UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterisation of the cellulose binding domain of endo-β-1,4-glucanase A from Cellulomonas fimi Din, Neena 1994

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

Item Metadata


831-ubc_1994-953365.pdf [ 3.89MB ]
JSON: 831-1.0088306.json
JSON-LD: 831-1.0088306-ld.json
RDF/XML (Pretty): 831-1.0088306-rdf.xml
RDF/JSON: 831-1.0088306-rdf.json
Turtle: 831-1.0088306-turtle.txt
N-Triples: 831-1.0088306-rdf-ntriples.txt
Original Record: 831-1.0088306-source.json
Full Text

Full Text

CHARACTERISATION OF THE CELLULOSE BINDINGDOMAIN OF ENDO-B-1,4-GLUCANASE A FROMCELL ULOMONA S FIMIbyNeena DinB .Sc., University College, LondonM.Sc., University of British ColumbiaA THESIS SUBMITTED IN PARTIAL FULFILLMENTOF THE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHIEOSOPHYinTHE FACULTY OF GRADUATE STUDIESDepartment of Microbiology and ImmunologyWe accept this thesis as conformingto the required standardTHE UNIVERSITY 0 BRITIS COLUMBIASeptember 94© Neena Din, 1994In presenting this thesis in partial fulfillment of therequirements for an advanced degree at the University of BritishColumbia, I agree that the Library shall make it freely availablefor reference and study. I further agree that permission forextensive copying of this thesis for scholarly purposes may begranted by the head of my department or by his or herrepresentatives. It is understood that copying or publication ofthis thesis for financial gain shall not be allowed without mywritten permission.(Signature)Department of______________________The University of British ColumbiaVancouver, CanadaDate / C\UAbstractEndoglucanase A (CenA) from the bacterium Cellulomonasfimi comprises acatalytic domain and a non-hydrolytic cellulose binding domain (CBD) which canfunction independently. The DNA fragment encoding CBDCenA was cloned andexpressed in E. coli and the polypeptide produced was characterised in an attempt toexamine the mechanisms by which CBDs adsorb to their substrate and to determine therole of CBDs in the process of cellulose hydrolysis.CBDCepA and related CBDs from other bacterial glycanases have a characteristicmotif of four highly conserved tryptophan residues. Each of two of these residues (W14and W68) was mutated in CBDCenA to give the mutant polypeptides CBDCenAW14Aand CBDCepAW68A. The binding affinities of CBDCenAW14A and CBDCenAW68Afor crystalline cellulose were reduced 50 and 30 fold respectively compared withCBDCenA. Data obtained from CD, 1H NMR and tryptophan fluorescence spectroscopyindicated that the mutant polypeptides have a similar conformation to CBDp.Tryptophan fluorescence data also suggested that W14 and W68 are exposed on thesurface of CBDCenA and hence positioned to interact with residues on the cellulosesurface upon adsorption of the polypeptide.Tryptophan fluorescence spectroscopy was used to examine the interaction ofCBDCenA with the soluble glucans cellohexaose, carboxymethylcellulose andhydroxyethylcellulose and the insoluble substrate, bacterial microcrystalline cellulose(BMCC). No interaction of CBDCenA with the soluble glucans was detected, howeverfluorescence quenching of CBDCenA was observed upon addition of BMCC.The isolated binding domain of CenA released small particles from cotton anddisrupted the structure of cellulose fibres. This is the first demonstration of thedispersive effects of a CBD and implies art active role for some CBDs in cellulosehydrolysis. The related CBDCex showed similar properties to CBDCenA in disruptingthe structure of cellulose fibres; however the mutant polypeptides CBDCenAW14A andCBDCenAW68A did not release small particles from cotton.A co-operative interaction between the isolated binding and catalytic domains ofCenA in the release of soluble and insoluble sugar from cotton fibres was observed. Thetwo domains interacted synergistically in the hydrolysis of cotton fibres. Thisintramolecular synergism is distinct from the previously described intermolecularsynergism observed between individual cellulases.inivTable of contentsPageAbstract.iiTable of Contents ivList of Tables viiList of Figures viiiList of Abbreviations xiAcknowledgments xiii1. Introduction 11.1 Cellulose 11.2 Cellulose binding domains 51.3 Objectives of this study 152. Materials and Methods 172.1 Chemicals, buffers and enzymes 172.2 Bacterial strains, plasmids and phage 172.3 Media and growth conditions 172.4 Recombinant DNA techniques 192.5 Screening for gene expression 212.6 Cell fractionation 222.7 Purification of proteins 232.8 Detection of proteins 252.9 Determination of protein concentration 262.10 Sedimentation equilibrium centrifugation 262.11 Binding Analyses 27V2.11.1 Avicel binding assays .272.11.2 Binding analysis with BMCC 272.12 Fluorescence Spectroscopy 282.12.1 Binding to soluble substrates 282.12.2 Binding to BMCC 302.12.3 Quenching by iodide 302.13 Circular dichroism 312.14 Nuclear magnetic resonance 312.15 Treatment of cellulose fibres with purified proteins 312.15.1 Small particle release from cotton 312.15.2 Synergy between CBD.PTCenA and p30 332.15.3 Surface degradation of Ramie fibres 332.15.4 Disruption of Ramie fibre structure 343. Results 353.1 Construction of expression vectors 353.1.1 Construction of the plasmid pUC18-CBDCenA, expressingthe cellulose binding domain of CenA 353.1.2 Construction of the plasmids expressing CBDCenAW14Aand CBDCenAW68A 353.2 Production and purification of CBDCenA, CBDCenAW14Aand CBDCe11AW68A 433.3 Binding studies 523.3.1 Estimation of adsorption parameters for CBDCe11A,CBDCenAW14A and CBDCenAW68A 523.3.2 Fluorescence quenching 593.4 Structural studies on CBDCenA, CBDCenAW14A and CBDCenAW68A 65vi3.4.1 Circular dichroisrn .653.4.2 1H NMR 653.4.3 Fluorescence data 673.5 Role of CBDCenA in the breakdown of cellulose 783.5.1 Effects of CBD.PTCenA, CBDCenA, p30 and CenA on cotton fibres 783.5.2 Effects of CBD.PTCenA, p30 and CenA on Rarnie fibres 823.5.3 Small particle release by CBDCenAW14A and CBDCenAW68A 923.5.4 Small particle release by Cex and CBDCex andtheir effects on Ramie fibres 923.5.5 Synergy between CBD and p30 974. Discussion 1054.1 Binding interaction of CBDCenA with cellulose 1054.2 Role of CBDCenA in the hydrolysis of cellulose 1105. References 120vuList of TablesTable Page1.1 Cellulose binding domain families .102.1 E. coli strains 182.2 Plasmids and phage 183.1 Equilibrium association constants for the binding of CBDCenA,CBDCenAW14A and CBDCeI1AW68A to BMCC 57vu’List of FiguresFigure Page1.1 The structure of cellulose .21.2 The structure of CenA and Cex 71.3 Amino-acid sequence of some family II CBDs from bacterial13-1 ,4-glycanases 123.1 The DNA sequence and amino-acid sequence of CBDCenA 363.2 Construction of the expression vector pUC18-CBDCenA 383.3 Screening for CBDCenA expression from E. coli JM1O1 transformants 403.4 Sequence analysis of mutants of CBDCenA 423.5 Detection of polypeptides CBDCenA, CBDCenAW14A andCBDCenAW68A from E. coli .JM1O1 clones by SDS-PAGE analysis 443.6 Detection of polypeptides CBDCenA, CBDCenAW14A andCBDCenAW68A from E. co/i JM1O1 clones by Western blot analysis 463.7 SDS-PAGE of E. co/i JM1O1: PUC18-CBDCenAosmotic shock fractions 483.8 Purification of CBDCenA on CF1TM 493.9 SDS-PAGE of CBDCenA, CBDCenAW14A and CBDCenAW68Apurified by CF1TM and size-exclusion chromatography 513.10 Semi-quantitative analysis of the adsorption of CBDCenA,CBDCenAW14A and CBDCenAW68A to BMCC 533.11 Adsorption isotherms of CBDCenA, CBDCenAW14A andCBDCenAW68A to BMCC 563.12 Double-reciprocal plots of adsorption data for CBDCenA,CBDCenAW14A and CBDCenAW68A 58ix3.13 Change in the relative fluorescence of CBDCenA on addition ofcellohexaose (G6) 613.14 Change in the relative fluorescence of CBDCenA on addition ofcarboxymethylcellulose (CMC) 623.15 Change in the relative fluorescence of CBDCenA on addition ofhydroxyethylcellulose 633.16 Change in the relative fluorescence of CBDCenA on addition of BMCC 643.17 Circular dichroism spectra of CBDCenA, CBDCenAW14A andCBDCenAW68A 663.18 One dimensional 1H NMR spectra of CBDCenA, CBDCe11AW 4Aand CBDCe11AW68A 683.19 Tryptophan fluorescence spectroscopy of CBDCenA, CBDCenAW14Aand CBDCenAW68A 693.20 Tryptophan fluorescence emission spectra of CBDCenA in the absenceand presence of BMCC 723.21 Tryptophan fluorescence emission spectra of CBDCenAW14A in theabsence and presence of BMCC 733.22 Tryptophan fluorescence emission spectra of CBDCenAW68A in theabsence and presence of BMCC 743.23 Spectra of W14 in the absence and presence of BMCC 763.24 Spectra of W68 in the absence and presence of BMCC 773.25 Small particle release from dewaxed cotton 793.26 Small particle production from dewaxed cotton by CBD.PTCenA,CenA or p30 after 2 minutes of incubation 803.27 Small particle production from dewaxed cotton by CBD.PTCenA 813.28 Phase contrast microscopy of small particles released from cottonby CenA, CBD.PTCCnA and p30 83x3.29 Scanning electron microscopy of particulate material released fromcotton by CenA, p30 and CBD.PTCenA 853.30 Scanning electron micrographs of Rarnie cotton fibres treated withCenA, CBD.PTCe11and p30 883.31 Disruption of Ramie fibre structure 903.32 Small particle release from cotton by CBDCenA, CBDCenAW14Aand CBDCenAW68A 933.33 Small particle production from dewaxed cotton by CBDex and Cex 943.34 Scanning electron microscopy of cellulose fibres treated withCBDcex and Cex 953.35 Release of sugars from cotton by CBD.PTCenA and p30 983.36 Release of sugars from cotton by p30, CBD.PTCe11Aand CenA 1013.37 Release of sugars from BMCC by p30, CBD.PTCenA and CenA 103xiList of Abbreviations[B] Concentration of bound ligand[F] Concentration of free ligand[N] Concentration of available binding sites[N0] Concentration of binding sites in absence of liganda Number of lattice units occupied by a single ligandAmp r Ampicillin resistanceBMCC Bacterial microcrystalline celluloseBSA Bovine serum albuminCBD Cellulose binding domainCBD.PTCenA Cellulose binding domain of CenA with thePro/Thr linkerBDCenA Cellulose binding domain of C.fimi endo-13-1,4-glucanase ACBDCenAW14A The cellulose binding domain of CenA in whichresidue tryptophan 14 has been changed to analanineCBDCenAW68A The cellulose binding domain of CenA in whichresidue tryptophan 68 has been changed to analanineCBDex The cellulose binding domain of C.fimi exo-13-1 ,4-xylanase/B- 1 ,4-glucanaseCBHI and CBFTII Cellobiohydrolases I and II from T. ReeseiCD Circular dichroismCenA C.fimi endo-B-1,4-glucanase ACex C. fimi exo-B- 1 ,4-xylanase/B- 1 ,4-glucanaseXIICMC CarboxymethylcellulosedNTP Deoxynucleoside triphosphateDTT DithiothreitolFITC Fluorescein isothiocyanate06 CellohexaoseGdmCl Guanidinium chlorideHEC HydroxyethylcelluloseIPTG Isopropyl-13-D-thiogalactosideKa Equilibrium association constantKr Relative equilibrium association constantkbp Kilobase pairskDa KilodaltonsKan r Kanamycin resistanceLB Luria-BertaniNMR Proton nuclear magnetic resonancep30 Catalytic domain of CenAPBS Phosphate-buffered salinePEG Polyethylene glycolPro)Thr linker Sequence linking the catalytic and bindingdomains of CenA and Cexrpm Revolutions per minuteSAXS Small angle X-ray scatteringSDS-PAGE S odium dodecyl suiphate-polyacrylamide gelelectrophoresisSEM Scanning electron microscopyTCA Trichioro acetic acidTYP Tryptone, yeast extract, phosphate mediumx1llACKNOWLEDGMENTSThis thesis work was supported by grants from the Natural Sciences andEngineering Research Council of Canada. First, I would like to sincerely thank my jointsupervisors Doug Kilbum and Tony Warren for their constant encouragement andenthusiasm over the years, and for providing me with the opportunity to take on thisinteresting project. Thanks also for their friendship and fantastic sense of humour,which created a great atmosphere in which to work. I am also greatful to Bob Miller andNeil Gilkes for their many useful ideas and input into this project. Special thanks go toPeter Tomme for sharing his vast knowledge of the cellulase field with me. I would alsolike to acknowledge Les Burtnick and Mike Weiss for introducing me to the techniquesof fluorescence spectroscopy and scanning electron microscopy, and for their excellentexpert advice.The following people deserve many thanks for their technical help: Ian Forsythefor his help with the work on CBDCenAW68A and binding analysis; Emily Kwan for herconstant supply of p30 and CenA; Dedreia Tull for dewaxing cotton; Les Hicks andCyril Kay for carrying out the equilibrium sedimentation centrifugation and finallyAndrew Stevenson for his help with the circular dichroism studies.Many warm thanks to all the past and present members of the cellulase lab who Ihave had the pleasure to work with, for their wonderful companionship and sense of fun.In particular thanks to Helen Smith and Patti Miller for always being there when youneeded them and to Jeff Greenwood and Edgar Ong for their special friendship and greatadvice.To my husband and best friend Terry, I can’t say thank you enough for all thesupport and love you’ve given me over the years. Thanks for keeping me sane!I dedicate this thesis to my parents for their constant belief in my abilities andtheir love.11. Introduction1.1 CelluloseCellulose is the major structural component of all higher plants and issynthesised by a number of other organisms, including fungi, bacteria, invertebrates andprotists (Richmond, 1991). Due to its vast abundance as a renewable resource, thetreatment of cellulose by cellulolytic enzymes to release sugars for practical purposeshas been of great interest to biotechnologists. Improvement of cellulose hydrolysis bycellulolytic enzymes requires a good understanding not only of the structure andfunction of these enzymes, but also of the structure of the substrate itself.Although some of the structural features of cellulose are debated, the chemicalstructure of cellulose is well defined. Cellulose is a linear polymer composed of 13-1,4linked glucose subunits, in which each glucose residue is rotated by 1800 relative to thepreceding glucose residue, forming the basic repeating unit of cellobiose (Figure 1.1).These cellulose molecules have degrees of polymerisation (DP) ranging from 500 to14,000 residues in higher plants. The cellulose chains are themselves joined by inter-and intramolecular hydrogen-bonding networks and van der Waal& forces (Figure 1.1).These physical forces hold the chains in insoluble aggregates called microfibrils. Nativecellulose microfibrils vary in size ranging from 2Onm in diameter for the alga Valoniamacrophysa to 3-4nm for wood (Chanzy, 1990). In most natural sources of cellulose,such as higher plant cells, these cellulose microfibrils are closely associated with othercell wall components, such as hemicellulose (composed mainly of xylans andglucomannans), pectin and lignin (Frey-Wyssling, 1976; Atalla et al., 1993; Rees, 1977;Alberts et aL, 1983).A 2HPanel A; the chemical structure of cellulose, indicating the glucose and cellobioseresidues. Panel B; the structure of cellulose I, ab projection, looking along the chainaxes. Van der Waal& contacts are made in the direction of the b axis. Panel C; celluloseI, ac projection. Panel D; cellulose I, with the dashed lines indicating the hydrogen-bonding network in the sheet parallel to the ac plane. Adapted from Blackwell (1982).BFigure 1.1 The structure of celluloseIELLOBIOSE i— GLUCOSE —C3Cellulose was one of the first substrates to be analysed by X-ray diffraction, however,due to difficulties in interpretation of diffraction patterns, the early models of nativecrystalline cellulose differed on whether glucan chains within microfibrils are organisedin a parallel or antiparallel fashion (Meyer and Mark, 1928; Meyer and Misch, 1937).Recent models using improved techniques of computer analysis of the X-ray datasupport the idea of the parallel arrangement of glucan chains (Gardner and Blackwell,1974; Blackwell, 1982; Sarko, 1986). Additional evidence of the parallel arrangementof glucan chains comes from silver labeling of reducing chain ends of Valoniamacrophysa cellulose (Hieta et al., 1984) and bacterial cellulose (Kuga and Brown,1988) and from observations that the enzymatic degradation of Valonia macrophysacellulose by enzymes from Trichoderma reesei which degrade cellulose from the non-reducing ends of the chains proceeds from one end of the crystals (Chanzy andHenrissat, 1985).Cellulose chains can undergo packing to form at least four different crystallinestructures, depending on the treatment of the cellulose sample (Marchessault and Sarko,1967; Blackwell and Marchessault, 1971). The predominant naturally occurringcrystalline form of cellulose is termed cellulose I. Figure 1.1 shows the model proposedfor cellulose I by Gardner and Blackwell, 1974. Cellulose I has now also beenrecognised to be composed of a mixture of two different crystalline forms, termed Iand 18, which are distinguished by their intermolecular hydrogen-bonding patterns.These two crystalline forms of cellulose I occur in different proportions in most nativecelluloses; the ratio being dependent on the origin of the cellulose. For example,cellulose produced by primitive organisms, such as Acetobacter, have higher amounts ofthe ‘a crystalline form, whereas celluloses from higher plants, such as cotton, havegreater proportions of the 18 form (Atalla and VanderHart, 1984; Atalla, 1989).Cellulose fibres can contain very highly ordered crystalline regions as well asless well ordered ‘amorphous’ regions (Coughlan, 1985). These amorphous regions are4areas where the crystalline structure is disrupted and can be caused by natural surfaceirregularities, chain end dislocations, chain-bonding irregularities such as an occasional1, 6 bond, or strained and twisted areas of the crystalline structure (Turková, 1978;Chanzy, 1990). The degree of crystallinity of cellulose fibres varies, depending on thesource of the fibres. For example Valonia macrophysa cellulose has a crystallinityestimated to be almost 100% (Kulshreshtha and Dweltz, 1973), whereas Ramie andcotton celluloses are estimated to be about 70% crystalline (Cheek and Roussel, 1989;Wood, 1988).A number of different types of cellulose were used in this study. Theircharacteristics will be described briefly. Cotton, which is 95-99% cellulose (Young,1986), is one of the most commercially important cellulosic fibres. Cotton fibrescontain a primary cell wall and 2-3 layers of secondary cell wall surrounding a centrallumen (Nevell and Zeonian, 1985). Native fibres have an outer waxy cuticle which isremoved during pretreatment (Wood, 1988). Ramie fibres are similar to cotton fibresbut have only a single wall and the waxy cuticle is replaced with gum (Cheek andRoussel, 1989). The walls of both fibres contain cellulose microfibrils, approximately25nm in diameter in cotton (Kolpak and Blackwell, 1975) and 5nm in Ramie (Chanzy,1990). CF1TM is a fibrous cellulose made from highly purified cotton. The fibresaverage 200pm in length and 13.5jim in diameter and the crystallinity is estimated at 75-85% (E. Heilweil, Whatman mc, personal communication).Soluble cellulose derivatives, for example carboxymethylcellulose (CMC) andhydroxyethylcellulose (HEC) are also available commercially and have a wide variety ofapplications (Davidson, 1980; Whister and BeMiller, 1973). These cellulose derivativesare produced through the reactions of the hydroxyl groups of the anhydroglucose ring.Their solubility depends on the nature of the substituent, the degree of substitution (DS)and the uniformity of distribution (Ott et al., 1954). CMC is a negatively chargedpolymer, whereas HEC is uncharged.5Other forms of cellulose used in this study include Avicel, a microcrystallinecellulose obtained by partial acid hydrolysis of wood fibres (Wood, 1988). Thiscellulose preparation consists of rod shaped particles and irregular aggregates and has adegree of crystallinity estimated at 50% (Kulshreshtha and Dweltz, 1973; Lee et al.,1982; Ooshima et al., 1983). Bacterial microcrystalline cellulose (BMCC), obtainedfrom the bacterium Acetobacter xylinum, is used as a substrate in binding analyses inthis study. Cellulose is synthesised by these organisms as microfibrils that form intoribbons outside the cell wall (Cannon and Anderson, 1991). In contrast to Avicel,BMCC prepared fromAcetobacterxylinum cultures (Hestrin, 1963) consists of auniform suspension of microfibrils of high purity and crystallinity (White and Brown,1981; Henrissat and Chanzy, 1986; Kuga and Brown, 1987).1.2 Cellulose binding domainsMicroorganisms that degrade cellulose typically secrete a whole arsenal ofenzymes which act synergistically to hydrolyse cellulose and associated plant cell wallpolysaccharides into their constituent sugars. The Gram positive bacterium,Cellulomonasfimi, is one such organism that utilises cellulose as a carbon source andsecretes a number of B-i ,4-glycanases which have been studied extensively in thislaboratory. The genes for a number of C.Jlmi cellulases have been cloned and expressedin E. coli for characterisation. They include five endo 13-1,4-glucanases (CenA, B, C, Dand E), an exo B-i,4-xylanase/B-1,4-glucanase (Cex) and a B-l,4-exo cellobiohydrolase(CbhA) (Langsford et al., 1984; Gilkes et al., 1984; Wong et al., 1986; Owalabi et al.,1988; Meinke et al., 1991; Moser et al., 1989; Coutinho et al., 1991; Meinke et al., 1993;Shen et al., 1994; ONeil1 et al., 1986; Meinke et al., 1994).6A comparison of B-i ,4-glycanases involved in the hydrolysis of cellulose revealsthat most have a modular structure, which include a catalytic domain and a substratebinding domain separated by a short linker peptide (Van Tilbeurgh et al., 1986; Knowleset al., 1987; Gilkes et al., 1988; Béguin, 1990; Tomme et al., 1988; Gillces et al., 1991).All of the enzymes studied from C.Jlmi to date typify B-1,4-glycanases in having thismuitidomain structure, comprising functionally distinct catalytic domains and cellulosebinding domains (CBDs). The enzymes CenB, C and D contain additional domains ofpresently undetermined function (Meinke et al., 1991a; Coutinho et al., 1991; Meinke etaL, 1993).CenA and Cex, two of the major extracellular enzymes produced by C.fimi, havebeen characterised in some detail (Langsford et al., 1984; Gillces et al., 1984a; Langsfordet al., 1987; Gilkes et al., 1988, 1989). The enzymes are similarly organised with eitheran amino-terminal or carboxy-terminal CBD separated from the catalytic domain by aproline/threonine linker sequence (Figure 1.2). The ProfThr linkers are virtuallyidentical in sequence and the CBDs share 50% homology, whereas the catalytic domainsare unrelated (Wong et al., 1986; O’Neill et al., 1986; Warren et al., 1986; Henrissat etaL, 1989). The two domain structure of these enzymes was first established byproteolysis. Both CenA and Cex bind tightly to cellulose (Gilkes et al., 1992). Anextracelluiar protease cleaves CenA and Cex at the C-terminal of the Pro/Thr linker togive a large fragment (the catalytic domain) which is catalytically active but does notbind to cellulose, and a smaller fragment (the cellulose binding domain or CBD) whichbinds to cellulose but has no hydrolytic activity (Gillces et al., 1988).Gene fragments encoding CBDCenA and CBDCex (the cellulose bindingdomains of CenA and Cex, respectively) have been cloned and expressed in E. coil.Binding analyses of the adsorption of these polypeptides and the proteins CenA and Cexto BMCC, indicate that the affinity of CenA and Cex for cellulose can be attributedalmost entirely to the interaction of the respective CBDs with the substrate (Gilkes et al.,7CENACBD F:/1: CATALYTIC DOMAINCATALYTIC DOMAIN p/ CBD COOH316 335 443Figure 1.2 The structure of CenA and CexThe domain structures of mature CenA and Cex are shown, indicating the cellulosebinding domains (CBDs) which are separated from the catalytic domains by theproline/threonine linkers (Pm. The numbers refer to the amino-acid positions in theproteins.1992; Ong et al., 1993; Din et al., 1994). The independent binding ability of the CBDsof CenA and Cex also has practical significance, The CBDs retain their cellulosebinding properties when fused to heterologous proteins, making them valuable tools forprotein purification and enzyme immobilisation (Greenwood et aL, 1989; Ong et al.,1991; Greenwood et al., 1992; Kilburn et aL, 1993).The fungus Trichoderma reesei is perhaps the most extensively studiedcellulolytic microorganism. Its cellulase system includes two secreted exoglucanases(cellobiohydrolases) and at least three endoglucanases (Henrissat et al., 1985; Knowleset al., 1987; Claeyssens and Tomme, 1989; Teeti et al., 1991). These enzymes also haveH2N1H2N1COOH112 134 418CEX8a common structural organisation of a large catalytic domain and a smaller cellulosebinding domain linked by a peptide composed of proline and hydroxy-side-chain amino-acids. The T. reesei CBDs share a 70% conserved sequence of about 30 amino-acids(Knowles et al., 1987). Proteolysis studies of cellobiohydrolases I and II (CBHI and II)have also demonstrated that the larger proteolytic fragment retains catalytic activity butno longer binds to cellulose, whereas the smaller fragment binds to cellulose, as does a33 amino-acid synthetic peptide corresponding to the conserved CBD sequence (Tommeet al., 1988; Johansson et al., 1989).The cellulose binding domains described to date can be grouped into sevenfamilies on the basis of amino-acid sequence similarities (Table 1.1; Dr. P. Tomme,unpublished data). Some of these have been shown to bind to cellulose; others areincluded only on the basis of similarity in amino-acid sequence. Family II contains atleast 28 members, all of bacterial origin, and all approximately 100 residues in length.The CBDs of CenA and Cex are typical members of this family. The CBDs of T. reeseicellulases are all members of family I.The prevalence of independent cellulose binding domains within cellulasesimplies that they play an important role in the degradation of cellulose. The expressionof the coding regions of the CBDs from CenA (Gilkes et al., 1992; Din et al., 1994),CenB (Dr. C. Ramirez, unpublished data), CenC (Coutinho et al., 1992) and Cex (Ong eta!., 1993) has allowed the detailed characterisation of these domains in terms of theirbinding properties and possible roles in hydrolysis of cellulose. This thesis describes thecloning and characterisation of the CBD of CenA.The mechanisms by which CBDs adsorb to cellulose are presentlyunknown. The understanding of the mechanism of adsorption and the structure of theCBDs should help to establish the role of CBDs in cellulose degradation. Small angle Xray scattering (SAXS) analysis shows that CenA and Cex are both tadpole-shaped, withthe catalytic domain forming the head region and the cellulose binding domain and the9Table 1.1 Cellulose binding domain familiesThe cellulose binding domains (or putative CBDs) described to date (June, 1994) havebeen placed into seven families based on amino-acid similarity. The names of theorganisms and the abbreviated names of the enzymes of which the CBDs have beendescribed are indicated in the two left columns. The location of the CBD within eachprotein is indicated as N-terminal (N), internal (I) and C-terminal (C), and the size of theCBD as the number of amino-acid residues (under the column labelled ‘AA’). Thecolumn headed ‘Binding’ indicates the CBDs for which binding of the polypeptide tocellulose has been determined. Table courtesy of Dr. Peter Tomme.10Family Organism Enzyme Location AA Binding ReferenceAgaricusbizporus Cell C 36 RaguzetaL(1993)Agaricus bisporus Ce13 N 36 Chow et at. (1994) Gene Bank onlyHumicola grisea Cbbl C 36 Azevedo eta?. (1990)Peniciliiumjanthinellum CbIil C 36 ChristophetaL(1993)Phanerochaete chrysosporium CbM C 36 Sims et aL (1988)Phanerochaele civysosporium CbbI-2 C 36 Covert et aL(1992)Phanerochaele chrysosporium Cbhl-3 C 36 Covert et aL(1992)Phanerochaete chrysosporium Cbhl-4 C 36 Vanden Wymelenberg et at. (1993)Trichoderma koningil Cbhl C 36 Wey et al.