas fimi endoglucanase CenC. Biochemistry. 37(10): 3529-3537 Creagh, A.L., Ong, E., Jervis, E., Kilburn, D . G , and Haynes, C A . (1996). Binding of the cellulose-binding domain of exoglucanase Cex from Cellulomonas fimi to insoluble microcrystalline cellulose is entropically driven. Proc. Natl. Acad. Sci. USA 93:12229-12234. Damude, H.G., Withers, S.G., Kilburn, D.G., Miller, R.C. and Warren, R.A.J. (1995). Site-directed mutation of the putative catalytic residues of endoglucanase CenA from Cellulomonas fimi. Biochemistry. 34(7): 2220-2224. Davies, G., and Henrissat, B. (1995). Structure and mechanisms of glycosyl hydrolases. Structure 3(9): 853-859. Davies, G., Sinnott, M.L. and Withers, S.G. (1998) Glycosyl transfer. Comprehensive biological catalysis. A Mechanistic Reference. Volume I. Academic Press. San Diego. Del Sal, G, Manfioletti, G, and Schneider, C. (1989). The CTAB-DNA precipitation method: a common mini-scale preparation of template D N A from phagemids, phages or plasmids suitable for sequencing. Biotechniques. 7(5): 514-520 Dower, W.J. (1987). Mol. Biol. Rep. (Bio-Rad Laboratories)!, 5. Durand, P., Lehn, P., Callebaut, I., Fabrega, S., Henrissat, B, and Mornon, J.-P. (1997). Active-site motifs of lysosomal acid hydrolases: invariant features of clan GH-A glycosyl hydrolases deduced from hydrophobic cluster analysis. Glycobiol. 7(2):277-284. Edelman, G.M. (1983). Cell adhesion molecules. Science 219:450. Edge, C.J., Rademacher, T.W., Wormald, M.R., Parekh, R.B., Butters, T.D., Wing, D.R. and Dwerk R.A. (1992). Fast sequencing of oligosaccharides: the reagent-array analysis method. Proc. Natl. Acad. Sci. USA 89:6338-6342. Egelseer, E.M., Leitner, K., Jarosch, M . , Hotzy, C , Zayni, S., Sleytr, U.B., and Sara, M . (1998). The S-layer proteins of two Bacillus stearothermophilus wild-type strains are bound via their 174 N-terminal region to a secondary cell wall polymer of identical chemical composition. J. Bacterid. 180(6):1488-1495. Estrada, P., Mata, I., Dominguez, J.M., Castillon, M.P., and Acebal, C. (1990). Kinetic mechanism of B-glucosidase from Trichoderma reesei QM 9414. Biochim. Biophys. Acta 1033:298-304. Fan, J.-Q., Takegawa, K., Iwahara, S., Kondo, A., Kato, L, Abeygunawardana, C , and Lee, Y.C. (1995). Enhanced transglycosylation activity of Arthrobacter protophormiae endo-B-N-acetylglucosaminidase in media containing organic solvents. J. Biol. Chem. 270:17723-17729. Feinberg, A.P., and Vogelstein, B. (1983). A technique for radiolabeling D N A restriction endonuclease fragments to high specific activity. Anal Biochem. 132: 6-13. Franco. T.T., Rodrigues, N.R., Serra, G.E., Panegassi, V.R., and Buckeridge, M.S. (1996). Characterization of storage cell wall polysaccharides from Brazilian legume seeds and the formation or aqueous two-phase systems. J. Chromatogr. B. Biomed. Appl. 680(1-2):255-261. Fujino, T., Beguin, P., and Aubert, J.-P. (1993a). Organisation of a Clostridium thermocellum gene cluster encoding the cellulosomal scaffolding protein CipA and a protein possibly involved in the attachement of the cellulosome to the cell surface. J. Bacteriol. 175:1891-1899. Fujino, T., Karita, S., and Ohmiya, K. (1993b). Nucleotide sequence of the celB gene encoding endo-l,4,-beta-glucanase-2, ORF1 and ORF2 forming a putative cellulase gene cluster of Clostridium josui. J. Ferment. Bioeng. 76:243-250. Gaboriaud, C , Bissery, V., Benchetrit, T., and Mornn, J.