Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An investigation of the activity and stability of caldocellum saccharolyticum β-glucosidase Stevenson, Andrew D. 1994

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

Item Metadata

Download

Media
831-ubc_1995-0415.pdf [ 3.99MB ]
Metadata
JSON: 831-1.0059593.json
JSON-LD: 831-1.0059593-ld.json
RDF/XML (Pretty): 831-1.0059593-rdf.xml
RDF/JSON: 831-1.0059593-rdf.json
Turtle: 831-1.0059593-turtle.txt
N-Triples: 831-1.0059593-rdf-ntriples.txt
Original Record: 831-1.0059593-source.json
Full Text
831-1.0059593-fulltext.txt
Citation
831-1.0059593.ris

Full Text

AN INVESTIGATION OF T H E ACTIVITY AND STABILITY OF CALDOCELLUM SACCHAROLYTICVM p-GLUCOSDDASE By ANDREW D. STEVENSON B.Sc, University of Toronto, 1989 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July 1994 © Andrew D. Stevenson, 1994 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of (^K<Soo \ & The University of British Columbia Vancouver, Canada Date DE-6 (2/88) 11 ABSTRACT The activity and stability of the thermostable Caldocellum saccharolyticum ji-glucosidase has been investigated and the properties compared to the fJ-glucosidase derived from the mesophile Agrobacterium faecalis, with which there exists a 50% sequence similarity. Previous work on the Agrobacterium P-glucosidase has indicated that the enzyme operates via a two-step mechanism, consisting of initial cleavage of the glycosidic bond with formation of a covalent glycosyl-enzyme intermediate (glycosylation), followed by hydrolysis of this intermediate to yield free enzyme and glycose (deglycosylation). The high degree of sequence similarity between the Agrobacterium and Caldocellum enzymes, combined with the kinetic data determined herein, strongly suggests that the Caldocellum jJ-glucosidase operates through the same mechanism. Values of k ^ and Kj^ were determined for enzymic hydrolysis of twelve substituted phenyl-P-D-glucopyranosides with leaving group pKa's ranging from 5.49 to 10.32. Values of log kcat shown no significant dependence upon leaving group pK S ) indicative of rate-limiting deglycosylation with all substrates except those with very poor leaving groups (pKa > 9.5). Substrates with poor leaving groups show a Pig of -0.2 suggesting very little cleavage of the glycosidic bond in the glycosylation transtion state. This contrasts with the Agrobacterium (3-glucosidase which shows a fiig of -0.7, indicating significant cleavage of the glycosidic bond in the transition state. There exists a good correlation between the Agrobacterium and Caldocellum P-glucosidase log kcat^M v a l u e s f ° r these substrates, indicative of similarities in the enzymic binding site. Values of kc a t and K \ j were determined for enzymic hydrolysis of six 4-nitrophenyl glycopyranosides. A reasonable correlation of the log kc a t /KM f ° r m e glycoside hydrolysis exists between the two enzymes, further indicating a similar binding site structure. An activation energy barrier of 7 kcal moi "1 was determined for the Caldocellum enzyme, this comparing reasonably well with a value of 9 kcal mol'l for the Agrobacterium ($-glucosidase, again indicating a similar conformation of the two enzymes. I l l The Caldocellum f3-glucosidase is significantly more stable to both thermal and chemical denaturation than the Agrobacterium enzyme, by at least 3 kcal mol'l. The stability of either enzyme was found to increase when trapped as the 2-fluoroglycosyl-enzyme intermediate, this increase in stability relating to the strength of the interactions with the substrate; again this increase in stability was at least 3 kcal moH. The Caldocellum enzyme's higher content of hydrophobic amino acids and a lower net charge, apparent from a comparison of the sequence of the two enzymes, are believed to be responsible for its greater stability. iv T A B L E OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF FIGURES vii LIST OF TABLES xi LIST OF SCHEMES '. xii LIST OF ABBREVIATIONS AND DEFINITIONS xni ACKNOWLEDGEMENT xv CHAPTER I: INTRODUCTION 1. Glycosidases and Glycoside Hydrolysis 1 2. ThermophUicity and Enzyme Thermostability 9 2.1 Factors Affecting Thermal Stability of Proteins 11 2.2 Attempts to Alter Thermostability of Proteins 13 3. Agrobacterium and Caldocellum P-Glucosidase 15 4. The Aims of This Study 17 CHAPTER II: RESULTS AND DISCUSSION 1. Synthesis 18 2. Extinction Coefficients 18 3. Kinetics 18 4. Aryl Glucoside Kinetics 20 5. 4-Nitrophenyl-Glycoside Kinetics 26 6. pH and Temperature Stability 30 7. Dependence of Rate upon pH and Temperature 31 8. Cysteine Titration 34 9. Chemical Denaturation Stability 35 10. Conclusions 40 CHAPTER HI: MATERIALS AND METHODS 1. Synthesis 42 1.1 General Procedures and Methods 42 1.2 General Compounds : 43 2. Enzymology 47 2.1 General Procedures 47 2.2 Enzyme Isolation and Purification 47 2.3 Measurement of Enzyme Concentration 48 2.4 Extinction Coefficients 49 2.5 Kinetic Measurements '. 49 2.6 Cysteine Titration 50 2.7 pH Dependent Stability and Activity 51 2.8 Temperature Dependent Stability and Activity 51 2.9 Circular Dichroism Measurements 52 APPENDIX I: BASIC CONCEPTS OF ENZYME CATALYSIS 1. Fundamental Equations of Enzyme Kinetics 54 2. Transition State Stabilization and Enzyme Catalysis 55 APPENDIX II: THE USE OF CIRCULAR DICHROISM IN THE STUDY OF PROTEINS 1. General Principles of Circular Dichroism Spectropolarimetry 58 2. Application of Circular Dichroism to the Study of Proteins 60 V I APPENDIX HI: GRAPHICAL REPRESENTATION OF KINETIC DATA 1. Lineweaver-Burk Plots for Hydrolysis of Aryl Glycosides by Caldocellum p-Glucosidase at 50°C in pH 6.5 MES Buffer 62 2. Lineweaver-Burk Plots for Hydrolysis of 4-Nitrophenyl-fJ-D-Glycopyranosides by Caldocellum pVGlucosidase at 50°C 71 3. Lineweaver-Burk Plots for Hydrolysis of 4-Nitrophenyl-fJ-D-Glycopyranosides by Caldocellum |3-Glucosidase in pH 6.5 MES Buffer... 76 APPENDIX TV: SEQUENCE ALIGNMENT 80 REFERENCES 83 vii LIST OF FIGURES Figure 1: Structural and electronic similarities of glucono-l,5-lactone to the oxocarbonium ion-like glycosyl cation . 7 Figure 2: Lineweaver-Burk plots for the hydrolysis of p-nitrophenyl-P-D-xyloside (a) and p-nitrophenyl-a-L-arabinoside (b) 21 Figure 3: Bronsted plot relating log (kcat) ^ d leaving group ability (pKa) for hydrolysis of aryl glucosides by Caldocellum (o) and Agrobacterium (A) P-glucosidase 24 Figure 4: Bronsted plot relating log (k^/KM) and leaving group ability (pKa) for hydrolysis of aryl glucosides by Caldocellum (o) and Agrobacterium (A) P-glucosidase : 25 Figure 5: Plot of log k ^ / K M values measured for the aryl glucoside series with Agrobacterium P-glucosidase versus the corresponding parameter measured for the Caldocellum enzyme 26 Figure 6: Glycone moeities used; R is a 4-nitrophenyl group 27 Figure 7: Plot of log kc a t/Kj^ values measured for the 4-nitrophenyl-glycoside series with Agrobacterium P-glucosidase (o) and Sulfolobus P-galactosidase (A) versus the corresponding parameter measured for the Caldocellum enzyme. Agrobacterium values measured for glycosides with axial C-4 hydroxyl groups are shaded 29 Figure 8: Plot of logarithim of the rate of activity loss Gog k) against pH for the Caldocellum P-glucosidase at 50°C (o) and 70°C(A) 30 Figure 9: Plot of the half-life for thermally induced activity loss against temperature for the Caldocellum (o) and Agrobacterium (A) enzymes 31 Figure 10: pH dependence of the hydrolysis of 4-nitrophenyl-P-D-glucoside by Caldocellum P-glucosidase. (a) Plot of log k ^ vs. pH. (b) Plot of log kc a t /KM vs. pH 32 VU1 Figure 11: Arrhenius plots relating rates of Caldocellum (o) and Agrobacterium(&) (5-glucosidase catalyzed hydrolysis of 4-nitrophenyl-p,-D-glucoside with temperature, (a) Plot of log vs. 1/T. (b) Plot of log kgat^M v s 34 Figure 12: Urea denaturation curves for Caldocellum (A) and Agrobacterium (o) [J-glucosidases and the Agrobacterium (J-glucosidase with the 2-fluoro-2 -deoxyglycosyl moeity bound in the active site (•) at 41 °C. The relative ellipticity at 222 nm is plotted as a function of denaturant concentration. The ellipticities in the absence of urea were defined as 100% 37 Figure 13: Urea denaturation curves for Caldocellum (5-glucosidase (o) and the Caldocellum fi-glucosidase with the 2-fluoro-2-deoxyglycosyl moeity bound in the active site (•) at 79°C. The relative ellipticity at 222 nm is plotted as a function of denaturant concentration. The ellipticities in the absence of urea were defined as 100% 38 Figure 14: Free energy diagram for a typical enzymic reaction (solid line) and its corresponding uncatalyzaed reaction (dashed line) 56 Figure 15: Free energy diagram of an enzymic reaction involving: (a) Maximum enzyme-substrate complemtarity at the ground state; Ob) Maximum enzyme-substrate complementarity at the transition state 57 Figure 16: Electric vectors of a right circularly polarized light at constant time and variable distance from the light source 58 Figure 17: The electric vectors of the left- and right-handed circularly polarized light sum to an electric vector which traces an elliptical path upon passing through an optically active medium in which the left- and right-handed circularly polarized light are unequally absorbed 59 Figure 18: Hydrolysis of 2-chloro-4-nitrophenyl-p>-D-glucopyranoside 62 Figure 19: Hydrolysis of 4-chloro-2-nitrophenyl-fi-D-glucopyranoside 62 ix Figure 20: Hydrolysis of 4-nitrophenyl-P-D-glucopyranoside 63 Figure 21: Hydrolysis of 3,5-dichlorophenyl-p,-D-glucopyranoside 63 Figure 22: Hydrolysis of 3-nitrophenyl-|}-D-glucopyranoside 64 Figure 23: Hydrolysis of 3-cyanophenyl-fJ-D-glucopyranoside 64 Figure 24: Hydrolysis of 3-chlorophenyl-pVD-glucopyranoside 65 Figure 25: Hydrolysis of 4-bromophenyl-P-D-glucopyranoside 65 Figure 26: Hydrolysis of 2-napthylphenyl-fi-D-gluc6pyranoside 66 Figure 27: Hydrolysis of phenyl-fJ-D-glucopyranoside 66 Figure 28: Hydrolysis of 4-methoxyphenyl-pVD-glucopyranoside 67 Figure 29: Hydrolysis of 3,4-dimethyl-fi-D-glucopyranoside 67 Figure 30: Hydrolysis of 4-nitrophenyl-fi-D-galactopyranoside 68 Figure 31: Hydrolysis of 4-nitrophenyl-P-D-fucopyranoside 68 Figure 32: Hydrolysis of 4-nitrophenyl-p,-D-mannopyranoside 69 Figure 33: Hydrolysis of 4-nitrophenyl-pVD-xylopyranoside 69 Figure 34: Hydrolysis of 4-nitrophenyl-a-L-arabinopyranoside 70 Figure 35: Hydrolysis of cellobiose 70 Figure 36: Hydrolysis of PNPG in pH 5.00 MES buffer 71 Figure 37: Hydrolysis of PNPG in pH 5.61 MES buffer 71 Figure 38: Hydrolysis of PNPG in pH 6.04 MES buffer 72 Figure 39: Hydrolysis of PNPG in pH 6.48 MES buffer 72 Figure 40: Hydrolysis of PNPG in pH 7.10 MES buffer 73 Figure 41: Hydrolysis of PNPG in pH 6.48 HEPES buffer 73 Figure 42: Hydrolysis of PNPG in pH 6.97 HEPES buffer 74 Figure 43: Hydrolysis of PNPG in pH 7.46 HEPES buffer 74 Figure 44: Hydrolysis of PNPG in pH 7.98 HEPES buffer 75 Figure 45: Hydrolysis of PNPG in pH 8.52 HEPES buffer 75 Figure 46: Hydrolysis of PNPG at 25.9°C 76 X Figure 47: Hydrolysis of PNPG at 37.0°C 76 Figure 48: Hydrolysis of PNPG at 43.1 °C 77 Figure 49: Hydrolysis of PNPG at 51.1 °C 77 Figure 50: Hydrolysis of PNPG at 54.9°C 78 Figure 51: Hydrolysis of PNPG at 60.2°C 78 Figure 52: Hydrolysis of PNPG at 68.0°C 79 Figure 53: Hydrolysis of PNPG at 75.1 °C : 79 Figure 54: Sequence alignment of the Agrobacterium (Abg) and Caldocellum (Cbg) fi-glucosidases. Aligned residues which are identically conserved are denoted by *. Aligned residues which are "similar" are denoted by A . Amino acids which are classed as "similar" are: AS,T; D,E; N, Q; R, K; I, L, M, V; F, Y, W. Residues implicated in catalysis are indicated by o 80 Figure 55: Multiple sequence alignment of Class 1 f3-glycohydrolases: fi-galactosidase from Sulfolobus solfataricus (a), P-glucosidases from Agrobacterium faecalis (b) and Caldocellum saccharolyticum (c), phospho-P-galactosidases from Lactobacillus casei (d), Streptococcus lactis (e) and Staphylococcus aureus (f), phospo-fi-glucosidase from E. coli (g) and E. chrysanthemi (h) and human LPH domans II (i), IQ (j) and IV (k). Residues conserved in at least 6 of the 11 sequences are boxed. Residues implicated in catalysis are indicated by o 82 xi LIST OF TABLES Table 1: Extinction Coefficients of Substituted Phenols and Phenyl Glycosides Under Direct Assay Conditions 19 Table 2: Extinction Coefficients of Substituted Phenols and Phenyl Glycosides Under Stopped Assay Conditions 20 Table 3: Michaelis-Menten Parameters for the Hydrolysis of Aryl Glucosides by Caldocellum (J-Glucosidase 23 Table 4: Michaelis-Menten Parameters for the Hydrolysis of 4-Nitrophenyl Glycosides by Caldocellum fi-Glucosidase 28 Table 5: Amino Acid Composition of Agrobacterium and Caldocellum (J-Glucosidase 39 Table 6: 4-Nitrophenol Extinction Coefficient From 25°C to 55°C at pH 6.5 for Direct Assay Measurements 52 xu LIST OF SCHEMES Scheme 1: Reaction catalyzed by a glycosidase; R may be an alkyl or aryl group or another sugar 1 Scheme 2: Stereochemical outcome of a glycosidase catalyzed hydrolysis 2 Scheme 3: Mechanism of hydrolysis of {5-glucosides by a p-glucosidase 3 Scheme 4: Active site directed labelling of P-glucosidase by Conduritol B epoxide 4 Scheme 5: Reaction of 2',4'-dinitrophenyl-2-deoxy-2-fluoro-p>-D-gluco-pyranoside with enzymic nucleophile and subsequent inactivation 5 Scheme 6: Hydrolysis vs. methanolysis of aryl P-galactosides in which (A) a common intermediate is not formed; (B) a common intermediate is formed 6 Scheme 7: Structures of nojirimycin 8 X U 1 LIST OF ABBREVIATIONS AND DEFINITIONS Abbreviations ABG Agrobacterium fi-glucosidase ala alanine AMPSO ,3-[(14-a e^memyl-2-hydroxyemyl)ammo]-2-hydr^ ^ arg arginine asp aspartic acid CAPSO 3-(cyclohexylamino)-2-hydroxy- 1-propanesulphonic acid CBG Caldocellum P-glucosidase CD circular dichroism DNA deoxyribonucleic acid DNP2FG 2\4,-dimtrophenyl-2-deoxy-2-fluoro-p,-D-glucopyranoside EDTA ethylene diamine tetraacetic acid glu glutamic acid HEPES N-(2-hydroxyemyl)piperazine-N'-(2-ethanesulphonic acid) ile isoleucine leu leucine lys lysine MES 2-(N-morpholino)ethanesulphonic acid m.p. melting point nmr nuclear magnetic resonance PAGE polyacrylamide gel electrophoresis PIPES piperazine-N,Nl-bis(2-ethanesulphomc acid) PNPG 4'-nitrophenyl-p'-D-glucopyranoside SBG Sulfolobus fi-galactosidase xiv SDS sodium dodecyl sulphate thr threonine TRIS tn^ (hyckoxymemyl)aminomethane tyr tyrosine val valine Kinetic and Physical Contants T Temperature (K) k Boltzmann constant h Planck's constant K M Michaelis-Menten constant (the apparent dissociation constant for all bound enzyme-substrate species) V r a a x Maximal rate of an enzyme-catalyzed reaction kc a t First-order rate constant for catalysis (turnover number) 1 CHAPTERI INTRODUCTION 1. Glycosidases and Glycoside Hydrolysis Glycosidases are a large, widespread class of enzymes which catalyze the hydrolysis of glycosidic linkages such as those between adjacent sugar residues (Scheme 1). While the natural substrate is typically a polyglycoside, the enzyme will catalyze alkyl or aryl glycoside hydrolysis. Scheme 1: Reaction catalyzed by a glycosidase; R (the aglycone) may be an alkyl or aryl group or another sugar residue. The glycosidase family of enzymes is subdivided into several groups based on the nature of the substrate and the specific characteristics of the hydrolysis catalyzed. First, the enzymes are categorized according to the sugar moiety, or glycone, they are most reactive towards (e.g. glucosidases are most reactive towards glucose, etc.) and further according to the ring form of the sugar that the enzyme prefers (e.g. furanose). Glycone specificity is often relaxed and many glycosidases have activity towards other glycosides than their optimum substrate. Second, the enzymes are classed according to the stereochemistry at the anomeric centre of the substrate (i.e. a or P) which they cleave. Specificity towards stereochemistry at the anomeric centre is near absolute. Finally, glycosidases are classed according to the relative product stereochemistry at the anomeric centre. Those glycosidases which release a product with the same stereochemistry as the substrate (e.g. (3 to P) are termed "retaining" while those with a different stereochemistry are termed "inverting" (Scheme 2). OH OR 2 OH Scheme 2: Stereochemical outcome of glycosidase catalyzed hydrolysis. A mechanism for retaining glycosidases was proposed by Koshland (1) in 1953 (Scheme 3). In this mechanism an enzymic nucleophile attacks at the anomeric centre on the opposite side from the aglycone to produce a covalent glycosyl enzyme intermediate, with general acid-assisted release of the aglycone. This intermediate is hydrolysed by general base-catalyzed attack of water to regenerate the free enzyme and release the product glycose with the same stereochemistry as the substrate sugar. Both the first step, termed glycosylation, and the second step, termed deglycosylation, proceed via transition states with significant oxocarbonium-ion character. Much of the rate acceleration in this mechanism occurs through non-covalent interactions between substrate and enzyme. Several of the features of this mechanism will be discussed below. The enzymic nucleophile is believed to be a carboxylate group in all cases and considerable evidence exists for this. 3-Dimensional structures of several glycosidases, determined from X-ray crystallographic data, indicate the presence of a suitably placed carboxylate nucleophile in hen egg white lysozyme (2) and bacteriophage T4 lysozyme (3), both glycosidases which catalyze the hydrolysis of cell wall polysaccharides, and in B. subtilis xylanase and B. circulans xylanase (4), glycosidases involved in the degradation of xylan. Further evidence is provided by the irreversible inactivation of sweet almond fi-glucosidase by Conduritol B 3 Scheme 3: Mechanism of hydrolysis of (3-glucosides by a P-glucosidase. 