Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Binding of substrates and inhibitors to human pancreatic alpha amylase Zhang, Xiaohua 2016

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

Item Metadata


24-ubc_2016_september_xiaohua_zhang.pdf [ 4.16MB ]
JSON: 24-1.0305050.json
JSON-LD: 24-1.0305050-ld.json
RDF/XML (Pretty): 24-1.0305050-rdf.xml
RDF/JSON: 24-1.0305050-rdf.json
Turtle: 24-1.0305050-turtle.txt
N-Triples: 24-1.0305050-rdf-ntriples.txt
Original Record: 24-1.0305050-source.json
Full Text

Full Text

Binding of Substrates and Inhibitors to Human Pancreatic Alpha Amylase  by Xiaohua Zhang B.S. (Hons) National University of Singapore, Singapore, 2008  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in The Faculty of Graduate and Postdoctoral Studies (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (VANCOUVER)  June 2016  © Xiaohua Zhang, 2016  ii Abstract Human pancreatic -amylase (HPA) is the enzyme responsible for hydrolyzing starch within the gut into shorter oligosaccharides. Selective inhibition targeted at only HPA could be used to modulate blood glucose levels for the treatment of diabetes and obesity. Montbretin A (MbA) is a potent (Ki = 8.1 nM) and specific inhibitor of HPA. Controlled degradation studies on MbA, coupled with inhibition analysis, identified an essential high-affinity core structure comprising the myricetin and caffeic acid moieties linked via a disaccharide, mini-MbA. X-ray structural analyses of the complex of MbA-HPA confirmed the importance of this core structure and revealed a novel mode of glycosidase inhibition wherein internal -stacking interactions between the myricetin and caffeic acid organize their ring hydroxyls for optimal hydrogen bonding to the catalytic residues of HPA. The simplified analogue mini-MbA therefore offers potential for new strategies for glycosidase inhibition and therapeutic development. As part of a search for selective, mechanism-based covalent inhibitors of HPA, chemo-enzymatic syntheses of oligoglycosyl epi-cyclophellitols are described. - Glucosyl epi-cyclophellitol, synthesized from epi-cyclophellitol by coupling of a glucosyl moiety using maltose phosphorylase, inactivated HPA stoichiometrically. X-ray crystallographic analysis of the covalent derivative so formed confirmed its reaction at the active site with the catalytic nucleophile Asp197. Another trisaccharide analogue 4’-O-methyl--maltosyl epi-cyclophellitol was synthesized enzymatically or by in situ elongation by HPA. Both of the inhibitors showed time-dependent inactivation of HPA, with the trisaccharide version being a better inactivator.  iii This new class of mechanism-based inhibitors will be useful as activity-based probes for amylases.  Several potential starch binding sites have been identified on the surface of HPA by crystallography, but their role, if any, in starch degradation is unknown. Through analysis of the binding of HPA mutants, modified individually at each site, to soluble and granule starch, two of these surface binding sites (SBSs) were shown to play a role in starch granule binding. A quite separate site was shown to be important for binding to, and cleavage of, soluble starch. Binding at SBSs was distinguished from binding to the active site by blocking the active site with the glycosyl epi-cyclophellitol mechanism-based inactivators.     iv Preface A version of Chapter 2 has been published in Nature Chemical Biology 2015. 11 (9): 691-696 with the title “The amylase inhibitor montbretin A reveals a new glycosidase inhibition motif” by Williams, L. K.; Zhang, X. H.; Caner, S.; Tysoe, C.; Nguyen, N. T.; Wicki, J.; Williams, D. E.; Coleman, J.; McNeill, J. H.; Yuen, V.; Andersen, R. J.; Withers, S. G. and Brayer, G. D.  Williams, L. K. and Zhang, X. H. contributed equally to the work. I was responsible for performing the degradation studies of MbA, characterizations of the subsequent fragments, inhibition kinetics and specificity tests of different glucosidases. N.T.N. crystallized the MbA–HPA complex; L.K.W. solved and analyzed its structure; S.C. solved the structure of HPA-mini-MbA; C.T. synthesized MbA-analogue inhibitor and performed kinetic studies thereon; J.W. performed kinetic studies on MbA variants; D.E.W. performed the ROESY NMR studies on MbA; J.C., V.Y., J.M., R.J.A., S.G.W. and G.D.B. designed and supervised the project.  Chapter 3 is based on work titled “Glucosyl epi-cyclophellitol allows mechanism-based inactivation and structural analysis of human pancreatic alpha-amylase” by Caner, S.; Zhang, X.H.; Jiang, J.; Chen, J.; Nguyen, N.T.; Overkleeft, H.; Brayer, G.D.; and Withers, S.G. in the recent publication in FEBS letters 2016. 590: 1143-1151. Caner, S. and Zhang, X.H. contributed equally to the work. I synthesized and characterized G-ECP, and performed kinetic and mass spectrometry studies thereon. H.O., G.D.B, and S.G.W supervised the project; J.J, and H.C synthesized ECP; N.T.N. crystallized wild type HPA and prepared the G-ECP/HPA complex; S.C. collected X-ray diffraction data, and solved the resultant structures; manuscript preparation was primarily done by S.C., G.D.B. and S.G.W. with contributions from all.  v Chapter 4 has not been published. The crystal structures were obtained by Dr. Brayer’s group at UBC. I conducted all the experiments including structure mapping, mutagenesis, kinetic testing and analyzed the results.    vi Table of Contents Abstract ............................................................................................................................................ii Preface ............................................................................................................................................ iv Table of Contents ............................................................................................................................ vi List of Tables .................................................................................................................................. xii List of Figures ................................................................................................................................ xiii List of Abbreviations ................................................................................................................... xviii Acknowledgements ....................................................................................................................... xxi Dedication .................................................................................................................................... xxii Chapter 1 General Introduction ...................................................................................................... 1 1.1 Glycoside hydrolases ......................................................................................................... 2 1.2 Classifications ..................................................................................................................... 3 1.3 Catalytic mechanisms of glycosidases ............................................................................... 5 1.4 -Amylase family ............................................................................................................... 7 1.5 Human pancreatic α-amylase ............................................................................................ 9 1.6 Starch ............................................................................................................................... 12  vii 1.7 Discovery of HPA inhibitors ............................................................................................. 15 1.7.1 Acarbose ................................................................................................................ 15 1.7.2 Flavonoids .............................................................................................................. 18 1.7.3 Proteinaceous inhibitors ........................................................................................ 22 1.7.4 Mechanism-based inhibitors ................................................................................. 24 1.8 Aims of the thesis ............................................................................................................ 28 Chapter 2 Structural and Mechanistic Characterization of the Amylase Inhibitor Montbretin A and Its Substructures .................................................................................................................... 29 2.1 Introduction to montbretin A .......................................................................................... 30 2.2 Aim of the project ............................................................................................................ 32 2.3 Results and discussions .................................................................................................... 33 2.3.1 Generation of MbA substructures by chemical and enzymatic modifications ..... 33 2.3.2 Kinetic analysis of the binding of MbA substructures to HPA ............................... 39 2.3.3 Inhibition studies for porcine pancreatic α-amylase ............................................. 42 2.3.4 Contributions from each fragment moiety ........................................................... 43 2.3.5 Comparison of the kinetic results with the X-ray structural analysis of HPA-MbA complex ........................................................................................................................... 44  viii 2.3.6 Crystal structure of mini-MbA and HPA complex .................................................. 46 2.3.7 Inhibition specificity of MbA and mini-MbA.......................................................... 48 2.4 Conclusion ........................................................................................................................ 50 Chapter 3 Chemoenzymatic Synthesis of Oligoglycosyl Epi-cyclophellitols Allows Mechanism-Based Inactivation and Structural Analysis of Human Pancreatic Alpha-Amylase ....................... 53 3.1 Introduction ..................................................................................................................... 54 3.2 Aim of the project ............................................................................................................ 58 3.3 Results and discussions .................................................................................................... 59 3.3.1 Chemoenzymatic synthesis of glucosyl epi-cyclophellitol ..................................... 59 3.3.2 Kinetic analysis of the inactivation of HPA by the two glucosyl epi-cyclophellitols ........................................................................................................................................ 61 3.3.3 Mass spectrometric analysis of the inactivated enzyme....................................... 63 3.3.4 Structural analysis of the inactivated enzyme. ...................................................... 64 3.3.5 Chemoenzymatic synthesis and characterization of a maltotriosyl epi-cyclophellitol ................................................................................................................... 67 3.3.6 In situ elongation of MeG2-ECP ............................................................................. 70 3.3.7 Comparisons of the two ECP-containing MBIs ...................................................... 73  ix 3.4 Conclusion ........................................................................................................................ 74 Chapter 4 Exploration of Starch Surface Binding Sites on Human Pancreatic Alpha-Amylase .... 77 4.1 Introduction ..................................................................................................................... 78 4.2 Aim of the project ............................................................................................................ 83 4.3 Results and discussions .................................................................................................... 85 4.3.1 Choice of mutants .................................................................................................. 85 4.3.2 Site-directed mutagenesis and mutant enzyme expression in Pichia ................... 92 4.3.3 Catalytic efficiencies of mutant proteins using CNP-G3 as a substrate ................ 97 4.3.4 Catalytic efficiencies of mutant proteins using soluble starch as a substrate ...... 98 4.3.5 Starch granule binding assay ............................................................................... 102 4.3.6 Starch granule binding assay using a MBI as an active-site blocking reagent .... 106 4.3.7 Proposed HPA binding mechanism with starch granules .................................... 114 4.3.8 Amylase evolution ............................................................................................... 115 4.4 Conclusion ...................................................................................................................... 117 Chapter 5 Experimental Procedures ........................................................................................... 120 5.1 Synthesis ........................................................................................................................ 121 5.1.1 General materials for synthesis ........................................................................... 121  x 5.1.2 General enzymatic degradation procedures for MbA degradation .................... 122 5.1.3 Enzymatic synthesis of glucosyl epi-cyclophellitols ............................................. 133 5.1.4 Enzymatic synthesis of 4’-O-methyl α-maltosyl epi-cyclophellitol (MeG2-ECP) . 135 5.1.5 Mass spectrometric analysis of intact protein .................................................... 135 5.2 Molecular biology .......................................................................................................... 136 5.2.1 General materials ................................................................................................ 136 5.2.2 Site directed mutagenesis ................................................................................... 136 5.2.3 Generation of pPic9K-HPA mutant vector ........................................................... 137 5.2.4 Screening for successful transformants by activity ............................................. 138 5.2.5 Protein expression and purification .................................................................... 139 5.3 Kinetics ........................................................................................................................... 140 5.3.1 General assay conditions ..................................................................................... 140 5.3.2 Enzymatic activity tests of naringinase................................................................ 141 5.3.3 Michaelis-Menten kinetics of HPA using CNP-G3 as a substrate ........................ 142 5.3.4 Dinitrosalicylic acid assay of soluble starch ......................................................... 142 5.3.5 Inhibitory assay .................................................................................................... 143 5.3.6 Inactivation assays ............................................................................................... 145  xi 5.3.7 Evaluation of kinetics of reactivation of HPA ...................................................... 146 5.3.8 Enzyme concentration measurement using NanoOrange assay ......................... 146 5.3.9 Adsorption assay of enzymes to starch ............................................................... 147 References .................................................................................................................................. 149 Appendices .................................................................................................................................. 163 Appendix A: Kinetics ............................................................................................................ 163 A1 Michaelis-Menten kinetics ...................................................................................... 163 A2 Enzyme kinetics in the presence of a reversible inhibitor ...................................... 168 A3 Enzyme kinetics in the presence of a mechanism-based inhibitor ......................... 174 Appendix B: Lineweaver-Burk plots of HPA inhibition by MbA derivatives ........................ 176     xii List of Tables Table 1-1 Flavonol and flavone structures. .................................................................................. 19 Table 2-1 Color-coded schematic diagrams of intact  MbA and derivatives, with associated Ki values for binding to HPA, and IC50 values for both HPA and PPA.. ...................................... 40 Table 2-2 Contributions of equilibrium free energy from each fragment .................................... 43 Table 2-3 Inhibition of HPA and other α-glycosidases by MbA and mini-MbA. ........................... 49 Table 4-1 Observed binding interactions of sugar analogues with neighbouring amino acid residues at different sites on the HPA surface. ..................................................................... 89 Table 4-2 Oligonucleotide sequences of the mutagenic primers ................................................. 93 Table 4-3 Catalytic efficiencies of HPA variants for hydrolysis of CNP-G3 and soluble starch. ... 97 Table 4-4 Apparent binding constants for HPA variants to the starch granules in the absence and presence of MbA. ......................................................................................................... 104 Table 4-5 Apparent binding constants for HPA variants to starch granules. ............................. 108     xiii List of Figures Figure 1-1 Glycosidase mechanisms. (a) Inversion mechanism, and (b) retention mechanism. ... 6 Figure 1-2 Three dimensional structure of human pancreatic α-amylase ................................... 10 Figure 1-3 Catalytic mechanism of human pancreatic α-amylase................................................ 11 Figure 1-4 Schematic representation of the different structural levels of the starch granule and the involvement of amylose and amylopectin.. .................................................................... 15 Figure 1-5 Structures for acarbose, miglitol and voglibose .......................................................... 16 Figure 1-6 Schematic drawing of the structure of acarbose and its expected mode of binding to HPA active site. ...................................................................................................................... 16 Figure 1-7 Chemical structure of Montbretin A. .......................................................................... 21 Figure 1-8 Crystal structure of PPA with tendamistat. ................................................................. 23 Figure 1-9 Structures of activated (a) 2-deoxy-2-fluoroglycoside; (b) trinitrophenyl 2-deoxy-2,2-difluoro glycoside; (c) 5-fluoro glycosyl fluorides; (d) mechanism of trapping of β-glucosidases by 2-deoxy-2 fluoro probe. R is aglycone. ........................................................ 25 Figure 1-10: In situ elongation-trapping strategy for HPA using the example of 5-fluoro-idosyl fluoride (5FIdoF) as the acceptor and 4’-O-methyl--maltosyl fluoride (MeG2F) as the donor. .................................................................................................................................... 27 Figure 2-1 The chemoenzymatic degradation strategy of montbretin A. .................................... 31  xiv Figure 2-2 Treatment of MbA to afford decaffeoylated product MbA-C. .................................... 34 Figure 2-3 Treatment of MbA with Abg to afford MbA-G. ........................................................... 35 Figure 2-4 Proposed scheme of MbA-C with treatment of Abg. .................................................. 36 Figure 2-5 Hydrolysis of naringin into prunin, rhamnose, naringenin and glucose by naringinase containing α-L-rhamnosidase and β-D-glucosidase activities. .............................................. 37 Figure 2-6 Treatment of MbA with heat-treated Naringinase to afford MbA-R. ......................... 37 Figure 2-7 Treatment of MbA with Naringinase to afford mixture of MbA-RX and MbA-GRX. ... 37 Figure 2-8 Treatment of MbA-C with heat-treated Naringinase to afford MbA-CR. ................... 38 Figure 2-9 Treatment of MbA-R with xylosidase to afford MbA-RX. ............................................ 39 Figure 2-10 Correlations of inhibition constants of MbA derivatives for HPA and PPA. .............. 42 Figure 2-11 The interactions of MbA adjacent to HPA active site residues and the novel stacking conformation formed by the aromatic rings of MbA ............................................................ 45 Figure 2-12 The interactions of mini-MbA inhibitor bound in the active site of HPA .................. 47 Figure 2-13 Structural (a) and corresponding schematic (b) representation of hydrogen bond interactions in the mini-MbA/HPA complexes ...................................................................... 47 Figure 2-14 Structural overlay of mini-MbA with MbA in the active site of HPA ......................... 48 Figure 3-1 (a) Mechanism of inactivation of retaining β-glucosidases by epoxides. Structure of (b) exocyclic epoxide; (c) conduritol B epoxide; (d) cyclophellitol and (e) epi-cyclophellitol. ... 55  xv Figure 3-2 (a) Formation of G-ECP catalyzed by LaMalP in sodium acetate buffer; (b) reaction of alkene precursor of ECP with βG1P; (c) illustration of the expected mechanism-based inhibition of G-ECP with HPA. ................................................................................................ 61 Figure 3-3 Residual activity of HPA in the presence of () 3.6 mM α-1.4 G-ECP, and () 6 mM α-1,3 G-ECP in phosphate buffer .............................................................................................. 63 Figure 3-4 Mass spectra of HPA (above) and HPA treated with α-1,4 G-ECP (below). ................ 64 Figure 3-5  (a) Simulated annealing omit map at the 3σ level (green mesh) around the bound G-ECP inactivator (yellow sticks) in the active site of HPA. Major and minor conformations are shown to demonstrate the two different binding modes. Superposition of the G-ECP ligand in the major conformation with the HPA inhibitors. (b) acarbose  and (c) MeG2-5FIdo. Structural (d) and schematic (e) representations of hydrogen bond interactions in the G-ECP-HPA complex.. ................................................................................................................ 66 Figure 3-6 Chemoenzymaic synthesis of MeG2-ECP using MeG2F and ECP by CGTase in sodium phosphate buffer ................................................................................................................... 69 Figure 3-7 Residual activity of HPA in the presence of 29 M MeG2-ECP in phosphate buffer .. 69 Figure 3-8 Schematic drawing of the structure of G-ECP, MeG2-5FIdoF and the expected mode of MeG2-ECP binding to HPA active site.. ............................................................................. 70 Figure 3-9 Proposed in situ elongation-trapping mechanism for HPA using ECP as the acceptor and MeG2F as the donor. ...................................................................................................... 71  xvi Figure 3-10 Mass spectra of HPA (above) and HPA treated with MeG2F and ECP (below). ........ 71 Figure 4-1 Snapshots of a variety of malto-oligosaccharide ligands in different positions on the HPA surface ............................................................................................................................ 83 Figure 4-2 Flow chart of project design ........................................................................................ 84 Figure 4-3 Zoomed view of structures of HPA at (a) Active site and SBS 2; (b) SBS 3; (c) SBS 4; (d) SBS 5; (e) SBS 6; (f) SBS 7; (g) SBS 8; (h) SBS 9 and (i) SBS 10.. .............................................. 88 Figure 4-4 Graphic representation of overlap extension polymerase chain reactions (OE-PCRs) 93 Figure 4-5 The map of (a) pPIc9 vector and (b) pPic9K vector. (c) Multiple plasmid integration in Pichia...................................................................................................................................... 95 Figure 4-6 Reducing sugar production in the supernatant of starch suspension, when incubated with HPA in the presence/absence MbA at 4 C. ................................................................ 105 Figure 4-7 Structural overlap of barley -amylase AMY1 and HPA ........................................... 108 Figure 4-8 Zoomed view of HPA active site with (a) MbA from PDB 4W93; (b) MeG2-ECP  ..... 110 Figure A-1 Graphical representation for the determination of the modes of inhibition with a Lineweaver-Burke plot. (a) Competitive inhibition, (b) noncompetitive inhibition and (c) uncompetitive inhibiton. ..................................................................................................... 172 Figure A-2 Graphical representation for the determination of the modes of inhibition with a Dixon plot. (a) Competitive inhibition, (b) noncompetitive inhibition and (c) uncompetitive inhibiton. .............................................................................................................................. 173  xvii Figure A-3: Lineweaver-Burk plot of inhibition of HPA by MbA-G (Ki = 9.1 ± 0.8 nM). .............. 176 Figure A-4 Lineweaver-Burk plot of inhibition of HPA by MbA-R (Ki =21.3 ± 2.2 nM). .............. 176 Figure A-5 Lineweaver-Burk plot of inhibition of HPA by MbA-RX (Ki = 42.4 ± 5.3 nM). ........... 176 Figure A-6: Lineweaver -Burk plot of inhibition of HPA by MbA-GR (Ki = 79.3 ± 9.5 nM) .......... 177 Figure A-7 Lineweaver-Burk plot of inhibition of HPA by MbA-GRX (Ki = 93.3 ± 7.6 nM) .......... 177 Figure A-8 Lineweaver-Burk plot of inhibition of HPA by MbA-C (Ki = 0.73 ± 0.09 µM). ........... 177 Figure A-9 Lineweaver-Burk plot of inhibition of HPA by MbA-CG (Ki = 2.2 ± 0.2 µM). ............. 178 Figure A-10 Lineweaver-Burk plot of inhibition of HPA by MbA-CR (Ki = 46.8 + 3.9 µM) .......... 178 Figure A-11 Lineweaver-Burk plot of inhibition of HPA by MbA-CGR (Ki = 128 ± 15 µM) .......... 178     xviii List of Abbreviations  5FGlcF  5-Fluoro-α-D-glucopyranosyl fluoride 5FIdoF 5-Fluoro--L-idosyl fluoride  Abg Agrobacterium sp. β-glucosidase  ACN  Acetonitrile AGE Affinity gel electrophoresis  BCA Bicinchoninic acid BMGY  Buffered Glycerol-complex Medium BMMY Buffered Methanol-complex Medium BSA Bovine serum albumin  CAZy Carbohydrate Active Enzyme CBE Conduritol B epoxide CBM Carbohydrate binding module CGTase Cyclodextrin glucanotransferase CNP-G3 2-Chloro-4-nitrophenyl α-maltotrioside CP Cyclophellitol DNP 2,4-Dinitrophenol DNS 3,5-Didinitrosalicyclic acid  EC Enzyme commission  ECP 1,6-epi-cyclophellitol  ESI Electrospray ionization G3 Maltotriose  G6 Maltohexaose  G-ECP Glucosyl epi-cyclophellitol GH Glycoside hydrolase GHIL D-gluconohydroximino-1,5-lactam HPA Human pancreatic -amylase   xix HPLC High performance liquid chromatography HRMS High resolution mass spectrometry HSA  Human salivary -amylase ITC Isothermal titration calorimetry kcat  Catalytic rate constant Kd Binding constant or equilibrium dissociation constant  ki Inactivation rate constant Ki  Dissociation constant Km  Michaelis constant LaMalP Lactobacillus acidophilus NCFM  MbA Montbretin A MBI Mechanism-based inhibitor MD Minimal Dextrose MeG2-ECP 4’-O-methyl-α-maltosyl epi-cyclophellitol MeG2F 4’-O-methyl-α-maltosyl fluoride MeOH Methanol MGAM Maltase-glucoamylase  MS Mass spectrometry NaOH Sodium hydroxide NAPS Nucleic Acid and Protein Service NMR Nuclear magnetic resonance OE Overlap extension PCR Polymerase chain reaction PDB Protein data bank PPA Porcine pancreatic α-amylase SBS Surface binding site SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis SI Sucrase-isomaltase   xx SPR Surface plasmon resonance  TLC Thin layer chromatography UV-Vis Ultraviolet-visible light Vmax Maximum velocity of an enzyme-catalyzed reaction     xxi Acknowledgements I would like to thank my supervisor, Dr. Stephen G. Withers, for his patience, encouragement and excellent guidance. Without him, this thesis could not have been accomplished. I feel so lucky to have been part of his research group. His enthusiasm for science and life have been and will continue inspiring me.  I would also like to thank the past and present members of the Withers group for all the fun times in lab and out. I am particularly indebted to Dr. Hong-Ming Chen for his supply of synthetic compounds, Dr. Lars Baumann for good discussions, Ms. Emily Kwan for her biochemical expertise, and Ms. Tianmeng Duo, Dr. Ching-Ching Yu, Dr. Tina Rasmussen and Dr.  Andrés Gonzalez Santana for their advice on correcting my Ph.D. thesis.  Also, I thank the departmental Nuclear Magnetic Resonance and Mass Spectrometry for their expertise, and Genome Canada, NSERC, as well as the University of British Columbia for funding these ventures.  Lastly, special thanks to my family, whose support, criticism and encouragement always helped me achieve my best. My heartfelt personal thanks goes to my parents, who are far away in China. Their support is always an important inspiration source for me.   xxii Dedication    谨以此文献给我的 父亲 张学武先生 母亲 董金芝女士 感谢他们对我学业的无限理解和全心支持!     1       Chapter 1 General Introduction    2 1.1 Glycoside hydrolases Glycoside hydrolases (GH, also known as glycosidases or glycosyl hydrolases) are a group of enzymes that catalyze the hydrolysis of glycosidic linkages in oligosaccharides, polysaccharides and glyco-conjugates, the most structurally diverse class of biopolymers. Glycosidases are found in essentially all domains of life, from archaebacteria to humans, and are largely responsible for the degradation of biomass (e.g. cellulose and hemicellulose), anti-bacterial defense strategies (e.g. lysozyme), and in normal cellular functions (e.g. trimming mannosidases involved in N-linked glycoprotein biosynthesis). As some measure of their importance to biology, around 1-3% of the average genome is dedicated to carbohydrate active enzymes, many of which are glycosidases. 1 Together with glycosyltransferases, glycosidases form the major catalytic machinery for the synthesis and breakage of glycosidic bonds. Since the structure of lysozyme was first solved in 1965,2 glycosidases have been major subjects of structural biology studies in order to understand the molecular details of substrate recognition and of catalysis. A large number of three-dimensional structures of glycosidases have been solved by X-ray diffraction and NMR methods. This accomplishment has been possible mostly for several reasons. One is that many glycosidases have high stability and are easy to isolate from natural sources and higher organisms. A second is that glycosidases are used in a number of  industrial applications for processing, such as fermentation, pulp and paper etc. 3 Moreover, glycosidases are involved in many biological process including cell-cell, or cell-virus recognition, immune response, and viral and parasitic infections. 4 Therefore, their function or dysfunction has been implicated in a number of different diseases, such as diabetes, influenza, tumor metastasis and Gaucher disease etc, leading to an interest in drug discovery  3 offering exciting new therapeutic opportunities. 5,6,7,8,9 As such, they have become interesting targets of industrial and pharmaceutical companies. 1.2 Classifications With an increasing number of three-dimensional protein structures being solved by X-ray diffraction and NMR studies, there has come a realization that, categorization solely by their enzyme commission (EC) numbers, conceived by the International Union of Biochemistry and Molecular Biology Enzyme Commission, is an unsatisfactory way to classify and predict glycosidase structural/functional relationships. This is primarily because the system relies on their substrate specificities and/or their molecular mechanism. 10,11 The major drawback stems from the fact that the EC number cannot correlate the structural features and evolutionary relationships of enzymes. Furthermore, the EC classification cannot be applied to enzymes with unidentified function. To overcome these limitations, an alternative classification system, the Carbohydrate Active Enzyme database (CAZy;, was introduced by Bernard Henrrissat in 1991. 12 CAZy is based on sequence similarity of carbohydrate-active enzymes, and represents the most thorough attempt to classify enzymes that work upon carbohydrate-based substrates. CAZy annotation consists of assigning module/domain families and subfamilies for six major CAZyme categories: glycoside hydrolases (GHs) and polysaccharide lyases (PLs) that cleave glycans, glycosyltransferases (GTs) that synthesize glycans, carbohydrate esterases (CEs), and auxiliary activities (AAs) that work as redox enzymes to act in conjunction with carbohydrate-active enzymes, as well as non-catalytic carbohydrate binding modules (CBMs).  4 Each of the enzyme classes is further divided into families that are designated by Arabic numerals (e.g. GH13). These are expanded into subfamilies on the basis of members of one family sharing more sequence/structural and functional characteristics that applicable for the entire family (e.g. GH13_24). In other words, subfamily members should share a more recent evolutionary ancestor. The classification of glycosidases based upon sequence similarity can greatly simplify studies of this class of enzymes since enzymes within the same family are thought to share structural and mechanistic similarities. As the number of solved crystal structures of glycosidases has grown, it has been found that some families having little sequence similarity are nonetheless structurally related. Another level of hierarchical classification in the CAZy database, “clans” numbered from A to N (e.g. GH-H), was thus introduced to group together families sharing catalytic machinery and adopting the same structural fold of the catalytic domain, but with significant difference in overall sequence.  Grouping the complex array of carbohydrate active enzymes by amino acid sequence does not always account for differences in substrate specificity. Family GH13, for instance, includes α-amylases, pullulanases, and trehalose synthases, as well as enzymes with many other functions. The converse is also true, with different enzymes having the same function being classified into different families. Today, CAZy encompasses more than 130 GH families and underpins the majority of functional, structural and mechanistic considerations of these proteins. 11,13 CAZy provides a direct relationship between sequence and folding similarities that allows reliable prediction of catalytic mechanism (retaining/inverting), active site residues and possible substrate  5 specificities. 11, 14,15,16 This allows generalizations to be made based on structural and mechanistic studies on representative members of a family. 1.3 Catalytic mechanisms of glycosidases The conceptual basis used by glycosidases to achieve glycosidic bond cleavage in a stereospecific manner was first proposed by Koshland.17 It was based on the stereochemical outcome of the reactions, and on known chemical mechanisms of similar non-enzymatic reactions. The anomeric configuration of the observed product is either inverted or retained relative to that of the substrate glycoside. In Koshland’s proposal, inverting glycosidases catalyze the attack of water in a single nucleophilic displacement mechanism. On the other hand, a retention reaction requires a two-step mechanism involving a stabilised intermediate to afford overall retention of the anomeric configuration (Figure 1-1). Both Koshland mechanisms proceed via transition states with substantial oxocarbenium ion character and involve two conserved active site residues, typically glutamic or aspartic acid, to facilitate catalysis. The vast majority of known glycosidases adopt one of these two classical mechanistic schemes. In the inversion mechanism, the reaction is completed in a single step, with the two adjacent carboxylic residues spaced 6-11 Å apart,18 functioning as a general acid and base catalyst, respectively. They position on opposite sides of the substrate glycoside, to accommodate both the substrate and a water molecule. During the reaction, one of the carboxylate residues protonates the scissile glycosidic oxygen atom to facilitate the departure of the aglycone leaving group, while the general base catalyst coordinates the nucleophile (i.e. water) to assist the attack at the anomeric centre. Overall, the inverting mechanism is completed in a single   6  Figure 1-1 Glycosidase mechanisms. (a) Inversion mechanism, and (b) retention mechanism.  displacement mechanism via an oxocarbenium ion-like transition state. Moreover, it does not involve formation of any covalent enzyme intermediate during the course of catalysis and induces the inversion of anomeric configuration of the starting material. The retention mechanism also involves a pair of essential carboxylic acid residues, which are normally closer at ~ 5.5 Å apart. 18 The mechanism is known as a double displacement reaction involving two steps: a glycosylation step and a deglycosylation step. Firstly, the enzyme is glycosylated by the concerted action of the two carboxylic acid residues in the enzyme’s active site. One of the residues functions as a nucleophile, attacking at the substrate’s anomeric center, while the aglycone leaves with the assistance of another carboxylic acid, acting as a general acid catalyst. A covalent glycosyl-enzyme intermediate that has an opposite anomeric configuration to that of the starting material is formed. In the second step, the deglycosylation  7 step, the carboxylic acid that first worked as a general acid catalyst, now in its deprotonated form, acts as the general base catalyst and activates an incoming water molecule, which further attacks the anomeric center of the glycosyl-enzyme intermediate to generate the hydrolyzed product with a net retention of anomeric stereochemistry. Both of these steps involve oxocarbenium ion-like transition states, which are greatly stabilized by finely tuned active site noncovalent interactions. As a result of a number of structural and mechanistic characterizations of glycosidases, entirely different mechanisms involving elimination and/or hydration steps have been revealed for some glycosidases within the last decade. These mechanisms may proceed either through cationic or anionic transition states. Several reviews have covered these types of atypical mechanism.3,19,20,21  Since they are not the focus of this thesis, they will not be covered in detail. 