-P. (1987). Hydrophobic cluster analysis. A n efficient new way to analyse amino acid sequences. FEBS Lett. 224:149-155. Gal, L. (1997). Etude du cellulosome de Clostridium cellulolyticum et de l'un de ses composants: la cellulase CelG. PhD thesis. Universite de Provence Aix-Marseille I. Gebler, J.C, Aebersold, R., and Withers, S.G. (1992). Glu-537, not Glu-461, is the nucleophile in the active site of (lacZ) beta-galactosidase from Escherichia coli. J. Biol. Chem. 267:11126-11130. Gehrmann, M.C., Opper, M . , Sedlacek, H.H. , Bosslet, K., and Czech, J. (1994). Biochemical properties of recombinant human B-glucuronidase synthesized in baby hamster kidney cells. Biochem. J. 301:821-828. Gherardini, F., Babcock, M . , and Salyers, A. (1985). Purification and characterization of two a-galactosidases associated with the catabolism of guar gum and other a-galactosides by Bacteroides ovatus. J. Bacteriol. 161:500-506. Gilbert, H.J. and Hazelwood, G.P. (1993). Bacterial cellulases and xylanases. J. Gen. Microbiol. 139:187-194. 175 Gilkes, N.R., Chanzy, H. , Kilburn, D .G , Miller, R.C. Jr., Sugiyama, J., Warren, R.A.J., and Henrissat, B. (1993). Visualization of the adsorption of a bacterial endo-beta-l,4-glucanase and its isolated cellulose-binding domain to crystalline cellulose. Internat. J. Biol. Macromol. 15:347 Gilkes, N.R., Henrissat, B., Kilburn, D.G., Miller, R .C, Warren, R.A.J. (1991). Domains in microbial (3-1,4-glycanases: Sequence conservation, function, and enzyme families. Microbiol. Rev. 55:303-315. Gilkes, N.R., Jervis, E., Henrissat, B., Tekant, B., Miller, R.C. Jr., Warrren, R.A.J., and Kilburn, D.G. (1992). The Adsorption of a bacterial cellulase and its two isolated domains to crystaliline cellulose. J. Biol. Chem. 267:6743-6749. Gilkes, N.R., Kilburn, D.G., Miller, R.C. Jr., and Warren, R.A.J. (1989) Structural and functional analysis of a bacterial cellulase by proteolysis. J. Biol. Chem. 264:17802-17808. Gilkes, N.R., Kwan, E., Kilburn, D.G., Miller, R .C, and Warren, R.A.J. (1997). Attack of carboxymethyl cellulose at opposite ends by two cellobiohydrolases from Cellulomonas fimi. J. Biotechnol. 57:83-90. Gilkes, N.R., Warren, R.A.J., Miller, R .C, and Kilburn, D.G. (1988). Precise excision of the cellulose-binidng domains from two Cellulomonas fimi cellulases by a homologous protease and the effects on catalysis. J. Biol. Chem. 263:10401-10407. Grodber, J., and Dunn, J.J. (1988). ompT encodes the Escherichia coli outer membrane protease that cleaves T7 RNA polymerase during purification. J. Bacteriol. 170:1245-1253 Grueninger-Leitch, F., D'Arcy, A., D'Arcy, B,. and Chene, C. (1996). Deglycosylation of proteins for crystallization using recombinant fusion protein glycosidases. Protein Sci. 5:2617-2622. Guglielmi, G., and Beguin, P. (1998). Cellulase and hemicellulase genes of Clostridium thermocellum from five independent collections contain few overlaps and are widely scattered across the chromosome. FEMS Microbiol. Lett. 161:209-215. Gusakov, A.V. , Protas, O.V., Chernoglazov, V . M . , Sinitsyn, A.P., KovalyshevaG.V., Shpanchenko, O.V., and Ermolova, O.V. (1991). Transglycosylation activity of cellobiohydrolase I from Trichoderma longibrachiatum on synthetic and natural substrates. Biochim. Biophys. Acta 1073:481-485. Hanahan, D. (1983). Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166:557-580. Harjunpaa, V., Teleman, A., Siika-Aho, M . , and Drakenberg, T. (1995). Kinetic and stereochemical studies of mannooligosaccharide hydrolysis catalysed by B-mannanases from Trichoderma reesei. Eur. J. Biochem. 234:278-283. 176 Hauzer, K., Ticha, M . , Horejsi, V., and Kocourec, J. (1979) Studies on lectins. XLIV. The p H dependence of lectin interactions with sugars as determined by affinity electrophoresis. Biochim. Biophys. Acta 583:103-109. Hendrixon, T.L., and Millane, R.P. (1993). In: Cellulosics:chernical, biochemical and material aspects. Kennedy, J.F., Phillips, C O . , Williams, P A . (Eds.), Ellis Horwood Limited. Henrissat, B. (1990). A classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem. J. 280:304-316. Henrissat, B. and Bairoch, A. (1993). New families in the classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem. J. 293:781-788. Henrissat, B., and Bairoch, A . (1996) Updating the sequence based classification of glycosyl hydrolases. Biochem. J. 316:695-696. Henrissat, B., Callebaut, I., Fabrega, S., Lehn, P., Mornon, J.