4 epoxide (5), which specifically labels an aspartate residue, believed to be the catalytic nucleophile by virtue of the specificity of the labelling and the similarity of the inactivator to the natural substrate (Scheme 4). Similar studies have been done on other glucosidases with related epoxides and produced similar labelling (6,7). / / Enzyme Scheme 4: Active site directed labelling of P-glucosidase by Conduritol B epoxide. The mechanism based inactivator 2',4'-dinitrophenyl-2-deoxy-2-fluoro-P-D-glucopyranoside reacts with the Agrobacterium P-glucosidase to trap the enzyme as the 2-deoxy-2-fluoro-glucosyl-enzyme intermediate (8) (Scheme 5). The presence of the electron withdrawing fluorine at C(2) destabilizes the oxocarbonium ion-like transition states, resulting in reduction in the rates of both glycosylation and deglycosylation. The good leaving group 2',4'-dinitrophenolate accelerates the glycosylation step but does not affect deglycosylation, leading to the accumulation of the glycosyl-enzyme intermediate and thus inactivation. Sequence determination of labelled peptide derived from the proteolysed inactivated enzyme has shown that Glu 358 is labelled by the 2-deoxy-2-fluoro-glucosyl moiety, suggesting that this is the catalytic nucleophile. 5 slow Enzyme Scheme 5: Reaction of 2\4'-dirtitrophenyl-2-deoxy-2-fluoro-pVD-glucopyranoside with enzymic nucleophile and subsequent inactivation. The Koshland mechanism proposes that a glucosyl-enzyme intermediate is formed in the reaction. Evidence for an intermediate comes from observation of the ratio of products formed in the presence of methanol in the solvolysis of several different aryl P-galactosides by f$-galactosidase (9). Methanol acts as a galactose acceptor and competes with water in the reaction. It was found that the ratio of methyl galactoside (the product of methanolysis) to galactose (the product of hydrolysis) was constant for a series of galactosides despite differences in rate with different leaving groups, indicating that a common intermediate is formed. The solvolysis of this intermediate is independent of the formation of the intermediate, occurring in a separate step (Scheme 6). l^F-NMR studies of the 2,,4'-dinitrophenyl-2-deoxy-2-fluoro-fJ-D-glucopyranoside-inactivated Agrobacterium P-glucosidase (see above) indicate that a covalent intermediate is formed with oc-stereochemistry (10) as is expected by the Koshland mechanism. While the Koshland mechanism proposes a covalent glucosyl-enzyme intermediate, others have proposed that an intimate ion pair forms between an oxocarbonium ion and an enzymic carboxylate rather than a true covalently bound species. Phillips (11) has proposed this mechanism for lysozyme and it is possible for this enzyme that an intimate ion pair is indeed the intermediate species, though it is as yet unproven. However, the lifetime of the intermediate in 6 H 2 0 — Galactose EGal-X M e 0 H Methyl Galactoside B ^ H 2 0 — Galactose Gal-X + E - EGal-X +- X + Gal-E MeOH*' Methyl Galactoside Scheme 6: Hydrolysis vs. methanolysis of aryl fi-galactosides in which (A) a common intermediate is not formed; (B) a common intermediate is formed. many glycosidases is on the order of 10" 3 to 10"! seconds (12) while the lifetime of a glucosyl cation free in aqueous solution is on the order of 10" 10 seconds (13). It would seem unlikely that the enzymic environment, no matter how favourable, would be able to stabilize the glucosyl cation to a sufficient degree so as to increase its lifetime by a factor of up to 10^ . The oxocarbonium ion character of the two transition states is another important feature of this mechanism. Evidence for this comes from kinetic isotope effects measured in cases where (individually) either glycosylation or deglycosylation are rate limiting. Significant a-deuterium kinetic isotope effects Qqj/kD of 1.15 to 1.20), indicative of sp^ character at the anoraeric centre in the transition state, have been observed for the hydrolysis of C(l)-deuterated {5-galactosyl pyridinium salts by fi-galactosidase (14), a reaction for which glycosylation is known to be the rate determining step. Similar results have been observed for the deglycosylation step. For example, kfjr/krj values of 1.20 to 1.25 were found for hydrolysis of C(l)-deuterated 2',4'-dinitrophenyl-fi-D-galactopyranoside by (J-galactosidase (14), a substrate for which deglycosylation was shown to be the rate tetenriining step. The indication of sp^ character at the 7 anomeric centre in the transition state is consistent with the formation of an oxocarbonium ion-like transition state. The positive isotope effects observed in this case indicate that at the deglycosylation transition state, C(l) is more sp^ -Uke than in the preceeding ground state, in this case the glycosyl-enzyme intermediate. This suggests that C(l) has sp^ hybridisation in the intermediate, consistent with the formation of covalent a-linked glycosyl-enzyme intermediate as proposed in the Koshland mechanism. Further evidence is derived from the study of inhibitors of these enzymes. Enzymes are believed to be optimized to bind and stabilize the transition states of the reactions they catalyze, thus a compound which mimics the shape and charge of the transition state will be bound tightly by the enzyme (Appendix I) (15). Such compounds are termed transition state analogues and are among the most potent inhibitors of an enzyme by virtue of their extremely tight binding to the substrate for the active site. Glucono-l,5-lactone is a potent inhibitor of P-glucosidase (16), binding 2 to 3 orders of magnitude better than the normal substrate. The coplanarity of C(5), the ring oxygen, C(l) and C(2) in this transition state analog makes it isosteric with the glycosyl cation (Figure 1). OH OH OH Gluconolactone Glucosyl Cation Figure 1: Structural and electronic similarities of glucono-l,5-lactone to the oxocarbonium ion-like glucosyl cation. The potent inhibitor nojirimycin (17) exists in several forms, and when protonated or dehydrated mimics the glucosyl cation (Scheme 7). The tight binding of these inhibitors suggests 8 that the transition state is oxocarbonium ion-like in nature. Scheme 7: Structures of nojirimycin. While a portion of the rate acceleration observed with glycosidases is likely due to attack of the nucleophilic carboxylate and the acid-catalyzed departure of the aglycone, much of the fate enhancement likely comes from non-covalent interactions between the enzyme and substrate. Non-covalent interactions which are optimized only at the transition state result in stabilization of the transition state relative to the ground state in which enzyme-substrate interactions are not optimally oriented. The relative stabilization of the transition state results in rate acceleration (Appendix I). Enzymes bind their substrates through a combination of non-covalent interactions such as hydrogen bonding, ionic bonding and van der Waal's attractions and it is these interactions which stabilize the transition state. The importance of such interactions has been demonstrated by studies of the E. coli f5-galactosidase hydrolysis rates for a series of deoxy analogues of 2',4'-dinitrophenyl-fi-D-galactopyranoside (18). Large decreases in enzyme catalyzed hydrolysis rates were observed when any of the four hydroxyl groups were deleted, despite the fact that spontaneous hydrolysis of the deoxy substrates was faster than that of the parent compound. The deoxy substrates are unable to undergo hydrogen bonding at the deoxy position and it is this loss of binding energy stabilization of the transition state which results in lower hydrolysis rates. The transition state is stabilized by approximately 4 kcal mol"* by each interaction at the 3-, 4- and 6- positions while the 2-position confers at least 8 kcal moH towards transition state stabilization. The enzyme 9 must be able to selectively bind the transition state while retaining affinity towards the ground state. Binding interactions must therefore occur in the ground state but not be optimized until the transition state is reached. The 2-position hydroxyl group undergoes the most extensive restructuring of any hydroxyl group in going from the ground state to the transition state and thus selectivity in binding the transition state relative to the ground state is greater with this interaction than with other interactions. This accounts for the much greater transition state stabilization observed with the 2-position interaction. Interactions at the 3-, 4- and 6- position appear to hold the rest of the sugar ring in position while the ring is distorted by transition state formation which is largely stabilized by the 2-position interaction. Similar results have been obtained in studies of A. wentii (3-glucosidase (19). Hydroxyl group deletion in this case has been shown to result in large decreases in enzyme catalyzed hydrolysis rates, the greatest rate decrease being observed with loss of the 2-position interaction. 2. Thermophilicitv and Enzyme Thermostability Microorganisms have evolved to be able to thrive in a wide range of extreme environmental conditions. Among these are organisms able to survive in conditions of high salinity, high or low pH and at extremes of temperature. Elevated temperature is among the most hostile of conditions in which microorganisms have evolved, for many of the macromolecules necessary for life are unstable under these conditions and will be readily denatured or destroyed at these temperatures. Despite these difficulties, studies have shown that microbial life can thrive at temperatures approaching or even exceeding the boiling point of water (20, 21). Clearly such organisms must possess unusual mechanisms which enable their survival. Thermophilic microorganisms depend, for their survival, upon a wide range of relatively subtle mechanisms which collectively allow survival at extreme temperatures. The biochemistry of thermophilic microorganisms is not significantly altered from that of their mesophilic (non-thermophilic) counterparts. This is illustrated by the fact that many mermophilic microorganisms are able to survive and grow at more normal temperatures as well as at temperature extremes, 10 indicative of the lack of any gross change in functioning of these organisms. Most bacterial genera include thermophilic species which resemble in many ways their mesophilic counterparts (22), further suggesting that mermophilicity results from small changes in the functioning of the microorganism. Thermophilicity is believed to primarily arise from three general mechanisms. First, changes in the lipid composition of mermophilic organisms towards a higher percentage of saturated and branched-chain fatty acids (23, 24) results in a higher melting point and greater stability of the membranes at higher temperatures (25). In addition, increased interactions of lipids with proteins in these microorganisms results in stabilization of the proteins, since a higher proportion of protein is membrane-bound in thermophilic microorganisms than in their mesophilic counterparts (25). Second, increased metabolic activity in thermophilic microorganisms results in a higher rate of macromolecule biosynthesis in order to replace heat-denatured cellular components (26). Third, and perhaps the most important, is the fact that the macromolecules have inherent thermostability which reduces the rate at which they are denatured by the extremes of their environment (27). This is best illustrated by the fact that almost all proteins isolated from thermophilic sources are inherently thermostable. Of interest then is an understanding of the factors which determine the thermostabUity of proteins. Active, functioning proteins exist in a specific folded conformation which exists by virtue of its greater stability than the unfolded state or a different folded state. This native conformation reflects a balance of a large number of small forces which favour the folded state over the unfolded state. At higher temperatures these interactions may be broken as a result of increased vibrational energy and collisions with water molecules resulting in a shift from the folded state to a denatured, inactive form (28). Thermostability results from a combination of the same factors which determine the folded state of a protein and which ultimately result from a protein's amino acid sequence. 11 2.1 Factors Affecting Thermal Stability of Proteins The factors which contribute to the stability of the folded state, and thus to protein thermostability, include non-covalent forces such as electrostatic interactions, hydrogen bonding, hydrophobicity and van der Waals interactions and covalent interactions such as disulphide bonding. Stability is also affected by the presence of bound metals or substrate. Electrostatic interactions can broadly be divided into specific and nonspecific charge interactions. Specific charge interactions between ion pairs occur when oppositely charged amino acid side chains are in close spatial proximity resulting in attraction between distant regions on the protein. Such interactions have been demonstrated to stabilize the folded state. Intra- and inter-subunit ion pairing has been demonstrated to be the main source of the difference in thermostability between the glyceraldehyde-3-phosphate dehydrogenases from lobster and B. stearothermophilus (29). Fersht has shown that engineering of a new ion pair into B. amyloliquefaciens ribonuclease (barnase) results in stabilization of the mutant by 0.5 kcal/mol (30). Nonspecific charge interactions have a destabilizing effect upon proteins. A high surface charge results in destabilization due to charge repulsion, since the high charge density in the folded state is rninimized by the unfolding of the protein. Conditions of high acidity or basicity increase the charge on the protein resulting in destabilization leading to denaturation of the protein. Proteins are thus most stable near their isoelectric point where the charge repulsions are minimal (31). Hydrogen bonding, which occurs when a hydrogen atom is shared between electronegative atoms, and van der Waal's attractions, which arise from fixed or induced dipoles, contribute to the stability of the folded state by forming bonds between otherwise distant regions of the protein. Proteins contain a large number of dipolar functionalities including carbonyl C=0 and amide NH groups of the peptide backbone and several side chains which can readily form hydrogen bonds and van der Waal's interactions. Hydrogen bonding plays a particularly important role in the stabilization of the elements of protein secondary structure, oc-helices and p-sheets. Side chain hydrogen bonding has been demonstrated to stabilize T4 lysozyme (32, 33) while site-12 directed mutagenesis work on this enzyme in which Val-157 was replaced by Thr resulted in increased stability as a result of better van der Waal's attractions (34). The hydrophobic effect is believed to be the most important factor affecting stability in most proteins. The hydrophobic effect essentially consists of an aversion of non-polar molecules for water resulting in the clustering of non-polar amino acids into the core of the protein away from the aqueous exterior. There is still some disagreement about the specific nature of this aversion. The effect seems to arise from the entropicaUy unfavourable ordering of water molecules required for the solvation of a non-polar solute. This ordering is a consequence of maximization of hydrogen bonding with other water molecules in the absence of such interactions with the solute. In order to mirumize interaction with the water the non-polar molecules associate, thereby reducing the surface area exposed to the aqueous media and reducing the •v. • / quantity of ordered water (35, 36, 37). The enthalpy and entropy for the solvation of non-polar molecules by water are strongly dependent upon temperature. The enthalpy of solvation becomes increasingly unfavourable as temperature increases while the unfavourable entropy for solvation diminishes. This results in an increase in the enthalpic contribution to the hydrophobic effect and a decrease in the entropic component until at sufficiently high temperatures the effect becomes largely driven by enthalpy (38, 39). Proteins contain a large number of non-polar, hydrophobic groups which are forced together by virtue of the hydrophobic effect and this contributes significantly to the stability of the folded state. The importance of the hydrophobic interactions is demonstrated by the fact that hydrophobic residues in the cores of proteins are more strongly conserved than those amino acids associated with other types of interactions (40). Thermal stability has been shown to correlate with the hydrophobicity of the protein. Proteins containing a higher proportion of hydrophobic amino acids are more stable than those of lower hydrophobic content (41, 42). Indeed, stability measurements of twelve different mutants of T4 lysozyme modified at the same position have shown a correlation of stability with hydrophobicity of the amino acid substitutions (43). 