Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Endoglucanase A from Cellulomonas fimi : determination of the amino acids directly involved in catalysis… Damude, Howard Glenn 1995

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

Item Metadata


831-ubc_1995-982764.pdf [ 2.75MB ]
JSON: 831-1.0088836.json
JSON-LD: 831-1.0088836-ld.json
RDF/XML (Pretty): 831-1.0088836-rdf.xml
RDF/JSON: 831-1.0088836-rdf.json
Turtle: 831-1.0088836-turtle.txt
N-Triples: 831-1.0088836-rdf-ntriples.txt
Original Record: 831-1.0088836-source.json
Full Text

Full Text

ENDOGLUCANASE A FROM CELLULOMONAS FIMI:DETERMINATION OF THE AMINO ACIDS DIRECTLY INVOLVED INCATALYSIS AND THEIR FUNCTIONbyHOWARD GLENN DAMUDEB.Sc., University of Waterloo, 1990A THESIS SUBMITUED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Department of Microbiology and Immunology)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRiTISH COLUMBIAApril 1995© Howard Glenn Damude, 1995In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)___________________________Department of /7) C 1’.’ i) i’O Io5 ,The University of British ColumbiaVancouver, CanadaDate 11p r / 199 SDE-6 (2/88)11AbstractThe overall objective of this study was to characterize the catalytic domain ofendoglucanase A (CenA) from Cellulomonasfimi. More specifically, the objective was toidentify the amino acids in the catalytic domain which are directly involved in acid and basecatalysis and/or in stabilizing the transition state.The catalytic domains of 13-1,4-glucanases can be grouped into families of related aminoacid sequences. CenA is a member of family 6. All enzymes from this family are believed tohydrolyze 13-1,4-glucosidic bonds using a general acid-base catalytic mechanism resulting ininversion of anomeric configuration at the scissile bond. Three-dimensional structures for twocellulases from family 6 have been determined by x-ray crystallographic analysis. Thesestructures show that there are four aspartate residues which are in a position to function as acidcatalyst, base catalyst and/or transition state stabilizers. These aspartates are conserved in allmembers of family 6. The roles of D216, D252, D287 and D392, the corresponding amino acidresidues in CenA, were determined. These aspartates have been systematically replaced withalanine and glutamate via site-directed mutagenesis and the resulting effects on activity, pHprofile, substrate specificity and overall structure have been determined.Changes in overall structure were monitored using circular dichroism spectroscopy andno significant differences between the wild-type and mutant proteins were found. Active sitestructure was also found to be intact as all proteins bound to a cellobiose affinity column.The kinetic parameters of the enzymes were determined on various substrates withdifferent leaving groups. Further kinetic analysis of mutants using small nucleophilic anions andusing a- and f3-cellobiosyl fluoride, as well as pH dependence studies, were also carried out.On the basis of these results, D252 and D392 are assigned as the acid and base catalysts,respectively, in CenA. Residue D287 appears to aid D252 in acid catalysis and D216 is notabsolutely required for catalysis.111Table of ContentsAbstractiiTable of ContentsjjjList of TablesviiList of FiguresviiiList of AbbreviationsxAcknowledgementsxii1. Introduction11.1 Importance of studying the structure-function relationships of cellulose andcellulases 11.2 Cellulose and cellulases 21.3 The catalytic mechanisms of glycoside hydrolases 31.4 Primary and tertiary structure similarities in cellulases 61.4.1 Amino acid sequence similarities 61.4.2 Structural similarities 81.5 Identification of catalytic residues 121.5.1 Chemical modification of amino acids 121.5.2 Affinity labeling 131.5.3 Site-directed mutation coupled with kinetic studies 141.6 Objectives 182. Materials and Methods192.1 Chemicals, media components and enzymes 192.2 Bacterial strains, plasmids and phage 192.3 Media and growth conditions 212.4 Recombinant DNA techniques 212.4.1 Site-directed mutation without phenotypic selection 222.4.2 Site-directed mutation using PCR 23iv2.5 Screening for mutants 252.6 Detection of proteins 262.6.1 Estimation of protein yields 262.6.2 N-terminal amino acid sequences 262.7 Protein production and purification 262.7.1 Cellulose affinity chromatography 262.7.2 Desorbants for CenA adsorbed to CFl cellulose 272.7.3 Purification of CenA by ion exchange chromatography 272.7.4 Purification by nickel affinity chromatography 282.8 Determination of protein concentrations 292.9 Confirmation of overall protein structure 292.9.1 Circular dichroism spectroscopy 292.9.2 Cellobiose affinity chromatography 292.10 Enzyme kinetics 292.10.1 Determinations of activities on CM-cellulose and PAS-cellulose 292.10.2 Determination of kinetic parameters for the hydrolysis of cellotrioseand cellotetraose 302.10.3 Determination of kinetic parameters for the hydrolysis of nitrophenylcellobiosides 302.10.4 Determination of kinetic parameters for the hydrolysis of x- and Bcellobiosyl fluorides 312.10.5 Hydrolysis of 2,4-DNPC in the presence of small anions 322.10.6 pH dependence studies 322.10.7 Determination of stereospecificity of hydrolysis for mutant forms ofCenA.N 33V3. Results 343.1 Re-sequencing of casA 343.2 Development of a non-denaturing purification scheme for CenA 343.2.1 Purification by cellulose affinity chromatography 413.2.2 Desorbents for CenA adsorbed to CFl cellulose 413.2.3 Purification by ion exchange chromatography 433.2.4 Purification by nickel affinity chromatography 463.3 Generation of CenA mutants 463.4 Production and purification of mutant proteins 543.5 Structural characterization 543.5.1 Monitoring overall structure using circular dichroism 543.5.2 Monitoring active site structure by cellobiose affinity chromatography. 593.6 Catalytic properties of CenA.N arid the mutants 593.6.1 Hydrolysis of CM-cellulose and PAS -cellulose 593.6.2 Hydrolysis of cellotriose and cellotetraose by wild-type CenA.N 623.6.3 Hydrolysis of nitrophenyl cellobiosides 623.6.4 Hydrolysis of (X- and f3-cellobiosyl fluoride 673.6.5 Hydrolysis of 2,4-DNPC in the presence of small anions 693.6.6 pH dependence studies 733.6.7 Determination of stereospecificity of hydrolysis for mutant forms ofCenA.N 744. Discussion 804.1 Effects of mutation of CenA on the hydrolysis of substrates having leavinggroups with various pKa values 804.1.1 Wild-type CenA.N 804.1.2 D252A 824.1.3 D392A 834.1.4 D287A 84vi4.1.5 D216A.854.1.6 Glutamate mutants 864.2 Azide effects 874.3 Hydrolysis of a- and B-cellobiosyl fluoride 884.4 Conclusions 905. References 916. Appendix 986.1 Basic concepts of enzyme kinetics 986.2 Example calculation for the determination of and KM 1006.3 Graphical representation of kinetic data 1036.3.1 Hydrolysis of cellotriose and cellotetraose by wild-type CenA.N 1036.3.2 Hydrolysis of nitrophenyl cellobiosides by CenA.N and mutants 1046.3.3 Hydrolysis of 13-cellobiosyl fluoride 109vi’List of TablesTable 1.1 Enzymes in family 6 of 13-i ,4-glucosidases 8Table 2.1 E. coli strains 20Table 2.2 Plasmids, phagemids and phage 20Table 2.3 Oligonucleotides used in site-directed mutation of cenA 22Table 2.4 Oligonucleotides used in site-directed mutation of cenA by PCR 25Table 2.5 Nitrophenyl cellobiosides and parameters for respective aglycones 31Table 3.1 Chemicals tested as desorbents for CenA adsorbed to CF1 cellulose 43Table 3.2 Hydrolysis of CM-cellulose and PAS-cellulose by CenA.N and mutants 60Table 3.3 Hydrolysis of cellotriose and cellotetraose by wild-type CenA 62Table 3.4 Hydrolysis of 2,4-DNPC by CenA.N and mutants 65Table 3.5 Hydrolysis of various nitrophenyl cellobiosides by CenA.N 67Table 3.6 Hydrolysis of a- and B-cellobiosyl fluoride by CenA.N and mutants 69Table 3.7 Hydrolysis of 2,4-DNPC by CenA.N and mutants in the presence of azide 73Table 3.8 P.Ka values for wild-type CenA.N, D216A, D252A and D287 74Table 6.1 Hydrolysis of 2,4-DNPC by wild-type CenA.N (0.46 jiM) 100VII’List of FiguresFigure 1.1 Typical inverting mechanism for hydrolysis of 8-glucosides by B-glucosidases. ... 5Figure 1.2 Proposed mechanism for the hydrolysis of a-cellobiosyl fluoride by CenA 7Figure 1.3 Structural topology comparison of Cbhll and E2 10Figure 1.4 Active site structure of E2 with cellobiose bound 11Figure 2.1 PCR mutagenesis of D392 24Figure 3.1 Corrected sequence of the casA gene 35Figure 3.2 Alignment of family 6 catalytic domains including corrected CasA aasequence 38Figure 3.3 Estimation of CenA concentration in culture supematant 42Figure 3.4 FPLC profile and SDS-PAGE of fractions from supematants containing CenApassed through an anion exchange column 44Figure 3.5 FPLC profile and SDS-PAGE of fractions from supematants containing CenApassed through an cation exchange column 45Figure 3.6 Construction of pTUgKRG-1.5cenA.N 47Figure 3.7 Purification of CenA.N by nickel affinity chromatography and by celluloseaffinity chromatography 48Figure 3.8 Construction of pTZ18R-1.6cenA 50Figure 3.9 Generation of mutants of cenA 51Figure 3.10 Relative CM-cellulase activities for mutant forms of CenA.N compared to thewild-type 53Figure 3.11 Purification of wild-type CenA.N and the mutant proteins 55Figure 3.12 CD spectra of CenA.N and mutants 57Figure 3.13 General mechanism for the hydrolysis of 8-1,4-glucosidic bonds 61Figure 3.14 Cleavages of various cello-oligosacharides by CenA 63Figure 3.15 General mechanism for the hydrolysis of a nitrophenyl cellobioside 64Figure 3.16 Structures of 2,4-DNPC, 2,5-DNPC, 3,4-DNPC and pNPC 66Figure 3.17 Hammett plot for hydrolysis of various nitrophenyl cellobiosides by CenA.N .... 68ixFigure 3.18 Plot of rate vs S0 for the hydrolysis of x-cellobiosyl fluoride by CenA.N 70Figure 3.19 Double reciprocal plots for S0 and S02 71Figure 3.20 Proposed mechanism for the hydrolysis of 2,4-DNPC in the presence of azidebyCenA 72Figure 3.21 Plot of kt vs pH for hydrolysis of 2,4-DNPC by CenA.N 75Figure 3.22 Plot of kcatlKM vs pH for the hydrolysis of 2,4-DNPC by CenA.N 76Figure 3.23 Stereospecificity of hydrolysis for mutant forms of CenA.N 78Figure 6.1 Determination of kcat and KM values 101Figure 6.2 Hydrolysis of cellotetraose by wild-type CenA.N 103Figure 6.3 Hydrolysis of cellotriose by wild-type CenA.N 103Figure 6.4 Hydrolysis of 2,4-DNPC by wild-type CenA.N 104Figure 6.5 Hydrolysis of 2,4-DNPC by D216A 104Figure 6.6 Hydrolysis of 2,4-DNPC by D216E 105Figure 6.7 Hydrolysis of 2,4-DNPC by D252A 105Figure 6.8 Hydrolysis of 2,4-DNPC by D252E 106Figure 6.9 Hydrolysis of 2,4-DNPC by D287A 106Figure 6.10 Hydrolysis of 2,4-DNPC by D287E 107Figure 6.11 Hydrolysis of 2,5-DNPC by wild-type CenA.N 107Figure 6.12 Hydrolysis of 3,4-DNPC by wild-type CenA.N 108Figure 6.13 Hydrolysis of pNPC by wild-type CenA.N 108Figure 6.14 Hydrolysis of B-cellobiosyl fluoride by wild-type CenA.N 109xList of Abbreviations2,4-DNPC 2’,4’-dinitrophenyl B-D-cellobioside2,5-DNPC 2’,5’-dinitrophenyl 13-D-cellobioside3,4-DNPC 3’,4’-dinitrophenyl B-D cellobiosideaa amino acidAmp ampicillinbp base pairCD circular dichroismCHAPS 1% 3-[(3-cholamidopropyl)-dimethylammonio]-1-propane-sulfonateCM-cellulose carboxymethylcelluloseDMSO dimethyl sulfoxideHBAH p-hydroxybenzoic acid hydrazideHPLC high performance liquid chromatographyIPTG isopropyl-13-D-thiogalactosidekb kilobase pairskcat enzyme turnover numberkDa kilodaltonsKM Michaelis constantKan kanamycinLB Luria-Bertani mediumMr relative molecular massPAS-cellulose phosphoric acid-swollen cellulosePCR polymerase chain reactionpNPC 4’-nitrophenyl B-D-cellobiosideSDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electrophoresisS0 initial substrate concentrationTB terrific brothTE ths-EDTATLC thin-layer chromatographyTYP tryptone, yeast extract, phosphatexiXIIAcknowledgementsI would first like to sincerely thank my supervisor, Tony Warren, for his encouragement,enthusiasm and input, while still allowing me total autonomy over my project. If it were not forhim, I would not have chosen to come to Vancouver to pursue my PhD and would probably haveended up as a lawyer! Special thanks go to Steve Withers who contributed greatly to this thesiswith his advice and interesting ideas. I am also grateful to him for including me as part of hisown lab both scientifically and socially. Thanks also to Bob Miller and Doug Kilbum for theirinput into this project and for their “real life” advice.I am especially grateful to the many members of the cellulase group, past and present,who have contributed greatly to this thesis through their knowledge, advice and friendship.Many thanks to my comrades from room 12, Don “Who the hell are you” Trimbur, CeliaRamirez, Andreas Meinke and Roger “Pool Ball” Graham, who taught me just about everything Iknow about DNA. Thanks also to Neena Din, Dave Nordquist, Jeff Greenwood, AndrewWierzba, Greg “3AM saviour” Doheny, Shen Hua, Al Macleod, Neil Gilkes, Emily Kwan, Sr.,Emily Akow, Jr., Mark Bray and Peter “Go back to your woods” Tomme. A very special thanksgoes to the veterans Ed “Golf Pro” Ong and “Mr. Vegas” Johnny Coutinho for all the interestingconversations and worldly advice. Many warm thanks to Pat “Thumbs up” Miller and Helen“Ski Bum” Smith for all their special “behind the scenes work”.Thanks to Dee Tull for the nitrophenyl cellobiosides, Lloyd Mackenzie for paminobenzyl 13-D-thiocellobioside and Vito Ferro for a- and 13-cellobiosyl fluoride. I am alsograteful to Mike Spezio for the crystal structure co-ordinates of E2. The following peopledeserve many thanks for their technical help and advice; Gary “Pizza and Suds” Lesnicki, DeanHildebrand, Curtis Braun, Trevor Andrews, Karen Rupitz, Thisbe Lindhorst, Lando Robillo,Primrose Gontier, Carrie Hirsch, Rosario Bauzon and Linda Dodds.Although too numerous to mention, I am very thankful to all the friends who helpedmake my time in Vancouver interesting and fun. I will miss you all!Last but not least, I am especially grateful to my family for their love, encouragement andsupport throughout these many years.1I. Introduction.1.1 Importance of studying the structure-function relationships of cellulose and cellulases.Contrary to the popular belief held by many of my non-scientific friends and relaves,cellulose is not fat on the human body. That is cellulite. Cellulose is the most abundantcarbohydrate on the earth. It is one of the major components of wood and other forms of plantbiomass. Cotton, paper and the indigestible fibre found in the foods we eat are all differentforms of cellulose. As well, many of the processed foods we buy contain different forms ofcellulose as fillers. This being known, the incredible importance of cellulose in nature, industryand our daily lives becomes obvious.The amount of cellulose produced on the earth annually has been estimated to be about 4x 109 tonnes (Coughian, 1990). This material does not accumulate because manymicroorganisms in nature possess the ability to degrade it, thus providing themselves with anenergy and carbon source while helping to cycle carbon in the environment. The enzymesproduced by microorganisms to degrade cellulose are called cellulases. In addition to their rolein carbon re-cycling in the environment, cellulases are also involved in plant/parasiteinteractions, plant defense mechanisms and plant cell growth and differentiation (Tomme et a!.,1995). These enzymes are not only important in nature, but also have great potential forbiotechnological applications in industry. Although not economically feasible now, cellulasesare likely to become important for producing fermentable sugars from cellulosic biomass becauseof its vast abundance (Saddler, 1993). These sugars might then be used to produce ethanolwhich could be used as a renewable energy source. Other promising applications in this areainclude the use of cellulases to non-chemically modify pulp fibre surface properties in theproduction of paper, or in the de-inking of recycled paper (Jeffries et al., 1994; Jeffries et al.,1992). Cellulases may also become important for textile treatment to reduce fuzz or enhance thesoftness, lustre and colour of fabrics made from cotton. They are already being used to softenand fade denim in the manufacture of stone-washed jeans (Lange, 1993). Cellulases have also7been used in detergents to assist in dirt removal and improve fabric appearance Tomme et at.,1995).Because of their importance in nature and their potential for applications in industry, themechanism of hydrolysis of cellulose by these enzymes has been studied extensively over theyears. The use of molecular biological techniques to engineer cellulases with altered propertieshas led to new insights into the structure and function of these enzymes. Continued studies in thisarea will undoubtedly make possible the creation of enzymes having new properties such asenhanced catalytic efficiency, altered substrate specificity, increased pH and temperature stabilityand increased activities in non-aqueous solvents.1.2 Cellulose and cellulases.Cellulose is a linear polymer of up to 10,000 glucose molecules joined by 13-1,4-glucosidicbonds (Blackwell, 1982). These polymers associate via hydrogen bonds and hydrophobicinteractions to form tightly packed, water insoluble, crystalline microfibrils. Interspersedthroughout the microfibrils are amorphous regions of a less ordered nature. Cellulose is difficultto degrade mainly because of its insolubility and inaccessibility to depolymerizing enzymes orcellulases.Cellulases are capable of hydrolyzing 13-1 ,4-glucosidic bonds and also serve to disrupt thecrystalline structure of cellulose (Beguin and Aubert, 1994). Enzymes which cleave the internal13-1,4 bonds are classed as endoglucanases (EC while those that cleave off cellobioseunits from the ends of cellulose chains are exoglucanases or cellobiohydrolases (EC enzymes act synergistically as the action of the endoglucanases on cellulose greatlyincreases the number of free non-reducing ends, which are sites for hydrolysis by theexoglucanases. The final product of cellulose degradation is generally cellobiose which can thenbe cleaved to glucose by 13-glucosidases (EC that degrade cellulose generally secrete many different cellulases whichact synergistically to hydrolyze the substrate. For instance, the gram positive soil bacteriumCellulomonasfimi produces an array of cellulases when grown on cellulose (Beguin et al., 1977).3The genes encoding four endoglucanases; CenA, CenB, CenC and CenD (Coutinho et a!., 1991;Meinke eta!., 1993a; Owolabi eta!., 1988; Whittle et at., 1982; Wong et al., 1986), twoexoglucanases, CbhA and CbhB (Meinke et at., 1994; Shen, unpublished results), and anexoglucanase/xylanase, Cex (OtNeill eta!., 1986), have been cloned in Escherichia coti andsequenced. These proteins all have the ability to bind to cellulose.The endoglucanase CenA has been well characterized. When produced in C. fimi, thisenzyme is glycosylated (gCenA) but the recombinant form (CenA) is not (Gilkes et at., 1984a;Gilkes et at., 1984b). CenA has the same activity as gCenA and both have been shown to bind tocellulose in the same manner (Langsford et at., 1984). The main function of glycosylationappears to be to protect the enzyme from proteolysis (Langsford et a!., 1987). CenA has amolecular weight of 44 kDa. As is the case for many cellulases, CenA is a modular proteincomposed of two domains, the catalytic domain and the cellulose binding domain (CBD),separated by a short linker sequence composed principally of prolyl and threonyl residues. Thebinding domain of CenA has allowed for its purification in one step by cellulose affinitychromatography. Proteolysis experiments have demonstrated that on its own, the catalyticdomain of CenA retains similar activity to that of the whole enzyme on soluble substrates andloses the ability to bind to cellulose (Gilkes et at., 1991). The CBD retains the ability to bind tocellulose but lacks any hydrolytic activity. CenA lacks the ability to efficiently hydrolyzecrystalline cellulose but is quite active on phosphoric acid swollen cellulose (PAS-cellulose) andcarboxymethyl cellulose (CM-cellulose) (Meinke et a!., 1993b).1.3 The catalytic mechanisms of glycoside hydrolases.All glycoside hydrolases (glycosidases) characterized to date catalyze hydrolysisstereoselectively. The configuration at the anomeric carbon of the scissile bond is either retaiiied(a to a, 13 to 13) or is inverted (a to 13, 13 to a) upon hydrolysis. CenA hydrolyzes the 8-1,4-glucosidic bond with inversion of anomeric configuration (Withers et a!., 1986).The accepted mechanisms for hydrolysis by glycosidases were first proposed byKoshland (Koshland, 1953) and have been described more recently by Sinnott (Sinnott, 1990).4Glycosidases acting by a retaining mechanism proceed via a double displacement mechanismwhile those that invert do so with a single displacement mechanism. Because CenA is aninverting enzyme, only this type of mechanism will be discussed in detail. A typical invertingmechanism is shown in Figure 1.1. The inverting mechanism involves protonation of theglycosidic oxygen of the scissile bond by an acidic amino acid residue (general acid catalyst). Ina concerted process, direct stereospecific attack at the anomeric carbon of the scissile bond by awater molecule also occurs. The nucleophilicity of this water molecule is greatly increasedthrough deprotonation by a basic amino acid residue (general base catalyst). The partiallypositive charge