UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Fluoro-glycosyl acridinones as sensitive active site titrating agents for glycosidases Duo, Tianmeng 2013

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

Item Metadata

Download

Media
24-ubc_2013_spring_duo_tianmeng.pdf [ 4.63MB ]
Metadata
JSON: 24-1.0073565.json
JSON-LD: 24-1.0073565-ld.json
RDF/XML (Pretty): 24-1.0073565-rdf.xml
RDF/JSON: 24-1.0073565-rdf.json
Turtle: 24-1.0073565-turtle.txt
N-Triples: 24-1.0073565-rdf-ntriples.txt
Original Record: 24-1.0073565-source.json
Full Text
24-1.0073565-fulltext.txt
Citation
24-1.0073565.ris

Full Text

FLUORO-GLYCOSYL ACRIDINONES AS SENSITIVE ACTIVE SITE TITRATING AGENTS FOR GLYCOSIDASES  by  Tianmeng Duo  B.Sc., The University of British Columbia, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  The Faculty of Graduate Studies  (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  January 2013  © Tianmeng Duo, 2013  Abstract  9H-(1,3-dichloro-9,9-dimethylacridin-2-one) (DDAO) is a fluorescent and chromophoric molecule with a relatively low pKa (5.3) and moderate water solubility. DDAO was evaluated as a leaving group for mechanism-based inhibitors and a suitable reporter molecule for active site titrations of glycosidases. The Koenigs-Knorr coupling of DDAO to a protected α-glucosyl bromide resulted in two connectivities of DDAO to the glycone – DDAOY β-D-glucopyranoside (DDAOY-Glc) with chlorines close to the glycone and DDAOR β-D-glucopyranoside (DDAOR-Glc) with the chlorines distal. Kinetic evaluation of DDAOY-Glc and DDAOR-Glc revealed that both compounds were excellent substrates for a model β-glucosidase from an Agrobacterium sp. (Abg), reacting at or near diffusion-controlled rates. Several novel fluorogenic DDAO 2-deoxy-2-fluoro-β-D-glycopyranosides were synthesized and kinetic parameters for inactivation of a variety of retaining glycosidases were determined. DDAOY 2-deoxy-2-fluoro-β-D-glucopyranoside (DDAOY-2FGlc) was found to be the most effective inactivator of Abg reported to date (ki/Ki = 106 mM-1 min-1), also reacting at, or near to, diffusion controlled rates and approximately 106-fold faster than the DDAOR analogue (DDAOR-2FGlc). Differences in reactivity are partially due to the different inherent reactivities of DDAOY and DDAOR aglycones since the rate constant for spontaneous hydrolysis of DDAOY-Glc was found to be 700 times greater than that for DDAOR-Glc. DDAOY 2-deoxy-2fluoro-β-D-galactopyranosides, D-xylopyranosides and cellobiosides were also shown to be effective inactivators of a variety of cognate glycosidases, including Escherichia coli βgalactosidase (LacZ), Trichoderma reesei endo-glucanase I (EG I), Cellulomonas fimi endoxylanase (Cex), and Bacillus halodurans β-xylosidase (β-Xyl). Finally, DDAOY-2FGlc was shown to be a valuable tool for determining the concentration of Abg by active site titration. Through fluorescence detection, enzyme concentrations down to 3 nM could be reliably measured. DDAOY fluorosugar inactivators and DDAOR glycosides may serve as excellent substrates and probes in metagenomic and directed evolution screens.  ii  Table of Contents Abstract........................................................................................................................................................ ii Table of Contents ....................................................................................................................................... iii List of Tables .............................................................................................................................................. vi List of Figures............................................................................................................................................ vii List of Schemes .......................................................................................................................................... xii List of Abbreviations ............................................................................................................................... xiii Acknowledgements .................................................................................................................................. xvi 1  Introduction ...................................................................................................................................... 1 1.1  1.2  1.3  2  Glycosidases ............................................................................................................................. 1 1.1.1  Glycosidase Classification ........................................................................................... 1  1.1.2  The Catalytic Mechanism of Retaining Glycosidases ................................................. 2  1.1.3  Michaelis-Menten Kinetics.......................................................................................... 