13 Covalent interactions between distant portions of the protein, specifically disulphide bonds between cysteine residues, like hydrogen bonding or electrostatic interactions, connect otherwise distant portions of the protein and prevent those connected points from unfolding away from each other, thereby increasing folded stability. While this is a stabilizing interaction in mesophile-derived proteins, this interaction is not as common in proteins from a thermophilic source, due to thermoinactivation which occurs at higher temperatures. This can result from cleavage of such disulphide bonds via a p l^imination mechanism to form thiocysteine and dehydroalanine residues (44). Metal binding and substrate binding both stabilize proteins by providing a means by which distant portions of the protein can interact through a bridging effect mediated by the metal or substrate. Ionic bonding or hydrogen bonding from several regions of the protein to the metal or substrate creates, in effect, a long bond between two regions of the protein which would otherwise not interact. The presence of calcium ions has been demonstrated to increase the thermostability of Bacillus thermoproteolyticus thermolysin by reducing chain flexibility and countering the destabilizing influence of a large number of negatively charged amino acids (45). Treatment of bovine, goat and human a-lactalbumin with EDTA, removing a bound calcium ion, was found to reduce the thermal stability by more than 20°C (46, 47). Lactase from Streptococcus thermophilus was found to be more stable in the presence of lactose, galactose and glucose but not other carbohydrates (48). Irreversible attachment of product to the active site of E. Coli dehydroquinase was found to increase the stability, raising the concentration of guanidine hydrochloride neccessary to denature the enzyme from 2 M to 4 M (49). 2.2 Attempts to Alter Thermostability of Proteins A variety of different approaches have been used to attempt to stabilize proteins. Random and site directed mutagenesis techniques have been used to produce mutant proteins containing one or more amino acid substitutions which have demonstrated increased stability. Such studies have demonstrated that even a single amino acid substitution can have a dramatic effect upon 14 stability. Random mutagenesis techniques typically involve growth of thermophilic bacteria with the gene for an essential non-thermostable protein inserted into the bacteria in a plasmid, in the presence of mutagenic chemicals. Survival depends upon the functioning of the non-thermostable protein, so the organism must mutate the thermostable protein so as to be able to survive. After exposure to high temperatures, those bacteria found to survive at the high temperature are isolated and the nature of the mutation(s) in the non-thermostable protein are examined. This approach has been used to engineer stability into B. pumilus chloroamphenicol acetyltransferase (Cat-86) (50). The thermophilic Bacillus stearothermophilus containing the non-thermostable enzyme from B. pumilus was grown at 58°C in the presence of the mutagen N-methyl-N'-nitro-N-nitrosoguanidine and the antibiotic chloroamphenicol. Survival depends on mutation of the Cat-86 enzyme such that this enzyme is stable at 58°C, as this enzyme confers resistance to the antibiotic. Mutation of alanine-203 to valine was observed to occur in all mutants found to be stable at this elevated temperature. Random mutagenesis has also been used to identify compensating substitutions which restore stability in mutants of T4 lysozyme, engineered so as to be less stable than the wild type enzyme (51). Insertion of such changes, by site directed mutagenesis, into the wild type enzyme will presumably introduce greater stability. Site directed mutagenesis involves the specific replacement of one or more amino acids in a protein with other known amino acids. This change is carried out by specific modification of the DNA coding for the enzyme. This approach typically attempts to specifically increase an intramolecular interaction which contributes to stability. This has been done by examination of a series of changes at structurally important amino acids or by inferring replacements from sequence comparison of related thermostable proteins. Site directed mutagenesis typically requires more detailed sequence and structural information than the random mutagenesis approach. The technique has become widely used in the past decade and numerous examples abound in the literature describing some increase in stability introduced with this technique. Previous mention has been made of increased stability observed in barnase when an additional ion-pair interaction was engineered into the enzyme (30). Venema has shown increases in stability of Bacillus 15 stearothermophilus neutral protease by individual mutation of methionine-186 to tryptophan, leucine-284 to tryptophan, cysteine-288 to leucine and cysteine-288 to isoleucine, mutations designed to give tighter packing of the hydrophobic core and increase van der Waals interactions (52). An increase of 10.7°C in the melting temperature (Tm) of Bacillus stearothermophilus L-lactate dehydrogenase was observed by replacement of arginine-171 with tyrosine and glutamine-102 with arginine (53). 3. Agrobacterium and Caldocellum ft-Glucosidase The wild type P-glucosidase enzyme from Agrobacterium faecalis was originally isolated by Han and Srinivasan (54). Subsequent investigation by Day and Withers (55) and Kempton and Withers (56) has extensively characterized the enzyme. The enzyme has been cloned and expressed in E. Coli (57) and the cloned product found to be identical to the wild type enzyme in amino acid content and properties. Most of the characterization of the enzyme has been on the cloned enzyme. The enzyme contains 458 amino acid residues, giving a monomer molecular weight of -50,000 Daltons, and it exists as a dimer in its active form. The enzyme is a retaining enzyme, suggesting the Koshland mechanism (Scheme 3) and evidence gathered from several studies strongly supports this mechanism for this enzyme. Kinetic studies show that the enzyme will act upon not only its natural substrate, cellobiose, but also aryl glucosides, galactosides, mannosides and other glycosides, indicating relatively relaxed specificity in both the glycone and aglycone sites. From the kinetic studies it has been revealed that the rate-determining step changes with the nature of the substrates involved. Bronsted plots of a series of aryl-glucoside substrates reveal a biphasic relationship between reaction rate and aglycone leaving group ability, indicating a change in the rate-determining step for this series. The most reactive substrates with the best leaving groups show rate limiting deglycosylation, while glycosylation is the rate-determining step for substrates with poor leaving groups (56). An a-deuterium kinetic isotope effect of kjj/kj} = 1.06 is found with substrates for which glycosylation is rate-limiting, while a a-deuterium kinetic 16 isotope effect of kjj/k£) = 1.11 is observed for the deglycosylation step. This change in the isotope effect further supports the observed change in the rate-determining step. Use of the mechanism based-inactivator, 2',4'-dinitrophenyl-2-deoxy-2-fluoro-f5-D-glucopyranoside (see above) results in covalent attachment of the inactivator to glutamate-358, strongly suggesting that this is the catalytic nucleophile. Site directed mutagenesis in which glutamate-358 was mutated to asparagine or glutamine, sterically conservative mutations which prevent the residue from acting as a nucleophile, resulted in complete loss of activity. Mutation to aspartate, which displaces but does riot remove the carboxylate, resulted in a 2500-fold reduction in the rate constant for the glycosylation step (58). Mutagenesis work has also indicated that Glu-170 plays a significant role in the mechanism, and is believed to be the acid-base catalyst Elimination of this group results in significant reductions in the rate of glycosylation for substrates with poor leaving groups and a reduction in the deglycosylation rate for all substrates. Sequence analysis reveals the Agrobacterium fi-glucosidase to be a member of a large family of |3-glycosidases which share a high level of sequence similarity (>20%). The family includes fi-glucosidase, P-galactosidase, phospho-fi-glucosidase and phospho-p-galactosidase enzymes derived from a wide range of sources (Archaebacteria to humans) (59,60). One member of this family is a retaining fi-glucosidase from the extreme thermophile Caldocellum saccharolyticum, a cellulolytic, anaerobic bacterium isolated from a 68°C geothermal pool in New Zealand (61). The Caldocellum fi-glucosidase shows a 50% sequence similarity to the Agrobacterium fi-glucosidase and an identical sequence in the region surrounding the catalytic glutamate nucleophile, suggesting Glu-355 as the catalytic nucleophile in the Caldocellum enzyme (62) (Appendix IV). Glu-170, believed to be the acid/base catalyst (63), is also conserved as is Tyr-298, another residue implicated in catalysis (64). The Caldocellum f3-glucosidase contains 455 amino acids and has a monomer molecular weight of 53,000. Whether the active enzyme exists in dimeric or monomelic form is unknown. Initial characterization of the enzyme has been done and suggests a broad substrate specificity similar to the Agrobacterium enzyme (65). 17 4. The Aims of This Study The aim of this study is to expand upon the characterization of the thermostable Caldocellum saccharolyticum P-glucosidase and to relate its properties to those of the Agrobacterium P-glucosidase, a non-thermostable enzyme with a high degree of sequence similarity. Specifically, kinetic behavior and stability of these enzymes will be investigated and compared. The rates of enzyme-catalyzed hydrolysis of a series of related substrates will be determined for the Caldocellum enzyme and related to those found for the Agrobacterium pV glucosidase in order to determine the extent of any relationship between the activities of these two enzymes. Thermal and chemical stability measurements will be performed in order to elucidate the extent of the Caldocellum P-glucosidase thermostability with respect to the Agrobacterium enzyme. The ultimate goal of this study is to probe similarities in the enzymic activity of the Agrobacterium and Caldocellum p-glucosidase and differences in their thermostabilities. 18 C H A P T E R H RESULTS AND DISCUSSION 1. Synthesis 4-Bromophenyl-fi-D-glucopyranoside was synthesized according to the method of Dr. L. Ziser (66), a modified Koenigs-Knorr (67) synthesis, from 2,3,4,6-O-acetyl-a-D-glucosyl bromide and 4-bromophenol. The low yield arose largely from difficulties in separating the product from the silver bromide byproduct. Attempts at using this technique for the synthesis of other glucosides were relatively unsuccessful and all other glucosides were synthesized by the Koenigs-Knorr (67) method from the protected glucosyl bromide and the appropriate phenol. Low yields achieved with this technique are due to a major hydrolytic side reaction resulting in production of 2,3,4,6-tetra-O-acetyl-D-glucopyranose. This impurity was successfully removed by silica gel column chromatography allowing for ready crystallization of the protected glucoside. The glucosides were deprotected using sodium methoxide in dry methanol according to the method of Zemplen (68). This technique gave good yields and the deprotected glycosides were readily crystallized and characterized. 2. Extinction Coefficients Extinction coefficients were determined for phenols and glucosides under conditions of the assays in which they were used for kinetic measurements. The assay wavelength was chosen as the point of maximum phenol absorbance. Values determined are shown in Tables 1 and 2. 3. Kinetics All kinetic measurements used relied upon spectrophotometric detection of released product. In all cases, enzyme concentration was chosen so as to provide sufficiently large absorbance changes to allow for accurate calculation of rate. In addition rates were sufficiently 19 Table 1: Extinction Coefficients of Substituted Phenols and Phenyl Glycosides Under Direct Assay Conditions Phenol Assay Eohenol x 1 0 - 3 eelucoside x l®" 3 Substituent Wavelength (run) (M"l cm-1) ( M - 1 • cm"1) 2-Chloro-4-nitro 400 14.30 0.000 4-Chloro-2-nitro 425 3.125 0.013 4-Nitro 400 6.751 0.000 3,5-Dichloro 280 1.364 0.613 3-Nitro 380 0.436 0.135 3-Cyano 292 2.646 1.638 3-Chloro 274 1.408 0.737 4-Bromo 288 0.969 0.205 2-Napthyl 325 • 1.438 0.242 4-Methoxy 288 2.339 1.546 3,4-Dimethyl 277 1.432 0.952 All extinction coefficients measured at 50°C in 50 mM MES buffer, pH 6.5. slow so as to result in less than 10% substrate usage over the period of assay, ensuring linear dependence of product release with time. Substrate concentrations employed generally ranged from one-seventh K \ j to seven times R e -values of kcat and were determined by a weighted fitting to the Michaelis-Menten equation using Grafit (69). Data are, however, presented in Appendix m in the form of Lineweaver-Burk (70) plots. This method is convenient for visual inspection but due to errors inherent in double-reciprocal analysis, kinetic values were not determined from these plots, except as noted below. 20 Table 2: Extinction Coefficients of Substituted Phenols and Phenyl Glycosides Under Stopped Assay Conditions Phenol Substitutent H 4-Methoxy 3,4-Dimethyl Assay Wavelength (run) 285 307 292 eohenol«10*3 (M-l cm-1) 2.470 2.956 2.121 £elucoside x l®" 3 (M" 1 cm"1) 0.000 0.015 0.008 All extinction coefficients measured at 50°C and pH 11.5 (conditions of stopped assay). The Lineweaver-Burk plot of the Caldocellum B-glucosidase catalyzed hydrolysis of 4-nitrophenyl-P-D-xyloside and 4-nitrophenyl-a-L-arabinoside shows biphasic behaviour (Figure 2) indicative of transglycosylation. Transglycosylation occurs when a second molecule of substrate, rather than water, attacks the glycosyl-enzyme intermediate, to produce an aryl disaccharide product by a mechanism analogous to the reverse of the normal reaction. This pathway occurs at higher concentrations of substrates and results in rate enhancement over the expected rate if deglycosylation is rate-limiting, giving the observed biphasic behaviour in the Lineweaver-Burk plot. Kinetic values for both the hydrolytic reaction and the transglycosylation reaction were estimated from the Lineweaver-Burk plot by separate extrapolation of the low and high concentration values. 