formed on the anomeric carbon in the transition state is stabilized throughresonance with the ring oxygen. This gives the transition state significant oxocarbonium ion-likecharacter (Figure 1.1) which may be further stabilized by electrostatic interactions with nearbycarboxylate side chains in addition to the general base. The product of hydrolysis has theopposite stereochemical configuration to the original substrate. In general, only glutamate andaspartate residues act directly as general acid or base catalysts in glycosidases (Sinnott, 1990;Zvelebil and Sternberg, 1988). Although these catalytic residues are essential for hydrolysis,much of the rate enhancement due to enzyme catalysis likely comes from hydrophobic andelectrostatic interactions between the substrate and amino acid residues farther removed from thescissile bond. These residues are responsible for binding the substrate and transition states andsignificantly lowering the activation energy of hydrolysis (Sinnott, 1990).Although all glycosidases hydrolyze by either a ‘retaining” or “inverting” mechanism asdescribed, some inverting glycosidases have the ability to hydrolyze certain artificial substrateshaving the “wrong” anoineric configuration. For these substrates, hydrolysis occurs with overallretention of anomeric configuration. This has been shown to occur for the hydrolysis of manyglycosyl fluorides having the wrong anomeric configuration, iiicluding B-maltosyl fluoride by 13-amylase (EC from sweet potato (Hehre eta!., 1979) and x-xylosyl fluoride by a 13-xylosidase (EC fromBacilluspumilus (Kasumi eta!., 1987). In all cases, the glycosylfluoride having the “correct” anomeric configuration was hydrolyzed with “normal” inversion ofconfiguration. Plots of rate versus initial substrate concentration (S0) for hydrolysis of glycosylfluorides of the “wrong” configuration generally curve upward which suggests substrate5\ Acid ‘. AcidHO P6-I +OHOHHO— HO6R‘ +H J{6BaseBase’O‘.. Acid ‘ AcidROHO+ OHHHORHO__________HOHO I OHOH PRODUCES LEAVEOH\ENZYME RESTOREDBase*0BaseFigure 1.1 Typical inverting mechanism for hydrolysis of B-glucosides by B-glucosidases. Rindicates a glucose residue.6activation. Also, the fact that the normally inverting mechanism shows overall retention ofconfiguration with these substrates suggests the occurrence of two successive inverting reactions.A mechanism to explain these observations has been proposed by Hehre eta!. (1979) (Figure1.2). CbhII, au exoglucanase from Trichoderma reesei, hydrolyses a-cellobiosyl fluoride(Konstantinidis eta!., 1993), but appears to do so with regular Michaelis-Menten type kinetics.Sinnott thus proposes that hydrolysis by CbhII occurs by a mechanism different than thatproposed by Hehre (Konstantinidis et al., 1993).1.4 Primary and tertiary structure similarities in cellulases.1.4.1 Amino acid sequence similarities.As with many other classes of proteins, glycosidases can be grouped into families basedon similarities in the amino acid sequences of their catalytic domains. Recently, 48 glycosidaseenzyme families have been described (Henrissat and Bairoch, 1993). The cellulase and xylanase(EC enzyme classes contain over 200 members and can be grouped into 12 differentfamilies. These families contain characteristic sequence motifs which have become important inthe assignment of new enzymes to the families. Presumably, sequence similarities reflect the factthat the enzymes within the family catalyze hydrolysis by a similar mechanism and have similaroverall tertiary structure (Claeyssens and Henrissat, 1992; Gebler eta!., 1992; Henrissat andBairoch, 1993). For all cellulases and xylanases tested, all enzymes in a family catalyzehydrolysis with the same stereochemistry, thus lending support to this hypothesis (Gebler et at.,1992; Schou eta!., 1993).CenA is a member of family 6 of cellulases and xylanases. An updated list of membersof family 6 is presented in Table 1.1 (Tomme eta!., 1995). All family 6 enzymes so far testedhydrolyze the B-1,4-glucosidic bond with inversion of anomeric configuration. An interestingfeature of family 6 is that it contains both endoglucanases and exoglucanases which showdefinite differences in amino acid sequence. The exoglucanases of family 6 have extra aminoacid insertions in the sequences (Meinke et a!., 1994; Rouvinen eta!., 1990; Spezio et at.,1993b).7OHHOZOHHO—1’--F‘S AHHOOHROOO1HOHO ‘OHHOFH‘S AHO\HO—HO-r-.HF LEAVES HO2OFFFRS F_ROLHO IHOFigure 1.2 Proposed mechanism for the hydrolysis of c-cellobiosyl fluoride by CenA. R groupscOHBHHO1Si,FH0OHHOHOFFOH0indicate a glucose unit.8Table 1.1 Enzymes in family 6 of B-i ,4-glucosidases.Organism Enzyme Activity Accession numberAgaricus bisporus Cel3a Exo L245 19Agaricus bisporus Cel3b Exo L24520Cellulomonasfimi CbhA Exo L25809Cellulomonasfimi CenA Endo M15823Cellulomonasfiavigena ORFB Endo (Al-Tawheed, 1 988)*Fusarium oxysporum B homolog Exo L29377Humicola insolens CbhII Exo (Schou et a!., 1993)*Humicola insolens EGVI Endo (Schou et at., 1993)*Microbispora bispora Ce1A Endo P264 14Micromonospora celtulolyticum McenA Endo (Lin et at., 1994)Phanerochaete chrysosporium CbhII Exo (Tempelaars et at., 1994)Streptomyces halstedii Cell Endo Z12 157Streptomyces sp.(KSM-9) CasA Endo L03218Trichoderma reesei CbhII Exo M16190/A03821/M55080Thermomonosporafusca E2 Endo M73321Thermomonosporafusca E3 Exo (Spezio eta!., 1993a)*An asterisk indicates that the full sequence has not been published or submitted to a database.The published carboxy terminal amino acid sequence of CasA does not align well withother members of this family and many amino acids believed to be catalytically important aremissing. Re-sequencing of a small portion of the sequence revealed errors in a region believed toencode part of the catalytic site (Gilkes eta!., 1991).1.4.2 Structural similarities.Currently, X-ray crystallographic analysis of about 24 cellulases and xylanases has beenreported or is in progress (Tomme et a!., 1995). Three dimensional structures have been solvedfor enzymes from families 6, 7, 9, 10, 11, 12 and 45 (Tomme eta!., 1995). Although the numberof solved structures is still relatively small, the prediction that members of a family have asimilar overall fold has been confirmed in families 6 and 11.9The X-ray crystallographic structures for two family 6 enzymes, CbhII, an exoglucanasefrom T. reesei, and E2, an endoglucanase from Thermomonosporafusca, have been solved(Rouvinen et at., 1990; Spezio et al., 1993a). In the case of CbhII, only the tx-carbon (proteinbackbone) co-ordinates are available. Both enzymes are comprised of a central cilB barrelstructure, similar to that of those phosphate isomerase. As expected, their topologies arevirtually identical (Figure 1.3). Although very similar overall, there are significant differences inthe two enzymes which is not surprising considering one is an endoglucanase while the other isan exoglucanase. The active site of CbhII is enclosed by two extended surface loops which forma tunnel about 20 A in length (Rouvinen et al., 1990). The C-terminal surface loop is absent inE2 while the N-terminal loop is present but is pulled back so that it no longer covers the activesite (Spezio et at., 1993a). This results in a cleft approximately 11 A deep and 10 A wide, ratherthan a tunnel. The tunnel-shaped active site in CbhII is thought to restrict access of longcellulose polymers, thus imparting exoglucanase activity only. In E2, this resthction is not aproblem so that the active site is fully accessible to the cellulose polymers and the enzymetherefore acts as an endoglucanase. Other surface loops present in CbhII, and corresponding toinsertions in the amino acid sequence of the exoglucanases of family 6, are also absent in E2 andpresumably the other endoglucanases.The active site of both enzymes appears to be composed of four subsites (A to D) withbond cleavage believed to occur between subsites B and C. In E2, a cellobiose molecule binds insubsites A and B. Six conserved carboxylic amino acid residues (D79, El 15, Dl 17, D156, E263,and D265) are located near the active site of E2 (Figure 1.4). The corresponding amino acids inCbhII are reported to be in similar positions with the exception of D175 (D79 in E2). This aminoacid is reported to be located in the active site of CbhII while in E2, this residue is well displacedfrom the catalytic center. Because the ioop containing D79 may move during substrate binding,the observed position in the structure does not rule this amino acid out as being catalyticallyimportant (Spezio, personal communication).10Figure 1.3 Structural topology comparison of the catalytic domains of Cbhll and E2. Thestructures for the a-carbons (peptide backbone) of CbhII (blue) and E2 (yellow) are shown.Cysteine residues forming disulfide bonds are shown in red. A) Cbhll and E2 viewed side byside. B) Overlay of Cbhll and E2.11Figure 1.4 Active site structure of E2 with cellobiose bound. The a-carbon backbone of E2 isindicated in purple. Amino acid side chains are shown in yellow. Cellobiose is shown in bluewith oxygen molecules in red. Conserved glutamates and aspartates near the active site arenumbered according to their position in E2. The numbers in brackets indicate the positions ofthe corresponding amino acids in CenA. A) Side view of E2. B) Top view of the active site ofE2.12Because the amino acid sequence of CasA differs from that of the other family 6enzymes, the group involved in solving the structure of CbhII proposed a different mechanismfor this enzyme (Rouvinen et at., 1990), while the E2 group proposed a radically differentalignment to force conservation of a putative general base (Spezio et a!., 1993a).1.5 Identification of catalytic residues.There are a number of approaches to identify and determine the importance and functionof amino acid residues in catalysis. Only when used together can they provide a fullunderstanding of the detailed mechanism of how an enzyme carries out hydrolysis. Twoimportant techniques, sequence alignment and X-ray crystallography, have already beendiscussed. Both methods can suggest important amino acids to be further studied, and X-raystructures can also suggest functions for amino acids. Other approaches commonly used toidentify catalytic residues are described below.15.1 Chemical modification ofamino acids.Chemical reagents which react specifically with certain amino acid side chains, such asthe thiol group of cysteine, the carboxylate groups of aspartate and glutamate and the imidazolering of histidine, can be used to determine the importance of these types of residues in catalysis.Amino acids located at the active site of an enzyme can also be identified by chemicalmodification in the presence and absence of substrates or substrate analogues which when boundin the active site of the enzyme, afford protection to the amino acid side chains being modified.Chemical modification has been used to identify many important amino acid residues inglycosidases. Bray and Clarke (1990) used this technique to provide evidence for the presence ofonly catalytically essential carboxy groups in the active site of xylanase A from Schizophyllumcommune, while eliminating the possible involvement of histidine, tyrosine or cysteine residuesin catalysis by the enzyme. Further modification studies also revealed that only one glutamateresidue was protected from modification by substrate and that this was a highly conservedresidue (Bray and Clarke, 1994). Other chemical modification studies with glucoamylase G213(EC from Aspergillus niger, an inverting enzyme, identified three carboxylic acidresidues in the active site (Svennson et at., 1990). Two of these residues were proposed to actdirectly in catalysis while the third was believed to play a major role in substrate binding.Modification of another inverting enzyme, Ce1D from Clostridium thermocetlum, identifiedseven possible carboxylate groups as being near the active site, but sequence comparisonssuggested that only two of these were catalytically important (Tomme et a!., 1992).Although this technique is very useful in suggesting amino acid residues located in theactive site, it often only narrows down the possibilities of catalytically active residues and doesnot provide information about the function of that amino acid in catalysis. Also, there are oftenproblems with reactivity of the chemical reagent in that essential but inaccessible groups may notbe modified by the reagent used or several different types of side chains may be modifieddepending on the specificity of the reagent and reaction conditions.Chemical modification of carboxylic acids in CenA has been attempted (Claeyssens,personal communication). Although modification did result in inactivation of the enzyme,showing the importance of carboxylates in catalysis, the substrates used failed to protect anycarboxylic acids from modification and thus the identity of catalytically active residues could notbe established.1.52 Affinity labeling.Substrates or substrate analogues having reactive groups added are often used as affinitylabels for glycosidases. The substrate portion of these affinity labels can bind in the active site ofthe enzyme, thus directing the reactive group to presumably catalytically active amino acidresidues. These groups generally react specifically with nucleophilic amino acids such asdeprotonated forms of aspartate or glutamate residues.Epoxides are one class of affinity labels that have been used successfully in identifyingcatalytically important carboxylic amino acid residues. They have been used mainly withretaining enzymes (Chen et at., 1987; Hoj et at., 1992; Hoj eta!., 1991; Macarron eta!., 1993)but have also been used with some inverting enzymes such as B-amylase from soybean (Nitta etat., 1989). Another type of affinity label, N-bromoacetyl-13-D-glucopyranosylamine, has been14used to identify a glutamate as the general acid/base catalyst of a cyanogenic 13-D-glucosidase(EC from cassava (Keresztessy eta!., 1994). This was in agreement with work carriedout on a similar B-glucosidase from Agrobacteriumfaecalis (Wang et at, unpublished data).These types of affinity labels have not been used with CenA.Although when all goes well, they are powerful tools in identifying amino acids locatedat the active site, affinity labels often react with groups distant from the active site, thus givingmisleading information. As well, while identifying active site residues, these inhibitors alone donot give information about the function of the labelled amino acid.A new class of mechanism based inhibitors (2-deoxy-2-fluoroglucosides) has beendeveloped by Withers and his group. These inhibitors are able to specifically label the catalyticnucleophile of retaining enzymes (Withers and Street, 1988; Withers and Street, 1989; Withers etat., 1988). The use of these inhibitors has provided the first direct evidence for the formation ofa covalent intermediate during hydrolysis by retaining enzymes. The presence of theelectronegative fluorine group at the C2 position destabilizes the oxocarbonium ion-liketransition state formed during hydrolysis. Although this results in a dramatic decrease in therates for both steps of hydrolysis, the reduction of rate for the glycosylation step is counteractedby the use of aglycon groups having very good leaving group abilities. In this way, the covalentglucosyl enzyme intermediate is trapped. The actual nucleophilic residues have been determinedeither using radioactively labeled inhibitors followed by proteolysis and peptide sequencing orusing mass-spectrometry to identify peptides having increased molecular weights (Tull et at.,1991; Withers et at., 1990, Miao et at, 1994). Unfortunatcly, this technique can not be used toidentify catalytic residues in inverting glycosidases as no covalent intermediates are formedduring hydrolysis (Withers, personal communication).1.5.3 Site-directed mutation coupled with kinetic studies.Another method for identifying the amino acids which are important for catalysis involveskinetic analysis of single residue mutant forms of the enzyme. In this way, a direct andquantifiable assessment of the contributions of individual amino acids toward catalysis can beobtained (Johnson and Benkovic, 1990). Catalytic residues are essential for activity and asignificant decrease in activity will result if they are critically altered. If an amino acid believedto be catalytically important is removed and there is no substantial effect on activity, then thiswould rule it out as being catalytically important. Unfortunately, in the absence of a detailedkinetic study, the reverse is not always true. A mutation which causes a protein to have loweractivity does not conclusively prove that the changed amino acid is catalytically important. Theactivity could be lower because the changed amino acid was involved in stabilization of enzymestructure or substrate binding. For this reason, structural studies of the mutant forms of anenzyme must be carried out in order to ensure that loss of activity is not due to major structuralvariations. Alone, this technique can only be used to rule out non-catalytic residues and suggestputative catalytic residues. This technique has been used to give information on catalyticallyimportant residues in many 13-glycosidases. For example, a number of mutants were made of theretaining 13-glucosidase from A.faecalis. Of the 22 different residues targeted, only five werefound to significantly decrease activity (Trimbur et a!., 1992). This work confirmed a glutamateresidue as being the nucleophile and suggested an aspartate residue as a candidate for the role ofacid-base catalyst. Affinity labelling studies and detailed kinetic analysis of mutants has sinceruled out the aspartate residue as acid/base catalyst (Keresztessy et at., 1994; Wang, unpublisheddata). Circular dichroism (CD) spectra of all the mutants were essentially indistinguishable fromthat of the native enzyme, indicating that no gross conformational changes had resulted from thesubstitutions (Withers et a!., 1992). In another retaining enzyme, a xylanase from Bacilluspumilus, carboxylic amino acid residues were mutated based on chemical modification studies(Ko et al., 1992), amino acid sequence similarities and three-dimensional structural data. Twoglutamate residues were identified as the best candidates for the essential catalytic residues of thisenzyme. Structure was confirmed by CD spectroscopy. In an inverting endoglucanase, Ce1Dfrom C. thermocellum, 12 of the carboxyl residues which are conserved throughout family 9 weremutated to alanine (Chauvaux eta!., 1992). Of these, only five were found to have less than 1%activity compared to the wild-type enzyme. Overall structural integrity was inferred from theability of the mutant forms of Ce1D to bind calcium ions in a similar way to the wild-typeenzyme. The location of the five glutamates in the crystal structure allowed two of them to beproposed as the acid and base catalyst, respectively. These studies complemented previous16chemical modification studies. Conserved aspartate and glutamate residues have also beenmutated in another inverting enzyme, 8-amylase from soybean (Totsuka and Fukazawa, 1993;Totsuka et a!., 1994). Two of the carboxylic amino acid residues are involved in the catalytic siteas their mutation completely eliminated activity. CD spectroscopy and cyclomaltohexaoseaffinity binding indicated no significant differences in structure.In order to fully analyze the effect of mutating putative catalytic residues, detailed kineticanalysis using a range of substrates and at varying pH values must be carried out. In the case ofinverting enzymes, few mutants have been subjected to detailed kinetic analysis. Exceptions arethe identical glucoamylases from A. awomori and A. niger. To date, over 17 amino acid residuesin this enzyme have been mutated and the mutants analyzed kinetically (Svensson and Sogaard,1993). Many of the mutations involved carboxylic acid residues which may function as the acidor base catalyst and/or be important in stabilizing the transition state. The residues were chosenbased on crystal structures, chemical modification and amino acid sequence alignments.Mutations, coupled with intensive kinetic analyses at various pH vlues ide1tifi d two glutamateresidues as the general acid and base catalysts (Frandsen et al., 1994; Sierks et a!., 1990; Sierksand Svensson, 1993; Sogaardet a!., 1993; Svensson and Sogaard, 1993). Other carboxylic acidresidues were implicated in the hydrogen bonding network in the active site of the enzyme.Another approach to identifying catalytic residues, used thus far in retaining enzymesonly, has been the combination of mutation and kinetic analysis in the presence and absence ofsmall anionic nucleophiles. In these studies, the putative catalytic nucleophile of B-glucosidasefrom A.faecalis (E358) and the putative acid/base catalyst of Cex from C.fimi (E233) weremutated to alanine (MacLeod et at., 1994a; Wang et at., 1994). The activity on various substrateswas significantly decreased but was restored when a small anionic nucleophile such as azide wasadded. The restored activity also resulted in the formation of products containing azide whichcould be identified by thin layer chromatography. Such kinetic studies not only allowed for thedirect identification of catalytic residues but also, in the case of the catalytic nucleophile mutant,resulted in a change from a retaining mechanism to an inverting mechanism. These types ofstudies have not been carried out with inverting enzymes.One further novel approach in identifying catalytic residues through mutational andkinetic analysis has been to use glycosyl fluorides having the “wrong” anomeric configurations assubstrates. As described earlier, these compounds are able to act as substrates by a mechanismwhich initially involves the condensation of two substrate molecules (Figure 1.2). A mutant of Bglucosidase from A. faecalis, in which the catalytic nucleophile has been changed to an alanine,can still catalyze the condensation of two B-glucosyl fluoride molecules by this mechanism(Wang et at., 1994). No subsequent hydrolysis of the product can occur as the nucleophile wouldbe required. This allows for the build up of the initial product, which can be identified. It alsoresults in a change of the stereochemistry of the enzyme from overall retaining to inverting.1718L6 Objectives.The primary objective of this study was to identify and determine the function ofcatalytically essential carboxylic amino acid residues from the catalytic domain of CenA. Theapproach taken was to locate putative catalytically important amino acid residues using sequencealignments and structural data from family 6 enzymes, then to mutate these residues and study theeffects on structure and catalysis. Specific objectives were the following:1) To re-sequence the casA gene in order to verify its homology with the other family 6enzymes.2) To develop a purification scheme for CenA which avoids the use of strong denaturants.3) To generate mutations in the catalytic domain of CenA at appropriate sites. Oncegenerated, the mutant proteins would be purified by the method developed for furthercharacterization.4) To study the effect of each mutation on overall structure.5) To study the effects of each mutation on activity using various substrates and, ifpossible, determine kinetic parameters (KM, kcat).6) To determine the pH profile of CenA and alanine mutants.7) To study the kinetics of the mutants in the presence of sodium azide and other smallanions.8) To use a.