1.4 -Amylase family Among the many enzymes that are widely used industrially, the -amylase clan (GH13, 70 and 77 also known as clan GH-H), is an important class of nearly 30 different hydrolytic enzymes that cleave the -linked glycosidic bonds in starch, amylose, amylopectin, glycogen and various maltodextrins to yield oligosaccharides, with retention of -anomeric configuration in the products. 12,22 The most complex is the GH13 family, which consists of hydrolases, transferases and isomerases with over 22 known specificities. 22 -Amylases share a number of common characteristics such as a ()8 barrel structure, the hydrolysis or formation of glycosidic bonds with the -configuration, and a number of conserved amino acid residues in the active site. 23,24  8 Most of the -amylases are metalloenzymes, which require calcium ions (Ca2+) for their activity, structural integrity and stability. 25, 26 -Amylases are produced by a diverse variety of organisms, including bacteria, fungi, plants and animals. In many mammals, they have been found to exist in the saliva, having been synthesized in the parotid gland where they begin the chemical process of digestion, and in the gut where starch is mainly digested by pancreatic -amylase. In bacteria and fungi, -amylases are secreted into their environment where they encounter starch that is hydrolyzed to maltodextrins and transported into the cells to be further processed. Plants also produce -amylases to degrade starch, their main energy storage form. 27 Because starch is the ultimate energy unit obtained via photosynthesis, starch digestion becomes a very important process in the utilization of solar energy, hence the wide distribution of -amylases in nature. -Amylases have diverse applications in industry, such as in the preparation of fermented foods, the production of paper and textiles, and the bioconversion process of starchy biomass to biofuels. 25,28,29 They are in increasing demand due to their important role in starch hydrolysis and the applications of this hydrolytic action. As an industrial enzyme of high demand, -amylases hold the maximum market share of enzyme sales, and represent approximately 30% of the world’s enzyme production. 30 The ubiquitous nature, ease of production and broad spectrum of applications make α-amylase an industrially important enzyme.  Further, since modulation of -amylases activity affects the utilization of carbohydrate as an energy source in humans, and cause several diseases, such as diabetes, obesity and hyperglycemia, control of the activity of the enzymes provides a valuable therapeutic approach.  9 31,32,33,34 Therefore, the spectrum of -amylases application has expanded into many other fields, such as clinical, and medicinal chemistry. 1.5 Human pancreatic α-amylase  Human -amylases (α-1,4 D-glucan glucanohydrolase, EC are endoenzymes that catalyze the hydrolysis of α-1,4-glycosidic linkages in starch and are present in both salivary and pancreatic secretions. In humans, the digestion of starch involves several stages. Initially, partial digestion of starch by the human salivary -amylase (HSA) results in degradation into shorter oligosaccharides. Since HSA is largely inactivated by the acid and the proteases in the stomach, these products and the remaining starch are extensively hydrolyzed by human pancreatic α-amylase (HPA) in the small intestine and excreted into the lumen. The resultant oligosaccharides, including maltose, maltotriose and a number of α-1,6 and α-1,4 oligoglucans, will be further hydrolyzed into glucose by other brush border enzymes, such as disaccharidases, maltase-glucoamylase (MGAM) and sucrase-isomaltase (SI) and finally enter the bloodstream. Hence, these enzymes play an important role in the production of glucose in the human lumen and are effective drug targets for the treatment of obesity and type II diabetes mellitus, which alone affects an estimated 180 million people worldwide. 35 Both HSA and HPA are composed of 496 amino acids in a single polypeptide chain with essential chloride and calcium binding ions. 36,37,38 The two isozymes are encoded as part of a multigene family on chromosome 1 and regulated so that the different isozymes are expressed solely in either the salivary glands or the pancreas. 39,40 They are highly homologous in terms of primary sequence with only 15 amino acid differences, corresponding to 97% similarity. 41 The high   10           Figure 1-2 Three dimensional structure of human pancreatic α-amylase. Also indicated are the relative positions of the three structural domains in the protein (Domain A residues 1-99 and 169-404; Domain B residues 100-168; Domain C residues 405-496) along with locations of the calcium and chloride binding sites. The  strands are shown in cyan. The HPA coordinates used were from PDB 1HNY.  homology in their sequences is reflected by a correspondingly high level of structural similarity, with both being composed of three domains.37,38 The crystal structure of HPA is shown in  Figure 1-2. Domain A (residues 1-99 and 169-404) is a (/)8 barrel containing the active site as well as the conserved chloride binding site. Three carboxylic acids (Asp197, Glu233, and Asp300), which are conserved in other amylase enzymes, are located at the bottom of the V-shaped active-site cleft. The chloride ion and calcium ion are located in close proximity to the V-shaped depression. While Ca2+ ions have been known for a long time to render amylase resistant to proteolytic attack, 26 the bound chloride acts as an allosteric effector, and removal of this ion results in a 30-fold decrease in activity. 36,42,43 Domain B is a large loop (residues 100-168) which occurs between the third -strand and the -helix of the barrel domain, and it forms a structurally crucial calcium binding site, which is believed to be responsible for stabilizing the active-site loop. Domain C (residues 405-496) is only loosely associated with an Domain A Domain B Domain C  11 anti-parallel -barrel type structure of unknown functions. Subsite mapping studies have clearly demonstrated the presence of at least five high affinity glucose-binding subsites within the active site, with three on the nonreducing side of the scissile bond and two sites on the reducing end side. 44,45  HPA catalyzes the hydrolysis of the α-1,4 glycosidic linkage in starch with net retention of anomeric configuration and is believed to proceed via a classical double displacement mechanism (Figure 1-3). 45 The first step involves a nucleophilic attack at the substrate’s anomeric center by the catalytic nucleophile Asp197 while the aglycone leaves with the assistance of the general acid catalyst Glu233. In the second step, the resultant glycosyl-enzyme intermediate is further hydrolyzed by an incoming water molecule with the aid of general base catalyst Glu233. Both of these steps proceed in the oxocarbenium ion-like transition states that are stabilized through noncovalent interactions.  Figure 1-3 Catalytic mechanism of human pancreatic α-amylase.  12 The sequence of HPA reveals a high degree of homology with 83% identity with porcine pancreatic α-amylase (PPA), 37, 46 with only 70 amino acids different to each other. Belonging to the same GH13 family as HPA, PPA shares a similar structure and mechanism with HPA, and catalyzes the hydrolysis of internal α-1,4-glycosidic bonds in starch. It is a single chain polypeptide with its amino-terminal end is blocked as a pyrrolidone carboxylic acid. 47 The crystal structure of PPA is very closely related to that of HPA, but with the difference being that PPA crystallizes in a lattice with a much more “open” active site. 48  1.6 Starch As the natural substrate of amylase, starch is the most common carbohydrate in the human diet and is produced by all green plants as an energy store. Depending on different botanical origins of plants, starch is synthesized in plastids and amyloplasts as semi-crystalline granules varying in shape (spherical, oval or polygonal), in size (1 µm-100 µm in diameter), in association of individual or granule clusters and in composition (α-glucan, lipid, moisture, protein and mineral content).30, 49,50,51 Regardless of the botanical origins, starch is a polysaccharide primarily composed of two types of polymers, the linear and helical amylose and the branched amylopectin, with generally 20 to 25% amylose and 75 to 80% amylopectin by weight. 52  Amylose is a relatively long, linear α-glucan containing around 99% -1,4- and 1% -1,6- linkages, with a molecular weight of approximately 1 x 105 –1 x 106 Da. 53,54 Because of the low degree of branching, dissolved amylose has a tendency to form insoluble semi-crystalline aggregates. 51 The polysaccharide chain of amylose may be folded into three different structures denoted A, B, and V. 50,51,55,56 Both the structures of A- and B-type amyloses are  13 double helices and they differ only in packing arrangement and water content. B-type amylose is packed in a hexagonal mode with the central channel surrounded by numerous water molecules (36 H2O/unit cell), while A-type is packed closer with only eight water molecules per unit cell. 57,58 The symmetry of the double helices also differs in A and B structures, since the repeat unit is a maltotriosyl unit in the A form and a maltosyl unit in the B form. 58 Generally, the A-type occurs preferentially in cereals and the B-form in tuber starch (e.g. potato). The polysaccharide chain of V-amylose is a generic term for amyloses obtained as a left-handed single helix with six to eight residues per turn, 59 59 59 59 59 59 59 59 59 and forms a central channel-like cavity which can co-crystallize with another hydrophobic guest molecule such as iodine, a fatty acid, or an aromatic compound. 50,51 Within this group of V-amylose, there are many different variations. Each is notated with V and then a subscript indicating the number of glucose units per turn. The most common is the V6 form, which has six glucose units a turn with 7.91- to 8.17-Å pitch height.60, 61  Amylopectin, on the other hand, is a much larger and highly branched polymer, with molecular weight of 1 x 107 –1 x 109 Da. 44,54 The linear successive glucose units are linked by α-1,4-glycosidic linkage with 5-6% of -1,6 bonds at the branch points, which have a profound effect on the physical and biological properties. The basic organization of the chains is described in terms of the A-, B- and C- chains as defined by Peat et al. 62 The outer A-chains are analogous to amylose, and are glycosidically linked at their potential reducing group through C6 of a glucose residue to an inner B-chain. The single C-chain per molecule likewise carries other chains as branches but contains the sole reducing terminal residue (Figure 1-4d). The current model of amylopectin structures indicates that five to eight A-chains cluster on the B-chain and that such  14 clusters occur every 25-35 residues eventually constituting the subdomains in the structure of amylopectin. 62  The native starch granule has a growth-ring structure composed of alternately arranged semi-crystalline and amorphous shells (Figure 1-4). 50,51 The semi-crystalline shells are mainly composed of clusters of branched-chain amylopectin while the amorphous shells are of long linear-chain amylose and low-molecular-mass amylopectin. 63 Both amylose chains and exterior chains of amylopectin can form double helices which may in turn associate to form crystalline domains. Several reviews on the structures of starch granules can be found. 50,51,52,55,58 Due to the complex structure of native starch granules, one would intuitively expect it represents a formidable challenge for enzymatic degradation in solution. Starch becomes soluble in water when heated. The granules swell and burst, the semi-crystalline structure is lost and the smaller amylose molecules start leaching out of the granule, forming a network that holds water, thereby resulting in more amorphous material and increasing the mixture's viscosity. 64 This gelatinization process during cooking makes the starch more digestible by mammalian -amylases. 65 Processing that disrupts the granule integrity and reduces the degree of crystallinity, increases the susceptibility to amylases. As such, enzymatic processing generally requires the presence of either specialized carbohydrate binding modules (CBMs) that are independently functional folding domains or in the form of surface binding sites (SBSs) on the surface of enzyme outside the active site. 66,67  15   Figure 1-4 Schematic representation of the different structural levels of the starch granule and the involvement of amylose and amylopectin (adapted from reference 50). (A) starch granule; (B) slice of a starch granule showing the growth rings consisting of alternating semi-crystalline and amorphous regions; (C) detail of the semi-crystalline region; (D) highly branched organization of the amylopectin molecule, and A-, B- and C-chains are shown; (E) structure of amylose.  Moreover, rapid digested and absorbed dietary starch results in a sharp increase in the postprandial blood glucose levels, and disorders of carbohydrate uptake may cause severe health problems, including type II diabetes, obesity as well as periodontal diseases, all of which threaten an increasing worldwide population.31,32 Such adverse consequences have led to the extensive study of enzyme/substrate or enzyme/inhibitor interactions in the starch digestion process, as discussed below.  1.7 Discovery of HPA inhibitors  1.7.1 Acarbose One common therapeutic strategy for the treatment of disorder of carbohydrate uptake is to retard the absorption of glucose through the inhibition of carbohydrate-hydrolyzing enzymes, such as the -glucosidases (MGAM and SI) and -amylases. The inhibition can delay carbohydrate digestion and prolong overall carbohydrate digestion time, causing a reduction in the rate of glucose absorption and consequently blunting the postprandial plasma glucose   16  Figure 1-5 Structures for acarbose, miglitol and voglibose    Figure 1-6 Schematic drawing of the structure of acarbose and its expected mode of binding to HPA active site (adapted from Brayer et al. 45 and Li et al. 68). The binding subsites in the active site cleft of HPA have been identified according to the convention indicated in the bottom portion of this diagram. Also shown is the expected binding mode for a normal starch substrate, where hydrolysis would occur between subsites -1 and +1.    17 rise.69 Most therapeutics currently in use were developed as inhibitors of these -glucosidases since this provided the added benefit of preventing the hydrolysis of common dietary sugars like sucrose into glucose while also blocking the hydrolysis of starch-derived oligosaccharides. 70 Indeed, inhibitors currently in clinical use, such as acarbose, miglitol and voglibose (Figure 1-5) are known to inhibit a wide range of α-glucosidases and -amylases.35, 71,72,73,74,76 Acarbose, a natural pseudo-tetrasaccharide extracted from Streptomyces sp. is one of the most studied glycosidase inhibitors currently in medical use for non-insulin-dependent Type II diabetes mellitus.71, 74-75 This structure is composed of a valienamine (unsaturated cyclitol) -N-linked to 4-amino-4,6-dideoxyglucose to form acarviosine. Acarviosine is further linked by an -1,4-O-glycosidic bond to a maltose moiety. The strong inhibition of amylases and related enzymes by acarbose is widely attributed to the half-chair planar conformation of the valienamine ring, which mimics that of the oxocarbenium ion-like transition state of the GH13 -glucosidases, and to the presence of strong electrostatic interactions between the carboxyl groups at the active site and the protonated nitrogen of the inhibitor. 45,73,76  When acarbose is bound to HPA in the crystal structure, it is modified by the enzyme through the addition of a maltosyl unit to the nonreducing end of the valienamine ring and the loss of a glucose moiety at the reducing end, leading to a larger and tighter binding inhibitor derivative as shown in Figure 1-6.45 The pseudo-pentasaccharide spans the -3 to +2 high-affinity binding subsites with the unhydrolyzable acarviosine group bound in subsites -1 and +1, 45 between which cleavage of substrates would normally occur (nomenclature –n/+n). 77 This is consistent with HPA’s substrate cleavage preferences of five binding subsites, all of which are occupied in the crystal structure of the HPA-acarbose complex. The proposed rearrangement mechanism by  18 which acarbose is enzymatically optimized upon binding to HPA was further described by Li et al. 68 Similar, but not identical, modifications have also been seen in structural studies of other -amylases in complexes with acarbose. 78,79,80  In fact, acarbose inhibits all of the intestinal -glucosidases with the exception of isomaltase, which cleaves -1,6 rather than -1,4 linked glucosidic bonds.81 As a consequence, the main adverse effects observed with acarbose are of gastrointestinal nature, including abdominal discomfort, flatulence and diarrhea,82,83 possibly due to the abnormal bacterial fermentation of undigested carbohydrates in the intestine. Indeed, these main side effects are common to -glucosidase inhibitors and often lead to therapy discontinuation.  To overcome these problems, specific inhibition of HPA is considered as a strategy for the alternative treatment of diabetes. As discussed previously, HPA is the enzyme at the top of the starch digestion pyramid. Selective inhibition targeted at only HPA, could be used to quantitatively modulate blood glucose levels by restricting or even shutting down starch degradation while still allowing hydrolysis of oligosaccharides, thereby minimizing the specificity problems that arise with currently available α-glucosidase inhibitors.  1.7.2 Flavonoids  In vitro studies on plant-extracts have shown inhibition of α-glucosidase and α-amylase activity or reduction of postprandial hyperglycemia. 84,85,86,87 The active ingredients of these extracts were found to contain flavonoids, which are an abundant class of natural phenolic compounds sharing a common structural skeleton consisting of two moieties: benzopyran (A and C rings)   19 Table 1-1 Flavonol and flavone structures. Adapted from Lo Piparo et al. 87  Compound  Group R3 R5 R6 R7 R3' R4' R5' Fisetin flavonol OH H H OH H OH OH Kampferol flavonol OH OH H OH H OH H Myricetin flavonol OH OH H OH OH OH OH Quercetagetin flavonol OH OH OH OH OH OH H Quercetin flavonol OH OH H OH OH OH H Isohamnetin flavonol OH OH H OH H OH OCH3 Rhamnetin flavonol OH OH H OCH3 OH OH H Acacetin flavone H OH H OH H OCH3 H Diometin flavone H OH H OH OH OCH3 H Eupafolin flavone H OH OCH3 OH H OH OH Genkwanin flavone H OH H OCH3 H OCH3 H Luteolin flavone H OH H OH OH OH H Scutellarein flavone H OH OH OH H OH H   and phenyl (B ring) groups (Table 1-1). Depending on the C ring type and the linkage between the benzopyran and phenyl groups, six groups of flavonoids have been categorized: flavones, flavonols, flavanones, isoflavones, flavanols and anthocyanidins. 88 Given the inadequacy of α-glucosidase inhibitors directed at diabetes, the use of optimized flavonoids could represent an alternative therapeutic route. Tadera et al. 89 tested several flavonoid compounds for their inhibitory activity against α-glucosidases and PPA. They showed that luteolin, myricetin and quercetin inhibited PPA with IC50 values lower than 500 µM, and  20 were able to establish a relationship between the inhibitory activity and the hydroxyl groups on the B ring: the hydroxyl substitution on the B ring enhanced the inhibitory activity.  Lo Piparo et al. 87 further investigated the structural requirements for inhibition of HSA by 19 flavonoids through in silico modeling. Their modelling suggested that the inhibitory activity depends on the formation of (1) hydrogen bonds between the hydroxyl groups in R7 and/or R4’ of the flavonoid ligand and the catalytic residues Asp197 and Glu233 of the active site, and (2) a possible π- π interaction between either the AC- or B-ring system and Trp59 to stabilize the interactions with the active site.    A more precise work published recently by Williams et al. 90 determined the high-resolution X-ray structure of myricetin in complex with HPA at 1.2 Å resolution. For the first time, they showed a novel binding mode of a flavonoid in the enzyme active site, distinct from those of other noncovalent inhibitors studied to date. In the crystal structure, myricetin partially fits within the extended substrate binding cleft of HPA through subsite -3 to -1 in the active site, and engages the catalytic residues Asp197 and Glu233 through hydrogen bonding to the R3’, R4’ and R5’-OH along the planar edge of its B-ring. Furthermore, the R3′-OH interacts with the putative nucleophilic water molecule that plays a role in the deglycosylation step of the catalytic reaction. The unique feature of myricetin interfering with the catalytic residues of HPA thereby illustrates the reason that it acts as a competitive inhibitor of HPA with Ki of 110 μM. 91 The result is also quite consistent with a previous study that treatment with myricetin (3 mg/12 h) resulted in the lowering of glycemia in diabetic rats by 50% after 2 days of treatment, while no significant effect was observed on the serum glucose levels in normal rats. 92 Also, there was  21 no indication of serious hepatotoxicity with myrecetin-treated rats and therefore suggested that myricetin could be of therapeutic potential in the management of hyperglyceria in dabetes. Another myricetin-containing molecule montbretin A (MbA in Figure 1-7) was identified by the Withers research group from natural product screening. MbA is a water-soluble glycosylated acyl flavonol comprising a myricetin core attached to two carbohydrate chains at O3 of the benzopyrone (C ring) and at C4’ of the phenyl moiety (B ring). One chain is made up of a xylose and a rhamnose, while the other contains another rhamnose, and two glucoses. 93 A caffeic acid moiety is attached to the C6 of the first glucose in the chain. It is found that MbA is a potent inhibitor of HPA (Ki = 8.1 nM) and has a high level of selectivity towards HPA when tested against a series of -glycosidases, including other enzymes from the GH13 family. 91  Figure 1-7 Chemical structure of Montbretin A.   22 The potency and specificity of MbA merit further study of its interactions with HPA, and structure-activity correlations. This could make MbA a good candidate for structural studies aimed at discovering new enzyme-inhibitor binding modes. We expect the extensive study of this molecule would allow us to clearly discover the key binding interactions with HPA.  1.7.3 Proteinaceous inhibitors Other amylase inhibitors include proteinaceous inhibitors 94 and small peptide mimics. 95 Tendamistat (HOE 467) is a 74 amino acid protein produced by Streptomyces tendae,96 which has an immunoglobulin-like fold consisting of two three-stranded antiparallel β-sheets stabilized by two disulfide bridges.97 Tendamistat forms a strong stoichiometric 1:1 complex with PPA (Figure 1-8), with an inhibition constant Ki of 9 x 10-12 M.98 The tight binding activity was in part explained by the multipoint interactions between the enzyme and tendamistat, which binds to an extended cleft in PPA and blocks substrate access to the active site in the crystal structure. 99 Four segments of its polypeptide chain in domains A and B, with a total of 15 amino acid residues, form various salt bridges, hydrogen bonds or hydrophobic interactions with PPA. Additional interactions involve residues in PPA that are further away from the active site on the surface of the cleft. A single β-loop that includes the Trp18-Arg19-Tyr20 motif, which is highly conserved in this class of inhibitors, interacts directly with the catalytic residues of PPA with strong electrostatic interactions. Schneider et al. 100 also suggested an induced-fit type of binding upon formation of the inhibitor-amylase complex by comparing the crystal structure of free tendamistat with its inhibitory complex form with PPA. Although tendamistat is a  23 promising drug for control of blood sugar levels, it has the intrinsic disadvantage of triggering a response from the immune system. Such a drawback makes it unsuitable for clinical trials.  Our group has recently identified and characterized another novel proteinaceous inhibitor, helianthamide, by screening natural product extracts from the UBC Marine Natural Products Extract Library, which contains 10,000 natural product extracts of marine origin. 101 The material with the greatest inhibitory activity was the product of the Caribbean Sea anemone, Stichodactyla helianthus. The active component was isolated and identified as helianthamide, which mimics the structure of -defensins. Kinetic experiments revealed that helianthamide is a reversible inhibitor of HPA with a Ki of 10 x 10-12 M and it appears to be specific for mammalian pancreatic α-amylase over other glycosidases.  Figure 1-8 Crystal structure of PPA with tendamistat. Green represents PPA and blue represents tendamistat structure. Adapted from PDB 1BVN.   24 Determination of the structure of the helianthamide/PPA complex shows helianthamide bound in a non-covalent complex with PPA. A third of helianthamide’s solvent accessible surface area is buried in contact with PPA, primarily within and around the amylase active site.  Further studies demonstrate that three aromatic residues of helianthamide (Tyr7, Tyr9 and His10) make all of the important polar contacts with the catalytic machinery of PPA and along with Val12, and create a non-polar interface to complement the hydrophobic ridges bordering PPA’s active site. Therefore, this novel YIYH motif of helianthamide may serve as inspiration for further drug development. Further studies are under investigation.  1.7.4 Mechanism-based inhibitors Besides the different classes of noncovalent inhibitors mentioned above, mechanism-based inhibitors (MBIs) are another class of molecules that ablate the enzyme activity through the formation of a covalent bond between the enzyme active site and some functionality on the inactivator, leading to a formation of a stable enzyme-inactivator complex. 102,103,104,105 Hence, the loss of the enzyme activity is mainly accounted for by physical blockage or chemical modification of the enzyme active site, so that no further catalytic cycles can occur.  This type of inactivator has been widely exploited for several applications in glycosidases. The most prevalent use is in the identification of active site residues. Together with mass spectrometry, site-directed mutagenesis and kinetic analysis, the function of the active site residues can be deciphered as either being catalytic or structural. 102,103 Also, highly specific inactivators can be used to selectively inactivate a target enzyme in complex biological systems while observing the effect of this “deletion” on the organism. This approach has been further  25 extended to the design of specific activity-based probes for the discovery and characterization of novel enzymes.102  The activated 2-deoxy-2-fluoro, 2-deoxy-2,2-difluoro and 5-fluoro glycosides (Figure 1-9 a-c) represent the most specific class of MBIs known for glycosidic activity. In 2-deoxy-2-fluoroglycosides, the replacement of the hydroxyl group by fluorine at C-2 slows both the glycosylation and deglycosylation steps in two ways (Figure 1-9 d). Firstly, introduction of fluorine removes or diminishes the key hydrogen bonding interactions found at the transition state between the hydroxyl group at C-2, and the enzyme active site (more than 8 kcal/mol). 106   Figure 1-9 Structures of activated (a) 2-deoxy-2-fluoroglycoside; (b) trinitrophenyl 2-deoxy-2,2-difluoro glycoside; (c) 5-fluoro glycosyl fluorides; (d) mechanism of trapping of β-glucosidases by 2-deoxy-2 fluoro probe. R is aglycone.  26  Secondly, fluorine is more electronegative than oxygen thus the electron-deficient transition states are further destabilized by the inductive effect. The consequence of the two effects is typically a 106 -107 -fold decrease of the rates of the two steps.107,108 Incorporation of a good leaving group, i.e. aglycones with low pKa values such as 2,4-dinitrophenolate or fluoride, to the anomeric position of these compounds increases the rate of the glycosylation step relative to the deglycosylation step, which results in the accumulation of a relatively stable 2-deoxy-2-fluoroglycosyl-enzyme intermediate. The formation of the covalent linkage and 1:1 stoichiometry of inactivator and enzyme have been supported by 19F-NMR and ESI/MS studies. 108-109 Moreover, the enzyme activity can be recovered either through hydrolysis of the covalent glycosyl-enzyme intermediate, or by transglycosylation onto a suitable sugar acceptor.108,110  While these active 2-deoxy-2-fluoro sugars have proven effective in trapping the intermediates during the double displacement mechanism followed by retaining -glucosidases, they have not proved generally useful for -glucosidases. 45 The reason was proposed as a consequence of stereoelectronic effect and the fluorine substituent is not sufficiently slowing in the deglycosylation step relative to glycosylation. 111 The inactivation for -glucosidases could be achieved to use either 2-deoxy-2,2-difluoroglycosides with extremely reactive leaving groups, such as the trinitrophenyl moiety,111 to allow accumulation of the intermediate; or another class of related compounds 5-fluoroglycosyl fluorides in which the location of the C5 fluorine is directly adjacent to the site of greatest charge development (O5) and lead to substantial destabilization, though the intermediates formed are often not very long-lived in this latter case.112 While these reagents work well for exo-glucosidases, their application to amylase  27 inhibition requires a minimum of a “disaccharide” structure. Such a 2,2-difluoro-maltosyl species was synthesized chemically and shown to inactivate HPA in a saturable fashion, but no three-dimensional structure of the complex could be determined.111, 113  A quite different approach was employed for the 5-fluoroglycosyl species in which HPA synthesized its own disaccharide inactivator “in crystallo” when incubated with a maltosyl donor and the C-5 epimeric 5-fluoro-idosyl fluoride as an acceptor (Figure 1-10). 114 Solution of the structure of this complex revealed a trisaccharide covalently attached to Asp197 in the active site of HPA. However, as expected for such a C5 epimeric species, the sugar in the -1 site adopted a distorted conformation, so the structure was of limited value for design of improved inhibitors. 114,115 Therefore, there is a need to develop a specific class of MBIs that targets HPA in a stoichiometry manner and can be used for X-ray study.  Figure 1-10: In situ elongation-trapping strategy for HPA using the example of 5-fluoro-idosyl fluoride (5FIdoF) as the acceptor and 4’-O-methyl--maltosyl fluoride (MeG2F) as the donor.   28 1.8 Aims of the thesis The search for safer, more specific and effective hypoglycemic agents has been an important area of investigation. Therefore, this thesis will focus on the search for inhibitors or inactivators of HPA, and will expand further to explore the surface interactions of HPA with its natural substrate starch granules. The first part of the thesis is to understand the role of each component of MbA in its interaction with HPA, and therefore to gain information about its potency and specificity. This information can be used to create new, derivative compounds that can act as lead compounds in the development of new drugs to treat Type II diabetes.  The second part is to synthesize selective MBIs of HPA and the approach can be extended to the design of specific activity-based probes for localization and quantitation of amylase in vivo when coupled to a fluorophore. The last part of the thesis will focus on the exploration of starch granule binding sites on the surface of HPA. The research can help to understand the interactions of HPA with starch granules in the starch digestion process. The working hypothesis of this thesis is that a strong synergistic program of mechanistic and inhibitor discovery studies targeting HPA, can lead to the development of novel therapeutics of improved efficacy for diseases, such as diabetes and obesity.     29     Chapter 2 Structural and Mechanistic Characterization of the Amylase Inhibitor Montbretin A and Its Substructures    30 2.1 Introduction to montbretin A  In order to search for new and more effective drugs that can target HPA specifically, a library of 50,000 plant natural product extracts was screened, and a potent and specific HPA inhibitor montbretin A (MbA in Figure 2-1) with nanomolar inhibition constant (Ki = 8.1 nM) was successfully discovered from the locally grown plant Crocosmia crocosmiiflora. 91 MbA is a perennial plant of the Iridaceae family native to South Africa, and originally was used as an antitumour agent in Japanese folk medicine.116 In the same screening library, MbA analogues (Figure 2-1) were also discovered that share a myricetin flavonol core, with glycosylation at the 3- and 4’- positions. A disaccharide containing a L-rhamnopyranosyl--1,4-xylopyranose is -linked at the 4’ position of the flavonol (B ring). D-glucopyranosyl--1,2-glucopyanosyl--1,2-rhamnopyranose is attached at C3 (C ring). The middle glucopyranosyl of this trisaccharide is joined to a 6-O-cinnamic ester. Montbretins B and C are derivatives in which the substituents at the 3 position of the cinnamic acid ester, are replaced by either hydrogen or methoxyl. 117 Indeed, the deoxygenation (Montbretin B) results in a 1000-fold reduction in affinity, corresponding to a strong (~ 4.2 kcal mol-1) hydrogen bond. An even greater reduction is caused by methylation of a caffeic acid hydroxyl (Montbretin C). 117 Similarly, Montbretin E, in which the 7-OH of myricetin (A ring) is glycosylated, binds well over 100-fold worse than MbA (unpublished results). Previously, many simple flavonols have been shown to inhibit amylases, although all of these bind with considerably weaker affinity and poorer specificity than does MbA. 84,85,86,87 Indeed, the myricetin core of MbA itself is a competitive inhibitor of HPA (Ki = 110 μM) and has been  31 shown to bind to the active site, while a caffeic acid derivative binds considerably more weakly (Ki = 1.3 mM) and non-competitively at locations remote from the active site of HPA.90, 117  Pre-clinical trials using MbA also show very promising results in controlling blood glucose levels in diabetic rats. Zucker diabetic fatty (ZDF) rats were treated with MbA at an average level of 7.5 mg/kg per day resulting in a significant decline (P<0.05) of the fasted plasma glucose final level to 12.4 mM, compared with the controls at 30.6 mM. 118 Moreover, no adverse effects or toxicity were noted in the rats, even after treatment for over 9 weeks.   Figure 2-1 The chemical structure of montbretin A (MbA). The specific points and agents of chemical and enzymatic cleavage used for fragmentation are shown in arrows, and the dotted rectangle delineates the structure of mini-MbA. The natural congener analogues Montbretin B, C and E, and their Ki values for inhibition of HPA are shown at the bottom.    32 Despite the understanding of MbA action at the physiological level, neither the origin of the affinity of MbA nor its high selectivity for HPA is understood. The unique specificity of MbA to HPA is especially important, given the non-specific nature of HPA inhibitors currently in medical use and the side effects that result as a possible consequence. To directly address these mechanistic issues and to guide further drug development, a better understanding of the contributions of each component of MbA is required. This may then allow development of smaller, more synthetically tractable analogues.  2.2 Aim of the project The aim of this project into the inhibition of HPA by MbA is to determine the basis of this inhibitor’s specificity and affinity by dissecting out the compound into smaller fragments and evaluating the contributions of each of the substructures and their binding modes to HPA kinetically. Together with crystallography work from our collaborator Dr. Brayer from UBC, the structural contribution of each component can be illustrated.  The information so gained will hence elucidate the interactions between MbA and HPA, and explain why small changes to the chemical structure in the montbretin family can alter their ability to inhibit HPA significantly. This will allow the development of a single, minimal unit that retains MbA’s powerful inhibitory action, and pave the way for further discovery of diabetes drugs.    33 2.3 Results and discussions 2.3.1 Generation of MbA substructures by chemical and enzymatic modifications In order to understand the role of each component of MbA in its interaction with HPA, a set of MbA substructures needs to be produced and tested. Chemical synthetic routes to synthesize the substructures are predicted to be inefficient, given the anticipated difficulties of multiple protection and purification steps, which are tedious and time-consuming and normally result in poor yields. To overcome the limitations, enzymatic approaches are more favorable in this study. Since we have access to large quantities of MbA (provided by Centre for Drug Research and Development, UBC), we propose to generate MbA fragments by a series of chemical or/and enzymatic degradations from the parent molecule, as shown in Figure 2-1. In this way, the stereo- and regio-chemistry of the untouched glycosidic bonds remains unchanged. In brief, treatment with sodium methoxide solution will cleave the caffeic ester linkage, yielding de-caffeoylated MbA, while treatment of MbA with β-glucosidase alone can remove a single glucose. Alternatively, initial treatment of MbA with rhamnosidase will cleave just the terminal rhamnose, while addition of xylosidase to that will remove both of the sugars attached on the right hand side. Further treatments using combinations of the degradation sequences described above will yield all the remaining analogues (for convention, the substructures generated will be named using the following abbreviations that C standing for caffeic acid, G for glucose, R for rhamnose and X for xylose). With this complete set of analogues in hand, the inhibition constants for each substructure can be determined. The information will help to identify the core structure that contributes to the strong inhibition of HPA.  34 A: Removal of caffeic acid to produce MbA-C.   