4. Aryl Glucoside Kinetics Previous work by Kempton and Withers (56) measured Agrobacterium P-glucosidase catalyzed hydrolysis rates for a series of aryl glucosides. In order to compare the catalytic ability of the Caldocellum B-glucosidase to the Agrobacterium enzyme, rates of hydrolysis of a similar series of aryl glucosides were determined for the Caldocellum enzyme. The low activity of the 21 I i • i i i i i i i i I 0 0 0 I • • • • I 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 2 4 6 8 10 12 14 16 18 20 22 24 26 1/JS1 (1/mM) 1/IS1 (1/mM) Figure 2: Lineweaver-Burk plots for the hydrolysis of p-nitrophenyl-fi-D-xyloside (a) and p-nitrophenyl-a-L-arabinoside (b). Caldocellum enzyme at 37°C necessitated the use of a higher temperature, 50°C, for the measurement of the rates of this series. At this temperature many of the most reactive substrates were too labile and thus relatively few substrates with very good leaving groups (i.e. low pK a of corresponding phenol) were used in this study. A temperature higher than 50°C would have caused increased spontaneous hydrolysis of even more substrates, resulting in even fewer substrates being useable in the study, thus ruling out use of temperatures higher than 50°C for most of this work, despite these temperatures being closer to the enzyme normal operating temperature. High background absorbances were found for several substrates due to high glucoside extinction coefficients. For these substrates the standard assay with a 1 cm pathlength cell could not be used due to inaccuracies associated with the high background. Two techniques were used for kinetic measurements in these cases. One method involved use of a 0.1 cm pathlength cell to reduce background absorbance, though this had the unfortunate effect of also reducing the absorbance change, making accurate measurements of rate more difficult The second method used employed a stopped assay technique. Treatment of the assay system after a set time with concentrated base, raising the pH to approximately 11.5, had the effect of denaturing the enzyme 22 and stopping the reaction. Further, the released phenol is deprotonated to the phenolate which has in most cases a higher extinction coefficient and absorbs at a longer wavelength than the corresponding phenol. The glucoside absorbance was unaffected by the pH and was generally lower at longer wavelengths, resulting in a dramatic increase in extinction coefficient difference between phenol(ate) and glucoside, and further having much lower background absorbance. Values obtained through these two techniques were generally found to agree well. Kinetic constants for the Caldocellum B-glucosidase catalyzed hydrolysis of several aryl glucosides are presented in Table 3. Plots of log kg a t and log k<.at/KM v s - leaving group pKa, representative of leaving group ability, are shown in Figures 3 and 4 respectively. Data obtained with Agrobacterium B-glucosidase (56) are presented on the same Figures. As can be seen in Figure 3 log k ^ is independent of leaving group pKa up to pKa 9.5 for the Caldocellum enzyme, above which a shallow dependence with a slope of Pig « -0.2 is observed, though with a poor correlation coefficient The plot of log ^c^t^M against leaving group pKa for the Caldocellum enzyme (Figure 4) shows a linear dependence with a slope of Pig = -0.3. The proposed mechanism of the Caldocellum B-glucosidase involves two distinct catalytic steps (glycosylation and deglycosylation) (Scheme 3). kc a t is the rate constant for the rate determining step while k ^ / K ^ is the apparent second order rate constant for free enzyme and free substrate proceeding to the first irreversible step. Since diffusion of the phenol leaving group away makes the first step (glycosylation) essentially irreversible, I C ^ / K M should be associated with the first transition state. The rate deterrnining step changes with the nature of the leaving group between the first step, when glycosylation is the rate-detennining step (with poor leaving groups), and the second step, when deglycosylation is rate-limiting (with good leaving groups). The parameter k ^ / K ^ should show a dependence upon leaving group pKa, since this rate constant reflects the glycosylation step, and this is indeed observed. The shallow slope (Pig = -0.3) shows only a small dependence however indicating relatively little charge build up in the phenolate oxygen at the transition state. This is likely due to either bond cleavage being limited at 23 Table 3: Michaelis-Menten Parameters for the Hydrolysis of Aryl Glucosides by Caldocellum P-Glucosidase Phenol p K a kcat K M Substituent (s"1) (mM) (s-1 mM" 1) 2-Chloro-4-nitro 5.49 171 ± 2 0.44 ±0 .02 390 ± 2 0 4-Chloro-2-nitro 6.45 206 ± 6 0.30 ±0 .03 690 ± 9 0 4-Nitro 7.18 1 6 2 ± 4 0.79 ±0.06 210 ± 2 0 3,5-Dichloro 8.19 187 ± 6 0.62 ±0.06 300 ± 4 0 3-Nitro 8.39 210 ± 6 0.86 ±0.08 240 ± 3 0 3-Cyano 8.57 2 0 4 ± 6 1.6 ±0 .1 130 ± 1 0 3-Chloro 9.02 220 ± 1 0 1.6 ± 0 . 2 130 ± 3 0 4-Bromo 9.34 158 ± 3 1.49 ±0.08 106 ± 8 2-Napthyl 9.51 77 ± 5 1.3 ± 0 . 2 6 0 ± 1 0 H 9.99 69 ± 2 6.5 ±0 .5 11 ± 1 4-Methoxy 10.20 120 ± 6 1.3 ± 0 . 2 90 ± 2 0 3,4-Dimethyl 10.32 9 6 ± 6 5.0 ±0 .8 1 9 ± 4 Phenol pK a values were taken from (71), (72), (73) and (74). the transition state or to a large degree of proton donation to the phenoate oxygen. The lack of dependence of k ^ upon leaving group ability for substrates with good leaving groups (pKa < 9.5) indicates that glycosylation is rate limiting in this case while with poor leaving groups, glycosylation is rate limiting as the rate becomes dependent upon leaving group ability (Pig = -0.2). The small dependence of the glycosylation step upon leaving group ability observed with the kcat data is consistent with the k ^ / K M data, differences likely being due to the wide scatter in the points from which the k ^ value for Pi g is derived. 24 1 a o 2.2 h 2.4 h 0.8 1.4 h 1.6 h 1.8 h 1.2 h 2 h 1 h 4 6 8 10 pKa Figure 3: Bronsted plot relating log (kcat) and leaving group ability (pKa) for hydrolysis of aryl glucosides by Caldocellum (o) and Agrobacterium (A) fi-glucosidase. The Agrobacterium fi-glucosidase shows, as expected, a biphasic plot of log kc a t against leaving group pK a (Figure 3). Substrates with relatively good leaving groups (pKa < 8) show no dependence of log on leaving group pK a . With poor leaving groups (pKa > 8), a significant dependence of kc a t upon leaving group ability is observed as shown by the slope of Pig = -0.7. Unexpectedly, a similar biphasic plot of log k ^ / K M against leaving group pK a is observed (Figure 4), with log kcju/Kj^ independent of pK a below approximately pK a 7. The breakpoint observed in the kcat^M date ^ unexpected as there should be no change in the step upon which k ^ / K M is dependent The lower than expected kc a t /K\f values with good leaving groups for the Agrobacterium P-glucosidase has been theorized to result from association of the enzyme and substrate becoming rate-limiting (56). A plot of the log kcat/KM values of the Agrobacterium fi-glucosidase vs. the Caldocellum enzyme values for the series of aryl glucosides is presented in Figure 5. As k ^ / K M values are a 25 Figure 4 Brosted plot relating log (kcat^M) ^ d leaving group ability (pKa) for hydrolysis of aryl glucosides by Caldocellum (o) and Agrobacterium (A) P-glucosidase. reflection of the structure of the transition state, the plot shown in Figure 5 is therefore a comparison of the transition state structures of the two enzymes (75). A correlation coefficient of 0.95 was found indicating considerable homology in the active site region of these two enzymes, suggesting the two enzymes interact with the substrate in a similar manner. This would seem to indicate that the type and nature of the binding interactions are relatively similar, not surprising considering the high sequence homology. The slope of 2.0 in this plot suggest that while the nature of the binding interactions are the same, the exact nature of the transition state differs. The greater sensitivity of the Agrobacterium P-glucosidase to changes in the leaving group ability indicates that there is a greater degree of bond breakage and charge development in the Agrobacterium enzyme's transition state than in the Caldocellum P-glucosidase transition state. This is roughly consistent with the relative slopes measured in the kc a t against leaving group pK a plots for glycosylation rate-limiting substrates in which a slope Pig » -0.2 for the Caldocellum 26 indicates much smaller degree of bond cleavage than in the Agrobacterium enzyme for which a slope fJig = -0.7 was measured. As the relative ratio of Pig values should be the same as the slope of the relative log Ckcat^M) plot, there exists a discrepancy between these two measures of transition state interaction. Due to the poor correlation of log k^t and pK a greater confidence is placed on the relative log (kcat/KM) values as a measure of transition state similarity and thus the large relative Pig values are believed to be due to this poor correlation. Thus, while the two enzymes share similar binding interactions, the transition state is more developed in the Agrobacterium. CBG log kcat/Km Figure 5: Plot of log k ^ / K M values measured for the aryl glucoside series with Agrobacterium P-glucosidase versus the corresponding parameter measured for the Caldocellum enzyme. 5. 4-Nitrophenyl-Glycoside Kinetics Rates of the Caldocellum P-glucosidase catalyzed hydrolysis of several 4-nitrophenyl glycosides were determined in order to investigate the specificity of the enzyme towards the 27 glycone portion of the substrate and for comparison with previous data collected for hydrolysis of these substrates by the Agrobacterium enzyme (56). The low background absorbance of the glycosides allowed all the kinetics in this series to be carried out with the standard assay at a fixed temperature (50°C). The different members of this series of substrates differ from the glucoside in the configuration of individual hydroxyl groups around the ring or in some cases the absence of such hydroxyl groups. As such they provide excellent tools for probing the binding interactions which are important for substrate binding and catalysis. The structures of the glycone moieties used in this study are shown in Figure 6. Glucoside Galactoside Fucoside Mannoside Xyloside L-Arabinoside Figure 6: Glycone moieties used; R is a 4-nitrophenyl group. A particularly broad specificity is observed from the data provided in Table 4, not unlike the specificity observed for the Agrobacterium enzyme. The enzyme is able to hydrolyze many glycosides reasonably efficiently (based upon kcat^M values) despite changes in structure. The C-2 epimer (mannoside) shows the biggest difference, a 280-fold drop in efficiency, but even this is not extreme compared to many enzymes. Other changes have even less effect, with the 28 Table 4: Michalis-Menten Parameters for the Hydrolysis of 4-Nitrophenyl Glycosides by Caldocellum P-Glucosidase 4-Nitrophenyl ^cat K M kcat/KM Glycoside Substrate (s"1) (mM) (s*1 mM" 1) P-D-Glucoside 162 ± 4 0.79 ±0.06 210 ± 2 0 P-D-Galactose 152 ± 2 1.98 ±0.08 77 ± 4 P-D-Mannoside 0.144 ±0.001 0.198 ±0.008 0.73 ±0 .03 p-D-Xyloside i) 7.2 ±0 .6 1.5 ± 0 . 3 5 ± 1 ii) 16.2 ±0 .6 6.9 ±0 .7 2.4 ± 0 . 3 P-D-Fucoside 147 ± 2 0.49 ± 0.02 300 ± 2 0 a-L-Arabinoside i) 4.4 ±0 .3 0.051 ±0.008 90 ± 2 0 ii) 13.5 ±0 .1 0.43 ±0 .02 31 ± 2 Cellobiose 2 4 0 ± 1 0 38 ± 5 6.4 ± 1 Parameters indicated (i) are for the hydrolysis reaction and (ii) are for the transglycosylation reaction. fucoside actually hydrolyzed 1.5 times more efficiently. A comparison is made with the Agrobacterium enzyme through a plot of corresponding log k ^ / K M values of the two enzymes against each other, and is presented in Fig 7. For purposes of comparison, log k ^ / K M values for the activity of the Sulfolobus solfataricus P-galactosidase (76) are plotted against the Caldocellum enzyme values and presented on the same figure. The Sulfolobus P-galactosidase is a member of the same family of enzymes as the Agrobacterium and Caldocellum P-glucosidases and shares considerable sequence similarities (Appendix IV) (59, 60). The correlation of 0.90 between the Agrobacterium and Caldocellum p-glucosidases and the wide scatter indicates that while there is a trend in the specificity, suggestive of some similarity in the binding site, there appear to be some differences in the manner of binding between the two enzymes. The most 29 -0.4 0 0.4 0.8 1.2 1.6 2 2.4 2.8 CBG log kcat/Km Figure 7: Plot of log kcat/K-M values measured for the 4-nitrophenyl-glycoside series with Agrobacterium B-glucosidase (o) and Sulfolobus B-galactosidase (A) versus the corresponding parameter measured for the Caldocellum enzyme. Agrobacterium values measured for glycosides with axial C-4 hydroxyl groups are shaded. significant difference would appear to occur with epimers at C-4. The Caldocellum enzyme seems much more readily able to accomodate changes in C-4 than the Agrobacterium enzyme. Either the Caldocellum enzyme has a fortuitously placed group able to interact positively with an axial C-4 hydroxyl group or some group in the Agrobacterium clashes with the axial C-4 hydroxyl making such binding unfavourable for steric reasons. Such a group need not be part of the normal catalytic machinery, thus there is no reason for it to be conserved. Unfortunately, omission of all of the glycosides with axial C-4 hydroxyls leaves too few points to produce a reasonable or meaningful correlation. 30 6. pH and Temperature Stability Measurement of the stability of the enzyme at different pH and temperature conditions was determined by incubating the enzyme under a range of conditions of pH and temperature followed by an assay under standard conditions in order to ensure that any variations in activity are attributed to denaturation under the conditions of the incubation. Assays were performed at 50°C in pH 6.5 buffer, conditions to which the enzyme was known to be stable, with saturating concentrations of 4-nitrophenyl-B-D-glucopyranoside as a substrate. The primary goal of the pH/stability study was to provide conditions under which enzyme stability was known in order to allow for accurate pH activity measurements to be made without enzyme denaturation during the time course of the assay. pH stability assays at both 50°C and 70°C showed the enzyme was stable to a broad range of pH and results are shown in Figure 8. -0.02 h •2 -0.04 <Jr -0.06 -0.08 ~ r 1 r g—o a g—fi—o—o-J 1 I L 8 10 PH Figure 8: Plot of logarithim of the rate of activity loss Gog k) against pH for the Caldocellum B-glucosidase at 50°C (o) and 70°C (A). 31 Thermal stability measurements showed reasonable stability of the enzyme to temperatures in excess of 85 °C. The half-life for activity loss as a function of temperature is shown in Figure 9 along with data for the Agrobacterium enzyme. The greater stability of the Caldocellum enzyme is clearly evident from this graph, since rapid thermal inactivation occurs with the Agrobacterium enzyme at 60°C, far below the >85°C required for similar denaturation of the Caldocellum. It is interesting to note that under the parent organism optimal growth temperature (37°C for Agrobacterium, 68°C for Caldocellum), the stability is very similar (4-5 hour half-life). CD 3 h to p O) o 1 h 20 40 60 80 100 Temperature Figure 9: Plot of the half-life for thermally induced activity loss against temperature for the Caldocellum (o) and Agrobacterium (A) enzymes. 