-P., Davies, G. (1995). Conserved catalytic machinery and the prediction of a common fold for several families of glycosyl hydrolases. Proc. Natl. Acad. Sci. USA 92:7090-7094. Henrissat, B., Teeri, T.T., and Warren, R.A.J. (1998). A scheme for designating enzymes that hydrolyse the polysaccharides in the cell walls of plants. FEBS Lett. 425:352-354. Higgins D., Thompson, J., Gibson, T.Thompson, J.D., Higgins, D . C , Gibson T.J. (1994). "CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting,position-specific gap penalties and weight matrix choice." Nucleic Acids Res. 22:4673-4680. Holmskov, U. , Fischer, P.B., Rothman, A., and Horjup, P. (1996). Affinity and kinetic analysis of bovine plasma C-type lectin collectin-43 (CL-43) interacting with mannan. FEBS Lett. 393:314-316. Hui , P.A., Neukom, H . (1964). Properties of galactomannans. Tappi. 47:39-42. Ichikawa, Y., Look, C C , wong, C.-H. (1992). Enzyme-catalyzed oligosaccharide synthesis. Anal. Biochem. 202:215-238. Irwing, D., Shin, D.H., Barr, B.K., Sakon, J., Karplus, P A . , and Wilson, D.B. (1998). Roles of the catalytic domain and the two cellulose-binding domains of Thermomonospora fusca E4 in cellulose hydrolysis. J. Bacteriol. 180:1709-1714. Jacobson, R.H., Zhang, X.-J., DuBose, R.F., and Matthews, B.W. (1994). Three-dimensional structure of beta-galactosidase from E. coli. Nature 369:761-766. Jain, S., Drendel, W.B., Chen, Z.-W., Mathews, S., Sly, W.S., and Grubb, J.H. (1996). Structure of human beta-glucuronidase reveals canditate lysosomal targeting and active-site motifs. Nature Struc. Biol. 3:375-380. 177 Jerseth, J. (1992). Strategies 5(1). Johnson, E. P., Tomme, P., Joshi, M . D., and Mcintosh, L. P. (1996a). Interaction of soluble cellooligosaccharides with the N-terminal cellulose-binding domain of Cellulomonas fimi CenC. 2. N M R and ultraviolet absorption spectroscopy. Biochem. 35:13895-13906. Johnson, P.E., Joshi, M.D., Tomme, P., Kilburn, D.G., and Mcintosh, L.P. (1996b). Structure of the N-terminal cellulose-binding domain of Cellulomonas fimi CenC determined by nuclear magnetic resonance spectroscopy. Biochem. 35:14381-14394. Kallweit, U. , Boernsen, K.O., Kresbach, G.M., and Widmer, H . M . (1996). Matrix compatible buffers for analysis of proteins with matrix-assisted laser desorption/ionization mass spectrometry. Rapid Com. Mass Spec. 10:845-849. Katz, G. (1965). The location and the significance of the O-acetyl groups in a glucomannan from Parana pine. TAPPI, 48, 34-41. Keller, L.E., Poole, D.M., Ferreira, L.M.A., Durrant, A.J., Hazlewood, G.P., Gilbert, H.J. (1990). Xylanase B and an arabinofuranosidase from Pseudomonas fluorescens subsp. cellulosa contain identical cellulose-binding domains and are encoded by adjacent genes. Biochem J. 272(2): 369-376. Kempton, J.B., and Withers, S.G. (1992). Mechansm of Agrobacterium B-glucosidase: Kinetic studies. Biochemistry 31:9961-9969. Kim, H.K., and Pack, M.Y. (1989). Cloning and expression of Cellulomonas fimi (3-glucosidase genes in Escherichia coli. Enzyme Microb. Technol. 11:313-316. Kleman-Leyer, K .M. , Glikes, N.R., Miller, R.C. Jr., and Kirk, K.T (1994). Changes in the molecular-size distribution of insoluble celulloses by the action of recombinant Cellulomonas fimi cellulases. Biochem. J. 32:463-469. Kohchi C, and Toh-e A. Cloning of Candida pelliculosa beta-glucosidase gene and its expression in Saccharomyces cerevisiae. Mol Gen Genet. 