5  Small Molecule Inhibitors of Glycosidases and Their Applications ........................................ 9 1.2.1  Non-covalent Glycosidase Inhibitors ........................................................................ 10  1.2.2  Covalent Inactivators of Glycosidases....................................................................... 11  1.2.3  Determination of Enzyme Concentrations................................................................. 15  1.2.4  Enzyme Kinetics with a Mechanism-based Inactivator............................................. 17  1.2.5  Small Molecule Latent Fluorophores ........................................................................ 18  1.2.6  Small Molecule Latent Fluorophores as Active Site Titrants .................................... 24  Aims of Thesis........................................................................................................................ 25  Results and Discussion ................................................................................................................... 26 2.1  Synthesis and Characterization............................................................................................... 26 2.1.1  Synthesis of DDAO ................................................................................................... 26  2.1.2  Characterization of DDAO ........................................................................................ 27  2.1.3  Synthesis of 2-Deoxy-2-fluoro Sugars ...................................................................... 31  iii  2.1.4  Koenigs-Knorr Glycosylation of DDAO and Per-O-acetylated α-Glucosyl Bromide ................................................................................................................................... 34  2.2  2.1.5  Koenigs-Knorr Glycosylation of DDAO and 2Fsugars............................................. 37  2.1.6  Elucidation of the Structures of DDAOY-2FGlc and DDAOR-2FGlc ..................... 39  2.1.7  UV/Vis Characterization of DDAOY and DDAOR 2FGlycosides ........................... 44  Enzymatic Evaluation............................................................................................................. 45 2.2.1  Testing DDAOY-2FGlc as an Inactivator of Abg ..................................................... 45  2.2.2  Testing DDAOR-2FGlc as an Inactivator of Abg ..................................................... 53  2.2.3  Testing DDAOY-2FGal and DDAOR-2FGal as Inactivators of Abg ....................... 55  2.2.4  Testing DDAOY-Glc, DDAOR-Glc and DAO-Glc as Substrates of Abg ................ 57  2.2.5  Measurement of Spontaneous Hydrolysis Rate Constants of DDAOY-Glc, DDAORGlc and DAO-Glc ...................................................................................................... 60  2.2.6  Testing DDAOY-2FGlc as an Inactivator of β-Glucocerebrosidase (GCase) ........... 65  2.2.7  Testing DDAOY-2FGal as an Inactivator of LacZ ................................................... 67  2.2.8  Testing DDAOY-2FXyl and DNP-2FXyl as Inactivators of B. halodurans C-125 βXylosidase (β-Xyl) .................................................................................................... 69  2.2.9  3  Testing DDAOY-2FCel as an Inactivator of Cellulases............................................ 71  2.3  Active Site Titrations of Abg ................................................................................................. 74  2.4  Summary of Inactivation Parameters ..................................................................................... 76  2.5  Conclusions and Future Directions ........................................................................................ 77  Materials and Methods................................................................................................................... 79 3.1  Generous Gifts ........................................................................................................................ 79  3.2  General Materials and Methods.............................................................................................. 79 3.2.1  General Procedure for Acetylation of Free Sugars .................................................... 79  3.2.2  General Procedure for Deacetylation......................................................................... 80  3.2.3  General Procedure for Synthesis of α-Glycosyl Bromides ........................................ 80  3.2.4  General Procedure for the Synthesis of 2Fglycosides ............................................... 80  iv  3.2.5  General Procedure for the Coupling of DDAO and Per-O-acetyl-2Fglycosyl Bromides.................................................................................................................... 81  3.3  Synthesis and Characterization............................................................................................... 