(1994)Trichoderma longibrachialum CbhI C 36 Gonzales et aL(1992)Trichoderma reesei Cbhl C 36 + Shoemaker et al.(1983)Trichoderma reesei CbhlI N 36 + Chen et aL (1987)Trichodermareesei EngI C 36 + Pentillä eta!. (1986)Trichoder,na reesei EngIl N 36 + Saloheimo et a!. (1988)Trichoderma viride Cbbl C 36 Cheng et a?. (1990)Butyrivibriofibrisolvens Endl C 95 Bcrger et aL (1989)Celtulomonasfimi CcnA N 106 + Wong etal, (1986)Ceflulomonasfiini CaiB C 103 + Meinke et at. (1991)Cellulomonasfimi CcnD C 105 + Meinkeetai(1993)Cellulomonasfini CenE C 104 + Shea et at. (unpubl.)CelluJomonasfimi CNA C 106 + Meinke eta!. (1994)Cellulomonasfimi Ccx C 106 + ONeill et at. (1986).Ce(lulonionasfimi XynD I/C 90 + Millward-Sadler et at. (1994)Cetlutomonasfiavigena CIIX C 106 Al-Tawheed (1988)Clostridium longisporum CeIA C 97 + Mittendocf and Thomson (1994)Clostridiu,n cellidovorans EngD C 108 + Hamamoto et at. (1992)Dictyostelium discoideum 270-1 la C 98 Giorda et at. (1990)270-1 lb 1 106 Giorda et at. (1990)Microspora bispora CeIA C 100 Yablouski€t at. (1988)Pseudomonasfluorescens EgIA C 100 Hall and Gilbert (1988)Pseudonionasfluorescens CdB N 102 + Gilbert et at. (1990)Pseudomonacfluorescens Ce1C N 99 + Ferreira et a!. (1991)Pseudomonasfluoresceas XynA N 101 + Hall el at. (1989)Pseudonwnasfluoresceris XynB/C N 99 + Kellet et at. (1990)Pseudomonasfluorescens XynD N 102 Feereira et at. (1993)Streptomyces lividans CeIA N 108 Theberge et aL (1992)Streptomyces lividans ChiC N 105 Fuji and Miyashita( 1993)Streptomyces tividans XInB C 86 Shareck et at. (1991)Thermomonosporafusca El C 96 + Irwin el at. (1994)Thermomonosporafusca E2 C 96 + Lao et at. (1991)Thermomonosporafusca E4 C 104 + Irwin et a?. (1994)Thermoinonosporafusca ES N 103 + Lao et at. (1991)Thermomonosporafusca XynA C 86 + Irwin el at. (1994)ifi Bacillus lautus CeIA C 150 + Hansen et at. (1992)Bacillus subtilis DLG Eng C 132 Robson and Chambliss (1986)Bacillus subtilis N-24 Eng C 132 Nakamura eta!. (1987)Bacillus subtilis PAP115 Eng C 132 MacKay e( at. (1986)Coidocdllurn sacciw.rotyticurn CeIB I 136 Saul et at. (1989)Ce(lulomonasfimi CenB 1 131 + Meinke eta!. (1991)Clostridium cellulovorans CbpA N 161 + Shoseyov eta!. (1992)Clostridium slercorarium CeIZC I 144 + Jauris eta!. (1990)Clostridiurn stercorarium CeIZC C 133 Jauris eta?. (1990)Clostridiurn ther,nocetlum C1pA 1 156 Grengross eta?. (1993)Clostridiurn thermocelturn CipB C 167 + Poole et aL (1992)Clostridiu,n lher,nocellwn Cell C 150 Hazlewood eta!. (1992)Erwinia carotovora CelV C 156 Cooper and Salmond (1993)IV Ce(lulornonasfimi CenC N x 2 148 + Coutinho et aL (1991 and 1992)Clostridium cellutolyticum CeICCE N 168 Bagnara-Tardifet at. (1992)Myxococcusxanthus EgI N 140 Quilletet at. (1994)Streptomyces reliculi Cell N 125 Schlochtenneier eta!. (1992)Thermornonosporafusca El N 141 Lao eta!. (1991)V Clostridium Ihermocellum CeIE I 240 + Duffant eta!. (1991)VI Erwinia chrysanthemi EgZ C 63 + Py el a!. (1991)VII Dictyoslilium discoidum CeIA N 152 + Ram alingasn et aL (1992)11ProfThr linker forming an extended tail region of approximately lOnm. It is presumedthat this tail region is a hairpin structure which is stabilised by a disuiphide bondbetween cysteine residues near each end of the CBDs (Pilz et at, 1990; Shen et al.,1991; N. R. Gilkes and M. Schmuck unpublished data). The overall dimensions of theseproteins are similar to those of CBH I and II from T. reesei (Abuja et at, 1988; Shen etat, 1991). The three dimensional structure of the synthetic binding domain of CBHIfrom T. reesei has been determined by NMR (Kraulis et al., 1989). It is wedge shaped,stabilised by two disulphide bonds, with one face being flat and hydrophilic and theother face being more hydrophobic. The only three dimensional structure of a bacterialCBD elucidated to date is that of the CBD of Cex, which has also been determined byNMR spectroscopy (G.Y. Xu, L. Kay and J. Carver, manuscript in preparation). Thesecondary structure is composed of an anti-parallel 13-stranded barrel. The overall shapeof the domain is more irregular or globular than that of the CBD of CBHI.Analysis of the amino-acid sequences of some of the family II CBDs reveals thatthey have several features in common, including large numbers of hydroxy-side-chainamino-acids, very few charged residues and several highly-conserved residues includingaromatics, glycine and asparagine (Figure 1.3). There are also two highly-conservedcysteine residues, which form a disulphide bond in the CBDs of CenA and Cex (Gilkesetal., 1991a).Aromatic residues are known to be important for several protein-carbohydrateinteractions (for example, Johnson et al., 1988; Svensson et al., 1989; Abbot and Feizi,1991; Spurlino et al., 1991; Brummell et al., 1993; Maenaka et al., 1994; Coutinho andReilly, 1994). Numerous van der Waals’ contacts are formed in protein-carbohydrateinteractions. Some of these arise by the stacking of aromatic residues against the facesof sugar rings. These stacking interactions are thought to be important in the stabilityand specificity of protein-carbohydrate complexes (Quiocho, 1986; Vyas, 1991;Spurlino et al., 1992). In the present work two of the four conserved tryptophan residues12Figure 1.3 Amino-acid sequence comparison of some family II CBDs from bacterial B-1 ,4-glycanasesThe CBDs (or putative CBDs) are from C.fimi CenA (Wong et al., 1986); Cex (O’Neillet al., 1986); CenB (Meinke et al., 1991), and CenD (Meinke et al., 1993); a translatedopen reading frame Cflx from C.fiavigena (Al-Tawheed, 1988); Clostridiwncellulovorans endoglucanase D (Harnamoto et al., 1992); Pseudomonas fluorescenssubsp. cellulosa endoglucanases A (Hall and Gilbert, 1988) and B (Gilbert et al., 1990),cellodextrinase C (Ferreira et al., 1991), and xylanases A (Hall et al., 1989), B (Kellet etal., 1990) and C (Kellet et al., 1990); Bulyrivibrio fibrisolvens endoglucanase 1 (Bergeret al., 1989); Microbispora bispora endoglucanase A (CeIA) (Yablonski et a!., 1988);Thermonospora fusca endoglucanases 2 and 5 (Lao et al., 1991); and Streptomyceslividans endoglucanase (Ce1A) (Theberge et al., 1992). The sequences fromPseudomonasfluorescens subsp. cellulosa xylanases B and C are identical. The namesof bacteria and enzymes are indicated on the left in abbreviated form. All sequences arenumbered from the first amino-acid residue of the mature enzymes. Positions of Ctermini, deduced from the occurrence of a stop codon in the corresponding gene, areindicated by asterisks. The consensus shows amino acid residues that are identical innine or more sequences, or residues that are similar in 11 or more sequences. Symbolsdenote consensus residues that are identical (•) or similar (+) in all 17 sequences.13CfiCenA A PG C - — K V DY A VT NQW PC C F C A N VIII Nb C - D P VS SW 34CfiCex p A C C -— Q V L W G V - N Q 44 N I C F I A N V I V K N I S S A P V D C 44 371CfiCen8 TPSC- TVVYS-TNSWNVGFTGSVKITNTCTTPL-TW 942CfiCenL) TGSC - -vVTYT-ANCWSCGFTAAVTLTNTCTTALSCW 632Cf1CIfX I C S C - - K V £ Y N A S S - 44 N T C F I A S V K V I N T C T I A L N C 44 240CceEngD QSAV- -EVTYAITNSWGSCASVNVTIKNNCTTPINCW 411Pf1EGA GGNC----QYVVTNQWNNCFTAVIRVRNNGSSAINRW 894Pf1EGB A A V C - - - - £ y K V T N H 44 G S G F I A S I K I T N N C S S I I N G 61PfICeIC - C C - - - - £ Y V V T N S 44 G S C F T A A I K I T N S I S S V I N C W 31PflXynA TATC - - - -SYNITNEWNTGYTCI)ITITNRGSSAINGW 33PflXynB/C A C - — - - T Y T I 1) 5 H 44 S I C F I A N I T L K N 0 T C A A I N N 44 31BfiEndl GALK - --AEYTI-NNWCSGYQVLIKVKNDSASRVDCW 429MbiCeIA CRAC--EATYALVNQWPCGFQAEVTVKNTCSSPINGW 395TfuEGE2 RLLCGVHGDVHDANEWNDGFQATVTVTANQN - - ITGW 343TtuEGES ACLT - -ATVTKESS-WDNGYSASVTVRNDTSSTVSQW 34S1iCe1A ATCC--KAEYTITSQWEGCFQAGVKITNLG-OPVSGW 34a a + + + aConsensus C Y N W C F A I V N C S S I C 44CfiCeaA KLOWTYTCfiCex TLTFSFPCfiCenB TLGFAFPCfiCenD TLCFAFPCflClfx TLTFPFACceEngo TLKWTMPPflEGA SVNWSYSPf1EGB SVSWNYTPflCe1C NVSWQYNPflXynA SVNWQYAPflXynB/C N V N 44 Q Y SDfiEndl TLKISKSIbiCelA TVQWTLPTfuEG2 TVTWTFTTfuEG5 EVVLTLPS1iCe1A TLGFTMP+Consensus I L WA -GQR1QLWNGTASTNCCQVSVTSLPWNC 70S -CQQVTQAWSSTVTQSCSAVTVRNAPWNG 407S -CQQVTQCWSATWSQTCTTVTATGLSWNA 978S -CQTLTQGWSARWAQSCSSVTATNEAWNA £68N -GQTVQQCWSADWSQSCTTVTAKNAAWNC 276I - NQT I TNMWSAS FVASGTTLSVTNAGYNG 447D-CSR1TNSWNANVTGNNPY -AASALGWNA 9290 -CSRVTSSWNAGLSGANPY-SATPVCWNT 96S - -NRVTNLWNPNLSGSNPY -SASNLS44NG 65I - - NR LSS SW NA NVS CS N P Y - S AS NLSW NC £7S --NRMTSCWNANFSCTNPY- NATNI4SWNC 65E - -VKIDSSWCVNIAEEGGYYVITPMSWNS 524S - G Q S I I Q L 44 N C 1) L S I S C S N V T V K N V S 44 N C 431D- C Q I I I N A 44 N A 0 V S I S C S S V I A K N V G H N G 379C - CT TV A Q V 44 N A Q H T S S G N S HI FT C V SW N S 70DAGQRLVQGWNATWSQSGSAVTACGVDWNR 71+ aCQ VTQ W AaS SC TA WNCfiCenACf iCexCfiCenBCf iCenDCflclfxCceEngDPflECAPf1ECBPf1Ce1CPflXynAPflXynI3/CBfiEndlMbiCelATfu EG2.Tfu EGSS1iCe1ASI PTSI PATLQPV LA PSLAATI AAN I Q PSI P1TIQPNIQPS lAPSLEPNVPATLSQTI PPTLAT+-CGTA-GCTA-CQST-CA S V-GQTVNCGTQ-C Q TA-CSSV-GQTV- CV S V-CQS I-SASV-GGST—CAST- CC TA-GAS AWAHTliPHIMNYSKGNCTNKNKNS FCFNGSQFCFNGS0 I C F N C SEl C F S GTDIG F NGA-SFGFNINEFCFQGTE FGVGNE F C FQG VS FCFQVNSFGLQGEDFGIQGSSFGFLGSE F C F VG LSSGFIAS0 LGFVGS-CSNPTPASFSLNGTT- CT N A APT A F S L NC T P-GTNTNPASFTVNGEV-CTNTAPATFTVCGAT-GTNNKPASFTLNGAT- C V L S K PIG FT VN GTEACSRQVPA---VTCSVSSRAQVPA---VTGAISGTVESPT- - -VNCAACC S A ER PS - - - VC CS IGSTAERPT---VTGAA- GS - I GTS - -- VNISV-GTCQLSSS ITCTGTCTVC *CC *CITR*CTVG *CIVIC *CQ *CGGQCTCC SC SC NSACSAS *+ a10644310127043124849611309810199551460394103108*CS -GEPTHCT- - - INGAPCDEGFT-GANPAPTSFTLNGATCSGSVConsensus T I C S FGF C C PT V G C14in CBDCenA (Figure 1.3) were targeted for site directed mutagenesis, in an attempt toassess their role in the binding interaction of the CBD with cellulose (Din et al., 1994).Previous results in which aromatic residues have been shown to be important in thebinding interaction of CBDs with their substrates include those from the T. reesei CBHI,where conserved tyrosine residues were mutated chemically and by site directedmutagenesis (Claeyssens and Tomme, 1989; Reinikainen et al., 1992) and from the CBDof a Pseudomonasfluorescens subsp. cellulosa xylanase (Xylanase A) where three offive conserved tryptophan residues were shown to be important for binding (Poole et aL,1993).Hydrogen-bonding is another dominant force in stabilising protein-carbohydratecomplexes (Quiocho, 1986; Vyas, 1991; Johnson et al., 1988). Water molecules mayalso contribute to the stability of a protein-carbohydrate complex by mediating H-bonding between the functional groups of the protein and sugar (Quiocho et al., 1989)Deletion of the CBD from some 13-1,4-glucanases reduces their activity oninsoluble cellulose but not on soluble substrates (Tomme et al., 1988; Gillces et al., 1988;Ghangas and Wilson, 1988; Hamamoto et al., 1992; Wilson, 1992). Discrete bindingdomains are also found in enzymes which hydrolyse other insoluble carbohydrates, suchas chitinases and amylases. Removal of the binding domains from some of theseenzymes also decreases their activities against insoluble substrates (Svenssen et al.,1989; Watanabe et al., 1990). These observations imply that binding domains contributein some way in the hydrolysis of insoluble polysaccharides.The role of CBDs in cellulose hydrolysis is at present unclear. The CBD maysimply increase the effective enzyme concentration on the substrate’s surface. However,a more active role of the binding domain in some cellulases has been proposed. It hasbeen recognised for some time that the efficient hydrolysis of crystalline celluloserequires the synergistic action of cellulolytic components with various specificities. Theoriginal model of synergistic enzymatic degradation of crystalline cellulose put forward15the idea that the first step in the process involves the action of a non-hydrolyticcomponent of the cellulase system, called Ci, which destabilises the cellulose structure,rendering it accessible to hydrolytic enzymes of the system, called C, (Reese et al.,1950). It has also been recognised that cellulose undergoing enzymatic attack showschanges in physical properties, including fibre swelling (Wood, 1989) and fragmentationto short fibres (Halliwell, 1965; Lui and King, 1967; Kyriacou et al., 1987; Walker et al.,1992), before any significant amounts of sugars are released. These data lent support tothe idea that the disaggregation of cellulose chains is an essential preliminary step in thehydrolysis of crystalline cellulose and that the binding domains of cellulases may beacting as the dispersing agents to liberate cellulose chains for hydrolysis by the catalyticdomains (Knowles et al., 1988; Reinikainen et al., 1992).The CBDs of CenA and Cex have been demonstrated to prevent the flocculationof microcrystalline bacterial cellulose (Gilkes et al., 1993), indicating that these CBDscan promote substrate dispersion. Evidence presented in this thesis shows that the CBDsof CenA and Cex cause the release of cellulose particles from cotton and the disruptionof cellulose fibres (Din et al., 1991). Data are also presented that show that the CBD ofCenA synergises with the catalytic domain, p30, in hydrolysing cotton fibres (Din et al.,1994a). Together these results provide the first demonstration of the mechanicaldisruptive properties of a CBD.1.3 Objectives of this studyThe overall objective of this study was to investigate the properties of the CBDof CenA independent of the catalytic domain. Specific objectives in terms of thecharacterisation of CBDCenA involved: (a) an evaluation of the role of two conservedtryptophan residues in the adsorption of the polypeptide to cellulose as part of an initialattempt to probe the mechanism by which CBDs interact with their substrate; (b) aninvestigation of the effect of CBDCenA on cellulosic substrates to determine if thispolypeptide has a functional role in cellulose hydrolysis other than simply to effectadsorption of CenA to the substrate.16172. Materials and methods2.1 Chemicals, buffers and enzymes.All the chemicals used were of analytical or HPLC grade. The buffers wereprepared according to Sambrook et al (1989). Buffers and conditions for restrictionenzymes were as recommended by the manufacturers. AviceP PH 101 was from FMCInternational and CFl was from Sigma. Bacterial microcrystalline cellulose (BMCC)was prepared from cultures of Acetobacter xylinum, as described by Hestrin (1963).2.2 Bacterial strains, plasmids and phageAll DNA manipulations and gene expression were carried out in the Escherichiacoli strains listed in Table 2.1. The phage and the plasmids that were used orconstructed during the course of this study are listed in Table Media and growth conditionsThe E. coil strains used in this study were grown in LB (Miller, 1972) and TYPbroth (16g tryptone, 16g yeast extract, 5g NaCl and 2.5gK2HPO4per litre). Solidmedium contained 1.5% (w/v) agar. The antibiotic used was ampicihin (BoebringerMannheim) at a final concentration of 100 .tg/rn1; isopropyl-B-D-thiogalactoside (IPTG,Sigma), at a final concentration of 0.1mM, was used to induce gene expression.Bacterial cultures were grown at 37°C, for a period of 16 hours, unless otherwise stated.18Table 2.1 E. coil strains.E.coli strain Genotype ReferenceJM1O1 supE thi A (lac-proAB) F’[tra D36 Yanisch-Perron etpro AB iaclq lacZ AM 15] al., 1985RZ1032 HfrKL16PO/45[lysA(61-62)] dutl Kunkel et al., 1987unglthilrelAl Zbd279::TnlO supE44Table 2.2 Plasmids and phagePlasmid Characteristics ReferencepUC18-1.6cenA Encodes CenA; expressed as lac Z Guo et al., 1988fusionM13mp18-1.6cenA Carries complete coding sequence of This studyceiiA; used for the production of singlestranded DNAPUC18CBDe, Encodes the cellulose binding domain This studyof CenAPUC18CBDCeMW14A Encodes the mutant polypeptide This studyCBDCenAW14A.PUC18CBDCepAW68A Encodes the mutant polypeptide This studyCBDCenAW68A19Table 2.2 ContinuedPhage Characteristics ReferenceM13K07 Kanr Vieira and Messing,19872.4 Recombinant DNA techniquesAll the recombinant DNA work was generally carried out as described bySambrook et al., 1989. DNA fragments were isolated from agarose gels and thenpurified using GeneClean or MerMaid kits (Bio 101, La Jolla, CA). Competentcells were prepared and transformed as described by Miller (1987).Oligonucleotides were synthesized on an Applied Biosystems automated DNAsynthesizer, model 380B (Oligonucleotide Synthesis Lab, U.B.C.) and then purified byreverse phase chromatography on Sep-Pak C-18 columns (Millipore).Growth of phage Ml 3 and the isolation of its single stranded and doublestranded DNA has been described in detail by Messing (1983) and Zoller et al (1983).The transfection of E. coli JM1O1 with replicative form (RF) DNA from M13 wascarried out as described by Zoller et al., (1983), using 5-bromo-4-chloro-3-indolyl-B-D-galactopyranoside (X-Gal) in the top agar to differentiate recombinant and nonrecombinant phage.The primer extension technique used in the construction of the vector expressingwild-type CBDCenA was as follows; 20 pmoles of single stranded M13mp18-1.6cenAwere mixed with an equimolar amount of primer in 200j.tl of buffer containing 50mMNaCl, 50mM Tris-HC1 (pH 8.0), and 10mM MgCl2. The DNA was denatured byheating the mixture to 90°C for 5 minutes and then allowed to anneal at 50°C for 2-4hours. Fifteen units of the Klenow fragment of E. coli DNA polymerase I were then20added to the mix, along with all four dNTPs (0.25mM final concentration) and DTT(5mM fmal concentration), to initiate primer extension. The reaction mixtures weredeproteinised by extraction with phenol/chloroform and chloroform and the DNA wasrecovered by ethanol precipitation.Single stranded uracil-containing DNA, used for site directed mutagenesis, wasprepared in E. coli RZ1032 (Kunkel et al., 1987). Single and double stranded DNAwere sequenced by the dideoxy method, using a Sequenase 2.0’ kit (United StatesBiochemical). Double-stranded DNA for sequencing was prepared from alkaline celllysates by caesium chioride/ethidium bromide ultracentrifugation. Single stranded PTZDNA was purified using the following procedure (Donald Trimbur, personalcommunication); 2m1 of TYP broth containing ampicillin and 5x108 pfu/ml of M13K07were inoculated with a single colony from an LB Amp plate and incubated at 37°C for 1hour. Kanamycin (final concentration 50igIrnl) was added and incubation wascontinued overnight (approximately 16 hours). 1 .5ml of the cultures were transferred to1.7m1 Eppendorf tubes and then centrifuged (14,800g, 10 minutes at room temperature)to pellet the cells. The supernatant was transferred to a new tube. Ammonium acetate(3.75M)- 2.5% PEG 6000 solution (200p1) was added and the tube was then incubatedon ice for 30 minutes. The phage particles were pelleted (15,000g, 10 minutes at 4°C)and the supernatant removed. The pellet was resuspended in 20il of TE buffer (10mMTris-HC1, 1mM EDTA, pH 8.0) and 200iil of 4M NaC1O4 added. This mixture wasincubated at room temperature for 5 minutes and then the DNA was bound to a glassfibre filter (GF/C, Whatman) by vacuum filtration. This was achieved by cutting theglass fibre paper into disks to fit into the wells of a 96 well microtitre plate. The wellsof the microtitre plate had holes in them, made with a hot 18- gauge needle, to allowliquid to pass through into a fitted container. The solution containing the DNA wasapplied to the filter paper disks and the microtitre plate was then connected to a vacuumpump. The filters were washed several times with 70% ethanol and finally removed21from the microtitre plate with sterile tweezers and allowed to air dry. They were thenplaced into 0.5m1 Eppendorf tubes, pierced at the base with a 21-gauge needle, and 20ilof 0. lx TE was added to elute the DNA from the filters. After 5 minutes incubation atroom temperature, the 0.5m1 tubes were placed into 1.7m1 Eppendorfs and spun brieflyin a microfuge, to collect the DNA solution in the larger tubes. The yield of DNA wasestimated from agarose gels.Site-directed mutagenesis was carried out as described below (Zhou et al., 1990).Single-stranded uracil-containing DNA was prepared and the mutagenic primer was thenannealed to the template (0.3pmoles of template and 6pmoles of primer in lOxSequenase buffer (400mM Tris-HC1, pH 7.5; 100mM MgCl2; 500mM NaC1) by heatingto 70°C for 5 minutes and slowly cooling to room temperature over 30 minutes.In vitro synthesis of the complementary strand was achieved by adding lOx 17 DNApolymerase buffer (500mM TrisHCl, pH 7.5; 100mM DTT; 70mM MgC12; 10mM ATP;5mM each dATP, dCTP, dGTP, dTTP), 1.5 units of T4 ligase and 2.5 units of 17 DNApolymerase. This mixture was incubated at room temperature for 5 minutes and then at37°C for 1-2 hours. E. coli JM1O1 was then transformed with 2-5pi of this mixture.Single-stranded DNA was prepared from the transformants and was screened for themutations by dideoxy sequencing.2.5 Screening for gene expressionIM1O1 colonies were screened for the expression of CBDe, as follows: 2m1 ofan overnight culture (16 hours, 37°C) of each of the clones to be screened were used toprepare clarified cell extracts (see section 2.6). BSA was added to these extracts, to afinal concentration of 1%, and then 10i1 were spotted onto a 9.0cm diameter Whatman541 filter paper. 