UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mechanistic investigation of family 4 glycosidases : a novel redox-elimination mechanism in the hydrolysis… Yip, Vivian Ling Yee 2007

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_2007-319673.pdf [ 35.53MB ]
Metadata
JSON: 831-1.0060173.json
JSON-LD: 831-1.0060173-ld.json
RDF/XML (Pretty): 831-1.0060173-rdf.xml
RDF/JSON: 831-1.0060173-rdf.json
Turtle: 831-1.0060173-turtle.txt
N-Triples: 831-1.0060173-rdf-ntriples.txt
Original Record: 831-1.0060173-source.json
Full Text
831-1.0060173-fulltext.txt
Citation
831-1.0060173.ris

Full Text

] Mechanistic Investigation of Family 4 Glycosidases: A Novel Redox-Elimination Mechanism in the Hydrolysis of Glycosides by VIVIAN LING YEE YIP B.Sc , University of British Columbia, 2002 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Chemistry) THE UNIVERSITY OF BRITISH COLUMBIA May 2007 © Vivian L.Y. Yip, 2007 Abstract n Abstract The chemical mechanisms of glycoside hydrolases, which catalyze the hydrolysis of glycosidic linkages, are some of the most extensively studied amongst enzymatic reactions. Decades of research have led to the widespread acceptance of the two general glycosidase mechanisms - . both involving nucleophilic displacement steps and oxocarbenium ion-like transition states - first proposed by Koshland in 1953. Glycoside hydrolase family 4 is an anomaly within this large class of enzymes. Not only does this family uniquely contain both a- and p-glycosidases, but members of family 4 also require both a divalent metal (Mn 2 + , Co 2 + , Ni 2 + ) and a N A D + cofactor for activity. The unusual cofactor requirement, coupled with the observation of solvent deuterium incorporation into C2 of the reaction product prompted the proposal of a mechanism involving key NAD+-mediated redox steps as well as elimination of the glycosidic oxygen. Mechanistic and structural analyses of BglT (a 6-phospho-p-glucosidase from Thermotoga maritima) and GlvA (a 6-phospho-a-glucosidase from Bacillus subtilis) were performed, revealing a common mechanism for both the a- and p-glycosidases in this family. The key steps include: 1) C3 hydride abstraction by the N A D + cofactor and consequent oxidation of the C3 hydroxyl; 2) abstraction of the C2 proton via general base catalysis; 3) a,p-elimination of the aglycone; 4) 1,4-Michael-like addition of water to the a,p-unsaturated intermediate; 5) reprotonation at C2; and finally 6) reduction of the C3 carbonyl via oxidation of the "on-board" NADH cofactor. Primary kinetic isotope effects and Bronsted relationships reveal that the C3 hydride and C2 proton abstractions are both partially rate-limiting and that the C l - O l linkage is cleaved rapidly. The evidence suggests that both BglT and GlvA utilize an ElCb-type mechanism. Currently, only family 4 glycosidases are known to utilize an elimination mechanism proceeding via anionic transition state(s). Therefore, they stand in stark contrast to other glycoside hydrolases that use "classical" nucleophilic displacement mechanisms and oxocarbenium ion-like transition states. Structural analyses suggest that a tyrosine residue found in close proximity to the C2 proton of the substrate is oriented ideally to act as the catalytic base for C2 deprotonation. Consistent with this data, mutants in which this tyrosine residue have been replaced by phenylalanine or alanine have significantly lower activity than the wild type enzyme. Direct evidence for the role of N A D + was obtained by Abstract reduction in situ using NaBKL leading to an inactive enzyme that could be reactivated by the addition of excess N A D + . This was accompanied by the expected UV-vis spectrophotometric changes. Furthermore, in the BglT Y241A mutant, the deprotonation/elimination step is slowed sufficiently that steady state accumulation of the reduced nicotinamide cofactor N A D H is observed during catalysis. Some clues as to the origins of this unusual class of enzymes come from their structural similarities to a-hydroxy organic acid dehydrogenases. Further, the proposed mechanism shows striking resemblances to those of NAD+-dependent sugar dehydratases and SAH-hydrolases. Table of Contents iv Table of Contents Abstract 1 1 Table of Contents iv List of Figures x n List of Tables x v i l List of Schemes x x List of Abbreviations x x ' Amino Acid Abbreviations x x " i Publications X X 1 V Acknowledgements x x v i Chapter 1 - Introduction 1 1.1 General Introduction 2 1.2 Enzymatic Carbohydrate Degradation 3 1.2.1 Glycosidases 3 1.2.1.1 Subsite Nomenclature 5 1.2.1.2 Glycosidase Mechanisms 6 1.2.1.2.1 Inverting Glycosidases 6 1.2.1.2.2 Retaining Glycosidases 7 1.2.1.2.3 Mechanistic Anomalies 8 1.2.2 Family 4 Glycosidases 10 1.2.2.1 An Eccentric Family 10 1.2.2.2 General Mechanistic Consideration 12 1.2.3 Other Carbohydrate-Degrading Enzymes Utilizing Elimination Mechanisms 13 1.2.3.1 Polysaccharide Lyases 13 1.2.3.2 a-Glucan Lyase 14 1.3 NAD+-Dependent Enzymes 18 1.3.1 Dehydrogenases 19 1.3.2 Epimerases 20 1.3.3X Dehydratases 22 Table of Contents v 1.4 Proposed GH4 Mechanisms 24 1.4.1 Proposed Mechanism #1 24 1.4.2 Proposed Mechanism #2 25 1.4.3 Proposed Mechanism #3 27 1.5 Specific Aims of Thesis 29 Chapter 2 - 6-Phospho-P-glucosidase Bg lT from Thermotoga maritima 31 2.1 Introduction and Specific Aims 32 2.2 Kinetic Characterization 33 2.2.1 Design of Substrates and Kinetic Assays 33 2.2.2, Dinucleotide Cofactor 37 2.2.2.1 Dinucleotide Cofactor Specificity and Determination of the/^d Value 37 2.2.2.2 Kinetic and Spectroscopic Investigation of Dinucleotide Cofactor Reduction 38 2.2.3 Divalent Metal Specificity and Determination of the Kd Value .. 40 2.2.4 Reducing Conditions 41 2.2.5 Results of Kinetic Characterization of BglT 43 2.3 Mechanistic Studies 46 2.3.1 Stereochemical Outcome Determination 46 2.3.2 Solvent Deuterium Isotopic Exchange 50 2.3.2.1 Solvent Deuterium Isotopic Exchange - Methanolysis... 50 2.3.2.2 Solvent Deuterium Isotopic Exchange - Substrate and Product 51 2.3.3 Active Site Architecture and Mechanistic Implications 55 2.3.4 pH-Dependence 58 2.3.5 Types of Elimination Reactions 61 2.3.6 Linear Free Energy Relationship 63 2.3.6.1 Synthesis of Substrates for Linear Free Energy Relationship 63 2.3.6.2 Linear Free Energy Relationship - Results and Discussion 65 Table of Contents VI 2.3.7 Kinetic Isotope Effects 70 2.3.7.1 Synthesis of Substrates for Kinetic Isotope Effect Measurements 70 2.3.7.2 Kinetic Isotope Effects - Results and Discussion 73 2.3.7.3 Kinetic Isotope Effects in D 2 0 Buffer 75 2.3.7.3.1 pD-Dependence 75 2.3.7.3.2 Solvent Kinetic Isotope Effect 77 2.3.7.3.3 Determination of Primary Kinetic Isotope Effects for 2[2H]4NPf3G6P (2.80) in D 2 0 Buffer 78 2.3.8 Proposed E l C b mechanism 79 2.3.9 Potential Inhibitors of BglT 80 2.3.9.1 Deoxy-6-phospho-P-D-glucopyranosides as Inhibitors ofBlgT 80 2.3.9.2 6-Phospho-oc-D-glucopyranosides as Inhibitors of BglT 82 2.3.9.3 Common Glycosidase Inhibtors 83 2.3.9.3.1 2-Deoxy-2-fluoro-6-phospho-p-D-glucosides as Inhibitors of BglT 83 2.3.9.3.2 Thioglycosides as Potential Inhibitors 84 2.3.9.4 Inhibition Studies - Results and Discussion 85 2.3.9.5 Potential Solvent Isotopic Exchange with Competitive Inhibitors 87 2.4 Mutagenesis Studies.. 89 2.4.1 Previous Mutagenesis Studies 89 2.4.2 Design of BglT Mutants and CD Spectroscopic Analysis 90 2.4.3 Dinucleotide Cofactor 91 2.4.3.1 Determination of the KA Value for N A D + 91 2.4.3.2 Kinetic and Spectroscopic Investigation of N A D + Reduction 92 2.4.4 Determination of the Kd Value for M n 2 + 93 Table of Contents vii 2.4.5 Michaelis-Menten Kinetic Analysis of BglT Y241F and BglT Y241A 94 2.4.6 Direct Observation of the N A D H Formation in BglT Y241A.... 96 2.4.7 pH-Dependence 98 2.4.8 Kinetic Isotope Effects 100 2.5 Conclusions 103 Chapter 3 - 6-Phospho-cc-glucosidase GlvA from Bacillus subtilis 105 3.1 Introduction and Specific Aims 106 3.2 Kinetic Characterization 106 3.2.1 Substrate Specificity 106 3.2.2 Dinucleotide Cofactor 111 3.2.2.1 Dinucleotide Cofactor Specificity and Determination of the K d Value I l l 3.2.2.2 Kinetic and Spectroscopic Investigation of Dinucleotide Cofactor Reduction 113 3.2.3 Divalent Metal Specificity and Determination of the Kd Value ..115 3.2.4 Reducing conditions 116 3.2.5 Results of Kinetic Characterization of GlvA 117 3.3 Mechanistic Studies 119 3.3.1 Stereochemical Outcome Determination 119 3.3.2 Solvent Deuterium Isotopic Exchange 126 3.3.3 Active Site Architecture and Mechanistic Implications 129 3.3.4 pH-Dependence 137 3.3.5 Linear Free Energy Relationship 139 3.3.5.1 Synthesis of Substrates for Linear Free Energy Relationship 139 3.3.5.2 Linear Free Energy Relationship - Aryl 6-Phospho-(3-D-Glucopyranosides 141 3.3.5.3 Linear Free Energy Relationship -6-Phospho-a-D-Glucopyranosides 144 3.3.6 Kinetic Isotope Effects 146 Table o f Contents v i i i 3.3.6.1 Preparation of Substrates for Kinetic Isotope Effect Measurement 146 3.3.6.2 Kinetic Isotope Effects - Results 148 3.3.7 Mechanistic Analysis via KfE and Linear Free Energy Relationship 149 3.3.8 Potential Inhibitors... 154 3.4 Proposed Mechanism and Conclusions 155 Chapter 4 - Thioglycosides and Elimination Mechanisms 157 4.1 Introduction - Thioglycosides and "Classical" Glycosidases 158 4.2 Cleavage of C-S linkages via Eliminative Mechanisms 160 4.3 Thioglycosides - Substrates for BglT from Thermotoga maritima 165 4.4 Conclusions 169 Chapter 5 - Materials and Methods 170 5.1 Generous Gifts 171 5.2 General Synthesis 171 5.2.1 General Methods 172 5.2.1.1 General Acetylation Methodology 172 5.2.1.2 General Procedure for Synthesis of Glycosyl Bromides . 173 5.2.1.3 General Zemplen Deprotection 173 5.2.1.4 General Deprotection Using HC1 ( s ) 173 5.2.1.5 Koenigs Knorr Reaction 173 5.2.1.6 Stannic Chloride-Catalyzed Glucosidation 174 5.2.1.7 General BglK-Catalyzed Phosphorylation of p-D-Glucosides 174 5.2.1.8 General Chemical Phosphorylation Using Diphenyl Chlorophosphate 175 5.2.1.9 General Procedure for Diphenyl Phosphate Deprotection 175 5.2.1.10 General Procedure for Chemical Phosphorylation Using Phosphorus Oxychloride 176 Table of Contents ix 5.2.2 Synthesis and Compound Characterization 176 5.3 Enzyme Kinetics 229 5.3.1 General Methods 229 5.3.1.1 Conditions for Measurement of Initial Rates 230 5.3.1.2 Conditions for the Substrate Depletion Method 230 5.3.1.3 Standard Procedures for Preparing Buffers and Enzyme in D 2 O Solutions 230 5.3.2 Enzyme Kinetics with BglT 231 5.3.2.1 Buffer Systems 231 5.3.2.2 BglT - Direct UV-vis assay 231 5.3.2.3 Glucose 6-Phosphate Dehydrogenase Coupled Assay.... 231 5.3.2.4 Abg Coupled assay 233 5.3.2.5 Determination of K6 Value for M n 2 + 233 5.3.2.6 Determination of Kd Value for N A D + 234 5.3.2.7 Kinetic and Spectroscopic Investigation of Cofactor Reduction 234 5.3.2.8 Stereochemical Outcome Determination by Methanolysis 235 5.3.2.9 Solvent Deuterium Isotopic Exchange 235 5.3.2.10 pH-Dependence 236 5.3.2.11 Bronsted Analysis 237 5.3.2.12 Deuterium Kinetic Isotope Effect Measurements 237 5.3.2.13 pD-Dependence 238 5.3.2.14 Solvent Kinetic Isotope Effect 239 5.3.2.15 Primary Kinetic Isotope Effect Measurements for 2[2H]4NPp>G6P in D 2 0 Buffer 239 5.3.2.16 Determination of K, Values 240 5.3.3 Enzyme Kinetics with GlvA 240 5.3.3.1 Buffer Systems 240 5.3.3.2 Michaelis-Menten Kinetics - 4NPocG6P and 4NP0G6P by the Direct UV-vis assay 241 Table of Contents x 5.3.3.3 Glucose 6-Phosphate Dehydrogenase Coupled Assay.... 241 5.3.3.4 Determination of Kd Value for M n 2 + 242 5.3.3.5 Determination of Kd Value for N A D + 242 5.3.3.6 Kinetic and Spectroscopic Investigation of Cofactor Reduction 243 5.3.3.7 Stereochemical Outcome Determination by Methanolysis 244 5.3.3.8 Solvent Deuterium Incorporation 244 5.3.3.9 pH-Dependence 245 5.3.3.10 Bronsted Analysis 246 5.3.3.11 Deuterium Kinetic Isotope Effect Measurements 247 5.3.3.12 Determination of K\ Values 248 5.4 Molecular Biology 248 5.4.1 General Materials and Methods 248 5.4.2 Overexpression of the Gene (bglt) Encoding BglT from Thermotoga maritima 249 5.4.3 Site-Directed Mutagenesis 249 5.4.3.1 Preparation of the Clone pET22bgltY241F 250 5.4.3.2 Preparation of the Clone pET22bgltY24IA 251 5.4.3.3 BglT Y241F Overexpression and Purification 252 5.4.3.4 BglT Y241A Overexpression and Purification 253 5.5 CD Spectroscopy 253 References 254 Appendices 265 Appendix 1 - Kinetic Isotope Effects 266 Appendix 2 - Linear Free Energy Relationships 272 Appendix 3 - Graphical Representation of BglT Data 275 Appendix 4 - Graphical Representation of GlvA Data 280 Appendix 5 - Graphical Representation of Inhibition Data 285 Appendix 6 - Ligand Binding Curves 293 Appendix 7 - Mass Spectrometry Data 295 Table of Contents xi Appendix 8 - Sequence Alignment 297 List of Figures xii List of Figures Figure 1.1 General hydrolysis reaction catalyzed by a glycosidase 3 Figure 1.2 General reactions catalyzed by retaining and inverting glycosidases 4 Figure 1.3 Illustration of the nomenclature used in describing sugar-binding subsites in glycosidases 6 Figure 1.4 General mechanism utilized by inverting glycosidases 7 Figure 1.5 General mechanism utilized by retaining glycosidases 8 Figure 1.6 General mechanism utilized by a retaining iV-acetyl-p-hexosaminidase 8 Figure 1.7 The G H l myrosinase mechanism in which L-ascorbate replaces a carboxylate amino acid residue as the general base catalyst 9 Figure 1.8 Comparison of the oxocarbenium ion-like transition states of glycosidases.... 10 Figure 1.9 Proposed elimination mechanism of chondroitin A C lyase 14 Figure 1.10 Proposed mechanism of the Gracilariopsis oc-1,4-glucan lyase from GH31 15 Figure 1.11 Proposed mechanism for an a-glucosidase-catalyzed hydration of D-glucal 17 Figure 1.12 Structures of the dinucleotide cofactors in their oxidized ( N A D + or NADP + ) and reduced (NADH or NADPH) forms 19 Figure 1.13 Proposed mechanism of L-lactate dehydrogenase 20 Figure 1.14 Malate dehydrogenase reaction 20 Figure 1.15 Proposed mechanism of the human UDP-galactose 4-epimerase 21 Figure 1.16 Proposed mechanism of GDP-D-mannose:GDP-L-galactose epimerase 22 Figure 1.17 Proposed mechanism of RmlB from Salmonella entehca serovar typhimurium 23 Figure 1.18 Proposed mechanism #1 for GH4 25 Figure 1.19 Proposed mechanism #2 for GH4 26 Figure 1.20 Proposed mechanism #3 for GH4 28 List of Figures xin Figure 2.1 Schematic for the BglT-catalyzed hydrolysis of C6'P for the G6PDH coupled assay 34 Figure 2.2 Schematic for the BglT-catalyzed hydrolysis of 4NPpC6 ,P (2.10) for the Abg and G6PDH coupled assays 35 Figure 2.3 BglT-catalyzed hydrolysis of the chromogenic substrate 4NPf3G6P (2.13) 36 Figure 2.4 Ligand binding curve of N A D + for BglT 37 Figure 2.5 Absorbance spectra of BglT in its oxidized (NAD + ) and reduced (NADH) forms 39 Figure 2.6 Assay of BglT in the oxidized (NAD + ) and reduced (NADH) states 40 Figure 2.7 Ligand binding curve of M n 2 + for BglT 41 Figure 2.8 Schematic of the BglT-catalyzed methanolysis of 4NPpG6P (2.13) for the stereochemical outcome determination 47 Figure 2.9 1 H NMR spectra of the BglT methanolysis reaction 48 Figure 2.10 1 H NMR spectra of the BglT methanolysis reaction at elevated temperatures 49 Figure 2.11 1 H NMR of the BglT-catalyzed methanolysis of 4NPpG6P (2.13) in D 2 0 buffer and 5 M C D 3 O D 51 Figure 2.12 ' H NMR of the BglT-catalyzed hydrolysis of 4NPpG6P (2.13) in D 2 0 buffer 53 Figure 2.13 1 H NMR analysis illustrating the solvent deuterium incorporation into compound 2.23 55 Figure 2.14 A diagram of the interaction of BglT with M n 2 + , N A D + and G6P 56 Figure 2.15 Stereoview of the active site of BglT 57 Figure 2.16 Proposed stabilization of the enediolate intermediate by the divalent metal ion 58 Figure 2.17 pH-Activity profiles of BglT 60 Figure 2.18 More O'Ferrall-Jencks diagram for E l , E2, and E1 C b mechanisms 61 Figure 2.19 Chemical representation of the potential E l , E2, and E l c b elimination in the BglT mechanism 62 Figure 2.20 Bronsted plots of the BglT-catalyzed cleavage of a series of List of Figures xiv aryl 6-phospho-p-D-glucopyranosides (2.13, 2.47—2.56) with the corresponding pKa values for the leaving group phenol 67 Figure 2.21 Bronsted plots of the BglT-catalyzed cleavage of a series of aryl 6-phospho-(3-D-glucopyranosides (2.13, 2.47—2.56), C6'P (2.2), and MepG6P (2.58) with the corresponding pATa values for the leaving group phenol 69 Figure 2.22 pL-Dependence of the kcJKu for BglT 76 Figure 2.23 Proposed E1 Cb mechanism of BglT 80 Figure 2.24 ' H NMR of potential solvent deuterium incorporation with competitive inhibitors 4NP2deoxypG6P (2.96) and 4NP2FpG6P (2.100) 88 Figure 2.25 CD spectra of BglT WT, BglT Y241F, and BglT Y241A at 50 °C 90 Figure 2.26 Absorbance spectra of BglT Y241A in its oxidized (NAD*) and reduced (NADH) forms 92 Figure 2.27 Assay of BglT mutants in the oxidized (NAD + ) and reduced (NADH) states 93 Figure 2.28 Amino acid residues in close proximity to proposed catalytic base Tyr241 and C2 of the substrate 96 Figure 2.29 Burst phase kinetics for the hydrolysis of 4NPPG6P (2.13) by BglT Y241A 97 Figure 2.30. The steady state accumulation of N A D H during the BglT Y241A-catalyzed hydrolysis of C6'P (2.2) NADH 98 Figure 2.31 pH-Activity profiles of BglT mutants 99 Figure 3.1 Potential substrates of GlvA 108 Figure 3.2 Direct UV-vis spectrophotometric assay for the GlvA-catalyzed hydrolysis of 4NPaG6P (2.17) 110 Figure 3.3 Direct UV-vis spectrophotometric assay for the GlvA-catalyzed hydrolysis of 4NPpG6P (2.13) 110 Figure 3.4 Illustration of the G6PDH coupled assay for measuring kinetic parameters for GlvA 111 List of Figures X V Figure 3.5 Ligand binding curves of N A D for GlvA 112 Figure 3.6 Absorbance spectra of GlvA with the dinucleotide cofactor in its oxidized (NAD + ) and reduced (NADH) states 113 Figure 3.7 Assay of GlvA in the oxidized (NAD + ) and reduced (NADH) states 114 Figure 3.8 Ligand binding curves of M n 2 + f o r GlvA 116 Figure 3.9 Schematic of the GlvA-catalyzed methanolysis of 4NPaG6P (2.17) for the determination of the stereochemical outcome 119 Figure 3.10 'H NMR spectra showing methanolysis of 4NPocG6P (2.17) by GlvA ..121 Figure 3.11 Schematic of the GlvA-catalyzed methanolysis of 4NPaG6P (2.17) in D2O/CD3OD 122 Figure 3.12 Schematic of the GlvA-catalyzed methanolysis of 4NPpG6P (2.13) for the stereochemical outcome determination : 123 Figure 3.13 Schematic of the GlvA-catalyzed methanolysis of 4NPpG6P (2.13) in D 2 O / C D 3 O D 123 Figure 3.14 ' H NMR spectra showing the methanolysis of 4NPpG6P (2.13) by GlvA 124 Figure 3.15 1 H NMR of the GlvA-catalyzed hydrolysis of 4NPaG6P (2.17) in D 2 0 buffer 127 Figure 3.16 1 H NMR spectra illustrating the solvent deuterium exchange for the GlvA-catalyzed hydrolysis of 4NPpG6P (2.13) 128 Figure 3.17 1 H N M R analysis illustrating the solvent deuterium incorporation into 1,5-anhydroglucitol 6-phosphate (2.23) 129 Figure 3.18 Overlap of the structures of BglT, AlgA, Agu4B from Thermotoga maritima, and GlvA from Bacillus subtilis 130 Figure 3.19 CD spectra of BglT and GlvA 130 Figure 3.20 Stereoview of the GlvA active site 131 Figure 3.21 Active site of BglT overlayed with that of GlvA 132 Figure 3.22 Active site architecture of GlvA 133 Figure 3.23 Sequence alignment of some 6-phospho-glycosidases from GH4 134 Figure 3.24 An overlay of the active sites of GlvA (blue) from Bacillus subtilis, and BglT (green) and AglA (pink) from Thermotoga maritima 136 List of Figures xvi Figure 3.25 pH-Activity profiles of GlvA 138 Figure 3.26 Bronsted plots of the GlvA-catalyzed hydrolysis of a series of aryl 6-phospho-(3-D-glucopyranosides at pH 7.5 (HEPES buffer) with the corresponding pKa values for the leaving group phenol 142 Figure 3.27 Bronsted plots of the GlvA-catalyzed hydrolysis of a series of aryl 6-phospho-(3-D-glucopyranosides at pH 8.4 (Tris buffer) with the corresponding pKa values for the leaving group phenol 143 Figure 3.28 Bronsted plots of the GlvA-catalyzed hydrolysis of a series of 6-phospho-a-D-glucopyranosides at pH 7.5 (HEPES buffer) with the corresponding pA'a values for the leaving group 145 Figure 3.29 Bronsted plots of the GlvA-catalyzed hydrolysis of a series of 6-phospho-a-D-glucopyranosides atpH 8.4 (Tris buffer) with the corresponding p/C values for the leaving group 146 Figure 3.30 Graphical representation of the reaction coordinate for the GlvA-catalyzed hydrolysis of a - and p-substrates 153 Figure 3.31 Proposed El C b mechanism for GlvA 156 Figure 4.1 List of common glycosidase inhibitors 158 Figure 4.2 Comparison of O-glucoside and thioglucoside 158 Figure 4.3 General glucosinolate structure 159 Figure 4.4 Proposed mechanisms of enzymes that catalyze C-S bond cleavage 164 Figure 4.5 Schematic for the BglT-catalyzed hydrolysis of C6'P (2.2), 4NPpC6'P (2.10), S-4NPpC6'P (4.3), and 5-C6'P (4.5) using the G6PDH coupled assay 167 List o f Tables xvn List of Tables Table 2.1 Test of dinucleotide cofactors for activation of BglT 38 Table 2.2 Test of divalent metal ions for activation of BglT 41 Table 2.3 Test of reducing reagents for activation of BglT 43 Table 2.4 Comparison of the kinetic parameters for the BglT-catalyzed hydrolysis of the natural substrate C6'P (2.2) and the chromogenic substrate 4NPpG6P (2.13) 44 Table 2.5 Comparison of the kinetic parameters for the hydrolysis of 4NPpC6'P (2.10) via the Abg and the G6PDH coupled assays 44 Table 2.6 Substrate specificity of BglT 46 Table 2.7 Summary of p/Gi, \>K*2, and pH o pt values obtained for BglT from the plots of kcat vs pH and of kcJKM vs pH 60 Table 2.8 Michaelis-Menten kinetic parameters for the hydrolysis of a series of aryl 6-phospho-p-D-glucosides by BglT at 50 °C, pH 7.5 66 Table 2.9 Michaelis-Menten kinetic parameters for the hydrolysis of C6'P (2.2) and MepG6P (2.58) by BglT as measured by the G6PDH coupled assay 68 Table 2.10 Summary of KJEs determined for BglT 74 Table 2.11 Summary of apparent pA'a values and p H o p t values obtained from the pH- and pD-activity profiles 77 Table 2.12 Summary of Michaelis-Menten kinetic parameters obtained using 4NPpG6P (2.13) in H 2 0 and D 2 0 77 Table 2.13 Solvent KlEs measured for the BglT-catalyzed hydrolysis of 4NPpG6P (2.13) 78 Table 2.14 Comparison of KlEs for 2[ 2H]4NPpG6P (2.79) in H 2 0 and D 2 0 78 Table 2.15 Approximate Kt values of the potential inhibitors of BglT 85 Table 2.16 Summary of N A D + dissociation constants with BglT WT, BglT Y241F, and BglT Y241A 92 Table 2.17 Summary of M n 2 + dissociation constants with BglT WT, BglT Y241F, and BglT Y241A 94 Table 2.18 Kinetic parameters determined for BglT WT, BglT Y241F, List o f Tables xvui and BglT Y241A 95 Table 2.19 Summary of apparent pKa and pH o p t values for BglT WT, BglT Y241F, and BglT Y241A 100 Table 2.20 KIEs measured for BglT WT, BglT Y241F, and BglT Y241A 103 Table 3.1 Substrate specificity of GlvA 109 Table 3.2 Dinucleotides tested by Thompson and co-workers for activation of GlvA I l l Table 3.3 Summary of dissociation constants for N A D + with GlvA 112 Table 3.4 Divalent metal ions tested by Thompson and co-workers for the activation of GlvA 115 Table 3.5 Dissociation constants measured for M n 2 + binding to GlvA 116 Table 3.6 Reducing reagents tested for activation of GlvA 117 Table 3.7 Summary of kinetic parameters for GlvA-catalyzed hydrolysis of substrates 4NPaG6P (2.17), 4NPpG6P (2.13), and MeaG6P (2.98) 118 Table 3.8 Summary of pKa values for GlvA as determined from the pH-dependence of kcar and from the pH-dependence of kcJKM 137 Table 3.9 Michaelis-Menten kinetic parameters for the hydrolysis of a series of aryl 6-phospho-p-D-glucophyranosides by GlvA at 37 °C in HEPES buffer (pH 7.5) 142 Table 3.10 Michaelis-Menten kinetic parameters for the hydrolysis of a series of aryl 6-phospho-p-D-glucophyranosides by GlvA at 37 °C in Tris buffer (pi I 8.4) 143 Table 3.11 Michaelis-Menten kinetic parameters for the hydrolysis of a series of 6-phospho-oc-D-glucopyranosides by GlvA at 37 °C in HEPES buffer (pH 7.5) 145 Table 3.12 Michaelis-Menten kinetic parameters for the hydrolysis of a series of 6-phospho-a-D-glucophyranosides by GlvA at 37 °C in Tris buffer (pH 8.4) 146 Table 3.13 Kinetic isotope effect measurements for deuterated substrates of GlvA obtained at pH 7.5 149 Table 3.14 Kinetic isotope effect measurements for deuterated substrates of List of Tables xix GlvA obtained at pH 8.4 149 Table 3.15 Approximate K, values of the potential inhibitors of GlvA 155 Table 4.1 Summary of kinetic parameters for the hydrolysis of O- and S-glucosides by BglT 168 List of Schemes xx List of Schemes Scheme 2.1 Enzymatic synthesis of the natural substrate C6'P (2.1) 33 Scheme 2.2 Chemoenzymatic synthesis of 4NPpC6'P (2.10) 35 Scheme 2.3 Enzymatic synthesis of 4NPpG6P (2.13) 36 Scheme 2.4 Synthesis of 4NPaG6P (2.17) 45 Scheme 2.5 Synthesis of aryl 6-phospho-p-D-glucopyranosides 64 Scheme 2.6 Synthesis of MepG6P (2.58) 65 Scheme 2.7 Synthesis of 3[2H]-D-glucose (2.64) 71 Scheme 2.8 Synthesis of deuterated chromogenic substrates (2.79—2.81) for KIE measurements 72 Scheme 2.9 Synthesis of 1 [ 2H]MepG6P (2.84) 73 Scheme 2.10 Synthesis of 4NP3deoxypG6P (2.94) 81 Scheme 2.11 Enzymatic synthesis of 4NP2deoxypG6P (2.96) 82 Scheme 2.12 Synthesis of MeaG6P (2.98) 83 Scheme 2.13 Enzymatic synthesis of 4NP2FpG6P (2.100) 84 Scheme 2.14 Enzymatic synthesis of S-4NPpG6P (2.102) 85 Scheme 2.15 Enzymatic synthesis of 4MUpG6P (2.108) 94 Scheme 3.1 Synthesis of PaG6P (3.10) 140 Scheme 3.2 Synthesis of 34DNPaG6P (3.14) 140 Scheme 3.3 Synthesis of 1 [ 2H]4NPaG6P (3.21), 2[ 2H]4NPaG6P (3.22), and 3[ 2H]4NPaG6P (3.23) for KIE measurements with GlvA 147 Scheme 3.4 Synthesis of 1 [ 2H]MeaG6P (3.26) 148 Scheme 4.1 Enzymatic synthesis of S-4NPpC6'P (4.3) and S-C6T (4.5) 166 List of Abbreviations xxi List of Abbreviations kcat - catalytic rate constant (turnover number) k\\lko - ratio of catalytic rate constants for protio and deutero substrates KXE - kinetic isotope effect KM - Michaelis constant of a substrate K\ - dissociation constant for an enzyme-inhibitor complex k0bS - pseudo-first order rate constant N M R - nuclear magnetic resonance ppm - parts per million SDS-PAGE - sodium dodecyl sulfate polyacrylamide gel electrophoresis UV - ultraviolet light vis - visible light WT - wild-type 10 - primary 2° - secondary GH4 - glycoside hydrolase family 4 N A D + - P-nicotinamide adenine dinucleotide N A D H - P-nicotinamide adenine dinucleotide, reduced N A D P + - P-nicotinamide adenine dinucleotide phosphate NADPH - p-nicotinamide adenine dinucleotide phosphate, reduced l[ 2H]4NPaG6P - 4-nitrophenyl l-[2H]-6-phospho-a-D-glucopyranoside 2[ 2H]4NPaG6P - 4-nitrophenyl 2-[2H]-6-phospho-a-D-glucopyranoside 3[2H]4NPocG6P - 4-nitrophenyl 3-[2H]-6-phospho-oc-D-glucopyranoside 4NPaG6P - 4-nitrophenyl 6-phospho-a-D-glucopyranoside 34DNPaG6P - 3,4-dinitrophenyl 6-phospho-a-D-glucopyranoside PocG6P - phenyl 6-phospho-a-D-glucopyranoside l[ 2H]4NPpG6P - 4-nitrophenyl l-[2H]-6-phospho-p-D-glucopyranoside 2[ 2H]4NPpG6P - 4-nitrophenyl 2-[2H]-6-phospho-P-D-glucopyranoside 3[2H]4NPpG6P - 4-nitrophenyl 3-[2H]-6-phospho-p-D-glucopyranoside List of Abbreviations xxii 4NPf5G6P - 4-nitrophenyl 6-phospho-p-D-glucopyranoside 24DNPf3G6P - 2,4-dinitrophenyl 6-phospho-p-D-glucopyranoside 25DNPpG6P - 2,5-dinitrophenyl 6-phospho-p-D-glucopyranoside 34DNPpG6P - 3,4-dinitrophenyl 6-phospho-P-D-glucopyranoside 4C2NPPG6P - 4-chloro-2-nitrophenyl 6-phospho-p-D-glucopyranoside 2NPpG6P - 2-nitrophenyl 6-phospho-p-D-glucopyranoside 35DCPpG6P - 3,5-dichlorophenyl 6-phospho-p-D-glucopyranoside 3NPpG6P - 3-nitrophenyl 6-phospho-p-D-glucopyranoside 4CNPpG6P - 4-cyanophenyl 6-phospho-p-D-glucopyranoside PPG6P - phenyl 6-phospho-p-D-glucopyranoside MeaG6P - methyl 6-phospho-a-D-glucopyranoside 1 [ 2H]MeaG6P - methyl l-[2H]-6-phospho-a-D-glucopyranoside MepG6P - methyl 6-phospho-p-D-glucopyranoside l[ 2H]MePG6P - methyl l-[2H]-6-phospho-p-D-glucopyranoside C6'P - cellobiose 6'-phosphate 4NP2deoxypG6P - 4-nitrophenyl 2-deoxy-6-phospho-p-D-glucopyranoside 4NP3deoxyPG6P - 4-nitrophenyl 3-deoxy-6-phospho-p-D-glucopyranoside 4NP2FpG6P - 4-nitrophenyl 2-deoxy-2-fluoro-6-phospho-p-D-glucopyranoside KIE - kinetic isotope effect ESI MS - electrospray ionization mass spectrometry UV-vis - ultraviolet-visible G6P - glucose 6-phosphate G6PDB - glucose 6-phosphate dehydrogenase N M W L - nominal molecular weight limit DTT - Dithiothreitol EC - Enzyme Commission (classification number) of the International Union'of Biochemistry H.PLC - High performance liquid chromatography TLC - Thin layer chromatography Amino Acid Abbreviations xxi n Amino Acid Abbreviations Ala A Alanine Cys C Cystine Asp D Aspartic acid Glu E Glutamic acid Phe F Phenylalanine Gly G Glycine His H Histidine lie I Isoleucine Lys K Lysine Leu L Leucine Met M Methionine Asn N Asparagine Pro P Proline Gin Q Glutamine Arg Pv Arginine Ser S Serine Thr T Threonine Val V Valine Tip W Tryptophan Tyr Y Tyrosine Publications xxiv Publications Parts of the material in this thesis were previously reported in the following publications: 1) Yip, V. L. V.. Thompson, J., and Withers, S. G. (submitted March 16 t h, 2007) Mechanism of GlvA from Bacillus subtilis: A detailed kinetic analysis of a 6-phospho-a-glucosidase from glycoside hydrolase family 4, Biochemistry. 2) Yip, V. L. Y., and Withers, S. G. (2006) Family 4 glycosidases carry out efficient hydrolysis of thiodisaccharides via an a,(3-elimination mechanism, Angewandte-Chemie International Edition, 118, 6325-6328. 3) Yip, V. L. Y., and Withers, S. G. (2006) Breakdown of oligosaccharides by the process of elimination, Current Opinion in Chemical Biology, 10, 147-155. 4) Yip, V. L. Y., and Withers, S. G. (2006) Mechanistic analysis of the unusual redox-elimination sequence employed by Thermotoga maritima BglT: A 6-phospho-p-glucosidase from glycoside hydrolase Family 4, Biochemistry, 45, 571-580. 5) Yip, V. L. Y., and Withers, S. G. (2006) Family 4 glycoside hydrolases are special: The first p-elimination mechanism amongst glycoside hydrolases, Biocatalysis and Biotransformation, 24, \6l-\16. 6) Varrot, A. , Yip, V. L. Y., L i , Y. , Rajan, S. S., Yang, X . , Anderson, W. F., Thompson, J., Withers, S. G., and Davies G. J. (2005) N A D + and metal-ion dependent hydrolysis by Family 4 glycosidases: Structural insight into specificity for phospho-p-D-glucosides, Journal of Molecular Biology, 346, 423-435. 7) Yip, V. L. Y., and Withers, S. G. (2004) Nature's many mechanisms for the degradation of oligosaccharides, Organic and Biomolecular Chemistry, 2, 2707-2713. Publications X X V 8) Rajan, S. S., Yang, X., Collart, R, Yip , V . L . Y . . Withers, S. G., Varrot, A. , Thompson, J., Davies, G. J., and Anderson, W. F. (2004) Novel catalytic mechanism of glycoside hydrolysis based on the structure of an NAD+/Mn2 +-dependent phospho-a-glucosidase from Bacillus subtilis, Structure, 12, 1619-1629. 9) Yip , V . L . V. , Varrot, A. , Davies, G. J., Rajan, S. S., Yang, X., Thompson, J., Anderson, W. F., and Withers, S. G. (2004) An unusual mechanism of glycoside hydrolysis involving redox and elimination steps by a Family 4 P-glycosidase from Thermotoga mar itima, Journal of the American Chemical Society, 126, 8354-8355. Acknowledgements xxvi Acknowledgements I would like to thank my supervisor Dr. Stephen Withers for his guidance through the past few years. His dedication and keen interest in the work is inspirational. 1 would also like to thank the many collaborators involved in various aspects of this project. Dr. Annabelle Varrot, Dr. Gideon Davies and Dr. Wayne Anderson for solving x-ray crystal structures. Dr. John Thompson for providing plasmids and protein preps for GlvA and BglK, which greatly facilitated the work. Many co-workers in the Withers lab have also made this a wonderful experience. 1 would like to thank: Dr. Ian Greig for many insightful discussions of my strange kinetic data and his continued interest in this project; Dr. Hannes Miiellegger for his valuable time and patience helping me with molecular biology and provision of Abg mutants; Dr. Hong Ming Chen for being the mini Sigma-Aldrich, and fishing out compounds he just might have around in the freezer; Dr. Susan Hancock for helpful suggestions and careful reading of the thesis; Dr. Michael Jahn for donating his compounds; Dr. Chris Tarling for compound donation and helpful suggestions and sharing his expertise in various aspects of the lab and work; Jacqueline Wicki for help with kinetics; Dr. Shouming He for mass spec analysis of protein samples. The assistance from Ms. Miranda Joyce is also greatly appreciated. I am also in debt to the service staff at Chemistry Department including: the NMR facility for time and guidance with N M R spectroscopy; the MS facility Dr. Yun Ling for providing mass spectrometric analysis for many compounds, and microanalysis by Mr. Minaz Lakha. I would also like to thank Dr. Fred Rosell for assistance with CD spectroscopy. Finally, thank you for my parents and Wendy for their love and support. This would not have been possible without you. Chapter I - Introduction Chapter 1 Introduction Chapter 1 - Introduction 2 Section 1.1. General Introduction. Carbohydrates are an example of nature's most phenomenal designs. At first glance, carbohydrates appear comparatively simple in chemistry when considered in the company of other biological molecules such as nucleic acids and protein, which contain a diversity of functional groups. While nucleic acids are famed as genetic blueprints, proteins are recognized for their abilities to enhance reaction rates such that biological processes are complete on a life-sustaining time-scale. In addition, proteins play important roles in recognition pathways, regulating many inter- and intra-cellular activities. The chemical properties of proteins also give them robust, stable conformations, which are utilized for providing structural integrity. In contrast, carbohydrates are traditionally regarded as possessing merely two major functions— either as reservoirs of metabolic energy or as structural support in plants and exoskeleton invertebrates. Glucose and starch are familiar forms found in food. Many daily-used objects, such as wood, paper and cotton, contain cellulose. It is currently known that carbohydrates conjugated to proteins and lipids are often found to be part of signal recognition systems that regulate intercellular communication as well as intracellular activity.1"5 Carbohydrates offer a unique scaffold for encoding information. Carbohydrates possess far greater structural diversity than polypeptides and oligonucleotides due to the large number of isomeric forms. Polypeptides and oligonucleotides are synthesized on templates and only possess 20 and 4 different natural functional moieties respectively. By contrast, it is estimated that a simple reducing hexasaccharide can have 10 1 2 isomeric forms.6 The large number of isomers arises from the many potential stereoisomers, side branching as well as the possible occurrence of different ring sizes. Furthermore, the glycosidic linkage is one of the most stable bonds found in natural polymers, being 100 times more stable under neutral aqueous conditions than the phosphodiester bond in D N A , 7 which is in turn 1,000 and 100,000 times more stable than the peptide bond in proteins and the phosphodiester bond of R N A respectively.8"10 The estimated half-life for spontaneous hydrolysis of a single glycosidic bond of polysaccharides such as cellulose is approximately 5 million years." As such, carbohydrates offer greater stability as well as structural diversity, which allows for superior encoding abilities to those of their polypeptide and oligonucleotide cousins. An Chapter 1 - Introduction 3 understanding of how carbohydrates mediate intercellular communication as well as intracellular events may lead to applications in biochemistry, medicine and biotechnology.1"5 The applications include potential cures for infections and diseases, such as diabetes, cancer, HIV, and influenza. Section 1.2. Enzymatic Carbohydrate Degradation. Since the glycosidic linkage is one of the most stable bonds found in natural polymers," the catabolism of glycosides must require some external assistance from enzymes for biological processes to proceed on a relevant timescale. Enzymatic degradation of glycosidic linkages is most often accomplished by glycoside hydrolases, also known as glycosidases, (E .C . 3.2.1.-). Some glycoside hydrolases have been reported to catalyze the hydrolysis of glycosides at rates of up to 1000 s"1, which translates to a rate enhancement of 10 l 7-fold when compared to the spontaneous hydrolysis of the equivalent bond." Thus, glycosidases are some of the most proficient enzymes known. Section 1.2.1. Glycosidases. Glycoside hydrolases catalyze the hydrolysis of glycosidic linkages via cleavage of the anomeric C - O bond between the glycone and the aglycone, as shown in Figure 1.1. In many cases, the aglycone is also a carbohydrate molecule, but the aglycone moiety can be a variety of other groups including an aromatic moiety, a phosphate group or a lipid. glycone aglycone Figure 1.1. General hydrolysis reaction catalyzed by a glycosidase. Glycosidases can be classified according to their glycone specificity, substrate anomeric configuration and the stereochemical outcome of the reaction catalyzed (see Figure 1.2). Firstly, if a glycosidase displays higher activity towards glucosides than Chapter 1 - Introduction 4 towards galactosides or mannosides, the enzyme is termed a glucosidase based on its glycone specificity. Secondly, since these enzymes are highly specific (with a few rare exceptions3) for the anomeric configuration of the substrate glycosidic bond, they are assigned as "a" or "(3" glycosidases accordingly. Thirdly, these enzymes are mechanistically grouped as either inverting or retaining glycosidases based on the stereochemical outcome of the reaction catalyzed. Hence, if the anomeric configuration of the resulting hemiacetal is identical to that of the substrate glycoside, the glycosidase is identified as a retaining enzyme. On the other hand, an enzyme that hydrolyzes an a-glycoside to give a hemiacetal with a p-anomeric configuration, or vice versa, is known as an inverting glycosidase (Figure 1.2). Figure 1.2. General reactions catalyzed by retaining and inverting glycosidases. Beyond the general identification of glycosidases, the C A Z Y website (http://afmb.cnrs-mrs.frZ-cazv/CAZY/index.html) developed by Dr. Bernard Henrissat '' The rare exceptions are found in Family 4 glycosidases, which will be discussed in Section 1.2.2. Chapter 1 - Introduction 5 catalogues all known glycosidases into different families based on sequence similarity. 1 2 1 3 To date, there are over 100 known glycosidase families (GH). One of the major benefits of this classification system is that primary sequence identity has been shown to be a remarkably reliable tool in the prediction of tertiary structure, and catalytic mechanism and machinery.14"16 With the exception of Family 4, all members within a family hydrolyze glycosides of the same anomeric configuration and do so with the same stereochemical outcome.b Section 1.2.1.1. Subsite Nomenclature. A subsite nomenclature system similar to that used to describe proteases has been adopted by researchers in studying substrate binding in glycosidases.18 Some glycosidases act on polysaccharides, and it was found that catalytic activity is often affected by binding properties of sugar units far away from the hydrolyzed linkage.19 Therefore, a consistent nomenclature for the sugar Tending subsites is necessary to aid in the understanding and characterization of enzyme-substrate interactions. The currently used system provides a convenient method for describing glycosidase subsites. The subsite immediately towards the reducing end of the substrate of the scissile bond is labeled sequentially starting from +1, +2, +3 to +n. Towards the nonreducing end of the substrate, the numbering continues in the direction - 1 , -2, -3 to -n away from the hydrolyzed glycosidic linkage. The scissile bond is between subsites -1 and +1, as shown in Figure 1.3. For glycosidases that cleave disaccharides, there are only two subsites (-1 and +1) as shown in Figure 1.3(a). As aforementioned, the aglycone does not necessarily have to be a sugar moiety. These glycosidases are still considered to have two subsites with the scissile bond between -1. and +1, where +1 does not bind to a sugar unit. The nomenclature is most applicable to glycosidases that hydrolyze polysaccharides where there are multiple subsites, as shown in Figure 1.3(b) for the hen egg-white lysozyme. ' b Although glycosidases from Family 39, which includes both oc-L-iduronidases and (i-D-xylosidases,17 also appear to be an exception to this rule, the anomeric configurations of the substrates are the same and the presence of a - and p- glycosidases in the same family arises due to the conventions of nomenclature of D-and L-sugars. Chapter 1 - Introduction 6 - 4 - 3 - 2 -1 +1 +2 non-reducing end reducing end Figure 1.3. Illustration of the nomenclature used in describing sugar-binding subsites in glycosidases. (a) Subsite nomenclature for glycosidases that cleave disaccharides or non-specific monoglycosidases; (b) Subsite nomenclature proposed tor hen egg-white lysozyme. 2 0' 2 1 Section 1 .2 .1 .2. Glycosidase Mechanisms. In 1953, Koshland first proposed two mechanisms of enzymatic glycoside hydrolysis based on the stereochemical outcome of the reaction catalyzed - a direct displacement mechanism for the inverting glycosidases and a double displacement mechanism for the retaining glycosidases.22 The proposed mechanisms have been vindicated by numerous thorough mechanistic and structural studies.23"27 Thus, there are two general and now widely accepted mechanisms that glycosidases utilize for catalysis, both involving oxocarbenium ion-like transition states.24"27 Both subgroups of enzymes generally have two catalytic carboxylic acid functionalities (either glutamate or aspartate residues) at their enzyme active sites. Section 1 . 2 . 1 . 2 . 1 . Inverting Glycosidases. The inverting glycosidases undergo a direct displacement mechanism involving two carboxylic acid residues that are separated by 6-12 A as shown in Figure 1.4. One of the catalytic residues acts as a general base, activating a water molecule bound in the Chapter 1 - Introduction 7 active site for direct displacement of the leaving group, which is activated by protonation of the glycosidic oxygen by the other carboxylic acid residue. The catalytic mechanism goes through an oxocarbenium ion-like transition state to give the resulting hemiacetal with an anomeric configuration different from that of the substrate. Figure 1 . 4 . General mechanism utilized by inverting glycosidases. Section 1.2.1.2.2. Retaining Glycosidases. Conversely, the retaining glycosidases utilize a double-displacement mechanism involving a covalent glycosyl-enzyme intermediate (Figure 1.5). One carboxylic acid group acts as the catalytic nucleophile, and the other carboxylic acid, located approximately 5 A away, act as the general acid/base catalyst. Proceeding through an oxocarbenium ion-like transition state, the general acid/base carboxyl group promotes the formation of the covalent glycosyl-enzyme intermediate with the nucleophile. Subsequently, the glycosyl-enzyme intermediate undergoes hydrolysis via base-assisted catalysis by the same acid/base amino acid residue. Chapter 1 - Introduction 8 Figure 1.5. General mechanism utilized by retaining glycosidases. Section 1.2.1.2.3. Mechanistic Anomalies. As expected, not all glycosidases conform exactly to these two mechanisms. However, there only seem to be slight variations upon the same theme. For example, some ./V-acetyl-p-hexosaminidases belonging to Families 18 and 20 as well as hyaluronidases in Family 56 utilize the oxygen atom on the jV-acetyl group of the substrate as an intramolecular nucleophile in favor of the carboxylate amino acid residue (Figure 1.6).2X"30 The mechanism therefore involves an oxazolinium ion intermediate. Figure 1.6. General mechanism utilized by a retaining yV-acetyl-p-hexosaminidase. Furthermore, a tyrosine residue, as opposed to an aspartate or a glutamate, has been implicated as the catalytic nucleophile in the Trypanosoma cruzi /ram-sialidase as well as in sialidases from Trypanosoma rangeli and Mlcromonospora viridifaciensM~33 Moreover, in the Family 1 myrosinase from Sinapis alba, the catalytic acid/base residue Chapter I - Introduction 9 does not contain a carboxylic acid functionality, but it is instead replaced by a glutamine residue.34"36 An exogenous ascorbic acid functions as the catalytic base and activates a water molecule for the nucleophilic attack at the anomeric center of the glycosyl-enzyme intermediate (Figure 1.7).34-36 Common to all glycosidases described thus far is that they proceed through oxocarbenium ion-like transition states, as shown in Figure 1.8. Currently, only Family 4 glycosidases, to be introduced in Section 1.2.2, display some unusual properties that suggest this family of glycosidases may utilize a mechanism different from those of the "classical" glycosidases. sinigrin: R = ally!; sinalbin: R = / > h y d r o x y l b e n z y l Figure 1.7. The GHI myrosinase mechanism in which L-ascorbate replaces a carboxylate amino acid residue as the general base catalyst/4"'6 Chapter 1 - Introduction 10 inverting g lycos idase retaining g lycos idase retaining A/-acetyl-p-hexosamin idase Figure 1 . 8 . Comparison of the oxocarbenium ion-like transition states of glycosidases. (a) Inverting glycosidases; (b) Retaining glycosidases; (c) Retaining 7V-acetyl-(i-hexosaminidases. Section 1 .2.2. Family 4 Glycosidases. Section 1.2.2.1. An Eccentric Family. As alluded to earlier, enzymes within a family hydrolyze only glycosides of the same anomeric configuration and do so with the same stereochemical outcome. The sole exception to this rule is glycoside hydrolase Family 4 (GH4), with some members specific for a-D-glycosides and some for p-D-glycosides. Moreover, analogous to the Family 1 glycosidases,15,37 some Family 4 members prefer phosphorylated substrates, while others hydrolyze non-phosphorylated substrates.15,37"39 A l l members of this family to date are from bacterial sources. This family comprises a diversity of enzymes, including 6-phospho-a-glucosidases (-10 % of the Family 4 sequences deposited in C A Z Y ) , 3 9 " 4 3 a-glucosidases (-26 %) , 4 4 , 4 5 a-galactosidases ( - 3 %) , 4 6 , 4 7 and 6-phospho-p-glucosidase ( - 3 5 %). 4 X GH4 enzymes are also unique in their requirement for both a divalent metal, such as M n 2 + or N i 2 + , as well as N A D + for activity. 3 8 , 3 9 , 4 1" 5 3 In many cases, reducing conditions are required as we l l . 3 K , 4 0 , 4 4 ' 4 : , , 4 7 At the onset of the current thesis research project, a number of studies had been reported on several GH4 members. 3 8 ' 3 9 , 4 1" 5 5 The early studies include those of 6-phospho-a-glucosidases (Fusobacterium mortiferum M a l H , 4 1 , 4 2 , 5 0 Klebsiella pneumoniae A g l B , 4 3 ' 4 9 Bacillus subtilis G l v A 3 9 , 3 3 ) , an a-glucuronidase (Thermotoga maritima Agu4A 5 2 ) , an a-glucosidase (Thermotoga maritima A l g A 3 8 , 4 5 ) , and an a-galactosidase (Escherichia coli M e l A 4 6 ' 4 7 , 5 5 ) . In 1999, the first p-member of GH4 (CelF, a 6-phospho-p-glucosidase from Escherichia coli) was also reported.48 Because very little was known about GH4, Chapter 1 - Introduction most early studies focused on optimizing purification procedures and establishing the substrate specificity, cofactor requirements and physicochemical properties. Several remarkable findings were reported. Curiously, Agu4A 5 2 is the only GH4 member reported thus far that does not require added N A D + for enzyme activation. However, the dinucleotide cofactor may still be necessary for activity, because the presence of a very tightly bound N A D + cannot be ruled out. MalH, 5 0 the 6-phospho-a-glucosidase from Fusobacterium mortiferum, is also unusual in that the enzyme hydrolyzes both a- and p-anomers of 4-nitrophenyl 6-phospho-glucosides, although p-glucosidic linkages are not cleaved if the aglycone is not an activated leaving group such as 4-nitrophenol. A requirement for metal ions and reducing conditions is not that unusual for glycosidases. For example, many amylases have been reported to show a dependence on Ca 2 + , which is mainly required for proper protein folding/ 6 , 5 7 and the E. coli (3-galactosidase has been reported to require a M g 2 + for proper arrangement of the active site glutamate.38"61 However, the a-glucosidases, Thermococcus AN1 enzyme and some Bacillus enzymes are inhibited by M n 2 + and Co 2 + . 6 2 " 6 4 In GH4, the M n 2 + ion has been suggested to promote tetramerization to the active enzyme conformation in GlvA,' but the metal has no effect on the oligomerization of A l g A . 3 8 Therefore, the exact role of M n 2 + in GH4 enzymes has not been identified. The requirement for N A D and the accommodation of both a- and p-glycoside cleaving ability within the same family are unprecedented and suggest a catalytic mechanism significantly different from those used by all other characterized glycoside hydrolases. N A D + is most commonly recognized for its role in redox chemistry, but no conversion of the cofactor to its reduced N A D H form has been detected in any GH4 member and only catalytic amounts are required. 3 9 ' 4 3 ' 4 8 While the cofactor requirements may well be necessary for maintaining the structural integrity of these enzymes, the astonishingly loose substrate specificity baffles any comparisons with the two "classical" glycosidase mechanisms. No mechanistic or structural information had been reported at the start of the current thesis project. Since that time, the x-ray crystal structures of oc-glucosidase AglA from Thermotoga maritima (PDB l O B B ) , 4 4 a-glucuronidase Agu4B from Thermotoga maritima (PDB 1VJT), and 6-phospho-P-glucosidase from Geobacillus stearothermophilus (PDB 1S6Y) were solved. Some kinetic analyses and mutagenesis Chapter 1 - Introduction 12 studies have also been reported,40'65 but to date no mechanistic proposal has been put forth except from the work presented herein.66"70 Therefore, the remainder of the thesis is written in accordance with the course of our discoveries, taking the reader through the same journey that led to the proposed novel mechanism of enzymatic glycoside hydrolysis. Section 1.2.2.2. General Mechanistic Considerations. The specific roles of the divalent metal and of the dinucleotide are unclear. The stringent requirement for the N A D + cofactor for activity is suggestive of redox chemistry, and this should be taken into consideration in the mechanistic analysis even though conversion of N A D + to N A D H has not been detected.39'45'48 Preliminary data had indicated that the substrate C2 proton is exchanged with solvent deuterium when the reaction is carried out in D 2 0 buffer.0 Neither of the classical glycosidase mechanisms involves cleavage of the C2-H2 linkage, which is suggested by the solvent deuterium incorporation. Since the adjacent C l - O l is also cleaved, an elimination step across C l and C2 is likely to be a part of the overall mechanism. However, the C2 proton of glycosides is unactivated and has a very high pA' a. 7 1 Cleavage of the C2-H2 bond may therefore require some activation. Considering that the N A D cofactor is required for activity and that the cofactor is not consumed during catalysis, it is possible that the N A D + is only transiently reduced to effect the oxidation of the C3 hydroxyl group. The formation of a carbonyl functionality at C3 would lower the pA'a of the C2 proton, activating it for deprotonation. This strategy of transient oxidation/reduction is utilized 72 74 by a number enzymes found in nature. Such enzymes include epimerases, decarboxylases,75 and dehydratases.76"78 The likelihood of an elimination step by GH4 can be considered upon examination of two other groups of carbohydrate-degrading enzymes. c The incorporation of solvent deuterium was first determined by Dr. Bernard Henrissat and brought to our attention through a personal communication. Initial results with an uncharacterized g h 4 6-phospho-p-glucosidase BGIu from Bacillus subtilis were reported in the M.Sc. thesis of a former UBC colleague F. L i (Withers laboratory). Chapter I - Introduction 13 Section 1.2.3. Other Carbohydrate-Degrading Enzymes Utilizing Elimination Mechanisms. Although the most common method of degradation of carbohydrates in nature involves hydrolysis, a number of enzymes have evolved to use elimination mechanisms involving either cationic or anionic transition states. These enzymes include polysaccharide lyases (E.C. 4.2.2.-) and oc-glucan lyases (E.C. 4.2.2.13). Section 1.2.3.1. Polysaccharide Lyases. The polysaccharide lyases (E.C. 4.2.2.-) are involved in the degradation of glycosaminoglycans and pectin. These enzymes have been known for many years and sequences for many such enzymes have been determined. Indeed, the C A Z Y website also maintains an up-to-date inventory of these lyases, classifying them into different families based on primary sequence similiarity.79 Since the early 1960s, it has been recognized that, even though both the bacterial and testicular hyaluronidases degrade hyaluronic acid, these two "hyaluronidases" are mechanistically distinct.80"83 The testicular hyaluronidase behaves as a typical glycosidase, hydrolyzing the substrate to give two saturated saccharides as the products and incorporating an l 8 0 label into C l ' of 1 8 the uronic acid-containing polymer when the enzymatic reaction is carried out in H2 O. 8 1 , 8 4 This technique has been used for identifying the scissile bond for other glycosidases.22 Meanwhile, the products of the bacterial hyaluronidase reaction were a saturated reducing sugar and an a, p-unsaturated carboxylic acid-containing sugar.83 Furthermore, when the reaction was carried out in H2 I 8 0, no 1 8 0 atoms were incorporated into either of the sugar products. The bond cleaved therefore is that between C4 and the glycosidic oxygen rather than that between C l ' and the glycosidic oxygen, with reaction involving an elimination process. As such, it is evident that these bacterial "hyaluronidases" are markedly distinct from glycosidases and have, in fact, eventually been given their own classification, along with enzymes involved in polyuronic acid polymer degradation, as the polysaccharide lyases. A detailed mechanism was proposed by Gacesa in 1978, which involves neutralization of the substrate C5 carboxylate functionality, proton abstraction at C5, and p-elimination of the 4-O-glycosidic bond. 8 5 The mechanism is now largely accepted. A detailed mechanistic investigation on Chapter 1 - Introduction 14 chondroitin A C lyase from Flavobacterium heparinum has shown that the enzyme utilizes a stepwise E l C b mechanism, as shown in Figure 1.9.86 An E l C b mechanism was also suggested based upon structural analyses for pectate lyases Pel9A from Erwinia chrysanthemi1 and Pell OA from CeUvibrio japonicus** In both cases, the absence of a candidate general acid in the vicinity of the glycosidic oxygen was presented as evidence for an E l C b mechanism on the basis that cleavage could only occur in such a case if C5 deprotonation precedes C4-04 cleavage. BH + Figure 1.9. Proposed elimination mechanism of chondroitin A C lyase. Section 1.2.3.2. a-Glucan Lyase. Similarly intriguing are the recently discovered a-l,4-glucan lyases, which are assigned by primary sequence similarity into GH31, a retaining oc-glycosidase family. 