81  3.4  Crystallization ........................................................................................................................ 96  3.5  Expression and Purification of Wild Type Abg ..................................................................... 97  3.6  Kinetic Analysis ..................................................................................................................... 97 3.6.1  Steady State Rate Determinations ............................................................................. 98  3.6.2  Standard Procedure for Indirect Inactivation ............................................................. 99  3.6.3  Inactivation in the Presence of a Competitive Inhibitor ............................................ 99  3.6.4  Direct Assays for Inactivation on a Standard UV/Vis Spectrophotometer ............. 100  3.6.5  Direct Assays for Inactivation by Stopped Flow ..................................................... 100  3.6.6  Rates of Spontaneous Hydrolysis ............................................................................ 100  3.7  pKa Determinations .............................................................................................................. 101  3.8  Extinction Coefficient Determinations ................................................................................. 101  3.9  Active Site Titrations of Abg ............................................................................................... 102  References ................................................................................................................................................ 103  v  List of Tables  Table 1: Kinetic parameters obtained from inactivation of Abg with 3.15 in the presence and absence of saturating amounts of substrate. ..................................................................................................... 49 Table 2: Kinetic parameters for inactivation of Abg by 3.17 using the substrate protection assay and comparison to known inactivation parameters. .............................................................................. 51 Table 3: Michaelis-Menten parameters for the hydrolysis of aryl β-glucosides by Abg. ........................... 58 Table 4: Rate constants and half-lives for spontaneous hydrolysis of select aryl β-glucosides at 50°C..... 61  Table 5: Kinetic parameters for the inactivation of LacZ by DDAOY-2FGal (4.15) compared to that by DNP-2FGal. ................................................................................................................................... 68 Table 6: Kinetic parameters for inactivation of β-Xyl by DNP-2FXyl and 5.15. ....................................... 70 Table 7: Kinetic parameters for the inactivation of Cex by 6.15 and DNP-2FCel. .................................... 72 Table 8: Characteristic properties of the four chief T. reesei cellulases. .................................................... 73 Table 9: Kinetic parameters for inactivation of selected glycosidases by DDAO 2Fglycosides. ............... 77 Table 10: Reaction conditions used in enzymatic assays............................................................................ 98  vi  List of Figures  Figure 1: General glycosidase-catalyzed hydrolysis reaction. ...................................................................... 1 Figure 2: The mechanism for a retaining β-glucosidase. This is a double displacement mechanism showing an “exploded” transition state. Figure adapted from Withers, 1987.5 ............................... 3 Figure 3: Amino acid side chains that function as catalytic nucleophiles. ................................................... 4 Figure 4: Oxazoline intermediate formed through the hydrolysis of N-acetylglucosaminides and Nacetylgalactosaminides. .................................................................................................................... 5 Figure 5: A typical Michaelis-Menten type curve illustrating the behaviour of initial rate of enzymatic reaction (ν0) as a function of substrate concentration, [S]. ............................................................... 7 Figure 6: Flowchart showing the relationship between different types of inhibition. ................................ 10 Figure 7: Examples of non-covalent glycosidase inhibitors. Pictured: acarbose (1.1), nojirimycin (1.2), isofagomine (1.3) and gluconolactone (1.4)................................................................................... 11 Figure 8: Some examples of inherently active affinity probes. Pictured: mannosyl bromoketone (1.5), Nbromoacetyl glucosamine (1.6), glucosyl triazene (1.7). ............................................................... 12 Figure 9: Examples of photoreactive affinity labels. .................................................................................. 13 Figure 10: An example of the mechanism of action of a latent quinone methide probe............................. 