Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Structural studies of the catalytic mechanism of bacillus circulans xylanase Sidhu, Gary 1999

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

Item Metadata

Download

Media
831-ubc_1999-0089.pdf [ 6.72MB ]
Metadata
JSON: 831-1.0099366.json
JSON-LD: 831-1.0099366-ld.json
RDF/XML (Pretty): 831-1.0099366-rdf.xml
RDF/JSON: 831-1.0099366-rdf.json
Turtle: 831-1.0099366-turtle.txt
N-Triples: 831-1.0099366-rdf-ntriples.txt
Original Record: 831-1.0099366-source.json
Full Text
831-1.0099366-fulltext.txt
Citation
831-1.0099366.ris

Full Text

S T R U C T U R A L S T U D I E S O F T H E C A T A L Y T I C M E C H A N I S M O F BACILLUS CIRCULANS X Y L A N A S E by Gary Sidhu B.Sc , The University of British Columbia, 1996 A THESIS SUBMITTED IN PART IAL F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF M A S T E R OF SCIENCE in T H E F A C U L T Y OF G R A D U A T E STUDIES T H E D E P A R T M E N T OF BIOCHEMISTRY A N D M O L E C U L A R B IOLOGY We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH C O L U M B I A January 1999 © Gary Sidhu, 1999 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. The University of British Columbia Vancouver, Canada Department of DE-6 (2/88) Abstract The goals of the work described in this thesis were to gain further insight into the structural aspects of the catalytic mechanism of the retaining P-l,4-xylanase from Bacillus circulans. This included the study of both the wild type and certain variants of this enzyme as well as glycosyl-enzyme catalytic intermediates. The 1.6 A resolution recombinant wild type enzyme structure determined in this thesis is very similar to the previously published 1.49 A resolution structure, especially with respect to the residues in the active site. In fact, differences between the two models are restricted to the conformations of mobile surface residues. The structure of Asn35Asp B C X , determined to a resolution of 1.55 A , confirms that the hydrogen bond between residue 35 and the acid/base catalyst, Glul72, is maintained despite the asparagine to aspartate substitution. This hydrogen bond is expected to monopolize the proton on the side chain carboxyl group of Glul72 at pH values above the pK a of Asp35 in a manner that will disrupt catalysis and thereby modify the pH optimum of the enzyme. In the 1.5 A resolution structure of the catalytically inactive Tyr69Phe variant, it is observed, rather surprisingly, that the conformations of residues hydrogen bonded to the phenolic oxygen of Tyr69 in the wild type enzyme, including the catalytic nucleophile, Glu78, change little. These results disprove an earlier hypothesis suggesting that the role of Tyr69 in this enzyme is to correctly position the catalytic nucleophile for attack of the substrate. The structures of the wild type and the Asn35Asp o glycosyl-enzyme intermediates were determined to a resolution of 1.8 A. The 2-fluoro-xylose residue bound in the -1 subsite adopts a 25B (boat) conformation, allowing atoms C5, 05, C I , and C2 of the sugar to achieve coplanarity as required at the oxocarbenium ii ion-like transition states of the double-displacement catalytic mechanism. Comparison of the glycosyl-enzyme intermediates to a mutant of this same enzyme non-covalently complexed with xylotetraose reveals a number of differences beyond the distortion of the sugar moiety. Most notably, a bifurcated hydrogen bond is formed in the glycosyl-enzyme intermediates involving the OH atom of Tyr69, the endocyclic oxygen atom (05) of the xylose residue in the -1 subsite, and atom OE2 of the catalytic nucleophile, Glu78. Since mutation of Tyr69 to phenylalanine produces an inactive enzyme, it is suggested that the interactions involving the phenolic oxygen of Tyr69, 05 of the proximal saccharide, and Glu78 0E2 are important for the catalytic mechanism of this enzyme and, through charge redistribution, serve to stabilize the oxocarbenium-like ion of the transition state. Overall, studies of the covalent glycosyl-enzyme intermediates of this xylanase provide insights into specificity, as contacts with 05 of the proximal saccharide exclude the hydroxymethyl group of glucose-based substrates, and the mechanism of catalysis, including aspects of stereoelectronic theory as applied to glycoside hydrolysis. iii Table of Contents Abstract ii Table of Contents iv List of Tables vii List of Figures viii List of Abbreviations x Acknowledgements xii 1.0 Introduction 1 1.1 Retaining Glycosidases 1 1.2 Family G / l 1 Xylanases 3 1.3 Bacillus circulans Xylanase 4 1.4 Enzyme-Intermediate Complexes 9 1.5 Thesis Objectives 11 2.0 Experimental Methods 12 2.1. Sources of Materials 12 2.2 Preparation of Protein Crystals 12 2.3 The Collection and Processing of X-ray Diffraction Data 15 2.4 General Aspects of X-ray Diffraction Theory 17 2.4.1 Dependence of X-ray Diffraction on Structure 17 2.4.2 The Phase Problem 19 2.4.3 Phase Information for B C X Data 19 iv 2.5 Refinement of Atomic Models 20 2.5.1 Assessing the Quality of Structural Refinement 21 3.0 The Roles of Residues 35 and 69 in B C X 24 3.1 Experimental Procedures 24 3.2 Results 26 3.2.1 The Structure of Wild Type B C X 26 3.2.2 The Structure of Asn35Asp B C X 30 3.2.3 The Structure of Tyr69Phe B C X 33 3.3 Discussion 34 3.3.1 The Low pH Optimum of Asn35Asp B C X 34 3.3.2 The Requirement of Tyr69 for B C X Activity 36 4 .0 Glycosyl-Enzyme Intermediates of B C X 3 8 4.1 Experimental Procedures 38 4.2 Results '. 41 4.2.1 The Glycosyl-Enzyme Intermediate on Wild Type B C X 41 4.2.2 The Glycosyl-Enzyme Intermediate on Asn35Asp B C X 46 4.3 Discussion 49 4.3.1 Conformations of the Disaccharide and Active Site Residues 49 4.3.2 Comparison of BCX-2FXb and B C X X b 50 4.3.3 Interactions at C5 of the Proximal Saccharide in BCX-2FXb 57 4.3.4 Low pH Activity of Asn35Asp B C X 58 4.3.5 Implications 59 v Bibliography 6 2 vi List of Tables 3.1: Data collection parameters for B C X and related variants 25 3.2: Refinement statistics for B C X and related variants 27 4.1: Data collection parameters for the glycosyl-enzyme intermediates 39 4.2: Refinement statistics for the glycosyl-enzyme intermediates 42 4.3: Hydrogen bonding interactions at the active site of various B C X structures 54 vii List of Figures 1.1: General reaction scheme for a retaining (3-glycosidase 2 1.2: The three-dimensional structure of Bacillus circulans xylanase, a family G / l 1 xylanase 5 1.3: The active site of B C X 6 1.4: Schematic representation of the formation of a glycosyl-enzyme intermediate species on B C X 10 2.1: Schematic diagram of the hanging drop vapour diffusion technique for the crystallization of proteins 14 2.2: An example of a typical Ramachandran plot obtained for structures in this thesis 23 3.1: Plots of the average positional deviations for wild type B C X and the Asn35Asp and Tyr69Phe B C X variants 28 3.2: A stereo diagram showing the conformations of residues in the active site of the wild type B C X structure determined in this thesis 31 3.3: A stereo diagram showing the conformations of residues 35, 78 and 172 in the wild type B C X and Asn35Asp B C X variant structures 32 3.4: A stereo drawing depicting the conformations of residues 69, 71, 78 and 172 in the active site of both wild type and Tyr69Phe B C X 35 4.1: The three-dimensional structure of the BCX-2FXb glycosyl-enzyme ' intermediate 43 4.2: Plots of the average positional deviations for atoms of the BCX-2FXb and the viii Asn35Asp-2FXb B C X structures 44 4.3: Stereo diagrams depicting the bound conformation of the 2FXb disaccharide of the glycosyl-enzyme intermediate formed on B C X via Glu78 45 4.4: Interactions of the bound 2FXb disaccharide with active site residues of B C X ..47 4.5: Interactions of the bound 2FXb disaccharide and Asp35 at the active site of Asn35Asp-2FXb B C X 48 4.6: A schematic representing the observed positioning and interactions of the water molecule expected to act as the nucleophile in the second displacement of the catalytic mechanism of B C X 51 4.7: A stereo diagram showing the conformations of the ligands for both the B C X -2FXb a n d B C X X b complex structures 53 ix List of Abbreviations A Angstrom unit (1 A = 0.1 nm) a, b, c Crystallographic unit cell axis or axis length a, (5, y Crystallographic unit cell angles B Isotropic thermal factor B C X Bacillus circulans xylanase d Distance Da Dalton DNPFXb 2',4'-dinitrophenyl 2-deoxy-2-fluoro-p-xylobioside Fo, Fc Observed and calculated structure factors, respectively h, k, I Miller indices of a reflection I Intensity of a reflection N M R Nuclear Magnetic Resonance r.m.s Root mean squared Tris HC1 Tris(hydroxymethyl)aminomethane hydrochloride x, y, z Positional parameters in the crystallographic unit cell 2FXb 2-fluoro xylobioside The conventions of the IUPAC-IUB Combined Commissions on Biochemical Nomenclature are followed for both three letter and one letter abbreviations for amino acids [J. Biol . Chem. 241, 527 - 533 (1966); J. Biol . Chem. 243, 3557 - 3559 (1968)], for designating atoms and describing the conformational torsion angles of the polypeptide x chain [J. Biol . Chem. 245, 6489 - 6497 (1970)], as well as for the nomenclature for ring forms of monosaccharides [Eur. J. Biochem. Ill, 295 - 298 (1980); Carbohydrate Res. 297, 1-90 (1997)]. Variant proteins are referred to with an abbreviation consisting of the three letter code for the original wild type residue, the residue number and the three letter code for the replacement residue. xi Acknowledgements Many people are responsible for the work described in the following pages. The assistance - including that unintentional - of Robert, Gunnar, Nham, Dave, Yaoguang, Tony, Y i l i , Gary, Michael, Steve, Lawrence, Manish, and Greg is especially noted. Most important, however, has been the support of my family and friends; this work is dedicated to them. xii Chapter 1 Introduction 1.1 Retaining Glycosidases Glycosidases are a ubiquitous group of enzymes that have been the focus of a variety of studies since pioneering crystallographic experiments involving hen egg-white lysozyme in the 1960's (Blake et al., 1965). These enzymes catalyze hydrolysis of the glycosidic bond in a manner that can either retain or invert the stereochemistry of the anomeric carbon. For those that retain stereochemistry, enzymatic hydrolysis takes place by way of a double-displacement mechanism and requires two amino acid residues: one that acts as a general acid/base and another that acts as a nucleophile (McCarter & Withers, 1994). In this mechanism, hydrolysis proceeds through a glycosyl-enzyme intermediate that is formed and hydrolyzed via transition states with substantial oxocarbenium ion character (Koshland, 1953). The first displacement occurs when the nucleophilic residue attacks the anomeric carbon and the aglycon leaves with the assistance of the general acid (Figure 1.1). The second displacement results from the nucleophilic attack by a water molecule from which a proton has been removed by the general base. While this mechanism is considered to apply to most retaining glycosidases, alternatives, such as the ion-pair model, have also been suggested (Imoto et al., 1972). Recent studies involving x-ray crystallographic as well as other experiments have provided insight into structural aspects of the catalytic mechanism of retaining p-1 Chapter I Introduction 2 Deglycosylation NUC Figure 1.1 General reaction scheme for a retaining (3-glycosidase. A B C represents the acid/base catalyst, N U C represents the nucleophile, and R and R' represent the non-reducing and reducing ends of the substrate, respectively. The acid/base catalyst is shown in blue, the nucleophile is shown in red, and the transition states for glycosylation and deglycosylation are shown in green. Chapter 1 Introduction 3 glycosidases. Catalysis by these enzymes now appears to involve distortion of the saccharide residue in the -1 subsite (Davies et al., 1997). Current opinions state that the saccharide binds to the active site in a twist-boat conformation, adopts a half-chair conformation at the transition state, and then returns to the conventional chair conformation at the glycosyl-enzyme intermediate (White & Rose, 1997). Hydrolysis of the intermediate would then occur via the reverse conformations. Hypotheses of saccharide distortion are based on the idea that the C5, 05, CI and C2 atoms of the saccharide are required to approach coplanarity at the oxocarbenium ion-like transition states of the catalytic mechanism. 