7 9 ' 8 9 While the GH31 a-glycosidases utilize the "classical" double displacement mechanism,9 0 , 9 1 a-l,4-glucan lyases cleave glycosidic linkages in starch and glycogen via an elimination mechanism (Figure 1.10). Additionally, the reaction carried out by a-1,4-glucan lyases is clearly unlike that of other polysaccharide lyases as is evident by the differences between the substrates and the reaction products.92"95 Chapter I - Introduction 15 Figure 1.10. Proposed mechanism of the Gracilariopsis a-l ,4-glucan lyase from GH31 y 4 The substrates for the currently known 13 families of polysaccharide lyases are acidic polysaccharides, and abstraction of the C5 proton is only possible due to the adjacent carboxylic acid functionality inherent in the substrate.85 Upon cleavage, a double bond between C4 and C5 is formed on the resulting nonreducing end sugar.86 On the other hand, no protons are activated for abstraction in the substrates for a-l,4-glucan lyase, since the natural substrates are starch and glycogen, which contain no carboxylic acids.~"~ 9~ Furthermore, the a-l,4-glucan lyase reaction produces 1,5-anhydrofructose, which in its enol form contains a double bond between C l ' and C2' . 9 2 " 9 5 It is worth noting that one of the major accomplishments of the glycosidase classification system developed by Henrissat is that primary sequence identity has shown itself to be a remarkably useful tool in the prediction of tertiary structure, and catalytic mechanism and machinery.14 Therefore, based on the high primary sequence identity to Family 31 retaining a-glucosidases,95,96 the mechanism of cx-l,4-glucan lyases is expected to share similarities with that of the "classical" retaining a-glucosidases. This presumption is supported by various studies, including inhibition experiments with acarbose,94'96 1-deoxynojirimycin,9 4'9 6 carbodiimides,97 and 5-fluoro-p-L-idopyranosyl fluoride,93 all of which are common inhibitors of retaining a-glucosidases.98'102 Inhibition by Chapter 1 - Introduction 16 carbodiimides, which selectively react with aspartic and glutamic acids, suggests that carboxylic acid groups are an important part of the catalytic machinery, similar to other glycosidases.97 Meanwhile, 1-deoxynojirimycin and acarbose are known to inhibit a-glucosidases by mimicking the oxocarbenium ion-like transition state structure.98"100 Inactivation by the mechanism-based inactivator, 5-fIuoro-fj-L-idopyranosyl fluoride, provided evidence for formation of a covalent glycosyl-enzyme intermediate, thereby allowing the identification of Asp553 as the catalytic nucleophile.9 3'9 4 This aspartate residue is strictly conserved within GH31 and its equivalent was also identified as the catalytic nucleophile in two GH31 a-glycosidases (YicI 9 1 from Escherichia coli and A g l A 9 0 from Aspergillus niger) using the analogous labeling experiment. It was shown that following the formation of the glycosyl-enzyme intermediate, the second step proceeds via an E2 elimination mechanism with substantial E l character.94 Thus, the C2 proton becomes acidic at the oxocarbenium ion-like transition state, thereby facilitating elimination and generation of the 1,5-anhydrofructose product.9 3'9 4 The role of general base in this ^.'^-elimination mechanism may well be played by the departing, anionic nucleophile as shown in Figure 1.10. The structure of the glycosyl-enzyme complex of the GH31 a-xylosidase YicI supports such a dual role for the Asp553 nucleophile.91 In contrast to what is seen with GH13 1 0 3 and GH38 1 0 4 a-glycosidases, the carbonyl oxygen of the nucleophile is found in close proximity to the substrate C2 proton rather than the endocyclic oxygen.91 Interestingly, retaining glycosidases have long been known to catalyze the hydration of glycal substrates via a covalent glycosyl-enzyme intermediate through the .syn-addition of the C2 proton and the enzymatic nucleophile (Figure 1.11) in a somewhat concerted process that is essentially the reverse of the process proposed for a-l,4-glucan lyases. 6 1 ' 9 4 ' 1 0 5" 1 0 7 Chapter 1 - Introduction 17 Figure 1.11. Proposed mechanism for an oc-glucosidase-catalyzed hydration of D-glucal.' 0 6 , 1 0 7 The discovery of the oc-l,4-glucan lyase is one of the first indications that lyases are in some way linked to glycosidases. As dictated by the International Union of Biochemistry and Molecular Biology (IUBMB), this enzyme is designated as a lyase based on the reaction catalyzed, and the enzyme is given the E.C. number E.C. 4.2.2.13. 1 0 8 Conversely, the classification system of glycosidases and lyases based upon sequence alignment, which has been shown to be a useful tool in structural and mechanistic elucidation, groups the a-l,4-glucan lyases under Family 31 of the glycosidases.14 Indeed, the oc-l,4-glucan lyase mechanism shares similarities, namely the formation of the covalent glycosyl-enzyme intermediate, with the retaining a-glucosidases. Beyond the sequence similarities, no observations can be made concerning structural resemblances, since no structural data are currently available for the Ot-1,4-glucan lyases. However, it is expected that they will only differ in quite subtle ways. It is fascinating to consider the evolution of the a-l,4-glucan lyases and the retaining a-glucosidases and to realize that the functional similarities between these two different Chapter I - Introduction 18 groups of enzymes may have resulted from divergent evolution from a common ancestor.95'96 Further evidence of the connection between these two seemingly different classes of enzymes is that 1,5-anhydrofructose was reported as a side product from maltose hydrolysis by the Family 31 mammalian glucosidase II, suggesting the presence of minor lyase activity in this glycosidase.109 Likewise, the nucleoside 2-deoxyribosyltransferase from Lactobacillus leishmanii was shown to also produce D-ribal, in addition to its normal transfer product.110 Moreover, the crystal structure of a recently discovered Family 82 t-carrageenase was found to have a rare protein fold." 1 The inverting glycosidase contains a right-handed parallel P-helix, which is known in some Family 1 polysaccharide lyases but is completely novel to the glycosidases."1'"2 These findings highlight the fact that assignment of function to enzyme based solely upon sequence analysis can be hazardous. Confirmation of function requires demonstration of the actual reaction catalyzed using purified proteins. These examples also demonstrate that lyase activity is sometimes found in glycosidases. Therefore, the proposed elimination step for GH4 enzymes is not a farfetched proposition, but should be considered a distinct possibility based on the solvent isotopic exchange observed. Section 1 .3. NAD+-Dependent Enzymes. Given an apparent role for N A D + in GH4 enzymes, it is useful to review other enzyme systems that also utilize a N A D + / N A D H cofactor. N A D + (P-nicotinamide adenine dinucleotide) is one of a number of common cofactors that serve as electron carriers for enzymes that catalyze redox reactions. The nicotinamide ring of NAD7NADH is the redox active moiety of the cofactor. In its oxidized N A D + form, the nicotinamide ring is in the pyridinium salt form. The C4 position of the nicotinamide ring can reversibly accept hydrides and thereby convert to its reduced N A D H form with the nicotinamide ring converted to a 1,4-dihydro-pyridine (Figure 1.12). The C4 position of the nicotinamide ring is prochiral, and the C4 protons of N A D H can be labeled as either pro-R or pro-S. Thus, the hydride transfer is stereospecific to or from either the re-ox si- face of the nicotinamide ring. N A D P + (P-Nicotinamide adenine dinucleotide phosphate) is the phosphorylated counterpart that can also mediate redox reactions via the nicotinamide ring. Chapter 1 - Introduction 19 n i co t i nam ide ring + H + + 2e" OR OH OH OH pro-S pro-R NH 2 H H , e e 0 O 1 I I I N' NH, OR OH OH OH NAD + : R = H oxidized form NADH: R = H NADPH: R = P 0 3 2 " reduced form Figure 1.12. Structures of the dinucleotide cofactors in their oxidized (NAD + or NADP + ) and reduced (NADH or NADPH) forms. The dinucleotide cofactor is most often used as an electron carrier effecting redox chemistry in enzymatic systems. Therefore, it is imperative that other enzyme mechanisms involving NAD+-mediated redox steps are surveyed. The following sections provide a brief list of such enzymes. Section 1.3.1. Dehydrogenases. The simplest examples are the a-hydroxy organic acid dehydrogenases, such as malate or lactate dehydrogenases. Dehydrogenases are known to utilize N A D + / N A D H cofactors for reversible redox chemistry. For example, in L-lactate dehydrogenase, the N A D + cofactor is utilized in the interconversion of lactate and pyruvate.113 While a base deprotonates the lactate hydroxyl, the C4 center of the nicotinamide ring removes a hydride from the substrate C2 converting the substrate to pyruvate (Figure 1.13). In the reverse direction, the hydride is transferred from N A D H to C2 of the substrate and the carbonyl is reprotonated. The proposed mechanism of malate dehydrogenase is very similar to that of L-lactate dehydrogenase, and the enzyme catalyzes the interconversion of malate and oxaloacetate (Figure 1.14).113 Chapter I - Introduction 20 lactate pyruvate O H O H O H O H Figure 1.13. Proposed mechanism of L-lactate dehydrogenase. N A D + NADH mal ate oxaloacetate Figure 1.14. Malate dehydrogenase reaction. Section 1.3.2. Epimerases. As mentioned in Section 1.1, carbohydrates are ideal molecules for mediating signal recognition pathways due to the large number of isomeric forms arising from potential stereoisomers, side branching and different ring sizes. In order to expand the number of sugar stereoisomers available, nature has evolved epimerases that can invert the stereochemistry of one or more of the many chiral centers in carbohydrate molecules to produce large numbers of stereoisomers. Epimerization of stereocenters typically, though not necessarily,114 involves deprotonation/reprotonation strategies. In a number Chapter 1 - Introduction 21 of cases, deprotonation is possible due to "activation" of the stereocenter by an adjacent carbonyl, carboxylic acid or ester functionality. The activation is sometimes inherent to the substrate as in the polysaccharide lyases.85 However, many carbohydrates do not contain any functional groups that could "activate" the stereocenter. A number of epimerases overcomes this hurdle by employing a N A D + cofactor in transient redox chemistry to carry out the inversion of stereochemistry via several different strategies. An excellent example in which the site of inversion is oxidized is found in the human UDP-galactose 4-epimerase (Figure 1.15).7 2 , 7 3 , 1 l v ' 2 0 Tyrl57 acts as a general base and deprotonates the C4 hydroxyl and the N A D + cofactor removes a hydride from C4. Subsequently, the UDP-4-keto intermediate rotates in the enzyme active site. The N A D H then transfers the hydride to the opposite face of the sugar ring, and Tyrl57 reprotonates the C4 oxygen, leading to epimerization of the C4 center. The N A D + cofactor is not consumed, but is only transiently reduced to NADH. Tyrl57 Tyrl57 H OUDP NAD—H OUDP UDP-D-galactose U DP-4-keto-D-glucose Tyrl57 Tyrl57 OUDP UDP-D-glucose UDP-4-keto-D-glucose Figure 1.15. Proposed mechanism of the human UDP-galactose 4-epimerase. I 15 Chapter I - Introduction 22 In the case of GDP-D-mannose:GDP-L-galactose epimerase, the enzyme utilizes the N A D + cofactor to transiently oxidize the hydroxyl group adjacent to the sites of inversion (Figure 1.16). 7 4' 1 2 1' 1 2 2 Hydride transfer occurs from C4 of the sugar ring to the N A D + cofactor.122 The formation of a 4-keto intermediate lowers the pKa values of protons that are on the carbons a to the carbonyl. In this case, deprotonation of C5 can occur, and reprotonation from the opposite side of the sugar ring would result in epimerization of the C5 center. Similarly, deprotonation of C3 is favorable, because the anion formed is stabilized by the adjacent C4 carbonyl functionality. A 4-keto intermediate has been trapped in the active site of a mutant GDP-D-mannose:GDP-L-galactose epimerase providing support for this proposed mechanism. " Upon inversion of the stereocenters at C3 and C5, the carbohydrate ring flips, and N A D H returns the hydride to the C4 position to reduce the carbonyl and to generate GDP-L-galactose. N A D + N A D - z H HO T - - - c r 7 ^ 0 G D P O .^ \ 0 l t e O H OH\>H OGDP :B V I GDP -L -ga lactose Figure 1.16. Proposed mechanism of GDP-D-mannose:GDP-L-galactose epimerase. 7 4 - 1 2 1 - 1 2 2 Section 1.3.3. Dehydratases. Dehydratases catalyze dehydration reactions: the removal of water. Some dehydratases require a N A D + cofactor for activity. An interesting example is found in the Salmonella enterica serovar typhimitrium dTDP-D-glucose 4,6-dehydratase Chapter I - Introduction 23 RmlB. ' The mechanism utilized by RmlB (Figure 1.17) is in fact very similar to that of GDP-D-mannose:GDP1ir-galactose epimerase. The N A D + abstracts a hydride from C4 of the substrate sugar ring, which lowers the pKa of the C5 proton. A glutamate base removes the C5 proton, and elimination across C5 and C6 occurs. A water molecule is released and a 4-keto-5,6-glucosene intermediate is formed. The N A D H cofactor donates a hydride to C6 of the substrate and C5 is reprotonated to yield dTDP-4-keto-6-deoxy-D-glucose. The mechanism of CDP-D-glucose 4,6-dehydratase76 is essentially the same as that of RmlB. The enzyme utilizes an N A D + cofactor to oxidize C4 of CDP-D-glucose to form a 4-keto intermediate. Subsequently, general base catalysis removes the C5 proton and elimination across C5 and C6 occurs to release a water molecule. Finally, the on-board N A D H donates a hydride to C6 and reduces the unsaturated bond between C5 and C6. Similar to the epimerases discussed in Section 1.3.2, the dinucleotide cofactors in RmlB and CDP-D-glucose 4,6-dehydratase are not consumed by the enzymatic reaction. The cofactor is only involved in transient redox chemistry and the same redox form (NAD + ) is ultimately regenerated at the end of the reaction cycle. Tyrl67| Figure 1.17. Proposed mechanism of RmlB from Salmonella enterica serovar typhimurium. Chapter 1 - Introduction 24 Section 1.4. Proposed GH4 Mechanisms. For glycoside hydrolase Family 4, preliminary data had shown that the C2 proton undergoes solvent deuterium incorporation and indicated that elimination across C2 and C l is likely involved. However, due to the high pKa value of the C2 proton,71 any proposed mechanism would need to suggest some form of activation for this proton. In light of the mechanisms of polysaccharide lyases, a-l,4-glucan lyase, a-hydroxy organic acid dehydrogenases, epimerases and dehydratases, several potential mechanisms are proposed for GH4. Section 1.4.1. Proposed Mechanism #1. GH4 enzymes could utilize a mechanism very similar to that used by a-1,4-glucan lyase. The first step could be equivalent to that of retaining glycosidases, where an active site amino acid residue acts as the catalytic nucleophile and attacks the anomeric center. The C l - O l linkage is thus cleaved with general acid catalysis to the departing oxygen (Figure 1.18), leading to the formation of a glycosyl-enzyme intermediate. As was discussed for a-l,4-glucan lyase, the pAfa of the C2 proton could be lowered at the oxocarbenium ion-like transition state and an appropriate base residue could deprotonate C2, leading to elimination of the C l - O l glycosyl-enzyme bond. Addition of water across the double bond between C l and C2 would result in an overall hydrolysis product. The proposed mechanism depicted in Figure 1.18 is favored because such a mechanism is closely related to those of "classical" retaining glycosidases and involves nucleophilic displacement steps. Furthermore, this mechanism is also similar to that of a-l,4-glucan lyase in which an elimination step occurs in the glycosyl-enzyme intermediate. This mechanism could account for solvent isotope exchange at C2, but it does not include a role for N A D + . Furthermore, the roles of the catalytic nucleophile and general acid^ase need to be reversed to hydrolyze substrates of the opposite anomeric configurations. Since there is no case in which both a- and p-glycosidases are not found in single "classical" glycoside hydrolase families, this role reversal for the two catalytic residues seems highly unlikely. Chapter 1 - Introduction 25 0 : B H - O . N a 2 0 3 P O - ^ j H O - X - - ^ 0 ^ - 0 - R OH ~ BH Na 2 0 3 P0-HO' H O - X - - ^ ^ - O H OH e o ^ o e e glycosylation ROH 1,2-addition r H o \ - - - - , V * \ e :B N a 2 0 3 P 0 ~ A H H O 0 OH elimination BH 0 ^ 0 N a 2 0 3 P O - A H O ' H0-HO \ cr i H e °Y0 Figure 1.18. Proposed mechanism #1 for G H 4 . Section 1.4.2. Proposed Mechanism #2. A second possibility (Figure 1.19) involves an E l reaction that is a variation of the standard nucleophilic displacement mechanism. A single catalytic residue could act as both as the general acid and general base. As part of the first step, it would provide general acid catalysis to the leaving group oxygen. If the catalytic residue is a carboxylic acid (as in the case of most other "classical" glycosidases"), the second oxygen on the carboxylic acid would now be activated for base catalysis and could deprotonate C 2 . Provided the carboxylic acid is correctly positioned, this residue could play the dual role of general acid and general base catalyst. A dual role has been suggested for the catalytic nucleophile in the .^-elimination in the a-l,4-glucan lyase 9 1 ' 9 4 and for the hydration of glycals by retaining glycosidases.6 1 , 1 0 5"1 0 7 Indeed, based on the stereochemistry of C 2 protonation, the carbonyl oxygen of the nucleophile is found directly above C 2 of the substrate in the G H 3 1 a-xylosidase YicI . 9 1 Therefore, a dual role for the catalytic residue in this case is a realistic proposition. Subsequently, the roles are reversed, the glycal intermediate is reprotonated and the carbonyl oxygen activates a water molecule, which Chapter 1 - Introduction 26 adds to C l . The overall result is hydrolysis of the glycosidic linkage. It should be kept in mind that a general acid/base residue other than Asp or Glu could function in this dual role given that the reaction would proceed via a more favorable 6-membered transition state rather than the 8-membered transition state required if the catalytic residue is an aspartate or a glutamate. This mechanism is only directly applicable to the p-glycosidases in GH4, because the .vyrc-elimination requires the one carboxylic acid group to play the dual role of general acid and general base. However, the general mechanism could presumably be applicable to oc-glycosidases provided that a general acid catalyst is present on the opposite face of the sugar ring. However, this proposed mechanism would require a very specific arrangement of amino acid residues about the substrate anomeric center. Therefore, unlike classical glycosidases grouped into the same family, the active site of a- and p-glycosidases in GH4 might be quite different from one another. Whether this disparity holds true for GH4 remains to be examined, but it should become readily apparent as x-ray crystallographic data becomes available for this family of enzymes. Finally, this proposed mechanism has another unsettling element. While proposed mechanism #2 could satisfactorily account for the solvent isotope incorporation at C2, it does not include a role for the N A D + cofactor. HO ,B 1 H (b) NajOjPO-^ H - V H syn-elimination Na 2 0 3 PO-^ ( ^ - H 1,2-addition ^ N a 2 0 3 PO. HO-V-A-CHI \ HO H O V ^ V ^ O - p V HO-V^ " H O - Y — - ^ - O H OH ROH HO Figure 1.19. Proposed mechanism #2 for GH4. (a) Catalytic residue is Asp or Glu; (b) Catalytic residue is a general acid/base residue. Chapter I - Introduction 27 Section 1.4.3. Proposed Mechanism #3. The third potential mechanism (Figure 1.20) is most favored, because it involves the participation of the N A D + cofactor in redox steps and accounts for the solvent isotopic exchange. The proposed mechanism involves as its first step NAD + -mediated oxidation of the C3 hydroxyl to a carbonyl. This reduces the pKa of the C2 proton, which allows deprotonation at C2 and elimination across the C2-C1 bond leading to departure of the aglycone. A water molecule then attacks the oc,p-unsaturated intermediate in a 1,4-Michael-like addition, reprotonating C2. The on-board N A D H cofactor reduces the C3 carbonyl to yield the overall hydrolyzed product. Exchange of the C2 proton is fully consistent with activation by the adjacent ketone at C3. It is interesting to note that the first steps of proposed mechanism #3 and the RmlB mechanism (Section 1.3.3) are the same. The main difference is in the oxidation of the on-board N A D H . In the RmlB mechanism,7 7'7 8 the hydride is transferred from N A D H to C6 to reduce the double bond. Meanwhile, we propose that GH4 utilizes the on-board NADH cofactor to reduce the C3 carbonyl back to its original hydroxyl form, and the dinucleotide cofactor is returned to its original redox state. Proposed mechanism #3 is most attractive because it accounts for the many extraordinary observations of GH4. During the deprotonation of C2, solvent deuterium incorporation can readily occur after C2 proton abstraction, while the formation of the N A D H may not have been detected3 9'4 5'4 8 due to the transient nature of the intermediate species and its formation in only catalytic amounts. Furthermore, this mechanism satisfies the varied substrate anomeric stereochemistry for GH4 enzymes. Because the fundamental steps in both of the classical glycosidase mechanisms involve nucleophilic displacement reactions at the anomeric center, these enzymes require a very specific arrangement of the active site amino acid residues around the substrate anomeric center. On the other hand, in the GH4 mechanism, the first two steps involving the N A D + -mediated oxidation at C3 and deprotonation at C2 could be the same for both a- and p-glycosidases. The only obligatory difference would be the position of the general acid catalyst that protonates O l , thereby assisting the departure of the aglycone. Conceivably, the critical steps in all GH4 enzymes would be the C3 oxidation and deprotonation at C2. Subsequently, either a trans-2,1 -elimination utilized by p-glycosidases or a cis-2,1-Chapter 1 - Introduction 28 elimination employed by a-glycosidases ensues readily upon the formation of enediolate intermediate. Thus, it is plausible that some members of GH4 have evolved to hydrolyze a-glycosidic linkages while others hydrolyze (3-glycosidic linkages by strategic placement of the general acid catalyst close to C l in either the a- or p-anomeric configuration. Conversely, depending on the leaving group ability of the aglycone, no general acid catalysis may be required at all. Thus, only the N A D + cofactor and the general catalytic base over C 2 need to be conserved in GH4 for catalysis, with general acid catalysis for the departure of the aglycone being only of secondary importance to this family of enzymes. This would be similar to the proposed E l C b mechanism utilized by polysaccharide lyases in which no general acid catalyst was identified.8 7'8 8 Figure 1.20. Proposed mechanism #3 for G H 4 . Chapter 1 - Introduction 29 This proposed redox-elimination-addition mechanism accounts for the solvent deuterium incorporation and can be readily adapted to both a- and P-glycosidases. In addition, it suggests a role for the essential N A D + cofactor. On the other hand, while both proposed mechanisms #1 and #2 would be consistent with the observation of solvent isotopic incorporation at C2, neither of the two mechanisms satisfactorily justifies the occurrence of both a- and P-glycosidases in GH4. Furthermore, neither proposed mechanism #1 or #2 includes a role for the N A D + cofactor. Section 1.5. Specific A i m s of Thesis. The direct and double displacement mechanisms for glycosidases were first introduced by Koshland,2 2 and have copious support from structural and mechanistic analyses on enzymes across the numerous glycosidase families. These studies were also able to reveal that there is a direct relationship between primary sequence, and enzyme structure and mechanism, which authenticates the value of the classification system developed by Henrissat. 1 4 ' ' 5 , 3 7 ' 7 9 The direct and double displacement mechanisms have served as canonical models for over 50 years. Not all glycoside hydrolases conform exactly to the two mechanisms. However, all mechanisms involve oxocarbenium ion-like transition states and nucleophilic displacement steps. The enzymes of GH4 may be the first glycosidases found to use a radically different mechanism of glycoside hydrolysis in over 50 years. The proposed research focuses on the mechanistic elucidation of Family 4 glycosidases. While all other families of glycosidases appear to follow the two general hydrolysis mechanisms or some slight variation thereof, GH4 enzymes display unique properties that suggest that this family may utilize a mechanism very different from the nucleophilic displacement mechanisms proposed by Koshland.2 2 The requirement for both a divalent metal ion and N A D + cofactor has not been reported for any other glycoside hydrolase family. Furthermore, GH4 is the only family to include both a- and P-glycosidases. The goal of the current study is a comprehensive investigation of the GH4 mechanism through kinetic isotope effects, linear free energy relationships, inhibition and mutagenesis studies coupled with structural analysis. These studies are aimed at Chapter 1 - Introduction 30 providing an understanding of the bond breaking and forming steps, and the identification of catalytic residues is attempted. The role of the N A D + cofactor is also explored. A comprehensive mechanistic analysis of GH4 should attempt to rationalize the unusual features of facile hydrolysis of a- and p-glycosides by enzymes in the same glycoside hydrolase family. Therefore, two representative GH4 enzymes were selected for the current mechanistic investigation: (1) BglT, a 6-phospho-p-glucosidase from Thermotoga maritima; and (2) GlvA, a 6-phospho-a-glucosidase from Bacillus subtilis. These choices were based, in part, on ongoing work by our collaborators (x-ray crystallographers Dr. Annabelle Varrot and Dr. Gideon Davies) aimed at structural analysis of Family 4 glycosidases. The work was also carried out in collaboration with Dr. John Thompson, whose group has reported the physicochemical properties of a number of GH4 enzymes, and in consultation with Dr. Shyamala Rajan and Dr. Wayne Anderson, who solved the GlvA structure at the same time. Chapter 2 — BglT from Thermotoga maritima 31 C h a p t e r 2 6 - P h o s p h o - p - g l u c o s i d a s e B g l T f r o m Thermotoga maritima Chapter 2 - BglT from Thermotoga maritima 32 S e c t i o n 2.1. I n t r o d u c t i o n a n d S p e c i f i c A i m s . To date, very little is known about the mechanism of GH4 enzymes. Although standard kinetic characterization using natural and synthetic substrates to obtain kinetic parameters (X'C A T and /CM) has been performed on a number of GH4 enzymes,38"55 no mechanistic information has been reported on any a- or p-glycosidase in this family. The current study aims to provide a comprehensive mechanistic examination of a representative p-glycosidase member of GH4. BglT (415 residues; 47627 Da), a thermostable 6-phospho-p-glucosidase from Thermotoga maritima, is an ideal candidate for our study. Our collaborators, Dr. Gideon Davies and Dr. Annabelle Varrot, solved the x-crystal structure of the enzyme. Contribution of the structural analysis with kinetic isotope effects, linear free energy relationships and mutagenesis studies provides a complete set of tools for accelerating our understanding of the proposed GH4 mechanism, which involves: 1) oxidation of the C3 hydroxyl; 2) C2 proton abstraction; 3) 2,1 -elimination of the aglycone moiety; 4) 1,4-Michael-like addition of a water molecule at C l ; 5) reprotonation of C2; and 6) reduction of the C3 carbonyl. It is important to note that the mechanistic proposal was formulated at the beginning of this thesis project based mainly on the atypical cofactor requirements and varied substrate specificity exhibited by GH4 enzymes. Since that time, the structures of several GH4 enzymes have been solved: a-glucosidase AglA (PDB 10BB), 4 4 oc-glucuronidase Agu4B (PDB 1VJT), 6-phospho-p-glucosidase (PDB 1S6Y), and 6-phospho-a-glucosidase GlvA (PDB 1U8X). Studies of some GH4 mutants have also been reported by other groups.40 Thus, some aspects of the work presented herein benefited from these accounts. In addition, the mechanistic investigation was concurrent with that of GlvA (Chapter 3). Therefore, comparison between these will be included here even though some of the data were not available at the onset of this study. Chapter 2 - BglT from Thermotoga maritima 33 Section 2.2. Kinetic Characterization. Section 2.2.1. Design of Substrates and Kinetic Assays. The mechanistic analysis of any enzyme depends heavily on a practical and reliable kinetic assay. The proposed natural substrate for BglT is cellobiose 6'-phosphate (C6'P or 2.2), which is not commercially available and the chemical synthesis from cellobiose (2.1) would be lengthy requiring a number of orthogonal protection strategies to leave the 6'-hydroxyl free for phosphorylation. Fortunately, our collaborator Dr. John Thompson generously donated the ATP-dependent p-glucoside kinase B g l K 1 2 3 d which regioselectively phosphorylates the C6'-position of 2.1 according to Scheme 2.1. 2.1 2.2 (C6'P) Scheme 2.1. Enzymatic synthesis of the natural substrate C6'P (2.1). (i) BglK, ATP. BglT catalyzes the hydrolysis of C6'P (2.2) to produce glucose 6-phosphate (G6P or 2.3) and glucose (Glc or 2.4). A convenient spectrophotometric coupled assay involving glucose 6-phosphate dehydrogenase (G6PDH) and N A D P + (as depicted in Figure 2.1) can be utilized to determine the kinetic parameters for the BglT-catalyzed hydrolysis of C6'P. The G6P liberated by the BglT reaction was converted to 6-phospho-D-glucono lactone (2.5) by added G6PDH, with concomitant reduction of N A D P + , which can be monitored at 340 nm (extinction coefficient, £NADPH = 6220 cm^M"1). Fortunately, there is no conflict in the use of N A D P + in this assay with the requirement of BglT with N A D + since BglT does not accept N A D P + or NADPH (see Section 2.2.2.1) and, in any case, both enzymes require the same cofactor redox form. In addition, measurement of only initial rate kinetics ensures no significant accumulation of the reduced form. d BglK. is a versatile P-glucoside kinase from Klebsiella pneumoniae and was used for the selective phosphorylation the C6 hydroxyl group of a variety of P-glucosides and p-cellobiosides used in this study. Chapter 2 - BglT from Thermotoga maritima 3 4 Figure 2.1. Schematic for the BglT-catalyzed hydrolysis of C6 ' P for the G 6 P D H coupled assay, (i) BglT; (ii) G 6 P D H , N A D P + . To examine the validity of this coupled assay, the kinetic parameters for the hydrolysis of a chromogenic disaccharide 4-nitrophenyl 6-phospho-fi-cellobioside (4NPpC6'P or 2.1.0) by BglT to produce G6P and 4-nitrophenyl P-D-glucopyranoside (4NPpGlc or 2.1.1) were measured using the G6PDH coupled assay as well as with a separate coupled assay involving an added p-glucosidase, Abg from Agrobacterium sp.124 4NPPC6'P was synthesized via the chemoenzymatic route shown in Scheme 2.2. Acetylation of cellobiose (2.1) in pyridine yielded 2.6, which was converted to its bromide derivative (2.7) and then to 2.8 via the Koenigs-Knorr reaction. " Zemplen deprotection126 of 2.8 using NaOMe/MeOH afforded 4-nitrophenyl p-cellobioside (2.9), which was phosphorylated by BglK to yield 4NPpC6'P, which was a suitable substrate for both coupled assays (Figure 2.2). Neither of the reaction products, G6P (2.3) nor 4NPPGlc (2.11), changes the absorbance reading at 340 nm, so there is no interference with detection of NADPH (A. l l i a x = 340 nm, 8NADPH - 6220 cm'M" 1 ) formation in the G6PDH coupled assay. In the Abg coupled assay, once 4NPPGlc is released from the BglT reaction it is hydrolyzed rapidly by Abg to produce Glc (2.4) and 4-nitrophenolate (2.12), and the reaction rate can be determined by monitoring the change in absorbance at 400 nm due to the release of the chromophore 2..12.124 Fortunately, Abg will not hydrolyze 4NPpC6'P (2.10), because it does not cleave 6'-phospho-sugars. Furthermore, Chapter 2 - BglT from Thermotoga maritima 35 the reaction product 4NPpGlc (2.11) is not consumed directly by BglT, because this enzyme is selective for phosphorylated substrates (see results in Section 2.2.5). Scheme 2.2. Chemoenzymatic synthesis of 4NP(JC6'P (2.10). (i) A c 2 0 , pyridine; (ii) 33% HBr in HOAc, A c 2 0 ; (iii) 4-nitrophenol, acetone, 1 M NaOH; (iv) NaOMe/MeOH; ( V ) BglK, ATP. N A D P H 2.4 (Glc) 2.12 2.5 400 nm 340 nm Figure 2.2. Schematic for the BglT-catalyzed hydrolysis of 4NPpC6'P (2.10) for the Abg and G6PDH coupled assays, (i) BglT; (ii) Abg; (iii) G6PDH, N A D P + . While these assays are useful for studying the "natural" substrate, an alternative set of substrates was required for mechanistic analysis in which leaving groups could be varied and isotopes incorporated. The use of synthetic chromogenic substrates in Chapter 2 - BglT from Thermotoga maritima 36 glycosidase activity assays has been well-established.124 Aryl glycosides are particularly popular because the rate of hydrolysis can be measured by directly monitoring the UV-vis absorbance for the release of the chromophore, the phenol leaving group. Furthermore, aromatic residues have been found in the -1 and +1 subsites for a number of glycosidases,127"130 where they form favorable hydrophobic Jt-stacking interactions with sugar subunits. 1 2 9 1 3 0 Aryl glycosides can therefore ideally mimic the same interaction in binding to the enzyme active site. The commercially available 4NPpGlc (2.11) was phosphorylated using BglK to generate the chromogenic 4-nitrophenyl 6-phospho-P-D-glucopyranoside (4NPpG6P or 2.13), which was used for the kinetic characterization of BglT. 2.11 (4NPPGIC) 2.13 (4NPpG6P) Scheme 2.3. Enzymatic synthesis of 4NPpG6P (2.13). (i) BglK, ATP. BglT catalyzes the cleavage of the C l - O l linkage of 4NPPG6P (2.13) to release G6P (2.3) and 4-nitrophenolate anion (2.12), which is readily detected by monitoring the change in absorbance at 400 nm (Figure 2.3). 2.13 (4NPPG6P) 2.3 (G6P) 2.12 ^-max = 400 nm Figure 2.3. BglT-catalyzed hydrolysis of the chromogenic substrate 4NPf>G6P (2.13). (i) BglT. This direct UV-vis assay was used to determine the optimal conditions for BglT activity. Optimal activity was obtained at pH 7.5, and therefore all subsequent kinetic analyses were performed in 50 mM HEPES adjusted to pH 7.5. The thermostable enzyme is active at 40-90 °C; thus, activity assays were conducted at 50 °C to minimize errors from Chapter 2 - BglT from Thermotoga maritima 37 changes in assay volumes due to evaporation at elevated temperatures or denaturation of G6PDH or Abg. Section 2.2.2. Dinucleotide Cofactor. Section 2.2.2.1. Dinucleotide Cofactor Specificity and Determination of the KA Value. Consistent with all other GH4 enzymes, BglT requires a dinucleotide cofactor for catalytic activity, the enzyme being completely inactive in the absence of N A D + . Thus, four different dinucleotide cofactors (NAD + , N A D H , N A D P + , NADPH) were investigated as potential activators. BglT was first dialyzed to remove any endogenous cofactor(s), and then assayed in the presence of each dinucleotide, of which only N A D + increased the rate of BglT-catalyzed hydrolysis of 4NP(3G6P (Table 2.1). Activity was then measured in the presence of various concentrations of N A D + as shown in Figure 2.4. A direct fit of the data to a simple hyperbolic binding equation yielded a value of 480 nM for the binding of N A D + to BglT. 1 — i — i — i — i — i — i — 1 — i 1 r 0 2 4 6 8 10 [NAD+] (uM) Figure 2.4. Ligand binding curve of N A D + for BglT. Chapter 2 - BglT from Thermotoga maritima 38 Dinucleotide Cofactor Act ivat ion of B g l T N H 2 0 OH OH OH OH N A D + + N H 2 H, H 0 k l j o o [ J N H ? N iy 1 ^ K / OH OH O H OH N A D H — N H 2 o % J ^ N 1 I W O O H O H OH e o-p-oe — N A D P + NH 2 H, H Q I D o O (J O OH O H OH © o-fj-oe N A D P H — Table 2.1. Test of dinucleotide cofactors for activation of BglT. Section 2.2.2.2. Kinetic and Spectroscopic Investigation of Dinucleotide Cofactor Reduction. A central feature of the proposed mechanism for BglT is the bound N A D + cofactor and its role in transient redox chemistry. Such essential "on-board" N A D + cofactors have been seen and characterized in a number of other enzymes, such as epimerases,7",•74 decarboxylases,75 and dehydratases,76"78 wherein they carry out a transient oxidation to acidify an adjacent proton or to permit epimerization via reduction Chapter 2 - BglT from Thermotoga maritima 39 from the opposite face (Section 1.3.2). Some confusion exists in the literature concerning the possibility that N A D + , as well as N A D H , could activate the enzymes in Family 4, 3 9 which is inconsistent with the proposed mechanism. Therefore, the exact role of the cofactor merited further investigation. Direct evidence for the involvement of N A D + in redox chemistry would generate substantial evidence in support of the proposed mechanism. The observation that dialysis of N A D + from the enzyme removed all catalytic activity and that full activity could be restored in a saturable fashion by titration with N A D + suggests an essential role for the cofactor. Further support for this, and direct proof that the reduced form (NADH) was inactive, was derived by reduction of an enzyme sample with sodium borohydride and demonstration that this enzyme form was devoid of activity. Analysis of these samples by UV-visible spectroscopy confirmed that the bound N A D + had indeed been reduced to NADH (kimx = 340 nm). The absorbance spectra (from 320 to 400 nm) of BglT were recorded under three different conditions: 10 uM of dialyzed BglT; 10 p M BglT incubated with 10 | l M N A D + ; and 10 p M BglT incubated with 10 U.M N A D + , and 10 mM sodium borohydride (Figure 2.5). The peak in absorbance at 340 nm corresponding to NADH appears upon reduction of the enzyme sample with sodium borohydride, and is consistent with the quantitative reduction of N A D + to N A D H . On the basis of the small Kd value of 480 nM for N A D + , more than 99% of BglT has N A D + or N A D H bound to its active site under these conditions, and very little remains free in solution. 0.16 | : 1 320 340 360 380 400 V\fehelength(nrn) Figure 2.5. Absorbance spectra of BglT in its oxidized (NAD + ) and reduced (NADH) forms. Absorbance spectra of 10 uM BglT (—); 10 U.M BglT incubated with 10 uM N A D + (---); and 10 uivl BglT incubated with 10 U.M N A D + and 10 mM sodium borohydride ( ). Chapter 2 - BglT from Thermotoga maritima 40 As shown in Figure 2.6, hydrolysis of 4NPf3G6P (2.13) by BglT was assayed in the oxidized (NAD + ) and reduced (NADH) state. The enzyme is completely inactive in the reduced state, which was achieved by quantitative reduction of the N A D + using 10 mM sodium borohydride. Importantly, addition of a fresh aliquot of N A D + to this sample (after sufficient time had elapsed for consumption of excess 'NaBH'4) restored the enzyme to full activity. This confirmed the essential role of N A D + and showed that loss of activity was not caused by other damage to the enzyme. It also confirmed that N A D H could not activate the enzyme, consistent with the proposed mechanism. The activation observed previously in G l v A 3 9 presumably arose from contaminating N A D + within the NADH sample. Only minute amounts of N A D + contamination of the readily oxidized NADH would be necessary to give rise to the observed activation of the enzyme given the low Kd of 37 uM for N A D + reported for G l v A 3 9 and the catalytic amounts of enzyme employed in those experiments. 1 1~—i r- 1 1 1 1 0 10 20 30 Time (min) Figure 2.6. Assay of BglT in the oxidized (NAD + ) and reduced (NADH) states. Observed rates of hydrolysis of 4NPpG6P by BglT via the detection of 4-nitrophenolate release at 400 nm. In each case, the assay volume was 200 uL, and the concentration of BglT used was 2.25 ug/mL. Control (—): standard BglT assay conditions, 50 mM HEPES (pH 7.5), 0.1 mM MnCI 2 , 10 uM N A D + , 10 mM 2-mercaptoethanol, and 0.1% (w/v) BSA at 50 °C. BglT assay conditions (---): BglT preincubated in 50 mM HEPES (pH 7.5), 0.1 mM MnCI 2 , 10 uM N A D + , 10 mM NaBH 4 , 10 mM 2-mercaptoethanol, and 0.1 % (w/v) BSA at 50 °C. No release of 4-nitrophenolate is observed until time = 18 min, when 2 nmol of N A D + was added. Section 2.2.3. Divalent Metal Specificity and Determination of the KA Value. The metal dependency of BglT was also investigated. An enzyme sample was dialyzed to remove any bound metal ions, and it was found to be catalytically inactive Chapter 2 - BglT from Thermotoga maritima 41 against the substrate 4NPpG6P. A number of metal ions (at 1 mM concentrations) were tested for their ability to activate samples of dialyzed enzyme (Table 2.2). Of the divalent metal ions tested, 1 mM M n 2 + was the best activator, followed by M g 2 + and Ca 2 + , which provided approximately 100-fold less activation. Divalent Meta l Ion Activat ion of Bg lT relative to 1 m \ l M n 2 + M n 2 + ( M n C l 2 ) 1.0 Z n 2 + (ZnS04) Enzyme precipitation Ca 2 + (CaCl 2 ) 5.5 x 10"5 M g 2 + (MgCl 2) 7.8 x 10"3 N i 2 + (NiCl 2 ) Enzyme precipitation C u 2 + (CuS04) Enzyme precipitation C o 2 + (CoCl 2) Solution turns brown with 2-mercaptoethanol Table 2.2. Test of divalent metal ions for activation of BglT. Subsequently, a ligand-binding curve was generated by measuring the rate of hydrolysis of 4NPPG6P in the presence of added M n 2 + . By fitting the data to a simple hyperbolic equation (Figure 2.7), a dissociation constant (Kd) of 32 uM was obtained. 0 0.05 0.1 0.15 0.2 [Mn^] (mM) Figure 2.7. Ligand binding curve of M n 2 + for BglT. Section 2.2.4. Reducing Conditions. The requirement for reducing conditions was investigated using three commonly used reducing reagents: tris(2-carboxyethyl)phosphine hydrochloride (TCEP), dithiothreitol (DTT), and 2-mercaptoethanol (Table 2.3). DTT and 2-mercaptoethanol Chapter 2 - BglT from Thermotoga maritima 42 are commonly used in enzyme assays to maintain thiol groups in the reduced state during enzyme manipulation. Oxidation of DTT is driven by the formation of an intramolecular disulfide bond in an energetically favorable six-membered ring. The efficiency, as well as stability and mild odor of DTT compared to 2-mercaptoethanol make DTT a more commonly used reducing reagent. TCEP is also used in the reduction of enzyme samples. In some cases, TCEP is preferred over DTT, because TCEP is stable over a wider pH range and is a stronger reductant than DTT. 1 3 1 In addition, DTT oxidation is catalyzed by some metal ions, 1 3 2 and the reagent forms precipitates with M n 2 + under basic conditions, which might complicate BglT assay conditions. While TCEP did not activate BglT, preincubation with either 2-mercaptoethanol or DTT activated the enzyme. Although DTT was found to activate BglT, activity could not be could not be reinstated to BglT in a saturable fashion by titration with DTT since, under the conditions used, the presence of even 5 mM DTT, along with MnCl 2 and 4NP(3G6P caused a precipitate to from in the assay solution, interfering with spectrophotometric analysis. On the other hand, up to 10 mM 2-mercaptoethanol could be used to activate BglT without compromising the enzyme assay. Consequently, 10 mM 2-mercaptoethanol was included in all subsequent activity assays. GH4 members, such as M e l A , 4 6 - 4 7 - 5 5 Agu4A, 5 2 Agu4B, 6 5 M a l H , 4 0 " 4 2 , 5 0 and A g l A , 3 8 ' 4 4 - 4 5 have also been reported to require addition of DTT for catalytic activity. A possible explanation for this was derived when the x-ray crystal structure of AglA from Thermotoga maritima was solved by Striiter and co-workers.44 It was noted that an absolutely conserved Cys, found at the active site, was oxidized to a sulfinic acid and this may have resulted in an inactive AglA sample.44 It is likely that the conserved active site Cys residue needs to be in its reduced thiol form for catalytic activity. Chapter 2 - BglT from TJiermotoga maritima 43 Reducing Reagent Activation of BglT O ^ O H / • HCI 1 fl O O TCEP HO OH + \ / SH HS DTT + 2-mercaptoethanol Table 2.3. Test of reducing reagents for activation of BglT. Section 2.2.5. Results of Kinetic Characterization of BglT. Based on the above findings, all kinetic assays of BglT included 1 uM N A D + , 0.1 mM M n 2 + , 10 mM 2-mercaptoethanol, and they were conducted in 50 mM HEPES buffer (pH 7.5) at 50 °C. The natural (C6'P, 2.2) and chromogenic (4NPf3G6P, 2.13) substrates were assayed with BglT via the G6PDH coupled assay (kmK - 340 nm, ENADPH = 6220 cirf'M"') and the direct UV-vis assay (A-max = 400 nm, Ae = 13791 M"'cm _ l) respectively. Kinetic parameters (summarized in Table 2.4) were determined on the basis of a direct fit of the data to the Michaelis-Menten equation (Appendix 3). Potential concerns regarding the reliability of the results obtained with an unnatural substrate, especially one that contains a much better leaving group than the natural substrate, are minimal, because, as is clear from this data, BglT binds and hydrolyzes 4NP(3G6P and C6'P in a similar fashion. Chapter 2 - BglT from Thermotoga maritima 44 Substrate ^cat (S ) KM (UM) ^cai/A'w ( s ' m M 1 ) C6'P (2.2) 0.61 69 8.8 (natural substrate) 4NPpG6P(2.13) 0.99 48 21 (chromogenic substrate) Table 2.4. Comparison of the kinetic parameters for the BglT-catalyzed hydrolysis of the natural substrate C6'P (2.2) and the chromogenic substrate 4NPpG6P (2.13). The kinetic parameters measured for the hydrolysis of 4NPpC6'P (2.10) via the Abg (A,™ = 400 nm, Ae = 13852 M'cm" 1) and G6PDH (X 1 1 i a x = 340 nm, £ N A D P H = 6220 M ' c n f 1 ) coupled assays are presented in Table 2.5. BglT binds 4NPpC6'P with high affinity, with KM values from the Abg and the G6PDH coupled assays being over 10-fold lower than those obtained for the natural (C6'P) and chromogenic (4NPpG6P) substrates, but the KJKM values for all three substrates (4NPpC6'P, C6'P and 4NPpG6P) are similar. It is noteworthy that the kinetic parameters for the BglT-catalyzed hydrolysis of 4NPPC6'P obtained from the G6PDH and the Abg coupled assays are in reasonable agreement, but not identical. This is largely a consequence of the low KM of BglT for this substrate, which makes the measurement of initial rates challenging with a coupled assay due to the need to accumulate sufficient product for the coupling enzymes to react efficiently. The values obtained may therefore be slight underestimates of the true KJKM values. Assay *cal («"') KM (UM) KJKM ( s ' m M 1 ) G6PDH coupled assay 0.017 1.0 16 A b g coupled assay 0.026 7.9 Table 2.5. Comparison of the kinetic parameters for the hydrolysis of 4NPpC6'P (2.10) via the Abg and the G6PDH coupled assays. It was important to establish the substrate specificity of BglT, because GH4 comprises a diversity of enzymes, including 6-phospho-a-glucosidases (-10% of the sequences deposited in C A Z Y ) , 3 9 4 3 a-glucosidases (~26%),44'4:> a-galactosidases (~3%)1 3 3, and 6-phospho-P-glucosidase (-35%).4 S Furthermore, a GH4 member, M a l H / 0 had also been reported to hydrolyze both a- and P-glycosidic linkages of substrates with activated leaving groups such as 4-nitrophenol. Therefore, substrate specificity was investigated using a number of aryl glycosides (listed in Table 2.6): 4NPpGlc (2.11), 4-Chapter 2 - BglT from Thermotoga maritima 45 nitrophenyl a-D-glucopyranoside (4NPaGlc or 2.14), 4-nitrophenyl (3-D-galactopyranoside (4NP(3Gal or 2.15), 4-nitrophenyl oc-D-galactopyranoside (4NPocGal or 2.16), and 4-nitrophenyl 6-phospho-oc-D-glucopyranoside (4NPocG6P or 2.17). Because BglK does not phosphorylate a-glucopyranosides, 4NPaG6P (2.17) was synthesized via the selective phosphorylation at 06 of the commercially available 4NPocGlc (2.14) using POCh 1 3 4 Al l other aryl glycopyranosides are commercially available. With the exception of 4NP(3G6P (2.13), none of the aryl glycosides tested were hydrolyzed by BglT even at high concentrations of enzyme (10-fold higher than those normally used for the hydrolysis of 4NP(3G6P). Clearly, BglT only hydrolyzes (3-glucosidic linkages and substrates must also possess a phosphate group at the C6 position. 2.14(4NPaGlc) 2.17(4NPaG6P) Scheme 2.4. Synthesis of 4NPaG6P (2.17). (i) POCl 3 , PO(OCH 3 ) 3 , H 2 Q. Chapter 2 - BglT from Thermotoga maritima 4 6 Potential Substrate Hydrolysis by Bg lT Potential Substrate Hydrolysis by Bg lT ^ O H 2.11 (4NPpGlc) — OH/OH H O A - ^ A ^ O - ^ / ^ 2.15(4NPpGal) , O P 0 3 N a 2 H O A » ^ - ^ ^ 0 - - _ / ^ , ^vT \ 1 2.13 (4INPpG6P) + OH/OH H O - \ ^ T " - - \ OH] N 0 2 2.16 (4NPaGal ) — -OH H O - ^ ~ ^ " ° \ HO~V*-^A OH N 0 2 2.14 (4NPaGlc) — , O P 0 3 N a 2 OH] 'XX N 0 2 2.17(4NPaG6P) — Table 2.6. Substrate specificity of BglT. Section 2.3. Mechanistic Studies. Section 2.3.1. Stereochemical Outcome Determination. As mentioned in Section 1.2.1.2, a mechanistic signature of glycosidases is the stereochemical outcome of the reaction catalyzed. Therefore, classification of BglT as either a retaining or an inverting glycosidase may provide insights into the reaction mechanism. Direct analysis by 'H NMR, which was successfully used for a number of other glycosidases,13:> was rendered problematic by the requirement for the paramagnetic M n 2 + ion for optimal activity. The unpaired electrons have a magnetic moment that is approximately 600 times greater than that of a proton.1 3 6 Therefore, the paramagnetic metal provides a very efficient relaxation source and causes the line broadening of ' H NMR signals. Additionally, direct 'H NMR determination of the stereochemical Chapter 2 - BglT from Thermotoga maritima outcome of most systems requires that the enzyme reaction be complete within 4 minutes, because the half-life for the mutarotation of a-D-glucose is approximately 7 minutes.137" 0 9 In the case of BglT, the reaction product G6P (2.3) mutarotates over 200 times faster than glucose,1 3 7"1 3 9 which made a stopped analysis involving M n 2 + removal impractical. The meager activation provided by M g 2 + or C a 2 + in comparison to that provided by M n 2 + also means that mutarotation of G6P product would preclude NMR analysis by substitution of M n 2 + with another metal. Instead, another approach involving the formation of a stable product was sought. The stereochemical outcome determination was achieved in a previous difficult case with a Family 39 a-L-iduronidase17 by use of a methanolysis reaction. Fortunately, BglT retained reasonable activity in 5 M methanol such that a mixture of the hydrolysis product and the stereochemically stable methyl 6-phospho-glucopyranoside were formed. Upon completion of reaction, the mixture was treated with alkaline phosphatase, which resulted in a mixture of oc-D-glucopyranose (aGlc or a-2.4), p-D-glucopyranose (pGlc or (5-2.4), and the stereochemically stable methyl D-glucopyranoside. Determination of the anomeric configuration of the product methyl D-glucopyranoside would reveal the stereochemical course of the BglT reaction. Prior to 'H NMR analysis, the enzymes (BglT and alkaline phosphatase) were removed from the reaction mixture using a centrifugal filter unit, and metal chelating resin was used for removal of the M n 2 + ion. The reaction scheme for this experiment is shown in Figure 2.8. 0-2.4 (pGIc) a-2.4 (aGlc) Figure 2.8. Schematic of the BglT-catalyzed methanolysis of 4NP(iG6P (2.13) for the stereochemical outcome determination, (i) BglT, buffer and 5 M methanol; (ii) alkaline phosphatase; (iii) mutarotation. Chapter 2 - BglT from Thermotoga maritima 48 Comparison of the 'H NMR spectrum of the reaction mixture (Figure 2.9) to those of standard samples of aGlc (a-2.4), p-Glc (p-2.4), 4NPpG6P (2.13), MepGlc (2.18), and methyl a-D-glucopyranoside (MeaGlc or 2.19) revealed that BglT-catalyzed the hydrolysis and methanolysis reactions to form glucose (2.4) and MefJGlc (2.18) respectively. In the spectrum shown in Figure 2.9(a), the doublets at 5.07, 4.49 and 4.22 ppm correspond to the anomeric proton signals of aGlc (a-2.4), pGlc (P-2.4) and MepGlc (2.18) respectively. D,0 (a) (b) (c) (d) (e) ppm(f1) 7.0 6.0 HI (2.13) D,0 D20 HI (2.18) HI (2.19) D20 5.0 J U U 4.0 Figure 2.9. 'H NMR spectra of the BglT methanolysis reaction. 'H N M R spectra of (a) reaction mixture; (b) a- and p-glucopyranose (a-2.4 and p-2.4 respectively); (c) substrate 4NPPG6P (2.13); (d) MepGlc (2.18); (e) MeaGlc (2.19). As shown in Figure 2.9(a), the D2O peak is very broad. Therefore, to ensure that the anomeric 'H signal of MeaGlc (2.19), if present in the reaction mixture, is not Chapter 2 — BglT from Thermotoga maritima 49 masked by the solvent D2O peak at 4.63 ppm, 'H NMR spectra of the reaction mixture along with standard solutions of MepGlc (2.18) and MeaGlc (2.19) were also collected at 50 °C. At higher temperatures, the D2O peak is shifted upfield, and it was shown unambiguously in Figure 2.10(a) that there are no proton signals at 4.68 ppm that would correspond to the anomeric proton of MeaGlc (2.19). The formation of only the p-anomer (MepGlc, 2.18) from the methanolysis indicates that BglT is a retaining glycosidase. Whether BglT utilizes the same double displacement mechanism employed by other "classical" retaining glycosidases requires further analysis. D,0 (a) (b) (c) (d) (e) 6.0 HI HI (2.18), HI (p-2.4) (a-2.4) HI (2.19) D,0 "i—1—1—1—1—r~i~">—r 5.0 4.0 ppm (f1) Figure 2.10. 'H N M R spectra of the BglT methanolysis reaction at elevated temperatures. 'H N M R spectra of : (a) reaction mixture, 50 °C; (b) MepGlc (2.18), 50 °C; (c) MeaGlc (2.19), 50 °C; (d) MepGlc (2.18), room temperature; (e) MeaGlc (2.19), room temperature. Chapter 2 - BglT from Thermotoga maritima 50 Section 2.3.2. Solvent Deuterium Isotopic Exchange. Section 2.3.2.1. Solvent Deuterium Isotopic Exchange - Methanolysis. The methanolysis reaction in the stereochemical outcome determination revealed an interesting finding. The methanolysis reaction was conducted in D2O buffer containing 5 M C D 3 O D with BglT that had been exchanged into deuterated buffer via repeated dilution and concentration using a centrifugal filter unit. Upon completion, the reaction mixture was treated with alkaline phosphatase. Then, the enzymes and metal ion were removed using a centrifugal filter unit and metal chelating resin respectively. The 'H NMR spectrum of the reaction mixture was collected at room temperature and at 50 °C (Figure 2.11). Analysis of the proton signals revealed similar reaction products to those determined previously in the methanolysis in 5 M C H 3 O H described in Section 2.3.1. The main difference is that the anomeric proton signals at 5.07, 4.49, and 4.22 ppm are singlets (corresponding to 2[2H]-a-D-glucopyranose (2[ 2H]aGlc or a-2.21), 2[2H]-p-D-glucopyranose (2[2H](3Glc or (3-2.21), and compound 2.20 respectively) and not doublets. The lack of first order ' H coupling for the anomeric protons indicates that the C2 proton had been exchanged for solvent deuterium. Chapter 2 - BglT from Thermotoga maritima 51 (a) 0 P O 3 N a 2 2.13 (4NPPC6P) 2.20 OH] a P-2.21 (2|2H|pGlc) H c r - \ - V-K\ HCK\*^--y--H OH a-2.2I (2| 2H|aGle) (b) fe) (d) (i) HI (2.19) I D,0 ppm(t1) 5.50 5.00 4.00 3.50 Figure 2.11. 