7. Dependence of Rate uoon pH and Temperature Michaelis-Menten parameters for hydrolysis of PNPG by Caldocellum fi-glucosidase at a series of pH values ranging from 5 to 8.5 were measured at a constant temperature of 50°C. The previous study had demonstrated that the enzyme is stable (>95% activity remaining over 5 32 minutes) at the range of pH values studied. MES and HEPES buffers were used for this study with duplicate measurements being made at intermediate pH values to ensure the absence of specific buffer effects. A stopped assay system was used for this study since this avoided problems of a pH-dependent product extinction coefficient encountered with a direct assay. Plots of log kf;at and log k ^ / K M against pH are presented in Figure 10. The pH dependence of log k^t reflects ionizations in the enzyme-substrate complex, thus with the substrate used in this study, reflects ionization in the deglycosylation step. The pH dependence of log kcat/KM reflects ionization in the free enzyme and thus reflects the glycosylation step. In both cases reactions are dependent upon a protonated group (pKa 7.0 for deglycosylation, pK a 6.8 for glycosylation) the deprotonation of which results in loss of activity. Activity is also dependent, in the opposite sense, upon a group with a pK a < 5. However, the instability of the enzyme at low pH values prevented accurate measurements at sufficiently low pH values to accurately determine this value. This pH dependence compares reasonably well with that of the Agrobacterium enzyme, although it is shifted to somewhat lower pH values since the Agrobacterium activity is dependent upon a group with a pK a of approximately 8. b Figure 10: pH dependence of the hydrolysis of 4-nitrophenyl-fi-D-glucoside by Caldocellum fi-glucosidase. (a) Plot of log kcat vs. pH. (b) p l o t o f l o 8 kcat^M v s - P H -33 It is possible to speculate as to the nature of the residues responsible for the observed ionizations, though the observed ionizations might not be due to discrete mechanistically important residues, but rather may reflect some remote ionizations which affect activity in some manner. The ionization at pK a < 5 may reflect the ionization of the nucleophile, as this must be in the deprotonated form to attack the anomeric centre, and for the glycosylation pH dependence this would seem to be likely. Alternatively, this ionization may reflect changes in the protonation state of the base catalyst, which also must be in the deprotonated form, and for the deglycosylation pH dependence this would seem to be more likely. In the glycosylation pH dependence, the ionization at pK a 6.8 may reflect ionization of the acid catalyst. However, by the principle of microscopic reversibility, the residue responsible for the acid catalysis must be the same as the base catalyst and this would indicate a change in pK a of approximately 2 pH units (6.8 to < 5) in the enzyme by binding the substrate. A large change in the pK a of the acid/base catalyst has been observed with the B. circulans xylanase (77), supporting the possibility that these ionizations are those of the acid/base catalyst in the B-glucosidase. Recent results on the Agrobacterium enzyme suggest that the ionization of pK a ~ 8 is that of the putative acid/base catalyst, Glu 170. Since this residue is conserved in the Caldocellum B-glucosidase, the difference in pK a is either due to the different temperature, or to a different environment around the same ionizable group (78). Michaelis-Menten parameters for hydrolysis of PNPG by Agrobacterium and Caldocellum B-glucosidases were determined at a series of temperatures within the stability range of each enzyme (>95% activity retained over 5 minutes). These are presented in the form of Arrhenius plots in Figure 11. A breakpoint near 50°C is seen in both the log k ^ and log k ^ / K ^ plots for the Caldocellum B-glucosidase. This breakpoint likely arises from some conformational change which occurs with a rise in temperature or a change in the rate-determining step. The exact nature of any conformational change is unknown, but if it is occurring must involve a shift from a 34 conformation which is very sensitive to temperature changes at low temperature, to one that is much less sensitive to temperature changes at high temperature. An approximate activation energy barrier of 7 kcal moH for the high temperature Caldocellum k ^ measurements compares reasonably well with the 9 kcal mol'l for the Agrobacterium enzyme. At low temperatures an approximate value of 19 kcal moH for the activation energy is observed. At temperatures near the normal operating temperatures of the respective enzymes, the similar thermal dependence of activity and activation energy observed is suggestive of a similar conformation or the same rate-determining step for the two enzymes. Figure 11: Arrhenius plots relating rates of Caldocellum (o) and Agrobacterium (A) f3-glucosidase catalyzed hydrolysis of 4-nitrophenyl-fi-D-glucoside with temperature, (a) Plot of log v s - (b) Plot of log kc at/K\i vs. 1/T. 8. Cysteine Titration A cysteine titration of the Agrobacterium and Caldocellum fi-glucosidase revealed 6.07 and 1.90 free cysteines per enzyme molecule, respectively, for the two enzymes. The primary sequence of these enzymes indicates the presence of 6 and 2 cysteines for the Agrobacterium and Caldocellum enzymes, respectively, indicating that both enzymes exist in the fully reduced state 35 and that there are no disulphide bonds present in either enzyme. The absence of such bonding rules out disulphide bonds influencing stability in either enzyme. 9. Chemical Denaturation Stability Resistance to chemical denaturation was measured by addition of urea at different concentrations to enzyme samples then monitoring the extent of unfolding by circular dichroism (CD) spectropolarimetry. Chemical denaturants such as urea or guanidine cause protein denaturation by reducing the hydrophobic effect, since non-polar residues are more easily solvated by urea or guanidine solutions than by water alone. Previous work (79, 80) has indicated that proteins that show increased thermal stability will show increased resistance to chemical denaturation, making chemical stability a useful probe of thermal stability. This is understandable as chemical stability is dependent upon the extent of the hydrophobic effect in the stabilization of the protein, a factor which has a strong effect upon thermal stability as well. In this study it was important to ensure that the denaturation of the protein was reversible in order to provide a true measure of stability. To a simple approximation, a protein will exist in one of two states, a folded state, F, and an unfolded state, U, which are in equilibrium. Under conditions in which the protein is normally active, the protein will exist primarily in the folded state, while denaturing conditions shift the protein to the unfolded state. The position of this equilibrium under a particular denaturing condition is a reflection of the stability of the protein. The unfolded state can undergo association and may ultimately precipitate out of solution, making the process irreversible. This situtation can be represented as follows: F = 5 = ^ U • A (1) In order to ensure that a stable equilibrium between F and U was truly being measured, it was important to nunirnize the extent of the association reaction to form A, as significant association would shift the equilibrium to the unfolded side. Initial chemical denaturation studies used 36 guanidine hydrochloride as a denaturant. Unfortunately, refolding to the folded state could not be demonstrated for this denaturant and the validity of measurements made with guanidine was suspect. Experiments with urea allowed for >75% recovery of activity after unfolding, suggesting that very little association is occurring. Urea is unfortunately not as good a denaturant as guanidine, resulting in some difficulty in effecting denaturation of more stable species. Urea denaturation studies, shown in Figure 12 and Figure 13, were carried out on both Agrobacterium and Caldocellum enzymes and upon these enzymes with the 2-deoxy-2-fluoro-D-glucosyl moiety bound in the active site. Both of these enzymes possess significant amounts of a-helix and have intense circular dichroism absorption bands centred at 222 nm. Lower a-helix absorption bands were not used due to absorption of urea at wavelengths lower than 210 nm. The unfolding of the enzyme, in particular the loss of secondary structure, is characterized by a reduction in the intensity of the absorbance of 222 nm as the structure of the enzyme becomes much more random. Plots such as Figure 12 & 13 indicate the quantity of urea necessary to cause denaturation, with greater stability being manifest in the greater amounts of urea being neccessary for denaturation to occur. Unfortunately it was not possible to demonstrate unfolding of the two different samples at the same temperature due to the relatively poor capacity of urea as a denaturant. From this work it is clear however that the Caldocellum enzyme is more stable to chemical denaturation than the Agrobacterium enzyme and further that the presence of the 2-fluoroglucosyl moiety in the active site increases the stability of both the Agrobacterium and Caldocellum enzymes. The two techniques used in this study to measure stability, urea denaturation monitored by circular dichroism and time-dependent thermally induced activity loss, provide different measurements of stability. Circular dichroism, as used in this experiment, is most sensitive to changes in secondary structure and the denaturation measured by this technique as discussed above represents loss of secondary structure. As activity is dependent upon tertiary structure, the measurement of activity loss with temperature monitors the denaturation of the tertiary structure. Thus, these two techniques will not neccessarily provide results which coincide with each other. 37 g. CD (D > f 40 h 20 h 100 ^- A A * * # A o I 80 60 T 1 1 -|— A A * A J o J L 4 6 Urea Concentration (M) 8 Figure 12: Urea denaturation curves for Caldocellum (A) and Agrobacterium (o) B-glucosidase and the Agrobacterium P-glucosidase with the 2-fluoro-2-deoxyglucosyl moiety bound in the active site (•) at 41°C. The relative ellipticity at 222 nm is plotted as a function of denaturant concentration. The ellipticities in the absence of urea were defined as 100%. While the greater stability of the Caldocellum P-glucosidase is evident, the source of the increased stability is not. Specific amino acid differences that result in the increased stability cannot be determined from the data measured and will likely require extensive mutagenesis work to elucidate. It is possible to speculate as to the nature of the difference in stability by examination of the amino acid composition and sequence of the two enzymes (Appendix IV). The most signficant changes in amino acid content between the two enzymes are indicated in Table 5. There appears to be a shift from small hydrophobic amino acids (Ala) in the Agrobacterium enzyme to larger hydrophobic amino acids (Ile/Leu/Val) in the Caldocellum enzyme. A sequence comparison reveals 14 substitutions of an alanine in the Agrobacterium 38 " I ' 1 " 1 ~l 1 ' 4 .» • • .-O -O ~ o o j I i I i I i ° 0 0 2 4 6 8 Urea Concentration (M) Figure 13: Urea denaturation curves for Caldocellum fi-glucosidase (o) and the Caldocellum J5-glucosidase with the 2-fluoro-2-deoxyglucosyl moeity bound in the active site (•) at 79°C. The relative ellipticity at 222 nm is plotted as a function of denaturant concentration. The ellipticities in the absence of urea were defined as 100%. enzyme by a larger hydrophobic amino acid in Caldocellum enzyme, while there are only six such substitutions in the other direction. Tighter packing of the protein interior and a larger hydrophobic effect are expected from these amino acid substitutions, suggesting that the hydrophobic effect is responsible in part for the greater stability of the Caldocellum enzyme. The thermostable Sulfolobus (3-galactosidase shows a similar shift to larger hydrophobic amino acids relative to the Agrobacterium enzyme, further indicating that an increased hydrophobic effect is responsible for the increased stability. The Agrobacterium fi-glucosidase has significant net negative charge due to the excess of acidic residues (Asp/Glu) over basic residues (Arg/Lys). The increase in number of lysine residues in the Caldocellum enzyme results in a much smaller charge imbalance and a lower net charge on the Caldocellum enzyme. The lower net charge should 100 0 80 d 'o Q_ UJ CD > 60 03 CD rr 40 39 result in decreased electrostatic destabilization of the Caldocellum enzyme and thus increased stability. The conformational rigidity of the Caldocellum enzyme would seem to be increased by the reduction in the number of flexible glycine residues and this would be expected to increase stability, though the decrease in proline residues would tend to suggest greater flexibility and thus the net effect of this change is difficult to assess. A great deal of speculation about the source of the Caldocellum enzyme's greater stability is possible, but until mutagenesis work is done to Table 5: Amino Acid Composition of Agrobacterium and Caldocellum 8-Glucosidase Amino Acid Agrobacterium Caldocellum Difference Ala 53 22 -31 Cys 6 2 - 4 Gin 6 19 +13 Gly 43 30 -13 De 16 33 +17 Lys 16 33 +17 Pro 26 16 -10 Arg/Lys 40 50 +10 Asp/Glu 62 63 +1 ne/Leu/Val 80 106 +26 corroborate this speculation, it is difficult to know for certain which changes are responsible for the greater stability. The presence of a 2-deoxy-2-fluoro-glucosyl moiety covalently bound to the nucleophilic carboxylate evidently results in considerable stabilization of both enzymes. Hydrogen bonding between enzyme and inactivator acts as a "bridge" between either side of the active site and creates in effect several very long bonds across the active site. The observed increase in stability is doubtless due to this effective increase in the number of intramolecular bonds in the enzyme. The extent of the stabilization conferred is difficult to assess. While approximate changes in the stability of proteins can be derived from denaturation plots, such measurements require a constant temperature for all measurements. Due to the relatively poor denaturation ability of urea, it was not possible to measure the chemical denaturation of the 2-deoxy-2-fluoroglycosyl 40 inactivated enzymes at the same temperature as the non-inactivated enzyme. Thus, any energy difference determined would have to account for changes in the quantity of chemical denaturant and changes in temperature, and would be very crude at best. An energy difference of greater than approximately 3 kcal moH can be assessed for the presence of the inactivator as this is the highest energy difference which could be determined from the urea denaturation curves, and the inactivated enzyme is known to be at least this stable. The value in such measurements lies in the fact that the extent of stabilization must correlate with the binding energy of the inactivator with the enzyme. Changes in the inactivator which result in changes in the binding energy should be reflected by changes in the stabilization of the enzyme. 10. Conclusions The results of the experiments presented herein further characterize the activity and stability of the thermostable Caldocellum saccharolyticum P-glucosidase, in particular as these relate to the properties of the P-glucosidase from the mesophile Agrobacterium faecalis with which the Caldocellum enzyme shares a 50% sequence similarity. Similarities between the activities of the Caldocellum and Agrobacterium enzymes towards aryl glycoside substrates combined with the similarity in the sequences of the two enzymes, supports a two-step mechanism which proceeds through a covalent glucosyl-enzyme intermediate in both cases. This mechanism is supported by the inactivation of the enzyme by the inactivator 2\4,-dinitrophenyl-2-deoxy-2-fluoro-p>-D-glucoside which traps the enzyme as the 2-deoxy-2-fluoroglucosyl-enzyme intermediate. The second step in this mechanism is rate-limiting with all substrates, save those with the poorest of leaving groups, as indicated by the near lack of dependence of log kf.at upon pK a of the substrate leaving group. The value Pig of -0.2 for the glucosylation step indicates either that the glycone-aglycone bond is not substantially cleaved at the transition state or that there is considerable proton donation, in contrast to the Pig of -0.7 for the Agrobacterium enzyme, which indicates significant bond cleavage at the transition state and 41 relatively little proton transfer. The Caldocellum enzyme, while having similar activity to the Agrobacterium enzyme, has been demonstrated to be significantly more stable to chemical and thermal denaturation, by at least 3 kcal moi"*. A similar increase in stability is observed when either enzyme is trapped as the 2-deoxy-2-fluoroglucosyl-enzyme intermediate, this increase in stability relating to the strength of the interactions with the substrate. 42 C H A P T E R m M A T E R I A L S A N D M E T H O D S 1. Synthesis 1.1 General Procedures and Methods Melting points (m.p.) were determined using a Laboratory Devices Mel-temp II melting point apparatus, and are uncorrected. lH-nuclear magnetic resonance (nmr) spectra were recorded on a 200 MHz Bruker AC-200. Chemical shifts are listed in the delta (8) scale. Compounds run in CDCI3 and C D 3 O D are referenced against internal deuterium signals. Micro-analyses were performed by Mr. P. Borda, Micro-analytical laboratory, University of British Columbia, Vancouver. Solvents and reagents were either reagent grade, certified or spectral grade. Dry methanol was distilled from magnesium methoxide prepared in situ by reaction of methanol with magnesium turnings in the presence of iodine. Substituted phenols were obtained from the following sources: Aldrich chemicals; 3-chloro, 3-cyano: Eastman Organic Chemicals; 4-bromo, 3,4-dimethyl, 4-methoxy. All chemicals were used as supplied without further purification. Thin-layer chromatography (tic) separations were performed using Merck Kieselgel 60 F-254 analytical plates. Compounds were detected visually (when possible) under U.V. light, or by charring with 5% sulphuric acid in methanol Column chromatography was performed according to the method of Clark-Still et al (81), using a silica gel column of Kiesegel 60 (180-230 mesh). Several compounds used in this work were synthesized by others in this laboratory: From J. Kempton; 4-Chloro-2-nitrophenyl-, 3,5-dichlorophenyl-, and 2-napthyl- B-D-glucopyranoside and 3'-nitrophenyl-2,3,4,6-tetra-0-acetyl-B-D-glucopyranoside: from M. Namchuk; 2',4'-dinitrophenyl 2-deoxy-2-fluoro-B-D-glucopyranoside and 2-deoxy-2-fluoro-B-D-glucopyranosyl fluoride. 43 Phenyl-B-D-glucopyranoside and all 4-iutrophenyl-|H)-glycopyranosides were obtained from Sigma Chemical Co. 2-Chloro-4-nitrophenyl-B-D-glucopyranoside was a gift of Dr. Marc Cloeyssens. 