5i.tl of each extract was first applied to the filter paper and allowed toair-dry before another 5j.il was applied to the same spot. The filter papers were again22air-dried. As negative and positive controls, cell extracts from E. coli JM1O1 harbouringpUC18 or pUC18-1.6cenA (expressing intact CenA, Guo et al., 1988) respectively, wereincluded on each filter. The filter papers were then placed into glass petri dishescontaining a 1/3000 dilution (diluted in phosphate-buffered saline containing 3% BSA)of the primary antibody, which was rabbit polyclonal antiserum raised against CenAEC2(Wong et a!., 1986, Whittle et al., 1982). This antibody had been preadsorbed withE. coli JM1O1 lysate (Helfman and Hughes, 1987): The filter papers were incubatedwith this antibody for 3 hours at room temperature on an orbital shaker. They were thenwashed (3 times, 10 minutes each) with PBS-0.5% Tween 20 followed by PBS. Afterthe washing steps, the filters were incubated with the secondary antibody, alkalinephosphatase conjugated goat anti-rabbit IgG (BRL), used at a dilution of 1/2000, for 3hours at 37°C. The filters were washed again with PBS-0.5% Tween 20 and PBS andthen developed as described by Blake et al., 1984. The filters were finally rinsed in PBSand air dried.2.6 Cell fractionationCell extracts for Avicel-binding assays and screening for gene expression(section 2.5) were made from 5m1 cultures (unless otherwise stated in the text) grownovernight at 37°C. Cultures were grown in medium containing l00ig/ml ampidihin and0.1mM IPTG. Cells were harvested by centrifugation (14,800xg for 10 minutes, 4°C)and the pellet was resuspended in 0.5m1 phosphate-azide buffer (50mM potassiumphosphate pH 7.0, 0.02% sodium azide). Cells were disrupted with a sonicator(SonifierTM model 350, Canlab) using three continuous pulses of 10 seconds each. Thecells were kept on ice during the whole procedure. The extract was clarified bycentrifugation at 14,800xg for 20 minutes at 4°C.23Cell extracts for large scale protein purification were prepared from cells grownin 20 litres of medium (containing l00.tgIml ampicillin) in a 35 litre batch fermentor(Chemap FZ 3000, Volketswil, Switzerland) at 37°C, for 16 hours. The 20 litres of LBwere inoculated with lOOml of an overnight culture, and IPTG (final concentration of0.1mM) was added an hour after the inoculation. Cells were harvested at 31,000xg(Sharples Inc.) and resuspended in 500m1 of phosphate-azide buffer. The cells weredisrupted with an APV Gaulin 30 CD homogeniser, at 16,000 psi. The cell debris wasremoved by centrifugation at 17,400xg, 4°C for 30 minutes. Streptomycin sulphate(final concentration 1.5%) was added to the supernatants to precipitate nucleic acids.After incubation overnight at 4°C, the precipitate was removed by centrifugation(17,400xg, 4°C for 30 minutes). The supernatants were again incubated overnight at4°C and clarified by centrifugation for 20 minutes.Cells were shocked osmotically as described by Nossal and Heppel (1966). Thecytoplasmic and periplasmic markers used as controls for this procedure were glucose-6-phosphate dehydrogenase and 13-lactamase, respectively. These enzymes were assayedas outlined in the Worthington enzyme manual (1988) and O’Callaghan et al., (1972).2.7 Purification of proteins.The polypeptides CBDCenA, CBDCenAW14A , and CBDCenAW68A werepurified from cell extracts by binding to CF1TM cellulose. The cellulose was firstwashed 10-15 times with distilled water to remove fines. This was achieved by mixing330g of the cellulose with water in a 4 litre beaker. The cellulose was allowed to settleand the supernatant was poured off. The cellulose was then equilibrated with 50mMpotassium phosphate-azide buffer. This prepared cellulose was mixed at roomtemperature with the clarified cell extract obtained from a 20 litre culture (see section2.6) in a 4 litre beaker. The mixture was stirred occasionally over a period of 1-3 hours24to allow binding of proteins to cellulose and was then filtered through a GFIC glass fibrefilter (Whatman) to remove all the proteins which had not bound to the cellulose. Thecellulose was resuspended in 1 litre of 1M NaCl (in phosphate-azide buffer), stirred for 1hour at room temperature and again filtered through a glass fibre filter. This NaCl washwas repeated and was followed by two phosphate- azide buffer washes. Bound proteinswere eluted with 500m1 of 6M guanidinium chloride (GdmCl) in 50mM potassiumphosphate, pH 7.0. This mixture was stirred occasionally over a period of 1-3 hours andthen filtered. The eluate was concentrated by ultrafiltration using a lkDa cut-offmembrane (Filtron OmegaTM series) and then centrifuged at 150,000xg, 4°C for 40minutes, using a Beckman Ti 50.2 rotor, to remove any cellulose fines. The supematantwas passed over a Superose 12TM size exclusion column (Pharmacia) using 6Mguanidinium chloride in 50mM potassium phosphate, pH 7.0, as the buffer. The columnwas controlled with a fast protein liquid chromatography (FPLCTM, Pharmacia) systemand the peak fractions were monitored by on-line absorbance readings at 280nm. Thepeak fractions corresponding to the desired polypeptides (i.e. CBDCenA,CBDCenAW14A, CBDCenAW68A) were pooled and the guanidinium chlorideexchanged with 50mM phosphate-azide buffer by ultrafiltration using a 1 kDa cut-offmembrane, until the concentration of guanidine hydrochloride was reduced to less than1pM.The purity of the polypeptide preparations was monitored by SDS-PAGE (seesection 2.8). The polypeptide solutions were then stored at 4°C.CenA, Cex and the polypeptides p30 and CBD.PTCenA were obtained fromEmily Kwan. CenA and Cex were prepared as described by Gilkes et al., (1988). p30was prepared as described by Pilz et al., (1990) and CBD.PTCenA as described by Gilkeset al., (1992). CBDex was obtained from Dr. Edgar Ong and was prepared as describedbyOngetal.,(1993).252.8 Detection of proteinsProteins were resolved by sodium dodecyl sulphate polyacrylamide gelelectrophoresis (SDS-PAGE) as described by Laemmli (1970). The gels were stainedwith Coomassie Brilliant Blue R250 (Sambrook et al. 1989). The molecular weightstandards used were as follows; myosin (205 kDa); 8-galactosidase (116-130 kDa);phosphorylase B (97.4kDa); bovine serum albumin (66 kDa); catalase (57.5kDa);glutamate dehydrogenase (53 kDa); alcohol dehydrogenase (45 kDa); ovalbumin (41kDa); glyceraldehyde-3-phosphate-dehydrogenase (36 kDa); carbonic anhydrase (29.4kDa); soybean trypsin inhibitor (20. lkDa); B-lactoglobulin (18.4 kDa); lysozyme (14.3kDa); cytochrome C (12.4 kDa); bovine trypsin inhibitor (6.2 kDa) and insulin, c and Bchains (2.9 kDa). Prestained molecular weight markers used were from BRL.Western blots were made as described by Towbin et al (1979). PBS, 3%BSAwas used in blocking and antibody binding and PBS, 0.05% Tween 20 was used in thewashing steps. The primary antibodies used were rabbit polyclonal antiserum againstCenAEC2 (Wong et al., 1986, Whittle et al, 1982) and rabbit polyclonal antiserumagainst CBDCenA (this study). These were used at dilutions of 1/3000 and 1/1000respectively. Both antibodies had been preadsorbed with E. coli cell lysates, asdescribed by Helfman et aL, (1987). The secondary antibody used was alkalinephosphatase-conjugated goat anti-rabbit IgG (BRL), used at 1/3000 dilution. Thewestern blots were developed as described by Blake et al., (1984).N-terminal sequencing analysis was carried out on polypeptides which had beenseparated by SDS-PAGE and then transferred onto ImmobilonTM -P membrane(Millipore corp). The transferred polypeptides were visualised with Coomassie blue(Matsudaira, 1990) and the desired bands excised from the membrane. The excisedbands were sequenced by automated Edman degradation using an Applied Biosystems26470A gas phase sequenator (Protein Microchemistry Facility, University of Victoria,Victoria, B.C.)2.9 Determination of protein concentrationProtein concentrations of cell extracts were determined as described by Bradford(1976). The concentrations of purified protein solutions were determined by absorbancemeasurements at 280nm (A280nm) using the extinction coefficients for thepolypeptides. The extinction coefficients (280nm; lmg.m11 1cm) were calculated fromthe Trp and Tyr content and the theoretical molecular masses of the polypeptides(Cantor and Schimmel, 1980). The molecular masses were calculated from the primarystructures of the proteins, deduced from DNA sequence analysis. The calculatedextinction coefficients of the proteins were: CBDCenA, 3.26; CBDCenAW14A, 2.78 andCBDCenAW68A, 2.78. The extinction coefficients for CBD.PTCenA, CenA, Cex, p30and CBDex (2.85, 2.64, 1.61, 2.38 and 2.3 respectively) were reported previously(Gillces et al., 1992, Pilz et al., 1990, and Ong et al., 1993). In all cases the predictedvalues were in good agreement with the extinction coefficients determinedexperimentally by the far-U.V. method of Scopes (Scopes, 1974).2.10 Sedimentation equilibrium centrifugationCentrifugation was performed using a Beckman model E analyticalultracentrifuge equipped with a photo-electric scanning system on polypeptide sampleswhich had been diluted in phosphate-azide buffer to achieve an A280 of approximately0.4. Mr values were determined using the sedimentation equilibrium techniquedescribed by Chervenka (1970).272.11 Binding analyses2.11.1 Avicel binding assaysAvicel (5mg) was used in each binding assay together with 100jil of the sampleto be screened for proteins which bound to cellulose (e.g. cell extracts prepared fromcultures grown on a small or large scale, as described in section 2.6). Before binding,the Avicel was washed three times with distilled water to remove fines by resuspendingin lml of water in a 1.7m1 Eppendorf tube and then centrifuging in a microfuge(14,800xg) for 2 minutes. The Avicel was then equilibrated with 50mM phosphateazide buffer. The binding was carried out on ice for one hour with mixing every 10-15minutes. The Avicel was pelleted in a microfuge for 10 minutes and the supernatantremoved (‘Avicel supernatant’). The pellet was then washed three times with 1M NaC1in phosphate-azide buffer and then three times with phosphate-azide buffer alone.Proteins which remained bound to the Avicel were recovered by boiling the Avicel inSDS-PAGE sample buffer for two minutes.2.11.2 Binding analysis with BMCCSemi-quantitative analysis of adsorption of CBDCenA, CBDCenAW14A andCBDCenAW68A to BMCC was carried out in a similar manner to Avicel binding assays(section 2.11.1). Purified polypeptide (100jig, in phosphate-azide buffer) was adsorbedto 1.5mg BMCC suspended in 1.5ml phosphate-azide buffer by incubation at 30°C for16 hours, with rotation at 2rpm. The suspension was centrifuged in a microfuge for 10minutes and the supernatant was removed. The unbound polypeptides in the supernatantwere precipitated with 10% trichloroacetic acid (TCA); 25 jig BSA were added to each28supernatant fraction as a carrier to facilitate recovery. The BMCC pellets were washedtwice with 1.5m1 of 1M NaC1 in phosphate-azide buffer, twice with phosphate-azidebuffer alone and twice with water, all at 30CC for 10 minutes. The desorbedpolypeptides from each of these washes were precipitated with TCA, as described. Allthe precipitated polypeptides were dissolved in loading buffer and analysed by SDSPAGE. The polypeptides which remained bound to the BMCC after the washes weredesorbed by resuspending and boiling the BMCC in SDS loading buffer. These toowere analysed by SDS-PAGE.For the estimation of adsorption parameters for the wild type and mutantCBDCenA polypeptides the procedure described by Gilkes et al (1992) was used. BMCC(1.5mg in 1.5m1 phosphate-azide buffer) was mixed with the purified polypeptide in2.Oml Eppendorf tubes (Island Scientific) at 30CC. The initial CBDCenA,CBDCenAW14A and CBDCenAW68A concentrations were in the range 2.60-5.74, 1.14-2.33 and 1.14-2.57 .tM respectively. The tube contents were mixed by rotation at 2 rpmfor 16 hours. The BMCC was then pelleted by centrifugation (14,800xg for 10 minutes)and the supernatant removed. The supernatants were centrifuged twice more in order toremove all BMCC particles. The concentration of the unbound protein [F] wasdetermined from the absorbance reading at 280nm and the extinction coefficients of theprotein (see section 2.9). The concentration of bound protein, [B], was determined fromthe difference between the initial protein concentration and [F].2.12 Fluorescence spectroscopy2.12.1 Binding to soluble substrates29The interaction of CBDCenA with cellohexaose (G6), carboxymethylcellulose(CMC) and hydroxyethylcellulose (HEC) was analysed by fluorescence spectroscopy asoutlined below.Samples from stock solutions of the substrate to be tested were added and mixedwith the polypeptide solution (initial concentration of 2.5tM in phosphate-azide buffer)in a fluorometer cell. The stock solutions used (all made up in phosphate-azide buffer)were as follows: lOOjiM, 0.5mM and 5mM G6 ; 0.1% and 1% HEC; 1% and 4% CMC.To avoid any problems with light scattering effects (this was more important forinvestigating the effect with BMCC- section 2.12.2) a shorter path length cuvette (1.Oml,turned sideways) was used. The fluorescence intensity of the solution at emissionwavelengths of 330nm and 350nm was then recorded on a Perkin Elmer LS-5Bluminescence spectrometer. The excitation wavelength was 295nm and the excitationand emission slits were both set at 5.Onm. To account for changes in intensity due todilution, a control was included where the same volumes of phosphate-azide bufferalone were added to the polypeptide solution and changes in intensity recorded. Anychanges in intensity due to the cellulose substrates themselves were also accounted forby taking into consideration intensity readings obtained from different concentrations ofthe cellulose substrate in buffer alone. Data were plotted by comparing the fluorescencedecrease observed upon addition of a given volume of substrate with that observed whenan equivalent volume of buffer was added to a solution of CBDCenA. This allowed fordilution effects on the emission intensities recorded after addition of the test substrates.To investigate changes in peak emission maxima of the CBDCenA solution uponaddition of the cellulosic substrates, the emission profile (from 3 10-450nm) of theCBDCenA solution was recorded in the presence of either a fixed concentration of thetest substrate or an equivalent volume of buffer in a 4.Oml cuvette. All the emissionprofiles of the polypeptide solutions were recorded using a model LS5O Perkin Elmerluminescence spectrometer (Perkin Elmer, Oak Brook, IL) interfaced to an IBM30personal system 2. Data were analysed with a fluorescence data manager programme(Perkin Elmer software). The excitation and emission slit widths were set at 5.Onmeach. In all cases the excitation wavelength was set at 295nm and the emission spectrarecorded from 3 l0-450nm.2.12.2 Binding to BMCCThe changes in fluorescence intensity of CBDCenA upon addition of differentconcentrations of BMCC were recorded as described in 2.12.1. Samples of BMCC wereadded from stock solutions of 20.ig/ml, 100ig/m1, 1mg/mi and 2mg/mi of BMCC inphosphate-azide buffer. As a control, the BMCC supernatant from a 2mg/mi suspensionof BMCC was also tested (as for the other substrates) for any effects on fluorescenceintensity of the CBDCenA solution. The BMCC supernatant was obtained by spinningdown the particulate material in a 2mg/mi suspension of BMCC (20 minutes, 14,800xg),removing the supernatant, and then spinning this supematant again to remove traces ofBMCC.The emission spectra of the polypeptides CBDCenA, CBDCenAW14A andCBDCenAW68A (1.25jiM in phosphate-azide buffer) in the presence of a fixed amountof BMCC suspension (final concentration of 50ig/rn1) or phosphate-azide buffer alonewere recorded as described in section Quenching by iodideThe emission spectra of 1.25.tM solutions of the polypeptides CBDCenA,CBDCenAW14A and CBDCenAW68A (in phosphate-azide buffer) were recorded in thepresence of 0.5M KC1 or 0.5M KI. The emission spectra of the polypeptide solutionswere recorded from 3lOnm to 450nm. KC1 was used to maintain the ionic strength at a31constant value. Na2S 203, at a final concentration of 10mM was included in the KIsolution to keep the iodide reduced. The excitation wavelength was 295nm and theexcitation and emission slits were both set at 5.Onm.2.13 Circular dichroism (CD)CD spectra (200-250nm) of l00.tg.ml1 solutions of the polypeptides CBDCenA,CBDCenAW14A and CBDCenAW68A (in phosphate-azide buffer) were recorded at 37°Cusing a Jasco 3720 spectropolarimeter (Japan Spectroscopic Co. Ltd., Tokyo, Japan).The spectropolarimeter was fitted with a l00il silicon quartz cell of 10mm path length.Data were collected using an IBM personal computer and analysed with Jasco J700software.2.14 Nuclear magnetic resonance (NMR)NMR spectra of the purified polypeptides CBDCenA, CBDCenAW14A andCBDCenAW68A were recorded for 1.5-2mM samples in water containing 10% D20.Spectra were acquired using a Varian Unity 500 spectrometer at 30°C (Lewis Kay,University of Toronto). Data processing was carried out on a Sun computer with Variannmrl software.2.15 Treatment of cellulose fibres with purified proteins2.15.1 Small particle release from cottonThe dewaxed cotton that was used in these experiments had been prepared asdescribed by T.M. Wood (1991) with the help of Deidre Tull in Dr. Steve Withers’s32laboratory. Various concentrations (as described in the text) of purified proteins andpolypeptides, (in 50mM potassium phosphate-azide buffer, in a total volume of 5m1)were shaken at 37°C, 300 rpm (New Brunswick Model G25 shaker incubator) with25mg of dewaxed cotton (which had been soaked in phosphate-azide buffer overnight)in siliconised glass vials to release small particles. Controls contained cotton and bufferalone or protein and buffer alone. The glass vials were removed at suitable intervals andthe contents agitated for one minute with a vortex mixer. One ml of the ‘supernatanCwas removed immediately and the absorption measured at 600nm against a phosphatebuffer blank. This sample was then returned to the vial and incubation was continueduntil the next reading. At the end of a given time (see text for details) the ‘supernatant&were removed from each of the vials, as described above and centrifuged (14,800xg, 30minutes) to pellet the particulate matter. These supematants were removed andcentrifuged again and the resulting clarified supernatants were assayed for solublecarbohydrate using orcinol-suiphuric acid (White and Kennedy, 1986). D-Glucose wasused as a standard in these assays. For carbohydrate determination, the pellets wereresuspended in 50mM phosphate-azide buffer and then used in orcinol-sulphuric acidassays. For microscopy, the pellets were washed several times with distilled water andthen treated with proteinase K (50jig/mI, 30° C, 20 minutes). They were washed againwith distilled water and finally resuspended in 0.5m1 of sterile distilled water. Samplesof these suspensions were analysed by phase contrast microscopy and scanning electronmicroscopy (SEM). For phase contrast microscopy, lOjil of the suspension were placedon a glass slide, covered with a coverslip and examined with a Zeis axiophotphotomicroscope 7082. For SEM analysis, 2Ojil of the suspension were allowed to airdry on an SEM aluminum stub. The stubs were coated with gold using a sputter coater(Nanotech Semprep II) and then examined with a Cambridge stereoscan microscope 250(Biosciences electron microscopy facility).332.15.2 Synergy between CBD.PTCenA and p30In experiments investigating the synergy between CBD.PTCenA and p30, thepolypeptide mixtures were incubated with 5mg of dewaxed cotton (which had beensoaked overnight in phosphate-azide buffer) in a total volume of imi phosphate-azidebuffer for 24 hours at 37°C, shaking at 300 rpm. Controls of cotton incubated withbuffer alone were included. At the end of the incubation period, the supernatants’ wereremoved, as described for the small particle release experiments (section 2.15.1), andwere centrifuged to separate soluble and particulate fractions. The soluble fractionswere centrifuged again to remove any traces of particulate material. The carbohydratecontents of the soluble and particulate fractions were assayed using orcinol-suiphuricacid (White and Kennedy 1986). The data obtained were all corrected for the controlscontaining cotton and buffer alone.With BMCC as the substrate the reaction mixtures contained 1mg BMC/ml andvarious concentrations of p30, CBD.PTCenA or CenA in one ml phosphate-azide buffer.BMCC incubated with buffer alone controls were included. The suspensions, in 1.7m1Eppendorf tubes, were rotated at 2 rpm at 37°C for 24 hours. The BMCC was thensedimented (14,800g, 30 minutes) and the supernatant removed and centrifuged again.The final supernatant was used for analysis of soluble sugar using orcinol-sulphuricacid. It should be noted that untreated BMCC is a suspension of fine particles and soonly the release of soluble sugar was measured because anytsmaller’ particulate materialreleased from the BMCC could not be separated from the starting material. All datawere corrected for the buffer and BMCC controls.2.15.3 Surface degradation of Ramie fibres34Ramie fibres were cut in 1.0 cm sections and soaked overnight in phosphateazide buffer. Fibre sections were then incubated with CenA, CBD.PTCenA, 1330,CBDCenA, Cex or CBDex, at lOjiM in phosphate-azide buffer, or incubated with bufferalone as controls. The reaction volume was 1 .Oml. After incubation for 72 hours at37°C without agitation, the fibres were washed with distilled water and treated withproteinase K (50.tg/ml, 37°C, 20 minutes) to remove any bound protein. The fibreswere washed again with sterile distilled water, dried, and attached to SEM aluminumstubs with AveryTM sticky tabs. They were coated with gold (section 2.15.1) andexamined by SEM.2.15.4 Disruption of Ramie fibre structureRamie cellulose fibres were pretreated with CenA, CBD.PTCenA or p30, asdescribed for SEM analysis (see section 2.15.3). Control fibres were incubated withphosphate-azide buffer alone. Following the proteinase K treatment and washing, thefibres were incubated with fluorescein isothiocyanate (FITC)-labeled CBDCenA,lOj.tg/ml in phosphate buffer, for 15 minutes at 20°C and washed with distilled water.The CBDCenA had been labeled with FITC as described by Hudson and Hay (1976).The fluorescein to protein ratio, as determined by absorbance, was 1:2. The fibres werethen examined as wet mounts with a confocal scanning microscope (Bio-Rad, MRC500) using an argon laser to excite the fluorescein. Images were reconstructed usingData View Software (Computing services, University of British Columbia).353. Results3.1 Construction of expression vectors3.1.1 Construction of the plasmid PUCl8CBDCenA, expressing the cellulose bindingdomain of CenA.In this study, CBDCenA corresponds to the amino acid and DNA sequenceindicated in Figure 3.1. The expression vector pUCl8-CBDCe,A, encoding thepolypeptide CBDp, was constructed as outlined in Figure 3.2. E. coli JM1O1transformants were screened for the expression of CBDCenA using their cell extracts,(Materials and Methods, section 2.5) as illustrated in Figure 3.3, and for the presence ofthe correct plasmid construct by restriction analysis of the DNA. The DNA from anyclones that gave positive results from both these analyses was sequenced using thereverse and universal primers (Pharmacia; recognising sequences in pUC18 which liejust upstream of the Sstl site and downstream of the Hindill site respectively, as shownin Figure 3.2-panel B) which allowed the sequencing of the entire CBDCenA insertwithin pUC18.The plasmid pUC1 8CBDCenA encodes CBDCenA (Alal-Thri 10, see Figure 3.1)plus the 31 amino-acid residue CenA leader polypeptide (Wong et al., 1986). In thisvector, expression of CBDCenA is controlled by the lac promoter.3.1.2 Construction of the plasmids expressing CBDCenAW14A and CBDCenAW68A.In order to investigate the role of the highly conserved tryptophan residues inCBDCenA (see Introduction -Figure 1.3), W14 and W68 were changed to alanine by sitedirected mutagenesis. Alanine was chosen for the substitution for the following reasons:36Figure 3.1 The DNA and amino-acid sequence of CBDCenA.The given sequence shows the DNA and amino-acid sequence of the CBD of CenA(Gilkes et al, 1988). The downward arrows indicate the leader sequence processing sitesin E. coli. The (Pro-Thr)4-Thr-(Pro-Thr)7 amino acid sequence (PIT linker) is overlinedand this links the CBD to the catalytic domain, p30 (in the direction of the largehorizontal arrow). The positions of the primers used for the construction of CBDCenA(primer #1) and for the mutant polypeptides CBDCe11AW 4A (primer #2) andCBDCepAW68A (primer #3) are shown by arrows. The sequences of the primers arecomplimentary to the underlined sequences in CBDCenA and primer extension occurs inthe direction of the arrows. Primer #4 was used in screening for CBDCenAW14Amutants (see text, section 3.1). Alal and Thri 10 indicate the first and last residues of themature CBDCenA sequence.MetSerThrArgArThrAlaAlaAlaLeoLeoAlaAlaAloAlaValAlaValGlyGlyLouThrAlaLeoThr?hrIhrAlaAleGirtATGICCACCCCCAGAACCCCCGCACCCCTGCTGCCCCCCCCCCCCGTCCCCGIGCCCCCICtGACCCCCCICACCACCACCCCCCCCCAGAlaAlProGiyCysArçjVolAspTyrAlaVa)ThrAsriGlriTrpProGlyGiyPheGiyAlaAsnVolThrliethrAsn1.00GlyAspCCCGCTCCCCCCTGCCCCGiGGAGTAGCCCGIG4CCAACGAGICCCCCCCCCCCTICCCCCCCAACGICACGATC4CCAACGIGCCCGAC2IProVolSeCSerTrpLysLeoAspTrpThrTyrThrAlaGlyGlnArgTieGinGinLeoTrpAsnGlyThrAlaSérThrAsriGlyGlyCCCGTCICCICCTGGAAGCTCGACICCACCTAC4CCGCAGGCCAGc3GAICC’AGCAGCTGTGGAACCCCACCCCCICG4CCAACCCCCCC4GirlVa)SerVa)ThrSerLeoProIrpAsnGlySerliePro11wGlyGlyThrAlaSerPheGlyPheAsnGlySerTrpAlaGiySerCAGGIGTCCGTC4CC4CCGIGCCCIGGAACGGC4CCATCCCC4CCCCCCCC4CCCCCICCTTCCCCTICAACCCCTCGTOGCCCCOGICC3I11oAsriProThrProAlaSerPheSerLeoAsnGlyThrThrCysThrGlyThrVolProThrThrSerProThrProThrProThrProThrAACCCC4CCCCGCCCICCTICTCGdCAACCCCACC4CCTGC4CCGGC4CCGIGCCC4CC4CCAGCCCC6CCCCC4CCCCCACGCCCACG—1ThrProThrProThr’ProThrProThrProThrProThrPro11w4CCCCCACGCCC4CCCCC4CCCCCACCCCC4CCCCCAGOCCC4CC38Figure 3.2 Construction of the expression vector pUC18-CBDCenAThe steps in the construction of the vector pUC18CBDe,, encoding the polypeptideCBDCepA, are shown in panel A.Fragment I was obtained from digestion of pUC18 with HindIll and SstI. The DNAsequence encoding cenA, carried on a 1.6kb Sstl fragment, had previously been clonedinto pUC18 (Guo et al., 1988). This fragment was isolated and cloned into M13mp18.This vector, M13mp18-l.6cenA, was used to make single stranded DNA for the primerextension reaction (see Materials and Methods, section 2.4). For this extension, anoligodeoxynucleotide (2lmer) complementary to the 3’ end of the DNA sequenceencoding CBDCenA (Primer #1, Figure 3.1) was annealed and extended on the templateusing the Kienow fragment of E. coli DNA polymerase I. Remaining single strandedDNA was eliminated by digestion with mung bean nuclease. The DNA was digestedwith SstI and the resulting 0.43kb fragment (fragment Ii) was recovered by agarose