1.2 Family G / l l Xylanases The endo-l,4-R-xylanases (E.C. 3.2.1.8) of family G / l l are enzymes involved in the hydrolysis of nature's most abundant hemicellulose, xylan. These low molecular weight (20 kDa) xylanases perform this function with net retention of anomeric configuration and thus employ the double-displacement catalytic mechanism of retaining glycosidases (Gebler et al., 1992). Interest in the mechanism of R-1,4-xylanases has been piqued with the realization that these enzymes are excellent targets for protein engineering experiments. Xylanolytic activity is useful in biotechnological applications such as the bleaching of hardwood kraft for paper manufacture and the processing of feed for livestock (Coughlan & Hazlewood, 1993, Paice et al., 1992). A substantial volume of both structural and functional information is now available on the family G / l l xylanases (Torronen & Rouvinen, 1997). The three-dimensional fold of these enzymes has been determined and the two acidic residues Chapter I Introduction 4 implicated in the double-displacement mechanism have been identified through a combination of sequence analysis, mutational studies, inhibition experiments, and structure determination (Figure 1.2) (Miao et al., 1994, Wakarchuk et al., 1992, Wakarchuk et al., 1994, Withers & Aebersold, 1995). In the case of Bacillus circulans xylanase (BCX), N M R experiments involving both the native enzyme and a glycosyl-enzyme intermediate have even allowed for the determination of pK a changes that accompany the enzymatic reaction (Mcintosh et al., 1996). Yet, it is the lack of information regarding the structural characteristics of the reaction pathway that prevents a thorough understanding of the detailed mechanism of xylanase catalysis. A structure of a catalytically inactive B C X mutant incubated with xylotetraose has been characterized, but only a xylobiose moiety could be observed at the active site (Wakarchuk et al., 1994). Structures of epoxyalkyl xylosides covalently attached to the active site residues of xylanase II from Trichoderma reesei are also available, but these include only one xylose residue (Havukainen et al., 1996). 1.3 Bacillus circulans Xylanase B C X is a typical member of the family G / l l xylanases. Its native structure (Campbell et al., 1993) and the aforementioned variant enzyme-substrate complex structure are well complemented by a wealth of functional data obtained from N M R spectroscopic experiments. These data, as well as those derived from mutational analysis, have allowed for the determination of important residues in the active site cleft (Figure 1.3) (Wakarchuk et al., 1994). Besides the catalytic nucleophile, Glu78, and the acid/base Chapter 1 Introduction 5 Figure 1.2 Three-dimensional structure of Bacillus circulans xylanase, a family G / l l xylanase. Shown in the active site cleft are residues Glu78 and Glul72, the nucleophile and the acid/base catalyst in the catalytic mechanism, respectively. Throughout this thesis, oxygen atoms are depicted in red, nitrogen atoms in blue, carbon atoms in grey, and fluorine atoms in green. Chapter 1 Introduction 6 Figure 1.3 Active site of B C X . Shown are some of the residues that appear to influence the catalytic ability of B C X . Also shown are hydrogen bonds (dashed yellow lines) between these residues and to nearby water molecules (red spheres). Chapter 1 Introduction 1 catalyst, Glul72, residues Trp9, Tyr69, Tyr80, A r g l l 2 , and Tyrl66 appear to contribute to the catalytic ability of B C X . From inspection of the available structures, it appears that Trp9, Tyr69, Tyr80, and Tyrl66 are involved in substrate binding. Trp9 forms a stacking interaction with the xylose residue bound in the -2 subsite while Tyr69 and Tyrl66 hydrogen bond to the hydroxyl groups of the saccharides on the non-reducing side of the scissile bond. Tyr80 is expected to bind to the xylose residue in the +1 or +2 subsite. Interestingly, while independent mutation of each of these three tyrosine residues to phenylalanine results in varying deleterious effects on catalysis, only the Tyr69Phe mutant exhibits no detectable enzyme activity (Wakarchuk et. al, 1994, Joshi & Mcintosh, 1998). Tyr69 OH donates a hydrogen bond to OE2 of the nucleophile, Glu78, and it has been suggested that this hydrogen bond is essential for the correct positioning of the nucleophile for attack of the substrate. A r g l l 2 is thought to modify the electrostatic environment of Glu78 such that its pIC, is lowered and its nucleophilicity thereby increased. Although the members of family G / l 1 are homologous at both the primary and tertiary levels of structure, they can be divided into two groups according to their p i values. The alkaline pi xylanases, of which B C X is a member, are most active in the pH range from 4 to 8, whereas the acidic pi xylanases are most active at pH 3 to 6 (Torronen & Rouvinen, 1997). It has been suggested that the residue equivalent to Asn35 of B C X , a residue hydrogen bonded to the acid/base catalyst, is critical in determining the active pH range of these enzymes. Interestingly, for all family G / l 1 xylanases active at pH 4 to 8, this residue is an asparagine, while for the xylanases active at pH 3 to 6, it is an aspartate. This correlation is rather counterintuitive as one would normally expect the Chapter I Introduction 8 presence of an acidic residue near the general acid/base catalyst to increase the local negative charge and thus increase the acid/base catalyst's pK a and shift the active pH range to a higher value. One recent theory attempting to resolve this conundrum suggests that, at higher pH values, the aspartate residue would be deprotonated and thus required to accept a hydrogen bond from the acid/base catalyst (Krengel & Dijkstra, 1996). In this arrangement, acid/base catalyst would be unable to donate its proton to the leaving group in the first part of the catalytic mechanism. At lower pH values, however, the protonated aspartate residue could donate its proton and free the acid/base catalyst to assist the reaction. Note that this theory assumes that the association of the aspartate residue and the acid/base catalyst observed in crystal structures of unliganded enzymes is maintained in the presence of substrate. Such a situation, while possible, is contrary to that seen for cyclodextrin glycosyltransferase, in which a similar interaction is broken upon substrate binding (Klein et al., 1992, Lawson et al., 1994). Further evidence of the importance of the residue at position 35 in determining pH optimum comes from the fact that the substitution of Asn35 for aspartate in B C X results in an enzyme whose activity profile shifts to a lower pH (Joshi & Mcintosh, 1998). Thus, this single substitution is enough to convert a high pH optimum xylanase to a low pH optimum xylanase. Understanding the nature of the relationship between the residue equivalent to Asn35 of B C X and the pH optima of the family G / l l xylanases should be considerably facilitated by the determination of Asn35 variant structures, especially that of Asn35Asp with bound ligand. Chapter I Introduction 9 1.4 Enzyme-Intermediate Complexes Structural information on glycosyl-enzyme intermediates can provide valuable insights into the catalytic mechanism of glycosidases. Such an intermediate can be trapped for B C X using the mechanism-based inhibitor 2',4'-dinitrophenyl 2-deoxy-2-fluoro-R-xylobioside (DNPFXb), as has been done in kinetic and spectroscopic studies (Figure 1.4) (Mcintosh et al., 1996, Miao et al., 1994). The electronegative fluorine atom at the 2-position of the xylose moiety slows both the formation and the hydrolysis of the intermediate by inductively destabilizing the oxocarbenium ion transition states and by eliminating important hydrogen bonding interactions at this position (Withers et al., 1988, Withers et al., 1987). The 2,4-dinitrophenyl leaving group, however, facilitates the first nucleophilic attack by Glu78 and thus allows the intermediate to form (second order rate constant kJK{ = 0.34 min"1 mM"1) (Miao et al., 1994). Recently, similar inhibitors have been used in x-ray crystallographic studies of other glycosyl-enzyme complexes including those formed on the catalytic domain of the p-1,4-glycanase Cex from Cellulomonas fimi (Notenboom et al., 1998a, Notenboom et al., 1998b, White et al., 1996), the S-glycosidase myrosinase from Sinapis alba (Burmeister et al., 1997), and the cellulase Cel5A from Bacillus agaradherens (Davies et al., 1998). The results of these experiments confirm a double-displacement mechanism by showing unambiguously the covalent linkage between the catalytic nucleophile and the sugar moiety. A l l three enzymes belong to the G H - A clan and, as such, possess the same (p/a)8 T I M barrel fold and conserved catalytic machinery. Information regarding glycosyl-enzyme intermediates of enzymes outside of this clan, such as the Chapter 1 Introduction 10 Glu172 Glu78 Glu172 Glu78 Figure 1.4 Schematic representation of the formation of a glycosyl-enzyme intermediate species on B C X . The fluorine atom at the 2-position of the proximal saccharide slows both formation and hydrolysis of the intermediate. The 2,4-dinitrophenyl leaving group, however, facilitates the first nucleophilic attack by Glu78, allowing the intermediate to form. Chapter I Introduction 11 family G / l 1 xylanases, is unavailable and may well provide new insights into glycosidase catalytic mechanisms. 1.5 Thesis Objectives The research described in this thesis had two primary objectives. The first was to use x-ray crystallographic techniques to determine the structures of variants of B C X with substitutions at key catalytic residues. Accordingly, I describe in Chapter 3 the structures of wild type B C X along with the Asn35Asp and Tyr69Phe variants. This work tests hypotheses concerning the nature of the ability of the residue at position 35 of B C X to modulate the pH optimum, and the absolute requirement of the phenolic oxygen of Tyr69 for enzyme catalysis. The second objective involved the use of the DNPFXb inhibitor to generate a glycosyl-enzyme intermediate on B C X with the intention of elucidating the structure of this species for both the wild type and the Asn35Asp variant. Structures of these species are presented in Chapter 4. The results obtained provide a unique glimpse into the structural aspects of the catalytic mechanism of enzymes from family G / l l and new insights into the role of substrate distortion in enzyme catalysis. Chapter 2 Experimental Methods 2.1 Sources of Materials The proteins used in this work were obtained from collaborators in the laboratory of Professor Lawrence P. Mcintosh (Departments of Biochemistry and Molecular Biology, and Chemistry, University of British Columbia). The DNPFXb inhibitor was obtained from the laboratory of Professor Stephen G. Withers (Departments of Biochemistry and Molecular Biology, and Chemistry, University of British Columbia). Protein mutagenesis, expression, and purification, as well as inhibitor synthesis, have been previously described (Wakarchuk et al., 1994, Ziser et al., 1995). 