1.2 General Compounds 2.3.4.6-Tetra-O-acetyl q-D-glucopyranosvl bromide fl i Compound (1) was prepared by the method of Haynes and Newth (82). 1,2,3,4,6-Penta-O-acetyl a-D-glucopyranose (25 g, 64 mmol) was added to a solution of 48% HBr in acetic acid (25 ml) and acetic anhydride (1 ml) then stirred for 90 minutes at which time the reaction was complete by tic. The solution was added to 500 ml of ice water and extracted against CHCI3 (4 x 50 ml). The pooled organic extract was washed with ice water (500 ml) followed by sodium bicarbonate (500 ml), after which the aqueous layer remained basic, and again with water (500 ml). The organic layer was dried over MgS(>4, filtered, and the solvent removed in vacuo to produce a white mass of crystals. M.p. 87-89° (lit. (83) m.p. 88-89°C). Aryl 2.3.4-Tetra-O-Acetyl ft-D-Glucopvranosides The Koenigs-Knorr (67) procedure of glucoside synthesis was used for all "glucosides except for the synthesis of 4-bromophenyl-2,3,4-tetra-0-acetyl pMD-glucopyranoside. The appropriately substituted phenol was suspended in 1 M NaOH (1 mmol phenol/ml base), compound (1) dissolved in acetone (approximately 0.4 mmol/ml) added to the phenol solution, and the mixture stirred for 16-24 hours at room temperature. After evaporation of the solvent the slurry was diluted with water and extracted with CH2CI2 O x 50 ml), the combined organic layers washed twice with NaOH (0.1 M), dried over MgS04, filtered, and solvent evaporated in vacuo to give a colourless gum. Most compounds were purified by a silica gel column after which they readily crystallized. 44 Peaggtyiation Protected glucosides were deprotected by a modified Zemplen method (68). The protected glucoside was suspended in dry methanol (0.2 mmol/10 ml), to which was added sodium methoxide (0.2 mmol/10 ml). The reaction was stirred at room temperature until complete, usually 1 to 2 hours, then excess base was neutralized by addition of Dowex 50W(H+) cation exchange resin, the resin removed by gravity filtration and washed several times with methanol. The methanol was evaporated in vacuo to give a crystalline product, which was then recrystallized from an appropriate solvent 3'-Nitrophenyl-B-D-glucopyranoside (3) The acetylated compound (2) was prepared by J. Kempton. This material (0.1 g, 0.21 mmol) was deacetylated with sodium methoxide/methanol and the product glucoside (3) was recrystallized from ethanol to produce white crystals (0.040 g, 0.13 mmol, 62%). M.p. 166-168°C (lit.(84) m.p. 166-168°C) ! H nmr (D20): 8 8.00 (m, 2 H, H(2'), H(4')), 8 7.50 (m, 2 H, H(5'), H(6')), 8 5.20 (d, 1 H, J l j 2 8 Hz, H(l)), 8 3.95 (dd, 1 H, J 6 a > 6 b 11 Hz, J 5 ) 6 a 2 Hz, H(6a)), 8 3.75-3.45 (5 H, H(2), H(3), H(4), H(5), H(6b)). Anal. calc. for C 1 2 H 1 5 N 0 8 ; C, 47.84%; H, 5.03%; N, 4.65%; Found: C, 47.45%; H, 5.07%; N, 4.50%. 4'-Bromophenyl-B-D- glucopyranoside (5) The acetylated compound (4) was prepared according to the method of Dr. L. Ziser (M6), a modified Koenigs-Knorr procedure. 4-Bromophenol (4.2 g, 24 mmol) was dissolved in dry acetonitrile (50 ml) to which was added 2,6-lutidine (2.3 ml, 23 mmol), Ag 2 0 (5 g, 22 mmol) and CaS04 (10 g). After 30 minutes stirring, compound (1) (5.0 g, 12 mmol) dissolved in acetonitrile (25 ml) was added slowly and allowed to stir in the dark for 16 hours. The solution was filtered through Celite and the solvent evaporated in vacuo, the residue dissolved in CH 2 C1 2 (100 ml) and washed with water (100 ml), 0.1 M HC1 (100 ml) and 0.1 M NaOH (100 ml). The organic layer was dried over MgS04, filtered and the solvent evaporated in vacuo. The residue was passed 45 through a silica gel column in 2:1 hexane/ethyl acetate; fractions containing the desired product were pooled and concentrated and the residue crystallized from ethanol to give white needles (1.1 g, 3.3 mmol, 28%). This material (0.52 g, 1.6 mmol) was deacetylated with sodium methoxide/methanol, and the product glucoside (5) was crystallized from ethanol to produce fine, white needles (0.20 g, 37%). M.p. 175-176°C (Lit (56) m.p. 175-176°C). *H nmr (D20): 8 7.50 (d, 2 H, J 9 Hz, H(3'), H(5')), 8 7.03 (d, 2 H, J 9 Hz, H(2'), H(6')), 8 5.05 (d, 1 H, J i > 2 8 Hz, H(l)), 8 3.90 (dd, 1 H, J 6 a , 6 b 12 Hz, J 5 , 6 a 2 Hz, H(6a)), 8 3.73 (dd, 1 H, J 6 a > 6 b 12 Hz, J 5 j 6 b 5 Hz, H(6b)), 8 3.60-3.40 (4 H, H(2), H(3), H(4), H(5)). Anal. calc. for Ci 2H 1 506Br; C, 43.01%; H, 4.52%; Found: C, 42.30%; H, 4.72%. 3'-Chlorophenyl-6-D-glucopvranoside (7) The acetylated compound (6) was prepared according to Koenigs and Knorr from (1) (2.5 g, 6.1 mmol) and 3-chlorophenol (1.62 g, 12.6 mmol), purified through a silica gel column (2:1 hexane/ethyl acetate) and crystallized from ethyl acetate/hexane to produce white crystals (0.7 g, I. 5 mmol, 25%). M.p. 131-132°C (Lit (84) m.p. 131.5-133.5°C). This material (0.25 g, 0.54 mmol) was deacetylated with sodium methoxide/methanol and the product glucoside (7) was recrystallized from ethanol to produce white crystals (0.12 g, 0.41 mmol', 76%). M.p. 181-182°C (Lit (84) m.p. 179-180°C). *H nmr (D20): 8 7.35-6.98 (4 H, H(2'), H(4'), H(5"), H(6')), 8 5.08 (d, 1 H, J l j 2 8 Hz, H(l)), 8 3.90 (dd, 1 H, J 6 a < 6 b 12 Hz, J 6 & j 5 2 Hz, H(6a)), 8 3.69 (dd, 1 H, J6a,6b 1 2 H z » J5,6b 6 H z « H(6b)), 8 3.60-3.40 (4 H, H(2), H(3), H(4), H(5)). Anal. calc. for C i 2 H 1 5 0 6 C l ; C, 49.58%; H, 5.20%; Found: C, 49.38%; H, 5.35%. 3'-Cyanophenyl-B-D-glycopyranoside(9) The acetylated compound (8) was prepared according to Koenigs and Knorr from (1) (2.5 g, 6.1 mmol) and 3-cyanophenol (1.5 g, 13 mmol), purified through a silica gel column (2:1 hexane/ethyl acetate) and recrystallized from ethyl acetate/hexane to produce white crystals (0.8 g, 1.78 mmol, 29%) M.p. 125-126.5°C (Lit (84) m.p. 126-127°C). This material (0.25 g, 0.55 46 mmol) was deacetylated with sodium methoxide/methanol and the product glucoside (9) was crystallized from ethanol to produce white crystals (0.10 g, 0.36 mmol, 65%). M.p. 162-163°C (Lit (84) m.p. 167-168°C). lH nmr (D20): 8 7.53-7.35(4 H, H(2'), H(4'), H(5'), H(6')), 8 5.13 (d, 1 H, J 1 > 2 7 Hz, H(l)), 8 3.91 (dd, 1 H, J 6 a > 6 b 12 Hz, J 5 > 6 a 2 Hz, H(6a)), 8 3.70 (dd, 1 H, J6a,6b 1 2 H z - J5,6b 5 Hz H(6b)), 8 3.60-3.40 (4 H, H(2), H(3), H(4), H(5)). Anal. calc. for C i 3 H 1 5 N 0 6 ; C, 55.51%; H, 5.38%; N, 4.98%; Found: C, 54.49%; H, 5.46%; N, 4.80%. 4'-Methoxyphenyl-B-D-glucopyranoside (11) The acetylated compound (10) was prepared according to Koenigs and Knorr from (1) (2.5 g, 6.1 mmol) and 4-methoxyphenol (1.5 g, 12 mmol), purified through a silica gel column (3:1 hexane/ethyl acetate) and recrystallized from ethyl acetate/hexane to produce white crystals (1.1 g, 2.4 mmol, 39%). M.p. 94-95°C. This material (0.50 g, 1.1 mmol) was deacetylated with sodium methoxide/methanol and the product glucoside (11) was crystallized from ethanol to produce fine white needles (0.22 g, 0.77 mmol, 70%). M.p. 175-175.5°C (Lit (84) m.p. 175-177° C); *H nmr (D20): 8 7.07 (d, 2 H, J 10 Hz, H(3'), H(5')), 8 6.83 (d, 2 H, J 10 Hz, H (2'), H(6')), 8 4.96 (d, 1 H, J 1 > 2 7 Hz, H(l)), 8 3.88 (dd, 1 H, J 6 a , 6 b 13 Hz, J 5 , 6 a 2 Hz, H(6a)), 8 3.76 (s, 3 H, H M e ) , 8 3.69 (dd, 1 H, J 6 a ? 6 b 12 Hz, J 5 > 6 b 6 Hz, H(6b)), 8 3.60 - 3.38 (4 H, H(2), H(3), H(4), H(5). Anal. calc. for C 1 3 H 1 8 C 7 ; C, 54.54%; H, 6.34%; Found: C, 54.26%; H, 6.24%. 3'.4'-Dimethylphenyl-B-D-glucopyranosidefl3) The acetylated compound (12) was prepared according to Koenigs and Knorr from (1) (2.5 g, 6.1 mmol) and 3,4-dimethylphenol (1.5 g, 12 mmol), purified through a silica gel column (2:1 hexane/ethyl acetate) and recrystallized from ethyl acetate/hexane to produce white crystals (0.7 g, 1.5 mmol, 25%). M.p. 118-118.5°C This material (0.28 g, 0.62 mmol) was deacetylated with sodium methoxide/methanol and the product glucoside (13) was crystallized from ethanol to produce white crystals (0.090 g, 0.31 mmol, 50%). M.p. 173-173.5°C; *H nmr (D20): 8 7.15 (d, 1 H, J5>t6- 8 Hz, H(5')), 8 6.91 (d, 1 H, J 2 ' j 6 ' 3 Hz, H(2')) 8 6.86 (dd, 1 H, J 5 - > 6 - 8 Hz, J 2 - > 6 ' 3 Hz, 47 H(6')), 8 5.05 (d, 1 H, J 1 > 2 7 Hz, H(l)), 8 3.90 (dd, 1 H, J6&t6h 12 Hz, J 5 > 6 a 2 Hz, H(6a)), 8 3.71 (dd, 1 H, J 6 a > 6 b 12 Hz, J 5 j 6 b 6 Hz, H(6b)), 8 3.62 - 3.40 (4 H, H(2), H(3), H(4), H(5), 8 2.20 (s, 3 H, H(Me)), 8 2.15 (s, 3 H, H(Me)). Anal. calc. for C i 4 H 2 0 O 6 ; C, 59.14%; H, 7.09%; Found: C, 58.84%; H, 7.13%. 2. Enzymology 2.1 General Procedures All absorbance measurements were made on a Pye-Unicam PU-8800 spectrophotometer or a Perkin Elmer Lambda 2 spectrophotometer. Temperatures were maintained by a circulating water bath. Unless otherwise specified, 1 cm pathlength quartz cells were used for absorbance measurements. 2.2 Enzyme Isolation and Purification Agrobacterium B-glucosidase was isolated and purified by T. Andrews, M. Namchuk and K. Rupitz from E. coli by a modification of the method utilized for isolation of the native enzyme from Agrobacterium (55). E. coli cell paste containing cloned enzyme was provided by Dr. R.A.J. Warren and Dr. D. Trimbur, Department of Microbiology, University of British Columbia. The pNZ1070 plasmid (62) containing the Caldocellum ji-glucosidase gene was generously provided by Dr. P L . Bergquist, Department of Cell Biology, University of Auckland. The cloned enzyme was expressed in E. coli JM101 by Dr. D. Trimbur. Cells containing the pNZ1070 plasmid were grown overnight in 2 ml of LB amp at 37°C from which one milllihtre of the culture was used to inoulate 1 L of M56 salts (85) supplemented with 0.6% glycerol, 0.2% casein and 100 ug/ml ampicillin. After 6 hours the culture was transfered to a 110-L fermentor containing 60 L of the same medium plus 0.1 mM IPTG, and grown until 2 ODgoo- Cells were collected by centrifugation at 31 OOOg. Enzyme was isolated and purified from these cells using a modification of the method used for preparation of the Agrobacterium enzyme. Cell extracts were prepared by grinding cell paste with 2.5 times their weight in alumina (86). The extraction 48 buffer was 50 mM sodium phosphate, pH 6.5, 10 mM 2-mercaptoethanol. The alumina was pelleted by centrifugation (10 min, 5000 g) and the supernatant retained. Nucleic acids were removed by precipitation with 1.5% streptomycin sulphate (3 h at 4°C) followed by centrifugation (20 min, 17000 g). The majority of the (mesophilic) E. coli proteins were precipitated from the extract by heating (1 h at 70°C) and pelleted by centrifugation (2 h, 17000 g). The resultant supernatant was dialyzed for 14 to 16 hours against 50 mM TrisHCl buffer, pH 8.0 (18 L buffer/L supernatant; dialysis buffer changed after 7 to 8 hours) and loaded onto a column of DE53 cellulose (2.6 x 60 cm) previously equilibrated with 50 mM TrisHCl buffer, pH 8.0. The bound proteins were eluted with a 2 x 1.4 L linear gradient of 0-0.5 M sodium chloride in the starting buffer. Fractions (10 ml) containing B-glucosidase (eluting at approximately 0.1 M sodium chloride) were pooled and concentrated using Amicon Centriprep 10 centrifuged ultrafiltration devices. The protein (in 1 ml fractions) was loaded onto a Sephacryl S-200 gel filtration high-resolution column (2.6 x 90 cm) equilibrated with 50 mM phosphate, pH 6.5. Fractions containing the highest protein concentrations were pooled and stored with 1 mM sodium azide. The purity of the protein was judged by SDS-PAGE. 2.3 Measurement of Enzyme Concentration Stock Caldocellum B-glucosidase enzyme concentration was determined from the 2,4-dinitrophenolate burst released upon treatment of the enzyme with the covalent inactivator 2',4'-dinitrophenyl-2-deoxy-2-fluoro-B-D-glucopyranoside (DNP2FG). The enzyme (approximately 2 mg/ml) in 50 mM sodium phosphate buffer, pH 6.5 was equilibrated at 37°C and DNP2FG added to a final concentration of 0.4 mM. A blank sample without enzyme was similarly prepared. Phenolate release was monitored in both samples spectrophotometrically at 400 nm until absorbance levelled off. From the absorbance difference, based upon an extinction coefficient of 10.91 cm"l mM'l and an enzyme molecular weight of 53 300 g moH, enzyme concentration was determined. 49 2.4 Extinction Coefficients All extinction coefficients were determined under conditions (temperature, pH, buffer) of the particular assay for which they are relevant using matched quartz cells. The assay wavelength at which extinction coefficients werre measured was chosen to be the point of maximal absorbance difference between phenol and glucoside, determined from a wavelength scan of the phenol and glucoside. Phenols and glucosides were dried in vacuo over P2O5, weighed, and dissolved in a known volume of buffer to create stock solutions. From the stock solutions a series of 5 solutions were prepared by diluting with bufferr to give absorbance values ranging from 0 to 0.2. Accurate absorbance values were measured at the assay wavelength and a plot of the measured absorbance values against the concentration of phenol or glucoside gave as the slope the extinction coefficient according to Beer's law: A = ebc (2) where A is the absorbance, E is the extinction coefficient (cm"* M"*), b is the cell path length (cm) and c is the concentration of the solution (M). 2.5 Kinetic Measurements Matched, 1-cm quartz cells were used in those experiments in which the glycoside background absorbance was relatively small (< 0.5); this includes all kinetics in which the phenol substituents were 2-chloro-4-nitro-, 4-chloro-2-nitro-, 4-nitro- and 3-nitro-. Due to high extinction coefficients for the glycoside in some cases, solutions containing high concentrations of glycoside were found to absorb too intensely to be accurately measured using a 1-cm pathlength cell. A 1-mm pathlength cell was used to reduce the background absorbance to levels low enough to allow for accurate measurements; this includes all kinetics in which the phenol substituents were 3,5-dichloro-, 3-cyano-, 3-chloro-, 4-bromo-, 2-napthyl-, 4-methoxy- and 3,4-dimethyl-. Rates of enzyme catalyzed hydrolysis were determined by incubating the appropriate concentration of substrate in buffer at 50°C in the appropriate cuvette located in the thermostated block of the spectrophotometer. Reaction was initiated by the addition of enzyme (typically 10 50 ul) from a syringe and the reaction was monitored at the appropriate wavelength. Rates were determined at 7-10 different substrate concentrations ranging from (whenever possible) approximately 0.15 times the value of the K \ j ultimately determined to 7 times the value. The extinction coefficient difference (AE) between phenol and glucoside was used to convert the observed rate of change of absorbance (AA/min) during enzymatic hydrolysis to rate of phenol release (v): v = AA/min-r Ae (3) Values of K M and kc a t were determined from these rates by computer fitting using the program GraFit (69). Rates of hydrolysis of substrates in which the phenol substituents were phenyl-, 4-methoxy- and 3,4-dimethyl were determined by a stopped assay system. Cells containing substrate in 50 mM MES/145 mM sodium chloride were equilibrated to 50°C and reaction initiated by addition of enzyme. The reaction was stopped by addition of 1.5 M sodium hydroxide causing the pH to rise to approximately 11. The extent of substrate hydrolysis was determined spectrophotometrically from the quantity of released phenolate. Cellobiose hydrolysis was determined using a stopped, coupled assay system. The appropriate concentration of cellobiose was incubated in buffer at 50°C and reaction initiated by the addition of enzyme from a syringe. The reaction was terminated after 4 min by quickly heating to 100°C and maintaining this temperature for 4 minutes. Glucose release was determined by means of a coupled assay using the method of Schaeter (87). As above, rates were determined at 7-10 different substrate concentrations ranging from approximately 0.15 to 7 times the value of the K M ultimately determined. 2.6 Cysteine Titration Free thiols in the Agrobacterium and Caldocellum B-glucosidase were determined by titration with 5,5'-dithio-bis(2-nitrobenzoic acid) Oillman's Reagent). (88) All measurements were carried out in 5.4 M guanidine hydrochloride/20 mM HEPES/0.15 mM Ellman's reagent, pH 7.0 51 at 37°C for the Agrobacterium B-glucosidase, pH 6.5 at 50°C for the Caldocellum B-glucosidase. The extinction coefficient for free thiol reaction with Ellman's reagent at 412 nm was determined from a L-cysteine standard curve and found to be 1.25 x 104 cnr 1 M" 1 at 37°C and pH 7.0, 1.13 x 104 cm-1 M" 1 at 50°C and pH 6.5. 