- and 13-cellobiosyl fluoride to conduct a detailed kinetic study of the mutantproteins.192. Materials and Methods2.1 Chemicals, media components and enzymes.All chemicals used were of analytical or HPLC grade and were obtained from Sigma (St.Louis, MO) or BDH (Poole, England). All media components were from Difco (Detroit, MI).Restriction endonucleases, polymerases, ligase and nucleotides were obtained from Pharmacia(Baie d’Urfé, Québec) or New England Biolabs (Beverly, MA). Sequenase Version 2.O wasfrom United States Biochemical (Oakville, ON). Avice1 PH1O1, a microcrystalline cellulose,was from FMC International (Cork, Ireland) and CF1, a fibrous cellulose, was from Sigma.Buffers and solutions used in this study were prepared as described by Sambrook et al., (1989).Cellotriose and cellotetraose were obtained from Seikagaku corporation (Tokyo, Japan). CM-cellulose and p-nitrophenyl B-D-cellobioside (pNPC) were from Sigma. PAS-cellulose wasprepared from Avice1 as described previously (Wood, 1988). Substrates 2,4’-dinitrophenyl BD-cellobioside (2,5-DNPC), 2’,5’-dinitrophenyl 13-D-cellobioside (2,4-DNPC), 3’,4-dinitrophenylB-D-cellobioside (3,4-DNPC), cx. and B-cellobiosyl fluorides and p-aminobenzyl 13-D-thiocellobioside were gifts from D. Tull, Dr. V. Feffo and L. Mackenzie from the Department ofChemistry, UBC.2.2 Bacterial strains, plasmids and phage.The E. coli strains used in this study are listed in Table 2.1. All piasrnids and phage aredescribed in Table 2.2. Bacterial stocks were maintained at -70°C in 10% DMSO. DNA wasstored in TE buffer or water at -20°C. Phage M13KO7 was stored in TYP medium at 4°C.20Table 2.1 E. coli strains.E. coli strain Genotype ReferenceJM1O1 supE thi tX(lac-proAB)(F traD36 proAB (Yanisch-Perron eta!., 1985)lacJqZAM 15)RZ1032 HfrKL16PO/45 (lysA(61-62)) dutl ungl (Kunkel eta!., 1987)thu relAl Zbd-279::TnlO supE44Table 2.2 Plasmids, phagemids and phage.Plasmid or phage Relevant characteristics Reference or sourcepUC18-1.6cenA contains 1.6 kb C.fimi DNA including cenA; Amp’ (Wong eta!., 1986)pTZ18R-1.6cenA Amp’ fi ori; expresses cenA from lac promoter as a This study!acZ fusion; used for production of single-strandedDNA and for expression of cenA and mutants;pTZ18R- as above; Nhe I site directly after sequence encoding (Nordquist, 1992)1.6cenA.N CenA leader peptide; used as intermediate forcloning into pTUgKRG-cenA.NM13mp18- contains 1.6kb C.firni DNA including cenA; used S. Hua, unpublished1.6cenA for production of single-stranded DNA resultspTUgKRG Kanr; !acI; tac promoter; cex leader peptide R. Graham and C.sequence followed by a sequence encoding a Ramirez,histidine tail and an Nhe I site; high expression vector unpublished resultspTUgKRG- as above; adds His tail at N terminus of CenA; used This study1.ScenA.N as high expression vector for cenA and mutantsM13K07 helper phage for preparation of single-stranded DNA; (Viera and Messing,Kanr 1987)212.3 Media and growth conditions.Luria-Bertani (LB) and terrific broth (TB) media were described previously (Sambrook etal., 1989). TYP medium contained the following (per litre): 16 g tryptone, 16 g yeast extract, 5 gNaC1, 2.5 gK2HPO4.The final pH of all media was adjusted to 7.0 All E. coil strainscontaining plasmids were grown in test tubes or shake flasks containing LB or TYP for 18 hoursat 37°C, 200 rpm. The medium was supplemented with ampicillin (Amp) at 100 igfmL for E.coli carrying pTZ18R based vectors or kanamycin (Kan) at 50 pg/mL for E. coil carryingpTUgKRG based vectors. For protein production, medium was also supplemented with 0.1 mMisopropyl-13-D-thiogalactoside (IPTG). Growth of E. coil was monitored spectrophotometricallyby measuring A60011m. E. coil strains containing M13 DNA were plated in LB top agar asdescribed previously (Sambrook et a!., 1989). Solid media contained 1.5% agar.2.4 Recombinant DNA techniques.Most recombinant DNA techniques were carried out as described by Sambrook et at.,(1989) unless otherwise stated. Double-stranded plasmid and M13 RF DNA were isolated fromE. coil by the small-scale alkaline lysis method (Sambrook et a!., 1989). Sequencing gradedouble-stranded DNA was isolated using a plasmid purification kit (Plasmid midi kit, Qiagen,Chatsworth, CA). Single-stranded DNA was isolated as described either by Trimbur et at.,(1992) from E. coii strains containing pTZ18R vectors or by Sambrook et at., (1989) from E. coiistrains containing M13 vectors. Restriction endonuclease digestions and ligations were carriedout according to the directions of the manufacturers in the buffers provided. DNA fragmentswere isolated from agarose gels and purified using a gel extraction kit (Qiaex gel extraction kit,Qiagen). Preparation and transformation of chemically competent E. coli strains was performedas described by Hanahan (1983). Electrocompetent E. coli strains were transformed using aGene Pulser II apparatus equipped with the Pulse Controller II unit (Biorad, Mississauga, Ont.).Electroporation was carried out as described by the manufacturers (Biorad).222.4.1 Site-directed mutation without phenorypic selection.Site-directed in vitro mutation was performed essentially as described by Kunkel et at.,(1987). E. co/i RZ1032 was used as host strain to produce single-stranded uracil-containingtemplate DNA from either pTZ18R-l.6cenA or M13mp18-l.6cenA. Single stranded pTZ18R-1.6cenA DNA was used to produce D252A and D252E. Single stranded M13mp18-1.6cenADNA was used to produce D216A, D216E, D287A, D287E and D392A. Syntheticoligodeoxyribonucleotide primers used for mutagenesis were synthesized by the UBC NucleicAcid and Protein Synthesis unit (NAPS). The oligonucleotides used in this study are shown inTable 2.3. Some of the oligonucleotide primers incorporated unique restriction sites into theDNA without altering the encoded amino acid sequence.Table 2.3 Oligonucleotides used in site-directed mutation of cenA.Name Oligonucleotide sequence (53’)* Mutation in Amino AcidcenA ChangeHDA PATCCCGGGCCGCGCCTGCGGATCCCACTCC GAC>>>>GCC D216AGGCGGTHDB PATCCCGGGCCGCGATGCGGATCçCACTCC GAC>>>>GAG D21 6EGGCGGTRD 1 PATCCTCGAGCCC(GA)(AC)(CG’)GCGCTCGCG GAC>>>>GCC D252ACAG GAC>>>>GAG D252EHDC PCGCGTCTACATCGCCGCGGGGCATGCGAAG GAC>>>>GCC D287ATGGCTGHD9 PCGCGTCTACATCG(ACGjGGCGGGCCACGCG GAC>>>GAG D287EHD1 1 PCCCGGCGAGTCCG(ACG)GGGCGCGTGCAAC GAC>>>>GCG D392A* P indicates that the oligonucleotide was 5’ phosphorylated- nucleotide changes are underlined. Silent mutations adding a restriction site are in italics.232.42 Site-directed mutation using PCR.PCR mutagenesis was used to generate D392E as outlined in Fig. 2.1. Oligonucleotideprimers used are described in Table 2.4.A PCR reaction mix was prepared by adding the following components in a 1.5 mLeppendorf tube: 15 iL of lOx Vent polymerase PCR buffer, 79.5 jiL H20, 30 .tL of dNTP mix(2 mM each of dATP, dCTP, dGTP and dTTP) and 1.5 iL of Vent polymerase. In a separate0.5 mL eppendorf tube, 1 iL each of PCR primers HD4 and HDE (20 pmol4LL) were combinedwith 1 tL of pTZ18R-1.6cenA DNA (10 ng/iL) and 5 pL of DMSO. A second 0.5 mLeppendorf tube contained 1 iL each of PCR primers HD12 and HDD (20 pmol4iL), 1 j.iL ofpTZ18R-1.6cenA DNA (10 ng4tL) and 5 jaL of DMSO. The 0.5 mL eppendorf tubes wereheated to 96°C for 1 minute (hot start) in the PCR thermocycler (TwinBlock, Ericomp, SanDiego, CA). At this time, 42 .tL of the PCR reaction mix were added to each and 2 drops ofmineral oil were placed on top of the mixtuies. The mixtures were heated to 95°C for 30 seconds,55°C for 30 seconds and 72°C for 30 seconds. This cycle was repeated a total of 20 times.Afterwards, the mixtures were heated to 72°C for 1 minute, then cooled to room temperature.PCR products were analyzed by electrophoresis through a 1.5% agarose gel.The two resulting PCR products were diluted to about 50 ng/iL, combined, and 1 iL wasadded to a 0.5 mL eppendorf tube along with 5 jaL of DMSO. The tube was heated at 96°C for 1minute in the PCR block as before. PCR primers HD4 and HD12 (flanking primers) were leftout of the hot start. After the hot start, 42 iL of the reaction mix were added; the samples werethen heated at 55°C for 30 seconds and 72°C for 54 seconds. At this time, 1 .tL of each flankingprimer (HD4 and HD12) was added and 2 drops of mineral oil were placed on top of the PCRmixture. The mixtures were heated to 95°C for 30 seconds, 55°C for 30 seconds and 72°C for 54seconds. This cycle was repeated a total of 20 times. Afterwards, PCR products were analyzed ona 1% agarose gel.24PCRTube#1 PCRTube#2HDD HD44_____HD12 HDE20 Cycles 20 Cycles95°C, 30 seconds 95°C, 30 seconds55°C, 30 seconds 55°C, 30 seconds72°C, 30 seconds 72°C, 30 seconds1) Purify, dilute andcombine fragments2) Denature (95°C) andanneal (55°C)1) Extend fragments(72°C, 54 seconds)2) Add flanking primersHD44HU 1220 Cycles95°C, 30 seconds55°C, 30 seconds72°C, 30 secondsFigure 2.1 PCR mutagenesis of D392.25Table 2.4 Oligonucleotides used in site-directed mutation of cenA by PCR.Name Oligonucleotide sequence (53!) Mutation in Amino AcidcenA ChangeHD4 GTCGACACCGTCGCGCAGGGCHDE GGGCCGCCGTFGCATGCGCCITCGGACTCGC GAC>>>>GAA D392ECGGGCAGCHDD CCCGGCGAGTCCGAGGCGCTGCAACGGC GAC>>>>GAG D392EGGCCCGHD 12 GTrGGGTAACGCCAGGGFIFrCCCAG- nucleotide changes are underlined. Silent mutations adding a restriction site are in italics.2.5 Screening for mutants.After in vitro mutation, the double stranded pTZ18R-1.6cenA DNA was transformed intoJM1O1. Trañsformants were plated on LB agar plates supplemented with Amp and 1% CM-cellulose and grown for 16-20 hours at 37°C. Transformants were screened for reduced CMcellulase activity by Congo Red staining (Gillces et a!., 1984a).Double-stranded M13mp18-1.6cenA DNA was also transformed into JM1O1 and platedon LB agar as described in section 2.3. Plaques were chosen randomly and double-stranded M13RF DNA was purified and screened by restriction site analysis if appropriate.If PCR products were of the correct size then they were purified by phenol/chloroformextraction and cut with BsiW I and Hind III. The resulting DNA fragment was purified, clonedinto pTUgKRG-1.5cenA and transformed into E. coli JM1O1. Transformants were then screenedby restriction site analysis of purified double-stranded DNA.All mutations were confirmed by DNA sequencing. Sequencing of DNA templates wascarried out as described previously Crabor and Richardson, 1987). Deaza-7-dGTP and deaza-7-dATP replaced dGTP and dATP, respectively. Modified Ti DNA polymerase (SequenaseVersion 2.0) was used at a reaction temperature of 43°C.262.6 Detection of proteins.2.6.1 Estimation ofprotein yields.E. coli strains carrying cenA and mutants of cenA were grown under the conditionsdescribed in section 2.3. After 18-24 hours, CenA had leaked into the medium (Guo et al.,1988). The protein yield was estimated by binding 1 mL of culture supematant to 10 mg ofAvicel at 4°C for 1 hour. The Avicel was centrifuged at 13 000 x g for 1 mm., washed with 1 MNaCI, then with 50 mM phosphate buffer, pH 7.0. Avicel was collected by centrifugation, thenboiled for 2 minutes in SDS loading buffer (Laemmli, 1970). The boundpolypeptides wereanalyzed by SDS-PAGE (Laemmli, 1970) with a Mini-PROTEAN apparatus (Biorad). Proteinbands were visualized by staining with Coomassie blue.An estimation of mutant protein yieldcould be made by comparison with a known quantity of wild-type CenA.2.62 N-terminal amino acid sequences.Proteins were separated by SDS-PAGE, and then transferredonto Immobilon-Pmembranes (Millipore Corp., Marlborough, MA). The transferred proteinswere visualized withCoomassie blue and the desired bands were excised from themembrane. N-terminal amino acidsequences were determined by automated Edman degradationusing an Applied Biosystems470A gas phase sequenator (Protein Microchemistry Facility, University of Victoria, Victoria,B.C.).2.7 Protein production and purification.2.7.1 Cellulose affinity chromatography.Both wild type CenA and D252A were initially purified by cellulose affinitychromatography. E. coli JM1O1 containing these pTZ18R-l.6cenA constructs was inoculatedinto 500 mL of LB medium supplemented with 100 .tg Amp/mL and incubated at 37°C, 220rpm. At an OD1yj of 1.5, IPTG was added to 0.1 mM and the culture was grownunder the sameconditions for 18-24 hours. Cells were separated from the culture supernatant by centrifugation27at 11 000 x g. CF1 cellulose (25 g) was added to the clarified supernatarit and mixed at 4°C for12 hours to allow for protein binding. CF1 cellulosewas then collected by sedimentation andpacked into a 1.6 x 20 cm FPLC column (XK16/20, Pharmacia). The column was washed with 5column volumes of 1 M NaC1 and the adsorbed proteinwas eluted with a gradient ofguanidinium HC1 (0 to 8 M) in potassium phosphate buffer (50 mM, pH 7.0). Absorbance of theeluate was measured continuously at A280nm and peakfractions were screened for protein bySDS-PAGE using a 10% gel. Fractions containing the pure protein were combined andconcentrated by ultrafiltration using a 10 kDa cutoff polysulfone membrane (PM 10, Amicon,Beverly, MA). Buffer was then exchanged with potassium phosphatebuffer (50 mM, pH 7.0) bydialysis, using the same apparatus. The concentrated proteins were centrifuged at 190 000 x gfor 30 minutes, then filtered through a 0.2 im polysulfone filter (Acrodisc®, Gelman Sciences,Ann Arbour, MI). Protein purity was assessed by SDS-PAGE. Proteins were stored at 4°C.2.7.2 Desorbents for CenA adsorbed to CFJTM cellulose.CenA (20 mg), purified as described in section 2.7.1, was bound to 20 g of CF1cellulose. Samples of 2 g of CF1 cellulose, containing 2 mg of CenA, were packed into 1 x 5cm columns (Econo-columns, Biorad) connected to a FPLC system (Pharmacia). The followingcompounds were then used in an attempt to desorb CenAfrom the cellulose: H20, 1%cellobiose, 20% ethanol, 20% glycerol, 0.1 N NaOH, 0.1N HC1, 3.2 M DMSO, 1% CM-cellulose, CHAPS and 6 M guanidinium HC1. A280nm of the eluate was monitored and positivepeak fractions were then assayed for protein using theBicinchoninic acid (BCA) protein assay(Pierce, Rockford, IL).2.7.3 Purification of CenA by ion exchange chromatography.E. coil carrying pTZ18R-1.6cenA was grown in 100 mLof LB medium as described insection 2.7.1. After growth, the cells were removed by centrifugation and the clarifiedsupematant was concentrated to 10 mL by ultrafiltrationusing a PM1O membrane. The mediumwas exchanged with potassium phosphate buffer (50 mM, pH 8.0) by dialysis. A separate 10 mL28portion of concentrated supernatant was exchanged with potassium phosphate buffer (pH 6.0).MacroPrep® Q and MacroPrep® S resins (Biorad) were prepared according to themanufacturers instructions and packed into XK16/20 FPLC columns. The MacroPrep® Qcolumn was equilibrated with 5 volumes of potassium phosphate buffer (50 mM, pH 8.0). Theconcentrated supernatant (5mL, pH 8.0) was loaded onto the column. The column was washedwith potassium phosphate buffer and bound proteins were eluted with a gradient of NaC1 (0 to 1M) in phosphate buffer (pH 8.0). The same procedure was carried out using MacroPrep® Scationic exchange resin, except that the potassium phosphate buffer and protein sample were atpH Purification by nickel affinity chromatography.E. coli JM1O1 containing the pTUgKRG-15cenA.N constructs was inoculated into 500mL of TB medium, supplemented with Kan (50 jig/mL) and IPTG (0.1 mM), and incubated at37°C; 220 rpm. In some cases, larger cultures of 10 L were grown in a fermenter (Chemap,FZ3000, Laval, Quebec),. Cultures were grown at 37°C for 18-24 h by which time CenA hadleaked into the medium (Guo et al., 1988). Cells were separated from culture medium bycentrifugation at 11 000 x g or by continuous flow centrifugation (T-1P centrifuge, Sharples,Scarborough, ON) The protein was precipitated from the clarified supernatant with 50%(NH4)2S0.Proteins were recovered by centrifugation at 11 000 x g and the pellet wasdissolved in binding buffer (5 mM imidazole, 500 mM NaC1, 20 mM Tris-HC1, pH 7.9). Afterfurther centrifugation at 190 000 x g, the clarified supernatant was loaded onto a nickel agarosecolumn (PET Nickel column, Novagen, Markham, ON). The column was washed with washbuffer (60 mM imidazole, 500 mM NaCI, 20 mM Tris-HC1, pH 7.9). Bound proteins were elutedwith a gradient of imidazole (0 to 500 mM) in Tris-HCIJNaC1 buffer (20 mM/500 mM, pH 7.5).The fractions were screened for CenA by SDS-PAGE using a 10% gel. Fractions containingCenA were combined and EDTA was added to 10 mM. The protein was concentrated byultrafiltration using a PM 10 membrane and the buffer was exchanged with potassium phosphatebuffer (50mM, pH 7) by dialysis. The concentrated protein solution was centrifuged at 190 000x g for 30 minutes and filtered through a 0.2 urn polysulfone filter. Proteins were stored at 4°C.292.8 Determination of protein concentrations.The concentration of purified CenA was determined by measuring A28011m using anextinction coefficient of 2.64 mL mg-1 cm-’ (Gilkes eta!., 1992). All other proteinconcentrations were measured using the BCA protein assay with wild-type CenA as the standard.2.9 Confirmation of overall protein structure.2.9.1 Circular dichroism spectroscopy.CD spectra were recorded with a spectropolarimeter (model J-720, JASCO, Tokyo,Japan) controlled with 1-700 software. The spectra were obtained at a protein concentration of2.2 mM in 1.25 mM potassium phosphate buffer, pH 7, 27°C using a 100 iL silicon quartz cellwith a 1 mm path length. They were recorded 3 times from 250-200 nm, at a scan rate of 20nm/mm. using a 2.0 s response and a sensitivity of 10 mdeg.2.92 Cellobiose affinity chromatography.A cellobiose affinity column was prepared by linking p-aminobenzyl B-Dthiocellobioside to Sepharose 4B (Tomme et al., 1988). Purified proteins (1 mg) were loadedonto the column in potassium phosphate buffer (50 mM, pH 7). After washing the column with50 mM phosphate buffer, bound proteins were eluted with 100 mM cellobiose in phosphatebuffer. A280nm of the eluate was measured continuously.2.10 Enzyme kinetics.2.10.] Determinations of activities on CM-cellulose and PAS-cellulose.Initial rates of hydrolysis of CM-cellulose were determined by incubating 2400 .tL of0.5% CM-cellulose in sodium citrate buffer (50 mM, pH 7.0) with an appropriate amount ofenzyme (diluted to 600 IlL in 50 mM citrate buffer, pH 7.0) at 37°C in a water bath. Aliquots30(500 iL) were removed at 0, 5, 10, 20 and 30 minute intervals and added to 1 mL of phydroxybenzoic acid hydrazide (HBAH) reagent (Miller et al., 1960) in glass test tubes. Testtubes were heated in a steam bath for 12 minutes and after cooling to room temperature, A455nmwas measured. A standard curve using glucose was also prepared in this way.Initial rates of hydrolysis of PAS-cellulose were determined by incubating 10 mg/mLPAS-cellulose in sodium citrate buffer (50 mM, pH 7.0) with an appropriate amount of enzyme(total volume 1500 jiL) at 37°C in a 2 mL eppendorf tube. Samples were mixed by rotatinginside a 37°C incubator. Aliquots (150 pL) were removed at 0, 1, 2, 3 and 4 hours andcentrifuged at 13 000 x g for 5 minutes. An aliquot of the clarified supernatant (50 pL) wasremoved and added to 1 mL of HBAH reagent for determination of reducing sugars as describedpreviously.2.10.2 Determination of kinetic parameters for