The presence of a simple ester linkage to the caffeic acid allows cleavage under mild basic conditions that do not affect glycosides. Treatment of MbA with alkaline sodium methoxide at ambient temperature yielded MbA-C and methyl caffeate (Figure 2-2). 116 The purified compound, isolated as a yellow powder, was characterized by high resolution mass spectrometry (HRMS) and nuclear magnetic resonance (NMR).   Figure 2-2 Treatment of MbA with sodium methoxide to afford decaffeoylated product MbA-C.  B: Enzymatic hydrolyses using Agrobacterium sp. β-glucosidase. Removal of the terminal glucose (Glc2) of MbA to generate MbA-G can be achieved using a β-glucosidase from Agrobacterium sp. (Abg) belonging to the GH family 1. Abg is a retaining enzyme that catalyzes the hydrolysis of non-reducing terminal β-D-glucosides by a double- displacement mechanism.119 Incubation of MbA with Abg at room temperature in sodium phosphate buffer (50 mM, pH 6.8) containing 0.1% bovine serum albumin (BSA) to stabilize the enzyme resulted in clean conversion to MbA-G, and only the terminal glucose was cleaved in this case (Figure 2-3). The product MbA-G is also a β-linked glucose residue, however it bears a  35 large 6-O-caffeic ester moiety and would therefore not be processed by Abg to a smaller fragment. 120  A similar approach was applied to MbA-C. Since two -linked glucose moieties are present in the molecule, we might expect to see two products generated in a sequential manner. TLC analysis showed that the starting material was fully converted into MbA-CG (Figure 2-4). However, there was no indication of formation of MbA-CGG even after increasing the reaction time or rising the enzyme concentration to the millimolar range. Quite possibly, MbA-CG is not cleaved by Abg because the rhamnose moiety does not bind productively in the +1 site. Another possibility is that the glucose liberated from the first cleavage inhibits the second Abg cleavage reaction (Ki = 6.4 mM). 119 To test this, MbA-CG was purified to remove any glucose generated and subjected to digestion by Abg again, but no cleavage was observed, so inhibition by glucose was not the problem. Surprisingly, however, MbA-CG was found to act as a non-competitive inhibitor of Abg (Ki = 2.4 M), presumably binding in a non-productive mode, thus is not hydrolyzed by Abg. Similar results were obtained with another β-glucosidase, in this case from sweet almond.  Figure 2-3 Treatment of MbA with Abg to afford MbA-G.  36  Figure 2-4 Proposed scheme of MbA-C with treatment of Abg.  C: Enzymatic cleavage of rhamnose moieties by naringinase  Naringinase from Penicillium decumbens is a debittering enzyme used in the commercial production of grapefruit juice. The commercial enzyme is a multienzyme complex consisting of two enzymes α-L-rhamnosidase (EC and β-D-glucosidase (EC (Figure 2-5).121 α-Rhamnosidase converts naringin, the most bitter component in grapefruit to the less bitter substance prunin by cleavage of the α-1,2 bond between rhamnose and glucose. Subsequent hydrolysis by the β-D-glucosidase breaks the prunin into its aglycone naringenin and glucose (Figure 2-5). Initially, naringinase was expected to cleave the terminal rhamnose moiety from MbA to produce MbA-R (Figure 2-6). However, the inherent glucosidase activity is a big concern. The hydrolytic activity of naringinase has pH and temperature optima of 3.5-4.5 and 50-60 °C, respectively.122 Initial attempts to incubate commercial naringinase with MbA at pH 4.0, 60 °C directly yielded two products, MbA-RX and MbA-GRX, arising from removal of both rhamnose and xylose in one case and one more terminal glucose in the other case (Figure 2-7). The loss of xylose and glucose is mostly likely caused by the inherent glucosidase activity of naringinase. A quick test with the corresponding para-nitrophenyl glycoside substrates shows that naringinase   37  Figure 2-5 Hydrolysis of naringin into prunin, rhamnose, naringenin and glucose by naringinase containing α-L-rhamnosidase and β-D-glucosidase activities.    Figure 2-6 Treatment of MbA with heat-treated Naringinase to afford MbA-R.   Figure 2-7 Treatment of MbA with Naringinase to afford mixture of MbA-RX and MbA-GRX. β-D-glucosidase α-L-rhamnosidase   38  indeed has rhamnosidase, glucosidase and xylosidase activities. Since the products have similar retention time and are difficult to separate by HPLC following the standard protocol, it was thought better to control the glucosidase activity and derive the specific product.  While one approach would be to isolate the rhamnosidase from the enzyme complex, previous research suggested that neither ion-exchange nor gel-filtration chromatography can completely separate the rhamnosidase activity from the glucosidase. 121c Another way is to inactivate the glucosidase activity by denaturation at elevated temperature, and this could be achieved by incubating the enzyme at 60 °C for 2 hours. 123Activity tests showed that the glucosidase activity fully vanished, while the rhamnosidase activity remained 80% that of the untreated enzyme complex. Having successfully controlled the enzyme activity, without further purification, the partially inactivated enzyme was added directly to a solution of MbA at room temperature. The reaction generated mainly MbA-R with only trace amounts of side products, allowing HPLC sample purification. Similarly, the approach was applied to MbA-C to yield MbA-CR (Figure 2-8).   Figure 2-8 Treatment of MbA-C with heat-treated Naringinase to afford MbA-CR.  39  Figure 2-9 Treatment of MbA-R with xylosidase to afford MbA-RX.  D: Enzymatic Cleavage by β-xylosidase To remove the exposed xylose, MbA-R was treated with several different xylosidases (also known as xylan β-1,4-xylosidase, EC, including Aspergillus niger β-xylosidase and Bacillus halodurans β-xylosidase. Only the treatment with Bacillus halodurans β-xylosidase worked well to give the specific product MbA-RX (Figure 2-9). As a member of GH family 52, Bacillus halodurans β-xylosidase hydrolyzes β-1,4-D-xylans by removing successive D-xylose residues from the non-reducing termini. In this enzymatic reaction, MbA-R was incubated with the enzyme at room temperature to afford MbA-RX quantitatively.  Overall, by combination of four different chemical/enzymatic methods, nine MbA derivatives have been generated successfully (Table 2-1).  2.3.2 Kinetic analysis of the binding of MbA substructures to HPA To evaluate the contributions of each substructure of MbA to the inhibition of HPA, inhibition constants (Ki) of all the derivatives for HPA and modes of inhibition were determined by measuring the initial rates of amylase reactions at different inhibitor and substrate  40 concentrations. Observed reaction rates were fitted directly to various inhibition models using GraFit version 5.0.13 and Ki values were determined (Table 2-1). Competitive inhibition provided the best fit in each case.            Table 2-1 Color-coded schematic diagrams of intact  MbA and derivatives, with associated Ki values for binding to HPA, and IC50 values for both HPA and PPA. MbA components are indicated by ‘G’ for glucose, ‘C’ for ethyl caffeate, ‘R’ for rhamnose, ‘M’ formyricetin and ‘X’ for xylose.  Cartoon structure Ki of HPA (nM) IC50 of HPA (nM) IC50 of PPA (nM) MbA  8.1 ± 0.8 7.1 ± 0.2 - MbA-G         9.1 ± 0.8 13.6 ± 0.3 6.2 ± 0.6 MbA-R  21.3 ± 2.2 46.2 ± 5.8 42.6 ± 0.09 MbA-RX  42.4 ± 5.3 82.3 ± 7.9 26.7 ± 1.2 MbA-GR         79.3 ± 9.5 101.8 ± 4.9 22.8 ±1.8 MbA-GRX         93.3 ± 7.6 137.3 ± 5.2 68.6 ± 2.1 MbA-C  730 ± 90 1740 ± 130 4320 ± 20 MbA-CG         2,240 ± 200 3,340 ± 80 2,100 ± 10 MbA-CR  46,800 ± 3,900 72,800 ± 5,800 15,300± 400 MbA-CGR         128,000 ± 15,000 199,800 ± 9,600 66,000 ± 4,400 Errors quoted are the fit errors produced by GraFit (Erithacus Software, London, UK).   41 As can be seen from Table 2-1, removal of the terminal glucose had very little effect on affinity; Ki of MbA-G remains approximately the same as the parent compound MbA (Ki = 9.1 nM for MbA-G compared to Ki = 8.1 nM for MbA). Likewise, sequential removal of the terminal rhamnose and xylose each resulted in only an approximate 2-fold loss in affinity. Consequently, a drastically minimized structure MbA-GRX could be generated (by sequential cleavage of the terminal sugars using naringinase, xylosidase and Abg), with an affinity loss of only approximately ten-fold. This “mini-MbA” structure comprises the two aromatic moieties linked via a glucosyl rhamnose disaccharide. These results would suggest that the “peripheral” sugar moieties (terminal glucose, rhamnose and xylose) contribute very little towards binding, and that it is the mini-MbA core that engages in the key interactions. The importance of the caffeic ester moiety to binding of MbA is highlighted by the 100-fold loss in affinity resulting from its base-catalyzed removal (MbA-C). Subsequent removal of sugar moieties from this structure is substantially more deleterious than their removal from the parent MbA, highlighting the importance of the two aromatic moieties in conformationally organizing the core structure.  Together with the results from Tarling’s work, 91 we anticipate that both myricetin and caffeic acid are essential structural moieties for the tight binding of MbA to HPA and that the sugar residues may act as linkers to these two units, possibly providing additional hydrogen bonding interactions. To our delight, the results were quite consistent with crystallography work by the Brayer group, and this will be discussed in detail in Section 2.3.5.    42 2.3.3 Inhibition studies for porcine pancreatic α-amylase  Since porcine pancreatic α-amylase (PPA) and HPA have a high degree of homology with 83% identity (Chapter 1.5),37, 46 and share similar crystal structures with the difference that PPA crystallizes in a lattice with a much more “open” active site.48 It was of interest to see if the MbA derivatives bind to PPA in a similar way to HPA and if those derivatives have some selectivity over other enzymes within the same GH family.   Figure 2-10 Correlations of inhibition constants of MbA derivatives for HPA and PPA.  The half maximal inhibitory concentration (IC50), which indicates how much of the inhibitor is required to inhibit a given enzymatic activity by half, was used as a simple measure to evaluate the binding affinity for PPA. IC50 values of these MbA derivatives for HPA were also measured for comparison. When substrate concentration used is at its Km of HPA (Km = 3.6 mM for 2-chloro-4-nitrophenyl α-maltotrioside), IC50 is approximate double of the Ki value for the corresponding inhibitor. Table 2-1 summarizes the inhibition constants of MbA derivatives for 0123450 1 2 3 4 5 6log IC50 PPA log IC50 HPA Correlations of Inhibition constants of MbA derivatives for both HPA and PPA  MbA-GMbA-RMbA-RXMbA-GRMbA-GRXMbA-CMbA-CGMbA-CRMbA-CGRlinear 43 both HPA and PPA. A measure of the similarity of the binding affinities for the two enzymes is provided by the plot in Figure 2-10. The straight trend line with a slope of 0.98 and correlation coefficient of 0.95 provided a direct graphical comparison and correlation. Clearly, it shows that these MbA derivatives inhibit the two enzymes in a very similar way. We would therefore anticipate their interaction pattern with the two enzymes would be similar as well.  2.3.4 Contributions from each fragment moiety Change in the inhibition constants observed for HPA can be related to changes in the equilibrium binding free energy (for dissociation constants), which can be calculated for each fragment for their thermodynamic contribution to binding based on the equation ΔΔG°= -RTln(K1/K2), where ΔΔG° is the change in free energy, R is the gas constant (1.986 calK-1mol-1), T is room temperature (298K), and K1 and K2 are the inhibition constants of MbA substructures. The changes in ΔΔG° are calculated by comparing each pair of MbA substructures, and the average ΔΔG° for each component is tabulated in Table 2-2. It is shown evidently that, with the exception of rhamnose 2, the other sugar moieties have only a slight effect in binding to HPA. In addition, it is also revealed that myricetin as well as caffeic acid is a key binding determinant in HPA binding.  Table 2-2 Contributions of equilibrium free energy from each fragment Components Average ΔΔG° (kcal/mol) Xylose 0.4 Terminal glucose (Glc2) 0.6 Terminal rhamnose (Rha2) 1.1 Caffeic acid 3.6  44 2.3.5 Comparison of the kinetic results with the X-ray structural analysis of HPA-MbA complex The structure of the HPA-MbA complex was determined in the group of Dr. Brayer (UBC) by co-crystallization of MbA with HPA and the X-ray structure was resolved by Dr. Williams to a resolution up to 1.35 Å. 124 Analysis of the crystallographic data established a close correlation between the structure and inhibition kinetic data we collected. The well-defined electron density revealed that MbA binds in the active site adjacent to, and interacting with the catalytic residues Asp197 and Glu233 (Figure 2-11a). In this location, it largely fills the extended active site binding cleft, occluding catalytic residues from access to bulk solvent. MbA binds to the active site in a folded conformation that is characterized by intramolecular hydrophobic interactions between the planar A-ring of the myricetin group and the aromatic ring of the caffeic ester. This conformation then aligns the phenolic hydroxyl groups of the caffeic acid and myricetin for optimal hydrogen bonding with the catalytic carboxylic acids, as well as for additional water-mediated interactions with other key active site residues such as Glu300. These hydrogen bonds are quite important since loss of just one of these interactions with the caffeic acid causes almost 1000-fold weaker inhibition by the Montbretin B and C. Although these caffeic-myricetin interactions would appear to be the most important inhibitor interactions, there are a total of 9 direct hydrogen bonds and 9 water-mediated hydrogen bonds in the complex, as shown schematically in Figure 2-11b. However, most of the non-catalytic residue interactions would appear to be relatively weak, entirely consistent with our degradation studies, which suggest that the terminal glucose and the myricetin-appended xylose and rhamnose each contribute relatively little to binding.   45 Interestingly, ROESY NMR analysis of MbA alone by Dr. Andersen’s group (UBC) indicated that its folded conformation is for the most part pre-formed in solution. 124 This pre-organized conformation is very similar to that of the bound species (Figure 2-11), differing only in a 180° flip of the plane of the caffeate ring. The novel stacked mode of MbA was not foreseen, in the earlier myricetin-HPA crystal strcture, 90 and it contrasts sharply with the binding mode of the transition state analog inhibitor acarbose or even of substrates such as malto-oligosaccharides. Substrates and acarbose bind along the active site groove, with the protonated inhibitor nitrogen developing electrostatic interactions with the active site carboxylates. 45 By contrast, none of the sugars of MbA bind in a similar fashion, and the two stacked phenolics enter the active site orthogonally to the valienamine moiety of acarbose to interact with Asp197, Glu233 and Asp300. The unique binding mode of this core and the relative innocence of the linker suggest that myricetin and caffeic acid moieties are essential for the tight binding property of MbA for HPA, and may serve as a template to design much simpler and more synthetically accessible molecules that target amylase.   Figure 2-11 (a) Close-up view illustrating the interactions of MbA adjacent to HPA active site residues and the novel stacking conformation formed by the aromatic rings of MbA; (b) Schematic drawing of this hydrogen bonding network between HPA and MbA. Protein residues with more than one hydrogen bond to MbA have been shaded in the same color. Water molecules are shown as red spheres. Gray dashed lines labeled with distances in angstroms show the proximity of amino acid residues or waters to the bound ligand. b a  46 2.3.6 Crystal structure of mini-MbA and HPA complex  Since the degradation study yielded an essential core structure mini-MbA (MbA-GRX) with only 10-fold weaker inhibition against HPA (Ki = 93.3 nM) than the parent molecule MbA (Ki = 8.1 nM), it was useful to confirm that binding of mini-MbA to HPA is similar to that of MbA, and to validate the novel inhibitory motif of mini-MbA for HPA. A high resolution structure of mini-MbA in complex with HPA was solved by Dr. Sami Caner (unpublished data) and compared to the MbA-HPA complex structure.  A simulated annealing omit electron density map of the HPA/mini-MbA complex clearly shows the well-defined electron density present for the ligand in the active site (Figure 2-12). The refined final occupancy of mini-MbA in the active site of HPA was found to be 88 %. All hydrogen bond interactions found in the complex were plotted in Figure 2-13. Comparison between the structures of the mini-MbA/HPA and MbA/HPA-complexes indicated that the overall protein fold was not significantly changed, as the C-α root mean square deviation of 0.245 Å reflects. Notably, when comparing B-factor distributions within these two bound inhibitors, their aromatic moieties are less flexible than their carbohydrate-linker parts, also supporting the dispensability of the latter. Moreover, the four missing xylose interactions in mini-MbA do appear to lead to an increase in the conformational flexibility of myricetin’s terminal ring B, thus propagating thermal motion to ring C. Indeed, an overall increase of 32% in the B-value of the myricetin moiety was determined when compared to MbA. Thus while terminal carbohydrate ring interactions appear to be less significant in binding to MbA, they may increase affinity of the inhibitor to the active site by reducing thermal motion of its core.   47   Figure 2-12 (a) Simulated annealing omit map (green mesh) around the bound mini-MbA inhibitor (yellow stick) is shown at the 3σ level in the active site of HPA. (b) Final refined 2Fo-Fc electron density map (blue mesh) at the 1σ level around the bound ligand in the active site of HPA. The catalytic nucleophile Asp197 and the acid/base catalyst Glu233 are shown in green stick representation. Water molecules involved in ligand binding are shown as red spheres. Distances between atoms are indicated in angstroms and are visualized by gray dashed lines.     Figure 2-13 Structural (a) and corresponding schematic (b) representation of hydrogen bond interactions in the mini-MbA/HPA complexes. Mini-MbA is represented as yellow stick three-dimensional models or by black chemical two-dimensional-drawings. Distances are indicated in Angstroms and are visualized by gray dashed lines. Water molecules involved in protein-ligand interaction are represented by red spheres.  b a b a  48  Figure 2-14 Structural overlay of mini-MbA with MbA in the active site of HPA. Mini-MbA is shown in yellow and the corresponding catalytic triad residues (Asp197, Glu233, Asp300) are shown in green. MbA is shown in magenta and its corresponding catalytic triad in blue.  A detailed assessment indicates that the structural differences between mini-MbA and MbA upon HPA binding are small. Nonetheless differences are present in the carbohydrate-linker of the mini-MbA complex where Rha1 is shifted by an average of 1.15 Å when comparing ring atoms C3-C5 and the corresponding methyl and hydroxyl groups, to that of MbA. To a lesser extent, the ring atoms of mini-MbA’s Glc1 are shifted by an average of 0.47 Å. Additional positional changes can be observed in the 3, 4, 5-trihydroxyphenyl ring of myricetin (B ring), where ring atom positions vary by an average of 0.52 Å (Figure 2-14).  2.3.7 Inhibition specificity of MbA and mini-MbA Although previous studies with a range of largely commercially available glycosidases had demonstrated that MbA was a specific amylase inhibitor,117 no data were initially collected on the inhibition of intestinal α-glucosidases or -amylases found in gut bacteria in those early studies. Hence, the inhibition of glycosidases that are related to starch digestion by both MbA  49 and mini-MbA were investigated to determine whether the two inhibitors showed the requisite specificity between human enzymes. Human maltase-glucoamylase (MGAM), a brush border membrane enzyme that plays a role in the final steps of digestion of starch, was tested. MGAM consists of two GH 31 catalytic subunits, the N-terminal subunit (ntMGAM) and C-terminal subunit (ctMGAM), which share 40% sequence identity and similar digestion activities against -1,4-linked substrates. 125 Another intestinal -glucosidase subunit C-terminal sucrase-isomaltase (ctSI) was also tested, which works together with MGAM to hydrolyze the mixture of linear α-1,4 and branched α-1,6 oligosaccharide substrates that typically make up terminal starch digestion products. 126  As shown in Table 2-3, neither MGAM nor SI was inhibited by MbA, even at inhibitor concentrations up to 500 µM, thereby clearly showing the desired specificity for HPA. This may, in part, be due to the fact that these human brush border α-glucosidases are all members of  Table 2-3 Inhibition of HPA and other α-glycosidases by MbA and mini-MbA. Enzyme GH Family Ki of MbA Ki of Mini-MbA HPA 13 8.1 ± 0.8 nM 93.3 ± 7.6 nM Roseburia inulinivorans Amy A 13 N.I.a N.I.b Butyrivibrio fibrisolvens Amy B 13 1.6 ± 0.2 µM N.D.c Saccharomyces cerevisiase Yeast α-glucosidase 13 N.I. a IC50 = 446 ± 70 M (est. value) ntMGAM 31 N.I. a N.I. b ctMGAM 31 N.I. a N.I.b ctSI 31 N.I. a IC50 = 75.5 ± 2.5 µM a No inhibition at concentrations of 500 µM b No inhibition at concentrations of 100 µM c Not determined.   50 CAZy family GH31, which is distinct in sequence and structure from the GH13 family to which HPA belongs. 125,127 Furthermore, analysis of representative gut bacteria revealed that one of the gut bacterial amylases (Amy A) was not inhibited at all, and the other only weakly inhibited by MbA (Ki = 1.6 µM for Amy B). Since it was possible that the “terminal” sugars (Glc2, Rha2 and Xyl) might be providing the specificity for HPA over the α-glucosidases by sterically occluding the binding of the core structure containing the two aromatics, inhibition studies on these enzymes were also performed with mini-MbA. As shown in Table 2-3, no inhibition of Amy A or MGAM was observed, and although inhibition of ctSI could now be seen, it remained very weak, at an IC50 of 75.5 µM. The great specificity of MbA and mini-MbA for HPA compared to other downstream human α-glucosidases could make them potential drug candidates for diabetes treatment through selective amylase inhibition, in the hope of less side effects caused by the currently available diabetes drugs of α-glucosidase inhibitors (e.g. acarbose). 2.4 Conclusion In this project, a total of nine MbA derivatives were successfully generated by sequential chemical and enzymatic cleavage reactions. For each substructure generated, detailed kinetic analysis for HPA reveal a competitive inhibition mode. The loss of the caffeic acid weakens binding 100-fold while removal of the terminal sugar moieties (Glc2, Rha2 and Xyl) has relatively small effects. A minimum scaffold MbA-GRX consisting of simply myricein and caffeic acid linked via two sugars remains a tight binding inhibitor for HPA only 10-fold weaker than MbA. Therefore, the myricetin and caffeic acid are believed to be essential structural moieties for the tight binding of MbA to HPA while the sugar residues act as linkers to these two units, possibly providing additional hydrogen bonding interactions. All the derivatives inhibit HPA and  51 PPA at comparable inhibitor concentrations. It is thus reasonable to expect that their interaction patterns with the two enzymes are similar.  Since MbA is such a potent inhibitor, and is easily isolated from the corms of a readily grown plant (Crocosmia crocosmiiflora), it has potential as a new agent for controlling blood glucose levels in diabetics and obese patients. Notably, its complex structure should limit uptake from the gut, keeping it in the correct locale for amylase inhibition and minimizing side effects arising from systemic bioavailability. MbA has superior efficacy to that shown by acarbose in controlling blood sugar levels, and does so while only inhibiting HPA and not the intestinal α-glucosidases. These properties thus allow testing of the hypothesis that medical administration of amylase-specific inhibitors will result in fewer undesirable side effects compared to less specific α-glucosidase.  Structures of complexes of HPA with MbA and mini-MbA determined by our collaborator support our initial hypothesis. X-ray structural analysis suggests that the potent inhibition by MbA arises from a completely new mode of glycosidase inhibition in which the two phenolic moieties, myricetin and caffeic acid, are pre-organized via hydrophobic stacking to optimally hydrogen bond with the active site carboxylic acids Asp197 and Glu233. By contrast, the sugars appear to play very little role in binding. The central glucosyl rhamnose moiety serves primarily as a linker to suitably orient the two aromatics, and itself has very few direct interactions with the protein. Likewise, the other sugars contribute very little to binding, as was shown by the small affinity losses upon their removal.   52 The unique binding mode of this core and the relative innocence of the linker suggest that much simpler and synthetically accessible mimics might be possible, and may serve as the template for the design of a broad new class of glycosidase inhibitors. To further optimize affinity, such inhibitors could have additional specificity elements built into or onto the linker itself and/or the structures of the phenolic components could be readily varied. In addition, it seemed plausible to substitute this linker in favor of a non-glycosylated variant suitable for pharmaceutical synthesis, therefore we can open up strategies to further improve inhibitor efficacy with the aim to reduce the side effects of current oral anti-diabetic drug therapies.    53     Chapter 3 Chemoenzymatic Synthesis of Oligoglycosyl Epi-cyclophellitols Allows Mechanism-Based Inactivation and Structural Analysis of Human Pancreatic Alpha-Amylase    54 3.1 Introduction While interest has generally focused on reversible inhibitors of glycosidases, irreversible inhibitors of glycosidases have proved to be extremely important in providing structural and mechanistic insights. 104,105,115 As noted earlier in Section 1.7.4, a mechanism-based inhibitor (MBI) can form a covalent bond with the active site nucleophile of a targeted enzyme and trap the glycosyl-emzyme intermediate that is formed during the enzyme-catalyzed reaction. Therefore, MBIs can be used to assist in the detection and identification of functionally related proteins (and homologues) in complex biological systems, such as a cell extract, living cells or in animal models. 102,103,115 When attached to reporter moieties, such as a fluorescent tag or a bioorthogonal ligation handle, these types of compounds are referrd as activity-based probes (ABPs). Design of selective probes for carbohydrate-processing enzymes has been aided greatly by X-ray crystallography 128 and the development of small molecule inhibitors. 129,130 By far the most developed group of ABPs to date are those that target glycoside hydrolases. 131 Our group has investigated several fluoroglycoside MBIs for glycosidases by the incorporation of a fluorine atom at the C2- or C5- position of the pyranose ring, thus destablizing the two transition states, and when partnered with a good leaving group resulting in a buildup of glycosyl-enzyme intermediate. 107,108,109,111,113 The enzymes have been demonstrated to be effectively labelled by the inhibitors and to show time-dependent inactivation. These fluoroglycoside MBIs have also been proved effective in vivo, selectively inactivating the expected glycosidases in several organs tested in rats, including the brain. 132 Therefore fluoroglycosides can be used to represent suitable scaffolds for the development of ABPs. However, the intrinsic disadvantages associated with this type of MBIs are the inherent  55 instablilty of the glycosyl-enzyme adduct and the lower efficiency of the reaction, which limit their use for retaining -glycosidases. 113-114 Epoxides, another class of MBIs, have proven to be MBIs in a number of retaining glycosidases. 102,115,133 Inactivation is believed to follow the binding of the inactivator in a mode resembling that of the corresponding sugar substrate. Activation of the epoxide by protonation of the oxirane by the catalytic acid/base residue is followed by attack by the catalytic nucleophile residue, effectively opening the epoxide and forming a stable covalent complex (Figure 3-1a). in most cases, covalent inhibition of the enzyme takes place only when the epoxide binds properly in the enzyme active site to allow protonation of the epoxide oxygen atom by a catalytic acid/base residue in close proximity. 102 As compared to the fluoroglycosides, inhibition of glycosidases by epoxide-containing inhibitors is fast and in most cases truly irreversible. No reactivation of the enzyme is observed after dialysis. 134   Figure 3-1 (a) Mechanism of inactivation of retaining β-glucosidases by epoxides. Structure of (b) exocyclic epoxide; (c) conduritol B epoxide; (d) cyclophellitol and (e) epi-cyclophellitol.   56 In general, the epoxide-containing glycosidase inactivators can be classified into the exocyclic epoxides and the endocyclic epoxides. Exocyclic epoxides (Figure 3-1b) are the compounds that have the epoxide group attached to a sugar via an alkyl spacer, while endocyclic epoxides incorporate an epoxide within a cyclitol ring, which itself is a carbocyclic mimic of a sugar (Figure 3-1 c-e). Of these, the endocyclic epoxide have been studied the most extensively. To date, a wide range of these endocyclic epoxides has been synthesized to “match” the specificities of a range of glycosidases. One that has seen considerable use is conduritol B epoxide (CBE in Figure 3-1 c), and it has been shown to be a potent MBI for several retaining glycosidases in various organisms, and has been employed to identify the catalytic nucleophiles in several glycosidases. 102, 115 CBE has been particularly used to study mammalian retaining -glycosidases - human lysosomal β-glucocerebrosidase (GCase). The deficiency of GCase can cause a pathological condition known as Gaucher disease, which is a common lysosomal storage disorder that can involve defects in blood, bone, neurological and liver development.135 CBE is shown to be a selective inactivator of human GCase, while have no inhibitory effect for other known mammalian β-glucosidases. 136 This allows selective ablation of GCase activity in cell homogenates, thus permitting analysis of other β-glucosidases in tissues. 115,137   Furthermore, the symmetry of CBE allows it to inactivate retaining α-glycosidases, and applications include the identification of active site residues in mammalian sucrase-isomaltase and human lysosomal α-glycosidase. 138,139,140 In general, it was found that β-glycosidases are inactivated much more effectively by CBE than are α-glycosidases, likely because trans opening of the epoxide by the β-glycosidase leads to the preferred trans-diaxial product. Ring opening with the α-glycosidases necessarily yields the less favored trans-diequatorial product. 102 On the  57 other hand, a major drawback of the structural symmetry of CBE is that it can bind in some enzyme active sites in more than one mode, thereby resulting in covalent binding to residues other than the catalytic nucleophile. This was observed  by the labelling of alternate residues in both human GCase 141 and almond β-glucosidase.142 In both cases, the labeled residue was mistakenly identified as the catalytic nucleophile, which was later corrected with a more specific class of reagent, the 2-deoxy-2-fluoro glycosides. 143,144 Since the inherent symmetry of CBE allows it to function as an inactivator of both α and β-glycosidases, it had been proposed that the introduction of a C5 hydroxylmethyl group could result in a more selective and potentially a more potent inactivator of β-glucosidases. 145 Indeed, this proved to be the case with the asymmetric cyclitol epoxide - cyclophellitol (CP in Figure 3-1 d), which was initially isolated as a natural product from a mushroom, Phellinus sp. 146 Consistent with its structural resemblance to a β-D-glucose, CP was found to be a potent, mechanism-based irreversible inhibitor of retaining β-glucosidases compared to CBE. 147,146,148 Similarly, CP was also used as a small-molecule inhibitor of mammalian GCases in vivo to induce a Gaucher-like state in both cell culture and an animal study. 147, 149 Its structural analogue with an azide group attached to the C-6 position and extended with a BODIPY fluorophore moieties, makes it an excellent ABP for GCase and allows immediate fluorescence scanning on SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis).150,151 Although extensive studies have been reported on the synthesis and use of CP analogues, 134 in only one case has an X-ray structure of an enzyme inactivated with CP been reported, this beings the β-glucosidase from Thermotoga maritima. 152   58 No equivalent -glucosidase inactivator has yet been discovered as a natural product. A diastereoisomer of CP, 1,6-epi-cyclophellitol (ECP as in Figure 3-1e), which has the epoxide trans to the hydroxymethyl group, thus down relative to the ring plane, has been synthesized and shown to be a potent -glucosidase inactivator of yeast α-glucosidase and α-mannosidase from jack beans.149,153 However, no reports have been published on the inactivation of polysaccharide-degrading enzymes, such as human pancreatic α-amylase (HPA), by CP or ECP, largely because such enzymes typically require at least a “disaccharide” moiety in the active site for productive binding and the chemical synthesis of such oligosaccharide versions is non-trivial. This is highly true for HPA, since its active site spans from -3 to +2 subsites. 45 As HPA is a crucial enzyme in the starch degradation pathway, it would be very interesting to design and synthesize an oligoglycosyl-ECP MBI to selectively label HPA via the formation of a relatively stable glycosyl-enzyme intermediate. Such reagents could be useful in the structural characterization of the amylase active site, and could also be beneficial for the design as probes of amylase biology.  3.2 Aim of the project The aim of the project is to synthesize an oligoglycosyl mechanism-based inactivator, with an epi-cyclophellitol moiety at the reducing end (i.e. glycosyl ECP and maltosyl ECP), via chemoenzymatic approaches. Determination of its inhibition mode through kinetic and crystallography studies could pave the way for subsequent structural analysis and further development of ABPs.   59 3.3 Results and discussions 3.3.1 Chemoenzymatic synthesis of glucosyl epi-cyclophellitol   First, a simple disaccharide version of MBI with an ECP moiety, glucosyl epi-cyclophellitol (G-ECP) was synthesized. Since chemical synthetic approaches would be challenging, especially given the relatively limited availability of ECP, itself synthesized via an eight step route from D-xylose, 154 enzymatic coupling approaches were attempted. These are also not without challenges since NDP-sugar-dependent glycosyl transferases that carry out the desired coupling are not readily available and in any case tend to have high specificity for acceptors. A route involving transglycosylation using an appropriate retaining glycosidase or transglycosidase was thought to be unlikely to work since either ECP itself, or the G-ECP product, would probably inactivate the coupling enzyme. To avoid these problems an inverting glycoside phosphorylase was chosen, on which no covalent glycosyl enzyme intermediate was formed and thus inactivation was less probable. An excellent candidate for such an enzyme was the GH65 maltose phosphorylase from Lactobacillus acidophilus NCFM (LaMalP), which has been shown to carry out -glucosyl transfer from maltose to a range of sugar acceptors in phosphate buffer. 155 The use of high maltose concentrations and phosphate buffers generated -glucose-1-phosphate in solution, which served as the donor for transfer to the acceptor. This specific approach was not attractive in our case since it also required high concentrations of the acceptor, which was not feasible for ECP. An alternative approach was to use -glucosyl fluoride as the donor in the presence of phosphate, the higher reactivity of the fluoride allowing lower concentrations to be employed. 156 While this approach did work, it led to partial degradation of the epoxide product, presumably via HF-promoted ring-opening.   60 The best approach was to use -glucose-1-phosphate, prepared by chemical synthesis, 157 while performing the reaction at as low a pH as possible to favour the reaction equilibrium (Figure 3-2a). Carrying out the reaction on a 10 mg (1.2 mL) scale using 9 µM of LaMalP and monitoring by TLC and MS yielded G-ECP (40 %) after a 6-hour reaction. 1D 1H-NMR analysis of the purified product, however, revealed the presence of a major and a minor species, presumably linkage isomers. HPLC purification was therefore performed to separate the two “disaccharide” products and subsequent two-dimensional NMR analysis (COSY, HSQC and NOESY) identified the two products as the -1,3-linked and -1,4-linked isomers in a 4:1 ratio, with the -1,4 G-ECP unfortunately being the minor species. A structural rational for this may be found in the crystal structure of wild type LaMalP, whose loop 3 (His413-Glu421), connecting α3 and α4 of the (α/α)6-barrel forms a barrier that prevents binding of the malto-oligosaccharides longer than maltose in the catalytic pocket. 158 In other words, the specificity of the acceptor for LaMalP is strict towards maltose, and it was possible that the oxirane moiety of ECP was adversely affecting the binding mode in LaMalP, leading to the undesirable isomer. Therefore, the possibility that the alkene precursor of ECP (bearing a double bond across the same carbons as the epoxide) might also function as an acceptor, was explored with the hope of an improved yield of the desired -1,4 isomer (Figure 3-2b). If so then the epoxidation reaction of the double bond could be performed after the coupling, rather than before. Unfortunately, even though coupling proceeded well, the ratio of isomers was not improved so this route was abandoned.   