2.2 Preparation of Protein Crystals The first step in the structural analysis of proteins using x-ray crystallographic methods is the preparation of diffraction quality crystals. This section describes some of the general characteristics of protein crystals and introduces the technique used to grow the B C X crystals from which the structures in this thesis were elucidated. For a more comprehensive treatment of this topic, the reader is referred to reviews by McPherson (1982), Carter (1990), or Ducruix & Giege (1992). A protein crystal consists of a three-dimensional array of ordered molecules held together by non-covalent intermolecular interactions. For our purposes, it is illustrative 12-Chapter 2 Experimental Methods 13 to consider a crystal as a volume containing individual building blocks arranged in a three-dimensional lattice. The building block, the smallest volume representative of the entire crystal, is referred to in this case as the unit cell. The unit cell's dimensional parameters and space group describe its size and internal symmetry, respectively. The unit cell itself is made up of a smaller motif (often a single protein molecule) that, when acted upon by the symmetry elements prescribed by the space group, generates the complete unit cell. The volume containing the minimum representative motif is known as the asymmetric unit and the goal of the x-ray diffraction experiment is to determine the positions of all atoms in this asymmetric unit. Obtaining single crystals suitable for x-ray crystallographic experiments is often a trial and error procedure. In essence, it is an attempt to find the best conditions under which to manipulate a supersaturated solution of the protein being studied. Protein forced out of solution under these conditions (supersaturation) wil l form either an amorphous precipitate or, in the desired result, single, well ordered crystals. Fortunately, a number of techniques that bring a protein solution to supersaturation in a manner that favours crystal formation have been developed. Common to these techniques is the fact that the protein solution is brought to supersaturation through some mode of diffusion, often with another solution containing a higher concentration of crystallizing agent. The most popular of such techniques is hanging drop vapour diffusion, which is illustrated in Figure 2.1. Crystals for all B C X variants described in this thesis were grown with the hanging drop vapour diffusion method as follows. Drops of protein solution (5-10 ul) at Chapter 2 Experimental Methods 14 Figure 2.1 Schematic diagram of the hanging drop vapour diffusion technique for the crystallization of proteins. Chapter 2 Experimental Methods 15 concentrations of -15 mg ml"1 were placed on siliconized (Sigmacote; Sigma Chemical Co., St. Louis, Missouri) microscope glass cover slips. The protein drop was then mixed with an aliquot of crystallizing solution (generally a 1:1 ratio by volume), and placed inverted over a well containing only the crystallizing solution (typically 1 ml). The well, to which high vacuum silicon grease (Dow Corning Co., Midland Michigan) was applied before contact with the cover slip, was then sealed to ensure a closed system. Through vapour diffusion, the protein solution gradually attained the same concentration of crystallizing reagent as that of the solution at the bottom of the well. In this way, it was possible to bring the protein solution to supersaturation and induce crystal growth. A typical hanging drop crystallization experiment for B C X was carried out using a twenty-four well Linbro plate (Flow Laboratories, McLean, Virginia). Such plates allow for the convenient scanning of a number of different crystallization conditions. For all variants of B C X , the parameters found to be most important to crystal growth included pH, buffer composition, concentration of crystallizing reagent, and the concentration of protein in the hanging drop. Initial scans to determine crystallization conditions were based on those that provided diffraction quality crystals for other variants of B C X (Wakarchuk et al., 1994). Crystals typically appeared in 3 - 4 days and grew to full size in 3 - 4 weeks (-0.3 x 0.4 x 0.5 mm). The precise conditions used to crystallize each variant are presented in the following chapters. 2.3 The Collection and Processing of X-Ray Diffraction Data For analysis, each crystal was mounted in a thin walled glass capillary (1.0 mm diameter) with short columns of mother liquor placed above and below the crystal to maintain Chapter 2 Experimental Methods 16 hydration during data collection. The glass capillary was then sealed with dental wax and placed on a goniometer head. Diffraction data were collected on a Rigaku R-AXIS IIC imaging plate area detector system using C u K a radiation supplied by a Rigaku RU300 rotating anode generator operating at 50 kV and 100 mA. Exposure time for each frame was typically 20 minutes depending on the size and quality of the crystal available. To maximize the amount of diffraction data obtained, the glass capillary was first mounted vertically and then offset at 45°. Each data set was collected using an oscillation angle of 1.2° over a range of O greater than or equal to 90°, and the crystal to detector distance was set between 54 and 78 mm depending upon the resolution of the diffraction data. Intensity data were processed with H K L (Otwinowski & Minor, 1997). Using this software, the diffraction intensities were integrated via a profile fitting algorithm. The level of background radiation was determined by measuring the observed intensity directly surrounding each measured and integrated peak. Partial diffraction spots that were spread over two data collection frames, but which belonged to one diffraction intensity, were added together after the peak integration step was completed. Corrections for background, Lorentz, and polarization effects were applied to the individual diffraction intensities measured on each frame. Following these corrections, the data frames were scaled to account for crystal decay and absorption effects. After scaling, multiple measurements of intensities were merged and all intensities were reduced to structure factor amplitudes. An estimate for the absolute scale for each diffraction data set was obtained using the method of Wilson (Wilson, 1942). This statistical approach compares the observed structure factors with those predicted for a comparable number of atoms in a random Chapter 2 Experimental Methods 17 distribution. Wilson plot calculations were performed using the CCP4 suite of software (Collaborative Computational Project Number 4, 1994). 2.4 General Aspects of X-ray Diffraction Theory This section presents a brief overview of the fundamentals of x-ray diffraction theory as applied to the work described in this thesis. The intent is to provide, to the reader who may be unfamiliar with these methods, the background necessary to follow the crystallographic aspects of these experiments. A more advanced discussion of these and other related topics may be found in a number of excellent texts (for example: Blundell & Johnson (1976), Stout & Jensen (1989), McRee (1993), and Drenth (1994)). 2.4.1 Dependence of X-ray Diffraction on Structure X-rays are electromagnetic radiation with wavelengths in the range of 10"7 - 1 0 " m. This short wavelength makes the interference or diffraction of x-rays sensitive to the atomic arrangement within molecules. Images of molecules, however, cannot be formed directly with x-rays. In contrast to the conventional microscope in which visible light scattered by an object is recombined to produce a magnified image, x-ray images are constructed computationally from the experimentally recorded intensities found in the diffraction pattern of the molecule of interest. There are no lens materials able to focus or bend x-rays. X-rays that interact with atoms are scattered in a discrete manner that is dependant on both the kind of atom involved and its position in space. For a continuous electron distribution, the structure factor F(S) can be defined as an integral in which the Chapter 2 Experimental Methods 18 total scattered wave F in the direction of the reciprocal space vector S is the summation of individually scattered waves by the continuous electron density p at position r integrated over the volume V. This can be expressed as follows: Periodic arrays of atoms, such as those found in crystals, restrict the observation of scattered diffraction intensities to a discrete set of reflections (the result of scattered waves meeting the conditions for constructive interference). The periodicity of the crystalline lattice thus allows for the replacement of the scattering vector S in Equation 2-1 with values specifying a discrete direction. In fact, under conditions of periodicity, the diffraction of x-rays can be considered as occurring from planes in this lattice. These planes are drawn through lattice points and designated by a set of three integers defined as the Miller indices h, k, and /. Each set of parallel and equidistant planes is considered an independent diffractor and produces a single reflection. This, together with the assumption that electron density is localized at atomic centres specified by real space coordinates (x, y, z), provides an equation (2 - 2) that relates the electron density present in the unit cell of a crystal p{x, y, z) to structure factors Fhkl: F(S) = jvp(r)e^rS)dV (2-1) p(*> y. z) = — IXX^MT*' r2m(hx + ky + fe) (2 -2 ) h k I Chapter 2 Experimental Methods 19 2.4.2 The Phase Problem The electron density equation (Equation 2-2) is extremely useful, but in itself is not sufficient to determine a molecular structure. The structure factor Fhkl consists of an amplitude IFAWI or Fm, which is proportional to the square root of the observed diffraction intensity Ihkl, as well as a non-measurable phase term am. This can be expressed as follows: P(*. * z) = ^ I I S | F M ; | ^ V - 2 m ( f a + ^ + w (2 -3) V h k I The inability to measure the phase angles of individual diffraction intensities poses a serious problem for macromolecular crystallography. To overcome this, a number of methods to estimate the necessary phase information have been derived. 2.4.3 Phase Information for BCX Data Initial phase estimates for the structures described in this thesis were all determined by what is essentially a molecular replacement procedure. However, since the variant protein crystal forms were isomorphous (same space group and comparable cell dimensions) to that which provided the original wild type B C X structure, this process was greatly simplified. In fact, under these conditions, one is able to use the known wild type protein structure to estimate initial phases for a crystallized variant protein directly. Thus, the starting model for a variant with an isomorphous crystal form consisted of the wild type B C X structure placed in the unit cell of the variant protein crystal. Following Chapter 2 Experimental Methods 20 placement, structure factors (including phases) were calculated from the positional and thermal parameters of the atoms within the unit cell in the following manner: In this expression, each atom j in the model protein structure is represented by its atomic position (Xj, yjt zj), an atomic scattering factor fp and a thermal parameter Bj related to the mean square displacement of the atom about its average position. The constant C represents a scale factor and shkl is equal to sinOA, for the reflection hkl. Phases determined in this manner were used as estimates for the experimentally measured structure factor amplitudes of the variant protein crystal. 