'H N M R of the BglT-catalyzed methanolysis of 4NPpG6P (2.13) in D 2 0 buffer and 5 M CD 3 OD. (a) Schematic of the BglT-catalyzed methanolysis reaction in D 2 0 buffer: (i) BglT, D 2 0 / C D 3 O D ; (ii) alkaline phosphatase; (iii) mutarotation; and 'H NMR spectra of (b) reaction mixture, room temperature; (c) reaction mixture, 50 °C; (d) a- and P-D-glucopyranose (a-2.4 and P-2.4 respectively), room temperature (e) MePGIc (2.18), room temperature; (t) MeaGIc (2.19), room temperature. Section 2.3.2.2. Solvent Deuterium Isotopic Exchange - Substrate and Product. The solvent deuterium isotope exchange phenomenon was more fully explored using G6P (2.3) and 4NPPG6P (2.13) under conditions favorable for BglT-catalyzed hydrolysis rather than methanolysis. As a control experiment to ensure that exchange did not occur after the hydrolysis reaction in a non-enzymatic process via labeling of the Chapter 2 - BglT from Thermotoga maritima 52 acyclic form; possible exchange into the reaction product G6P (2.3) was investigated. As shown in Figure 2.12, no such exchange was observed with G6P (2.3) or 4NP(3G6P (2.13) when incubated with the D 2 0 buffer and cofactors in the absence of enzyme. On the other hand, solvent deuterium isotopic exchange into C2 of the hydrolysis product G6P (2.3) was observed when the reaction of BglT with 4NPpG6P (2.13) was carried out in D2O buffer in the absence of methanol, as revealed by the observation of singlets at 5.13 and 4.54 ppm (corresponding to 2[2H]-a-D-glucose 6-phosphate (2[2H]aG6P or a-2.22) and 2[2H]-p-D-glucose 6-phosphate (2[2H]pG6P or p-2.22) respectively) for the anomeric protons in the H N M R spectrum (Figure 2.12). In each case, the Mn was removed using metal chelating resin prior to NMR analysis. The C2 proton of the G6P with BglT in D2O buffer under the same reaction conditions also resulted in exchange of the C2 proton. Exchange of the C2 proton is therefore a direct consequence of the catalytic mechanism. Chapter 2 - BglT from Thermotoga maritima 53 m OP03Na 2.13(4NPpG6P) p-2.22 (2[2H|PG6P) ,OP03Na2 a-2.22 (2|2H|aG6P) OP0 3Na 2 O H i P-2.3 (PG6P) OP0 3Na 2 HO-\^»-P«A^H OHI OH a-2.3 (aG6P) no H p-2.22 (2I2H|PG6P) OP03Na2 HO--OH| OH a-2.22 (2|2H|aG6P) (C) (d) HI (a-2.22) HI i (a-2.22) HI (2.13) ( e ) T \ HI ! '~\ (a-2.3) 5.0 D ,0 | D ,0 0 , 0 / I ChO 48 ppm V HI (P-2.22) 1 HI (P-2.22) J ^ V 4.6 Figure 2.12. 'H NMR of the BglT-catalyzed hydrolysis of4NPpG6P (2.13) in D 2 0 buffer, (a) Schematic of the BglT-catalyzed hydrolysis of 4NPpG6P (2.13) in D 2 0 buffer: (i) BglT, D 2 0; (ii) mutarotation; (b) Schematic of the solvent deuterium incorporation into G6P: (i) BglT, D 20; (ii) mutarotation; and 'H NMR spectra of: (c) BglT hydrolysis of4NPpG6P (2.13) in D 2 0 buffer; (d) aG6P (a-2.3) and PG6P (|}-2.3) incubated in D zO buffer with BglT; (e) Control experiment: 4NPpG6P incubated in D 2 0 buffer without BglT; (f) Control experiment: aG6P (a-2.3) and pG6P (P-2.3) incubated in D 2 0 buffer without BglT. A number of possible reasons for this exchange should be considered. In accordance with the proposed mechanism, the activation of the C2 proton could be provided by the adjacent C3 ketone formed via the NAD+-mediated oxidation of the hydroxyl group. Alternatively, considering that the deuterium exchange is observed in the reducing sugar product G6P (2.3), it is possible that the activation of the C2 proton could be due to the presence of the aldehyde at C l of the sugar in its acyclic form. Chapter 2 - BglT from Thermotoga maritima 54 Deuterium incorporation in the methanolysis reaction suggested that deuterium incorporation is a direct consequence of the enzymatic reaction, because the methyl glucoside product formed is chemically stable and does not undergo spontaneous hydrolysis or ring-open into an acyclic form. Thus, deuterium incorporation is not a result of non-enzymatic equilibrium exchange into the reaction product in the assay solution (Figure 2.12). Moreover, this experiment provided justification of the stereochemical outcome determination from the methanolysis reaction, because it confirms that the methanolysis product is formed by the enzyme catalyzed process and not an indiscriminate nucleophilic attack by MeOH at the anomeric center. Possible solvent deuterium incorporation into the unreacted substrate (4NP(3G6P) was also investigated via mass spectrometry. Therefore, 4NP(3G6P was reacted with BglT in D 2 0 buffer. Aliquots of the reaction mixture were taken at various rime points prior to reaction completion and the reaction was quenched by flash freezing at -78 °C. The samples were analyzed by mass spectrometry and no deuterium incorporation into unreacted substrate was detected. This indicates that reprotonation of the C2 anion and release from the active site are slower than the elimination step, as will be discussed in Section 2.3.6.3. As a yet more rigorous exploration of the phenomenon of solvent deuterium incorporation at the C2 position, a substrate analogue 1,5-anhydroglucitol 6-phosphate (2.23), which cannot undergo any ring-opening or glycosidic bond cleavage process was subjected to analysis. The 'H NMR spectra of 2.23 that had been incubated in D2O buffer with and without BglT are shown in Figure 2.13. After incubation of 2.23 in the presence of BglT, the C2 proton was partially exchanged for solvent deuterium, as is demonstrated by the emergence of a doublet for the axial-Hl and for the H3 protons (overlapped with the triplet signals of 2.23), and the slight decrease in the H2 signal upon inclusion of BglT in the D2O buffer system. The absence of a leaving group at the anomeric center and the inability of the substrate to ring-open remove two possible side reactions that could lead to the exchange process, thereby directly supporting a mechanism involving formation of the C3 ketone during the course of the enzymatic reaction. Chapter 2 - BglT from Thermotoga maritima 55 2.23 2.24 H 6 a , H6b , H i e 3.90 3.70 3.50 3.30 3.10 2.90 ppm Figure 2.13. 'H N M R analysis illustrating the solvent deuterium incorporation into compound 2.23. (a) Schematic of the solvent deuterium incorporation into 2.23: (i) BglT, D 2 0 buffer; and 'H NMR spectrum of: (b) 2.23 incubated in D 2 0 buffer with BglT; and (c) Control experiment: 2.23 incubated in D 2 Q buffer without BglT. Section 2.3.3. Active Site Architecture and Mechanistic Implications. Based on the data presented thus far, proposed mechanism #3 involving the oxidation-elimination-addition sequence was formulated. However, further structural and mechanistic data were required to substantiate the mechanistic proposal. Fortunately, our collaborators, Dr. Annabelle Varrot and Dr. Gideon Davies, solved the x-ray crystal structure for the BglT-product complex (PDB 1UP6) around the same time as we were obtaining key results in the current mechanistic study. Furthermore, the first x-ray crystal structure of a GH4 enzyme, AglA (PDB lOBB) from Thermotoga maritima, had been solved by Strater and co-workers44 just a few months prior. However, as mentioned before, the crystal structure of AglA is that of an inactive enzyme and no metal ions were bound to the AglA crystal.44 As such, no mechanistic proposal was formed and it was difficult to draw conclusions about active site residues from that work. Nonetheless, some useful information can be derived, because the overall structure is very similar to that of BglT. Chapter 2 - BglT from Thermotoga maritima 56 In the case of BglT, the enzyme is a mixed a/(3 protein composed of 12 (3-strands and 16 a-helices. The N-terminal region contains a glycine-rich sequence characteristic of the dinucleotide-binding Rossman fold. 1 4 0 As with A g l A , 4 4 the BglT fold shows no structural similarity to any known glycosidases, but it displays a strong resemblance to lactate/malate dehydrogenases known to utilize a N A D / N A D H cofactor for redox chemistry.1 4 1"1 4 3 The enzyme also showed some structural similarities to dehydratases. Of particular interest to the current study was the active site architecture (Figures 2.14). Arg87 \ NH !l Lys41 Figure 2.14. A diagram of the interaction of BglT with M n 2 + , N A D + and G6P. The M n 2 + interacts with NE2 from His 192 (2.8 A), SH from Cysl62 (2.7 A), N07 of the nicotinamide ring (2.3 A) and hydroxyls 02 and 03 of G6P (2.3 and 2.5 A respectively). For the adenosyl portion of the N A D + , the AN1 atom is hydrogen bonded Chapter 2 - BglT from Tfiermotoga maritima 57 to the hydroxyl group of Tyrl23 (2.7 A). Atoms 02 and 03 of the ribose moiety are hydrogen bonded to OD1 of Asp36 (2.5 A) and NZ of Lys41 (2.7 A), respectively. The A O l oxygen atom of the phosphate group interacts with the NE atom of Arg80 (2.9 A), while Glu269 also interacts with this phosphate group, suggesting that Glu269 is protonated in the complex. In the nicotinamide moiety, the N O l and N02 oxygen atoms of the phosphate group are hydrogen bonded with NH2 of Arg8() (3.1 A) and OH of Ser 10 (2.7 A), respectively. The 02 hydroxyl of the ribose group interacts with OE2 of Glul03 (2.8 A). The 03 hydroxyl group of G6P is hydrogen bonded to the ND2 of Asnl40 (2.8 A) and NE2 of His 192 (3.1 A). The 02 hydroxyl group is hydrogen bonded to the ND2 of Asn 163 (2.5 A). The 02P atom of the phosphate group interacts with the NE of Arg261 (2.7 A), whilst the 03P interacts with the NH1 (2.6 A) and NH2 of Arg87 (3.0 A). Figure 2.15. Stereoview of the active site of BglT. The figure was generated using Swiss-PDB Viewer. 1 4 4 Inspection of the active site (Figures 2.14 and 2.15) in light of the mechanistic proposal reveals that the N A D + cofactor, with C4 of the nicotinamide ring poised at 4.2 A beneath C3 of G6P, is perfectly positioned for C3 hydride abstraction thus oxidation of the C3 hydroxyl. Tyr241 is located directly over C2 and can potentially act as a catalytic base. The importance of the Tyr241 residue is corroborated by the pH-activity profile (see Section 2.3.4) and subsequent mutagenesis studies (Section 2.4). Furthermore, Tyr241 lies in close proximity (~ 3.4 A) to the glycosidic oxygen atom. Tyr241 may in fact play a dual role, providing general acid catalysis to the departing OI as well as C2 deprotonation, as suggested for the tyrosine catalytic base of family PL8 chondroitin/hyaluronan lyases.145 It is interesting to note also that the M n 2 + is chelated to 02 and 03 of G6P. While, the Mn" + may play an important role in substrate binding, Chapter 2 — BglT from Thermotoga maritima 58 mechanistically, the Mn probably plays an important role in the stabilization of the proposed enediolate intermediate thereby promoting C2 deprotonation. This would account for the absolute requirement for the divalent metal for catalytic activity. The high density of negative charge generated by proton abstraction reactions in some enzymatic systems is thought to be made possible via metal-mediated stabilization of the anionic species.71*146 The finding that the divalent metal is chelated to 02 and 03 of G6P in the crystal structure supports this proposal (Figure 2.16). This is most readily recognized in mandelate racemase in which the divalent M g 2 + is coordinated to both the carboxylate oxygen and to the a-hydroxyl group, 1 4 6" 1 4 8 in a manner that is very similar to that proposed for the BglT reaction intermediate. 3-kcto intermediate cncdiolatc intcnncdiatc Figure 2.16. Proposed stabilization of the enediolate intermediate by the divalent metal ion. (a) Proposed electrostatic stabilization by M g 2 + in the mandelate racemase mechanism.' 2 2 ' 1 4 6; (b) Proposed electrostatic stabilization of the enediolate intermediate by M n 2 + in the proposed GH4 mechanism. Section 2.3.4. pH-Dependence. The pH-dependence of activity can provide significant insight into the nature of the enzyme active site. Because many enzymatic reactions require general acid/base catalysis, important acid and base functionalities must be in the correct ionization states for catalysis. Even though any enzyme contains numerous ionizable functionalities, the Chapter 2 - BglT from Thermotoga maritima 59 pH-dependence of the activity of an enzyme can often be attributed to one or two ionizable groups within its active site, because usually only the ionization of groups directly involved in catalysis at the enzyme active site would affect enzyme activity. Thus, the pH-dependence of the rates of many enzymatic reactions takes the form of single or double ionization curves. Therefore, valuable clues to the identities of the catalytic residues can be derived from pH-activity profiles. The pH-stability of BglT was determined by measuring the loss of activity, as measured under standard assay conditions (pH 7.5), after the enzyme sample had been incubated in different buffers at various pH conditions. No loss of activity was found for enzyme samples incubated in the pH range of 4-12 over a period equivalent to assay times, thereby defining the limits of pH conditions to be used in investigating the pH-dependence of kcat and kcJK.M- The plot of k^IKu vs pH was generated by the substrate depletion method using 4NP(3G6P (2.13) as substrate, and the data were fit to a classical bell-shaped curve. The plot of kcJKu vs pH reflects ionizations of the free enzyme and the free substrate. Ionizations of the E*S (enzyme-substrate) complex are manifested in the pH-dependence of kcax, which was investigated by measuring the initial rates of hydrolysis of 4NPpG6P under saturating substrate concentrations at various pH values. Although BglT is stable under basic conditions up to pH 12, the assay solution forms precipitates under these conditions; thus, no data points were collected at pH >10. Plots of km vs pH and kcJKM vs pH are shown in Figure 2.17, and both sets of data were fit to classical bell-shaped curves. The fit and error analysis were performed with the program, GraFit4. 1 4 9 The pKa values deduced and the p H o p t obtained are summarized in Table 2.7. Chapter 2 - BglT from Thermotoga maritima 60 Figure 2.17. pH-Activity profiles of BglT. (a) Plot of kcJKM vs pH; (b) Plot of kc.dl vs pH. P#a. pK:>2 pH o p t pH-dependence of kcM 7.0 ±0.1 8.2 ±0.1 7.6 pH-dependence of 7.1 ±0.1 9.3 ±0.1 8.0 Table 2.7. Summary ofpKd], pKta, and pH o p i values obtained for BglT from the plots of £ c a l vs pH and of kcJKM vs pH. The pATai value of approximately 7 could correspond to that of Tyr241 since this residue would need to be deprotonated to serve as the catalytic base. The normal pKa of 10 for Tyr could easily be lowered by 2-3 pKa units in the enzyme active site. Alternatively, since BglT displays selectivity for p-D-glucosides with a phosphate group at 06, this pATai value could potentially correspond to that of the phosphoryl moiety of the substrate. The pAfa2 value of approximately 9 may represent the ionization of the conserved Arg rcsidue(s) that, by x-ray crystallographic analysis, are found to be within hydrogen bonding distance of the phosphate group of the substrate. Again, the normal p/Ca of 12 for Arg could be lowered to 9, and it is reasonable to expect that the Arg residue(s) would need to be protonated in order to form electrostatic interactions with the substrate phosphate moiety. A cautionary note is warranted, because it was later discovered that kinetic isotope effects were pH-dependent (Section 2.4.8), suggesting that the rate-limiting step changes with pH. Thus, the measured pKa values are kinetic pKa's and not only represent the pA"a values of ionizable goups but also include contributions Chapter 2 - BglT from Thermotoga maritima 61 from the pH-dependence of all the rate constants affecting the reaction. Therefore, the interpretation that the measured pKa values represent the ionization of active site residues may be an oversimplification. Section 2.3.5. Types of Elimination Reactions. There are 3 fundamental elements in any 1,2-elimination mechanism. In the context of the proposed BglT mechanism, the elimination step involves: 1) Cleavage of the C2-H2 bond to generate the enolate intermediate; 2) Cleavage of the C l -Ol linkage and departure of the leaving group; 3) Formation of an unsaturated bond between C l and C2. The timing of the first two steps and the nature of the transition state structures can be used to distinguish between the three general classifications of 1,2-elimination reactions: E l , E2 and E l c b . The More O'Ferrall-Jencks diagram (Figure 2.18) provides a representation of the different classifications of the three elimination mechanisms.1 5 0'1 5 1 Chapter 2 - BglT from Thermotoga maritima 62 E l c b carbanion carbocation E l Figure 2.19. Chemical representation of the potential E l , E2, and E l c b elimination in the BglT mechanism. As shown in Figures 2.18 and 2.19, an E l mechanism involves the rate-limiting departure of the Cl-Ol-l inked leaving group and formation of a carbocation at C l , followed by the rapid cleavage of the C2-H2 linkage in a subsequent step. An E2 elimination mechanism by definition entails a concerted process, with cleavage of the C l - O l and C2-H2 bonds occurring to the same extent along the entire reaction coordinate. Finally, for an El Cb mechanism, an anionic intermediate is formed in the first step, which involves the rate-limiting C2 deprotonation. The lone pair of electrons then donates into the C1-C2 linkage forming the unsaturated double bond followed by rapid cleavage of the C l - O l bond. The description of elimination mechanisms given in this section is a simplified introduction and the three types of elimination mechanisms are treated as though they are distinct from one another. However, it must be noted that in reality many elimination mechanisms do not fit neatly into the description of an E l , El Cb or E2 mechanism. In Chapter 2 - BglT from Thermotoga maritima 63 fact, it is a matter of some debate whether there is in fact a distinct separation of a stepwise El Cb and a concerted E2 mechanism.1~2'153 Thus, although cleavage of the C2-H2 and C l - O l bonds occurs to the same extent along the entire reaction coordinate in an E2 elimination by definition, it is not necessarily a synchronous process in reality. The following sections describe how linear free energy relationships and kinetic isotope effects were used to study the nature of the bond-forming and bond-breaking steps in the proposed elimination mechanism for BglT. Section 2.3.6. Linear Free Energy Relationship. Section 2.3.6.1. Synthesis of Substrates for Linear Free Energy Relationship. Useful mechanistic data can often be obtained from the analysis of linear free energy or Bransted relationships. These studies provide information on the electronic effect of a substituent on a chemical reaction, in particular by probing the charge development at various centers in the transition state of the reaction. Such information can aid in the understanding of reaction mechanisms. The theory behind linear free energy relationships is explained in detail in Appendix 2. Linear free energy relationship analysis was performed by varying the reactivity of the leaving group at C l . The initial study began with the investigation of a series of aryl 6-phospho-p-D-glucopyranosides, whose kinetic parameters can be easily measured in a direct spectrophotometric assay. A number of substrates (compounds 2.47—2.56) were selected for this study and were synthesized according to Scheme 2.5. Glucose was acetylated in pyridine to form 2.25. The oc-bromide (2.26) was synthesized by reaction of 2.25 in HBr dissolved in acetic acid. The aryl 2,3,4,6-tetra-O-acetyl p-D-glucopyranosides (2.27—2.36) were synthesized via the Koenigs-Knorr reaction, and those aryl P-D-glucopyranosides with leaving group pKa values below 6 were deprotected with HCl(g) dissolved in anhydrous methanol. Al l other aryl P-D-glucosides were deprotected using the Zemplen deacetylation procedure. Phosphorylation of all compounds (2.37—2.46) was accomplished enzymatically with BglK and ATP to yield the desired aryl 6-phospho-p-D-glucopyranosides (2.47—2.56). Chapter 2 - BglT from Thermotoga maritima 64 OAc AcO~\^—»^»A OAc I 2.4: R = H 2.25: R = Ac OP0 3Na 2 OH 2.26 2.47 (24DNPfiG6P) : R 1 2.48 (25DNPPG6P) : R 1 2:49 (34DNPPG6P) : R' 2.50 (4C2NPPG6P) : R 1 2.51 (2NPPG6P) : R 1 2.52 (35DCPPG6P) : R 1 2.53 (3NPpG6P) : R 1 2.54 (4CNPPG6P) : R 1 2.55 (PPG6P) : R ' 2.56 (4fBuPpG6P) : R 1 = 2,4-dinitrophenyl = 2,5-dinitrophenyl = 3,4-dinitrophenyl = 4-chloro-2-nitrophenyl = 2-nitrophenyl = 3,5-dichlorophenyl = 3-nitrophenyl = 4-cyanophenyl = phenyl = 4-f-butylphenyl 2.27: R 1 = 2,4-dinitrophenyl, R 2 = Ac 2.28: R 1 = 2,5-dinitrophenyl, R 2 = Ac 2.29: R 1 = 3,4-dinitrophenyl, R 2 = Ac 2.30: R 1 = 4-chloro-2-nitrophenyl, R 2 = A c 2.31: R 1 = 2-nitrophenyl, R 2 = Ac 2.32: R 1 = 3,5-dichlorophenyl, R 2 = Ac 2.33: R 1 = 3-nitrophenyl, R 2 = Ac 2.34: R 1 = 4-cyanophenyl, R 2 = Ac 2.35: R 1 = phenyl, R 2 = Ac 2.36: R 1 = 4-f-butylphenyl, R 2 = Ac 2.37: R 1 = 2,4-dinitrophenyl, R 2 = H 2.38: R 1 = 2,5-dinitrophenyl, R 2 = H 2.39: R 1 = 3,4-dinitrophenyl, R 2 = H 2.40: R' = 4-chloro-2-nitrophenyl, R 2 = H — * " 2.41: R 1 = 2-nitrophenyl, R 2 = H 2.42: R 1 = 3,5-dichlorophenyl, R 2 = H 2.43: R 1 = 3-nitrophenyl, R 2 = H 2.44: R 1 = 4-cyanophenyl, R 2 = H 2.45: R 1 = phenyl, R 2 = H 2.46: R 1 = 4-/-butylphenyl, R 2 = H Scheme 2.5. Synthesis of aryl 6-phospho-p-D-glucopyranosides. (i) A c 2 0 , pyridine; (ii) 33% HBr, AcOH, A c 2 0 ; (iii) NaOH, acetone and for 2.27: 2,4-dinitrophenol; for 2.28: 2,5-dinitrophenol; for 2.29: 3,4-dinitrophenol; for 2.30: 4-chloro-2-nitrophenol; for 2.31: 2-nitrophenol; for 2.32: 3,5-dichlorophenol; for 2.33: 3-nitrophenol; for 2.34: 4-cyanophenol; for 2.35: phenol; for 2.36: 4-f-butylphenol; (iv) For 2.37, 2.38, 2.39: HCI(„), MeOH; for 2.40, 2.41, 2.42, 2.43, 2.44, 2.45, and 2.46: NaOMe/MeOH; (v) BglK, ATP. The kinetic parameters for the hydrolysis of C6'P (2.2) and methyl 6-phospho-(3-D-glucopyranoside (MepG6P or 2.58) were also included in the Bronsted analysis to determine whether rate-limiting steps are the same for substrates with and without activated leaving groups. The kinetic parameters for C6'P were reported in Section 2.2.5. Chapter 2 - BglT from Thermotoga maritima 65 As shown in Scheme 2.6, the selective phosphorylation of the primary C6 hydroxyl of Me(3Glc (2.18) was achieved using diphenyl chlorophosphate, followed by catalytic hydrogenation to yield MepG6P (2.58). 0 „ H O - ^ V ~ ^ °\ H O - - X - " « ^ - » \ - - O M e O H O P O ( O P h ) 2 HO---vA Q H O - \ « - - ^ « \ ^ - - O M e OH 2.57 . O P 0 3 N a 2 HO- OMe OH 2 .58 (MepG6P) 2.18 (MePGlc ) Scheme 2.6. Synthesis of Me(3G6P (2.58). (i) (PhO) 2POCl, pyridine; (ii) Pt0 2 , H 2 ( g ) , MeOH Section 2.3.6.2. Linear Free Energy Relationship - Results and Discussion. Using the series of aryl 6-phospho-p-D-glucopyranosides (2.47 to 2.56) synthesized (Section 2.3.6.1), a full Bronsted analysis was performed. The phenol leaving groups have a broad range of reactivity, with pKa values ranging from 3.96 to 10.37.1 2 4 The initial rate of hydrolysis for each substrate by BglT was followed spectrophotometrically at the wavelength of maximal absorbance difference between the released phenol and the respective aryl 6-phospho-P-D-glycopyranoside. Enzyme concentration was adjusted such that less than 10% of the total substrate was consumed, ensuring linear rates. The difference in extinction coefficient (Ae) between the aryl 6-phospho-p-D-glucoside and the phenol released at pH 7.5 and 50 °C was determined by the total enzymatic hydrolysis of a solution of the aryl 6-phospho-fj-glucoside of known concentration (Table 2.8). Each substrate displayed Michaelis-Menten kinetics, and values of kcat and KM determined for each substrate are presented in Table 2.8. The logarithms of & c a t and kcJKM were calculated, and each was plotted against the pKa of the corresponding leaving group. The Bronsted plots thereby produced are shown in Figure 2.20. Both plots are essentially flat and indicate that neither /^ cat nor Acat/^ Mvi are significantly dependent on the phenol leaving group ability of these aryl 6-phospho-fj-D-glucopyranosides. Chapter 2 - BglT from Thermotoga maritima 66 A r y l 6-phospho-P-D-•jlucopyranoside Phenol pKa (nm) Ae (cm 1 M" 1) kcat ( s 1 ) KM (UM) ( s ' l n M " ' ) 2 4 D N P p G 6 P (2.47) 3.96 400 10282 2.29 44.4 52 2 5 D N P P G 6 P (2.48) 5.15 443 4383 1.95 16.6 117 3 4 D N P p G 6 P (2.49) 5.36 400 15982 1.15 31.4 37 4 C 2 N P p G 6 P (2.50) 6.45 428 4164 1.69 15.0 113 4 N P 0 G 6 P (2.13) 7.18 400 13791 0.99 48.6 20 2 N P p G 6 P (2.51) 7.22 413 3719 1.21 17.4 70 3 5 D C P p G 6 P (2.52) 8.19 285 1799 2.33 41.4 56 3 N P p G 6 P (2.53) 8.39 380 312 4.69 72.9 64 4 C N P p G 6 P (2.54) 8.49 272 8101 1.01 45.4 22 P p G 6 P (2.55) 9.99 270 1850 0.79 31.9 25 4rBu P P G 6 P (2.56) 10.37 276 1156 1.21 58.3 21 T a b l e 2.8. Michaelis-Menten kinetic parameters for the hydrolysis of a series of aryl 6-phospho-|3-D-glucosides by BglT at 50 °C, pH 7.5. Error analyses for &ca, and KM values were performed using the program GraFit. The error was < 15% in each case. Chapter 2 - BglT from Thermotoga maritima 67 (b) 1\ •41 Slope = -0.08 o o — r 4 pKa ~1 1 1 r-6 8 ~I ' 10 12 Figure 2.20. Bronsted plots of the BglT-catalyzed cleavage of a series of aryl 6-phospho-p-D-glucopyranosides (2.13, 2.47—2.56) with the corresponding p/C, values for the leaving group phenol, (a) Bronsted plot for the correlation of log(/cCaI) vs p/C,; and (b) Bronsted plot for the correlation of logik^Jvs pKtl. The lack of dependence of rate on leaving group could result, if: 1) elimination of the C l - O l linkage is not rate-limiting; or 2) elimination is rate-limiting, but there is no significant buildup of negative charge on the glycosidic oxygen for these substrates at the transition state due to efficient proton donation via general acid catalysis Subsequent measurement of kinetic isotope effects will help to distinguish between these two possibilities. The kinetic parameters, kcat and KM, for Me(3G6P (2.58) were determined by the G6PDH coupled assay, monitoring for the oxidation of the G6P product and formation of NADPH (£NADPH = 6220 cm'lvl"1). The data obtained were fit to a standard Michaelis-Menten curve to calculate the kinetic parameters (Appendix 3). The kcat and KM values measured for the BglT-catalyzed hydrolysis of MefJG6P and C6'P (2.2), and the pKa values of the two leaving groups are summarized in Table 2.9. Chapter 2 - BglT from Thermotoga maritima 68 Substrate Leaving group pKA *c»t ( S 1 ) KM (uM) (s 'mM 1 ) C6'P 13.00 0.61 69 8.8 (2.4) MepGoP (2.58) 15.50 0.23 3.8 x 102 0.61 Table 2.9. Michaelis-Menten kinetic parameters for the hydrolysis of C6'P (2.2) and MepG6P (2.58) by BglT as measured by the G6PDH coupled assay. Error analyses for and KM values were performed using the program GraFit. The error was < 15% in each case. Comparison of the rates of BglT-catalyzed hydrolysis of aryl 6-phospho-p-D-glucopyranosides, C6'P, and MepG6P must be made with care since the leaving groups are of different chemical type (phenol versus alcohol) and the affinity of MepG6P is likely to be very different. Nonetheless, the logarithms of £ C a t and kcJKu for C6'P and MePG6P were calculated, and each was plotted against the pKa of the leaving group along with those obtained for the aryl 6-phospho-p-D-glucopyranosides. The Bronsted plots thereby produced are shown in Figure 2.21, and they were used to determine whether the rate-limiting steps for the hydrolysis of C6'P (2.2) and MepG6P (2.58) are the same as those for aryl 6-phospho-P-D-glucopyranosides. The plots of log (&cat) against p/^a and log (kcJKu) against pKa are very similar. Chapter 2 - BglT from Thermotoga maritima 69 (a) 6 4 2 el o 00 o -2-1 -4 -6 Slope = -0.03 - i — \ — i — | — i | i—| r | i | i 4 6 8 10 12 14 1 (b) 6 -4 -o Slope = -0.08 T 1 1 1 1 1 1 1 1 | I 1 I 2 4 6 8 10 12 14 1 Figure 2.21. Bronsted plots of the BglT-catalyzed cleavage of a series of aryl 6-phospho-p-D-glucopyranosides (2.13, 2.47—2.56), C6'P (2.2), and MepG6P (2.58) with the corresponding p/c:a values for the leaving group. (—): Linear fit of the data for aryl 6-phospho-p-D-glucopyranosides (2.13, 2.47—2.56); (—): Linear fit of the data for C6'P (2.2) and MepG6P (2.58) (a) Bronsted plot for the correlation of log(£c u,) vs p/C,; and (b) Bronsted plot for the correlation of l o g ^ / ^ M ) vs p/C,-Even with the inclusion of substrates with poorer leaving groups, glucose and methanol, the values of kcat and kcat/KM are only moderately sensitive to changes to the leaving group pKa. The kcat value for the most reactive substrate (3NP(3G6P, 2.53) is only approximately 20-fold higher than that of the least reactive substrate (MepG6P). The maximal value of kcaXIKu obtained is roughly 200-fold greater than that of MepG6P. However, a comparison based on these values can be misleading since substrate binding must be accounted for, and the glycosidases have been known to show preferential binding to aryl glycosides since the +1 subsite often contains an aromatic group that forms favorable binding interactions with the aglycone.129 Therefore, MepG6P is not expected to display the same binding properties as either aryl 6-phospho-p-D-glucopyranosides or C6'P. On inspection, the data points for C6'P and MePG6P appear to fall below slopes of the linear fit generated based on the data obtained for the aryl 6-phospho-P-D-glucopyranosides only, though, at least for C6'P, the values fall within the error limits of the line. This could indicate a change in the rate-limiting step with much poorer leaving groups. Therefore, the plots in Figure 2.21 include two different linear fits: one for the aryl 6-phospho-P-D-glucopyranosides only (2.13, 2.47—2.56) and one Chapter 2 - BglT from Thermotoga maritima 70 for C6'P (2.2) and MepG6P (2.58) only. With the limited number of substrates and the small change in the kcal and kcat /KM values despite the inclusion of the MefiG6P, it is not readily apparent as to whether the Bronsted plots presented in Figure 2.21 are actually biphasic and whether or not there is a change in the rate-limiting steps. Kinetic isotope effects, to be discussed in the next section, will help to determine if breakage of the C l -O l bond is rate-limiting for aryl 6-phospho-P-D-glucopyranosides and i f this step becomes rate-limiting for substrates without activated leaving groups, such as C6'P and MepG6P. Section 2.3.7. Kinetic Isotope Effects. Section 2.3.7.1. Synthesis of Substrates for Kinetic Isotope Effect Measurements. Kinetic isotope effect (KIE) measurements are an indispensable tool in the elucidation of reaction mechanisms. Isotopic substitution has no effect on the qualitative chemical reactivity of the substrate, but reaction rates may be affected based on the differences in the mass of the different isotopes. KIEs are therefore useful in providing information on bond-forming and bond-breaking steps. The theory behind KIEs is described in detail in Appendix 1. Primary (1°) KIEs in particular provide the most useful information on the rate-limiting steps involving bond breakage. Secondary (2°) KIEs present themselves when bonds to the isotopically substituted atoms are not broken but affect the reaction rates due to a change in hybridization at the site of substitution. Examination of the proposed BglT mechanism suggests that KIE measurements could be valuable in the dissection of the various steps of the reaction mechanism. Thus, primary KIEs should be detected for substrates with deuterium incorporated into C2 or C3 if either the C3 hydride abstraction or the C2 deprotonation steps are rate-limiting. If the departure of the leaving group is rate-limiting, a 2° KIE could be detected for substrates containing a deuterium atom at C l since the anomeric center would be undergoing rehybridization as part of a rate-limiting step. 4-Nitrophenyl l-[2H]-6-phospho-p-D-glucopyranoside (1 [ 2H]4NPpG6P or 2.79), 4-nitrophenyl 2-[2H]-6-phospho-p-D-glucopyranoside (2[2H]4NPpG6P or 2.80), and 4-nitrophenyl 3-[2H]-6-phospho-p-D-glucopyranoside (3[2H]4NPpG6P or 2.81) were synthesized for these studies as follows: 3-[2H]-D-Glucose (2.64) was synthesized from Chapter 2 - BglT from Thermotoga maritima 71 the commercially available l,2:5,6-di-0-isopropylidene-a-D-ribo-3-hexulofuranulose hydrate (2.59) (Scheme 2.7). The dihydrate 2.59 was reduced using sodium borodeuteride. Due to steric hindrance, the deuteride attacks C3 stereoselectively to form 2.60. The stereochemistry at C3 was inverted by the sequence of tosylation followed by SN2 displacement of the tosyl group with sodium benzoate to form 2.62. Finally, the benzoate and isopropylidene protecting groups were removed under basic and acid conditions respectively to produce 3-[2H]-D-glucopyranose (2.64). . I 2.59: R 1 , R 2 = O H 2.64 .'. * 2.60: R 1 = D, R 2 = O H " . 2 .61: R 1 = D, R 2 = O T s " ' 2 .62: R 1 = O B z , R 2 = D 1 V ' 2.63: R 1 = O H , R 2 = 3 Scheme 2.7. Synthesis of 3[2H]-D-glucose (2.64). (i) NaBD 4 , 95% EtOH; (ii) TsCI, pyridine; (iii) sodium benzoate, DMF; (iv) NaOMe/MeOH; (v) Amberlite IR-I20 (H + -resin), H 2 0 . l-[2H]-D-glucopyranose (2.65) and 2-[2H]-D-glucopyranose (2.66) were purchased from a commercial source. The deuterated glucose isotopomers (2.64—2.66) were acetylated in pyridine and reacted with HBr in acetic acid to form the respective a-bromides (2.70—2.72). Then, the 4-nitrophenyl group was introduced to 2.70—2.72 via the Koenigs-Knorr reaction to yield 2.73—2.75. Removal of the acetyl protecting groups via Zemplen deprotection followed by enzymatic phosphorylation at C6 by BglK according to Scheme 2.8 yielded the desired compounds 1 [ 2H]4NPpG6P (2.79), 2[ 2H]4NPpG6P (2.80), and 3[ 2H]4NPpG6P (2.81). Chapter 2 - BglT from Thermotoga maritima 72 2.65: R 2 , R 3 , R 4 = H, R 1 = D 2.66: R 1 , R 3 , R 4 = H, R 2 = D 2.64: R 1 , R 2 , R 4 = H, R 3 = D 2.67: R 2 , R 3 = H, R 1 = D, R 4 = A c 2.68: R 1 , R 3 = H, R 2 = D, R 4 = A c 2.69: R 1 , R 2 = H, R 3 = D, R 4 = A c 2.70: R 2 , R 3 = H, R 1 : 2.71: R 1 , R 3 = H, R 2 = 2.72: R 1 , R 2 = H, R 3 = D, R 4 = A c D, R 4 : D, R 4 : A c A c OP03Na2 2.79(1[2H]4NPPG6P): R 2 , R 3 = 2.80 (2[2H]4NPPG6P): R 1 , R 3 = 2.81 (3[2H]4NPPG6P): R 1 , R 2 = H, R H, R 2 H, R 3 2.73: R 2 , R3 = H, R 1 = D, R 4 = A c 2.74: R 1 , R 3 = H, R 2 = D, R 4 = A c 2.75: R 1 , R 2 = H, R 3 = D, R 4 = A c = D : D 2.76: R 2 , R 3 , R 4 = H, R 1 2.77: R 1 , R 2.78: R 1 , R R 4 : H, R z H, R 3 • D = D = D Scheme 2.8. Synthesis of deuterated chromogenic substrates (2.79—2.81) for KIE measurements, (i) A c 2 0 , pyridine; (ii) 33% HBr in acetic acid, A c 2 0 ; (iii) 4-nitrophenol, NaOH, acetone; (iv) NaOMe/MeOH; (v) BglK, ATP. As a probe of a possible change in the rate-limiting step for Me(3G6P (2.58), methyl l-[2H]-6-phospho-P-D-glucopyranoside (1 [ 2H]MepG6P or 2.85) was also synthesized (Scheme 2.9). Compound (2.82) was synthesized by the silver carbonate-catalyzed glycosidation of methanol using the deuterated oc-bromide 2.70. Subsequently, 2.82 was deprotected using the Zemplen method and was selectively phosphorylated at C6 using diphenyl chlorophosphate. Deprotection of 2.84 by catalytic hydrogenation afforded the desired product 1 [ 2H]MepG6P (2.85). Chapter 2 - BglT from Thermotoga maritima 73 .OAc O R 2 AcO' AcO-\ I OAc | Br 2.70 OR11 D 2.82: R 1 = R 2 = Ac 2.83: R 1 = R 2 = H 2.84: R 1 = H , R 2 = P O ( O P h ) 2 H O " H O -xOP0 3Na 2 \ OH ~ C H , 2.85 (1[2H]MepG6P) Scheme 2.9. Synthesis of l [ 2H]MepG6P (2.85). (i) A g , C 0 3 , MeOH, C H 2 C l 2 , 4 A molecular sieves; (ii) NaOMe/MeOH; (iii) PO(OPh>2CI, pyridine; (iv) Pt0 2 , H 2 ( g ) , MeOH. Section 2.3.7.2. Kinetic Isotope Effects - Results and Discussion. Initial rates of hydrolysis of protio- and deuterio-substituted-4-nitrophenyl 6-phospho-p-D-glucopyranosides (4NPpG6P, 1 [ 2H]4NPpG6P, 2[ 2H]4NPpG6P, 3[2H]4NPpG6P) were measured by the direct UV-vis assay, monitoring the release of 4-nitrophenol. Initial rates of hydrolysis of the protio- and deuterio-methyl 6-phospho-p-D-glucopyranosides (MepG6P, l[ 2H]MepG6P) were determined by the G6PDH coupled assay. Kinetic isotope effects upon V n i a x were determined via standard initial rate measurements using saturating substrate concentrations. For each compound, a substrate concentration of > 10 KM was employed ensuring the measurement of V 1 1 i a x . Initial rates were measured in alternation for the protio and deuterio substrates. A set of values for (^cat)H/(/<cat)D was calculated from the data by dividing the rate for the protio substrate by the rate for the deuterio substrate in each case. Kinetic isotope effects upon kCJKM were measured by the substrate depletion method. The direct UV-vis assay was used for monitoring the substrate depletion of protio- and deuterio-4-nitrophenyl 6-phospho-p-D-glucopyranosides (4NPpG6P, l[ 2H]4NPpG6P, 2[ 2H]4NPpG6P, 3[2H]4NPpG6P) at substrate concentrations of < 1/5 KM, while the G6PDH coupled assay was used for the protio- and deuterio-substituted-methyl 6-phospho-p-D-glucopyranosides (MepG6P, 1 [ 2H]MePG6P). Values of (A:cat//CM)H/('tcat/^M)D for each isotopically substituted substrate pair were calculated by dividing the first-order rate constant for the protio substrate by the first-order rate constant for the deuterio substrate. In all cases, values presented represent the average of Chapter 2 - BglT from Thermotoga maritima 7 4 at least 7 pairs of measurements. The standard deviations for each set of data were calculated, and the error analysis for (^catWC&caOD and (A c a t /A^M )H / ( £ca t /A^ i )D were determined from the data. Al l KIEs upon &ca, and J W & i u are summarized in Table 2.10. Substrate KIE l| 2H]4NPflG6P (2.79) 1.00 ± 0 . 0 1 1.01 ± 0 . 0 4 None 2| 2H|4NPPG6P (2.80) 1.84 ± 0 . 0 2 2.03 ± 0 . 0 1 1° 3| 2H]4NPPG6P (2.81) 1.63 ± 0 . 0 1 1.91 ± 0 . 0 3 l I 2 H|MefiG6P (2.85) 1.25 ± 0.04 1.16 + 0.06 2° Table 2.10. Summary o f KlEs determined for BglT. Small but significant primary KIEs were measured for both 2[2H]4NPpG6P (2.80) and 3[2H]4NPpG6P (2.81) indicating that both the C2-H2 and C3-H3 bond cleavages are rate-limiting steps in the enzyme mechanism. The 1° KIEs are smaller than the theoretical maximum value of 6-7 but well within the range of those measured for similar reactions."4 One contributing factor to the small KIE is that there are 2 partially rate-limiting steps and therefore neither KIE is fully expressed. Additionally, if Tyr241 plays the proposed dual role of C2 deprotonation and of general acid catalyst, the nonlinear arrangement of the catalytic base, C2 and H2 can result in a reduction of the 10 KIE. 1 2 ' 4 For example, the intramolecular proton abstraction during pyrolytic elimination of amine oxides has been reported to result in a small 1° KIE due to this nonlinear arrangement of the appropriate atoms. •"' The small 1° KIE could also result from an asymmetric transition state, as often occurs in highly endothermic or exothermic reactions.157 No 2° KIE is observed for l[2H]4NPpG6P (2.79), which indicates that the elimination step is fast and not rate-limiting for aryl 6-phospho-p-D-glucopyranosides. Because a 1° KIE is observed for 2[2H]4NPpG6P (2.80) and no 2° KIE is observed for l[2H]4NPpG6P (2.79), the KIEs rule out both E2 and E l mechanisms, but are consistent with an E l c b mechanism in which the deprotonation step is rate-limiting followed by the rapid departure of the leaving group. The KIEs measured for the 4-nitrophenyl 6-phospho-P-D-glucopyranosides indicate that, subsequent to the slow C3 Chapter 2 - BglT from Thermotoga maritima 75 hydride abstraction, a second slow, partially rate-limiting step involving C2 deprotonation occurs, followed by rapid elimination of the aglycone. This suggests that BglT utilizes an ElCb-type mechanism for hydrolysis of glycosidic linkages. It is important to note that these KIEs were all obtained using 4-nitrophenyl 6-phospho-p-D-glucopyranoside substrates containing an activated leaving group, which could make the elimination step more facile than for the natural substrate C6'P. The Bronsted studies presented in Section 2.3.6 showed that MePG6P was cleaved somewhat more slowly than 4NPPG6P possibly indicating that the elimination step becomes rate-limiting. Indeed, as presented in Table 2.10, a 2° KIE was determined for the hydrolysis of l[ 2H]MepG6P (2.85) indicating that cleavage of the C l - O l linkage in MepG6P is rate-limiting. This indicates that the Bronsted plots shown in Figure 2.21 are actually slightly biphasic with a change in the rate-limiting step from C2-deprotonation to cleavage of the C l - O l linkage as the leaving group ability decreases from that of the various phenols in the aryl 6-phospho-p-D-glucopyranosides to the methyl group in MepG6P. Section 2.3.7.3. Kinetic Isotope Effects in D 2 O Buffer. Section 2.3.7.3.1. pD-Dependence. The measurement of small but significant primary kinetic isotope effects on the hydrolysis of 3[ 2H]4NPpG6P (2.81) and 2[ 2H]4NPpG6P (2.80) provided compelling evidence for both the oxidation at C3 and for the proton abstraction at C2, as well as suggesting that both steps were partially rate-limiting. However, the possibility remained that the true deuterium kinetic isotope effect for 2[ 2H]4NPPG6P was much larger, but that its value had been suppressed by exchange of the C2 deuterium for solvent H 2 0 during the measurement since it was shown that the C2 proton efficiently exchanges with solvent from the solvent deuterium exchange experiments (Section 2.3.2). This would be a significant concern if the reprotonation of the anion generated at C2 occurs at a greater rate than that of the elimination step. The lack of solvent deuterium incorporation into C2 of the unreacted substrate 4NPPG6P (2.13) suggests that no such exchange occurs (Section 2.3.2). However, it is not conclusive proof since the observation of exchange Chapter 2 - BglT from Thermotoga maritima 76 would require that the reduction at C3 also occur more rapidly than elimination, which is not necessarily the case. Measurement of the kinetic isotope effect for 2[2H]4NP(3G6P (2.80) in D 2 0 should provide a more stringent test since any exchange in that case would leave the deuterium atom in place, and the full KIE should be observed. However, it was first necessary to determine whether ionizations in the enzyme, thus possibly the pH o p t , are significantly different in D 2 O buffer vs H 2 0 buffer. Since the pH-dependence of £ C at and of kcJKu were found to be very similar (see Section 2.3.4), only the pD-dependence of kcJKu was investigated to determine the p L o p t for kinetic analysis. As discussed in Section 2.3.4, because precipitates form in the assay solution under basic conditions, no data points could be obtained above pH 10. The results in Figure 2.22 clearly show a shift in the acid limb of the pH profile, and a similar shift in the basic limb—though the basic limb is now less well-defined within the pH stability range of the enzyme. This results in a small shift in the pD o p t from pH o pt (Table 2.11). Such a shift of approximately 0.6 unit in the two apparent pKa values is consistent with the typical solvent isotope effects observed for the ionization of a number of general acids, such as citric acid, salicylic acid, 2-mercaptoethanol, boric acid, formic acid, and 4-nitrophenol.158 However, BglT is reasonably active at pL 8.1 in both cases, thereby ensuring that kinetic isotope effect measurements at this pL will be meaningful. 4 6 8 1 0 pl-Figure 2.22. pL-Dependence of the kcM/KM for BglT. The empty circles represent the data obtained in H 2 0, and the filled squares present the data obtained in D 2 Q. Chapter 2 - BglT from Thermotoga maritima 77 P#al P*a2 pH0pt pH-activity profile 7.1 ± 0 . 1 9.3 ± 0 . 1 8.2 pD-activity profile 7.6 ± 0 . 1 9.9 ± 0 . 1 8.8 Table 2.11. Summary of apparent pK,t values and pH o p , values obtained from the pH- and pD-activity profiles. Section 2.3.7.3.2. Solvent Kinetic Isotope Effect. The kinetic parameters for the BglT-catalyzed hydrolysis of 4NP(3G6P were determined in D 2 O buffer at pD 8.1 using the direct UV-vis assay. This pD was chosen because, as noted in Section 2 .3.7.3.1, at this value rates are reasonably close to optimal in both H 2 O and D 2 O (Figure 2.22). Michaelis-Menten kinetic behavior was observed, and the kcat and KM values at pD 8.1 in D 2 O and in HbO are compared in Table 2.12. The values obtained were used to determine the solvent KIE (Table 2.13). Solvent KIEs are usually observed when a proton transfer from solvent or from a solvent-exchangeable group occurs in the rate-determining step. It is not surprising that a small solvent KIE was determined as several proton transfers are occurring in the proposed mechanism. The key purpose of this part of the study was to determine whether the KIE measured for 2[ 2 H]4NP(3G6P in Section 2.3.7.2 was diminished due to exchange of the C 2 deuterium for solvent H 2 0 . Thus, the solvent KIE itself was not of great significance to the current study. The small solvent deuterium KIE measured indicates that the hydrolysis reaction is not greatly altered in D 2 O , thereby allowing a reasonable comparison of the magnitude of the KIE obtained in H 2 O and that obtained in D 2 O . Aca,(s"') KM (M-M) Acar/Kv, ( s 'mM 1 ) 0.91 40 23 D 2 0 0.72 70 10 Table 2.12. Summary of Michaelis-Menten kinetic parameters obtained using 4NPPG6P (2.13) in PbO and DoO. Chapter 2 - BglT from Thermotoga maritima 78 (Aoal)H 2()/(A:ial)l) 20 1.3 (A'«a</A^M)H;<)/(Acal//ifM)l),() 2.2 Table 2.13. Solvent KIEs measured for the BglT-catalyzed hydrolysis of 4NP(3G6P (2.13). Section 2.3.7.3.3. Determination of Primary Kinetic Isotope Effects for 2[ 2H]4NPpG6P (2.80) in D 2 0 Buffer. The KIE on kcat and on kcJKM for 2[2H]4NPpG6P (2.80) was determined in D 2 0 buffer exactly as described in H 2 0 buffer (see Section 2.3.7.2). Al l KIEs upon kcat and &cat/^ Mvi are summarized in Table 2.14. KIE conditions (Acai Wreath) (A'tal/A VI )ll/( A\a|/A vi )l> 2[2H]4NPPG6P (2.80) 1.84 ± 0 . 0 2 2 . 0 3 ± 0 . 0 1 H 2 0 buffer 2[2H]4NPpG6P (2.80) 1.62 ± 0 . 0 4 1.60 ± 0 . 0 9 D 2 0 buffer Table 2.14. Comparison of KIEs for 2[ 2H]4NPpG6P (2.79) in H , 0 and D 2 0 . The deuterium kinetic isotope effects on kcat and kcJKM for 2[2H]4NPpG6P (2.80) in D 2 0 were slightly smaller than those measured in H 2 0 (Table 2.14). The slight decrease is probably a consequence of other proton transfer events becoming slightly rate-limiting in D 2 0 as evidenced by the solvent KIE (Section 2.3.7.3.2), but this does not obscure the analysis of the KIE for 2[ 2H]4NPpG6P. This confirms that the deuterium KIE observed for 2[ 2H]4NPpG6P is indeed a small 1° KIE. The values for the 1° KIE obtained in H 2 0 were not suppressed by exchange of the C2 deuterium for solvent H 2 0 . Clearly, no significant reprotonation of the anion occurs during catalysis, presumably because cleavage of the C l - O l linkage occurs more rapidly than reprotonation of the enediolate Chapter 2 - BglT from Thermotoga maritima 7 9 intermediate. The KIEs obtained for 2[2H]4NPpG6P are indeed smaller than the theoretical maximum of 6 or 7, and this may result because: 1) cleavage of the C2-H2 linkage is only partially rate-limiting; 2) non-linear arrangement of the Tyr241 oxygen, C2 and H2; 3) an asymmetric transition state, which more closely resembles the structure of the reactants or of the products. Section 2.3.8. Proposed E 1 c b Mechanism. In summary, the solvent deuterium exchange measurements, x-ray crystallographic data, Bronsted analysis and KIEs support the proposed mechanism shown in Figure 2.23. The on-board N A D + effects a transient and partially rate-limiting oxidation of the substrate at C3, thereby acidifying the C2 proton. Partially rate-limiting proton abstraction follows, generating a metal-stabilized enediolate, which rapidly undergoes elimination to generate the bound a,(3-unsaturated ketone intermediate. The reaction is completed by addition of water to the same face of the Michael acceptor as that of the departing aglycone. Finally, the on-board N A D H cofactor reduces the C3 carbonyl, and the overall hydrolysis product G6P is formed with net retention of configuration at the substrate anomeric center. Thus, BglT employs a stepwise E l C b mechanism, involving anionic transition states, in stark contrast to all other currently characterized glycosidase families, which catalyze hydrolysis through cationic, oxocarbenium ion-like transition states. Chapter 2 - BglT from Thermotoga maritima 80 Figure 2.23. Proposed E\tb mechanism of BglT. Section 2.3.9. Potential Inhibitors of B g l T . Enzyme inhibitors often prove to be useful probes of enzyme mechanisms, particularly when used in conjunction with x-ray crystallographic analyses. Therefore, several classes of compounds were prepared and their inhibitory properties against BglT were examined. Strategies towards inhibitor design were developed with the enzyme structure and the proposed mechanism in mind. Section 2.3.9.1. Deoxy-6-phospho-|3-D-glucopyranosides as Inhibitors of Bg lT . A number of deoxy-6-phospho-p-D-glucopyranosides could potentially inhibit BglT activity. For example, 4-nitrophenyl 3-deoxy-6-phospho-p-D-glucopyranoside (4NP3deoxypG6P or 2.94) is not expected to be hydrolyzed by BglT, because oxidation Chapter 2 — BglT from Thermotoga maritima 81 at C3 is now impossible, and the C3 keto-intermediate cannot be formed. Thus, if it binds to the enzyme it will not be converted, and should act as an inhibitor. 4NP3deoxypG6P (2.94) was synthesized from diacetone glucose (2.86) as shown in Scheme 2.10. Compound 2.86 was reacted with phenyl thiochloroformate to give the xanthate derivative (2.87) and then deoxygenated using tributyl tin hydride to produce 2.88 using a Barton—McCombie deoxygenation.159 The diisopropylidene protecting groups in 2.88 were then removed by acid-catalyzed hydrolysis to form 3-deoxy-D-glucopyranose (2.89). Compound 2.89 was acetylated, converted to the a-bromide (2.91), and 4-nitrophenyl 2,4,6-tri-0-acetyl-3-deoxy-p-D-glucopyranoside (2.92) was synthesized via the Koenigs-Knorr reaction. Deprotection via the Zemplen method1 2 6 yielded 4-nitrophenyl 3-deoxy-p-D-glucopyranoside (2.93). In contrast to what was found with other p-D-glucopyranosides used in this study thus far, the kinase BglK does not phosphorylate 2.93. Consequently, the chemical phosphorylation approach with POCl 3 was utilized to synthesize 4NP3deoxypG6P (2.94). OR v Ad ,OAc 2.86: R = OH 2.89: R = H 2.91 IV 2:87: R - OC(=S)OPh 2:90: R = Ac 11 * 2:88: R = H VI , O P 0 3 N a 2 ,0R 2.94 (4NP3deoxyHG6P) 2.92: R = Ac VII 2:93: R = H Scheme 2.10. Synthesis of 4N'P3deoxypG6P (2.94). (i) D M A P , CH 2 C1 2 , PhOC(=S)Cl; (ii) Bu 3SnH, AIBN, toluene, reflux; (iii) H 3 0 + / H 2 0 ; (iv) Ac 2 G\ pyridine; (v) 33% HBr in HOAc, A c 2 0 ; (vi) NaOH, acetone, 4-nitrophenol; (vii) NaOMe/MeOH; (viii) POCI 3 , PO(OCH;,)3, H 2 Q. Chapter 2 - BglT from Thermotoga maritima 82 The use of a 2-deoxy-6-phospho-P-D-glucopyranoside as an inhibitor is based upon the fact that, in the crystal structure, the C2 hydroxyl group of the G6P product is chelated to the octahedral M n 2 + . Hence, the absence of a C2 hydroxyl group in 4-nitrophenyl 2-deoxy-6-phospho-P-D-glucopyranoside (4NP2deoxypG6P or 2 . 9 6 ) could result in incorrect binding of the essential metal ion. This could affect the oxidation step since the deprotonation of the C3 hydroxyl is mediated by a metal chelated hydroxide. Likewise, C2 deprotonation may only occur if the enediolate intermediate is fully stabilized by a bound metal ion. Therefore, if 2 . 9 6 binds to the enzyme, it should not be cleaved and may be a useful inhibitor. Thus, 4-nitrophenyl 2-deoxy-p-D-glucopyranoside ( 2 . 9 5 ) , which had been previously synthesized by Dr. M . N . Namchuk, 1 6 0 was phosphorylated using BglK according to Scheme 2.11 to yield the desired inhibitor 4NP2deoxypG6P. 2.95 2.96 (4NP2deoxyPG6P) Scheme 2.11. Enzymatic synthesis of 4NP2deoxy PG6P (2.96). (i) BglK, ATP. S e c t i o n 2 . 3 . 9 . 2 . 6 - P h o s p h o - o c - D - g l u c o p y r a n o s i d e s as I n h i b i t o r s o f B g l T . The inhibitory properties of methyl 6-phospho-a-D-glucopyranoside ( 2 .98 ) were also investigated. As discussed earlier, a 6-phospho-a-glucosidase, MalH, 5 0 from GH4 hydrolyzes both a- and p-glycosides containing activated leaving groups such as 4-nitrophenol.50 BglT, on the other hand, catalyzes the hydrolysis of only 4NPpG6P ( 2 . 1 3 ) , and not of the oc-anomer 4NPaG6P ( 2 . 17 ) . Inspection of the x-ray crystal structure reveals that the active site is too small to allow aryl 6-phospho-D-glucosides of the oc-anomeric configuration to bind productively. The testing of MeaG6P ( 2 . 9 8 ) as a potential substrate provides a simple approach to assess the ability of BglT to bind any type of 6-phospho-a-D-glycopyranoside. Because the aglycone of MeaG6P is considerably smaller than that of 4NPaG6P ( 2 . 1 7 ) , MeaG6P ( 2 . 9 8 ) has a greater Chapter 2 - BglT from Thermotoga maritima 83 probability of binding to the active site of BglT. Therefore, MeaG6P (2.98) was synthesized via the selective phosphorylation of the primary C6 hydroxyl of MeocGlc (2.19) using diphenyl chlorophosphate and subsequently deprotected via catalytic hydrogenation (Scheme 2.12). OH OPO(OPh)2 0P03Na2 H C r ^ \ - ^ - - " Q »- H C ^ V ^ - - - " ^ . H C r ^V - ^ - ^ - Q , HO-\*»^r--A H O - A ^ - ^ O i . HO-\^«--X»>A OMe OMe OMe 2.19(MeaGlc) 2.97 2.98 (MeaG6P) Scheme 2.12. Synthesis of MeaG6P (2.98). (i) (PhO) 2POCl, pyridine; (ii) PtQ 2, H 2 , MeOH. Section 2.3.9.3. Common Glycosidase Inhibitors. Section 2.3.9.3.1. 2-Deoxy-2-fluoro-6-phospho-pVD-glucosides as Inhibitors of BglT. Activated fluorosugars have been used extensively for inactivating retaining glycosidases. 2 3" 3 3 , 9 0 , 1 6 1* 1 6 2 Because "classical" glycosidases utilize nucleophilic displacement mechanisms proceeding via oxocarbenium ion-like transition states, the incorporation of an electronegative fluorine atom adjacent to the anomeric center, where substantial positive charge develops, destabilizes the transition state, thereby slowing down both the formation of the glycosyl-enzyme intermediate (glycosylation) and its hydrolysis. By attaching activated leaving groups at the anomeric center, the inhibitor undergoes the glycosylation process at a reasonable rate, but the second step is sufficiently slowed such that the glycosyl-enzyme intermediate accumulates and the enzyme is inhibited. Because the proposed mechanism of BglT does not involve a cationic transition state but rather anionic intermediates, glycosides with an electron-withdrawing fluorine at C2 might not inhibit BglT activity. In fact, C2 deprotonation may be enhanced due to stabilization of the "enediolate" intermediate. However, they also may not even bind to the enzyme since the 2-hydroxyl normally coordinates the M n 2 + . As shown in Scheme 2.13, 4-nitrophenyl 2-deoxy-2-fluoro-6-phospho-(3-D-glucopyranoside (4NP2FpG6P or 2.100) was synthesized via the enzymatic phosphorylation of 4-nitrophenyl 2-deoxy-2-fluoro-P-D-glucopyranoside (2.99) (previously prepared Dr. I. P. Street163), and the inhibitory properties were tested. C h a p t e r 2 - B g l T f r o m Thermotoga maritima 84 2.99 2.100 (4NP2FpC6P) Scheme 2.13. E n z y m a t i c syn thes is o f 4 N P 2 F p G 6 P (2.100). ( i ) B g l K , A T P . Section 2.3.9.3.2. Thioglycosides as Potential Inhibitors. Thioglycosides, in which the glycosidic oxygen has been replaced by a sulfur atom, have proved to be stable analogues of the parent glycosides and have been extensively employed in number of insightful structural studies through analyses of the glycosidase/thioglycoside complexes. 