61  Figure 3-2 (a) Formation of G-ECP catalyzed by LaMalP in sodium acetate buffer (50 mM, pH 4.0) at room temperature to give -1,3-linked and -1,4-linked products in a ratio of 4:1; (b) reaction of alkene precursor of ECP with βG1P; (c) illustration of the expected mechanism-based inhibition of G-ECP with HPA. The inactivator reacts with HPA and forms a covalent species that blocks access to the catalytic Asp197 in the active site.  3.3.2 Kinetic analysis of the inactivation of HPA by the two glucosyl epi-cyclophellitols Irreversible inhibitors display time-dependent inhibition and their potency therefore cannot be accurately characterised by IC50 or Ki values. This is because the amount of active enzyme at a given concentration of irreversible inhibitor will be different depending on how long the inhibitor is pre-incubated with the enzyme. Instead, time-dependent inactivation studies should be done, yielding a pseudo first-order rate constant of inactivation kobs. The observed rate constant kobs for loss of activity in the presence of inhibitor at concentration [I] can be described by the rapid formation of a non-covalent complex, having the apparent dissociation constant Ki, which gives the covalent glycosyl-enzyme complex with an inactivation rate constant ki. 102   62 Due to the challenges of synthesizing sufficient amounts of G-ECP, it was not possible to do full kinetic analysis of the inactivation process. Thus, time-dependent inactivation of HPA activity was measured only at a single concentration of inactivator. The inactivation was monitored by incubation of HPA in the presence of -1,4 G-ECP (3.6 mM) or -1,3 G-ECP (6 mM). Residual enzyme activity was determined by addition of an aliquot (10 l) of the inactivation mixture to a solution of 2-chloro-4-nitrophenyl α-maltotrioside (CNP-G3). kobs for inactivation is determined by direct fitting the residual activity of HPA versus time to an equation describing a single exponential decay with offset in GraFit 7.0.0.  Incubation of HPA in the presence of the -1,4 isomer of G-ECP resulted in time-dependent loss of activity with kobs of 0.001 min-1. This is therefore a relatively slow inactivator, probably because the binding of a “disaccharide” moiety at the -2 and -1 subsites of HPA active site is not optimal. By contrast no inactivation was seen with the -1,3 isomer of G-ECP for over 48 hours (Figure 3-3). This complete lack of inactivation by the 1,3-linked isomer simplified further crystallography work since the mixture of isomers could then be used, avoiding the challenging HPLC separation of relatively valuable materials.  The catalytic incompetence of the trapped covalent intermediate could be demonstrated by attempted reactivation experiments, in which G-ECP treated HPA was first freed of excess inactivator and then incubated in the same HPA buffer, and aliquots were assayed at time intervals. At 30 °C, no reactivation was observed, even after 24 hours. Since many glycosidases carry out transglycosylation more rapidly than hydrolysis, the effect of addition of 20 mM maltose was investigated on reactivation rates. No such turnover of the enzyme was observed  63 indicating the covalent linkage between the enzyme and the inactivator is quite stable, as could be expected, given the absence of a ring oxygen to assist cleavage.   Figure 3-3 Residual activity of HPA in the presence of () 3.6 mM α-1.4 G-ECP, and () 6 mM α-1,3 G-ECP in phosphate buffer (50 mM sodium phosphate, pH 7.0) containing sodium chloride (100 mM) at 30 °C.      3.3.3 Mass spectrometric analysis of the inactivated enzyme   A sample of HPA was first incubated with the mixture of G-ECPs, until no enzyme activity remained. The excess inactivators were removed by use of an Amicon 10 kDa nominal cut-off centrifugal concentrator, and the resulting mixture was subjected to ESI-MS analysis, along with an untreated sample of HPA. As can be seen in the mass spectra in Figure 3-4 the mass of the treated enzyme (m/z = 56,412) is greater than that of the untreated enzyme (m/z = 56,073) by the anticipated mass of the ECP (m/z = 338), indicating stoichiometric labelling of the enzyme with a single molecule of inactivator α-1.4 G-ECP. Time (hr)0 10 20 30 40 50Residual activity of HPA (%)020406080100 64  Figure 3-4 Mass spectra of HPA (above) and HPA treated with α-1,4 G-ECP (below).   3.3.4 Structural analysis of the inactivated enzyme.  A high resolution structure (1.23 Å) was acquired in Dr. Brayer’s group from the crystal of HPA soaked in G-ECP isomers, and only α-1,4 G-ECP was observed to bind to the active site of HPA, quite consistent with the kinetic results of the two isomers. As can be seen from the omit maps of Figure 3-5a, the structure of α-1,4 G-ECP bound to the active site of HPA revealed two slightly different binding modes. By omitting the ligand from the G-ECP/HPA model, a simulated annealing omit map was calculated to demonstrate covalent binding of each of the major and  65 minor structures of G-ECP. In these two different modes, G-ECP maintains almost identical overall positions in the active site, with the differences residing in two different conformations of the catalytic nucleophile Asp197. The major conformation of Asp197 is consistent with previously reported structures of HPA inhibitor complexes and appears to be stabilized by hydrogen bonding interactions with the C6-OH of ECP. 114,124,159 This hydroxyl group is in approximately the same position as the endocyclic oxygen of the glucose moiety in the normal glycosyl-enzyme intermediate, which has been seen in close proximity to the carbonyl oxygen of the catalytic nucleophile in several structures of GH13 glycosyl-enzymes 160,161,162 The minor conformation of Asp197 is an unusual one for the nucleophile, with the χ-1 angle relatively rotated by ~20° towards Glu233 thereby reducing the distance between the carboxylic groups of Asp197 and Glu233 from 5.4 Å to 4.2 Å. It appears to be stabilized by hydrogen bonding with the C2-OH of ECP.  The ECP moiety is clearly covalently linked to Asp197 with the carboxylate having attacked the epoxide from the “beta” face and opened the epoxide in a trans diequatorial fashion, as expected from the amylase catalytic mechanism. The cyclitol ring sits in a normal chair conformation and is engaged in multiple hydrogen bonding interactions, as seen Figure 3-5 d-e. Indeed, 10 direct and 3 water-mediated hydrogen bond interactions were observed in the G-ECP-HPA complex major conformation, and one less interaction in the minor conformation. Comparison of the structure of the G-ECP complex (major conformation) with those of the non-covalent inhibitor acarbose and the trapped covalent 4’-O-methylmaltosyl-5-fluoroidosyl-HPA (MeG2-5FIdo-HPA) intermediate is instructive (Figure 3-5 b-c). In both cases the distal glucosyl moieties, in the -2 binding subsite, overlay almost perfectly indicating that the enzyme is  66 recognizing the oligosaccharides in a similar fashion. Differences are found only in the -1 site, and are those anticipated on simple geometric considerations. Thus C1 (normal nomenclature of glucose) of G-ECP overlays well with C1 of the covalently bound fluoroglycoside inhibitor, since both are attached to the oxygen of Asp197. However, the equivalent carbon atom of the  Figure 3-5  (a) Simulated annealing omit map at the 3σ level (green mesh) around the bound G-ECP inactivator (yellow sticks) in the active site of HPA. Major and minor conformations are shown to demonstrate the two different binding modes. Superposition of the G-ECP ligand in the major conformation with the HPA inhibitors. (b) acarbose (PDB 1CPU) and (c) MeG2-5FIdo (PDB 3IJ7) in magenta stick representation. The catalytic triad residues (Asp197, Glu233 and Asp300) of HPA are shown in green with the corresponding overlaid residues shown in blue. Substrate binding subsites in the active site of HPA are labeled according to the convention used for glycosidases. The asterisk indicates the substrate cleavage site, which takes place between subsites -1 and +1 of HPA. Structural (d) and schematic (e) representations of hydrogen bond interactions in the G-ECP-HPA complex. Involved amino acids are highlighted in the schematic representation by colored spheres. Hydrogen bonds between protein residues and ligands are indicated by yellow or grey dashed lines. Distances in angstroms are denoted in (e). Water molecules involved in ligand and protein interactions are represented by red spheres and labeled with Roman numerals.   67 valienamine moiety of acarbose bound in the -1 subsite sits approximately 1.8 Å further away due to the inherent half chair conformation of valienamine’s. Accordingly, the ring plane of the valienamine moiety is tilted approximately 35° relative to that of the cyclitol of G-ECP. Interestingly the ring plane of the 5-fluoro-idosyl species is also tilted by 35° despite the covalent attachment, but in this case that is a consequence of the ring distortion occasioned by the inverted stereochemistry at C5. 3.3.5 Chemoenzymatic synthesis and characterization of a maltotriosyl epi-cyclophellitol Although -1,4 G-ECP was demonstrated to be a MBI covalently linked to the catalytic nucleophile of HPA, occupying the subsites -2 to -1 in the active site, its slow inactivation limits its use as an effective inactivator of HPA. Either a long incubation time with the enzyme or a very high concentration of the valuable inactivator is required to fully inactivate HPA. The successful demonstration of its inactivation did suggest that extended versions could be more potent inactivators by analogy with the faster turnover of longer substrates. 45 Presumably, the longer version of MBI can occupy additional sugar binding subsites in the enzyme’s catalytic pocket. Therefore, a “trisaccharide” version of amylase inactivator that incorporates an ECP moiety and occupies at least three subsites was synthesized and described below, as an extension of the previous work.  As discussed in Section 3.3.1, LaMalP is not suitable for synthesizing products longer than a disaccharide due to the structural constraints in the enzyme’s active site. Therefore, an alternative enzyme, cyclodextrin glucanotransferase (CGTase) of the GH13 family, was used for the chemoenzymatic synthesis of the trisaccharide version of MBI. CGTase is capable of  68 catalyzing the synthesis of non-reducing cyclodextrins from starch and other polysaccharides, and has been shown to be highly effective in transferring sugar moieties to saccharide analogues without suffering severe product inhibition. 163,164  An activated glycosyl fluoride donor, such as maltosyl or maltotriosyl fluoride donors is normally used as an elongating reagent, but the reaction can be complicated by self-transglycosylation of the fluoride donors to form a variety of elongated malto-oligosaccharyl fluorides. To simplify both the enzymatic synthesis and subsequent kinetic analysis of the elongated species, a blocked version of the fluoride donor in which the 4’-hydroxyl group at the non-reducing end was methylated, was used to synthesize the maltotriosyl version of MBI. Therefore, 4’-O-methylmaltosyl fluoride (MeG2F) was incubated with ECP at room temperature in the presence of CGTase. After the initial transglycosylation, the 4’-O-methyl group prevented further transglycosylation reactions, resulting in a homogeneous trisaccharide analogue 4’-O-methyl--maltosyl epi-cyclophellitol (MeG2-ECP) as shown in Figure 3-6. The reaction was monitored by TLC until ECP was fully consumed and CGTase was removed by ethanol precipitation. The crude product was analyzed by ESI-MS, and the expected mass was observed (detected m/z = 537.1732; expected with sodium adduct, m/z = 537.1795). The crude product was subsequently tested as an inactivator of HPA directly, and a time-dependent inactivation was observed, with kobs = 0.007 min-1 at an estimated inactivator concentration of 29 M (Figure 3-7). This indicates the trisaccharide formed contains the expected -1,4 linked product, since only the -1,4-linked reagent can inactivate HPA. The structure of the glycosyl-enzyme complex would presumably resemble that of G-ECP and those reported in literatures, 114, 163 in which the trisaccharide analogue spans the -3 to -1 subsites in the active site of HPA.  69 The residue in the -1 subsite is expected to be an ECP moiety, covalently attached to the HPA catalytic residue Asp197, whereas the -2 and -3 subsites are occupied by glucosyl moieties (Figure 3-8). Compared to G-ECP (kobs = 0.001 min-1 at [G-ECP] = 3.6 mM), MeG2-ECP is a much better inactivator (kobs = 0.007 min-1 at [MeG2-ECP] = 29 M). The higher inactivation efficiency observed than G-ECP being due to the occupation of the additional sugar binding sites at the active site region.    Figure 3-6 Chemoenzymaic synthesis of MeG2-ECP using MeG2F and ECP by CGTase in sodium phosphate buffer (50 mM, pH 7.0) containing sodium chloride (100 mM) at 30 C.       Figure 3-7 Residual activity of HPA in the presence of 29 M MeG2-ECP in phosphate buffer (50 mM sodium phosphate, pH 7.0) containing sodium chloride (100 mM) at 30 °C.      Time (hr)0 1 2 3 4 5Residual activity of HPA (%)5060708090100 70  Figure 3-8 Schematic drawing of the structure of G-ECP, MeG2-5FIdoF and the expected mode of MeG2-ECP binding to HPA active site. The binding subsites in the active site cleft of HPA have been identified according to the convention indicated in the bottom portion of this diagram (Brayer et al. 45). Also shown is the expected binding mode for a normal starch substrate, where hydrolysis would occur between subsites -1 and +1.  3.3.6 In situ elongation of MeG2-ECP Having shown that CGTase could elongate ECP and form a useful trisaccharide version of HPA inactivator, it was of interest to see if HPA itself could carry out the elongation and form its own inactivator in situ. This seems possible in light of the previous use of HPA to elongate inhibitors such as D-gluconohydroximino-1,5-lactam (GHIL), generating a remarkably better reversible inhibitor MeG2-GHIL, with the elongated inhibitor binding roughly 1000-fold tighter than the original GHIL. 163 Later, the same in situ strategy was applied to 5FIdoF, producing a trisaccharide inactivator MeG2-5FIdoF that indeed covalently bound to HPA and resulted in a time-dependent loss of enzyme activity. 114 The structure of the resulting glycosyl-enzyme   71  Figure 3-9 Proposed in situ elongation-trapping mechanism for HPA using ECP as the acceptor and MeG2F as the donor.  Figure 3-10 Mass spectra of HPA (above) and HPA treated with MeG2F and ECP (below).  72 complex was determined by carrying out the reaction “in crystallo”, and revealed extensive hydrogen-bonding interactions with the enzyme subsites.  Incubation of ECP and MeG2F (28 mol each) with HPA (1.6 nmol) at 30 °C did indeed result in a time-dependent inactivation with complete loss of enzyme activity in 4 hours. The proposed mechanism is shown in Figure 3-9. Evidence for formation of an elongated covalent MeG2-ECP-HPA glycosyl-enzyme complex was obtained by mass spectrometry. Both the wild type and the trapped HPA complex were analyzed by ESI-TOF and the individual masses were determined. ESl mass spectrometry showed that the wild type HPA was expressed in a series of forms with the parent mass being 56,063 Da (Figure 3-10). The masses differ by approximately 160 amu; this is the typical mass for a sugar residue. Since the enzyme was express in Pichia pastoris, which is known to glycosylate foreign proteins heterogeneously, 123,165 the mass observed suggests the presence of 1-2 sugar residues on the glycosylated enzyme forms, given the mass of nonglycosylated protein predicted to be 55,889 Da. In addition, the sample was kept in fridge for a certain time, and the other masses may possibly arise from glycation/glycosylation as well as some other unknown process.   Upon treatment of MeG2F and ECP, all three peaks of the wild type enzyme were shifted up by 514 ( 1) Da, which matches, within error, to the mass of predicated covalent attachment of MeG2-ECP (m/z = 514.5), indicating that all three species were being active form. It is also noteworthy that the mass spectra of the MeG2-ECP-HPA revealed that the inactivator covalently linked to HPA in a stoichiometry of 1:1 ratio, and essentially no free enzyme was detected, indicating nearly complete covalent labeling, consistent with the complete loss of  73 enzyme activity observed. The resultant trapped covalent intermediate MeG2-ECP-HPA was also found to be stable, with no reactivation of the enzyme seen when the complex was incubated with 20 mM maltose for a period of over 10 hours.  3.3.7 Comparisons of the two ECP-containing MBIs The synthesis of G-ECP by LaMalP unfortunately resulted in the formation of two isomers with the undesired -1,3 linked isomer being the major one, presumable because of preferred binding of ECP moiety in the “1,3” mode in the acceptor site. In order to gain better yields of the desired -1,4 product, it may be possible to mutate LaMalP to alter binding modes, or another type of maltose phosphorylase could be used. Our group is working on the screening of potential novel maltose phosphorylases from a metagenomic library, and the identified hits may prove more useful.   Since the subsites of HPA span through -3 to +2, monosaccharide or disaccharide analogues generally have very weak affinity for HPA’s active site. 163 Our result is quite in accordance that the disaccharide version of MBI, -1,4 G-ECP, inactivates HPA with a kobs of 0.001  min-1 at 3.6 mM, while the introduction of the extra glucose mimic at the non-reducing end makes MeG2-ECP a better inactivator, with a kobs = 0.007 min-1 at 0.029 mM. The higher inactivation efficiency observed for MeG2-ECP is apparently due to the additional occupation of the -3 subsite in the active site. The accelerated inactivation rate of HPA by MeG2-ECP could be arise from the activation energy of the reaction being lowered by the binding of an additional sugar moiety.   74 Similarly, Numao et al. 163 showed that MeG2-GHIL was a competitive inhibitor of HPA with an apparent dissociation constant Ki of 25 μM, 3 orders of magnitude lower than that for the monosaccharide inhibitor GHIL (Ki = 18 mM), clearly demonstrating that extension of GHIL generates a better inhibitor for HPA. This phenomenon had also been seen previously in the fact that, the relative rates of hydrolysis of HPA (kcat/Km) varied significantly for different oligosaccharide; for example, kcat/Km of HPA increased from 98 s-1 mM-1 to 830 s-1 mM-1 when the substrates were changed from maltosyl fluoride to maltotriosyl fluoride. 45 This 4-fold increase is a consequence of filling the -3 subsite, and points to substantial stabilization arising from the additional sugar residues. Several other studies have also shown that oligosaccharide-based molecules are better inhibitors for porcine pancreatic α-amylase, the sequence- homologe of HPA. 164,166,167  Further elongation to synthesize an oligosaccharide version of MBIs with more than three sugar residues is not likely to be fruitful, since maltotetraose was observed to bind in two productive modes with the preferred cleavage mode being when the -2 to +2 subsites of HPA were filled. 45 As such, a “tetrasaccharide” version, maltotetraosyl epi-cyclophellitol, would probably function as a substrate rather than an inactivator, and would be cleaved by the enzyme before inactivating. 3.4 Conclusion Although HPA is an important pharmaceutical target for controlling blood glucose levels, very few reports have been published on the development of novel MBIs of this enzyme, in large part because of the difficulties encountered in the synthesis of oligosaccharide-based  75 inactivators. In this project, a glucosyl moiety was -linked to epi-cyclophellitol using a chemo-enzymatic approach with a maltose phosphorylase, allowing the assembly of a new class of mechanism-based amylase inactivators that formed a covalent glycosyl-enzyme intermediate with HPA. Furthermore, a trisaccharide version was synthesized using a GH13 starch-processing enzyme, CGTase. By taking advantage of the intrinsic ability of HPA to catalyze transglycosyl-ation reactions, the inactivator could be elongated in situ by preincubating HPA with ECP and an activated glycosyl donor. Accurate measurement of the mass of the unmodified native enzyme and the inactivated enzymes gave the increase in mass caused by reaction with the inhibitors and showed the stoichiometry of the reactions. Both of the MBIs inactivate HPA in a time-dependent manner, with the trisaccharide version being a better inactivator. This is because the additional glucosyl moiety provides a substantial increase in affinity. Crystallographic analysis of the covalent complex of G-ECP with the enzyme confirmed the occupation of the inactivator in the -2 to -1 subsites, and the stoichiometric, covalent nature of the reaction. The precise structure of the elongated species MeG2-ECP and its interactions with HPA could be determined crystallographically by incubating HPA crystals with the ECP and MeG2F moieties in future studies and can provide further insights into the interactions within the enzyme active site. The binding of the inactivator would be expected to be similar to that of G-ECP, showing the labeling of the catalytic nucleophile Asp197 by the oxirane, and binding of the non-reducing end sugar moiety in the -3 subsite. Beyond the use in structural analysis of amylases, the selectivity and potency of the class of ECP-containing MBIs together with the stability of glucosyl-enzyme adduct, make it an excellent probe lead for ABPs for monitoring amylases in vivo. The X-ray structure reveals that, as would  76 be expected, a reporter group could be attached at O4’ or even O3’ of the distal (non-reducing end) sugar without encumbrance, allowing direct fluorescent monitoring of the enzyme in vivo or identification of novel amylases by “pull-down” experiment using bioorthogonal linkers. Furthermore, the strategy of the in situ elongation can be applied for high-throughput screening of potential probes for amylases. Different glucose analogues with either an azide, amine or thiol functional group etc at the C3’-, C4’- or C6’- position of the non-reducing end can be used as donor compounds though chemical syntheses of the donor molecules could be challenging. These donor molecules are then incubated with ECP in the presence of HPA in a 96-well plate, and only the suitable donors (recognized by HPA) can be transglycosylated to the ECP moiety and form a covalent glycosyl-enzyme intermediate with HPA. By mass spectrometry analysis and residual enzyme activity test as described in this Chapter, successful elongated probes can be identified. Those functional groups at the non-reducing end can be further reacted with other reporter group for fluorescent visualization or identification. The probe leads of interest can be synthesized by CGTase individually and used for amylase family. Moreover, if the acceptor is changed to an aziridine sugar analogue, allowing the attachment of different fluorescent tags or reporter groups at the reducing end, the probe library can be further expanded.      77     Chapter 4 Exploration of Starch Surface Binding Sites on Human Pancreatic Alpha-Amylase    78 4.1 Introduction Starch, the major component of human diet, is the most abundant storage polysaccharide in higher plants and is synthesized in the plastids, where it forms insoluble granules of different sizes, shapes and degrees of crystallinity. 168 Starch has two major components: the linear amylose composed largely of -1,4 linked glucose units and the branched amylopectin mainly formed by -1,4 linked glucosyl residues but also containing occasional branching -1,6 linkages. The overall structure of starch granule is a growth ring structure consisted of semi-crystalline and amorphous regions. 50,51,52 Due to its complex structure, a range of amylolytic enzymes is needed to degrade starch, including -amylases, -amylases, limit dextrinases, isoamylases and -glucosidases. 30 These represent endo-glycosidases and exo-glycosidases as well as -1,6 debranching activities. 30 By far, the majority of these starch-degrading enzymes fall into GH13, the -amylase family.  The hydrolysis of native starch granules by -amylases involves an enzyme in solution acting on a solid substrate. Thus, the surface area accessible to the enzyme and the efficiency of adsorption onto the surface are important parameters. 169 The targeting and binding of amylolytic enzymes to starch granules commonly depends either on non-catalytic binding sites on auxiliary domains that are independent folding domains known as carbohydrate binding modules (CBMs), or on surface binding sites (SBSs) situated on the catalytic domain. 67,170,171 Approximately 7% of all GHs in the CAZy database contain at least one CBM, 13 and more than one CBM can be present to increase their catalytic efficiency. Extensive research has been done to elucidate the biological roles of these autonomously folding and functioning modules, removal of which can reduce or even abolish binding and activity towards insoluble substrates.  79 67 SBSs are much more challenging to identify since they are found on the catalytic domain outside the enzyme active site, but are just as important as CBMs for the enzyme’s functions. 171,172,173 Most logically, SBSs can ensure the association of the enzyme with their polysaccharide substrate, 174,175,176 can guide the substrate chain into the active site 175,177,178 and can possibly assist in disruption of the natural structure of the solid substrate to facilitate catalysis.179,180,181 Allosteric regulation is also possible when a SBS is occupied by carbohydrate ligands that affect the properties of the active site conformationally.182,183,184 Moreover, SBSs can facilitate processive action (multiple attack mechanism), 175,185 in which an initial endo-attack is followed by the hydrolysis of additional glucosidic bonds before the enzyme-substrate complex dissociates. 186 Additional roles ascribed to SBSs include passing products on to neighbouring carbohydrate binding modules, and anchoring to the cell wall of the parent microorganism etc. 187,188 The majority of SBSs discovered so far have been deduced through crystal structures of enzymes in complex with substrates or substrate analogues.171 About half of the >45 enzymes with an identified SBS are from GH13 and a few are from GH77, both belonging to clan GH-H of the CAZy classification. Within the GH13 family, SBS-containing enzymes are found within 17 of the 36 subfamilies.172 The existence of SBSs was first described for porcine pancreatic -amylase (PPA) of GH13_24 by Loyter & Schramm, 189 who investigated the interactions of HPA with macromolecular limit dextrins of glycogen and also determined a binding stoichiometry of two for maltotriose by aid of equilibrium dialysis. Despite this discovery 50 years ago, the characterization of SBSs has not progressed far. 170 The crystal structure of PPA was not available until 1980,190 and crystallographic studies with substrate or substrate analogues  80 bound to different regions of the enzyme surface were only completed relatively recently. 79, 181, 191,192,193,194,195,196 The most thoroughly investigated SBSs in GH13 belong to barley α-amylase isozymes (AMY1 and AMY2) of GH13_6. 174,175,197,198,199,200 The first binding site in AMY2 was discovered by differential chemical modification of the enzyme in complex with -cyclodextrin and shown to contain two adjacent tryptophan residues (Trp278 Trp279). A second SBS on the C-terminal -sheet domain was identified in the AMY1 isozyme, which has approximately 80% sequence identity with AMY2. The two sites are referred to as the “starch granule binding site” and the “sugar tongs”. In addition, multiple SBSs were reported in several other GH13 enzymes. Human salivary -amylase of GH13_24 was shown to have three SBSs 178 and Neisseria polysaccharea amylosucrase of GH13_4 to have two SBSs, 201 suggesting that synergistic effects involving multiple residues might participate in starch binding to amylases.  The increasing awareness of SBSs in recent years has driven the establishment of a variety of analytical tools for the identification and characterization of the binding sites. The utilization of protein crystals with oligosaccharides (co-crystallization or soaking experiments) bound on the enzyme surface is an attractive approach for the identification of novel SBSs. However, the approach is likely to result in crystallographic artifacts due to either the high ligand concentrations employed or crystal lattice constraints. 173 Retardation affinity gel electrophoresis (AGE) turns out to be another relatively simple and informative screening method to identify novel SBSs, 202 and involves native gel electrophoresis done either in the presence or absence of soluble polysaccharides. Since enzyme migration is retarded by polysaccharide interactions, the migration of an interacting enzyme is slower when the polysaccharide is present than that of the control without polysaccharide. An apparent binding  81 constant or equilibrium dissociation constant Kd, for the enzyme complex formation can also be estimated. 203 In the case of insoluble polysaccharides, AGE is replaced by pull-down or co-precipitation experiments, which are referred to as adsorption assays. 178,204,205 Sequence comparison with closely related SBS-containing proteins may also be used for the identification of SBSs, but is not applicable to GH families, such as GH13 that have low sequence homology. 172 Next, to validate the capacity of the identified sites to bind to oligosaccharides, site-directed mutagenesis of residues within putative SBSs, followed by characterization of functional properties of the enzyme variants, forms the basis of a straightforward strategy. Several characterization methods, such as surface plasmon resonance (SPR) or isothermal titration calorimetry (ITC) can be used to measure the dissociation constants and associated thermodynamic parameters quantitatively. 175,202,206,207 NMR analysis is also able to resolve contributions from interactions at the active site and SBSs, 177 but has not been commonly implemented with the -glucan active enzymes, since they are typically too large for NMR analysis. A combination of mutational analysis with SPR and ITC is usually the most powerful strategy for the elucidation and characterization of the roles of key functional residues within SBSs. 173 A challenge often faced in the strategy above is that of indiscrimination between the active site and binding sites, as they both have binding affinity for polysaccharide substrates. It is therefore necessary to have a means of blocking the active site, and/or the SBSs. Mutagenesis of a single amino acid may not be sufficient, since binding to the active site occurs along the entire length of the substrate binding area and modification of one subsite may not have a great impact on binding with the others. In order to eliminate the effects of carbohydrate  82 interactions at the enzyme’s active site and discriminate between the active site and SBSs, it is better to inactivate the enzyme’s hydrolytic ability and block the catalytic pocket simultaneously. Thus, mechanism-based covalent inhibitors (MBIs) or other tight-binding specific inhibitors can be used to block polysaccharide binding at the active site. While this approach may work in principle for any carbohydrate-active enzymes, the availability of inhibitors is enzyme-dependent, thus the synthesis of suitable functionalized inhibitors is required. In reality, only very few suitable reactive carbohydrate MBIs are available. 102,115, 134,206  While the SBSs in other GH13 enzymes have been probed, no studies have been reported on how human pancreatic -amylase (HPA) binds to and degrades its natural substrate of starch granules. Crystallographic studies performed on HPA in which oligosaccharides or oligosaccharide precursors were soaked into crystals of HPA have in several cases identified binding modes for these ligands outside the active site (Brayer unpublished data). Likewise, recent crystallographic analyses of HPA utilizing 4’-O-methyl--maltosyl fluoride (MeG2F) and 5-fluoro--L-idosyl fluoride (5FIdoF) have discovered the sugar analogues bound on the HPA surface at remote locations that are at a distance from the active site. 114 Since these soaking experiments are typically performed using high ligand concentrations (typically 50 mM), it is not clear whether the binding modes so identified have any biochemical or physiological relevance. In a similar vein it is quite possible that other “true” binding modes have been missed if they are occluded by crystal lattice packing interactions. Nevertheless, these observed binding modes provide an excellent starting point for identification and validation of SBS on HPA. Figure 4-1 presents a composite of ligand binding modes observed at distinct surface sites on HPA at ten different regions. Site 1 is located in the active site cleft, so is the primary binding site, while   83  Figure 4-1 Snapshots of a variety of malto-oligosaccharide ligands in different positions on the HPA surface. Site 1 is located in the active site cleft, and sites 2-10 are remote from this region and possibly bind elements of the starch granules. These sites were identified from a series of experiments. They have not been seen to all be occupied at one time.   sites 2-10 are remote from this region and possibly bind elements of the starch granules. It is hypothesized that one or more SBSs in HPA may be involved in starch binding, and potential interactions at these sites may play a significant role in starch-binding activity.  4.2 Aim of the project As ingestion of starch by HPA affects post-prandial glucose levels, it is of interest to mechanistically probe how HPA degrades its complex, high molecular weight, and often insoluble substrates - starch granules. In this chapter, to elucidate the mode of HPA-starch granule binding and metabolism, potential SBSs of HPA are examined through structural mapping, and enzyme-ligand interactions, such as hydrogen bonding and hydrophobic interactions. The strategy outlined in Figure 4-2 is followed. The important amino acids on the  10  84  Figure 4-2 Flow chart of project design  85 enzyme surface are first identified and a series of single mutants are constructed in which key amino acid side chains in each SBS have been individually mutated to either alanine to remove key interactions, or to a bulky residue such as arginine to preclude binding at that site. Successfully expressed mutants are then assayed using the small molecule substrate 2-chloro-4-nitrophenyl α-maltotrioside (CNP-G3), which does not require SBSs for turnover. If the mutant shows wild type levels of activity, it is assumed that the enzyme is correctly folded. Kinetic parameters for hydrolysis of soluble starch by such mutants are subsequently determined, to access the effects of remote mutation on that class of polymeric substrate. Finally, the binding affinity of wild type and mutants to insoluble starch is assayed in the presence of an amylase inhibitor that blocks the active site. This should assess the contribution of each such SBS to starch granule degradation.  4.3 Results and discussions  4.3.1 Choice of mutants The composite X-ray structures of Figure 4-1 clearly present a subset of mono- and oligo-saccharide molecules directly involved in binding interactions with HPA. The overall conformation of proteins in the HPA/ligands complex is essentially the same as in the native structure. 37,114 All atoms of the model have well-defined density and the interactions between HPA and the ligands can be clearly identified in detail. Two types of ligands were identified, oligosaccharides such as maltotriose (G3) or maltohexaose (G6), and sugar analogues such as MeG2F or 5FIdoF. Around these bound sugar units, arginine, glutamic acid and aspartic acid polar-planar side chains, main-chain polar groups and the ammonium side chain of lysine  86 residues form the shell of the network of hydrogen bonds, which is considered as one of the major features of protein-carbohydrate interactions.208 Hydrophobic stacking interaction between the sugar and aromatic residues is also important. Such interaction occurs in all known surface binding sites of amylolytic enzymes, 178,192,209 and mutation of the aromatic residues that the -glucan curls around has resulted in decreases of activity in all cases. 175,176,184,199a,210, 211 Therefore, these interactions are elucidated in each SBS in my study as described below. Active site and SBS 2 The complex structure in Figure 4-3a revealed that a single maltopentaose G5 molecule occupied the active site from subsite -3 to subsite +2. At the extended SBS 2 that is 4.3 Å away from subsite +2, a maltotriose G3 molecule Glc516-518 was clearly observed. O1 of Glc516 was hydrogen bonded to ND2 of Asn152, and interactions of the bound ligand with neighbouring amino acids are summarized in Table 4-1. The enzyme surface at this position suggests the presence of a potential starch-binding site, and Asn152 could be an interesting amino acid to study for binding interaction.  SBS 3 At another location close to the calcium binding site, which is 12.4 Å away from the catalytic nucleophile Asp197, distinctive curved electron density corresponding to a G6 molecule was observed at SBS 3. G6 was anchored at the entrance of an open depression at the bottom of which the calcium ion is enmeshed (Figure 4-3b), lying at an average distance of 2.3 Å above the indole ring of aromatic residue Trp203, and forming numerous interactions with other amino acid residues through direct and water-mediated hydrogen bonds (Table 4-1). Interestingly, G6   87  a: Active site and SBS 2 b: SBS 3 c: SBS 4 d: SBS 5 e: SBS 6 f: SBS 7 T84R80 D197 N152 Glc516 W203  N152 H201 E369I367 Glc519 Asn250 W284Y276  88  Figure 4-3 Zoomed view of structures of HPA at (a) Active site and SBS 2; (b) SBS 3; (c) SBS 4; (d) SBS 5; (e) SBS 6; (f) SBS 7; (g) SBS 8; (h) SBS 9 and (i) SBS 10. Sugar analogues are shown in purple sticks. Residues of wild type HPA are shown in cyan and structure of HPA bound with ligands is represented in green. The calcium ion is represented as a red ball. (j) Structural overlay of barley α-amylase AMY1 (1HT6, orange) and HPA (1HNY, green) at domain C. The key amino acids are shown in sticks, denoted with b and h, respectively.  adopted a left-handed helix with average O4’---O4 distances of 4.45 Å, possibly arising from the ordered glucose residues of a longer and flexible oligosaccharide or -glucan chain segment that curls around Trp203. It is quite plausible that the association with cyclodextrin-like G6 reflects the binding by the enzyme of helical turns of natural substrates. This architecture is common for SBSs that bind to -1,4 linked D-glucose chains. 