2.5 Refinement of Atomic Models An initial structural model with associated phase estimates can be improved through an iterative process involving a series of alternating cycles of computational least squares fitting of atomic coordinates to the observed diffraction data and manual adjustments to the polypeptide chain based on electron density maps. The progress of structural refinement can be monitored by calculating the crystallographic R-factor, a measure of the agreement between the observed and calculated structure factor amplitudes. This can be expressed as follows: F(K k, I) = C^fjis^e rB,^,)2 2m(_hXj + + kj) (2-4) R-factor = \hkl\L o,hkl (2-5) Chapter 2 Experimental Methods 21 In this equation, Fohkl are the structure factors from the observed data while FcMl are the structure factors calculated from the refined structural model. A low R-factor implies that there is good agreement between the experimental data and the model structure. Generally, for well defined protein structures, the crystallographic R-factor falls between 10% and 20%. Manual adjustments to structural models are primarily based on the assessment of F0 - Fc, 2F() - Fc and fragment deleted Fa - Fc difference electron density maps. Especially useful in the initial characterization of mutation sites and the search for solvent molecules are F0 - Fc difference maps. Maps of the 2Fa - Fc variety superimpose difference electron density on the current refinement model and thus are helpful in delineating features of polypeptide chain structure. Fragment deleted F„ - Fc difference electron density maps are particularly useful in confirming more subtle structural features, as the density of a deleted fragment is not biased by model phases for that region. A l l variant proteins described in the following chapters were refined by least-squares methods using both X-PLOR (Briinger, 1992) and the CCP4 Suite (Collaborative Computational Project Number 4, 1994). Specific details of refinement for each structure studied in this thesis are available in the appropriate chapter. 2.5.1 Assessing the Quality of Structural Models During the course of refinement, the stereochemistry of each B C X structural model was examined using the program P R O C H E C K (Laskowski et al., 1993). Of particular Chapter 2 Experimental Methods 22 interest in these analyses were Ramachandran plots (for example see Figure 2.2), peptide bond planarity, alpha carbon chirality, side chain %, angles, and contact distances between non-bonded atoms. In the case of significant deviations from ideality, the residue involved was closely examined using various difference electron density maps and adjusted accordingly. Error estimates for atomic coordinates were obtained by the method of Luzzati (Luzzati, 1952), a technique based on the dependence of the crystallographic R-factor with resolution. Chapter 2 Experimental Methods 23 Figure 2.2 An example of a typical Ramachandran plot (Ramachandran & Sasisekharan, 1968) obtained for structures completed as a part of this thesis. The data shown is from the final refined structure of BCX-2FXb (see Chaper 4). The coloring shows the most favoured (orange), allowed (brown), generously allowed (yellow), and disallowed (white) regions of Phi/Psi space based on an analysis of 118 high resolution protein structures. Glycine residues are indicated as triangles, while all other residues are represented as squares. ) Chapter 3 The Roles of Residues 35 and 69 in BCX i 3.1 Experimental Procedures Crystals of wild type, Asn35Asp and Tyr69Phe B C X were grown at room temperature (21 °C) using the hanging drop vapour diffusion method as described in Chapter 2. The reservoir solutions contained 17% (NH 4 ) 2 S0 4 , iO m M NaCl and 40 m M Tris HCI at pH 7.5. The hanging droplets consisted of 5 ul of protein solution (-15 mg ml"1) mixed with i 5 pi of reservoir solution. Diffraction quality crystals appeared after approximately one month. Diffraction data for all three proteins were collected on a Rigaku R-AXIS IIC imaging plate area detector system using CuK„ radiation supplied by a Rigaku RU300 rotating anode generator operating at 50 kV and 100 mA. Each diffraction data frame was exposed for 20 minutes during which time the crystal was oscillated through 1.2°. Intensity data were integrated, scaled, and reduced to structure factor amplitudes with the H K L suite of programs as outlined in Section 2.2 (Otwinowski & Minor, 1997). Data collection statistics are provided in Table 3.1. Since the wild type and variant protein crystals were isomorphous with those of the original wild type structure (Campbell et; al., 1993), this wild type B C X model, having each mutated residue truncated to alanine, was used to initiate structural i refinement. Each starting model was subjected to rigid body, simulated annealing, 2 - t Chapter 3 The Roles of Residues 35 and 69 in BCX Table 3.1: Data collection parameters for BCX and related variants. 25 Parameters Wild Type Asn35Asp Tyr69Phe Space group P2.2.2, P2 i2,2 l P2.2.2, Cell dimensions (A) a 43.95 44.05 44.00 b 52.71 52.69 52.75 c 78.49 78.61 78.49 Number of measurements 113039 150984 128221 Number of unique reflections 24823 27301 30126 Mean I/cf 18.4 (5.5) 21.1 (7.0) 16.2 (3.6) Merging R-factor (%)*•* 6.8(13.3) 5.7(15.2) 6.4 (19.9) Resolution range (A) oo.- 1.6 oo - 1.55 oo - 1.5 'Values in parentheses are for data in the highest resolution shell (1.66 - 1.60 A in wild type BCX, 1.61 - 1.55 A in Asn35Asp BCX, and 1.55 - 1.50 A in Tyr69Phe BCX). Chapter 3 The Roles of Residues 35 and 69 in BCX 26 positional, and individual isotropic B-factor refinement using X - P L O R and the CCP4 Suite of software until convergence was realized (Briinger, 1992, Collaborative Computational Project Number 4, 1994). At this point, F0-Fc difference electron density maps were calculated and the residue substituted at each mutation site was built into the observed density with the program O (Jones et al., 1991). Structural models were then refined further with X - P L O R using standard topology and parameter libraries. In the later stages of structural refinement, each model was examined with F0-Fc, 2F0-FC and fragment-deleted difference electron density maps covering the entire course of the polypeptide chain. Based on these maps, manual adjustments were made and solvent molecules were identified. The validity of solvent molecules was assessed based on both hydrogen bonding potential to protein atoms and the refinement of a thermal factor of less than 75 A 2 . The coordinate errors estimated from Luzzati plots (Luzzati, 1952) are 0.19 A for wild type B C X , 0.18 A for Asn35Asp B C X , and 0.17 A for Tyr69Phe B C X . Refinement statistics are provided in Table 3.2. 3.2 Results 3.2.1 The Structure of Wi ld Type B C X The structure of wild type B C X was determined to a resolution of 1.6 A with a conventional R-factor of 19.2% (Tables 3.1 and 3.2). As assessed from the final refinement parameters and the program P R O C H E C K (Laskowski et al., 1993), the stereochemistry of the structural model is excellent. Differences between this structure and the one determined previously for B C X (Campbell et al., 1993) are minimal and Chapter 3 The Roles of Residues 35 and 69 in BCX Table 3.2: Refinement statistics for B C X and related variants 27 Parameters Wild Type Asn35Asp Tyr69Phe Number of reflections 22887 25023 27832 Resolution range (A) 10-1.6 10-1.55 10-1.5 Completeness within range (%) 92.5 91.8 92.7 Number of non-hydrogen protein atoms 1427 1427 1426 Number of non-hydrogen solvent atoms 128 146 148 Average thermal factors (A2) Protein 12.1 13.2 12.5 Solvent 33.5 36.6 34.7 Final refinement R-factor (%)* 19.2 19.4 18.8 Stereochemistry bonds(A) angles (°) 0.007 1.159 r.m.s. deviations 0.007 1.210 0.007 1.155 + ,\Fo, hkl — Fc, hkl\ *R-factor = 1 yLhki\ Fo' hkl Chapter 3 The Roles of Residues 35 and 69 in BCX a « • i i i i i i i i i | i i i i i i i i i | i i i i i i i i i | I I I I I I I I I | I I I I ! I I I I | I I I I I I 1 M | I I I M 1 I I I | I I I I I 1 I I I | 1 I I I I I'l T T | " T I o < - -20 40 60 80 100 120 140 160 180 Residue Number Figure 3.1 continued on next page Chapter 3 The Roles of Residues 35 and 69 in BCX 29 Figure 3.1 continued. Plots of the average positional deviations for wild type B C X and the Asn35Asp and Tyr69Phe B C X variants. In a) the average positional deviations of atoms of the wild type B C X structure determined in this thesis from the original wild type B C X structure are shown. In b) the average positional deviations of the Tyr69Phe B C X variant from wild type B C X (this thesis) are shown. Frame c) shows the average positional deviations of the Asn35Asp B C X variant from wild type B C X (this thesis). In each diagram, thick lines represent the average deviations of main chain atoms, thin lines represent the average deviations of side chain atoms, and the dashed horizontal line represents the overall average positional deviation for all main chain atoms. Chapter 3 The Roles of Residues 35 and 69 in BCX 30 restricted to the conformations of disordered side chains on the surface of the molecule (Figure 3.1). The r.m.s. deviation for all atoms between these two models is 0.21 A. Note that the density at residue 61 suggests that two conformations are available for its peptide backbone. The one that is more defined has been chosen. As shown in Figure 3.2, the active site of the wild type B C X model determined as part of this thesis is very similar to that of the original 1.49 A resolution structure. In fact, positional differences for the atoms of many active site residues are less than the overall estimated coordinate error of the refined structure. Nonetheless, some variation is observed in the active site solvent structure since the water molecule observed to be hydrogen bonded to Tyr69 in the original B C X structure is missing in the present determination (Figures 1.3 and 3.2). Note that all further references to the wild type B C X structure in this thesis refer to the recently determined 1.6 A model. 3.2.2 The Structure of Asn35Asp B C X The Asn35Asp B C X structure was determined to a resolution of 1.55 A with a crystallographic R-factor of 19.4% (Tables 3.1 and 3.2). This structure, despite the introduced substitution, is quite similar to that of wild type B C X and there is an overall r.m.s. deviation of 0.20 A for all atoms between these two structural models. Average positional deviations are presented in Figure 3.1. The side chain of Asp35 adopts approximately the same conformation as that of the originally present Asn35. However, these two isosteric residues do exhibit differences in %2 with Asp35 being rotated -16° with respect to Asn35 (Figure 3.3) Chapter 3 The Roles of Residues 35 and 69 in BCX 31 Figure 3.2 A stereo diagram showing the conformations of residues in the active site of the wild type B C X structure determined in this thesis. Oxygen atoms are shown in red, nitrogen atoms are shown in blue, and carbon atoms are shown in gray. Hydrogen bonds are depicted by yellow dashed lines, and water molecules are shown as red spheres. Chapter 3 The Roles of Residues 35 and 69 in BCX 32 Phe36 N Phe36 N Asn/Asp35 Glul72 Asn/Asp35 Glul72 Glu78 Glu78 Figure 3.3 A stereo diagram showing the conformations of residues 35, 78 and 172 in the wild type B C X and Asn35Asp B C X variant structures. Coordinates were superimposed by least-squares fitting all main chain atoms. Those carbon atoms belonging to the wild type B C X structure are shown in light grey while those belonging to the Asn35Asp variant are shown in dark grey. Hydrogen bonds are shown as dashed yellow lines. Chapter 3 The Roles of Residues 35 and 69 in BCX 33 Many of the other active site residues, including the acid/base catalyst, Glul72, are also similar in conformation for both the Asn35Asp and wild type B C X structures. A small rotation in %3 (~8°) of Glul72 is observed, but this is not enough to disrupt its interaction with the side chain of residue 35. Correspondingly, the distance between Asn35 ND2/Asp35 OD2 and Glul72 OE2 changes only very slightly (3.17 A in wild type B C X and 3.25 A in Asn35Asp BCX) with the replacement of Asn35 by an aspartate. Note, however, that Asp35 OD2 in the variant structure forms an additional hydrogen bond with the amide nitrogen of Phe36 (d = 2.92 A; Figure 3.3). This interaction is not observed in the wild type enzyme as neither Asn35 ND2 or Phe36 N are able to accept hydrogen bonds. The exchange of asparagine for aspartate at position 35 appears to increase the thermal motion of both residues 35 and 172 as determined from normalized isotropic thermal factors. The average thermal factor for residue 35 increases from 11 A 2 to 17 A 2 , while for residue 172 this parameter increases from 11 A 2 to 14 A 2. A l l other residues in the active site region appear structurally unaffected in the Asn35Asp variant. 