gelelectrophoresis. A l6mer oligodeoxynucleotide was synthesised which had a stop codonat its 5’ end. This was annealed to a complementary 20 mer to form a linker with bluntand Hindill ends (fragment III). The three fragments, I, II and III were ligated toproduce pUC18-CBDCe,, as shown in panel B. The shaded area in the CBDCenAsequence represents the leader peptide.39AMl 3mpl 8-1 .GCenApUC18‘Sstl,HindIIIHindBI______F—H____tacZ’ Ampr IacZ’BIPrimer extensionMung bean nucleaseSstlSstl L1 CBD BluntHUNKER5’mo GGC CCG AGC CGC A 3’ATC CCG CGC TCG GCG TTCGAIII5,LAC P0EoRI SstlCBD CATALYTIC DOMAINSstlPrimer (5’ GCT GGT CGT CGG CAC GGT GCC 3’)•SstI40+Figure 3.3 Screening for CBDCenA expression from E. colt JM 101 transformantsClarified cell extracts from E. colt JM1O 1 transformants were screened for theexpression of CBDCenA using the procedure described in Materials and Methods,section 2.5. Cell extracts from E. colt JM1O1:pUC18 and JM1O1:pUC18-1.6cenA wereincluded in equivalent amounts as negative and positive controls respectively. The filtershown above was used to screen for expression of CBDCenA from ten JM1O1transformants. The primary antibody used was rabbit polyclonal anti-CenAEC2 (Whittleet al., 1982; Wong et al., 1986). Colonies expressing CBDCenA were easily detectedusing this assay. The positions of the negative and the positive controls are markedas ‘-‘ and ‘+‘ respectively.I41the side chain of alanine is short and contains no poiar or functional groups, therefore itshould not introduce extra steric constraints. The ways in which the vectors expressingthe mutated polypeptides CBDCenAW14A and CBDCenAW68A were constructed aredescribed below.The 0.44kb pair (approximately) SstI/HindIlI fragment from pUC18-CBDe,(Figure 3.2) was inserted into PTZ18U (also cut with SstI/HindIII) to give the plasmidPTZ18UCBDCeM. This plasmid PTZl8U-CBDCe,A DNA was used to transform E.coil RZ1032 to allow for the production of uracil-containing single-stranded DNA to beused for site-directed mutagenesis (Zhou et aL, 1990). The oligonucleotides 5’GCCGCC GGG CGC CR3 GTT GGT Y (primer #2, Figure 3.1) and 5GCT GCC GTT CGCGGG CAG GCT 3’ (primer #3, Figure 3.1) were used to generate the mutationsCBDCenAW14A and CBDCenAW68A, respectively. The underlined residues in theoligonucleotides mark the position of the desired change in amino acid (tryptophan toalanine) in the mutagenesis procedure. The mutations were detected by DNAsequencing using primer #4 (Figure 3.1) to screen for mutations of W14 and theuniversal primer for mutations of W68, with a single track loading method, as describedby Zhou et al., 1989. Figure 3.4 illustrates this method in screening for mutations ofW14. Any clones picked up in this way were then loaded and sequenced using all fourtermination mixes to confirm the change.The plasmids carrying the mutations of the first and fourth conserved tryptophanin CBDe, changed to an alanine were designated PTZ18U-CBDCenAWJ4A andPTZ18UCBDCenAW68A, respectively. The mutated CBDCenA sequences from theseplasmids were then subcloned into pUC18 using Sstl and Hindlil restriction sites (seeFigure 3.2) to give the plasmids PUC18-C’BDCenAWJ4A and PUC18CBDCenAW68A.The CBDCenA insert from these plasmids was sequenced (using the reverse and universalprimers-see above) to ensure that no changes other than the desired mutation hadSite-directed mutation of residues W14 and W68 to alanine was performed asdescribed in Materials and Methods, section 2.4. Single stranded DNA prepared fromthe resulting E. coli JM1O1 transformants was then used as template in a dideoxysequencing reaction but only the reaction with the A termination mixture was loaded andelectrophoresed. In this way any mutant clones missing a T, as a result of the changefrom TGG, encoding tryptophan, to GCG, encoding alanine, could be distinguishedfrom the wild type sequences. The autoradiogram above shows the results from 10clones being screened for the W14 to alanine change, using primer #4 (Figure 3.1) andindicates that clone #2 is missing a T at the marked position. Single-stranded DNAsfrom any clones identified in this way were then isolated and sequenced using all 4termination mixes.1 Iio 42Figure 3.4 Sequence analysis of mutants of CBDCenA.43occurred within the CBDCeM gene. As for the CBDCenA, expression of CBDCenAWI4Aand CBDCen,4W68A in these plasmids is under the control of the lac promoter.3.2 Production and purification of CBDCe11A, CBDCenAW14A andCBDCe11AW68A.The polypeptides CBDCenA, CBDCenAW14A and CBDCenAW68A weredetected in cell extracts of E. coil JM1O1 clones by SDS-PAGE and Western blotanalysis (Figures 3.5 and 3.6) after Avicel binding assays had been carried out on thecell extracts (see Materials and Methods, section 2.11.1).N-Terminal amino-acid sequencing determined that all three polypeptidescontained approximately equal proportions of two N-Terminal variants, resulting fromthe leader peptide being processed at two sites in E. coli (Figure 3.1). The intact enzymeCenA produced in E. coli is processed at the same sites (Gilkes et al., 1988).Osmotic shock analysis of E. coil JM 101 :pUC1 8-CBDCA (Figure 3.7) showedthat the majority of CBDCenA was in the periplasmic fraction of the cells. Greater than90% of the total activity assayed for the marker enzymes glucose-6-phosphatedehydrogenase and fi-lactamase was present in the cytoplasmic and periplasmicfractions, respectively, as would be expected if the cells had been fractionatedsuccessfully.CBDCenA, CBDCenAW14A and CBDCenAW68A were purified on a large scaleby batch affinity chromatography on CF1TM using clarified cell extracts from 20 litrecultures (see Materials and Methods, section 2.6). A one-step purification on CF1TMafforded polypeptides which were contaminated with higher molecular weight proteins(Figure 3.8), and so a second step was introduced into the purification procedure.44Figure 3.5 Detection of polypeptides CBDCenA, CBDCenAW14A andCBDCe11AW68A from E. coli JM1O1 clones by SDS-PAGE analysis.Panel A- Clarified cell extract (100jil), prepared from a l.5m1 overnight culture ofJM101:pUC18-CBDce, (see Materials and Methods, section 2.6), was incubated with5mg of washed Avicel for an Avicel-binding assay (see Materials and Methods, section2.11.1). After binding, the Avicel was pelleted and the supernatant removed (‘Avicelsupernatant’). The Avicel pellet was washed and then resuspended and boiled in 30 piof SDS loading buffer. The Avicel extract and Avicel supernatant were analysed bySDS-PAGE using a 12% acrylamide gel. The gel was stained with Coomassie blue.Lane 1- JM1O1:pUC18-CBDCenA Avicel extract. Lane 2-JM1O1:pUCl8-CBDCenAAvicel supematant. Lane 3- molecular weight markers. The Avicel supernatantcorresponds to approximately 1/6 th of the Avicel extract. The large arrow head refersto the position of CBDCenA.Panel B- Clarified cell extracts were prepared from 5m1 overnight cultures ofJM1O 1 :pUC1 8-CBDe,W14A and JM 101 :pUC 1 8-CBDCenAW68A. In addition, asnegative controls, equivalent amounts of JM1O1:pUC18 and JM1O1 cell extracts wereincluded. Avicel binding assays were carried out on lOOpi of these extracts with 5mg ofwashed Avicel. After binding, the Avicel was pelleted and the supernatant removed.The washed pellets were then resuspended and boiled in 30pJ of SDS loading buffer.The Avicel extracts and Avicel supernatants were analysed by SDS-PAGE using a 12%acrylamide gel. The gel was stained with Coomassie blue. Lane 1-high molecularweight markers. Lane 2- JM1O1:pUC18-CBDCe,AW 4A Avicel extract. Lane 3-JM 101 :pUC1 8-CBDe,W14A Avicel supernatant. Lane 4- JM 101 :pUC1 8-CBDCenAW68A Avicel extract. Lane 5 -JM 101 :pUC 1 8-CBDCe,AW68A Avicelsupernatant. Lane 6- JM1O1:pUC18 Avicel extract. Lane 7- JM1O1:pUC18 Avicelsupernatant. Lane 8- JM1O1 Avicel extract. Lane 9- TM 101 Avicel supernatant. Lane10- low molecular weight markers. The Avicel supernatants correspond toapproximately 1/6 th of the Avicel extracts. The large arrow head refers to the positionsof CBDCenAW14A and CBDCenAW68A.a44444 ouI463.6 Detection of polypeptides CBDCenA, CBDCenAW14A and CBDCenAW68A fromE. coil JM1O1 clones by Western blot analysis.Panel A- Clarified cell extracts were prepared from 5ml overnight cultures ofJM1O1:pUC18-CBDCeM and JM101:pUC18. Equivalent amounts of these were usedin Avicel binding assays (Materials and Methods, section 2.11.1). The Avicel pelletswere resuspended and boiled in 30 il of SDS loading buffer. The cell extracts, Avicelextracts and Avicel supernatants were resolved by SDS-PAGE on a 14% acrylamide geland analysed by Western blot, using antiserum to CenAEC2 as the primary antibody(Materials and Methods, section 2.8). Lane 1- JM101:pUC18 cell extract. Lane 2-JM1O1:pUC18-Avicel supernatant. Lane 3- JM1O1:pUC18 Avicel extract. Lane 4-JM101:pUC18-CBDce, cell extract. Lane 5- JM101:pUC18-CBDce, Avicelsupernatant. Lane 6- JM101:pUC18-CBDc, Avicel extract. The Avicel supernatantsand cell extracts correspond to approximately 1/6 th of the Avicel extracts.The large arrow head marks the position of CBDCenA.Panel B- Clarified cell extracts were prepared from overnight cultures of JM1O1:pUC18-CBDe,i , JM101:pUC18-CBDce,W14A and JM101:pUC18-CBDc,W68A, asdescribed in legend of Figure 3.5 and l00pl used in Avicel binding assays. Equivalentvolumes of the cell extracts, Avicel extracts and Avicel supernatants were resolved bySDS-PAGE and analysed by Western blot using antiserum to CBDCenA as the primaryantibody. Lane 1- JM101:pUC18-CBDc,AW14A Avicel extract. Lane 2-3M 101 :pUC1 8-CBDe,W14A Avicel supernatant. Lane 3- JM 101 :pUC1 8-CBDCepjWl4A cell extract. Lane 4- JM1O1:pUC18-CBDCeMW68A Avicel extract.Lane 5- JMl01:pUC18-CBDcAW68A Avicel supernatant. Lane 6-JM1O1:pUC18-CBDepjW68A cell extract. Lane 7-JM101:pUC18-CBDce, Avicel extract. Lane 8-JM101:pUC18-CBDce, Avicel supernatant. Lane 9- JM1O1:pUC18-cBDCeM cellextract. The large arrow head refers to the positions of CBDCej-A, CBDCenAW14A andCBDCenAW68A.48kDaFigure 3.7 SDS-PAGE of E. coil JM1O1:pUC18-CBDCeM osmotic shock fractions.JM1O1:pUC18-CBDce, (200mls) was grown up overnight and the cells subjected toosmotic shock (as described in Materials and Methods, section 2.6) to obtain theperiplasmic and cytoplasmic fractions. Avicel binding assays were carried out usinglOOpi of these fractions with 5mg of washed Avicel (see Materials and Methods, section2.11.1). After the washing steps, the Avicel pellet was resuspended and boiled in SDSloading buffer. Avicel extracts of the periplasmic and cytoplasmic fractions were thenanalysed by SDS-PAGE on 12% acrylamide gels. The gel was developed by Coomassieblue stain. Lane 1- cytoplasmic fraction-Avicel extract. Lane 2- periplasmic fractionAvicel extract. Lane 3- molecular weight markers. Equivalent amounts of theperiplasmic and cytoplasmic fractions were loaded. The large arrow head marks theposition of CBDCenA.—2O.149Figure 3.8 Purification of CBDe Ofl CF1TMThe cell extract from a 20 litre culture of JM101:pUC18-CBDc, was treated withstreptomycin sulphate and then clarified as described in Materials and Methods, section2.6. The clarified extract was incubated with CF1TM overnight at 4°C. The cellulosewas recovered and washed with high salt buffer (1M NaC1 in phosphate-azide buffer),and low salt buffer (phosphate-azide buffer). 6M Guanidine hydrochloride was thenmixed with the cellulose to desorb bound proteins. The eluate was concentrated andthen exchanged with phosphate-azide buffer by ultrafiltration (Materials and Methods,section 2.7). The purity of the polypeptide was analysed on a 14% SDS acrylamide gel,stained with Coomassie blue. Lane 1-molecular weight markers. Lane 2- cell extract,after streptomycin sulphate treatment (corresponding to 7.5il of total cell extract). Lane3- total of 20jig of CF1TM purified CBDCenA. Lane 4- total of 5jig of CF1TM purifiedCBDCenA. The large arrow head marks the position of CBDCenA.50Figure 3.8 illustrates the purification of CBDCenA butthe results obtained for thepolypeptides CBDCenAW14A and CBDCe11AW68A were similar in that these too werecontaminated with other proteins after this step in the purification. The high molecularweight contaminants were removed by size-exclusion chromatography, as described inMaterials and Methods, section 2.7. The resulting purified polypeptides werehomogeneous, as judged by SDS-PAGE analysis (Figure 3.9). Using this protocol, thetypical yields of the polypeptides from a 20 litre culture were up to 150mg of CBDCenA,10mg of CBDCenAW14A and up to 20mg of CBDCe11AW68A.The molecular weights of CBDCenA, CBDCenAW14A and CBDCenAW68A,calculated from the DNA-derived protein sequence (Cantor and Schimmel, 1980) were11,294, 11,179 and 11,179 respectively, in agreement with the apparent Mr values ofthese polypeptides, as determined by SDS-PAGE (1 l,300Da).The extinction coefficients (280nrn; lmg.m111cm) determined experimentallyfor CBDCenA, CBDCenAW14A and CBDCenAW68A were 2.9, 2.6 and 2.6, respectively.These values were in agreement with the predicted extinction coefficients of 3.26, 2.78and 2.78 respectively.Equilibrium sedimentation centrifugation of CBDCenA, CBDCenAW1 4A andCBDCenAW68A showed that these polypeptides were monodisperse and had apparentMrs of 11,450, 11,300 and 11,400 Da respectively, in agreement with their calculatedmolecular weights.51kDa.“116.2O.1-12.4Figure 3.9 SDS-PAGE of CBDCenA, CBDCenAW14A and CBDCenAW68A purified byCF1TM and size exclusion chromatography.The polypeptides were expressed in E. coli JM1O1 and purified by cellulose affinitychromatography and size exclusion chromatography (see Materials and Methods, section2.7). The gel contained 14% acrylamide and was stained with Coomassie blue. Lanes1-3 were loaded with 15.tg of CBDCe11A, CBDCenAW14A and CBDCenAW68A,respectively. Lane 4- molecular weight markers.523.3 Binding studies3.3.1 Estimation of adsorption parameters for CBDCenA, CBDCenAW14A andCBDCenAW68A.A semi-quantitative analysis of the affinities of purified CBDCenA,CBDCenAW14A and CBDCenAW68A was carried out by examining the adsorption ofthese polypeptides to BMCC and their desorption in high salt, 50mM phosphate-azidebuffer and water (Figure 3.10). After incubation for 16 hours at 30°C, most of theCBDCenA had adsorbed to BMCC and remained bound during the various washingsteps. However, relatively smaller proportions of CBDCenAW14A and CBDeW68Abound to BMCC and the majority was desorbed in the washing steps (Figure 3.10),showing that they had lower affinity for BMCC than wild type CBDCenA.The adsorption parameters of a single ligand reacting with one lattice unit, in theabsence of positive or negative co-operative effects, are usually analysed using theLangmuir adsorption isotherm (Stuart and Ristroph, 1985, Steiner et al., 1988).However, the dimensions of the CBD of CenA (Pilz et al., 1990, Shen et aL, 1991)indicate that it must occupy more than one repeating cellobiose lattice unit on thecellulose surface (Henrissat et al., 1988). Gillces et al (1992) therefore suggested thatthe cellulose surface can be considered as an array of overlapping potential binding sites.Hence the concentration of binding sites would need to be determined by a probabilityfunction, where the probability of finding a free binding site is dependent on both theconcentration and configuration of bound ligand on the cellulose surface. However, ifadsorption at only very low concentrations of bound protein is considered, then thiscomplication is avoided. This is because at these lower concentrations there is adecreased probability of any two neighbouring bound protein molecules excluding thebinding of a third protein molecule. The modified Langmuir equation then becomes533.10 Semi-quantitative analysis of the adsorption of CBDCe11A, CBDCenAW14A andCBDCenAW68A to BMCC.Samples of CBDCenA (Panel A), CBDCenAW14A (panel B) or CBDCenAW68A (panelC) were adsorbed to BMCC as described in Materials and Methods, section 2.11.2.Unbound polypeptides were removed by centrifugation and precipitated with TCA afterfirst adding BSA as a canier. The precipitated polypeptides were boiled in SDS bufferand analysed by SDS-PAGE (lane 2). The BMCC pellets were washed twice with 1MNaC1 in phosphate-azide buffer and twice with phosphate-azide buffer alone and twicewith water. The desorbed polypeptides from these three sets of washes wereprecipitated with TCA and analysed as described above (lanes 3 and 4, lanes 5 and 6 andlanes 7 and 8 respectively). Polypeptides which remained bound to the BMCC wereremoved by resuspending and boiling the pellets in SDS loading buffer (lane 9). Equalvolumes were loaded in all lanes and correspond to one quarter of the total sample.Lane 1-molecular weight markers. The arrow heads mark the positions of thepolypeptides CBDCenA, CBDCenAW14A and CBDCenAW68A. The 66kDa band isBSA.5497.4*66*I.36.iikDa 123456789_z__ __20.1*”.12.4*i’ —K — - - -‘—Ne1-:1— —12 3 4 5 6 7 B 91’—-• ---r-•BAC55[N0] Ka [F][B]=________(1)1+aKa[F]where [B] is the concentration of bound ligand (mol.g cellulose-’), [F] is theconcentration of free ligand (molar), [N0] is the concentration of binding sites in theabsence of ligand (mol.g cellulose 1), a is the number of lattice units (i.e., cellobiosylresidues) occupied by a single ligand molecule and Ka is the equilibrium associationconstant (litres.mol1) If equation (1) is re-arranged in a double reciprocal form weobtain the following;1 a (2)[B]— Ka[No] [F] [No]This form of the equation emphasises data for the lower range of ligand concentrations.The slope of a plot of 1/[B] versus 1/[F] would give a value for lIKa[No], where [No]Kais defined as the relative equilibrium constant (Kr). This value of Kr allows us tocompare the affinities of various related ligands for a given preparation of cellulose, as[N0] will be constant. The absolute value of Ka cannot be determined unless [No] isknown. However, if a value of 101 I.tmol lattice residues.g cellulose’ is used for [N0],which is the estimated concentration of cellobiosyl residues exposed on the 110 face ofBMCC microfibril bundles (Gilkes et al, 1992), then Ka can be calculated. This analysiswas used to compare the relative affinities of CBDCenA and the mutant polypeptidesCBDCenAW14A and CBDCenAW68A for BMCC.The equilibrium adsorption isotherms for CBDCenA, CBDCenAW14A andCBDCenAW68A are shown in Figure 3.11. Saturation of BMCC by CBDCenA was5610-I II—HG) 8(I)-1II I06Ci—IL.A .AA0 I0 1 2 3[F] (pM ligand)Figure 3.11 Adsorption isotherms of CBDCenA, CBDCenAW14A and CBDCepAW68Ato BMCC.The giaph shows the equilibrium adsorption isotherms for CBDCenA (s),CBDCenAW14A () and CBDCenAW68A (A) on bacterial microcrystalline cellulose.Free and bound polypeptide concentrations [F] and [B], respectively were determinedafter equilibration at 30°C for 16 hours (see Materials and Methods, section 2.11.2). Thecellulose concentration was 1.g.1; initial CBDCenA, CBDCenAW14A andCBDCenAW68A concentrations were in the range 2.60-5.74, 1.14-2.33 and l.14-2.57j.tMrespectively. Data points for CBDCenA are means of six replicates; those forCBDCenAW14A and CBDCenAW68A are means of three replicates. The standard errorsin both dimensions are indicated by vertical and horizontal bars.57approached but not attained at the highest total protein concentration used (5.74iM).Adsorption parameters for CBDCenA, CBDCenAW14A and CBDCenAW68A wereobtained using the model described above. The relative equilibrium constants (Kr) andthe equilibrium association constants (Ka) were determined from the slope of a plot of1/[B] vs. 1/[F] for the three polypeptides (Figure 3.12). Table 3.1 shows that the Kr andKa values of CBDpjW14A and CBDCenAW68A for BMCC were reduced 50- and 30-fold respectively, relative to CBDCenA.Polypeptide Kr Kal.g1 14L,noMCBDCe 30.3 ± 6.7 0.30 ± 0.07CBDCenAW14A 0.60 ± 0.11. 0.006 ± 0.001CBDCenAW68A 1.08 ± 0.08 0.011 ± 0.001Table 3.1 Equilibrium association constants for the binding of CBDCe11A,CBDCenAW14A and CBDCenAW68A to BMCC.The relative equilibrium association constant (Kr) and equilibrium association constant(Ka) for each polypeptide was calculated from the adsorption data shown in Figure 3.12.Values are shown ± the accumulated standard error. The values for Ka were calculatedusing [No] = 101 .tmol lattice residues.g cellulose1(see text).583C(t5C)20E0)U)0D0)C) 10)cci00 101/[F] (pM ligand 1)Figure 3.12 Double-reciprocal plots of adsorption data for CBDCenA, CBDCepAW14Aand CBDCeW68A.The adsorption data for CBDCenA ( )‘ CBDCenAW14A () and CBDCenAW68A(A) are plotted in double-reciprocal form (1/[B] versus lI[F]), with standard errors inboth dimensions indicated by vertical and horizontal bars. The Kr values listed in Table3.1 were estimated from the slopes of these plots.2 4 6 8593.3.2 Fluorescence quenchingThe total fluorescence of a protein, in the absence of energy transfer betweenaromatic residues, should be the sum of the contributions of the tyrosine andthetryptophan residues. This means that each tyrosine and tryptophan residue contributesto the overall protein spectrum and emission yield. The individual contribution by eachresidue is dependent on its local environment (Lehrer and Fasman, 1967). The emissionof tyrosine is in most cases negligible in comparison to the tryptophan emission (Konev,1967) as the absorbance at the wavelength of excitation and the quantum yield ofemission of tryptophan residues are considerably larger than the respective values fortyrosine and phenylalanine. Tryptophan fluorescence can be investigated selectively byexcitation at wavelengths of 295nm (Szabo, 1989). According to a model proposed byBernstein et al (1973), the emission spectrum of each tryptophan depends on itsenvironment, according to whether the residue is buried in a non-polar region (from 330-332nm), partially exposed to solvent (up to 342nm) or completely exposed to solvent(up to 352nm).Quenching of the inherent fluorescence of a protein upon binding to substratecan be attributed to modifications of the microenvironment of one or several of thetryptophan residues upon conformational change of the protein. This spectroscopicmethod was used to investigate the interaction of the CBDCenA with both soluble andinsoluble substrates. At present it is not known how many sugar units occupy thebinding site of CBDCenA during adsorption to cellulose, hence it was of interest toinvestigate the interaction of the CBDCenA with oligosaccharides and soluble substrates.The emission maximum of CBDCenA was around 330nm (see Figure 3.19,section 3.4.3); however, it was observed that the cellulosic substrates themselves emitsome fluorescence at this wavelength, whereas they emit very little fluorescence at350nm. Therefore, changes in fluorescence at both emission wavelengths ëre60monitored upon addition of the substrates to the polypeptide solution. There were nosignificant differences between the data obtained at the two wavelengths. All figuresshow the data obtained using the fluorescence changes observed at the 350nm emissionwavelength, and take into account dilution effects and fluorescence from the substratesthemselves (see Materials and Methods, section 2.12.1).The first soluble substrate investigated was the oligosaccharide cellohexaose(G6). There seemed to be little change in the fluorescence of CBDCenA with theaddition of the G6 over a final concentration range of 0.5iM to 45OjiM (Figure 3.13).Other soluble substrates investigated were carboxymethylcellulose (CMC) andhydroxyethylcellulose (HEC) (Figures 3.14 and 3.15). Again, there were no significantchanges in the fluorescence emission of CBDCenA with the addition of these substrates.In addition, there were no shifts of the emission peak maximum of the CBDCe11Asolution (335nm, Figure 3.19) upon mixing of a fixed high concentration (1mM G6,0.5% CMC and 0.5% HEC) of these substrates with the polypeptide solution.The fluorescence changes upon addition of BMCC, a substrate with whichCBDCenA is known to interact, were then investigated. In this case, the fluorescenceemission of CBDCA was shown to decrease in the presence of BMCC (Figure 3.16,and see section 3.4.3), indicating that the polypeptide was interacting with this substrate.A maximum fluorescence decrease was observed upon addition of approximately3igBMCCIm1. As a control the supernatant alone from the BMCC suspension wastested, to see if this was causing the decrease in the fluorescence. There was no decreasein fluorescence or change in the maximum emission wavelength of the CBDCenAsolution upon addition of this supernatant61© 120©c- ‘‘u• 110100 100 -90 9080 80. 70 70-.