14 Figure 11: Examples of mechanism-based inactivators. Pictured: conduritol-β-epoxide (1.10), conduritol aziridine (1.11), 2-deoxy-2-fluoroglucopyranosyl fluoride (1.12), 2,4-dinitrophenyl 2-deoxy-2fluoroglucopyranoside (1.13). ........................................................................................................ 14 Figure 12: 7-Hydroxycoumarin. ................................................................................................................. 19 Figure 13: A water-soluble 7-hydroxycoumarin derivatized by hemicyanine. ........................................... 19 Figure 14: A coumarin-based quinone methide precursor as an imaging agent of β-galactosidase activity. ........................................................................................................................................................ 20 Figure 15: A coumarin-derived fluorogenic latent quinone methide as a probe of β-glucosidase activity. 21 Figure 16: Resorufin (1.18), resorufin α-D-mannoside (1.19), resorufin β-D-cellobioside (1.20). ............ 21 Figure 17: DAO and DAO-derivatives as probes of enzyme activity......................................................... 22 Figure 18: DDAO (1.23) and fluorogenic substrates DDAO phosphate (1.24) and DDAO β-Dgalactopyranoside (1.25). ............................................................................................................... 23  vii  Figure 19: The two tautomeric forms of DDAO. ........................................................................................ 28 Figure 20: Absorbance spectra (solid lines) of protonated (pH 2) and deprotonated (pH 12) DDAO; fluorescence emission spectrum at 600 nm excitation wavelength. ............................................... 28 Figure 21: Determination of pKa of DDAO by fluorescence using excitation/emission wavelengths of 600 nm/655 nm. The data was fitted onto a titration curve by GraFit 5. .............................................. 29 Figure 22: The linear range of fluorescence response for DDAO at a concentration range of up to 5 μM. Conditions: 10 nm/10 nm excitation and emission slit width, 600V (left), 5 nm/5 nm excitation and emission slit width, 640 V (right). ........................................................................................... 30  Figure 23: Non-linearity of fluorescence intensity as DDAO concentration increases. A self-quenching phenomenon is observed. ............................................................................................................... 30 Figure 24: Coupling constants and identification of axial or equatorial fluorines. The blue arrows show vicinal couplings. The size of the arrows indicates the relative magnitude of the coupling constants. ........................................................................................................................................ 33 Figure 25: The thermodynamic products of bromination of per-O-acetylated 2-deoxy-2-fluoro-Dcellobiose. ...................................................................................................................................... 34 Figure 27: The two major resonance contributors of deprotonated DDAO. ............................................... 35  Figure 28: 1H-NMR spectra of per-O-acetylated DDAOY-Glc (above) and per-O-acetylated DDAOR-Glc (below) in CDCl3............................................................................................................................ 36 Figure 29: A portion of the HMBC spectrum of DDAOR-2FGlc, showing the aromatic carbon range from 130 ppm to 180 ppm. The most downfield carbon resonance at 174 ppm, represented by the blue arrow, correlates to the carbon of the carbonyl functional group of the aglycone (C2). The red arrow indicates the carbon resonance of C7 of the aglycone. The structure and numbering scheme of DDAOR-2FGlc are shown......................................................................................................... 40 Figure 30: A portion of the HMBC spectrum of DDAOY-2FGlc, showing the aromatic carbon range from 120 ppm to 195 ppm. The most downfield carbon resonance at 189.0 ppm, represented by the blue arrow, correlates to the carbon of the carbonyl functional group of the aglycone (C2). The red arrow indicates the carbon resonance of C7 of the aglycone. The structure and numbering scheme of DDAOY-2FGlc are shown. .......................................................................................... 41 Figure 31: Crystal structure of per-O-acetylated 9H-(1,3-dichloro-9,9-dimethylacridin-7-one-2-yl) 3,4,6tri-O-acetyl-β-D-2-deoxy-2-fluorogalactopyranoside (per-O-acetylated DDAOY-2FGal). Colour  viii  code: gray, carbon; red, oxygen; blue, nitrogen; yellow, fluorine; green, chlorine. Hydrogens have been omitted for clarity. ................................................................................................................. 