3.2.3 The Structure of Tyr69Phe B C X The 1.5 A resolution structure of the Tyr69Phe variant of B C X was determined with a R-factor of 18.8% (Tables 3.1 and 3.2). The substitution of phenylalanine at position 69 has little effect on the overall three-dimensional fold of B C X with the r.m.s. deviations for main chain atoms being 0.09 A and for all atoms 0.23 A, relative to the wild type enzyme. Average positional deviations for these two structures are shown in Figure 3.1. Chapter 3 The Roles of Residues 35 and 69 in BCX 34 As illustrated in Figure 3.4, the side chain of Phe69 is found in the same position as that of the tyrosine residue in wild type B C X . Upon replacing Tyr69 with phenylalanine, two hydrogen bonds in the active site of B C X involving Tyr69 OH are lost. One of these hydrogen bonds is formed with the side chain of the catalytic nucleophile, Glu78, while the other is formed with the ring nitrogen of Trp71. Nonetheless, the conformations of these two residues changes only slightly in Tyr69Phe B C X (Figure 3.4). With respect to the wild type B C X structure, %2 and %3 of Glu78 change by only -8° and -11°, respectively. For Trp71, the NE1 side chain atom moves only 0.25 A away from Phe69. The isotropic thermal factors of residues 69 and 78 do, however, exhibit more significant change. The average thermal factor increases from 8 A 2 to 12 A 2 for atoms of residue 69, and from 9 A 2 to 12 A 2 for atoms of Glu78. 3.3 Discussion 3.3.1 The Low p H Optimum of Asn35Asp B C X In the structure of Asn35Asp B C X , the interaction between Glul72 and residue 35 is maintained (Figure 3.3). In fact, a hydrogen bonding distance only slightly longer than seen in the naturally occurring low pH optimum xylanases is observed between Asp35 OD2 and Glul72 OE2. Interestingly, kinetic studies have determined that the pH optimum drops from 5.7 to 4.3 with the substitution of Asn35 by an aspartate in B C X (Joshi & Mcintosh, 1998). The observed conformations of Glul72 and Asp35 support the hypothesis of Krengel and Dijkstra (1996) which states that the low pH activity of Chapter 3 The Roles of Residues 35 and 69 in BCX 35 Figure 3.4 A stereo drawing depicting the conformations of residues 69, 71, 78 and 172 in the active site of both wild type and Tyr69Phe B C X . The coordinates were superimposed by least-squares fitting all main chain atoms. Carbon atoms belonging to the wild type structure are shown in light grey and those of the Tyr69Phe variant are shown in dark grey. Hydrogen bonds are shown as dashed yellow lines. Chapter 3 The Roles of Residues 35 and 69 in BCX 36 aspartate containing xylanases results from the fact that, at higher pH values, Asp35 will ionize and form a hydrogen bond interaction with Glul72. This interaction will tie up the proton on Glul72 and disrupt the catalytic activity of B C X . Recent measurement of the pK a ' s of residues 35 and 172 in Asn35Asp B C X confirms that Asp35 (pK, 3.5) wil l ionize at a lower pH than Glul72 (pK a 8.4) (Joshi & Mcintosh, 1998). Note, however, that the hypothesis of Krengel and Dijkstra also requires that the interaction between Asp35 and Glul72 be maintained in the presence of substrate. The structure of Asn35Asp B C X with bound substrate could thus provide important data concerning the validity of this hypothesis. The structure of one such complex has been completed and is discussed in Chapter 4. An interesting consequence of the described asparagine to aspartate mutation is that the side chain of residue 35 is now able to accept an additional hydrogen bond. A new hydrogen bond is, in fact, formed between Asp35 OD2 and Phe36 N (d = 2.92 A). This hydrogen bond may be important at the pH optimum of Asn35Asp B C X (pH 4.3) in stabilizing the Asp35 - Glul72 interaction, although in the naturally occurring low pH optimum xylanase from Aspergillus niger, no such interaction is observed (Krengel & Dijkstra, 1996). 3.3.2 The Requirement of Tyr69 for B C X Activity The loss of the hydrogen bond from Tyr69 to Glu78 in Tyr69Phe B C X was expected to alter the position of the nucleophile's side chain in a manner that would disrupt catalysis. This hypothesis was based on the observation that the substitution of tyrosine by phenylalanine at residue 69 in B C X completely abrogates activity (Wakarchuk et al., Chapter 3 The Roles of Residues 35 and 69 in BCX 37 1994, Joshi & Mcintosh, 1998). However, as seen in the structure of Tyr69Phe B C X , the conformation of the side chain of Glu78 changes little with the loss of this hydrogen bond (Figure 3.4). Thus, the correct placement of Glu78 must be dictated by other interactions, likely including one with the nearby side chain of Gin 127. A hydrogen bond interaction between Glu78 OE1 and Glnl27 NE2 is found in both the wild type (d = 2.65 A) and Tyr69Phe variant (d = 2.71 A) B C X structures. The current structural studies suggest that the importance of the O H atom of Tyr69 to the catalytic activity of B C X is expressed via some mechanism other than the positioning of the catalytic nucleophile, Glu78. While it is clear that the loss of the O H atom of Tyr69 disrupts two hydrogen bonds and slightly elevates the thermal motion of residues 69 and 78, it seems unlikely that such small changes are sufficient to account for a complete loss of activity. These results imply that the function of Tyr69 in catalysis may involve interactions with bound substrate. Furthermore, because the structure of the Glul72Cys variant of B C X complexed with a xylobiose ligand shows no such interactions, these may occur at the transition state (Wakarchuk et al., 1994). Additional insight into the role of Tyr69 has been recently obtained by looking at the structure of wild type B C X complexed with a mechanism based inhibitor. These results are presented in Chapter 4. Chapter 4 Glycosyl-Enzyme Intermediates of BCX 4.1 Experimental Procedures Crystals of both wild type and Asn35Asp B C X were grown at room temperature (21 °C) using the hanging drop vapour diffusion method as described in Chapter 2. The reservoir solutions contained 17% (NH 4 ) 2 S0 4 , 10 mM NaCl and 40 mM Tris HCI at pH 7.5. The hanging droplets consisted of 5 ul of protein solution (-15 mg ml"1) mixed with 5 pi of reservoir solution. Diffraction quality crystals appeared after one month. Crystals were then soaked in a 75 pi aliquot of well solution which was combined with 25 ul of 1.5 mM DNPFXb, 100% (NH 4 ) 2 S0 4 , 10 mM NaCl and 40 mM Tris HCI at pH 7.5 for -24 hours just prior to data collection. Diffraction data for both the wild type and Asn35Asp glycosyl-enzyme intermediates were collected on a Rigaku R-AXIS IIC imaging plate area detector system using CuKo radiation supplied by a Rigaku RU300 rotating anode generator operating at 50 kV and 100 mA. Each diffraction data frame was exposed for 20 minutes during which time the crystal was oscillated through 1.2°. Intensity data were integrated, scaled, and reduced to structure factor amplitudes with the H K L suite of programs (Otwinowski & Minor, 1997). Data collection statistics are provided in Table 4.1. Since both forms of inhibitor soaked crystals retained isomorphous unit cells, the wild type B C X and Asn35Asp B C X structural models were used to phase their respective Chapter 4 Glycosyl-Enzyme Intermediates of BCX 39 Table 4.1: Data collection parameters for glycosyl-enzyme intermediates. Parameters - BCX-2FXb Asn35Asp-2FXb Space group P2.2.2, P2.2.2, Cell dimensions (A) a 43.87 43.83 b 52.79 52.74 c 78.09 78.80 Number of measurements 137937 99311 Number of unique reflections 17727 17579 Mean I/or 13.6 (4.9) 25.8 (11.1) Merging R-factor (%)t-* 7.3 (30.3) 5.7 (14.2) o Resolution range (A) oo - 1.8 oo - 1.8 ''"Values in parentheses are for data in the highest resolution shell (1.86 - 1.80 A for both BCX-2FXb and Asn35Asp-2FXb B C X ) . Emerge = £ ^ ' = ° 1 " 1 w Chapter 4 Glycosyl-Enzyme Intermediates of BCX 40 glycosyl-enzyme intermediate diffraction data sets. These starting models were subjected to rigid body, simulated annealing, positional, and individual isotropic B-factor refinement using X-PLOR and the CCP4 Suite of software until convergence was realized (R-factor of -23%) (Briinger, 1992, Collaborative Computational Project Number 4, 1994). At this point, for each glycosyl-enzyme intermediate, F0-Fc difference electron density maps were calculated and the 2FXb saccharide was built into the density observed extending from the side chain of Glu78 with the program O (Jones et al., 1991). The models were then refined further with X-PLOR using a standard carbohydrate topology and parameter library. Since the observed electron density clearly indicated that the conformation of the proximal saccharide was distorted, the dihedral angle restraints on this residue were removed. The length of the new glycosidic bond formed to Glu78 was restrained to an o ideal value of 1.43 A using a weight similar to that applied to other sugar bond lengths in the ligand. Atoms of the bound disaccharide were refined at full occupancy. The structural models were examined periodically during refinement with F0-Fc, 2F0-FC, and fragment-deleted difference electron density maps. Manual adjustments were made as necessary and solvent molecules were identified. The validity of solvent molecules was assessed based on both hydrogen bonding potential to appropriate protein atoms and refinement of a thermal factor of less than 75 A 2 . The coordinate error estimated from a Luzzati plot (Luzzati, 1952) is 0.18 A for the intermediate formed on wild type B C X and 0.19 A for the intermediate formed on Asn35Asp B C X . Chapter 4 Glycosyl-Enzyme Intermediates of BCX 41 4 .2 Results 4 .2 .1 The Glycosyl-Enzyme Intermediate on W i l d Type B C X The structure of the glycosyl-enzyme intermediate formed on wild type B C X (BCX-2FXb) was determined to a resolution of 1.8 A with a conventional R-factor of 18.9% (Table 4.1, Table 4.2, and Figure 4.1). This enzyme structure exhibits excellent stereochemistry and the bound disaccharide atoms are well defined (average fi-factor of 17.6 A 2). Differences observed in the BCX-2FXb structure relative to that of the free enzyme are primarily restricted to the flexible loop region composed of residues 111 to 125 at the entrance of the active site (Figure 4.2). This loop shifts by up to 1 A leading to a widening of the active site cleft. The final refined enzyme-intermediate structure shows the 2-deoxy-2-fluoro-xylobiose (2FXb) ligand covalently bonded to the nucleophile, Glu78, via an oc-anomeric linkage with the CI atom of the proximal saccharide, syn to the newly formed ester group (Figure 4.3). The proximal saccharide occupies the -1 subsite (Davies et al., 1997) of B C X , and the new bond between the CI atom and the OE2 atom of Glu78 is 1.45 A in length. The observation of this covalent glycosyl-enzyme intermediate supports the proposal for a double-displacement mechanism as shown in Figure 1.1. The xylose residue covalently bonded to Glu78 is heavily distorted from the conventional 4 C , (chair) conformation and adopts a 2 5 B (boat) conformation instead (Figure 4.3). The 2'5B conformation of this proximal saccharide allows its C5, 05, C I , and C2 atoms to achieve a nearly planar geometry (0.05 A r.m.s. deviation from planarity versus 0.22 A r.m.s. deviation for the same atoms in the distal saccharide). Chapter 4 Glycosyl-Enzyme Intermediates of BCX 42 Table 4.2: Refinement statistics for BCX-2FXb and Asn35Asp-2FXb. Parameters BCX-2FXb Asn35Asp-2FXb Number of reflections 17492 16559 Resolution range (A) 10-1.8 10-1.8 Completeness within range (%) 99.0 94.5 Number of non-hydrogen