178,192,209 It is therefore expected  W388R387K322R389Q390g: SBS 8 h: SBS 9 W434S437 N412 i: SBS 10 j N461 hW434 bY380  89 Table 4-1 Observed binding interactions of sugar analogues with neighbouring amino acid residues at different sites on the HPA surface.  Surface Binding Site Ligands Contact HPA Residues Mutations Reasons 2 G3 N152, G239 N152W  End of active site groove 3 5FIdoF or G6 K140, D153, T155, R158, D159, W203, G205, D206, D246 W203A Hydrophobic interactions 4 G3 I367, E369 E369R H-bonding interactions 5 5FIdoF R80, T84, N220 R80 H-bonding interactions 6 5FIdoF N250 N250R H-bonding interactions 7 G3 K261, E272, Y276, N279, W284 Y276A, W284A Hydrophobic interactions 8 MeG2F K322, R387, W388, R389, Q390 W388 Hydrophobic interactions 9 G3 N412, W434, S437 W434A H-bonding interactions 10  5FIdoF N459, N461A, K466 N461A H-bonding interactions  that a mutation at Trp203 would result in a decrease in substrate binding affinity as reported in other studies. 175,176,184,199a,210,211 SBS 4  A G3 molecule G519-521 was found at SBS 4 (Figure 4-3c). O6 of Glc519 was bonded to a neighbouring water molecule, and O1 was hydrogen bonded to the main chain nitrogen of Glu369 and the oxygen of Ile367 at a distance of 2.2 and 2.8 Å, respectively. Since the major interactions are through hydrogen bonding to the main chains of the amino acids, mutation of either of the residues to a simple alanine should cause no difference in oligosaccharide binding. Therefore, mutation of the glutamic acid residue to a positively charged arginine is proposed in  90 the expectation that difference in surface charge could lead to changes in the binding affinity for oligosaccharides.  SBSs 5 and 6 Two 5FIdoF moieties were found bound to HPA, but at remote locations (SBSs 5 and 6 in Figure 4-3d-e) on the enzyme surface that are far too distant to influence the hydrolytic activity in the active site region. At SBS 5, O1 of 5FIdoF was hydrogen bonded to NH1 of Arg80, which underwent an 80 ° flip towards the ligand upon binding, compared to the wild type HPA structure, while O2 of the same sugar analogue was hydrogen bonded to OG1 of Thr84 at a distance of 2.2 Å. At SBS6, O6 of 5FIdoF was hydrogen bonded to ND2 of Asn250. At both sites, hydrogen bonding between ligand and the enzyme are the major contribution, and the amino acids involved (Arg80 and Asn250) are therefore noteworthy to explore. SBS 7 At SBSs 7, another G3 molecule made extensive hydrophobic contacts from stacking of the apolar faces of the sugars onto the flat surfaces of aromatic tryptophan and tyrosine rings (Tyr276 and Trp284) as shown in Figure 4-3f. A noticeable ligand-induced conformational change of ~ 0.6 Å was observed, involving Trp284. Though this site is situated at an average distance of 30 Å from the active site, it may participate in multivalent binding of polysaccharide substrates as reported for barley -amylase, whose “starch granule binding site” (Trp278 Trp279) is situated on the side of the catalytic ()8-barrel domain. 198 Therefore, this SBS 7 of HPA is considered as a potential binding region, and the key residues are investigated.   91 SBS 8 A single MeG2F was observed lying above the plane of the indole ring of residue Trp388, at an alternate remote SBS 8 located 20 Å from the active site (Figure 4-3g). The indole ring of the tryptophan was nearly parallel to the glucose ring plane. Hydrophobic stacking of the aromatic residues occurred involving the ring of Trp388 abutting the glucose ring of the ligand. The binding site architecture was completed by residues Gln390, Arg389, Arg387 and Lys322, which act like the pincers to dock the disaccharide analogue. The tryptophan ring, polar side chains and main chain polar groups act as a platform associated with the pincers serving to dock the ligand. Mutation of the aromatic tryptophan ring would be expected to eliminate the hydrophobic interaction as well as hydrogen bonds with the surrounding amino acids. SBSs 9 and 10 Both SBSs 9 and 10 were located in the non-catalytic domain C (Figure 4-3h-i), whose function is unknown in GH-H members and may have diverse roles in each enzyme. A G3 molecule at SBS 9 formed hydrogen bonds directly with neighbouring amino acids (Asn412, Trp434 and Ser437), while 5FIdoF at SBS 10 was hydrogen bonded to Asn461. Interestingly, structural overlay analysis revealed that Trp434 of HPA and the “sugar tongs” residue Tyr380 of barley -amylase 199b are actually located in the same region of the C-terminal five-stranded anti-parallel -sheet domain, with the side chains facing in different directions (Figure 4-3j).  Of all the SBSs discussed above, interactions between the natural sugar molecules G3 and the enzyme surface at SBSs 2, 3, 4, 7 and 9 appear to be more reliable for the identification of potential surface binding sites, since the bound ligands mimic the actual structure of the  92 substrate. On the other hand, SBSs observed with sugar analogues (5FIdoF and MeG2F) could be less likely candidates since binding of oligosaccharides is far more likely to signify a site that normally binds to a polymeric structure while difluorinated derivatives are far from a perfect mimic. However, knowing the possible crystal lattice restraints that might limit binding modes and given that this sugar contains a vicinal triol of the same configuration as glucose, these sites were still examined. Therefore, the key amino acids at the SBSs discussed above were mutated by site-directed mutagenesis to examine their contributions to the surface binding. As such, in the nine potential SBSs, the mutations of the amino acid residues that are involved in binding of starch to HPA are designed and summarized in Table 4-1. In each SBS, a mutation is designed in which the key amino acid has been individually mutated to either a small alanine, or to a bulky hydrophilic arginine residue, or to a large hydrophobic tryptophan. Since a previous study on human salivary -amylase showed the disorder of the protein structure when multiple amino acids at different SBSs were mutated, 178 only single mutations on the enzyme surface were initially planned. 4.3.2 Site-directed mutagenesis and mutant enzyme expression in Pichia HPA wild type was expressed in Pichia as a secreted protein as reported by Rydberg. 165 In order to introduce the single amino acid mutations, site-directed mutagenesis using overlap extension polymerase chain reactions (OE-PCRs) was performed (Figure 4-4). The overlap extension method, first described by Higuchi, 212 involves two flanking primers (primers A and D) on either end of the mutant sequence, and two internal primers (primers B and C) that contain the mismatched bases and hybridize at the mutation region. In the first round, two separate   93  Figure 4-4 Graphic representation of overlap extension polymerase chain reactions (OE-PCRs) Table 4-2 Oligonucleotide sequences of the mutagenic primers Primer Name Primer Sequence Mutants Generated A: 5’ AOX1 GACTGGTTCCAATTGACAAGC  D: 3’ AOX1 CCGTTTACCGTAAGACTGTAGGA  N152Wc GT GGA GAT ATC GAG AAT TAC TGG GAT GCT ACT CAG GTC AGA GA N152W N152Wb GTA ATT CTC GAT ATC TCC ACT TCC W203Ac CTTGATGCTTCCAAGCACATG GCTCCT GGA GAC ATA AAG GCA W203A W203Ab CAT GTG CTT GGA AGC ATC AAG E369Rc AT AAT GGA GTA ATT AAA AGA GTT ACT ATT AAT CCA GAC E369R E369Rb TTT AAT TAC TCC ATT ATT ATT  R80Ac GGA AAT GAA GAT GAA TTT GCT AAC ATG GTG ACT AGA TGT R80A R80Ab AAA TTC ATC TTC ATT TCC AGA N250Rc AGC AGT GAC TAC TTT GGT AGA GGC CGG GTG ACA GAA TTC N250R N250Rb ACC AAA GTA GTC ACT GCT Y276Ac AAT GGA GAG AAG ATG TCT GCT TTA AAG AAC TGG GGA G Y276A Y276Ab AGA CAT CTT CTC TCC ATT CCA C W284Ac AAG AAC TGG GGA GAA GGT GCT GGT TTC GTA CCT TCT GAC W284A W284Ab ACC TTC TCC CCA GTT CTT TAA W388Ac TGG GTC TGT GAA CAT CGA GCA CGC CAA ATA AGG AAC ATG W388A W388Ab TCG ATG TTC ACA GAC CCA GTC W434Ac GTT TTC AAC AAT GAT GAC GCT TCA TTT TCT TTA ACT TTG C W434A W434Ab GTC ATC ATT GTT GAA AAC AAT G N461Ac GGA GAT AAA ATT AAT GGC GCT TGC ACA GGC ATT AAA ATT N461A N461Ab GCC ATT AAT TTT ATC TCC AGA  94 PCRs are used to create the AB and CD oligos from the target template, respectively. Because primers B and C have terminal complementary ends, AB and CD will then hybridize in a subsequent “fusion” reaction in which the overlapping ends anneal, with the aid of primers A and D. The resulting product AD containing the mutated internal sequence can be obtained, and subcloned into a shuttle vector upon further purification.  With pPic9-HPA as the template, two universal primers 5’ AOX1 and 3’ AOX1 were used as primer A and D for all the mutants. The oligonucleotide sequences of primers for each mutant are shown in Table 4-2. The mutagenesis was successful for all the designed genes, and the resulting fragment AD was cloned individually into another Pichia vector pPic9K, which is a selective Pichia system allowing isolation of multiple copy integration of desired genes (occurs spontaneously in Pichia at a frequency between 1% and 10%) so as to increasing yields of expressed protein (Invitrogen catalog no. V175-20). As shown in Figure 4-5, pPic9k is identical to pPic9 except for the presence of the bacterial kanamycin gene (kan from Tn903) that confers resistance to Geneticin® in Pichia (Invitrogen catalog no. V175-20). The level of Geneticin® resistance roughly depends on the number of kanamycin genes integrated. A single copy of pPic9K integrated into the Pichia genome confers resistance to Geneticin® to a level of ~0.25 mg/mL, and multiple integrated copies of pPic9K can increase the Geneticin® resistance level up to 4 mg/mL. Because of the genetic linkage between the kanamycin gene and the gene of interest in pPic9K, one can infer that Geneticin® resistance clones contain multiple copies of desired gene. Secreted protein expression may increase because of a gene dosage effect. Thus, the presence of the kanamycin gene allows for in vivo  95 screening of transformants and can be used as a tool to detect pPic9K transformants that harbor multiple inserts of the gene. Figure 4-5c shows multiple insertion of expression cassette linked to the kan gene. However, this screening method requires tedious work and a long time, and there is no guarantee that multiple copies will actually increase the yields of expressed   Figure 4-5 The map of (a) pPIc9 vector and (b) pPic9K vector. (c) Multiple plasmid integration in Pichia. Detailed information can be found at a c b  96 protein. In other words, finding a colony with good screening results against Geneticin® does not necessarily mean that it will express more protein. Since mutation on the surface would not be expected to kill hydrolytic activity, an alternative screening method proceeding directly with substrate to see if any of the colonies overexpress the desired proteins. The principle for this screening is that a colony with higher hydrolytic activity and lower cell density can produce a relatively high yield of protein. First, the recombinant cells were grown on an agar plate until viable colonies appeared. The colonies were picked onto a 96-well plate by a Colony picker (Genetix, QP1x2xT), and were grown in 200 L cultures individually. Once the mutant proteins were secreted, the cell supernatant mixtures were directly assayed against CNP-G3. CNP-G3 is a small soluble substrate for α-amylases, and is hydrolyzed to release the chromophore 2-chloro-nitrophenol (CNP), which can be measured spectrophotometrically. The resulting absorbance increase per minute is directly related to the α-amylase activity in the sample. Based on the selection criteria of high ratio of (hydrolysis rate/cell optical density), ten colonies were selected per mutant. Because poor aeration in the 96-well plate may cause problems with protein expression, and lead to unreliable results, the selected colonies underwent a second round of screening, in which cells were grown in 5 mL growth medium. The same selection criteria applied, and the best one or two colonies were chosen for protein expression. Eight HPA mutant proteins were obtained at greater than 99% purity, with a cell culture yield of over 2 mg/L. Unfortunately, only fragmented versions of R80A and W388A were observed upon  97 purification, probably due to incorrect folding. All the mutants were analyzed subsequently by sequence analysis, SDS-PAGE and ESI-TOF, and their identities were confirmed. 4.3.3 Catalytic efficiencies of mutant proteins using CNP-G3 as a substrate With the purified mutant enzymes in hand, catalytic efficiencies for CNP-G3 hydrolysis were assayed. Since CNP-G3 binds to the active site only, its hydrolysis rate should not be affected by the surface mutation. Therefore, catalytic efficiencies of all the mutants would be expected to be similar if their structures are folding correctly.  As shown in Table 4-3, all the mutants and the wild type have similar kinetic parameters (kcat, Km, and kcat/Km values) for CNP-G3 hydrolysis, indicating that the structures of the mutants retain functional active sites, and the mutations do not cause severe changes in the binding manner of small substrates in the active site. Although W203A has the smallest kcat/Km among all the mutants, the three-fold difference from the wild type is not considered as very  Table 4-3 Catalytic efficiencies of HPA variants for hydrolysis of CNP-G3 and soluble starch.  Protein CNP-G3 hydrolysis Starch hydrolysis kcat x 103 (min-1) Km  (mM) kcat/Km x 102 (mM-1min-1) kcat x103  (s-1) Km (mg/mL) kcat/Km x103 ( HPA wt 1.25  0.03 3.22  0.27 3.89  0.03 2.07  0.04 0.66  0.05 3.12  0.24 N152W 0.88  0.02 2.21  0.16 3.99  0.03 1.03  0.04 0.57  0.07 1.83  0.25 W203A 0.52  0.01 3.96  0.20 1.32  0.01 0.55  0.02 1.14  0.08 0.48  0.03 E369R 0.63  0.04 3.75  0.71 1.68  0.03 1.68  0.03 0.55  0.09 3.07  0.49 N250R 0.46  0.03 2.66  0.54 1.74  0.04 0.69  0.01 0.30  0.06 2.31  0.55 Y276A 1.07  0.03 3.12  0.32 3.43  0.04 1.85  0.02 0.81  0.17 2.28  0.50 W284A 0.97  0.01 2.87  0.15 3.37  0.02 1.45  0.05 0.86  0.09 1.70  0.19 W434A 1.06  0.01 3.63  0.23 2.93  0.02 1.57  0.01 1.17  0.30 1.34  0.36 N461A 0.60  0.03 2.06  0.42 2.90  0.06 1.69  0.01 0.56  0.09 3.00  0.52 Errors were calculated from GraFit program directly.  98 as the wild type and were used to test their catalytic efficiencies for soluble starch and binding affinities for starch granules in the following study. 4.3.4 Catalytic efficiencies of mutant proteins using soluble starch as a substrate To study starch hydrolysis by the recombinant enzymes, soluble starch was used as the substrate because its amorphous structure is more digestible by -amylases than the crystalline form of starch granules. 65 Inhomogeneous starch polymer contains fractions that differ in susceptibility to digestion by amylases, and generally fractions with more amorphous material exposed on the surface and easy access to the enzymes are more rapidly digested (RD). In the hydrolysis reaction by HPA, RD-starch is converted to products quickly. As the reaction proceeds, the concentration of RD-starch decreases, and the rate of digestion slows. This is most probably because that slow diffusion of amylase into the polymer becomes a rate-limiting step. 213,214 Since the specific forces between the enzyme active site and accessible glycosidic linkages lead to the formation of a productive Michaelis-Menten complex, the catalytic efficiency for starch hydrolysis would be affected by the binding ability of amylases to RD-starch.  For homogenous substrates such as CNP-G3, Michaelis-Menten constant Km is the concentration of substrate at which the reaction rate is half of Vmax, while Km for inhomogeneous starch is defined as the concentration of available RD-starch that will support an initial rate of reaction at Vmax/2. Therefore, it would be expected that Km values for starch hydrolysis would be larger at a later stage of reaction than at an initial stage. For each amylase surface mutant protein, the ability of the enzyme variant to attach itself to the surface of the RD-starch would vary to a different extent if the mutated site is involved in binding interactions,  99 and a higher Km value may be resulted from a weaker binding of the mutant protein to the accessible glycosidic linkages of starch. In this study, the rates of soluble starch hydrolysis were determined by measuring the concentrations of reducing sugar produced per reaction time. Starch solution was first prepared at different concentrations (0-5 mg/mL) by gentle boiling of soluble potato starch (Sigma-Aldrich@ S-5651). Initial rates for each concentration of starch and each enzyme variant were measured within 5 minutes. The concentrations of reducing sugar generated were quantified using the colorimetric 3,5-dinitrosalicylic acid (DNS) assay,215 and were calibrated using a maltose standard curve. Plotting the concentrations as a function of time yielded a straight line, with the slope as the reaction rate. The reaction rates and the corresponding starch concentrations were further fit into a Michaelis-Menten equation. Kinetic parameters Km and kcat were calculated by GraFit 7.0.0, and are shown in Table 4-3.  The mutants and the wild type HPA exhibited similar Km values, except for W203A and W434A with the Km values being double that of wild type enzyme. The Trp203 residue is spatially located closer to subsite +2, where Tyr151 is present, and is sequentially closer to His201, which binds to an essential calcium ion, while the Trp434 residue is located in the region of non-catalytic domain C, relatively far from the active site. It is suggested that the mutation of the tryptophan residues may eliminate hydrophobic interactions, change the affinity of the enzyme to the saccharides and less digestible starch would be accessed by the enzyme, thereby resulting in a higher Km value.  100 The turnover number kcat, the number of substrate molecules converted to product per enzyme molecule per unit time, is another important kinetic parameter to examine. For amylase action on soluble starch, the measured kcat value will reach the maximum possible when there are no structural constraints to access of the susceptible substrate glycosidic linages. 216 In other words, kcat can be affected if the pathway where the substrate entered into or the product released from the active site is hindered. In this work, most of the HPA mutants have kcat values around 1.0 x 105 min-1, which are within the same range as PPA (1.0 to 1.6 x 105 min-1) when it was tested with different sources of starch. 217,218 The exceptions in kcat were observed for W203A and N250R at SBSs 3 and 6 respectively, both having lower kcat values than the rest of the mutants. More likely, explanation is that the tryptophan residue is involved in the initial global positioning of polymeric substrates through hydrophobic stacking interactions, and mutation weakens the binding affinity for starch and destroys oligosaccharide binding at the site. The disruption of the protein-ligand interaction thus affects the proper position of the substrate polymeric chain near the active site region, and slow diffusion of the substrate to the active site becomes a rate limiting step, thereby resulting in a low kcat value. The assumption is quite consistent with another study that Trp203 was proposed to serve as an anchorage region for the natural substrates of PPA on their way to the active site. 192,219 Asn250 is almost on the opposite side of the enzyme catalytic pocket and is about 15 Å away from the catalytic residue Asp197. Thus, its influence in the catalytic site would be expected to be less significant than that of Trp203, and that is consistent with the kcat values obtained. The specific constant kcat/Km can also be used to compare the relative reactivity of different mutants. As shown in Table 4-3, kcat/Km of the wild type enzyme and most of the mutants are  101 very similar, except for W203A, which has a six-fold lower kcat/Km than the wild type. It is known that aromatic residues such as tryptophan and tyrosine often contribute to sugar binding by the interaction between hydrogen of C-H in sugars and  electrons of the aromatic ring, the so-called CH- stacking interaction, with the predominant arrangement being a face-to-face geometry. 220,221,222 However, the interaction between Trp203 and substrate may not be due to the face-to-face CH- stacking, as can be calculated from the free energy differences between the wild type and the mutant enzymes using the equation below. 223  ∆∆G0 =  −RTLn(𝑘𝑐𝑎𝑡𝐾𝑚) 𝑚𝑢𝑡(𝑘𝑐𝑎𝑡𝐾𝑚) 𝑤𝑡 Where the subscripts mut and wt refer to the mutant and the wild type enzyme. Thus, the loss of binding energy change ∆∆G of W203A, caused by the tryptophan to alanine mutation,  corresponds to about 1.1 kcal/mol, while the stabilizing interaction energy of one complete CH- stacking effect between a fucose unit and a benzene ring with a face-to-face geometry was estimated at 3.0 kcal/mol. 224 Therefore, the loss of binding energy is too small to regard the interaction as complete CH- stacking. It is more likely that the interaction between Trp203 and the sugar moiety is due to partial CH- stacking, like the T-shaped edge-to-face geometry, or the intermolecular force, such as London dispersion force. A similar energy change was observed in -glucosidase of GH31 when a phenylalanine residue was mutated to alanine. 225  The kinetic results for the hydrolysis of the soluble starch shows that W203A has exceptional hydrolytic activities among all the tested enzyme, with the highest Km value and the lowest kcat and kcat/Km values. This suggests that Trp203 at SBS 3 may play an important role in the  102 hydrolysis of soluble starch. Such saccharide binding at remote SBSs, in addition to the active site, might be necessary for the enzyme to exhibit optimal hydrolytic activity. Besides, the aromatic residue may serve as the initial binding site of the enzyme through partial CH- stacking interaction and position the polymeric substrates near the active site for easy access.  4.3.5 Starch granule binding assay  AGE, ITC and SPR are well known approaches for assessing binding affinities of polysaccharides to proteins, but they are more applicable to the study of soluble polysaccharides. As an instead, adsorption assay is normally used for the study of insoluble polysaccharides, such as starch granules. In this study, adsorption of purified enzymes to starch granules was assayed by a procedure slightly modified from that previously reported. 226 Potato starch was first washed in deionized water and ethanol to remove any soluble small oligosaccharides, and was used to prepare a set of suspensions (0-200 mg/mL). The resulting starch suspensions were pre-chilled at 4 °C, followed by incubating with recombinant HPA mutant proteins (final concentration around 100 nM) on a rotor for 60 mins. The mixtures were then centrifuged to separate all the insoluble particles, including starch granules and amylase bound to starch. Unbound enzyme in the supernatant was quantified subsequently. The percentage of bound proteins (normalized with a protein control not incubated with starch granules) was plotted against the starch concentrations, and the data were fitted to a Langmuir adsorption isotherm (one-site binding model): 𝑏 =  𝐵max X [𝑆][𝑆]+𝐾d (where b is the bound enzyme fraction, [S] the starch granule concentration, Bmax the maximum fraction of enzyme bound and Kd the dissociation constant).   103 There are a few ways to measure protein concentration in the solution, either by absorbance at 280 nm, the Bradford assay or the bicinchoninic acid (BCA) assay. However, these methods have an intrinsic drawback in that the lowest detection limit is 0.5 g/mL, which is far higher than the free enzyme amount in this study. These methods are therefore not applicable. Another method using either CNP-G3 or amylose as substrates in order to assay the unbound enzyme concentration in the supernatant was not successful. The reaction progress curves had substantial downward concavity, suggesting that the enzyme catalyzed both the hydrolysis and transglycosylation. This yielded an elongated version of substrate, which bound more tightly. Further, when cleaved, no chromophore would be released. The same transglycosylation reaction was also reported in a previous study. 163 As such, the observed rate underestimated the true amount of total enzyme. Alternatively, NanoOrange, a merocyanine dye that produces a large increase in fluorescence quantum yield upon interaction with proteins, was used. The dye provides a very sensitive and easy assay for protein quantitation and can measure protein concentrations as low as 10 ng/mL. The free protein concentration in different starch suspensions could then be measured accurately and fit to the adsorption isotherm subsequently, from which Kd for each enzyme variant was calculated using GraFit 7.0.0. Values determined are shown in Table 4-4. The ability of HPA and its surface mutants to bind to starch granules was compared. The measured Kd was 5.85 mg/mL for HPA wild type, and very similar for most of the mutants, with the exception of Y276A and W284A, which have three-fold and five-fold higher Kd values respectively. This indicates Tyr276 and Trp284 may play a role on the binding to starch granules. However, the Kd values for all the enzyme surface mutants were much higher than that of the  104 Table 4-4 Apparent binding constants for HPA variants to the starch granules in the absence and presence of MbA. Surface binding site Proteins Starch granule binding Kd (mg/mL) Kd 1 (mg/mL) 1 HPA wt 5.85  1.25 0.38  0.06 D197N 0.96  0.11 0.96  0.14 2 N152W 5.96  0.87 0.42  0.07 3 W203A 7.22  1.18 1.18  014 4 E369R 5.32   0.86 0.63  0.01 6 N250R 6.07  1.42 0.74  0.23 7 Y276A 16.74  2.56 5.88  0.55 W284A 29.11   6.73 22.29  2.90 9 W434A 4.63  0.59 0.66  0.06 10 N461A 3.87  0.82 1.02  0.13 1 The values were measured after treatment with MbA Errors were calculated from GraFit program directly.  active site mutant D197N (0.96 mg/mL), which is odd. This perhaps suggests that the wild type and other mutants are significantly hydrolyzing the starch granules and liberating oligo-saccharides that compete for the surface binding sites, thereby raising the Kd values. Consistent with this idea, Kd of starch granules for barley -amylase was 0.47 mg/mL, and increased six-fold (2.85 mg/mL) when 0.5 mM -cyclodextrin, was added to both inhibit binding and suppress activity on starch granules. 180  To supress the hydrolytic activity of the enzymes while keeping their binding affinity for starch granules via any SBS, a tight-binding specific inhibitor monbretin A (MbA, Ki = 8.1 nM),91 discussed in detail in Chapter 2, was added to inhibit the enzyme activity. Each enzyme variant (20 M) was incubated with MbA (400 M) at room temperature for an hour prior to adding to starch suspensions for the adsorption assay. Since the concentration of MbA was 20 times more than that of the enzyme and 50,000 times higher than its Ki, over 99% of the enzyme was  105 supposed to be fully inhibited. The free enzyme concentration in each starch suspension was re-measured and the Kd values are shown in the right column of Table 4-4. As expected, almost all the enzyme variants showed a reduction in Kd, compared to the previously measured ones when no inhibitor was present, demonstrating the competition for binding sites from starch hydrolyzed oligosaccharides.  In order to ensure that no hydrolysis of starch occurred during the adsorption assay when MbA –treated enzyme was used, the concentrations of sugar reducing ends in the supernatant were measured, after incubation with either HPA alone or MbA-treated HPA. The data were compared with the starch controls that contained no enzymes, and are plotted in Figure 4-6. As shown by the purple line, without any MbA present, the concentration of sugar reducing ends far exceeded the level of the control sample at every starch concentration, indicating that HPA Figure 4-6 Reducing sugar production in the supernatant of starch suspension, when incubated with HPA in the presence/absence MbA at 4 C. -50501502503504500 10 20 30 40 50Reducing end (uM) Starch concentration (mg/mL) Starch aloneHPA + StarchHPA + MbA +Starch 106 could hydrolyze starch granules at low temperature and the oligosaccharides released would subsequently interfere with starch granules for enzyme surface binding, as seen from the Kd results. Addition of MbA greatly reduced the hydrolytic activity, as shown by the cyan line, leaving some uncertainty as to whether the observed Kd reflected real values. In order to address this, a mechanism-based amylase inhibitor that can covalently derivatize an active site residue and preclude any binding interaction at the active site simultaneously, is required. 4.3.6 Starch granule binding assay using a MBI as an active-site blocking reagent In chapter 3, I introduced two epoxide-based inactivators of HPA that would be useful. Of these, the most attractive one was the “in-situ-extended” “trisaccharide” 4’-O-methyl-α-maltosyl epi-cyclophellitol (MeG2-ECP) formed by HPA-catalyzed extension of epi-cyclophellitol (ECP) using 4’-O-methyl maltosyl fluoride (MeG2F). This was the fastest inactivator and the easiest to produce. Pre-treatment of HPA in this way prevented enzyme hydrolysis, thereby eliminating competition from small oligosaccharides, and fully block the active site (data not shown). This should allow accurate measurement of Kd for individual SBS, and any changes in Kd would be related to the binding affinity for starch granules only. The in situ elongation methodology was applied to other surface mutants, and Kd of starch granules was re-measured after the enzyme had been fully inactivated (Table 4-5). Kd of most of the MeG2-ECP-treated enzymes fell to the range previously reported for active site mutant D197N and essentially all the mutants exhibited higher values than the wild type, though in most cases not much higher. A difference in Kd was observed for each enzyme variant when pre-treated with either MbA or MeG2-ECP, indicating that the non-covalent HPA/MbA  107 complex may dissociate in the presence of starch granules. Consequently, free active enzyme is liberated to hydrolyze the glycosidic bonds of starch and generate oligosaccharides that can compete for the binding site and complicate the binding results. Therefore, the covalent nature of MeG2-ECP that reacts with HPA catalytic residue Asp 197 and forms a stable glycosyl-enzyme intermediate, allows a more accurate measurement of binding constants without any ambiguity. Significant changes in Kd were observed at SBS 7, at which residues Tyr276 and Trp284A are juxtaposed next to each other. Y276A has a 4-fold higher Kd of 1.69 mg/mL and W284A 10-fold higher Kd than that of wild type HPA. The increase in Kd could be due to the two aromatic residues appearing to form a rigid platform for sugar binding through hydrophobic interaction, thus the removal of any residue at this site leads to lower affinity for starch granules. The two mutants also showed some lower hydrolytic activity for soluble starch compared to that of wild type (Table 4-3), probably because SBS 7 mediates the attachment of the polysaccharide chains to the enzyme and binding at this site prior to the hydrolysis reaction is required. Interestingly, a structural overlay of HPA and barley -amylase AMY1 shows that SBS 7 is located relatively close to the “starch granule binding site” of barley -amylase AMY1, which contains two adjacent tryptophan residues Trp278 and Trp279 (Figure 4-7). The distinctive binding sites of two amylases are both situated on the side of the highly conserved catalytic ()8-barrel domain, suggesting SBS 7 of HPA is critical for efficient binding to starch granules. Moreover, the same aromatic residues have been proposed to bind to the polymeric substrate in other mammalian amylases. 178,192 The site harboring the aromatic residue Trp284 has been observed as a secondary saccharide-binding site in PPA, 192 while the double mutant Y276A/W284A of human salivary -amylase showed an intermediate reduction in the ability of granule starch  108 Table 4-5 Apparent binding constants for HPA variants to starch granules. Surface binding site Protein Starch granule binding Kd1 (mg/mL) Kd2 (mg/mL) 1 HPA wt 0.38  0.06 0.45  0.09 2 N152W 0.42  0.07 2.62  0.34 3 W203A 1.18  014 0.57  0.12 4 E369R 0.63  0.01 0.87  0.06 6 N250R 0.74  0.23 1.40  0.30 7 Y276A 5.88  0.55 1.69  0.46 W284A 22.29  2.90 4.68  0.74 9 W434A 0.66  0.06 0.42  0.08 10 N461A 1.02  0.13 1.54  0.11 1 The values were measured using MbA-treated enzymes.  2 The values were measured using MeG2-ECP-treated enzymes.  Errors were calculated from GraFit program directly  Figure 4-7 Structural overlap of barley -amylase AMY1 (1HT6, orange) and HPA (1HNY, green). Catalytic nucleophile Asp197 of HPA, aromatic amino acid residues at SBS 7 of HPA and in the “starch granule binding site” of AMY1 are represented in sticks, denoted with h and b.  bW279   hW284   hY276 bW278   hD197  109 binding, compared to the wild type.178 Our findings on the Y276A and W284A mutants are quite consistent with those studies, and the conspicuous location of aromatic residues and the potential stacking interactions may play a significant role in the starch granule binding. Similarities in the three-dimensional molecular structures and interaction patterns with carbohydrates in these eukaryoyic amylases, 37 suggests that SBS 7 with its aromatic residues is a contributing site for amylase binding to oligosaccharide substrates. Furthermore, it is worth mentioning that the location of two aromatic residues relative to the active site (~30 Å). Since the involvement of multiple surface sites distal to the active site in starch binding has been proposed in barley -amylases, 199a the existence of SBS 7 may support the multivalency binding of -amylases to long amylose chains, which assists in localization of the enzyme on starch granules to enhance processivity. 199a  Another increase in Kd was observed for N152W, which is 6-fold higher than that of the wild type. Asn152 at SBS 2 is located close to the +2 subsite of the enzyme catalytic pocket, and the mutation of the asparagine residue into a bulky hydrophobic tryptophan could cause a surface deformation and impede binding of polysaccharides at that location. Since SBS 7 is proposed as the starch binding site prior to hydrolysis reaction, and extension of the carbohydrate chain to the active site by passing through SBS 2 is possible. It is therefore reasonable to propose that the substrate may enter the active site from SBS 2, and occupy the subsites for subsequent hydrolysis reaction. The mutated tryptophan residue may block the access of the substrate into the active site, resulting in a reduced enzyme hydrolytic activity. The assumption is consistent with the nearly 2-fold lower hydrolytic activity of N152W for soluble starch than that of wild type (Table 4-3). On the other hand, Kd of MbA-treated N152W did not show any significant  110 change from the wild type enzyme. This odd difference could be explained by the different binding interactions of the two amylase inhibitors within the enzyme active site. As mentioned in Chapter 2 and 3, MbA binds to HPA in a folded conformation and largely fills the extended active site binding cleft, while MeG2-ECP binds to HPA covalently through subsites -1 to -3 (Figure 4-8). When N152W is treated with MbA, the deformed structure of the mutant protein may reorganize itself by induced-fit type of binding upon formation of the inhibitor-amylase complex, as proposed by Schneider. 100 The reorganized complex structure may highly mimic the conformation of the wild type HPA, with MbA at the active site near SBS 2, thereby minimizing the (proposed) effect of surface deformation caused by the mutation, and form possible CH- stacking interaction through tryptophan residue. As such, the binding interaction of N152W to the starch granules would resemble that of wild type HPA, with similar dissociation constants as observed in this study. While for the MeG2-ECP-treated enzyme, the presence of the inhibitor may not transform the protein structure or influence the binding interaction at SBS 2, and Kd can reflects the binding affinity to starch granules at that site.   Figure 4-8 Zoomed view of HPA active site with (a) MbA from PDB 4W93; (b) MeG2-ECP (modelled based on crystal structure of 5FIdoF-HPA (PDB 3IJ7) and G-ECP-HPA (PDB 5EMY)). Residue Asn152 is colored in magentas. b a  111 Since the catalytic efficiencies of W203A (at SBS 3) for CNP-G3 and soluble starch are the lowest among all the mutants (Table 4-3), it would be reasonable to expect its binding constant for starch granules would be much higher than the wild type. However, Kd of the mutant W203A does not differ much from the wild type enzyme, indicating that Trp203 may not being involved in starch granule binding. This suggests that HPA may have different binding mechanisms for soluble substrates and starch granules. More likely, Trp203 is proposed to assist in the initial positioning of the soluble substrates through stacking interactions, or guide the polymeric chain into the active site through a cooperative mechanism. Therefore, mutation of this residue causes a significant reduction in its catalytic efficiencies. By contrast, starch granules may bind to a distinctive site (SBS 7) other than SBS 3, and do not require Trp203 for surface binding.   E369R at SBS 4 exhibits little or no changes on either the binding affinity or the catalytic efficiencies. The crystal structure at this site (Figure 4-3c) showed that a maltotriose molecule G3 was hydrogen bound to the main chain nitrogen atom of Glu369. Apparently, mutation to the arginine residue did not cause a severe disruption to the strong hydrogen bonding interaction, and change of the surface charge did not preclude the substrate binding at the site as well. Therefore, SBS 4 is predicted not to play a significant role in the binding to starch granules and in the hydrolysis of soluble substrates. N250R at SBS 6 was shown to have a similar Kd value as the wild type HPA. As seen in X-ray structure (Figure 4-3e), 5FIdoF was bound at a location 3.2 Å away from Asn250, and hydrogen bonding to main chain ND2 through O6 of ligand. Mutation of the neutral asparagine residue into a positively charged arginine may not weaken the hydrogen bonding interaction between  112 the main chain nitrogen and the ligand, therefore little change in Kd was seen. Besides, the ligand-enzyme crystal structure was acquired at high concentration of 5FIdoF, therefore it would be no surprise to see some false positive binding on the protein surface. Moreover, the structure and conformation of the monosaccharide moiety does not truly mimic those of the natural substrate, and may also lead to some bias in the crystallographic work. Hence, SBS 6 is not considered as an important binding site. Though HPA shares a similar three-dimensional structure with barley -amylase AMY1 and their binding sites are close in the ()8-barrel domain as shown in Figure 4-7, their binding behavior for oligosaccharide substrates in domain C is quite distinct. The Tyr380 residue of barley -amylase AMY1 is located in the non-catalytic domain C and was confirmed as a key residue, forming numerous contacts with the oligosaccharides. Mutation of the residue clearly demonstrated its importance for adsorption to starch granules and binding with -cyclodextrin. 