the hydrolysis of cellotriose and cellotetraose.Initial rates of hydrolysis of cellotriose and cellotetraose were determined by incubatingappropriate substrate concentrations in potassium phosphate buffer (5 mM, pH 7.0) at 37°C untilthermally equilibrated. Reactions were initiated by the addition of 20 IlL of enzyme (totalvolume 1000 iL); after 30 minutes, reactions were stopped by addition of 100 I.tL ofH2S04 (1M). Quantification of cellobiose production was carried with an HPLC system (model DX500,Dionex, Sunnyvale, CA) using an anion exchange column (Carbopac PAl) and a pulsedamperometric detector (ED4O). A gradient of NaOH, sodium acetate and H20 was used asdescribed previously (Van Nifterik et a!., 1993). Data were analyzed using the Dionex Peaknetsoftware package and kinetic parameters were determined using the program GraFit 3.0 (Sigma).2.10.3 Determination of kinetic parametersfor the hydrolysis of nitrophenyl cellobiosides.The nitrophenyl cellobiosides used in this study and the pKa, extinction coefficients andmaximal absorption wavelengths of the respective aglycone units are listed in Table 2.5. Allsteady state kinetic studies for nitrophenyl cellobioside hydrolysis were performed by recordingchanges in absorbance using either a Hitachi U 2000 spectrophotometer (Tokyo, Japan), with atemperature-controlled cell holder, or a Pye-Unicam PU-8800 spectrophotometer, equipped with313.96 400 109102,5-DNPC 5.15 400 11 0093,4-DNPC 5.36 440 4 288pNPC 7.18 400 7280a Values were determined previously (Kempton and Withers, 1992).2.10.4 Determination ofkinetic parameters for hydrolysis of a— andJ3-cellobiosyl fluorides.Initial rates of hydrolysis of both a- and f3-cellobiosyl fluoride were determined byincubating appropriate substrate concentrations in potassium phosphate buffer (50 mM, pH 7.0)at 30°C until thermally equilibrated and reactions were initiated by the addition of 20 .tL ofenzyme (total volume 500 .tL). Release of fluoride was monitored using a fluoride electrodea temperature-controlled circulating water bath. Hydrolysis was monitored at appropriatewavelengths using extinction coefficients listed in Table 2.5 (Kempton and Withers, 1992).Initial rates of hydrolysis were determined by incubating solutions of the appropriate substrateconcentrations in 50 mM potassium phosphate buffer (pH 7.0, 0.1% BSA) at 37°C within thespectrophotometer until thermally equilibrated. Reactions were initiated by the addition ofenzyme, and release of phenol product was monitored at the appropriate wavelength.Approximate values of KM and keat were determined by measuring initial rates of hydrolysis at 4to 5 substrate concentrations covering a wide range. Accurate values were then measured using7 to 10 different substrate concentrations, which generally ranged from 0.2 to 5 times the KMvalue. Values for KM and kcat were determined by nonlinear regression analysis using theprogram GraFit 3.0.An example of calculation of KM and kcat values is given in the Appendix.Table 2.5 Nitrophenyl cellobiosides and parameters for respective aglyconesa.Nitrophenyl pKa Wavelength (nm) Extinction coefficient (Mcm1)cellobioside2,4-DNPC32(Huoride Ion Combo-electrode, VWR, Toronto, ON) coupled to a pH meter (VWR). Theelectrode was calibrated using standard solutions of sodium fluoride (10 mM, 1 mM and 0.1mM). Kinetic parameters were determined using the program GraFit Hydrolysis of2,4-DNPC in the presence ofsmall anions.Initial rates of hydrolysis of 2,4-DNPC (1 mM) in the presence of sodium azide, sodiumformate, sodium acetate and sodium isothiocyanate were determined in exactly the same way asdescribed in section 2.10.3. Anion concentrations ranged from 100 mM to 1000 mM. Productsof hydrolysis were analyzed by thin layer chromatography on silica gel plates using ethylacetate:methanol:H20 (7:2:1) as mobile phase.2.10.6 pH dependence studies.Kinetic parameters for hydrolysis of 2,4-DNPC were determined for wild-type CenA,D216A, D252A and D287A over the range pH 4.5 to 9.5 at increments of 0.5 pH units. Sodiumsuccinate buffer (50 mM, 0.1% BSA, 150 mM NaC1) was used from pH 4.5 to 6.0, potassiumphosphate buffer (50 mM, 0.1% BSA, 150 mM NaC1) was used from pH 6.0 to 8.0 andglycylglycine buffer (50 mM, 0.1% BSA, 150 mM NaCl) was used from pH 8.0 to 9.5. A 2xstock (111.1 mM buffer, 0.22% BSA and 333.3 mM NaC1) of each buffer was prepared and 50mL of each was added to 40 mL of H20. NaOH or HC1 was added to bring the buffer to thecorrect pH and H20 was added to a total of volume 100 mL. The actual pH of each buffer wasdetermined using a pH meter (Accumet 925 pH Ion Meter, Fischer Scientific, Ottawa, ON).Molar extinction coefficients were determined for 2,4-DNPC from pH 4.5 to 6.5 by makingdilutions of a stock solution of 2,4-DNPC in each buffer. Initial rates of hydrolysis weredetermined exactly as in section 2.10.3 by adding 900 pL of the appropriate buffer, 10 to 90 iLof substrate and 10 .tL of enzyme to a 1 mL cuvette. Rates were measured for 6 to 10 differentsubstrate concentrations, which generally ranged from 0.2 to 5 times the KM. ‘<cat KM and pKavalues were determined using GraFit Determination of stereospecificity of hydrolysis for mutantforms of CenA N.Hydrolysis of cellopentaose (5 mM) in potassium phosphate buffer (5 mM, pH 7.0) wasinitiated by the addition of appropriate amounts of CenA.N and each of the mutants. Reactionswere incubated at 22°C for 5 minutes. Production of cellotriose was monitored by HPLC on aShimadzu system with an SIL6B auto injector, a RID6A refractive index detector, and a CR501chromatopac integrator/plotter (Fisher). Cellodextrins were separated on a Dextro-Pak column(8 x 100 mm, 4 jiM particle size, Waters) equipped with a resolve C18 Guard-Pak column.Water was used as the mobile phase at a flow rate of 1 ml.min.33343. Results3.1 Re-sequencing of casA.CasA is an endo-B- 1 ,4-glucanase from Streptomyces KSM-9 belonging to family 6 of B1,4-glucanases. The residue conservation pattern of CasA was significantly different from theother members of the family. Residues believed to be catalytically important, such as anaspartate corresponding to Asp4Ol from Cbh U, were not conserved. This led to the suggestionthat CasA may function by a fundamentally different mechanism than the other members offamily 6 (Rouvinen et at., 1990). A previous analysis of a portion of the casA gene revealedsequencing errors in the region encoding part of the catalytic site (Gilkes et at., 1989).Additional errors in the original sequence were suspected, based on sequence comparison of theC terminus of CasA with other members of family 6. For this reason, the casA coding regionwas completely re-sequenced. Many errors were revealed in the original sequence (Figure 3.1).The revised CasA aa sequence was aligned with the sequences of the catalytic domains of theother members of family 6 (Figure 3.2). This new alignment shows that an aspartatecorresponding to Asp4Ol in Cbh II is indeed conserved throughout family 6. Cysteine residuesconserved in all other members, where they serve to form disulfide bonds, are also readilyapparent in the revised sequence (Damude et at., 1993).3.2 Development of a non-denaturing purification scheme for CenA.Previously, wild type CenA was purified by cellulose affinity chromatography using 8 Mguanidinium HC1 for desorption. The mechanism of elution is by complete denaturation of thecellulose-binding domain of CenA. For the wild type protein, purification involving completedenaturation of CenA was shown not to result in a loss in activity for hydrolysis of CM-cellulose.This may not hold true for mutant proteins. Loss in activity of a mutant of CenA must bedirectly attributable to the single change made and not to improper folding. In order to minimizeloss in activity due to incorrect re-folding, purification by another scheme was desirable.35Figure 3.1 Corrected sequence of the casA gene. The original (old) and the corrected (new)nucleotide sequences are given, with the corresponding deduced aa sequences above and belowthe nucleotide sequences, respectively. Numbers at the end of the line refer to positions of thelast aa in the line in the corrected aa sequence of the pro-enzyme. The dashed line (--) indicatesthe start of the pro-enzyme.36MEN PR T T PT PT P L R R R R SE Rold GTGGAGAACCCCAGAACCACGCCCACCCCCACACCTCTCCGCCGGCGGCGGTCGGAGCGCnew GTGGAGAACCCCAGAACCACGCCCACCCCCACACCTCTCCGCCGGCGGCGGTCGGAGCGCMEN PR T T PT PT P L R R R R SE RR AR G G R V L TA L T G VT L LAG Lold CGTGCGCGCGGCGGACGCGTGCTCACCGCGCTCACCGGCGTGACGCTGCTCGCCGGTCTGnew CGTGCGCGCGGCGGACGCGTGCTCACCGCGCTCACCGGCGTGACGCTGCTCGCCGGTCTGR AR G G R V L TA L T G VT L LAG LA IA PA AT GASPS PAPPAS PAold GCGATCGCGCCGGCCGCGACCGGCGCGTCCCCCTCGCCCGCCCCGCCCGCATCGCCCGCCnew GCGATCGCGCCGGCCGCGACCGGCGCGTCCCCCTCGCCCGCCCCGCCCGCATCGCCCGCCA IA PA AT GASPS P APP A SPAP 5 A D S G T A 0 A--G T T A L P S M E Lold CCGTCCGCGGACTCCGGCACCGCGGACGCCGGCACGACCGCCTTGCCCTCCATGGAGCTCnew CCGTCCGCGGACTCCGGCACCGCGGACGCCGGCACGACCGCCCTGCCCTCCATGGAGCTCP S A D S G T A D A--G T T A L P 5 M E L (10)YR A E AG V H AWL DAN PG D HR Aold TACCGGGCAGAAGCCGGCGTCCACGCATGGCTCGACGCGAACCCCGGCGACCACCGTGCAnew TACCGGGCAGAAGCCGGCGTCCACGCATGGCTCGACGCGAACCCCGGCGACCACCGTGCAY R A E A G V H A W L D A N P G D H R A (30)P LIVER I G SEP E A V W F A GAYold CCGCTGATCGTGGAGCGGATCCGCTCGGAACCGGAGGCGGTCTGGTTCGCCGGCGCGTACnew CCGCTGATCGCGGAGCGGATCGGCTCGCAACCGCAGGCGGTCTGGTTCGCCGGCGCGTACP L I A E R I G S Q PQ A V W F A G A Y (50)N PG TI T Q Q V A E V TSR R Q Q PPold AACCCGGGCACGATCACCCAACAGGTCGCCGAGGTCACCTCCCGGCGGCAGCAGCCGCCCnew AACCCGGGCACGATCACCCAACAGGTCGCCGAGGTCACCT--CGGCGGCAGCAGCCGCCNP G TI T Q Q VA E VT S A A A A A (69)G Q L P V V VP Y MI P F RD C G N H Sold GGCCAACTGCCCGTGGTGGTGCCCTACATGATCCCGTTCCGGGACTGCGGGAACCATTCCnew GGCCAACTGCCCGTGGTGGTGCCCTACATGATCCCGTTCCGGGACTGCGGGAACCATTCCG Q L P V V V P Y M I P F R 0 C G N H S (89)G G GAPS F A A Y A E W S G L F A AGold GGCGGCGGAGCACCGAGCTTCGCCGCGTACGCCGAGTGGAGCGGGCTCTTCGCGGCGGGGnew GGCGGCGGAGCACCGAGCTTCGCCGCGTACGCCGAGTGGAGCGGGCTCTTCGCGGCGGGGG G G A P S F A A Y A E W S G L F A A G (109)L G SEP V V V V L E PD Al P LID Cold CTCGGCTCCGAGCCGGTCGTGGTGGTGCTCGAGCCCGATGCGATTCCGCTGATCGACTGCnew CTCGGCTCCGAGCCGGTCGTGGTGGTGCTCGAGCCCGATGCGATTCCGCTGATCGACTGCL G S E P V V V V L E P D A I P L I D C (129)L D N Q Q RAE R LA A LAG LA E A Vold CTCGACAACCAGCAGCGGGCGGAACGGCTGGCCGCGCTGGCAGGCCTCGCCGAGGCCGTCnew CTCGACAACCAGCAGCGGGCGGAACGGCTGGCCGCGCTGGCAGGCCTCGCCGAGGCCGTCL 0 N Q Q R A E R L A A L A G L A E A V (149)T DAN PEAR V Y Y DV G H SAW H Aold ACCGACGCCAACCCCGAGGCCCGTGTCTACTACGACGTCGGCCACTCGGCCTGGCACGCGnew ACCGACGCCAACCCCGAGGCCCGTGTCTACTACGACGTCGGCCACTCGGCCTGGCACGCGr D A N P E A R V Y Y 0 V C H S A W H A (169)37PA A IA PT LV E AGILE H GAG Iold CCCGCCCCCATCGCACCGACGCTGGTGGACGCGGGCATCCTGGAGCACGGAGCCGGCATCnew CCGGCCGCCATCGCACCGACGCTGGTGCAGGCGGGCATCCTGGAGCACGGAGCCGGCATCP A A I A P T L V E A G I L E H G A G I (189)A TN I S N YR T T T D ETA Y A S A Vold GCCACCAACATCTCCAACTACCGGACCACCACCGACGAGACGGCCTACGCGTCCGCGGTCnew GCCACCAACATCTCCAACTACCGGACCACCACCGACGAGACGGCCTACGCGTCCGCGGTCA T N I S N Y R T T T D E T A Y A S A V (209)IA EL C G G L GA V V D TSR N C N Gold ATCGCCGAGCTGGGCGGCGGTCTCGGCGCTGTGGTCGACACCAGCCGCAACGGCAACGGCnew ATCGCCGAGCTGGGCCGCGGTCTCGGCGCTGTGGTCGACACCAGCCGCAACGGCAACGGCI A E L G G G L G A V V D T S R N G N G (229)P T A A DLV N T R Told CCG ACC-GC--GGCCGACTTGGT-G--AAC-ACCCGGACCGnew CCGCTCGGCAGCGAGTGGTGCGACCCGCCCGGCCGA-TTGGTCGGCAACAACCC-GACCGP L C S E W C D P P G R L V G N N P T (248)V T R C PG V D A F LW IT C P V Told TGA- —CCCGGTGT-CCCGGCGTGGACGCCTTCCTCTGGATCACG-TGCCCGGTGA-CT-Gnew TGAACCCCGGTGTCCCCGGAGTGGACGCCTTCCTCTGGATCAAGCTGCCCGGTGAACTCGV N P G V P G V D A F L W I K L P G E L (268)DC GD C P V F S P P K L Q Lold ACGG-TGGCCACCGCCCGG TC-TTCTCCCCC-CCTAAG-C-TAC-AGCTnew ACGGCTG-CGACGGCCCGGCCGGGTCGTTCTCCCCCGCC-AAGGCGTACGAGCTGGCCGGD G C D C P A G S F S P A K A Y E L A G (288)P R K P A AG R CC RD T IVold ACCCC-GCA---A-GCCCGCC-GCCGGCCCCCGGTGCCGGGACACG--ATCGTnew AGCCTGACACCCCCGCACCGACGGCCGCCCGCCGGCCGCGGGT-CCCGGACACCCATCGTG * (289)R S AR Q Q Q T R PP G K PG Lold CCGGTCCGCGCGTCAGCAGCAGAC-GCGAC CCCCGGGGAACCCCGGGCTCCnew CCGGTCCGCGCGTC -CCACCAGACGGCCACGGCCGGTCGCCCCGGGGAA-CCCGGGCGACPA G RD SIR HG A AG *old CGGCCGGTCGCGACAGCATCCGG-CACGGTGCGGCCCGCTGACGGCGAGCGCCCGGTCGCnew CGGCCGGTCGCGACAGCATCCGGCCACGGTGCGGCCGGCTGACGGCGAGCGCCCGGTCGCold CGGGCACCCTTCCGCTGATCACGTCGGTGAGCTTGACTCCCCGCACCGAGGTCGCGGCGAnew CGGGCACCCTTCCGCTGATCA-GTCGGTGAGGTTGACTCCCCGCACCGAGGTCGCGCCCAold TCTCGGCGCCGATCTCGCTCAGCACCTGGGCCCGGATGGTGCCGTCGACnew TCTCGGCCCCGATCTCGGTCAGCACCTGGGCCGGGATGGTGCCGTCGAC38Figure 3.2 Alignment of family 6 catalytic domains including corrected CasA aa sequence. Theintroduction of gaps to improve the alignment is indicated by dashes. Numbers at the end ofeach line refer to positions of the last aa in each line in the aa sequence of the pro-enzymes. Anasterisk at the end of the sequence indicates the carboxyl terminus of the enzyme. A dot at theend of the sequence indicates the theoretical end of the catalytic domain. All fully conservedacidic residues are indicated by a + below the sequence. The residues mutated in this study areindicated in bold. Cysteine residues forming disulfide bonds are indicated by a number abovethe sequence corresponding to the disulfide bond. Accession numbers (Genbank or SWISSPROT) are indicated in parenthesis. CfiCenA, Cellulomonasfimi CenA (M15823); TfuEgE2,Thermomonosporafusca E2 (M73321); MbiCe]A, Microbispora bispora CeIA (P26414);MceMcenA, Micromonospora cellulolyticum McenA (no accession number); ShaCell,Streptomyces halstedii Cell (Z12157); SspCasA, Streptomyces sp CasA (L03218); CfiCbhA,Cellulomonasfimi CbhA (L25809); TreCbhll, Trichoderma reesei CbhII (Ml6 190, A03821,M55080); AbiCel3A, Agaricus bisporus Ce13A (L24519); AbiCel3B, Agaricus bisporus Ce13B(L24520); FoxCel, Fusarium oxysporum Cel (L29377); PchCbhII, Phanerochaetechrysosporium CbhII (no accession number).39Cf iCenA TVTPQPTSGFYVDPTTQCYRAWQAASGTDK----ALLEK----IALTPQAYWVGNWADASH-- 186TfuEgE2 -----NDSPFYVNPNMSS-AEWVRNNPNDPR--TPVIRDR----IASVPQGTWFA-HHNPGQ-- 49MbiCelA YDSPFYVDPQSNA-AKWVpNDpR-TpVIRDR- IAAVPTGRWFANYNPST- 49MceMcerA VAGTVSGSLLYRDPSSAV-VRWVAANPGDFR-AAVIREK----IASQPQARWYA-NFNPST-- 61ShaCel 1 ADPTTMTNGFYADPDSSA-SRWAAANPGDGR-AAAINAS----IANTPMARWFGSW--SGA---- 53SspCasA GTTALPSMELY-RAEAGV-HAWLDANPGDHR-APLIAER----IGSQPQAVWFAGAYNPGT-- 54Cf i CbhA APVHVDNPYAGAVQYVNPTWAASVNAAAGRQSADPALAAKMRTVAGQPTAVWMDRI SAl TGNA 63TreCbhl I --TYSGNPFVGVTPWANAYYASEVSSLAIP-SLTGAMATAAAAVAKVPSFMWLDTLDKTPL-- 144AbiCel3A PTSGAGNPYTGKTVWLSPFYADEVAQAAADIS-NPSLATKAASVAKIPTFVWFDTVAKVPD-- 143AbiCel3B PTSOAGNPYTGKTVWLSPFYADEVAQAAADIS-NPSLATKAASVAKIPTFTWFDTVAKVPD-- 143FoxCel 1 - -AASDNPYAGVDLWANNYYRSEVMNLAVP-KLSGAKATAAAKVADVPSFQWMDTYD}-{ISL-- 159PchCbhII PPPSANNPWTGFQIFLSPYYANEV-A1AAKQIMDPTLSSKAASVANIPTFTWLDSVAKIPD-- 1581Cf iCenA ----AQAEVADYTGRAVAAGKTPM--LVVYAIPGRD-CGSHSGGGVSE SEYAR-WVD 235TfuEgE2 ----ITGQVDALMSAAQAAGKIPI--LVVYNAPGRD-CGNHSSGGAPS HSAYRS-WID 99MbiCelA ----VRAEVDAYVGAAAAAGKIPI--MVVYANPNRD-CGGPSAGGAPN HTAYRA-WID 99MceMcenA ----IQSEVSAFIGAANSAQQIPV--LSVYEITNRD-CGCAHAGGAPD LNQYQT-WVS 111ShaCel 1 --- -IGTAAGAYAGMDGRDKLPI--LVAYNIYNRDYCGGHSAGGAAS PSAYAD-WIA 104SspCasA ----ITQQVAEVTSAAAAAGQLPV--VVPYMIPFRD-CGN}ISGGGAPS FAAYAE-WSG 104Cf iCbhA DGNGLKFHLDNAVAQQKAAGVPLVFNLVIYDLPGRD-CFALASNGELPATDACLARYKSEYID 125TreCbhl I - -MEQTLADIRTANKNCCNYAGQ-FVVYDLPDRD-CAALASNGEYSIADGGVAKYKN-YID 200AbiCel3A----LGGYLADARS--KNQLVQ----IVVYDLPDRD-CAALASNGEFSLANDGLNKYKN-YVD 194AbiCel3B----LGGYLADAQS--KNQLVQ----IVVYDLPDRD-CAALASNGEFSLANDGLNKYKN-YVD 194FoxCel 1 - --MEDTLADIRKANKAGGKYAGQ-FVVYDLPNRD-CAAAASNCEYSLDKDGANKYKA-YIA 215PchCbhII----LGTYLASASALGKSTGTKQLVQIVIYDLPDRD-CAAKASNCEFSIAJGQANYEN-YID 215+1Cf iCenA TVAQGI KGNP-IVILEPDALAQLGD C---SGQGDRVGFLKYAAKSLTLKG-- 281TfuEgE2 EFAAGL KNRPAYIIVEPDLISLMSS CMQHVQQ-EVLETMAYAGKALKAGSSQ 150MbiCelA EIAAGL RNRPAVIILEPDALPIMTN CMSPSEQAEVQASMAYAGKKFKAASSQ 151MceMcenA NFARGL GNQT-VIILETDSLALQT CLSTSELNARNQALSTATQTIKSANPN 162ShaCell RFAGGI AARPAVVILEPDSLGDYG CMNPAQIDEREAMLTNALVQFNRQAPN 155SspCasA LFAAGL GSEPVVVVLEPDAIPLI-D CLDNQQRAERLAALAGLAEAVTDANPE 155Cf iCbhA PIADLLDNPEYESIRIAATIEPDSLPNLTTNISEPACQQAA--PYYRQGVKYALDKLHAI-PN 185TreCbhl I TIRQIV--VEYSDIRTLLVIEPDSLANLVTNLGTPKCANAQ--SAYLECINYAVTQLNL--PN 257AbjCel3A QIAAQI--KQFPDVSVVAVIEPDSLANLVTNLNVQKCAJ’JAQ--SAYKEGVIYAVQKLNAVG-- 251AbiCel3B QIAAQI--KQFPDVSVVAVIEPDSLANLVTNLNVQKCANAQ--SAYKEQVIYAIQKLDAVG-- 251FoxCel 1 KIKGIL -QNYSDTKVILVIEPDSLANLVTNLNVDKCAKAE--SAYKELTVYAIKELNL--PN 272PchCbhII QIVAQI--QQFPDVRVVAVIEPDSLANLVTNLNVQKCANAK--TTYLACVNYALTNLAKVG-- 272++Cf jCenA ARVYIDAGHAKWL---SVDTAVNRLNQVGF-EYAV GFALNTSNYQ 322TfuEgE2 ARIYFDAGHSAWH---SPAQMASWLQQADISNSAH GIATNTSNYR 192MbjCelA AKVYFDAGHDAWV---PADEMASRLRGADIANSA DGIALNVSNYR 193MceMcerjA AKVYLDGCHSTWN---SANDTANRLRAAGVQYA DGFFTNVSNFN 203ShaCell TWVYMDAGNPRWA---DAATMARRLHEAGLRQAH CFSLNVSNYI 196SspCasA ARVYYDVGHSAWH---APAAIAPTLVEAGILEHGA CIATNISNYR 197Cf CbhA VYNYIDIGHSGWLGWDSNAGPSATLFAEVAKSTTAGFAS IDGFVSDVANTTPLEEPLLSDSSL 248TreCbhII VANYLDAGHAGWLGWPANQDPAAQLFANVYKNASSPRA-LRGLATNVANYN 307AbiCel3A VTMYIDAGHACWLGWPANLSPAAQLFAQIYRDAGSPRN-LRGIATNVANFN 301AbiCel3B VTMYIDAGHAGWLGWPANLSPAAQLFAQIYPDAGSPRI\I-LRGIATNVANFN 301FoxCeli VSMYLDAGHGGWLGWPANIGPAAKLYAQIYKDAGKPSR-VRCLVTNVSNYN 322PchCbhII VYMYMDAGHAGWLGWPANLSPAAQLFTQVWQNAGKSPF-IKGLATNVANYN 322+40Cf iCenA -TTADYQQIsQRLGK-KFVIDTSRNGNGSNG 356TfuEgE2 WTADEVAYAKAVLSAIGNP-SLRAVIDTSRNGNGPAG 228MbiCelA YTSGLISYAKSVLSAIGAS----HLRAVTDTSRNGNCPLG 229MceMcenA PTSSEANFGRAVISALNGM-GISQKRQVIDTSRNG-GAAG 247haCe11 TTAENTAYGNAVNNELAAR-YGYTKPFVVDTSRNGNGSNG 235SspCasA TTTDETAYASAVIAELGGC LGAVVDTSRNCNGPLG 232Cf iCbhA TINNTPIRSSKFYEWNFDFDEIDYTAHMHRLLVAAGFPSIGMLVDTSRNGWGGPNRPTSITA 311TreCbhII --GWNITSPPSYTQGNAVYNEKLYIHAIGPLLANHGWSNAF-FITDQGRSGKQPTG 360AbiCel3A —-ALRASSPDPITQGNSNYDEIHYIEALAPMLSNAGFPAH--FIVDQGRSGVQNIR 353AbiCel3B --ALRASSPDPITQCNSNYDEIHYIEALAPMLSNAGFPAH--FIVDQGRSGVQNIR 353FoxCell --GWKLSTKPDYTESNPNYDEQRYINAFAPLLAQEGWSNVK-FIVDQGRSGKQPTG 375PchCbhII --ALQAASPDPITQGNPNYDEIHYINALAPLLQQAGWDAT--FIVDQGRSGVQNIR 374+2Cf iCenA EWCNPRGRALGERPVAVNDG---SGLDALLWVKLPGESDCA 394TfuEgE2 NEWCDPSGRAIGTPSTTNTCD---PMIDAFLWIKLPGEADG 266MbiCelA SEWCDPPGRATGTWSTTDTGD---PAIDAFLWIKPPGEADG 267MceMcenA DWCADDNTDRRIGQYPTTNT-GDANIDAYLWVKPPGEADG 280ShaCell EWCNPSGRRIGTPTRTGGG AEMLLWIKTPGESDGN 270SspCasA SEWCDPPGRLVGNNPTVNPGV---PGVDAFLWIKLPGELDG 270Cf iCbhA STDVNAYVDANRVDRRVHRGAWCNPLGAGIGRFPEATPSGYAASHLDAFVWIKPPGESDGAST 374TreCbhII QQQWCDWCNVIGTGFGIRPSANTGD---SLLDSFVWVKPGGECDGTSD 405AbiCel3A DQWGDWCNVKGAGFGQRPTTNTGS---SLIDAIVWVKPGGECDGTSD 397AbiCel3B DQWGDWCNVKGAGFGQRPTTNTGS---SLIDAIVWVKPCGECDGTSD 397FoxCell QKAQGDWCNAKGTGFGLRPSTNTGD---ALADAFVWVKPGGESDCTSD 420PchCbhII QQWGDWCNIKCACFGTRPTTNTQS---QFIDSIVWVKPGCECDQTSN 418+ ++2Cf iCenA CNGG PAAGQWWQEIALEMARNARW* 418TfuEgE2 CIAG AGQFVPQAAYEMAIAAGGT. 289MbiCelA CIAT PGVFVPDRAYELAMNAAPPTY. 292MceMcenA CATR GSFQPDLAFSLANGVPN. 310ShaCell CGVG SGSTAGQFLPEVAYK1IIYGY* 294SspCasA CDGP AGSFSPAKAYELAGG* 289Cf jCbhA DIPNDQGKRFDRNCDPTFVSPKLNNQLTGATPNAPLAGQWFEEQFVTLVKNAYPVIG. 431TreCbhII SSAP----RFDSHCALP DALQPAPQAGAWFQAYFVQLLTNANPSFL* 447AbiCel3A NSSP----RFDSHCSLS DAHQPAPEAGTWFQAYFETLVANANPAL* 438AbiCel3B SSSP----RFDSHCSLS DAHQPAPEAGTWFQAYFETLVANANPAL* 438FoxCell TSAA-----RYDYHCGLD DALKPAPEAGTWFQAYFEQLLDNANPSFL* 462PchCbhII SSSP----RYDSTCSLP DAAQPAPEAGTWFQAYFQTLVSAANPP_L* 459413.2.1 PurWcation by cellulose affinity chromatography.In order to test other schemes, it was first necessary to purify CenA by cellulose affinitychromatography using the standard procedure. This was carried out as described in section 2.7.1.Approximately 50 mg of CenA, estimated by binding to Avicel and SDS-PAGE (Figure 3.3),was obtained in the culture supernatant of a 10-L fermentation after 24 hours. Affinitychromatography on CF1 cellulose yielded about 45 mg of pure enzyme with a specific activityon CM-cellulose (2.7%) of about 2400 mm-1,comparable to the specific activities of otherpreparations (E. Kwan, personal communication). 