‘60— 60•0 100 200 300 400 500 0 20 40 60 80 100 120Concentration of G6 (tiM)Figure 3.13 Change in relative fluorescence of CBDCenA on addition of cellohexaose(G6).The relative fluorescence of CBDCenA at 350nm in the presence of differentconcentrations of G6 was recorded as described in Materials and Methods, section2.12.1. The fluorescence quenching is expressed as the % decrease in fluorescencerelative to a control where buffer was added to the polypeptide solution instead of thesugar. The excitation wavelength was 295nm. The excitation and emission slits were5.Onm. Panel A shows the changes in the fluorescence of CBDCenA in the presence ofup to 450p.M G6. Panel B shows the changes in fluorescence of CBDCenA in thepresence of 0- 120i.LM G6. Data points represent the mean of two replicates.62120-‘S0.3Concentration of CMC (%)Figure 3.14 Change in relative fluorescence of CBDCenA on addition ofcarboxymethylcellulose (CMC).The relative fluorescence of CBDCenA at 350nm in the presence of differentconcentrations of CMC was recorded as described in Materials and Methods, section2.12.1. The fluorescence quenching is expressed as the % decrease in fluorescencerelative to a control where buffer was added to the polypeptide solution instead of thesugar. The excitation wavelength was 295nm. The excitation and emission slits were5.Onm. Data points represent the mean of two replicates.631200110100900.00Concentration of hyclroxyethylcellulose (%)Figure 3.15 Change in relative fluorescence of CBDCenA on addition ofhydroxyethylcellulose (HEC)The relative fluorescence of CBDCenA at 350nm in the presence of differentconcentrations of HEC was recorded as described in Materials and Methods, section2.12.1. The fluorescence quenching is expressed as the % decrease in fluorescencerelative to a control where buffer was added to the polypeptide solution instead of thesugar. The excitation wavelength was 295nm. The excitation and emission slits were5.Onm. Data points represent the mean of two replicates.0.02 0.04 0.06 0.08 0.1064—©010090 9080 80C) 7060 600 10 20 30 40 50 60 0 2 4 6 8 10 12Concentration of BMCC (jig/mi)Figure 3.16 Change in relative fluorescence of CBDCenA on addition ofBMCC.The relative fluorescence of CBDCenA at 350nm in the presence of differentconcentrations of BMCC suspension was recorded as described in Materials andMethods, section 2.12.1. The fluorescence quenching is expressed as the % decrease influorescence relative to a control where buffer was added to the polypeptide solutioninstead of the sugar. The excitation wavelength was 295nm. The excitation andemission slits were 5.Onm. Panel A shows the changes in the fluorescence in thepresence of up to 50p.g/ml of BMCC suspension. Panel B shows the changes influorescence of CBDCA in the presence of 0-lOig/ml of BMCC. Data pointsrepresent the mean of two replicates.A 120110100B70653.4 Structural studies on CBDCe11A,CBDCe11AW 4A andCBDCenAW6SA.A number of analyses were carried out on purified CBDCenA, CBDCe11AW 4Aand CBDCenAW68A in order to compare the structural integrity of the mutantpolypeptides to the wild type. These are discussed below. Fluorescence spectroscopywas also used to obtain information on the possible location of the residues W14 andW68 m the polypeptide CBDCenA.3.4.1. Circular dichroism (CD)Circular dichroism (CD) spectroscopy is a technique which is sensitive to theoverall conformation and folding of polypeptides chains and so is a valuable tool forevaluating protein structure in solution. CD gives information about the unequalabsorption of right- and left-handed circularly polarised light by optically activemolecules. CD bands of proteins occur in two spectral regions; the far U.V. or amideregion (170-250nm) contains information about the peptide bonds and secondarystructure of a protein and the CD bands in the near-U. V. region (250-300nm) arecontributed by the aromatic amino-acids. Figure 3.17 shows that the spectra ofCBDCAW14A and CBDCenAW68A were similar to the spectrum for CBDCenA. Allthree spectra show an absence of alpha-helical structure (Yang et al., 1986).3.4.2. 1H NMRNuclear magnetic resonance (NMR) is a powerful tool for macromolecularstructural determination at atomic resolution. This technique, in addition to X-ray660EFigure 3.17 Circular dichroism spectra of CBDCepA, CBDCenAW14A andCBDCepAW68A.Spectra were obtained from CBDCenA ( ), CBDCenAW14A () andCBDCAW68A (.—_... ) dissolved in 50mM potassium phosphate buffer, pH 7.0, at afinal concentration of 1O0Ig.m14(Materials and Methods, section 2.13). [o] representsthe mean residue effipticity.Wavelength (nm)67crystallography, has allowed the determination of an increasing number of proteinstructures (Wüthrich, 1976, 1986). NMR can also be used in studies of conformationalchanges in proteins and was used to determine if the conformations of the mutantpolypeptides CBDCenAW14A and CBDCenAW68A were greatly different from that ofCBDCenA. Figure 3.18 shows the methyl region of the one dimensional 1H NMRspectra of the three polypeptides. All three polypeptides produced similar dispersespectra with upfield shifted methyl peaks (indicative of folded polypeptides), suggestingthat they have very similar conformations in solution. The lines of the spectrum forCBDCenA are significantly broader than those of the spectra of the mutant polypeptides,which may be a consequence of the wild-type polypeptide occurring as a mixture ofmonomers and higher molecular weight aggregates (Dr. Lawrence McIntosh, personalcommunication).3.4.3. Fluorescence dataTryptophan fluorescence data showed that the emission spectra of CBDCe11A,CBDCenAW14A and CBDCenAW68A all had maxima near 330nm (Figure 3.19),indicative of tryptophan residues in stably folded cores (Lehrer and Fasman, 1967).Each spectrum also showed definite contributions from solvent-exposed residues in the350nm region. In CBDCenAW14A and CBDCenAW68A there was a decrease in theemission intensity at 350nm, relative to that at 330nm, consistent with the loss of anexposed tryptophan residue from the surface of each mutant polypeptide.Quantitatively, the loss of one of the six tryptophan residues within CBDCenAdid not decrease the observed total emission proportionally. This is expected becauseaccessibility to quenching by solvent molecules, proximity of amino-acid side chainsthat can act as quenchers such as cysteine, asparagine or glutamine residues and possible68ppmFigure 3.18 One dimensional 111 NMR spectra of CBDCenA, CBDCenAW14A andCBDCenAW68AOne dimensional 1H NMR spectra of the polypeptides CBDCenAW68A (Panel A),CBDCepAW14A (panel B) and CBDCepA (panel C) were obtained as outlined inMaterials and Methods, section 2.14. Panels A, B and C show the methyl regions of thespectra.7.5 7.0 0.5 0.0 -0.569Figure 3.19 Tryptophan fluorescence spectroscopy of CBDCenA, CBDCenAW14A andCBDCenAW68A.Tryptophan fluorescence spectra of CBDCenA (panel A), CBDCenAW68A (panel B) andCBDCenAW14A (panel C) with O.5M KC1 (upper curves) or KI (middle curves); thelower curve in each panel shows the difference spectrum (see Materials and Methods,section 2.12.3). The excitation wavelength was 295nm. The excitation and emissionslits were set at 5.Onm. The spectra shown above are representative of the spectraobtained from repeated experiments.0r)ECCVV>Ctf-)CY)(I)ELu00ry-)ODUODSOJOflI471inter-tryptophan energy transfer would affect the contribution of each tryptophan residueto the total emission from the polypeptide (Harris and Hudson, 1990).The interiors of proteins are generally considered to be non-polar and becausetryptophan residues are also non-polar, we would expect some of these to be in theinterior of proteins, and hence inaccessible to quenchers in the external aqueous phase.In consequence, quenching by any substance which cannot penetrate into the interior ofproteins can be used to determine fluorescence due to any tryptophan residues located atthe surface of proteins (Lakowicz, 1983). One such substance is iodide, which ischarged and hydrated and hence is unable to penetrate proteins. The fluorescence oftryptophan residues exposed to the solvent at the surface of CBDCenA and its twomutants was susceptible to quenching by iodide. Figure 3.19 shows emission spectra ofthe polypeptides in the presence of KI or KCI. The difference between the emissionspectrum of the polypeptides in KI and KCI is also shown (Figure 3.19-lower curves).The difference spectra for all three polypeptides was characteristic of the spectrum of anexposed tryptophan residue, having maximal emission near 350nm. The magnitudes ofthe difference spectra of CBDCenAW14A and CBDCenAW68A were smaller than that ofCBDCenA, suggesting that the mutant polypeptides had relatively fewer exposedtryptophan residues.The changes in fluorescence characteristics of the mutant and wild typeCBDCenA upon interaction with BMCC were also investigated. The mutantpolypeptides had emission maxima around 330nm, like the wild type polypeptide, with adistinct shoulder at 350nm (Figure 3.19). The decrease in fluorescence of CBDCenAupon addition of BMCC had little effect on the maximum wavelength of emission(Figure 3.20, 335nm in buffer and 336.5nm in BMCC), whereas distinct changes in thespectrum envelope were observed with the two mutant polypeptides upon addition ofBMCC (Figures 3.21 and 3.22). The emission peak maxima for CBDCenAW14A were33 mm and 337nrn in buffer and in the presence of BMCC suspension respectively. Thea)C.)G)0Cl)0Da)>CuEmission wavelength (nm)410Figure 3.20 Tryptophan fluorescence emission spectra of CBDCenA in the absence andpresence of BMCC.The emission spectra of 1.25p.M CBDCenA solution in the absence ( ) or presence( ) of 50.tg/m1 BMCC suspension (see Materials and Methods, section 2.12.2).The excitation wavelength was 295nm. The excitation and emission slits were set at5.Onm. The spectra shown above are representative of the spectra obtained fromrepeated experiments.7273C)1)C)C.)U)C)L..0C)‘ICua)410Emission wavelength (nm)Figure 3.21 Tryptophan fluorescence emission spectra of CBDCe11AW 4A in theabsence and presence of BMCC.The emission spectra of 1.25$tM CBDCenAW14A in the absence ( ), or presence( ) of 5Oig/ml BMCC suspension (see Materials and Methods, section 2.12.2).The excitation wavelength was 295nm. The excitation and emission slits were set at5.Onm. The spectra shown above are representative of the spectra obtained fromrepeated experiments.310 330 350 370 390C.)C.)C’)0I0‘4-G)>4-.Cud)—1Emission wavelength (nm)410Figure 3.22 Tryptophan fluorescence emission spectra of CBDCenAW68A in theabsence and presence of BMCC.The emission spectra of 1.25jiM CBDCenAW68A solution in the absence ( ), orpresence ( ) of 5Otg/m1 of BMCC suspension (see Materials and Methods, section2.12.2). The excitation wavelength was 295nm. The excitation and emission slits wereset at 5.Onm. The spectra shown above are representative of the spectra obtained fromrepeated experiments.74Th75emission peak maxima for CBDCenAW68A were 332nm and 336nm in buffer and in thepresence of BMCC suspension respectively. Martineau et al (1990) proposed a model inwhich, if it is assumed that the fluorescence of a protein is due to emission fromtryptophan residues only and there is no energy transfer between these residues, thedifference spectrum between the wild-type protein and the mutant lacking a tryptophanresidue would represent the spectrum of that missing tryptophan. This model was usedwith the mutant polypeptides CBDCenAW14A and CBDCe11AW68A to obtain emissionspectra of tryptophan residues 14 and 68.The difference spectra of CBDCenAW14A and CBDCenAW68A in the presenceand absence of BMCC are shown in Figures 3.23 and 324. In the absence of BMCC,the residues W14 and W68 had emission maxima around 340-345nm, indicative ofexposed residues and in the presence of BMCC the residues have blue-shifted emissionmaxima of 338-340nm. This suggested that the tryptophan residues were in a morehydrophobic environment in the presence of BMCC, probably as a result of theirinteraction with the substrate. These results correlate directly with those obtained fromthe KI quenching experiments, in that W14 and W68 were inferred to be on the surfaceof CBDCenA.a)0CG)0C,)a)10II>Cua)Figure 3.23 Spectra of W14 in the absence and presence of BMCC.410The emission spectra of residue W14 in the absence (------) or presence ( ) ofBMCC were obtained by plotting the difference of the spectra obtained for CBDCepA(Figure 3.20) with that of the spectra obtained for CBDeWl4A (Figure 3.21).76Emission wavelength (nm)770CG)C)U)0‘4-G)>•1-’0.410The emission spectra of residue W68 in the absence ( ) or presence ( ) ofBMCC were obtained by plotting the difference of the spectra obtained for CBDe&(Figure 3.20) with that of the spectra obtained for CBDp.pW68A (Figure 3.22).Emission wavelength (nm)Figure 3.24 Spectra of W68 in the absence and presence of BMCC.310 330 350 370 390783.5 Role of CBDCe11Ain the breakdown of cellulose.Cellulose binding domains in cellulases are proposed to play an important role inthe degradation of cellulose (see Introduction). The production of the independentcellulose binding domain with and without the Pro/Thr linker (CBDCenA, this study andCBD.PTCenA, Gilices et a!., 1992) and the catalytic domain, p30, (prepared from CenAby proteolysis with C.fimi protease; Gilkes et al, 1988), made it possible to compare theeffects of the individual components of CenA on cellulose with those of the intactenzyme, and hence to study the effect of the cellulose binding domain itself on cellulose.3.5.1 Effects of CBD.PTCe11A,p30 and CenA on cotton fibres.When dewaxed cotton fibres were incubated with l0iM CenA, p30 andCBD.PTCenA, small bundles of microfibrils, which will be referred to as small particles,were released from the cotton. These small particles were visible to the naked eye as ahaziness in the supematant, after the bulk of the cotton had settled. They werequantified by measuring the turbidity of the supernatants at 600nm (Figure 3.25). CenAand p30 released small particles from cotton throughout the period of incubation.BD.PTCenA also released small particles from cotton, but the release wasdiscontinuous and fewer particles were released. It was seen that most of the particlerelease by CBD.PTe occurred within two minutes of incubation (Figure 3.26) andthat the release was concentration dependent (Figure 3.27).Determination of the soluble carbohydrate in the clarified supernatants showedthat CenA and p30 both released equivalent amounts (0.11 mg/nil and 0.105mg/mirespectively) after 72 hours of incubation. No soluble carbohydrate could be detected in793.02.0A6001.00.030Time (Hours)Figure 3.25 Small particle release from dewaxed cotton.A600 readings were taken of the supernatants from 25mg of dewaxed cotton incubatedwith 1OpM CenA (7), CBD.PTCenA () or p30 (<‘)),in phosphate-azide buffer(Materials and Methods, section 2.15.1). Controls included cotton only (i).CBD.PTCe11Aonly (A), p30 only (fl ) and CenA only (0) . The data points representthe average value obtained from repeated experiments. It should be noted that the datapoints for p30 alone are co-linear with the horizontal axis.0 10 2080Cotton + CBD.PTCotton + p30Cotton + CenACottonFigure 3.26 Small particle production from dewaxed cotton by CBD.PTCenA,CenA or p30 after 2 minutes of incubation.A600 readings were taken of the supematants from 25mg of dewaxed cotton incubatedwith lOj.tM CenA, CBD.PTCenA, or p30, in phosphate-azide buffer (see Materials andMethods, section 2.15.1). CenA, p30 and CBD.PTCenA controls without cotton gavenegligible A63 readings. The data shown represent the average values obtained from0.00 0.10 0.20 0.30A600repeated experiments.810.800.700.600.50A600 0.400.300.200.10 I0.00 0.10 0.20 0.30CBD.PTCenA concentration (mg/mi)Figure 3.27 Small particle production from dewaxed cotton by CBD.PTCenA.Various concentrations of CBD.PTep, were incubated with (El) or without (0)cotton for 72 hours and A600 readings taken of the supernatants at the end of this period(see Materials and Methods, section 2.15.1). The data points represent the average valueobtained from repeated experiments.03082the clarified supernatants from the CBD.PTCenA and the buffer control incubation(<0.01 mg/mi).Phase contrast microscopy showed that the particles released by CenA weremostly small and irregular (Figure 3.28). There were some larger particles present (upto 0.5mm in length) and scanning electron microscopy (SEM) of these (Figure 3.29)showed that they had very rough surfaces. Particles released by p30 were fibres up to2n-im in length and SEM analysis showed that the fibres, in contrast to those released byCenA, had very smooth surfaces (Figure 3.28 and 3.29). The particles released byCBD.PTCenA were very similar to those released by CenA (Figure 3.28), but there werefar fewer larger fibres. Again, these larger fibres were seen to have very disruptedsurfaces by SEM analysis (Figure 3.29).In order to ensure that the smaller, irregularly-shaped particles released by CenAand CBD.PTCe11Awere in fact cotton and not protein aggregates, some of the materialwas boiled for 5 minutes with SDS-urea buffer (0.0625M Tris, pH 6.8. 10% glycerol,2% SDS, 0.05% B-mercaptoethanol and 1M urea). Phase contrast microscopy showedthat there were no changes in appearance of the small particles after the boiling step.3.5.2 Effects of CBD.PTCenA, CBDCe11A,p30 and CenA on Ramie fibres.Ramie fibres are highly crystalline (70-74%) fibres (Cheek and Roussel, 1989)which can be up to 20cm in length, and hence are easily examined microscopically.These fibres therefore provided a means of investigating the effect of CBD.PTCenA,CBDCenA, p30 and CenA on individual cellulose fibres.Ramie fibres were incubated with buM of CBD.PTCenA, CBDCe11A,p30, CenAor buffer for 72 hours at 37°C (Materials and Methods, section 2.15.3) and the treatedfibres were then examined by SEM. Distinct changes in the surfaces of the fibres were83Figure 3.28 Phase contrast microscopy of small particles released from cotton by CenACBD.PTCenA and p30.Phase contrast microscopy of cotton fibres incubated with phosphate-azide buffer alone(A), and small particles released from cotton by p30 (B), CenA (C) or CBD.PTp(I)) after 72 hours in a shaking water bath at 37 °C, 300 rpm (Materials and Methods,section 2.15.1). The concentrations of the polypeptides were lOj.iM in phosphate-azidebuffer. The bar = 20.tm. Panels A, B and C courtesy of Dr. Neil Gilkes.U•-85Figure 3.29 Scanning electron microscopy of particulate material released from cottonby CenA,p30 and CBD.PTCenA.Electron micrographs of small particles released from cotton by CenA (B), p30 (C) andCBD.PTCe11A(D). Panel A shows the appearance of the starting material after beingincubated in buffer alone. The reaction conditions are as outlined in the legend to Figure3.28.co0V87observed (Figure 3.30). CenA caused extensive degradation along the length of the fibre(Figure 3.30- panel A). The microfibril arrangement of the fibre was exposed by theaction of the protein. CBD.PTcenAafld CBDCenA-treated fibres showed similar areas ofdisruption, but these were confined to ‘hot spots’ along the length of the fibres (Figure3.30- panel B). The surfaces of the untreated fibres were relatively smooth with a fewnatural irregularities (D). Fibres treated with p30, however, were devoid ofirregularities, and had highly polished surfaces (Figure 3.30- panel C). This smoothingeffect by the catalytic domain alone was in sharp contrast to the disruptive effects of theintact enzyme and its binding domain.The penetration of a fluorescein labelled probe, FITC-labelled CBDCenA, into afibre was used to determine the extent to which the internal structures of the fibres weredisrupted after treatment with CBD.PTCenA, p30 and CenA. Fibres were treated withthe polypeptides as for the SEM analysis, but after the proteinase K treatment wereincubated with fluorescein-labelled CBDCenA. The concentration of fluoresceinylatedCBD used was less than one-tenth the concentration of polypeptide used for pretreatment of the fibres (10ig/ml). Transverse optical sections of Ramie fibres wereexamined for the presence of the labelled CBD by confocal microscopy (Figure 3.31).The FITC-CBDCenA is yellow, but the intensity of the colour varies depending on theamount of the FITCCBDCenA present (as shown on the panel on the left of the figure)The control fibre, which had been incubated in phosphate-azide buffer alone prior to theF1TCCBDCenA incubation, showed little binding of the label and this was confined topatches on its surface (Figure 3.31-panels A and B). Treatment of a fibre with CenAincreased the access of the labelled probe to the interior of the fibre, indicating thatCenA had disrupted the structural integrity of the fibre (Figure 3.31-panels E and F).CBD.PTCenA (Figure 3.31-panels C and D) also disrupted the fibre structure but to alesser extent than CenA. The label had penetrated at areas where the fibre was88Figure 3.30 Scanning electron micrographs of Ramie cotton fibres treated with CenA,CBD.PTen and P3°Electron micrographs of Ramie cotton fibres treated with CenA (A) CBD.PTCe11(B)and p30 (C). A control fibre which had been incubated with the phosphate-azide bufferalone is shown in (D). The fibre treated with CBDCenA is not shown but wasindistinguishable from the CBD.PTCenA-treated fibre. Polypeptide concentrations wereall l0.tM in phosphate-azide buffer. Incubations were for 72 hours at 37°C (Materialsand Methods, section 2.15.3).90Figure 3.31 Disruption of Ramie fibre structureRamie fibres were pretreated with 10M CenA, CBD.PTCenA, or p30, then labelledwith FITC-CBDCenA and examined by confocal microscopy (see Materials andMethods, section 2.15.4). (A) and (B) show the control fibre, which had not beenpretreated, (incubated with phosphate-azide buffer alone) after incubation with thelabelled CBDCenA. (A) shows a composite of 9 slices through the fibre and (B) shows amid-section slice. (C) and (D) show, respectively, the composite image and mid-sectionslice from the fibre which had been pretreated with CBD.PTCenA. (E) and (F) show thecomposite and mid-section slice from the CenA pretreated fibre and (G) and (H) showthe composite and mid-section slice from the p30 pretreated fibre. Note that the FITCCBDCenA is yellow, but the intensity of the colour varies depending on the amount ofF1TC-CBDCenA present (panel on the left).I:lu1.A IIIII—=cIQCC-‘N3=7_fl-‘ 0==0=0==C,===000=0000=000000CDCDCCCDCCCDCD•++++++++00000000-‘-‘N3NN3N1692exfoliated. Fibres which had been pretreated with p30 showed little binding of thelabelled probe and only on the surface of the smooth fibres (Figure 3.31-panels G andH).3.5.3 Small particle release by CBDCenAW14A and CBDCenAW68A.In light of the reduced affinities of the mutant polypeptides, CBDCenAW14A andCBDCenAW68A, for cellulose, it was of interest to investigate whether thesepolypeptides facilitated the release of small particles from cotton, as was observed withthe wild-type polypeptide (Section 3.5.1). Figure 3.32 shows that the mutantpolypeptides did not release any particulate material from cotton for the duration of theincubation period. Any increase in the A600 readings could be accounted for by theprotein alone controls.3.5.4 Small particle release by Cex and CBDex and their effects on Ramie fibresCex is an enzyme from C.Jinii which hydrolyses both 13-1,4-xylosidic and 13-1,4-glucosidic bonds and contains a CBD related to that of CenA (see Introduction), hence itwas of interest to compare the effect of Cex and its CBD (CBDCex) on cotton and Ramiefibres with the effects seen with CenA and the CBD of CenA.When dewaxed cotton was incubated with 1OtM Cex and CBDex (Materialsand Methods, section 2.15.1) it was apparent that these polypeptides also released smallparticles from the cotton fibres (Figure 3.33), with CBDex releasing fewer particlesover the incubation period than the intact enzyme, as had been observed for the CBD ofCenA. Scanning electron microscopy of the small particles released by Cex andCBDex (Figure 3.34, panels C and A respectively) showed that these small particlesalso had very disrupted surfaces.930.20C(0< 0.100.0030Time (Hours)Figure 3.32 Small particle release from cotton by CBDCenA CBDCe11AW 4A andCBDCenAW68A.25mg of dewaxed cotton were incubated with 10iM of CBDCenA (D),CBDCe11AW 4A (7) or CBDCenAW68A (A) in a shaking water bath (37 °C, 300 rpm)for 24 hours, as described in Materials and Methods, section 2.15.1. A600 readings weretaken of the supernatants at the times indicated. Controls were included where thereaction vials contained cotton with buffer alone (0) and protein alone without thecotton; CBDCenA without cotton (c), CBDCe11AW 4A without cotton () andCBDCenAW68A without cotton (s). The data points represent the average valueobtained from repeated experiments.0 10 20942.0A6001.00.030Time (Hours)Figure 3.33 Small particle production from dewaxed cotton by CBDex and CexA600 readings were taken of the supernatants from 25mg of dewaxed cotton incubatedwith lOi.LM CBDCex, (7) or Cex (<)),in a shaking water bath (37°C, 300rpm) for 24hours, as described in Materials and Methods, section 2.15.1. The final reaction volumewas 5m1 in phosphate-azide buffer. Controls included cotton only (0)’ CBDCex only(LI) and Cex only (A) . The data points represent the average value obtained fromrepeated experiments.0 10 2095Figure 3.34 Scanning electron microscopy of cellulose fibres treated with CBDCex andCex.Electron micrographs of small particles released from cotton by CBDc (A) or Cex(C). Incubations were for 72 hours in a shaking water bath at 37°C, 300rpm (Materialsand Methods, section 2.15.1), using 1O.tM of CBDCex and Cex in phosphate-azidebuffer. Panels B and D show electron micrographs of Ramie cotton fibre treated with10.tM CBDex and Cex respectively. Incubations were at 37°C for 72 hours (Materialsand Methods, section 2.15.3)96C)97Ramie fibres treated with Cex and CBDCex also showed similar results to those that hadbeen observed for CenA and its CBD in that they caused disruption of the Ramie fibre.Cex had caused less marked degradation of the entire fibre than had been observed withCenA (Figure 3.34- Panel D) but the CBDcex-treated fibre showed similar areas ofdisruption to that of CBDCenA (Figure 3.34- Panel B) at hot spots along the length ofthe fibre.3.5.5 Synergy between CBD and p30The discoveries that the CBD