203: 89-94 Kormos, J. M . (1998). Master thesis. Functional and mutational analysis of the cellulose-binding domain CBDN1 from Cellulomonas fimi B-glucanase C (CenC). University of British Columbia. Koshland, D.E. (1953). Stereochemistry and the mechanism of enzymatic reactions. Biol. Rev. 28:416-436. Krawiec, S., and Riley, M . (1990). Organisation of the bacterial chromosome. Microbiol. Rev. 54:502-539. Leamli, U.K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. 178 Leatherbarrow, R.J. (1992). GraFit version 3.0. Erithacus Software Ltd., Staines, United Kingdom. Legler, G. (1990). Glycoside hydrolases: Mechanistic information from studies with reversible and irreversible inhibitors. Adv. Carbohydr. Chem. Biochem. 48:319-385. Leipprandt, J.R., Kraemer, S.A., Haithcock, B.E., Chen, H. , Dyme, J.L., Cavanagh, K.T., Friderici, K .H. , and Jones, M.Z. (1996) Caprine beta-mannosidase: sequencing and characterization of the cDNA and identification of the molecular defect of caprine beta-mannosidosis. Genomics 37:51-56. Lemaire, M . , Miras, I., Gounon, P., and Beguin, P. (1998). Identification of a region responsible for binding to the cell wall within the S-layer protein of Clostridium thermocellum. Microbiol. 144:211-217. Lemaire, M . , Ohayon, H. , Gounon, P., Fujino, T., and Beguin, P. (1993). OlpB, a new outer layer protein form Clostridium thermocellum, and binding of its S-layer-like domains to components of the cell envelope. J. Bacteriol. 177:2451-2459. Letzhava, A., Mizukami, T., Kajitani, T., Kameoka, D., Redenbach, M . , Shinkawa, H. , Nimi, O., and Kinashi, H . (1995). Physical map of a linear chromosome of Streptomyces griseus. J. Bacteriol. 177:6492-9498. Lever, M . (1973). Colorimetric and fluorometric carbohydrate dtermination with p-hydroxybenzoic acid hydrazide. Biochem. Med. &:274-281. Lever, M . (1977). Carbohydrate determination with 4-hydroxybenzoic acid hydrazide (PAHBAH): effect of bismuth on the reaction. Anal. Biochem. 81:21-27. Linder, M . , Mattinen, M.-L., Kontteli, M . , Lindeberg, G., Stahlberg, J., Drakenberg, T., Reinikainen, T., Petterson, G., and Annila, A. (1995). Identification of functionally important amino acids in the cellulose-binding domain of Trichoderma reesei cellobiohydrolase I. Protein Sci. 4:1056-1064. Lu, J. (1997). Collectins: collectors of microorganisms for the innate immune system. BioEssays 19(6):509-518. Luthi, E., Jasmat, N.B., Grayling, R.A., Love, D.R., and Bergquist, P. (1991). Cloriing, sequence analysis, and expression in Escherichia coli of a gene coding for a beta-mannanase from the extremely thermophilic bacterium Caldocellum saccharolyticum. Appl. Environ. Microbiol. 57:694-700. Mackenzie, L.F., Brooke, G.S., Cutfield, J.F., Sullivan, P.A., and Withers, S.G. (1997). Identification of Glu-330 as the catalytic nucleophile of Candida albicans exo-B-l,3-glucanase. J. Biol. Chem. 272:3161-3167. Mackenzie, L.F., Wang, Q., Warren, R.A.J., Withers, S.G. (1998). Glycosynthases: Mutant glycosidases for oligosaccharide synthesis. J. Am. Chem. Soc. 120:5583-5584. 179 MacLeod, A . M . , Lindhorst, T., Withers, S.G. and Warren, R.A.J. (1994). The acid/base catalyst in the exoglucanase/xylanase (Cex) from Cellulomonas fimi. Gene 121:143-147. Mahuran, D.J. (1995). (^hexosaminidase: Biosynthesis and processing of the normal enzyme, and identification of mutations causing Jewish Tay-Sachs Disease. Clinical Biochemistry 28:101-106. Margolles-Clark, E., Tenkanen, M . , Luonteri, E., and Penttilae (1996) Three alpha-galactosidase genes of Trichoderma reesei cloned by expression in yeast. Eur. J. Biochem. 240: 104-111. Matsudaira, P. (1990). Limited N-terminal sequence analysis. Methods Enzymol. 