2.7 pH Dependent Stability and Activity The pH stability of the Caldocellum B-glucosidase was estimated by incubating samples of the enzyme at 50°C and 70°C in 50 mM buffer/145 mM sodium chloride at a series of pH values, removing aliquots at different times and assaying under standard conditions. Buffers used for incubation were acetate (pH 4.0, 5.0, 5.5), MES (pH 5.5, 6.0, 6.5), PIPES (pH 6.5, 7.0), HEPES (pH 7.0, 7.5, 8.0), AMPSO (pH 8.5, 9.0) and CAPSO (pH 9.5, 10.0). Standard assay conditions were 10 mM 4-nitrophenyl-B-D-glucopyranoside (PNPG) as a substrate in 50 mM phosphate buffer, pH 6.5 at 50°C. Subsequent determinations of and K M were made at pH values at which the enzyme was stable (>95% activity) over a 5 minute period. The pH dependence of the enzyme activity was investigated by determining and K M for PNPG in 50 mM MES or HEPES buffer/145 mM sodium chloride over a series of pH values from 5.00 to 8.52 at 50°C using the stopped assay technique described above. An extinction coefficient of 18.2 mM'lcm"! for the 4-nitrophenol was measured under these conditions. 2.8 Temperature Dependent Stability and Activity Thermal stability was estimated by incubating samples of the enzyme in 50 mM MES buffer/145 mM sodium chloride, pH 6.5 at a series of temperatures from 50° to 90°C, removing aliquots at different times and assaying under standard conditions. Standard assay conditions were 10 mM 4-nitrophenyl-B-D-glucopyranoside as a substrate in 50 mM phosphate buffer, pH 6.5 at 50°C. 52 Subsequent determination of and K M were made at temperatures at which the enzyme was stable (>95% activity) over a 5 minute period. The thermal dependence of enzyme activity was investigated by determining kc a t and K \ j in 50 mM MES buffer/145 mM sodium chloride, pH 6.5 over a series of temperatures ranging from 25° to 75°C. The standard assay described above was used for measurement of activity up to 60°C. The stopped assay described above was used for activity measurements beyond 60°C, with absorbance measurements for all samples made at 50°. The value of 18.2 mM'icm'i for the 4-nitrophenol measured above was used for the extinction coefficient in the stopped assay measurements. Table 6: 4-Nitrophenol Extinction Coefficient From 25°C to 75°C at pH 6.5 for Direct Assay Measurements Temperature (°C) £oheno! x 1 0 * 3 ( M - 1 c™"1) 25.9 4.354 37 4.964 43.1 5.573 51.1 6.335 54.9 7.359 2.9 Circular Dichroism Measurements All circular dichroism measurements were made using a Jasco J-710 spectropolarimeter using a 1-mm path length thermostated cylindrical quartz cell connected to a circulating water bath which maintained the cells at the appropriate temperature. The buffer employed for all circular dichroism measurements was 50 mM MES buffer, pH 6.5 containing urea (0 to 8 M) (Sigma, Molecular Biology Reagent) added to effect denaturation. All urea solutions were prepared fresh each day. Protein concentrations varied from 0.05 to 0.1 mg/ml for different experiments; samples were allowed to thermally equilibrate for 1 hour prior to CD measurements. 53 Incubation for periods of up to 4 hours showed no significant changes in CD measurements after 1 hour. CD spectra were obtained over the region from 260 to 210 nm with data recorded at 0.25 nm intervals with a time constant of 1 s and a 1 nm constant spectral bandwidth. Data was averaged over four repetitive scans. Data was smoothed and normalized for protein concentration and optical path length and the mean residue ellipticity, [0], was calculated by using a mean residue weight of 111.3 for the Agrobacterium 6-glucosidase and 117.1 for the Caldocellum 6-glucosidase and expressed in degcm /^dmoL 54 APPENDIX I BASIC CONCEPTS OF ENZYME CATALYSIS 1. Fundamental Equations of Enzyme Kinetics The Michaelis-Menten treatment of enzyme kinetics is based on the following scheme: K M c^at E + S = £ : ES - — - E + P The catalytic reaction is divided into two steps. First, the enzyme and substate combine in a v rapid, reversible process to give an enzyme-substrate complex, ES, held together by physical forces, no chemical changes having taken place. In the second step, bound substrate reacts to produce product and is released from the enzyme (89). The rate equation derived from this scheme (the Michaelis-Menten equation) is as follows: v = ([E] 0[S]k c a t)/(KM + [S]) (4) v is the rate of reaction (measured by substrate breakdown or product release), [E] 0 is the total enzyme concentration, [S] is the substrate concentration, k ^ is the catalytic constant and Kjyf is termed the Michaelis constant At low concentrations of substrate, where [S] « K M the equation is approximated by v = (k c a t/KM)[E] 0[S] (5) and v is linearly dependent upon substrate concentration. At high concentrations of substrate, where [S] » K M , V approaches a limiting value, vmax = fccatfElo (6) K M is the apparent dissociation constant and may be treated as the overall dissociation constant of all enzyme bound species by virtue of the relationship 55 K M = «E][S])/Z[ES] (7) KM is a measure of the amount of enzyme that is bound in any form and reflects the stability of the enzyme-substrate complex. Lower values of K M indicate tighter binding df substrate to enzyme. kcat/KM can be derived from equation 4 and at low substrate concentrations where [E]0 is approximated by the amount of free enzyme, k^t/KM becomes the apparent second order rate constant for reaction of free enzyme with free substrate. kcat^M Is m u s m indicator of the catalytic efficiency of the enzyme. For purposes of graphical representation, a Lineweaver-Burk plot is frequently used for the display of data. This technique converts from the Michaelis-Menten equation into a linear plot which can highlight deviations in the data from expected behaviour. The relationship is derived by inverting both sides of the equation 1/v = 1/V m a x + KM/(Vm a x[S]) (8) A graph of 1/v vs. 1/[S] gives a straight line with a slope of ^Mf^max' a y-intercept of 1/V max. and an x-intercept of -1/KM-2. Transition State Stabilization and Enzvme Catalysis Enzymes represent a class of highly specific catalysts capable of greatly accelerating rates of reaction and possessing high specificity for a particular substrate. As with all catalysts, enzymes function by lowering the transition state energy barrier for product formation, thereby increasing the rate of reaction. Enzymes achieve their specificity by having a well defined active site with several binding sites in the correct spatial orientation to match each binding group on the substrate. The changing nature of the substrate through the course of the reaction dictates that the enzyme can only be complementary to one form of substrate. Maximal efficiency is achieved by the enzyme if it is complementary to the transition state structure of the substrate rather than to the ground state. 56 Figure 14 shows how a typical enzyme reaction can be defined in terms of free energy changes along the pathway. E + S K M kcat T ES » E • P Reaction Coordinate Figure 14: Free energy diagram for a typical enzymic reaction (solid line) and its corresponding uncatalyzed reaction (dashed line). Kinetic parameters relate to the free energy changes which occur along the reaction pathway as follows, for the sample case shown in Figure 14: kc a t = (kT/h)exp(-AGc*/RT) (9) W M = (kT/h)exp(-AGx*/RT) (10) where k is the Boltzmann constant and h is Planck's constant K \ j is inversely related to the energy difference AGES-The effect of stabilization of the ground state and the transition state upon the free energy diagram is depicted in Figure 15. If the ground state is stabilized by some additional binding interaction which contributes A G R to the binding energy, mis has the effect of increasing AGES and A G C ^ by a value of AGR as indicated in Figure 15. An increase in AGES results in a decrease in Kj^, resulting in tighter binding and from equation 9 the increase in AGC$ can be seen 57 to decrease and result in a slower reaction. If the transition state is stabilized by this additional binding interaction, then this will have the effect of decreasing A G ^ and A G j i by A G R as indicated in Figure IS, resulting in an increase in both k ^ and kcat/^M (from equations 9 and 10), resulting in faster reaction. Thus, transition state stabilization as opposed to ground state stabilization results in an increase in both enzymic rate and enzyme efficiency and thus is of the utmost importance to the catalytic ability of enzymes. a ACr Figure 15: Free energy diagram of an enzymic reaction involving: (a) Maximum enzyme-substrate complimentarity at the ground state; (b) Maximum enzyme-substrate complementarity at the transition state. 58 APPENDIX H THE USE OF CIRCULAR DICHROISM IN THE STUDY OF PROTEINS 1. General Principles of Circular Dichroism Sp^tropolajunetry Circular dichroism (CD) and the related phenomenon of optical rotary dispersion (ORD) are valuable tools in the elucidation of structure and conformation of proteins. Both CD and ORD are derived from the interaction of polarized light with asymmetric molecules. The use of ORD for technical reasons developed before the use of CD, despite the fact that for many purposes CD spectra have advantages over ORD spectra. For mis reason, only the phenomena of CD will be detailed here. Light is composed of oscillating electric and magnetic fields. Light can be polarized such that the electric field is confined to a single plane, as is the magnetic field, and these planes are perpendicular to each other. This linearly polarized light can be resolved into two circularly polarized components the electric vectors of which separately describe a left- or right-handed helix (Figure 16). The magnetic vectors similarly describe a helix. The summation of the left-and right-handed helix produces plane-polarized light. Figure 16: Electric vectors of a right circularly polarized light at constant time and variable distance from the light source (90). When linearly polarized light is passed through an optically active material it will be partly absorbed if its wavelength is close to one of the material's natural frequencies. One of the 59 circularly polarized components of the light will be absorbed to a greater extent than the other, creating elliptically polarized light (Figure 17). The angle whose tangent is the ratio of the minor axis to major axis of the elliptically polarized light is defined as the ellipticity (6). This unequal absorption of left- and right-handed circularly polarized light is the phenomena of circular dichroism. 6 = arctan (B'O/A'O) Figure 17: The electric vectors of the left- and right-handed circularly polarized light sum to an electric vector which traces an elliptical path upon passing through an optically active medium in which the left- and right-handed circularly polarized light are unequally absorbed (90). A CD experiment consists of alternately passing left- and right-handed circularly polarized light through a sample and determining the difference in absorbance between the left- and right-handed circularly polarized light AA = A j _ - A R (11) 60 Ultimately, this difference is absorption can be related to the difference in extinction coefficients for the left- and right-handed circularly polarized light The absorption difference is generally converted to the molar ellipticity [6 ], the common method for reporting the intensity of CD bands G = 2.303AA.180/4JC (12) [ei = 100 8/Cd (13) where C is the concentration (in molL'l) and d is the pathlength (in cm"l); the units of [6] are deg-cm -^dmol"!. Compounds which are not optically active will absorb left- and right-handed circularly polarized light to an equal extent and produce therefore a net CD of 0. 2. Application of Circular Dichroism to the Study of Proteins The CD of proteins is primarily the CD of the amide chromophore. The amide chromophore has as its first peak absorbance a band at approximately 220 nm with several other absorption bands at lower wavelengths. Non-amide bands which occur primarily at longer wavelengths can provide additional information but are generally much less intense. The CD spectrum measured for a particular molecule is a reflection of the excited state of the molecule and is often a sum of several different absorbing groups. The wavelength and intensity of each band reflects the environment of the group responsible for the band. In a polymer and absorption band may be the sum of a large number of absorbing groups and will reflect and be sensitive to the environment of all of the groups which comprise it The well-ordered nature of protein secondary structures results in a particularly large additive effect of multiple absorbing groups in regular and defined spatial orientation, to produce intense bands which reflect the extent and nature of the secondary structure. This makes CD particularily sensitive to the study of protein secondary structure as it can be used to accurately measure the quantity of a particular secondary structure and is useful for monitoring changes in 61 existing secondary structure. Based upon theoretical model spectra and empirically derived sample spectra for particular secondary structures, it is possible to determine quantities of particular secondary structure types within most proteins. In particular for a-helices, such approximations can produce very reliable estimates of protein secondary structure content. The technique is also very useful for measuring changes in structure which occur as a result of a mutation in the protein or changes in environmental conditions. 62 APPENDIX m GRAPHICAL REPRESENTATION OF KINETIC DATA 1. Lineweaver-Burk Plots for Hydrolysis of Aryl Glycosides by Caldocellum ft-Glucosidase at 50°C in pH 6.5 MES Buffer 1/[S] (1/mM) Figure 18: Hydrolysis of 2-chloro-4-nitrophenyl-p>-D-glucopyranoside. Enzyme concentration = 2.70 x lf>3 mg/ml. I i I < I i I * I * I i I i I * I i I | 1/[S] (1/mM) Figure 19: Hydrolysis of 4-chloro-2-nitrophenyl-f3-D-glucopyranoside. Enzyme concentration = 2.67 x 10-4 mg/ml. 63 0.05 I 1 1 r 0 I i I i I i I I 0 2 4 6 8 1/[S] (1/mM) Figure 20: Hydrolysis of 4-nitrophenyl-B-D-glucopyranoside. Enzyme concentration = 2.91 x HHmg/ml. 0 I • i • i i i i i i i i i i i i i • i • i 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 1/[S] (1/mM) Figure 21: Hydrolysis of 3,5-dichlorophenyl-B-D-glucopyranoside. Enzyme concentration = 6.98 x 10"4 mg/ml. 64 § 0.022 § 0.02 0 2 4 6 1/[S] (1/mM) Figure 22: Hydrolysis of 3-nitrophenyl-f3-D-glucopyranoside. Enzyme concentration 1.07 x l O 3 mg/ml. 0.06 0.04 h £ 0.02 0 2 4 6 8 • l/[S](l/mM) Figure 23: Hydrolysis of 3-cyanophenyl-fJ-D-glucopyranoside. Enzyme concentration 3.82 x 10-3 mg/ml. 65 0.05 j 1 -r 0 2 4 6 8 ' . 1/[S] (1/mM) Figure 24: Hydrolysis of 3-chlorophenyl-B-D-glucopyranoside. Enzyme concentration 3.82 x 10-3mg/ml. 0.05 | 1 1——i 1 1 1 1 1 1——| 0.04 o I i i i I i i i I i_ 0 2 4 1/[S] (1/mM) Figure 25: Hydrolysis of 4-bromophenyl-B-D-glucopyranoside. Enzyme concentration 3.82 x 10-3 mg/ml. 66 0 2 4 6 8 10 1/[S] (1/mM) Figure 26: Hydrolysis of 2-napthyl-p>-D-glucopyranoside. Enzyme concentration = 5.07 x lf>3 mg/ml. 0.14 0.12 0.1 nmg/u 0.08 0.06 0.04 0.02 o. i i i i i i i i i i i i i i i i I 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 1/[S] (1/mM) Figure 27: Hydrolysis of phenyl-P-D-glucopyranoside. Enzyme concentration = 5.46 x lf>3 mg/ml. 67 b 1/tS] (1/mM) ' l/[S] (1/mM) Figure 28: Hydrolysis of 4-methoxyphenyl-B-D-glucopyranoside. Enzyme concentration = 3.82 x 10 3 mg/ml in direct assay (a); 1.90 x 10 - 3 mg/ml in stopped assay (b). l/IS](l/mM) l/[S](l/mM) Figure 29: Hydrolysis of 3,4-dimemylphenyl-B-D-glucopyranoside. Enzyme concentration = 3.82 x 10"3 mg/ml in direct assay (a); 1.90 x 10"3 mg/ml in stopped assay (b). 