1 6 4 , 1 6 5 , 6 Since normal glycosidases are known to effect hydrolysis by acid/base-catalyzed mechanisms involving oxocarbenium ion-like transition states, ' the resistance of the thioglycosidic bond to cleavage has been ascribed to the lower proton affinity of sulfur over that of oxygen, resulting in inefficient general acid catalysis to the departing aglycone. 1 6 X 1 7 0 Even though BglT utilizes an elimination mechanism, it was of interest to determine whether the enzyme is inhibited by thioglycosides. Thus 4-nitrophenyl 6-phospho-p-D-thioglucopyranoside (5,-4NP(3G6P or 2.102) was synthesized via the BglK-catalyzed phosphorylation of 4-nitrophenyl thio-p-D-glucopyranoside (2.1.01) (previously synthesized by Dr. M . Jahn1 7 1) as shown in Scheme 2.14. e T h e one e x c e p t i o n is a s p e c i a l i z e d g r o u p o f S - g l y c o s i d a s e s o f p lan t o r i g i n , c a l l e d the m y r o s i n a s e s ( E . C . 3.2.3.1), w h i c h s p e c i f i c a l l y h y d r o l y z e g l u c o s i n o l a t e subst ra tes, a n i o n i c 1 - th io -P -g lucosides. B y s e q u e n c e a l i g n m e n t , the m y r o s i n a s e s are assoc ia ted w i t h the F a m i l y 1 g l y c o s i d a s e s , w h i c h ca ta l yze the h y d r o l y s i s o f p - ( 9 -g l ycos ides w i t h re tent ion o f the substrate a n o m e r i c c o n f i g u r a t i o n . W h i l e F a m i l y 1 g l y c o s i d a s e s con ta i n a c o n s e r v e d g lu tamate as the ca ta ly t i c n u c l e o p h i l e , in the m y r o s i n a s e s the g l u t a m i c a c i d that se rves as the ac id /base cata lys t is rep laced by a g l u t a m i n e res idue . It has been p r o p o s e d that the m y r o s i n a s e s are ab le to ca ta l yze the c leavage o f g l u c o s i n o l a t e , because the subst rates c o n t a i n i n h e r e n t l y g o o d l e a v i n g g r o u p s thus d o not requ i re genera l a c i d ass is tance for l e a v i n g g r o u p depar ture . A b o u n d ascorbate a n i o n appears to f u n c t i o n as the genera l base cata lyst for h y d r o l y s i s o f the g l y c o s y l - e n z y m e in te rmed ia te . A s i m i l a r e x p l a n a t i o n is g i v e n for the G H 8 4 h u m a n O - G l c N A c a s e , w h i c h is p r o p o s e d to ca ta l yze the h y d r o l y s i s o f ac t i va ted t h i o g l y c o s i d e s via a ve ry d i s s o c i a t i v e t rans i t ion state and w i t h o u t genera l a c i d ca ta l ys i s to pro tonate the th io la te l e a v i n g g r o u p / 5 ' 1 6 6 Chapter 2 - BglT from Thermotoga maritima S 5 ^ "N02 2.101 2.102 (S-4NPpG6P) Scheme 2.14. Enzymatic synthesis of S-4NPpG6P (2.102). (i) BglK, ATP. Section 2.3.9.4. Inhibition Studies - Results and Discussion. Each compound was tested as substrate first using standard assay conditions. The enzyme concentration was increased to 10-fold more than that necessary for hydrolysis of 4NP(3G6P, and no hydrolysis was observed. Approximate K\ values for the potential inhibitors were determined by measuring the reduction in the rate of BglT-catalyzed hydrolysis of a fixed concentration of the chromogenic substrate 4NPpG6P (2.13), in the presence of varying concentrations of inhibitor. The experiments were repeated at different concentrations of 4NPJ3G6P, and the data were graphed as a Dixon plot (1/v v.v [competitive inhibitor]). The lines from the different concentrations of 4NP|3G6P intersected in the Dixon plots (Appendix 5). Additionally, at the same point as a horizontal line drawn through 1/Fmax, the intersection occurs at an inhibitor concentration equal to -K, The results of the inhibition studies are summarized in Table 2.15 Inhibitor Approximate ( p M ) Type of inhibition 4NP3deoxypG6P (2.94) 200 Competitive 4NP2deoxyPG6P (2.96) 50 Competitive 4 N P 2 F p G 6 P (2.100) 6 Competitive 5 - 4 N P p G 6 P (2.102) — ~ M e a G 6 P (2.09) 3000 Non-competitive Table 2.15. Approximate K, values of the potential inhibitors of BglT. The results from the competitive inhibition studies confirm that 4NP3deoxypG6P (2.94), 4NP2deoxypG6P (2.96), and 4NP2FpG6P (2.100) bind to the active site of BglT. As expected, BglT does not hydrolyze 4NP3deoxypG6P, but binds it considerably more Chapter 2 - BglT from Thermotoga maritima 86 weakly than its parent substrate, presumably because the hydroxyl group at C3 is important for binding to the M n 2 + . Likewise, neither 4NP2deoxypG6P nor 4NP2F(3G6P were hydrolyzed by BglT, presumably for the reasons discussed earlier. Interestingly, however, both function as good inhibitors with the 4NP2deoxy(3G6P (2.96) binding with .similar affinity to the parent 4NPpG6P (2.13) and 4NP2FpG6P (2.100) binding 6-fold more tightly. Presumably, the 2-hydroxyl group is not very important for ground state binding, but crucial for binding at the transition state. MeaG6P (2.98), however, was a non-competitive inhibitor of BglT. A noncompetitive inhibitor binds to the enzyme and the enzyme-substrate complex, usually with equal affinity. In many cases, non-competitive inhibitors do not bind to the enzyme active site. This observation is consistent with the x-ray crystallographic data, which revealed that the +1 subsite of BglT cannot accommodate substrates of an oc-anomeric configuration. Whether the mechanism utilized by BglT can in fact cleave a-glycosidic linkages similar to M a l H 5 0 cannot be fully investigated since the restricted enzyme active site does not allow for binding of substrates of a-anomeric configuration even with a small aglycone, such as a methyl group in MeaG6P. Surprisingly, ,S-4NPpG6P (2.102) was found to be hydrolyzed by BglT with similar kinetic parameters to those of its oxygen-containing counterpart, 4NPpG6P (2.13). This indicates that thioglycosides might, in general, function as substrates for this enzyme. This is of interest as it was recently demonstrated that chondroitin A C lyase, which also operates via an elimination mechanism, also cleaves thioglycosides,86'172 suggesting that thioglycosides can be cleaved if the enzyme uses an anionic elimination mechanism, but not i f it employs a standard oxocarbenium ion-like mechanism. However, some caution is necessary before drawing a broad conclusion since the use of an activated aryl leaving group could be misleading in this case. Previous studies with a "normal" family 1 p-glucosidase, Abg, " and a GH84 human 0-GlcNAcase 1 6 6 have shown that activated aryl thioglycosides are cleaved reasonably efficiently. Therefore, proper investigation of the hydrolysis of thioglycosides required further investigation with thioglycoside analogues of the natural substrate, which will be discussed in Section 4.3. Chapter 2 - BglT from Thermotoga maritima 87 Section 2.3.9.5. Potential Solvent Isotopic Exchange with Competitive Inhibitors. Possible solvent deuterium incorporation into the competitive inhibitors (4NP3deoxypG6P, 4NP2deoxypG6P and 4NP2FpG6P) was also investigated. The inhibitors and the chromogenic substrate (4NPpG6P) were each individually incubated in D 2 0 buffer with and without BglT. Potential solvent deuterium incorporation into 4NP2deoxypG6P and 4NP2FpG6P was investigated by 'H NMR (Figure 2.24) after removal of enzyme and M n 2 + (exactly as described in Section 2.3.2.2). Due to the small amount of 4NP3deoxypG6P available, potential solvent deuterium incorporation was investigated by mass spectrometry. No solvent deuterium incorporation in 4NP3deoxypG6P, 4NP2deoxypG6P, or 4NP2FpG6P was detected, while BglT-catalyzed the complete hydrolysis of 4NPpG6P with deuterium incorporation at C2 under the same conditions. This result is exactly as expected for 4NP3deoxypG6P since a ketone cannot be formed at C3, thus the C2 proton does not become acidic. For 4NP2deoxypG6P and 4NP2FPG6P, the lack of deuterium incorporation into C2 suggests that the initial oxidation at C3, and hence C2 deprotonation does not occur, even though the inhibitors bind to the active site. Deprotonation may not be possible either due to improper binding of the M n 2 + or simply due to the lack of transition state stabilization that was previously afforded by the enediolate intermediate. In summary, the competitive inhibitors are able to bind to the active site of BglT, but the enzyme does not carry out the essential C2-deprotonation step in the proposed El Cb mechanism. Chapter 2 - BglT from Thermotoga maritima 88 i i i i i i " i i i " ~ i i i i i i i 5.50 5.00 4.50 4.00 ppm (f1) Figure 2.24. 'H N M R analysis of potential solvent deuterium incorporation into the competitive inhibitors 4NP2deoxypG6P (2.96) and 4NP2F0G6P (2.100). (a) Schematic of the solvent deuterium incorporation into 4NP2deoxy(3G6P (2.96): (i) BglT, D 2 0 buffer; and 'H NMR spectra of: (b) 4NP2deoxyPG6P (2.96) incubated in D 2 0 buffer with BglT; (c) 4NP2deoxypG6P (2.96) incubated in D 2 0 buffer without BglT; (d) Schematic of the solvent deuterium incorporation into 4NP2FpG6P (2.100): (i) BglT, D 2 0 buffer; and 'H N M R spectra of: (e) 4NP2F|3G6P (2.96) incubated in D 2 0 buffer with BglT; (t) 4NP2FpG6P (2.96) incubated in D 2 0 buffer without BglT. Chapter 2 - BglT from Thermotoga maritima 89 Section 2.4. Mutagenesis Studies. Section 2.4.1. Previous Mutagenesis Studies. The identities of several important amino acid residues have been inferred from kinetic analyses of mutants and from structural analyses with a number of GH4 members. These include the amino acid residues involved in binding the N A D + cofactor in the Rossman fold, 4 3 a strictly conserved Cys that is chelated to the divalent metal 4 0 ' 6 8 , 7 0 and an Asp residue adjacent to the active site Cys. 4 0 Other catalytic residues are suggested based largely on structural analyses. Currently, the structures of GlvA and BglT provide the most useful information, because both are catalytically active and are complexed with all the necessary cofactors as well as the reaction product.68"70 The other structures for GH4 enzymes do not contain either the dinucleotide cofactor or the metal cation or both, leading to concerns about identification of residues. More structural data and kinetic analyses with mutants will be necessary before catalytic residues can be more firmly identified. However, from the preliminary structural analyses, Asp260 has been proposed to act as the catalytic base in C2 deprotonation in AglA based on the available structure and our proposed mechanism, 4 4 , 6 9 ' 7 0 whereas a Tyr likely plays the same role in the 6-phospho-glycosidases (such as GlvA and BglT). 6 8 " 7 0 The preference for Tyr in the 6-phospho-glycosidases was suggested to be due to the need to circumvent the potential electrostatic repulsion between the C6 phosphate group of the substrate and an Asp . 6 9 , 7 0 Indeed, other glycosidases also seem to avoid electrostatic repulsions in similar manners, with sialidases and /ran.v-sialidases employing a Tyr nucleophile instead of an Asp or Glu, presumably to minimize interactions with the substrate carboxylate functionality.3 3 , 1 7 4 The role of the general acid involved in protonation of the departing aglycone oxygen has been assigned to an Asp (in GlvA) or a His (in AglA) residue. Interestingly, for BglT, which undergoes a .^-elimination, the Tyr that is assigned the function of general base is positioned a mere 3.4 A away from the glycosidic oxygen, and so the Tyr likely bears the dual role of general base in C2 deprotonation as well as general acid in assisting C l - O l bond cleavage/formation.69 This may seem unlikely at first consideration, but the discrete El Cb mechanism makes this role attractive. Furthermore, the pH-activity profile revealed an important ionizable group with a pKa Chapter 2 - BglT from Thermotoga maritima value of around 7, which could correspond to the Tyr residue. Therefore, site-directed mutagenesis of Tyr241 for further kinetic analysis is presented in the next sections. Section 2.4.2. Design of BglT Mutants and C D Spectroscopic Analysis. Kinetic analysis of mutants in which key amino acid residues have been replaced often provides the most direct approach for examining the role of these residues in catalysis. Therefore, site-directed mutagenesis of Tyr241 to its most closely related residue (Phe) and to Ala was carried out for an extensive analysis of the role of the proposed catalytic base. The two mutants BglT Y241F and BglT Y241A were cloned and overexpressed in E. coli, and kinetic analysis, using the tools employed for investigating the BglT WT mechanism, was performed to more fully establish the role of Tyr241 in the catalytic mechanism. Circular dichroism (CD) spectra of BglT WT, BglT Y241F, and BglT Y241A were measured to ensure that the mutants were correctly folded. -40 J Wavelength (nm) Figure 2.25. C D spectra of BglT WT, BglT Y241F, and BglT Y241A at 50 °C. Each spectra was collected after the enzyme samples had been equilibrated at 50 °C for 20 minutes. Chapter 2 — BglT from Thermotoga maritima 91 As shown in Figure 2.25, the CD spectra of BglT WT, BglT Y241F and BglT Y241A collected at 50 °C are very similar, indicating that the two mutants are conformationally stable at the assay temperature. This indicates that any loss of activity measured for BglT Y241A and BglT Y241F relative to BglT WT is likely not due to misfolding or instability of the mutant enzymes. Section 2.4.3. Dinucleotide Cofactor. Section 2.4.3.1. Determination of the Kd Value for NAD+. As was done with BglT WT, BglT Y241F and BglT Y241A were dialyzed against large volumes of buffer to remove any bound N A D + . Upon dialysis, BglT Y241F was found to be completely inactive in the absence of N A D + , full activity (measuring the hydrolysis of 4NPpG6P (2.13) via the direct spectrophotometric assay) could be restored to the mutant in a saturable fashion by titration with N A D + . A Kd value of 383 nM (Table 2.16) was obtained based on a direct fit of the data to a hyperbolic equation (Appendix 6). On the other hand, even after extensive dialysis, the N A D + remained tightly bound to BglT Y241A. Since activity did not decrease further, a ligand binding curve could not be generated for this mutant. This was shown to be the case by adding sodium borohydride to a sample of BglT Y241A to which no additional N A D + had been included. The UV-vis absorbance spectra, shown in Figure 2.26, revealed a peak at 340 nm as expected for bound N A D H , and the difference in absorbance from that of the untreated enzyme sample corresponded to the presence of N A D H in equimolar concentration to the enzyme sample. Chapter 2 - BglT from Thermotoga maritima 9 2 0 - | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — I 250 300 350 400 Wavelength (nm) Figure 2.26. Absorbance spectra of BglT Y241A in its oxidized (NAD*) and reduced (NADH) forms. Absorbance spectra of 10 uM untreated BglT Y241A (—), 10 uM BglT Y241A incubated with 10 mM sodium borohydride (—). A'c(nM) BglT WT 480 BglT Y241F 383 BglT Y241A Not determined (very tightly bound) Table 2.16. Summary of N A D + dissociation constants with BglT WT, BglT Y241F, and BglT Y241A. Section 2.4.3.2. Kinetic and Spectroscopic Investigation of NAD+ Reduction. As was done with BglT WT, the role of the N A D + cofactor in transient redox chemistry was examined by measuring enzyme activity in the presence of N A D H , generated in situ via the addition of sodium borohydride. In the case of BglT Y241 A, it was difficult to determine whether the dinucleotide cofactor was quantitatively reduced by inspection of the absorbance spectrum alone because the N A D + cofactor could not be removed by dialysis. However, because samples of both BglT Y241F and BglT Y241A were completely inactive in the sodium borohydride-treated samples, it was assumed that 10 mM sodium borohydride is sufficient for the quantitative reduction of the N A D ' cofactor. Significantly, full activity was rapidly restored to these mutant enzyme samples upon addition of a fresh aliquot of N A D + . The requirement for the N A D + cofactor in its correct redox form for both mutants is thus established, and the restoration of full activity shows that the loss of activity did not result from enzyme denaturation. Activity assays Chapter 2 - BglT from Thermotoga maritima 93 of the mutants in the oxidized (NAD + ) and reduced (NADH) states are presented in Figure 2.27. (a) < (b) l 1 l 1 l 1 l 1 l 1 l 1 l 1 6 8 10 12 14 16 18 20 Time (min) S3 CO X> j ~ O < Figure 2.27. Assay of BglT mutants in the oxidized (NAD + ) and reduced (NADH) states, (a) Observed rates of hydrolysis of 4NPpG6P (2.13) by BglT Y24IF via the detection of 4-nitrophenolate release at 400 nm. Control (—): standard BglT Y241F assay conditions, 50 mM HEPES (pH 7.5), 0.1 mM MnCI 2 , 10 uM N A D + , 10 mM 2-mercaptoethanol, and 0.1% (w/v) BSA at 50 °C. BglT Y241F assay conditions (---): BglT Y24IF preincubated in 50 mM HEPES (pH 7.5), 0.1 mM MnCI 2 , 10 uM N A D + , 10 mM NaBH 4 , 10 mM 2-mercaptoethanol, and 0.1% (w/v) BSA at 50 °C. No release of 4-nitrophenolate is observed until time = 10 min, when 2 nmol of N A D + was added, (b) Observed rates of hydrolysis of 4NP(iG6P (2.13) by BglT Y241A via the detection of 4-nitrophenolate release at 400 nm. Control (—): standard BglT Y241A assay conditions, 50 mM HEPES (pH 7.5), 0.1 mM M n C l 2 , 10 u M N A D + , 10 mM 2-mercaptoethanol, and 0.1% (w/v) BSA at 50 °C. BglT Y24IA assay conditions (—'): BglT Y241A preincubated in 50 mM HEPES (pH 7.5), 0.1 mM M n C l 2 , 10 pM N A D + , 10 mM NaBH 4 , 10 mM 2-mercaptoethanol, and 0.1% (w/v) BSA at 50 °C. No release of 4-nitrophenolate is observed until time = 18 min, when 2 nmol of N A D + was added. Section 2.4.4. Determination of the Kd Value for M n ^ . The divalent metal ion binding properties of BglT Y241F and BglT Y241A were also investigated. Both mutants are completely inactive in the absence of M n 2 + and full activity could be restored in a saturable fashion by titration with rvln2+. The rates of hydrolysis of 4NPpG6P were measured at various concentrations of M n 2 + , and the data were fit to a standard ligand binding curve (Appendix 6). Dissociation constants were calculated and are summarized in Table 2.17. Chapter 2 - BglT from Thermotoga maritima 94 Ac ftlM) WT BglT 32 BglT Y 2 4 1 F 14 BglT Y 2 4 1 A 40 Table 2.17. Summary of M n 2 + dissociation constants with BglT WT, BglT Y24IF, and BglT Y241A. Section 2.4.5. Michaelis-Menten Kinetic Analysis of BglT Y24TF and BglT Y241A. Michaelis-Menten kinetics parameters for BglT Y241F were determined by measuring the initial rates of hydrolysis of 4NPpG6P spectrophotometrically at 400 nm. The activity of BglT Y241F was found to be approximately 100-fold lower than that of BglT WT (Table 2.18), while BglT Y241A has dramatically lower activity. The Michaelis-Menten kinetic parameters for BglT Y241A were measured using two different substrates: 4NPpG6P (2.13) and 4-methylumbelliferyl 6-phospho-(3-D-glucopyranoside (4MUpG6P or 2.108). 4NPpG6P was assayed with BglT Y241A using the standard direct UV-vis spectrophotometric assay. However, because the activity of BglT Y241A is over 1000-fold lower than that of the wild-type enzyme, a more sensitive stopped fluorimetric assay involving 4MUpG6P (2.108) was developed. 4MUPG6P was synthesized via the enzymatic phosphorylation of 4-methylumbelliferyl p-D-glucopyranoside (2.107) as shown in Scheme 2.15. 2.107 2.108 (4MUBC6P) Scheme 2.15. Enzymatic synthesis of 4MUfJG6P (2.108). (i) BglK, ATP. A standard curve of 7-hydroxy-4-methylcoumarin at various concentrations (absorption maximum at 360 nm and emission maximum at 450 nm) in 1 M NaOH was generated and used to correlate the fluorescence intensity to the rate of hydrolysis. In the stopped fluorimetric assay, BglT WT was incubated with various concentrations of 4MUPG6P under the standard assay conditions, aliquots were removed every minute for 5 minutes, quenched with 1 M NaOH, and the fluorescence measured to determine the initial rates of hydrolysis. The data were fit to the Michaelis-Menten equation and the Chapter 2 - BglT from Thermotoga maritima 95 kinetic parameters determined (Table 2.18). 4MUpG6P was also assayed against the WT enzyme via the direct UV-vis assay at 365 nm (Ae = 4847 cm"'M"') also monitoring for the release of the 7-hydroxy-4-methylcoumarin. The kinetic parameters obtained from the direct UV-vis assay are in agreement with those obtained from the stopped fluorimetric assay. The stopped fluorimetric assay for the BglT Y241A-catalyzed hydrolysis of 4MU]3G6P was conducted at pH 9.0, and the kinetic parameters determined are presented in Table 2.18. BglT Y241A has slightly higher kcaX and KM values for the hydrolysis of 4MUf3G6P than for 4NPpG6P, which is similar to the trend observed for BglT WT. Furthermore, the kinetic parameters obtained from the stopped fluorimetric assays are similar to those obtained for 4NPpG6P, indicating that the direct UV-vis assay produced reliable data and could be used in subsequent kinetic analyses of BglT Y241 A. Kinetic parameters Continuous UV assay Relative to BglT W T Stopped fluorimetric assay Relative to BglT W T Continuous U V assay Chromogenic substrate 4NPpG6P (2.13) Fluorogenic substrate 4MUPG6P (2.108) Fluorogenic substrate 4MUPG6P (2.108) B»l 1 N\ 1 J 0.99 1.0 1.0 1.0 1.2 K M (MM) 48 1.0 94 1.0 83 k . / K v i f s ' m M ' ) 21 1.0 11 1 0 14 A;, , (s ' ) 0.013 0.013 /CM(uM) 62 1.3 — — — k c J K M (s-'mM"') 0.21 0.010 — - — A , „ ( s ' ) 9.6 x 10"4 9.7 x 10"4 l . l x l 0 3 1.1 x 10"3 — /tM(pM) 134 2.8 289 3.1 -W^Cs 'mlvl ' - 1 ) 7.2 x 10"3 3.4 x 10"4 3.8 x 10"3 3.5 x IQ-4 -Table 2.18. kinetic parameters determined for BglT WT, BglT Y24IF, and BglT Y241A. The error in each case was < 15%. The 100-fold and 1000-fold lower activities, for the BglT Y241F and BglT Y241A mutants respectively, indicate that Tyr241 is important for catalysis. The higher activity of BglT Y241F versus BglT Y241A may result if the active site geometry is better maintained and an active site water molecule can act as a substitute for the Chapter 2 - BglT from Thermotoga maritima % phenolate anion of Tyr241, though its p/Ca could be quite different. Alternatively, through slight adjustments of the active site, a nearby amino acid residue might replace the role of Tyr241 in BglT WT and carry out the C2 deprotonation. Amino acid residues within 5 A of Tyr241 in BglT WT are shown in Figure 2.28. Of course, it is possible that the position of Thrl04 (6.4 A from C2 of the substrate) or Hisl92 (3.3 A from Cl) is readjusted in BglT Y241F such that either amino acid residue is close enough to C2 of the substrate to carry out the deprotonation. Consequently, further mechanistic studies were performed to probe these questions. Figure 2.28. Amino acid residues in close proximity to proposed catalytic base Tyr241 and C2 of the substrate. Section 2.4.6. Direct Observation of N A D H Formation in BglT Y241A. While measuring the initial rates of hydrolysis of 4NP(3G6P by BglT Y241A, an unusual burst phase was observed (Figure 2.29). Consideration of the mechanism suggests that the burst phase could be the result of the rapid initial formation of one of the intermediate species following 4-nitrophenolate release. Chapter 2 - BglT from Thermotoga maritima 97 0 2 4 6 8 Time (min) Figure 2.29. Burst phase kinetics for the hydrolysis of 4NP H G6P (2.13) by BglT Y241 A. Accumulation of any intermediate species would mostly likely necessitate that there is a build-up of N A D H as well. Therefore, the reaction of BglT Y241A with C6'P (2.2) was monitored at 340 nm to monitor changes in the N A D ' /NADH ratio. This substrate was chosen because it is spectrophotometrically silent. It is important to note that this experiment was performed at high enzyme concentrations, with no additional N A D + , thus any spectrophotometric changes observed at 340 nm would only result from the accumulation of endogenous NADH. As shown in Figure 2.30, the absorbance at 340 nm of a sample of BglT Y241A in the assay buffer at pH 9.0 was monitored. Upon addition of the substrate C6'P at time - 10 min, accumulation of N A D H was indeed observed, and the change in absorbance indicates that N A D H is formed in equimolar amounts to the BglT Y241A in the assay sample. Previous attempts at the direct observation of transient NADH for BglT WT had failed, indicating that the NADH species only transiently formed and that the cofactor is very tightly bound at the active site (also suggested by the low Kd for NAD H ) . In the current study, the N A D ' concentration was nearly the same as the BglT Y241A concentration. No large excess of N A D + was added to the enzyme assay, because this mutant was found to bind the dinucleotide tightly and added N A D + did not activate enzyme activity. Thus, no free NADH should be found in solution and any accumulation of NADH would have to be bound to the enzyme. The observation of NADH formation is gratifying, because this Chapter 2 - BglT from Thermotoga maritima 9>8 experiment provided further support for the participation of the N A D in redox chemistry. Presumably, the steady-state accumulation of N A D H is possible for BglT Y241A, because the some step subsequent to the oxidation is slower in the mutant enzyme than in BglT WT. Furthermore, since no accumulation of NADH is seen in the BglT WT assays, this suggests that the oxidation step is rate-limiting, which is consistent with the primary KIE determined for 3[2H]4NPpG6P. 0.9 - i 1 0.88 -1 0.86 -< o 0.84 -4 0.78 - | 1 1 1 1 1 1 0 10 20 30 Time (min) Figure 2.30. The steady state accumulation of NADH during the BglT Y241 A-catalyzed hydrolysis of C6'P (2.2). The absorbance of a sample of BglT Y24l A is monitored at 340 nm. The reaction is initiated by the addition of C6'P (2.2) at time = 10 min at which point steady-state accumulation of N A D H is observed. Section 2.4.7. pH-Dependence. The pH-activity profiles (Figure 2.31) of BglT Y241F and BglT Y241A are expected to differ appreciably from that of BglT WT if Tyr241 functions as the proposed catalytic base residue. The pH-dependence of kcJKM for BglT Y241F was therefore measured by the substrate depletion method at low substrate concentrations of 4NPp*G6P. The pH-dependence of kcat vs pH was obtained by measuring the initial rates of hydrolysis of 4NP(3G6P under saturating substrate concentrations. For BglT Y241A, the stopped fluorimetric assay was used to measure the initial rates of hydrolysis of 4MU(3G6P (saturating substrate concentrations), yielding values of kC2X under different pH conditions. The low activity of BglT Y241A did not allow for a reliable and Chapter 2 - BglT from Thermotoga maritima 99 convenient method for monitoring reaction rates by the substrate depletion method. Therefore, only the pH-dependence of kcat was measured for this mutant. Figure 2.31. pH-Activity profiles of BglT mutants. The filled squares represent the data of the mutant in each case, and the open circles represent the data obtained for BglT WT scaled to fit each plot, (a) BglT Y241 F, ktJKM vs pH; (b) BglT Y241 F kal vs pH; (c) BglT Y241A kcill vs pH. Each set of data were fit to classic bell-shaped curves and the pA^i, pAa2 and p H o p t values obtained are summarized in Table 2.19. This implies the involvement of two ionizable groups of different pATa over the range. Shifts in pA'ai of 0.3 and 0.6 pH units were observed for BglT Y241F in kCSLt and kcJKu respectively. For pA^, a shift of 1 pH unit was measured for the A'ca t vs pH plot. For the BglT Y241A mutant, the pA'ai and pATa2 were both shifted by more than 1 unit higher than that for BglT WT. The shift in the pKa Chapter 2 - BglT from Thermotoga maritima 100 values for BglT Y241F and BglT Y241A are consistent with the removal of the active Tyr base. Therefore, a shift in the acidic limb of the pH-activity profiles observed for BglT Y241F and BglT Y241A supports our contention that the pATai value corresponds to that of the active site Tyr. p H data based on ".'cat or ktAiIK\\ Ptfa. pKa2 p H o p t Bg lT W T A'cat 7 . 0 1 0 . 1 8.2 ± 0 . 1 7.6 7.1 ± 0 . 1 9.3 ± 0 . 1 8.2 Bg lT ktat 7.4 + 0.1 9.2 ± 0 . 1 8.3 Y241F kcaJ KM 7.7 ± 0 . 1 9.3 ± 0 . 1 8.5 B g l T kcat 8.6 ± 0 . 2 9.6 ± 0.2 9.1 Y241A i tat/ KM - — — Table 2.19. Summary of apparent pA'„ and p H o p l values for BglT WT, BglT Y241F, and BglT Y241A. Section 2.4.8. Kinetic Isotope Effects. KJEs were measured using l[ 2H]4NPpG6P (2.79), 2[ 2H]4NPpG6P (2.80), and 3[2H]4NPpG6P (2.81) for BglT Y241F and BglT Y241 A, and all data are summarized in Table 2.20 along with those values obtained for BglT WT. KIEs were initially measured at pH 7.5 for BglT WT and BglT Y241F. Inspection of the KIEs suggests that BglT Y241F utilizes a mechanism similar to that of BglT WT to catalyze the hydrolysis of 4NPpG6P (2.13). Primary KIEs are measured for 2[ 2H]4NPpG6P (2.80) and 3[ 2H]4NPpG6P (2.81), and no secondary KIE is observed for l[ 2H]4NPpG6P (2.79). This suggests that the C3 hydride abstraction and the C2 deprotonation steps are partially rate-limiting followed by a rapid cleavage of the C l - O l linkage consistent with an El Cb mechanism. BglT Y241F may well in fact be able to catalyze the hydrolysis reaction using a similar mechanism. Replacement of the Tyr with a Phe is not likely to dramatically change the active site architecture. It is conceivable that an active site water molecule could replace the role of the phenolate anion of Tyr as the catalytic base in C2 deprotonation, although it is surprising that the pH-dependence was so similar if this is the case. KIEs were measured for BglT Y241A under more basic conditions (since its Chapter 2 - BglT from Thermotoga maritima 101 pH 0pt = 9.1 is higher) and the data suggest that the nature of the individual steps of the mechanism may be different from that of BglT WT. For BglT Y241A, a significantly larger 1° KIE was measured for 2[ 2H]4NPpG6P (2.80) than for BglT WT (Table 2.20). Meanwhile, a very small 1° KIE was measured for 3[2H]4NPpG6P (2.81) and an inverse KIE for l[ 2H]4NPpG6P (2.79). The very small 1° KIE measured for 3[2H]4NPpG6P (2.81) and the large 1° KIE determined for 2[ 2H]4NPpG6P (2.80) suggest that the activation barrier to the C2 deprotonation has increased dramatically. This is consistent with the removal of the catalytic Tyr base, such that oxidation of the C3 hydroxyl is no longer a rate-determining step. The inverse KIE measured for 1 [ 2H]4NPpG6P is an a-2 KIE and it likely represents the rehybridization of the C l sp2 to a sp3 center during the 1,4-Michael-like addition, which may have become a rate-limiting process for BglT Y241A, and this may account for the steady-state accumulation of N A D H (Section 2.2.4.6). Keeping in mind that the KIEs for BglT Y241A were measured under much more alkaline conditions than those determined for BglT WT and BglT Y241F, it was important to examine the effect of pH upon KIEs for all systems. Therefore, KIEs were determined at 3 different pH values. For BglT WT, the KIEs at pH 7.5 and pH 8.4 are similar. 1° KIEs were measured for 2[ 2H]4NPpG6P and 3[ 2H]4NPpG6P, and no KIE was determined for l[ 2H]4NPpG6P. At pH 6.5, 1° KIEs were still measured for 2[ 2H]4NPpG6P and 3[ 2H]4NPpG6P, and kinetic analysis of 1 [ 2H]4NPpG6P revealed a 2° KIE. In particular, the 1° KIE for 2[ 2H]4NPpG6P increases with decreasing pH, consistent with a change in the ionization of the Tyr241, which needs to be deprotonated for C2 proton abstraction, as the pH drops below the p/Ca value (suggested to be approximately 7 from Section 2.2.4). As the concentration of Tyr241 in its deprotonated form decreases under more acidic assay conditions, the C2 proton abstraction step becomes more rate-limiting. This may also affect the 1,4-Michael-like addition step, which accounts for the 2° KIE determined for 1 [ 2H]4NPpG6P. Under acidic conditions, the addition step may become slower, because the catalytic base residue located around the substrate anomeric center may become protonated and thus unable to activate a water molecule for addition to the a,p-unsaturated or enediolate intermediate. Similar results Chapter 2 - BglT from Thermotoga maritima 102 were observed for BglT Y241F. 1° KIEs were observed for 2[2H]4NP(3G6P and 3[2H]4NPf3G6P under all three pH conditions, and a possible 2° KIE was measured for l[ 2H]4NPpG6P at pH 6.5. Again, the 1° KIE for 2[ 2H]4NPpG6P increases with increasingly acidic conditions for the reasons similar to those provided for BglT WT. Although Tyr241 has been replaced by a Phe, the KIE suggests that BglT Y24IF utilizes a mechanism similar to that of BglT WT, and perhaps an active site water molecule replaces the role of the phenolate anion of Tyr as the catalytic base in C2 deprotonation. As the pH drops below the p/^ai, the concentration of the general base responsible for C2 deprotonation decreases, thereby slowing that catalytic step. The KIEs for BglT Y241A were determined under 3 different pH conditions (pH 8.0, 9.0 and 10.0), and a similar trend for 2[2H]4NP(3G6P is observed for this mutant. The C2 deprotonation step slows down significantly under acidic conditions consistent with a reduction in concentration of the enzyme in the correct ionization state required for catalysis. Due to a limited availability of 1 [2H]4NP(3G6P, the KIE for this compound was not measured for BglT Y241A. Similar to BglT WT and BglT Y241F, only minor variations in the small 1° KIE were determined for 3[2H]4NP(3G6P. Most importantly, the KIEs measured for 2[ 2H]4NPpG6P for the wild-type enzyme and both mutants show a remarkable dependence upon pH. The data suggest that C2 deprotonation is the most sensitive to the pH of the assay solutions and that the general base catalysis plays a crucial role in C2 deprotonation. Chapter 2 - BglT from Thermotoga maritima 103 Substrate (*cat)H/(^cat)D at pH 6.5 M E S KIE (AcatVC&caOD at pH 7.5 HEPES KIE (ACat)H/(A'cat)D at pH 8.4 Tris KIE pH 6.5 M l ES pH 7.5 H E >ES pH 8.4 Tris l[ 2H]4NPpG6P (2.79) 1.16 ± 0.02 2° 1.00 ± 0.01 None 1.00 ± 0 . 0 1 None 2[ 2H]4NPPG6P (2.80) 4.89 ± 0 . 0 5 1° 1.84 ± 0 . 0 2 1° 1.47 ± 0.03 1° 3[ 2H]4NPpG6P (2.81) 1.85 ± 0 . 0 8 1° 1.63 ± 0 . 0 1 1° 1.61 ± 0 . 0 2 1° pH 6.5 M E S pH 7.5 HEPES pH 8.4 Tris lI 2H]4NPpG6P (2.79) 1.03 ± 0 . 0 3 None or 2° 1.00 ± 0 . 0 1 None 1.00 ± 0 . 0 1 None 2[ 2H]4NPpG6P (2.80) 4.2 ± 0 . 2 1° 2.06 ± 0.03 1° 1.50 ± 0 . 0 1 1° 3| 2 Hl4NPpG6P (2.81) 1.49 ± 0 . 0 7 1° 1.72 ± 0 . 0 1 1° 1.66 ± 0.03 1° BglT Y241A v .. . , * - . - J L ? , * ^>^*** r <£-pH 8.0 Tris J pH 9.0 Tris pH 10.0 CAPS l[ 2H]4NPpG6P (2.79) — — 0.93 ± 0.02 Inverse — -2[2H14NPPG6P (2.80) 5.8 ± 0 . 5 1° 2.54 ± 0 . 0 7 1° 1.65 ± 0 . 0 8 1° 3[2H14NPPG6P (2.81) 1.18 + 0.04 1° 1.04 ± 0 . 0 1 1° 1.05 ± 0 . 0 3 1° Table 2.20. KIEs measured for BglT WT, BglT Y241F, and BglT Y241 A. Section 2.5. Conclusions. This study represents the first detailed mechanistic analysis of a GH4 enzyme using a number of tools. Similar to other GH4 members, BglT, a 6-phospho-P-glucosidase, requires both N A D + and M n 2 + for catalytic activity. The enzyme was found to be a retaining glycosidase based on 'H NMR product analysis of the methanolysis reaction. However, solvent deuterium exchange of the C2 proton along with 1° KIEs measured for 2[ 2H]4NPpG6P (2.80) and 3[ 2H]4NPpG6P (2.81) indicate that the C2-H2 and C3-H3 linkages are cleaved during the enzymatic reaction. This result suggests that BglT does not utilize the double displacement mechanism employed by "classical" Chapter 2 - BglT from Thermotoga maritima 104 retaining glycosidases, but it supports the proposed redox'-elimination mechanism. The lack of a 2° KIE for 1 [ 2H]4NPpG6P (2.79) and the flat linear free energy relationship derived from aryl 6-phospho-p-D-glucopyranosides indicates an El Cb mechanism following the partially rate-limiting oxidation of the C3 hydroxyl. A 2° KIE was measured for the BglT-catalyzed hydrolysis of 1 [ 2H]MepG6P, indicating that the elimination step becomes rate-limiting when the substrate does not contain an activated phenolate leaving group. Examination of the active site architecture demonstrated that C4 of the nicotinamide ring in the dinucleotide cofactor is perfectly oriented for C3 hydride abstraction, providing support for the proposed oxidation step. Analysis of the x-ray crystallographic data further revealed that Tyr241 may act as the general base responsible for C2 deprotonation. Approximate p/Ca values of 7 and 9 were obtained from the pH-activity profiles. The p/Ca value of 9 could correspond to that of Arg87 and/or Arg261, which forms favorable electrostatic interactions with the substrate phosphate moiety. The pA^ value of 7 is suggested to represent that of the catalytic base Tyr241. Mutation of Tyr241 led to a significant decline in catalytic activity in the 2 mutants BglT Y241F and BglT Y241A, which supports the proposed role of Tyr241 in the catalytic mechanism. Interestingly, a burst phase was observed while measuring the initial rates of hydrolysis by BglT Y241A. By direct observation of the N A D H formation, this burst phase was shown to correspond to the steady-state accumulation of an intermediate species. This result can be rationalized by reasoning that in the BglT Y241A mutant the rate-limiting steps are different such that C3 oxidation is no longer a major rate-determining step (but a subsequent step is) and leads to this accumulation of an intermediate species. Together, the data presented here indicate that BglT utilizes the proposed redox-elimination-addition mechanism to hydrolyze 6-phospho-p-D-glucosides. Chapter 3 - GlvA from Bacillus subtilis 105 C h a p t e r 3 6 - P h o s p h o - o c - g l u c o s i d a s e G l v A f r o m Bacillus subtilis Chapter 3 - GlvA from Bacillus suhtitis 106 Section 3.1. Introduction and Specific Aims. GlvA is a 6-phospho-a-glucosidase from Bacillus subtil is, a Gram-positive spore-forming organism. Much of the groundwork for the kinetic analysis of this 449 residue, 50,513 Da enzyme has been reported by Thompson and co-workers.' This work established the substrate specificity, cofactor specificity, and physiochemical properties of this enzyme.39 In their study, mutagenesis of 3 conserved carboxylic acid residues and the kinetic analysis of these mutants was also performed. In addition, our collaborators Dr. Annabelle Varrot and Dr. Gideon Davies had reported the preliminary crystallographic analysis of GlvA. 5 3 Although these reports did not provide enough information for any mechanistic proposals, they greatly facilitated our endeavors towards understanding the mechanism of this enzyme. Our collaborator Dr. Jack Thompson generously provided a supply of active enzyme. Plenty of kinetic and mechanistic data had been accumulated for BglT (Chapter 2) by this time and those experiments provided a useful guide for the study of GlvA. The many tools used in the mechanistic analysis of BglT were also employed for studying the mechanism of GlvA. It was our primary goal to determine whether GlvA, chosen as a representative oc-glycosidase in GH4, also utilizes the redox-elimination-addition sequence proposed for the (3-members in this family and to provide the first report of the mechanistic analysis of an a-member of GH4. Chapter 3 provides a detailed kinetic investigation to further the understanding of the bond forming/breaking steps involved in the GH4 mechanism. Section 3.2. Kinetic Characterization. Section 3.2.1. Substrate Specificity. A previous report by Thompson and co-workers had established the substrate specificity of GlvA by testing a number of potential substrates, which are listed in Table 3.1. The natural substrate is proposed to be maltose 6'-phosphate (3.1), although trehalose 6-phosphate (3.3) was also hydrolyzed by the enzyme." Thompson and co-workers studied GlvA using the chromogenic substrate 4NPocG6P (2.17), and the fluorogenic substrate 4-methylumbelliferyl 6-phospho-a-D-glucopyranoside (4MUaG6P, 3.2) (Figure 3.1). The enzyme was found to be selective for phosphorylated glucosides, because GlvA does not hydrolyze 4-nitrophenyl a-D-glucopyranoside (4NPaGlc, 2.14), Chapter 3 - GlvA from Bacillus subtilis 107 4-nitrophenyl a-D-galactopyranoside (4'NPocGal, 2.16), or 4-nitrophenyl a -D-mannopyranoside (4NPaMan, 3.4).39 Furthermore, neither 4-nitrophenyl 6-phospho-a-D-galactopyranoside (4NPaGal6P, 3.5) nor 4-nitrophenyl 6-phospho-oc-D-mannopyranoside (4NPaMan6P, 3.6) were hydrolyzed by GlvA. J Although the physicochemical properties of GlvA had been determined in the previous study, the kinetic characterization was reinvestigated in the current study using the more convenient direct UV-vis assay using 4NPaG6P (2.17) as well as MeaG6P (2.98). In addition, because MalH, 5 0 a 6-phospho-a-glycosidase from GH4, has been reported to hydrolyze both 4NPaG6P (2.17) and 4NPpG6P (2.13), the substrate specificity of GlvA was further explored in the current study to test for its ability to hydrolyze p-glycosidic linkages using the following compounds: 4NPpG6P (2.13), MepG6P (2.58), and C6'P (2.2). Indeed, GlvA can efficiently hydrolyze 4NPpG6P (2.13). Chapter 3 - GlvA from Bacillus subtilis 108 2.16 (4NPaGal)f 3.4 (4NPaMan)' 3.5 (4NP<xGal6P)f 3.6 (4NPaMan6P)f HO-HO-^OP03Na2 -O OH ' 0 " O _ 2.13 (4NPBG6P) .OPOjNa, OH 2.58 (MeBG6P) Figure 3.T. Potential substrates of GlvA. 1 ' Data obtained from Thompson and co-workers. Chapter 3 - GlvA from Bacillus subtilis 109 Potential Substrate Hydrolysis by BglT C6'P (2.2) 4NPPG6P + (2.13) 4NPaGlc' (2.14) 4NPaGal r (2.16) 4NP<xG6P' (2.17) + MepG6P (2.58) MeaG6P + (2.98) Mal6'P f (3.1) 4MUaG6P' (3.2) + + Trehalose 6-phosphate + (3.3) 4NPaMan' (3.4) 4NPaGal6P' (3.5) 4NPaMan6P f (3.6) Table 3.1. Substrate specificity of GlvA.* The substrate specificity of GlvA was clearly demonstrated by analysis with the series of compounds (data obtained from current study and from Thompson et. al. 3 9) listed in Table 3.1. Except for 4NPocG6P (2.17), 4NPpG6P (2.13), and 4MUaG6P (3.2), none of the aryl glycosides were hydrolyzed by GlvA. Clearly, GlvA is selective for glucosides with a phosphate group at C6. Furthermore, GlvA can catalyze the hydrolysis of 6-phospho-a- and 6-phospho-P-glucosides with activated leaving groups. Therefore, the kinetic characterization of GlvA was carried out with 4NPaG6P (2.17), 4NPPG6P (2.13), and MeaG6P (2.98). The rates of hydrolysis of chromogenic substrates (4NPaG6P and 4NPPG6P) can be conveniently measured via a direct UV-vis spectrophotometric assay, monitoring the release of 4-nitrophenolate anion as depicted in Figures 3.2 and 3.3. Chapter 3 - GlvA from Bacillus subtilis 110 -0P0 3Na 2 H O ' ^ V ^ ' 0 \ HO-~v««--r»«A O H | 2.17(4NPaG6P) H 0 - " \ \ ^ ° \ Ho-~x»^r»«\ O H | 0 + ~N02 O H - N O , 2.3 (G6P) 2.12 Figure 3.2. Direct UV-vis spectrophotometric assay for the GlvA-catalyzed hydrolysis of 4NPocG6P (2.17). ( i )GlvA. ,.OP03Na2 H O - - \ « * « - ^ \ _ - - O . O H •~N02 2.13(4NPpG6P) H O ~ \ ^ T ^ A O H | 2.3 (G6P) G + 2.12 ? W - 400 nm Figure 3.3. Direct UV-vis spectrophotometric assay for the GlvA-catalyzed hydrolysis of 4NP(3G6P(2.13). (i) GlvA. Kinetic parameters for the hydrolysis of MeaG6P (2.98) were measured using the G6PDH coupled assay depicted in Figure 3.4. This substrate was chosen to test for the ability to hydrolyze 6-phospho-a-D-glucosidic linkages that do not involve an activated leaving group such as an aryl group. Al l results were obtained at pH 7.5 and at pH 8.4, because GlvA was found to display optimal activity at these two pH values with the kinetic parameters kcal and kcJK.M respectively (see Section 3.3.4). Chapter 3 - GlvA from Bacillus subtilis 2.98 (MeccG6P) 2.3 (G6P) 2.4 NADPH 2.5 *max - 340 nm Figure 3.4. Illustration of the G6PDH coupled assay for measuring kinetic parameters for GlvA. (i) GlvA; (ii) G6PDH, N A D P \ Section 3.2.2. Dinucleotide Cofactor. Section 3.2.2.T. Dinucleotide Cofactor Specificity and Determination of the Value. Thompson and co-workers had investigated the dinucleotide cofactor specificity of G l v A . 3 9 Of the cofactors tested (listed in Table 3.2), only N A D + and NADH provided any substantial activation of GlvA. N A D + was the best activator, but GlvA retained approximately 60% of its activity in the presence of N A D H . 3 9 Dinucleotide Cofactor Activation of BglT NAD + + NADH + NADP + — NADPH — NMN — ATP — FAD — FMN — Table 3.2. Dinucleotides tested by Thompson and co-workers for activation of GlvA. ' Chapter 3 - GlvA from Bacillus subtilis 112 The Ki values were reported to be 37 uM and 2.0 x 102 p.M for N A D + and N A D H respectively.39 Clearly, N A D + is the preferred dinucleotide cofactor based on the superior activation and the lower Kd value obtained for N A D + . 3 9 In the current study, the dinucleotide binding was reinvestigated using the direct UV-vis spectrophotometric assay. A dialyzed sample of GlvA was assayed with 4NPaG6P (2.17) as substrate in the presence of various concentrations of N A D + , with the enzyme being completely inactive in the absence of N A D + . The observation that dialysis of N A D + from the enzyme removed all catalytic activity and that full activity could be restored in a saturable fashion by titration with N A D + suggests an essential role for the cofactor. Ligand binding curves (Figure 3.5) were generated based on a direct fit of the data to a simple hyperbolic binding equation and the Kd values obtained are summarized in Table 3.3. A 0.1 mM concentration of N A D + was sufficient to provide full activity, and it was the concentration included in all subsequent kinetic analyses of GlvA. [NAD ] (mM) [NAD +] (mM) Figure 3.5. Ligand binding curves of N A D + for GlvA. (a) pH 7.5; (b) pH 8.4. Kd (NAD*) (p:M) pH 7.5 17 pH 8.4 29 Table 3.3. Summary of dissociation constants for NAD"" with GlvA. Chapter 3 - GlvA from Bacillus subtilis Section 3.2.2.2. Kinetic and Spectroscopic Investigation of Dinucleotide Cofactor Reduction. The report that N A D H activated G l v A 3 9 posed a conundrum in light of the proposed mechanism, because N A D H should be unable to carry out the initial oxidation of the C3 hydroxyl and thus render the enzyme inactive. However, the activity measured could have been an artifact due to a small amount of contaminating N A D + since N A D H is readily oxidized to N A D + in aqueous solution. Therefore, the activation of GlvA by N A D H was reinvestigated by generating NADH in situ using sodium borohydride to reduce the N A D + cofactor. As shown in Figure 3.6, the absorbance spectra (from 300 to 375 nm) of GlvA were recorded under three different conditions: 100 p M GlvA, 100 p M GlvA incubated with 100 | i M N A D + , and 100 p M GlvA incubated with 100 p M N A D + and 10 mM sodium borohydride (Figure 3.6). The peak in absorbance at 340 nm corresponding to NADH appears upon reduction of the enzyme sample with sodium borohydride, and is consistent with the quantitative reduction of N A D + to N A D H . On the basis of the small ATd value of < 30 p M for N A D + , more than 90% of GlvA has N A D + or N A D H bound to its active site under these conditions, and very little remains free in solution. 300 320 340 360 Wavelength (nm) Figure 3.6. Absorbance spectra of GlvA with the dinucleotide cofactor in its oxidized (NAD + ) and reduced (NADH) states. Absorbance spectra of the 100 uM GlvA (—), 100 u M GlvA incubated with 100 uM N A D + (—), and 100 uM GlvA incubated with 100 u M N A D + and 10 mM sodium borohydride ( ) at pH 7.5. Chapter 3 - GlvA from Bacillus subtilis The catalytic activity of the enzyme sample treated with sodium borohydride was compared to that of a sample activated with N A D + under standard assay conditions (Figure 3.7). Treatment of the assay sample with sodium borohydride resulted in a complete loss of activity towards the hydrolysis of 4NPocG6P (2.17). It was critical to note that full activity was restored to the borohydride-treated sample upon addition of N A D + , confirming that loss of activity was not due to enzyme denaturation or dramatic changes in pH (Figure 3.7). This experiment provided direct proof that NADH does not activate GlvA, in contrast to the results reported by Thompson and co-workers.39 The activation observed in their study was most likely due to contaminating amounts of N A D ' resulting from oxidation of the NADH sample. 0 20 40 60 0 10 20 30 Time (min) Time (min) Figure 3.7. Assay of GlvA in the oxidized (NAD + ) and reduced (NADH) states. Observed rates of hydrolysis of 4NPaG6P (2.13) by GlvA via the detection of 4-nitrophenolate release at 400 nm. In each case, the assay volume was 200 uL, and the concentration of GlvA used was 3.3 ug/mL. (a) pH 7.5: Control (—): standard GlvA assay conditions, 50 mM HEPES (pH 7.5), I mM M n 2 + , 0.1 mM N A D + , 10 mM 2-mercaptoethanol, and 0.1% (w/v) BSA at 37 °C. GlvA assay conditions (—): GlvA preincubated in 50 mM HEPES (pH 7.5), I mM M n 2 + , 0.1 mM N A D + , 10 mM NaBH 4 , 10 mM 2-mercaptoethanol, and 0.1% (w/v) BSA at 37 °C. No release of 4-nitrophenolate is observed until time = 18 min, when 2 nmol of N A D + was added. The activity was rapidly restored, and the catalytic rate is essentially the same as the sample without added NaBH 4 ; (b) pH 8.4: Control (—•): standard GlvA assay conditions, 50 mM Tris (pH 8.4), 1 mM M n 2 + , 0.1 mM N A D + , 10 mM 2-mercaptoethanol, and 0.1% (w/v) BSA at 37 °C. GlvA assay conditions (—): GlvA preincubated in 50 mM Tris (pH 8.4), 1 mM M n 2 + , 0.1 mM N A D + , 10 mM NaBH 4 , 10 mM 2-mercaptoethanol, and 0.1% (w/v) BSA at 37 °C. No release of 4-nitrophenolate is observed until time = 8 min, when 2 nmol of N A D + was added. The activity was rapidly restored, and the catalytic rate is essentially the same as the sample without added NaBH 4 . Chapter 3 - GlvA from Bacillus subtilis 1 15 Section 3.2.3. Divalent Metal Ion Specificity and Determination of the Ad Value. The metal ion dependency of GlvA was investigated by Thompson and co-workers, who found that, of a number of divalent metals tested (listed in Table 3.4), Mn 2* provided the greatest activation of the enzyme.3 Divalent Metal Ion Activation of GlvA relative to 1 mM M n 2 + M n 2 + 1.0 Fe 2 + 0.34 N i 2 + 0.21 Co 2 + 0.21 Sr 2 + 0.13 Ca 2 + 0.06 Mg2"1" 0.04 C d 2 + 0.02 Z n 2 + No detectable activity Table 3.4. Divalent metal ions tested by Thompson and co-workers for the activation of GlvA.' Therefore, Mn 2 + was chosen for activation of GlvA in the current study. A sample of GlvA that had been dialyzed to remove any bound metal ions was found to be devoid of activity. Full activity was restored upon addition of M n 2 + in a titratable fashion. Ligand-binding curves were generated by measuring the rate of hydrolysis of 4NPocG6P (2.17) in the presence of different concentrations of added M n 2 + at pH 7.5 and at pH 8.4. By fitting the data to a simple hyperbolic equation (Figure 3.8), dissociation constants were obtained, and these are summarized in Table 3.5. The M n 2 + concentration was kept at 1 mM (at saturating concentrations) for the activation of GlvA for all subsequent kinetic analyses. Chapter 3 - GlvA from Bacillus subtilis 1 16 tfd (Mn 2 + ) (UM) pH 7.5 54 pH 8.4 99 Table 3.5. Dissociation constants measured for Mn~ + binding to GlvA. Section 3.2.4. Reducing Conditions. Many GH4 enzymes (BglT, M e l A , 4 6 , 4 7 ' 5 5 Agu4A, 5 2 Agu4B, 6 5 M a l H , 4 0 4 2 ' 5 0 and A g l A 3 8 ' 4 4 ' 4 5 ) require the addition of reducing agents for enzyme activity. Although the previous report by Thompson and co-workers had not indicated that anerobic or reducing conditions were required for GlvA, it was found that the enzyme loses activity over time, but that full activity could be restored upon addition of 2-mercaptoethanol. TCEP and DTT were also tested for their ability to activate GlvA (Table 3.6). Both DTT and 2-mercaptoethanol activated GlvA (using 4NPaG6P (2.17) as the substrate), while TCEP did not. However, 10 mM 2-mercaptoethanol was utilized in all subsequent kinetic analysis of GlvA, because DTT is incompatible with the assay conditions, causing precipitates with M n 2 + under basic conditions. As discussed in Chapter 2, the reducing conditions are likely necessary to keep the absolutely conserved active site Cys in its thiol form. AglA from Thermotoga maritima44 was found to be inactive when the active site Cys was oxidized to a sulfinic acid, as revealed by structural analysis.44 Chapter 3 - GlvA from Bacillus subtilis 117 Reducing Reagent Activation of BglT TCEP — DTT + 2-mercaptoethanol + Table 3.6. Reducing reagents tested for activation of GlvA. Section 3.2.5. Results of Kinetic Characterization of GlvA. The report by Thompson and co-workers provided a general guideline for the kinetic characterization of G l v A . 