180 The Trp434 residue of HPA is in close proximity with Tyr380 in their three-dimensional structures (Figure 4-3j), but its alanine mutant did not show changes in either the hydrolytic activities or the binding affinity, indicating the site may not have any significant impact on substrate hydrolysis or granule binding. The difference in the binding affinities of the two enzymes could be explained by the unknown functions of domain C in GH-H members that appears to show some variability in sequence and length between amylases. 227 We can also argue that the difference in starch binding is subtle since the two aromatic residues are not the same. Further, the result also demonstrates that not all the aromatic residues on the protein surface are involved in CH-π stacking interaction with the ligand, and that structural comparison  113 to identify potential SBSs is not reliable for proteins within the same GH family. Therefore, Trp434 is considered as a false positive hit for granule binding.  Asn461 is also located at SBS 10 in domain C, and its mutant N461A has a Kd value slightly higher than that of the wild type, but exhibits no effect on the catalytic efficiencies of CNP-G3 or soluble starch. As shown in Section 4.3.1 (Figure 4-3h), this amino acid was chosen as a potential SBS because it was hydrogen bonded to a sugar analogue 5FIdoF in the crystal structure. The high concentration of the ligand used in the crystallization experiment, and the conformation differences between the ligand and the natural substrate make SBS 10 less attractive in this study.  To summarize, binding of HPA to starch granules appears to be mediated primarily by SBS 7, a classic polysaccharide binding site with interactions mediated via Tyr276 and Trp284. Interaction with soluble starch, by contrast, appear to occur primarily at SBS 3 and involves sequestration of helical turns of the starch polymer around a tryptophan axis (Trp203). Interestingly both sites are on the same face of the HPA molecule as the active site, thus a model in which HPA is anchored to the surface of the granule via SBS7 is easily envisaged.  Binding in such a way would allow loose chain ends to engage both with the active site groove, with which it is nicely aligned, as well as with SBS 3. Interactions at the other sites, with the possible exception of SBS 2, which is an extension of the active site in any case, were not very significant. Indeed these other sites are quite possibly artifacts of their initial method of identification: crystallographic soaking studies at high ligand concentrations.     114 4.3.7 Proposed HPA binding mechanism with starch granules  The processing of starch by enzymes is hindered by the compact helical structure of the polymeric substrate. The disruption of such a regular structure is usually accomplished by a cooperative mechanism through the utilization of CBMs and/or SBSs. SBSs have been demonstrated to play a variety of roles in polysaccharide converting enzymes. 171,172,173 However, how these binding sites work together to weaken the secondary structure of starch polymers, and whether they function cooperatively through the interaction with starch remains unknown. Of the number and distribution of SBSs discussed above, I hereby propose the binding mechanism of HPA to starch granules via SBSs. First, starch polymeric chains may wind around the protein surface, and interact with various amino acids though hydrogen bonding and CH- stacking interactions for optimal binding affinity. Such binding would highly damage the crystalline regions of the starch granules, and expose more accessible glycosidic linkages on the surface. Then SBSs (especially SBS 7 where Tyr276 and Trp284 are juxtaposed next to each other) may work to assist in holding the starch molecule, and help to guide the polymer chain into the active site for efficient hydrolysis. Optimal catalytic action occurs when -glucan residue enters the active site from the +2 site and occupies each of the five subsites. HPA catalyzes the hydrolysis of the -1,4 glycosidic bond and the product is expelled from subsite -3. The breaks in polysaccharide chains lead to further local destabilisation of crystallinity and favour continued amylolysis. After dissociation of the product, SBSs may work cooperatively to shuttle along the glycan chain so that the subsites are newly occupied, so to enhance the  115 processivity. Alternatively, the polymeric chain may leave and active site, and another section of chain can associate with HPA, with the whole binding and hydrolysis process repeating.  4.3.8 Amylase evolution Changes in human diet, such as starting to use milk products after the domestication of animals or eating increasing amounts of starch-rich foods, have resulted in new evolutionary adaptations. Perry et al. reported that the copy number of the human salivary -amylase gene (AMY1) was correlated positively with salivary amylase protein level and that individuals from populations with high-starch diets had, on average, more AMY1 gene copies than those with traditionally low-starch diets. 228 They also hypothesized that increased consumption of starch-rich foods during human evolutionary history gave individuals with multiple copies of the AMY1 gene an “advantage.” 228 This advantage comes from the increased ability to begin the chemical breakdown of energy-rich starches in the mouth. Even a small increase in the mean number of copies of the AMY1 gene in a population, can result in an increase in survival probability over time, especially in times when other food sources are scarce - a classic example of evolution by natural selection.  An extension of the study to include other primates revealed that chimpanzees possess fewer copies of AMY1 and lower salivary amylase protein levels than do humans. 228 This implied that the diet of chimpanzees probably contains less starch and therefore the greater copy number of AMY1 in humans may reflect a response to selective evolutionary pressure. Presumably, early hominids would have consumed considerable amounts of uncooked and minimally processed food that would have contained native starch granules, which are digested much  116 more slowly than the heat-treated material, and perhaps higher copy numbers of genes were required in order to produce the enzyme in quantity so that the relatively non-reactive starch was broken down sufficiently fast to contribute usefully to an individual’s energy requirement.  Belonging to the same multigene family, HPA encoded in AMY2 is the isozyme of human salivary amylase. The structures of the two genes demonstrated that AMY1 gene was derived from a pre-existing pancreatic amylase gene. 229 It is therefore reasonable to speculate that AMY2 is or has been subject to strong pressures of diet-related natural selection pressures, and evolves differently from AMY1. The selective advantage to account for this multiple independent evolution is proposed that since the enzymatic activities of pancreatic and salivary amylases are quite similar, and all mammalian species produce pancreatic amylase, there is no obvious advantage to duplication of the digestive activity per se in the two organs. 229 Other than the copy numbers of HPA gene, it is possible that the SBSs of HPA evolve to adapt to the high starch diet over time. Presumably HPA had zero or single binding site when it was first expressed; with increased consumption of starch-rich food, having more SBSs might mean that those individuals are able to digest more starch in a shorter time, and are able to obtain more energy from starch than individuals with less SBSs. Either the point mutations, small or large scale insertions or deletions of the gene can introduce new SBSs. This is quite possible since the human genome mutation rate is estimated to be ~1.1×10−8 per position per generation. 230 The hypothesis can be tested by measuring the binding affinities of surface mutants of other mammalian pancreatic amylases. Independent evolution of pancreatic amylase in human and other mammals can then be compared and variety of pancreatic amylases can be disclosed. This should provide considerable insight into mammalian evolutionary history.  117 4.4 Conclusion  A few potential SBSs were discovered in the previous crystallographic studies to bind to carbohydrate substrates/analogues on the surface of HPA. In the present study, these sites were analyzed by structural mapping and site-directed mutagenesis, and a total of eight surface mutant proteins were successfully expressed. All the mutant proteins were folded in a correct conformation and their catalytic efficiencies for soluble starch highlight the importance of Trp203 as a binding site for soluble starch that can guide the substrate to the active site.  The use of a covalent amylase inactivator that was assembled “in situ” by taking advantage of the intrinsic ability of HPA to catalyze transglycosylation reaction between ECP and an activated donor (MeG2F), solved the problem of small oligosaccharides competition arising from the hydrolysis reaction and allowed a more accurate measurement of binding constants. Indeed, this approach can discriminate the binding contributions of the SBSs from the active site, and is superior to all the studies reported today in that regard. Two distinctive sites (SBSs 2 and 7) are identified, located either near the subsite +2 or at a remote position 30 Å distant from the active site, to participate in the starch granule binding, while the rest of the SBSs are considered as crystallographic artifacts. It is reasonable to suggest that HPA may have two different mechanisms in binding to soluble substrates and starch granules. Trp203 may not participate in the latter case. Tyr276 and Trp284 are most probably required as a binding site prior to the hydrolysis reaction while Asn152 may work as a starch anchorage point where the chain enters into the active site.   118 This thorough analysis of the complexation mode of HPA with starch granules shows the ability of the enzyme to offer multiple carbohydrate binding sites. Hydrophobic interactions between sugars and aromatic residues appears to be an important feature for surface binding. Individual binding sites in the multivalent protein-carbohydrate interactions are of moderate affinity and a rather small difference between comparable binding events possibly elicits differences in mobilization of starch granules. These results reflect an elusive structure/function relationship of the surface site. Analysis among the GH13 members also signifies the SBS 7 to be an important starch binding site, but the exact biological significance and the mechanisms behind its actions remain to be elucidated and subject to further investigation.  Further studies may include analysis of double alanine mutants in which two of the amino acids Asn152, Tyr 276 and Trp284 are mutated pairwise, and fully characterized to reveal if the binding effect of multiple sites is additive. Also, HPA and the mutant proteins could be labelled with a fluorophore using a commercial dye or an ECP-containing probe. Upon incubation with starch granules, their interaction could be visualized directly using a confocal laser scanning microscopy. The intensity of the image could then be correlated to the binding affinities of the enzyme variants to the starch granules. Moreover, studies on the surface sites of other amylases will contribute to our understanding of the roles of SBSs in polysaccharide processing enzymes.  To our knowledge, this is the first study of the starch binding abilities of HPA mediated by amino acid residues. Consistent with many other studies, SBSs represent a functional feature confined to a well-defined non-catalytic structural site. This study has provided new details on  119 the individual roles played by the SBSs of HPA, and the information is important not only for clarifying the mode of action of HPA, but also improving the understanding of the growing number of other enzymes identified as possessing SBSs.     120     Chapter 5 Experimental Procedures    121 5.1 Synthesis 5.1.1 General materials for synthesis All reagents were purchased from commercial suppliers (Sigma-Aldrich®, or Thermol Fisher Scientific®) unless otherwise specified, and were used without purification. MbA was a generous gift from CDRD (Center for Drug Research and Development). Methanol was dried over magnesium. Deionized water was prepared using a Millipore-Direct QTM 5 Ultrapure Water System. Agrobacterium sp. β-glucosidase and Bacillus halodurans β-xylosidase, were previously expressed in house. Naringinase from Penicillium decumbens was purchased from Sigma-Aldrich. Lactobacillus acidophilus NCFM maltose phosphorylase was expressed as previously reported. 158 Cyclodextrin glycosyl transferase (CGTase) was a generous gift from Amano Enzyme Inc (Nishiki, Japan). Epi-cyclophellitol (ECP) and 4’-O-methyl -maltosyl fluoride (MeG2F) was synthesized by Hong-Ming Chen as described previously. 154, 231 Analytical thin layer chromatography (TLC) was performed (on Merck pre-coated 0.2 mm aluminum-backed sheets of Silica Gel 60F254) and TLC plates were visualized using UV light (254 nm) and by exposure to 10% sulfuric acid in ethanol or 10% ammonium molybdate in 2 M H2SO4, followed by charring. High Performance Liquid Chromatography (HPLC) was performed on Agilent 1260 Infinity Bio-inert Quaternary LC using a Agilent Eclipse XDB-C18 column (9.4 x 250 mm, 5 µm) or a PrepHT XDB-C18 column (21.2 x 100 mm, 5 µm) at room temperature using a gradient from 5% to 60% aqueous acetonitrile (ACN) over 20 minutes with a flow rate of 3 mL/min to afford different MbA fragments. The effluent was monitored by UV absorption at 254 nm and peak fractions were collected according to the chromatogram. 1H-NMR and 13C- 122 NMR spectra were acquired on a Bruker 300 MHz, 400 MHz Inverse, 400 MHz Direct or 600 MHz spectrometer. Chemical shifts are reported in parts per million (ppm ) relative to the relevant residual solvent peak (eg. CD3OD and D2O). Low resolution mass spectra were acquired on a Waters ZQ Mass Detector equipped with ESCI ion source and Waters 2695 HPLC for sample delivery, while high resolution mass spectra were submitted to the University of British Columbia mass spectrometry facility for analysis on a Waters/Micromass LCT with electrospray ionization and time of flight detection (ESI-TOF) in either positive or negative mode.  5.1.2 General enzymatic degradation procedures for MbA degradation General procedure for enzymatic hydrolysis by Agrobacterium sp. β-glucosidase (Abg). Deglucosylation was catalyzed by Abg in a solution of sodium phosphate buffer (50 mM, pH 6.8) containing 0.1% bovine serum albumin (BSA). The reaction mixture was incubated at 37°C until completion. Enzyme was precipitated by the addition of methanol solution, and the solution was filtered (Millex® Millipore Express® PES Membrane filter unit, 0.22 µM). The reaction mixture was evaporated under vacuum and redissolved in H2O. The resulting solution was purified by HPLC to yield the deglucosylated product. General procedure for enzymatic cleavage by naringinase (rhamnosidase). Naringinase was incubated in 50 mM acetate buffer (pH 6.4) at 60°C for 2 hours, and was added to the reactant solution (50 mM acetate buffer, pH 5.6). Upon consumption of the starting material, enzyme was precipitated by MeOH and removed by Millex® filter. The reaction mixture was evaporated and redissolved in H2O, then purified by HPLC.   123 General procedure for enzymatic cleavage by xylosidase. Previously de-rhamnosylated glycoside was dissolved in buffer (50 mM sodium phosphate, 50 mM sodium chloride, pH 6.8), and B. halodurans β-xylosidase was added and left to react at room temperature overnight. Enzyme was precipitated by MeOH and removed by Millex® filter. The reaction mixture was evaporated, redissolved in H2O and purified by HPLC. Synthesis and characterizations of MbA fragments  MbA-G (1)  Removal of glucose was performed according to the general procedure for enzyme hydrolysis by Abg. MbA (49 mg) was dissolved in 400 µL sodium phosphate buffer (50 mM, pH6.8) with 100 µL of 8 mg/mL Abg. The reaction mixture was incubated at 37°C with addition of a further 100 µL of enzyme after 24 hours. Upon completion and removal of Abg, the resulting solution was purified by HPLC to yield 1 as a light yellow powder in 70% yield.  1H NMR (400 MHz, methanol-d4):  7.40 (d, J = 15.8 Hz, 1H), 6.91 (s, 2H), 6.87 (s, 1H), 6.76 (d, J = 8.2 Hz, 1H), 6.68 (d, J = 8.1 Hz, 1H), 6.23  (s, 1H), 6.19 (s, 1H), 6.05 (d, J = 15.8 Hz, 1H), 5.77 (s, 1H), 4.83 (m, 3H), 4.53 (d, J = 2.1 Hz, 1H), 4.49 (d, J = 7.8 Hz, 1H), 4.35 (d, J = 4 Hz, 1H), 4.14 -  124 4.23 (m, 2H), 3.94 - 4.00 (m, 1H), 3.83 - 3.86 (m, 1H), 3.79 (br.s., 1H), 3.70 - 3.76 (m, 2H), 3.55 - 3.64 (m, 2H), 3.47 - 3.54 (m, 1H), 3.42 – 3.45 (m, 2H), 3.36 - 3.40 (m, 2H), 3.29 - 3.31 (m, 2H), 1.30 (d, J = 6.2 Hz, 3H), 1.08 (d, J = 6.0 Hz, 3H)  13C NMR (75 MHz, methanol-d4): δ 179.76, 169.08, 165.97, 163.25, 158.50, 158.10, 151.89 (2C), 149.49, 147.13, 146.83, 137.32, 136.28, 128.82, 127.73, 123.03, 116.40, 115.15, 114.87, 109.94 (2C), 107.41, 107.30, 106.26, 102.71, 100.05, 99.97, 95.00, 83.34, 77.77, 75.76, 75.50, 75.30, 75.24, 75.01, 74.02, 73.59, 72.51, 72.21, 72.07, 71.86, 71.72, 70.24, 64.65, 64.36, 18.09, 17.94 HRMS (m/z): [M]+ calcd. for C47H54O28Na, 1089.2699; found, 1089.2687 MbA-R (2)  Removal of rhamnose was carried out according to the general procedure for enzymatic cleavage by naringinase. Naringinase (30 µL of 5.6 mg/mL) was incubated at 60°C for 2 hours to inactivate the glucosidase activity, then was added to 4 mL 17 mM MbA solution in sodium acetate buffer (50 mM, pH 5.6).  After one day the reaction mixture was purified by HPLC to give 2 in 73% yield.  125 1H NMR (400 MHz, methanol-d4):  7.32 (d, J = 15.8 Hz, 1H), 6.84 (s, 2H), 6.78 (d, J = 1.8 Hz, 1H), 6.67 (dd, J = 8.2, 2.1 Hz, 1H), 6.58 (d, J = 8.2 Hz, 1H), 6.15 (d, J = 2.1 Hz, 1H), 6.10 (d, J = 1.8 Hz, 1H), 5.97 (d, J = 16.1 Hz, 1H), 5.59 (s, 1H), 4.74 - 4.76 (m, 2H), 4.50 (m, 3H), 4.40 (dd, J = 11.9, 2.13 Hz, 1H), 4.23 (d, J = 2.7 Hz, 1H), 4.12 (dd, J = 11.9, 5.5 Hz, 1H), 3.85 – 3.94 (m, 2H), 3.64 - 3.69 (m, 2H), 3.47 - 3.55 (m, 4H), 3.37 - 3.42 (m, 3H), 3.27 - 3.35 (m, 4H), 3.18 - 3.20 (m, 1H), 1.00 (d, J = 6.1 Hz, 3H)  13C NMR (100 MHz, methanol-d4):  179.73, 169.05, 166.26, 163.24, 158.52, 157.95, 151.91 (2C), 149.54, 147.20, 146.75, 137.43, 136.22, 128.77, 127.69, 123.07, 116.41, 115.12, 114.83, 109.89 (2C), 107.34, 106.60, 106.17, 105.59, 102.67, 100.16, 95.12, 84.72, 84.38, 79.23, 77.91, 77.35, 77.26, 76.00, 75.46, 74.80, 74.08, 72.30, 71.96, 71.43, 71.04, 70.86, 67.46, 64.21, 62.48, 17.84 HRMS (m/z): [M]+ calcd. for C47H54O29Na, 1105.2648; found, 1105.2677 MbA-RX (3)   126 Xylose cleavage followed the general procedure for enzymatic cleavage by xylosidase. 2 (43.7 mg) was dissolved in 6 ml buffer (50 mM sodium phosphate, 50 mM sodium chloride, pH 6.8) and 15 µL of 3.3 mg/mL β-xylosidase was added and incubated at room temperature overnight. Upon HPLC purification 3 was produced in 40% yield. 1H NMR (300 MHz, methanol-d4): 7.34 (d, J = 16.0 Hz, 1H), 6.92 (s, 2H), 6.83 (dd, J = 6.7, 1.8 Hz, 1H), 6.71 (dd, J = 8.3, 1.9 Hz, 1H), 6.59 (d, J = 8.0 Hz, 1H), 6.18 (d, J = 2.3 Hz, 1H), 6.09 (d, J = 2.1 Hz, 1H), 6.02 (d, J = 15.8 Hz, 1H), 5.40 (s, 1H), 4.52 (d, J = 7.5 Hz, 1H), 4.36 (d, J = 7.8 Hz, 1H), 4.16 - 4.20 (m, 3H), 3.85 – 3.97 (m, 3H), 3.76 (dd, J = 9.6, 3.7 Hz, 1H), 3.67 (dd, J = 12.5, 4.5 Hz, 1H), 3.50 - 3.55 (m, 2H), 3.37-3.39 (m, 1H), 3.27 - 3.34 (m, 4H), 1.01 (d, J = 6.2 Hz, 3H) 13C NMR (75 MHz, methanol-d4): 179.92, 169.11, 165.86, 163.29, 158.72, 158.51, 149.58 (2C), 147.21, 147.07, 146.81, 137.89, 137.06, 127.76, 123.17, 122.11, 116.42, 115.04, 114.92, 109.51 (2C), 106.63, 106.02, 105.69, 102.79, 99.91, 94.89, 84.84, 84.58, 79.29, 77.98, 77.83, 76.06, 75.42, 74.22, 72.41, 71.95, 71.05, 70.93, 63.89, 62.48, 17.79 HRMS (m/z): [M]+ calcd.  for C42H45O25, 949.2250; found, 949.2272 MbA-GR (4)   127 4 was synthesized by the general procedure for enzyme hydrolysis by Abg. 2 (13.5 mg) was dissolved in 5 mL sodium phosphate buffer (50 mM, pH 6.8) with 80 µL of 8 mg/mL Abg at 37°C. Upon completion of reaction the mixture was purified by HPLC to yield 4 as a light yellow powder in 35% yield. 1H NMR (300 MHz, methanol-d4):  7.31 (d, J = 16.0 Hz, 1H), 6.82 (s, 2H), 6.78 (d, J = 2.1 Hz, 1H), 6.67 (dd, J = 9.0, 1.8 Hz, 1H), 6.58 (d, J = 8.2 Hz, 1H), 6.15 (d, J = 2.3 Hz, 1H), 6.10 (d, J = 2.1 Hz, 1H), 5.95 (d, J = 16.0 Hz, 1H), 5.67 (s, 1H), 4.72 – 4.74 (m, 1H), 4.37 - 4.45 (m, 2H), 4.25 (dd, J = 3.5, 1.3 Hz, 1H), 4.10 (dd, J = 12.0, 5.6 Hz, 1H), 3.91 (dd, J = 11.4, 5.0 Hz, 1H), 3.75 (dd, J = 9.7, 3.5 Hz, 1H), 3.45 - 3.54 (m, 3H), 3.34 - 3.40 (m, 3H), 3.27 - 3.31 (m, 3H), 3.18 – 3.21 (m, 1H), 0.99 (d, J = 6.2 Hz, 3H) 13C NMR (75 MHz, methanol-d4): 179.85, 169.09, 166.05, 163.34, 158.57, 158.15, 151.34 (2C), 149.55, 147.17, 146.79, 137.41, 136.31, 128.85, 127.77, 123.04, 116.41, 115.16, 114.89, 109.92 (2C), 107.47, 107.38, 106.29, 102.81, 100.05, 95.00, 83.63, 77.80, 77.30, 75.56, 75.32, 74.84, 73.63, 72.11, 71.92, 71.88, 70.89, 67.50, 64.37, 17.94 HRMS (m/z): [M]+ calcd. for C41H44O24Na, 943.2120; found, 943.2134  128 Mini-MbA: MbA-GRX (5)  5 was prepared according to the general procedure for enzyme hydrolysis by Abg. 3 (13.5 mg) was dissolved in 3 mL sodium phosphate buffer (50 mM, pH 6.8) with 100 µL of 8 mg/mL Abg and incubated at 37°C. After 24 hours another 100 µL of enzyme was added, then upon completion, the enzyme was precipitated with MeOH and removed by filtration. The reaction mixture was evaporated, redissolved in H2O and purified by HPLC to yield 5 in 30% yield.  5 could be generated in a one-pot reaction by incubating MbA (10 mg) with Abg (30 µL, 8 mg/mL), naringinase (20 µL, 5.6 mg/mL ) and xylosidase (35 µL, 3.3 mg/mL) in 1.5 mL buffer (50 mM sodium phosphate, 50 mM sodium chloride, pH 6.8) at ambient temperature. The enzymes were precipitated with MeOH and removed by filtration. The reaction mixture was evaporated, redissolved in H2O and purified by HPLC to yield 5 in 40% yield. 1H NMR (300 MHz, methanol-d4):  7.41 (d, J = 16.0 Hz, 1H), 6.98 (s, 2H), 6.90 (d, J = 2.1 Hz, 1H), 6.79 (dd, J = 8.2, 1.8 Hz, 1H), 6.68 (d, J = 8.2 Hz, 1H), 6.24 (d, J = 2.1 Hz, 1H), 6.17 (d, J = 2.1 Hz, 1H), 6.09 (d, J = 16.0 Hz, 1H), 5.60 (s, 1H), 4.33 - 4.40 (m, 3H), 4.20 – 4.24 (m, 2H), 3.85 - 3.95 (m, 3H), 3.34 - 3.43 (m, 3H), 1.09 (d, J = 6.2 Hz, 3H)  129 13C NMR (75 MHz, methanol-d4): 179.91, 169.37, 165.92, 163.20, 159.07, 159.01, 151.93 (2C), 149.25, 147.05, 146.81, 137.05, 136.96, 127.80, 127.65, 123.14, 116.67, 115.10, 114.93, 111.79, 109.62 (2C), 105.60, 104.22, 102.40, 99.94, 94.89, 85.00, 83.88, 75.50, 75.35, 75.30, 71.60, 71.40, 60.85, 17.87  HRMS (m/z): [M]- calcd. for C36H35O20, 787.1722; found, 787.1712 MbA-C (6)  MbA (20 mg) was dissolved in 10 mL dry MeOH under N2 in a round bottom flask, and 0.1 mL of sodium methoxide solution was added dropwise. The reaction was worked up after two hours by addition of Amberlite IR-120 (H+) to the solution until it was slightly acidic (pH ~5). The polymer was filtered under reduced pressure, and the solvent was evaporated under vacuum. The mixture was redissolved in water, extracted with ether, then the aqueous layer was concentrated and purified by HPLC to yield 6 as a light yellow powder in 50% yield.  130 1H NMR (300 MHz, methanol-d4):  6.93 (s, 2H), 6.39 (d, J = 2.1 Hz, 1H), 6.23 (d, J = 2.1 Hz, 1H), 5.62 (s, 1H), 4.79 - 4.82 (m, 2H), 4.58 (d, J = 7.8 Hz, 1H), 4.49 (d, J = 7.8 Hz, 1H), 4.20 (d, J = 3.0 Hz, 1H), 4.14 (dd, J = 11.5, 4.7 Hz, 1H), 3.92 - 3.97 (m, 2H), 3.63 - 3.77 (m, 7H), 3.52 - 3.60 (m, 4H), 3.35 - 3.44 (m, 6H), 3.24 – 3.26 (m, 3H), 1.28 (d, J = 6.2 Hz, 3H), 0.99 (d, J = 6.2 Hz, 3H) 13C NMR (75 MHz, methanol-d4): 179.70, 166.24, 163.28, 158.76, 158.68, 152.06 (2C), 137.20, 136.30, 128.62, 109.87 (2C), 107.18, 106.72, 106.17, 105.62, 102.63, 100.20, 99.94, 95.05, 85.30, 83.81, 79.23, 77.97, 77.83, 77.78, 76.09, 75.65, 75.24, 74.93, 74.01 (2C), 72.49, 72.35, 72.20, 71.90, 70.97, 70.34, 70.24, 64.58, 62.42, 62.15, 18.09, 17.88 HRMS (m/z): [M]+ calcd. for C44H57O30, 1065.2935; found, 1065.2908 MbA-CG (7)  7 was generated from 6 according to the general procedure for enzyme hydrolysis by Abg. 6 (10 mg) was dissolved in 1.5 mL sodium phosphate buffer (50 mM, pH 6.8) with 10 µL of 8 mg/mL Abg at 37°C. After 24 h another 10 µL of enzyme was added. Upon reaction completion and workup, the resulting solution was purified by HPLC to yield 7 in 50% yield.   131 1H NMR (400 MHz, methanol-d4): 6.92 (s, 2H), 6.41 (d, J = 1.8 Hz, 1H), 6.25 (d, J = 2.1 Hz, 1H), 5.70 (s, 1H), 4.83 (m, 2H), 4.44 (d, J = 7.6 Hz, 1H), 4.26 (dd, J = 3.5, 1.4 Hz, 1H), 4.15 (dd, J = 11.7, 5.0 Hz, 1H), 3.93-3.99 (m, 1H), 3.82 (dd, J = 9.3, 3.5 Hz, 2H), 3.77 - 3.79 (m, 2H), 3.67 - 3.72 (m, 3H), 3.53 - 3.64 (m, 3H), 3.35 - 3.45 (m, 3H), 3.22 - 3.27 (m, 3H), 1.30 (d, J = 6.1 Hz, 3H), 1.00 (d, J = 5.8 Hz, 3H) 13C NMR (100 MHz, methanol-d4): 179.75, 166.72, 163.29, 159.01, 158.01, 152.17 (2C), 137.08, 136.46, 128.71, 109.89 (2C), 107.40, 107.31, 106.12, 102.77, 100.35, 100.00, 95.12, 82.84, 78.05, 77.97, 75.75, 75.49, 75.27, 75.01, 74.05, 73.54, 72.53, 72.22, 72.11, 71.87, 71.03, 70.26, 64.61, 62.44, 18.09, 17.96 HRMS (m/z): [M]+ calcd. for C38H48O25Na, 927.2382; found, 927.2363 MbA-CR (8)  8 was prepared according to the general procedure for enzymatic cleavage by naringinase. Naringinase (20 µL of 5.6 mg/mL) was incubated at 60°C for 2 hours, then was added to a  132 solution of 6 (5 mg/mL) in sodium acetate buffer (50 mM, pH 5.6). After one day the reaction mixture was purified by HPLC to give 8 in 73% yield. 1H NMR (300 MHz, methanol-d4):  6.94 (s, 2H), 6.40 (d, J = 2.1 Hz, 1H), 6.24 (d, J = 2.1 Hz, 1H), 5.64 (s, 1H), 4.83 (d, J = 7.3 Hz, 3H), 4.59 (d, J = 7.5 Hz, 1H), 4.50 (d, J = 7.5 Hz, 1H), 4.22 (d, J = 2.5 Hz, 1H), 3.93 - 4.02 (m, 2H), 3.68 - 3.77 (m, 3H), 3.54 - 3.63 (m, 3H), 3.40 - 3.48 (m, 4H), 3.35 - 3.39 (m, 3H), 3.22 - 3.28 (m, 3H), 0.99 (d, J = 5.7 Hz, 3H)  13C NMR (75 MHz, methanol-d4): 179.79, 166.33, 163.39, 158.87, 158.77, 152.13 (2C), 137.25, 136.34, 128.66, 109.82 (2C), 107.30, 106.80, 106.20, 105.70, 102.70, 100.20, 95.01, 85.40, 83.90, 79.353, 78.04, 77.88 (2C), 77.21, 76.16, 74.78, 74.00, 72.44, 71.92, 71.02, 70.87, 70.37, 67.42, 62.46, 62.20, 17.89 HRMS (m/z): [M]+ calcd. for C38H48O26Na, 943.2332; found, 943.2348 MbA-CGR (9)   133 9 was prepared according to the general procedure for enzymatic cleavage by naringinase. Naringinase (30 µL of 5.6 mg/mL) was incubated at 60°C for 2 hours, then added to a solution of 7 (7.5 mg/mL) in sodium acetate buffer (50mM, pH 5.6). After one day, the reaction mixture was purified by HPLC to give 9 in 30% yield. 1H NMR (300 MHz, methanol-d4):  6.82 (s, 2H), 6.28 (d, J = 1.8 Hz, 1H), 6.12 (d, J = 1.8 Hz, 1H), 5.60 (d, J = 1.1 Hz, 1H), 4.72 (d, J = 7.3 Hz, 2H), 4.32 (d, J = 7.8 Hz, 1H), 4.16 (d, J = 1.6 Hz, 1H), 3.89 (dd, J = 11.3, 4.9 Hz, 1H), 3.68 - 3.73 (m, 2H), 3.60 – 3.61 (m, 3H), 3.44 - 3.55 (m, 3H), 3.32 - 3.38 (m, 1H), 3.25 - 3.29 (m, 2H), 3.10 - 3.19 (m, 1H), 0.90 (d, J = 5.3 Hz, 3H)  13C NMR (75 MHz, methanol-d4):  179.68, 167.25, 163.33, 158.90, 152.19 (2C), 137.09, 136.46, 128.74, 109.86 (2C), 107.46, 107.33, 105.96, 102.79, 100.51, 95.23, 82.86, 77.98, 77.27 (2C), 75.49, 74.81, 74.05, 73.57, 72.11, 71.89, 71.02, 70.88, 67.46, 62.42, 17.93 HRMS (m/z): [M]+ calcd. for C32H38O21Na, 781.1803; found, 781.1798 5.1.3 Enzymatic synthesis of glucosyl epi-cyclophellitols  The synthesis of glucosyl epi-cyclophellitols (G-ECPs) was carried out according to Nakai 232 with modifications. LaMalP (9 µM) was incubated with -glucose-1-phosphate (50 mM) and ECP (50 mM) in sodium acetate buffer (0.3 mL, 50 mM, pH 4.0) at room temperature for 6 hours. The reaction was monitored by TLC. The enzyme was inactivated by methanol, and removed by centrifugation. The resulting reaction mixture was purified by HPLC with a light scattering detector (Sedex, France) using a TSKgel Amide-80 column (Tosoh, 4.6 x 250 mm, 5 µm) at a  134 constant flow (1.0 mL/min) of mobile phase ACN/water (v/v) 80/20 to afford the α-1, 3 and α-1,4 G-ECP products at retention time of 14 and 16 mins with a ratio of 4:1, respectively.  α-1,3 G-ECP (10)  1H NMR (400 MHz, H2O-d4): 5.20 (d, J = 3.9, 1H), 4.03 (dd, J = 8.2, 1.8, 1H), 4.01-3.99 (m, 1H), 3.93 (dd, J = 11.3, 3.2, 1H), 3.82-3.78 (m, 3H), 3.74 (dd, J = 9.5, 9.5, 1H), 3.62-3.56 (m, 2H), 3.56-3.53 (m, 1H), 3.51-3.45 (m, 2H), 3.36 (d, J = 4.1, 1H), 2.12-2.07 (m, 1H) 13C NMR (75 MHz, H2O-d4): 100.88, 82.50, 73.99, 72.80, 72.68, 70.86, 70.75, 70.18, 61.11, 61.05, 58.90, 56.13, 45.01 HRMS (m/z): [M]+ calcd. for C13H22O10Na, 361.1111; found, 361.1110. α-1, 4 G-ECP (11)  1H NMR (400 MHz, H2O-d4): 5.18 (d, J = 3.9, 1H), 3.99 (dd, J = 8.8, 1.5, 1H), 3.94 (dd, J = 11.4, 2.8, 1H), 3.89-3.76 (m, 4H), 3.72-3.67 (m, 2H), 3.60 (dd, J = 9.9, 3.9, 1H), 3.53-3.48 (m, 2H), 3.42 (dd, J = 9.3, 9.3, 1H), 3.38 (dd, J = 8.7, 4.5, 1H), 2.27-2.23 (m, 1H)  135 13C NMR (75 MHz, H2O-d4): 100.75, 79.92, 73.44,72.97, 72.52, 71.72, 70.65, 69.33, 60.45, 60.31, 57.31, 55.25, 42.82  HRMS (m/z): [M]+ calcd. for C13H22O10Na, 361.1111; found, 361.1110. 5.1.4 Enzymatic synthesis of 4’-O-methyl α-maltosyl epi-cyclophellitol (MeG2-ECP) Enzymatic coupling of MeG2F (0.72 mg, 2 mol) and ECP (0.41 mg, 2.3 mol) in phosphate buffer (40 L, 50 mM sodium phosphate, 100 mM NaCl, pH 7.0), was accomplished by incubating the two compounds in the presence of CGTase (1 L) at 30 C for 12 h. The reaction was monitored by TLC until fully consumption of ECP. Upon completion, the reaction mixture was treated with an equal volume of ethanol, and the precipitated enzyme was removed by centrifugation. The resulting crude product was directed used for subsequent analysis.  5.1.5 Mass spectrometric analysis of intact protein The mass spectra were recorded using the NanoAquity Ultra Performance LC System™ and a GS-2 Q-Tof ™ quadrupole/orthogonal acceleration TOF mass spectrometer from Waters Corporation, Micromass MS Technologies (Manchester, UK). Proteins were loaded onto a C4 trap cartridges (300 μm × 5 mm, LC Packings/Dionex distributed by Kovalent AB, Hägersten, Sweden), and eluted with a gradient of 0-90 % solvent B in solvent A over the course of 10 mins at a flow rate of 5 μl/min (solvent A: 5% ACN(aq), 0.1% FA: solvent B: 95% ACN(aq), 0.1% FA). The sample flow from the C4 cartridge was coupled directly to the Q-Tof ESI interface consisting of the Z spray source fitted with an electrospray probe (source voltage 3.0 kV, source temp 80 °C, desolvation temp 140 °C, desolvation gas flow 600 L/h, cone voltage 40 V, cone gas flow  136 50 L/h). TOF MS data were acquired over the m/z range 400–2000 at a resolution > 30,000 FWHM (full width at half maximum). All data were collected using a scan time of 5 s.  5.2 Molecular biology 5.2.1 General materials  2-Chloro-4-nitrophenyl α-D-maltotrioside (CNP-G3) was purchased from Genzymes (Cambridge, USA). All commercially available vectors, restriction enzymes, T4 DNA ligase, fast alkaline phosphatase and Pichia pastoris strains were acquired from Thermo Fisher Scientific. pPic9-HPA vector and endoglycosidase F-cellulose binding domain fusion protein (endo-F) were provided by Ms. Emily Kwan. All chromatographic resins were purchased from GE Healthcare (Little Chalfont, UK). All oligodeoxy-ribonucleotides were acquired from Nucleic Acid and Protein Service (NAPS) Unit at University of British Columbia (Vancouver, Canada). All protein mass was analyzed by Proteomics Core Facility UBC Centre for High-Throughput Biology.  5.2.2 Site directed mutagenesis Each PCR reaction for AB and CD fragments was performed in a total volume of 50 μL containing the pPic9-HPA template (10 ng), the pair of primers (20 pmol), each of the four deoxynucleotides (20 pmol) and Pfu polymerase (1.25 units) in 1X Phusion HF buffer (Fermentas). The reaction was repeated 25 times by cycling the temperature of the reaction mixture using a GeneAmp PCR system 2400 (Perkin-Elmer), with a temperature cycle 98 °C (10 seconds)-56 °C (30 seconds)-72 °C (35 seconds). Both AB and CD fragments were purified by a QIAquick gel extraction kit (Qiagen) to separate them from wild type template DNA. To get the  137 AD fragment, PCR conditions were carried out as described above with slight differences: (1) both AB and CD fragments (20 ng) were used as templates and (2) the temperature cycle was modified to 98°C (10 seconds) - 60°C (30 seconds) - 72°C (50 seconds). The resulting AD fragment was gel purified and digested with restriction enzymes SacI and NotI in 1X FastDigest buffer at 37 °C for one hour, followed by purification by a QIAquick PCR kit (Qiagen). 5.2.3 Generation of pPic9K-HPA mutant vector The pPic9K vector was first treated with SacI and NotI (1 unit each), followed by an addition of fast alkaline phosphatase (1 unit) to give a fragment of 1027 bp and the desired 8249 bp fragment containing the kanamycin resistance gene. The larger fragment was then gel purified to ligate with the fragment AD which was purified as mentioned in 5.2.3. Ligation of the AD fragment and linearized pPic9K vector at a ratio of 2:1 using T4 DNA ligase (0.3 unit) at room temperature for an hour, yielded the desired pPic9K-HPA mutant vectors. After gel purification, the recombined plasmid contained both the kanamycin resistance gene and the α-factor secretion sequence of the amylase. The purified plasmid (8 μL) was transformed into Escherichia coli (Topp 10) by electroporation. The transformed cells were plated on LB-Amp plates overnight. Generally, each transformation resulted in 20-60 colonies. 10 colonies were picked from the plate and grew up in liquid LB medium overnight, and the plasmid was isolated using QIAprep Spin Miniprep kit (Qiagen). The isolated plasmids were sequenced at the region of interest for the desired mutations, by the UBC NAPS unit. 3-6 colonies were sequenced per mutation reaction and 90-100% of the plasmids contained the desired gene.   138 Electroporation was then used to transform the Pichia pastoris strain GS-115 with the pPic9K-HPA mutant plasmid (linearized with SacI) according to the Invitrogen protocol (Invitrogen catalog no. V175-20). The transformed cells were then plated on MD plates (Minimal Dextrose, 1.34% yeast nitrogen base, 4x10-5 % biotin, and 2% dextrose) and grown at 30 °C for two days.  5.2.4 Screening for successful transformants by activity The colony picker (Genetix, QP1x2xT) was used to transfer 90 yeast cell colonies from MD-plate to a 96-well assay plate (Costar@, 3370) in 200 μL BMGY (Buffered Glycerol-complex Medium, 1% yeast extract (w/v), 2% peptone (w/v), 100 mM potassium phosphate, pH 6.0, 1.34% yeast nitrogen base (w/v), 1% glycerol (v/v), 0.006% antiform C) and grown at 30 °C for one day. 10 μL of the culture medium were transferred into 96 deep well plate (Axygen@ Scientific, P-DW-20-C) in 900 μL BMGY medium, which was changed to 300 μL BMMY (Buffered Methanol-complex Medium, 1% yeast extract (w/v), 2% peptone (w/v), 100 mM potassium phosphate, pH 6.0, 1.34% yeast nitrogen base (w/v), 0.5% methanol (v/v) with 0.006% antifoam C) after one day.  The cells were further grown in BMMY medium for two days with MeOH (30 μL of 50% (v/v)) induction twice a day. Once proteins were expressed, the cells were first centrifuged at 4000 rpm, 4 °C for 10 minutes. Each of the culture supernatants (20 μL) containing the mutant proteins was transferred into CNP-G3 solution (74 mM) in HPA buffer (50 mM sodium phosphate, 100 mM sodium chloridel, pH = 7.0) in another 96-well plate. The rate of hydrolysis was measured by plate reader (BioTek, Synergy H1 Hybrid Reader) at 400 nm. The selected mutants were cultured in a 5 mL tube as  139 previously described and the crude protein supernatants were analyzed by the SDS-PAGE (Bio-rad, 4–15% Mini-PROTEAN®TGX™ Gel) for bands at the desired molecular weight. 5.2.5 Protein expression and purification  The protocol used for expression and purification of HPA was first developed by Dr. Edwin Rydberg 233 and was adapted as follows. A selected colony from the 96-well master plate was used to inoculate into 60 mL of BMGY medium in a baffled flask and this was grown at 30 °C overnight. 10 mL of cell culture was transferred to another 600 mL of BMGY medium and grown in a 2 L baffled flask at 30 °C for one day. The cells were collected at 5000 rpm, 4 °C for 15 minutes and re-suspended in 200 mL of BMMY. 2 mL of 50% (v/v) methanol in water was added twice a day into the cell culture and continued to grow at 30 °C for two days. The cells were harvested at 6000 rpm, 4 °C for 20 minutes and the supernatant was filtered through a glass microfibre filter GF/C (WhatmanTM, Cat No.1822125), followed by concentration with an Omega 76 mm 10 K membrane (Pall Filtron, Cat No. MO001076) to about 20 mL. The volume of the supernatant was adjusted with the loading buffer (50 mM sodium chloride, 100 mM potassium phosphate, pH = 7.5) and loaded onto the Phenyl-Sepharose CL-4B column. The column was washed with additional loading buffer (5 × bed volume) followed by elution with de-ionized water. All the fractions were checked on SDS-PAGE and positive fractions were pooled and concentrated to 4 – 8 mL using an Amicon Ultra-15 centrifugal filter (Millipore, 30K). Concentrated buffer was added to this deep green solution to make a final buffer concentration of 20 mM potassium phosphate, 25 mM sodium chloride, pH = 7.0 (cutting buffer). Endo-F fusion protein (1:100 (w/w)) was added and the mixture was left at room temperature  140 overnight. The solution was then passed through a HiTrapTM Q Sepharose Fast Flow anion exchange column that was pre-equilibrated with the loading buffer. The flow-through, which contained the HPA mutant protein, was concentrated by an Amicon 30K. Enzyme concentrations were determined spectrophotographically by a Varian Cary 300 UV/Vis spectrometry using an A0.1% of 2.24 at 280 nm for both wild type and variant HPA. 165  5.3 Kinetics 5.3.1 General assay conditions Unless specified, all reagents were purchased from commercial suppliers (Sigma-Aldrich®, Thermol Fisher Scientific®), and were used without purification. All kinetic studies were performed at 30 °C in HPA buffer (50 mM sodium phosphate, 100 mM NaCl, pH 7.0), unless otherwise noted. NanoOrange protein quantitation kit was purchased from Invitrogen (Cat No. N-6666). Porcine pancreatic -amylase (PPA) was purchased from Sigma-Aldrich. Human gut bacterial amylases were isolated by Tara Hill, human N- and C-terminal maltase-glucoamylase (ntMGAM and ctMGAm), and C-terminal sucrose-isomaltase (ctSI) were donated by Professor David R. Rose (University of Waterloo, Canada). Hydrolysis of CNP-G3 by either HPA or PPA was monitored by the increase of absorbance at 400 nm using a Varian CARY 300/4000 spectrophotometer equipped with a circulating water bath. All enzyme kinetic data were processed using the program GraFit 7.0.0 (Erithacus Software Limited).     