68 0 I 1 1 1 1 1 1 I I L 0 2 4 1/[S] (1/mM) Figure 30: Hydrolysis of 4-nitrophenyl-P-D-galactopyranoside. Enzyme concentration 5.46 x lfr 4 mg/ml. 0 I i i- i j i i i i i i i , • i • i • i • I 0 2 4 6 8 10 12 14 16 18 20 l /[S](l /mM) Figure 31: Hydrolysis of 4-nitrophenyl-|5-D-fucopyranoside. Enzyme concentration 3.28 x 104 mg/ml. 69 0 2 4 6 8 10 12 14 16 18 20 22 l/[S](l/mM) Figure 32: Hydrolysis of 4-nittophenyl-fJ-D-mannopyranoside. Enzyme concentration 3.28 x 10-2 mg/ml. if 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1/[S] (1/mM) Figure 33: Hydrolysis of 4-nitrophenyl-p,-D-xylopyranoside. Enzyme concentration 2.98 x 10-3 mg/ml. 70 0.00 1 i i i • i • i . i • i • i • i • I 0 2 4 6 8 10 12 14 16 18 20 22 24 26 1/[S] (1/mM) Figure 34: Hydrolysis of 4-nitrophenyl-a-L-arabinopyranoside. Enzyme concentration 2.98 x 10-3 mg/ml. 0 0.1 0.2 0.3 0.4 1/[S] (1/mM) Figure 35: Hydrolysis of cellobiose. Enzyme concentration = 1.00 x 10-3 mg/ml. 71 2. Lineweaver-Burk Plots for Hydrolysis of 4-Nitrophenyl-5-D-Glucopyranoside by Caldocellum ft-Glucosidase at 50°C I _i I i _ I i i • i 0 2 4 6 8 1/[S] (1/mM) Figure 36: Hydrolysis of PNPG in pH 5.00 MES buffer. Enzyme concentration = 1.35 lfr 4 mg/ml. I i J i I i i i i 0 2 4 6 8 1/[S] (1/mM) Figure 37: Hydrolysis of PNPG in pH 5.61 MES buffer. Enzyme concentration = 1.35 10-4 mg/ml. 72 I i I 1 I • i • i 0 2 4 6 8 1/[S] (1/mM) Figure 38: Hydrolysis of PNPG in pH 6.04 MES buffer. Enzyme concentration = 1.35 x 10-4 mg/ml. I i _ I i I •• i • i 0 2 4 6 8 1/[S] (1/mM) Figure 39: Hydrolysis of PNPG in pH 6.48 MES buffer. Enzyme concentration = 1.35 x 1(H mg/mL 73 0 2 4 6 8 1/[S] (1/mM) Figure 40: Hydrolysis of PNPG in pH 7.10 MES buffer.' Enzyme concentration = 1.35 x 10"4 mg/ml. I i I i I i 1 i 0 2 4 6 8 1/[S] (1/mM) Figure 41: Hydrolysis of PNPG in pH 6.48 HEPES buffer. Enzyme concentration = 1.35 xlO" 4 mg/ml. 74 0 I i I i I i I I 0 2 4 6 8 1/[S] (1/mM) Figure 42: Hydrolysis of PNPG in pH 6.97 HEPES buffer. Enzyme concentration = 1.35 x 10"4 mg/ml. 1/[S] (1/mM) Figure 43: Hydrolysis of PNPG in pH 7.46 HEPES buffer. Enzyme concentration = 1.35 x 10"4 mg/ml. 75 1/[S] (1/mM) Figure 44: Hydrolysis of PNPG in pH 7.98HEPES buffer. Enzyme concentration = 1.35 x 10"4 mg/ml. 0 1 2 3 4 1/[S] (1/mM) Figure 45: Hydrolysis of PNPG in pH 8.52 HEPES buffer. Enzyme concentration = 1.35 x 10-4 mg/ml. 76 3. Lineweaver-Burk Plots for Hydrolysis of 4-NitrophenyI-pVD-Glucopyranoside by Caldocellum ft-Glucosida.se in pH 6.5 MES Buffer 0.6 | 1 1 r 0 2 4 6 1/[S] (1/mM) Figure 46: Hydrolysis of PNPG at 25.9°C. Enzyme concentration = 2.607 x 10"3 mg/ml. 1/[S] (1/mM) Figure 47: Hydrolysis of PNPG at 37.0°C. Enzyme concentration = 5.221 x 10'4 mg/ml. 77 l/[S](l/mM). Figure 48: Hydrolysis of PNPG at 4 3 . P C Enzyme concentration = 5.221 x 10"4 mg/ml. 0 2 4 6 1/[S] (1/mM) Figure 49: Hydrolysis of PNPG at 51.1°C. Enzyme concentration = 2.373 x 10 - 4 mg/ml. 78 0.06 0.04 I? £ 0.02 \-J I L J I L 2 4 6 1/[S] (1/mM) Figure 50: Hydrolysis of PNPG at 54.9°C. Enzyme concentration = 2.373 x 10'4 mg/ml. i 1 1 1 r J I L 4 6 8 1/[S] (1/mM) 10 Figure 51: Hydrolysis of PNPG at 60.2°C. Enzyme concentration = 2.23 x 10"4 mg/ml. 79 Figure 53: Hydrolysis of PNPG at 75.1°C. Enzyme concentration = 2.23 x 10"4 mg/ml. 80 APPENDIX IV SEQUENCE ALIGNMENT Abg - 0 I^ DDPhrrAARFPCDFLFCVATASFQIECSTICADCRKPSIWDAFCNWPGHVFCRHNCDIACDHYNRWEEDLDLIKEMCVE Cbg - 0 MDMSFPKGFLWGMTASYQIECAV^ErXKCESIWDRFTHQKRNILYCHNCDVACDHYHRFEEDVSLMKELGLK « * > > * * A * . * * * * A * • * * * _ _ * * A . . * * * * . * A . . . * * * * A * * * * * . * A * * * A . * A * * A * A . Abg - 79 AYRFSLAWRIIPrXIFGPINEKGLDFYDRLVDGCKARGIIC^ Cbg - 73 AYRFSIAWTRIFPrXJFGTVNQKGLEFYDRLINKLVENGIEFWTLYHWDLPQKLQDIGGWANPEIVNYYFDYAMLVINR * * * * * A * * * * ***** A* ***A*****A * * * * * * * * * * * * * * * A * * A * Abg - 158 LGDRLDAVATFNEPWCAWLSHLYGVHAPGERhWEAALAAMHHINU^HGF_GVEASRHVAPKVPVGLVLNAHSAIPASD cbg - 152 YKDKVKKWCTFNEPYCTAFLGYFHGIHAFK:IKDFKVAMDWHSIJILSH_FKWKAVKENNIDVEVGITLNLTPVYLQTE . . * A * * * * * A * . . A * * A * * * * . A * A . . A * . A . * A * . * . . * . * . A * . * * A . * * AA Abg - 236 G EADLKAAERAFQFH NGAFFDPVFKGEYP AEMMEALGDRMPWEAEDLGIISQKI DWWGLNY Cbg - 230 RLGYKVSEIEREMVSLSSQLDN_QLFLDPVLKGSYPQKLLDYLVQKDI LD S_QKALSMQQEVKENFIFPDFLGINY A A * . * " . * * . * * A . A A A . * . A * A . * A * * Abg - 298 YTWRVADDATPGVEFPAT_^PAPAVSDV_l(TDIGViEWAPALHT Cbg - 304 YTRAVRLYDENSSWIFP IRWEHPA GEYTEMGWEVFPQGLFDLLIWIKESYPQIPIYTTENGAAYNDIVTEDGK * * * * * * * A A * * * * A * * A A * * * * * * * * * . * * . . * . * . * . Abg - 372 VNDQPRLDYYAEHLGIVADLIRIX.YFt1RGYFAWSlJ1DNFEWAEGYRMRFGLVHVDYETQVRTVKNSGKWYSALASGFP-Cbg - 377 VHDSKRIEYLKQHFEAARI(AIENGVDLRGYFWSLMDNFEWAMGYTKRFGIIWDYETQKRIKKDSFYFYQQYIKENS-* * * A A * . . . * * . . * . . A * * * * . * * * * * * * * * * . * * . . * * * A A . * * * * * * . * . , * . * . . A * . , . . . , . , Figure 54: Sequence alignment of the Agrobacterium (Abg) and Caldocellum (Cbg) |5-glucosidases. Aligned residues which are identically conserved are denoted by *. Aligned residues which are "similar" are denoted by A . Amino acids which are classified as "similar" are: A, S, T; D, E; N, Q; R, K; I, L, M, V; F, Y, W. Residues implicated in catalysis are indicated by o. Alignment performed by BLAST program (91). 81 G[rJTlWjrr*[Ss w v a D R I 11 t V s Q V V 8 G 0 1 p ren M r G B V T G M I C - - -r t i 0 X A H X L X G n N G • - -F L O X Q G R r X P I I I D N I M Y T A t n E H Y M y T A 0 P H G V M G K M E P A I L C X s 0 F Q G I r G I Z V T R 0 P G 0 s n KRP L N T T E c Q U l - - • - — r T H T P c S M V X D N A ? G - - — r s R T P L R V I N 0 A I G - -E N G D V 0 P T P E P IDIV [DJV e v ID A l J V [ A j c l O j y H R y H R x y HITIY Y[NJAY y H R y y H R y y HfK~v 0 L K I y H y H m: J U L D I "D~T|D D[2.S DT]A D L D L E Off A N M l P X . P K P I M Q T G T P - r — - r - A -y - — — - y -D E - -D E " " H G - - - - -RN T T " 103 K E N S P V Z S V D L N E S X L R E M D N Y ATJ1H 9 6 . G P I N E 8 E . - - G T VINJQ 6 S - - G E V t [ 95 C Z 86 G E 96 V E 98 - - - A E 467 S S 990 - - - - - - - - - - - - - - - - - - - - S S 1463 » > R H I I D[R"T1V D[R L{Z HTIL F a 153 H P I N "DH o P I b 126 H W D L P L T L M G D c 118 H W D L P 0 K L Q D I d 116 H T | D PF P E R L H E A e 125 H F 1 D T P E A I H S N t 116 K ID. T P E V A< H K D 9 126 H y E M P y G 1, V X N y h 128 H y E M y G L V E X H i 49? H W D L P Q A L Q D H j 1020 H w o L P 0 A L 0 D Z k 1493 H H D L P Q T L 0 D V O R V R R G O F T G P TjnrtLEl G D |G G W A S S R T V y E[F|A R R S T A H AlFjQ R HiA[S]P E I V N y_y_F D 0 F D F TT G M G N G N G N G G G[tlwjl."'T'b E M L D D F S V A Z D V D E H E R I D D S RLL|R E V A N V M A V Z N DlTlL A S G D R 1* D K IpJ K (VTT S E - V X P E P E 0 H R H - V E 'vyTj 1% v x v x V X a 203 E y A b 165 A V A c 157 K w I a 154 y w I e 163 y w T / 154 y w T 9 166 L w L h 168 R w L 1 •536 L w V 1 1059 F H M * 1532 F w I y AfTlMJN E P T F N E P T F N E ? E | T -r F N E T F N E T T N E j -T F(H]ET1 I F N E P TfTlN E P • r » - i 8 i - M E - A A L - F X - V A M - R F D X T F - D {• A K V F - D F E X V r - E A - E v y - X A - A i y - P G V A S F - P G N A P Y R P G T A P X a 250 Y D A 1 X s V S X X _ s v G I z - Y A N T s y T T L b 210 V E A S R H V A p X V P V G L V - N A H S A TIP A c 202 . V K A V X E N N i D V E V G : T - t N L T P V TIL Q d 200 V N L y X s M 0 i G G' Q i C i V - H "A L Q T V y p y e 209 V K L y K D X G y X G E I G V V - H A L P T X y p y / 200 V K L F X D G G y X G E I G V V - H A L P I X y P F 9 210 V K A C H S L L p E A K Z G N M - L L O G L V y p L h 212 i 582 V K A c H 0 M Z p D A 0 z G N M — 1. I. G A K If y p L w H H y N s H H R P Q Q 0 G H V G Z v Fl N S D W A E j 1105 * 1579 V H T y D E X y R 0 E 0 X G V Z S Z. s S T H H A E w K L y N D V y R A S 0 G G V I s Z T T S S D M A B m 250 2*3 240 250 24. 243 250 Hi 114 J i :S23 X K H N 0 N £ N H N H N R R R E M L S L H G r L Y A!F|F o 1:3 L | : LIS F T 2 F r\r c o F;A H P , r A H p ITJA H P E I T S t E C Q M V R S X K I t l . . ^ Q X L L 0 T « !|TO I T I A 1 V Y S O K T M E G V rls R E T H 1 C V r T i c i n j u r Y P G T M H R Y if PA : U T O iY P jD T M X N X [YJN E V M K T R V V L 0 A A A E 0 X P E X T T A P V LsA T I N K I P L P L P 3 . . . . . . 0 E V K E H r P Q E M X A I - S 0 F Q A L - E 0 T A I t L R P T r T r T B A E E E E B S E D E D A 0 A A A A X 0 X. R F R X 3:9 294 295 294 303 294 295 297 f>i ::39 :i~3 L R K L DIW : o!w H F O H T v N • V G i t M t Y G!: IN t r G : V! N T ] Y G i : | N r Y '• UL Y Y T r T i Y s r s jf Y V V T R V A V R L M K V V I V A K A E S G Y L T L P G T C D R C I R N S L S X . A - - - - -O O A T P O V E F P A T t t P A P A V -Y D E N S S M i r P I R N E H F A G - - -A Y H G X S E T X H N C D G T K G 3 S V A R L 0 G V G E E X . . F D A F S G E ? E 2 I H N G X G E X C 3 S K Y 0 I X G V G * R V A P D C X O C E S E f T H N A T G D X G G S X Y Q L X G V G Q R E F D V O E S I N K N . A Q C N t l M K I P l i P - - -O E A Q L E K T R G N I L N M V P N P A P O M T C I P S Y O T I G G - - - -X T P R L N P P S Y E D D Q E . . . . L N Y A T A I S S F D A D R G * 355 N L P T SfOlF G fTTJE F F [ f ]E -:* E V Y A P -324 S D V X T L O J : G 323 E Y - - E M G W! E V F T 339 G I E T l T 3 W S I Y p R -« 343 Y V P R T D W D W.I I Y ? E -*• 339 D V P R T O W 3 W M I Y p 0 -9 327 H L K S s TIW'G w a i D p .1 329 Y L E S s z'jr 3 H Q 1 D p v 709 F S Q H V N H V H P Q ! T S S j 1232 M A E E E "51? S H P S T A M N k 1706 V A S I A OJR S JiJ ? D S G S F -nmY - ? Q T R L - -rritmy MIX L - P T C J Y T T E - i f F T Y I T E v p vTiY[VJT E Y X K . T l T t E Y H X ' I Y I ? E ? t i t 11 _ T _ N N N N N N N E N Y f T i L SpTIl Y E HTJG QIHIF E x T j T T s - 'E N V V M N N H L H L H L H L T T Y I Y I - E l k A A A A A A_ I Y R N E!G v b~v RFD G!TT"M H;5"cprN V V 0 GjA N[V O G V pit D G VfTlV D GJAN , o | T | v V v ID TmplTlR IDfKTV PlLIR TTT] H S M S H S H S H S HfG WO I A R| 5 V Aft* S T V|W S -rw]s .15 A A S H S X A E W,-- c p s I T T s f Q l i L - |"5riL N|G YIT V T ) -[wjA T |GfF |TlE[R R F C T l H F G l» r G L X Y L T X H V D Y E T 0 Y V O Y E T Q F V D F P T Q Y V D F D T 0 Y V D F E T 0 Y V D R P P N Y V D R D D A H V r S O S H V F N N T F V N Y S D P G E G S L T G H C S L E 8 X - - - $ H R P S I P Figure 55: Multiple sequence alignment of Class 1 p^glycohydrolases: p-galactosidase from Sulfolobus solfataricus (a), {J-glycosidases from Agrobacterium faecalis (b) and Caldocellum saccharolyticum (c), phospho-|}-galactosidases from Lactobacillus casei (d), Streptococcus lactis (e) and Staphylococcus aureus (f), phospho-p-glucosidase from E. coli (g) and K chrysanthemi (h) and human LPH domans II (i), IE (j) and IV (k). Residues conserved in at least 6 of the 11 sequences are boxed. Residues implicated in catalysis are indicated by o. Alignment taken from El Hassouni, et al. (59). 83 REFERENCES (1) Koshland, D.E. (1953) Biol. Rev. 28,416. (2) Imoto, T., Johnson, L.N., North, A.C.T., Phillips, D.C. & Rupley, J.A. (1972), The Enzymes (Boyer, P. ed), vol 7,3rd edit., pg 665, Academic Press, New York. (3) Anderson, W.F., Grutter, M.G., Remington, S.J., Weaver, L.H. & Matthews, B.W. (1981) 7. Mo/, flu?/. 147,523. (4) Wakarchuk, W.W., Campbell, R.L., Sung, W.L., Davoodi, J. & Yaguchi, M. (1994) Protein Science 3,467. (5) Legler, G. & Harder, A. (1978) Biochim. Biophys. Acta. 524,102. (6) Bause, E. & Legler, G. (1974) Hoppe-Seyler's Z. Physiol. Chem. 335,438. (7) Herrchen, M. & Legler, G. (1984) Eur. J. Biochem., 138,527. (8) Withers, S.G., Street, LP. & Dolphin, D.H. (1987) J. Amer. Chem. Soc. 109,7530. (9) Stokes, T.M. & Wilson, LB. (1972) Biochemistry 11,1061. (10) Street, LP. & Withers, S.G. (1988) J. Amer. Chem. Soc. 110, 8551. (11) Blake, C.C.F., Mair, G.A., North, A C T . , Phillips, D.C. & Sarma, V.R (1967) Proc. R. Society London Ser. B. 167,365. (12) Weber, J.P. & Fink, A.L. (1980) J. Biol. Chem. 255,9030 (13) Chiang, Y. & Kresge, A J . (1985) J. Amer. Chem, Soc. 107,6363. (14) Sinnott, M.L. & Withers, S.G. (1974) Biochem. J. 143,751. (15) Pauling, L. (1946) Chem, Eng. News 24,1375. (16) Legler, G., Sinnott, M.L. & Withers, S.G. (1980) J. Chem, Soc. Perk. Trans. II, 1376. (17) Legler, G. & Julich, E. (1984) Carbohydrate Res. 128, 61. (18) McCarter, J.D., Adam, M J . & Withers, S.G. (1992), Biochem, J. 286,721. (19) Roeser, K.R. & Legler, G. (1981) Biochim. Biophys. Acta 657,321. (20) Brock, T.D. & Darland, G.K. (1970) Science, 169,1316. (21) Stetter, K.O. (1986) Thermophiles: General and Applied Microbiology (Brock, T.D., 84 ed), pg 33, John Wiley & Sons, New York. (22) Singleton, R. & Amelunxen, R.E. (1973) Bacteriol. Rev. 37,320. (23) Ray, P.H., White, D.C. & Brock, T.D. (1971) 7. Bacteriol. 106, 25. (24) Ray, P.H., White, D.C. & Brock, T.D. (1971) 7. Bacteriol. 108,227. (25) Wisdom, C. & Welker, N.E. (1973) 7. Bacteriol. 114,1336. (26) Inoue, K. (1977) 7. Gen. Appl. Microbiol. 23, 53. (27) Amelunxen, R.E. & Lins, M. (1968) Arch. Biochem. Biophys. 125,765. (28) Ward, O.P. & Moo-Young, M. (1988) Biotech. Adv. 6 39. (29) Argos, P., Rossmann, M.G., Grau, J.M., Zuber, H., Frank, G. & Tratschin, J.D. (1979) Biochem, 18,5698. (30) Serrano, L., Horovitz, A., Avron, B., Bycroft, M. & Fersht, A.R. (1990) Biochemistry 1990,9343. (31) Privalov, P.L. & Kechinashvili, N.N. (1974) 7. Moi. Biol. 86,665. (32) Alber,T. (1987) AfafMr.? 330, 41. . (33) Grutter, M.G. (1987) 7. Moi. Biol. 197,315. (34) Dang, L.X., Merz, K.M. & Kollman, P.A. (1989) 7. Am. Chem. Soc. I l l , 8505. (35) Geiger, A., Rahman, A. & Stillinger, F.H. (1979) 7. Chem, Phys. 70,263. (36) Stillinger, F.H. (1980) Science 209,451. (37) Ravishankar, G., Mezei, M. & Beveridge, D.L. (1982) Faraday Symp. Chem, Soc. 17, 79. (38) Privalov, P.L. & Gill, S.J. (1988) Adv. Protein. Chem. 39,191. (39) Baldwin, R.L. (1986) Proc. Natl. Acad. Sci. U.SA. 83, 8069. (40) Lira, W.A. & Sauer, R.T. (1989) Nature 339, 31. (41) Bull, H.B. & Breese, K. (1973) Arch. Biochem. Biophys. 158,681. (42) Ikai, A. (1980) 7. Biochem. 88,1895. (43) Matsumura, M., Becktel, W.J. & Matthews, B.W. (1988) Nature 334,406. (44) Feeney, R.E. (1980) Chemical Deterioration of Proteins (Whitaker, J.R. & Fujimaki, M. 85 eds), pg 1, Amer. Chem. Soc., Washington. (45) Weaver, L.H., Kester, W.R., Ten Eyck, L.F. & Matthews, B.W. (1976) Enzymes and Proteins from Thermophilic Microorganisms (Zuber, H. ed), pg 31, Birkhauser Verlag, Basel. (46) Hiraoka, Y., Segawa, T., Kuwajima, K., Sugai, S., Murai, N. (1980) Biochem. Biophys. Res. Commun. 95,1098. (47) Segawa, T., Sugai, S. (1983) J. Biochem. 93,1321. (48) Greenberg, N.A., Wilder, T. & Mahoney, R.R. (1985), J. Dairy Res., 52,439. (49) Reilly, A., Morgan, P., Davis, K., Kelly, S.M., Green, J., Rowe, A.J., Harding, S.E., Price, N.C., Coggins, J.R. & Kleanthous, C. (1994) J. Biol. Chem. 269,5523. (50) Turner, S.L., Ford, G.C., Mountain, A., Moir, A. (1992), Protein Eng. 5(6), 535. (51) Matthews, B.W. (1993), Protein Eng. 6 (supplement), 6. (52) Eijsink, V.G.H., Dijkstra, B.W., Vriend, G., van der Zee, J.R., Veltman, O.R., van der Vinne, B., van den Burg, B., Kempe, S. & Vemema, G. (1992) Protein Eng. 5(5), 421. (53) Kallwass, H.K.W., Surewicz, W.K., Parris, W., Macfarlane, E.L.A., Luyten, M.A., Kay, C M . , Gold, M. , Jones, J.B. (1992), Protein Eng. 5(8), 769. (54) Han, Y.W. & Srinivasan, V.R (1968) J. Bacteriol. 100,1355. (55) Day, A G . & Withers, S.G. (1986) Can. J. Biochem. Cell Biol. 64,914. (56) Kempton, J.B. & Withers, S.G. (1992) Biochemistry, 31,9961. (57) Warkarchuk, W.W., Kilburn, D.G., Miller, R C . & Warren, R A J . (1986) Molec. Gen. Genetics 205,146. (58) Withers, S.G., Rupitz, K., Tribur, D. & Warren, R.A.J. (1992) Biochemistry 31,9979. (59) El Hassouni, M., Henrissat, B., Chippaux, M. & Barras, F. (1992) J. Bacteriol 174,765. (60) Henrissat, B. (1991) Biochem. J. 280,309. (61) Reynolds, P.H.S., Sissons, C.H., Daniel, R.M. & Morgan, H.W. (1986) Appl. Environ. Microbiol. 51,12. • (62) Love, D.R., Fischer, R. & Bergquist, P.L.(1988) Moi. Gen. Genet. 213, 84. 86 (63) Wang, Q. & Withers, S.G., unpublished. (64) Gebler, J. & Withers, S.G., unpublished. (65) Plant, A.R., Oliver, J.E., Patchett, M.L., Daniel, R.M. & Morgan, H.W. (1988) Arch. Biochem. Biophys. 262,181. (66) Ziser, L., personal communication (67) Conchie, J. & Lewy, G.A. (1963) Methods in Carbohydrate Chemistry U, 335. (68) Sinnott, M.L. & Souchard, I.J.L. (1973) Biochem. J. 133, 89. (69) Leatherbarrow, RJ. (1990) GraFit Version 2.0, Erithacus Software Ltd., Staines, U.K. (70) Lineweaver, H. & Burk, D. (1934) J. Amer. Chem, Soc. 56,658. (71) Barlin,G.B. &Perrin,D.D., (1966) Q.Rev. Chem. Soc. 20, 75. (72) Ba-Saif, S.A. & Williams, A. (1988) J. Org. Chem. 53,2204. (73) Kortum, G., Vogel, W. & Andrussow, K. (1961) Pure Appl. Chem. 1,450. (74) Robinson, R.A., Davis, M.M., Paabo, M. & Bower, V.E. (1960) J. Res. Natl. Bur. Stand., Sect A 64, 347. (75) Withers, S.G. & Rupitz, K. (1990) J. Amer. Chem. Soc. 29, 6405. (76) Nucci, R., Moracci, M., Vaccaro, C , Vespa, N. & Rossi, M. (1993) Biotecnol. Appl. Biochem. 17,239. (77) Korner, M., personal communication. (78) Wang, Q„ personal communication. (79) Daniel, R.M. (1986) Protein Structure, Folding and Design (Oxender, D.L., ed) pg 291, A.R. Liss, New York. (80) Cowan, D.A., Daniel, R.M. & Morgan, H.W. (1985) Trends Biotecnol. 3, 68. (81) Clark-Still, W., Kahn, M. & Mitra, M. (1977) J. Org. Chem. 43,2923. (82) Haynes, L J . & Newth, F.J. (1955) Adv. Carbohydrate Chemistry 10,213. (83) Wolfrom, M.L. & Thompson, A. (1963) Methods in Carbohydrate Chemistry II, 213. (84) Nath, R.L. & Rydon, H.N. (1954) Biochem. J. 57,1. (85) Gerhardt, P., Murray, R.G.E., Costilow, R.N., Nester, E.W., Wood, W.A., Kreig, N.R. & 87 Phillips, G.B. (1981) Manual of Methods for General Bacteriology, American Socieity for Microbiology, Washington, D.C. (86) Schleif, R.F. & Wensink, P.C. (1981) Practical Methods in Molecular Biology, Springer, New York. (87) Schachter, H. (1963) Methods in Carbohydrate Chemistry I, 3. (88) Deakin, H. (1963) Biochem. J. 89,296. (89) Fersht, A. (1985) Enzyme Structure and Mechanism, pg 99, W..H. Freeman and Company, New York. (90) Beychok, S. (1966) Science 154,1288. (91) Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J. (1990) J. Moi. Biol. 215,403. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059593/manifest

Comment

Related Items