3 9 Similar to other GH4 members, GlvA required both M n 2 + and N A D + for activity. Based on the dissociation constants obtained and the investigation of physiochemical properties, 1 mM M n 2 + , 0.1 mM N A D + , and 10 mM 2-mercaptoethanol were added to all GlvA assay solutions. The kinetic parameters for the GlvA-catalyzed reactions were measured at both pH 7.5 (50 mM HEPES) and pH 8.4 (50 mM Tris), for the reasons explained in Section 3.3.4. Because GlvA can hydrolyze both 4NPaG6P (2.17) and 4NPpG6P (2.13), these substrates were assayed with GlvA via the direct UV-vis assay. The hydrolyses of both substrates were monitored at 400 nm (the wavelength of maximal absorbance difference between the 4-phenolate leaving group and the respective substrate). Kinetic parameters (summarized in Table 3.7) were determined on the basis of a direct fit of the data to the Michaelis-Menten equation. Various concentrations of MeaG6P (2.98) were assayed with GlvA via the G6PDH coupled assay (A-max - 340 nm, ENADPH - 6220 cm'M" 1 ) . The data were fit to the Michaelis-Menten equation (Appendix 4), and the kinetic parameters obtained from the fits are listed in Table 3.7. Chapter 3 - GlvA from Bacillus subtilis 118 Conditions Substrate A M ( P - M ) KJKM (s'mlVI"1) 4NPaG6P (2.17) 0.70 52 13 pH 7.5 4NPf3G6P (2.13) 0.52 54 9.6 MeaG6P (2.98) 0.41 6.1 x 102 0.67 4NPaG6P (2.17) 0.85 14 61 pH 8.4 4NP0G6P (2.13) 0.34 24 14 MeaG6P (2.98) 1.1 5.0 x 102 2.2 Table 3.7. Summary of kinetic parameters for GlvA-catalyzed hydrolysis of substrates 4NPaG6P (2.17), 4NPPG6P (2.13), and MeaG6P (2.98). Error analysis was performed using the program GraFit. The error was < 15% in each case. Thus, the GlvA catalytic mechanism was investigated using both 6-phospho-a-and 6-phospho-(3-D-glucosides. As summarized in Table 3.7, GlvA catalyzes the hydrolysis of 4NPocG6P and 4NPpG6P with similar kinetic parameters. The &c a t values for 4NPaG6P are slightly larger than those for 4NP(3G6P. The KM values are also smaller for 4NPaG6P than for 4NP|3G6P, which results in higher kCJKM values for 4NPaG6P than for 4NPpG6P at both pH 7.5 and pH 8.4. Comparison of the kinetic parameters for the hydrolysis of MeaG6P with those measured for aryl 6-phospho-oc- and aryl 6-phospho-p-D-glucopyranosides is complicated by the very different chemical reactivities of these compounds. The Accat value determined for MeccG6P is similar to those obtained for 4NPocG6P, not differing by more than 4-fold, but the /CM value for MeccG6P is more than 10-fold larger than those measured for either 4NPocG6P or 4NPpG6P. This could be consistent with the formation of favorable interactions between the aglycone moieties of 4NPaG6P and 4NPpG6P and the +1 subsite, but not in the case of the methyl 6-phospho-glucosides. As a consequence, the kCJKM is about 20-fold smaller for MeocG6P than that determined for 4NPaG6P. Chapter 3 - GlvA from Bacillus subtilis Section 3.3. Mechanistic Studies. Section 3.3.1. Stereochemical Outcome Determination. The methanolysis reaction (Section 2.3.1) used to determine the stereochemical outcome for BglT was applied to GlvA. Fortunately, GlvA was also found to remain reasonably active in 5 M MeOH. 4NPaG6P (2.17) was therefore incubated with GlvA in the standard assay buffer system containing 5 M MeOH, until completion of reaction when the mixture was further treated with alkaline phosphatase. The reaction scheme is shown in Figure 3.9. Subsequently, the enzymes were removed by a centrifugal filter unit and the M n 2 + ion was removed using metal chelating resin. Product analysis via ' H NMR revealed a mixture of the hydrolysis product aGlc (a-2.4), pGlc (p-2.4), and the methanolysis product. The p-anomer, MepGlc (2.18), of the methanolysis reaction was not formed, and only MeaGlc (2.19) was detected. ,0H ,0H H O ' O H O " — ^ ~ V " Q H O ~ X — - ^ A -« — H O ~ V ^ ^ V - O H O H ] O H OH a-2.4 (aGlc) $-2.4 (3Glc) Figure 3.9. Schematic of the GlvA-catalyzed methanolysis of 4NPctG6P (2.17) for the determination of the stereochemical outcome, (i) GlvA, buffer and 5 M methanol; (ii) alkaline phosphatase; (iii) mutarotation. Figure 3.10 shows the ' H NMR spectra of the GlvA-catalyzed methanolysis reaction and compares the anomeric proton signals to those of standard solutions of MeaGlc (2.19), MepGlc (2.18), and glucose (a-2.4 and P-2.4). As shown in Figure 3.10(d), no ' H NMR signal at 4.22 ppm, corresponding to the anomeric proton of MepGlc (2.18), was observed in the *H NMR spectra of the reaction mixture. Since, the D 2 0 peak at 4.63 ppm is very broad, the ] H NMR spectrum was collected at 50 °C to Chapter 3 - GlvA from Bacillus subtilis 120 shift the D 2 0 peak upfield, thereby revealing the doublet signal at 4.68 ppm, which corresponds to the anomeric proton of MeaGlc (2.19). Chapter 3 - GlvA from Bacillus subtilis 121 Hi (a) (P-2.21) D , 0 HI (a-2.21) HI (3.71 1 > J b2o (0 (g) (h) (') HI (2.18) HI (2.19) D ,0 JUJ D ,0 HI (2.18) JUU HI (2.19) D ,0 r 7.0 6.0 5.0 4.0 ppm (t1) Figure 3.10. 'H NMR spectra showing the methanolysis of 4NPocG6P (2.17) by GlvA. (a) Methanolysis in D z O buffer and 5 M C D 3 O D , 'H N M R collected at 50 °C; (b) Methanolysis in D 2 0 buffer and 5 M C D 3 O D , 'H N M R collected at room temperature; (c) Methanolysis in H 2 0 buffer and 5 M MeOH, 50 °C; (d) Methanolysis in H 2 0 buffer and 5 M MeOH, room temperature; (e) aGlc (a-2.4) and PGlc 0-2.4), room temperature; (f) MepGlc (2.18), 50 °C; (g) MeaGlc (2.19), 50 °C; (h) MepGlc (2.18), room temperature; (i) MeaGlc (2.19), room temperature. The 'H N M R of the control experiment containing 4NPaG6P in D 2 0 buffer without enzyme is shown in Figure 3.15. Chapter 3 - GlvA from Bacillus subtilis 122 When the methanolysis reaction was performed in D?0 buffer containing 5 M C D 3 O D using a GlvA sample that was exchanged into D?0 buffer, the C2 proton was found to have exchanged with solvent deuterium to yield the methanolysis and hydrolysis products 3.7 and 2[2H]-D-glucopyranose (2.21) respectively (Figure 3.11). This result was verified by the singlet peaks at 5.07 and 4.49 ppm (Figures 3.10 (a) and (b)) for a-2.21 and (5-2.21 respectively, and at 4.68 ppm for 3.7 This result indicates that, similar to BglT (Section 2.3.2), the C2-H2 bond is cleaved during the course of the enzymatic reaction, as is further explored in Section 3.3.2. Figure 3.11. Schematic of the GlvA-catalyzed methanolysis of 4NPocG6P (2.17) in D 2 0 /CD 3 OD. (i) GlvA, D 2 0 buffer and 5 M CD,OD; (ii) alkaline phosphatase; (iii) mutarotation. The same mixture of reaction products was recovered when the GlvA methanolysis reaction was performed with 4NPf3G6P (2.13) as the substrate (Figure 3.12). The experiments were carried out independently in MeOH/HbO (Figure 3.12) and CD3OD /D2O (Figure 3.13) exactly as described for the methanolysis of 4NPaG6P. The 'H NMR spectra for the reaction mixtures and the standard sample solutions are shown in Figure 3.14. The 'H NMR for the control experiment containing 4NPaG6P in D2O buffer without enzyme is shown in Figure 3.15. Chapter 3 - GlvA from Bacillus subtilis 123 a-2.4 (aGlc) B-2.4 (pGlc) Figure 3.12. Schematic of the GlvA-catalyzed methanolysis of 4NPJ3G6P (2.13) for the stereochemical outcome determination, (i) GlvA, buffer and 5 M methanol; (ii) alkaline phosphatase; (iii) mutarotation. a-2.21 (2[2H]aGlc) 0-2.21 (2[2H]BGIc) Figure 3.13. Schematic of the GlvA-catalyzed methanolysis of 4NPpG6P (2.13) in D 2 0 / C D 3 O D . (i) GlvA, D 2 0 buffer and 5 M CD 3 OD; (ii) alkaline phosphatase; (iii) mutarotation. Chapter 3 - GlvA from Bacillus subtilis 124 7.0 6.0 5.0 4.0 ppm(t1) Figure 3.14. 'H N M R spectra showing the methanolysis of 4NP(3G6P (2.13) by GlvA. (a) Methanolysis in D , 0 buffer and 5 M CD 3 OD, 50 °C; (b) Methanolysis in D 2 0 buffer and 5 M C D 3 O D , room temperature; (c) Methanolysis in H 2 0 buffer and 5 M MeOH, 50 °C; (d) Methanolysis in H 2 0 buffer and 5 M MeOH, room temperature; (e) aGlc (a-2.4) and |3Glc (0-2.4), room temperature; (f) MepGlc (2.18), 50 °C; (g) MeaGIc (2.19), 50 °C; (h) MepGlc (2.18), room temperature; (i) MeaGIc (2.19), room temperature. The 'H N M R of the control experiment containing 4NPpG6P in D 2 0 buffer without enzyme is shown in Figure 3.16. Chapter 3 - GlvA from Bacillus subtilis 125 Thus, GlvA catalyzes the hydrolysis of 4NPocG6P (2.17) with net retention of substrate anomeric configuration, but the hydrolysis of 4NP(3G6P (2.13) proceeds with inversion of stereochemical configuration at C l . This observation is consistent with the proposed mechanism. The elimination step of the 4-nitrophenol leaving group from either 4NPaG6P or 4NPpG6P may not require general acid catalysis and may proceed to form the same enediolate intermediate regardless of the substrate anomeric configuration. However, the 1,4-Michael-like addition to the a,P-unsaturated intermediate (common to both the hydrolysis of 4NPocG6P and 4NPPG6P) requires a properly positioned water molecule and most likely a general base catalyst to activate the water molecule for nucleophilic attack. Therefore, the anomeric configuration of the methanolysis product is independent of substrate anomeric configuration, in contrast to what is found with "classical" glycosidases. However, since GlvA hydrolyzes 6-phospho-a-glucopyranosides without "activated" leaving groups (such as Mal6'P (3.1) and MeaG6P (2.98)) but not C6'P (2.2) or methyl 6-phospho-p-D-glucopyranoside (2.58), GlvA is certainly more appropriately classified as a 6-phospho-a-glucosidase than as a 6-phospho-p-glucosidase. The ability to hydrolyze p-glycosidic linkages is exceptional and applies only to activated 6-phospho-p-D-glucopyranosides. The solvent deuterium incorporation experiment was also performed for the methanolysis reaction (in D 2 0 buffer and 5 M CD 3 OD) with 4NPocG6P (2.17) and with 4NPpG6P (2.13). Product analyses of both sets of reaction mixtures (Figures 3.10 and 3.14) showed that solvent deuterium had been incorporated into C2 of the reaction products G6P (a—2.21 and p-2.21) as well as compound 3.7. Confirmation of this exchange was provided by the singlet 'H NMR peaks for the anomeric protons in the products as shown in Figures 3.10 and 3.14. The results suggest that GlvA utilizes the same mechanism for the hydrolysis and methanolysis of 4NPocG6P and 4NPpG6P. Thus, the production of only MeaGIc (2.19) from the methanolysis reaction suggests that GlvA is a retaining glycosidase. As of current, both BglT and GlvA have been shown to catalyze the hydrolysis of glycosidic linkages with retention of the substrate anomeric configuration, suggesting that GH4 can be classified as a family of retaining glycosidases. Chapter 3 - GlvA from Bacillus subtilis 126 Section 3.3.2. Solvent Deuterium Isotopic Exchange. A sample of GlvA was exchanged into D 2 0 buffer via repeated concentration of the sample using a centrifugal filter unit and dilution into D 2 0 . Solvent deuterium incorporation experiments at C2 Were performed on a number of compounds. G6P (2.3), 4NPaG6P (2.17), 4NPpG6P (2.13) and 1,5-anhydroglucitol 6-phosphate (2.23) were incubated in D 2 0 buffer with and without GlvA. Each compound was incubated for the amount of time required to hydrolyze 4NPaG6P (2.17) at the same fixed GlvA concentration. Upon completion, the enzyme and M n 2 + were removed and the samples were subjected to 'H N M R analysis. As shown in Figure 3.15, 4NPaG6P (2.17) was hydrolyzed to give a mixture of 2[ 2H]aG6P (a-2.21) and 2[2H]pG6P (P-2.21), as determined by the singlets for the anomeric protons at 5.07 and 4.49 ppm respectively. C2 of G6P (2.3) was also exchanged with solvent deuterium upon incubation with GlvA in D 2 Q buffer, resulting in 2[2H]G6P (2.21). Chapter 3 - GlvA from Bacillus subtilis 127 5.6 5.4 5.2 ^ 5.0 4.8 4.6 Figure 3.15. 'H N M R of the GlvA-catalyzed hydrolysis of 4NPaG6P (2.17) in D , 0 buffer, (a) Schematic of the GlvA-catalyzed hydrolysis of 4NPaG6P (2.17) in D 2 0 buffer: (i) GlvA, D 2 0 buffer; (ii) mutarotation; (b) Schematic of the solvent deuterium incorporation into G6P: (i) GlvA, D 2 0 buffer; (ii) mutarotation; and 'H NMR spectra of: (c) Control experiment: 4NPocG6P (2.17) incubated in D 2 0 buffer without GlvA; (d) GlvA hydrolysis of 4NPocG6P (2.17) in D 2 0 buffer; (e) Control experiment: G6P (2.3) incubated in D 2 0 buffer without GlvA; (e) G6P (2.3) incubated with GlvA in D 2 0 buffer. The solvent deuterium incorporation was also observed for the GlvA-catalyzed hydrolysis of 4NPpG6P (2.13) as shown in Figure 3.16. Chapter 3 - GlvA from Bacillus subtilis 128 i r r i I T ii i i i 5.00 4.50 ppm (t1) Figure 3.16. 'H NMR spectra illustrating the solvent deuterium exchange for the GlvA-catalyzed hydrolysis of 4NPfiG6P (2.13). (a) Schematic of the solvent deuterium incorporation via GlvA-catalyzed hydrolysis of 4NPfiG6P (2.13): (i) GlvA, D 2 0 buffer; (ii) mutarotation; 'H NMR spectra of: (b) GlvA hydrolysis of 4NPf3G6P (2.13) in D 2 0 buffer; (c) Control experiment: G6P (2.3) incubated in D 2 0 buffer without GlvA; (d) Control experiment: 4NPpG6P (2.13) incubated in D 2 0 buffer without GlvA. Finally, as shown in Figure 3.17, in the case of 1,5-anhydroglucitol 6-phosphate (2.23) a slight decrease in the intensity of the C2 proton and the emergence of doublets at H l a and H3 point to only partial deuterium incorporation to yield a mixture of 2.23 and 2.24 under the same reaction conditions. No such deuterium incorporation was detected when any of the four compounds G6P (2.3), 4NPccG6P (2.17), 4NP(3G6P (2.13) or 1,5-anhydroglucitol 6-phosphate (2.23) was incubated in D 2 0 buffer without GlvA (shown in Figures 3.15 (e), 3.15 (c), 3.16 (d) and 3.17 (c) respectively). Therefore, the solvent deuterium incorporation is a direct consequence of the reaction mechanism, akin to what was observed for BglT. Chapter 3 - GlvA from Bacillus subtilis 129 2.23 2.24 H6a, H6b, Hie l H2 H4 H3 H5 I l ia 3.50 p p m v . , ; Figure 3.17. 'H NMR analysis illustrating the solvent deuterium incorporation into 1,5-anhydroglucitol 6-phosphate (2.23). (a) Schematic of the solvent deuterium incorporation into 2.23: (i) GlvA, D 2 0 buffer; and ' H NMR spectrum of: (b) 2.23 incubated in D 2 0 buffer with BglT; and (c) Control experiment: 2.23 incubated in D 2 0 buffer without BglT. Section 3.3.3. Active Site Architecture and Mechanistic Implications. The x-ray crystal structure of GlvA (PDB 1NRH and 1U8X) was solved independently by our two collaborators (Dr. Wayne Anderson's group at Northwestern University and Dr. Gideon Davies' group at York University). The availability of the crystallographic data greatly facilitated our understanding of GlvA. The global structure is very similar to those of the reported GH4 structures such as A g l A 4 4 (PDB lOBB) and BglT (PDB 1UP6) (Section 2.3.3) from Thermotoga maritima. An overlay of all available GH4 structures is shown in Figure 3.18. The similarities between the CD spectra of BglT and of GlvA (Figure 3.19) are further indications that these two enzymes contain similar secondary structural components. Chapter 3 - GlvA from Bacillus subtilis 130 As stated in Chapters 1 and 2, the crystal structures of Agu4B (PDB 1VJT) from Thermotoga maritima and the 6-phospho-P-glucosidase (PDB 1SY6) from Geobacillus stearothermophilus do not provide useful information about the active sites. No ligands Chapter 3 - GlvA from Bacillus subtilis 131 are bound in the structure of the 6-phospho-p-glucosidase from Geobacillus stearothermophilus. In the Agu4B structure, only the N A D + cofactor is found at the enzyme active site, but the nicotinamide ring is not shown, probably because it is not well-defined in the crystal structure. The AglA structure is that of an inactive enzyme 44 with no divalent metal ion. Therefore, the active site architecture of GlvA is only compared to that of BglT, with obvious reservations made with respect to the different substrate specificities of these two enzymes. Of particular mechanistic relevance is that the N A D + cofactor is found to be in close proximity to the reaction product bound to the active site (Figure 3.20). Figure 3.20. Stereoview of the GlvA active site. This figure was generated using Swiss-PDB Viewer. 1 4 4 Figure 3.21 shows an overlay of the active sites of BglT and GlvA. Clearly, the geometry of the GlvA active site residues is very similar to that of BglT. Chapter 3 - GlvA from Bacillus subtilis 132 Figure 3.21. Active site of BglT overlayed with that of GlvA. The BglT structure is shown in green. For GlvA, C atoms are depicted in grey, N in blue, O in red, P in yellow, S in orange, and M n 2 + in pink. This figure was generated using Swiss-PDBViewer. 1 4 4 The geometric arrangement of the active site architecture is shown in Figure 3.22. C4 of the nicotinamide ring is 3.56 A away from C3 of G6P and in ideal geometric arrangement for hydride transfer between the carbohydrate and the cofactor to occur. As 2+ in BglT, the absolutely conserved active site Cysl71 is found to be chelated to the Mn" , which is also coordinated to 02 and 03 of G6P. The +1 subsite of GlvA is quite large and may explain the ability to bind and hydrolyze both aryl 6-phospho-a-D-glucosides and aryl 6-phospho-p-D-glucopyranosides. Also of interest is that Asp 172 is found 3.15 A away from the glycosidic oxygen of the bound G6P. This Asp 172 residue could be a catalytic residue providing general acid catalysis to the glycosidic oxygen and assisting the departure of the aglycone in the elimination step. The M n 2 + is octahedrally bound by SH from Cysl71 (2.4 A), NE2 of His202 (2.9 A), N07 of the nicotinamide ring (2.9 A) Chapter 3 - GlvA from Bacillus subtilis 133 of N A D + , hydroxyls 02 and 03 of G6P (2.6 and 2.3 A respectively), and an active site water (or hydroxide) molecule (2.1 A). Interestingly, the relative locations of the N A D + and the substrate are such that in GlvA and BglT the hydride moiety would be transferred to the re-face of the N A D + cofactor, whereas in the case of AglA it would appear to be to the .v/'-face. This difference arises from the different orientation of the N A D + in the active site in the two cases. This may also have been a consequence of an inactive AglA enzyme. Figure 3.22. Active site architecture of GlvA. Glycine rich region of the Rossman fold Conserved Asp c rs ^ -xi 1 10' * 20 30 I GO 70 80 90 100 110 120 130 8 #ff i • * • *f—• * * • • • • • 1 5; » 3 § BglT HRIflVIGGGSSYTPELVKGLLDISEDVRIDEVIFYuIDEEKQK—IVVDFVKRLVKDR—FK-VLISBTFEGRVVDRKYVIFQFRPGGLKGRENDEGIPLKYGLIGQETTGVGGFSRRLRrlFPI s n CelF SQKLKVVTIGGGSSYTPELLEGFIKRYHELPVSELHLVDVEGGKPKLOIIFDLCQRHIDHflGVPMK-LYKTLORREflLKDflOFVTTQLRVGQLPflRELOERIPLSHGYLGQETHGflGGLFKGLRTIPV 2. f £ Glvfi HKKKSFSIVIRGGGSTFTPGIVLHLLDHLEEFPIRKLKLYONDKERQD-RIflGRCOVFIREKRPDIE-FflflTTDPEEflFTOVDFVHflHIRVGKYflllRflLOEQIPLKYGVVGQETCGPGGIRYGHRSIGG | 3 6 w halH (IKKFSVVIRGGGSTFTPGIVLMLLDHHDKFPIRKLKFY0NDKERQR--IVflGRCEIILKEKRPEIE-FLRTTNPKEIlFTOVDFVMRHIRVGKYRHRELOEKIPLKYGVVGQETCGPGGIflYGttRSIGG 1 5 5 RglB HKKFSVVIRGGGSTFTPGIVLHLLflHQDRFPLRSLKFY0NDGRRQE~TIRERCKVILKEQRPEIE-FSYTTOPQflflFTOVOFVttflHIRVGKYPMREQDEKIPLRHGVLGQETCGPGGIRYGnRSIGG m % % % Pagl MKKYSICIVGGGSRYTPDHLflMLCNQKERFPLRKIVLYDNESERQE~TVGNYflKILFKEYYPELEEVIHTTOEKEflFEDIOFRLMQIRflGRLKMREKOEKISLKHGCLGQETCGflGGFflYGLRSVPfl M %> I l Consensus , . k . . ! v . i G G G S s X T P e l v . g l l . . . e . . p . . e l . l M B . I ! . e k q k . . i ! . d . ck r . . k l t k... . t .d..efl . .Dad/!v..q.RvG.l . .RE .DE . IpLkhG.lGQET .G.GG..,g*'R . ip. p S C: n I S - V i 131 140 150 1G0 Conserved active site Cys Conserved, active site His 2 3 0 m m m 8 g' « | 4 • • * f—t 1 , i - t * + • • 1 ft I ^ J P BglT VEEYVDTVRKTS-NRTIVNFTHPSGHITEFVRHYLEYE-KFIGLCNVPIHFIREIflEHFSflR-LEDVFLKYYGLNHLSFIEKVFVK--GEOVTEKVFEHL KLKLS---HIPOEDFPTHFYDSVRL-^ ? 3 CelF IFDIVKOVEELCPNRHVINFTHPflGMVTERVYRHTGFK-RFIGVCNIPIGMKMFIRBVLMLKBSDDLSIDLFGLNHHVFIKDVLIH--GKSRFflELLDGVflS-GQLKflSSVKHIFDLPFSEGLIRSLHL-3 OQ 3 § Glvfl VLEILDYHEKYSPDRUMLNYSNPRfllVRERTRRLRPNS-KILNICOHPVGIEORHRQILGLSSRKEMKVRYYGLHHFGHHTSIQBQE-GNBLHPKLKEHVSQYGYIPKT-ERERVEflSHNBTFRKRRBVQ 3 cc t>, I HalH VIEILDYHEKYSPNRHHLNYSNPflfllVRERTRKLRPNS-KILNICOHPIGIETRMREILGLESRKEMTVKYYGLHHFGHHSDIRBKD-GNOLHPKLKEHVKKYGYVRENGDTQHTDflSHNBTFflKflKDVY S b RglB VLELVDYMEKYSPNRMHLNYSHPflfllVREflTRRLRPNfl-KILNICDttPIGIEGRHRQIVGLKDRKQHRVRYYGLHHFGHHTSIEOLD-GHDLnPKLREYVflKYGYVPPSNDP-HTEflSHNDTFflKRKDVQ g W fc » Pag l VI0LIKSIRTYSPKCMILNYSNPflflIVREflTKRVFPN0YRIINIC0HPIflIHBIYRRVLGLK-RRDLEPKYFGLNHFGHFTHIL0KKTGEHYLPKLREILKTP-V0VqTEPLFQEKSHKSTFEFHSQI1I •a 3 3 . o Consensus ! . i . v d . v e k , s p n a u . , N ! £ t N P a g , ! t E a w r r . , , , . , k f i g . C t , P I g , , . , i a . , 1 . 1 k , . , # , . . k y / ! G L N H , , f i , , ! . , k , , G , d , . , k l , 8 , v , , , g , l k , s , , , n i , t t . , f , . , f , , s , , l . 3> 2" a Proposed catalytic base (Tyr) m 3 1 f J 3 2 0 3 3 0 M | 350 3G0 370 380 390 7^ T T + + + + + + + + + I 8 ^ St J ? o BglT IVNPYLRYYLMEKKHFK KISTHELRRREVHKIEKELFEKYRTRV--EIPEELTKRGGSHYSTRRRHLIR0LETDEGKIHIVNTRNNGSIENLPBDYVLEIPCYV-RSGRVHTLSQGKGOH 0 f. CelF LPCSYLLYYFKQKEHLRIEMGEYYKGGRRRQVVQKVEKQLFELYKHPELKVKPKELEQRGGRYYSBflRCEVIHRIYNOKQREHYVNIPHHGQIONIPRDHRVErlTCKLGRDGRTPHPRITHFOD 3 S i 3" TO -c < o Glvfl RRDPDTLPNTYLQYYLFPDDMVKKSHPNHTRRHEVtlEGRERFIFSQCDMITREQSSEHSEIKIDD-—HRSYIVDLflRRIRYHTGERMLLIVENNlifllRHFDPTRNVEVPCIVGSNGPEP-ITVGTIPQ ft I > HalH flVDPSTLPNTYLKYYLFPDYVVEHSNKEYTRRNEVHDGREKFVFGECKKVIEHQSTKGCKHEIDE-—HflSYIVDLflRRISYNTHERrlLLIVPNNGSIENFOSTGIIVEIPCIVGSNGPEP-LTHGKIPQ I f | RglB RLBPQTMPNTYLKYYLFPDYVVflHSNPERTRflNEVHDHREKNVFSflCRRIIflRGKSTRGBLEIDE-—HRSYIVDLRTRIRFNTQERHLLIVPNNGRIHHFDR0RMVEIPCLVGHNGPEP-LTVGOIPH f i l l Pag l HDYDEYLPNTYLQYYLYPRKHRNKENPEYTRRNEVHDGHEKETYERHHKIISLGKIHGTKYELTSBVGCHREYIVDLflTRIRNHTHEIFLIITENKGTIHHVSKBmiVEVPCRVGSNGVEP-LVVGSIPR W 1" &*w Consensus l p n . Y L . Y Y L . k . n ey r a . . v . K . e k . l f e k y p k e l . , r g g . . y s . a a . J i . a i . J . , , . h . v n < . n n G J . N , p . d . . v E , p C . v g r \ G . , p . l , , g 4 , d . > f t 1§ g 3 ' _ 8; 391 400 410 420 430 440 450 4G04G3 U | I f |> | + + + * + + +--| TO r > 3 BglT FflLSFIHflVKtlYERLTIERYLKRSKKLRLKRLLSHPLGPDVEBRKDLLEEILERHREYVKLG g a | 0 CelF KVtlGLIHTIKGFEIRRSNRRLSGEFHDVLLRLNLSPLVHSDRBRELLRREHILflHEKHLPHFRDCIRELKKRH P f g i Glvfl FQKGLttEQQVSVEKLTVERWflEKSFQKLHQRLILSKTVPHRRVRRLILEDLVEflNKBFMPELDQSPTRIS Q 2 cr P TO HalH FQKGLMEQQVSVEKLVVERHKEKSYQKLHQniTLSRTVPSRKVRKQILDELIEVNKBYHPELH RglB FQKGLHSQQVflVEKLVVDRMEQRSYHKLHQfllTLSKTVPSRSVRKfllLDBLIRRHKOYHPELH Pag l FYKGLHEHQYRYEKLSVDRCLEGSYQKRLQRLVLHRTVVHTDVRKELLKBLIERNKGYHNELH j H Consensus f , , g l i h . , k . . E J , J R J , , s , , , , l , R l , l s p l v p , , , d f i k , l l J , i e R n , , y . p . . , . , . £h =r< s?> K O O S 3 3 Chapter 3 - GlvA from Bacillus subtilis 135 Because the substrate specificities of the enzymes within GH4 differs widely, sequence alignment can be misleading. Thus, only the sequence alignment of 6-phospho-glycosidases is shown in Figure 3.23. More sequence alignments are given in Appendix 8. In particular, AlgA from Thermotoga martima is significantly larger than other GH4 members making any sequence alignment with this enzyme difficult to interpret. The glycine-rich region (GXGS) of the N-terminus in the Rossman binding fold is absolutely conserved amongst GH4 enzymes. Asp41, which makes extensive contacts with the ribose of N A D + in GlvA, is absolutely conserved. In addition, in the work reported by Thompson and co-workers,39 mutation of this residue to either a glycine or a glutamate resulted in a 100-fold decrease in catalytic activity." The corresponding Asp residue in BglT (Figure 2.14) also forms hydrogen bonding interactions with the ribose of the N A D + . Other absolutely conserved residues include Cysl71 (GlvA numbering) and His202, both of which are shown to be responsible for chelating the Mn 2 + ion in BglT and GlvA. Tyr265, proposed to be the catalytic base that abstracts the C2 proton, is also absolutely conserved amongst the 6-phospho-glycosidases, although this is not the case in the GH4 members that are not selective for substrates with a phosphate group at the C6 position (Appendix 8). The structure of AglA is most readily compared to those of GlvA and BglT. Spatially, Asp260 in AglA occupies the same position as the proposed catalytic Tyr base in BglT and GlvA (Figure 3.24). Consequently, Asp260 is proposed to act as the catalytic base for AglA. The tyrosine residue that is conserved in the 6-phospho-glycosidases is not found in AglA. Chapter 3 - GlvA from Bacillus subtilis 136 Figure 3.24. An overlay of the active sites of GlvA (blue) from Bacillus subtilis, and BglT (green) and AglA (pink) from Thermotoga maritima. This figure was generated using Swiss-PDB Viewer. 1 4 4 This is consistent with the contention that a tyrosine residue is used as a catalytic base instead of a carboxylate residue in GlvA and BglT because of potential unfavorable electrostatic interactions between the substrate C 6 phosphate and a carboxylate residue in the 6-phospho-a- and 6-phospho-p-glycosidases. A similar situation is seen in the case of sialidases and /mn.v-sialidases, where the catalytic nucleophile has been identified through covalent trapping of the glycosyl-enzyme intermediate to be a tyrosine residue rather than the usual aspartate/glutamate.33 These enzymes most likely evolved to use a tyrosine residue to diminish the electrostatic repulsion between the catalytic nucleophile and the substrate carboxylate functionality at the anomeric center of the sialic acid substrate.33 Most notably, it has been proposed that the chondroitin A C lyase, whose substrate also contains a negatively charged moiety, also uses a tyrosine as a catalytic base for a-proton abstraction. 8 5" 6 1 4 5- 1 7" Chapter 3 - GlvA from Bacillus subtilis 137 Section 3.3.4. pH-Dependence. The pH-stability of GlvA was determined by measuring the loss of activity after the enzyme sample had been incubated in different buffers at various pH conditions, then assayed under standard conditions. No loss of activity over the time taken for normal assays was found for enzyme samples incubated in the pH range of 4-12. Thus, the enzyme was found to be stable in this pH range, thereby defining the limits for investigating the pH-dependence of £ c a t and kcJKu-The pH-dependence of kcat was determined by measuring the initial rates of hydrolysis under saturating concentrations of 4NPocG6P ( 2 . 1 7 ) at various pH values. The differences in extinction coefficients, Ae, between 4NPaG6P ( 2 .17 ) and the 4-nitrophenolate leaving group under various pH conditions were determined by the total enzymatic hydrolysis of a 4NPocG6P solution of known concentration. Rates were measured at two different substrate concentrations at each pH value to ensure substrate saturation in the determination of Vniax- The feat values were calculated from the observed data, and they were plotted against the pH of the assay solution and fit to a bell-shaped curve (Figure 3.25(a)). The plot of k^J KM VS pH was generated by the substrate depletion method using 4NPocG6P as substrate. The data were also fit to a classical bell-shaped curve (Figure 3.25(b)). The pATai, pKA2 and pH o pt obtained from the pH-activity profiles are summarized in Table 3.8. The errors were determined based on the fit to the bell-shaped curve generated using the GraFit program.1 4 9 P A : „ P A U 2 pH o p t pH-dependence of ACat 7.0 ± 0 . 1 8.2 ± 0 . 1 7.6 pH-dependence of 7.7 ± 0 . 1 8.9 ± 0 . 0 1 8.3 Table 3.8. Summary of pKd values for GlvA as determined from the pH-dependence of k„M and from the pH-dependence of kcJKM. From analysis of the kcM vs pH-activity profile, the enzyme was found to have a p H o p t of 7.6, and pATai and pKA2 values of 7.0 ± 0.1 and 8.2 ± 0.1 respectively. Meanwhile, the plot of kcJKM vs pH revealed that GlvA exhibited optimal activity at pH 8.3. From the fitted curve, the pATai and pKA2 values were calculated to be 7.7 ± 0.1 and Chapter 3 - GlvA from Bacillus subtilis 138 8.9 ± 0.01 respectively. Because two different pH o p t were obtained, all subsequent kinetic analyses were performed at pH 7.5 (HEPES) and at pH 8.4 (Tris). Because the pH-dependence of kcJKM reflects the ionization of either the free enzyme or the free substrate, the two apparent pA"a values could very well represent ionizable groups within the active site that are critical to catalytic activity. The values obtained are reasonably close to those reported for BglT (see Section 2.3.4). As proposed for BglT, the pATai value of 7.7 may correspond to that of the active site Tyr residue responsible for C2 deprotonation: Tyr265 in GlvA and Tyr241 in BglT. However, because GlvA is also selective for phosphorylated substrates, this pATa value could potentially correspond to that of the phosphoryl moiety of the substrate. The pKa2 value may represent the ionization of the conserved Arg95 and/or Arg285 that by x-ray crystallographic analysis are found to be within hydrogen bonding distance to the phosphate group of the substrate. The pH-dependence of kcM usually reflects the ionizations in the enzyme-substrate complex, which may explain the slight differences observed between the two pXTa values determined from the kC3X and kcJKu plots. The interpretation that the two pKa values of approximately 7 for Tyr265 or the C6 phosphate group, and around 9 for the arginine residue(s) located around the phosphate moiety on the substrate still appears to be the most compelling explanation based upon examination of the active site architecture (Section 3.3.3). Chapter 3 - GlvA from Bacillus subtilis 139 Section 3.3.5. Linear Free Energy Relationships. Section 3.3.5.1. Preparation of Substrates for Linear Free Energy Relationship. The linear free energy relationship for the GlvA-catalyzed cleavage of the C l - O l linkage was studied for the hydrolysis of a series of 6-phospho-a-D-glucopyranosides and a series of 6-phospho-p-D-glucopyranosides. Aryl 6-phospho-p-D-glucopyranosides had already been prepared for analysis of the Bronsted relationship in BglT (Section 2.3.6). The synthesis of 6-phospho-a-D-glucopyranosides proved to be more challenging, since BglK only phosphorylates • p-D-glucopyranosides and not a-D-glucopyranosides. Selective phosphorylation of the C6 primary hydroxyl group in the presence of three secondary hydroxyl groups can be achieved in several ways. One of these methods uses the phosphorylating reagent diphenyl chlorophosphate, which was used to synthesize both 2.58 and 2.98 (Schemes 2.6 and 2.12). However, this method is incompatible with the synthesis of aryl 6-phospho-D-glucopyranosides, because the deprotection requires hydrogenation or acidic conditions, both of which are incompatible with nitrophenol leaving groups at the anomeric center. The phosphorylation method described by Wilson and Fox 1 3 4 is the most commonly used method for the synthesis of aryl 6-phospho-a-D-glucopyranosides, and was selected for the synthesis of the target compounds. Since the rates of BglT-catalyzed hydrolysis of a series of aryl 6-phospho-p-D-glucopyranosides were essentially independent of leaving group ability, and a similar situation was likely for the GlvA-catalyzed hydrolysis of aryl 6-phospho-a-D-glucopyranosides, only three aryl 6-phospho-a-D-glucopyranosides were chosen for the Bronsted analysis: phenyl 6-phospho-a-D-glucopyranoside (PaG6P, 3.10), 4NPaG6P (2.17), and 3,4-dinitrophenyl 6-phospho-a-D-glucopyranoside (34DNPaG6P, 3.14). PaG6P (3.10) was synthesized according to Scheme 3.1. Compound 2.25 was coupled with phenol via stannic chloride-catalyzed glycosidation.179 Subsequently, 3.8 was deprotected by the Zemplen method, and selective phosphorylation of 3.9 was achieved using phosphorus oxychloride to yield the desired product PaG6P (3.10). Chapter 3 - GlvA from Bacillus subtilis 140 2.25 I 3.8: R = A c 3.10 ' — - 3.9: R = H Scheme 3.1. Synthesis of PocG6P (3.10). (i) phenol, SnCI4, CH 2 C1 2 , 60 °C; (ii) NaOMe/MeOH; (iii) POCI 3, H 2 Q, PO(OCH 3 ) 3 . The synthesis of 34DNPaG6P (3.14) is illustrated in Scheme 3.2. Compound 3.11 was synthesized by reaction of 2.25 in HF-pyridine, then glycosidation of 3.11 with 3,4-dinitrophenol using BF3-etherate. The acetyl protecting groups of 3.12 were removed using HCl(g) dissolved in methanol and finally selective phosphorylation at C6 was achieved using phosphorus oxychloride to produce 34DNPaG6P (3.14).1 3 4 Scheme 3.2. Synthesis of 34DNPaG6P (3.14). (i) HF, pyridine; (ii), BF 3 -Et0 2 , 3,4-dinitrophenol, acetonitrile; (iii) H C l ( g ) , MeOH; (iv) POCl 3 , PO(OCH 3 ) 3 , H 2 0 . Chapter 3 - GlvA from Bacillus subtilis 141 Section 3.3.5.2. Linear Free Energy Relationship - Aryl 6-Phospho-|3-D-glucopyranosides. Rates of hydrolysis of each aryl 6-phospho-p-D-glucopyranoside (2.13, 2.47— 2.55) by GlvA at pH 7.5 and pH 8.4 was monitored at the wavelength of maximal absorbance difference between the substrate and the respective phenol leaving group (listed in Tables 3.9 and 3 .10) with the difference in extinction coefficient, Ae, being determined by the total enzymatic hydrolysis of a known concentration of substrate under these conditions. Each set of data showed saturation kinetics and were fit to the Michaelis-Menten equation to determine the kinetic parameters kcat and KM, which are summarized in Tables 3.9 and 3.10. The log kcat and log kcJKM values were calculated and plotted against the pATa of the leaving groups. As shown in Figures 3.26 and 3.27, the data points for the log kcJKM vs pATa plots are more scattered than the data for log kCat. This disparity is expected because the natural substrate is a 6-phospho-a-D-glucopyranoside. The km represents the activation energy between the enzyme substrate complex and the transition state, so dependencies on binding to the various substrates are present in both the ground state and the transition state, thus partially cancel. Meanwhile kcJKM represents the conversion from the free enzyme to the transition state, so no cancellation occurs. Therefore, the different phenol structures are irrelevant for A:cat, while the kcJKu represents the steps from the free enzyme to some transition state structure and would include some binding component. Therefore, the binding of 6-phospho-f3-D-glucopyranosides may be very sensitive to small changes to the aglycone moiety. Nonetheless, the Bronsted plots shown in Figures 3.26 and 3.27 indicate essentially no dependence of the rate on leaving group ability. Such a flat linear free energy relationship can result if cleavage of the C l - O l linkage is not rate-limiting or if it is rate-limiting, but general acid catalysis provides sufficient proton donation to the departing oxygen such that little negative charge accumulates at O l during departure of the aglycone. Chapter 3 - GlvA from Bacillus subtilis 142 A r y l 6-phospho-P-D-glucopyranoside Phenol PA* (nm) Ae (IVT'cm"1) kCM (s"1) AM ( U M ) ^cat/Ajvi ( s ' m M 1 ) 2 4 D N P 0 G 6 P (2.47) 3.96 400 9871 0.27 35 7.5 2 5 D N P p G 6 P (2.48) 5.15 443 4821 0.25 1.3 x 10 2 2.0 3 4 D N P P G 6 P (2.49) 5.36 400 14951 0.42 2.9 1.5 x 10 2 4 C 2 N P p G 6 P (2.50) 6.45 428 4498 0.18 67 2.7 4 N P p G 6 P (2.13) 7.18 400 12618 0.52 54 ••• - • v 9.6 2 N P p G 6 P (2.51) 7.22 413 2155 0.29 2.9 x 10 2 1.0 3 5 D C P P G 6 P (2.52) 8.19 285 1930 0.66 27 : ... • 24 3 N P P G 6 P (2.53) 8.39 380 849 0.54 34 16 4 C N P p G 6 P (2.54) 8.49 272 6758 0.64 1.2 x 102 5.5 P P G 6 P (2.55) 9.99 270 1044 0.27 1.1 x 102 2.5 Table 3.9. Michaelis-Menten kinetic parameters for the hydrolysis of a series of aryl 6-phospho-p-D-glucopyranosides by GlvA at 37 °C in HEPES buffer (pH 7.5). Error analysis was performed using the program GraFit. The error was < 15% in each case. (b) ^ o oo -2 O -4 H Slope = 0 ~i i — r 4 6 i—i—r~ 8 10 Figure 3.26. Bronsted plots of the GlvA-catalyzed hydrolysis of a series of aryl 6-phospho-fi-D-glucopyranosides at pH 7.5 (HEPES buffer) with the corresponding pA"a values for the leaving group phenol, (a) Bronsted plot for the correlation of log (£ c ; r t) v.v pK:i; (b) Bransted plot for the correlation of log (k^JKu) vs pK.d. Chapter 3 - GlvA from Bacillus subtilis 143 A r y l 6-phospho-f}-D-glucopyranoside Phenol Amax (nm) Ae (ivrW1) kcat (S"') KM ( p M ) kcatl KM or'mivr1) 24DNPPG6P (2.47) 3.96 400 9903 0.63 41 16 25DNPPG6P (2.48) 5.15 443 4035 0.46 73 6.3 34DNPpG6P (2.49) 5.36 400 14720 0.36 1.9 1.9x 10" 4C2NPpG6P (2.50) 6.45 428 4152 0.52 75 6.8 4NPpG6P (2.13) 7.18 400 17308 0.34 24 14 2NPpG6P (2.51) 7.22 413 4581 0.40 3.0 xlO2 1.3 35DCPpG6P (2.52) 8.19 285 2387 0.47 3.8 1.2 x 102 • ' H m 3NPPG6P (2.53) 8.39 380 1289 0.97 8.9 1.1 x 102 4CNPpG6P (2.54) 8.49 272 14008 0.32 28 11 PpG6P (2.55) 9.99 270 488 1.1 23 52 Table 3.10. Michaelis-Menten kinetic parameters for the hydrolysis of a series of aryl 6-phospho-P-D-glucopyranosides by GlvA at 37 °C in Tris buffer (pH 8.4). Error analysis was performed using the program GraFit. The error was < 15% in each case. o 1 -4 H (b) -4 A Slope = 0 10 Figure 3.27. Bronsted plots of the GlvA-catalyzed hydrolysis of a series of aryl 6-phospho-(i-D-glucopyranosides at pH 8.4 (Tris buffer) with the corresponding pKa values for the leaving group phenol, (a) Bransted plot for the correlation of log (A:aU) v.v pK_,; (b) Bransted plot for the correlation of log (kcJKM) vs pKa. Chapter 3 - GlvA from Bacillus subtilis 144 Section 3.3.5.3. Linear Free Energy Relationship - 6-Phospho-a-D-glucopyranosides. Bronsted plots were also generated for the GlvA-catalyzed hydrolysis of a small series of 6-phospho-a-D-glucopyranosides: 34DNPaG6P (3.14), 4NPaG6P (2.17), PaG6P (3.10), and MeaG6P (2.98). Initial rates for the hydrolysis of the aryl 6-phospho-a-D-glucopyranosides were measured at the wavelength of maximal absorbance difference between the substrate and the respective phenol leaving group (listed in Tables 3.11 and 3.12) as described in Section 3.3.5.2. Reaction rates for the hydrolysis of MeaG6P (2.98) were measured using the G6PDH coupled assay, monitoring for the formation of NADPH (e = 6220 IvT'crrf1). Each set of data was fit to the Michaelis-Menten equation to determine the kinetic parameters kcat and A M , which are summarized in Tables 3.11 and 3.12. As shown in Figures 3.28(a) and 3.29(a), the plots of log kCM against the pKa of the leaving group were essentially flat and indicate that the kcat does not display any significant dependence upon the aglycone leaving group ability at either pH 7.5 or pH 8.4. Because of the very different chemical reactivity of MeaG6P compared to those of aryl 6-phospho-a-D-glucopyranosides, the linear regression was fit only to the data points obtained for the aryl 6-phospho-a-D-glucopyranosides in each case. Again, the flat linear free energy relationship is indicative of one of two possible situations: either cleavage of the C l - O l linkage is not rate-limiting and/or very little negative charge develops at OI during the rate-limiting step. The plots shown in Figure 3.28(b) and 3.29(b) indicate that the kcJKM shows a slight dependence upon the leaving group ability, with the shallow slopes of-0.13 (pH 7.5) and -0.32 (pH 8.4) suggesting that there is only minor negative charge development at the glycosidic oxygen at the transition state during cleavage of the C l - O l linkage. However, because only 4 data points were collected, it is unclear whether there is a rate dependence of the leaving group ability across the range when this group is varied from methanol to 3,4-dinitrophenol or whether the Bronsted plots in Figures 3.28(b) and 3.29(b) are in fact biphasic, which would indicate a change in the rate-limiting steps for the different substrates used in this study and rate-limiting C-O bond cleavage only for MeaG6P (2.98). Chapter 3 - GlvA from Bacillus subtilis 145 6-Phospho-a -D-glucopyranoside Leav ing group p A . A-max (nm) Ae (ivr'cm"') * c a t (S"') KM ( U M ) kcat/KM ( s ' m M " ' ) 3 4 D N P a G 6 P (3.14) 5.36 400 14951 0.72 12 £ 1 61 4 N P a G 6 P (2.17) 7.18 400 12602 0.70 52 14 P a G 6 P (3.10) 9.99 270 1044 0.90 69 13 M e a G 6 P (2.98) 15.5 ~ - 0.42 6.1 x 10 2 0.69 Table 3.11. Michaelis-Menten kinetic parameters for the hydrolysis of a series of 6-phospho-a-D-glucopyranosides by GlvA at 37 °C in HEPES buffer (pH 7.5). Error analysis was performed using the program GraFit. The error was < 15% in each case. (a) DO -2 1 O -4 1 Slope = 0.02 i — | — i — | — i — | — i | — i | i | r 2 4 6 8 10 12 14 16 pA a (b) „ 2 X •~H 0 -4" -6 H Slope = -0.13 il | I | i | I | i | i | r 2 4 6 8 10 12 14 16 Figure 3.28. Bransted plots of the GlvA-catalyzed hydrolysis of a series of 6-phospho-a-D-glucopyranosides at pH 7.5 (HEPES buffer) with the corresponding pA'a values for the leaving group. The linear regression was only tit to the data obtained for aryl 6-phospho-a-D-glucopyranosides. Open circles = aryl 6-phospho-a-D-glucopyranosides; filled squares = MeaG6P. (a) Bransted plot for the correlation of log ( feat) v*v pA a ; (b) Bransted plot for the correlation of log (k^/KM) vs pKd. Chapter 3 - GlvA from Bacillus subtilis 146 6-Phospho-a-D-glucoside Leaving group pKa ^max (nm) Ae (IVL'cm"') Acar ( S1 ) KM (pM) KMIKM (s'mlVl"1) 34DNPaG6P (3-14) 5.36 400 14720 0.81 4.0 2.0 x 10 2 4NPaG6P (2.17) 7.18 400 17259 0.85 14 60 PaGoP (3.10) 9.99 270 488 0.26 38 6.9 MeaGoP (2.98) 15.5 — - 1.1 5.0 x 102 2.2 Table 3.12. Michaelis-Menten kinetic parameters for the hydrolysis of a series of 6-phospho-a-D-glucopyranosides by GlvA at 37 °C in Tris buffer (pH 8.4). Error analysis was performed using the program GraFit. The error was < 15% in each case. 2 4 6 8 10 12 14 16 2 4 6 8 10 12 14 16 Figure 3.29. Bransted plots of the GlvA-catalyzed hydrolysis of a series of 6-phospho-a-D-glucopyranosides at pH 8.4 (Tris buffer) with the corresponding pA'a values for the leaving group. The linear regression was only fit to the data obtained for aryl 6-phospho-a-D-glucopyranosides. Open circles = aryl 6-phospho-a-D-glucopyranosides; filled squares = MeaG6P. (a) Bransted plot for the correlation of log (A'cu,) v.v p/C,; (b) Bransted plot for the correlation of log (KJKU) vs pKd. Section 3.3.6. Kinetic Isotope Effects. Section 3.3.6.1. Preparation of Substrates for Kinetic Isotope Effect Measurement. As discussed in Section 2.3.6.1, KIEs are possible in this system for compounds with deuterium substitution at the substrate C l , C2 and C3 positions. In the proposed mechanism, the C2-H2 and C3-H3 linkages are cleaved. If these bonds are cleaved in the rate-limiting steps, 1° KIEs may be measured for substrates with deuterium substitutions at C2 and C3. If the rate-limiting step involves rehybridization at C l , a 2° KIE will be Chapter 3 - GlvA from Bacillus subtilis 147 observed for substrates with a deuterium substitution at C l . Therefore KIEs were measured using 1 [2H]4NP(3G6P (2.79), 2[2H]4NP(3G6P (2.80), and 3[ 2H]4NPpG6P (2.81). The corresponding deuterio 4-nitrophenyl 6-phospho-a-D-glucopyranosides (3.21 to 3.23) were also prepared according to Scheme 3.3. Coupling of the deuterio 1,2,3,4,6-penta-O-acetyl-p-D-glucopyranoses (2.64—2.66) with 4-nitrophenol was achieved via stannic chloride-catalyzed glycosidation. 1 7 9 - 1 8 0 Zemplen deprotection of 3.15—3.17 yielded the corresponding deuterio 4-nitrophenyl a-D-glucopyranosides (3.18—3.20), which were selectively phosphorylated on the primary hydroxyl group at C6 using phosphorous oxychloride to yield the desired compounds: 4-nitrophenyl l[ 2H]-6-phospho-a-D-glucopyranoside (1 [ 2H]4NPaG6P or 3.21), 4-nitrophenyl 2[2H]-6-phospho-a-D-glucopyranoside (2[2H]4NPocG6P or 3.22), and 4-nitrophenyl 3[2H]-6-phospho-a-D-glucopyranoside (3[ 2H]4NPaG6P or 3.23). 2.67: R 2 , R 3 = H, R'= D 2.68: R1, R 3 = H, R 2 = D 2.69: R', R 2 = H, R 3 = D 3.15: R 2 , R 3 = H, R 1 = D, R" = Ac 3.16: R', R 3 = H, R 2 = D, R 4 = Ac 3.17: R \ R 2 = H, R 3 = D. R 4 = Ac 3.18: R 2, R 3 , R 4 = H, R 1 = D 3.19: R1, R 3 . R 4 = H. R 2 = D 3.20: R \ R 2 , R 4 = H, R 3 = D 3.21: R 2, R 3 = H, R' = D 3.22: R 1, R 3 = H. R 2 = D 3.23: R 1 . R 2 = H, R 3 = D Scheme 3.3. Synthesis of l[ 2H]4NPaG6P (3.21), 2[ 2H]4NPaG6P (3.22), and 3[ 2H]4NPaG6P (3.23) for KIE measurements with GlvA. (i) SnCI 4, 4-nitrophenol, CH 2 C1 2 , 60 °C; (ii) NaOMe/MeOH; (iii) POCl 3 , PO(OCH 3 ) 3 , H 2 0 . Compound 3.26 was synthesized according to Scheme 3.4. Compound 2.65 was subjected to methanolysis under acidic conditions to yield methyl l[ 2H]-a-D-glucopyranoside (3.24). Diphenyl chlorophosphate was utilized for the selective phosphorylation of 3.24 at C6. Then, deprotection of 3.25 by catalytic hydrogenation afforded the desired product methyl 1 [2H]-6-phospho-a-D-glucopyranoside (l[ 2 H]MeaG6Por 3.26). Chapter 3 - GlvA from Bacillus subtilis 148 1 *- 3.25: R 1 = H, R 2 = PO(OPh) 2 Scheme 3.4. Synthesis of I [2H]MeocG6P (3.26). (i) MeOH, H + ; (ii) PO(OPh)2CI, pyridine; (iii) PtO,, H2(g), MeOH. Section 3.3.6.2. Kinetic Isotope Effects - Results. KIEs upon 'kctA and k0JKM were determined at both pH 7.5 and at pH 8.4. The KIEs upon /ccat were obtained by measuring the initial rates of hydrolysis of the protio and deuterio compounds under saturating substrate conditions. The KIEs upon kcJK.M were determined by the substrate depletion method at substrate concentrations far below A^M. In each case, the rates for the protio and deuterio chromogenic substrates (4NPpG6P, l[ 2H]4NPpG6P, 2[ 2H]4NPpG6P, 3[ 2H]4NPpG6P, 4NPaG6P, l[ 2H]4NPaG6P, 2[ 2H]4NPaG6P, 3[ 2H]4NPaG6P) were measured in alternation via the direct UV-vis assay, monitoring the release of 4-nitrophenolate at 400 nm. Rates for the protio and deuterio methyl 6-phospho-a-D-glucopyranosides (MeaG6P, l [ 2H]MeaG6P) were likewise obtained in alternation, but using the G6PDH coupled assay. Values of ( £ c a t ) i V ( £ c a t ) D and (kcJK.M)Hl(kcJKM)D were calculated from the data by dividing the rate for the protio substrate by the rate for the deuterio substrate in each case. Each pair of measurements was repeated at least 7 times, and, the error analysis was based on the standard deviation for each set of data. Al l KIE data are summarized in Tables 3.13 and 3.14. Chapter 3 - GlvA from Bacillus subtilis 149 Substrate (*cat)H/(^cat)D (pH 7.5) KIE ( A c a t / A M ) h / ( A c a t / K\\)x> (pH 7.5) KIE l[2HJ4NPaG6P (3.21) 0.99 ± 0.02 None 1.01 ±0.04 None 2l2H]4NPaG6P (3.22) 1.21 ±0.02 1° 1.28 ±0.07 i° 3I2H]4NPaG6P (3.23) 1.73 ±0.02 1° 1.59 ±0.05 1° l| 2H]MeaG6P (3.26) 0.96 ± 0.02 Inverse or None 0.98 ± 0.04 None 1[2H]4NPPG6P (2.79) 1.00 ±0.01 None 1.00 ±0.06 None 2[2H14NPpG6P (2.80) 3.52 ±0.02 1° 6.4 ±0.5 1° 3[2H14NPpG6P (2.81) 1.70 ±0.02 1° 1.89 ±0.09 1° Table 3.13. Kinetic isotope effect measurements for deuterated substrates of GlvA obtained at pH 7.5. Substrate (ACat)H/(ACat)D (pH 8.4) KIE (ACat/ A"M)H/(ACat/ nxuto (pH 8.4) KIE l| 2H]4NPaG6P (3.21) 0.88 ±0.01 Inverse • 1.0 rt 0.1 None 2| 2H|4NPaG6P (3.22) 1.14±0.01 1° 1.30 ±0.09 1° 3[2H]4NPaG6P (3.23) 1.82 ±0.02 1° 2.2 ±0.3 1° l[ 2H]MeaG6P (3.26) 0.93 ±0.01 Inverse 0.94 ± 0.03 Inverse l[ 2H|4NPpG6P (2.79) 0.89 ±0.01 Inverse 0.99 ± 0.03 None 2[2HJ4NPpG6P (2.80) 1.76 ±0.02 1° 3.2 ±0.2 1° 3[2H]4NPpG6P (2.81) 1.89 ±0.02 1° 2.2 ±0.1 1° Table 3.14. Kinetic isotope effect measurements for deuterated substrates of GlvA obtained at pH 8.4. Section 3.3.7. Mechanistic Analysis via K I E and Linear Free Energy Relationship. Careful analysis of the rather large body of kinetic data on the GlvA-catalyzed hydrolysis of both the a- and p-D-glucopyranosides provides insights not only into the general mechanism followed, but also some clue as to the general shape of the reaction Chapter 3 - GlvA from Bacillus subtilis 150 coordinate diagram. Although the KIEs and Bronsted analyses of the 6-phospho-P-D-glucosides are in agreement with those observed for 6-phospho-oc-D-glucopyranosides, some uncertainties regarding the interpretation of the data for the 6-phospho-P-D-glucopyranosides is merited particularly in light of the fact that GlvA is unable to hydrolyze the glycosidic linkage of naturally occurring phospho-p-D-glucopyranosides. The first general conclusion is that the overall mechanism proposed is fully supported by the kinetic data. Firstly, the measurement of primary KIEs for both the 3[ 2H]4NPaG6P and 3[ 2H]4NPpG6P, at both pH values (kcat and kcJKM), is completely consistent with the proposed oxidation at C3, and with this step being at least partially rate-limiting. Indeed it is difficult to envisage any other mechanism that would give rise to such an isotope effect at C3. Secondly, the measurement of isotope effects that are clearly primary for the 2[ 2H]4NPPG6P is fully consistent with a deprotonation event at C2 that is at least partially rate-limiting. Those measured at C2, for 2[ 2H]4NPaG6P, are considerably smaller but still consistent with such a mechanism. Thirdly, the absence of any dependence of kcat on leaving group ability is consistent with relatively rapid, and non-rate-limiting glycosidic bond cleavage. This is fully supported by the absence of normal secondary deuterium KIEs for the l-[2H]-6-phospho-D-glucopyranoside substrates. Overall, these data strongly support a redox-activated El Cb mechanism for this enzyme. A closer look at the data, and particularly a comparison of A:cat and kcaxIKu data for the 6-phospho-oc- and 6-phospho-p-D-glucopyranosides provides further insights, particularly into the general shape of the reaction coordinate diagrams for the two substrates. It is worth noting that the most useful sets of kinetic isotope effects to evaluate are those determined at the pH optimum for that parameter. This would mean the KIEs on kcat determined at pH 7.5 and the KIEs on kcJKM measured at pH 8.4. Indeed the pairs of values measured in these cases are quite similar, suggesting that the two kinetic parameters may well be reflecting the same chemical step at their respective pH o pt. GlvA catalyzes the hydrolysis of 4NPaG6P and 41MPpG6P with similar kcat values, which is consistent with the notion that steps that do not critically depend on the substrate anomeric configuration are rate-limiting for these two classes of substrate; Chapter 3 - GlvA from Bacillus subtilis 151 namely oxidation at C3 and deprotonation at C2. A key difference in the data for the two substrates is in the values of the KIEs for 2[ 2H]4NPaG6P and 2[ 2H]4NPpG6P. Very large KIEs are seen for 2[ 2H]4NPpG6P, particularly at the lower pH, suggesting that this deprotonation step is more fully rate-limiting than is the case for the 2 f H]4NPocG6P (though surprisingly for this interpretation the KIE measured for 3[2H]4NPpG6P is not correspondingly lowered). Interestingly, at the higher pH value, the KIEs for the 2[2H]4NP[3G6P is lower than that measured at the lower pH value. This was expected since proton abstraction should be faster in the presence of a higher concentration of the conjugate base under more basic conditions, making the other step more rate-limiting. Values of KIEs for 2[ 2H]4NPaG6P are much lower; indeed on the borderline with values of large secondary deuterium KIEs that would be associated with a rate-limiting breakage of the glycosidic bond. However, this latter scenario cannot be the case, at least for kcat, since no dependence of kcat upon aglycone pATa was observed (Figures 3.28 & 3.29). It is therefore probable that these are small primary KIEs whose lower magnitude either reflects a smaller contribution of that step to the rate-limiting process or a highly asymmetric transition state for deprotonation, which results from either early or, more probably, a late transition state. Given its natural substrate specificity, GlvA presumably evolved to cleave 6-phospho-oc-D-glucopyranosides, for which its acid and base catalysts are optimally positioned. When operating on 6-phospho-p-D-glucopyranosides, the acid catalyst assisting C l - O l cleavage is certainly out of position, hence the lack of action on unactivated substrates of this anomeric stereochemistry, such as the MepG6P and C6'P. It is also likely that the aryl aglycone of the 6-phospho-P-D-glucopyranosides may somewhat displace the base catalyst, making deprotonation less efficient, thus more fully rate-limiting. The tempting alternative explanation that the mechanism for cleavage of 6-phospho-a-D-glycosides is more E2-like, given the trans-periplanar arrangement of the C2-H2 bond and the glycosidic bond, is rendered untenable by the absence of any leaving group dependence on kcai. This result is consistent with the ElCb-type mechanism previously proposed for BglT where oxidation of the C3 hydroxyl and deprotonation at C2 are both partially rate-limiting, while cleavage of the C l - O l linkage is rapid and not a rate-determining step. Chapter 3 - GlvA from Bacillus subtilis 152 Because C6'P and MepG6P are not hydrolyzed by GlvA, the Bransted plots generated from kcat (Figures 3.26a & 3.27a) and kcJKM (Figures 3.26b and 3.27b) were only available for leaving groups with pKa values ranging from 3.96-9.99. Both log kcal and log kcJK.M vs pKa at both pH 7.5 and 8.4 produced flat linear free energy relationships. The log kcat/KM vs pKa plots displayed a high level of scattering, which can be attributed to the large variation in KM observed for the aryl 6-phospho-fi-D-glucopyranosides. The flat linear free energy relationship in conjunction with the lack of a secondary KIE for l[ 2H]4NPpG6P ((kcat)H/(kcat)D) suggests that cleavage of the C l - O l linkage is not rate-limiting for aryl 6-phospho-P-glucoside substrates, and is therefore consistent with the proposed El Cb mechanism. The results shown in Figures 3.28a and 3.29a indicate that kcat does not display any significant dependence upon the aglycon leaving group ability at either pH 7.5 or pH 8.4. The plots shown in Figure 3.28b and 3.29b display small dependences of kcat/KM upon leaving group ability. However, comparison and interpretation of data based upon kcJKu requires prior consideration of the meaning of that parameter in this case. Normally kcat/KM is assumed to reflect the first irreversible step, which would appear to be cleavage of the glycosidic bond. However, if this step is fast compared to the prior proton transfer step, as seems to be the case, the deprotonation step would also be rendered essentially irreversible. Likewise, i f deprotonation is faster than the redox process then kcJKu may also reflect contributions from the oxidation step. This could explain why substantial KIEs are seen in kcat/KM for both 2-deuterio and 3-deuterio substrates, which otherwise would not be expected to exhibit primary KIEs if kcJKM purely reflected the bond cleavage step. Indeed, as noted above, when the KIEs on kcat at pH 7.5 and kcat/KM at pH 8.4 are compared, they are indeed very similar within each isotopomer pair, consistent with the same step being reflected. Such a situation demands a general reaction coordinate diagram like that shown in Figure 3.30 for the first half of the reaction, up to the formation of the a,p-unsaturated intermediate. The assignment of energy levels in this diagram is arbitrary: their measurement would be non-trivial, since it would require determination of partition rates for all the species along the coordinate. However, the generalized diagram is a useful way of thinking about rate constants and how they differ for the two substrates. The barrier heights are shown as being similar across the coordinate, as Chapter 3 - GlvA from Bacillus subtilis 153 required by the presence of isotope effects on H3, H2 and, even in one case, also on HI. However the highest barrier for the 6-phospho-a-D-glycopyranoside substrates is that for the oxidation step, while the highest for the 6-phospho-(3-D-glycopyranoside substrates is that for the proton abstraction, as dictated by the large primary KIE measured for 2[ 2H]4NPpG6P. I • Reaction Coordinate Figure 3.30. Graphical representation of the reaction coordinate for the GlvA-catalyzed hydrolysis of a- and p-substrates. The origin of the inverse secondary deuterium KIE measured at high pH for the 1 [ 2H]MeaG6P (on both &c a t and kcJKM) and for 1 [2H]4NPocG6P on & c a t (non-optimal pH) is unclear. Interestingly, the kcat values measured in these three situations are amongst the highest of all the /ccat values measured. This suggested at the outset that a step subsequent 2 3 to C l - O l bond cleavage in which a sp hybridized center is converted to a sp center. For example, addition of water to the Michael acceptor intermediate might be the rate-limiting step. However, methanol competition experiments (not shown) revealed no rate enhancements under conditions where methanol is known to act as a superior nucleophile, as was seen in the use of methanol as a superior acceptor in NMR-based stereochemical outcome determinations.69 Therefore, this step cannot be rate-limiting. Chapter 3 - GlvA from Bacillus subtilis 154 It is important to stress that the ability to hydrolyze both aryl 6-phospho-OC-glucosides and aryl 6-phospho-P-glucosides is unique to some GH4 enzymes, and may only be possible because of the inherent redox-elimination mechanism. The important catalytic residues are those responsible for the redox steps at C3 and deprotonation at C2. Rapid departure of the aglycone may render general acid catalysis expendable in some cases. However, "classical" glycoside hydrolases utilizing nucleophilic displacement mechanisms may have stringent requirements for catalysis and hence absolute geometric arrangement of the catalytic residues around the substrate anomeric center. Therefore, no other glycoside hydrolase families have yet been found to catalyze the hydrolysis of substrates of different anomeric configurations. Section 3.3.8. Potential Inhibitors. A number of compounds were tested as inhibitors of GlvA. These include C6'P (2.2), 4NP2deoxypG6P (2.96), 4NP2FpG6P (2.100), and MepG6P (2.58). Each potential inhibitor was first tested as a substrate under standard GlvA assay conditions. The enzyme concentration was increased up to 10 times necessary for the hydrolysis of 4NPaG6P, but none of the inhibitors were hydrolyzed by GlvA. Approximate Kt values for the potential inhibitors were determined by measuring the reduction in the rate of GlvA-catalyzed hydrolysis of a fixed concentration of the chromogenic substrate 4NPaG6P (2.17), in the presence of varying concentrations of inhibitor. The experiments were repeated at different concentrations of 4NPaG6P. The data were graphed on a Dixon plot (1/v vs [competitive inhibitor]). The lines from the different concentrations of 4NPaG6P intersected in the Dixon plots (Appendix 5) at an inhibitor concentration equal to -Ki The results of the inhibition studies are summarized in Table 3.15 Chapter 3 - GlvA from Bacillus subtilis 155 Inhibitor Approx imate K, (uM) Type of inhibit ion C 6 ' P (2.2) 300 Competitive MePG6P (2.58) 750 Competitive 4NP2deoxypG6P (2.96) 100 Competitive 4 N P 2 F 0 G 6 P (2.100) 150 Competitive Table 3.15. Approximate AT; values of the potential inhibitors of GlvA. The results from the competitive inhibition studies confirm that 4NP2deoxypG6P (2.96), 4NP2FpG6P (2.100), C6'P (2.2) and MepG6P (2.58) bind to the active site of GlvA. GlvA does not hydrolyze 4NP2deoxypG6P but binds it somewhat less tightly than 4NPpG6P (2.13) and 4NPaG6P (2.17), assuming KM = K D . This is most likely because the 2-hydroxyl is important for binding to the M n 2 + . Likewise, the fact that neither 4NP2deoxypG6P nor 4NP2FpG6P were hydrolyzed probably related to the poor stabilization of the enediolate intermediate in the absence of the second hydroxyl for metal coordination. Because GlvA and BglT are proposed to utilize a common mechanism for hydrolysis, it was not surprising that these competitive inhibitors of BglT were also able to inhibit GlvA activity. Interestingly, both C6'P (2.2) and MePG6P (2.58) were also competitive inhibitors of GlvA, indicating that they do indeed bind to the active site. As noted, the lack of substrate activity in these cases is probably a consequence of the absence of a correctly located acid catalyst, since 6-phospho-P-D-glucopyranosides with good leaving groups that do not need acid catalysis are indeed cleaved. The fact that GlvA binds C6'P more tightly than MePG6P can be rationalized by the possibility of favorable interactions in the +1 subsite with the sugar ring in C6'P, while no such interactions are possible in the case of MepG6P. Section 3.4 Proposed Mechanism and Conclusions. From the linear free energy relationship and KJEs, GlvA is proposed to utilize an El Cb mechanism for the hydrolysis of aryl 6-phospho-a- and aryl 6-phospho-p-D-Chapter 3 - GlvA from Bacillus subtilis 156 glucopyranosides. The proposed mechanism is shown in Figure 3.30. The proposed El Cb mechanism of GlvA is evidently the same as that described for BglT. Importantly, the key catalytic residues for this mechanism are those that interact with the C2 and C3 positions of the substrate, rather than those that interact with the anomeric center or the glycosidic oxygen. Indeed, acid/base catalysis of aglycone departure and water attack will be of much less importance for this mechanism than for the standard nucleophilic displacement mechanisms of "classical" glycosidases. This likely explains why GH4 contains both a- and p-glycosidases, and why some members, such as GlvA, can hydrolyze both a- and P-glycosides containing activated leaving groups. Figure 3.31. Proposed E l c b mechanism for GlvA. Chapter 4 - Thioglycosides and Elimination Mechanisms 157 C h a p t e r 4 T h i o g l y c o s i d e s a n d E l i m i n a t i o n M e c h a n i s m s Chapter 4 - Thioglycosides and Elimination Mechanisms 158 Section 4.1. Introduction - Thioglycosides and "Classical" Glycosidases. Considerable efforts have been extended towards the development of glycosidase inhibitors, both as stable substrate analogues for structural and mechanistic studies and for potential therapeutic and industrial applications.'81"186 As illustrated in Figure 4.1, some of the commonly known glycosidase inhibitors9 8"1 0 2 include acarbose,94'96 1-deoxynojirimycin,9 4 , 9 6 carbodiimides,97 and fluorosugars.93 Amongst those developed, thioglycosides, in which the glycosidic oxygen has been replaced by a sulfur atom (Figure 4.2), have proved to be stable analogues of the ground state substrate and have been employed in a number of insightful structural studies.' 