of CenA disrupted the surface of cotton and Ramiecellulose fibres and released small particles from these substrates and that the CBDs ofCenA and Cex prevented the flocculation of microcrystalline bacterial cellulose (Gilkeset al., 1993) led to the investigation of whether the CBD acts co-operatively with thecatalytic domain of CenA in attacking cellulose.Increasing concentrations of CBD.PTCenA in the presence of a fixedconcentration of p30 released increasing amounts of soluble sugar from cotton (Figure3.35, Panel A); CBD.PTCenA itself did not produce soluble sugar from the substrate. Infact, an equimolar mixture of CBD.PTCe11Aand p30 (i.e. l0iM p30: l0!IM CBD)released almost twice as much soluble sugar as the same amount of p30 alone. Both p30and CBD.PTCenA alone liberated particulate material from cotton (Figure 3.35, PanelB). Mixtures of p30 and CBD.PTCenA in the various molar ratios (1:1, 1:2 and 1:4) allreleased more insoluble carbohydrate than the combined amounts released by theindividual components. This indicated that the two polypeptides were actingsynergistically to release more particles from the cotton. The total amount ofcarbohydrate, both soluble and particulate, released by the mixtures was also greaterthan the combined amount released by p30 and CBD.PTCenA alone (Figure 3.35 , PanelC). There was a steady increase in the total sugar released (both soluble and particulate)98Figure 3.35 Release of sugars from cotton by CBD.PTCejA and p30.Reaction mixtures contained various concentrations of p30 and CBD.PTCenA and 5mgof dewaxed cotton, in phosphate-azide buffer. All reaction mixtures were set up induplicate. The suspensions were incubated at 37°C, 300rpm for 24 hours and then theclarified supernatants (A) and the particulate material (B), were assayed forcarbohydrate using orcinol-suiphuric acid (see Materials and Methods, sections 2.15.1and 2.15.2). Panel C shows the total soluble and particulate sugars released. The sugarrelease expected, if the effects of the two polypeptides was additive, is given (‘Expectedeffect’) to compare the sugar release obtained experimentally (‘Observed’). Error barsrepresent standard deviations of the measurements.1OiMp30:80)IMCBD1OjiMp3O:40MCBO1OiMp3O:2O.tMCBD1OiMp3O:1OjMCBD80iMCBD401.tMCBD201iMCBD1OtMCBD10iMp300100200SoIubecarbohydrate(jig/mi)A[IBJ El 1CHSExpectedDObserved01002000‘100200Particulatecarbohydrate(jig/mi)Total carbohydrate(jig/mi)300100from cotton with the addition of increasing amounts of CBD to a fixed amount of p30 upto a molar ratio of CBD to p30 of 4:1. Above this ratio, the synergistic effect decreased(Figure 3.35, Panels A, B and C).In order to keep the final total molar concentration of protein in each reactionmixture the same, p30 and CBD.PTCenA were mixed with cotton in the proportions of25E.tM:25pM, 10IM:4O!1M, and 40jiM:l0iM. These mixtures were also compared with25iM of the intact enzyme, CenA. All three mixtures released more soluble sugar fromcotton than either polypeptide alone at the same final molar concentration (Figure 3.36,Panel A). Similarly, the 25j.iM p30 plus 25pM CBD.PTCenA and lOj.iM p30 plus 40j.tMCBD.PTCenA released more particulate matter from cotton than the individualpolypeptides alone. Although all three combinations of p30 and CBD.PTCenA were notas effective as the intact enzyme, CenA, in releasing sugars from cotton, the twoindividual domains clearly acted synergistically on this substrate (Figure 3.36, Panel C).In experiments analogous to those with cotton, p30 and CBD.PTCenA did not actsynergistically on bacterial microcrystalline cellulose (BMCC); CBD.PTCenA did notenhance the release of sugars or particulate matter from BMCC by p30 (Figure 3.37).101Figure 3.36 Release of sugars from cotton by p30, CBD.PTCe11Aand CenA.The hydrolysis reactions were set up as described in the legend of Figure 3.35. Allreactions were set up in duplicate. Error bars represent standard deviations of themeasurements.5OtMCBD1 O1iMp3O:4OiMCBD25KMp3O:25iMCBD4OiMp30:lOjiMCBD5OjtMp3O25pMCenAABCII0-10020001002000100200300Soub!ecarbohydrate(tg/mI)Particulatecarbohydrate(jig/mi)Total carbohydrate(jig/mi)F’J103Figure 3.37 Release of sugars from BMCC by p30, CBD.PTCenA and CenA.The reaction mixtures contained 1mg/mi BMCC and various concentrations of p30,CBD.PTCenA or CenA in phosphate-azide buffer. All reactions were set up in duplicate.The suspensions were incubated at 37°C for 24 hours and the clarified supernatantsanalysed using orcinol-suiphuric acid (see Materials and Methods, section 2.15.2). Errorbars represent standard deviations of the measurements.10450.tMp30 A50MCBD40iM CBD: l0iM p3025.tM CBD: 25iM p301 0M CBD: 40pM p30251.tM p30251iM CBDlOj.tM p301OMCBD25MM CenAI •5 10 15 20 25 30Soluble carbohydrate (jig/mi)251iM CBD:50iM p3050M p3025jiM CBD:40iM p304ORM p3025 M CSD: 301iM p3030.tM p3025M CBD5 10 15 20 25 30Soluble carbohydrate (jig/mi)0BHIt01054. Discussion4.1 Binding interaction of CBDCenA with celluloseThe biological activity of a protein or polypeptide depends on its structure andhow it interacts with other molecules, therefore analysis of the structure of thecellulolytic enzymes produced by an organism can provide insight into the mechanismof degradation of the substrate. In the absence of a three-dimensional structure, site-directed mutagenesis is a valuable tool which can be used to probe the function ofconserved residues within a protein or polypeptide. In an initial attempt to better ourunderstanding of the mechanisms by which CBDs interact with their substrates, two ofthe conserved tryptophan residues within CBDCenA were targeted for mutagenesis. Themutant polypeptides, CBDCenAW14A and CBDCenAW68A have significantly reducedequilibrium association constants (50- and 30-fold respectively) relative to the wild typeCBDCenA. The reduced affinities of these polypeptides suggest that W14 and W68 mayhave important roles in the binding interaction of CBDCenA with its substrate.Any interpretation of the roles of W14 and W68 requires confirmation of thestructural integrity of the mutant polypeptides CBDCenAW14A and CBDCe11AW68A, asit is possible that these residues play a role in the folding of CBDCenA. The CD and one-dimensional 1H NMR spectra of the mutant polypeptides were similar to those for thewild type polypeptide and equilibrium sedimentation centrifugation indicated that allthree polypeptides are monomers. Tryptophan fluorescence data also revealed that allthree polypeptides have emission maxima of around 330nm, which is characteristic ofburied tryptophan residues in stably folded structures (Lehrer and Fasman, 1967). Thesedata all strongly suggest that the reduced affinities of CBDCenAW14A andCBDCenAW68A for BMCC can be attributed to the loss of the respective tryptophanresidues rather than unfolding or aggregation of the mutant polypeptides.106The structure of the periplasmic maltodextrin binding protein (MBP) from E.coli, determined by high resolution crystallographic analysis, shows a large number ofaromatic residues in its ligand binding site (Spurlino et al., 1991). However, before thestructure of this protein was available, a model was designed to analyse the role oftryptophan residues in the activity and fluorescence properties of the MBP. This modelwas developed by comparing the tryptophan fluorescence properties of wild type tomutants of the MBP, in which one of each of the eight tryptophan residues within theMBP had been changed to alanine (Martineau et al., 1990). Many of the predictions forthe involvement of specific tryptophans in the binding interaction proved to be correctwhen the crystal structure of the MBP complexed with maltose was obtained (Spurlinoet al., 1991, 1992). An analogous approach for CBDCenA and the mutant polypeptidesCBDCenAW14A and CBDCenAW68A predicts that residues W14 and W68 are on thesurface of CBDCenA. The interaction of CBDCe11Awith BMCC induced changes in thefluorescence properties of these two tryptophans, shifting their emission maxima toshorter wavelengths, suggesting that their environment becomes less polar. Thisobservation implies that these tryptophans are directly involved in the bindinginteraction of CBDCenA with BMCC, with the loss of water from the binding siteplacing these tryptophans in a more hydrophobic environment. The quenching offluorescence by the addition of potassium iodide also suggests that residues W14 andW68 are located on the surface of CBDCe11A. All this evidence is supported by threedimensional NMR data for the related CBD of Cex, where three of the conservedtryptophan residues, including those which correspond to W14 and W68, are exposed tosolvent and a fourth is partially exposed (G.Y. Xu, L. Kay and J. Carver, manuscript inpreparation).High resolution X-ray crystallography of several carbohydrate binding proteinshas demonstrated the importance of van der Waals’ contacts of aromatic side chains onsugar rings in protein-carbohydrate interactions (Quiocho and Vyas, 1984; Vyas et al.,1071988; Johnson et al., 1988; Spurlino et al., 1992). Proteins which bind carbohydratesmust distinguish between a large number of different sugar structures. Hence it wasproposed that the stacking interaction of aromatic residues with sugar rings providesspecificity by preventing the binding of particular sugar epimers through sterichindrance or polarity of the hydroxyls (Quiocho, 1986, 1988).The structures of several glycosidases show aromatic residues in the active orbinding sites of these enzymes (for example, Kraulis et al., 1989; Rouvinen et al., 1990;Juy et al., 1992; Speizo et al., 1993; Coutinho and Reilly, 1994; Maenaka et al., 1994;Wakarchuk et al., 1994). In several cases the structures of these enzymes whencomplexed with ligands indicate direct van der Waals’ contacts between aromaticresidues and sugar rings. For example, three tryptophan residues in the active site ofCBHII stack with the faces of glycosyl residues of the complexed ligand (Rouvinen eta!., 1990), and a Bacillus circulans xylanase complexed with xylotetraose reveals astacking interaction of an active site tryptophan residue with one of the xylose rings(Wakarchuk et al., 1994). Similarly, analysis of a Therniomonosporafuscaendoglucanase complexed with cellobiose shows tluee tryptophans lining the active sitecleft, one of which is observed to stack with the surface of the non-reducing glycosylring (Speizo et al., 1993). The X-ray structure of a Bacillus circulans cyclodextringlycosyltransferase (an enzyme involved in the conversion of starch) indicates extensivehydrophobic stacking of bound maltose molecules with aromatic residues in the bindingsite of the enzyme (Lawson et al., 1994). Some of these aromatic residues are highlyconserved within starch binding domains from a wide variety of starch convertingenzymes (Svensson et al., 1989). The 3D structure of the CBD of CBHI shows that thespacing of conserved tyrosine residues exposed along one edge of the wedge shapedCBD is equal to that between repeating cellobiose units in the cellulose crystal(Reinikainen et al., 1992).108Gilkes et al (1992) proposed a model for the binding of intact enzymes andCBDs to cellulose, and calculated that CBDCenA covers about 39 cellobiosyl residues onthe cellulose surface. All the data obtained in this present study support the proposalthat W14 and W68 are exposed on the surface of the CBD and are involved ininteracting simultaneously with these cellobiosyl residues, probably through van derWaals’ stacking interactions of the types described above. However, without a three-dimensional structure of CBDCe11A,it is not possible to specify the exact placement orinvolvement of these residues in the binding interaction of the polypeptide with itssubstrate. The structures of only two CBDs have been determined, the CBD of CBHIfrom the fungus T. reesei (Kraulis et al., 1989) which belongs to the family I CBDs (Dr.P. Tomme, unpublished data) and the CBD of Ccx from C.flmi (G.Y. Xu, L. Kay and 3.Carver, manuscript in preparation) which belongs to the family II CBDs, of whichBDCenA is also a member. These structures demonstrate the presence of conservedaromatic residues on the surfaces of both polypeptides.There are two other conserved tryptophan residues within CBDCenA. These mayalso be important for the binding interaction in a manner similar to W14 and W68.Amino-terminal deletions of CenA by a C.fimi protease gives the polypeptides p45 andp43, missing two (W14 and W34) and three (W14, W34 and W50) of the conservedtryptophan residues respectively. These polypeptides still bind to cellulose, althoughmore weakly than CenA. Peptides in which N-terminal truncations remove all of theconserved tryptophans do not bind to cellulose significantly, and neither does theisolated catalytic domain (Gilkes et al., 1989). Other highly conserved amino-acids inCBDCenA include the aromatic residues Y8, F18, F80 and F82. These could also beinvolved in stacking interactions with sugar units of the substrate.Hydrogen-bonding interactions are also significant forces in determiningspecificity and stability in protein-carbohydrate complexes (Quiocho, 1986, 1988).Three-dimensional determinations of the arabinose binding protein and galactose109binding protein from E. coli complexed with their substrates indicate that all of the H-bonding groups of the saccharides are involved in interactions with charged and polaramino-acid side chains (Quiocho, 1986). Similarly, the three-dimensional structure ofthe MBP with bound maltose shows that all the disaccharide hydroxyl groups areinvolved in hydrogen bonds, either directly or via water molecules, with polar andaromatic amino-acid side chains (Spurlino et al., 1991). CBDCe11Acontains few chargedamino-acid residues, therefore it seems highly likely that the large number of hydroxyside chain amino-acids in CBDCenA participate in H-bonding interactions. Conservedtryptophan, tyrosine and asparagine residues might also participate in such interactions.The role of the strictly conserved glycine residues (G 17 and G8 1) in CBDCenA isunclear at present, but their importance may be in helping to maintain the conformationof the polypeptide. Mutagenesis studies and the three-dimensional structure ofCBDCenA should help in elucidating the roles of these conserved residues and of theother conserved residues within the polypeptide.Tryptophan fluorescence of CBDCenA is not affected significantly by theaddition of the soluble glucans cellohexaose (G6), carboxymethylcellulose (CMC) orhydroxyethylcellulose (HEC), implying that CBDCenA has very low or no affinity forthese substrates. Alternatively, the binding interaction between CBDCenA and thesesoluble glucans might not bring about the change in the polarity of the fluorescenceemitting residues, which is seen with the binding of the insoluble substrate BMCC. Thebinding of CBDCenA to a single-chain substrate may only induce small, local structuralmodifications of the polypeptide compared to the larger changes observed when thepolypeptide binds to a crystalline substrate where the interaction is likely with severaladjacent chains. It should also be noted that CMC has a negative charge, which mightexplain why no interaction between CBDCenA and this sugar was observed. Howeverthere is other evidence to suggest that CBDCenA does not bind to these solublesubstrates. One-dimensional 1H NMR was used to probe the interaction of CBDCenA110with the oligosaccharides G2-G6 (i.e. cellobiose to cellohexaose). No changes in thespectra of CBDCe were observed on the addition of these oligosaccharides (P.Johnson, personal communication). CBDCenA and CBDex were also used incompetitive binding assays with phosphoric acid swollen cellulose and BMCC to showthat neither polypeptide interacts with HEC or CMC (Dr. P. Tomme, personalcommunication). Competitive binding assays and one dimensional 1H NMR show thatthe binding domain of CenC (Ni), which binds amorphous but not crystalline cellulose(Coutinho et al., 1992), does interact with some of these soluble glucans (P. Johnson andDr. P. Tomme, personal communication). The interactions of CBDs with substratessuch as CMC and HEC are difficult to measure using one dimensional 1H NMR due tothe viscosity of the sugar solutions, but micro-calorimetry techniques can be used forsuch purposes (Dr. Charles Haynes, personal communication). These investigations arecurrently under way. Additional investigations of the interactions of C.fimi CBDs witha range of soluble glycans are also being undertaken. The data compiled from theseexperiments may show that some C.fimi CBDs, such as CBDCenA interact only withinsoluble substrates, including amorphous and crystalline cellulose, whereas others suchas CBDen interact only with soluble and amorphous substrates and not crystallinecellulose. The significance of such differences in the binding specificities of differentCBDs is not understood, but may be related to the roles of these CBDs in the hydrolysisof cellulose, as will be discussed in section Role of CBDCenA in the hydrolysis of celluloseAlthough cellulase systems from a variety of different organisms are currently beingstudied, much of the information relating to the characterisation of cellulases is at thebiochemical level and there is relatively little information on how these enzymes alterthe morphology of their substrates during the process of hydrolysis.inIt has been recognised for some time that cellulose is fragmented duringenzymatic attack. During the early stages in the breakdown of native cellulose,cellulolytic fungi release very short, insoluble fibres before producing significantamounts of soluble sugar (Marsh, 1957; King, 1966; Liu and King, 1967; Halliwell andRiaz, 1970; Kyriacou et al., 1987). Short fibre formation has also been demonstrated bycellulases of bacterial origin (Poulsen and Peterson, 1992; Walker et al., 1992). Thesignificance of fragmentation in the overall process of cellulose degradation is not clear,but an increase in the surface area of the substrate could enhance its rate of hydrolysis(King, 1966). The effects of cellulase systems on the morphology of the substrates havealso been examined by electron microscopy (White and Brown, 1981; Chanzy andHenrissat, 1983; Sprey and Bochem, 1993; Hoshino et aL, 1993). Most of theseinvestigations have concentrated on the effect of whole enzymes or enzyme mixtures onthe cellulose substrates and in many cases results have been confounded by thepossibility of minor contaminants in enzyme preparations (for example, Halliwell andRiaz, 1970; Chanzy and Henrissat, 1983; Kyriacou et al., 1987). Molecular cloningtechniques provide a means of obtaining pure enzymes or enzyme fragments withoutcontaminating background activities. The cenA gene of C.fimi has been cloned andexpressed in E. coli (Wong et al., 1986). CenA is cleaved by an extracellular C.fimiprotease into two fragments: the catalytic domain, p30 and the cellulose binding domainwith the Pro,’Thr linker, CBD.PTCeiA (Gilices et al., 1988). The gene fragment encodingCBD.PTCenA has also been cloned and expressed in E. coli (Gilkes et al., 1992). Theavailability of the independent catalytic domain and binding domain of CenA hasallowed the effects of these individual components on cellulose to be investigated. Moreimportantly, it has allowed analysis of the role of the cellulose binding domain of CenAin the hydrolysis of cellulose.The isolated CBD.PTCenA has no detectable hydrolytic activity (Gillces et al.,1988); however it causes roughening of the surface of Ramie and cotton fibres and the112release of small particles from cotton. The way in which CBD.PTCenA facilitates suchdisruption of cellulose fibres is not understood. Examination of the substrates showsthat cotton fibres have a central lumen surrounded by 2-3 layers of secondary cell wall, aprimary cell wall and an outer waxy cuticle (Nevell and Zeonian, 1985). Ramie fibreshave only a single cell wall, protected by an outer layer of gum (Cheek and Roussel,1989). Fibres which have been pretreated to remove wax or gum have numerous surfaceirregularities and some transverse fissures, presumably as a result of mechanicaldamage. These surface irregularities may comprise partially exfoliated microfibrillarsheets (i.e. small particles) which are still bound non-covalently to underlying fibres orare attached at their ends to more coherent microfibrils of the fibre structure. It can bepostulated that CBD.PTCenA binds to and penetrates the fibre at surface discontinuitiesand sloughs off these non-covalently associated cellulose fragments. Further penetrationof CBD.PTCenA could exfoliate the fibre structure to release microfibrils which remainbound to the fibre and cause roughening of the surface.Over 40 years ago Reese et al (1950) proposed that a non-hydrolytic componentof the cellulase system, termed C1, makes the substrate more accessible for thehydrolytic enzymes, or C components of the system. C1 was proposed to be a proteinthat binds to cellulose and disrupts the bonding networks in the crystal lattice, ultimatelyto collapse the ordered structure of the substrate (Liu and King, 1967; Rautela and King,1968). The swelling, defibrillation and production of small particles from cotton havebeen attributed to particular hydrolytic components of the cellulolytic system whichwere labeled as C1. (Wood, 1989), but a strictly non-hydrolytic component has neverbeen identified. The observed properties of CBD.PTCenA indicate that it acts like the C1factor proposed by Reese et al (1950), in that the binding of this polypeptide tocellulose, which likely involves the H-bonding and hydrophobic stacking interactionsdiscussed earlier, causes the disruption of the substrate. However, at present it is notknown if CBD.PTCenA has true Ci activity i.e. the capacity to disrupt bonds which113stabilise the 8-1,4 glucan chains within the cellulose crystal. To demonstrate this, theeffect of the CBD on a cellulosic substrate must be examined at the atomic level,perhaps using techniques such as atomic force spectroscopy, 13C solid state NMR orRaman spectroscopy.The three-dimensional structure of the synthetic CBD of CBHI has beendetermined and shows that the domain has a wedge-like structure, one face being flatand hydrophilic and the other face being more hydrophobic (Kraulis et al., 1989). Theintroductory description of the structure of crystalline cellulose indicated that cellulosechains are joined together by H-bonding networks in one plane to form sheets ofcellulose and that these sheets are held together in layers by van der Waals’ forces. Itwas therefore proposed that both faces of the CBHI CBD could interact with thesubstrate and in this way facilitate the disruption of the crystalline structure. Similarlythe three-dimensional structure of CBDCe11Amay provide some insight as to how thepolypeptide interacts with crystalline cellulose and disrupts fibre structure. Earlierdiscussions (section 4.1) argued that conserved tryptophan residues within CBDCenA arelocated on the surface of the polypeptide and are positioned to interact with thesubstrate. The 3D structure of the related CBDex shows three conserved tryptophanresidues exposed on the surface of this polypeptide, with a fourth being partiallyexposed. The location of these conserved tryptophan residues on the surface of thedomains is consistent with the model in which these residues interact with the substratesurface. The size and location of these residues may allow them to slot into the spacebetween the layers of cellulose sheets to disrupt the existing van der Waals’ forces and inthis way collapse the sheet-to-sheet bonding networks in the crystalline structure of thesubstrate (Ted Atkins, personal communication).The mutant polypeptides CBDCenAW14A and CBDCenAW68A did not releasesmall particles from cotton. This may be explained as a combined resultant of theirreduced affinity for cellulose and a direct consequence of the loss of the respective114tryptophan residues which interact with the cellulose chains as proposed above. It isalso of interest to note that derivatives of CBD.PTCenA with two cellulose bindingdomains in tandem, released more small particles from cotton than CBD.PTCenA (D.Nordquist, 1992).Removal of the CBD from CenA by proteolysis gives the catalytic domain, p30.This polypeptide exhibits decreased affinity for crystalline substrates but its capacity tohydrolyse soluble substrates is unaltered (Gilkes et al., 1988). The action of p30 oncellulose appears to be different from that of the intact enzyme CenA or the isolatedbinding domain CBD.PTCenA. The latter two polypeptides roughen the surface of fibreswhereas the cellulose is polished by p30. It is likely that p30 hydrolyses glycosidicbonds on the surface of the cellulose fibre where the surface is disrupted or has brokenaway from the crystalline structure, giving the observed smoothing effect. The fragmentsreleased by p30 from cotton are also different from those released by CBD.PTCenA,being mostly short fibres as opposed to small particles. These may arise by hydrolysisof the fibres at regions of mechanical damage which extend across the entire fibre width(Wood, 1991).The extensive penetration of FITC-CBD probe following CenA pretreatment islikely a consequence of the hydrolysis of disordered regions within the Ramie fibre;however it appears that internal hydrolysis is dependent on concomitant disruption bythe CBD since penetration was not observed following treatment