141 5.3.2 Enzymatic activity tests of naringinase Rhamnosidase activity using 4-nitrophenyl--L-rhamonside as a substrate The rhamnosidase activity of naringinase was tested according to previously reported protocol with modification. 121b 4-Nitrophenyl--L-rhamonside (1.75 mM) was incubated in sodium acetate buffer (50 mM, pH 5.6) at 25 C, and 10 L of naringinase (5.6 mg/mL) was added to initiate the reaction. The rate of the reaction was measured by monitoring the increase of absorbance at 400 nm using a Varian CARY 300 spectrophotometer. Glucosidase activity using 2,4-dinitrophenyl--D-glucoside as a substrate 2,4-Dinitrophenyl--D-glucoside (30 mM) was incubated in sodium acetate buffer (50 mM, pH 5.6) at 25 C, and 10 L of naringinase (5.6 mg/mL) was added to initiate the reaction. The rate of the reaction was measured by monitoring the increase of absorbance at 400 nm using a Varian CARY 300 spectrophotometer. Xylosidase activity using 4-nitrophenyl--D-xyloside as a substrate 4-Nitrophenyl--D-xyloside (1.9 mM) was incubated in sodium acetate buffer (50 mM, pH 5.6) at 25 C, and 10 L of naringinase (5.6 mg/mL) was added to initiate the reaction. The rate of the reaction was measured by monitoring the increase of absorbance at 400 nm using a Varian CARY 300 spectrophotometer.     142 5.3.3 Michaelis-Menten kinetics of HPA using CNP-G3 as a substrate Michaelis-Menten parameters of HPA variants for CNP-G3 were measured by monitoring the increase of absorbance at 400 nm using a Varian CARY 300 spectrophotometer. Quartz cuvettes (200 μL) with a path length of 1 cm were used. The concentration of enzyme used was about 20 nM for all assays. Initial rates were measured using 6 – 8 different CNP-G3 concentrations ranging from 1 – 20 mM in HPA buffer at 30 °C, and fit to the Michaelis-Menten equation by non-linear regression using the program GraFit 7.0.0. to obtain Vmax (kcat) and Km. 5.3.4 Dinitrosalicylic acid assay of soluble starch Determination of the hydrolysis rate of soluble starch by an enzyme was performed by quantifying the concentrations of sugar reducing ends using dinitrosalicylic acid (DNS) assay. 215 Firstly, a stock of soluble starch solution (10 mg/mL) was made in HPA buffer and kept gentle boil until all the solids were dissolved. Starch solutions at various concentration (0-6 mg/mL) were prepared by diluting the stock solution with HPA buffer. Then purified enzyme samples (final concentration of 3-8 nM depending on enzyme activity) were added to 4 mL diluted starch solution, and incubated for an appropriate length of time at 30 °C. At different time points, 500 µL of the mixture was withdrawn and quenched immediately by addition of an equal volume of DNS solution (10 mg/mL 3, 5-dinitrosalicyclic acid and 300 mg/mL sodium potassium tartrate in 0.4 M sodium hydroxide solution). The color was developed by boiling the stopped samples for 5 minutes, followed by cooling to room temperature. The absorbance was measured spectrophoto-metrically at 540 nm, and plotted against the reaction time. Extrapolating the curve gave a straight line with the slope as the initial rate of starch hydrolysis.  143 The rates were plotted against substrate concentration and fit to a Michaelis-Menten equation on GraFit 7.0.o to obtain Vmax (kcat) and Km. 5.3.5 Inhibitory assay IC50 measurements An approximate inhibition constant was obtained first by measurement of reaction rates with a fixed concentration of substrate at its Km value and at least five inhibitor concentrations. For IC50 of HPA, samples with various inhibitor concentrations were first incubated with the enzyme (4 nM) at 30 °C for 10 min, and CNP-G3 (4 mM) was added as the substrate to initiate the reaction. The release of the chloronitrophenolate anion was monitored continuously at 400 nm and the initial rate was measured. The rates were plotted against inhibitor concentrations and IC50 was calculated by GraFit 5.0.13. IC50 with other enzymes were measured in a similar way with inhibitor concentration up to 500 µM and fixed substrate concentration (PNPαGlc: ntMGAM 7.2 mM, ctMGAM 1.5 mM, ctSI 3.6 mM in 100 mM 2-(N-morpholino)ethanesulfonic acid, pH 6.5 at 37 °C and yeast α-glucosidase 0.36 mM in 50 mM sodium phosphate, pH 7.0 at 37 °C; CNP-G3: porcine pancreatic -amylase: 0.6 mM; riAmy 2.4 mM; DNP-G3: bfAmy 1.0 mM in HPA buffer at 30 °C). The initial rates measured were plotted against inhibitor concentrations and IC50 values were calculated by GraFit 5.0.13. Inhibition constants (Ki) of MbA substructures for HPA The inhibition constants (Ki) of MbA substructures for HPA were determined by measuring the rates of reactions at different inhibitor and substrate concentrations. Inhibitors at concentra- 144 tions ranging from 1/5 to 5 times the Ki were first incubated with HPA (4 nM) at 30 °C for 10 min, and CNP-G3 was used as the substrate in the range of 1/5 to 5 times the Km. Initial reaction rates for the release of chloronitrophenolate were measured by Varian Cary 4000 UV/Vis spectrophotometer at 400 nm. Observed reaction rates were fitted directly to modified Michaelis-Menten equations describing reaction in the presence of competitive, non-competitive and uncompetitive type of inhibition by non-linear regression using GraFit. Competitive inhibition provided the best fit in each case, and only the Ki values fit by GraFit were reported. Inhibition constants (Ki) of MbA-3 for Abg The inhibition constant (Ki) of MbA-3 for Abg were determined by measuring the rates of reactions at different inhibitor and substrate concentrations. Inhibitor at concentration ranging from 1/5 to 5 times the Ki were incubated with Abg (4.5 nM) at 37 °C for 10 min, and 4-nitrophenyl--D-glucoside was used as the substrate in the range of 1/5 to 5 times the Km (0.078 mM). Initial reaction rates for the release of chloronitrophenolate were measured by Varian Cary 4000 UV/Vis spectrophotometer at 400 nm. Observed reaction rates were fitted directly to modified Michaelis-Menten equations describing reaction in the presence of competitive, non-competitive and uncompetitive type of inhibition by non-linear regression using GraFit. Non-competitive inhibition provided the best fit, and only the Ki values fit by GraFit were reported.     145 5.3.6 Inactivation assays General inactivation method Compounds designed as potential mechanism-based inhibitors were tested for their ability to induce time-dependent loss of enzyme activity. Samples at varied inhibitor concentrations were incubated with enzymes at the optimal temperature, using the buffer described for standard Michaelis-Menten kinetics. To test for residual enzyme activity, aliquots of reaction mixture were removed at time points and were added to a pre-incubated substrate buffer solution at a substrate concentration in a large excess over Km. Data for single inactivator concentration were fit to an equation describing the first order decay with offset and kobs was calculated by GraFit 7.0.0. Inactivation by G-ECPs  Samples of HPA (150 nM) were incubated in the presence of G-ECP (3.6 mM α-1,4 or 6 mM α-1,3) in HPA buffer (50 mM sodium phsophate, 100 mM NaCl, pH 7.0) at 30 °C. Aliquots (10 μL) of these inactivation mixtures were removed at time intervals and diluted into assay cells containing 200 µL CNP-G3 (10 mM) pre-incubated at 30°C, and the absorbance change was monitored at 400 nm. Data for each inactivator were fit to an equation describing first order decay with offset in Grapfit 7.0.0 to give kobs. Kinetic evaluation of MeG2-ECP as an HPA inactivator Samples of HPA (75 nM) were incubated in the presence of MeG2-ECP (29 M) in the HPA buffer (50 mM NaPi, 100 mM NaCl, pH 7.0) at 30 °C. Aliquots (4 μL) of the inactivation mixture were removed at time intervals and diluted into assay cells containing 200 µL CNP-G3 (5 mM)  146 pre-incubated at 30°C, and absorbance change was monitored at 400 nm. Data for each inactivator were fit to an equation describing first order decay with offset in Grafit 7.0.0. 5.3.7 Evaluation of kinetics of reactivation of HPA Fully inactivated enzyme was freed of excess inactivator by 10-fold dilution with HPA buffer (50 mM sodium phosphate, 100 mM sodium chloride, pH 7.0), and the diluted reaction mixture was concentrated at 4 °C using an Amicon 30 kDa nominal cut-off centrifugal concentrator to a volume of approximately 35 μL. This procedure was repeated a total of five times. The resultant solution was then diluted to 200 μL with either buffer alone or with 100 mM maltose. Reactivation was monitored by removal of aliquots (10 μL) at appropriate time intervals and assaying using CNP-G3 under standard assay conditions. 5.3.8 Enzyme concentration measurement using NanoOrange assay  The free enzyme concentration was determined using an Invitrogen NanoOrange® assay kit, performed according to the manufacturer’s instructions (MP-06666). The concentrated NanoOrange® protein quantitation diluent was first diluted 10-fold in distilled water to make a 1X protein quantitation diluent. Prior to running the experiment, the NanoOrange® protein quantitation reagent was diluted 500-fold into the 1X protein quantitation diluent to make a 1X NanoOrange working solution. BSA standard solutions were prepared by diluting into 1X NanoOrange working solution. The standard samples were heated at 95 °C for 10 minutes, prevented from photodegradation by covering it with aluminum foil. After cooling to room temperature, the solution was transferred into a full-volume methacrylate cuvette  147 (FisherbrandTM), and the fluorescence was measured directly with an excitation/emission wavelength of 485/590 nm using the fluorimeter (Varian Cary Eclipse). A standard curve of fluorescence versus concentration of BSA was generated. To perform a protein assay, protein sample was diluted into 1 mL 1X NanoOrange working solution, and treated as described for BSA standards. The measured fluorescence of the protein sample was substituted into the BSA standard curve, and the protein concentration was determined. 5.3.9 Adsorption assay of enzymes to starch General adsorption assay The adsorption of purified enzymes to starch granules was assayed according to a procedure slightly modified from the previously reported. 180 Potato starch (Sigma-Aldrich@ S-5651) was first washed three times in deionized water, twice in ethanol and air dried. 234 It was further made into suspension mixtures at various concentrations of 0-200 mg/mL in HPA buffer (50 mM sodium phosphate, 100 mM sodium chloride, pH 7.0) and cooled at 4 °C for at least 10 minutes. Recombinant HPA mutant proteins (final concentration around 100 nM) were incubated with the pre-chilled starch mixtures (1 mL) on a rotor at 4 °C for 60 minutes. The mixtures were then centrifuged (13.6 krpm, 4 °C, 10 min) to separate all the starch, including bound amylase, from free enzymes. 500 L of the resulting supernatant was concentrated to a volume of approximately 10 µL using Amicon 30K. The concentrated enzyme solution was diluted with 1X NanoOrange working solution to achieve a final volume of 1 mL. The samples were treated as described in Section 5.3.8, the fluorescence of the unbound enzyme was  148 measured at 485/590 nm. The activity (expressed as the percentage of bound proteins when compared with a no-starch control) was plotted against the starch concentrations, and the data were fitted to a one-site binding model. 𝑏 =  𝐵𝑚𝑎𝑥 X [𝑆][𝑆] + 𝐾d Where b is the bound enzyme fraction, [S] the starch concentration, and Bmax the maximum fraction of enzyme bound. Modified protocol with MbA as inhibitor HPA and its surface mutants (20 M) were incubated with MbA (400 M) in HPA buffer at room temperature for one hour, prior to the addition (5 L) of the enzyme complex into pre-chilled starch suspensions (1 mL). The adsorption assay was followed the protocol of general adsorption assay. Modified protocol with in situ elongation and inactivation of HPA variant  In situ elongations were accomplished by pre-incubating HPA (90 g, 1.6 nmol) with MeG2F (10 mg, 28 mol) and ECP (5 mg, 28 mol) in 200 L HPA buffer at 30 °C, until no residual activity of enzyme could be detected. The excess elongation reagents were removed by Amicon 30K, and the inactivated enzyme (12.5 L) was added into the pre-chilled starch suspensions (1 mL), followed by the general adsorption assay.   149 References 1. Davies, G. J.; Gloster, T. M.; Henrissat, B., Recent structural insights into the expanding world of carbohydrate-active enzymes. Curr. Opin. Struct. Biol. 2005, 15 (6), 637-645. 2. Blake, C. C. F.; Koenig, D. F.; Mair, G. A.; North, A. C. T.; Phillips, D. C.; Sarma, V. R., Structure of Hen Egg-White Lysozyme - a 3-Dimensional Fourier Synthesis at 2 Å Resolution. Nature 1965, 206 (4986), 757-761. 3. Brás, N. F.; Fernandes, P. A.; Ramos, M. J.; Cerqueira, N. M. F. S. A., Chapter 6 Glycosidases – A Mechanistic Overview. InTech: 2012; p 117-134. 4. Bojarova, P.; Kren, V., Glycosidases: a key to tailored carbohydrates. Trends Biotechnol. 2009, 27 (4), 199-209. 5. Brás, N. F.; Cerqueira, N. M. F. S. A.; Ramos, M. J.; Fernandes, P. A., Glycosidase inhibitors: a patent review (2008-2013). Expert Opin. Ther. Pat. 2014, 24 (8), 857-874. 6. Asano, N.; Nash, R. J.; Molyneux, R. J.; Fleet, G. W. J., Sugar-mimic glycosidase inhibitors: natural occurrence, biological activity and prospects for therapeutic application. Tetrahedron-Asymmetr 2000, 11 (8), 1645-1680. 7. Asano, N., Naturally occurring iminosugars and related compounds: Structure, distribution, and biological activity. Curr. Top. Med. Chem. 2003, 3 (5), 471-484. 8. Butters, T. D.; Dwek, R. A.; Platt, F. M., Therapeutic applications of imino sugars in lysosomal storage disorders. Curr. Top. Med. Chem. 2003, 3 (5), 561-574. 9. de Melo, E. B.; Gomes, A. D.; Carvalho, I., alpha- and beta-Glucosidase inhibitors: chemical structure and biological activity. Tetrahedron 2006, 62 (44), 10277-10302. 10. C. Webb, E., Enzyme Nomenclature 1992. Recommendations of the Nomenclature Committee of the International Union of Biochemistry and Molecular Biology on the Nomenclature and Classification of Enzymes. Academic Press: San Diego, 1992; p 1-862. 11. Henrissat, B.; Davies, G., Structural and sequence-based classification of glycoside hydrolases. Curr. Opin. Struct. Biol. 1997, 7 (5), 637-644. 12. Henrissat, B., A Classification of Glycosyl Hydrolases Based on Amino-Acid-Sequence Similarities. Biochem. J. 1991, 280, 309-316. 13. Lombard, V.; Ramulu, H. G.; Drula, E.; Coutinho, P. M.; Henrissat, B., The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res. 2014, 42 (D1), D490-D495. 14. Gebler, J.; Gilkes, N. R.; Claeyssens, M.; Wilson, D. B.; Beguin, P.; Wakarchuk, W. W.; Kilburn, D. G.; Miller, R. C.; Warren, R. A. J.; Withers, S. G., Stereoselective Hydrolysis Catalyzed by Related Beta-1,4-Glucanases and Beta-1,4-Xylanases. J. Biol. Chem. 1992, 267 (18), 12559-12561. 15. Henrissat, B.; Bairoch, A., Updating the sequence-based classification of glycosyl hydrolases. Biochem. J. 1996, 316, 695-696. 16. Henrissat, B.; Callebaut, I.; Fabrega, S.; Lehn, P.; Mornon, J. P.; Davies, G., Conserved Catalytic Machinery and the Prediction of a Common Fold for Several Families of Glycosyl Hydrolases. Proc. Natl. Acad. Sci. U. S. A. 1995, 92 (15), 7090-7094. 17. Koshland, D. E., Stereochemistry and the Mechanism of Enzymatic Reactions. Biol. Rev. Camb. Philos. Soc. 1953, 28 (4), 416-436. 18. Zechel, D. L.; Withers, S. G., Glycosidase mechanisms: Anatomy of a finely tuned catalyst. Acc. Chem. Res. 2000, 33 (1), 11-18. 19. Yip, V. L.; Withers, S. G., Nature's many mechanisms for the degradation of oligosaccharides. Org. Biomol. Chem. 2004, 2 (19), 2707-2713.  150 20. Yip, V. L.; Withers, S. G., Breakdown of oligosaccharides by the process of elimination. Curr. Opin. Chem. Biol. 2006, 10 (2), 147-155. 21. Jongkees, S. A. K.; Yoo, H.; Withers, S. G., Mechanistic Investigations of Unsaturated Glucuronyl Hydrolase from Clostridium perfringens. J. Biol. Chem. 2014, 289 (16), 11385-11395. 22. MacGregor, E. A., An overview of clan GH-H and distantly-related families. Biologia (Bratisl.) 2005, 60, 5-12. 23. Takata, H.; Kuriki, T.; Okada, S.; Takesada, Y.; Iizuka, M.; Minamiura, N.; Imanaka, T., Action of Neopullulanase. Neopullulanase Catalyzes Both Hydrolysis and Transglycosylation at Alpha-(1-4)-Glucosidic and Alpha-(1-6)-Glucosidic Linkages. J. Biol. Chem. 1992, 267 (26), 18447-18452. 24. Kuriki, T.; Imanaka, T., The concept of the alpha-amylase family: Structural similarity and common catalytic mechanism. J. Biosci. Bioeng. 1999, 87 (5), 557-565. 25. Sivaramakrishnan, S.; Gangadharan, D.; Nampoothiri, K. M.; Soccol, C. R.; Pandey, A., alpha-amylases from microbial sources - An overview on recent developments. Food Technol. Biotechnol. 2006, 44 (2), 173-184. 26. Stein, E. A.; Hsiu, J.; Fischer, E. H., -Amylases as calcium-metalloenzymes. I. Preparation of calcium-free apoamylases by chelation and electrodialysis. Biochemistry 1964, 3 (1), 56-61. 27. Robyt, J. F., Essentials of Carbohydrate Chemistry. Springer-Verlag: New York, 1998; p 328-333 and 337-343. 28. Li, C. F.; Du, M. F.; Cheng, B.; Wang, L. S.; Liu, X. Q.; Ma, C. Q.; Yang, C. Y.; Xu, P., Close relationship of a novel Flavobacteriaceae alpha-amylase with archaeal alpha-amylases and good potentials for industrial applications. Biotechnol Biofuels 2014, 7. 29. Gupta, R.; Gigras, P.; Mohapatra, H.; Goswami, V. K.; Chauhan, B., Microbial alpha-amylases: a biotechnological perspective. Process Biochem. 2003, 38 (11), 1599-1616. 30. van der Maarel, M. J. E. C.; van der Veen, B.; Uitdehaag, J. C. M.; Leemhuis, H.; Dijkhuizen, L., Properties and applications of starch-converting enzymes of the alpha-amylase family. J. Biotechnol. 2002, 94 (2), 137-155. 31. Jenkins, D. J. A.; Wolever, T. M. S.; Taylor, R. H.; Barker, H.; Fielden, H.; Baldwin, J. M.; Bowling, A. C.; Newman, H. C.; Jenkins, A. L.; Goff, D. V., Glycemic Index of Foods - a Physiological-Basis for Carbohydrate Exchange. Am. J. Clin. Nutr. 1981, 34 (3), 362-366. 32. Jenkins, D. J. A.; Ghafari, H.; Wolever, T. M. S.; Taylor, R. H.; Jenkins, A. L.; Barker, H. M.; Fielden, H.; Bowling, A. C., Relationship between Rate of Digestion of Foods and Post-Prandial Glycemia. Diabetologia 1982, 22 (6), 450-455. 33. Meyer, B. H.; Muller, F. O.; Kruger, J. B.; Clur, B. K.; Grigoleit, H. G., Effects of Tendamistate (an Alpha-Amylase Inactivator), Guar and Placebo on Starch Metabolism. S. Afr. Med. J. 1984, 66 (6), 222-223. 34. Truscheit, E.; Frommer, W.; Junge, B.; Mueller, L.; Schmidt, D. D.; Wingender, W., Chemistry and biochemistry of microbial -glucosidase inhibitors. Angew. Chem. 1981, 93 (9), 738-55. 35. Scheen, A. J., Is there a role for -glucosidase inhibitors in the prevention of type 2 diabetes mellitus? Drugs 2003, 63 (10), 933-951. 36. Levitzki, A.; Steer, M. L., Allosteric Activation of Mammalian Alpha-Amylase by Chloride. Eur. J. Biochem. 1974, 41 (1), 171-180. 37. Brayer, G. D.; Luo, Y.; Withers, S. G., The structure of human pancreatic alpha-amylase at 1.8 Å resolution and comparisons with related enzymes. Protein Sci. 1995, 4 (9), 1730-42. 38. Ramasubbu, N.; Paloth, V.; Luo, Y. G.; Brayer, G. D.; Levine, M. J., Structure of human salivary alpha-amylase at 1.6 Å resolution: Implications for its role in the oral cavity. Acta Crystallogr D 1996, 52, 435-446. 39. Gumucio, D. L.; Wiebauer, K.; Caldwell, R. M.; Samuelson, L. C.; Meisler, M. H., Concerted Evolution of Human Amylase Genes. Mol. Cell. Biol. 1988, 8 (3), 1197-1205.  151 40. Groot, P. C.; Bleeker, M. J.; Pronk, J. C.; Arwert, F.; Mager, W. H.; Planta, R. J.; Eriksson, A. W.; Frants, R. R., Human Pancreatic Amylase Is Encoded by 2 Different Genes. Nucleic Acids Res. 1988, 16 (10), 4724-4724. 41. Nishide, T.; Emi, M.; Nakamura, Y.; Matsubara, K., Corrected sequences of cDNAs for human salivary and pancreatic alpha-amylases. Gene 1984, 28 (2), 263-70. 42. Numao, S.; Maurus, R.; Sidhu, G.; Wang, Y.; Overall, C. M.; Brayer, G. D.; Withers, S. G., Probing the role of the chloride ion in the mechanism of human pancreatic alpha-amylase. Biochemistry 2002, 41 (1), 215-225. 43. Maurus, R.; Begum, A.; Williams, L. K.; Fredriksen, J. R.; Zhang, R.; Withers, S. G.; Brayer, G. D., Alternative catalytic anions differentially modulate human alpha-amylase activity and specificity. Biochemistry 2008, 47 (11), 3332-3344. 44. Braun, C.; Meinke, A.; Ziser, L.; Withers, S. G., Simultaneous High-Performance Liquid-Chromatographic Determination of Both the Cleavage Pattern and the Stereochemical Outcome of the Hydrolysis Reactions Catalyzed by Various Glycosidases. Anal. Biochem. 1993, 212 (1), 259-262. 45. Brayer, G. D.; Sidhu, G.; Maurus, R.; Rydberg, E. H.; Braun, C.; Wang, Y.; Nguyen, N. T.; Overall, C. M.; Withers, S. G., Subsite mapping of the human pancreatic alpha-amylase active site through structural, kinetic, and mutagenesis techniques. Biochemistry 2000, 39 (16), 4778-91. 46. Pasero, L.; Mazzei-Pierron, Y.; Abadie, B.; Chicheportiche, Y.; Marchis-Mouren, G., Complete amino acid sequence and location of the five disulfide bridges in porcine pancreatic alpha-amylase. Biochim. Biophys. Acta. 1986, 869 (2), 147-57. 47. Kluh, I., Amino acid sequence of hog pancreatic -amylase isoenzyme I. FEBS Lett. 1981, 136 (2), 231-4. 48. Payan, F.; Haser, R.; Pierrot, M.; Frey, M.; Astier, J. P.; Abadie, B.; Duee, E.; Buisson, G., The three-dimensional structure of -amylase from porcine pancreas at 5Å resolution - the active-site location. Acta Crystallogr. 1980, B36 (2), 416-21. 49. Jane, J. L.; Kasemsuwan, T.; Leas, S.; Zobel, H.; Robyt, J. F., Anthology of Starch Granule Morphology by Scanning Electron-Microscopy. Starch 1994, 46 (4), 121-129. 50. Buleon, A.; Colonna, P.; Planchot, V.; Ball, S., Starch granules: structure and biosynthesis. Int. J. Biol. Macromol. 1998, 23 (2), 85-112. 51. Perez, S.; Bertoft, E., The molecular structures of starch components and their contribution to the architecture of starch granules: A comprehensive review. Starch 2010, 62 (8), 389-420. 52. Tester, R. F.; Karkalas, J.; Qi, X., Starch - composition, fine structure and architecture. J. Cereal Sci. 2004, 39 (2), 151-165. 53. Mua, J. P.; Jackson, D. S., Fine structure of corn amylose and amylopectin fractions with various molecular weights. J. Agric. Food Chem. 1997, 45 (10), 3840-3847. 54. Buleon, A.; Gallant, D. J.; Bouchet, B.; Mouille, C.; DHulst, C.; Kossmann, J.; Ball, S., Starches from A to C - Chlamydomonas reinhardtii as a model microbial system to investigate the biosynthesis of the plant amylopectin crystal. Plant Physiol. 1997, 115 (3), 949-957. 55. J.N. BeMiller, R. L. W., Starch: Chemistry and Technology. 3rd Ed. ed.; Elsevier: New York, 2009; p 1-853. 56. Galliard, T., Starch: Properties and Potential. John Wily & Sons: New York, 1987; p 55-78. 57. Imberty, A.; Chanzy, H.; Perez, S.; Buleon, A.; Tran, V., The Double-Helical Nature of the Crystalline Part of a-Starch. J. Mol. Biol. 1988, 201 (2), 365-378. 58. Imberty, A.; Buleon, A.; Tran, V.; Perez, S., Recent Advances in Knowledge of Starch Structure. Starch 1991, 43 (10), 375-384. 59. Sarko A, Z. P., ACS Symposium Series, editors. In: Fiber Diffraction Methods. Am. Chem. Soc: Washington, DC, 1980.  152 60. Rappenecker, G.; Zugenmaier, P., Detailed Refinement of the Crystal-Structure of Vh-Amylose. Carbohydr. Res. 1981, 89 (1), 11-19. 61. Murphy, V. G.; Zaslow, B.; French, A. D., Structure of V-Amylose Dehydrate - Combined X-Ray and Stereochemical Approach. Biopolymers 1975, 14 (7), 1487-1501. 62. Peat, S.; Whelan, W. J.; Thomas, G. J., Evidence of Multiple Branching in Waxy Maize Starch. J Chem Soc 1952,  (Nov), 4546-4548. 63. Tang, H. J.; Mitsunaga, T. H.; Kawamura, Y., Molecular arrangement in blocklets and starch granule architecture. Carbohydr. Polym. 2006, 63 (4), 555-560. 64. Tester, R. F.; Morrison, W. R., Swelling and Gelatinization of Cereal Starches .1. Effects of Amylopectin, Amylose, and Lipids. Cereal Chem. 1990, 67 (6), 551-557. 65. Gallant, D. J.; Bouchet, B.; Buleon, A.; Perez, S., Physical Characteristics of Starch Granules and Susceptibility to Enzymatic Degradation. Eur. J. Clin. Nutr. 1992, 46, S3-S16. 66. Nielsen, M. M.; Seo, E. S.; Dilokpimol, A.; Andersen, J.; Abou Hachem, M.; Naested, H.; Willemoes, M.; Bozonnet, S.; Kandra, L.; Gyemant, G.; Haser, R.; Aghajari, N.; Svensson, B., Roles of multiple surface sites, long substrate binding clefts, and carbohydrate binding modules in the action of amylolytic enzymes on polysaccharide substrates. Biocatal. Biotransform. 2008, 26 (1-2), 59-67. 67. Guillen, D.; Sanchez, S.; Rodriguez-Sanoja, R., Carbohydrate-binding domains: multiplicity of biological roles. Appl. Microbiol. Biotechnol. 2010, 85 (5), 1241-1249. 68. Li, C. M.; Begum, A.; Numao, S.; Park, K. H.; Withers, S. G.; Brayer, G. D., Acarbose rearrangement mechanism implied by the kinetic and structural analysis of human pancreatic alpha-amylase in complex with analogues and their elongated counterparts. Biochemistry 2005, 44 (9), 3347-3357. 69. Cheng, A. Y. Y.; Fantus, I. G., Oral antihyperglycemic therapy for type 2 diabetes mellitus. Can. Med. Assoc. J. 2005, 172 (2), 213-226. 70. (a) Sels, J. P.; Huijberts, M. S.; Wolffenbuttel, B. H., Miglitol, a new alpha-glucosidase inhibitor. Expert Opin Pharmacother 1999, 1 (1), 149-56; (b) van de Laar, F. A., Alpha-glucosidase inhibitors in the early treatment of type 2 diabetes. Vasc Health Risk Manag 2008, 4 (6), 1189-95; (c) Baron, A. D., Postprandial hyperglycaemia and alpha-glucosidase inhibitors. Diabetes Res Clin Pract 1998, 40 Suppl, S51-5. 71. Scheen, A. J., Drug treatment of non-insulin-dependent diabetes mellitus in the 1990s. Achievements and future developments. Drugs 1997, 54 (3), 355-368. 72. Robyt, J. F., Inhibition, activation, and stabilization of alpha-amylase family enzymes. Biologia (Bratisl.) 2005, 60, 17-26. 73. Yoon, S. H.; Robyt, J. F., Study of the inhibition of four alpha amylases by acarbose and its 4(IV)-alpha-maltohexaosyl and 4(IV)-alpha-maltododecaosyl analogues. Carbohydr. Res. 2003, 338 (19), 1969-1980. 74. Ferey-Roux, G.; Perrier, J.; Forest, E.; Marchis-Mouren, G.; Puigserver, A.; Santimone, M., The human pancreatic -amylase isoforms: isolation, structural studies and kinetics of inhibition by acarbose. Biochim. Biophys. Acta. 1998, 1388 (1), 10-20. 75. Brogard, J. M.; Willemin, B.; Blickle, J. F.; Lamalle, A. M.; Stahl, A., Alpha-glucosidase inhibitors: a new therapeutic approach in diabetes and functional hypoglycemia. Rev. Med. Intern. 1989, 10 (4), 365-74. 76. Mosi, R.; Sham, H.; Uitdehaag, J. C. M.; Ruiterkamp, R.; Dijkstra, B. W.; Withers, S. G., Reassessment of acarbose as a transition state analogue inhibitor of cyclodextrin glycosyltransferase. Biochemistry 1998, 37 (49), 17192-17198. 77. Davies, G. J.; Wilson, K. S.; Henrissat, B., Nomenclature for sugar-binding subsites in glycosyl hydrolases. Biochem. J. 1997, 321, 557-559.  153 78. Qian, M. X.; Haser, R.; Buisson, G.; Duee, E.; Payan, F., The Active-Center of a Mammalian Alpha-Amylase - Structure of the Complex of a Pancreatic Alpha-Amylase with a Carbohydrate Inhibitor Refined to 2.2 Å Resolution. Biochemistry 1994, 33 (20), 6284-6294. 79. Gilles, C.; Astier, J. P.; MarchisMouren, G.; Cambillau, C.; Payan, F., Crystal structure of pig pancreatic alpha-amylase isoenzyme II, in complex with the carbohydrate inhibitor acarbose. Eur. J. Biochem. 1996, 238 (2), 561-569. 80. Brzozowski, A. M.; Davies, G. J., Structure of the Aspergillus oryzae alpha-amylase complexed with the inhibitor acarbose at 2.0 Å resolution. Biochemistry 1997, 36 (36), 10837-10845. 81. Lebovitz, H. E., Alpha-glucosidase inhibitors. Endocrinol. Metab. Clin. N. Am. 1997, 26 (3), 539-&. 82. Gyemant, G.; Kandra, L.; Nagy, V.; Somsak, L., Inhibition of human salivary alpha-amylase by glucopyranosylidene-spiro-thiohydantoin. Biochem. Biophys. Res. Commun. 2003, 312 (2), 334-339. 83. Chakrabarti, R.; Rajagopalan, R., Diabetes and insulin resistance associated disorders: Disease and the therapy. Curr. Sci. 2002, 83 (12), 1533-1538. 84. McDougall, G. J.; Shpiro, F.; Dobson, P.; Smith, P.; Blake, A.; Stewart, D., Different polyphenolic components of soft fruits inhibit alpha-amylase and alpha-glucosidase. J. Agric. Food Chem. 2005, 53 (7), 2760-2766. 85. Mogale, M. A.; Lebelo, S. L.; Thovhogi, N.; de Freitas, A. N.; Shai, L. J., alpha-Amylase and alpha-glucosidase inhibitory effects of sclerocarya birrea [(A. Rich.) Hochst.] subspecies caffra (Sond) Kokwaro (Anacardiaceae) stem-bark extracts. Afr. J. Biotechnol. 2011, 10 (66), 15033-15039. 86. Rubilar, M.; Jara, C.; Poo, Y.; Acevedo, F.; Gutierrez, C.; Sineiro, J.; Shene, C., Extracts of Maqui (Aristotelia chilensis) and Murta (Ugni molinae Turcz.): Sources of Antioxidant Compounds and alpha-Glucosidase/alpha-Amylase Inhibitors. J. Agric. Food Chem. 2011, 59 (5), 1630-1637. 87. Lo Piparo, E.; Scheib, H.; Frei, N.; Williamson, G.; Grigorov, M.; Chou, C. J., Flavonoids for controlling starch digestion: Structural requirements for inhibiting human alpha-amylase. J. Med. Chem. 2008, 51 (12), 3555-3561. 88. Havsteen, B., Flavonoids, a Class of Natural-Products of High Pharmacological Potency. Biochem. Pharmacol. 1983, 32 (7), 1141-1148. 89. Tadera, K.; Minami, Y.; Takamatsu, K.; Matsuoka, T., Inhibition of alpha-glucosidase and alpha-amylase by flavonoids. J. Nutr. Sci. Vitaminol. 2006, 52 (2), 149-153. 90. Williams, L. K.; Li, C. M.; Withers, S. G.; Brayer, G. D., Order and Disorder: Differential Structural Impacts of Myricetin and Ethyl Caffeate on Human Amylase, an Antidiabetic Target. J. Med. Chem. 2012, 55 (22), 10177-10186. 91. Tarling, C. A.; Woods, K.; Zhang, R.; Brastianos, H. C.; Brayer, G. D.; Andersen, R. J.; Withers, S. G., The search for novel human pancreatic alpha-amylase inhibitors: High-throughput screening of terrestrial and marine natural product extracts. ChemBioChem 2008, 9 (3), 433-438. 92. Ong, K. C.; Khoo, H. E., Effects of myricetin on glycemia and glycogen metabolism in diabetic rats. Life Sci. 2000, 67 (14), 1695-1705. 93. Asada, Y.; Hirayama, Y.; Furuya, T., Acylated flavonols from Crocosmia crocosmiiflora Phytochemistry 1988, 27 (5), 1497-1501. 94. O'Connell, J. F.; Bender, R.; Engels, J. W.; Koller, K. P.; Scharf, M.; Wuthrich, K., The nuclear-magnetic-resonance solution structure of the mutant alpha-amylase inhibitor [R19L] Tendamistat and comparison with wild-type Tendamistat. Eur. J. Biochem. 1994, 220 (3), 763-70. 95. (a) Etzkorn, F. A.; Guo, T.; Lipton, M. A.; Goldberg, S. D.; Bartlett, P. A., Cyclic Hexapeptides and Chimeric Peptides as Mimics of Tendamistat. J. Am. Chem. Soc. 1994, 116 (23), 10412-25; (b) Matter, H.; Kessler, H., Structures, Dynamics, and Biological Activities of 15 Cyclic Hexapeptide Analogs of the -Amylase Inhibitor Tendamistat (HOE 467) in Solution. J. Am. Chem. Soc. 1995, 117 (12), 3347-59; (c) Alam, N.; Gourinath, S.; Dey, S.; Srinivasan, A.; Singh, T. P., Substrate-inhibitor interactions in the kinetics  154 of alpha-amylase inhibition by ragi alpha-amylase/trypsin inhibitor (RATI) and its various N-terminal fragments. Biochemistry 2001, 40 (14), 4229-33. 96. Aschauer, H.; Vertesy, L.; Nesemann, G.; Braunitzer, G., The Primary Structure of the Alpha-Amylase Inhibitor Hoe-467a from Streptomyces-Tendae 4158 - a New Class of Inhibitors. Hoppe. Seylers Z. Physiol. Chem. 1983, 364 (10), 1347-1356. 97. (a) Pflugrath, J. W.; Wiegand, G.; Huber, R.; Vertesy, L., Crystal structure determination, refinement and the molecular model of the alpha-amylase inhibitor Hoe-467A. J. Mol. Biol. 1986, 189 (2), 383-386; (b) Kline, A. D.; Braun, W.; Wuthrich, K., Determination of the Complete 3-Dimensional Structure of the Alpha-Amylase Inhibitor Tendamistat in Aqueous-Solution by Nuclear Magnetic-Resonance and Distance Geometry. J. Mol. Biol. 1988, 204 (3), 675-724. 98. Vertesy, L.; Oeding, V.; Bender, R.; Zepf, K.; Nesemann, G., Tendamistat (Hoe 467), a Tight-Binding Alpha-Amylase Inhibitor from Streptomyces-Tendae 4158 - Isolation, Biochemical-Properties. Eur. J. Biochem. 1984, 141 (3), 505-512. 99. Wiegand, G.; Epp, O.; Huber, R., The Crystal-Structure of Porcine Pancreatic Alpha-Amylase in Complex with the Microbial Inhibitor Tendamistat. J. Mol. Biol. 1995, 247 (1), 99-110. 100. Konig, V.; Vertesy, L.; Schneider, T. R., Structure of the alpha-amylase inhibitor tendamistat at 0.93 Å. Acta Crystallogr D 2003, 59, 1737-1743. 101. Tysoe, C. W., L. K.; Keyzers, R.; Nguyen, N.T.;  Tarling, C;  Wicki, J.; Goddard-Borger, E. D.; Aguda, A. H.; Perry, S.; Foster, L. J.; Andersen, R. J.; . Brayer, G. D.; and Withers, S. G., Potent Human α-Amylase Inhibition by the β-Defensin-like Protein Helianthamide. ACS Cent. Sci. 2016, 2 (3), 154-161. 102. Legler, G., Glycoside Hydrolases - Mechanistic Information from Studies with Reversible and Irreversible Inhibitors. Adv. Carbohydr. Chem. Biochem. 1990, 48, 319-384. 103. Withers, S. G.; Aebersold, R., Approaches to Labeling and Identification of Active-Site Residues in Glycosidases. Protein Sci. 1995, 4 (3), 361-372. 104. Mosi, R. M.; Withers, S. G., Trapping of alpha-glycosidase intermediates. Method Enzymol 2002, 354, 64-84. 105. Wicki, J.; Rose, D. R.; Withers, S. G., Trapping covalent intermediates on beta-glycosidases. Methods Enzymol. 2002, 354, 84-105. 106. Mccarter, J. D.; Adam, M. J.; Withers, S. G., Binding-Energy and Catalysis - Fluorinated and Deoxygenated Glycosides as Mechanistic Probes of Escherichia-Coli (Lacz) Beta-Galactosidase. Biochem. J. 1992, 286, 721-727. 107. Street, I. P.; Rupitz, K.; Withers, S. G., Fluorinated and Deoxygenated Substrates as Probes of Transition-State Structure in Glycogen-Phosphorylase. Biochemistry 1989, 28 (4), 1581-1587. 108. Street, I. P.; Kempton, J. B.; Withers, S. G., Inactivation of a Beta-Glucosidase through the Accumulation of a Stable 2-Deoxy-2-Fluoro-Alpha-D-Glucopyranosyl Enzyme Intermediate - a Detailed Investigation. Biochemistry 1992, 31 (41), 9970-9978. 109. Withers, S. G.; Street, I. P., Identification of a Covalent Alpha-D-Glucopyranosyl Enzyme Intermediate Formed on a Beta-Glucosidase. J. Am. Chem. Soc. 1988, 110 (25), 8551-8553. 110. Withers, S. G.; Warren, R. A. J.; Street, I. P.; Rupitz, K.; Kempton, J. B.; Aebersold, R., Unequivocal Demonstration of the Involvement of a Glutamate Residue as a Nucleophile in the Mechanism of a Retaining Glycosidase. J. Am. Chem. Soc. 1990, 112 (15), 5887-5889. 111. Braun, C.; Brayer, G. D.; Withers, S. G., Mechanism-based inhibition of yeast alpha-glucosidase and human pancreatic alpha-amylase by a new class of inhibitors. 2-Deoxy-2,2-difluoro-alpha-glycosides. J. Biol. Chem. 1995, 270 (45), 26778-81. 112. McCarter, J. D.; Withers, S. G., Unequivocal identification of Asp-214 as the catalytic nucleophile of Saccharomyces cerevisiae alpha-glucosidase using 5-fluoro glycosyl fluorides. J. Biol. Chem. 1996, 271 (12), 6889-94.  155 113. Zhang, R.; McCarter, J. D.; Braun, C.; Yeung, W.; Brayer, G. D.; Withers, S. G., Synthesis and testing of 2-deoxy-2,2-dihaloglycosides as mechanism-based inhibitors of alpha-glycosidases. J. Org. Chem. 2008, 73 (8), 3070-7. 114. Zhang, R.; Li, C.; Williams, L. K.; Rempel, B. P.; Brayer, G. D.; Withers, S. G., Directed "in situ" inhibitor elongation as a strategy to structurally characterize the covalent glycosyl-enzyme intermediate of human pancreatic alpha-amylase. Biochemistry 2009, 48 (45), 10752-64. 115. Rempel, B. P.; Withers, S. G., Covalent inhibitors of glycosidases and their applications in biochemistry and biology. Glycobiology 2008, 18 (8), 570-586. 116. Asada, Y.; Hirayama, Y.; Furuya, T., Acylated flavonols from Crocosmia crocosmiiflora. Phytochemistry 1988, 27 (5), 1497-501. 117. Tarling, C. A.; Woods, K.; Zhang, R.; Brastianos, H. C.; Brayer, G. D.; Andersen, R. J.; Withers, S. G., The search for novel human pancreatic -amylase inhibitors: high-throughput screening of terrestrial and marine natural product extracts. ChemBioChem 2008, 9 (3), 433-438. 118. Yuen, V. G.; Coleman, J.; Withers, S. G.; Andersen, R. J.; Brayer, G. D.; Mustafa, S.; McNeill, J. H., Glucose lowering effect of montbretin A in Zucker Diabetic Fatty rats. Mol. Cell. Biochem. 2016, 411 (1-2), 373-381. 119. Day, A. G.; Withers, S. G., The purification and characterization of a beta-glucosidase from Alcaligenes faecalis. Biochem. Cell Biol. 1986, 64 (9), 914-22. 120. Namchuk, M. N.; Withers, S. G., Mechanism of Agrobacterium -glucosidase: kinetic analysis of the role of noncovalent enzyme/substrate interactions. Biochemistry 1995, 34 (49), 16194-202. 121. (a) Dunlap, W. J.; Hagen, R. E.; Wender, S. H., Preparation and properties of rhamnosidase and glucosidase fractions from a fungal flavonoid glycosidase preparation, "Naringinase C-100". J. Food Sci. 1962, 27, 597-601; (b) Romero, C.; Manjon, A.; Bastida, J.; Iborra, J. L., A method for assaying the rhamnosidase activity of naringinase. Anal. Biochem. 1985, 149 (2), 566-71; (c) Young, N. M.; Johnston, R. A. Z.; Richards, J. C., Purification of the -L-rhamnosidase of Penicillium decumbens and characterization of two glycopeptide components. Carbohydr. Res. 1989, 191 (1), 53-62. 122. Ting, S. V., Fruit Glycoside Hydrolysis, Enzymatic hydrolysis of naringin in grapefruit. J. Agric. Food Chem. 1958, 6 (7), 546-9. 123. Nersissian, A. M.; Mehrabian, Z. B.; Nalbandyan, R. M.; Hart, P. J.; Fraczkiewicz, G.; Czernuszewicz, R. S.; Bender, C. J.; Peisach, J.; Herrmann, R. G.; Valentine, J. S., Cloning, expression, and spectroscopic characterization of Cucumis sativus stellacyanin in its nonglycosylated form. Protein Sci. 1996, 5 (11), 2184-2192. 124. Williams, L. K.; Zhang, X. H.; Caner, S.; Tysoe, C.; Nguyen, N. T.; Wicki, J.; Williams, D. E.; Coleman, J.; McNeill, J. H.; Yuen, V.; Andersen, R. J.; Withers, S. G.; Brayer, G. D., The amylase inhibitor montbretin A reveals a new glycosidase inhibition motif. Nat. Chem. Biol. 2015, 11 (9), 691-696. 