182:602-613. McCarter, J.D., Adam, M.J., and Withers, S.G. (1992). Binding energy and catalysis. Biochem. J. 286:721-727. McCarter, J.D., Adam, M.J., Hartman, N.G., and Withers, S.G. (1994). In vivo inhibition of (3-glucosidase and B-mannosidase activity in rats by 2-deoxy-2-fluoro-B-glycosyl fluorides and recovery of activity in vivo and in vitro. Biochem. J. 301:343-348. McCarter, J.D., and Withers, S.G. (1994). Mechanism of enzymatic glycoside hydrolysis. Curr. Opin. Struct. Biol. 4:885-892. McCleary, B. (1983). B-D-mannosidase from Helix pomatia. Carb. Res. 111:297-310. McCleary, B.V. (1978). Soluble, dye-labeled polysaccharides for the assay of endohydrolases. Methods in Enzymol. 160:74-86. McCleary, B.V., Clark, A . H . , Dea, I.C.M., and Rees, D.A. (1985). The fine structure of carob and guar galactomannans. Carb. Res. 139:237-260. McCleary, B.V., Taravel, F.R., and Joseleau, J.-P. (1983a). Characterisation of the oligosaccharides produced on hydrolysis of galactomannan with B-D-mannanase. Carb. Res. 118:91-109. Mcintosh, L.P., Hand, G., Johnson, P.E., Joshi, M.D., Koerner, M . , Plesniak, L.A., Ziser, L., Wakarchuk, W.W., and Withers, S.G. (1996). The pKa of the general acid/base carboxyl group of a glycosidase cycles during catalysis: A 13C-NMR study of Bacillus circulans xylanase. Biochem. 35:9958-9966. Meier, H . (1958). On the structure of cell walls and cell wall mannans from ivory nuts and from dates. Biochim. Biophys. Acta 28:229-240. Meinke, A. , Gilkes, N.R., Kilburn, D.G., Miller, R.C. Jr., and Warren, R.A.J. (1993). Cellulose-binding polypeptides from Cellulomonas fimi: Endoglucanase D (CenD), a family A (3-1,4-glucanase. J. Bacteriol. 175:1910-1918. 180 Meinke, A. , Gilkes, N.R., Kilburn, D.G., Miller, R G . Jr., and Warren, R.A.J. (1991). Multiple domains in endoglucanase B (CenB) from Cellulomonas fimi: function and relatedness to domains in other polypeptides. J. Bacteriol. 173:7126-7135. Meinke, A. , Gilkes, N.r., Kwan, E., Kilburn, D.G., Warren R.A.J., and Miller, R.C. Jr. (1994). Cellobiohydrolase A (CbhA) from the cellulolytic bacterium Cellulomonas fimi is a B-1,4,-exocellobiohydrolase analogous to Trichoderma reesei CBHII. Mol Microbiol. 12:413-422. Meinke, A., Schmuck, M . , Gilkes, N.R., Kilburn, D.G., Miller, R .C, and Warren, R.A.J. (1992). The tertiary structure of endo-B-l,4-glucanase B (CenB), a multidomain cellulase from the bacterium Cellulomonas fimi. Glycobiol. 2:321-326. Mendoza NS, Arai M , Sugimoto K, Ueda M , Kawaguchi T, Joson L M (1995) Cloning and sequencing of beta-mannanase gene from Bacillus subtilis N M - 39. Biochim.Biophys.Acta 1243:552-554. Miao, S., McCarter, J.D., Grace, M . , Grabowski, G., Aebersold, R., and Withers, S.G. (1994). Identification of Glu 340 as the active site nucleophile in human glucocerebrosidase by use of electrospray tandem mass spectrometry. J. Biol. Chem. 269:10975-10978. Millward-Saddler, S.J., Davidson, K., Hazlewood, G.P., Black, G.W., Gilbert, H.J., and Clarke, J.H. (1995). Novel cellulose-binding domains, NodB homologes and conserved modular architecture in xylanases from the aerobic soil bacteria Pseudomonas fluorescens subs, cellulosa and Cellvibio mixtus. Biochem. J. 312:39-48. Millward-Sadler, S.J., Hall, J., Black, G.W., Hazlewood, G.P., and Gilbert, H.J. (1996) Evidence that the Piromyces gene family encoding endo-l,4-mannanases arose through gene duplication. FEMS Microbiol. Lett. 141:183-188. Milward-Sadler, S.J., Poole, D.M., Hernissat, B., Hazlewood, G.P., Clarke, J.H., and Gilbert, H.J. (1994). Evidence for a general role for high-affinity non-catalytic cellulose-binding domains in microbial plant cell wall hydrolases. Mol. Microbiol. 11:375-382. Mizuno, Y., Kozutsumi, Y., Kawasaki, T., and Yamashina, I. (1981) Isolation and characterization of a mannan-binding protein from rat liver. J. Biol. Chem. 256:4247-4252. Morris, D.D., Reeves, R.A., Gibbs, M.D., Saul, D.J., and Berquist, P.L. (1995). Correction of the B-mannanase domain of the celC pseudogene from Caldocellosiruptor saccharolyticus and activity of the gene product on kraft pulp. Appl. Environ. Microbiol. 