with p30 alone.Presumably the effect of CenA is the net result of the co-operative activities ofCBD.PTCenA and p30. The resultant rough surface of cellulose fibres treated with CenAsuggests that the polishing activity of p30 lags behind the disruptive effects of the CBD.Synergy experiments with the isolated CBD and p30 show that these two domains do infact interact co-operatively when hydrolysing cotton to give greater total sugar releasethan the individual domains alone.115It has been known for some time that cellulases act synergistically in degradingcrystalline cellulose. Two types of synergy have been described previously; thatbetween endoglucanases and exoglucanases (endo-exo synergism) and that betweenexoglucanases (exo-exo synergism). Endo-exo synergism was first described in the1970’s when it was suggested that randomly acting endoglucanases of a cellulase systemwould first hydrolyse amorphous regions of the cellulose fibres generating non-reducingends for the exoglucanases (Wood and McCrae, 1978; Eriksson, 1978). Exo-exosynergism was demonstrated between CBHI and CBHII from T. reesei (Fagerstam andPettersson, 1980; Henrissat et al., 1985). The results of this present study show clearlythat there is intramolecular synergism between the cellulose binding and catalyticdomains of CenA in the hydrolysis of cotton fibres. This is the first time that inter-domain synergism has been described, and it represents a third type of synergism seen inthe hydrolysis of cellulose.From the discussion presented above of the action of the individual domains oncotton and Ramie fibres, a model for synergy between p30 and CBD.PTCenA can bepostulated in which CBD.PTCenA releases particulate material and cellulose chain ends,thereby facilitating hydrolysis by p30. Mixtures of p30 and CBD.PTCenA are not aseffective as CenA itself in liberating particles or soluble sugars from the substrate. Ifsome particles are attached both covalently and non-covalently to the rest of the fibrestructure, close association of the domains should make the intact enzyme more efficientat releasing these particles. Furthermore, when p30 and CBD.PTCe11Aare separated,they may compete for sites in the cotton fibres, such as areas where the crystallinestructure is disrupted, thereby decreasing their overall efficiency relative to CenA. Thiswould also explain the decrease in synergy seen at high CBD concentrations. It wouldalso follow that the ratio of p30 to CBD.PTCe11A,as well as their concentrations, mightbe important factors in determining the optimal degree of synergism obtained with aparticular substrate, as has been demonstrated from previous studies investigating116synergism between individual cellulases (Henrissat et al., 1985; Tornme et al., 1988;Wood and McCrae, 1986; Woodward et al., 1988).There was no synergy between p30 and CBD.PTCenA in the hydrolysis ofBMCC. The reasons for this likely relate to the differences between the structures ofcotton and BMCC. Henrissat et al (1985) proposed that the degree of synergismbetween cellulase components depends on several structural and ultrastructural featuresof the substrate including degree of crystallinity, degree of polyrnerisation and particlesize. Cotton microfibrils are much less coherent than bacterial cellulose microfibrils,partly due to differences in packing of the glucan chains within the microfibrils (Kolpakand Blackwell, 1975), which is much more disordered in cotton microfibrils than inbacterial cellulose microfibrils, affecting the degree of crystallinity of the cellulose. IfCBD.PTCenA disrupts cellulose where the intercrystalline contacts are relatively weak,then cotton would be a more readily accessible substrate than the bacterial cellulose,perhaps allowing easier penetration of CBD.PTCe11A. It has been proposed that theCBDs of CenA and Cex prevent flocculation of BMCC by preventing reassociation ofcellulose chains (Gilkes et al., 1993), and therefore if CBD.PTCenA penetrates themicrofibril structure of cotton more readily than that of BMCC, it could loosen up thestructure further making this more accessible to p30. If the microfibrils of bacterialcellulose are held in a much tighter conformation, then the ability of CBD.PTCenA topenetrate the microfibrils is decreased, translating into decreased synergy betweenCBD.PTCenA andp3O.Cotton cellulose consists of thick bundles of microfibrils with widths of 0.1-O.5p.m and several cm long (Kolpak and Blackwell, 1975), whereas BMCC preparedfrom Acetobacter xylinum pellicles (Hestrin, 1963) consists of smaller bundles ofmicrofibrils 20-5Onm wide and several trn long (White and Brown, 1981; Henrissat andChanzy, 1986). Therefore, the decreased effectiveness of CBD.PTCenA On BMCCmight be due simply to the larger surface area of this substrate compared to cotton. The117dispersive action of the CBD would then be insignificant on this substrate compared tocotton.The synergy observed between CBD.PTCenA and p30 of CenA may explain whymany cellulases have multi-domain structures with distinct CBDs and catalytic domains.The CBD would not only bring the catalytic domain into close association with thesubstrate but also function to disorder the substrate to allow the attached catalyticdomain easier access. All of the C.fimi enzymes described to date have discrete CBDs(Wong et al., 1986; Meinke et al., 1991, 1991a; Coutinho et al., 1991, 1992; Meinke etat, 1993, 1994; Shen et al., unpublished data). The CBD of Cex behaves similarly tothat of CenA in releasing small particles from cotton and disrupting the structure ofRamie fibres. Similar studies have not yet been carried out with the CBDs of the otherC.fimi glycanases so it also remains to be seen if intramolecular synergism is commonto all of these enzymes. It would be of particular interest, for instance, to examine theeffects of the isolated CBDs of CenB and CenD as both enzymes readily solubilisecrystalline cellulose (K. Kleman-Leyer, N. R. Gilkes, R. C. Miller and T. K. Kirk,manuscript in press). CenB contains two cellulose binding domains, neither of which isneeded for hydrolysis of soluble substrates; however, removal of one of these domainsreduces the activity of the enzyme (Meinke et al., 1991a).This study is the first to examine the effects of isolated CBDs on cellulose.However, there is evidence from other sources that CBDs may play a vital role in thehydrolysis of crystalline cellulose. Removal of CBDs from fungal cellulases reducestheir activity on insoluble cellulose, but not on soluble cellulose (Van Tilbeurgh et al.,1986; Tomme et al., 1988) and a dispersive role of these CBDs has also been suggested(Knowles et al., 1988). A truncated form of a B. subtilis endoglucanase missing its CBDis more active than the intact enzyme on soluble substrates but less active on insolublecellulose (Hefford et al., 1992). Similar results are obtained when the CBDs are118removed from an endoglucanase from T.fisca and an endoglucanase from C.cellulovorans (Hamamoto et al., 1992; Wilson, 1992).It is quite conceivable that not all of the CBDs described to date have the samefunctional roles in the hydrolysis of cellulose. The significance of CBDs in enzymessuch as chitinases (Fuji and Miyashita, 1993) and hernicellulases (Hall and Gilbert,1988; Kellet et aL, 1990; Ferreira et al., 1993; Miliward-Sadler et al.,1994; Irwin et al.,1994) remains to be established. Removal of CBDs from two hemicellulases fromPseudomonasfluorescens subsp. cellulosa did not alter their catalytic activity (Ferreiraet a!., 1990, 1991). The role of these CBDs may be to target the enzymes to the plantcell wall. Alternatively if the CBDs in such enzymes act to disperse crystalline celluloseas well as to facilitate close association with the substrate, this would render the plantcell wall accessible to a whole range of cell wall hydrolases which could then actsynergistically to degrade the different substrates exposed during hydrolysis (Ferreira etat, 1993). The effects of isolated CBDs from hemicellulases on cellulose remain to beinvestigated, therefore it is not known if they act like the CBDs of CenA or Cex indisrupting the substrate. However, it is interesting to note that the CBD of Cex, which isa mixed function 13-1,4-xylanase/B-l,4-glucanase, also releases small particles fromcotton and disrupts the structure of cellulose fibres.A previous study has shown that the paired CBDs of CenC, N1N2 and Ni aloneadsorb to regenerated cellulose (considered to be amorphous or composed of celluloseII, Ooshima et al., 1983; Sarko, 1986) but not to crystalline cellulose (Coutinho et al.,1992) whereas CBDex and CBDCenA bind to both forms of cellulose (Ong et al., 1993;E. Kwan, personal communication). Furthermore, substitution of CBDCenA withCBDCenC in CenA gave a hybrid enzyme which does not bind to crystalline cellulose,and is less active on both BMCC and Avicel and more active on CMC (solublecellulose) than CenA (Coutinho et al., 1993). Therefore, it seems that the CBD of CenCmay not have the same role in cellulose hydrolysis as that of CBDCenA or CBDex. This119observation raises the point that some CBDs may act to make the substrate moreaccessible to enzymes whose CBDs do not recognise more crystalline forms of thesubstrate (Coutinho et al., 1992)Extensive studies of the properties of CBDs of C.fimi enzymes and those ofenzymes from other organisms are currently being undertaken and the results of suchstudies should help to clarify the role of CBDs in the overall scheme of cellulosehydrolysis.Therefore in summary, it has been demonstrated that two conserved tryptophanresidues within CBDCenA are involved in the binding interaction of the domain with itssubstrate. These residues, along with other conserved aromatic residues withinCBDCenA are proposed to make van der Waals’ contacts with sugar residues on thesubstrate surface. In addition it has been shown that the CBD of CenA has a disruptiveeffect on cotton fibres and that the domain interacts synergistically with the catalyticdomain of CenA in hydrolysing cotton fibres. The CBD of CenA seems to have a nonhydrolytic activity similar to that proposed for the Ci component of the cellulase systemproposed by Reese et al (1950).1205. ReferencesAbbott, W.M. and Feizi, T. (1991). Soluble l4kDa B-galactoside-specific bovine lectin:evidence from mutagenesis and proteolysis that almost the complete polypeptide chainis necessary for integrity of the carbohydrate recognition domain. J. Biol. Chem., 266,5552-5557.Abuja, P.M., Pilz, I., Claeyssens, M. and Tomme, P. (1988). Domain structure ofcellobiohydrolase II as studied by small angle X-ray scattering: close resemblance tocellobiohydrolase I. Biophys. Res. Commun., 156, 180-185.Al-Tawheed, A.R. (1988). Molecular characterisation of cellulase genes fromCellulomonasfiavigena. MSc. thesis, Trinity College, London.Alberts, B., Bray, D., Lewis, V., Raff, M., Roberts, K. and Watson, J.D. (1983).Molecular Biology of the Cell. Garland Publishers, New York.Atalla, R.H. and VanderHart, D.L. (1984). Native cellulose: a composite of two distinctcrystalline forms. Science, 223, 283-285.Atalla, R.H. (1989). Pattern of aggregation in native celluloses: implication of recentspectroscopic studies. In: Cellulose: structural and functional aspects (Kennedy, J.F.,Phillips, G.O. and Williams, P.A., Eds), pg 61. Ellis Horwood Ltd. Publishers.Atalla, R.H., Hackney, J.M., Uhlin, I. and Thompson, N.S. (1993). Hemicelluloses asstructure regulators in the aggregation of native cellulose. Tnt. J. Biol. Macromol. 15,109-112Azevedo, M.D., Filipe, M.S., Astolfi-Filho, S. and Radford, A. (1990). Cloning,sequencing and homologies of the cbh-1 (exoglucanase) gene from Humicola grisea var.thermoidea. J. Gen. Microbiol., 136, 2569-2576.Bagnarda-Tardif, C., Gaudin, C., Belaich, A., Hoest, P., Citard, T. and Belaich, J.P.(1992). Sequence analysis of a gene cluster encoding cellulases from Clostridiumcelluloyticum. Gene, 119, 17-28.Béguin, P., (1990). Molecular biology of cellulose degradation. Ann. Rev. Microbiol.,44, 2 19-248.Berger, E., Jones, W.A., Jones, T.D. and Woods, D.R. (1989). Cloning and sequencingof an endoglucanase (Endi) from Butyrivibrio solvens H17c. Mol. Gen. Genet. 219,193-198.Bernstein, E.A., Vendenkina, N.S and Ivkova, M.N. (1973). Fluorescence and thelocation of tryptophan residues in protein molecules. Photochem. Photobiol. 18, 263-279.Blackwell, J. and Marchessault, R.H. (1971). Infrared spectroscopy of cellulose. In:Cellulose and cellulose derivatives (Bikales, N. and Segals, L.E. Eds), pg 1. WileyInterscience, New York.121Blackwell, J. (1982). The macromolecular organisation of cellulose and chitin. In:cellulose and other natural polymer systems (Brown, R.M., Jr., Ed), pg 403. PlenumPress, New York and London.Blake, M.S., Johnston K.H., Russell-Jones, G.J. and Gotschilch, E.C. (1984). A rapidsensitive method for detection of alkaline phosphatase-conjugated anti-antibody onWestern blots. Anal. Biochem. 136, 175-179.Bradford, M. (1976). A rapid and sensitive method for the quantitation of microgramquantities of proteins utilising the principle of protein-dye binding. Anal. Biochem. 72,248-254.Brummell, D.A., Sharma, V.P., Anand, N.N., Bilous, D., Dubuc, G., Michniewiecz, J.,Mackenzie, C.R., Sadowska, 3., Sigurskjold, B.W., Sinnott, B., Young., M.N., Bundle,D.R. and Narang, S.A. (1993). Probing the combining site of an anti-carbohydrateantibody by saturation-mutagenesis: role of the heavy chain CDR3 residues.Biochemistry, 32, 1180-1187.Cannon, R.E. and Anderson, S.M. (1991). Biogenesis of bacterial cellulose. Crit. Rev.Microbiol., 17, 435-447.Cantor, C.R. and Schimmel, P.R. (1980). Biophysical Chemistry, Part 2. SanFrancisco: W.H. Freeman, pg 380.Chanzy, H., Henrissat, B., Vuong, R. and Schtilein, M. (1983). The action of 1,4-13-D-glucan cellobiohydrolase on Valonia cellulose microcrystals. FEBS Lett., 153, 113-118.Chanzy, H. and Henrissat, B. (1985). Unidirectional degradation of Valonia cellulosemicrocrystals subjected to cellulase action. FEBS Lett., 184, 285-288.Chanzy, H. (1990). Aspects of cellulose structure. In: Cellulose sources andexploitation. Industrial utilisation, Biotechnology and Physio-chemical Properties(Kennedy, J.F., Philips, G.O. and Williams, P.A., Eds), pg 4. Ellis Horwood, NewYork.Cheek, L. and Roussel, L. (1989). Mercerization of Ramie: comparisons with flax andcotton. Part I: Effects on physical, mechanical, and accessibility characteristics. TextileRes. J., 59, 478-483.Chen, C.M., Grizaldi, M. and Stafford, D.W. (1987). Nucleotide sequence and deducedprimary structure of cellobiohydrolase II from Trichoderma reesei. BiolTechnology, 5,274-278.Cheng, C., Tsukagoshi, N. and Udaka, 5. (1990). Nucleotide sequence of thecellobiohydrolase gene from Trichodernia viricie. Nucleic Acids Res., 18, 5559-556 1.Chervenka, C.H. (1970). A manual of methods for the analytical ultracentrifuge, SpincoDivision of Beckman Instruments, Palo Alto, CA.Christoph, A.K., Weigel, T.O. and Schulz, G. (1993). Gene, 124, 57-65.122Claeyssens, M. and Tomme, P. (1989). Structure-function relationships of cellulolyticproteins from Trichoderma reesei. In: Trichoderma reesei cellulases: biochemistry,genetics, physiology and application (Kubicek, C.P et al., Eds), pg 1. Proceedings,Tricell, Royal Society of Chemistry.Cooper, V.J.C. and Salmond, G.P.C. (1993). Molecular analysis of the major cellulase(Ce1V) of Erwinia carotovora: evidence for a evolutionary ‘mix-match’ of enzymedomains. Mol. Gen. Genet. 241, 341-350.Coughian, M.P. (1985). The properties of fungal and bacterial cellulases with commenton their production and application. Biotechnology and Genetic Engineering Reviews,3, 39-109.Coutinho, J.B., Moser, B., Kilburn, D.G., Warren, R.A.J. and Miller, R.C. Jr. (1991).Nucleotide sequence of the endoglucanase C gene (cenC) of Cellulomonasfimi, its highlevel expression in Escherichia coli, and characterisation of its products. Mo!.Microbiol. 5, 1221-1233.Coutinho, J.B., Gilkes, N.R., Warren, R.A.J, Kilburn, D.G., and Miller, R.C., Jr. (1992).The binding of Cellulomonasfimi endoglucanase C (CenC) to cellulose and sephadex ismediated by the N-terminal repeats. Mol. Microbiol., 6, 1243-1252.Coutinho, J.B., Gilkes, N.R., Kilburn, D.G.,Warren, R.A.J. and Miller, R.C., Jr. (1993).The nature of the cellulose binding domain affects the activities of a bacterialendoglucanase on different forms of cellulose. FEMS Microbiol. Lett., 113, 211-218.Coutinho, P.M. and Reilly, P.J. (1994). Structure-function relationships in the catalyticand starch binding domains of glucoamylase. Protein Eng., 7, 393-400.Covert, S.F., Vandenwymelenberg, A. and Culleri, D. (1992). Structure, organisation,and transcription of a cellobiohydrolase gene cluster from Phanerochaetechrysosporium. Appl. Environ. Microbiol. 58, 2168-2175.reighton,T.E. (1984). Proteins. W.H. Freeman and Company. New YorkDavidson, R.L. (1980). Handbook of water soluble gums and resins. McGraw-Hill,New York.Din, N., Gilkes, N.R., Tekant, B., Miller, R.C., Jr., Warren, R.A.J. and Kilbum, D.G.(1991). Non-hydrolytic disruption of cellulose fibres by the binding domain of abacterial cellulase. Bio/Technology, 9, 1096-1099.Din, N., Forsythe, I.J., Burtnick, L.D., Gilkes, N.R., Miller, R.C., Jr., Warren, R.A.J. andKilbum, D.G. (1994). The cellulose binding domain of endoglucanase A (CenA) fromCellulomonasjlmi: evidence for the involvement of tryptophan residues in binding.Mol. Microbiol., 11, 747-755.Din, N., Gilkes, N.R., Miller, R.C., Jr., Warren, R.A.J. and Kilburn, D.G. (1994a). C1-C revisited: intramolecular synergism in a cellulase. Proc. Natl. Acad. Sci. U.S.A., inpress123Durrant, A.J., Hall, J., Hazeiwood, G.P. and Gilbert, H.J. (1991). The non-catalytic C-terminal region of endoglucanase E from Clostridium thermocellum contains a cellulosebinding domain. Biochem. J., 273, 28 9-293.Eriksson, K.E. (1978). Enzyme mechanisms involved in cellulose hydrolysis by the rotfungus Sporotrichum pulverulentum. Biotechnol. Bioeng., 70, 317-332.Fägerstam, L.G. and Pettersson, L.G. (1980). The 1,4-B-glucan cellobiohydrolase ofT.reesei QM9414. FEBS Lett., 119, 97-101.Ferreira, L.M.A., Durrant, A.J., Hall, J., Hazelwood, G.P and Gilbert, H.J. (1990).Spatial separation of protein domains is not necessary for catalytic activity or substratebinding in a xylanase. Biochem. J., 269, 26 1-264.Ferreira, L.M.A., Hazelwood, G.P., Parker, P.J. and Gilbert, H.J. (1991). Thecellodexthnase from Pseudomonasfluorescens subsp. cellulosa consists of multiplefunctional domains. Biochem. J., 279, 793-799.Ferreira, L.M.A., Wood, T.M., Williamson, G., Faulds, C., Hazelwood, G.P., Black, G.and Gilbert, H.J. (1993). A modular esterase from Pseudomonasfluorescens subsp.cellulosa contains a non-catalytic cellulose binding domain. Biochem. J., 294, 349-355.Frey-Wyssling, A. (1976). The Plant Cell Wall. Gebruder Borntrager, Berlin.Fuji, T. and Miyashita, K. (1993). Multiple domain structure in a chitinase gene (chiC)of Streptomyces lividans. J. Gen. Microbiol., 139, 677-686.Gardner, K.H and Blackwell, J. (1974). The structure of native cellulose. Biopolymers,13, 1975-2001.Ghangas, G.S. and Wilson, D.B. (1988). Cloning of the Thermornonosporafusca E2gene in Streptomyces lividans: affinity purification and functional domains of the clonedproduct. Appi. Environ. Microbiol. 54, 2521-2526.Gilbert, H.J., Hall, J., Hazelwood, G.P. and Ferreira, L.M.A. (1990). The N-terminalregion of an endoglucanase from Pseudomonasfluorescens subsp. cellulosa constitutesa cellulose binding domain that is distinct from the catalytic center. Mol. Microbiol., 4,759-767.Gilkes, N.R., Kilbum, D.G., Langsford, M.L., Miller, R.C., Jr., Wakarchuk, W.W.,Warren, R.A.J., Whittle, D.J. and Wong, W.K.R. (1984). Isolation and characterisationof Esherichia co/i clones expressing cellulase genes from Cellulomonasfimi. J. Gen.Microbiol., 130, 1377-1384.Gillces, N.R., Langsford, M. L., Kilburn, D.G., Miller, R.C., Jr. and Warren,R.A.J.(1984a). Mode of action and substrate specificities of cellulases from clonedbacterial genes. J. Biol. Chem., 259, 10455-10459.Gilkes, N.R., Warren, R.A.J., Miller, R.C., Jr. and Kilbum, D.G. (1988). Preciseexcision of the cellulose binding domains from two Cellulomonasfimi cellulases by ahomologous protease and the effects on catalysis. J. Biol. Chem., 263, 1040 1-10407.124Gilkes, N.R., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J. (1989). Structural andfunctional analysis of a bacterial cellulase by proteolysis. J. Biol. Chem., 264, 17802-17808.Gillces, N.R., Henrissat, B., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J. (1991).Domains in microbial 13-1,4-glycanases: sequence conservation, function and enzymefamilies. Microbial Rev., 55, 305-3 15.Gillces, N.R., Claeyssens M., Aebersold, R., Henrissat, B., Meinke, A., Morrison, H.D.,Kilbum, D.G., Warren, R.A.J. and Miller, R.C., Jr. (1991a). Structural and functionalrelationships in two families of 13-1,4-glycanases. Eur. J. Biochem. 202, 367-377.Gilkes, N.R., Jervis, E., Henrissat, B., Tekant, B., Miller, R.C., Jr., Warren, R.A.J. andKilburn, D.G. (1992). The adsorption of a bacterial cellulase and its two isolateddomains to crystalline cellulose. J. Biol. Chem., 267, 6743-6749.Gilkes, N.R., Kilbum, D.G., Miller, R.C., Jr., Warren, R.A.J., Sugiyama, J., Chanzy, H.,and Henrissat, B. (1993). Visualisation of the adsorption of a bacterial endo-13-1,4-glucanase and its isolated cellulose binding domain to crystalline cellulose. mt. J. Biol.Macromol. 15, 347-35 1.Giorda, R., Ohmachi, T., Shaw, D.R. and Ennis, H.L. (1990). A shared internalthreonine-glutamic acid-threonine-proline repeat defines a family of Dictyosteiumdiscoideum spore germination specific proteins. Biochemistry, 29, 7264-7269.Gonzales, R., Ramon, D. and Perez-Conzalez, J.A. (1992). Cloning, sequence analysisand yeast expression of the egi gene from Trichoderma iongibrachium. Appi.Microbiol. Biotechnol., 38, 370-375.Greenwood, J.M., Gilkes, N.R., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J.(1989). Fusion to an endoglucanase allows alkaline phosphatase to bind to cellulose.FEBS Lett., 244, 127-13 1.Greenwood, J.M., Ong, E., Gilkes, N.R., Warren, R.A.J., Miller, R.C., Jr. and Kilburn,D.G. (1992). Cellulose-binding domains: potential for purification of complex proteins.Prot. Eng. 5, 361-365.Grengross, U.T., Romaniec, M.P.M., Kobayashi, T., Huskisson, N.S. and Demain, A.L.