43 Figure 32: Crystal structure of per-O-acetylated 9H-(1,3-dichloro-9,9-dimethylacridin-2-one-7-yl) 3,4,6tri-O-acetyl-β-D-2-deoxy-2-fluorogalactopyranoside (per-O-acetylated DDAOR-2FGal). Colour code same as Figure 31. ................................................................................................................. 43 Figure 33: Absorbance spectra of DDAOY-2FGlc (3.15) and DDAOR-2FGlc (3.16). The λmax of DDAOY-2FGlc is 400 nm; λmax of DDAOR-2FGlc is 458 nm. Spectra were obtained in 50 mM sodium phosphate buffer, pH 6.8, 37°C. The concentrations used were not identical. .................. 44  Figure 34: The structure of DNP-2FGlc, an efficient inactivator of Abg. .................................................. 46 Figure 35: Inactivation of Abg with 3.15. a) Plot of normalized residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation. b) Plot of observed inactivation rate constants vs. inactivator concentration. ............................................................... 46 Figure 36: Inactivation of Abg by 3.15 monitored in the presence of saturating amounts of PNP-Glc. Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation, at a concentration of a) 0.556 mM, b) 1.35 mM, and c) 3.70 mM of PNP-Glc. d) Plot of observed inactivation rate constants vs. inactivator concentration at the indicated substrate concentrations.................................................................................................. 48 Figure 37: Inactivation of Abg by 3.17 monitored in the presence of saturating amounts of PNP-Glc. Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation, at a concentration of 0.56 mM (a), 0.83 mM (b), and 1.1 mM (c) of PNP-Glc. d) Plot of observed inactivation rate constants vs. inactivator concentration at the indicated substrate concentrations.................................................................................................. 50 Figure 38: Inactivation of Abg with 3.15: plot of the observed rate constants of inactivation vs. concentration of inactivator............................................................................................................ 52 Figure 39: Inactivation of Abg by 3.16. a) Plot of normalized residual enzyme activity vs. time at a range  of inactivator concentrations. b) Plot of observed inactivation rate constants vs. inactivator concentration fit to a Michaelis-Menten-like curve. ...................................................................... 54 Figure 40: Glycosylation transition state of 3.15 (top left) and 3.16 (top right). The glycosylation of Abg proceeds via a late transition state. Once ionized, the DDAO moieties of 3.15 and 3.16 are identical through electron delocalization and can be represented by the resonance hybrid structure (below). .......................................................................................................................................... 55  ix  Figure 41: Inactivation of Abg with 4.15 and 4.16. a) Plot of normalized residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an equation for exponential decay. b) Plot of observed inactivation rate constants vs. concentration of 4.15. ..................................................... 56 Figure 42: Structures of selected aryl glycosides. Compounds 3.20, 3.21 and 3.22 were tested as substrates of Abg in this work. ....................................................................................................... 57 Figure 43: Brønsted plots of log(kcat) vs. aglycone pKa for the hydrolysis of a series of aryl glucosides by Abg. Data taken from Kempton64 and Namchuk38. ........................................................................ 58 Figure 44: Brønsted plots of log(kcat/Km) vs. aglycone pKa for the hydrolysis of a series of aryl glucosides  by Abg. Data taken from Kempton and Namchuk.64,38 .................................................................. 59 Figure 45: A plot of the spontaneous hydrolysis of several compounds over time in pH 6.80 sodium phosphate buffer, 50°C. a) Absorbance vs. time; b) Rate of hydrolysis (A/min) vs. time. ............ 61 Figure 46: Plot of log(kobs) for hydrolysis of 3.20 at 50°C at a range of pH values. ................................... 62 Figure 47: Glucosides containing aryl nucleofuges as possible substrates (R=OH) or inactivators (R=F) of Abg to probe the contribution of steric strain to reaction rate........................................................ 63 Figure 48: Transition states for hydrolyses of DDAOY-Glc (3.20, top left) and DDAOR-Glc (3.21, top right) are depicted with the negative charge concentrated on the phenolate. Resonance of  DDAOY and DDAOR into a hybrid structure (below) is slow; hence the two transition states are unequal. The DDAOY anion may be more stabilized due to the inductive effects of nearby electron-withdrawing chloro groups (blue arrow). ........................................................................ 