125. Sim, L.; Willemsma, C.; Mohan, S.; Naim, H. Y.; Pinto, B. M.; Rose, D. R., Structural basis for substrate selectivity in human maltase-glucoamylase and sucrase-isomaltase N-terminal domains. J. Biol. Chem. 2010, 285 (23), 17763-70. 126. Van Beers, E. H.; Buller, H. A.; Grand, R. J.; Einerhand, A. W.; Dekker, J., Intestinal brush border glycohydrolases: structure, function, and development. Crit. Rev. Biochem. Mol. Biol. 1995, 30 (3), 197-262. 127. Sim, L.; Quezada-Calvillo, R.; Sterchi, E. E.; Nichois, B. L.; Rose, D. R., Human intestinal maltase-glucoamylase: Crystal structure of the N-terminal catalytic subunit and basis of inhibition and substrate specificity. J. Mol. Biol. 2008, 375 (3), 782-792. 128. Vocadlo, D. J.; Davies, G. J., Mechanistic insights into glycosidase chemistry. Curr. Opin. Chem. Biol. 2008, 12 (5), 539-555. 129. Heightman, T. D.; Vasella, A. T., Recent insights into inhibition, structure, and mechanism of configuration-retaining glycosidases. Angew Chem Int Edit 1999, 38 (6), 750-770.  156 130. Gloster, T. M.; Vocadlo, D. J., Developing inhibitors of glycan processing enzymes as tools for enabling glycobiology. Nat. Chem. Biol. 2012, 8 (8), 683-694. 131. Stubbs, K. A., Activity-based proteomics probes for carbohydrate-processing enzymes: current trends and future outlook. Carbohydr. Res. 2014, 390, 9-19. 132. McCarter, J. D.; Adam, M. J.; Hartman, N. G.; Withers, S. G., In vivo inhibition of beta-glucosidase and beta-mannosidase activity in rats by 2-deoxy-2-fluoro-beta-glycosyl fluorides and recovery of activity in vivo and in vitro. Biochem. J. 1994, 301 ( Pt 2), 343-8. 133. Witte, M. D.; van der Marel, G. A.; Aerts, J. M. F. G.; Overkleeft, H. S., Irreversible inhibitors and activity-based probes as research tools in chemical glycobiology. Org. Biomol. Chem. 2011, 9 (17), 5908-5926. 134. Kallemeijn, W. W.; Witte, M. D.; Wennekes, T.; Aerts, J. M., Mechanism-based inhibitors of glycosidases: design and applications. Adv. Carbohydr. Chem. Biochem. 2014, 71, 297-338. 135. Vitner, E. B.; Platt, F. M.; Futerman, A. H., Common and Uncommon Pathogenic Cascades in Lysosomal Storage Diseases. J. Biol. Chem. 2010, 285 (27), 20423-20427. 136. Daniels, L. B.; Glew, R. H.; Radin, N. S.; Vunnam, R. R., A Revised Fluorometric Assay for Gauchers-Disease Using Conduritol-Beta-Epoxide with Liver as the Source of Beta-Glucosidase. Clin. Chim. Acta 1980, 106 (2), 155-163. 137. Hays, W. S.; Wheeler, D. E.; Eghtesad, B.; Glew, R. H.; Johnston, D. E., Expression of cytosolic beta-glucosidase in guinea pig liver cells. Hepatology 1998, 28 (1), 156-163. 138. Hermans, M. M. P.; Kroos, M. A.; Vanbeeumen, J.; Oostra, B. A.; Reuser, A. J. J., Human Lysosomal Alpha-Glucosidase - Characterization of the Catalytic Site. J. Biol. Chem. 1991, 266 (21), 13507-13512. 139. Quaroni, A.; Gershon, E.; Semenza, G., Affinity Labeling of Active-Sites in Sucrase-Isomaltase Complex from Small-Intestine. J. Biol. Chem. 1974, 249 (20), 6424-6433. 140. Braun, H.; Legler, G.; Deshusses, J.; Semenza, G., Stereospecific Ring-Opening of Conduritol-B-Epoxide by an Active-Site Aspartate Residue of Sucrase-Isomaltase. Biochim. Biophys. Acta 1977, 483 (1), 135-140. 141. Dinur, T.; Osiecki, K. M.; Legler, G.; Gatt, S.; Desnick, R. J.; Grabowski, G. A., Human Acid Beta-Glucosidase - Isolation and Amino-Acid-Sequence of a Peptide Containing the Catalytic Site. Proc. Natl. Acad. Sci. U. S. A. 1986, 83 (6), 1660-1664. 142. Legler, G.; Harder, A., Amino-Acid Sequence at Active-Site Beta-Glucosidase-a from Bitter Almonds. Biochim. Biophys. Acta 1978, 524 (1), 102-108. 143. Miao, S. C.; Mccarter, J. D.; Grace, M. E.; Grabowski, G. A.; Aebersold, R.; Withers, S. G., Identification of Glu(340) as the Active-Site Nucleophile in Human Glucocerebrosidase by Use of Electrospray Tandem Mass-Spectrometry. J. Biol. Chem. 1994, 269 (15), 10975-10978. 144. He, S. M.; Withers, S. G., Assignment of sweet almond beta-glucosidase as a family 1 glycosidase and identification of its active site nucleophile. J. Biol. Chem. 1997, 272 (40), 24864-24867. 145. Caron, G.; Withers, S. G., Conduritol Aziridine - a New Mechanism-Based Glucosidase Inactivator. Biochem. Biophys. Res. Commun. 1989, 163 (1), 495-499. 146. Atsumi, S.; Umezawa, K.; Iinuma, H.; Naganawa, H.; Nakamura, H.; Iitaka, Y.; Takeuchi, T., Production, Isolation and Structure Determination of a Novel Beta-Glucosidase Inhibitor, Cyclophellitol, from Phellinus Sp. J. Antibiot. 1990, 43 (1), 49-53. 147. Atsumi, S.; Iinuma, H.; Nosaka, C.; Umezawa, K., Biological activities of cyclophellitol. J. Antibiot. 1990, 43 (12), 1579-85. 148. Withers, S. G.; Umezawa, K., Cyclophellitol - a Naturally-Occurring Mechanism-Based Inactivator of Beta-Glucosidases. Biochem. Biophys. Res. Commun. 1991, 177 (1), 532-537. 149. Atsumi, S.; Nosaka, C.; Ochi, Y.; Iinuma, H.; Umezawa, K., Inhibition of Experimental Metastasis by an Alpha-Glucosidase Inhibitor, 1,6-Epi-Cyclophellitol. Cancer Res. 1993, 53 (20), 4896-4899.  157 150. Witte, M. D.; Kallemeijn, W. W.; Aten, J.; Li, K. Y.; Strijland, A.; Donker-Koopman, W. E.; van den Nieuwendijk, A. M. C. H.; Bleijlevens, B.; Kramer, G.; Florea, B. I.; Hooibrink, B.; Hollak, C. E. M.; Ottenhoff, R.; Boot, R. G.; van der Marel, G. A.; Overkleeft, H. S.; Aerts, J. M. F. G., Ultrasensitive in situ visualization of active glucocerebrosidase molecules. Nat. Chem. Biol. 2010, 6 (12), 907-913. 151. Witte, M. D.; Walvoort, M. T. C.; Li, K. Y.; Kallemeijn, W. W.; Donker-Koopman, W. E.; Boot, R. G.; Aerts, J. M. F. G.; Codee, J. D. C.; van der Marel, G. A.; Overkleeft, H. S., Activity-Based Profiling of Retaining beta-Glucosidases: A Comparative Study. ChemBioChem 2011, 12 (8), 1263-1269. 152. Gloster, T. M.; Madsen, R.; Davies, G. J., Structural basis for cyclophellitol inhibition of a beta-glucosidase. Org. Biomol. Chem. 2007, 5 (3), 444-6. 153. Tai, V. W. F.; Fung, P. H.; Wong, Y. S.; Shing, T. K. M., Kinetic-Studies on Cyclophellitol Analogs - Mechanism-Based Inactivators. Biochem. Biophys. Res. Commun. 1995, 213 (1), 175-180. 154. (a) Li, K. Y.; Jiang, J.; Witte, M. D.; Kallemeijn, W. W.; Donker-Koopman, W. E.; Boot, R. G.; Aerts, J. M.; Codee, J. D.; van der Marel, G. A.; Overkleeft, H. S., Exploring functional cyclophellitol analogues as human retaining beta-glucosidase inhibitors. Org. Biomol. Chem. 2014, 12 (39), 7786-91; (b) Li, K. Y.; Jiang, J. B.; Witte, M. D.; Kallemeijn, W. W.; van den Elst, H.; Wong, C. S.; Chander, S. D.; Hoogendoorn, S.; Beenakker, T. J. M.; Codee, J. D. C.; Aerts, J. M. F. G.; van der Marel, G. A.; Overkleeft, H. S., Synthesis of Cyclophellitol, Cyclophellitol Aziridine, and Their Tagged Derivatives. Eur. J. Org. Chem. 2014,  (27), 6030-6043. 155. Nakai, H.; Dilokpimol, A.; Abou Hachem, M.; Svensson, B., Efficient one-pot enzymatic synthesis of alpha-(1,4)-glucosidic disaccharides through a coupled reaction catalysed by Lactobacillus acidophilus NCFM maltose phosphorylase. Carbohydr. Res. 2010, 345 (8), 1061-1064. 156. Tsumuraya, Y.; Brewer, C. F.; Hehre, E. J., Substrate-Induced Activation of Maltose Phosphorylase - Interaction with the Anomeric Hydroxyl Group of Alpha-Maltose and Alpha-D-Glucose Controls the Enzymes Glucosyltransferase Activity. Arch. Biochem. Biophys. 1990, 281 (1), 58-65. 157. MacDonald, D. L., Method in carbohydrate chemistry, Volume VI: General carbohydrate method. Elsevier Science: New York and London, 2012; p 389-395. 158. Nakai, H.; Baumann, M. J.; Petersen, B. O.; Westphal, Y.; Schols, H.; Dilokpimol, A.; Hachem, M. A.; Lahtinen, S. J.; Duus, J. O.; Svensson, B., The maltodextrin transport system and metabolism in Lactobacillus acidophilus NCFM and production of novel alpha-glucosides through reverse phosphorolysis by maltose phosphorylase. FEBS J. 2009, 276 (24), 7353-7365. 159. Williams, L. K.; Li, C.; Withers, S. G.; Brayer, G. D., Order and disorder: differential structural impacts of myricetin and ethyl caffeate on human amylase, an antidiabetic target. J. Med. Chem. 2012, 55 (22), 10177-86. 160. Kobayashi, M.; Saburi, W.; Nakatsuka, D.; Hondoh, H.; Kato, K.; Okuyama, M.; Mori, H.; Kimura, A.; Yao, M., Structural insights into the catalytic reaction that is involved in the reorientation of Trp238 at the substrate-binding site in GH13 dextran glucosidase. FEBS Lett. 2015, 589 (4), 484-489. 161. Uitdehaag, J. C. M.; Mosi, R.; Kalk, K. H.; van der Veen, B. A.; Dijkhuizen, L.; Withers, S. G.; Dijkstra, B. W., X-ray structures along the reaction pathway of cyclodextrin glycosyltransferase elucidate catalysis in the alpha-amylase family. Nat. Struct. Biol. 1999, 6 (5), 432-436. 162. Jensen, M. H.; Mirza, O.; Albenne, C.; Remaud-Simeon, M.; Monsan, P.; Gajhede, M.; Skov, L. K., Crystal structure of the covalent intermediate of amylosucrase from Neisseria polysaccharea. Biochemistry 2004, 43 (11), 3104-3110. 163. Numao, S.; Damager, I.; Li, C. M.; Wrodnigg, T. M.; Begum, A.; Overall, C. M.; Brayer, G. D.; Withers, S. G., In situ extension as an approach for identifying novel alpha-amylase inhibitors. J. Biol. Chem. 2004, 279 (46), 48282-48291. 164. Uchida, R.; Nasu, A.; Tokutake, S.; Kasai, K.; Tobe, K.; Yamaji, N., Synthesis of new N-containing maltooligosaccharides, alpha-amylase inhibitors, and their biological activities. Chem. Pharm. Bull. 1999, 47 (2), 187-193.  158 165. Rydberg, E. H.; Sidhu, G.; Vo, H. C.; Hewitt, J.; Cote, H. C.; Wang, Y.; Numao, S.; MacGillivray, R. T.; Overall, C. M.; Brayer, G. D.; Withers, S. G., Cloning, mutagenesis, and structural analysis of human pancreatic alpha-amylase expressed in Pichia pastoris. Protein Sci. 1999, 8 (3), 635-43. 166. Takada, M.; Ogawa, K.; Saito, S.; Murata, T.; Usui, T., Chemo-enzymatic synthesis of galactosylmaltooligosaccharidonolactone as a substrate analogue inhibitor for mammalian alpha-amylase. J. Biochem. (Tokyo) 1998, 123 (3), 508-515. 167. Takada, M.; Ogawa, K.; Murata, T.; Usui, T., Inhibition of human alpha-amylases by synthetic 63-O-benzyl- and 63-O-beta-D-galactosyl maltotrionolactones. J. Carbohydr. Chem. 1999, 18 (2), 149-163. 168. Perez, S.; Bertoft, E., The molecular structures of starch components and their contribution to the architecture of starch granules: a comprehensive review. Starch/Staerke 62 (8), 389-420. 169. Bertoft, E.; Manelius, R., A Method for the Study of the Enzymatic-Hydrolysis of Starch Granules. Carbohydr. Res. 1992, 227, 269-283. 170. Cockburn, D. S., B., Surface binding sites in carbohydrate active enzymes: an emerging picture of structural and funcational diversity. In SPR Carbohydrate Chemistry - Chemical and Biological Approaches. Royal Society of Chemistry: Cambridge, United Kingdom, 2013; Vol. 39, p 204-221. 171. Cuyvers, S.; Dornez, E.; Delcour, J. A.; Courtin, C. M., Occurrence and functional significance of secondary carbohydrate binding sites in glycoside hydrolases. Crit. Rev. Biotechnol. 2012, 32 (2), 93-107. 172. Cockburn, D.; Wilkens, C.; Ruzanski, C.; Andersen, S.; Nielsen, J. W.; Smith, A. M.; Field, R. A.; Willemoes, M.; Abou Hachem, M.; Svensson, B., Analysis of surface binding sites (SBSs) in carbohydrate active enzymes with focus on glycoside hydrolase families 13 and 77-a mini-review. Biologia (Bratisl.) 2014, 69 (6), 705-712. 173. Wilkens, C., Cockburn, D., Andersen, S., Svensson, B., Analysis of Surface Binding Sites (SBS) within GH62, G13, and GH77. J. Appl. Glycosci 2015, 62, 87-93. 174. Ye, Z. M.; Miyake, H.; Tatsumi, M.; Nishimura, S.; Nitta, Y., Two additional carbohydrate-binding sites of beta-amylase from Bacillus cereus var. mycoides are involved in hydrolysis and raw starch-binding. J. Biochem. (Tokyo) 2004, 135 (3), 355-363. 175. Nielsen, M. M.; Bozonnet, S.; Seo, E. S.; Motyan, J. A.; Andersen, J. M.; Dilokpimol, A.; Abou Hachem, M.; Gyemant, G.; Naested, H.; Kandra, L.; Sigurskjold, B. W.; Svensson, B., Two Secondary Carbohydrate Binding Sites on the Surface of Barley alpha-Amylase 1 Have Distinct Functions and Display Synergy in Hydrolysis of Starch Granules. Biochemistry 2009, 48 (32), 7686-7697. 176. Sevcik, J.; Hostinova, E.; Solovicova, A.; Gasperik, J.; Dauter, Z.; Wilson, K. S., Structure of the complex of a yeast glucoamylase with acarbose reveals the presence of a raw starch binding site on the catalytic domain. FEBS J. 2006, 273 (10), 2161-2171. 177. Ludwiczek, M. L.; Heller, M.; Kantner, T.; McIntosh, L. P., A secondary xylan-binding site enhances the catalytic activity of a single-domain family 11 glycoside hydrolase. J. Mol. Biol. 2007, 373 (2), 337-354. 178. Ragunath, C.; Manuel, S. G. A.; Venkataraman, V.; Sait, H. B. R.; Kasinathan, C.; Ramasubbu, N., Probing the Role of Aromatic Residues at the Secondary Saccharide-Binding Sites of Human Salivary alpha-Amylase in Substrate Hydrolysis and Bacterial Binding. J. Mol. Biol. 2008, 384 (5), 1232-1248. 179. Allouch, J.; Helbert, W.; Henrissat, B.; Czjzek, M., Parallel substrate binding sites in a beta-agarase suggest a novel mode of action on double-helical agarose. Structure 2004, 12 (4), 623-632. 180. Bozonnet, S.; Jensen, M. T.; Nielsen, M. M.; Aghajari, N.; Jensen, M. H.; Kramhoft, B.; Willemoes, M.; Tranier, S.; Haser, R.; Svensson, B., The 'pair of sugar tongs' site on the non-catalytic domain C of barley alpha-amylase participates in substrate binding and activity. FEBS J. 2007, 274 (19), 5055-5067. 181. Larson, S. B.; Day, J. S.; McPherson, A., X-ray Crystallographic Analyses of Pig Pancreatic alpha-Amylase with Limit Dextrin, Oligosaccharide, and alpha-Cyclodextrin. Biochemistry 2010, 49 (14), 3101-3115.  159 182. Skov, L. K.; Mirza, O.; Sprogoe, D.; Dar, I.; Remaud-Simeon, M.; Albenne, C.; Monsan, P.; Gajhede, M., Oligosaccharide and sucrose complexes of amylosucrase - Structural implications for the polymerase activity. J. Biol. Chem. 2002, 277 (49), 47741-47747. 183. Fujii, K.; Minagawa, H.; Terada, Y.; Takaha, T.; Kuriki, T.; Shimada, J.; Kaneko, H., Function of second glucan binding site including tyrosines 54 and 101 in Thermus aquaticus amylomaltase. J. Biosci. Bioeng. 2007, 103 (2), 167-173. 184. Diaz, A.; Martinez-Pons, C.; Fita, I.; Ferrer, J. C.; Guinovart, J. J., Processivity and Subcellular Localization of Glycogen Synthase Depend on a Non-catalytic High Affinity Glycogen-binding Site. J. Biol. Chem. 2011, 286 (21), 18505-18514. 185. Albenne, C.; Skov, L. K.; Tran, V.; Gajhede, M.; Monsan, P.; Remaud-Simeon, M.; Andre-Leroux, G., Towards the molecular understanding of glycogen elongation by amylosucrase. Proteins-Structure Function and Bioinformatics 2007, 66 (1), 118-126. 186. Robyt, J. F.; French, D., Multiple Attack Hypothesis of Alpha-Amylase Action - Action of Porcine Pancreatic Human Salivary and Aspergillus Oryzae Alpha-Amylases. Arch. Biochem. Biophys. 1967, 122 (1), 8-&. 187. Patrick, W. M.; Nakatani, Y.; Cutfield, S. M.; Sharpe, M. L.; Ramsay, R. J.; Cutfield, J. F., Carbohydrate binding sites in Candida albicans exo-beta-1,3-glucanase and the role of the Phe-Phe 'clamp' at the active site entrance. FEBS J. 2010, 277 (21), 4549-4561. 188. Koropatkin, N. M.; Smith, T. J., SusG: A Unique Cell-Membrane-Associated alpha-Amylase from a Prominent Human Gut Symbiont Targets Complex Starch Molecules. Structure 2010, 18 (2), 200-215. 189. Loyter, A.; Schramm, M., Multimolecular Complexes of Alpha-Amylase with Glycogen Limit Dextrin - Number of Binding Sites of Enzyme and Size of Complexes. J. Biol. Chem. 1966, 241 (11), 2611-&. 190. Payan, F.; Haser, R.; Pierrot, M.; Frey, M.; Astier, J. P.; Abadie, B.; Duee, E.; Buisson, G., 3-Dimensional Structure of Alpha-Amylase from Porcine Pancreas at 5 Å Resolution - Active-Site Location. Acta Crystallographica Section B-Structural Science 1980, 36 (Feb), 416-421. 191. Larson, S. B.; Greenwood, A.; Cascio, D.; Day, J.; Mcpherson, A., Refined Molecular-Structure of Pig Pancreatic Alpha-Amylase at 2.1 Å Resolution. J. Mol. Biol. 1994, 235 (5), 1560-1584. 192. Qian, M. X.; Haser, R.; Payan, F., Carbohydrate-Binding Sites in a Pancreatic Alpha-Amylase-Substrate Complex, Derived from X-Ray Structure-Analysis at 2.1 Å Resolution. Protein Sci. 1995, 4 (4), 747-755. 193. Machius, M.; Vertesy, L.; Huber, R.; Wiegand, G., Carbohydrate and protein-based inhibitors of porcine pancreatic alpha-amylase: Structure analysis and comparison of their binding characteristics. J. Mol. Biol. 1996, 260 (3), 409-421. 194. Qian, M. X.; Spinelli, S.; Driguez, H.; Payan, F., Structure of a pancreatic alpha-amylase bound to a substrate analogue at 2.03 angstrom resolution. Protein Sci. 1997, 6 (11), 2285-2296. 195. Qian, M. X.; Nahoum, V.; Bonicel, J.; Bischoff, H.; Henrissat, B.; Payan, F., Enzyme-catalyzed condensation reaction in a mammalian alpha-amylase. High-resolution structural analysis of an enzyme-inhibitor complex. Biochemistry 2001, 40 (25), 7700-7709. 196. Qian, M. X.; Ajandouz, E.; Payan, F.; Nahoum, V., Molecular basis of the effects of chloride ion on the acid-base catalyst in the mechanism of pancreatic alpha-amylase. Biochemistry 2005, 44 (9), 3194-3201. 197. Gibson, R. M.; Svensson, B., Identification of Tryptophanyl Residues Involved in Binding of Carbohydrate Ligands to Barley Alpha-Amylase 2. Carlsberg Res. Commun. 1987, 52 (6), 373-379. 198. Kadziola, A.; Sogaard, M.; Svensson, B.; Haser, R., Molecular structure of a barley alpha-amylase-inhibitor complex: Implications for starch binding and catalysis. J. Mol. Biol. 1998, 278 (1), 205-217. 199. (a) Robert, X.; Haser, R.; Mori, H.; Svensson, B.; Aghajari, N., Oligosaccharide binding to barley alpha-amylase 1. J. Biol. Chem. 2005, 280 (38), 32968-32978; (b) Robert, X.; Haser, R.; Gottschalk, T. E.;  160 Ratajczak, F.; Driguez, H.; Svensson, B.; Aghajari, N., The structure of barley alpha-amylase isozyme 1 reveals a novel role of domain C in substrate recognition and binding: A pair of sugar tongs. Structure 2003, 11 (8), 973-984. 200. Nielsen, J. W.; Kramhoft, B.; Bozonnet, S.; Abou Hachem, M.; Stipp, S. L. S.; Svensson, B.; Willemoes, M., Degradation of the starch components amylopectin and amylose by barley alpha-amylase 1: Role of surface binding site 2. Arch. Biochem. Biophys. 2012, 528 (1), 1-6. 201. Albenne, C.; Skov, L. K.; Mirza, O.; Gajhede, M.; Feller, G.; D'Amico, S.; Andre, G.; Potocki-Veronese, G.; van der Veen, B. A.; Monsan, P.; Remaud-Simeon, M., Molecular basis of the amylose-like polymer formation catalyzed by Neisseria polysaccharea amylosucrase. J. Biol. Chem. 2004, 279 (1), 726-734. 202. Ludwiczek, M. L.; Heller, M.; Kantner, T.; McIntosh, L. P., A Secondary Xylan-binding Site Enhances the Catalytic Activity of a Single-domain Family 11 Glycoside Hydrolase. J. Mol. Biol. 2007, 373 (2), 337-354. 203. Takeo, K.; Nakamura, S., Dissociation Constants of Glucan Phosphorylases of Rabbit Tissues Studied by Polyacrylamide Gel Disk Electrophoresis. Arch. Biochem. Biophys. 1972, 153 (1), 1-7. 204. Warren, F. J.; Butterworth, P. J.; Ellis, P. R., The surface structure of a complex substrate revealed by enzyme kinetics and Freundlich constants for alpha-amylase interaction with the surface of starch. Biochim. Biophys. Acta 2013, 1830 (4), 3095-3101. 205. Cockburn, D.; Nielsen, M. M.; Christiansen, C.; Andersen, J. M.; Rannes, J. B.; Blennow, A.; Svensson, B., Surface binding sites in amylase have distinct roles in recognition of starch structure motifs and degradation. Int. J. Biol. Macromol. 2015, 75, 338-345. 206. Cuyvers, S.; Dornez, E.; Abou Hachem, M.; Svensson, B.; Hothorn, M.; Chory, J.; Delcour, J. A.; Courtin, C. M., Isothermal titration calorimetry and surface plasmon resonance allow quantifying substrate binding to different binding sites of Bacillus subtilis xylanase. Anal. Biochem. 2012, 420 (1), 90-92. 207. Cuyvers, S.; Dornez, E.; Rezaei, M. N.; Pollet, A.; Delcour, J. A.; Courtin, C. M., Secondary substrate binding strongly affects activity and binding affinity of Bacillus subtilis and Aspergillus niger GH11 xylanases. FEBS J. 2011, 278 (7), 1098-1111. 208. Quiocho, F. A., Protein-Carbohydrate Interactions - Basic Molecular-Features. Pure Appl. Chem. 1989, 61 (7), 1293-1306. 209. Lawson, C. L.; Vanmontfort, R.; Strokopytov, B.; Rozeboom, H. J.; Kalk, K. H.; Devries, G. E.; Penninga, D.; Dijkhuizen, L.; Dijkstra, B. W., Nucleotide-Sequence and X-Ray Structure of Cyclodextrin Glycosyltransferase from Bacillus-Circulans Strain-251 in a Maltose-Dependent Crystal Form. J. Mol. Biol. 1994, 236 (2), 590-600. 210. Przylas, I.; Terada, Y.; Fujii, K.; Takaha, T.; Saenger, W.; Strater, N., X-ray structure of acarbose bound to amylomaltase from Thermus aquaticus - Implications for the synthesis of large cyclic glucans. Eur. J. Biochem. 2000, 267 (23), 6903-6913. 211. Wilkens, C.; Cuesta-Seijo, J. A.; Palcic, M.; Svensson, B., Selectivity of the surface binding site (SBS) on barley starch synthase I. Biologia (Bratisl.) 2014, 69 (9), 1118-1121. 212. Higuchi, R.; Krummel, B.; Saiki, R. K., A General-Method of Invitro Preparation and Specific Mutagenesis of DNA Fragments - Study of Protein and DNA Interactions. Nucleic Acids Res. 1988, 16 (15), 7351-7367. 213. Patel, H.; Day, R.; Butterworth, P. J.; Ellis, P. R., A mechanistic approach to studies of the possible digestion of retrograded starch by alpha-amylase revealed using a log of slope (LOS) plot. Carbohydr. Polym. 2014, 113, 182-188. 214. Dhital, S.; Shrestha, A. K.; Gidley, M. J., Relationship between granule size and in vitro digestibility of maize and potato starches. Carbohydr. Polym. 2010, 82 (2), 480-488. 215. Bernfeld, P., Amylases, Alpha and Beta. Method Enzymol 1955, 1, 149-158.  161 216. Butterworth, P. J.; Warren, F. J.; Ellis, P. R., Human alpha-amylase and starch digestion: An interesting marriage. Starch-Starke 2011, 63 (7), 395-405. 217. Slaughter, S. L.; Ellis, P. R.; Butterworth, P. J., An investigation of the action of porcine pancreatic alpha-amylase on native and gelatinised starches. Bba-Gen Subjects 2001, 1525 (1-2), 29-36. 218. Tahir, R.; Ellis, P. R.; Butterworth, P. J., The relation of physical properties of native starch granules to the kinetics of amylolysis catalysed by porcine pancreatic alpha-amylase. Carbohydr. Polym. 2010, 81 (1), 57-62. 219. Buisson, G.; Duee, E.; Payan, F.; Haser, R., Alpha-amylase tertiary structures and their interactions with polysaccharides. Food Hydrocoll. 1987, 1 (5-6), 399-406. 220. Kandra, L.; Abou Hachem, M.; Gyemant, G.; Kramhoft, B.; Svensson, B., Mapping of barley alpha-amylases and outer subsite mutants reveals dynamic high-affinity subsites and barriers in the long substrate binding cleft. FEBS Lett. 2006, 580 (21), 5049-5053. 221. Lee, H. S.; Kim, M. S.; Cho, H. S.; Kim, J. I.; Kim, T. J.; Choi, J. H.; Park, C.; Lee, H. S.; Oh, B. H.; Park, K. H., Cyclomaltodextrinase, neopullulanase, and maltogenic amylase are nearly indistinguishable from each other. J. Biol. Chem. 2002, 277 (24), 21891-21897. 222. Tung, J. Y.; Chang, M. D. T.; Chou, W. I.; Liu, Y. Y.; Yeh, Y. H.; Chang, F. Y.; Lin, S. C.; Qiu, Z. L.; Sun, Y. J., Crystal structures of the starch-binding domain from Rhizopus oryzae glucoamylase reveal a polysaccharide-binding path. Biochem. J. 2008, 416, 27-36. 223. Wilkinson, A. J.; Fersht, A. R.; Blow, D. M.; Winter, G., Site-Directed Mutagenesis as a Probe of Enzyme Structure and Catalysis - Tyrosyl-Transfer Rna-Synthetase Cysteine-35 to Glycine-35 Mutation. Biochemistry 1983, 22 (15), 3581-3586. 224. Wimmerova, M.; Kozmon, S.; Necasova, I.; Mishra, S. K.; Komarek, J.; Koca, J., Stacking Interactions between Carbohydrate and Protein Quantified by Combination of Theoretical and Experimental Methods. PLoS One 2012, 7 (10), 1-9. 225. Tagami, T.; Okuyama, M.; Nakai, H.; Kim, Y. M.; Mori, H.; Taguchi, K.; Svensson, B.; Kimura, A., Key aromatic residues at subsites +2 and +3 of glycoside hydrolase family 31 alpha-glucosidase contribute to recognition of long-chain substrates. Bba-Proteins Proteom 2013, 1834 (1), 329-335. 226. Nielsen, M. M.; Seo, E. S.; Bozonnet, S.; Aghajari, N.; Robert, X.; Haser, R.; Svensson, B., Multi-site substrate binding and interplay in barley alpha-amylase 1. FEBS Lett. 2008, 582 (17), 2567-2571. 227. Pujadas, G.; Palau, J., Evolution of alpha-amylases: Architectural features and key residues in the stabilization of the  ()8 scaffold. Mol. Biol. Evol. 2001, 18 (1), 38-54. 228. Perry, G. H.; Dominy, N. J.; Claw, K. G.; Lee, A. S.; Fiegler, H.; Redon, R.; Werner, J.; Villanea, F. A.; Mountain, J. L.; Misra, R.; Carter, N. P.; Lee, C.; Stone, A. C., Diet and the evolution of human amylase gene copy number variation. Nat. Genet. 2007, 39 (10), 1256-1260. 229. Meisler, M. H.; Ting, C. N., The Remarkable Evolutionary History of the Human Amylase Genes. Crit. Rev. Oral Biol. Med. 1993, 4 (3-4), 503-509. 230. Roach, J. C.; Glusman, G.; Smit, A. F. A.; Huff, C. D.; Hubley, R.; Shannon, P. T.; Rowen, L.; Pant, K. P.; Goodman, N.; Bamshad, M.; Shendure, J.; Drmanac, R.; Jorde, L. B.; Hood, L.; Galas, D. J., Analysis of Genetic Inheritance in a Family Quartet by Whole-Genome Sequencing. Science 2010, 328 (5978), 636-639. 231. Damager, I.; Numao, S.; Chen, H. M.; Brayer, G. D.; Withers, S. G., Synthesis and characterisation of novel chromogenic substrates for human pancreatic alpha-amylase. Carbohydr. Res. 2004, 339 (10), 1727-1737. 232. Nakai, H.; Petersen, B. O.; Westphal, Y.; Dilokpimol, A.; Abou Hachem, M.; Duus, J. O.; Schols, H. A.; Svensson, B., Rational engineering of Lactobacillus acidophilus NCFM maltose phosphorylase into either trehalose or kojibiose dual specificity phosphorylase. Protein Eng. Des. Sel. 2010, 23 (10), 781-787.  162 233. Rydberg, E. H.; Sidhu, G.; Vo, H. C.; Hewitt, J.; Cote, H. C. F.; Wang, Y. L.; Numao, S.; MacGillivray, R. T. A.; Overall, C. M.; Brayer, G. D.; Withers, S. G., Cloning, mutagenesis, and structural analysis of human pancreatic alpha-amylase expressed in Pichia pastoris. Protein Science 1999, 8 (3), 635-643. 234. Glaring, M. A.; Baumann, M. J.; Abou Hachem, M.; Nakai, H.; Nakai, N.; Santelia, D.; Sigurskjold, B. W.; Zeeman, S. C.; Blennow, A.; Svensson, B., Starch-binding domains in the CBM45 family - low-affinity domains from glucan, water dikinase and alpha-amylase involved in plastidial starch metabolism. Febs J 2011, 278 (7), 1175-1185. 235. Michaelis, L.; Menten, M. L., The kinetics of the inversion effect. Biochem. Z. 1913, 49, 333-369.     163 Appendices  Appendix A: Kinetics  A1 Michaelis-Menten kinetics Michaelis and Menten proposed a two-step kinetic model to describe the behaviour of an enzyme in the presence of substrate. 235 The model involves an enzyme E binding to a substrate S reversibly to form an enzyme-substrate complex ES, in an equilibrium defined by by competition between the rates of association (k1) and dissociation (k-1). The chemical step forms product P and regenerates enzyme E, with an associated rate constant k2. This may be represented as   The central concept in the above scheme is the existence of a non-covalent complex: ES, before the enzyme turns over the substrate to generate the product. This ES complex sometimes is referred to as the Michaelis complex, in memory of Prof. Leonor Michaelis.  The initial rate of reaction 𝑣0 can be expressed as a function of product formation.  𝑣0 =𝑑[𝑃]𝑑𝑡= 𝑘2[𝐸𝑆] Equation A1.1  Since in most of the cases, the rate of the chemical reaction is much slower than the rate for E, S and ES to reach equilibrium, an assumption can be made to greatly simplify the calculation  164 and this is that under the steady state conditions, the rates of ES formation and depletion are identical. 𝑑[𝐸𝑆]𝑑𝑡=  𝑘1[𝐸][𝑆] −  𝑘−1[𝐸𝑆] − 𝑘2[𝐸𝑆] = 0 Equation A1.2 Since the total enzyme concentration [E]0 is the sum of the concentrations of free enzyme [E] and the concentration of the enzyme bound in the ES complex, [ES].   [E]0 = [E] + [ES]                                                   Equation A1.3 Solving for [ES] using Equation A1.2 and A1.3 gives [𝐸𝑆] =𝑘1[𝐸]0[𝑆]𝑘1[𝑆] +  𝑘2 +  𝑘−1 Equation A1.4 By substitution of the expression for [𝐸𝑆] from Equation A1.4 into Equation A1.1, one obtains 𝑣0 ==𝑘2[𝐸]0[𝑆]𝑘−1 + 𝑘2𝑘1+ [𝑆] Equation A1.5 The ratio of the rate constants 𝑘−1+ 𝑘2𝑘1 is defined as 𝐾m, the Michaelis constant, while the rate constant  k2 is defined as the catalytic constant, or kcat, (the turnover number). Therefore, Equation A1.5 can be simplified into a more general format, also known as the Michaelis-Menten equation: 𝑣0 =𝑑[𝑃]𝑑𝑡=𝑘cat[𝐸]0[𝑆]𝐾m + [𝑆] Equation A1.6  165 The Michaelis -Menten equation satisfyingly rationalizes the relationship between enzymatic rate and the substrate concentration in most of the cases. kcat and Km can be used to characterize the enzyme catalytic efficiency. kcat is the maximum number of substrate molecules converted to product per enzyme molecule per unit time. Mathematically, the Michaelis constant Km is the substrate concentration at which the reaction rate is at half-maximum, and is an inverse measure of the substrate's affinity for the enzyme—as a small Km value indicates a tighter binding of the substrate to the enzyme. The initial rate of enzymatic reaction 𝑣0 is dependent on both the substrate concentration [S] and total enzyme concentration [E]0. When enzyme is saturated with substrate ([S] >> Km), the value of Km becomes negligible with respect to substrate concentration and 𝑣0 asymptotically approaches its maximum value 𝑉max, maximum kinetic rate for a fixed amount of enzyme. Therefore, the Michaelis-Menten equation can be simplified to  𝑣0 = 𝑉𝑚𝑎𝑥 = 𝑘cat[𝐸]0 Equation A1.7 When [S] = Km, the initial rate 𝑣0 is half the 𝑉max value. 𝑣0 =𝑘cat[𝐸]02=𝑉𝑚𝑎𝑥2 Equation A1.8  At low concentrations ([S] << Km), [S] becomes negligible with respect to Km and 𝑣0 becomes directly proportional to [S], and the reaction becomes to a first-order reaction. 𝑣0 =𝑘cat𝐾m  [𝐸]0 [𝑆] Equation A1.9   166 The constant 𝑘𝑐𝑎𝑡𝐾𝑚 is referred to as the catalytic efficiency or specificity constant. It is a very useful index for measurement of how efficiently an enzyme converts a substrate into product and comparison of the relative specificity of one enzyme acting on different substrates. The Michaelis-Menten approach can be expanded to a more complex enzyme system where two distinct reaction steps occur, such as the double displacement mechanism for retaining glycosidases in this thesis. As shown in the following scheme, E binds reversibly to S to form the Michaelis-Menten complex ES, with an association rate constant (k1). The first step results in the formation of the covalent-enzyme intermediate (E-P) as well as release of the aglycone (ROH) and is governed by the glycosylation rate constant k2. Deglycosylation, associated with rate constant k3, involves attack of water on E-P to form free E and P.    Under steady-state conditions, the rate of formation and rate of depletion of [ES] and [E-P] are the same: 𝑑[𝐸𝑆]𝑑𝑡=  𝑑[𝐸𝑃]𝑑𝑡= 0  Equation A1.10 k1[E][S] = k2 [ES] + k-1[ES]                                    Equation A1.11 k2[ES] = k3[E-P]                                             Equation A1.12  167 The total enzyme concentration [E]0 is equal to the sum of the concentrations of all enzyme species.  [E]0 = [E] + [ES] + [E-P]                                      Equation A1.13 Substitution of the expression for [E] from Equation A1.13 into Equation A1.11 above, then arrangement to solve for [ES] yields: [𝐸𝑆] =𝑘1[𝐸]0[𝑆] − 𝑘1[𝐸𝑃][𝑆]𝑘1[𝑆] + 𝑘2 +  𝑘−1 Equation A1.14 Substitution of Equation A1.14 into Equation A1.12, then rearrangement to solve for [E-P] gives: [𝐸𝑃] =𝑘1𝑘2𝑘3[𝐸]0[𝑆]𝑘1[𝑆] + 𝑘2 +  𝑘−1 +𝑘1𝑘2𝑘3[𝑆] Equation A1.15 Therefore, the initial rate 𝑣0 is  𝑣0 =  𝑑𝑃𝑑𝑡=  𝑘3[𝐸𝑃] =𝑘1𝑘2[𝐸]0[𝑆]𝑘1[𝑆] + 𝑘2 +  𝑘−1 +𝑘1𝑘2𝑘3[𝑆] Equation A1.16 Equation A1.16 can be rearranged in the form of a Michaelis-Menten equation:  𝑣0 =𝑘2𝑘3 𝑘2 + 𝑘3[𝐸]0[𝑆]𝑘3(𝑘−1 + 𝑘2)𝑘1(𝑘2 +  𝑘3)+ [𝑆] Equation A1.17  168 The kcat value is actually the overall reaction rate of glycosylation (k2) and deglycosylation (k3) steps, and is represented by the constant  𝑘2𝑘3 𝑘2+ 𝑘3; Km by 𝑘3(𝑘−1+𝑘2)𝑘1(𝑘2+ 𝑘3).  When k2 >> k3 (deglycosylation step is the rate-limiting step): 𝑘𝑐𝑎𝑡 =𝑘2𝑘3 𝑘2 +  𝑘3≈𝑘2𝑘3 𝑘2= 𝑘3 Equation A1.17 When k3 >> k2 (glycosylation step is the rate-limiting step): 𝑘𝑐𝑎𝑡 =𝑘2𝑘3 𝑘2 +  𝑘3≈𝑘2𝑘3 𝑘3= 𝑘2 Equation A1.18 Interestingly, from the above equations, specially designed substrates which has either glycosylation or deglycosylation step rate-limiting can be used for enzyme kinetics. The overall rates measured experimentally will correspond to the rate of the slower step. A2 Enzyme kinetics in the presence of a reversible inhibitor Three modes of reversible enzyme inhibition can be distinguished by their effects on the kinetic behavior of enzymes, namely, competitive, noncompetitive or uncompetitive inhibition. i. Competitive Inhibition A competitive inhibitor competes with a normal substrate for binding to the enzyme. It therefore reduces the apparent affinity of the enzyme for its substrate (increase Km). A large excess of substrate can overcome the effect of the inhibitor, so 𝑉𝑚𝑎𝑥 is not affected.  169  Here, Ki is the dissociation constant for the enzyme-inhibitor complex EI. 𝐾𝑖 =[𝐸][𝐼][𝐸𝐼] Equation A1.19 The total concentration of enzyme is equal to [𝐸]0 = [𝐸] + [𝐸𝑆] + [𝐸𝐼].              Equation A1.20    Under the steady state conditions: 𝑑[𝐸𝑆]𝑑𝑡= 𝑘1[𝐸][𝑆] − 𝑘−1[𝐸𝑆] − 𝑘𝑐𝑎𝑡[𝐸𝑆] = 0 Equation A1.21 Solving for [E] using Equation A1.20 and A1.21 [𝐸] =[𝐸]01 +[𝑆]𝐾𝑚+[𝐼]𝐾𝑖 Equation A1.22  𝑣0 =  𝑉𝑚𝑎𝑥[𝑆]𝐾𝑚 (1 +  [𝐼]𝐾𝑖) + [𝑆] Equation A1.22  Therefore, the addition of a competitive inhibitor increases the apparent 𝐾𝑚 by a factor of (1 +[𝐼]𝐾𝑖). The value of 𝑉𝑚𝑎𝑥 is not affected by the inhibitor, since at high [𝑆], the inhibitor is displaced from the enzyme.  170 ii. Noncompetitive inhibition Noncompetitive inhibition occurs when the inhibitor and the substrate can bind simultaneously to the enzyme instead of competing for the same binding site.    If we assume that the 𝐾𝑖 does not change upon binding of the inhibitor, then under steady state conditions, the following equation can be derived: 𝑣0 =𝑉𝑚𝑎𝑥1 +  [𝐼]𝐾𝑖[𝑆]𝐾𝑚 + [𝑆] Equation A1.23 As a result, the apparent 𝑉𝑚𝑎𝑥 decreases by a factor of (1 +  [𝐼]𝐾𝑖), and the apparent 𝐾𝑚 is not affected.  iii. Uncompetitive inhibitions An uncompetitive inhibitor binds only to the enzyme-substrate complex.  171  Again, using the steady state approximation of 𝑑[𝐸𝑆]𝑑𝑡= 0 and 𝑣0 = kcat[ES], one obtains 𝑣0 =𝑉𝑚𝑎𝑥1 +  [𝐼]𝐾𝑖[𝑆]𝐾𝑚1 +  [𝐼]𝐾𝑖+ [𝑆] Equation A1.24  An uncompetitive inhibitor decreases the value of both the 𝐾𝑚 and 𝑉𝑚𝑎𝑥 by a factor (1 +  [𝐼]𝐾𝑖). The three types of reversible inhibition can be distinguished from each other by observing their corresponding Lineweaver-Burke plot (1/𝑣 vs. 1/[S]) or a single reciprocal Dixon plot (1/V vs. [I]).  172  Figure A-1 Graphical representation for the determination of the modes of inhibition with a Lineweaver-Burke plot. (a) Competitive inhibition, (b) noncompetitive inhibition and (c) uncompetitive inhibiton.  173  Figure A-2 Graphical representation for the determination of the modes of inhibition with a Dixon plot. (a) Competitive inhibition, (b) noncompetitive inhibition and (c) uncompetitive inhibiton.   174 A3 Enzyme kinetics in the presence of a mechanism-based inhibitor  In the presence of a covalent mechanism-based inhibitor I, a retaining glycosidase E reversibly reacts with it with a dissociation constant k-1/k1, also known as Ki. The chemical glycosylation step forms a covalent glycosyl-enzyme intermediate E-I, with an inactivation rate constant ki. The intermediate can be subsequently hydrolysed to form E and product I-OH.  If the reactivation rate constant k3 is negligible compared to k2, this scheme can be simplified into:  In most of the cases, [I] >> [E], the time-dependent inactivation is very similar to Michaelis-Menten kinetics, and the initial rate of inactivation 𝑣0 can be deduced as  𝑣0 =  𝑑[𝐸 − 𝐼]𝑑𝑡=𝑘𝑖[𝐸]0[𝐼] 𝐾𝑖 + [𝐼]= 𝑘𝑜𝑏𝑠[𝐸]0 Equation A1.19 𝑘𝑜𝑏𝑠 =  𝑘𝑖[𝐼]𝐾𝑖 + [𝐼] Equation A1.20  Since the apparent rate constant of inactivation 𝑘𝑜𝑏𝑠 is dependent on the inactivation rate constant (ki), the apparent dissociation constant (Ki) and the inactivator concentration ([I]), the  175 above process can be regarded as a pseudo first-order reaction for the enzyme. Therefore, exponential decay of enzymatic activity can be observed, when enzyme is incubated with mechanism-based inactivators. By plotting the 𝑘obs against different inactivator concentrations [I], the values of ki and Ki can also be extrapolated, which are important indicators of how good an inactivator is.  In these case when Ki >> [I], [I] becomes negligible with respect to Ki, and 𝑘obs becomes directly proportional to inactivator concentration: 𝑘𝑜𝑏𝑠 =  𝑘𝑖[𝐼]𝐾𝑖 Equation A1.21  Instead of obtaining individual values of ki and Ki, only the second-order inactivation constant ki/Ki can be accurately determined.    176 Appendix B: Lineweaver-Burk plots of HPA inhibition by MbA derivatives  Figure A-3: Lineweaver-Burk plot of inhibition of HPA by MbA-G (Ki = 9.1 ± 0.8 nM).    Figure A-4 Lineweaver-Burk plot of inhibition of HPA by MbA-R (Ki =21.3 ± 2.2 nM).      Figure A-5 Lineweaver-Burk plot of inhibition of HPA by MbA-RX (Ki = 42.4 ± 5.3 nM).  1 / [CNPG3] mM0 0.2 0.4 0.6 0.8 11 / V 020406080100120V @ [I] = 0V @ [I] = 6.92 nMV @ [I] = 12.11 nMV @ [I] = 17.3 nMV @ [I] = 24.22 nMV @ [I] = 32.87 nM1 / [CNPG3] mM0 0.2 0.4 0.6 0.8 11 / Vo0100200300400500V @ [I] = 0V @ [I] = 6.675 nMV @ [I] = 26.7 nMV @ [I] = 53.4 nMV @ [I] = 100.125 nMV @ [I] = 166.875 nM1 / [CNPG3] mM-0.2 0 0.2 0.4 0.6 0.8 11 / V -20020406080100120 V @ [I] = 0V @ [I] = 19.25 nMV @ [I] = 38.5 nMV @ [I] = 61.6 nMV @ [I] = 77 nMV @ [I] = 115.5 nMV @ [I] = 154 nM 177    Figure A-6: Lineweaver -Burk plot of inhibition of HPA by MbA-GR (Ki = 79.3 ± 9.5 nM)      Figure A-7 Lineweaver-Burk plot of inhibition of HPA by MbA-GRX (Ki = 93.3 ± 7.6 nM)        Figure A-8 Lineweaver-Burk plot of inhibition of HPA by MbA-C (Ki = 0.73 ± 0.09 µM). 1 / [CNPG3] mM0 0.2 0.4 0.6 0.8 11 / V 0306090120150180V @ [I] = 0V @ [I] = 12.3 nMV @ [I] = 49.2 nMV @ [I] = 92.25 nMV @ [I] = 184.5 nM-0.2 0 0.2 0.4 0.6 0.8 1 1.21 / V 020406080100120V @ [I] = 0V @ [I] = 21.7 nMV @ [I] = 54.25 nMV @ [I] = 86.8 nMV @ [I] = 130.2 nMV @ [I] = 217 nM1 / [CNPG3] mM0 0.5 1 1.51 / V0 (A400/min)050100150200V @ [I] = 0V @ [I] = 0.25 uMV @ [I] = 0.75 uMV @ [I] = 1.5 uMV @ [I] = 3 uM 178     Figure A-9 Lineweaver-Burk plot of inhibition of HPA by MbA-CG (Ki = 2.2 ± 0.2 µM).        Figure A-10 Lineweaver-Burk plot of inhibition of HPA by MbA-CR (Ki = 46.8 + 3.9 µM)        Figure A-11 Lineweaver-Burk plot of inhibition of HPA by MbA-CGR (Ki = 128 ± 15 µM)  1 / [CNPG3] mM-0.2 0 0.2 0.4 0.6 0.8 11 / V 0100200300400V @ [I] = 0V @ [I] = 0.337 uMV @ [I] = 0.632 uMV @ [I] = 1.475 uMV @ [I] = 2.951 uMV @ [I] = 5.901 uM1 / [CNPG3] mM-0.2 0 0.2 0.4 0.6 0.8 1 1.21 / V 050100150200V @ [I] = 0V @ [I] = 9.68 uMV @ [I] = 18.15 uMV @ [I] = 36.3 uMV @ [I] = 66.55 uMV @ [I] = 133.1 uM1 / [CNPG3] mM0 0.2 0.4 0.6 0.8 1 1.21 / V 0306090120150180V @ [I] = 0V @ [I] = 14.43 uMV @ [I] = 50.51 uMV @ [I] = 101  uMV @ [I] = 202 uMV @ [I] = 360.8 uM


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items