3 0 ' 1 6 4 ' 6^' 1 8 7" 1 9 0 O H O H O H N=C=N H O ' H F " O R carbodiimides 2-deoxy-2-fluoro glucoside Figure 4.1. List of common glycosidase inhibitors. O H H O -H O R 2-deoxy-2,5-difluoro glucoside O H O H S-glucoside O-glucoside or thioglucoside Figure 4.2. Comparison of O-glucoside and thioglucoside. Chapter 4 - Thioglycosides and Elimination Mechanisms 159 Glycosidases are known to effect hydrolysis by acid/base catalyzed nucleophilic 22 24 27 displacement mechanisms involving oxocarbenium ion-like transition states. ' " The departure of the aglycone in both the "classical" retaining and inverting glycosidase 22 24 27 mechanisms is proposed to require general acid catalysis. ' " Consequently, the resistance of the thioglycosidic bond to cleavage by "classical" glycosidases has been ascribed to the lower proton affinity of sulfur than oxygen, resulting in inefficient general acid catalysis to the departing aglycone.1 6 8"1 7 0 One special exception is the myrosinases that are mainly of plant or igin, 3 5 , 3 6 ' 1 9 1 " 1 9 3 although fungal194,19"^ and bacterial1 9 6 myrosinases are also known. These enzymes serve the specific purpose of hydrolyzing glucosinolar.es, which are anionic 1-thio-p-glucosides. The general glucosinolate structure is shown in Figure 4.3. R = alkyl/thioalkyl, aliphatic, olefins, alcohols, aromatic, ketones, or indole Figure 4.3. General glucosinolate structure. By sequence alignment, the myrosinases are associated with the Family 1 glycosidases, which catalyze the hydrolysis of p-0-glycosides with net retention of the substrate anomeric configuration.12'3"'36 The ability of the myrosinases to cleave thioglycosidic linkages has been attributed to the inherently good leaving group ability of the aglycone in these compounds. Consequently, they do not require general acid assistance for departure of the aglycone moiety and concomitant formation of the glycosyl-enzyme intermediate.35 Thus, in order to minimize unfavorable electrostatic interactions between the anionic substrate and the glutamate residue, the glutamate that functions as the general acid/base is replaced by a glutamine residue.3^'36 An exogenous ascorbate anion has been proposed to function as the general base catalyst, which activates a water molecule for the nucleophilic attack on the glycosyl-enzyme intermediate.34"36 Somewhat similarly, the GH84 human 0-GlcNAcase is proposed to Chapter 4 - Thioglycosides and Elimination Mechanisms 160 hydrolyze activated thioglycosides through a very dissociative transition state and without general acid catalysis to protonate the thiolate leaving group.1 6 6 Section 4.2. Cleavage of C-S linkages via Eliminative Mechanisms. Given the proposed redox-elimination-addition GH4 mechanism rather than one involving nucleophilic displacements, this family of enzymes could potentially cleave thioglycosidic linkages. In fact, enzymes that routinely cleave carbon-sulfur linkages do so through elimination mechanisms.197 As shown in Figure 4.4, excellent examples include cysteine C(3-Sy lyases, 1 9 8 ' 1 9 9 (3-cystathionase,200 S-(p-aminoethyl)-cysteine lyase, 5-alkylcysteine lyase, ' S-adenosyl homocysteine hydrolase (AdoHCyase), 2 0 4 and the more recently uncovered ribosyl homocysteinase.205 (a) N H ; Chapter 4 - Thioglycosides and Elimination Mechanisms 161 Chapter 4 - Thioglycosides and Elimination Mechanisms 162 Chapter 4 - Thioglycosides and Elimination Mechanisms 164 (e) Figure 4.4. Proposed mechanisms of enzymes that catalyze C-S bond cleavage, (a) cysteine Cp-Sy lyase;1 9 8 (b) methionine-y-lyase; (c) P-cystathionase;200 (d) S-adenosyl homocysteine hydrolase (AdoHCyase); 2 0 4 (e) ribosyl homocysteinase, LuxS . 2 0 5 ' 2 0 6 Most significantly, AdoHCyase 2 0 4 utilizes an N A D + cofactor to oxidize the ribose C3 hydroxyl group of S-adenosyl homocysteine, thereby lowering the pKa of the C4 proton and facilitating deprotonation hence a,fj-elimination of the thiol. The Michael-acceptor then undergoes 1,4-nucleophilic attack by a water molecule, followed by reduction of the C3 ketone by the "on-board" N A D H , yielding adenosine and homocysteine and returning the enzyme to its initial catalytic state. On the other hand, the mechanism utilized by LuxS is fascinating because it formally does not involve a redox activation step.2 0 1"2 0 8 Intriguingly, chondroitin A C lyase, known to utilize an El Cb mechanism, has also been shown to cleave C-S linkages readily.1 7 2 Chapter 4 - Thioglycosides and Elimination Mechanisms 165 The proposed GH4 mechanism of glycoside hydrolysis is essentially analogous to that of AdoHCyase, involving redox chemistry and anionic transition states. The similarities of these mechanisms raised the question as to whether GH4 enzymes, in contrast to all other glycosidases, can efficiently hydrolyze thioglycosidic bonds. Not only is this of fundamental mechanistic interest, but also it is highly relevant to the design of specific glycosidase inhibitors as potential therapeutics. Section 4.3. Thioglycosides - Substrates for BglT from Thermotoga maritima. As discussed in Section 2.3.9.4, S-4NPpG6P (2.102) is not an inhibitor of BglT; rather the compound was readily hydrolyzed. This is not surprising in light of the many lyases that have evolved to cleave C-S linkages. l 7 2- 1 9 s" 2 ( b Various concentrations of S-4NPPG6P were assayed with BglT via the direct UV-vis assay (km&x - 400 nm, Ae = 10,846 ivr'crrf'). The kinetic parameters were determined on the basis of a direct fit of the data to the Michaelis-Menten equation (Appendix 3), and the data are compared to those obtained for the corresponding (9-linked chromogenic substrate, 4NPPG6P (2.13), in Table 4.1. While these results are suggestive, the use of an activated aryl leaving group could be misleading in this case, because previous studies with a "normal" Family 1 p-glucosidase, Abg , 1 7 3 and a GH84 human O-GlcNAcase 1 6 6 have shown that activated aryl thioglycosides could indeed be cleaved reasonably efficiently. In order to properly test whether Family 4 enzymes can cleave thioglycosides, it was necessary to study the hydrolysis of non-activated thiodisaccharide substrates that closely mimic the natural substrate C6'P (2.2). Accordingly, thiodisaccharide analogues of the natural substrate for the 6-phospho-p-glucosidase were prepared. In order to simplify assays, as well as synthetic routes, the following compounds were synthesized: 4-nitrophenyl 4-deoxy-4-thio-6'-phospho-p-cellobioside (S'-4NPpC6 /P or 4.3) and 4-deoxy-4-thio-6'-phospho-p-cellobioside (S-C6'P or 4.5). In addition, the kinetic parameters for the oxygen counterparts had been determined using convenient kinetic assays as described in Section 2.2. Chapter 4 - Thioglycosides and Elimination Mechanisms 166 4.4 (S-C) 4.5 (S-C6T) Scheme 4.1. Enzymatic synthesis of S-4NPpC6'P (4.3) and S-C6'P (4.5). (i) Abg E l 70Q; (ii) cellulase Onozuka R- l 0 from Trichoderma viride; (iii) BglK, ATP. Synthesis of 5'-4NP(3C6'P (4.3) was achieved by the chemoenzymatic route shown in Scheme 4.1 starting with 4-nitrophenyl 4-deoxy-4-thio-(3-D-glucopyranoside (4.1). The key step involved an enzymatic coupling using the thioglycoligase technology recently developed in our group to produce the thiodisaccharide, 4-nitrophenyl 4-deoxy-4-thio-p-cellobioside (S-4NpPC or 4.2). 2 0 9 The free disaccharide, 4-deoxy-4-thio-cellobiose (S-C or 4.4) (Scheme 1), was prepared from 5-4NPpC by removal of the 4-nitrophenyl group at C l using the commercially available cellulase Onozuka R-10 from Trichoderma viride. The disaccharide products 5-4NPPC and S-C were then selectively phosphorylated at 06 ' using the kinase BglK. C6'P (2.2) and 4NPC6'P (2.10) had been prepared previously as described in Section 2.2. As shown in Figure 4.5, G6P (2.3) is a product in the BglT-catalyzed hydrolysis of each of C6'P (2.2), 4NPpC6'P (2.10), 5-4NPpC6'P (4.3) and S-C6'P (4.5). Thus, the kinetic parameters for the hydrolysis of all four compounds can be determined by the Chapter 4 - Thioglycosides and Elimination Mechanisms 167 G6PDH coupled assay, in which the formation of G6P was coupled to the reduction of N A D P + to NADPH (kmM = 340 nm, 8NADPH = 6220 cm"'M"'). .OP032- OH HO' OH X 0 H 2.2 (C6'P): X = 0; R = H 2.10(4NPpC6'P) : X = 0; R = 4-nitrophenyl 4.3 (S-4NPpC6'P) : X = S; R = 4-nitrophenyl 4.5 (5-C6'P) : X = S; R = H OH HX" HO- OR OH 2.4 (Glc): X = O; R = H 2.11 (4NPfjGlc) : X = 0, R = 4-nitrophenyl 4.6 : X = S; R = 4-nitrophenyl 4.7 : X = S; R = H ^ O P O 3 2 -HO-H O ^ ^ - r - \ ^ < OH 2.3 (G6P) •OH HO' N A D P H 2.5 Figure 4.5. Schematic for the BglT-catalyzed hydrolysis of C6'P (2.2), 4NP[iC6'P (2.10), 5-4NP(iC6'P (4.3), and 5-C6'P (4.5) using the G6PDH coupled assay, (i) BglT; (ii) G6PDH, N A D P + . Each substrate displayed Michaelis-Menten saturation kinetics. The G6PDH coupled assay was used to assay 5,-4NP(3C6'P (4.3) and iS-C6'P (4.5), at varying substrate concentrations. Kinetic parameters (summarized in Table 4.1) were determined on the basis of a direct fit of the data to the Michaelis-Menten equation (Appendix 3). To further ensure that the release of thiosugars, 4.6 and 4.7, did not interfere with the G6PDH coupled assay, approximate K\ values for the potential inhibitors were determined by measuring the reduction in the rate of BglT-catalyzed hydrolysis of a fixed concentration of the chromogenic substrate 4NPpG6P (2.13), in the presence of varying concentrations of each disaccharide. The experiments were repeated at different concentrations of 4NPpG6P. The data were graphed on a Dixon plot (1/v vs [competitive inhibitor]). The lines from the different concentrations of 4NP(3G6P intersected in the Dixon plots (Appendix 5). Additionally, a horizontal line drawn through 1/Kma* in the Dixon plot intersects the experimental lines at an inhibitor concentration equal to -Afj. The results of the inhibition studies are summarized in Table 4.1. The approximate K, values corresponded well with the /CM values obtained from the coupled assay, indicating Chapter 4 - Thioglycosides and Elimination Mechanisms 168 that the coupled assay was reliable for the kinetic characterization of all disaccharides used in this study. Substrate A'cat (S ) A M (UM) A'ca/A|Vl (s1 mM'1) Ai(uM) 4NPpG6P (2.13) 1.9 41 46 — S-4NPPG6P (2.102) 1.6 41 39 — 4NPpC6'P (2.10) 0.017 1.0 8.0 17 1.0 S-4NPpC6'P (4.3) 0.072 9.0 5.0 C6'P (2.2) 0.61 69 8.8 70 S-C6T (4.5) 0.53 37 14 33 Table 4.1. Summary of kinetic parameters for the hydrolysis of O- and S-glucosides by BglT. Error analysis was performed using the program GraFit. The error was < 15% in each case. The kinetic parameters for the disaccharide substrates are similar, with the kC3X and K M values for S-4NPpC6'P (4.3) being larger than those for 4NPpC6'P (2.10), resulting in similar kcJKu values. Meanwhile, the kc3t and KM for C6'P (2.2) are both slightly larger than those for S-C6'P (4.5), resulting in similar kcJKM values. The similarities of the KM and K, values in each case confirms that reaction is indeed occurring through the same active site. Furthermore, product analyses via NMR and MS showed that the expected products (G6P and 4-deoxy-4-thio-D-glucopyranose 4.6) are indeed formed when 5"-C6'P (4.5) is reacted with BglT. In addition, solvent deuterium incorporation into the product (resulting in 2[2H]G6P (2.22)) strongly suggests that BglT utilizes the same mechanism in the hydrolysis of thiodisaccharides as that of 0-glycosidic linkages. Overall, there is no significant difference between the rates of hydrolysis of O-Chapter 4 - Thioglycosides and Elimination Mechanisms 169 and 5-glycosidic linkages as catalyzed by BglT. Hence, we have shown that BglT is the v first glycosidase capable of efficiently cleaving unactivated thioglycosides. Section 4.4. Conclusions. These findings clearly indicate a fundamental difference in the BglT mechanism compared to that of most glycosidases, which cannot cleave unactivated thioglycosides. The efficient hydrolysis of thiodisaccharides therefore provides further supporting evidence for the novel glycosidase mechanism proposed for Family 4 enzymes. Cleavage of such thioglycosides at rates comparable to those of their oxygen counterparts is quite reasonable for the anionic mechanism proposed, but not for reactions via oxocarbenium ion-like transition states. The majority of glycosidases do not hydrolyze thioglycosidic linkages, and those that have been reported to possess thioglycosidase activity only react with thioglycosides containing highly activated leaving groups, as clearly demonstrated by the Bransted plot determined for thioglycoside hydrolysis in the case of O-GlcNAcase, which is stabilized in this case by the electron-withdrawing groups on the aglycone.166 The cleavage of the activated thioglycosidic linkages by O-GlcNAcase does not rely on general acid catalysis and there is substantial negative charge development on the sulfur atom at the transition state.166 Since the cleavage of the glycosidic linkage is not rate-limiting for BglT, 6 7 it is reasonable that substitution of the glycosidic oxygen with a sulfur atom does not significantly affect the overall rate. On the basis of these results, it is clear that thioglycosides should not be employed in any inhibition strategies for Family 4 enzymes whether this be for structural or mechanistic studies or as part of any biological control strategy. Indeed, the ability to cleave a thiodisaccharide could be a useful diagnostic of whether a new glycosidase belongs to Family 4. Finally, these findings raise the question of whether C6'P (2.2) is in fact the natural substrate for BglT, or whether this enzyme has evolved to cleave some other, as yet undiscovered, thioglycoside substrate. The striking similarities between the mechanisms utilized by GH4 enzymes and by AdoHCyase suggest that other possibilities should be kept in mind. Chapter 5 - Materials and Methods 170 C h a p t e r 5 M a t e r i a l s a n d M e t h o d s Chapter 5 - Materials and Methods 171 Section 5.1. Generous Gifts. The enzymes BglK and GlvA were generous gifts from Dr. John Thompson. 4-Nitrophenyl (3-D-thioglucopyranoside was donated by Dr. Michael Jahn. Dr. Hongming Chen kindly provided 4-nitrophenyl 4-deoxy-4-thio-(3-D-glucopyranoside. Dr. Johannes Miillegger generously provided the E170Q Abg mutant enzyme. Dr. Chris A. Tarling donated 1,5-anhydroglucitol 6-phosphate, which was synthesized from 1,5-anhydroglucitol. 4-Nitrophenyl 2-deoxy-p-D-arabino-hexopyranoside (4-nitrophenyl 2-deoxy-p-D-glucopyranoside) was previously synthesized by Dr. M . N . Namchuk 1 6 0 and 4-nitrophenyl 2-deoxy-2-fluoro-p-D-glucopyranoside was prepared by Dr. I. P. Street.163 The plasmid containing the T. maritima genomic D N A was obtained from ATCC (ATCC 43589) was cloned into Pet22b vector and was generously provided by our collaborators Dr. Annabelle Varrot and Dr. Gideon Davies. p-Glucosidase from Agrobacterium sp (Abg) was cloned and expressed by Ms. Karen Rupitz. Section 5.2. General Synthesis. Unless otherwise stated, all reagents were purchased from commercial suppliers (Sigma, Aldrich, Fluka, CMS, and Cambridge Isotopes) and were used without further purification. Flash chromatography was performed with 230-400 mesh silica gel. TLC was performed on Merck pre-coated 60 F-254 silica plates and visualized using UV radiation at 254 nm and further developed with 10% H 2 S 0 4 in MeOH or 10 % ammonium molybdate in 2 M H2SO4 followed by heating. Al l N M R spectra were collected with a Bruker AC200 (200 MHz), a Bruker AV-300 (300 MHz), a Bruker WH-400 (400 MHz), a Bruker AV-400 (400 MHz), or a Bruker AMX-500 (500 MHz) spectrometers. Chemical shifts are reported on the 8 scale in parts per million and were referenced to solvents such as C D C I 3 , C D 3 O D , and D 2 0 . Al l 3 1 P NMR spectra were externally referenced to 85 % H3PO4 in H 2 0 . 1 9 F NMR signals were externally referenced to C F 3 C O O H . The abbreviations in describing multiplicity are: s-singlet, d-doublet, t-triplet, and m-multiplet. Mr. Minaz Akber Lakha (Mass Spectroscopy Lab, Department of Chemistry, University of British Columbia) performed all elemental analyses. Melting points were recorded using a Laboratory Devices Mel-Temp II melting point apparatus and are uncorrected. Anhydrous solvents were prepared as follows: CH2CI2 and pyridine Chapter 5 - Materials and Methods 172 were distilled over CaH 2 ; THF was distilled over sodium benzophenone; methanol was distilled over magnesium and iodine. DMF and THF were dried successively over 4 A molecular sieves (3 x). Deionized water, purified with a Millipore Direct-Q™ 5 Ultrapure Water System, was used for all aqueous solutions. Low and high resolution mass spectra were collected by the mass spectrometry laboratory at the University of British Columbia. Section 5.2.1. General Methods. Section 5.2.1.1. General Acetylation Methodology. General acetylation procedures were taken from known methods. There are two general methods, one of which yields a mixture of a- and (3-anomers and the other yields the p-anomer as the major product Anomeric Mixture. The free sugar (1.0 equiv., final concentration ~ 0.05 M) was added to a stirred, ice cold mixture of acetic anhydride and pyridine (in a ratio of 1:1.3). Once the D-glucose was completely dissolved, the solution was stirred at room temperature overnight. Upon completion, as indicated by TLC analysis, the reaction mixture was concentrated under reduced pressure. The viscous syrup was then diluted with C H C I 3 , washed with 1 M HC1 (1 x), water (1 x), saturated NaHC0 3 (1 x), brine (1 x), dried over MgS04, filtered, and concentrated under reduced pressure. Pure product was obtained as white needles by recrystallization from ethanol or from a mixture of ethyl acetate and petroleum ether. $-Anomer. A solution of sodium acetate (25 g) and acetic anhydride (350 mL) was brought to a boil and the sugar (50 g) was then slowly added without heating. The acetylation reaction was exothermic, and the reaction mixture continued boiling once reaction had been initiated. After all the sugar had been added and the reaction had subsided, the reaction mixture was reheated to a full boil. The solution was immediately cooled and poured into ice-water. After 3 hours with occasional stirring, crystalline material was isolated by suction filtration. Pure product was obtained by recrystallization from ethanol. Chapter 5 - Materials and Methods 173 Section 5.2.1.2. General Procedure for Synthesis of Glycosyl Bromides.210 The starting material (1.0 equiv.) was dissolved in a solution of 33 % HBr in acetic acid (in a ratio of 1 g of starting material to 1 mL of HBr solution) and 0.1 equiv. of A C 2 O was added to the reaction mixture. The reaction vessel was tightly stoppered and the reaction was stirred at room temperature until completion by TLC analysis. Ice cold C H 2 C I 2 was added and the solution was washed with ice cold H 2 O (2 x) and sat. NaHCO} until the aqueous layer is neutral. The organic fraction was dried over MgSO*?, filtered, and concentrated under reduced pressure. The syrup was readily recrystallized in either EtOH or from a mixture of petroleum ether/EtOAc to yield a pure product as a white crystalline solid. The product was stable for several months when stored at -20 °C. Section 5.2.1.3. General Zemplen Deprotection.126'2" The starting material was dissolved in dry methanol and a catalytic amount of sodium methoxide was added. The reaction mixture was stirred at room temperature until TLC indicated reaction completion. The solution was then neutralized with Amberlite IR-120 (H+) resin. The resin was removed by filtration and the filtrate was concentrated under reduced pressure. For most compounds, the reaction yielded a relatively pure product and did not require further purification. Section 5.2.1.4. General Deprotection using HCI(g).212 The reactant was dissolved in dry methanol and the solution was cooled to 0 °C. HCl(g) was bubbled into the reaction mixture, then stirred at 4 °C until the reaction had reached completion by TLC analysis. The reaction mixture was then concentrated under reduced pressure. For most compounds, the reaction yielded a relatively pure product and did not require further purification. Section 5.2.1.5. Koenigs-Knorr Reaction. The a-bromide 2.26 (1.0 equiv.) was dissolved in acetone (such that [2.26] = 1 M). A 1 M solution of the appropriate phenol (2.0 equiv.) (in 1 M NaOH) was added to Chapter 5 - Materials and Methods 174 the reaction. The reaction mixture was stirred at room temperature overnight, and concentrated under reduced pressure upon reaction completion by TLC analysis. The resulting paste was redissolved in water and extracted with CHCL (3 x). The organic layer was then washed with sat. NaHC0 3 (3 x), dried over MgS0 4 , filtered, and concentrated under reduced pressure. The resulting syrup was recrystallized in EtOH or purified via flash chromatography. The Koenigs-Knorr reaction was utilized when the product p-glucopyranoside was desired. Section 5.2.1.6. Stannic Chloride-Catalyzed Glucosidation. The appropriate phenol reagent was dissolved in dry toluene and concentrated under reduced pressure. The procedure was repeated 3 times and the phenol was kept under vacuum overnight to remove any moisture from the commercial phenol reagents. The starting material (usually 1,2,3,4,6-penta-O-acetyl-p-D-glucopyranose) (1.0 equiv.) was dissolved in dry dichloromethane (volume chosen such that the concentration of the starting material was at 0.1 M), and SnCL (2.2 equiv.) and the appropriate phenol (1.5 equiv.) was added. The reaction mixture was stirred under No (g) at 70 °C for approximately 5 hours. Upon completion, the reaction mixture was cooled to 0 °C. An ice cold solution of saturated NaHC0 3 was added to the solution until the bubbling subsided. The mixture was extracted with 1 x C H 2 O 2 . Then, the organic layer was washed with sat. NaHC0 3 (4 x), dried over MgSC»4, filtered, and concentrated under reduced pressure. The crude product was purified by flash chromatography. The stannic chloride glucosidation yields both a- and (3-glucopyranosides, but the conditions described above have been optimized to maximize the yield of the a-anomer. Section 5.2.1.7. General BglK-Catalyzed Phosphorylation of P-D-Glucosides.123 ATP (1.1 equiv.) was dissolved in deionized water (volume was chosen such that the [ATP] ~ 40 mM) and was adjusted to pH 7.5 using 3 M N H 4 O H . The starting material (1.0 equiv.) was dissolved in 50 mM HEPES (pH 7.5) (volume was chosen such that the [starting material] ~ 400 mM), and this solution was added to the ATP solution. The enzymatic reaction was initiated by the addition of ~ 0.5 mg BglT, and the reaction Chapter 5 - Materials and Methods 175 mixture maintained at pH 7.5 with periodic adjustments via the addition of 3 M N H 4 O H . After 24 hours at room temperature, the solution was adjusted to pH 8.4 and 0.75 M Ba(OAc) 2 (3.0 equiv.) was added. The barium salts of ADP and of the remaining ATP formed fluffy white precipitates, which were removed by centrifugation. The supernatant was filtered through a 0.22 pm Millipore filter. EtOH (4 volumes) was added to the resulting solution and was stirred at 4 °C overnight. The solid product Was collected as its barium salt by centrifugation. The product was redissolved in 5 mL of deionized water and passed through an Amberlite IR-120 (Na+) column twice and lyophilized. The resulting product was purified via HPLC (C18 preparative column) with 5 % acetonitrile in water as the elution solvent. The pure fractions were pooled and lyophilized. Section 5.2.1.8. General Chemical Phosphorylation Using Diphenyl Chlorophosphate. The starting material (1.0 equiv., final concentration = 0.25 M) was dissolved in dry pyridine and cooled to 0 °C. Diphenyl chlorophosphate (1.1 equiv.) was added dropwise to the reaction mixture. The mixture was allowed to warm up to room temperature and was stirred under positive N 2 (g) overnight. 1 M H Q was added to the reaction mixture and the product was extracted into EtOAc (3 x). The organic fractions were pooled and washed with sat. NaHCCh (1 x), sat. NaCl (1 x), dried over MgS04, filtered, and concentrated under reduced pressure. The product was purified via flash chromatography. Section 5.2.1.9. General Procedure for Diphenyl Phosphate Deprotection. The procedure utilized was that described by Cawley et a l . 2 1 3 The starting material (1.0 equiv., final concentration = 0.01 M) was dissolved in dry methanol. Hydrogenation was performed in the presence of a catalytic amount of Pt0 2 . Upon completion by TLC analysis, the catalyst was removed by filtration through a Celite cake, and the filtrate was neutralized by 3 M NH 4 OH. The solution was concentrated under reduced pressure, and redissolved in a minimum amount of H 2 Q, passed through an Chapter 5 - Materials and Methods 176 Amberlite IR-120 (Na+) column twice, and then lyophilized to yield a relatively pure compound. Section 5.2.1.10. General Procedure for Chemical Phosphorylation using Phosphorus Oxychloride. 1 3 4 The starting material (1.0 equiv.) was dissolved in PO(OCH3)3 to a concentration of 0.8 M , and the solution was cooled to 0 °C. POCl 3 (1.0 equiv.) and H 2 0 (0.2 equiv.) were added to a second equal volume of PO(OCH3)3, and this mixture was cooled to 0 °C and added dropwise to the starting material. The reaction was stirred at 0 °C for several hours until completion by TLC analysis. The reaction mixture was adjusted to pH 9 using ice cold 3 M N H 4 O H , 0.75 M Ba(OAc) 2 (3 equiv.) and four volumes of EtOH were added to the reaction mixture, and the product precipitated as its barium salt. The product was collected by centrifugation and the supernatant was discarded. The precipitate was redissolved in a minimal volume of H 2 0 and passed through an Amberlite IR-120 (Na+) column twice and lyophilized. The resulting product was purified via HPLC (C18 preparative column) with 5 % acetonitrile in water as the elution solvent. The pure fractions were pooled and lyophilized. Section 5.2.2. Synthesis and Compound Characterization. Cellobiose 6'-phosphate (C6'P or 2.2). C6'P (150 mg, 0.44 mmol, 1.0 equiv.) was synthesized from cellobiose (2.1) by enzymatic phosphorylation with BglK (Section 5.2.1.7). The compound was purified via a C18 reverse phase HPLC column and lyophilized to give white crystals (41 mg, 0.088 mmol, 20 %): ' H N M R (400 MHz, D 2 0 ) 5 5.05 (d, J a l , a 2 3.7 Hz, aHl ) , 4.49 (d, Jp, j P 2 8.0 Hz, pHl) , 4.34 (d, J , . 2 . 7.9 Hz, H l % 3.85-3.64, 3.48-3.33, 3.19-3.11; , 3 C N M R (100 MHz, D 2 0) 5 102.73 (a,pCl % 95.58 (pCl), 91.66 (aCl) , 79.25, 79.13, 75.24 (d, Jy,P 6.9 Hz, C5% 74.89, 74.63, 74.21, 73.68, 73.21, 71.23, 71.01, 69.95, 68.89, 68.86, 62.73 (d, Chapter 5 - Materials and Methods 177 J6'.p 3.8 Hz, C6'), 60.03, 59.87; 3 I P N M R (162 MHz, D 2 0 ) 5 4.91 (t, JH6-,p 5.7 Hz); ESI-MS (high res.) m/z: calc. for [Ci 2H 220i4pNa 2] + 467.0543; Found: 467.0545; Anal. calc. For C i 2 H 2 iOi4PNa 2 -2H 2 0: C, 28.70; H, 5.02; Found: C, 28.61; H, 5.21. 1,2,2:3,3:4 ',6,6 -Octa-O-acetyl-cellobiose (2.6). Compound 2.6 was synthesized from cellobiose (2.1) (10.0 g, 29.3 mmol, 1.0 equiv.) via the general acetylation procedure using A c 2 0 and pyridine (Section 5.2.1.1). The final product was recrystallized in EtOAc and petroleum ether as a white crystalline solid in quantitative yield and was not further purified (17.1 g, 25.2 mmol, 86 %). ' U N M R (300 MHz, CDCI3) 8 6.17 (d, J 3.7 Hz), 5.59 (d, J 8.3 Hz), 5.36 (t, J 9.8 Hz), 5.18-4.94 (m), 4.84 (t, J 8.2 Hz), 4.46-4.38 (m), 4.29 (dd, J 12.5, 4.4 Hz), 4.07-3.95 (m), 3.79-3.58 (m), 2.04, 2.01, 1,95, 1.93, 1.90 (8 Ac); ESI-MS (low res) m/z: calc. 678; Found: 701.2 [M + Na] +. 2,2 ',3,3:4:6,6-Hepta-0-acetyl-a-cellobiosyI bromide (2.7). The oc-bromide 2.7 (2.6 g, 3.9 mmol, 1.0 equiv.) was synthesized from 2.6 by reaction in a solution of HBr dissolved in acetic acid, using the general procedure described in Section 5.2.1.2 The product was recrystallized from EtOAc and petroleum ether as a white solid (2.3 g, 3.3 mmol, 85 %): ' H N M R (400 MHz, CDC13) 5 6.49 (1 H, d, J,, 2 4.1 Hz, HI), 5.49 (1 H, t, J 2 , 3 = J 3 , 4 9.6 Hz, H3), 5.11 (1 H, t, h,y = h\4- 9.3 Hz, H3-), 5.04 (1 H, t, J 3 ' , 4 ' = J 4 . 5 ' 9.5 Hz, H4'), 4.90 (1 H, t, h',r = h'.r 9.3 Hz, H2'), 4.72 (1 H, dd, J,, 2 4.1, J 2,3 10.0 Hz, H2), 4.52-4.48 (2 H, m, HI ' , H6 a), 4.33 (1 H, dd, J«sa',6b' 12.5, JyM> 4.4 Hz, H6 a '), 4.18-4.11 (2 H, m, H5, H6 b), 4.01 (1 H, dd, J^b-12.5, iy.sw 2.0 Hz, H6 b '), 3.80 (1 Br Chapter 5 - Materials and Methods 178 H, t, J3,4 = J4.5 9.6 Hz, H4), 3.66-3.62 (1 H, m, H5), 2.10 (3 H, 1 s, Ac), 2.05 (6 H, 1 s, 2 Ac), 2.00 (6 H, 1 s, 2 Ac), 1.97 (3 H, 1 s, Ac), 1.94 (3 H, 1 s, Ac); ESI-MS (low res) calc: 698, 700; Found: 721.0, 723.1 [M+Na]+. 4-Nitrophenyl 2,2 ',3,3 ',4:6,6-hepta-0-acetyl-j3-cellobioside (2.8). r i f l r OAc The a-bromide (2.7) (2.3 g, 3.3 mmol, 1.0 equiv.) was reacted with 4-nitrophenol using the FCoenigs-Knorr reaction described in Section 5.2.1.5. The product was recrystallized from EtOAc and petroleum ether as a pale yellow solid (1.4 g, 1.8 mmol, 54 %): ' H N M R (400 MHz, CDC13) 8 8.18 (2 H, d, J Ar2.Ar3 = W A H . 9.2 Hz, Ar3, Ar5), 7.03 (2 H, d, J A r 2 ,Ar3 = J A r 5 .Ar6 9.2 Hz, Ar2, Ar6), 5.04 (1 H, t, J 8.8 Hz), 5.22-5.12 (3 H, m), 5.06 (1 H, t, J = J 9.7 Hz, H4), 4.93 (1 H, t, J 8.0 Hz), 4.53-4.51 (2 H, m), 4.36 (1 H, dd, J 12.5, J 4.4 Hz), 4.11 (1 H, dd, J 12.0, J 5.5 Hz), 4.04 (1 H, dd, J 12.5, J 2.1 Hz), 3.88-3.79 (2 H, m), 3.64-3.68 (1 H, m), 2.08 (3 H, 1 s, 1 Ac), 2.07 (3 H, 1 s, 1 Ac), 2.04 (6 H, 1 s, 2 Ac), 2.03 (3 H, 1 s, 1 Ac), 2.00 (3 H, 1 s, 1 Ac), 1.97 (3 H, 1 s, 1 Ac); ESI-MS (low res) m/z: calc: 757; Found: 780.0 [M+Na]+. 4-Nitrophenylfi-cellobioside (2.9). Zemplen deprotection (Section 5.2.1.3) of 2.8 (1.1 g, 1.3 mmol, 1.0 equiv.) afforded a quantitative yield of 2.9 as a white solid (0.59 g, 1.3 mmol, 100 %): ' H N M R (400 MHz, D 2 0 ) 8 8.10 (2 H, d, J A r 2 . A r 3 = J A r5,Ar6 9.3 Hz, Ar3, Ar5), 7.08 (2 H, d, J A r 2 , A r3 — JAr5,Ar6 9.3 Hz, Ar2, Ar6), 5.13 (1 H, d, J,, 2 7.8 Hz, HI), 4.38 (1 H, d, J,',2< 7.8 Hz, HI'), 3.86-3.51 (8 H, m), 3.38-3.32 (2 H, m), 3.27 (1 H, t, J 9.1 Hz), 3.20-3.16 (1 H, m); ESI-MS (low res) m/z: calc: 463; Found: 486.1 [M+Na]+. Chapter 5 - Materials and Methods 179 4-Nitrophenyl 6'-phospho-(3-ceUobioside (4NPpX6 T or 2.10). ,OH HO-Prepared by the general enzymatic phosphorylation of 2.9 (50 mg, 0.11 mmol, 1.0 equiv.) using BglK as described in Section 5.2.1.7. After HPLC, the purified fractions were lyophilized to yield white crystalline solids (22 mg, 0.034 mmol, 31 %): ' H N M R (400 MHz, D 2 0) 5 8.10 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 9.3 Hz, Ar3, Ar5), 7.08 (2 H, d, J A r 2 , A r 3 = J A r5,Ar6 9.3 Hz, Ar2, Ar6), 5.14 (1 H, d, J,, 2 7.8 Hz, HI), 4.38 (1 H, d, iy,r 7.9 Hz, HI'), 3.87-3.83 (3 H, m, H6 a, H6 a ' , H6 b '), 3.71-3.34 (8 H, m, H2, H3, H4, H5, H3', H4', H5', H6 b), 3.22 (1 H, t, Ji', 2- 7.9 Hz, H2'); 1 3 C N M R (100 MHz, D 2 0) 8 161.54 (C), 142.48 (C), 125.98 (CH), 116.34 (CH), 102.79 (CIO, 99.04 (Cl) , 78.72, 75.35 (d, i5\p 7.3 Hz, C50, 74.96, 74.87, 73.95, 73.25, 72.31, 68.84, 62.55 (d, J 6-, P 4.3 Hz, C60, 59.75 (C6); 3 I P N M R (162 MHz, D 2 0 ) 8 4.83 (1 P, t, W i w . p 5.1 Hz); ESI-MS (high res) m/z: calc. for [ C ] s H 2 5 N O , 6 P N a 2 ] + : 588.0706; Found: 588.0707; Anal. calc. For C l x H 2 4 N O i 6 P N a 2 - 4 H 2 0 : C, 32.79; H, 4.89; N , 2.12; Found: C, 32.54; H, 5.12; N , 2.17. 4-Nitrophenyl 6-phospho-{3-D-glucopyranoside (4NP/3G6P or 2.13). Prepared by the general enzymatic phosphorylation of the commercially available 4NPpGlc (2.11) (50 mg, 0.17 mmol, 1.0 equiv.) using BglK as described in Section 5.2.1.7. After HPLC, the purified fractions were lyophilized to yield white crystalline solids (22 mg, 0.071 mmol, 42 %): 'H N M R (500 MHz, D 2 0 ) 8 8.16-8.12 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 9.3 Hz, Ar3, Ar5), 7.15-7.12 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 9.3 Hz, Ar2, Ar6), 5.17-5.12 (1 H, d, J , , 2 7.6 Hz, HI), 3.98-3.93 (1 H, m, H6 a), 3.89-3.85 (1 H, m, H6b), 3.64-3.50 (4 H, m, H2, H3, H4, H5); , 3 C N M R (100 MHz, D 2 0 ) 8 161.64 (C), 142.30 (C), 126.00 (2 CH), 116.34 (2 CH), 99.54 (Cl) , 75.77 (d, J 5 , p 7.0 Hz, C5), 74.63 (C3), 72.87 (C2), Chapter 5 - Materials and Methods 180 68.43 (C4), 62.14 (d, J 6 , P 4.5 Hz, C6); 3 I P N M R (162 MHz, D 2 0 ) 5 5.21 (t, JH6.p 5.9 Hz); ESI-MS (high res) m/z calc. For [Ci 2 H| 5 NOnP]": 380.0383, Found: 380.0391; Anal, calc. for C 1 2 H | 4 N N a 2 0 i , P 3 H 2 0 : C, 30.07; H, 4.21; N , 2.92; Found: C, 30.26; H, 4.13; N , 2.96. 4-Nitrophenyl 6-phospho-a-D-glucopyranoside (4NPaG6P or 2.17). Prepared by the general chemical phosphorylation of the commercially available 4NPaGlc (2.14) (90 mg, 0.30 mmol, 1.0 equiv.) using POCI3 as described in Section 5.2.1.10. After HPLC, the purified fractions were lyophilized to yield white crystalline solids (19 mg, 0.045 mmol, 15 %): ! H N M R (400 MHz, D 2 0 ) 5 8.1 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 9.2 Hz, Ar3, Ar5), 7.15 (2 H, d, J A r 2 . A r 3 = J A r 5 , A r 6 9.2 Hz, Ar2, Ar6), 5.66 (1 H, d, J,,2 3.6 Hz, HI), 3.92-3.87 (1 H, m, H6 a), 3.83-3.78 (1 H, m, H3), 3.69-3.57 (4 H, m, H2, H4, H5, H6 b); l 3 C N M R (100 MHz, D 2 0 ) 8 161.35 (C), 142.24 (C), 125.96 (2 CH), 116.65 (2 CH), 96.72 (Cl) , 72.51 (d, J 5 , P 7.2 Hz, C5), 72.21, 70.99, 68.17, 61.92 (d, J 6 , P 4.0 Hz, C6); 3 , P N M R (162 MHz, D 2 0 ) 8 4.80 (t, JH6,p 5.8 Hz); ESI-MS (high res) m/z: calc. for [ C , 2 H l 5 N 0 1 i P N a 2 ] + 426.0178; Found: 426.0180; Anal. calc. For C , 2 H i 4 N N a 2 O n P - 3 H 2 0 : C, 30.07; H, 4.21; N , 2.92; Found: C, 30.47; H , 3.93; N , 2.70. 1,2,3,4,6-Penta-O-acetyl-D-glucopyranose (2.25). OAc D-Glucose (100 g, 555 mmol, 1.0 equiv.) was added to a stirred, ice cold mixture of A c 2 0 (500 mL) and pyridine (650 mL). The reaction was performed as described in Section 5.2.1.1. The clear colorless syrup was recrystallized in EtOH to give 2.25 as a white crystalline solid (204 g, 523 mmol, 94 %). The a/p stereoisomeric ratio was 90:10. (a-Chapter 5 - Materials and Methods 181 anomer): 'H NMR (300 MHz, CDCh) 5 6.31 (1 H, d, J,,2 3.7 Hz, HI), 5.45 (1 H, t, J 2, 3 = J 3 . 4 9.8 Hz, H3), 5.12-5.06 (2 H, m, H2, H4), 4.25 (1 H, dd, J6a/,b 12.5, J 5 , 6 a 4.1 Hz, H6 a), 4.12-4.05 (2 H, m, H5, H6 b), 2.16 (3 H, s, Ac), 2.07 (3 H, s, Ac), 2.02 (3 H, s, Ac), 2.01 (3 H, s, Ac), 2.00 (3 H, s, Ac); ESI-MS (low res) m/z calc: 390; Found: 413.1 [M + Na] + The (3-anomer was synthesized from D-glucose (10 g, 56 mmol, 1.0 equiv.) according to the general acetylation procedure described in Section 5.2.1.1. The product was recrystallized from ethanol (18 g, 46 mmol, 80%) ((3-anomer): 'H NMR (300 MHz, C D C I 3 ) 5 5.68 (1 H, d, J,, 2 = 8.2 Hz, HI), 5.22 (1 H, t, J 2 , 3 = J3.4 9.3 Hz, H3), 5.13-5.06 (2 H, m, H2, H4), 4.26 (1 H, dd, J 6 a ,6 b 12.5, J 5 .6 a 4.5 Hz, H6 a), 4.07 (1 H, dd, J 6 a,6b 12.5, J 5 , 6 b 2.2 Hz, H6 b), 3.81 (1 H, ddd, J 4 , 5 9.9, J 5,6a 4.5, J5.6b 2.2 Hz, H5), 2.08 (3 H, s, Ac), 2.05 (3 H, s, Ac), 2.01 (3 H, s, Ac), 2.00 (3 H, s, Ac), 1.98 (3 H, s, Ac); 1 3 C NMR (75 MHz, C D C I 3 ) 8 170.53 (C=0), 170.02 (C=0), 169.32 (C=0), 169.17 (C^O), 168.88 ( O O ) , 91.66 (Cl) , 72.75 (C3), 72.68 (C5), 70.20 (C4), 67.72 (C2), 61.41 (C6), 20.75 (Ac), 20.64 (Ac) , 20.50 (3 Ac); ESI-MS (low res) m/z: calc: 390; Found: 413.1 [M + Na] +; Anal. calc. for C , 6 H 2 2 0 i , : C, 49.23; H, 5.68; Found: C, 49.42; H, 5.66. 2,3,4,6-Tetra-O-acetyl-a-D'gluCopyranosyl bromide (2.26) (or Acetobromoglucose). 30.0 g (76.5 mmol, 1.0 equiv) of 2.25 was dissolved in 30 mL 33 % HBr in acetic acid and 3 mL of acetic anhydride was added. The reaction was completed according to the general procedure described in Section 5.2.1.2. (28.5 g, 69.3 mmol, 90 %): 'H NMR (400 MHz, CDCI3) 5 6.59 (1 H, d, J,, 2 4.1 Hz, Hi ) , 5.54 (1 H, t, J 2 , 3 = J3,4 9.7 Hz, H3), 5.14 (1 H, t, J 3 , 4 = J 4 , 5 9.7 Hz, H4), 4.82 (1 H, dd, J 2 , 3 9.7, J,, 2 4.1 Hz, H2), 4.33-4.26 (2 H, m, H5-H6 a), 4.11 (1 H, dd, j 6 a , 6 b 12.3, J5.6b 1.9 Hz, H6 b), 2.08 (3 H, s, Ac), 2.07 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.01 (3 H, s, Ac); , 3 C NMR (100 MHz, CDCI3) 5 170.34, 169.68, 169.63, 169.31, 86.48 (Cl) , 72.04 (C5), 70.48 (C2), 70.06 (C3), 67.07 (C4), 60.84 (C6), 20.52 (Ac), 20.51 (Ac), 20.48 (Ac), 20.41 (Ac); ESI-MS (low res) m/z calc: 410, 412; Found: 433.2; 435.1 [M + Na] +. .OAc Br Chapter 5 - Materials and Methods 182 The a-bromide 2.26 (7.0 g, 17 mmol, 1.0 equiv.) was reacted with 2,4-dinitrophenol (6.3 g, 24 mmoles, 2.0 equiv.) using the Koenigs-Knorr reaction described in Section 5.2.1.5. The product was recrystallized from EtOH as a pale yellow solid (1.3 g, 2.5 mmol, 15 %): ' H N M R (400 MHz, CDCh) 8 8.67 (1 H, d, J A r 3 j A r 5 2.7 Hz, Ar3), 8.40 (1 H, dd, J A r 5 , A r 6 9.2, J A r 3 , A r 5 2.7 Hz, Ar5), 7.46 (1 H, d, J A r 5 , A r 6 9.2 Hz, Ar6), 5.32-5.26 (3 H, m, HI, H2, H3), 5.17 (1 H, t, J 3 , 4 = U.s 9.7 Hz, H4), 4.27-4.20 (2 H, m, H6 a, H6 b), 3.96-3.92 (1 H, m, H5), 2.09 (3 H, s, Ac), 2.06 (3 H, s, Ac), 2.04 (3 H, s, Ac), 2.02 (3 H, s, Ac); l 3 C N M R (100 MHz, CDC13) 5 170.27, 170.05, 169.01, 153.54, 142.35, 140.36, 128.51, 121.35, 118.84, 99.45 (Cl) , 72.77, 71.88, 70.20, 67.75, 61.60 (C6), 20.60 (Ac), 20.49 (2 Ac), 20.40 (Ac); ESI-MS (low res) m/z calc: 514; Found: 537.2 [M + Na] +. 2,5-Dinitrophenyl 2,3,4,6-tetra-O-acetyl-p^D-glitcopyranoside (2.28). Acetobromoglucose 2.26 (5.0 g, 12 mmol, 1.0 equiv.) was reacted with 2,5-dinitrophenol (4.5 g, 24 mmol, 2.0 equiv) as described for 2.27. The product was recrystallized from EtOH as a pale yellow solid (2.1 g, 4.1 mmol, 33 %). ' H N M R (300 MHz, CDC13) 8 8.30 (1 H, d, J A r 4 , A r 6 1.7 Hz, Ar6), 8.00 (2 H, m, Ar3, Ar4), 5.25 (1H, d, J,, 2 7.9 Hz, HI), 3.95 (1 H, t, J 3 , 4 = J 4 , 5 9.7 Hz, H4), 3.70 (1 H, t, J 2 , 3 = J 3 , 4 9.7 Hz, H3), 3.60-3.30 (4 H, m, H2, H5, H6 a, H6 e), 2.07 (3 H, s, Ac), 2.05 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.02 (3 H, s, Ac); ESI-MS (low res) m/z calc: 514; Found: 537.0 [M + Naf . .OAc Chapter 5 - Materials and Methods 183 3,4-Dinitrophenyl 2,3,4,6-tetra-0-acetyl-/3-D-glucopyranoside (2.29). .OAc AcO ' Acetobromoglucose 2.26 (7.0 g, 17 rnmol, 1.0 equiv.) was reacted with 3,4-dinitrophenol (6.3 g, 34 mmol, 2.0 equiv.) as described for 2.27. The product was recrystallized from EtOH as a pale yellow solid (1.1 mg, 2.1 g, 14 %): ' H N M R (300 MHz, CDC13) 5 7.99 (1 H, d, J A r 5 , A r 6 9.0 Hz, Ar5), 7.39 (1 H, d, J A r 2 , A r 6 2.6 Hz, Ar2), 7.24 (1 H, dd, 9.0, J A r2,Ar6 2.6 Hz, Ar6), 5.31-5.09 (4 H, m, HI , H2, H3, H4), 4.18 (2 H, d, J 4.2 Hz, H6 a, H6 b), 3.98 (1 H, m, H5), 2.06 (3 H, s, Ac), 2.05 (3 H, s, Ac), 2.04 (3 H, s, Ac), 2.02 (3 H, s, Ac); 1 3 C N M R (100 MHz, CDC13) 5 170.47, 170.04, 169.33, 169.10, 159.66, 127.20, 120.21, 112.45, 98.08 (Cl) , 72.82, 72.18, 70.71, 67.83, 61.89 (C6), 20.58 (Ac), 20.53 (3 Ac): ESI-MS (low res) m/z calc: 514; Found: 537.3 [M + Na] +. 4-Chloro-2-nitrophenyl 2,3,4,6-tetra-O-acetyl-p^-D-glucopyranoside (2.30). Acetobromoglucose 2.26 (5.0 g, 12 mmol, 1.0 equiv) was reacted with 4-chloro-2-nitrophenol (4.2 g, 24 mmol, 2.0 equiv.) (24 mL NaOH, 12 mL acetone) as described for 2.27. The product was recrystallized from EtOH as a pale yellow solid (0.99 g, 2.0 mmol, 16 %): *H N M R (400 MHz, CDCI3) 8 7.78 (1 H, d, J A r 3 , A r 5 2.5 Hz, Ar3), 7.47 (1 H, dd, J A r 5 , A r 6 8.9, J A r 3 , A r 5 2.5 Hz, Ar5), 7.30 (1 H, d, J A r 5, A l -6 8.9 Hz, Ar6), 5.30-5.05 (4 H, m, HI-4), 4.26-4.18 (2 H, m, H6 a, H6 b), 3.83 (1 H, m, H5), 2.10 (3 H, s, Ac), 2.07 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.02 (3 H, s, Ac); ESI-MS (low res) m/z calc: 503; Found: 526.0 [M + Na] +. .OAc Chapter 5 - Materials and Methods 184 2-Nitrophenyl 2,3,4,6-tetra-O-acetyl-fi-D-glucopyranoside (2.31). Compound 2.31 was prepared from acetobromoglucose 2.26 (6.0 g, 15 mmol, 1.0 equiv.) via the Koenigs-Knorr reaction by coupling with 2 nitrophenol (4.1 g, 30 mmol, 2.0 equiv.) in 30 mL NaOH and 15 mL acetone. The bright orange solution was recrystallized in EtOH as pale yellow crystals (3.6 g, 7.7 mmol, 52 %): *H N M R (300 MHz, CDC13) 8 7.77 (1 H, dd, J A l ,3 , A r 4 8.1, J A r 3 , A r 3 1.6 Hz, Ar3), 7.53-7.48 (1 H, ddd, JAr4.Ar5 9.0, J A r 5 , A r 6 7.5, J A r 3.Ar5 1-6 Hz, Ar5), 7.33 (1 H, dd, J A r 5 .Ar6 7.7, J A r 4 , A r 6 1-1 Hz, Ar6), 7.22-7.16 (1 H, m, Ar4), 5.30-5.09 (4 H, m, HI , H2, H3, H4), 4.29-4.18 (2 H, m, H6 a, H6 b), 3.88-3.82 (1 H, m, H5), 2.10 (3 H, s, Ac), 2.07 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.02 (3 H, s, Ac): , 3 C N M R (75 MHz, CDC13) 8 170.43 (C=0), 170.17 (C=0), 169.30 (2 lines, 2 C=0), 149.22 (C), 141.64 (C), 133.66 (CH), 125.13 (CH), 123.88 (CH), 119.98 (CH), 100.20 (Cl) , 72.34 (2 lines), 70.54, 68.17, 61.80 (C6), 20.64 (2 Ac), 20.55 (2 Ac); ESI-MS (low res) m/z calc: 469; Found: 492.1 [M + Na] +. 3,5-Dichlorophenyl 2,3,4,6-tetra-0-acetyl-/3-D-glucopyranoside (2.32). ci Acetobromoglucose 2.26 (3.0 g, 7.3 mmol, 1.0 equiv.) was reacted with 3,5-dichlorophenol (2.38 g, 15 mmol, 2.0 equiv.) 7 mL acetone and 15 mL NaOH using the Koenigs-Knorr reaction (Section 5.2.1.5). The product was recrystallized from EtOH as a white solid (1.5 g, 3.0 mmol, 41 %): ' H N M R (400 MHz, CDCb) 8 7.06 (1 H, t, JAr2,Ar4 — JA r 4,A l-6 1.7 Hz, Ar4), 6.89 (2 H, d, J A r 2 ,Ar4 — JAr4,Ar6 1.7 Hz, Ar2, Ar6), 5.31-5.19 (2 H, m, H2, H3), 5.09 (1 H, t, J 3 , 4 = J 4 , 5 9.9 Hz, H4), 5.03 (1 H, d, j u 7.5 Hz, HI), 4.20-4.18 (2 H, m, H6 a ,H6 b), 3.91-3.85 (1 H, m, H5), 2.09(3 H, s, Ac), 2.05 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.01 (3 H, s, Ac); l 3 C N M R (100 MHz, CDC13) 8 170.58, 170.14, 169.40, Chapter 5 - Materials and Methods 185 169.20, 157.46, 135.45, 123.59, 115.88 (2 lines), 98.62 (Cl) , 72.50, 72.35, 70.92, 68.22, 62.07, 20.67 (Ac), 20.61 (Ac), 20.57 (2 Ac); ESI-MS (low res) m/z calc: 492; Found: 515.0 [M + Na] +; Anal. calc. For C 2 o H 2 2 C l 2 0 1 0 : C, 48.70, H , 4.50; Found: C, 48.84, H, 4.36. 3-Nitrophenyl 2,3,4,6-tetra-O-acetyl-fd-D-glucopyranoside (2.33). Compound 2.33 (1.5 g, 3.2 mmol, 43.9 %) was prepared from 2 . 2 6 (3.0 g, 7.3 mmol, 1.0 equiv.) via the Koenigs-Knorr reaction by coupling the a-bromide (2 .26 ) with 3-nitrophenol (2.8 g, 15 mmol). Recrystallization in EtOH yield the pure product as pale yellow crystals: ' H N M R (300 MHz, CDC13) 8 7.95-7.92 (1 H, m, Ar4), 7.84 (1 H, t, J A r 2 , A r 4 2.2 Hz, Ar2), 7.45 (1 H, t, J A r 4 , A r 5 = JA r 5,A,-6 8.2 Hz, Ar5), 7.31-7.27 (1 H, m, Ar6), 5.34-5.09 (4 H, m, HI , H2, H3, H4), 4.20-4.19 (2 H, m, H6 a, H6 b), 3.97-3.91 (1 H, m, H5), 2.07 (3 H, s, Ac), 2.05 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.01 (3 H, s, Ac); l 3 C N M R (75 MHz, CDC13) 8 170.65, 170.12, 169.39, 169.20, 156.98, 149.10, 130.21, 123.75, 118.24, 111.24, 98.64 (Cl) , 72.51 (2 carbons), 70.97, 68.14, 62.01, 20.59 (Ac), 20.56 (3 Ac); ESI-MS (low res) m/z calc: 469; Found: 492.0 [M + Na] +; Anal. calc. for C 2 oH23NO, 2 : C, 51.18, H, 4.94, N , 2.98; Found: C, 51.50, H, 4.85, N , 2.85. 4-Cyanophenyl 2,3,4,6-tetra-0-acetyl-(3-D-ghiCopyranoside (2.34). Acetobromoglucose 2 . 2 6 (3.0 g, 7.3 mmol, 1.0 equiv.) was reacted with 4-cyanophenol (1.7 g, 15 mmol, 2.0 equiv.) as described for 2 .27 . The product was recrystallized from EtOH as a pale yellow solid (969 mg, 2.2 mmol, 30 %): lti N M R (300 MHz, CDCI3) 8 7.59 (2 H, d, J A r 2 .Ar3 = W A K S 8 .8.HZ , Ar3, Ar5), 7.03 (2 H, d, J A r 2 , A r 3 = -Wre 8.8 Hz, .OAc .OAc Chapter 5 - Materials and Methods 186 Ar2, Ar6), 5.30-5.14 (4 H, m, HI , H2, H3, H4), 4.26 (1 H, dd, J 6 a , 6 b 12.4, J 5 , 6 5.5 Hz, H6 a), 4.15 (1 H, dd, J 6 a , 6 b 12.4, J 5,6 b2.4 Hz, H6b), 3.88 (1 H, ddd, J 4 . 5 9.8, J 5 , 6 a 5.5, J 5 , 6 b 2.4 Hz, H5), 2.04 (3 H, s, Ac), 2.03 (6 H, s, 2 Ac), 2.02 (3 H, s, Ac); E S I - M S (low res) m/z calc: 449; Found: 472.0 [M + Na] +, 488.5 [M + K ] + . Phenyl 2,3,4,6-tetra-O-acetyl-fi-D-gliicopyranosicle (2.35). Acetobromoglucose 2.26 (5.0 g, 12 mmol, 1.0 equiv.) was reacted with 4-cyanophenol (2.3 g, 24 mmol, 2.0 equiv.) as described for 2.27. The resulting colorless syrup was recrystallized in EtOH to give white solids (2.7 g, 6.3 mmol, 52 %): ' H N M R (300 MHz, CDC13) 8 7.24 (2 H, t, J A r 2 , A r 3 = J A r3.Ar4 = J Ar4,Ar5 = JAr5.Ar6 7.3 Hz, Ar3, Ar5), 7.05-6.94 (3 H, m, Ar2, Ar4, Ar6), 5.27-5.04 (4 H, m, HI , H2, H3, H4), 4.25 (1 H, dd, J 5 . 6 a 5.4, J 6 a , 6 b 12.3 Hz, H6 a), 4.11 (1H, dd, J 5 ,6 b 2.4, ^ ' 1 2 . 3 Hz, H6 b), 3.82 (1 H, ddd, J 4 , 5 9.8, J 5 , 6 a 5.4, J 5 . 6 b 2.4 Hz, H5), 2.03 (3 H, s, Ac), 2.01 (3 H, s, Ac), 2.00 (3 H, s, Ac), 1.99 (3 H, s, Ac); E S I - M S (low res) m/z calc: 424; Found: 447.0 [M + Na] + Anal. calc. For C 2 o H 2 4 0 1 0 : C, 56.60; H, 5.70; Found: C, 55.73; H, 5.55. 4-Butylphenyl 2,3,4,6-tetra-O-acetyl-pZ-D-gluCopyranoside (2.36). Acetobromoglucose 2.26 (3.0 g, 7.3 mmol, 1.0 equiv.) was reacted with t-butylphenol (2.2 g, 15 mmol, 2.0 equiv.) via the Koenigs-Knorr reaction (Section 5.2.1.5). The N M R (300 MHz, CDCI3) 8 7.29 (2 H, d, J A r 2 . A r 3 = JArs^rf 8.8 Hz, Ar3, Ar5), 6.90 (2 H, d, J A r2,Ar3 = JAT5^T6 8.8 Hz, Ar2, Ar6), 5.28-5.03 (4 H, m, HI , H2, H3, H4), 4.27 (1 H, dd, J 6 a .6b 12.3, J5.6a 5.2 Hz, H6 a), 4.15 (1 H, dd, J ^ b 12.3, J 5 , 6 B 2.4 Hz, H6b), 3.82 (1 H, ddd, .OAc .OAc product was recrystallized from EtOH as a pale yellow solid (1.0 g, 2.1 mmol, 29 %): H Chapter 5 - Materials and Methods 187 J4.5 9.7, J 5 , 6 a 5.2, J 5 . 6 b 2.4 Hz, H5), 2.06 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.01 (3 H, s, Ac), 1.27 (9 H, s, 3 CH 3 ) ; , 3 C N M R (100 MHz, CDC13) 5 170.62 (C=0), 170.29 (C=0), 160.41 (C=0), 169.33 (C=0), 154.60 (C), 146.19 (C), 126.36 (2 CH), 116.43 (2 CH), 99.23 (Cl) , 72.77, 71.93, 71.17, 68.32, 61.95, 34.22 (t-Bu, C), 31.43 (3 CH 3 ) , 20.69 (Ac), 20.62 (2 Ac), 20.59 (Ac); ESI-MS (low res) m/z cai: 480; Found: 503.3 [M + Na] +; Anal . calc. For C24H320,o: C, 59.99; H, 6.71; Found: C, 59.86; H, 6.69. 2,4-Dinitrophenyl /3-D-glucopyranoside (2.37). Deprotection of 2.27 (0.50 g, 0.97 mmol, 1.0 equiv.) was performed in H C l ( g ) dissolved in MeOH to yield 2.37 (crude yield = 0.32 g, 0.92 mmol, 95 %). The reaction mixture was dried under reduced pressure, and bright yellow crystals were not further purified. 2,5-Dinitrophenyl f3-D-glucopyranoside (2.38). Deprotection of 2.28 (0.50 g, 0.97 mmol, 1.0 equiv.) was carried out in HCl ( g) dissolved in MeOH to yield 2.38. The reaction mixture was dried under reduced pressure, and bright yellow crystals were not further purified. Crude yield: 0.29 g, 0.82 mmol, 85 %. Chapter 5 - Materials and Methods 188 3,4-Dinitrophenyl /3-D-glucopyranoside (2.39). Deprotection of 2.29 (0.50 g, 0.97 mmol, 1.0 equiv.) was accomplished in anhydrous methanol containing dissolved HCI ( g). The reaction mixture was concentrated under reduced pressure and the product was purified by flash chromatography (7:2:1 EtOAc:MeOH:H 20) as a yellow solid (0.27 g, 0.78 mmol, 80 %): ' H N M R (400 MHz, D 2 0 ) 5 8.02 (1 H, d, J A r 5,ArC, 9.1 Hz, Ar5), 7.49 (1 H, d, J A r 2 j A r 6 2.6 Hz, Ar2), 7.32 (1 H dd, JAr5.Arts9.1, J A r 2 , A r 6 2.6 Hz, Ar6), 5.14-5.12 (1 H, m, HI), 3.78-3.34 (6 H, m, H2, H3, H4, H5, H6 a, H6 b); l 3 C N M R (100 MHz, (CD 3 ) 2 CO) 5 161.26, 144.96, 135.81, 128.30, 120.25, 113.22, 100.28 (Cl) , 76.85, 75.78, 73.02, 69.63, 60.86 (C6); ESI-MS (low res) m/z cal: 346; Found: 370.1 [M + Na] +. 4-Chloro-2-nitrophenyl (3-D-glucopyranoside (2.40). Zemplen deprotection of 2.30 (0.80 g, 1.6 mmol, 1.0 equiv.) afforded 2.40 as a yellow solid (0.54 g, 1.6 mmol, 100 %): ' H N M R (300 MHz, D 2 0 ) 5 7.91 (1 H, d, J A r 3 , A r 5 2.6 Hz, Ar3), 7.57 (1 H, dd, J Ar3,Ar5 2.6, J A r 5 , A r 6 9.1 Hz, Ar5), 7.30 (1 H, d, JA,-5,Ar6 9.1 Hz, Ar6), 5.10 (1 H, d, J 1 > 2 7.3 Hz, HI), 3.80 (1 H, dd, J 6 a , 6 b 12.4, J 5 i < 5 a 1.9 Hz, H6 a), 3.64 (1 H, dd, J 6 a , 6 b 12.4, J 5 , 6 b 5.5 Hz, H6 b), 3.55-3.36 (4 H, m, H2, H3, H4, H5); ESI-MS (low res) m/z calc: 335; Found: 358.3 [M + Na] + Chapter 5 - Materials and Methods 189 2-Nitrophenyl fi-D-glucopyranoside (2.41). Compound 2.31 (3.5 g, 7.5 mmol, 1.0 equiv.) was subjected to the Zemplen deprotection as described in the general synthetic methods to yield a bright yellow powder (2.3 g, 7.5 mmol, 100 %): ' r l N M R (300 MHz, D 2 0) 8 7.80 (1 H, dd, J A r 3 , A r 4 8.2, J A r 3 j A r 5 1.5 Hz, Art) , 7.54 (1 H, dt, J A r 4 , A r 5 = W r e 8.2, J A r 3 . A r 5 1.5 Hz, Ar5), 7.29 (1 H, d, J A r 5 , A r 6 8.0 Hz, Ar6), 7.12 (1 H, dt, J A r 3 , A r 4 = J A r 4 , A r 5 8.2, J A r 4 , A r 6 0.9 Hz, Ar4), 5.10 (1 H, d, J,, 2 7.4 Hz, HI), 3.79 (1 H, dd, J 6 a .6b 12.4, J5M 2.0 Hz, H6 a), 3.62 (1 H, dd, J ^ b 12.4, h,6b 5.5 Hz, H6b), 3.54-3.34 (4 H, m, H2, H3, H4, H5); ESI-MS (low res) m/z calc: 301; Found: 324.0[M + Na]+. 3,5-Dichlorophenyl fd-D-gluCopyranoside (2.42). Zemplen deprotection of 2.32 (1.5 g, 3.0 mmol, 1.0 equiv.) afforded 2.42 as a white solid (0.98 g, 3.0 mmol, 100 %), which was not further purified: ESI-MS (low res) m/z calc: 325; Found: 348.3 [M + Na] +. 3-Ntirophenyl fd-D-glucopyranoside (2.43). Zemplen deprotection of 2.33 (1.5 g, 3.2 mmol, 1.0 equiv.) in 100 mL NaOMe afforded 2.43 as a yellow solid (0.93 g, 3.1 mmol, 98 %): ' H N M R (300 MHz, D 2 0 ) 5 7.84-7.79 ci Chapter 5 - Materials and Methods 190 (2 H, m, Ar2, Art) , 7.43 (1 H, t, J A r 4 , A r 5 = J A r5,Ar6 8.2 Hz, Ar5), 7.37-7.34 (1 H, m, Ar6), 5.08-5.06 (1 H, m, HI), 3.81-3.32 (6 H, m, H2, H3, H4, H5, H6 a, H6 b); l 3 C N M R (100 MHz, D 2 0 ) 5 156.68, 148.56, 130.46, 123.28, 117.93, 111.31,99.92 (Cl) , 76.09, 75.30, 72.64, 69.23, 60.30 (C6); ESI-MS (low res) m/z calc.: 301; Found: 324 [M + Na] +; Anal . calc. For C^HisNO*: C, 47.84, H, 5.02, N , 4.65; Found: C, 48.56, H, 5.10, N , 4.93. 4-Cyanophenyl fi-D-glucopyranoside (2.44) Zemplen deprotection of 2.34 (0.97 g, 2.2 mmol, 1.0 equiv) afforded 2.44 as a white solid (0.46 g, 1.5 mmol, 70 %): *H N M R (400 MHz, D 2 0 ) 5 7.60 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 8.5 Hz, Ar3, Ar5), 7.07 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 8.5 Hz, Ar2, Ar6), 5.07 (1 H, d, J,, 2 7.1 Hz, HI), 3.78 (1 H, dd, J 6 a , 6 b 12.4, J 5 , 6 a 2.1 Hz, H6 a), 3.60 (1 H, dd, J 6 a . 