2,4-DNPC had never been used as a substratefor CenA before. It was hydrolyzed by CenA with a KM value of 0.15 mM and a kcat value of 21mind.3.2.2 Desorbentsfor CenA adsorbed to CF1 cellulose.The first attempt to purify CenA under non-denaturing conditions was by celluloseaffinity purification using desorbents other than guanidinium HCI. The nature of binding ofCenA to cellulose is poorly understood. For this reason, desorbents were chosen for their abilityto interfere with common protein binding forces such as hydrogen bonding and hydrophobicinteractions. As well, some cellulose-like desorbents were chosen to try and compete the proteinoff of the cellulose. The chemicals listed in section 2.7.2 were tested as desorbents of wild typeCenA. The only chemical tested that was successful in desorbing CenA, other than guanidiniumHC1, was 0.1 N NaOH (Table 3.1). Because this solution has a pH of 13 and probably acts todenature the protein, it was not used as a desorbent in this study.42Figure 3.3 Estimation of CenA concentration in culture supernatant. Volumes of 100 .tL and 1mL of culture supernatant were bound to 10 mg of Avicel at 4°C for 1 hour. The Avicel wasseparated from the supernatant by centrifugation and the supernatants were removed. The pelletswere washed three times with 1M NaC1, then three times with 50 mM phosphate buffer (pH 7.0).The pellets were resuspended in 40 jiL of SDS-PAGE loading buffer, boiled for 2 minutes,centrifuged, then 20 iL of clarified loading buffer were loaded onto the gel. Known amounts ofCenA were also loaded onto the gel as standards.Lane 1, molecular mass standards with sizes indicated; lane 2, CenA (1 jig); lane 3; CenA (5 jig);lane 4, CenA (10 jig); lane 5, CenA (25 jig); lane 6, bound supernatant (100 jiL); lane 7, boundsupematant (1 mL).kDa 1234 5 6 743Table 3.1 Chemicals tested as desorbents for CenA adsorbed to CF1 cellulose.Chemical DesorptionH20 No1% cellobiose No1% CM-cellulose No0.1NNaOH Yes0.1NHC1 No20 % glycerol No20 % ethanol No3.2 M DMSO No1% CHAPS No6 M guanidinium HC1 Yes3.2.3 Purification by ion exchange chromatography.On the basis of the amino acid sequence, the calculated p1 of CenA is 6.3 (MacProMass,Beckman Research Institute, Duarte, CA). At pH 8.0, CenA should be negatively charged andshould bind to an anion exchange resin such as MacroPrep® Q. Culture supernatant containingCenA was adjusted to pH 8.0 and passed through a column of MacroPrep® Q resin as desóribedin section 2.7.3. The pH was not raised higher than 8.0 to prevent denaturation of the protein.Two separate Agonm peaks were identified by FPLC (Figure 3.4A); one which did not bind thecolumn (Peak 1) and one that bound to the column and which could be eluted with a gradient ofNaC1 (Peak 2). Fractions containing protein from each peak were monitored by SDS-PAGE(Figure 3.4B). Figure 3.4B shows that CenA did not bind to MacroPrep® Q resin at a pH of 8.0while many of the contaminating proteins did bind. CenA was also found to migrate very slowlyin a non-denaturing PAGE gel at pH 8.8 (data not shown). It seems that the p1 of CenA issignificantly higher than the calculated value of 6.3.A sample of supernatant was adjusted to pH 6.0. At this pH, CenA should be positivelycharged and should bind to a cation exchange resin such as MacroPrep® S. When the samplewas passed through a column of MacroPrep® S resin as described in section 2.7.3, neither CenAnor the contaminating proteins bound to the resin (Figure 3.5). The lack of success with theseresins led to the choosing of another purification method.Peak 1APeak 2-, ———1—19• ....j .J IIUUU IUUUUUUUnuuu,uuau,i— i.••••• •.: i..:. ..i r..—.. —.. ._ ...l..:i:..:..I::.:..Iii::r .Ii::i:,...i:ii:zi....I...:.,:.....,_i::,.i:_ii,..J—,l.....ji::,::i...i:iFractions200—116—66 —CenA—*45—31 —22 —Figure 3.4 FPLC profile and SDS -PAGE of fractions from supernatants containing CenA passedthrough an anion-exchange column. Culture supernatant containing CenA and exchanged withphosphate buffer (pH 8.0), was passed through an anion-exchange column as described inSection 2.7.3. A) Two peaks were observed in the FPLC profile; one that bound to the columnand eluted with a gradient of NaC1 (Peak 2) and one which did not bind the column (Peak 1). B)Peak fractions were analyzed for CenA by SDS-PAGE. Each sample (10 tL) was combinedwith 10 i.LL of SDS-PAGE loading buffer, boiled for 2 minutes and the 20 IlL was loaded ontothe gel. Lane 1, molecular mass standards with sizes indicated; lane 2, pre-columnsupematantlpH 8.0 (10 IlL); lane 3, Peak 1, fraction 1 (10 IlL); lane 4, Peak 1, fraction 3 (10 tL);lane 5, Peak 1, fraction 5 (10 tL); lane 6, Peak 2, fraction 2(10 j.iL); lane 7, Peak 2, fraction 4(10 IlL); lane 8, Peak 2, fraction 6 (10 ilL).44135II.,,..246kDaIi’B12345678_45Peak 1Fractions 1 3kDa200—11666 —CenA -45 —31 —22 —57AFigure 3.5 FPLC profile and SDS-PAGE of fractions from supematants containing CenA passedthrough a cation-exchange column. Culture supematant containing CenA and exchanged withphosphate buffer (pH 6.0), was passed through a cation-exchange column as described inSection 2.7.3. A) Only one peak was observed in the FPLC profile as all of the protein emergedin the flow through fractions. B) Fractions were collected and analyzed by SDS-PAGE. Eachsample (10 j.tL) was combined with 10 tL of SDS-PAGE loading buffer, boiled for 2 minutesand the 20 j.i.L was loaded onto the gel.Lane 1, molecular mass standards with sizes indicated;lane 2, supernatant (1 mL) bound to Avicel (10 mg)/ 20 p.L loaded; lane 3, pre-columnsupematanl/pH 6.0 (10 j,iL); lane 4, fraction 1 (10 tL); lane 5, fraction 3 (10 tL); lane 6, fraction5 (10 i.iL); lane 7, fraction 7 (10 iL).j1h]ikiu. I LLi..__Ii i ti1 i—t’ i—i’ ;i1:r—: ‘ :B12345678-ar463.2.4 Purification by nickel affinity chromatography.The construction of pTUgKRG-1.5ceizA.N, a high expression vector encoding CenA withan N-terminal hexahistidine tail, is shown in Figure 3.6. This form of CenA (CenA.N) wasproduced and purified according to the procedure described in section 2.7.4. Approximately 30mgfL of protein was produced in the culture supernatant after 18 hours of growth. Afterpurification from 500 mL of supernatant, 12 mg of protein remained. CenA.N had the sameaffinity for Avice1 as wild-type CenA. Nickel affinity chromatography was as effective ascellulose affinity chromatography for purification (Figure 3.7). The specific activity of CenA.Nfor the hydrolysis of CM-cellulose was 2000 min’. The KM was 0.16 mM and the kcat was 21mm-’ for the hydrolysis of 2,4-DNPC, virtually identical to the values for CenA. Thus, theaddition of a histidine tail to the N terminus of CenA had no affect on its physical properties. Allmutants of cenA were subcloned into pTUGKRG to give the cenA.N fusions. The mutantproteins could then be purified by nickel affinity chromatography.3.3 Generation of CenA mutants.A number of amino acids are conserved in the catalytic domains of family 6 cellulases,presumably because they play important roles in catalysis or in maintaining correct structure.Glutamic and aspartic acids are involved in the catalytic mechanisms of many glucosidases and,among the 27 carboxylic residues present in the catalytic domain of CenA, 8 are conserved in allfamily 6 proteins. Of these, D216, D252, D287 and D392 are possibly involved in catalysis,based on the location in or near the active site of the corresponding residues in CBHII from T.reesei (Rouvinen et al., 1990). These residues were chosen as candidates for m.itagenesis. Allwere mutated to alanine to completely remove acidic functionality. Alanine was also chosenbecause it is a small amino acid and distortions in the protein structure due to steric effects wouldbe minimized. These residues were also changed to glutamate in order to determine what theeffect would be of decreasing the spacing of the acidic functionality within the active site by onemethylene unit.47Hindlil1) Digest each plasmid with NheI/Hindlll.2) Separate DNA fragments by agarosegel electrophoresis.I IPurify 4.6 kb vector fragment.MiulBsIW1Figure 3.6 Construction of pTUgKRG-l.5cenA.N. Plasmids pTZ18R-1.6cenA.N (Nordquist,1992) and pTUgKRG (Graham et a!, unpublished data) were digested with Nhe I and Hind Ill.The 1.5 kb fragment containing the cenA gene was isolated and ligated to the 4.6 kb fragment ofthe pTIJgKRG vector to give pTUgKRG-1.5cenA.N (6.1 kb).Purify 1.5 kb fragmentcontainingcenA.48kDa 1 2 3 4 5 6 7116=”-—45 — — —31—22— — —14—Figure 3.7 Purification of CenA by nickel affinity chromatography and by cellulose affinitychromatography. Each protein was diluted so that the specified amount of protein was containedin 10 tL; 10 iL of loading buffer was added, the sample was boiled for 2 minutes and the full 20ji.L was loaded into each lane of a SDS-PAG.Lane 1, molecular mass standards with sizes indicated; lane 2, nickel affmity purified CenA.N (1ig); lane 3; nickel affinity purified CenA.N (5 .tg); lane 4, nickel affinity purified CenA.N (25fig); lane 5, cellulose affinity purified CenA (1 gig); lane 6, cellulose affinity purified CenA (5rig); lane 7, cellulose affinity purified CenA (25 .tg).49The plasmid pTZ18R-L6cenA was constructed as outlined in Figure 3.8. Mutation ofD252 was carried out as described in section 2.4.1. and mutants were screened as described insection 2.5. A total of 460 transformants were screened for reduction or loss of activity on CM-cellulose. Of the 460 screened, 17 were found to have smaller halos and 22 were found to haveno dctcctable halos, giving a phenotypic mutation frequency of only about 8%. Only 33 of theputative mutant transformants, as well as some of those having wild type CM-cellulase activity,were chosen for DNA sequencing. Of the 33, 8 were found to have the mutation D252A and 7were found to be D252E. The remaining colonies were mainly deletion mutants or were wildtype in the area sequenced. From these results the useful mutation frequency was only about 3%.This low mutation frequency could be a consequence of the redundant oligonucleotide primersused including the wild type sequence. For this reason, subsequent mutagenic primers used didnot include the wild type sequence. Also, the production of single-stranded DNA from thephagemid pTZ18R-1.6cenA was found to be very inconsistent and yields were very low. Forthis reason, M13mp18-1.6cenA was used to produce single-stranded template DNA for the othermutants. This phage was found to consistently produce large amounts of single-stranded DNA.Mutation frequencies using this system were found to be higher (10 to 30%). All mutants exceptD392E were obtained in this way. Mutant D392E was obtained as described in section 2.4.2.PCR mutagenesis was found to be highly efficient. Of the 16 transformants chosen forrestriction site analysis, 15 were found to be mutants (94%).DNA sequencing of a 272 bp BsiWI-HindIII fragment for the D392 mutants and a 456 bpMluI-Bgill fragment for all other mutants showed them to contain only the mutation of interest(data not shown). The sequenced mutant cassettes were subcloned from pTZ18R-l.6cenA,M13mp18-1.6cenA or the PCR product into pTZ18R-1.6cenA.N, replacing the equivalent wildtype fragment of DNA as shown in Figure 3.9. An Nhe I-Hind III fragment was then subclonedinto pTUgKRG-l.5cenA.N in order to introduce a histidine tail at the N terminus and for highlevel gene expression, as shown in Figure 3.9. The pTUgKRG-1.5cenA.N mutants weretransformed into E. coli JM1OI, plated on CM-cellulose/Kan/TPTG plates, grown at 37°C for 18hours and stained with Congo red. Figure 3.10 shows the relative activities of each of themutants on CM-cellulose.501) Digest each fragment with Sad2) Separate DNA fragments by agarosegel electrophoresis.I IPurify’ 2.7 kb vector fragment.HindlilPurify 1 .6 kb DNA fragmentcontaining cenA.Figure 3.8 Construction of pTZ18R-1.6cenA. Plasmids pTZ18R (Mead et al, 1986) and pUC18-1.6cenA (Wong eta!., 1986) were digested with Eco Ri and Hind Ill. The 1.6 kb fragmentcontaining the cenA gene was isolated and ligated with the 2.9 kb fragment of the pTZ18Rvector to give pTZ18R-1.6cenA (4.4 kb).Sad1pTZ18R-1.6cenA4.4 kbPstl51Figure 3.9 Generation of mutants of cenA. In vitro mutagenesis was performed as described insection 2.4. For the D392 mutants, a 272 bp Bsi W I-Hind III fragment was isolated either fromthe PCR product or M13mp18-1.6cenA and ligated to a4.1 kb fragment from pTZ18R-1.6cenA.N. For all other mutants, a 456 bp MIu I-Bsi WI fragment was isolated and ligated to a4.0 kb fragment from pTZ18R-1.6cenA.N. For each mutant, the mutant cassette replacing thewild type DNA was sequenced to confirm that only the desired mutation was present. Aftersequence confirmation, a 1.5 bp Nhe I-Hind III fragment was isolated from each mutant form ofpTZ18R-1.6cenA.N and ligated with a 4.6 kb vector fragment from pTUgKRG. The positions ofthe mutations are indicated by an asterisk.52Bsi WI Hind IIII D392PCR Product1) Cut PCR product andvector with Bsi WI and Hind II2)Purify vitro mutagenesis/MIu1Bs1W1pTZISR-1.6cenA4.4 kb M13mp18-1.6cenA8.8 kbHindu 1 )Cut with appropriaterestriction enzymes(Mlu /BsiWI or BsiWVHindIlI).2)Purify fragments.3)Ligate 3)LigateMIulBsIWIt392$aflPstlFlindill1) Cut both plasmidswith NheI/HindIII.2)Purify fragments.3)LigateMiul53Figure 3.10 Relative CM-ceilulase activities for the mutant forms of CenA.N compared to thewild-type. Each E. colt TM 101 strain containing the different forms of pTIJgKRG-1.5cenA.Nplasmid was streaked onto an LB/KanJCM-celluloseflPTG plate and grown at 37°C for 18 hours.Plates were stained with Congo red. From top to bottom, left to right; wild-type CenA.N,D252E, D252A, D216E, D216A, D392E, D392A, D287E, D287A.543.4 Production and purification of mutant proteins.CenA.N and all of the mutant proteins were produced and purified as described insections 2.7.4. Purification of these proteins was greatly facilitated by use of the N-terminalhistidine tag encoded by the pTIJgKRG vector. This allowed for one step purification on anickel column with elution of the protein under non-denaturing conditions. No significantdifferences in behaviour of proteins were noted during purification, although expression levelsvaried for each mutant gene. Each mutant protein ran as a single band on SDS-PAGE, at thesame position as the native enzyme. The purity of each protein was assessed by SDS-PAGE(Figure 3.11). Final concentrations of each protein in solution ranged from about 5 mg/mL to 25mg/mL3.5 Structural characterizationIn order to confirm that any changes in activity resulted directly from the mutationintroduced, and not from any major structural changes, the structural integrities of the mutantsand of CenA.N were compared.3.5.1. Monitoring overall structure using circular dichroism.The peptide bonds of proteins absorb UV light from about 190 to 230 nm. Proteins arealso optically active at these wavelengths. Optically active molecules tend to absorb right- andleft-handed circularly polarised light unequally. Although the difference in absorption isgenerally very small, it can be measured with a sensitive instrument. This technique is calledcircular dichroism (CD) spectroscopy. The CD spectrum of the peptide bonds of proteins is verysensitive to the local environment of the peptide bond. Changes in protein conformation wouldaffect the local environment and would show up as changes in the CD spectrum of the protein.CD spectra of CenA.N and all of the mutants are presented in Figure 3.12. All mutantshad spectra similar to that of CenA.N, indicating that no major changes in overall structure werecaused by the mutations.55Figure 3.11 Purification of wild-type CenA.N and the mutant proteins. Before purification, asample of culture supernatant (1 mL) was bound to Avicel(l0 mg) at 4°C for 1 hour. Thepellet was resuspended in 40 IlL of SDS-PAGE loading buffer, boiled for 2 minutes and 20 .tL ofclarified loading buffer was loaded onto the gel. CenA.N was also precipitated from the culturesupernatant (1 mL) with (NH4)2S0 (50%). The pellet was dissolved in 20 p.L of phosphatebuffer and 10 .tL of this was combined with 10 p.L of SDS-PAGE loading buffer. The samplewas boiled for 2 minutes and the 20 p.L was loaded onto the gel. The purity of CenA.N and themutant forms of CenA.N was analyzed by SDS-PAGE. Each purified protein was diluted so thatthe correct amount of protein was contained in 10 IlL, 10 .tL of loading buffer was added, thesample was boiled for 2 minutes and the entire 20 p.L was loaded into each lane.Lane 1, molecular mass standards with sizes indicated; lane 2, Avicel bound CenA.N (from anequivalent of 500 pL of supematant); lane 3, (NH4)2S0precipitated CenA.N (from anequivalent of 500 IlL of supematant); lanes 4 to 20, purified proteins; lane 4, CenA.N (2 pg);lane 5, CenA.N (10 tg); lane 6, D216A (2 rig); lane 7, D216A (10 jig), lane 8, D216E (2 pg);lane 9, D216E (10 jig); lane 10, D252A (2 jig); lane 11, D252A (10 jig); lane 12, D252E(2jig); lane 13, D252E (10 jig); lane 14, D287A (2 jig); lane 15, D287A (10 jig); lane 16, D287E(2 jig); lane 17, D287E (10 jig); lane 18, D392A (2 jig); lane 19, D392A (10 jig); lane 20,D392E (2 jig); lane 21, D392E (10 jig).56kDa200—116—66 —45 —31 —22 —2 3 45 67 89 10 1120 2112 13 14 15 16 17I18 1957Figure 3.12 CD spectra of CenA.N and mutants. The data are expressed in terms of meanresidual ellipticity (0).A) Wild type CenA.N (thick—), D216A (thick ---), D252A (thin —), D287A (thin ---), andD392A(thin” ).B) Wild type CenA.N (thick—), D216E (thick ---), D252E (thick “), D287E (thin —) andD392E (thin---).585A— —-—10 I I I I200 210 220 230 240 250Wavelength (nm)5E-5 /—10 I I I200 210 220 230 240 250Wavelength (nm)593.52 Monitoring the active site structure by cellobiose affinity chromatography.Integrity of the active site structure was investigated by passing the various mutantproteins through a cellobiose affinity column. CenA.N was found to bind tightly to the columnand could only be eluted with 100 mM cellobiose. The cellulose-binding domain from CenA.Ndid not adsorb to the column, showing that binding to cellobiose was dependent on the catalyticdomain alone. All of the mutant proteins also bound to the affinity column and could beremoved with 100 mM cellobiose.3.6 Catalytic properties of CenA.N and the mutants.In this study, two general classes of substrates were used in kinetic studies to identifycatalytic residues; cellulose-like substrates having glucosidic leaving groups of high pKa valueand chemically synthesized model substrates having leaving groups (aglycones) of low pKavalue. CenA lacks the ability to efficiently hydrolyze crystalline forms of cellulose but is quiteactive on amorphous cellulose and various cellodextrins (Meinke et al., 1993b; Neil Gillces,personal communication). For this reason, PAS-cellulose, CM-cellulose, cellotetraose andcellotriose were chosen as substrates having glucosidic groups of high pKa value. Choice ofsubstrates having aglycones of low pKa value was limited by the inherently low rates ofhydrolysis of these substrates by CenA, as well as by their availability. In this study variousdinitrophenyl cellobiosides as well as a- and 8-cellobiosyl fluoride were used.3.6.1 Hydrolysis of CM-cellulose and PAS-cellulose.The 13-1,4-glucosidic bonds of PAS-cellulose and CM-cellulose are very stable andalmost no hydrolysis occurs in the absence of enzyme. In order for hydrolysis to occur, theglucosidlic bond being cleaved must go through a transition state where the glucosidic oxygencontains considerable negative charge and the anomeric carbon contains considerable positivecharge (Figure 3.13). Such a transition state is highly unstable. The positive charge on theanomeric carbon is poorly stabilized by surrounding water molecules but can be partiallystabilized by resonance stabilization with the ring oxygen. The negative charge on the glycosidic60oxygen is also poorly stabilized by the surrounding water molecules and cannot be effectivelystabilized by the departing glucoside (Figure 3.13). For this reason, the rates of spontaneoushydrolysis are quite low. The pKa values for deprotonation of the hydroxyl groups of glucosidesare also very high (>14) because of the inability of the glucoside to stabilize the negative chargeformed during deprotonation.Because these substrates are so stable, all of the catalytically important amino acids in theenzyme that stabilize the transition state during hydrolysis, such as the acid and base catalyst, arerequired for full activity. If an amino acid is removed, then the activity should decreaseproportionally to the importance of that amino acid in hydrolysis. The relative activities on bothPAS-cellulose and CM-cellulose were determined for wild type CenA.N and the mutants (l’able3.2). Accurate kcat and KM values could not be determined because of inherent variability of thesubstrate. Wild type activities on both substrates werewithin 10% of those observed previously(Emily Kwan, personal communication). Decreases in activities for each mutant are similar onboth substrates.Table 3.2 Hydrolysis of CM-cellulose and PAS-cellulose by CenA.N and the mutants.CM-cellulose PAS-celluloseEnzyme Rate Fold Decrease Rate____________ Fold DecreaseCenA.N 1700---- 640D216A 13 130 0.50 1300D216E 1.2 1400 0.40 1600D252A 0.0078 220000 ND NDD252E 0.60 2900 0.40 1600D287A 0.10 17000 0.050 13000D287E 530 3.0 470 1.4D392A 0.050 34000 0.020 32000D392E 1.8 940 ND NDRates for hydrolysis of CM-cellulose (4 mg/mi) and PAS-cellulose (10 mg/mL) were determinedat 37°C in 50 mM citrate buffer, pH 7.0. Rates are expressed as imoles of reducing sugarreleased per mm per pmole of enzyme. Fold decreases in activities are calculated as Rate ofCenA.N / Rate of mutant. ND = not determined.61H20OHH20‘4ROH200HORFigure 3.13 General mechanism for the hydrolysis of B-1,4-glucosidic bonds. R indicates acontinuing chain of glucose residues.HQ HQ6H6623.6.2 Hydrolysis of cellotriose and cellotetraose by wild-type CenA.N.Cellotriose and cellotetraose are similar to CM-cellulose and PAS-cellulose in that thesame 13-1,4-glucosidic bond is cleaved. The pKaS of the leaving groups for these substrates(glucose and cellobiose) are also very high (> 14). CenA.N cleaves cellotriose to cellobiose andglucose while cellotetraose is cleaved to cellobiose only (Figure 3.14) (Claeyssens and Henrissat,1992). The kt and KM values for hydrolysis of these substrates had not been determinedbecause of difficulties in accurately quantifying the products. In this study, products wereseparated and quantified using an HPLC (Dionex) equipped with a PAD detector (Dionex). Thissystem allowed for efficient separation of the products and very accurate quantification. Thekeat and KM values for these substrates are presented in Table 3.3. Cellotriose is a poor substratefor CenA.N with a kt value of only 16 mm4. Cellotetraose is significantly better, with a kcatvalue which is 755 fold higher (13000 min1). The KM value for cellotriose is also about 5 foldlarger than for