protein atoms 1427 1427 Number of non-hydrogen ligand atoms 18 18 Number of non-hydrogen solvent atoms 117 129 Average thermal factors (A2) Protein 14.4 12.3 Ligand 17.6 23.2 Solvent 34.7 34.0 Final refinement R-factor (%)* 18.9 19.3 Stereochemistry r.m.s. deviations bonds (A) 0.007 0.007 angles (°) 1.128 1.136 *R-factor = ^ M \ F o , h k i - Fchki Chapter 4 Glycosyl-Enzyme Intermediates of BCX 43 residues 111-125 Glul72 Glu78 Figure 4.1 The three-dimensional structure of the B C X - 2 F X b glycosyl-enzyme intermediate. Along with the bound inhibitor, the side chains of the two glutamate residues implicated in the double-displacement catalytic mechanism of this enzyme are shown. Relative to native B C X , atoms in the polypeptide chain loop composed of residues 111 - 125 to the left of the bound inhibitor have shifted out from the active site cleft to accomodate the disaccharide moiety. Figure 4.2 Plots of the average positional deviations for atoms of a) the wild type B C X -2FXb and b) the Asn35Asp-2FXb glycosyl-enzyme intermediate from atoms of the uncomplexed wild type B C X structure. In each diagram, thick lines represent the average deviations of main chain atoms, thin lines represent the average deviations of side chain atoms, and the dashed horizontal line represents the overall average positional deviation for all main chain atoms. Chapter 4 Glycosyl-Enzyme Intermediates of BCX 4 5 Figure 4.3 Stereo diagrams depicting the bound conformation of the 2FXb disaccharide of the glycosyl-enzyme intermediate formed on B C X via Glu78. In a) the disaccharide ligand is shown superimposed on an F„ - Fc difference electron density map calculated before the inclusion of the ligand atoms in the refinement model and with Glu78 omitted from Fc. This map is contoured at the 2a level. In b) an alternative direction of view is shown for Glu78 and the proximal residue of the disaccharide which adopts a 25B conformation. Chapter 4 Glycosyl-Enzyme Intermediates of BCX 46 These results mark the first crystallographic observation of such significant distortion in a glycosyl-enzyme complex. The distal xylose residue in the -2 subsite maintains a 4 C , conformation and makes van der Waals contacts with the main chain atoms of Serl 17 and a stacking interaction with the side chain atoms of Trp9 (Figure 4.4). This bound conformation is similar to that observed in the non-covalent BCX-xylobiose ( B C X X b ) complex structure (Wakarchuk et al., 1994). Stacking interactions of the type observed have been seen in other carbohydrate-protein complex structures (Vyas, 1991). 4.2.2 The Glycosyl -Enzyme In termedia te on Asn35Asp B C X The structure of the glycosyl-enzyme intermediate formed on Asn35Asp B C X (Asn35Asp-2FXb) was determined to a resolution of 1.8 A with a crystallographic R-factor of 19.3% as outlined in Tables 4.1 and 4.2. As seen for the BCX-2FXb complex, the active site cleft of Asn35Asp-2FXb widens to accommodate the bound disaccharide (Figure 4.2). In fact, the Asn35Asp-2FXb structure exhibits many of the same features observed for the B C X -2FXb complex. These include 2FXb being covalently bonded to Glu78 via an oc-anomeric configuration, a proximal saccharide which is distorted to a 2,5B conformation, a distal saccharide which maintains a 4 C , conformation and stacks on the side chain of Trp9, and similar isotropic thermal factor profiles including those for residue 35. However, one important difference between the two complex structures is the length of the interaction o between residue 35 and the OE2 atom of Glul72. This distance is reduced from 3.28 A in BCX-2FXb to 2.74 A in the corresponding Asn35Asp-2FXb complex (Figure 4.5). In the unliganded Asn35Asp B C X structure, it was seen that a new hydrogen bond Figure 4.4 Interactions of the bound 2FXb disaccharide with active site residues of B C X . In a) these are shown in the form of a stereo diagram whereas in b) they are depicted as a schematic. Panel c) shows the interactions formed with the hydroxyl group of Tyr69. Hydrogen bonds are depicted by long dashes while the non-hydrogen bond interaction described in the text is depicted by short dashes. Hydrogen atoms were added using HBPLUS (McDonald & Thornton, 1994) Chapter 4 Glycosyl-Enzyme Intermediates of BCX 48 Figure 4.5 Interactions of the bound 2FXb disaccharide and Asp35 at the active site of Asn35Asp-2FXb B C X . In a) the conformations of the disaccharide, residue 35, Glul72, and Glu78 in both the Asn35Asp-2FXb (dark grey) and BCX-2FXb (light grey) glycosyl-enzyme intermediates are shown in the form of a stereo diagram. Coordinates were superimposed by least-squares fitting all main chain atoms. In b) a schematic detailing the interactions and distances observed in the Asn35Asp-2FXb glycosyl enzyme intermediate is shown. Hydrogen bonds are depicted by long dashes, while the non-hydrogen bond interaction described in the text is depicted by short dashes. Chapter 4 Glycosyl-Enzyme Intermediates of BCX 49 between Asp35 0D2 and Phe36 N was formed, helping to maintain the interaction between Asp35 and Glul72 (Figure 3.4). This hydrogen bond is also present in the Asn35Asp-2FXb glycosyl-enzyme intermediate at an almost equal distance to that of Asn35Asp B C X (d = 2.92 A in Asn35Asp B C X , and d = 2.91 A in Asn35Asp-2FXb B C X ) (Figure 4.5). 4.3 Discussion 4.3.1 Conformations of the Disaccharide and Active Site Residues The planarity observed for the C5, 05, C I , and C2 atoms of the proximal saccharide in the 2,5 B conformation is of particular interest. As noted earlier, both steps of the reaction catalyzed by B C X proceed through transition states with substantial oxocarbenium ion character in which a partial double bond develops between 05 and CI (Figure 1.1) (Sinnott, 1990). This requires C5, 05, C I , and C2 to approach coplanarity at the transition state. Since such planarity is realized at the glycosyl-enzyme intermediate, the formation and hydrolysis of this species should be considerably facilitated. Although the rearrangements involved in generating a 2 , 5B conformation from a 4 C , conformation may appear considerable, they can occur with little perturbation of the relative positions of most atoms of the proximal saccharide. If the rearrangement proceeds via a 4 C , -> 2H3 -> 2 5 0 -> 25B itinerary (Stoddard, 1971), the relative movement is confined mostly to atoms C5 and 05. In this case, the dihedral angle defined by atoms C I , C2, C3, and C4 is not required to undergo much change (this angle is observed to be 48° for the proximal saccharides in BCX-2FXb and Asn35Asp-2FXb, and 57° in the distal saccharides). Note that the C5 and 05 atoms of xylose are not encumbered by the functional groups found on atoms C I , C2, C3, and C4, and thus are able to experience the Chapter 4 Glycosyl-Enzyme Intermediates of BCX 50 movement prescribed by the aforementioned rotational itinerary without a large energetic cost. The acid/base catalyst, Glul72, occupies a similar position and conformation in the BCX-2FXb complex to that found in the wild type B C X structure (Campbell et al., 1993). In the BCX-2FXb complex, the OE2 atom of Glul72, which donates and abstracts protons during catalysis, makes a 3.10 A hydrogen bond to a water molecule situated 3.92 A away from the CI atom of the proximal saccharide (Figure 4.6). This water molecule is also hydrogen bonded to the phenolic oxygen atom of Tyr80 (d = 2.74 A) as well as to a second o solvent molecule (d = 3.17 A), and is a likely candidate for the nucleophile in the deglycosylation step of the reaction. According to the proposed mechanism and recent pK a measurements, the OE2 atom of Glul72 is deprotonated under experimental conditions (pK a = 4.2) and thus must accept a hydrogen bond from the nearby water molecule (Joshi et al., 1997, Mcintosh et al., 1996). Tyr80, which appears to be donating a hydrogen bond (d = 2.83 A) to the deprotonated OE1 atom of Glul72, must then also accept a hydrogen bond from the water molecule. In this position, one of the lone pairs of the tetrahedral oxygen atom of this water molecule is directed towards the second solvent molecule while the other lone pair points in the general direction of the proximal saccharide. This placement is ideal for an attack on the CI atom of the proximal saccharide residue during the deglycosylation reaction (Figure 4.6). 4 . 3 . 2 Comparison of B C X - 2 F X b and B C X X b It is instructive to compare the enzyme-ligand interactions observed in the B C X - 2 F X b Chapter 4 Glycosyl-Enzyme Intermediates of BCX. 51 Figure 4.6 A schematic representing the observed positioning and interactions of the water molecule (labeled with an asterisk) expected to act as the nucleophile in the second displacement of the catalytic mechanism of B C X . Also drawn are the interactions formed with the proximal saccharide residue. Lines with long dashes represent hydrogen bonds, the line with short dashes represents the non-hydrogen bond interaction described in the text, and the R group represents saccharide units bound in the -2 subsite or further removed from the active site. Chapter 4 Glycosyl-Enzyme Intermediates of BCX 52 glycosyl-enzyme intermediate with those of the B C X Xb non-covalent complex for which the xylobiose moiety is bound in the same two subsites (-1, -2) (Wakarchuk et al., 1994). Note that both xylose residues in the B C X X b complex maintain 4 C , conformations thus placing the nucleophilic oxygen atom of Glu78 3.35 A from the anomeric carbon of the proximal saccharide (Figure 4.7). The requirement to shorten this distance to the length of a glycosidic bond must be an important driving force for the conformational rearrangement of the proximal xylose residue. Of particular interest when comparing the BCX-2FXb and B C X X b structures are interactions with Tyr69 (Table 4.3). In the non-covalent complex, this residue donates a strong hydrogen bond (d = 2.60 A) to the nucleophilic oxygen atom (OE2) of Glu78 and accepts a hydrogen bond from the 2-position OH of the distal xylose moiety. In contrast, in the covalent intermediate, the hydrogen bond donated to the OE2 atom of Glu78 is much weaker (d = 2.99 A), consistent with the ether character of its partner. Furthermore, a new interaction for Tyr69 is formed in BCX-2FXb with the endocyclic oxygen (05) of the proximal xylose moiety (Figure 4.4). The nature of this interaction is of interest since the hydroxyl group of Tyr69 is very important for catalysis. In fact, the Tyr69Phe variant of B C X exhibits no detectable enzyme activity (Joshi & Mcintosh, 1998, Wakarchuk et al., 1994). Since, in the BCX-2FXb structure, Tyr69 OH donates a hydrogen bond to OE2 of Glu78, it cannot also donate to 05 of the proximal xylose residue unless its proton is able to alternately occupy two sites or form a bifurcated hydrogen bond (Baker & Hubbard, 1984). A detailed analysis of the hydrogen bonding geometry about Tyr69 OH suggests that the latter is, in fact, quite possible. Both of the potential hydrogen Chapter 4 Glycosyl-Enzyme Intermediates of BCX 53 Figure 4.7 A stereo diagram showing the conformations of the ligands for both the J3CX-2FXb and B C X X b complex structures. The side chain of Glu78 is also shown. Coordinates were superimposed by least-squares fitting the main chain atoms of residues 1 - 1 1 0 and 126 - 185. The carbon atoms belonging to the B C X - X b non-covalent structure are depicted in dark grey, while those belonging to the BCX-2FXb glycosyl-enzyme intermediate are depicted in light grey. Chapter 4 Glycosyl-Enzyme Intermediates of BCX Table 4.3: Interactions at the active site of B C X . 