6 b 12.4, J 5 , 6 b 5.7 Hz, H6 b), 3.52 (1 H, m, H5), 3.47-3.45 (2 H, m, H2, H3), 3.35 (1 H, m, H4); ESI-MS (low res) m/z calc: 281; Found: 304.1 [M + Na] +. Phenyl pZ-D-gluCopyranoside (2.45). Compound 2.35 (1.0 g, 2.4 mmol, 1.0 equiv.) was deprotected by the Zemplen method. The reaction mixture was concentrated under reduced pressure to give the pure product (0.59 g, 2.3 mmol, 96 %) as a white crystalline solid: ' H N M R (300 MHz, D 2 0) 8 7.22 (2 H, t, J A r 2 , A r 3 = J A r 3 , A r 4 = J A r 4 . A r 5 = J A r 5 ,A r 6 7.6 Hz, Ar3, Ar5), 6.98-6.94 (3 H, m, Ar2, Art , Ar6), 4.91 (1 H, d, J,, 2 7.3 Hz, HI), 3.88 (1 H, dd, J6a,r>b 12.3, J 5 , 6 a 2.3 Hz, H6 a), 3.73 (1 H, dd, J 6 a , 6 b 12.3, J 5,6 a 5.8 Hz, H6 b), 3.43-3.23 (4 H, m, H2-5); ESI-MS (low res) m/z calc: 256; Found: 279.0 [M + Naf . Chapter 5 - Materials and Methods 191 4-Butylphenyl (3-D-gluCopyranoside (2.46). Zemplen deprotection of 2.36 (0.42 g, 0.87 mmol, 1.0 equiv.) afforded 2.46 as a white solid (0.23 g, 0.73 mmol, 84 %): ' H N M R (400 MHz, D z O) 5 7.29 (2 H, d, JA r2,Ar3 = J A r 5 , A r 6 8.9 Hz, Ar3, Ar5), 6.92 (2 H, d, J A l 2 , A r 3 = J A r 5 ,Ar6 8.9 Hz, Ar2, Ar6), 4.91 (1 H, d, J,, 2 7.5 Hz, HI), 3.74 (1 H, dd, J 6 a . 6 b 12.4, i5M 2.0 Hz, H6 a), 3.56 (1 H, dd, J 6 a / ) b 12.4, J 5 , 6 b 5.8 Hz, H6 b), 3.45-3.27 (4 H, m, H2, H3, H4, H5); 1 3 C N M R (100 MHz, D z O) 8 126.64, 116.20, 100.22 (Cl) , 75.95, 75.42, 72.80, 69.32, 60.42 (C6), 33.49 (C), 30.46 (3 CH 3 ) ; ESI-MS (low res) m/z calc: 314; Found: 337.3 [M + Naf. 2,4-Dinitrophenyl 6-phospho-$-D-glucopyranoside disodium salt (24DNP0G6P or 2.47). Standard BglK-catalyzed phosphorylation of 2.37 (0.32 g, 0.92 mmol, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give pale yellow crystals of 2.47 (66 mg, 0.14 mmol, 15 %): ' H N M R (400 MHz, D 2 0 ) 8 8.70 (1 H, d, J A R 3 , A R 5 2.8 Hz, Ar3), 8.37 (1 H, dd, J A R 5 , A R 6 9.4, J A R 3 , A R 5 2.8 Hz, Ar5), 7.49 (1 H, d, J A R 5 , A R 6 9.4 Hz, Ar6), 5.26 (1 H, d, J,. 2 7.6 Hz, HI), 3.96-3.84 (2 H, m, H6 a, H6 b), 3.63-3.47 (4 H, m, H2, H3, H4, H5); L 3 C N M R (100 MHz, D 2 0) 8 154.23 (C), 141.30 (C), 138.62 (C), 129.86 (CH), 122.03 (CH), 117.59 (CH), 100.14 (Cl) , 76.02 (d, J 5 , P 6.5 Hz, C5), 74.63, 72.59, 68.26, 62.13 (d, J6,P 4.3 Hz, C6); 3 I P N M R (162 MHz, D 2 0 ) 8 4.88 (1 P, t, JH6,p 5.9 Hz); ESI-MS (high res) m/z: calc for [C ,2Hi 4 N 2 0] 3 PNa 2 ] + 471.0029; Found: 471.0028; Anal. calc. For Ci 2 H , 3 N2O 1 3 PNa2 -0 .5H 2 O: C, 30.08; H, 2.94; N , 5.85; Found: C, 30.25; H, 3.19; N , 6.15. Chapter 5 - Materials and Methods 192 2,5-Dinitrophenyl 6-phospho-f3-D-glucopyranoside disodhim salt (25DNPf3G6P or 2.48). Standard BglK-catalyzed phosphorylation of 2.38 (0.29 g, 0.82 mmol, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give pale orange crystals of 2.48 (15 mg, 0.029 mmol, 3.5 %): ' H N M R (400 MHz, D 2 0) 8 8.06 (1 H, d, JAr4,Ar6 1.7 Hz, Ar6), 7.97-7.92 (2 H, m, Ar3, Ar4), 5.25 (1 H, d, J,, 2 7.3 Hz, HI), 3.99-3.94 (1 H, m, H6 a), 3.86-3.81 (1 H, m, H6 b), 3.65-3.46 (4 H, m, H2, H3, H4, H5); , 3 C N M R (100 MHz, D 2 0 ) 8 150.28 (C) , 149.39 (C) , 143.34 ( C ) , 126.50 (CH), 117.87 (CH), 112.65 (CH), 100.34 (Cl) , 75.93 (d, J 5 , P 7.0 Hz, C5), 74.50, 72.67, 68.03, 61.99 (d, J6,p 4.5 Hz, C6); 3 1 P N M R (162 MHz, D 2 0 ) 5 5.04 (1 P, t, JH6,p 7.4 Hz); ESI-MS (high res) m/z: calc. for [ C | 2 H i 4 N 2 O i 3 P N a 2 ] + : 471.0029; Found: 471.0030; Anal. calc. For C i 2 H , 3 N 2 0 1 3 P N a 2 H 2 0 : C , 29.52; H, 3.10; N , 5.74; Found: C , 29.35; H, 3.39; N , 5.82. 3,4-Dinitrophenyl 6-phospho-f3-D-glucopyranoside disodivm salt (34DNPfiG6P or 2.49). Standard BglK-catalyzed phosphorylation of 2.39 (0.84 g, 0.24 mmol, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give 2.49 as a pale yellow solid (18 mg, 0.034 mmol, 14 %): ' H N M R (400 MHz, D 2 0) 5 8.70 (1 H, d, J A r 2 , A r 5 2.8 Hz, Ar2), 8.37 (1 H, dd, J A r 5 , A r6 9.4, J A r 2 , A r 5 2.8 Hz, Ar5), 7.49 (1 H, d, J A r 5 , A r 6 9.4 Hz, Ar6), 5.26 (1 H, d, J,, 2 7.6 Hz, HI), 3.96-3.84 (2 H, m, H6 a, H6 b), 3.63-3.47 (4 H, m, H2, H3, H4, H5); , 3 C N M R (100 MHz, D 2 0) 8 154.23 (C), 14.1.30 (C), 138.62 (C), 129.86 (CH), 122.03 (CH), 117.59 (CH), 100.14 (Cl) , 76.02 (d, J 5 , P 6.5 Hz, C5), 74.63, 72.59, 68.26, 62.13 (d, J 6 , P 4.3 Hz, C6); 3 1 P N M R (162 MHz, D 2 0 ) 8 4.88 (1 P, t, JH6.p 5.9 Hz); ESI-MS (high res) m/z: calc. for [ C i 2 H , 4 N 2 O i 3 P N a 2 ] + 471.0029; Found: C h a p t e r 5 - M a t e r i a l s a n d M e t h o d s 193 471.0028; Anal. calc. For C 1 2 Hi3N 2 O ,3PNa 2 0.5H 2 O: C, 30.08; H, 2.94; N , 5.85; Found: C, 30.25; H, 3.19; N , 6.15. 4-Chloro-2-nitrophenyl 6-phospho-f3-D-glucopyranoside disodium salt (4C2NPJ3G6P or Standard BglK-catalyzed phosphorylation of 2.40 (0.11 g, 0.33 mmole, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give a white solid (29 mg, 0.056 mmol, 17 %): *H N M R (400 MHz, D 2 0 ) 8 7.86 (1 H, d, J A r 3 , A r 5 2.6 Hz, Ar3), 7.53 (1 H, dd, J A r 5 , A r t i 9.1, J A , - 3 , A r 5 2.6 Hz, Ar5), 7.30 (1 H, d, W ^ . l Hz, Ar6), 5.08(1 H, d, J,, 2 7.5 Hz, HI), 3.94-3.88 (1 H, m, H6 a), 3.84 (1 H, ddd, J 12.2, J 5.7, J 1.5 Hz, H6 b), 3.59-3.43 (4 H, m, H2, H3, H4, H5); l 3 C N M R (75 MHz, D 2 0 ) 8 149.81 (C), 141.27 (C), 136.35 (CH), 128.76 (C), 126.84 (CH), 120.45 (CH), 102.16 (Cl) , 77.39 (d, J5.p 6.8 Hz, C5), 76.27, 74.27, 69.91, 63.73 (d, J 6,P4.1 Hz, C6); 3 , P N M R (162 MHz, D 2 0 ) 8 5.03 (1 P, t, J H6.P 8.1 Hz); ESI-MS (high res) m/z: calc. for [C | 2 H, 4 ClNNa 2 Oi , P ] + 459.9788; Found: 459.9786; Anal. calc. For C , 2 Hi3ClNNa 2 0 , ,P -2 .5H 2 0 : C, 28.53; H, 3.57; N , 2.77; Found: C, 28.94; H, 3.96; N , 3.00. 2-Nitrophenyl 6-phospho-f5-D-gluCopyranoside disodium salt (2NP/3G6P or 2.51). Standard BglK-catalyzed phosphorylation of 2.41 (0.23 g, 0.75 mmol, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give 2.51 as a pale yellow solid (30 mg, 0.070 mmol, 9.4 %): ' H N M R (400 MHz, D 2 0 ) 8 7.75 (1 H, dd, J A r 3 ,Ar4 8.2, J A r 3 ,Ar5 1 -5 Hz, Ar3), 7.53-7.49 (1 H, m, Ar5), 7.28 (1 H, d, J A r 5 , A r 6 8.2 Hz, Ar6), 7.07 (1 H, t, J A r 3 . A r 4 = J A r 4 . A o 8.2 Hz, Ar4), 5.09 (1 H, d, J,, 2 7.4 Hz, HI), 3.96-3.90 (2 H, m, H6 a), 3.84 (1 H, ddd, J 12.2, J 5.6, J 1.3 Hz, H6 b), 3.60-3.45 (4 H, m, H2, H3, 2.50). Chapter 5 - Materials and Methods 194 H4, H5); , 3 C N M R (100 MHz, D 2 0 ) 5 149.37 (C), 139.52 (C), 135.23 (CH), 125.55 (CH), 122.96 (CH), 117.24 (CH), 100.47 (Cl) , 75.81 (d, J 5 , P 6.8 Hz, C5), 74.71, 72.80, 68.36, 62.17 (d, J 6 , P 3.8 Hz, C6); 3 , P N M R (162 MHz, D 2 0 ) 5 5.17 (1 P, t, J H 6 .P 6.2 Hz); ESI-MS (high res) m/z: calc. for [C i 2 H, 5 NOi ,PNa 2 ] + 426.0178; Found: 426.0173; Anal, calc. For C i 2 H i 4 N N a 2 O n P - 3 H 2 0 : C, 30.07; H, 4.21; N , 2.92; Found: C, 30.47; H, 4.38; N , 3.00. 3,5-Dichlorophenyl 6-phospho-fi-D-gluCopyranoside disodium salt (35DCPfiG6P or 2.52). Standard BglK-catalyzed phosphorylation of 2.42 (0.26 g, 0.80 mmol, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give the product as a white solid (49 mg, 0.11 mmol, 14 %): ' H N M R (400 MHz, D 2 0) 5 7.05 (1 H, t, J A r 2 , A r 4 = J A r 4 . A r 6 1.6 Hz, Ar4), 6.96 (2 H, d, J A r 2 , A r 4 = J A r 4 . A r 6 1.6 Hz, Ar2, Ar6), 4.96 (1 H, d, J,. 2 7.6 Hz, HI), 3.98-3.92 (1 H, m, H6 a), 3.82 (1 H, ddd, J 12.4, 5.6, 1.6 Hz, H6 b), 3.62-3.41 (4 H, m, H2, H3, H4, H5); l 3 C N M R (75 MHz, D 2 0 ) 5 158.87 (C), 136.44 (CH), 124.53 (2 C), 116.89 (2 CH), 101.55 (Cl) , 77.19 (d, J 5 , P 7.0 Hz, C5), 76.12, 74.41, 69.79, 63.58 (d, J 6 , P 3.2 Hz, C6); 3 , P N M R (162 MHz, D 2 0) 8 5.03 (1 P, t, J H 6 , P 5.7 Hz); ESI-MS (high res) m/z: calc. for [Ci 2 H]40 9 PNa 2 Cl 2 ] + 448.9548; Found: 448.9540; Anal. calc. For C,2H 1 3Oc,PNa 2Cl 2-3H 20: C, 28.65; H, 3.81; Found: C, 28.83; H, 3.98. 3-Nitrophenyl 6-phospho-j3-D-gluCopyranoside disodium salt (3NPJ3G6P or 2.53) Standard BglK-catalyzed phosphorylation of 2.43 (0.22 g, 0.74 mmol, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give 2.53 as a ci Chapter 5 - Materials and Methods 195 white solid (56 mg, 0.13 mmol, 17 %): ' H N M R (400 MHz, D 2 0 ) 8 7.80 (1 H, ddd, JAr4,Ar5 8.3, J A r 2 . A r 4 2.1, J A r 4 , A r 6 1-3 Hz, Ar4), 7.75 (1 H, t, J A r2.Ar4 = J A r2 ,Ar6 2.1 Hz, Ar2), 7.41 (1 H, t, J A r4.Ar5 = J A r5,Ar6 8.3 Hz, Ar5), 7.36 (1 H, ddd, J A r 5 . A r 6 8.3, J A r 2 , A r 6 2.1, J A r 4 , A r 6 1.3 Hz, Ar6), 5.09-5.07 (1 H, m, HI), 3.97-3.92 (1 H, m, H6 a), 3.83 (1 H, ddd, J 12.3, 5.5, 1.6 Hz, H6b), 3.62-3.54 (2 H, m, H5, H3), 3.52-3.46 (2 H, m, H2, H4); 1 3 C N M R (100 MHz, D 2 0 ) 5 156.71 (C), 148.55 (C), 130.59 (CH), 123.17 (CH), 117.94 (CH), 111.45 (CH), 100.06 (Cl) , 75.68 (d, J5.P 6.9 Hz, C5), 74.63, 72.92, 68.32, 62.08 (d, J 6 , P 4.5 Hz, C6); 3 I P N M R (162 MHz, D 2 0 ) 8 5.04 (1 P, t, JH6,p 6.3 Hz); ESI-MS (high res) m/z: calc. for [ C 1 2 H 1 5 N 0 1 i P N a 2 ] + 426.0178; Found: 426.0175; Anal. calc. For C 1 2 H 1 4 N N a 2 0 i i P - 3 H 2 0 : C, 30.07; H, 4.21; N , 2.92; Found: C, 30.47; H, 4.61; N , 2.77. 4-Cyanophenyl 6-phospho-f3-D-glncopyranoside disodium salt (4CNPfiG6P or 2.54). Standard BglK-catalyzed phosphorylation of 2.44 (0.40 g, 1.4 mmol, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give 2.54 as a white solid (47 mg, 0.11 mmol, 7.6 %): ' H N M R (400 MHz, D 2 0 ) 8 7.60 (2 H, d, J A r 2 , A r 3 = - W o 7.5 Hz, Ar3, Ar5), 7.09 (2 H, d, J A r 2 > A r 3 = J A r 4 , A r 5 7.5 Hz, Ar2, Ar6), 5.08-5.06 (1 H, m, HI), 3.95-3.84 (2 H, m, H6 a, H6 b), 3.56-3.47 (4 H, m, H2, H3, H4, H5); L 3 C N M R (100 MHz, D 2 0) 5 159.94 (C), 134.51 (2 CH), 119.64 (C), 116.81 (2 CH), 104.88 (CN), 99.35 (Cl) , 75.58 (d, J5.p 7.1 Hz, C5), 74.72, 72.80, 68.41, 62.28 (d, J6.p 4.3 Hz, C6); 3 1 P N M R (162 MHz, D 2 0 ) 8 4.28 (1 P, t, JH6,p 5.2 Hz); ESI-MS (high res) m/z: calc. for [Ci 3 H 1 5 NOoPNa 2 ] + 406.0280; Found: 406.0279; Anal. calc. For C 1 3 Hi4NNa 2 0 9 P-2H 2 0 : C, 35.38; H, 4.08; N , 3.17; Found: C, 35.72; H, 4.00; N , 3.16. Chapter 5 - Materials and Methods 196 Phenyl 6-phospho-pZ-D-glucopyranoside disodium salt (PpTG6P or 2.55). -OP0 3Na 2 Standard BglK-catalyzed phosphorylation of 2.45 (0.22 g, 0.86 mmole, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give the product as a white solid (26 mg, 0.068 mmol, 7.9 %): ' H N M R (400 MHz, D 2 0 ) 8 7.25-7.21 (2 H, m, Ar3, Ar5), 7.00-6.96 (3 H, m, Ar2, Ar4, Ar6), 4.97 (1 H, d, J , , 2 7.6 Hz, HI), 3.95-3.89 (1 H, m, H6 a), 3.82 (1 H, ddd, J 12.4, J 5.6, J 1.6 Hz, H6 b), 3.59-3.41 (4 H, m, H2, H3, H4, H5); l 3 C N M R (100 MHz, D 2 0 ) 8 156.44 (C), 129.84 (2 CH), 123.18 (CH), 116.43 (2 CH), 100.19 (Cl) , 75.52 (d, J 5 , P 6.9 Hz, C5), 74.82, 73.06, 68.51, 62.21 (d, J6.P 4.3 Hz, C6); 3 , P N M R (162 MHz, D 2 0 ) 8 5.80 (1 P, t, W 8.3 Hz); ESI-MS (high res) m/z: calc. for [C, 2 H,60 9 PNa 2 ] + 381.0327; Found: 381.0325; Anal. calc. For C 1 2 H, 5 Na 2 0 9 P- !/ 2 H 2 0: C, 37.03; H, 4.14; Found: C, 37.19; H, 4.76. 4-Butylphenyl 6-phospho-fi-D-glucopyranoside disodium salt (4tBuPfKj6P or 2.56). Standard BglK-catalyzed phosphorylation of 2.45 (0.74 g, 0.24 mmol, 1.0 equiv.). After HPLC purification, the pure fractions were collected and lyophilized to give 2.56 as a white solid (6.8 mg, 0.016 mmol, 6.5 %): ' H N M R (400 MHz, D 2 0 ) 8 7.31 (2 H, d, J A r2,Ar3 = J A r5 .Ar6 8.6 Hz, Ar3, Ar5), 6.95 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 8.6 Hz, Ar2, Ar6), 4.93 (1 H, d, J , , 2 7.2 Hz, HI), 3.95-3.89 (1 H, m, H6 a), 3.87-3.83 (1 H, m, H6 b), 3.57-3.40 (4 H, m, H2, H3, H4, H5), 1.12 (9 H, s, 3 CH 3 ) ; l 3 C N M R (100 MHz, D 2 0 ) 8 154.30 (C), 146.56 (C), 126.65 (2 CH), 116.25 (2 CH), 100.49 (Cl) , 75.48 (d, J 5 . P 6.9 Hz, C5), 74.90, 73.06, 68.60, 62.37 (d, J 6 , P 4.4 Hz, C6), 33.50 (C), 30.51 (3 CH 3 ) ; 3 1 P N M R (121 MHz, D 2 0 ) 8 4.24 (1 P, t, J H 6 , P 5.7 Hz); ESI-MS (high res) m/z: calc. for [C , 6H 2 4 0 9 PNa 2 ] + Chapter 5 - Materials and Methods 197 437.0953; Found: 437.0952; Anal . calc. For C, 6 H 2 3 09PNa 2 -2H 2 0: C, 40.69; H, 5.76; Found: C, 40.98; H , 6.04. Methyl 6-O-diphenylphospho-fi-D-glucopyranoside (2.57). C H 3 OH The starting material (2.18) (5 g, 26 mmol, 1.0 equiv.) was reacted with 5.9 mL (28 mmol, 1.1 equiv.) diphenyl chlorophosphate according to the general procedure described in Section 5.2.1.8. The product was purified by flash chromatography (7:2:1 EtOAc: MeOH:H 2 0) to yield a white solid (9.6 g, 23 mmol, 87 %): ' H N M R (400 MHz, Acetone) 5 7.43-7.39 (4 H, m, Ar), 7.32-7.22 (6 H, m, Ar), 4.66-4.61 (1 H, m), 4.51 (1 H, d, J 3.5 Hz, OH), 4.46-4.40 (2 H, m), 4.31 (1 H, d, J 2.8 Hz, OH), 4.21 (1 H, d, J,, 2 7.8 Hz, HI), 3.58-3.54 (1 H, m), 3.44-3.30 (5 H, m), 3.19-3.12 (1 H, m); l 3 C N M R (100 MHz, Acetone) 8 151.59 (C), 151.52 (C), 130.74 (4 CH), 126.25 (2 CH), 121.00 (CH), 120.98 (CH), 120.95 (CH), 120.94 (CH), 104.83 (Cl) , 77.58, 75.30 (d, J5,P 6.9 Hz, C5), 74.54, 70.68, 69.32 (d, J6,P 6.0 Hz, C6), 56.88 (CH 3); 3 , P N M R (162 MHz, Acetone) 8 -11.35 (1 P, t, J P ,6H 4.4 Hz, 6P); ESI-MS (high res) m/z: calc. for [C , 9 H 2 3 0 9 PNa] + 449.0977, Found: 449.0974. Methyl 6-phospho-fi-D-glucopyranoside disodium salt (MefiG6P or 2.58). Compound 2.57 (9.0 g, 21 mmol, 1.0 equiv.) was deprotected by catalytic hydrogenation as described in Section 5.2.1.9 to yield a gummy white solid (5.6 g, 19 mmol, 91 %): ' H N M R (400 MHz, D 2 0 ) 5 4.22 (1 H, d, J u 2 8.0 Hz, Hi ) , 4.01-3.96 (1 H, m, H6 a), 3.93-3.88 (1 H, m, H6 b), 3.40-3.30 (6 H, m, H3, H4, H5, CH 3 ) , 3.13-3.08 (1 H, m, H2); 1 3 C N M R (100 MHz, D 2 0 ) 8 103.44, 75.60, 74.75 (d, J 5 , P 8.0 Hz, C5), 73.17, 69.06, 63.88 (1C, d, J6,P 4.9 Hz, C6), 57.41; 3 I P N M R (121 MHz, D 2 0 ) 8 1.86; ESI-MS (high res) Chapter 5 - Materials and Methods 198 m/z: calc. for [CvHuNa^oP]"" 319.0171, Found: 319.0173; Ana l . calc. For C7H13O9-P N a 2 » 2 . 2 5 H 2 0 C, 23.44; H, 4.92; Found: C, 23.31; H, 4.71. l,2:5,6-Di-0-isopropylidene-a-D-ribo-3-hexulofuranulose hydrate (9.7 g, 35 mmoles, 1.0 equiv.) was added to 75 mL of 95 % EtOH and was stirred at 0 °C for 5 min. Subsequently, 2.4 g of NaBD 4 (18 mmoles, 0.5 equiv.) was added to the reaction mixture in two portions. The reaction mixture was allowed to warm up and was stirred at room temperature for 2.5 hrs. Ice was then added to the reaction mixture and acetic acid was added dropwise until bubbling subsided. The reaction mixture is then dried under reduced pressure. 100 mL of 5 % NaCl ( a q) was added and the product was extracted with 4 x methylene chloride, which was then washed with 1 x sat. NaCl, dried over MgS04, filtered and dried under reduced pressure to give relatively pure product as white crystals (8.7 g, 0.033 mmol, 94 %): ' H N M R (300 MHz, C D C I 3 ) 5 5.76 (1 H, d, ha 3.8 Hz, HI), 4.56 (1 H, d, ha 3.8 Hz, H2), 4.29 (1 H, m, H5), 4.08-3.96 (2 H, m, H6 a,6 b), 3.79 (1 H, d, J4.5 4.6 Hz, H4), 2.51 (1 H, broad s, OH), 1.56, 1.44, 1.36, 1.34 (12 H, 4 s, 4 CH 3 ) ; , 3 C N M R (75 MHz, C D C I 3 ) 5 112.77 (C), 109.78 (C), 103.89 (Cl) , 79.63 (C4), 78.86 (C2), 76.93 (C3), 75.5 (C5), 65.78 (C6), 26.52 (CH3), 26.45 (CH 3), 26.25 (CH 3), 25.22 (CH 3); E S I - M S (low res) m/z calc: 261; Found: 284.3 [M + Na] +. 1,2:5,6-Di-0-isopropylidene-a-D-3['H]-allofuranose (2.60). Chapter 5 - Materials and Methods 199 1,2:5,6-Di-04sopropylidene-3-0-toluene-p-sulfonyl-a-D-3[2H]-alIofuranose (2.61). Compound 2.60 (4.4 g, 17 mmoles, 1.0 equiv.) was dissolved in 60 mL of dry pyridine and cooled to 0 °C. /?-Toluenesulfonyl chloride (6.4 g, 34 mmol, 2.0 equiv.) was dissolved in 60 mL of dry pyridine, cooled to 0 °C and added to the starting material dropwise. The reaction mixture was allowed to warm up to room temperature and was stirred overnight under positive N 2 (g) pressure at room temperature. Upon completion, the reaction mixture was concentrated under reduced pressure. The compound was redissolved in 50 mL chloroform. The mixture was washed successively with water (1 x), sat. NaHC0 3 (1 x), sat. NaCl (1 x), and then dried over MgS0 4 . The crude sample was concentrated under reduced pressure and purified by flash chromatography (9:1 CHCl 3 :MeOH) to yield 2.61 (6.5 g, 16 mmol, 92 %) as a white crystalline solid: 'H NMR (300 MHz, CDC13) 5 7.85 (2 H, d, J 8.4 Hz, Ar), 7.32 (2 H, d, J 8.4 Hz, Ar), 5.74 (1 H, d, J,, 2 3.7 Hz, HI), 4.62 (1 H, d, J,, 2 3.7 Hz, H2), 4.18-4.12 (2 H, m, H4, H5), 3.92 (1 H, dd, j 6 a , 6 b 8.4, h M 6.5 Hz, H6 a), 3.76 (1 H , dd, j 6 a . 6 b 8.4, J 5 , 6 b 6.4 Hz, H6b), 2.43 (3 H, s, CH 3 ) , 1.51, 1.31, 1.28, 1.27 (12 H, 4 s, 4 CH 3 ) ; l 3 C NMR (75 MHz, CDC13) 8 145.14 (C), 133.20 (C), 129.63 (2 CH), 128.32 (2 CH), 113.58 (C), 109.90 (C), 103.82 (Cl) , 77.90 ,(C2), 76.89, 74.65 (C4, C5), 65.18 (C6), 26.63 (CH 3), 26.58 (CH 3), 26.05 (CH 3), 25.05 (CH 3), 21.64 (CH 3); ESI-MS ( l o w res) m/z calc.: 415; F o u n d : 438.4 [M + Na] + Chapter 5 - Materials and Methods 2 0 0 3-O-Benzoyl-l ,2:5,6-di-0-isopropylidene-a-D-3 [2 H]-glucofuranose (2.62). o 6.5 g (16 mmol, 1.0 equiv.) of the starting material was dissolved in dry DMF. This mixture was stirred vigorously and 25 g (0.17 mol, 11 equiv.) of sodium benzoate was added. The reaction mixture was refluxed under positive N2(g) pressure overnight. Upon completion, the reaction mixture was cooled, and water was added to dissolve the NaOBz and the product was extracted into CHCI3 (2 x). The organic fractions were combined and dried over MgSCM, filtered and concentrated under reduced pressure to yield a brown syrup, which was not purified and directly used for the subsequent reaction. 1,2:5,6-Di-O-isopropylidene-a-D-3[2H]-allofuranose (2.63). A crude sample of 2.62 was dissolved in 300 mL of dry MeOH. A catalytic amount of NaOMe was added to the reaction and the mixture was stirred at room temperature for approximately 30 minutes until completion by TLC analysis. The reaction was quenched with Amberlite IR-120 (H+) resin, which was removed by filtration. The filtrate was concentrated under reduced pressure to yield a pale yellow syrup. The crude product was not purified and was used directly in the subsequent reaction. Chapter 5 - Materials and Methods 201 3'["HJ-D-gluCopyranose (2.64). A crude sample of compound 2.63 was dissolved in water and 2 g of Amberlite IR-120 (H+) resin was added. The suspension was stirred at 80 °C for 5 hours, until completion by TLC analysis. The resin was removed by filtration and the filtrate was concentrated under reduced pressure to yield the crude product, which was used directly for the acetylation reaction to generate 2.69. 1,2,3,4,6-Penta-O-acetyl-1 -[2H]-/3-D-glucopyranose (2.67) •OAc This compound was generated by the general acetylation reaction of 2.65 (2.0 g, 11 mmol, 1.0 equiv.) in NaOAc/Ac20. The product was recrystallized in EtOH as white crystals (3.5 g, 9.1 mmol, 83 %): ' H N M R (300 MHz, CDCh) 5 5.23 (1 H, t, J 9.4 Hz, H3), 5.14-5.08 (2 H, m, H2, H4), 4.27 (1 H, dd, J 6 a,6b 12.5, J 5 , 6 a 4.5 Hz, H6 a), 4.09 (1 H, dd, J 6 a , 6 b 12.5, J 5 , 6 b 2.1 Hz, H6 b), 3.82 (1 H, ddd, J 4 , 5 10.1, J5.6a 4.5, J 5 . 6 b 2.1 Hz, H5), 2.09 (3 H, 1 s, 1 Ac), 2.07 (3 H, 1 s, 1 Ac), 2.01 (6 H, 1 s, 2 Ac), 1.99 (3 H , 1 s, 1 Ac); , 3 C N M R (75 MHz, CDCh) 5 170.57 (C=0), 170.07 (C=0), 169.36 (C=0), 169.22 (C=0), 168.93 (C=0), 72.76 (C3/5), 72.67 (C3/5), 70.14 (C4), 67.72 (C2), 61.42 (C6), 20.78 (CH 3), 20.68 (CH 3), 20.54 (3 CH 3 ) ; ESI-MS (high res) m/z: calc. for [C , 6 H 2 iDOi ,Na] +: 414.1123 Found: 414.1115; Anal . calc. For d ^ D O , , : C, 49.11; H, 5.88; Found: C, 49.42; H, 5.79. C h a p t e r 5 - M a t e r i a l s and M e t h o d s 202 1,2,3,4,6-Penta-0-acetyl-2-[2H]-D-glucopyranose (2.68). .OAc Acetylation of 2.66 (1.9 g, 11 mmol, 1.0 equiv.) in N a O A c / A c 2 0 as described for •compound 2.67. The product was recrystallized in EtOH as shiny white crystals (3.6 g, 9.2 mmol, 84 %) as the anomeric mixture: ' H N M R (300 MHz, CDC13) 5 6.24 (1 H, s, aHl ) , 5.39 (1 H, d, J a 3,4 9.5 Hz, aH3), 5.18 (1 H, d, J p 3 . 4 9.4 Hz, pH3), 5.08-5.02 (2 H, m, aH4, pH4), 4.24-4.17 (m, aH6 a , pH6 a), 4.05-3.99 (m, aH5, aH6 b , pH6 b), 3.81-3.75 (1 H, m, pH5), 2.15, 2.10, 2.03, 2.01, 2.00, 1.96, 1.95, 1.93 (20 H, 10 A c ) ; ESI-MS (high res) m/z: calc. for [ C i 6 H 2 i D O n D N a ] + 414.1123; Found: 414.1107; Anal. calc. for C , 6 H 2 , D O u : C , 49.11; H, 5.88; Found: C, 49.26; H, 5.70. I, 2,3,4,6-Penta-0-acetyl-3-[2 H]-D-gluCopyranose (2.69). This compound was generated by the general acetylation reaction of crude 2.64 in NaOAc/Ac 2 0. The product was recrystallized in EtOH (3.5 g, 8.9 mmol, 57 % over three steps from 2.62) as the anomeric mixture: ' H N M R (300 MHz, C D C I 3 ) 5 6.29 (1 H, d, Jai.2 3.7 Hz, aHl ) , 5.68 (1 H, d, J p u 8.3 Hz, pHl) , 5.12-5.05 (4 H, m, aH2, aH4, pH2, pH4), 4.28-4.20 (2 H, m, aH6 a , pH6 a), 4.11-4.03 (3 H, m, aH6 b , pH6 b, aH5), 3.80 (1 H, m,pH5), 2.14, 2.07, 2.05,2.04,2.02,2.01,2.00, 1.99, 1.98, 1.97 (30 H, 10 s, 10 Ac) l 3 C N M R (75 MHz, C D C I 3 ) 5 170.51 ( O O ) , 169.93 (2 C=0), 169.55 (C=0), 169.44 (C=0), 169.09 (C=0), 168.85 (C=0), 168.65 (C=0), 168.33 (C=0), 168.20 (C=0), 91.64 (pCl), 89.01 (aCl) , 72.18 , 69.82, 69.41, 68.73, 67.42, 67.35, 61.07, 59.78, 20.79 (2 CH 3 ) , 20.73 (2 CH 3 ) , 20.61 (2 CH 3 ) , 20.48 (2 CH 3 ) , 20.36 (2 CH 3 ) ; ESI-MS (high res.) m/z: calc. for [C , 6 H 2 1 DO, ,Na] + : 414.1112 [M + Na] +; Anal. calc. for C , 6 H 2 1 D O i i : C, 49.11; H, 5.88; Found: C, 49.08; H, 5.61. .OAc OAc D Chapter 5 - Materials and Methods 203 2,3,4,6-Tetra-O-acetyl-1-['HJ-a-D-glucopyranosyl bromide (2.70). -OAc AcO' D Br The a-bromide 2.70 was obtained from 2.67 (1.0 g, 2.6 mmol, 1.0 equiv.) as described for 2.26. Shiny white crystals (1.0 g, 2.4 mmol, 93 %) were recrystallized from EtOH: ' H N M R (400 MHz, CDCU) 5 5.54 (1 H, t, J 2 , 3 = J 3 , 4 9.9 Hz, H3), 5.14 (1 H, t, J 3 , 4 = J 4 . 5 9.9 Hz, H4), 4.81 (1 H, d, J 2 , 3 10.0 Hz, H2), 4.33-4.25 (2 H, m, H5, H6 a), 4.13-4.10 (1 H, m, H6 b), 2.09, 2.08, 2.03, 2.02 (12 H, 4 s, 4 Ac); l 3 C N M R (100 MHz, CDC13) 5 170.49, 169.83, 169.78, 169.44, 72.09 (C5), 70.52 (C2), 70.13 (C3), 67.15 (C4), 60.93 (C6), 20.65 (2 CH 3 ) , 20.61, 20.54; ESI-MS (high res) m/z: calc. for [ C 1 4 H 1 8 D B r 0 9 N a ] + 434.0173, 436.0152, Found: 434.0172, 436.0153; Anal. calc. For C , 4 H l 8 D B r 0 9 : C, 40.79; H, 4.85; Found: C, 41.06; H, 4.78. 2,3,4,6-Tetra-0-acetyl-2-[2H]-a-D-glucopyranosyl bromide (2.71). The a-bromide 2.71 was obtained from 2.68 (1.0 g, 2.6 mmol, 1.0 equiv.) as described for 2.26. Shiny white crystals (0.95 g, 2.3 mmol, 88 %) were recrystallized from EtOH: ' r l N M R (300 MHz, CDC13) 5 6.59 (1 H, s, HI), 5.54 (1 H, d, J 3 , 4 9.4 Hz, H3), 5.17-5.11 (1 H, m, H4), 4.34-4.25 (2 H, m, H5, H6 a), 4.13-4.09 (1 H, m, H6 b), 2.09, 2.08, 2.03, 2.01 (12 H, 4 s, 4 Ac); , 3 C N M R (75 MHz, CDC13) 5 170.45 (C=0), 169.80 (C=0), 169.75 (C=0), 169.42 (C=0), 86.48 (Cl) , 72,14 (C5), 70.55 (C2), 70.13 (C3), 67.21 (C4), 60.96 (C6), 20.63 (3 lines), 20.53 (4 Ac); ESI-MS (High res.) m/z calc. for [ C 1 4 H 1 8 D B r 0 9 N a f 434.0173, 436.0152; Found: 434.0163, 436.0150 [M + Na] +; Anal. calc. for C, 4 H,xDBr0 9 : C, 40.79; H, 4.85; Found: C, 40.32; H, 4.71. .OAc Br Chapter 5 - Materials and Methods 204 2,3,4,6-Tetra-0-acetyl-3-[2H]-a-D-glucopyranosyl bromide (2.72). .OAc A c O ^ A ^ - * ^ A I OAc) D Br Shiny white crystalline 2.72 (1.0 g, 2.4 mmol, 92 %) is prepared from 2.64 (1.0 g, 2.6 mmol, 1.0 equiv.) as described for 2.62: ' H N M R (400 MHz, CDCI3) 5 6.59 (1 H, d, J,, 2 4.0 Hz, HI), 5.14 (1 H, d, J4.5 10.1 Hz, H4),4.81 (1 H, d, J,, 2 4.0 Hz, H2), 4.33-4.25 (2 H, m, H5, H6 a), 4.13-4.09 (1 H, m, H6 b), 2.08, 2.07, 2.03, 2.01 (12 H, 4 s, 4 Ac); l 3 C N M R (100 MHz, CDCI3) 5 169.83 ( O O ) , 169.79 ( O O ) , 169.46 (2 O O ) , 86.52 (Cl) , 72,10 (C5), 70.51 (C2), 67.06 (C4), 60.92 (C6), 20.64, 20.63, 20.60, 20.53 (4 Ac); ESI-MS (high res) m/z calc. for [C i 4 Hi 8 DBr0 9 Na] + 434.0165, 436.0154; Found: 434.0163, 436.0150 [M + Na] +; Anal. calc. for C 1 4 H,xDBr0 9 : C, 40.79; H, 4.85; Found: C, 40.52; H,4.88. 4-Nitrophenyl 2,3,4,6-tetra-0-acetyl-l-[2H]-fi-D-glucopyranoside (2.73). 2.73 was prepared from 2.70 (1.0 g, 2.4 mmol, 1.0 equiv.) using the Koenigs-Knorr reaction. The product was recrystallized from EtOH as a white solid (0.48 g, 1.0 mmol, 42 %): ' H N M R (400 MHz, CDCI3) 5 8.20 (2 H, d, J A r 2 , A r 3 = W A K S 9.3 Hz, Ar3, Ar5), 7.05 (2 H, d, J A r 2 ,Ar3 = J A r5 . A r6 9.3 Hz, Ar2, Ar6), 5.33-5.27 (2 H, m, H2, H3), 5.18-5.14 (1 H, m, H4), 4.26 (1 H, dd, J 6 a , 6 b 12.4, J 5 , 6 a 5.5 Hz, H6 a), 4.16 (1 H, dd, 12.4, J 5 , 6 b 2.4 Hz, H6 b), 3.91 (1 H, ddd, J 4 , 5 10.0, J 5 > 6 a 5.5, J 5 . 6 b 2.4 Hz, H5), 2.05 (3 H, s, Ac), 2.04 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.02 (3 H, s, Ac); l 3 C N M R (100 MHz, CDC13) 5 170.12 (2 O O ) , 169.31 (2 O O ) , 161.12(C), 143.26(C), 125.78 (2 CH), 116.60(2 CH), 72.41 (C3 & C5), 70.86 (C2), 68.00 (C4), 61.80 (C6), 20.65 (2 Ac), 20.56 (2 Ac); ESI-MS (high res) m/z: calc for [ C 2 0 H 2 2 D N O i 2 N a ] + 493.1181; Found: 493.1186; Anal . calc. for O j H ^ D N O i r V y ^ O : C, 50.10; H, 5.01; N , 2.92; Found: C, 50.03; H, 4.98; N , 3.08. .OAc Chapter 5 - Materials and Methods 205 4-Nitrophenyl 2,3,4,6-tetra-0-acetyl-2-[2H]-/3-D-glvcopyranoside (2.74). 2.1 A\ was prepared from 2.71 (0.95 g, 2.3 mmol, 1.0 equiv.) using the Koenigs-Knorr reaction. The product was recrystallized from EtOH as shiny white crystals (0.55 g, 1.2 mmol, 51 %): 'B N M R (300 MHz, CDC13) 8 8.20 (2 H, d, J A r 2 . A r 3 = W A ^ 9.2 Hz, Ar3, Ar5), 7.05 (2 H, d, J A r 2 ,Ar3 = JAFS^J 9.2 Hz, Ar2, Ar6), 5.30 (1 H, d, j 3 , 4 9.4 Hz, H3), 5.20-5.13 (2 H, m, HI , H4), 4.27 (1 H, dd, J 6 a .6b 12.4, hM 5.4 Hz, H6 a), 4.15 (1 H, dd, J 6 a ,6 b 12.4, J 5 ,6 b 2.4 Hz, H6 b), 3.91 (1 H, ddd, J 4 , 5 9.9, J 5 , 6 a 5.4, J 5 > 6 b 2.4 Hz, H5), 2.05 (3H, s, Ac), 2.04 (3H, s, Ac), 2.03 (3H, s, Ac), 2.02 (3H, s, Ac); l 3 C N M R (100 MHz, CDC13) 5 170.38 (C=0), 170.12 ( O O ) , 169.32 ( O O ) , 169.14 ( O O ) , 161.14 (C), 143.29 (C), 125.78 (2 CH), 116.62 (2 CH), 98.03 (Cl) , 72.43 (C5), 72.35 (C3), 68.02 (C4), 61.81 (C6), 20.64 (CH 3), 20.55 (3 CH 3 ) ; ESI-MS (high res) m/z calc. for [C2oH 2 2 DNO ,2Na] + 493.1181; Found: 493.1167; Anal. calc. for C20H22DNC2: C, 51.08; H, 5.11; N , 2.98; Found: C, 51.70; H, 4.89; N , 3.15. 4-Nitrophenyl 2J,4,6-tetra-0-acetyl-3-[2H]-fi-D-glucopyranoside (2.75). 2.75 was prepared from 2.72 (1.0 g, 2.4 mmol, 1.0 equiv.) using the Koenigs-Knorr reaction. The product was recrystallized from EtOH as shiny white crystals (0.46 g, 0.98 mmol, 41 %): ' H N M R (400 MHz, CDC13) 8 8.20 (2 H, d, J A r 2 .Ar3 = J A ^ 9.3 Hz, Ar3, Ar5), 7.05 (2 H, d, J A r 2 .Ar3 = W A K S 9.3 Hz, Ar2, Ar6), 5.29 (1 H, d, J,, 2 7.8 Hz, H2), 5.20 (1 H, d, J,, 2 7.8 Hz, HI), 5.16 (1 H, d, J4,510.1 Hz, H4), 4.27 (1 H, dd, J 6 a ,6b 12.4, J 5 , 6 a 5.5 Hz, H6 a), 4.16 (1 H, dd, J 6 a , 6 b 12.4, J 5 , 6 b 2.4 Hz, H6 b), 3.91 (1 H, ddd, J 4 , 5 10.1, J 5,6a 5.5, j 5 , 6b 2.4 Hz, H5), 2.06, 2.05, 2.04, 2.03 (12 H, 4 s, 4 Ac); , 3 C N M R (100 MHz, CDC13) 8 170.41 ( O O ) , 170.14 ( O O ) , 169.34 ( O O ) , 169.18 ( O O ) , 161.12 (C), 143.26 (C), 125.79 (2 CH), 116.60 (2 CH), 98.04 (Cl) , 72.41 (C5), 70.84 (C2), 67.91 (C4), 61.79 Chapter 5 - Materials and Methods 206 (C6), 20.66, 20.59, 20.57 (2 CH 3 ) ; ESI-MS (low res) m/z calc.: 470; Found: 493.0 [M + Naf ; Anal . calc. for C i 2 H 2 2 D N O , 2 : C, 51.08; H, 5.11; N , 2.98; Found: C, 51.56; H, 4.95; N , 3.12. 4-Nitrophenyl l-[2H]-f3-D-glucopyranoside (2.76). Zemplen deprotection of 2.73 (0.48 g, 1.0 mmol, 1.0 equiv.) afforded 2.76 as a white solid (0.29 g, 0.95 mmol, 95 %): ' H N M R (400 MHz, D 2 0 ) 5 8.11 (2 H, d, JA r2,Ar3 = JA r5.Ar6 9.3 Hz, Ar3, Ar5), 7.09 (2 H, d, J A r 2 , A r3 = - W r f 9.3 Hz, Ar2, Ar6), 3.79 (1 H, dd, J 6 a ,6 b 12.3, J 5 , 6 a 2.0 Hz, H6 a), 3.61 (1 H, dd, J 6 a , 6 b 12.3, J 5.6 b 5.7 Hz, H6 b), 3.57 (1 H, m, H5), 3.49-3.47 (2 H, m, H2, H3), 3.39-3.36 (1 H, m, H4); ESI-MS (high res) m/z: calc. for [Ci 2 Hi 4 DNOsNa] + : 325.0758, Found: 325.0761; Ana l . calc. for C , 2 H i 4 D N C v C, 47.68; H, 5.30; N , 4.63; Found: C, 46.91; H, 5.19; N , 4.60. 4-Nitrophenyl 2-[2 H]-f3-D-glucopyranoside (2.77). Zemplen deprotection of 2.74 (0.55 g, 1.2 mmol, 1.0 equiv.) afforded 2.77 as a white solid (0.32 g, 1.1 mmol, 89 %): l H N M R (400 MHz, CDC13) 8 8.09 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 9.3 Hz, Ar3, Ar5), 7.07 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 9.3 Hz, Ar2, Ar6), 5.09 (1 H, s, HI), 3.77 (1 H, dd, ) 6 a , 6 b 12.3, J5M 1.8 Hz, H6 a), 3.59 (1 H, dd, J 6 a .6 b 12.3, J 5 . 6 b 5.7 Hz, H6 b), 3.55-3.50 (1 H, m, H5), 3.46 (1 H, d, J 3 , 4 9.4 Hz, H3), 3.34 (1 H, t, J 3 , 4 J 4 , 5 9.4 Hz, H4); l 3 C N M R (100 MHz, CDC13) 5 161.56, 142.45, 125.96 (2 CH), 116.30 (2 CH), 99.24 (Cl) , 76.11, 75.19, 69.15, 60.28 (C6); ESI-MS (high res) m/z calc. for [C, 2 H 1 4 DNOxNa] + : 325.0758; Found: 325.0760; Ana l . calc. for C , 2 H 1 4 D N O x : C, 47.68; H, 5.30; N , 4.63; Found: C, 47.51; H, 5.15; N , 4.73. Chapter 5 - Materials and Methods 207 .OH HO' Zemplen deprotection of 2 . 75 (0.46 g, 0.98 mmol, 1.0 equiv.) afforded 2 . 7 8 as a white solid (0.29 g, 0.95 mmol, 97 %): ' H N M R (400 MHz, CDC13) 8 8.05 (2 H, d, J A r 2 . A r 3 = J A r 5 , A r 6 9.3 Hz, Ar3, Ar5), 7.05 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 9.3 Hz, Ar2, Ar6), 5.07 (1 H, d, Ji,2 7.9 Hz, HI), 3.77 (1 H, dd, J 6 a , 6 b 12.4, J 5 ,6 a 1.9 Hz, H6 a), 3.59 (1 H, dd, J6a,6b 12.4, J 5 , 6 b 5.7 Hz, H6 b), 3.52 (1 H, m, H5), 3.45 (1 H, d, J,, 2 7.9 Hz, H2), 3.34 (1 H, d, J 4 , 5 9.8 Hz, H4); 1 3 C N M R (100 MHz, CDC13) 8 161.33, 142.09, 125.65 (2 CH), 116.50 (2 CH), 99.05 (Cl) , 75.85 (C5,C3), 72.,27 (C2), 68.83 (C4), 60.05 (C6); ESI-MS (high res) m/z calc. for [C i 2 H 1 4 DNO x Na] + 325.0758; Found: 325.0757; Anal. calc. for C!2Hi4DNO><: C, 47.68; H, 5.30; N , 4.63; Found: C, 46.58; H, 5.05; N , 4.57. 4-Nitrophenyl 6-phospho-1 -[2 HJ-fi-D-glucopyranoside disodium salt (J/2H/4NPplG6P or Compound 2.76 (0.29 g, 0.95 mmol, 1.0 equiv.) was phosphorylated with BglK as described in Section 5.2.1.7. 2.79 was purified by HPLC (C18 column) using 5 % acetonitrile in water as the elution solvent and the pure fractions were lyophilized to yield a white solid (0.12 g, 0.29 mmol, 30 %): 'H N M R (400 MHz, D 2 0 ) 8 8.05 (2 H, d, JAr2,Ar3 = JAr5.Ar6 9.3 Hz, Ar3, Ar5), 7.06 (2 H, d, J A r 2 , A r 3 = JAT5,AT6 9.3 Hz, Ar2, Ar6), 3.98-3.93 (I H, m, H6 a), 3.88-3.84 (1 H, m, H6 b), 3.59-3.49 (4 H, m, H2, H3, H4, H5); l 3 C N M R (100 MHz, D 2 0 ) 8 161.67 (C), 142.29 (C), 126.12 (2CH), 116.40 (2CH), 99.18 (t, J i . D 23.4 Hz, C l ) , 75.68 (d, J 5 . p 76.5 Hz, C5), 74.58 (C3), 72.78 (C2), 68.46 (C4), 62.54 (d, J 6 , P 3.9 Hz, C6); 3 l P N M R (162 MHz, D 2 0) 8 5.41 (t, J H 6 , p 6.1 Hz); ESI-MS (high res) m/z: calc. for [ C 1 2 H 1 4 D N O n P N a 2 ] + 427.0241; Found: 427.0245; Anal. calc. For C | 2 H 1 3 D N N a 2 O i l P 2 H 2 0 : C, 31.18; H, 4.11; N , 3.03; Found: C, 31.40; H, 4.55; N , 3.36. 2.79). Chapter 5 - Materials and Methods 208 2.80). Compound 2.77 (0.32 g, 1.1 mmol, 1.0 equiv.) was phosphorylated with BglK as described in Section 5.2.1.7. 2.80 was purified by HPLC (C18 column) using 5 % acetonitrile in water as the elution solvent and the pure fractions were lyophilized to yield a white solid (0.10 g, 0.23 mmol, 21 %): ' H N M R (300 MHz, D 2 0 ) 8 8.08 (2 H , d, J A r2,Ar3 = W r 6 9.3 Hz, Ar3, Ar5), 7.09 (2 H, d, J A r2,Ar3 = J A r 5 . A r 6 9.3 Hz, Ar2, Ar6), 5.10 (1 H, s, HI), 3.93-3.81 (2 H, m, H6 a, H6 b), 3.60-3.48 (3 H, m, H3, H4, H5); , 3 C N M R (100 MHz, D 2 0) 8 161.67 (C), 142.54 (C), 126.07 (2CH), 116.43 (2CH), 99.49 (Cl) , 75.84 (d, J 5 , P 6.9 Hz, C5), 74.65 (C3), 72.50 (t, J 2 ) D 23.3 Hz, C2), 68.45 (C4), 62.12 (d, J 6 , P 4.4 Hz, C6); 3 I P N M R (162 MHz, D 2 0) 5 5.07 (t, JH6,p 6.0 Hz); High-resolution ESI-MS (high res) m/z calc. for: [Ci 2 H 1 4 DNOi,P]" 381.0446;, Found: 381.0441; Ana l , calc. for C 1 2 H I 3 D N N a 2 0 , i P - 4 H 2 0 : C, 28.91; H, 4.62; N , 2.81; Found: C, 29.27 ; H, 4.96 ; N , 2.92. 4-Nitrophenyl 6-phospho-3-[2H]-f3-D-glucopyranoside disodium salt (3l2H]4NPfiG6P or Compound 2.78 (0.29 g, 0.95 mmol, 1.0 equiv.) was phosphorylated with BglK as described in Section 5.2.1.7. 2.81 was purified by HPLC (C18 column) using 5 % acetonitrile in water and the pure fractions were lyophilized to yield a white solid (0.097 g, 0.23 mmol, 24 %): ' H N M R (400 MHz, D 2 0) 8 8.10 (2 H, d, J Ar2,Ar3 — JAr5,Ar6 9.3 Hz, Ar3, Ar5), 7.10 (2 H, d, J A r 2 ,Ar3 = W r 6 9.3 Hz, Ar2, Ar6), 5.11 (1 H, d, J,, 2 7.9 Hz, HI), 3.94-3.89 (1 H, m, H6 a), 3.82 (1 H, m, H6 b), 3.59-3.54 (2 H, m, H4, H5), 3.49 (1 H, d, J 1 > 2 2.81). -2 Chapter 5 - Materials and Methods 209 7.9 Hz, H2); l 3 C N M R (100 MHz, D 2 0 ) 5 161.59 (C), 142.49 (C), 126.01 (2CH), 116.34 (2CH), 99.43 (Cl) , 75.78 (d, J5,p 7.0 Hz, C5), 74.39 (C3), 72.78 (C2), 68.26 (C4), 62.18 (d, J6,p 4.4 Hz, C6); 3 , P N M R (162 MHz, D 2 0) 5 5.04 (t, JH6,p 6.1 Hz); ESI-MS (high res) m/z calc. for [Ci 2 H, 4 DNO, iP]~ 381.0446, found: 381.0451; Anal . calc. For C ^ H n D N N ^ O n P S . S ^ O : C, 30.01; H, 4.95; N , 2.67; Found: C, 26.96; H, 4.51; N , 2.46. Methyl 2,3,4,6-tetra-O-acetyl-1 -[2H]-/3-D-glucopyranoside (2.82). Compound 2.70 (1.0 g, 2.4 mmol, 1.0 equiv.) was dissolved in 12 mL of anhydrous CHC1 3 and stirred at room temperature with 4 A molecular sieves under N 2 ( g ) for 15 min. 0.5 mL of methanol (12 mmol, 5.0 equiv.), 2.0 g A g 2 C 0 3 (7.2 mmol, 3.0 equiv.), several crystals of 12, and 4 A molecular sieves were added to 12 mL of anhydrous CHC1 3 and stirred at room temperature under N 2 (g) for 15 min. The mixture containing methanol and silver carbonate was cannulated into the suspension of 2.70 and 4 A molecular sieves. The reaction mixture was stirred at room temperature overnight, then diluted with EtOAc and filtered through a Celite cake. The filtrate was concentrated under reduced pressure and purified via flash chromatography (a mixture of petroleum ethenEtOAc starting from a ratio of 3:1 to 1:1). The product was isolated as a white crystalline solid (0.52 g, 1.4 mmol, 62%): ' H N M R (300 MHz, CDC13) § 5.09 (1 H, t, J 3 . 4 = J4.5 9.4 Hz, H4), 4.97 (1 H, t, J2.3 = J3.4 9.4 Hz, H3), 4.85 (1 H, d, J 2 , 3 9.4 Hz, H2), 4.17 (1 H, dd, J 6 a ,6b 12.3, J 5.6a 4.7 Hz, H6 a), 4.03 (1 H, dd, J 6 a , 6 b 12.3, J 5 , 6 b 2.4 Hz, H6 b), 3.61 (1 H, ddd, J 4 , 5 9.4, J 5,6a 4.7, J 5 , 6 b 2.4 Hz, H5), 3.39 (3 H, s, CH 3 ) , 1.97 (3 H, s, Ac), 1.93 (3 H, s, Ac), 1.91 (3 H, s, Ac), 1.88 (3 H, s, Ac); l 3 C N M R (75 MHz, CDC13) 5 170.35 (C=0), 169.93 (C=0), 169.14 (C=0), 169.09 (C=0), 72.63, 71.51, 70.94, 68.21, 61.69, 56.66 (CH 3), 20.41 (2 Ac), 20.31 (2 Ac); ESI-MS (high res) m/z: calc for [C, 5 H 2 ,DOioNa] + : 386.1173, Found: ,OAc D 386.1174. Chapter 5 - Materials and Methods 210 Methyl l-[2H]-J3-D-glucopyranoside (2.83). Zemplen deprotection of 2.82 (0.52 g, 1.4 mmol, 1.0 equiv.) afforded 2.83 as shiny white crystals (0.27 g, 1.4 mmol, 100 %): ' H N M R (300 MHz, D 2 0 ) 5 3.74 (1 H, dd, J6a,6b 12.3, UM 2.2 Hz, H6 a), 3.53 (1 H, dd, J6a.6b 12.3, J5.6b 5.8 Hz, H6b), 3.38 (3 H, s, CH 3 ) , 3.33-3.15 (3 H, m, H3, H4, H5), 3.07 (1 H, d, J 2 , 3 9.2 Hz, H2); l 3 C N M R (75 MHz, D 2 0) 5 75.97, 75.83, 73.10, 69.72, 60.82 (C6), 57.20 (CH 3); ESI-MS (high res) m/z: calc.for [C 7 H, 3 D06Na] + : 218.0751, Found: 218.0752. Methyl 1'-['' H]-6-diphenylphospho-/3-D-glucopyranoside (2.84). Compound 2.84 (0.51 g, 1.2 mmol, 89 %) was synthesized from 2.83 (0.27 g, 1.4 mmol, 1.0 equiv.) according to the general chemical phosphorylation methods using diphenyl chlorophosphate as described in Section 5.2.1.8: ' H N M R (400 MHz, Acetone) 5 7.44-7.39 (4 H, m, Ar), 7.33-7.24 (6 H, m, Ar), 4.67-4.56 (2 H, m), 4.48-4.39 (2 H, m), 3.60-3.55 (1 H, m), 3.49-3.31 (5 H, m), 3.23-3.19 (1 H, m), 2.99 (1 H, s, OH); l 3 C N M R (75 MHz, C D 3 O D ) 5 151.79 (C), 151.70 (C), 131.09 (4 CH) , 126.87 (2 CH) , 121.23 (2 CH) , 121.17 (2 CH), 104.76 (t, J , , D 24.6 Hz, C l ) , 77.75, 75.70 (d, J5,P 6.7 Hz, C5), 74.76, 70.95, 69.81 (d, J 6 , p 5.9 Hz, C6), 57.33 (CH 3 ) ; 3 1 P N M R (121 MHz, C D 3 O D ) 5 -12.18 (1 P, t, JP,H6 6.4 Hz, 6P); ESI-MS (high res) m/z: calc. for [C,oH 2 2DO()PNa]+ : 450.1040, Found: 450.1042. D Chapter 5 - Materials and Methods 211 Methyl 1-["HJ-6-diphenylphospho-^-D-glucopyrano side (ll2H]MepG6P or 2.85). Phenyl ester protecting groups of 2.84 (0.45 g, 1.1 mmol, 1.0 equiv.) were deprotected via catalytic hydrogenation as described in Section 5.2.1.9 to yield the product as a white gummy solid (0.35 g, 1.1 mmol, 100%): ' H NMR (400 MHz, D 2 0 ) 5 3.97-3.87 (2 H, m, H6 a, H6 b), 3.42-3.32 (6 H, m, H3, H4, H5, CH 3 ) , 3.13 (1 H, d, J 2 , 3 8.8 Hz, H2); , 3 C N M R (75 MHz, D 2 0) 5 103.05 (t, J,,D24.3 Hz, C l ) , 75.49, 75.00 (d, J 5 , P 7.7 Hz, C5), 73.19, 69.05, 63.42 (1C, d, h.p 4.5 Hz, C6), 57.39 (CH 3); 3 , P N M R (162 MHz, D 2 0) 8 2.39; ESI-MS (high res) m/z: calc. for [C 7H, 3G\)PNa 2] +: 320.0234, Found: 320.0233; Anal, calc. For C 7H 1 2D09PNa2 '2H 20 C, 23.67; H, 5.07; Found: C, 24.00; H, 5.22. 1,2:5,6-Di-0-isopropylidene-3-0-phenylthiocarbonate-a-D-glucofuranose (2.87) l,2:5,6-Di-0-isopropylidene-oc-D-glucofuranose (2.86) (1.0 g, 3.9 mmol, 1.0 equiv.) was dissolved in 100 mL dry CH 2 C1 2 . To this solution, 1.4 g of 4-(dimethylamino)pyridine (DMAP) (12 mmol, 3.0 equiv.) and 0.58 mL of PhOC(S)Cl (4.2 mmol, 1.1 equiv.) was added. The reaction mixture was stirred at room temperature under positive N2(g) pressure overnight. The reaction was diluted with 100 mL CH 2 C1 2 , washed successively with 2 M HC1 (1 x), sat. NaHC0 3 (1 x), H 2 0 (until aqueous layer is neutral), and sat. NaCl (1 x). The organic fraction was dried over MgSC>4, filtered and concentrated under reduced pressure. The product was purified by flash chromatography (3:1 petroleum ether:EtOAc) as a colorless oil (1.4 g, 3.6 mmol, 93 %): ' H N M R (400 MHz, CDC13) 5 7.41 (2 H, t, J 7.4 Hz, Ph), 7.30 (1 H, t, J 7.4 Hz, Ph), 7.12-7.10 (2 H, m, Ph), 5.95 (1 H, ,OP0 3 Na 2 D Chapter 5 - Materials and Methods 212 d, J1.2 3.8 Hz, HI), 5.62(1 H, d, J 1.8 Hz, H3), 4.76(1 H,d , J,, 2 3.8 Hz, H2), 4.30-4.29 (2 H, m, H4, H5), 4.12-4.08 (1 H, m, H6 a), 4.06-4.03 (1 H, m, H6b), 1.53(3 H, s, CH 3 ) , 1.43 (3 H, s, CH 3 ) , 1.35 (3 H, s, CH 3 ) , 1.33 (3 H, s, CH 3 ) ; I 3 C N M R (75 MHz, CDC13) 5 193.60 (C=S), 153.18 (C), 129.54 (2 CH), 126.71 (CH), 121.67 (2 CH), 112.42 (C), 109.38 (C), 104.95 (Cl) , 85.03 (C3), 82.84 (C2), 79.63 (C4), 72.23 (C5), 67.03 (C6), 26.83 (CH 3) , 26.59 (CH 3), 26.20 (CH 3), 25.24 (CH 3); ESI-MS (low res) m/z calc.: 396; Found: 419.2 [M+Na]+ I, 2:5,6-Di-O-isopropylidene-a-D-ribo-hexofuranose (2.88). Compound 2.87 (1.4 g, 3.6 mmol, 1.0 equiv.) was coevaporated with dry toluene several times, then dissolved in 100 mL dry toluene. 4.8 mL of Bu 3SnH (18 mmol, 5.0 equiv.) and 0.12 mg of 2,2'-azobisisobutyronitrile (A1BN) (0.72 mmol, 0.2 equiv.) was added to the reaction mixture, which was refluxed under positive N 2 (g> overnight. The reaction was cooled and concentrated under reduced pressure. The crude product was passed through a short silica gel column (6:1 petroleum ether:EtOAc) to remove organic tin byproducts. The eluate was concentrated under reduced pressure and was not further purified but used directly for the subsequent reaction. 3-Deoxy D-riho-hexopyranose (2.89). Crude 2.88 was dissolved in water. 2 g of Amberlite 1R-120 (H +) resin was added and the suspension was refluxed for 2 hours. The cation exchange resin was removed by O H Chapter 5 - Materials and Methods 2 1 3 filtration and the filtrate was concentrated under reduced pressure. The crude product was isolated as a white gum and was not further purified. 1,2,4,6-Tetra-O-acetyl 3-deoxy-D-ribo-hexopyranose (2.90) OAc Compound 2.90 was synthesized according to the general acetylation method described in Section 5.2.1.1. The crude product (0.86 g, 2.6 mmol, 72 % over 3 steps) was precipitated in EtOH and was not further purified. 2,4,6-Tri-O-acetyl 3-deoxy-a-D-ribo-hexopyranosyl bromide (2.91) Br A crude sample of compound 2.90 (0.86 g, 2.6 mmol, 1.0 equiv.) was reacted in 33 % HBr in acetic acid using the general procedure for the synthesis of glycosyl bromide as described in Section 5.2.1.2. The crude product was recrystallized in EtOH and was not further purified. 4-Nitrophenyl 2,4,6-tri-O-aCetyl 3-deoxy-(3-D-ribo-hexopyranoside (2.92) Compound 2.92 (0.33 g, 0.81 mmol, 31 % over 2 steps) was synthesized from a crude sample of 2.91 according to the general Koenigs-Knorr reaction, and was isolated as a white solid: ' H N M R (400 MHz, CDC13) 5 8.19 (2 h, d, J A R 2 , A R3 = J Ars ,Ar6 9.3 Hz, Ar3, Ar5), 7.07 (2 H, d, J A r 2 , A r 3 = J A ^ A * 9.3 Hz, Ar2, Ar6), 5.19 (1 H, d, J,. 2 7.0 Hz, HI), 5.07 (1 H, ddd, J 2,3al2.2, J i , 2 7.0, J 2 , 3 B 5.2 Hz, H2), 4.92-4.86 (1 H, m, H4), 4.20-4.18 (2 H, m, H6 a, H6 b), 3.94-3.90 (1 H, m, H5), 2.70-2.64 (1 H, m, H3 a), 2.06 (6 H, 1 s, 2 Ac), 2.02 (3 Chapter 5 - Materials and Methods 214 H, 1 s, 1 Ac), 1.79-1.71 (1 H, m, H3 b); l 3 C N M R (100 MHz, CDCb) 5 170.51, 169.48, 169.36, 161.35, 142.92, 125.73, 116.46, 98.84 (Cl) , 75.27, 67.74, 65.29, 62.48 (C6), 32.0 (C3), 20.87 (2 CH 3 ) , 20.69; ESI-MS (high res) m/z: calc for [Cij iH 2iNOioNa] +: 434.1063, Found: 434.1055; Anal. calc. For C I 8 H 2 , N 0 1 o »1 . 5 H 2 0 C, 49.32; H, 5.48; N , 3.20; Found: C, 49.41; H, 5.25; N , 3.90. Zemplen deprotection of 2.92 (0.33 g, 0.81 mmol, 1.0 equiv.) afforded 2.93 as a white solid (0.23 g, 0.81 mmol, 100 %): ' H N M R (400 MHz, D 2 0) 5 8.07 (2 H, d, J A r 2 .Ar3 = JAr5,Ar6 9.3 Hz, Ar3, Ar5), 7.06 (2 H, d, W A * = JATS^ 9.3 Hz, Ar2, Ar6), 5.02 (1 H, d, Ji,2 7.8 Hz, HI), 3.76 (1 H, dd, J 6 a , 6 b 12.2, J 5 , 6 a 1.6 Hz, H6 a), 3.70-3.64 (1 H, m, H2), 3.61-3.48 (3 H, m, H4, H5, H6 b), 2.37-2.32 (1 H, m, H3 a), 1.58-1.49 (1 H, m, H3 b); , 3 C N M R (100 MHz, D 2 0) 5 161.66 (C), 142.30 (C), 125.96 (CH), 116.30 (CH), 101.27 C l ) , 79.85, 67.20, 63.86, 60.47 (C6), 37.76 (C6); ESI-MS (high res) m/z: calc. For [Ci 2 Hi5N0 7 Na] + 308.0746, Found: 308.0745; Anal. calc. For C n H i s N C V l . S H z O C, 46.15; H, 5.77; N , 4.49; Found: C, 46.40; H, 5.74; N , 4.28. 4-Nitrophenyl 3-deoxy-6-phospho-fi-D-ribo-hexopyranoside disodium salt (4-Nitrophenyl 3-deoxy-6-phospho-/3-D-glucopyranoside disodium salt) (4NP3deoxypXi6P or 2.94). Compound 2.93 (0.15 g, 0.53 mmol, 1.0 equiv.) was phosphorylated chemically using phosphorus oxychloride as described in Section 5..2.1.10. The compound was purified via HPLC (C18 reverse phase). The pure fractions were collected and lyophilized to 4-Nitrophenyl 3-deoxy-pZ-D-ribo-hexopyranoside (2.93). Chapter 5 - Materials and Methods 215 yield a pale yellow solid (15 mg, 0.037 mmol, 7.0 %): ' H N M R (400 MHz, D 2 0) 8 8.21 (2 H, J A r 2 , A r 3 = J A r 5 , A r 6 9.2 Hz, Ar3, Ar5), 7.12 (2 H, d, J A r 2 ,A r 3 = J A r 5 .A r 6 9.2 Hz, Ar2, Ar6), 5.07 (1 H, d, J,, 2 8.0 Hz, HI), 3.95-3.70 (4 H, m, H2, H4, H6», H 6 b ) , 3.95-3.55 (1 H, m, H5), 2.37-2.32 (1 H, m, H3 e), 1.58-1.50 (1 H, m, H3 a); , 3 C N M R (100 MHz, D 2 0) 8 161.80 (C), 142.54 (C), 126.19 (2 CH), 116.46 (2 CH), 101.55 (Cl) , 79.57, 67.55, 63.29, 62.38, 37.11 ( C 3 ) ; 3 , P N M R ( 1 6 2 M H z , D 2 0) 8 5.21 (1 P, t, J P , 6 H 6.5 Hz, 6P); ESI-MS (high res) m/z: calc. for [ C 1 2 H i S N N a 2 O 1 0 P ] + 410.0299, Found: 410.0228. 4-Nitrophenyl 2-deoxy-6-phospho-f5-D-arabino-hexopyranoside disodium salt (4-Nitrophenyl 2-deoxy-6-phospho-fi-D-gluCopyranoside disodium salt) (4NP2deoxypG6P or 2.96). 2.96 (29 mg, 0.070 mmol, 40 %) was synthesized by the BglK-catalyzed phosphorylation of 2.95 (50 mg, 0.18 mmol, 1.0 equiv.) as described in Section 5.2.1.7: *H N M R (400 MHz, D 2 0 ) 8 8.08 (2 H, d, J A r 2 . A r 3 = W r 6 9.3 Hz, Ar3, Ar5), 7.05 (2 H, d, J A r 2 , A r 3 = J A l 5 , A r 6 9.3 Hz, Ar2, Ar6), 5.04 (1 H, dd, J 9.6, 1.8 Hz, HI), 3.95-3.91 (I H, m, H6 a), 3.84-3.79 (1 H, m, H6 b), 3.73-3.67 (1 H, m, H3), 3.50-3.46 (2 H, m, H4, H5), 2.32 (1 H, ddd, J 2 a > 2 e 12.3, J 4.8, 1.8 Hz, H2 e), 1.76-1.68 (1 H, m, H2 a); 1 3 C N M R (100 MHz, D 2 0) 8 161.38 (C), 142.21 (C), 125.97 (2 CH), 116.19 (2 CH), 96.81 (Cl) , 75.82 (d, J 5,p 7.0 Hz, C5), 69.78 (C4), 69.43 (C3), 62.27 (d, J«s,P 4.2 Hz, C6), 37.75 (C2); 3 I P N M R (121 MHz, D 2 0) 8 5.85 (1 P, t, J H 6 , p 6.0 Hz); ESI-MS (high res) m/z: calc. for [C i 2 Hi 5 NOi 0 PNa 2 ] + 410.0229; Found: 410.0222; Anal . calc. For Ci 2 H, 4 NOioPNa 2 1.5H 2 0: C, 33.03; H, 3.90; N , 3.21; Found: C, 33.20; H, 3.98; N , 3.08. Chapter 5 - Materials and Methods 216 Methyl 6-diphenylphospho-a-D-glucopyranoside (2.97). OPO(OPh)2 OH | \ H 3 Commercially available methyl a-D-glucopyranoside (5.0 g, 26 mmol, 1.0 equiv.) was chemically phosphorylated using diphenyl chlorophosphate according to the general procedure described in Section 5.2.1.9. The product was purified by flash chromatography (7:2:1 EtOAc:MeOH:H 20) to yield a white solid (8.9 g, 21 mmol, 80 %): ' H N M R (300 MHz, Acetone) 5 7.44-7.21 (10 H, m, Ar), 4.64 (1 H, d, J,, 2 3.6 Hz, HI), 4.61-4.56 (1 H, m, H6 a), 4.48-4.40 (1 H, m, H6 b), 3.94 (3 H, broad singlet, 3 OH), 3.79-3.74 (1 H, m, H5), 3.65 (1 H, t, J 2 , 3 - J3.4 9.2 Hz, H3), 3.39-3.31 (5 H, m, H2, H4, OMe); , 3 C N M R (75 MHz, Acetone) 5 151.85 (C), 151.76 (C), 130.78 (4 CH) , 126.25 (2 CH) , 121.09 (2 CH) , 121.03 (2 CH) , 101.05 (Cl ) , 75.07, 73.35, 71.53 (d, J 5 , P 7.0 Hz, C5), 71.08, 69.47 (d, J 6 , P 5.9 Hz, C6), 55.52 (CH 3 ) ; 3 , P N M R (121 MHz, Acetone) 5 -10.48 (1 P, t, Jp,6H 7.1 Hz, 6P); ESI-MS (high res) m/z: calc. For [C,aH2 3Oc)PNa]+ 449.0977, Found: 449.0979. Methyl 6-phospho-a-D-glucopyranoside disodium salt (MeaG6P or 2.98). 5 g of 2.98 (12 mmol, 1.0 equiv.) was deprotected by catalytic hydrogenation using the general procedure described in Section 5.2.1.9. The reaction yielded the pure product (3.5 g, 11 mmol, 92%): ' H N M R (400 MHz, D 2 0 ) 8 4.52 (1 H, d, J,, 2 3.7 Hz, HI), 3.81-3.68 (2 H, m, H6 a, H6 b), 3.44-3.24 (4 H, m, H2, H3, H4, H5), 3.13 (3 H, s, CH 3 ) ; , 3 C N M R (100 MHz, D 2 0 ) 5 99.30 (Cl) , 72.61, 71.17, 70.84 (d, i5y 7.4 Hz, C5), 68.79, 62.83 (d, J 6 , P 4.4 Hz, C6), 55.02; 3 , P N M R (121 MHz, D 2 0 ) 5 3.90 (1 P, t, J P , 6 H 4.6 Hz, 6P); ESI-MS (high res) m/z: calc. for [CvHuOoPNa^ 319.0171, Found: 319.0170. Anal. calc. For C 7 Hi30 9 PNa 2 •0 .5H 2 O C, 25 69; H, 4.28; Found: C, 25.59; H, 4.58. o. Chapter 5 - Materials and Methods 217 4-Nitrophenyl 2-deoxy-2-jluoro-6-phospho-j5-D-glucopyranoside disodium salt (4NP2F/3G6P or 2.100). 2.100 (18 mg, 0.043 mmol, 52 %) was synthesized by the BglK-catalyzed phosphorylation of 2.99 (25 mg, 0.083 mmol, 1.0 equiv.) as described in Section 5.2.1.7: ' H N M R (400 MHz, D 2 0 ) 8 8.14 (2 H, d, J A r 2 ,Ar3 = JAr5,Ar6 9.3 Hz, Ar3, Ar5), 7.14 (2 H, d, J A r2,Ar3 = JAr5.Ar6 9.3 Hz, Ar2, Ar6), 5.45 (1 H, dd, J,, 2 7.7, J, , F 2.8 Hz, HI), 4.40 (1 H, ddd, J 2 , F 51.3, J 2 , 3 9.0, J,, 2 7.7 Hz, H2), 3.98 -3.92 (1 H, m, H6 a), 3.89-3.77 (2 H, m, H3, H6 b), 3.69-3.61 (2 H, m, H4, H5); L 3 C N M R (100 MHz, D 2 0 ) 8 161.11 (C), 142.66 (C), 126.02 (2 CH), 116.46 (2 CH), 97.00 (d, J, , F 24.4 Hz, C l ) , 91.58 (d, J 2 , F 183.6 Hz, C2), 75.83 (d, J 5 , P 7.2 Hz, C5), 73.20 (d, J 3 , F 18.2 Hz, C3), 67.93 (d, J 4 . F 8.7 Hz, C4), 61.88 (d, J6,P 4.1 Hz, C6); 3 I P N M R (121 MHz, D 2 0 ) 8 5.84 (1 P, t, W 6.3 Hz); , 9 F N M R (282 MHz, D 2 0 ) 8 -123.48 (1 F, dd, J 50.8, 14.1 Hz, F2); ESI-MS (high res) m/z: calc. for [C ] 2 Hi4NO 1 0 FPNa 2 ] + 428.0135; Found: 428.0134; , Anal. calc. For C 1 2 H 1 3 N N a 2 0 , , P F 4 H 2 0 : C, 28.87; H, 4.24; N , 2.81; Found: C, 28.87; H, 4.09; N , 2.98. 4-Nitrophenyl 6-phospho-/3-D-thioglucopyranoside disodium salt (S-4NP/5G6P or 2.102). Compound 2.1.01 (30 mg, 0.095 mmol, 1.0 equiv.) was phosphorylated with BglK as described in Section 5.2.1.7. After purification via HPLC (C18 reverse phase), 2.102 was isolated as a white solid (12 mg, 0.028 mmol, 30 %): *H N M R (300 MHz, D 2 0) 8 8.03 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 8.9 Hz, Ar3, Ar5), 7.48 (2 H, d, J A r2,Ar3 = J A r5,Ar6 8.9 Hz, Ar2, Ar6), 4.90 (1 H, d, J,, 2 9.6 Hz, Hi ) , 3.94-3.79 (2 H, m, H6 a, H6 b), 3.59-3.32 (4 H, m, H2-5); L 3 C N M R (100 MHz, D 2 0 ) 8 144.57 (C), 142.11 (C), 127.50 (2CH), 122.81 (2CH), 84.59 (Cl) , 78.26, 75.38, 70.61, 67.22, 61.16 (C6); 3 I P N M R (162 MHz, D 2 0) 8 3.61 (t, Chapter 5 - Materials and Methods 218 V P 5.9 Hz); ESI -MS (high res) m/z: calc. for [Ci 2Hi 5NOioSP]~ 396.0154 Found: 396.0150 Anal. calc. For C i 2 H, 4 NNa 2 OioSP4H 2 0: C, 28.08; H, 4.32; N , 2.73; Found: C, 27.85; H, 4.10; N , 2.80. 4-Methylumbelliferyl 6-phospho-fi-D-glucopyranoside disodium salt (4MUfiG6P or 2.108). The commercially available 4-methylumbelliferyl P-D-glucopyranoside (2.107) (100 mg, 0.30 mmol, 1.0 equiv.) was phosphorylated using BglK as described in Section 5.2.1.7. The product was purified via reverse phase HPLC to yield while crystals (36 mg, 0.077 mmol, 26 %): ' H N M R (300 MHz, D 2 0 ) 8 7.58 (1 H, d, J 8.8 Hz), 6.96 (1 H, dd, J 8.8, J 2.4 Hz), 6.95 (1 H, d, J 2.4 Hz), 6.09 (1 H, s), 5.11-5.08 (1 H, m, HI), 3.98-3.84 (2 H, m, H6 a, H6 b), 3.60-3.50 (4 H, m, H2, H3, H4, H5), 2.28 (3 H, s, CH 3 ) ; 1 3 C N M R (75 MHz, D 2 0) 8 164.68, 159.56, 156.35, 153.99, 126.86, 115.43, 113.87, 111.38, 103.72,99.86 (Cl) , 75.89 (d, J5,P 9.2 Hz, C5), 74.88, 73.10, 68.59, 62.40 (d, J6.p 5.1 Hz, C6), 18.09 (CH 3); 3 I P N M R (121 MHz, D 2 0 ) 8 5.00 (1 P, t, J6H,P 5.4 Hz); ESI-MS (high res) m/z: calc. For [ C , 6 H i 7 O n N a 3 P ] + 485.0202, Found: 485.0203. Anal. calc. For C , 6 H i 7 O i i P N a 2 • 4 H 2 0 C, 35.97; H, 4.72; Found: C, 36.51; H, 4.64. Phenyl 2,3,4,6-tetra-O-acetyl-a-D-glucopyranoside (3.8) 3.8 was synthesized from 2.25 (1.0 g, 2.6 mmol, 1.0 equiv.) according to the general procedure described in Section 5.2.1.6. The product was purified by flash chromatography (3:1 to 1:1 petroleum ether:EtOAc) to yield a white crystalline solid •OAc Chapter 5 - Materials and Methods 219 (0.59 g, 1.4 mmol, 54 %): l H N M R (300 MHz, CDC13) § 7.30 (2 H, m, Ar), 7.05 (3 H, m, Ar), 5.69 (2 H, m, HI , H3), 5.14 (1 H, t, J 3 > 4 = J 4 , 5 10.2 Hz, H4), 5.03 (1 H, dd, J 2 , 3 10.2, J 1 < 2 3.5 Hz, H2), 4.19 (1 H, dd, J ^ b 12.2, J5.<* 4.5 Hz, H6 a), 4.08 (1 H, ddd, J 4 , 5 10.2, J 5 , 6 a 4.5, J 5 , 6 b 2.1 Hz, H5), 3.99 (1H, dd, J 6 a , 6 b 12.2, J 5 , 6 b 2.1 Hz, H6 b), 2.04 (3 H, s, Ac), 2.03 (3 H, s, Ac), 2.02 (3 H, s, Ac), 2.01 (3 H, s, Ac); ESI-MS (low res) m/z cai: 424; Found: 447.0 [M + Na] +. Phenyl a-D-glucopyranoside (3.9). Zemplen deprotection of 3.8 (0.59 g, 1.4 mmol, 1.0 equiv.) yielded the desired white crystalline product (0.35 g, 1.4 mmol, 100%): ' H N M R (300 MHz, D 2 0 ) 5 7.27 (2 H, t, J 7.3 Hz, Ar), 7.15 (2 H, dd, J 1.1, 8.8 Hz Ar), 6.91 (I H, t, J 7.3 Hz, Ar), 5.47 (1 H, d, J,, 2 3.7 Hz, HI), 3.77 (1 H, t, J 2 , 3 = J 3 , 4 9.2 Hz, H3), 3.64-3.56 (3 H, m, H5, H6 a, H6 b), 3.48 (1 H, dd, J 2 , 3 9.2, J,, 2 3.7 Hz, H2), 3.34 (1H, t, J 3 , 4 = J 4 , 5 9.2 Hz, H4); ESI-MS (low res) m/z calc: 256; Found: 279.0 [M + Na] +. Phenyl 6-phospho-a-D-glucopyranoside disodium salt (PaG6P or 3.10). Compound 3.9 (0.20 g, 0.78 mmol, 1.0 equiv.) was phosphorylated chemically according to the procedure described in Section 5.2.1.10. The product was purified by HPLC (C18 reverse phase) and the pure fractions were lyophilized to yield a white solid (39 mg, 0.1 mmol, 13 %): ' H N M R (300 MHz, D 2 0 ) 5 7.27-7.24 (2 H, m, Ar), 7.06-7.00 (3 H, m, Chapter 5 - Materials and Methods 220 Ar), 5.50 (1 H, d, J,, 2 3.7 Hz, HI), 3.95-3.87 (1 H, m), 3.80 (1 H, t, J 9.3 Hz), 3.70-3.56 (4 H, m); l 3 C NMR (100 MHz, D 2 0) 5 155.83 (C), 129.80 (2 CH), 123.02 (C), 117.05 (2 CH), 97.04 (Cl) , 72.31, 72.11 (1 H, d, J 5. P 7.0 Hz, C5), 71.13, 68.34, 61.90 (1 H, d, J 6. P 4.2 Hz, C6); 3 , P NMR (162 MHz, D 2 0) 6 5.49 (1 P, t, J P , 6 H 6.5 Hz, 6P); ESI-MS (high res) m/z: calc. for [C , 2 Hi6Na 2 0 9 P] + 381.0327, Found: 381.0325. 2,3,4,6-Tetra-O-acetyl-a-D-glucopyranosylfluoride (3.11). Compound 2.25 (5.9 g, 15 mmol, 1.0 equiv.) was added to 25 g of HF/pyridine. The reaction mixture was allowed to stir at room temperature under positive N 2 (g,. The reaction was complete after 2 hours, and ice cold sat. NaHCCb was added to the reaction mixture until bubbling subsided. The product was extracted with C H C I 3 (3 x). The organic fractions were pooled and washed successively with sat. NaHCO.i (1 x), water (1 x), and sat. NaCl (1 x), dried over MgSC>4, filtered and concentrated under reduced pressure. The product was recrystallized from EtOH as a white crystalline solid (4.1 g, 12 mmol, 78 %):'H NMR (300 MHz, CDC13) 5 5.73 (1 H, dd, J,. 2 2.8, J, , F 52.9 Hz, HI), 5.48 (1 H, t, J2,3 = J3.4 9.9 Hz, H3), 5.14 (1 H, t, J 3 , 4 = J4.5 9.9 Hz, H4), 4.94 (1 H, ddd, J 2 , F 24.2, J 2 , 3 9.9, J,, 2 2.8 Hz, H2), 4.30-4.11 (3 H, m, H5, H6 a, H6 b), 2.09, 2.08, 2.02, 2.01 (12 H, 4 s, 4 Ac); l 9 F NMR (188 MHz, CDCI3) 8 -73.72 (dd, J , , F 1 52.6, J 2 , F 1 23.5 Hz, F l ) ; ESI-MS (low res) m/z calc: 350; Found: 373.0 [M + Naf . Anal. calc. For C4H19FO9 C, 48.00; H, 5.47; Found: C, 48.07; H, 5.49. .OAc Chapter 5 - Materials and Methods 221 3,4-Dinitrophenyl 2,3,4,6-tetra-O-acetyl-a-D-glucopyranoside (3.12). 3,4-Dinitrophenol was coevaporated in dry toluene several times and dried under high vacuum overnight. The next day, the starting material 3 . 1 1 (0.50 g, 1.4 mmol, 1.0 equiv.) was dissolved in 1.2 mL of boron trifluoride-diethyl etherate (9.4 mmol, 6.5 equiv.) and 2 mL of acetonitrile, and 3,4-dinitrophenol (0.13 g, 0.71 mmol, 0.5 equiv.) was added. The mixture was stirred at room temperature for 4 hours under positive N 2 (g). Upon completion, 20 mL of sat. NaHCC>3 was added. The organic layer was extracted with CHC1 3 (1 x ) , washed with water (1 x ) , sat. NaCl (1 x ) , dried over MgSCM, and concentrated under reduced pressure. The product was purified by flash chromatography to yield a pale yellowish solid (0.74 g, 0.50 mmol, 68 %): ' H N M R (300 MHz, CDCI3) 5 8.03 (1 H, d, h\6- 9-1 Hz, H5'), 7.53 (1 H, d, h\e 2.6 Hz, H2'), 7.38 (1 H, dd, J 5 . , 6 . 9.1, h',6- 2.6 Hz, H6'), 5.85 (1 H, d, J 1 > 2 3.6 Hz, HI), 5.64 (1 H, dd, J 2 , 3 10.2, J 3 > 4 9.5 Hz, H3), 5.14 (1 H, dd, J4.5 10.2, J3.4 9.5 Hz, H4), 5.07 (1 H, dd, J 2 , 3 10.2, J,, 2 3.6 Hz, H2), 4.21 (1 H, dd, J 6 a , 6 b 12.5, J 5 , 6 a 5.1 Hz, H6 a), 4.06-3.96 (2 H, m, H5, H6 b), 2.07 (3 H, s, OAc), 2.05 (2 H, s, OAc), 2.03 (6 H, s, 2 x OAc); ESI-MS (low res) m/z calc: 514; Found: 537.2 [M+'Naf. Chapter 5 - Materials and Methods 222 3,4-Dinitrophenyl a-D-glucopyranoside (3.13). Compound 3.12 (0.74 g, 0.50 mmol, 1.0 equiv.) was deprotected according to the procedure described in Section 5.2.1.4. Upon completion, the reaction mixture was concentrated under reduced pressure and was not further purified but used directly for the subsequent reaction. 3,4-Dinitrophenyl 6-phospho-a-D-glucopyranoside disodium salt (34DNPaG6P or 3.14). A crude sample of compound 3.13 (~0.18 g, 0.52 mmol, 1.0 equiv.) was phosphorylated according to the procedure described in Section 5.2.1.4. HPLC purification (reverse phase C18) yielded a pale yellow solid as the product (18 mg, 0.037 mmol, 7.2 %): 'H N M R (400 MHz, D 2 0 ) 5 8.03 (1 H, d, JA r 5,A,-6 9.2 Hz, Ar5), 7.55 (1 H, d, J A r 2 , A r 6 2.4 Hz, Ar2), 7.39 (1 H, dd, JA r5,Ar6 2.4 Hz, Ar6), 5.68 (1 H, d, J,, 2 3.6 Hz, HI), 3.91-3.85 (1 H, m), 3.77 (1 H, t, J 9.3 Hz), 3.69-3.53 (4 H, m); 1 3 C N M R (100 MHz, D 2 0 ) 8 160.41 (C), 144.67 (C), 135.39 (C), 128.06 (CH), 119.92 (CH), 113.36 (CH), 97.54 (Cl) , 72.91 (d, J 5 . P 6.5 Hz, C5), 72.24, 70.95, 68.17, 62.09 (1C, d, J 6 , P 3.8 Hz, C6); 3 , P N M R (162 MHz, D 2 0 ) 8 5.55 (1 P, t, Jp,6H 6.5 Hz, 6P); ESI-MS (high res) m/z: calc. for [ C i 2 H 1 4 N 2 N a 2 0 , 3 P ] + : 471.0029, Found: 471.0026. Chapter 5 - Materials and Methods 223 4-Nitrophenyl 2,3,4,6-tetra- O-acetyl-1 -[2H]- a-D-glucopyranoside (3.15) 3.15 (1.5 g, 3.2 mmol, 66 %) was prepared from 2.67 (1.9 g, 4.9 mmol, 1.0 equiv.) as described in Section 5.2.1.6. The product was purified via flash chromatography (3:1 petroleum ethenEtOAc) to yield 3.15 as a white .crystalline solid: ' H N M R (300 MHz, CDC13) 5 8.21 (2 H, d, J A r 2 > A r 3 = J A r 5 . A r 6 9.3 Hz, Ar3, Ar5), 7.19 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 9.3 Hz, Ar2, Ar6), 5.66 (1 H, d, J2.3 = J3.4 9.7 Hz, H3), 5.15 (1 H, t, J 3 , 4 = h,s 9.8 Hz, H4), 5.07 (1 H, d, J 2 , 3 9.7 Hz, H2), 4.22 (1 H, dd, W b 12.5, J 5.