61:2262-2269. Murata, T., Akimoto, S., Horimoto, M . , Kusui, T. (1997). Galactosyl transfer onto p-nitrophenyl (3-D-glucoside using B-D-galactosidase from Bacillus cirulans. Bio. Biotech. Biochemisty 61:1118-1120. Nakamura, K, Kashiwagi, S., and Takeo, K. (1992) Characterization of the interaction between human plasma fibronectin and collagen by means of affinity electrophoresis. J. Chrom. 597:351-356. 181 Namchuk, M.N. , and Withers, S.G. (1995). Mechanism of Agrobacterium B-glucosidase: Kinetic analysis of the role of noncovalent enzyme/substrate interactions. Biochemistry 34:16194-16202. Neufeld, E.F. (1991). Lysosomal storage diseases. Annu. Rev. Biochem. 60:257-280. Neustroev, K.N. , Krylov, A.S., Firsov, L.M. , Abroskina, O.N., and Khorlin, A.Y. (1991). Isolation and properties of B-mannosidase from Aspergillus awamori. Biokihimiya 56:1406-1412. Nidetzki, B., Steiner, W., Hayn, M . , and Claeyssens, M . (1994). Cellulose hydrolysis by the cellulases from Trichoderma reesei: a new model for synergistic interaction. Biochem. J. 298:705-710. Nieduszinski, I., and Marchessault, R.H. (1972). the crystalline structure of poly-B,D(l,4)mannose: mannan I. Can. J. Chem. 50:2130-2138. Nielsen, H. , Engelbrecht, J., Brunak, S., and von Heijne, G. (1997). Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Prot. Eng. 10:1-6. Notenboom, V., Birsan, C , Warren, R.A.J., Withers, S.G., and Rose, D.R. (1998). Exploring the cellulose/xylan specificity of the beta-l,4-glycanase cex from Cellulomonas fimi through crystallography and mutation. Biochemistry. 37: 4751-4758 O' Neill, G., Goh, S.H., Warren, R.A.J., Kilbrun, D.G., and Miller, R.C.Jr. (1986). Structure of the gene encoding the exoglucanase of Cellulomonas fimi. Gene 44:325-330. Olabarria, G., Carrascosa, J., Pedro, M.A., and Berenguer, J. (1996) A conserved motif in S-layer proteins is involved in peptidoglycan binding in Thermus thermophilus. J. Bacteriol. 178:4765-4772. Pace, N.C., Vajdos, F., Fee, L., Grimsely, G., and Gray, T. (1995). How to measure and predict the molar absorption coefficient of a protein. Protein Sci. 4:2411-2423. Posci, I., Taylor, S.A., Richardson, A.C., Smith, B.V., and Price, R.G. (1993). Comparison of several new chromogenic galactosidases as substrates for various B-D-galactosidases. Biochim. Biophys. Acta 1163:54-60. Puis, J., and Schuseil, J. (1993). Chemistry of hemicelluloses: relationship between hemicellulose structure and enzymes required for hydrolysis, ppl-28, In: Coughlan, M.P. and Hazlewood, G.P. (Eds.), Hemicellulose and hemicellulases, Portland press, London. Raven, J.A. (1996). The role of autotrophs in global C02 cycling. In: Microbial growth on CI Compounds (Lidstrom, M.E. and Tabita, F.R., eds.), pp. 351-358. Kluwer, Dordrecht. 182 Ries, W., Hotzy, C , Schocher, I., Sleytr, U.B., and Sagra, M . (1997). Evidence that the N -terminal part of the S-layer protein from Bacillus stearothermophilus PV72/p2 recognizes a secondary cell wall polymer. J. Bacteriol. 179:3892-3898. Rol, F. (1973). Locust bean gum, pp 323-337. Whistler, R.L., Miller, J.N. (Eds.), Industrial gums, Academic press, New York. Rouvinen, J., Bergfors, T., Teeri, T., Knowles, J.K.C., and Jones, T.A. (1990) Three-dimensional structure of cellobiohydrolase II from Trichoderma reesei. Science. 249(4967): 380-386 Sakon, J., Irwin, D., Wilson, D.B., and Karplus, P.A. (1997). Structure and mechanism of endo/exocellulase E4 from Thermomonospora fusca. Nature Struct. Biol. 4:810-817. Saloheimo, A. , Henrissat, B., Hoffren, A . M . , Teleman, O., and Penttila, M . (1994). A novel, small endoglucanase gene, egl5, from Trichoderma reesei isolated by expression in yeast. Mol. Microbiol. 13:219-228. Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual. 2 n d ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Sanjeev, J., Drendel, W.B., Chen, Z.W., Mathews, F.S., Sly, W.S. and and Grubb, J.H. (1996). Structure of human B-glucuronidase reveals candidate lysosomal targeting and active-site motifs. Nature Struct. Biol. 3:375-380. Sawadogo, M . , and Van Dyke, M.W. (1991). A rapid method for the purification of deprotected oligodeoxynucleotides.NAR 19:674. Shen, H . , Gilkes N.R., Kilburn, D.G., Miller, R.C, and Warren, R.A.J. (1995). Cellobiohydrolase B, a second exocellobiohydrolase from the cellulolytic bacterium Cellulomonas fimi. Biochem. J. 311:67-74. Sinnott, M.L. (1990). Catalytic mechanisms of enzymic glycosyl transfer. Chem. Rev. 90:1171-1202. Spezio, M . , Wison, D.B., and Karplus, P.A. (1993) Crystal structure of the catalytic domain of a thermophilic cellulase. Biochemistry 32:9906-9912 Stackerbrandt, E., and Prauser, H . (1991). Assignment of the genera Cellulomonas, Oerskovia, Promicromonospora and Jonesia to Cellulomonadacaea fam. nov. system. Appl. Microbiol. 14:261-265. Stalbrand, H. , Saloheimo, A., Vehmaanpera, J., Henrissat, B., and Pentilla, M . (1995). Cloning and expression in Saccharomyces cerevisiae of a Trichoderma reesei beta-mannanase gene containing a cellulose-binding domain. Appl. Environ. Microbiol. 61(3):1090-1097. Stalbrand, H. , Siika-aho, M . , Tenkanen, M . , and Viikari, L. (1993). Purification and characterization of two beta-mannanases from Trichaderma reesei. J. Biotech. 29:229-242. 183 Stewart, B.J., and Leatherwood, J.M. (1976). Derepressed synthesis of cellulase by Cellulomonas. J. Bacteriol. 128:609-615. Street, LP., Kempton, J.B., and Withers, S.G. (1992). Inactivation of a B-glucosidase through the accumulation of a stable 2-deoxy-2-fluoro-a-D-glucopyranosyl-enzyme intermediate: a detailed investigation. Biochem. 31: 9970-9978. Stryer, L. (1988). Carbohydrates. In: Biochemistry pp.331-348. W.H. Freeman and Company, New York. Takeo, K. (1984). Affinity electrophoresis: Principles and applications. Electrophoresis 5:187-195. Takeo, K., and Nakamura, S. (1972). Dissociation constant of glucan phosphorylase of rabbit tissues studied by polyacrylamide gel disc electrophoresis. Arch. Biochem. Biophys. 10:1684. Talbot, G., and Sygusch, J. (1990). Purification and charaterization of thermostable beta-mannanase and alpha-glactosidase from Bacillus stearothermophilus. Appl. Environ. Microbiol. 56:3505-3510. Teather, R.M., and Wood, P.J. (1982). Use of Congo Red - polysaccharide interactions in enumeration and characterization of cellulolytic bacteria from the bovine rumen. Appl. Environ. Microbiol. 43:777-780. Teeri, T.T., Lehtovaara, P., Kauppinen, S., Salovuori, I., and Knowles, J. (1987). Homologous domains in Trichoderma reesei cellulolytic enzymes: gene sequence and expression of cellobiohydrolase II. Gene 51:43-52. Tenkanen, M . , Makkonen, M . , Perttula, M . , Viikari, L., and Teleman, A. (1997). Action of Trichoderma reesei on galactoglucomannan in pine kraft pulp. J. Biotech. 57:191-204. Tews, I., Perrakis, A., Oppenheim, A., Dauter, Z., Wilson, K.S. and Vorgias, C.E. (1996). Bacterial chitobiase structure provides insight into catalytic mechanism and the basis of Tay-Sachs disease. Nature Struct. Biol. 3:638-648. Thiem, J. (1995). Application of enzymes in synthetic carbohydrate chemistry. FEMS Microbiol. Rev. 16:193-211. Thomas, M.G. (1994). A procedure for second-round differential screening of cDNA libraries. Biotechniques. 16:988-989. Timell, T.E. (1967). Recent progress in the chemistry of wood hemicelluloses. Wood Sci. Technol. 1:45-70. Tomme, P, Kwan, .E., Gilkes, N.R., Kilburn, D.G., and Warren, R.A.J. (1996a). Characterization of CenC, an enzyme from Cellulomonas fimi with both endo- and exoglucanase activities. J. Bacteriol. 178:4216-4223. 