54 Interaction Distances (A) BCX-2FXb Asn35Asp-2FXb B C X X b Tyr69 OH - Glu78 OE2 2.99 2.85 2.60 Tyr69 OH - distal X y l 02 2.80 2.90 2.88 Tyr69 OH - proximal X y l 05 2.95 3.08 4.04 Glu78 OE1 - proximal X y l 02 (F2)f* 2.78 2.66 3.00 Arg l 12 N E - proximal X y l 02 (F2)f 3.17 3.01 2.97 Arg l 12 NH2 - proximal X y l 03 3.26 3.05 3.15 Pro 116 0 - proximal X y l 03 2.68 2.66 2.56 Tyrl66 OH -distal X y l 02 2.97 3.07 3.24 Tyrl66 OH-dis ta l X y l 03 2.90 2.92 2.82 f Atoms in parentheses reflect substitutions made at the 2-position of the DNPFXb ligand. *This interaction is not a hydrogen bond in glycosyl-enzyme intermediate structures and is discussed in the text. Chapter 4 Glycosyl-Enzyme Intermediates of BCX 55 bonds donated by Tyr69 OH have reasonable distances and angles, as presented in Figure 4.4. The hydrogen bond interaction of Tyr69 OH with the endocyclic 05 oxygen atom of the proximal saccharide may be required to assist conformational rearrangement and stabilize the boat conformation in the reaction intermediate. The involvement of Tyr69 in such a bifurcated hydrogen bond may also explain its necessity for catalysis. Interestingly, a study of the hydrogen bonding geometry of Tyr330 in the glycosyl-enzyme intermediate of Sinapis alba myrosinase shows similar interactions of the phenolic hydroxyl group with the OE2 atom of the catalytic nucleophilic, Glu409, and the endocyclic oxygen of the sugar moiety (Burmeister et al., 1997). Clearly, however, a hydrogen bond to the ring oxygen of the proximal saccharide would be counter-catalytic since it would serve to destabilize the oxocarbenium ion-like transition state. It seems likely, therefore, that as the transition state is approached the hydrogen bond donated by Tyr69 OH becomes asymmetric, favouring Glu78 0E2. Simultaneously, the interaction between Tyr69 OH and the partially positively charged 05 of the proximal xylose residue becomes a direct oxygen - oxygen contact, unmediated by a proton. A dipolar interaction could then form between the two oxygen atoms, concomitant with improved proton donation from Tyr69 to the partially negatively charged 0E2 atom of Glu78, and, as such, could stabilize the transition state. This interaction may play an important role in catalysis and provide yet another explanation of the requirement of Tyr69 for enzyme activity. Another interaction that changes substantially between the B C X X b and B C X -2FXb structures is that between OE1 of Glu78 and the 2-substituent of the proximal sugar (OH in the non-covalent complex and F in the covalent species). Despite the fact that the Chapter 4 Glycosyl-Enzyme Intermediates of BCX 56 interaction between the fluorine and the oxygen in the covalent complex must be destabilizing, the distance between these two atoms is shorter than seen for the analogous yet stabilizing hydrogen bonding interaction in the B C X X b non-covalent complex (Table 4.3). The shortening of this distance in BCX-2FXb appears to be a consequence of the formation of the covalent bond at the anomeric centre and the 4 C , -> 25B conformational change in the proximal saccharide. Similar interactions have been seen in other 2-fluoro sugar glycosyl-enzyme complexes and may be an important component of catalysis with the natural substrate (Burmeister et al., 1997, Notenboom et al., 1998a, White et al., 1996). The suggestion is that a short, strong hydrogen bond is formed at this position in the intermediate and that this interaction is even shorter and stronger at the transition state where it is optimized geometrically and electrostatically. This is consistent with the very strong transition state interactions (typically 20 - 40 kJ mol"1) measured at the 2-position for a number of glycosidases and further bolstered by the recent observation of an even shorter (2.37 A) interaction between OE1 of the nucleophile and the hydroxyl group at C2 of the substrate in the cellobiosyl-enzyme intermediate formed on a double mutant of Cellulomonas fimi Cex (Namchuk & Withers, 1995, Notenboom et al., 1998b). It is noteworthy that an essentially identical interaction is seen in the present case even though B C X is from a different structural clan and the proximal saccharide is distorted. The other residue interacting with the 2-position fluorine in the B C X - 2 F X b complex is the highly conserved A r g l l 2 (Figure 4.4). This may suggest an important role for this residue in hydrogen bonding to the 2-hydroxyl of natural substrates as has been suggested for Asnl26 in Cellulomonas fimi Cex or Asnl86 in Sinapis alba myrosinase (Burmeister et al., 1997). It is, therefore, surprising that mutation of this residue causes Chapter 4 Glycosyl-Enzyme Intermediates of BCX 57 only relatively small deleterious effects on catalysis (R112K has 83% of wild type activity and R l 12N has 35% of wild type activity) (Wakarchuk et al., 1994). For B C X then, the most important interaction with the 2-substituent would appear to be that with the OE1 atom of Glu78. The planar organization of the reaction centre and the interactions with Tyr69 in BCX-2FXb appear to generate a much more reactive glycosyl-enzyme intermediate than those seen in the clan GH-A enzymes. This is consistent with kinetic observations since the deglycosylation step never became rate-limiting for wild type B C X even when studied with highly reactive aryl xylobiosides (Tull & Withers, 1994). By contrast, the deglycosylation step for Cex is rate-limiting for all aryl cellobiosides and xylobiosides studied, implying a much slower deglycosylation step relative to glycosylation. 4 . 3 . 3 Interactions at C5 of the Proximal Saccharide in BCX-2FXb Unlike the P-l,4-glycanases, such as Cellulomonasfimi Cex, which act on both xylan and cellulose, the family G / l l (3-1,4 xylanases are specific for xylan (Torronen & Rouvinen, 1997, Tull & Withers, 1994). Both the BCX-2FXb glycosyl-enzyme intermediate and B C X X b non-covalent structures provide insight into this specificity. If the C5 atom of a xylose residue carried a hydroxymethyl group, as do the glucose residues of cellulose, there would be significant steric problems associated with accommodating the saccharide units into the active site cleft of B C X . Modeling experiments with both complex structures suggest that a hydroxymethyl group on the distal saccharide would clash with the carbonyl oxygen of Serl 17, while for the proximal saccharide such a group would clash with the CG2 atom of Val37. In fact, the C5 atom of the proximal saccharide in the glycosyl-Chapter 4 Glycosyl-Enzyme Intermediates of BCX 58 enzyme intermediate makes two van der Waals interactions with the protein, contacting the CG2 atom of Val37 and the CZ3 atom of Trp9 at distances of 3.71 A and 3.59 A, respectively. The ND2 atom of Asn35 is also quite close (d = 3.69 A). These contacts with C5 are more extensive than seen between any other atom of the 2FXb ligand and the protein, and together, appear to shape this portion of the active site cleft such that it limits substrates to those that are unsubstituted at the C5 position. Note also that achieving the 2'5B conformation observed in the glycosyl-enzyme intermediate would be extremely difficult for glucose. This conformation would force the bulky hydroxymethyl group axial with substantial steric costs, both between the sugar and the protein, and within the sugar itself due to the "bowsprit" interactions with H2. 4 .3.4 Low p H Activity of Asn35Asp As illustrated in Figures 3.3 and 4.5, a hydrogen bonding distance between Asp35 OD2 and Glul72 OE2 is maintained and of even shorter length in the Asn35Asp-2FXb glycosyl-enzyme intermediate of B C X . This observation supports earlier studies which suggest that the low pH activity of the Asp35-like xylanases results from an inability of Glul72 OE2 to donate its proton to the leaving group at pH values above the pK a of Asp35 (Krengel & Dijkstra, 1996). For the Asn35Asp-2FXb glycosyl-enzyme intermediate complex of B C X , as mentioned in Chapter 3, it is anticipated that when Asp35 OD2 is deprotonated, its hydrogen bond with Glul72 will include and monopolize the proton on Glul72 OE2 and thereby disrupt catalysis. N M R studies of the Asn35Asp-2FXb B C X intermediate have been unable to determine pK a values for the side chain groups of Asp35 and Glul72 (Joshi & Mcintosh, Chapter 4 Glycosyl-Enzyme Intermediates of BCX 59 1998). Results of deuterium labeling experiments, however, suggest that the two residues may share a proton, consistent with the fact that the distance between Asp35 and Glul72 is -0.5 A shorter in Asn35Asp-2FXb than in Asn35Asp B C X . The formation of a low-barrier hydrogen bond between these residues at the transition state, such as those that have recently been implicated in the mechanism of serine proteases (Cassidy et al., 1997), has been hypothesized but not yet demonstrated (Joshi & Mcintosh, 1998). 4.3.5 Implications The observation of such striking conformational distortion for a sugar residue in a glycosyl-enzyme intermediate is unique. Furthermore, the nature of this conformational distortion is of considerable interest in terms of its implications for the catalytic mechanism of the family G / l l xylanases. The distortion to the 2,5B conformation places C5, 05, CI and C2 in a planar arrangement that would considerably facilitate the formation of oxocarbenium ion character at this centre in the transition state. Interestingly, the same 2,5B conformation was proposed previously as the conformation of the cc-glucoside substrate bound to yeast oc-glucosidase and identical arguments were forwarded concerning C5, 05, CI and C2 (Hosie & Sinnott, 1985). It was also pointed out that, in this conformation, there is no lone-pair on the ring oxygen that is anti-periplanar to the scissile bond whereas in the 4 C , conformation such an arrangement does exist. The first step (glycosylation) for an a-glycosidase and the second step (deglycosylation) for a p-glycosidase have identical stereochemical outcomes (cc-»P), and therefore stereoelectronic theory would apply in an equivalent manner to each. The direct observation of the saccharide bound to B C X in a Chapter 4 Glycosyl-Enzyme Intermediates of BCX 60 2,5B conformation, therefore, provides substantial evidence against the importance of the antiperiplanar-lone-pair hypothesis of Deslongchamps (Deslongchamps, 1983). Complexes of glycosidases with substrates or products in which a sugar ring is substantially distorted have been observed previously. The best defined are those in which the substrate is bound across the cleavage site with occupancy of both the -1 and +1 subsites, leading to the suggestion that interactions with the +1 subsite are driving the distortion (Davies et al., 1998, Sulzenbacher et al., 1996, Tews et al., 1996). However, the degree of distortion in these complexes, as well as that seen in a hen egg-white lysozyme product complex (Strynadka & James, 1991), is not as great as seen in the present example. It is important to note that since +1 subsite interactions do not exist in the glycosyl-enzyme intermediate complexes, such interactions cannot be important in driving the distortion observed in the covalent intermediates. The van der Waals interactions of the C5 atom of the proximal xylose residue suggest not only that the active site is designed specifically for xylan hydrolysis but also that this atom may have a role in achieving the 25B conformation. As this atom makes multiple contacts with the protein and occupies a similar position in both the B C X X b and the BCX-2FXb structures, it may provide a "hinge" for the action required to mediate distortion from the 4 C , to the 25B conformation (Figure 4.5). It is also likely that the lack of a hydroxymethyl group at C5 is key to the ability to undergo this distortion, suggesting that the hydrolysis of glucose-based substrates by other (3-1,4-glycosidases may follow a different course. In the Asn35Asp-2FXb complex, it is clear that the interaction between Asp35 and Glul72 is maintained, and therefore, the substitution of Asn by Asp at this position Chapter 4 Glycosyl-Enzyme Intermediates of BCX 61 provides an effective method for sequestering the proton required for catalysis and lowering the pH optimum for activity. The exact nature of this interaction in Asn35Asp-2FXb, however, is still