(1993). Sequencing of a Ciostridium thermocellum gene (cipA) encoding thecellulosomal SL-protein reveals an unusual degree of internal homology. Mol.Microbiol., 8, 325-334.Guo, Z., Arfman, N., Ong, E., Gilkes, N.R., Kilburn, D.G., Warren, R.A.J. and Miller,R.C., Jr. (1988). Leakage of Ceilulonionasfirni cellulases from Escherichia coii.FEMS Microbiol. Lett., 49, 279-283.Hall, J. and Gilbert, H.J. (1988). The nucleotide sequence of a carboxymethylcellulasegene from Pseudomonasfluorescens subsp. ceilulosa. Mol. Gen. Genet., 213, 122-117.Hall, J. Hazelwood, G.P., Huskisson, N.S., Durrant, A.J. and Gilbert, H.J. (1989).Conserved serine-rich sequences in xylanases and cellulases from Pseudomonasfluorescens subsp. celluiosa: internal signal sequence and unusual protein processing.Mol. Microbiol., 3, 1211-1219.125Halliwell, G. (1965). Hydrolysis of fibrous cotton and reprecipitated cellulose bycellulolytic enzymes from soil micro-organisms. Biochem. J., 95, 270-28 1.Halliwell, G. and Riaz, M. (1970). The formation of short fibres from native celluloseby components of Trichoderma koningii cellulase. Biochem. J., 116, 35-42.Hamamoto, T., Foong, F., Shoseyov, 0. and Doi, R.H. (1992). Analysis of functionaldomains of endoglucanases from Clostridium cellulovorans by gene cloning, nucleotidesequencing and chimeric protein construction. Mol. Gen. Genet., 231, 472-479.Hansen, C.K., Diderichsen, B. and Jorgensen, P.L. (1992). celA from Bacillus lautusPL236 encodes a novel cellulose binding endo-13-1,4-glucanase. J. Bacteriol., 174,3522-3531.Harris, D.L. and Hudson, B.S. (1990). Photophysics of tryptophan in bacteriophage T4lysozyme. Biochemistry, 29, 5276-5285.Hazeiwood, G.P., Davidson, K., Laurie, J.I., Huskisson, N.S. and Gilbert, H.J. (1992).Gene sequence and properties of Cell, a family E endoglucanase from Clostridiumthermocellum. J. Gen. Microbiol., 139, 307-316.Hefford, M.A., Laderoute, K., Willick, G.E., Yaguchi, M. and Seligy, V. (1992).Bipartite organisation of the Bacillus subtlis endo-B-1,4-glucanase revealed by C-terminal mutations. Protein Eng., 5, 433-439.Helfman, D.M. and Hughes, S.H. (1987). Use of antibodies to screen cDNA expressionlibraries prepared in plasmid vectors. Methods Enzyrnol., 152, 45 1-547.Henrissat, B., Driguez, H., Viet, C. and Schülein, M. (1985). Synergism of cellulasesfrom Trichoderma reesei in the degradation of cellulose. Bio/’T’echnology, 3, 722-725.Henrissat, B. and Chanzy, B. (1986). Enzymatic breakdown of cellulose crystals. In:Cellulose structure, modification and hydrolysis (Young, R.A. and Rowell, R.M. Eds),pg 337. John Wiley and Sons, New York.Henrissat, B., Vigny, B., Buleon, A., and Perez, S. (1988) Possible adsorption sites ofcellulases on crystalline cellulose. FEBS Lett., 231, 177-182Henrissat, B., Claeyssens, M., Tomme, P., Lemesle, L. and Momon, J.P. (1989).Cellulase families revealed by hydrophobic cluster analysis. Gene, 81, 83-95.Hestrin, S. (1963). Bacterial Cellulose. In: Methods in Carbohydrate Chem. 3 (Whistler,R.L. Ed), pg 4. Academic Press, New York.Hieta, K., Kuga, S. and Usuda, M. (1984). Electron staining of reducing ends evidencesa parallel-chain structure in Valonia cellulose. Bioploymers, 23, 1807-18 10.Hoshino, E., Sasaki, Y., Mori, K., Okazaki, M., Nisizawa, K. and Kanada, T. (1993).Electron microscopic observations of cotton cellulose degradation by exo- and endotype cellulases from Irpex lacteus. J. Biochern., 114, 236-245.Hudson, L. and Hay, F.C. (1976). Practical Immunology. Blackwell ScientificPublications, Oxford, U.K.126Irwin, D., Jung, E.D. and Wilson, D.B. (1994). Characterisation and sequence of aThermonosporafusca xylanase. Appi. Environ. Microbiol., 60, 763-770.Jauris, S., Rucknagel, K.P., Schwarz, W.H., Kratzsch, P., Bronnenmeir, K. andStaudenbauer, W.L. (1990). Sequence analysis of the Clostridium stercorarium celZgene encoding a thermoactive cellulase (Avicelase I): identification of catalytic andcellulose binding domains. Mol. Gen. Genet., 223, 258-267.Johansson, G., Stahlberg, I., Lindeberg, G., Engstrom, A. and Pettensson, G (1989).Isolated fungal cellulase terminal domains and a synthetic minimum analogue bind tocellulose. FEBS Lett., 243, 389-393.Johnson, L.N., Cheetham, J., McLaughlin, P.1., Acharya, K.R., Barford, D. and Phillips,D.C. (1988). Protein-oligosaccharide interactions: Lysozyrne, phosphorylase, amylases.Curr.Top. Microbiol. Irnrnunol., 139, 81-134.Juy, M., Amit, A.G., Aizari, P.M., Poijak, R.J., Claeyssens, M., Béguin, P. and Aubert,J.P. (1992). Three dimensional structure of a thermostable cellulase. Nature, 357, 89-91.Kellet, L.E., Poole, D.M. Ferreira, L.M.A., Durrant, A.J., Hazelwood, G.P. and Gilbert,H.J. (1990). Xylanase B and an arabinofuranosidase from Pseudomonasfluorescenssubsp. cellulosa contain identical cellulose binding domains and are encoded by adjacentgenes. Biochem. J., 227, 369-376.Kilbum, D.G., Assouline, Z., Din, N., Gilkes, N.R., Ong, E., Tomme, P. and Warren,R.A.J. (1993). Cellulose binding domains: properties and applications. In: Proceedingsof the second Tricel symposium on ‘T. reesei cellulases and other hydrolases’(Svominen, P and Reinikainen, T. Eds), pg 281. Foundation for Biotechnical andIndustrial Fermentation Research 8, Espoo Finland.King, K.W. (1966). Enzymatic degradation of crystalline hydrocellulose. Biochem.Biophys. Res. Comm., 24, 295-298.Knowles, J., Lehtovaara, P. and Teen, T. (1987). Cellulase families and their genes.Trends. Biotech., 5, 255-26 1.Knowles, J., Teen, T., Lehtovaara, P. Penttilä, M. and Saloheimo, M. (1988). The useof gene technology to investigate fungal cellulolytic enzymes. In: Proceedings of FEMSsymposium No.43. Biochemistry and Genetics of Cellulose degradation (Aubert, J.P.,Béguin, P and Millet, J. Eds.), pg 153. Academic press, London and New York.Kolpak, RJ. and Blackwell, J. (1975). Deformation of cotton and bacterial cellulosemicrofibnils. Textile Res. J., 45, 568-572.Konev, S.V. (1967). Fluorescence and phosphorescence of proteins and nucleic acids.Plenum Press, New York.Kraulis P.J., Clore, G.M., Nilges, M., Jones T.A., Petterson, G., Knowles, J. andGronenbom, A.M. (1989). Determination of the three-dimensional solution structure ofthe C-terminal domain of cellobiohydrolase I from Trichoderma reesei. A study usingnuclear magnetic resonance and hybrid distance geometry-dynamical simulatedannealing. Biochemistry, 28, 7241-7257.127Kuga, S. and Brown, R.M., Jr. (1987). Practical aspects of lattice imaging of cellulose.3. Electron Microsc. Technol., 6, 349-356.Kuga, S. and Brown, R.M., Jr. (1988). Silver labeling of the reduced ends of bacterialcellulose. Carbohydrate Research, 180, 345-350.Kulshreshtha, A.K. and Dweltz, N.E. (1973). Paracrystalline lattice disorder incellulose. I. Reappraisal of the application of the two-phase hypothesis to the analysis ofpowder X-ray diffractograms of native and hydrolysed cellulosic materials. J. Polym.Sci. Polym. Phys. Ed., 11, 487-497.Kunkel, T.A., Roberts, J.D. and Zakour, R.D. (1987). Rapid and efficient site-specificmutagenesis without phenotypic selection, Methods Enzymol., 154, 367-382.Kyriacou, A., Mackenzie, C.R. and Neufeld, R.J. (1987). Detection and characterisationof the specific and non-specific endoglucanases of Trichoderma reesei: evidencedemonstrating endoglucanase activity by cellobiohydrolase II. Enzyme Microb.Technol., 9, 25-32.Laemmli, U.K. (1970). Cleavage of structural proteins during the assembly of the headof bacteriophage T4. Nature, 227, 680-685.Lakowicz, J.R. (1983). Principles of Fluorescence Spectroscopy. Plenum Press. NewYork and LondonLangsford, M.L., Gilkes, N.R., Wakarchuk, W.W., Kilburn, D.G., Miller, R.C., Jr., andWarren, R.A.J. (1984). The cellulase system of Cellulomonasfirni. J. Gen. Microbiol.,130, 1367-1376.Langsford, M.L., Gilkes, N.R., Singh, B., Moser, B., Miller, R.C., Jr., Warren, R.A.J.and Kilburn, D.G.(1987). Glycosylation of bacterial cellulases prevents proteolyticcleavage between functional domains. FEBS Lett., 225, 163-167.Lao, G., Ghangas, G.S., Jung, E.D. and Wilson, D.B. (1991). DNA sequences of threei3-1,4-endoglucanase genes from Thermomonosporafusca. 3. Bacteriol., 173, 3397-3407.Lawson, C.L., Van Montfort, R., Strokopytov, B., Rozeboom, H.J., Kalk, K.H., deVries,G.E., Penninga, D., Dijkhuizen, L. and Dijkstra, B.W. (1994). Nucleotide sequence andX-ray structure of cyclodextrin glycosyltransferase from Bacillus circulans strain 251 ina maltose-dependent crystal form. J. Mol. Biol., 236, 590-600.Lee, S.B., Shin, H.S. and Ryu, D.D.Y. (1982). Adsorption of cellulase on cellulose:effect of physiochemical properties of cellulose on adsorption and rate of hydrolysis.Biotechnol. Bioeng., 24, 2137-2153.Lehrer, S.S., and Fasman, G.D., (1967). Fluorescence of lysozyrne and lysozymesubstrate complexes; separation of tryptophan contributions by fluorescence differencemethods. J. Biol. Chem., 242, 4644-4651Lui, T.H. and King, K.W. (1967). Fragmentation during enzymatic degradation ofcellulose. Arch. Biochem. Biophys., 120, 462-464.128MacKay, R.M., Lo, A., Willick, G., Zuker, M., Baird, S., Dove, M., Moranelli, F. andSeligy, V. (1986). Structure of a Bacillus subtilis endo-13-1,4-glucanase gene. NucleicAcids Res., 14, 9159-9 170.Maenaka, K., Kawal, G., Watanabe, K., Sunada, F. and Kumagai, I. (1994). Functionaland structural role of a tryptophan generally observed in protein-carbohydrateinteraction. J. Biol. Chem., 269, 7070-7075.Marchessault, R.H. and Sarko, A. (1967). X-ray structure of polysaccharides. Adv.Carbohydr. Chem., 22, 421-482.Marsh, P.B. (1957). Microscopic observations on cotton fibres subjected to enzymaticdegradation. Textile Res. J., 28, 913-916.Martineau, P., Szmelcman, S., Spurlino, J.C., Quiocho, F.A. and Hofnung M. (1990).Genetic approach to the role of tryptophan residues in the activities and fluorescence ofa bacterial periplasmic maltose-binding protein. J. Mol. Biol. 214, 337-352Matsudaira, P. (1990). Limited N-terminal sequence analysis. Methods Enzymol. 182,602-613.Meinke, A., Braun, C., Gilkes, N.R., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J.(1991). Unusual sequence organisation in CenB, an inverting endoglucanase fromCellulomonasfimi. J. Bacteriol., 173, 308-3 14.Meinke, A., Gilkes, N.R., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J. (1991a).Multiple domains in endoglucanase B (CenB) from Cellulomonasfinii: functions andrelatedness to domains in other polypeptides. J. Bacteriol., 173, 7126-7135.Meinke, A., Gilkes, N.R., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J. (1993).Cellulose-binding polypeptides from Cellulonionasfimi: endoglucanase D (CenD), afamily A B-1,4-glucanase. J. Bacteriol., 175, 1910-1918.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 Cellulomonasfimiis a 13-1,4-exocellobiohydrolase analogous to Trichoderma reesei CBH II. Mol.Microbiol., 12, 413-422.Messing, J. (1983). New M13 vectors for cloning. Methods Enzymol., 101, 20-79.Meyer, K.H. and Mark, H. (1928). The structure of the crystalline components ofcellulose. Ber. Deutsch. Chem. Ges. 61B, 593.Meyer, K.H. and Misch, L. (1937). The constitution of the crystalline part of cellulose.VI. The positions of the atoms in the new spacial model of cellulose. Helv. Chim. Acta.20, 232.Miller, H. (1987). Practical aspects of preparing phage and plasmid DNA: growth,maintenance and storage of bacteria and bacteriophage. Methods Enzymol., 152, 145-170.Miller, J.H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor laboratoryPress, Cold Spring Harbor.129Miliward-Sadler, S.J., Poole, D.M., Henrissat, B., Hazelwood, G.P., Clarke, J.H. andGilbert, H.J. (1994). Evidence for a general role for high affinity non-catalytic cellulosebinding domains in microbial plant cell wall hydrolases. Mol. Microbiol., 11, 375-382.Mittendorf, V. and Thomson, J.A. (1994). Cloning of an endo-(l,4)-8-glucanase gene,celA from rumen bacterium Clostridium sp. ‘Iongisporurn’ and characterisation of itsproduct, Ce1A, in Escherichia co/i. J. Gen. Microbiol. 139, 3233-3242.Moser, B., Gilkes, N.R., Kilburn, D.G., Warren, R.A.J. and Miller, R.C., Jr. (1989).Purification and characterisation of endoglucanase C of Cel/u/ornonasfimi, cloning ofthe gene and analysis of in vivo transcripts of the gene. Appl. Environ. Microbiol., 55,2480-2487.Nakamura, A., Uozumi, T. and Beppu, T. (1987). Nucleotide sequence of a cellulasegene of Bacillus subtilis. Eur. J. Biochem., 164, 3 17-320.Nevell, T.P and Zeonian, S.H. (1985). Cellulose chemistry fundamentals. In: Cellulosechemistry and its applications (Nevell, T.P. and Zeonian, S.H. Eds.), pg 15. John Wileyand Sons, New York.Nordquist, D.A. (1992). Characterisation of double binding domain derivatives of CenAfrom Cellulomonasfimi. MSc. thesis, University of British Columbia, Vancouver.Nossal, N.G. and Heppel, L.A. (1966). The release of enzymes by osmotic shock fromEscherichia co/i in exponential phase. The J. Biol. Chem., 241, 3055-3062.O’Callaghan, C., Morris, A., Kirby, S.M., and Shingler, A.H. (1972). Novel method fordetection of 8-lactamase by using a chromogenic cephalosporin substrate. Antimicrob.Agents. Chemother., 1, 283-288.O’Neill, G.P., Goh, S.H., Warren, R.A.J., Kilburn, D.G. and Miller, R.C., Jr. (1986).Structure of the gene encoding the exoglucanase of Ce//ulonionasfimi. Gene, 44, 325-330.Ong, E., Gilkes, N.R., Miller, R.C., Jr., Warren, R.A.J. and Kilburn, D.G. (1991).Enzyme immobilisation using a cellulose-binding domain: properties of a B-glucosidasefusion protein. Enzyme Microb. Technol., 13, 59-65.Ong, E., Gilkes, N.R., Miller, R.C., Jr., Warren, R.A.J. and Kilbum, D.G. (1993). Thecellulose binding domain (CBDex) of an exoglucanase from Celluinonasfimi:production in Escherichia co/i and characterisation of the polypeptide. Biotechnol.Bioeng., 42, 40 1-409.Ooshima, H., Sakata, M. and Harano, Y. (1983). Adsorption of cellulase fromTrichoderma viride on cellulose. Biotechnol. Bioeng., 25, 3103-3114.Ott, E., Spurlin, H.M. and Grafflin, M.W. (1954). Cellulose and cellulose derivatives.Second edition. Part II. Interscience, New York.Owalabi, J.B., Béguin, P., Kilbum, D.G., Miller, R.C., Jr. and Warren, R.A.J. (1988).Expression in Escherichia coli of the Ce//ulornonasfIrni structural gene forendoglucanase B. Appl. Environ. Microbiol., 54, 5 18-523.130Pentillä, M., Lehtovaara, P., Nevalainen, H., Bhikhabhai, R. and Knowles, 3. (1986).Homology between cellulase genes of Trichoderma reesei: complete nucleotidesequence of the endoglucanase I gene. Gene, 45, 253-263.Pilz, I., Schwarz, E., Kilburn, D.G., Miller, R.C., Jr., Warren, R.A.J., and Gilkes, N.R.(1990). The tertiary structure of a bacterial cellulase determined by small-angle X-ray-scattering analysis. Biochem. J., 271, 277-280.Poole, D.B, Morag, E., Lamed, R., Bayer, E.A., Hazeiwood, G.P. and Gilbert, H.J.(1992). Identification of the cellulose binding domain of the cellulosome subunitSlfrom Clostridium therniocellum YS. FEMS Microbiol. Lett., 99, 181-186Poole, D.M. Hazelwood, G.P., Huskisson N.S., Virden, R. and Gilbert, H.J. (1993). Therole of conserved tryptophan residues in the interaction of a bacterial cellulose bindingdomain with its ligand. FEMS Micobiol. Lett., 106, 77-84.Poulsen, O.M. and Peterson, L.W. (1992). Degradation of microcrystalline cellulose:synergism between different endoglucanases of Cellulomonas sp. ATCC 21399.Biotechnol. Bioeng., 39, 121-123.Py, B., Bortoli-German, I., Haiech, J., Chippaux, M. and Barra, F. (1991). CellulaseEGZ of Erwinia chrysanthemi : structural organisation and importance of His 98 andGlu 133 residues for catalysis. Protein Eng., 4, 325-333.Quillet, L., Barray, S., Labedan, B., Petit, F. and Guespin-Michel, 3. (1994). In Press.Quiocho, F.A. and Vyas, N.K. (1984). Novel steriospecificity of the L-arabinosebinding protein. Nature, 310, 38 1-386.Quiocho, F.A. (1986). Carbohydrate-binding proteins: tertiary structures and protein-sugar interactions. Ann. Rev. Biochem., 55, 287-3 15.Quiocho, F.A. (1988). Molecular features and basic understanding of protein-carbohydrate interactions. The arabinose-binding protein-sugar complex. Cuff. Top.Microbiol. Immunol., 39, 135-149.Quiocho, F.A., Wilson, D.K. and Vyas, N.K. (1989). Substrate specificity and affinityof a protein modulated by bound water molecules. Nature, 340, 404-407.Raguz, S., Yague, E., Wood, D.A. and Thurston, C.F. (1992). Isolation andcharacterisation of a cellulose-growth specific gene from Agaricus bisporus. Gene, 119,183-190.Ramalingam, R., Blume, I.E. and Ennis, H.L. (1992), The Diclyostelium discoideumspore germination-specific cellulase is organised into functional domains. J. Bacteriol.,174, 7834-7837.Rautela, G.S and King, K.W. (1968). Significance of the crystal structure of cellulose inthe production and action of cellulase. Arch. Biochem. Biophys., 123, 589-60 1.Rees, D.A. (1977). Polysaccharide Shapes (Outline Studies in Biology), pg 41.Chapman and Hall, London.131Reese, E.T., Sui, R.G.H. and Levinson, H.S. (1950). The biological degradation ofsoluble cellulose derivatives and its relationship to the mechanisms of cellulosehydrolysis. 3. Bacteriol. 59, 485-497.Reinikainen, T., Ruohonen, L., Nevanen, T., Laaksonen, L., Kraulis, P., Jones, T. A.,Knowles, 3. and Teeri, T. (1992). Investigation of the functions of mutated cellulosebinding domain of Trichoderma reesei cellobiohydrolase I. Proteins, 14, 475-482.Richmond, P.A. (1991). Occurrence and functions of native cellulose. In: Biosynthesisand Biodegradation of Cellulose (Haigler, C.H. and Weirner, Eds), pg 5. M. Dekker,New York.Robson, L.M. and Chambliss, G.H. (1986). Cloning of the Bacillus subtlis DLG 13-1,4-glucanase gene and its expression in Escherichia coli and Bacillus subtlis. J. Bacteriol.,165, 612-619.Rouvinen, J., Bergfors, T., Teen, T., Knowles, J. and Jones, T.A. (1990). Threedimensional structure of cellobiohydrolase II from Trichoderina reesei. Science, 249,380-386.Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989). Molecular Cloning. A LaboratoryManual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor.Sarko, A. (1986). Recent X-ray crystallographic studies of celluloses. In: Cellulosestructure, modification and hydrolysis (Young, R.A. and Powell, R.M. Eds), pg 29.John Wiley and Sons, New York.Saul, D.J., Williams, L.C., Love, D.R., Chamley, L.W. and Bergquist, P.L. (1989).Nucleotide sequence of a gene from Caldocellum saccharolyticum encoding forexoglucanase and endoglucanase activity. Nucleic Acid Research, 17, 439-441.Scopes, R.K. (1974). Measurement of protein by spectrometry at 205nm. Anal.Biochem., 59, 277-282.Shareck, F., Roy, C., Yaguchi, M., Morosoli, R. and Kluepfel, D. (1991). Sequence ofthe three genes specifying xylanases in Streptornyces lividans. Gene, 107, 75-82.Shen, H., Schmuck, M., Pilz, I., Gilkes, N.R., Kilburn, D.G., Miller, R.C., Jr., andWarren, R.A.J. (1991). Deletion of the linker connecting the catalytic and cellulose-binding domains of endoglucanase A (CenA) of Cellulomonasfimi alters itsconformation and catalytic activity. J. Biol. Chem., 266, 11335-11340.Shen, H., Tomme, P., Gilkes, N.R., Kilburn, D.G., Warren, R.A.J. and Miller, R.C., Jr.(1994). Steriochemical course of hydrolysis catalysed by CellulomonasfImi, Cen E, amember of a new family of 13-l,4-glucanases. Biochem. Biophys. Res. Commun., 199,1223-1228.Schlochtermeier, A., Walter, S., Schröder, J., Moorman, M. and Schrempf, H. (1992).The gene encoding the cellulase (Avicelase) Cell from Streptomyces reticuli andanalysis of protein domains. Mol. Microbiol., 6, 3611-3621.Shoemaker, S., Schweickaert, V., Ladner, M., Gelfand, D., Kwok, S., Myambo, K. andInnis, M. (1983). Molecular cloning of exo-cellobiohydrolase I derived formTrichoderma reesei strain L27. Bio/Technology, 1, 69 1-696.132Shoseyov, M., Tagaki, M., Goldstein, M.A. and Doi, R.H. (1992). Primary sequenceanalysis of Clostridium cellulovorans cellulose binding protein A. Proc. Nat. Acad. Sci.USA, 89, 30483-30487.Sims, P.F.G., James, C. and Broda, P. (1988). The identification, molecular cloning andcharacterisation of a gene from Phanerochaete chiysosporium that shows stronghomology to the exocellobiohydrolase gene from Trichodern2a reesei. Gene, 74, 411-422.Speizo, M., Wilson, D.B. and Karplus, P.A. (1993). Crystal structure of the catalyticdomain of a thermophilic endocellulase. Biochemistry, 32, 9906-99 16.Sprey, B. and Bochem, H.P. (1993). Formation of cross-fractures in cellulosemicrofibril structure by an endoglucanase-cellobiohydrolase complex from Trichodermareesei. FEMS Microbiology Letters, 106, 239-244.Spurlino, J.C., Lu, G.Y. and Quiocho, F.A. (1991). The 2.3-A resolution structure of themaltose-or maltodextrin-binding protein. A primary receptor of bacterial activetransport and chemotaxis. J. Biol. Chem., 266, 5202-52 19.Spurlino, J.C., Rodseth, L.E. and Quiocho, F.A. (1992). Atomic interactions in protein-carbohydrate complexes: tryptophan residues in the periplasmic maltodextrin receptorfor active transport and chemotaxis. J. Mol Biol., 266, 15-22.Steiner, W., Sattler, W. and Esterbauer, H. (1988). Adsorption of Trichoderma reeseicellulase on cellulose: experimental data and their analysis by different equations.Biotechnol. Bioeng., 32, 853-865.Stuart, J.Y. and Ristroph, D.L. (1985). Analysis of cellulose-cellulase adsorption data: afundamental approach. Biotechnol. Bioeng., 27, 1056-1059.Svensson, B., Jespersen, H., Sierks, M.R. and MacGregor, E.A. (1989). Sequencehomology between putative raw starch binding domains from different starch degradingenzymes. Biochem. J., 264, 309-3 11.Szabo, A.G. (1989). The fluorescence properties of aromatic amino-acids: their role inthe understanding of enzyme structure and dynamics. In: The enzyme catalysis process.(Cooper, A., Houben, J.L. and Chien, L.C. Eds). Plenum Publishing Corporation.Teen, T., Lehtovaara, P., Kauppinen, S., Salovuori, I. and Knowles, J.K.C. (1987).Homologous domains in Trichoderma reesei cellulolytic enzymes: gene sequence andexpression of cellobiohydrolase II. Gene, 51, 43-52.Teeti, T., Penttilä, M., Keränen, S., Nevalainen, H., Knowles, J. (1991). Structure,function and genetics of fungal cellulases. In: Biotechnology of filamentous fungi (Ball,C. and Finkelstein, D. Eds), pg 419. Butterworth, London.Théberge, M., Lacaze, P., Shareck, F., Morosoli, R. and Kluepfel, D. (1992).Purification and characterisation of an endoglucanase from Strep tomyces lividans 66 andDNA sequence of the gene. Appl. Environ. Microbiol., 58, 8 15-820.Tomme, P., Van Tilbeurgh, H., Pettersson, G., Van Damme, J., Vandekerckhove, J.,Knowles, J., Teen, T. and Claeyssens, M. (1988). Studies on the cellulolytic system of133Trichoderma reesei QM9414. Analysis of domain function in two cellobiohydrolasesby limited proteolysis. Eur. J. Biochem. 170, 575-58 1.Towbin, H., Staehelin, T. and Gordan, J. (1979). Electrophoretic transfer of proteinsfrom polyacrylamide gels to nitrocellulose sheets: procedure and some applications.Proc. Nati. Acad. Sci. USA, 76, 4350-4354.Turková, J. (1978). Affinity chromatography, pg 153. Elsevier Scientific PublishingCo., Amsterdam, Holland.Van Tilbeurgh, H., Tomme, P., Claeyssens, M., Bhikhabhai, R. and Pettersson, G.(1986). Limited proteolysis of the cellobiohydrolase I from Trichoderma reesei,separation of functional domains. FEBS Lett., 204, 223-227.Vanden Wymelenberg, A., Covert, S.F. and Cullen, D. (1993). Identification of thegene encoding the major cellobiohydrolase of the white rot fungus Phanerochaetechiysosporium. Appi. Environ. Microbiol., 59, 3492-3494.Vieira, J. and Messing, J. (1987). Production of single-stranded plasmid DNA.Methods Enzymol., 160, 3-11.Vyas, N.K., Vyas, M.N. and Quiocho, RA. (1988). Sugar and signal-transducer bindingsites of the Escherichia coli galactose chernoreceptor protein. Science, 242, 1290-1295.Vyas, N.K. (1991). Atomic features of protein-carbohydrate interactions. Cuff. OpinionStruct. Biol., 1, 732-740.Wakarchuk, W.W., Campbell, R.L., Sung, W.L., Davoodi, J. and Yaguchi, M. (1994).Mutational and crystallographic analyses of the active site residues of Bacillus circulansxylanase. Protein Science, 3, 467-475.Walker, L.P., Wilson, D.B., Irwin, D.C., McQuire, C. and Price, M. (1992).Fragmentation of cellulose by the major Thermomonosporafusca cellulases,Trichoderma reesei CBHI, and their mixtures. Biotechnol. Bioeng., 40, 1019-1026.Warren, R.A.J., Beck, C.F., Gilkes, N.R., Kilburn, D.G., Langsford, M.L., Miller, R.C.,Jr., O’Neill, G.P., Scheufens, M and Wong, W.K.R. (1986). Sequence conservation andregion shuffling in an endoglucanase and an exoglucanase from Cellulomonasfimi.Protein Struct. Funct. Genet., 1, 335-34 1.Watanabe, T., Oyanagi, W., Suzuki, K. and Tanaka, H. (1990). Chitinase system ofBacillus circulans WL-12 and importance of chitinase Al in chitin degradation. J.Bacteriol., 173, 4017-4022.Wey, T.T., Hseu, T.H. and Huang, L. (1994). Molecular cloning and sequence analysisof the cellobiohydrolase I gene from Trichodernia koningii. Current Microbiol., 28, 31-39.Whister, R.L. and BeMiller, J.N. (1973). Industrial gums, polysaccharides and theirderivatives. Second edition. Academic Press, New York and London.White, A.R. and Brown, R.M., Jr. (1981). Enzymatic hydrolysis of cellulose: visualcharacterisation of the process. Proc. Natl. Acad. Sci. U.S.A., 78, 1047-1051.134White, C.A. and Kennedy, J.F. (1986). Carbohydrate Analysis: A Practical Approach,p38. Chaplin, M.R. and Kennedy, J.F.(Eds.). Marcel Dekker, New York.Whittle, D.J., Kilburn, D.G., Warren, R.A.J. and Miller, R.C., Jr. (1982). Molecularcloning of a Cellulomonasfimi cellulase gene in Escherichia coli. Gene, 17, 139-145.Wilson, D.B. (1992). Biochemistry and genetics of actinomycete cellulases. Crit. Rev.Biotechnol., 12, 45-63.Wong, W.K.R., Gerhard, B., Guo, Z.M., Kilburn, D.G., Warren, R.A.J. and Miller, R.C.,Jr. (1986). Characterisation and structure of an endoglucanase gene cenA ofCellulomonasfImi. Gene, 44, 315-324.Wood, T.M. and McCrae, S.I. (1978). The cellulase of Trichoderma koningii.Purification and properties of some endoglucanase components with special reference totheir action on cellulose when acting alone or in synergism with the cellobiohydrolase.Biochem. J., 171, 61-72.Wood, T.M. and McCrae, S.I. (1986). The cellulase of Penicillium pinophilum.Biochem. 3., 234, 93-99.Wood, T.M. (1988). Preparation of crystalline, amorphous and dyed cellulosesubstrates. Methods in Enzymology, 160, 19-25.Wood, T.M.(1989). Mechanisms of cellulose degradation by enzymes from aerobic andanaerobic fungi. In: Enzyme systems for lignocellulose degradation (Coughlan, M.P.Ed), pg 12. Elsevier Applied Science, New York.Wood, T.M.(1991). Fungal cellulases. In: Biosynthesis and biodegradation of cellulose(Haigler, C.H. and Weimer, P.S. Eds), pg 491. Marcel Dekker, New York.Woodward, J., Lima, M. and Lee, N.E. (1988). The role of cellulase concentration indetermining the degree of synergism in the hydrolysis of microcrystalline cellulose.Biochem. J., 255, 895-899.Worthington, C.C., (1988). Worthington Enzyme Manual. Enzymes and relatedbiochemicals. Worthington Biochemical Corporation. Freehold, New Jersey.Wüthrich, K. (1976). NMR in biological research: peptides and proteins. NorthHolland Publishing C., Amsterdam.Wüthrich, K. (1986). NMR of proteins and nucleic acids. John Wiley and Sons, Inc.,New York.Yablonski, M.D., Bartley, T., Elliston, K.O., Kahrs, S.K., Shalita, Z.P. and Eveleigh,D.E. (1988). Characterisation and cloning of the cellulase complex of Microbisporabispora. In: Biochemistry and genetics of cellulose degradation (Aubert, J.P., Béguin, Pand Millet, J. Eds), pg 249. FEMS Symposium 43. Academic Press, London.Yang, J.T., Wu, C.S,C, and Martinez, H.M. (1986). Calculation of protein conformationfrom circular dichroism. Methods in Enzymol, 130, 208-269.135Yanisch-Perron, C., Vieira, J. and Messing, J. (1985). Improved M13 phage cloningvectors and host strains: nucleotide sequences of the Ml3mp and pUC vectors. Gene,33, 103-119.Young, R.A. (1986). Structure, swelling and bonding of cellulose fibres. In: Cellulose:structure, modification and hydrolysis (Young, R.A. and Rowell, R.M. Eds), pg 91.John Wiley and Sons, New York.Zhou, C., Abaigar, L. and Jong, A.Y. (1990). A protocol for using T7 DNA polymerasein oligonucleotide site-directed mutagenesis. BioTechniques, 8, 503-506.Zhou, C., S.H. Huang and A.Y. Jong. (1989). Molecular cloning of Saccharomycescerevisiae CDC6 gene. J. Biol. Chem., 265, 9022-9029.Zoller, M.J. and Smith, M. (1983). Oligonucleotide-directed mutagenesis of DNAfragments cloned into M13 vectors. Methods Enzyrnol., 100, 468-500.


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