184 Tomme, P., Creagh, A.L., Kilburn, D.G., and Haynes, C A . (1996b). Interaction of polysaccharides with the N-terminal cellulose-binding domain of Cellulomonas fimi CenC. 1. Binding specificity and calorimetric analysis. Biochemistry 35:13885-13894. Tomme, P., Warren, R.A.J., Gilkes, N.R. (1995). Cellulose hydrolysis by bacteria and fungi. Adv. Microb. Physiol. 37:1-81. Toone, E.J., Simon, E.S., Bednarski. M.D., and Whiteside, G.M. (1989). Enzyme catalyzed synthesis of carbohydrates. Tetrahedron. 45(259):5365-5422. Torronen, A . and Rouvinen, J. (1995). Structural comparison of two major endo-1,4-xylanases from Trichoderma reesei. Biochemistry 34:847-856. Towbin, H . , Sthaehelin, T., and Gordon, J. (1970). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354. ' Trevor, J.T. (1996). Genome size in bacteria. Antonie van Leuwenhoek 69:293-303. Tull, D., Withers, S.G., Gilkes, N.R., Kilburn, D.G., Warren, R.A.J, and Aebersold, R. (1991). Glutamic acid 274 is the nucleophile in the active site of a "retaining'' exoglucanase from Cellulomonas fimi. J. Biol. Chem. 266(24):15621-15625. van Emmerik LC, Kuijper EJ, Fijen CA, Dankert J, and Thiel S (1994) Binding of mannan-binding protein to various bacterial pathogens of meningitis. Clin.Exp.Immunol. 97:411-416. Wakarchuk, W.W., Kilburn, D.G., Miller, R.C. Jr., and Warren, R.A.J. (1984). J. Gen. Microbiol. 130:1385-1389. Warren, R.A.J. (1996). Microbial hydrolysis of polysaccharides. Annu. Rev. Microbiol. 50:183-212. White, A. , Withers, S.G., Gilkes, N.R. and Rose, D.R. (1994). Crystal structure of the catalytic domain of the b-l,4,-glycanase Cex from Cellulomonas fimi. Biochem. 33:12456-12552. Williamson, M.P., LeGal-Coeffet, M.-F., Sorimachi, K., Furniss, C.S.M., Archer, D.B., Williamson, G. (1997) Function of conserved tryptophans in the Aspergillus niger glucoamylase 1 starch binding domain. Biochemistry. 36(24): 7535-7539 Withers, S.G., and Aebersold, R. (1995). Approaches to labeling and identification of active site residues in glycosidases. Protein Sci. 4:361-372. Withers, S.G., Mackenzie, L., and Wang, Q. (1998). Methods and compositions for the synthesis of oligosaccharides, and the products formed thereby. United States Patent. Patent Number:5,716,812. 185 Withers, S.G., Rupitz, K., Trimbur, D., and Warren, R.A.J. (1992). Mechanistic consequences of mutation of the active site nucleophile Glu 358 in Agrobacterium B-glucosidase. Biochemistry 31:9979-9985. Wong, W.K.R., Gerhard, B., Guo, Z.M. , Kilburn, D.G., Warren, R.A.J., and Miller, R.C.Jr. (1986). Charaterization and structure of an endoglucanase gene cenA of Cellulomonas fimi. Gene 44:315-324. Wong-Madden, S.T., and Londry, D. (1998). New England Biolabs. Xiuzhu, D., Schyns, P.J.Y.M.J., and Stams, A.J.M. (1991). Degradation of galactomannan by a Clostridium butyricum strain. Antonie van Leuwenhoek 60:109-114. Xu, G.Y., Ong, E., Gilkes, N.R., Kilburn, D.G., Muhandrian, D.R., Harris-Brandts, M . , Carver, J.P., Kay, L.E., Harvey, T.S. (1995). Solution structure of a cellulose binding domain from Cellulomonas fimi by nuclear magnetic resonace spectroscopy. Biochemistry 34:6993-7009. Yanish-Perron, C , Vierira, J., and Messing, J. (1985). Improved M13 phage cloning vectors and host strains: Nucleotide sequences of M13mpl8 and pUC19 vectors. Gene 33:103-119. Yasukochi, T., Fukase, K., Suda, Y., Takagaki, K., Endo, M . , and Kusumoto, S. (1997). Enzymatic synthesis of 4-methylumbelliferyl glycosides of trisaccharide and core tetrasaccharide, Gal(pl-3) Gal(pl-4) Xyl and GlcA(pl-3) Gal(pl-3) Gal(pl-4) Xyl, corresponding to the linkage region of proteoglycans. Bull. Chem. Soc. Jpn., 70:2719-2725. Yui, T., Miyawaki, K., Yada, M . , and Ogawa, K. (1997). A n evaluation of crystal structure of mannan I by X-ray powder diffraction and molecular mechanics studies. Int. J. Biol. Macromol. 21(3): 243-250. 186 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items