cellotetraose. Substrate inhibition with cellotetraose occurred at concentrationsgreater than 0.75 mM (6.6 x KM). Unfortunately, analysis of mutant forms of CenA.N could notbe carried out with these substrates due to restrictions on the availability of the Dionex system.Table 3.3 Hydrolysis of cellotriose and cellotetraose by CenA.N.Substrate kcat (min4) KM (mM) kcat I KM (min4mM1)cellotriose 16 0.58 28cellotetraose 13000 0.11 1100003.6.3 Hydrolysis of nitrophenyl cellobiosides.Nitrophenyl cellobiosides are substrate analogues of cellotriose in which the glucoseresidue at the reducing end has been replaced with a nitrophenol group. The nitrophenyl groupsstabilize the negative charge build-up on the glucosidic oxygen of the scissile bond duringhydrolysis by resonance stabilization (Figure 3.15) and therefore the nitrophenyl cellobiosidesare generally more reactive.OHOH OH HOIlLCELLOTETRAOSEOHHO2Z7jZ7OH HO OHCELLOfl{IOSE0HFigure 3.14 Cleavages of various cello-oligosacharides by CenA. Points of hydrolysis are63CELLOPENTAOSEindicated by an arrow.641120—OH11201120— 0•sis of a ittoph1Yl cellobi05- In this figurepigure 3.15 General mecha11sm for the hYclrolYa glucose residue.pNPC is used as an exan1Pl R grouP indicates11Q8—ROiiq865In this study, 2,4-DNPC, 2,5-DNPC, 3,4-DNPC and pNPC were tested as substrates forCenA.N (Figure 3.16). Because 2,4-DNP can most effectively stabilize any negative chargebuild-up on the glucosidic oxygen, 2,4-DNPC is the most reactive of these substrates and has thelowest pKa. The kcat and KM values for hydrolysis of 2,4-DNPC by wild type CenA.N and allmutants are shown in Table 3.4. CenA.N had similar KM and kcat values for the hydrolysis of2,4-DNPC and cellotriose. Where determined, all mutant forms of CenA.N had similar KMvalues except for D252A and D252E. The k and KM values could not be determined forD392A and D392E because the activities were too low.Table 3.4 Hydrolysis of 2,4-DNPC by CenA.N and mutants.Enzyme kcat (min1) KM (iiiM) kcat I KM (minmM)CenA.N 22 0.17 130D216A 1.2 0.38 3.2D216E 0.050 0.11 0.50D252A 18 0.0 10 1400D252E 1.1 0.030 34D287A 0.60 0.10 5.9D287E 8.8 0.14 62D392A <0.00101 ND NDD392E 0.00781 ND ND1Rates were determined at 1 mM DNPC only. The value for D392A was an estimate based onthe amount of protein added and the lowest rate that could be accurately measured.Nitrophenyl cellobiosides with aglycones having different pKa values can be used assubstrates to determine if there is a correlation between the stabilization of the transition state bythe enzyme and the build up of negative charge on the oxygen of the nitrophenyl leaving group.Plots of -log (k) and -log (k/KM) vs pKa having a slope of 1 would tend to indicate that thenitrophenolate oxygen is fully deprotonated at the transition state, while a slope of 0 wouldindicate that the nitrophenolate oxygen is fully protonated at the transition state Tu1l andWithers, 1994). The kcat and KM values for nitrophenyl cellobiosides with aglycones havingdifferent pKa values were determined for wild type CenA.N and are shown in Table 3.5.:02,4-DNPC+/R/0:2,5-DNPC:0:3,4-DNPCpNPCFigure 3.16 Structures of 2,4-DNPC, 2,5-DNPC, 3,4-DNPC and pNPC. These compounds wereused as substrates for CenA. R group indicates a glucose residue.66+/...67Table 3.5 Hydrolysis of various nitrophenyl cellobiosides by CenA.N.Enzyme Substrate pKa keat (min1) KM (mM) keat / KM (mM1min)CenA 2,4-DNPC 3.96 22 0.17 1302,5-DNPC 5.15 0.70 0.48 1.53,4-DNPC 5.36 0.043 0.13 0.34pNPC 7.18 0.047 0.41 0.11Only four aryl cellobiosides were used due to the inherently low activity of CenA onthese types of substrates. A plot of -log(kt) and -log(k/Kij) vs pKa for the various substratestested is shown in Figure 3.17. Although the reliability of the linear fit is not very good becauseonly four data points were used, reasonable correlation coefficients of about 0.8 and 0.9,respectively, were obtained. The approximate slopes for these curves were also about 0.8 and0.9, respectively.3.6.4 Hydrolysis of a- andJ3-cellobiosyl fluoride.8-Cellobiosyl fluoride is quite reactive, mainly because of the low pKa of its fluorideleaving group. Because 8-cellobiosyl fluoride is very reactive and is of the correct anomericconfiguration, it was tested as a substrate for CenA.N. As expected, hydrolysis of 8-cellobiosylfluoride by CenA.N exhibited regular saturation kinetics with kcat and KM values of 55 min1and 0.6 mM, respectively. These values are very similar to the values for 2,4-DNPC. The rate ofspontaneous hydrolysis for 8-cellobiosyl fluoride was quite significant at 2 x iO mintM1.The rate of spontaneous hydrolysis was subtracted from the rate of enzyme catalyzed hydrolysisin all calculations for all enzymes. The relative rate of hydrolysis of L-cellobiosyl fluoride by thealanine mutants was also determined (Table 3.6). The actual kcat and KM values were notdetermined for these mutant forms of CenA.N due to lack of 8-cellobiosyl fluoride.a-Cellobiosyl fluoride has the wrong anomeric configuration to be a typical substrate forCenA.N but may be hydrolyzed by the Hehre mechanism described in section = 0.789 * x + -3.89 r = 0.82160-log (k)0.004.0 5.0 6.0 7.0PKa8.0y = 0.902 * x + -5.1 r = 0.890000- log (kcat/Km)-1.004.0 5.0 6.0 7.0PKa8.0Figure 3.17 Hammett plots for the hydrolysis of various nitrophenyl cellobiosides by CenA.N.69This was indeed the case as a plot of rate vs S0 for a-cellobiosyl fluoride gave an upwardlyconcave curve (Figure 3.18) and a plot of 1/rate vs 1/S02 gave a straight line (Figure 3.19),suggesting that hydrolysis was substrate activated. The rate of spontaneous hydrolysis for acellobiosyl fluoride was 1 x 10 min1I.tM and again, this was taken into account in allcalculations. The relative rates of hydrolysis of a-cellobiosyl fluoride by the alanine mutants areshown in Table 3.6.Table 3.6 Hydrolysis of a- and 13-cellobiosyl fluoride by CenA.N and mutants.Enzyme Rate on a- cellobiosyl fluoride Rate on 8-cellobiosyl fluorideCenA 3.4 37D216A 0.15 0.80D252A < 0.001 20D287A < 0.00 1 2.3D392A < 0.00 1 < 0.80Relative rates were determined on a-cellobiosyl fluoride (25 mM) and 13-cellobiosyl fluoride (1mM) and are reported as j.LM of fluoride released per minute per tM of enzyme. The relativerate indicates rate at that particular substrate concentration.3.65 Hydrolysis of2,4-DNPC in the presence of small anions.In some retaining glucosidases in which the general acid/base catalyst has been mutatedto an alanine or glycine, activity can be restored upon addition of small anions such as azide(MacLeod et a!., 1994a). In some cases, this also occurs when the catalytic nucleophile has beenreplaced with an alanine or glycine (Wang et a!., 1994). These small anions are able to replacewater or the catalytic nucleophile, respectively, in the hydrolytic reaction. In so doing, theanions become covalently attached to the substrate and the product can be detected by thin layerchromatography (mc). Theoretically, if the general base of an inverting enzyme was mutated toan alanine, then a small anion, such as azide, should be able to take the place of water in a similarway (Figure 3.20). The result would be a rate enhancement as well as the production of acellobiosyl azide, which would be detectable by TLC.704030 - -CE -20 - -a:10 - -0 I I I I I I0 2000 4000 6000 8000S0(pM)Figure 3.18 Plot of rate vs S0 for the hydrolysis of cz-cellobiosyl fluoride by CenA.N.71‘ I8——-E-(Es- 02-0 I I 10 0.002 0.004 0.0061/S0 (pM1)I I I I I I I I8-y = 1 .8e5x + 0.3, (r)=0.9967(Es02—0I I I I I I i0 2e-005 4e-00511(S02) (pM2)Figure 3.19 Double reciprocal plots for S0 and S02. A) Plot of 1/rate vs 1/S0 for the hydrolysisof a-cellobiosyl fluoride by CenA.N. B) Plot of 1/rate vs 1/S02for the hydrolysis of acellobiosyl fluoride by CenA.N.72o c-HO 1J.OHR’OQ — -N—N\Cnem1 Base \(netaI Bases’ \ imitated toimitated to an alanine.analanine.All sAROHO” O.R’OHR’O-HORN—N PRODUCES LEAVEENZYME RESTOPEDcnem1BaseGneia1Base initated toimitated toan alanine.Figure 3.20 Proposed mechanism for the hydrolysis of 2,4-DNPC in the presence of azide byCenA.N. In this case, azide should take the place of H20 as the nucleophile and directly reactwith the anomeric carbon. R group indicates 2,4-DNP. R’ indicates a glucose residue.73In order to test whether small nucleophiles couldtake the place of water in the reactionmechanism, and thus help to identify the general base catalyst, hydrolysis of 2,4-DNPC (1 mM)by all alanine mutants was measured in the presence of various concentrations of azide (Table3.7). Azide inhibited CenA.N, D252A and D216A, butonly slightly. No hydrolysis by D392Acould be detected either in the absence or presence of azide, even when very high concentrationsof enzyme (100 .i.M) were used. The activity of D287A,however, was enhanced by azide, with amaximum of about 4 fold increase at 500 mM.The reaction products were screened by TLC, butno new products, such as cellobiosyl azide, were observed. In addition, rates of hydrolysis of2,4-DNPC by the alanine mutants were measuredin the presence of 500 mM formate, acetateand isothiocyanate. Rate enhancement was only observed for D287A in the presence of formateand no new products were observed for any ofthe enzymes (data not shown).Table 3.7 Hydrolysis of 2,4-DNPC by CenA.Nand mutants in the presence of azide.Enzyme Rate (min1)0 mlvi azide. 100 mM azide 250 mM azide500 mlvi azide 1000 mM azide18 15 1614 9.20.90 0.92 0.920.80 0.4817 15 1311 7.50.56 1.1 1.92.5 pH dependence studies.Removing the acidic moiety from the targeted amino acids could affect other amino acidsinvolved in glucoside hydrolysis and affect the pH dependence of the reaction. In this study, 2,4-DNPC was used in order to determine both k< and kcatjKM values at different pH values forCenA.N as well as D216A, D252A and D287A.D392A was not used as it had been previouslyfound to have no activity. Changing buffers fromlow to high pH did not affect enzyme activityas rates of hydrolysis in each buffer at overlappingpH values were identical. Enzyme activityremained constant over time at each pH value indicating all enzymes were stable under theseCenAD216AD252AD287AD392AHydrolysis of 2,4-DNPC (1 mM) was determined at 37°Cand rates are expressed as iM DNPreleased per minute per i.LM of enzyme. Dashedline indicates no activity detected.74conditions. The increase in ionic strength by addition of 150 mM NaCI caused a small decreasein activity on all proteins at pH 7.0 (Table 3.4).If the proper assumptions are met (Alberty and Massey, 1954; Peller and Alberty, 1959),the pH-dependence ofk can sometimes provide pKa values of a particular enzyme-substratecomplex, and that of kcat/KM can and usually does provide pKa values of the free reactant state(Brocklehurst, 1994). A plot of kcat vs pH is shown in Figure 3.21. The pH profile curve wasbell shaped for all proteins except D252A, where activity did not decrease significantly withincreasing pH. Although the pH optimum of CenA.N had previously been assumed to be 7.0(Langsford, 1988) these results showed that the optimal pH was actually 6.5. A plot of kcat/KMvs. pH is shown in Figure 3.22. The results were similar to those in Figure 3.21, except that thecurves were all shifted to the left (acidic) by 0.5 to 1 pH unit. pKa values were determined foreach enzyme and are listed in Table 3.8.Table 3.8 pKa values for wild-type CenA.N, D216A, D252A and D287.Enzyme kcat VS pH kcat/KM vs pH.a (lower limb) j (upper limb) p (lower limb) p (upper limb)6.3 6.7 5.7 5.96.3 6.4 5.8 5.3D287A 6.3 4.8 Determination ofstereospecci1y of hydrolysis for mutantforms of CenA .N.If the base catalyst of an inverting enzyme such as CenA were moved closer to theanomeric carbon of the scissile bond of the substrate, for instance by mutation from aspartate toglutamate, then the possibility exists that this amino acid could act as a nucleophile, thuschanging the mechanism of CenA from inverting to retaining.302001004 61DpHFigure 3.21 Plot of kcat vs pH for hydrolysis of 2,4-DNPC by CenA.N. The pH profi1 curveswere fitted to the data points directly using GraFit 3.0.8760.30.2E0.108 9pHFigure 3.22 Plot of kcatlKM vs pH for the hydrolysis of 2,4-DNPC by CenA.N. The p11 profilecurves were fitted to the data points directly using GraFit 3.0.4 5 67In order to test this hypothesis, all mutants were analyzedfor the ability to hydrolyzecellopentaose by a mechanism resulting in overall retentionof anomeric configuration.Cellopentaose is cleaved into cellobiose and cellotriose byCenA (Figure 3.14). The individual xand B anomers of cellotriose can be separated on a Dextro-Pac HPLC column, and therefore theanomeric configuration of the cellotriose produced duringhydrolysis can be determined (Braunet at, 1993). Although the resolution of Dextro-Pac column is quite poor, the anomers ofcellothose can be resolved (Figure 3.23). HPLC profiles are shown for CenA.N and D392A onlybut all mutants gave similar results as the wild-type CenA.N, indicating that no change inmechanism occurred.7778Figure 3.23 Stereospecificity of hydrolysis for mutant forms of CenA.N. A) HPLC profile offully mutorotated cellotriose. B-Cellotriose is thermodynamically more stable than the cccellotriose and is therefore the predominant anomer. a-Cellotriose appears as a shoulder peak onthe right of 13-cellotriose in the HPLC profile. B) HPLC profile of cellopentaose hydrolyzed bywild-type CenA.N after 5 minutes. Initially, cc-cellotriose is the predominant anomer indicatinghydrolysis occurs by inversion of anomeric configuration. 13-Cellotriose appears as a shoulderpeak on the left of a-cellotriose. C) HPLC profile of cellopentaose hydrolyzed by wild-typeCenA.N after 30 minutes. Cellotriose has mutorotated toward its equilibrium state where the 13-anomer predominates. D) HPLC profile of cellopentaose hydrolyzed by D392E after 5 minutes.Initially, a-cellotriose is the predominant anomer indicating hydrolysis occurs by inversion ofanomeric configuration. 13-Cellotriose appears as a shoulder peak on the left of cz-cellotriose.ffer4—Buffer_B-G34—Buffer+-G2C-)-sa-G3S-ufferG2•13-G3cz-G34a-G3804. Discussion4.1 Effects of mutation of CenA on the hydrolysis of substrates having leaving groups withVUOUS pKa values.Hydrolysis by inverting glycosidases such as CenA requires both acid and base catalysis.A proton transfer occurs from the acid catalyst to the glycosidic oxygen as the glycosidic linkageis cleaved, thereby facilitating bond cleavage through stabilization of the leaving group. In aconcerted process, the general base catalyst removes a proton from water, thus making it morenucleophilic, as the water attacks the anomeric center of the glucoside. Because these catalyticresidues are mechanistically important, their modification will affect the rate of catalysis. Theextent to which the rate is affected will depend on the substrate being hydrolyzed. Substrateshaving aglycones of high pKa value and, therefore, poor leaving group ability, will be affectedgreatly by removal of the acid catalyst. Those with leaving groups of low pKa, which need littleor no protonic assistance for departure, will be affected very minimally, if at all. This has beenshown in the case of retaining glycosidases (Kempton and Withers, 1992; MacLeod et al., 1994a;Tull and Withers, 1994). By contrast, removal of the base catalyst should have an effect on allsubstrates as deprotonation of water is crucial for catalysis.4.1.1 Wild-type CenA.N.The substrates PAS-cellulose, CM-cellulose, cellotetraose and cellotriose have leavinggroups of very high pKa and their rate of spontaneous hydrolysis is negligible. Both CM-cellulose and cellotetraose are excellent substrates for CenA.N with enzymatic rateenhancements estimated at approximately l010 to 102-fold over spontaneous hydrolysis(Konstantinidis et al., 1993). The for cellotriose is 760-fold lower than for cellotetraose andthe KM value is about 5-fold higher. Based on crystal structures from E2 from T. fusca andCbhll from T. reesei (Rouvinen eta!., 1990; Spezio et al., 1993b), as well as on binding studieswith small soluble substrate analogues (Van Tilbeurgh et a!., 1989; Van Tilbeurgh et a!., 1986),the active site of family 6 enzymes consists of 4 subsites (A to D), with hydrolysis occurring81between subsites B and C. Presumably, CM-cellulose and cellotetraose are able to completelyfill the binding subsites while cellotriose binds in subsites A through C. In CenA, binding in allsubsites is very important for full realization of catalytic potential because when subsite Dremains empty, both KM and k values are significantly affected. In kinetic studies with CbhII,the kcat for hydrolysis of cellotetraose (198 miw1)is only 100-fold greater than for cellotriose(0.18 min1)(Konstantinidis et at., 1993). The lesser requirement for binding in subsite D maybe a consequence of a fundamental difference between endoglucanases and exoglucanases.The nitrophenyl cellobioside substrates (2,4-DNPC, 2,5-DNPC, 3,4-DNPC and pNPC)have leaving groups with pKa values that are considerably lower than those for CM-cellulose,cellotetraose or cellotriose and should therefore be good substrates for CenA. The most reactiveof these, 2,4-DNPC, has a kcat value only 1.4-fold higher than that for cellotriose, while the KMis virtually unchanged. As the pKa of the leaving group increases, the kcat decreasessignificantly, while KM values are not affected significantly. For instance, the least reactive,pNPC, has a which is 470-fold lower than for 2,4-DNPC and cellotriose even though the pKaof the leaving group of this substrate is much lower than that of cellothose. Presumably, becausethe only difference between cellotriose and the nitrophenyl cellobiosides is the substitution insubsite C, binding in this subsite is also crucial for full catalytic activity. Obviously, in the caseof 2,4-DNPC, the loss of binding in subsite C is made up for by the very high reactivity of thissubstrate.Plots of -log (kcat) and -log (kcat/KM) VS pKa of the leaving group can be used todetermine if there is a correlation between the stabffization of the transition state by the enzymeand the build up of negative charge on the oxygen of the nitrophenyl leaving group. Althoughthe correlation is not completely satisfactory, the slopes of the curves for the plots of -log (k)and -log (k/KJ versus pKa are quite high, at about 0.8 and 0.9, respectively. This trendreflects a large degree of negative charge accumulation on the nitrophenolate oxygen at thetransition state and indicates that stabilization of the transition state by acid catalysis isunimportantThe pH profile of wild type CenA.N on 2,4-DNPC gave a two limbed curve indicatingthat CenA.N depends on the presence of a protonated group and a deprotonated group for full82activity (Figure 3.21, Figure 3.22). Although interpretation of pH dependence data is a veryrisky pastime (Kempton and Withers, 1992), the protonated group is probably indicative of thegeneral acid catalyst (pKa 5.9, free enzyme; pKa 6.7, substrate bound enzyme) of the invertingmechanism, while the deprotonated group may be indicative of the general base (pKa 5.7, free;pKa 6.3, substrate bound). The pKa values of these two titratable groups are very similar for bothfree enzyme and substrate-bound enzyme. These similarities may reflect the fact that both aminoacids must donate and accept protons, respectively, during hydrolysis at the optimum pH of 6.5.If the pKa value of the general base were too low, then it might not act as a good proton acceptorand thus not remove a proton from water during hydrolysis while if the pKa of the acid were toohigh, it would not easily donate a proton to the substrate. The pKa values of the acid and basecatalyst are higher than expected for either glutamate or aspartate residues. These high pKaSprobably result from the presence of nearby negatively charged or hydrophobic groups in theenzyme, both of which would destabilize the deprotonated form of the acids and thus raise theirpKa. The binding of enzyme to substrate appears to raise the apparent pKa value of these groupseven further and this is most likely due to the exclusion of water from the active site by thesubstrate, thus making it a more hydrophobic environment.4.12 D252AThe kt value of D252A on CM-cellulose, a substrate with a sugar leaving group andthus requiring acid catalysis, is reduced at least 2 x 105-fold relative to wild type. This is by farthe largest decrease in activity observed. Whether the residual activity is from the mutant or isdue to contaminating wild type protein could not be determined; thus, the value determined mustbe considered as a minimum estimate. Because the activity was so low on CM-cellulose, thismutant was not assayed on PAS-cellulose. In this respect, CenA.N conforms with the patterntypical for carbohydrases, where the acid catalyst seems most critical for catalysis on naturalsubstrates. In contrast to CM-cellulose, the keat value of D252A on DNPC is almost the same asthat of the wild type enzyme. This is consistent with a role of this residue as general acid catalystsince the rate of hydrolysis of a substrate not requiring acid catalysis is not reduced while the rateof a substrate requiring acid catalysis is significantly reduced. Interestingly, the KM value for83DNPC with this mutant is considerably lower than that with wild type. The origin of thisimproved binding is unclear, though it is tempting to attribute it to improved interactions of thearomatic leaving group with