&4.7 Hz, H6 a), 4.05-3.98 (2 H, m, H5, H6 b), 2.05 (3 H, s, Ac), 2.04 (3 H, s, Ac), 2.02 (3 H, s, Ac), 2.01 (3 H, s, Ac); 1 3 C N M R (100 MHz, C D C I 3 ) 5 170.38 (C=0), 170.09 (C=0), 170.07 (C=0), 169.45 ( O O ) , 160.56 (C), 143.16 (C), 125.87 (2 CH), 116.50 (2 CH), 94.09 (Cl) , 69.58, 68.66, 67.96, 61.34 (C6), 20.67, 20.63, 20.59, 20.55; ESI-MS (High res.) m/z calc. for [C 2 oH 2 2 DNO, 2 Na] + : 493.1181; Found: 493.1192. 4-Nitrophenyl 2,3,4,6-tetra-0-acetyl-2-[2HJ-a-D-glucopyranoside (3.16) 3.16 (1.6 g, 3.5 mmol, 69 %) was prepared from 2.68 (2.0 g, 5.1 mmol, 1.0 equiv.) as described in Section 5.2.1.6. The product was purified via flash chromatography (3:1 petroleum ethenEtOAc) to yield 3.16 as a white crystalline solid: ' H N M R (300 MHz, CDCI3) 5 8.21 (2 H, d, J A r 2 , A r 3 = J A r 5 .Ar6 9.3 Hz, Ar3, Ar5), 7.19 (2 H, d, J A r 2 , A r 3 = J A r 5 .Ar6 9.3 Hz, Ar2, Ar6), 5.82 (1 H, s, HI), 5.67 (1 H, d, J3.4 9.8 Hz, H3), 5.15(1 H, t, J 3 , 4 = J4.5 9.8 Hz, H4), 4.22 (I H, dd, J ^ b 12.5, J5,<sa 4.7 Hz, H6 a), 4.05-3.98 (2 H, m, H5, H6 b), -OAc Chapter 5 - Materials and Methods 224 2.05, 2.04, 2.02, 2.01 (12 H, 4 s, 4 Ac); l 3 C N M R (100 MHz, CDC13) 5 170.38 ( O O ) , 170.09 ( O O ) , 170.07 ( O O ) , 169.45 ( O O ) , 160.56 (C), 143.16 (C), 125.87 (2 CH), 116.50 (2 CH), 94.09 (Cl) , 69.58, 68.66, 67.96, 61.34 (C6), 20.67, 20.63, 20.59, 20.55; ESI-MS (high res) m/z calc. for [C 2oH22DNO | 2 Na] + : 493.1181; Found: 493.1192. 3.17 (1.4 g, 3.0 mmol, 59 %) was prepared from 2.67 (2.0 g, 5.1 mmol, 1.0 equiv.) as described in Section 5.2.1.6. The product was purified via flash chromatography (3:1 petroleum ether:EtOAc) to yield the pure product as a white crystalline solid: ' H N M R (300 MHz, CDC13) 8 8.21 (2 H, d, J A r 2 j A r 3 = J A r5,Ar6 9.3 Hz, A r3 , Ar5), 7.19 (2 H, d, J A r2 , A r3 = J A r 5 , A r 6 9.3 Hz, Ar2, Ar6), 5.82 (1 H, d, J,,2 3.6 Hz, HI), 5.14 (1 H, t, J 4 , 5 9.8 Hz, H4), 5.07 (1 H, d, J2,3 10.2, J,, 2 3.6 Hz, H2), 4.22 (1 H, dd, J 6 a ,6 b 12.5, J 5,6a 4.7 Hz, H6 a ) , 4.05-3.98 (2 H, m, H5, H6 b ) , 2.05, 2.04, 2.02, 2.01 (12 H, 4 s, 4 Ac); , 3 C N M R (100 MHz, CDC13) 8 170.38 ( O O ) , 170.09 ( O O ) , 170.07 ( O O ) , 169.45 ( O O ) , 160.56 (C), 143.16 (C), 125.87 (2 CH), 116.50 (2 CH), 94.09 (Cl) , 69.58, 68.66, 67.96, 61.34 (C6), 20.67, 20.63, 20.59, 20.55; ESI-MS (high res) m/z calc. for [C 2oH22DNO ! 2 Na] + : 493.1181; Found: 493.1192. 4-Nitrophenyl 1-[2H]-a-D-glucopyranoside (3.18). Zemplen deprotection of 3.15 (1.5 g, 3.2 mmol, 1.0 equiv.) yielded 0.96 g of 3.18 (3.2 mmol, 100 %) as a white crystalline solid: ' H N M R (400 MHz, CDCI3) 8 8.09 (2 H, d, J A r2,Ar3 = W r 6 9.3 Hz, Ar3, Ar5) , 7.13 (2 H, d, J A r 2 i A r 3 = - W r f 9.3 Hz, Ar2, Ar6), 3.77 4-Nitrophenyl 2,3,4,6-tetra-0-acetyl-3-[2H]- a-D-glucopyranoside (3.17) .OAc Chapter 5 - Materials and Methods 225 (1 H, d, J 3 , 4 9.1 Hz, H3), 3.59-3.48 (3 H, m, H5, H6 a, H6 b), 3.45 (1 H, s, H2), 3.36 (1 H, t, J3,4 = J4,s 9.1 Hz, H4); 1 3 C N M R (100 MHz, CDC13) 5 161.24, 142.21, 125.89 (2 CH), 116.59 (2 CH), 96.43 (Cl) , 72.71 (2 lines), 69.02, 60.04 (C6); ESI-MS (High res.) m/z calc. for [Ci2H,4DNO><Na]+: 325.0758; Found: 325.0761; Anal. calc. For Ci2H , 4DM08V4 H 2 0 : C, 46.30; H, 5.47; N , 4.50; Found: C, 46.88; H, 5.09; N , 4.37 . 4-Nitrophenyl 2-[2H]-a-D-glucopyranoside (3.19). Zemplen deprotection of 3.16 (1.6 g , 3.4 mmol, 1.0 equiv.) yield 0.98 g of 3.19 (3.2 mmol, 95 %) as a white crystalline solid: ' H N M R (400 MHz, CDC13) 5 8.09 (2 H, d, J A r 2 , A r 3 = JAr5.Ar6 9.3 Hz, Ar3, Ar5), 7.13 (2 H, d, J A r 2 . A r 3 = J A r 5 . A r 6 9.3 Hz, Ar2, Ar6), 5.64 (1 H, s, HI), 3.77(1 H, d, J3,49.1 Hz, H3), 3.59-3.48 (3 H, m, H5, H6 a, H6 b), 3.36(1 H, t, J 3 > 4 = J4.5 9.1 Hz, H4); 1 3 C N M R (100 MHz, CDC13) 5 161.24, 142.21, 125.89 (2 CH), 116.59 (2 CH), 96.43 (Cl) , 72.71 (2 lines), 69.02, 60.04 (C6); ESI-MS (High res.) m/z calc. for [C | 2 H, 4 DNO x Na] + : 325.0758; Found: 325.0760; Anal. calc. For C i 2 H i 4 D N ( V / 2 H 2 0 : C, 46.30; H, 5.47; N , 4.50; Found: C, 46.05; H, 5.55; N , 4.21. 4-Nitrophenyl 3-[2H]-a-D-glucopyranoside (3.20). Zemplen deprotection of 3.17 (1.4 g , 3.0 mmol, 1.0 equiv.) yielded 0.88 g of 3.20 (2.9 mmol, 97 %) as a white crystalline solid: ' l l N M R (400 MHz, D 2 0 ) 5 8.05 (2 H , d, J A r 2 , A r 3 = J A r5.Ar6 9.3 Hz, Ar3, Ar5), 7.05 (2 H, d, J A r 2 , A r 3 = J A r 5 , A r 6 9.3 Hz, Ar2, Ar6), 5.67 (1 H, d, J,, 2 7.9 Hz, HI), 3.77 (1 H, dd, J 6 a , 6 b 12.4, J 5 , 6 a 1.9 Hz, H6 a), 3.59 (1 H, dd, J 6 a ,6 b .OH Chapter 5 - Materials and Methods 226 12.4, J 5 , 6 b 5.7 Hz, H6 b), 3.52 (1 H, in, H5), 3.45 (1 H, d, J,. 2 7.9 Hz, H2), 3.34 (1 H, d, J 4 . 5 9.8 Hz, H4); , 3 C N M R (100 MHz, CDC13) 5 161.33, 142.09, 125.65 (2 CH), 116.50 (2 CH), 99.05 (Cl) , 75.85 (C5), 72.,27 (C2), 68.83 (C4), 60.05 (C6); ESI-MS (High res.) m/z calc. for [C^HuDNOxNaf: 325.0758; Found: 325.0746. 4-Nitrophenyl 1 -['HJ-6-phospho-a-D-glucoside disodium salt (1l2H]4NPaG6P or 3.21). 2.18 (0.20 g, 0.66 mmol, 1.0 equiv.) was phosphorylated chemically according to the procedure outlined in Section 5.2.1.10. The product was purified by HPLC (C18 reverse phase). The pure fractions were pooled and lyophilized to yield a white solid (42 mg, 0.10 mmol, 15 %): ' H N M R (400 MHz, D 2 0) 5 8.12 (2 H, d, J A r 2 , A r 3 = J A r 5,Ar6 9.3 Hz, Ar3, Ar5), 7.15 (2 H, d, J A r 2 , A r 3 = J A r 5,Ar6 9.3 Hz, Ar2, Ar6), 3.92-3.86 (1 H, m, H6 a), 3.81 (1 H, t, J2,3 = J3,4 9.3 Hz, H3), 3.69-3.57 (4 H, m, H2, H4, H5, H6 b); l 3 C N M R (100 MHz, D 2 0 ) 5 161.35 (C), 142.24 (C), 125.95 (2 CH), 116.66 (2 CH), 72.50 (d, h.p 6.7 Hz, C5), 72.21, 70.90, 268.19, 61.91 (d, J 6 ,p 5.8 Hz, C6); 3 I P N M R (121 MHz, D 2 0 ) 5 5.54 (t, JHC.P 6.1 Hz); ESI-MS (high res) m/z: calc. for [C, 2 H] 4 DNOi,PNa 2 ] + 427.0241; Found: 427.0239. 4-Nitrophenyl 2-[2 H]-6-phospho-a-D-glucoside disodium salt (2f2H/4NPaG6P or 3.22). 2.19 (0.33 g, 1.0 mmol, 1.0 equiv.) was phosphorylated chemically according to the procedure outlined in Section 5.2.1.10. The product was purified by HPLC (C18 reverse phase). The pure fractions were pooled and lyophilized to yield a white solid (47 mg, 0.11 mmol, 11 %): ' H N M R (400 MHz, D 2 0) 8 8.13 (2 H, d, J A r 2 , A r 3 = J ^ w 9.3 Hz, Chapter 5 - Materials and Methods 227 Ar3, Ar5), 7.17 (2 H, d, JA r2.Ar3 = 9.3 Hz, Ar2, Ar6), 5.67 (1 H, s, HI), 3.90 (1 H, ddd, J 12.4, 7.7, 2.6 Hz, H6 a), 3.81 (1 H, d, J 3 , 4 8.8 Hz, H3), 3.66-3.58 (3 H, m, H4, H5, H6 b); , 3 C N M R (100 MHz, D 2 0 ) 8 161.34 (C), 142.23 (C), 125.95 (2 CH), 116.64 (2 CH), 96.69 (Cl) , 72.54 (d, J 5 , P 6.9 Hz, C5), 72.13, 68.16, 61.84 (d, J 6 , P 4.6 Hz, C6); 3 I P N M R (162 MHz, D 2 0 ) 8 4.98 (t, J H 6 .p 5.2 Hz); ESI-MS (high res) m/z: calc. for [Ci 2 Hi 4 DNO, ,PNa 2 ] + 427.0241; Found: 427.0243; Anal. calc. For C^HnDNNa^nP-SHaO: C, 30.01; H, 4.38; N , 2.92; Found: C, 29.27; H, 3.96; N , 2.92. 4-Nitrophenyl 3-[2H]-6-phospho-a-D-glucoside disodium salt (3f2HJ4NPaG6P or 3.23). 2.19 (0.28 g, 0.93 mmol, 1.0 equiv.) was phosphorylated chemically according to the procedure outlined in Section 5.2.1.10. The product was purified by HPLC (C18 reverse phase). The pure fractions were pooled and lyophilized to yield a white solid (23 mg, 0.056 mmol, 6.0 %): 'H N M R (400 MHz, D 2 0) 8 8.13 (2 H, d, J A r 2 , A r 3 = JA r5,Ar6 9.3 Hz, Ar3, Ar5), 7.16 (2 H, d, J A r 2 , A r 3 = J A r 5 . A r 6 9.3 Hz, Ar2, Ar6), 5.67 (1 H, d, J,, 2 3.7 Hz, HI), 3.90 (1 h, ddd, J 12.2, 7.6, 2.6 Hz, H6 a), 3.69-3.58 (4 H, m, H2, H4, H5, H6 b); L 3 C N M R (75 MHz, D 2 0) 8 162.85 (C), 143.75 (C), 127.46 (2 CH), 118.17 (2 CH), 98.24 (Cl) , 74.05 (d, J5.p 7.1 Hz, C5), 72.45, 69.61, 63.41 (d, J6,p 5.4 Hz, C6); 3 I P N M R (162 MHz, D 2 0 ) 8 4.88 (t, JH 6,P 6.1 Hz); ESI-MS (high res) m/z: calc. for [C , 2 H, 4 DNOnPNa 2 ] + 427.0241; Found: 427.0239; Anal. calc. For C i 2 H 1 3 D N N a 2 O n P - 5 H 2 0 : C, 27.91; H, 4.84; N , 2.71; Found: C, 26.96; H, 4.51; N , 3.22. Chapter 5 - Materials and Methods 228 Methyl 1 [2HJ-a-D-glucopyranoside (3.24). Compound 2.65 (0.8 g, 4.4 mmol, 1.0 equiv.) was dissolved in 2 mL of anhydrous MeOH, and 0.2 g of Amberlite IR-120 (VC) was added. The reaction mixture was refluxed for 48 hours. Upon completion, the reaction mixture was cooled to room temperature and filtered. The filtrate was concentrated under reduced pressure and recrystallized in EtOH to yield a shiny white crystalline solid (0.63 g, 3.2 mmol, 73 %): ' H N M R (300 MHz, D 2 0) 8 3.71 (1 H, dd, J&. 6 b 12.3, J 5 , 6 a 2.2 Hz, H6 a), 3.62-3.46 (3 H, m, H3, H4, H5), 3.39 (1 H, d, J 2 , 3 9.7 Hz, H2), 3.26-3.20 (4 H, m, H4, CH 3 ) ; l 3 C N M R (75 MHz, D 2 0) 8 98.93 (t, J , , D 25.5 Hz, C l ) , 73.18, 71.61, 71.21, 69.65, 60.67 (C6), 55.07 (CH 3); ; ESI-MS (high res.) m/z: calc for [C 7 H 1 3 D0 6 Na] + : 218.0751, Found: 218.0753. Methyl 1 [2HJ-6-diphenylphospho-a-D-gluCopyranoside (3.25). Compound 2.65 (0.50 g, 2.6 mmol, 1.0 equiv.) was phosphorylated according to the procedure outlined in Section 5.2.1.8. The product was purified by flash chromatography (7:2:1 EtOAc:MeOH:H 20) to yield a clear colorless syrup (0.87 g, 2.2 mmol, 87 %) : ' H N M R (400 MHz, Acetone) 8 7.43-7.40 (4 H, m, Ar), 7.32-7.29 (4 H, m, Ar), 7.26-7.22 (2 H, m, Ar), 4.63-4.58 (1 H, m, H6 a), 4.48-4.42 (1 H, m, H6 b), 3.79-3.75 (1 H, m, H5), 3.67 (1 H, t, J 2 , 3 = J 3 , 4 9.1 Hz, H3), 3.43-3.34 (2 H, m, H2, H4), 3.30 (3 H, s, OMe); l 3 C N M R (100 MHz, Acetone) 8 151.74 (C), 151.67 (C), 130.76 (4 CH), 126.24 (2 CH), 121.04 (2 ,OPO(OPh) 2 Chapter 5 - Materials and Methods 229 CH), 120.99 (2 CH), 75.01, 73.17, 71.42 (d, J 5 , P 6.9 Hz, C5), 70.99, 69.45 (d, J6,p 6.0 Hz, C6), 55.45 (CH 3); 3 1 P N M R (121 MHz, Acetone) 8 -10.67 (1 P, s, 6P); ESI-MS (high res) m/z: calc. for [C i 9 H 2 2 D0 9 PNa] + : 450.1040; Found: 450.1042. Methyl 1 [2H]-6-phospho-a-D-glucopyranoside disodium salt (l[2HjMeaG6P or 3.26). Compound 3.25 (0.87 g, 2.2 mmol, 1.0 equiv.) was dissolved in 100 mL of dry methanol, and subjected to hydrogenation in the presence of a catalytic amount of Pt0 2 . Upon completion, the reaction mixture was filtered through a Celite cake. The filtrate was neutralized with the addition of 3 M NH 4 OH and concentrated under reduced pressure to yield the pure product (0.65 g, 2.0 mmol, 92 %) as a white gummy solid: ' H N M R (400 MHz, D 2 0 ) 8 3.92-3.86 (1 H, m, H6 a), 3.77-3.73 (1 H, m, H6 b), 3.53-3.39 (4 H, m, H2, H3, H4, H5), 3.24 (3 H, s, CH 3 ) ; 1 3 C N M R (100 MHz, D 2 0) 8 98.85 (t, J , , D 24.9 Hz, C l ) , 72.37, 71.04-70.94 (2 C), 68.54, 62.13 (d, J6.p 3.9 Hz, C6), 54.85 (CH 3); 3 , P N M R (162 MHz, D 2 0 ) 8 5.58 (1 P, t, J P , 6 H 6.5 Hz, 6P); ESI-MS (high res) m/z: calc. for [C 7 H 1 3 D0 9 PNa 2 ] + ; 320.0234; Found: 320.0235; Anal. calc. For C 7 H | 2 D 0 9 P N a 2 • 3 H 2 0 : C, 22.65; H, 4.89; Found: C, 22.51; H, 4.50. Section 5.3. Enzyme Kinetics. Section 5.3.1. General Methods. A l l kinetic assays were conducted in 1 cm pathlength matched quartz cuvettes with a Cary 300 UV-vis spectrometer equipped with a circulating water bath, or a Cary 4000 UV-vis spectrometer with a Cary Temperature Controller attached. Al l data fitting was performed with GraFit 4.0 or Cary WinUV, Kinetics Application, Version 3.00 (182). Chapter 5 - Materials and Methods 230 Section 5.3.1.1. Conditions for Measurement of Initial Rates. Conditions for measurement of initial rates - The concentration of enzyme used for each substrate was chosen such that less than 10 % of the total substrate is consumed, ensuring linear rates. The enzyme was preincubated with the assay buffer mixture for 5 minutes and the reaction was initiated by the addition of the appropriate substrate. For the direct UV-vis assay, the initial rate of hydrolysis was followed spectrophotometrically upon addition of the appropriate aryl 6-phospho-D-glucoside at the wavelength of maximal absorbance difference between the released phenol and the respective aryl 6-phospho-D-glucoside. For the G6PDH coupled assay, the reaction was initiated with the addition of substrate and the initial rate of hydrolysis was followed spectrophotometrically at 340 nm, monitoring for the formation of NADPH (8NADPH = 6220 M ~ W ) . Section 5.3.1.2. Conditions for the Substrate Depletion Method. Conditions for the substrate depletion method2 1 4 - For the direct UV-vis assay, the kcJKu analyses were performed by the depletion method using low substrate concentrations and by monitoring the change in absorbance at the wavelength of maximal absorbance difference between the released phenol and the respective aryl 6-phospho-D-glucoside over approximately 30 minutes until the reaction was complete. For the G 6 P D H coupled assay, the substrate depletion method was employed using substrate concentrations far below the Ku and by monitoring for the formation of NADPH (ENADPH = 6220 M'cm" ' , A,™* = 340 nm). The data sets were fit to a first order equation, and the kcJK-M values were obtained by dividing the pseudo-first order rate constant obtained by the enzyme concentration. Section 5.3.1.3. Standard Procedures for Preparing Buffers and Enzyme in D 2 O Solutions. A l l buffers and chemicals were lyophilized twice from 99.9 % D2O. Prior to kinetic assays, BglT was exchanged into deuterated buffer solutions via repeated (three times) dilution and concentration using a centrifugal filter unit (Millipore) with a nominal molecular weight limit (NMWL) of 10 kDa. The pD was determined by measuring the Chapter 5 - Materials and Methods 231 pH using a standard glass pH electrode. A value of 0.4 was added to the pH reading to obtain the pD value, such that pD = pH (measured) + O.4. 2 1 5 Section 5.3.2. Enzyme Kinetics with BglT. Kinetic assays with BglT was initiated with the addition of the substrate after the enzyme had been preincubated in the appropriate assay buffer at 50 °C for 5 minutes. Section 5.3.2.1. Buffer Systems. The following buffer systems were employed for assaying BglT: Buffer A: 50 mM HEPES, 0.1 mM MnCl 2 , 1 uM N A D + , 10 mM 2-mercaptoethanol, and 0.1 %(w/v) BSA at pH 7.5 Buffer B: (All reagents were lyophilized twice from 99.9 % D 2 0 ) 50 mM HEPES, 0.1 mM MnCl 2 , 1 uM N A D + , 10 mM 2-mercaptoethanol (D 6), 0.1 % (w/v) BSA atpD 8.1. Section 5.3.2.2. BglT - Direct UV-vis Assay. Activity of BglT was measured with 4NP(3G6P by monitoring the release of the 4-nitrophenolate anion at 400 nm. The enzyme (final concentration of 2.87 Ug/mL) was preincubated in Buffer A at 50 °C, and the initial linear rate of increase in absorbance at 400 nm measured upon addition of 4NP(3G6P to a final assay volume of 200 p,L. The difference in extinction coefficients, A£, between 4NPPG6P and the 4-nitrophenolate anion at pH 7.5, 50 °C was determined to be 13,791 M"'cm"'. The concentration of 4NP(3G6P was varied and 8-12 data points were collected. The catalytic parameters were determined based on a direct fit of the data to the Michaelis-Menten equation. Section 5.3.2.3. Glucose 6-Phosphate Dehydrogenase Coupled Assay. C6'P was synthesized as described by Thompson and coworkers I 2 3 . Reaction rates were measured using a glucose 6-phosphate dehydrogenase (G6PDH) coupled assay. Al l experiments were carried out at 50 °C in Buffer A , and 2 mM N A D P + and 20 units of G6PDH. The activity of BglT was measured spectrophotometrically by monitoring the formation of the NADPH cofactor at 340 nm. Initial rates were measured Chapter 5 - Materials and Methods 232 upon addition of C6'P to the reaction mixture (final volume = 200 uL), by monitoring the increase in absorbance at 340 nrri. 8-12 Data points were collected under the following conditions: C6'P (final substrate concentration range = 10-1600 p M , final enzyme concentration = 5.63 pg/mL). A molar extinction coefficient of 6,220 M"'cm"' (Ae of NADPH) "was used for the calculation of initial rates of substrate hydrolysis, and the catalytic parameters were determined based on a direct fit of the data to the Michaelis-Menten equation. The concentration of BglT was doubled for two data points and the observed rate was also doubled, ensuring that concentrations of BglT, G6PDH and N A D P + used for the coupled assay were not rate-limiting factors. Furthermore, BglT was assayed with 4NPpG6P in the presence of 2 mM N A D P + , and the enzyme was also assayed with 4NPpG6P in the presence of 20 units of G6PDH. In each case the observed rate was the same as when BglT was assayed alone, indicating that the presence of N A D P + and G6PDH did not affect the activity of BglT. Rates of hydrolysis of 4NP(3C6'P, S-4NPpC6'P, C6'P and S-C6'P were individually assayed using the glucose 6-phosphate dehydrogenase (G6PDH) coupled assay in Buffer A. 8-12 Data points were collected for each substrate over the following ranges: 4NPPC6'P (final concentration range = 1-100 pM, final enzyme concentration = 45 iig/mL), S'-4NPpC6'P (final concentration range = 1-160 p M , final enzyme concentration = 22.5 mg/mL), C6'P (final concentration range — 10-1600 p M , final enzyme concentration = 5.63 pg/mL), and S'-C6/P (final concentration range = 10-250 p M , final enzyme concentration = 5.6 mg/mL). A molar extinction coefficient of 6,220 M'cm" 1 for the oxidation of NADP to NADPH was used in calculations, and the catalytic parameters were determined based on a direct fit of the data to the Michaelis-Menten equation. Several controls were run to ensure that the coupled assay was meaningful. Firstly, the concentration of BglT was doubled for reaction at two different substrate concentrations and the observed rates were also doubled, ensuring that the BglT reaction was truly the rate-limiting process under study. Secondly, BglT Was assayed with 4NPpG6P in the presence of 2 mM NADP, and the enzyme was also assayed with PNPG6P in the presence of 20 units of G6PDH. In each case the observed rate was the Chapter 5 - Materials and Methods 233 same as when BglT was assayed alone, indicating that the presence of NADP and G6PDH did not affect the activity of BglT. Section 5.3.2 .4. Abg Coupled Assay. Rates of hydrolysis of 4NPpC6'P by BglT were also determined by use of the GH1 P-glucosidase, Abg as a coupling enzyme. Al l experiments were carried out at 50 °C in Buffer A and 8.5 Ug/mL Abg. BglT activity was measured spectrophotometrically by monitoring the release of 4-nitrophenolate anion at 400 nm. BglT (final concentration of 90 ug/mL) was preincubated with the above solution at 50 °C, and the initial linear rate of increase in absorbance at 400 nm was measured upon addition of 4NPpC6'P (concentration ranging from 1-350 U.M) to a final assay volume of 200 uL. The difference in extinction coefficients, Ae, between 4-nitrophenyl p-D-glucoside and the 4-nitrophenolate anion at pH 7.5, 50 °C was determined to be 13,791 IVr'cm"1, and the catalytic parameters were determined based on a direct fit of the data to the Michaelis-Menten equation. The concentration of BglT was doubled for two data points and the observed rate was also doubled, ensuring that the concentration of Abg used was not the rate-limiting factor. Furthermore, BglT was assayed with 4NPpG6P in the presence of Abg and the observed rate was the same as when BglT was assayed alone, indicating that Abg did not hydrolyze 6-phospho-p-D-glucosides and that the presence of Abg did not affect the activity of BglT. Section 5.3.2.5. Determination of Kd Value for M n 2 + . 1 mL BglT (1 mg/mL) was first dialyzed against 5 x 3 L 50 mM HEPES at pH 7.5 to remove any bound divalent metals prior to manipulation. Samples of dialyzed BglT (final concentration of 6.5 Ug/mL) and M n 2 + (varied from 20 uM to 1 mM) were preincubated in 50 mM HEPES (pH 7.5), 1 U.M N A D + , 10 mM 2-mercaptoethanol, and 0.1 % (w/v) BSA at 37 °C for 5 minutes, then 4NPpG6P (final concentration > 10 A:M) was added to the reaction mixture to give a final volume of 200 uL and initial rates measured. The reaction rate was plotted against the concentration of M n 2 + to generate ligand binding curves. Chapter 5 - Materials and Methods 234 Section 5.3.2.6. Determination of K& Value for NAD+. 1 mL BglT (1 mg/mL) was first dialyzed against 5 x 3 L 50 mM HEPES at pH 7.5 to remove any bound N A D + prior to manipulation. Samples of dialyzed BglT (final concentration of 2.25 pg/mL) and N A D + (varied from 100 nM to 10 pM) were preincubated in 50 mM HEPES, 0.1 mM MnCl 2 , 10 mM 2-mercaptoethanol, and 0.1 % (w/v) BSA at pH 7.5 for 5 minutes, then 4NP(3G6P (final concentration = 615 pM > 10 KM) was added to the reaction mixture to give a final volume of 200 pL and initial rates measured. The reaction rate was plotted against the concentration of N A D + to generate a ligand binding curve and a Kd value of 480 nM for the binding of dinucleotide cofactor to BglT determined. Section 5.3.2.7. Kinetic and Spectroscopic Investigation of Cofactor Reduction. For the kinetic analysis, a sample of dialyzed BglT (final concentration = 2.25 pg/mL = 47 nM ) and N A D + (final concentration = 10 pM) was preincubated in Buffer A for 5 minutes, then 4NP(3G6P was added to the reaction mixture (final volume = 200 pL) to a final concentration of 615 p M (> 10 KM). In a second reaction, a sample containing dialyzed BglT (final concentration = 2.25 pg/mL = 47 nM), NAD + (f inal concentration = 10 pM), and N a B H 4 (final concentration = 10 mM) was preincubated in Buffer A for 5 minutes, and 4NP(3G6P (final concentration of 615 pM) was added to a final assay volume of 200 pL. The change in A400 was monitored for 18 minutes prior to the addition of 10 pL of 200 pM N A D + to rescue enzyme activity. The change in A400 was then measured for another 20 minutes. BglT activity was assayed in the presence of N A D + and the activity was compared to that obtained in the presence of N A D H (quantitatively reduced from N A D + using sodium borohydride). The enzyme was completely inactive in the presence of N A D H (reduced form) alone. Upon addition of excess N A D + , full activity was rapidly restored, demonstrating that the loss of activity upon reduction is due solely to having the dinucleotide cofactor in the wrong redox state rather than protein denaturation or dramatic changes in pH. The absorbance spectra (from 250 nm to 400 nm) of the following three dialyzed enzyme samples were measured in 1 mL, 1 cm pathlength - matched quartz cuvettes: 10 Chapter 5 - Materials and Methods 235 (lM BglT; 10 uM BglT incubated with 10 uM N A D + ; and 10 uM BglT incubated with 10 uM N A D + and 10 mM sodium borohydride. Measurement of sample pH after reaction confirmed that 10 mM sodium borohydride did not significantly alter the pH of the solution. The absorbance spectra (from 320 nm to 400 nm) of BglT were recorded under three differing conditions: 10 (lM BglT; 10 uM BglT incubated with 10 uM N A D + ; and 10 U.M BglT incubated with 10 uM N A D + and 10 mM sodium borohydride. The peak in absorbance at 340 nm corresponding to NADH appears upon reduction with 10 mM sodium borohydride, and is consistent with the quantitative reduction of N A D + to N A D H . Based on the small Kd value of 480 nM for the N A D + , over 99 % of BglT has N A D + or NADH bound to its active site under these conditions and very little remains free in solution. Section 5.3.2.8. Stereochemical Outcome Determination by Methanolysis. The enzymatic reaction was performed in Buffer A and 5 M MeOH. 4NP|3G6P (20 mg) was incubated at 50 °C in 5 mL of the above solution and the reaction was monitored by TLC (2:1:1 l-butanol:H20:acetic acid). Upon completion, the enzyme was removed using a 10 kDa NMVVL centrifugal filter unit. 100 mg of Chelex 100 resin (BioRad) was added to the filtrate and stirred at room temperature for 30 minutes to remove metal ions. The suspension was then filtered and the solution was lyophilized, then redissolved in D 2 0 for 'H NMR analysis at room temperature and at 50 °C. In a second set of reactions, all reagents and enzyme was exchanged into D 2 0 buffer. The enzymatic reaction was performed under the following conditions: 50 mM HEPES pH 7.5, 1 mM MnCl 2 , 1 uM N A D + , 0.4 mg/mL BglT, and 5 M CD 3 OD. The ' H NMR spectra were compared to those obtained for standard samples of MeocG6P, Me(3G6P, and G6P. Section 5.3.2.9. Solvent Deuterium Isotope Exchange. A l l buffers and chemicals were lyophilized twice from 99.9 % D 2 0 . BglT was exchanged into deuterated buffer solutions via repeated (three times) dilution and concentration using a centrifugal filter unit (Millipore) with a nominal molecular weight limit (NMWL) of 10 kDa. The enzymatic reactions were performed in Buffer B and 0.4 Chapter 5 - Materials and Methods 236 mg/mL BglT. 4NPpG6P (20 mg) was incubated at 50 °C in 5 mL of the above solution and the reaction was monitored by TLC. Upon completion, the enzyme was removed using a 10 kDa N M W L centrifugal filter unit. 100 mg of Chelex 100 resin was added to the filtrate and stirred at room temperature for 30 minutes. The Chelex resin was removed via filtration, and the solution was lyophilized then redissolved in D 2 0 for ' H N M R analysis. In an identical control experiment, the above solution was incubated at 50 °C without enzyme. Two other controls involved 'H NMR analysis of reaction mixtures in which the product glucose 6-phosphate (G6P) had been incubated with the buffer and cofactors in the presence and absence of BglT. The same deuterium exchange and control experiments were performed with 1,5-anhydroglucitol 6-phosphate, 4NP2FpG6P, and 4NPpG6P. Section 5.3.2.10. pH-Dependence. BglT was incubated at a series of pH values ranging from pH 2.0-14.0 for 30 minutes and periodically aliquots were removed for assay at pH 7.5. These studies revealed that BglT was stable in the range pH 4.0-12.0 for 30 minutes. The pH-dependence of kcJKu was then determined by measurement of kcat/KM at a series of pH values using the substrate depletion method. All experiments were carried out at 50 °C in 50 mM NaCl, 1 mM MnCl 2 , 1 p M N A D + , 10 mM mercaptoethanol, 0.1 % (w/v) BSA and 20 mM AcOH/NaOAc (pH 4.0-4.5), or 20 mM MES (pH 5.9-6.5), or 20 mM HEPES (pH 6.5-8.2), or 20 mM CHES (pH 8.4-9.6). The enzyme (final concentration of 5.73 pg/mL) is pre-incubated with the above solution at 50 °C for 5 minutes, and pNPPGlc6P (final concentration of 6.15 p M , at a concentration that was equal to 0.15 KM) was used to initiate the enzymatic reaction. The absorbance at 400 nm was monitored until the substrate was depleted. The data set was fitted to a first order equation, and the kcat/KM values were obtained by dividing this value by the enzyme concentration. The pH-dependence of kcat/KM was fitted to a bell-shaped curve representing 2 ionizable functionalities using the program GraFit. The pH-dependence of V n i a x was then determined by measuring the initial rate of hydrolysis of 4NPpG6P (at saturating concentrations of substrate) at a series of pH values. Al l experiments were carried out at 50 °C in 50 mM NaCl, 0.1 mM MnCl 2 , 1 p M Chapter 5 - Materials and Methods 237 N A D + , 10 mM 2-mercaptoethanol, 0.1 % (w/v) BSA containing either: 20 mM AcOH/NaOAc (pH 4.0-4.5), 20 mM MES (pH 6.1-6.7), 20 mM HEPES (pH 6.5-8.2), or 20 mM CHES (pH 8.4-10.0). The enzyme (final concentration of 4.1 Ug/mL) was pre-incubated with the above solutions (final assay volume = 200 U.L) at 50 °C for 5 minutes, and 4NP(3G6P (final concentration = 1.34 mM or 2.67 mM) was used to initiate the enzymatic reaction. Two different substrate concentrations were used in each buffer solution to ensure substrate saturation and that the observed rate corresponds to V n i a X . The differences in extinction coefficients Ae between pNPpG6P and the 4-nitrophenolate anion at the different buffer solutions were determined by the method described by Kempton and Withers 1 2 4 . The V i r i a x values were calculated from the observed data. The pH-dependence of V n i a x was fit to an equation describing a reaction governed by two essential ionizations. Data fitting was performed using the program GraFit. Section 5.3.2.11. Bronsted Analysis. Each aryl 6-phospho-p-D-glucoside was assayed in Buffer A and initial rates were measured. The typical substrate concentrations used range from 0.7-7 KM, and 7-10 data points were collected for each substrate. The concentration of BglT used in the final assay volume of 200 uL varied from 2.25-4.5 u.g/mL. The difference in extinction coefficients Ae between the aryl 6-phospho-p-D-glucoside and the phenol released at pH 7.5, 50 °C was determined by the method described by Kempton and Withers.1 2 4 The catalytic parameters (kcat and KM) were determined based on a direct fit of the data to the Michaelis-Menten equation. Logarithms of the kcat and kcat/KM values were plotted against the leaving group pKa values. The Bronsted coefficient was obtained from the slope of this plot. Section 5.3.2.12. Deuterium Kinetic Isotope Effect Measurements. A l l experiments were carried out at 50 °C in 1 cm path length-matched quartz cuvettes under the following conditions: 50 mM HEPES, 1 mM MnCl 2 , 0.001 mM N A D + , 10 mM mercaptoethanol, and 0.1 % (w/v) B S A at pH 7.5, 50 °C. For kH/kD measurements, BglT (final concentration of 2.87 u.g/mL) was preincubated with the above solution for 5 minutes, and substrate was added to the reaction mixture to give a Chapter 5 - Materials and Methods 238 final volume of 1 mL. Final substrate concentration was chosen to be 615 p M (over 10 times KM) to measure catalytic rates. Measurements for substrates, 4NPpG6P, 3[2H]4NPpG6P , 2[ 2H]4NPpG6P and 1 [ 2H]4NPpG6P, were repeated 10 times each, and KlEs were calculated from the data. For (A:Cat/^M')H/(^cat/^M)D measurements, BglT (final concentration 5.73 pg/mL) was preincubated with the buffer and cofactors at 50 °C for 5 minutes, and PNPpG6P, 3[ 2H]4NPpG6P, 2[ 2H]4NPpG6P, or 1 [ 2H]4NPpG6P (final concentration of 6.15 p M , at a concentration that was equal to 0.15 KM was used individually to initiate the enzymatic reaction. The absorbance at 400 nm was monitored until the substrate was depleted. The data set was fitted to a first order curve, and the kCJKM values were obtained by dividing this value by the enzyme concentration. Each measurement was repeated 8 times, and the (kCJ KM\\l(kCJ KM)D value was calculated. Section 5.3.2.13. pD-Dependence. BglT and all reagents were exchanged into D 2 0 buffer. BglT was incubated at a series of pD values ranging from pD 3.0-13.0 for 30 minutes, and periodically aliquots were removed for assay at pD 7.5. These studies revealed that BglT was stable in the range pD 4.0-10.0. The pD-dependence of kCJKM was then determined by measurement of kCJKM at a series of pD values using the substrate depletion method. A l l experiments were carried out at 50 °C in 50 mM NaCl, 0.1 mM MnCl 2 , 1 p M N A D + , 10 mM 2-mercaptoethanol (D 6), 0.1 % (w/v) BSA containing either: 20 mM AcOD/NaOAc (pD 4.0-4.5), 20 mM MES (pD 6.1-6.7), 20 mM HEPES (pD 6.5-8.2), or 20 mM CHES (pD 8.4-9.4). The enzyme (final concentration of 20.7 pg/mL or 5.2 pg/mL) was pre-incubated with the above solutions (final assay volume = 200 pL) at 50 °C for 5 minutes, and 4NPPG6P (final concentration = 5.85 p M = 0.08 KM) was used to initiate the enzymatic reaction. The kCJKM values were obtained by dividing the pseudo-first order rate constant by the enzyme concentration. The pD-dependence of kCAT/KM was fit to an equation describing a reaction governed by two essential ionizations. Data fitting were performed using the program GraFit. Chapter 5 - Materials and Methods 239 Section 5.3.2.14. Solvent Deuterium Kinetic Isotope Effect. Buffers, chemicals and the enzyme stock solution were prepared in D2O. The enzyme (final concentration = 1.03 ug/mL) was preincubated in Buffer B, and initial rates were measured upon addition of 4NPpG6P to a final assay volume of 200 pL. The assay pD was chosen to be 8.1 at which the rate is optimal and independent of small changes in pD. The difference in extinction coefficients, Ae, between 41MPpG6P and the 4-nitrophenolate anion at pD 8.1, 50 °C was determined to be 12,612 M"'cm~', and the catalytic parameters were determined based on a direct fit of the data to the Michaelis-Menten equation. Section 5.3.2.15. Primary Kinetic Isotope Effect Measurements for 2[2r1]4NPf3G6P in D 2 0 Buffer. Buffers, chemicals and the enzyme stock solution were prepared in D2O. A l l experiments were carried out in Buffer B (final assay volume = I mL). For (&Cat)H/(&cat)D measurements, initial rates were measured for 2[2lT]4NPpG6P or 4NPpG6P using BglT (final concentration of 4.1 pg/mL), at a final substrate concentration of 580 pM (> 8 KM). Alternating between the protio and deuterio substrates, linear initial rates were measured for 4NPpG6P and 2[ 2H]4NPpG6P, and the KIE (kc,t)H/(kcat)D was calculated by dividing the rate constant for the protio-substrate by the rate constant for the deuterio-substrate. The set of experiments was repeated 10 times, and the average KIE value was calculated from the data. Measurement of (ACSI/KM^/^W^VOD was performed using the substrate depletion method in which BglT (final concentration 8.2 pg/mL) was preincubated in Buffer B (final assay volume = 1 mL), with 4NPpG6P or 2[ 2H]4NPpG6P (final concentration = 5.85 pM = 0.08 KM) alternately. The resulting data set was fit to a first order curve, and the kCJKM values were obtained by dividing the pseudo-first order rate constant by the enzyme concentration. The set of experiments was repeated 9 times, and the average (kcat/'KM)HI(kcatlKM)O value was calculated from the data by dividing the first order rate constant for the protio-substrate by the first order rate constant for the deuterio-substrate in each case and averaging the set of values obtained. Chapter 5 - Materials and Methods 240 Section 5.3.2.16. Determination of K\ Values. Kinetic studies were performed at 50 °C in 50 mM HEPES, 0.1 mM MnCl 2 , 1 p M N A D + , 10 mM 2-mercaptoethanol, and 0.1 % (w/v) BSA at pH 7.5, using 2.8 pg/mL BglT (final assay volume = 200 pL). Approximate K\ values were determined by measuring the reduction in BglT activity as measured by the hydrolysis of 4NP(3G6P, in the presence of the disaccharide substrates. BglT was added to the assay mixtures containing a fixed concentration of 4NPpG6P and varying amounts of 4NPC6'P ( , S-4NPC6T, C6'P or iS-C6'P. As the hydrolysis of the disaccharides does not release 4-nitrophenolate anion, the disaccharides can be considered as competitive inhibitors against the 4NPpG6P substrate, and the rate of hydrolysis of 4NPpG6P can be monitored spectrophotometrically at 400 nm. The experiments were repeated at different concentrations of PNPG6P. The data was graphed on a Dixon plot (1/v v.v [competitive inhibitor]). A horizontal line drawn through 1/Kmax in the Dixon plot intersects the experimental lines at an inhibitor concentration equal to -KL The same experiments were performed with the inhibitors MeaG6P, 4NP2FpG6P, and 4NP2FpG6P to determine the K\ values, by monitoring for the decrease in BglT activity against the substrate, 4NPpG6P. Section 5.3.3. Enzyme Kinetics with GlvA. Unless stated otherwise, GlvA is preincubated in the assay buffer at 37 °C for 5 minutes prior to the addition of substrate to initiate the enzymatic reaction. Section 5.3.3.1. Buffer Systems. Buffer C: 50 mM HEPES, 1 mM MnCl 2 , 0.1 mM N A D + , 10 mM 2-mercaptoethanol, and 0.1 %(w/v) BSA atpH 7.5 Buffer D: 50 mM Tris-HCl, 1 mM MnCl 2 , 0.1 mM N A D + , 10 mM 2-mercaptoethanol, and 0.1 % (w/v) B S A at pH 8.4. Chapter 5 - Materials and Methods 241 Section 5.3.3.2. Michaelis-Menten Kinetics - 4NPaG6P and 4NP|3G6P by Direct UV-vis Assay. All experiments were carried out in either Buffer C or Buffer D. Activity of GlvA was measured with 4NP(3G6P ( l n i a x 400 nm, A£ (pH 7.5) =12618 cm"'M"', Ae (pH 8.4) = 17308 cm"'M"') or 4NPocG6P ( ? w 400 nm, Ae (pH 7.5) = 12602 cm"'M"', Ae (pH 8.4) = 17259 cm'M"') by monitoring the release of the 4-nitrophenolate anion. The initial linear rate of increase in absorbance at 400 nm is measured upon addition of the substrate. The difference in extinction coefficients Ae between the substrate and the 4-nitrophenol released at pH 7.5 and pH 8.4 were determined by the method described by Kempton and Withers 1 2 4 , and the catalytic parameters were determined based on a direct fit of the data to the Michaelis-Menten equation. Section 5.3.3.3. Glucose 6-Phosphate Dehydrogenase Coupled Assay. Reaction rates were measured using a glucose 6-phosphate dehydrogenase (G6PDH) coupled assay. Al l experiments were carried out at 37 °C in Buffer C or Buffer D, and 2 mM N A D P + and 20 units of G6PDH. The activity of GlvA was measured spectrophotometrically by monitoring the formation of the NADPH cofactor at 340 nm. Initial rates were measured upon addition of MeaG6P to the reaction mixture (final volume ~ 200 U.L). 8-12 data points were collected: MeaG6P (final substrate concentration range = 100-1700 U.M, final enzyme concentration = 6.5 ug/mL (pH 7.5) or 3.3 ug/mL (pH 8.4)). The molar extinction coefficient of 6,220 M " W (Ae of NADPH) was used for the calculation of initial rates of substrate hydrolysis, and the catalytic parameters were determined based on a direct fit of the data to the Michaelis-Menten equation. The concentration of GlvA was doubled for two data points and the observed rate was also doubled, ensuring that concentrations of GlvA, G6PDH and N A D P + used for the coupled assay were not rate-limiting factors. Furthermore, GlvA was assayed with 4NPaG6P in the presence of 2 mM N A D P + , and the enzyme was also assayed with 4NPaG6P in the presence of 20 units of G6PDH. In each case the observed rate was the same as when GlvA was assayed alone, indicating that the presence of N A D P + and G6PDH did not affect the activity of GlvA . In addition, approximate values were Chapter 5 - Materials and Methods 242 determined by measuring the reduction in GlvA activity as measured by the hydrolysis of 4NPaG6P in the presence of MeaG6P. GlvA was added to the assay mixtures containing a fixed concentration of 4NPocG6P and varying amounts of MeaG6P. Since MeaG6P and its hydrolyzed products are not chromogenic, inclusion of MeaG6P does not interfere with the detection of the 4-nitrophenolate anion liberated by the hydrolysis of 4NPaG6P and MeaG6P can be considered as a competitive inhibitor. (See Determination ofK^) Section 5.3.3.4. Determination of K$ Value for Mn . 1 mL GlvA (1 mg/mL) was first dialyzed against 5 x 3 L 50 mM HEPES at pH 7.5 to remove any bound divalent metals prior to manipulation. Samples of dialyzed GlvA (final concentration of 6.5 pg/mL) and M n 2 + (varied from 20 p M to 1 mM) were preincubated in 50 mM HEPES (pH 7.5) or 50 mM Tris-HCl (pH 8.4), 0.1 mM N A D + , 10 mM 2-mercaptoethanol, and 0.1 % (w/v) B S A at 37 °C for 5 minutes, then 4NPaG6P (final concentration > 10 KM) was added to the reaction mixture to give a final volume of 200 pL and initial rates measured. The reaction rate was plotted against the concentration of Mn to generate ligand binding curves. Section 5.3.3.5. Determination of /fd Value for N A D + . 1 mL GlvA (1 mg/mL) was first dialyzed against 5 x 3 L 50 mM HEPES at pH 7.5 to remove any bound N A D + prior to manipulation. Samples of dialyzed GlvA (final concentration of 6.5 pg/mL) and NAD +(varied from 5 p M to 100 pM) were preincubated in 50 mM HEPES (pH 7.5) or 50 mM Tris-HCl (pH 8.4), 0.1 mM MnCl 2 , 10 mM 2-mercaptoethanol, and 0.1 % (w/v) BSA at 37 °C for 5 minutes, then 4NPaG6P (final concentration > 10 KM) was added to the reaction mixture to give a final volume of 200 pL and initial rates measured. The reaction rate was plotted against the concentration of N A D + to generate ligand binding curves. Chapter 5 - Materials and Methods 243 Section 5.3.3.6. Kinetic and Spectroscopic Investigation of Cofactor Reduction. For the kinetic analysis, a sample of dialyzed GlvA (final concentration = 3.3 U.g/mL = 65 nM) and N A D + (final concentration = 100 u M ) was preincubated in Buffer C for 5 minutes, then 4NPaG6P (final concentration of 669 U.M > 10 K M ) was added to the reaction mixture (final volume = 200 U.L). In a second reaction, dialyzed GlvA (final concentration = 3.3 Ug/mL = 65 nM), N A D + (final concentration = 100 | lM) , and NaBH 4 (final concentration = 10 mM) was preincubated in Buffer C for 5 minutes, and 4NPaG6P (final concentration of 669 uM) was added to a final assay volume of 200 | iL . The change in A400 was monitored for 18 minutes prior to the addition of 10 U.L of 200 U.M N A D + to rescue enzyme activity. The change in A400 was then measured for another 20 minutes. The same experiment was repeated in Buffer C or Buffer D. The absorbance spectra (from 250 nm to 400 nm) of the following three dialyzed enzyme samples were measured in 1 mL, 1 cm pathlength - matched quartz cuvettes: 100 u M GlvA; 100 u M GlvA incubated with 100 j iM N A D + ; and 100 \iM GlvA incubated with 100 u M N A D + and 10 mM sodium borohydride. Measurement of sample pH after reaction confirmed that 10 mM sodium borohydride did not significantly alter the pH of the solution. The absorbance spectra (from 320 nm to 400 nm) of GlvA were recorded under three differing conditions: 10 u M GlvA; 10 (lM GlvA incubated with 10 p.M N A D + ; and 10 | l M GlvA incubated with 10 U.M NAD + and 10 mM sodium borohydride. The peak in absorbance at 340 nm corresponding to N A D H appears upon reduction with 10 mM sodium borohydride, and is consistent with the quantitative reduction of N A D + to N A D H . Based on the small Kd value of 16.6 u M (pH 7.5) for the N A D + , over 90 % of GlvA has N A D + or N A D H bound to its active site under these conditions and very little remains free in solution. GlvA activity was assayed in the presence of N A D + and the activity was compared to that obtained in the presence of N A D H (quantitatively reduced from N A D + using sodium borohydride). The enzyme was completely inactive in the presence of NADH (reduced form) alone. Upon addition of excess N A D + , full activity was rapidly restored, demonstrating that the loss of activity upon reduction is due solely to having the dinucleotide cofactor in the wrong redox state rather than protein denaturation or any significant changes in pH. Chapter 5 - Materials and Methods 244 Section 5.3.3.7. Stereochemical Outcome Determination by Methanolysis. A reaction mixture (5 mL) containing 25 mM Tris-HCl, pH 7.5, 1 mM MnCl 2 , 0.1 mM NAD, 9.7 pg/mL enzyme, 10 mg 4NPaG6P and 5 M MeOH was incubated at 37°C for 24 hours. Alkaline phosphatase (20 pL of 1 mg/mL; Sigma) was added to cleave the sugar phosphate and the reaction mixture incubated at 37°C for 2 hours. The enzymes were then removed using a 10 kDa N M W L centrifugal filter unit, and the Mn ion was removed using metal chelating resin. The filtrate was lyophilized and redissolved in D 2 0 for ' H NMR analysis. The composition of the reaction mixture was determined by comparing the 'H NMR spectra of the standard solutions of MeaGlc, Me(3Glc and Glc. "H NMR spectra were collected at room temperature and at 50 °C to shift the solvent D 2 0 peak downfield, because the anomeric proton of MeaGlc is very close to that of the solvent D 2 0 peak at 4.63 ppm at room temperature. The same methanolysis reaction was conducted with 4NPpG6P as substrate. Another set of methanolysis experiments was conducted with either 4NPaG6P or 4NPpG6P as substrate in D 2 0 buffer and 5 M CD 3 OD. 'H NMR analyses were also performed analogous to those described above. Section 5.3.3.8. Solvent Deuterium Incorporation. All buffers and chemicals were lyophilized twice from 99.9% D 2 0 . GlvA was exchanged into deuterated buffer solutions via repeated (three times) dilution and concentration using a centrifugal filter unit (Millipore) with a nominal molecular weight limit (NMWL) of 10 kDa. The enzymatic reactions were performed under the following conditions: 50 mM Tris-HCl, pH 7.5, 1 mM MnCl 2 , 0.1 mM NAD, 0.8 mg/mL enzyme. pNPaG6P (20 mg) was incubated at 37°C in 5 mL of the above solution and the reaction was monitored by TLC. After 24 hours additional enzyme (2 mg) was added. After a further 24 hours the enzyme was removed using a 10 kDa N M W L centrifugal filter unit and 100 mg of Chelex 100 resin (BioRad) added to the filtrate and stirred at room temperature for 2 hours to remove metal ions. Filtration of the resin left a solution that was lyophilized, then redissolved in D 2 0 for 'H NMR analysis. An identical control experiment, but without enzyme, was carried out. The same experiment was carried out with 4NPPG6P (20 mg) as substrate, and the equivalent control experiment without Chapter 5 - Materials and Methods 245 enzyme was also conducted. Two other controls involved 'H NMR analysis of reaction mixtures in which the product G6P had been incubated with the buffer and cofactors in the presence and absence of GlvA. Section 5.3.3.9. pH-Dependence. GlvA was incubated at a series of pH values ranging from 2.0-14.0 for 30 minutes and periodically aliquots were removed for assay at pH 7.5. These studies revealed that GlvA was stable in the pH range 4.0-10.0. The pH-dependence of kcJKM was then determined by measurement of kcJKM at a series of pH values using the substrate depletion method. Al l experiments were carried out at 37 °C in 50 mM NaCl, 1 mM MnCl 2 , 0.1 mM N A D + , 10 mM 2-mercaptoethanol, 0.1 % (w/v) BSA containing either: 20 mM AcOH/NaOAc (pH 4.0-4.5), 20 mM MES (pH 6.1-6.7), 20 mM HEPES (pH 6.5-8.2), or 20 mM CHES (pH 8.4-9.4). The enzyme (final concentration of 13.1 ug/mL or 26.2 U-g/mL) was pre-incubated with the above solutions (final assay volume = 1000 U.L) at 37 °C for 5 minutes, and 4NPaG6P (final concentration = 6.7 U.M) was used to initiate the enzymatic reaction. The kcJKM values were obtained by dividing the pseudo-first order rate constant by the enzyme concentration. The pH-dependence of kcJKM was fit to an equation describing a reaction governed by two essential ionizations. Data fitting was performed using the program GraFit. The pH-dependence of V n i a x was then determined by measuring the initial rate of hydrolysis of 4NPaG6P at a series of pH values. Al l experiments were carried out at 37 °C in 50 mM NaCl, 1 mM MnCl 2 , 0.1 mM N A D + , 10 mM 2-mercaptoethanol, 0.1 % (w/v) BSA containing either: 20 mM AcOH/NaOAc (pH 4.0-4.5), 20 mM MES (pH 6.1-6.7), 20 mM HEPES (pH 6.5-8.2), or 20 mM CHES (pH 8.4-10.0). The enzyme (final concentration of 4.1 ug/mL) was pre-incubated with the above solutions (final assay volume - 200 uL) at 37 °C for 5 minutes, and 4NPaG6P (final concentration = 1.34 mM or 2.67 mM) was used to initiate the enzymatic reaction. Two different substrate concentrations were used in each buffer solution to ensure substrate saturation and that the observed rate corresponds to Vmax- The difference in extinction coefficients Ae between pNPocG6P and the 4-nitrophenolate anion at the different buffer solutions were determined by the method described by Kempton and Withers. 1 2 4 The Vmax values Chapter 5 - Materials and Methods 246 were calculated from the observed data. The pH-dependence of V n i a x was fit to an equation describing a reaction governed by two essential ionizations. Data fitting was performed using the program GraFit. Section 5.3.3.10. Bronsted Analysis. Each aryl 6-phospho-p-D-glucoside was assayed in Buffer C and Buffer D and initial rates were measured. The typical substrate concentrations used range from 0.7-7 KM, and 7-10 data points were collected for each substrate. The concentration of GlvA used in the final assay volume of 200 pL varied from 1.6-16 pg/mL. The difference in extinction coefficients, Ae, between the aryl 6-phospho-p-D-glucoside and the phenol released at pH 7.5 and at pH 8.4 were determined by the method described by Kempton and Withers,1 2 4 and the Ae values were used in the calculation of the initial rates of substrate hydrolysis. The catalytic parameters (kcat and KM) were determined based on a direct fit of the data to the Michaelis-Menten equation. Logarithms of the kcat and k^JKu values were plotted against the leaving group pKa values. The Bronsted coefficients were obtained from the slope of the plots. 4NPaG6P, 34DNPaG6P and PaG6P were assayed in Buffer C and Buffer D and initial rates were measured. The typical substrate concentrations used range from 0.7-7 A"M, and 7-10 data points were collected for each substrate. The concentration of GlvA used in the final assay volume of 200 pL varied from 1.0-13 pg/mL. The difference in extinction coefficients Ae between the aryl 6-phosho-a-D-glucoside and the phenol released at pH 7.5 and at pH 8.4 were determined by the method described by Kempton and Withers. 1 2 4 The catalytic parameters (kcal and KM) were determined based on a direct fit of the data to the Michaelis-Menten equation. Bronsted plots were generated by plotting the logarithms of the kcat and kcat/KM values against the leaving group pKa values. The kinetic parameters for the hydrolysis of MeocG6P were included for this plot and the methanol leaving group was assumed to have a pKa of 15.5. Chapter 5 - Materials and Methods 247 Section 5.3.3.11. Deuterium Kinetic Isotope Effect Measurements. Deuterium kinetic isotope effects for 1 [2H]4NPccG6P, 2[ 2H]4NPaG6P, 3[ 2H]4NPaG6P, MeaG6P, l[ 2H]4NPpG6P, 2[ 2H]4NPpG6P and 3[2H]4NPpG6P were measured at two different substrate concentrations to give isotope effects on kcai and on kcJKu-For k^/ko measurements, GlvA (final concentration of 3.3 mg/mL (pH7.5) or 1.6 pg/mL (pH 8.4)) and substrate (final concentration was chosen such that it was at least equivalent to 10 K M ) were used to obtain kcat values from initial rates. Reactions were carried out at 1 mL assay volumes. Measurements for substrates, 4NPpG6P, 1[2H]4NPPG6P, 2[2H]4NPpG6P, 3[2H]4NPpG6P, 4NPocG6P, 1 [ 2H]4NPaG6P, 2[ 2H]4NPaG6P, and 3[ 2H]4NPaG6P, were repeated at least 6 times each in Buffer A and in Buffer B in alternation. Initial rates for the substrates containing the 4-nitrophenol leaving group were measured by monitoring the reaction spectrophotometrically at 400 nm. For MeocG6P and l[ 2H]MeaG6P, initial rates were measured using the G6PDH coupled assay, using 6.5 pg/mL GlvA, 20 units G6PDH, 2 mM N A D P + , 8.0 mM substrate. Each measurement was repeated at least 6 times in Buffer A and in Buffer B and the initial rate was monitored spectrophotometrically at 340 nm for NADPH formation. In each case, the (A^at)m/(^at)D value was calculated by dividing the rate for the protio substrate by the rate for the deuterio substrate. For (A:Cat/A"ivi)H/(^cat/^M)D measurements, GlvA (final concentration chosen such that the reaction was complete in less than 1 hour) Was assayed with 4NPPG6P, l[ 2H]4NPpG6P, 2[2H]4NPpG6P, 3[2H]4NPpG6P, 4NPaG6P, l[ 2H]4NPaG6P, 2[ 2H]4NPaG6P, and 3[ 2H]4NPaG6P (final concentration was chosen such that it was less than 0.2 KM). The absorbance at 400 nm was monitored until the substrate was depleted. The data set was fitted to a first order curve, and the kcJKu values were obtained by dividing this value by the enzyme concentration. For MeaG6P and l[2H]MeocG6P, the substrate depletion method was used with the G6PDH coupled assay and the change in absorbance at 340 nm was monitored for the formation of NADPH. Each measurement was repeated 6 times, and the (kcJK.M)^l(kcJKu)o value was Chapter 5 - Materials and Methods 248 calculated by dividing the pseudo-first order rate for the protio substrate by that obtained for the deuterio substrate. Section 5.3.3.12. Determination of Ai Values. C6'P, MepG6P, 4NP2deoxypG6P and 4NP2FpG6P were found to be competitive inhibitors of GlvA. Kinetic studies were performed in Buffer C or Buffer D and the appropriate amount of GlvA (final assay volume = 200 p.L). Approximate K\ values were determined by measuring the reduction in the rates of hydrolysis of 4NPaG6P, in the presence of different concentrations of inhibitor. GlvA was added to assay mixtures containing a fixed concentration of 4NPaG6P and varying amounts of C6'P, MeaG6P, MepG6P, 4NP2deoXypG6P, or 4NP2FpG6P, and the rate of hydrolysis of 4NPaG6P can be monitored spectrophotometrically at 400 nm. The experiments were repeated at different concentrations of 4NPaG6P. The data was graphed on a Dixon plot (1/v vs [competitive inhibitor]). A horizontal line drawn through 1/Fmax in the Dixon plot intersects the experimental lines at an inhibitor concentration equal to -K\_ Section 5.4. Molecular Biology. Section 5.4.1. General Materials and Methods. 1 % agarose (Bio-Rad, CA, USA) gel was used for all D N A electrophoresis. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (10 % SDS-PAGE) (stained with Coomassie blue) was performed to analyze the homogeneity of proteins. Molecular weight markers used were a pre-stained SDS PAGE standard (Bio-Rad), contained myosin 209 kDa, p-galactosidase 124 kDa, bovine serum albumin 80 kDa, ovalbumin 49.1 kDa, carbonic anhydrase 34.8 kDa, soybean trypsin inhibitor 28.9 kDa, lysozyme 20.6 kDa, and aprotinin 7.1 kDa. The Escherichia coli strains used in this study were TOP10 for D N A isolation and BL21 for protein expression. Yeast extracts and Tryptone were purchased from Difco Laboratory, M l , USA. 40% Acry1amide/N,N'-methylene bisacrylamide and N,N,N',N'-tetramethylethylenediamine (TEMED) were purchased from Bio-Rad, CA, USA. Al l other chemicals and reagents were purchased from Sigma Chemical Co. unless otherwise noted. For the study of BglT and BglT mutants, mass Chapter 5 - Materials and Methods 249 spectra were recorded using an ABI MDS-SCIEX API QSTAR Pulsar /' mass spectrometer (Sciex, Thornhill, ON) at the University of British Columbia. Gene sequencing was performed by the NAPS Unite (Nucleic Acid Protein Service Unit) at the University of British Columbia. Oligonucleotide primers were purchased from Integrated D N A Technologies. 5.4.2. Overexpression of the Gene (bglt) Encoding BglT from Thermotoga maritima. The plasmid containing the T. maritima genomic D N A cloned into Pet22b vector was generously provided by our collaborators Dr. Annabelle Varrot and Dr. Gideon Davies. The clone was transformed into electrocompetent BL21 cells. The cells were grown in Luria broth media supplemented with 100 pg/mL ampicillin ( L B A m p ) for Pet22b. For expression of the bglt gene, cells were grown in Luria broth media supplemented with 100 pg/mL ampicillin at 37 °C to an OD o^onm of approximately 0.6 before induction with 0.1 mM IPTG. Overnight cultures were harvested by centrifugation (10 min at 5000 rpm) and the pellet from 1 L of culture resuspended in 40 mL 50 mM HEPES pH 8.0. The cells were lysed by French cell press. The cells were centrifuged 30 minutes at room temperature and 18000 rpm in order to remove cell debris and intact cells. The supernatant was collected and incubated for 30 min at 70 °C to denature the non-thermally stable E. coli proteins. After centrifugation at 18000 rpm for 30 min, the clarified supernatant was recovered and loaded on a 5 mL HiTrap D E A E FF ion exchange column (Amersham). The column was washed with 10 volumes of 50 mM HEPES pH 8.0 before elution of the protein with a step gradient involving increasing concentrations of 50 mM HEPES containing 500 mM NaCl at pH