unknown. Further spectroscopic studies may be required to determine the protonation state of these two residues. Some of the most important interactions at the active site of B C X appear to be those between Tyr69 OH and the endocyclic (05) oxygen of the proximal sugar ring and OE2 of Glu78, as well as those between the 2-position of the proximal sugar and OE1 of Glu78. The interactions between these partners form an extended network which serves to redistribute charge as it is developed at the transition state while at the same time stabilizing the planar structure required. The structure of the B C X glycosyl-enzyme intermediate with its flattened oxocarbenium-like conformation provides insight into the mechanism of xylan hydrolysis and serves to highlight some of the important sites of interaction in a manner that may be useful for the engineering of xylanases for biotechnological applications. Bibliography Baker, E. N . , and Hubbard, R. E. (1984) Hydrogen bonding in globular proteins. Prog. Biophys. Mol. Biol. 44, 97-179. Blake, C. C , Koenig, D. F., Mair, G. A. , North, A . C , Phillips, D. C , and Sarma, V . R. (1965) Structure of hen egg-white lysozyme. A three-dimensional Fourier synthesis at 2 A resolution. Nature 206, 757-761. Blundell, T. L . , and Jonhson, L . N . (1976) Protein Crystallography, Academic Press, London. Briinger, A . T. (1992) X-PLOR Version 3.1: A system for x-ray crystallography and NMR, Yale University Press, New Haven. Burmeister, W. P., Cottaz, S., Driguez, H., Iori, R., Palmieri, S., and Henrissat, B. (1997) The crystal structures of Sinapis alba myrosinase and a covalent glycosyl-enzyme intermediate provide insights into the substrate recognition and active-site machinery of an S-glycosidase. Structure 5, 663-675. Campbell, R. L . , Rose, D. R., Wakarchuk, W. W., To, R., Sung, W., and Yaguchi, M . (1993) in Proceedings of the Second TRICEL Symposia on Trichoderma reesei and Other Hydrolases (Suominen, P., and Reinikainen, T., Eds.) pp 63-72, Foundations for Biotechnical and Industrial Fermention, Helsinki. Carter, C. W. J. (1990) Protein and nucleic acid crystallization, Vol . 1, Academic Press, Duluth. Cassidy, C. S., Lin, J., and Frey, P. A . (1997) A new concept for the mechanism of action of chymotrypsin: the role of the low-barrier hydrogen bond. Biochemistry 36, 4576-4584. Collaborative Computational Project Number 4. (1994) The CCP4 Suite: programs for protein crystallography. Acta. Crystallogr. D50, 760-763. Coughlan, M . P., and Hazlewood, G. P. (1993) (3-1,4-D-xylan-degrading enzyme systems: biochemistry, molecular biology and applications. Biotechnol. Appl. Biochem. 17, 259-289. Davies, G. J., Wilson, K. S., and Henrissat, B . (1997) Nomenclature for sugar-binding subsites in glycosyl hydrolases. Biochem. J. 321, 557-559. Bibliography 63 Davies, G. J., Mackenzie, L . F., Varrot, A. , Dauter, M . , Brzozwski, A . M . , Schulein, M . , and Withers, S. G. (1998) Snapshots along an enzymatic reaction coordinate: analysis of a retaining P-glycoside hydrolase. Biochemistry 37, 11707-11713. Deslongchamps, P. (1983) Stereoelectronic effects in organic chemistry, 1st ed., Pergamon Press, Oxford; New York. Drenth, J. (1994) Principles of protein x-ray crystallography, Springer-Verlag, New York. Ducruix, A . , and Giege, R. (1992) Crystallization of nucleic acids and proteins. A practical approach, Oxford University Press, New York. Gebler, J., Gilkes, N . R., Claeyssens, M . , Wilson, D. B., Beguin, P., Wakarchuk, W. W., Kilburn, D. G. , Miller, R. C., Jr., Warren, R. A . , and Withers, S. G. (1992) Stereoselective hydrolysis catalyzed by related p-l,4-glucanases and p-l,4-xylanases. J. Biol. Chem. 267, 12559-12561. Havukainen, R., Torronen, A. , Laitinen, T., and Rouvinen, J. (1996) Covalent binding of three epoxyalkyl xylosides to the active site of endo-l,4-xylanase II from Trichoderma reesei. Biochemistry 35, 9617-9624. Hosie, L . , and Sinnott, M . L . (1985) Effects of deuterium substitution a and p to the reaction centre, l s O substitution in the leaving group, and aglycone acidity on hydrolyses of aryl glucosides and glucosyl pyridinium ions by yeast a-glucosidase. A probable failure of the antiperiplanar-lone-pair hypothesis in glycosidase catalysis. Biochem. J. 226,437-446. Imoto, T., Johnson, L . N . , North, A . C. T., Phillips, D. C., and Rupley, J. A . (1972) in The Enzymes (Boyer, P. D., Ed.) pp 666-868, Academic Press, New York. IUPAC-IUB JCBN (1966) Abbreviations and symbols for chemical names of special interest in biological chemistry. Revised tentative rules. J Biol Chem 241, 527-533. IUPAC-IUB JCBN (1968) A one-letter notation for amino acid sequences. Tentative rules. JBiol Chem 243, 3557-3559. IUPAC-IUB JCBN (1970) Abbreviations and symbols for the description of the conformation of polypeptide chains. / . Biol. Chem. 245, 6489-6497. IUPAC-IUB JCBN (1980) Conformational nomenclature for five and six-membered ring forms of monosaccharides and their derivatives: recommendations. Eur. J. Biochem. Ill, 295-298. Bibliography 64 IUPAC-IUB JCBN (1997) Nomenclature of carbohydrates. Recommendations. Carbohydrate Res. 297, 1-90. Jones, T. A. , Zhou, J.-Y., Cowan, S. W., and Kjeldgaard, M . (1991) Improved methods for building protein molecules in electron density maps and the location of errors in these maps. Acta. Crystallogr. A47, 110-119. Joshi, M . D., and Mcintosh, L . P. (1998) personal communication. Joshi, M . D., Hedberg, A. , and Mcintosh, L . P. (1997) Complete measurement of the pK, values of the carboxyl and imidazole groups in Bacillus circulans xylanase. Protein Sci. 6, 2667-2670. Klein, C , Hollender, J., Bender, H . , and Schulz, G. E. (1992) Catalytic center of cyclodextrin glycosyltransferase derived from X-ray structure analysis combined with site-directed mutagenesis. Biochemistry 31, 8740-8746. Koshland, D. E. (1953) Stereochemistry and the mechanism of enzymatic reactions. Biol. Rev. 28,416-436. Krengel, U . , and Dijkstra, B . W. (1996) Three-dimensional structure of Endo-l,4-p-xylanase I from Aspergillus niger. molecular basis for its low pH optimum. J. Mol. Biol. 263, 70-78. Laskowski, R. A . , MacArthur, M . W., Moss, D. S., and Thornton, J. M . (1993) PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283-291. Lawson, C. L . , van Montfort, R., Strokopytov, B. , Rozeboom, H . J., Kalk, K. H . , de Vries, G. E., Penninga, D., Dijkhuizen, L . , and Dijkstra, B . W. (1994) Nucleotide sequence and X-ray structure of cyclodextrin glycosyltransferase from Bacillus circulans strain 251 in a maltose-dependent crystal form. J. Mol. Biol. 236, 590-600. Luzzati, P. V . (1952) Traitement statistique des erreurs dans la determination des structures cristallines. Acta. Crystallogr. A50, 803-810. McCarter, J. D., and Withers, S. G. (1994) Mechanisms of enzymatic glycoside hydrolysis. Curr. Opin. Struct. Biol. 4, 885-892. McDonald, I. K. , and Thorton, J. M . (1994) Satisfying hydrogen bonding potential in proteins. J. Mol. Biol. 238, 777-793. Bibliography 65 Mcintosh, L . P., Hand, G., Johnson, P. E., Joshi, M . D., Korner, M . , Plesniak, L . A . , Ziser, L . , Wakarchuk, W. W., and Withers, S. G. (1996) The pK a of the general acid/base carboxyl group of a glycosidase cycles during catalysis: a 1 3 C - N M R study of Bacillus circulans xylanase. Biochemistry 35, 9958-9966. McPherson, A . (1982) Preparation and analysis of protein crystals, John Wiley & Sons, New York. McRee, D. (1993) Practical protein crystallography, Academic Press, San Diego. Miao, S., Ziser, L . , Aebersold, R., and Withers, S. G. (1994) Identification of glutamic acid 78 as the active site nucleophile in Bacillus subtilis xylanase using electrospray tandem mass spectrometry. Biochemistry 33, 7027-7032. Namchuk, M . N . , and Withers, S. G. (1995) Mechanism of Agrobacterium (3-glucosidase: kinetic analysis of the role of noncovalent enzyme/substrate interactions. Biochemistry 34, 16194-16202. Notenboom, V . , Birsan, C., Warren, R. A . J., Withers, S. G., and Rose, D. R. (1998a) Exploring the cellulose/xylan specificity of the p-l,4-glycanase cex from Cellulomonas fimi through crystallography and mutation. Biochemistry 37, 4751-4758. Notenboom, V. , Birsan, C , Nitz, M . , Rose, D. R., Warren, R. A . J., and Withers, S. G. (1998b) Insights into transition state stabilization of the p-l,4-glycosidase Cex by covalent intermediate accumulation in active site mutants. Nat. Struct. Biol. 5, 812-818. Otwinowski, Z., and Minor, W. (1997) Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307-326. Paice, M . G. , Gurnagul, N . , Page, D. H . , and Jurasek, L . (1992) Mechanism of hemicellulose-directed prebleaching of kraft pulp. Enzyme Microb. Technol. 14, 272-276. Ramachandran, G. N . , and Sasisekharan, V . (1968) Conformation of polypeptides and proteins. Adv. Protein Chem. 23, 283-438. Sinnott, M . L . (1990) Catalytic mechanism of enzymic glycosyl transfer. Chem. Rev. 90, 1171-1202. Stoddard, J. F. (1971) Stereochemistry of Carbohydrates, John Wiley and Sons. Stout, G. H. , and Jensen, L . H. (1989) X-ray structure determination. A practical guide., John Wiley & Sons, New York. Bibliography 66 Strynadka, N . C , and James, M . N . (1991) Lysozyme revisited: crystallographic evidence for distortion of an N-acetylmuramic acid residue bound in site D. / . Mol. Biol. 220, 401-424. Sulzenbacher, G., Driguez, H . , Henrissat, B. , Schiilein, M . , and Davies, G. J. (1996) Structure of the Fusarium oxysporum endoglucanase I with a nonhydrolyzable substrate analogue: substrate distortion gives rise to the preferred axial orientation for the leaving group. Biochemistry 35, 15280-15287. Tews, I., Perrakis, A . , Oppenheim, A. , Dauter, Z. , Wilson, K. S., and Vorgias, C. E. (1996) Bacterial chitobiase structure provides insight into catalytic mechanism and the basis of Tay-Sachs disease. Nat. Struct. Biol. 3, 638-648. Torronen, A . , and Rouvinen, J. (1997) Structural and functional properties of low molecular weight endo-l,4-fi-xylanases. J. Biotech. 57, 137-149. Tull, D., and Withers, S. G. (1994) Mechanisms of cellulases and xylanases: a detailed kinetic study of the exo-R-l,4-glycanase from Cellulomonas fimi. Biochemistry 33, 6363-6370. Vyas, N . K . (1991) Atomic features of protein-carbohydrate interactions. Curr. Op. Struct. Biol. 1,132-1 AO. Wakarchuk, W., Methot, N . , Lanthier, P., Sung, W., Seligy, V . , Yaguchi, M . , To, R., Campbell, R., and Rose, D. (1992) in Xylan and xylanases (Visser, J., Ed.) pp 439-442, Elsevier Science, Amsterdam. Wakarchuk, W. W., Campbell, R. L. , Sung, W. L. , Davoodi, J., and Yaguchi, M . (1994) Mutational and crystallographic analyses of the active site residues of the Bacillus circulans xylanase. Protein Sci. 3, 467-475. White, A. , Tull, D., Johns, K. , Withers, S. G., and Rose, D. R. (1996) Crystallographic observation of a covalent catalytic intermediate in a P-glycosidase. Nat. Struct. Biol. 3, 149-154. White, A . , and Rose, D. R. (1997) Mechanism of catalysis by retaining |3-glycosyl hydrolases. Curr. Opin. Struct. Biol. 7, 645-651. Wilson, A . J. C. (1942) Determination of absolute scale from relative x-ray intensity data. Nature 150, 151-152. Withers, S. G. , Street, I. P., Bird, P., and Dolphin, D. H . (1987) 2-Deoxy-2-fluoroglucosides: A novel class of mechanism-based glucoside inhibitors. / . Am. Chem. Soc. 109, 7530-7531. Bibliography 67 Withers, S. G. , Rupitz, K. , and Street, I. P. (1988) 2-Deoxy-2-fluoro-D-glycosyl fluorides. A new class of specific mechanism-based glycosidase inhibitors. J. Biol. Chem. 263, 7929-7932. Withers, S. G., and Aebersold, R. (1995) Approaches to labeling and identification of active site residues in glycosidases. Protein Sci. 4, 361-372. Ziser, L . , Setyawati, I., and Withers, S. G. (1995) Syntheses and testing of substrates and mechanism-based inactivators for xylanases. Carbohydrate Res. 274, 137-153. 

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.831.1-0099366/manifest

Comment

Related Items