a less polar active site.Unfortunately, a pKa value for the acid catalyst in the free enzyme could not be estimatedwith this mutant as the KM values could not be accurately determined at high pH. The pH profileof the substrate-bound form of this mutant is very interesting in that removal of the acidfunctionality at D252 results in the loss of the basic limb. Because this substrate does not requireacid catalysis, the activity on 2,4-DNPC should not be affected with increasing pH and this isindeed the case. Eventually, at high enough pH, the activity should again decrease as otherimportant amino acid residues become deprotonated. A similar result was observed for acatalytic residue of glucoamylase from A. niger (Frandsen eta!., 1994).The crystal structures for both CbhII and E2 also show that the corresponding residues,D221 and Dl 17, respectively, are located in a position to act as the general acid (Rouvinen et a!.,1990; Spezio et a!., 1993b). In E2, the carboxyloxygen of Dl 17 is 4.5 A from the hydroxylgroup on the reducing end of cellobiose. Mutation of D221 in CbhII showed that this residue isessential for activity on CMC, as all activity is lost in the D221A mutant (Rouvinen et at., 1990).4.13 D392AThe mutant D392A shows the second largest decrease in activities. The keat values onCM-cellulose and PAS-cellulose are reduced about 3 x 104-fold, while virtually no activity isdetectable with 2,4-DNPC. Because this residue is important for activity both onsubstrateshaving very poor leaving groups (CM-cellulose and PAS-cellulose), and on substrates withexcellent leaving groups (2,4-DNPC), it is likely that it acts as the general base catalyst in theproposed mechanism. The incoming water molecule is not very nucleophilic and thereforecannot displace the leaving group on its own, even with acid catalysis. The general base isrequired to deprotonate the incoming water molecule thus making it more nucleophiic.The inverting enzymes Ce1D from C. thermocellwn, and 8-amylase from Bacilluspolymyxa and soybean are also very sensitive to mutation of catalytic residues (Chauvaux et at.,1992; Totsuka and Fukazawa, 1993; Totsuka et a!., 1994). The catalytic base mutant of Cell)84has k< reduced by 3 orders of magnitude while the base mutants of bacterial and soybean Bamylase are 4 orders of magnitude lower and completely inactivated, respectively. The putativegeneral base catalyst mutant of glucoamylase has unusually high residual activity (Frandsen etaL, 1994). This is in contrast to other amylolytic enzymes of the a-amylase family which arecompletely inactivated by mutation of any of the three invariant catalytic site carboxyl groups.This high activity might be attributable to rearrangement of the hydrogen bonding networkaround the mutated glutamate residue, allowing productive orientation of a water molecule.Alternatively, the high activity of the mutant could indicate that the amino acid is a poor basecatalyst, with the mechanism depending much more on acid catalysis.The crystal structures of both Cbhfl and E2 show that the residues corresponding to D392of CenA, D401 and D265, respectively, are in a position to act as general base (Frandsen et at.,1994; Rouvinen et at., 1990; Spezio et at., 1993b). In E2, D265 is 5.0 A from the anomericcarbon on the reducing end of cellobiose and 5.7 A from the hydroxyl group, allowing enoughspace for a water molecule. This residue also forms a salt bridge with R221 and is therefore verylikely in the deprotonated form (Spezio et al., 1993b).Unfortunately, the extremely low activity of the D392A mutant of CenA precludedanalysis of its pH dependence.4.1.4 D287AThe third largest decrease in activity upon mutation was that for D287A, where kcatvalues were reduced 2 x io- and 1 x 104-fold on CM-cellulose and PAS-cellulose, respectively.This residue is obviously important for activity on these substrates but is not as crucial as D252or D392. The keat value for 2,4-DNPC is reduced only 37-fold while the KM value remainsvirtually unchanged. Because there is a much larger decrease in activity on substrates requiringacid catalysis, it is likely that this residue in some way aids the acid catalyst. Perhaps this is doneby providing the correct environment to raise the pKa of D252, thus keeping it protonated andable to act as an acid. Indeed, the structure for E2 reveals that the corresponding residue, D156,is only 3.8 A from the putative acid, D117 and thus may well play a role in modulating theplacement or the acidity of the acid catalytic group. This residue appears to be too buried to act85as the acid catalyst itself as it is 9.6 A from the hydroxyl group on the reducing end of cellobiose(Spezio et at., 1993b).If indeed this residue is important in providing the correct environment to raise the pKa ofthe acid catalyst, then its mutation would be expected to lower the pKa of the acid. This occursin glucoamylase, where the apparent pKa of the general acid catalyst is depressed significantly intwo mutants (D176 and E180) when the carboxylate functionality is removed (Sierks andSvensson, 1993; Sogaard et at., 1993). For CenA.N, the mutation D287A seems to lower thepKa values of the acid catalyst and general base in the substrate-bound enzyme from 6.7 to 6.3and 6.3 to 5.7, respectively (Table 3.8). It may be that this residue has an effect on the acidcatalyst as well as the general base or may reflect the fact that the acid limb pKa has actuallybeen reduced from 6.7 to below the pKa of the lower limb (6.3). pKa values for the acid andbase catalyst in the free enzyme are less reliable due to the low activity of this mutant.4.15 D216A.The kcat values for D216A are reduced for all substrates tested, but only by 18- 132- and1380-fold on 2,4-DNPC, CM-cellulose and PAS -cellulose, respectively. The 10-fold loweractivity on PAS-cellulose compared to CM-cellulose is interesting and may be a consequence ofthe importance of this residue for hydrolysis of less soluble substrates. Because this mutantretains considerable activity, it is unlikely that this residue is directly involved in catalysis. TheKM value of this mutant on 2,4-DNPC is approximately the same as for the wild type enzyme,which may indicate that D216 is unimportant in binding this substrate. Mutation of thecorresponding residue in CbhII reduces the rate of hydrolysis of cellotrióse some 5-fold(Rouvinen et at., 1990), which is much less than the effect seen for CenA.N on CM-cellulose andPAS-cellulose. Recent data suggests that the decrease in activity for the corresponding mutant inCbhII is actually much greater than 5-fold, although actual numbers are not available (Ollerman,personal communication). Mutation of one of the conserved aspartate residues in soybean Bamylase, an inverting enzyme, also has little effect on activity (Totsuka and Fukazawa, 1993;Totsuka et a!., 1994).86In the structure of Cbhll, the corresponding amino acid, D175, is positioned near D221,the acid catalyst, and could serve to raise its pKa (Rouvinen et al., 1990). In the structure for E2,the corresponding amino acid, D79, is 1 iA away from the acid catalyst, Dl 17, and is thereforeunlikely to have such an effect upon it (Spezio et a!., 1993b). Certainly the catalytic prOpertiesof the D216A mutant of CenA.N are more consistent with an indirect role for this residue, assuggested by the structure of E2. It may also be that the structure for E2 changes when an actualsubstrate is bound, thus bringing D79 as close to Dl 17 as it is in CbhII. Further analysis mayreveal a role for this amino acid, which is likely to be important as it is conserved throughout thefamily.The pH profile of this mutant is similar to that of the wild-type enzyme, suggesting thatD216 has no direct affects on the pKaS of the general acid or base catalyst.4.1.6 Glutamate mutants.Mutation of D216, D252, D287 and D392 to glutamate decreases keat for all substrates.The extra methylene group could shift the carboxyl group so that it is no longer in a position tofunction correctly. It may also cause steric congestion as there may not be enough room in theenzyme to accommodate the larger side chain. This effect seems quite large for D252 and D392,which suggests that the positioning of the acid functionality of these groups is crucial foractivity. This is consistent with these amino acids being important for catalysis. Mutation ofputative catalytic aspartate and glutamate residues in soybean 8-amylase, an inverting enzyme, toglutamate and aspartate, respectively, completely eliminated activity (Totsuka et at., 1994).Mutation of catalytic glutamate residues in retaining enzymes to aspartates decreases activity100- to 2500-fold (MacLeod et at., 1994a; Navas and Béguin, 1993; Withers et al., 1992). In theretaining xylanase from B. pumilus, mutation of conserved aspartate residues to glutamate alsoresults in large decreases in activity of about 1500- to 11500-fold (Ko et at., 1992). Mutationfrom aspartate to glutamate would be expected to decrease activities more than from glutamate toaspartate due to the additional steno effects. The decrease in activity of D287E is quite small onall substrates, suggesting that this residue is not catalytically crucial or that there is more space inthis area of the active site to accommodate this change. The effect on D216 was difficult to87explain as mutation to glutamate resulted in very significant decreases in keat on both CM-cellulose and 2,4-DNPC, while mutation to alanine did not.4.2 Azide effects.Although the two mechanisms, inverting and retaining, utilized by glycosidases areseparate and distinct, there are some significant similarities. The first step of the retainingmechanism involves a proton transfer from the general acid catalyst to the glycosidic oxygen asthe glycosidic linkage is cleaved, thereby facilitating bond cleavage through stabilization of theleaving group. At the same time, a nucleophilic amino acid residue attacks the anomeric centerof the glucoside, forming a covalent bond. The inverting mechanism is very similar to the firststep of the retaining mechanism as it involves proton transfer from a general acid catalyst to theglycosidic oxygen. But, in the inverting mechanism, this carboxylate group is farther from theanomeric carbon, allowing room for a water molecule to enter. It therefore functions as a generalbase catalyst, removing a proton from water and producing a nucleophilic hydroxide ion whichattacks the anomeric carbon directly to form a covalent bond. Increasing thenucleophile/anomeric carbon distance by mutation of the nucleophilic glutamate to aspartate in8-glucosidase from A.faecalis did not cause a change in mechanism from retention to inversionof anomeric configuration (Withers et a!., 1992). Nor does increasing this distance by mutationof the general base aspartate to glutamate in CenA.N change the mechanism from inversion toretention of anomeric configuration (Figure 3.23). This inability to change reaction mechanismis likely a consequence of the distances being incorrect. The average separation betweencatalytic aspartates and glutamates in retaining 8-glycosidases is about 5.3 A while for invertingenzymes this distance is 9.5 A (Wang et al., 1994). Moving the carboxylate group by onemethylene results in a change in distance of about only 2 A while a change of at least 4.2 A maybe required. Furthermore, although the distance between the nucleophile and the anomericcarbon is crucial, other factors may well play a role in determining the stereochemical outcomeof the reaction.88The mechanism of hydrolysis by the B-glucosidase from A.faecalis can be changed fromretaining to inverting through the clever use of mutations and small anionic nucleophiles (Wanget al., 1994). The nucleophilic glutamate residue was eliminated by mutation to alanine,reducing activity by about 107-fold. However, addition of small anionic nucleophiles such asazide restores activity, with a-cellobiosyl azide being produced. Azide presumably substitutesfor the catalytic nucleophile, attacking the anomeric carbon directly to form the covalent product.This is the first example of a change in the catalytic mechanism of an enzyme, and it alsoconfirms that the glutamate residue is the catalytic nucleophile.Given the similarities in mechanism, this same technique was employed in an attempt toidentify the general base catalyst in CenA. Unfortunately, none of the CenA.N alanine mutantsproduced a-cellobiosyl azide, when reacted in the presence of azide. This may reflect afundamental difference between inverting and retaining enzymes. Because the distance from thegeneral base to the anomeric carbon is greater in inverting enzymes than in retaining enzymes, itmay be that the azide is held too far from the anomeric carbon to act as an effective nucleophile.The ability of azide to partially restore activity to D287A without formation of cellobiosyl azideindicates that the mutation creates enough space for azide to enter. The partial restoration ofactivity probably results from the negative charge introduced by the azide. A similar effect isseen in Cex from C.flmi (MacLeod, 1994b) when D123A is mutated to alanine. This residue isneither the acid/base nor the nucleophile (White et al., 1994), but it is located in close proximityto these residues.4.3 Hydrolysis of a- and 8-cellobiosyl fluoride.The kt for the hydrolysis of 8-cellobiosyl fluoride by CenA is 2.5-fold higher than for2,4-DNPC, while the KM value is about 3.5-fold higher. Although 2,4-DNPC does seem to bindslightly better than 8-ceilobiosyl fluoride, the use of this substrate confirms that the activityobserved on substrates having good leaving groups comes mostly from the reactivity of thesubstrate and not from increased binding in subsite C. a-Cellobiosyl fluoride is hydrolyzed byCenA.N with non-Michaelis-Menten type kinetics (Kasumi et at., 1987). This is contrary to89results for CbhII of family 6, which hydrolyzes a-cellobiosyl fluoride with typical MichaelisMenten type kinetics (Konstantinidis et at., 1993). This again may reflect fundamentaldifferences between exoglucanases and endoglucanases.In theory, these substrates could be used to differentiate between the acid and basecatalyst. The acid catalyst mutant, D252A, should hydrolyze 8-cellobiosyl fluoride atapproximately the same rate as the wild-type enzyme. However, it should hydrolyze acellobiosyl fluoride considerably more slowly, since hydrolysis by the Hehre mechanism requiresthis residue to deprotonate the 4-hydroxyl of the incoming a-cellobiosyl fluoride. Similarly, thebase catalyst mutant should not hydrolyze 8-cellobiosyl fluoride because a general base catalystwould be required to deprotonate water. However, it should hydrolyze a-cellobiosyl fluoride asthe presence of the base catalyst would not be required to protonate the departing fluoride, a verygood leaving group, while the acid catalyst would still be present to deprotonate the 4-hydroxylof the incoming a-cellobiosyl fluoride. Furthermore, once the intermediate a-cellotetraosylfluoride is formed, it should not be hydrolyzed efficiently because general base catalysis wouldbe required. a-Cellotetraosyl fluoride should be detectable in the reaction mixture.The acid catalyst mutant, D252A, did indeed hydrolyze 8-cellobiosyl fluoride at a rateonly 2-fold lower than that of the wild-type enzyme, much as observed with 2,4-DNPC.Furthermore, the rates of hydrolysis of a-cellobiosyl fluoride by this mutant were drasticallyreduced, as would be expected for an acid catalyst mutant. Both D216A and D287A hydrolyzed8-cellobiosyl fluoride with rates 46- and 16-fold lower than the wild-type enzyme, respectively.These values were also very similar to those found for 2,4-DNPC, as would be expected. WhileD287A was virtually inactive on a-cellobiosyl fluoride, D216A retained considerable activity.These results are similar to those found for the hydrolysis of CM-cellulose and further show thatD216 is not crucial for catalysis. Also, as expected for the base catalyst, D392A did nothydrolyze B-cellobiosyl fluoride. Unfortunately, no activity could be detected on a-cellobiosylfluoride either. Furthermore, no a-cellotetraosyl fluoride was detected in this reaction mixture orany of the other reaction mixtures for that matter. These results are in contrast to similar studiesfor 8-glucosidase from A.faecatis (Wang eta!., 1994). In that case, when the nucleophiicglutamate is replaced with alanine, the mutant enzyme hydrolyzes a-glucosyl fluoride with the90formation of a-cellobiosyl fluoride, an inverting type mechanism. Obviously, the mechanisms ofthese two enzymes are fundamentally different and observations made with retaining enzymesare not necessarily transferable to inverting enzymes.4.4 Conclusions.The use of substrates with different leaving groups in kinetic studies of selected mutants,along with pH dependence studies, allows elucidation of the roles of the conserved aspartateresidues in the active site of CenA.N. D252 is important for hydrolysis of substrates requiringacid catalysis but is not important for hydrolysis of substrates not requiring acid catalysis. ThepH profile of D252A indicates loss of the acid catalyst. D392 is important for hydrolysis of allsubstrates tested. Therefore, D252 and D392 are the acid and base catalyst, respectively. D287 isimportant for hydrolysis of substrates requiring acid catalysis but to a lesser degree than D252.This amino acid was not crucial for hydrolysis of substrates not requiring acid catalysis and thepH profile of D287A indicated a change in pKa of the acid and base catalyst-. D287 plays somerole in catalysis, probably by maintaining the correct environment in the active site for the acidand base catalyst to function. D216 is not crucial for catalysis on any of the substrates tested andis not likely to be directly involved in the catalytic mechanism.915. ReferencesAlberty, R.A. and Massey, V. (1954). On the interpretation of the pH variation of the maximuminitial velocity of an enzyme-catalyzed reaction. Biochim. Biophys. Acta, 13, 347-352.Al-Tawheed, A.R. (1988). Molecular Characterization of Cellulase Genes from Cellulomonasflavigena. M. Sc. Thesis, Trinity College, Dublin.Béguin, P., Eisen, H. and Roupas, A. (1977). Free and cellulose-bound cellulases in aCellulomonas species. 3. Gen. Microbiol. 101, 191-196.Béguin, P. and Aubert, 1.-P. (1994). The biological degradation of cellulose. FEMS Microbiol.Rev. 13, 25-58.Blackwell, J. (1982). The macromolecular organization of cellulose and chitin. In Cellulose andOther Natural Polymer Systems. Biogenesis, Structure, and Degradation, ed. R.M. Brown Jr.Plenum Press, New York; pp. 403-428.Braun, C., Meinke, A., Ziser, L. and Withers, S. (1993). Simultaneous high-performance liquidchromatograghic determination of both the cleavage pattern and stereocheniical outcome of thehydrolysis reactions catalysed by various glycosidases. Anal. Biochem. 212,259-262.Bray, M.R. and Clarke, A.J. (1990). Essential carboxy groups in xylanase A. Biochem. 3. 270,91-96.Bray, M.R. and Clarke, A.J. (1994). Identification of a glutamate residue at the active site ofxylanase A from Schizophyllum commune. Eur. J. Biochem. 219, 821-827.Briggs, G.E. and Haldane, J.B.S. (1925) Biochem. 1. 19, 338-339.Brocklehurst, K. (1994). A sound basis for pH-dependent kinetic studies on enzymes. Prot. Eng.7, 291-299.Chauvaux, S., Béguin, P. and Aubert, 3.-P. (1992). Site-directed mutagenesis of essentialcarboxylic residues in Clostridium thermocellum endoglucanase Ce1D. J. Biol. Chem. 267, 4472-4478.Chen, C.M., Gritzali, M. and Stafford, D.W. (1987). Nucleotide sequence and deduced primarystructure of cellobiohydrolase II from Trichoderma reesei. Bio/technol 5, 274-278.Claeyssens, M. and Henrissat, B. (1992). Specificity mapping of cellulolytic enzymes:classification into families of structurally related proteins confirmed by biochemical analysis.Prot. Sci. 1, 1293-1297.Comish-Bowden, A. (1981). Fundamentals of Enzyme Kinetics. Butterworth and Co.,Southampton, pp. 16-37.Coughian, M. (1990). Cellulose degradation by fungi. In Microbial Enzymes andBiotechnology, ed. W.M. Fogarty and C.T. Kelly. Elsevier Applied Science, London; pp. 1-36.Coutinho, J.B., Moser, B., Kilburn, D.G., Warren, R.A.J. and Miller, R.C., Jr. (1991). Nucleotidesequence of the endoglucanase C gene (cenC) of Cellulomonasfimi, its high-level expression inEscherichia coli, and characterization of its products. Molec. Microbiol. 5, 1221-1233.Damude, H.G., Gilkes, N.R., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J. (1993).Endoglucanase CasA from ailcalophilic Streptomyces strain KSM-9 is a typical member of familyB of 8-1,4-glucanases. Gene 123, 105-107.92Fersht, A. (1985) Enzyme Structure and Mechanism, 2nd ed., W.H. Freeman and Co., New York,pp. 98-120.Frandsen, T.P., Dupont, C., Lehmbeck, J., Stoffer, B., Seirks, M.R., Honzatko, R.B. andSvensson, B. (1994). Site-directed mutagenesis of the catalytic base glutamic acid 400 inglucoamylase from Aspergilius niger and of tyrosine 48 and glutamine 401, both hydrogen-bonded to the y-carboxylate gioup of glutamic acid 400. Biochemistry. 32, 13808-138 16.Gebler, 3., Gilkes, N.R., Claeyssens, M., Wilson, D.B., Béguin, P., Wakarchuk, W.W., Kilburn,D.G., Miller, R.C., Jr., Warren, R.A.J. and Withers, S.G. (1992). Stereoselective hydrolysiscatalyzed by related 13-1,4-glucanases and 13-1,4-xylanases. 3. Biol. Chem. 267, 12559-12561.Gilkes, N.R., Kilburn, D.G., Langsford, M.L., Miller, R.C., Jr., Wakarchuk, W.W., Warren,R.A.J., Whittle, D.J. and Wong, W.K.R. (1984a). Isolation and characterization of Escherichiacoil clones expressing cellulase genes from Celiulomonasfimi. J. Gen. Microbiol. 130, 1377-1384.Gillces, N.R., Langsford, M.L., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J. (1984b).Mode of action and substrate specifities of cellulases from cloned bacterial genes. J. Biol. Chem.259, 10455-10459.Gilkes, 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.Gilkes, N.R., Claeyssens, M., Aebersold, R., Henrissat, B., Meinke, A., Morrison, H.D., Kilburn,D.G., Warren, R.A.J. and Miller, R.C., Jr. (1991). Structural and functional relationships in twofamilies of B-1,4-glycanases. Eur. J. Biochem. 202, 367-377.Gillces, N.R., Jervis, E., Henrissat, B., Tekant, B., Miller, R.C., Jr., Warren, R.A.J. and Kilburn,D.G. (1992). The adsorption of a bacterial cellulase and its two isolated domains to crystallinecellulose. J. Biol. Chem. 267, 6743-6749.Guo, S., Arfman, N., Ong, E., Gilkes, N.R., Kilburn, D.G., Warren, R.A.J., and Miller, R.C., Jr.(1988). Leakage of Ceilulomonasfimi cellulases from Escherichia coil. FEMS Microbiol. Lett.49, 279-283.Hanahan, D. (1983). Studies on transformation of Escherichia coil with plasmids. J. Mol. Biol.166, 557-580.Hehre, E.J., Brewer, C.F. and Genghof, D.S. (1979). Scope and mechanism of carbohydraseaction. J. Biol. Chem. 254, 5942-5950.Henrissat, B. and Bairoch, A. (1993). New families in the classification of glycosyl hydrolasesbased on amino acid sequence similarities. Biochem. J. 293, 78 1-788.Hoj, P.B., Rodriguez, E.B., Iser, J.R., Stick, R.V. and Stone, B.A. (1991). Active site-directedinhibition by optically pure epoxyallcyl cellobiosides reveals differences in active site geometry oftwo 1,3-1,4-B-D-glucan 4-glucanohydrolases. J. Biol. Chem. 266, 11628-11631.Hoj, P.B., Condron, R., Traeger, J.C., McAuliffe, J.C. and Stone, B.A. (1992). Identification ofglutamic acid 105 at the active site of Bacillus amylo1iqufaciens 1,3-1,4-13-D-glucan 4-glucanohydrolase using epoxide-based inhibitors. 3. Biol. Chem. 267, 25059-25066.93Jeffries, T.W., Patel, R.N., Sykes, M.S. and Klungness, J. (1992). Enzymatic solutions toenhance bonding, bleaching and contaminant removal. In Material Interactions Relevant toRecycling of Wood-based Materials, 266, ed. R.M. Rowell, T.L. Laufenberg and J.K. Rowell.Materials Research Society, Pittsburg, pp. 277-287.Jeffries, T.W., Klungness, J., Sykes, M.S. and Rutledge-Cropsey, R.R. (1994). Comparison ofenzyme-enhanced with conventional deinking of xerographic and laser-printed paper. Tappi J.77, 173-179.Johnson, K.A. and Benkovic, S.J. (1990). Analysis of protein function by mutagenesis. In TheEnzymes: Mechanisms of Catalysis, 19, ed. D.S. Sigman and P.D. Boyer. Academic Press,Toronto.Kasumi, T., Tsumuray, Y., Brewer, C.F., Kersters-Hilderson, H., Claeyssens, M. and Hehre, E.J.(1987). Catalytic versatility of Bacillus pumilis 13-xylosidase: Glycosyl transfer and hydrolysispromoted with a- and B-D-xylosyl fluoride. Biochemistry. 26, 30 10-3016.Kempton, J.B. and Withers, S.G. (1992). Mechanism of Agrobacterium 8-glucosidase: kineticstudies. Biochemistry. 31, 9961-9969.Keresztessy, Z., Kiss, L. and Hughes, M.A. (1994). Investigation of the active site of thecyanogenic B-D-glucosidase (Linamarase) from Manihot esculenta Crantz (Cassava). H.Identification of Glu-198 as an active site carboxylate group with acid catalytic function. Arch.Biochem. Biophys. 315, 323-330.Ko, E.P., Akatsuka, H., Moriyama, H., Shinmyo, A., Hata, Y., Katsube, Y., Urabe, I. and Okada,H. (1992). Site-directed mutagenesis at aspartate and glutamate residues of xylanase fromBacillus pwnilus. Biochem. J. 288, 117-121.Konstantinidis, A.K., Marsden, I. and Sinnott, M.L. (1993). Hydrolyses of a— and B—cellobiosylfluorides by cellobiohydrolases of Trichoderma reesei. Biochem. J. 291, 833-838.Koshland, D.E (1953). Stereochemistry and the mechanism of enzymatic reactions. Biol. Rev.28,416-436.Kunkel, T.A., Roberts, J.D. and Zakour, R.A. (1987). Rapid and efficient site-specificmutagenesis without phenotypic selection. Methods Enzymol. 154, 367-382.Laemmli, U.K. (1970). Cleavage of structural proteins during the assembly of the head ofbacteriophage T4. Nature 227,680-685.Lange, N.K. (1993). Applications of cellulases in the textile industry. In Trichoderma reeseiCellulases and Other Hydrolases, ed. P. Suominen and T. Reinilcainen. Foundation forBiotechnical and Industrial Fermentation, Helsinki; pp. 263-272.Langsford, M.L., Gilkes, N.R., Wakarchuk, W.W., Kilbum, D.G., Miller, R.C., Jr. and Warren,R.A.J. (1984). The cellulase system of Cellulomonasfimi. J. Gen. Microbiol. 130, 1367-1376.Langsford, M.L., Gilkes, N.R., Singh, B., Moser, B., Miller, R.C., Jr., Warren, R.A.J. andKilbum, D.G. (1987). Glycosylation of bacterial cellulases prevents proteolytic cleavage betweenfunctional domains. FEBS Lett. 225, 163-167.Langsford, M.L. (1988). The purification and characterization of two cellulose-binding,glycosylated cellulases from the bacterium Cellulomonasfimi. Ph.D. thesis, University ofBritish Columbia.94Lin, F., Marchenko, G. and Cheng, Y. (1994). Cloning and sequencing of an endo-B-1,4-glucanase gene mcenA from Micromonospora cellulolyticuni 86W-16. J. of md. Microbiol. 13,344-350.Macarron, R., van Beeumen, J., Henrissat, B., de la Mata, I. and Claeyssens, M. (1993).Identification of an essential glutamate residue in the active site of endoglucanase Ill fromTrichoderma reesei. FEBS Lett. 316, 137-140.MacLeod, A.M., Lindhorst, T., Withers, S.G. and Warren, R.A.J. (1994a). The acid/base catalystin the exoglucanase/xylanase (Cex) from Cellulomonasfimi is Glu 127 : evidence from detailedkinetic studies of mutants. Biochemistry 33, 6371-6376.MacLeod, A.M. (1994b). Structure-function relationships in a B-1,4-glycanase (Cex) fromCellulomonasfimi: Identification of catalytic residues. Ph.D. thesis, University of BritishColumbia.Mead, D.A. Szczesha-Skorupa, E., and Kemper, B. (1986). Single-stranded DNA ‘blue’ T7promoter plasmids: a versatile tandem promoter system for cloning and protein engineering.Protein Eng. 1: 67-7 1.Meinke, A., Gilkes, N.R., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J. (1993). Cellulose-binding polypeptides from Cellulomonasfimi endoglucanase D (CenD), a family A 8-1,4-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 Cellulomonasflmi is a 13-1,4-exocellobiohydrolase analogous to Trichoderma reesei CBHII. Mol. Microbiol. 12,414-422.Miao, S., Ziser, L., Aebersold, R. and Withers, S.G. (1994). Identification of glutamic acid 78 asthe active site nucleophile in Bacillus subtilis xylanase using electrospray tandem massspectroscopy. Biochemistry 33,7027-7032.Miller, G.L., Blum, R., Glennon, W.E. and Burton, A.L. (1960). Measurement ofcarboxymethylcellulase activty. Anal. Biochem. 1, 127-132.Navas, J. and Béguin, P. (1993). Site-directed mutagenesis of conserved residues of Clostridiumthermocellum endoglucanase CelC. Biochem. Biophys. Res. Commun. 189, 807-8 12.Nitta, Y., Isoda, Y., Toda, H. and Sakiyama, F. (1989). Identification of glutamic acid 186affinity-labeled by 2,3-epoxypropyl cz-D-glucopyranoside in soybean 13-amylase. J. Biochem.105, 573-576.Nordquist, D. (1992). Characterization of double binding domain derivatives of CenA fromCellulornonasfimi. M.Sc. thesis, University of British Columbia.O’Neill, G., Goh, S.H., Warren, R.A.J., Kilburn, D.G. and Miller, R.C., Jr. (1986). Structure ofthe gene encoding the exoglucanase of Cellulomonosflmi. Gene 44, 325-330.Owolabi, J.B., Béguin, P., Kilburn, D.G., Miller, R.C., Jr. and Warren, R.A.J. (1988). Expressionin Escherichia coli of the Cellulomonasfimi structural gene for endoglucanase B. Appi. Environ.Microbiol. 54, 5 18-523.Peller, L. and Alberty, R.A. (1959). Multiple intermediates in steady state enzyme kinetics. Themechanism involving a single substrate and product. J. Am. Chem. Soc. 81, 5907-59 10.Rouvinen, J., Bergfors, T., Teeti, T., Knowles, J.K. and Jones, T.A. (1990). Three-dimensionalstructure of cellobiohydrolase II from Trichoderma reesei. Science 249, 380-386.95Saddler, J.N. (1993). Bioconversion ofForest and Agricultural Plant Residues. CABInternational, Oxford.Sambrook, 3., Fritsch, E.F. and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual,2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.Schou, C., Rasmussen, G., Kaltoft, M.B., Henrissat, B. and Schülein, M. (1993).Stereochemistry, specificity and kinetics of the hydrolysis of reduced cellodextrins by ninecellulases. Eur. 3. Biochem. 217, 947-953.VSierks, M.R., Ford, C., Reilly, P.3. and Svensson, B. (1990). Catalytic mechanism of fungalglucoamylase as defined by muagenesis of Asp 176, Glu 179 and Glu 180 in the enzyme fromAspergillus awamori. Protein Eng. 3, 193-198.Sierks, M.R. and Svensson, B. (1993). Functional roles of the invariant apstartic acid 55, tyrosine306 and aspartic acid 309 in Glucoamylase from Aspergillus awamori studied by mutagenesis.Biochem. 32, 1113-1117.Sinnott, M.L. (1990). Catalytic mechanisms of enzymic glycosyl transfer. Chem. Rev. 90, 1171-1202.Sogaard, M., Kadziola, A., Haser, R. and Svensson, B. (1993). Site-directed mutagenesis ofhistidine 93, aspartic acid 180, glutamic acid 205, histidine 290, and aspartic acid 291 at theactive site and tryptophan 279 at the raw starch binding site in barley a-amylase 1. 3. Biol. Chem.268, 22480-22484.Spezio, M., Irwin, D., Karplus, P.A., Zhang, S., Taylor, 3. and Wilson, D.B. (1993a). Structureand function of Thermomonosporafusca endoglucanase E2. In Genetics, Biochemistry andEcology of Lignocellulose Degradation, ed. K. Shimada et al. Uni Publishers, Tokyo; pp. 308-317.VSpezio, M., Wilson, D.B. and Karplus, P.A. (1993b). Crystal structure of the catalytic domain ofa thermophilic endocellulase. Biochemistry 32, 9906-9916.Svensson, B., Clarke, A.J., Svendsen, I. and Moller, H. (1990). Identification of carboxylic acidresidues in glucoamylase G2 from Aspergillus niger that participate in catalysis and substratebinding. Eur. J. Biochem. 188, 29-38.Svensson, B. and Sogaard, M. (1993). Mutational analysis of glycosylase function. J.Biotechnol. 29, 1-37.Tabor, S. and Richardson, C.C. (1987). DNA sequence analysis with a modified bacteriophageT7 DNA polymerase. Proc. Natl. Acad. Sci. U.S.A 84, 4767.Tempelaars, C., Birch, P.R.J., Sims, P.F.G. and Broda, P. (1994). Isolation, characterization, andanalysis of the expression of the cbhll gene of Phanerochaete chrysosporium. Appl. Environ.Microbiol. 60, 4387-4393.Tomme, P., Van Tilbeurgh, H., Pettersson, 0., Van Damme, 3., Vandekerckhove, J., Knowles, J.,Teeri, T. and Claeyssens, M. (1988). Studies on the cellulolytic system of Trichodenna reeseiQM 9414. Analysis of domain function of two cellobiohydrolases by limited proteolysis. Eur. 3.Biochem. 170, 575-58 1.Tomme, P., van Beeumen, 3. and Claeyssens, M. (1992). Modification of catalytically importantcarboxyl residues in endoglucanase D from Clostridium thermocellum. Biochem. J. 285, 319-324.96Tomme, P., Warren, R.A.J. and Gilkes, N.R. (1995). Cellulose hydrolysis by bacteria and fungi.Adv. Microbial Physiol. in press.Totsuka, A. and Fukazawa, C. (1993). Expression and mutation soybean 8-amylase inEscherichia coli. Eur. J. Biochem. 214,787-794.Totsuka, A., Nong, V.H., Kadokawa, H., Kim, C., Itoh, Y. and Fukazawa, C. (1994). Residuesessential for catalytic activity of soybean B-amylase. Eur. 3. Biochem. 221, 649-654.Trimbur, D.E., Warren, R.A.J. and Withers, S.G. (1992). Region-directed mutagenesis ofresidues suurounding the active site nucleophile in 13-glucosidase from Agrobacteriumfaecalis.J.Biol. Chem. 267, 10248-10251.Tull, D., Withers, S.G., Gillces, N.R., Kilburn, D.G., Warren, R.A.J. and Aebersold, R. (1991).Glutamic Acid 274 is the nucleophile in the active site of a “retaining” exoglucanase fromCellulomonasfimi. J. Biol. Chem. 266, 15621-15625.Tull, D. and Withers, S.G. (1994). Mechanisms of cellulases and xylanases: a detailed kineticstudy of the exo-B- 1 ,4-glycanase from CellulomonasfImi. Biochemistry 33, 6363-6370.Van Nifterik, L., Xu, 3., Laurent, J.L. and Mathieu, J. (1993). Analysis of cellulose and kraft pulpozonolysis products by anion-exchange chromatography with pulsed amperometric detection. J.Chromatog. 640, 335-343.Van Tilbeurgh, H., Tomme, P., Claeyssens, M., Bhikhabhai, R. and G., p. (1986). Limitedproteolysis of the cellobiohydrolase I from Trichoderma reesei. FEBS Lett. 204, 223-227.Van Tilbeurgh, H., Loontiens, F.G., Engelborgs, Y. and Claeyssens, M. (1989). Studies of thecellulolytic system of Trichoderma reesei QM 9414. Eur. 3. Biochem. 184, 553-559.Viera, 3. and Messing, J. (1987). Production of single stranded plasmid DNA. MethodsEnzymol. 160, 3-11.Wang, Q., Graham, R.W., Trimbur, D., Warren, R.A.J. and Withers, S.G. (1994). Changingenzymatic reaction mechanisms by mutagenesis: conversion of a retaining glucosidase to aninverting enzyme. J. Am. Chem. Soc. 116, 11594-11595.White, A., Withers, S.G., Gillces, N.R. and Rose, D.R. (1994) Crystal structure of the catalyticdomain of the B-1,4-glycanase Cex from Cellulomonasflmi. Biochemistry 33, 12546-12552.Whittle, D.J., Kilburn, D.G., Warren, R.A.J. and Miller, R.C., Jr. (1982). Molecular cloning of aCellulomonasfimi cellulase gene in Escherichia coli. Gene 17, 139-145.Withers, S.G., Dombroski, D., Berven, L.A., Kilburn, D.G., Miller, R.C., Jr., Warren, R.A.J. andGilkes, N.R. (1986). Direct 1H N.M.R. determination of the stereochemical course of hydrolysescatalyzed by glucanase components of the cellulase complex. Biochem. Biophys. Res. Commun.139,487-494.Withers, S.G., Street, I.P. and Percival, M.D. (1988). Fluorinated carbohydrates as probes ofenzyme specificity and mechanism. In ACS Symposium Series, Vol. 3’l4pp. 59-77.Withers, S.G. and Street, I.P. (1988). Identification of a covalent a-D-glucopyranosyl enzymeintermediate formed on a 8-glucosidase. 3. Am. Chem. Soc. 110, 855 1-8553.Withers, S.G. and Street, I.P. (1989). B-glucosidases: Mechanism and inhibition. In ACSSymposium Series, Vol. 399 pp. 597-607.97Withers, S.G., Warren, R.A.J., Street, I.P., Rupitz, K., Kempton, J.B. and Aebersold, R. (1990).Unequivocal demonstration of the involvement of a glutamate residue as a nucleophile in themechanism of a retaining glycosidase. J. Am. Chem. Soc. 112, 5887-5889.Withers, S.G., Rupitz, K., Trimbur, D.E. and Warren, R.A.J. (1992). Mechanistic consequencesof mutation of the active site nucleophile G1u358 in Agrobacterium 8-glucosidase. Biochemistry31, 9979-9985.Wong, W.K.R., Gerhard, B., Guo, Z.M., Kilburn, D.G., Warren, R.A.J. and Miller, R.C., Jr.(1986). Characterization and structure of an endoglucanase gene cenA of Cellulomonasfimi.Gene 44, 3 15-324.Wood, T.M. (1988). Preparation of crystalline, amorphous and dyed cellulase substrates. Meth.Enzymol. 160, 19-25.Yanisch-Perron, C., Vieira, J. and Messing, J. (1985). Improved M13 phage cloning vectors andhost strains: nucleotide sequences of the Ml3mp and pUC vectors. Gene 33, 103-119.Zvelebil, M.J. and Sternberg, M.J.E. (1988). Analysis and prediction of the location of catalyticresidues in enzymes. Protein Eng. 2, 127-138.986. Appendix6.1 Basic concepts of enzyme kinetics.The mechanism for hydrolysis of 8-1,4-glucosides by CenA is by inversion of anomericconfiguration (Figure 1.1). This mechanism involves the initial binding of substrate to CenAfollowed by hydrolysis involving a single displacement by water with release of products.k1 k2E÷S<>ES >E÷Pk1The rate equation for this mechanism is the same as the generalized mechanism derivedby Briggs and Haldane (1925). In this derivation, three assumptions are made. Initial rates aremeasured so that the concentration of product is assumed to be negligible, the enzymeconcentration is assumed to be negligible compared to that of the substrate and it is assumed thata steady state would be reached in which the concentration of intennediate (ES) was constant,ie. d(ES)/dt = 0. These assumptions are generally quite reasonable (Comish-Bowden, 1981).V=kcat[EJ [SI kcat=k2KM+SKM = k2 + k..1k1Vmax = ‘<cat [E0]The constant kt is often called the turnover number of the enzyme because it representsthe maximum number of substrate molecules converted to products per active site per unit time,or the number of times the enzyme “turns over” per unit time (Fersht, A, 1985). With CenA,when all binding steps are fast, kcat is simply the first-order rate constant for the chemicalconversion of the ES complex to the EP complex (ie. = k2).99The constant KM is the substrate concentration at which the rate is half maximal(v = Vm /2). For the simple Michaelis-Menten mechanism, KM = K, the true dissociationconstant of the enzyme-substrate complex. For some purposes, KM may be treated as anapparent dissociation constant where the concentration of free enzyme in solution may becalculated form the relationship KM = [E][Sj I 2[ES].The importance of kcat/KM is that it relates the reaction rate to the concentration of free,rather than total, enzyme. At low substrate concentration, the enzyme is largely unbound and[E}=[E0j. At such concentrations, the reaction rate is given by v = [E][S] (ktfKJ andkcat/KM is an apparent second-order rate constant which relates the reaction rate to theconcentration of the free enzyme and free substrate. Because ‘cat/”M is a measure of thecatalytic efficiency of the enzyme and defines the specificity for competing substrates, it is alsoreferred to as the specificity constant.1006.2 Example calculation for the determination of kcat and KM.In this example, keat and KM were determined by fitting the Michaelis-Menten equationdirectly to the data shown in Table 6.1 using the program GraFit 3.0. These data were obtainedfor the initial rates of hydrolysis of 2,4-DNPC by wild-type CenA.N (0.46 tM) as described insection 2.11.3.Table 6.1 Hydrolysis of 2,4-DNPC by wild-type CenA.N (0.46 i.IM).S0 (pM) V0 (AOD/min)30.5 0.01691.6 0.039109.9 0.044152.6 0.053213.7 0.063305.2 0.07 1549.4 0.085769.2 0.0911098.9 0.096The KM (I.IM) arid (AOD/min) values are shown in Figure 6.1. The rate ofproduction of 2,4-DNP (Vm) (I.LM/min) was calculated in the following way:AC/mm = (AOD/min) x 1000 / (e. 1)where: = extinction coefficient ( = cell path length (cm)AC/mm = change in concentration per minute (pM/min)1010. 200 400 600 800 1000 1200[2,4-DNPC] (jiM)Enzyme KineticsSimple weightingReduced Chi squared = 3.304e-007Variable Value Std. Err.V max 0.1108 0.0006 —_________Km 167.7320 2.9176I ‘ I ‘ I ‘ I ‘ II I i I i I I i IFigure 6.1 Determination of kcat and KM values.102Using a cell of pathlength 1 cm and assuming an extinction coefficient for 2,4-DNP (pH7.0) of 10.9,the rate of production of 2,4-DNP (Vm) is 10.2 tM/min.The kcat value is calculated in the following way:kcatVm/Eowhere: Vm = calculated initial rate of hydrolysisE0 = concentration of wild-type CenASo in this example, k is 22 mm4 and KM is 170 I.LM.6.3 Graphical representation of kinetic data.6.3.1 Hydrolysis of cellotriose and cellotetraose by wild-type CenA.N.2C11G4 (pM1)0 0.001 0.002 0.003 0.004 0.00511G3 (pM:1)Figure 6.3 The Lineweaver-Burke plot for the hydrolysis of cellotriose (G3).Enzyme = 0.16 pM1030.012 0.016 0.008Figure 6.2 The lineweaver-Burk plot for the hydrolysisEnzyme=1.1 xlO4pM1. cellotetraose (G4).6.32 Hydrolysis of nitrophenyl cellobiosides by CenA .N and mutants.60.4020260240220.9 20014012010080601/[2,4-DNPC] (pM1)1/[2,4-DNPC] (pM1)Figure 6.5 The Lineweaver-Burke plot for the hydrolysisEnzyme = 1.6 pM; = 400 nm; & = 10.9 mM1 cm-10.041040 0.01 0.02 0.03Figure 6.4 TheEnzyme = 0.46Lineweaver-Burke plot for the hydrolysis of 2,4-DNPC.pM; = 400 nm; & = 10.9 mM cm0 0.002 0.004 0.006 0.008 0.01of 2,4-DNPC by D216A.- 40CE20Figure 6.7 The Lineweaver-Burke plot for the hydrolysisEnzyme =0.36 pM; = 400 nm; Ae = 10.9 mM’ cm1of 2,4-DNPC by D252A.10516001400120011;::6004000.006 0.012 0.018 0.0241/[2,4-DNPC] (pM1)Figure 6.6 The Lineweaver-Burke plot for the hydrolysis of 2,4-DNPC by D21 6E.Enzyme = 5.2 pM; = 400 nm; ze = 10.9 mM1 cm1600.030 0.04 0.08 0.12 0.16 0.2 0.241/[2,4-DNPC] (pM1)106c6004002000 0.02 0.04 0.06 0.08 0.11/[2,4-DNPC] (IJM1)Figure 6.8 The Lineweaver-Burke plot for the hydrolysis of 2,4-DNPC by D252E.Enzyme = 0.59 pM; A. = 400 nm; & = 10.9 mM1 cm1200180160E12010080600 0.041/[2,4-DNPC] (pM1)Figure 6.9 The Lineweaver-Burke plot for the hydrolysis of 2,4-DNPC by D287A.Enzyme = 3.5 pM; A. = 400 nm; AE = 10.9 mM1 cm10.01 0.02 0.03C1070 0.02 0.048060C>201/[2,4-DNPC] (1JM)Figure 6.10 The Lineweaver-Burke plot for the hydrolysis of 2,4-DNPC by D287E.Enzyme = 0.83 pM; ?. = 400 nm; & = 10.9 mM1 cm1800060004000200000 0.005 0.015 0.021/[2,5-DNPC] (pM’)Figure 6.11 The. Lineweaver-Burke plot for the hydrolysis of 2,5-DNPC by wild-typeCenA.N. Enzyme =0.5 pM; = 440 nm; & = 4.3 mM1 cm10.01CEC21080.005 0.01 0.015500040003000200010000 0.021/[3,4-DNPC] (iJM1)Figure 6.2 The Lineweaver-Burke plot for the hydrolysis of 3,4-DNPC by wild-typeCenA.N. Enzyme = 1.8 pM; ). = 400 nm; Ae = 11.0 mM1 cm16000400020000.002 0.004 0.006 0.008 0.01 0.0121/[pNPC] (pM1)Figure 6.13 The Lineweaver-Burke plot for the hydrolysis of pNPC by wild-typeCenA.N. Enzyme = 2.73 pM; 2 = 400 nm; .e = 7.2 mM1 cm1CE1096.33 Hydrolysis ofJ3-cellobiosylfluoride.14121086420.005 0.01 0.015 0.021 I[13-cellobiosyl fluo ride] (jiM’)Figure 6.14 The Lineweaver-Burke plot for the hydrolysis of f-ceIlobiosyI fluoride.Enzyme=1.lpM


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