64 Figure 49: Transition state for the hydrolysis of DDAOY-Glc (3.20) showing halogen bonding interactions (hash lines) between the ortho-chlorines and the phenoxide...................................... 64 Figure 50: Glucosylceramide, a glycolipid and substrate of GCase. .......................................................... 65 Figure 51: Inactivation of GCase by 3.15. a) Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation. b) Plot of observed inactivation rate constants vs. inactivator concentration fitted to a Michaelis-Menten equation... 66 Figure 52: Efficient inactivators of GCase containing phosphate-based aglycones with two alkyl chains.67 Figure 53: Inactivation of LacZ by DDAOY-2FGal (4.15). a) Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation. b) Plot of observed inactivation rate constants vs. inactivator concentration fit to a Michaelis-Menten equation. ...... 68 Figure 54: Inactivation of β-Xyl by a) 5.15 and b) DNP-2FXyl. Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation. ......................... 69  x  Figure 55: Inactivation of β-Xyl by DDAOY-2FXyl (5.15) and DNP-2FXyl. Plot of observed inactivation rate constants vs. inactivator concentration fit to a Michaelis-Menten curve. ............................... 70 Figure 56: Inactivation of Cex by 6.15; plot of observed inactivation rate constants (kobs) vs. inactivator concentration. ................................................................................................................................. 71 Figure 57: Inactivation of T. reesei EG I by 6.15. Plot of initial rate of inactivation as a function of inactivator concentration, fitted to a Michaelis-Menten type equation. ......................................... 73 Figure 58: Active site titrations of Abg by DDAOY-2FGlc (3.15) and DNP-2FGlc (3.17). DDAO and DNP released was monitored a UV/Vis spectrophotometer at 600 nm and 400 nm respectively,  and plotted against Abg concentration. .......................................................................................... 74 Figure 59: Active site titration of Abg by 3.15 using the UV/Vis spectrophotometer. The DDAO released was monitored by absorbance at 600 nm and plotted as a function of Abg concentration. ........... 75 Figure 60: Active site titration of Abg by 3.15 on stopped flow at excitation/emission wavelengths of 600 nm/656 nm, 435 V. a) DDAO fluorescence release at a series of Abg concentrations as a function of time. No significant turnover of the covalent glycosyl-enzyme intermediate is observed. b) Plot of fluorescence response at 656 nm vs. Abg concentration. .......................................................... 75 Figure 61: Active site titration of Abg by 3.15 on the Cary Eclipse fluorimeter. Fluorescence response was measured at excitation/emission wavelengths of 600 nm/656 nm at 600 V. .......................... 76 Figure 62: Calibration curve of DDAO in the standard fluorimeter at excitation/emission wavelengths of 600 nm/656 nm. ........................................................................................................................... 102  xi  List of Schemes  Scheme 1: A simple reaction catalyzed by an enzyme in the presence of substrate. .................................... 5 Scheme 2: A kinetic scheme of the reaction catalyzed by a retaining glycosidase in the presence of substrate. .......................................................................................................................................... 8 Scheme 3: Kinetic scheme of a retaining glycosidase reacting with a mechanism-based inactivator. ....... 17 Scheme 4: Synthesis of DDAO (1.23) from commercially available starting materials. i) Mg0, CH3I, I2, Et2O; ii) NaOH, H2O, THF, 0°C; iii) Na2S2O4 (aq); iv) 2 M HCl, reflux; v) NaIO4 (aq). ............. 26  Scheme 5: General scheme for the synthesis of 2Fglycosides. i) Ac2O, 0.1 eqv NaOAc; ii) HBr, AcOH; iii) Zn, AcOH, H2O; iv) Selectfluor™, MeCN/H2O or MeNO2/H2O; v) Ac2O, pyr, DCM; vi) HBr, AcOH. ............................................................................................................................................ 32 Scheme 6: Koenigs-Knorr coupling of per-O-acetylated α-glucosyl bromide to DDAO to generate two glycoside products.......................................................................................................................... 34 Scheme 7: General scheme of the Koenigs-Knorr coupling of 2-deoxy-2-fluoro sugars to DDAO to generate 2-deoxy-2-fluoro β-glycosides......................................................................................... 37 Scheme 8: Deacetylation of per-O-acetylated DDAO 2Fglycosides. ......................................................... 38  xii  List of Abbreviations  Abg  Agrobacterium sp. β-glucosidase  Ac  Acetyl group  AcOH  Acetic acid  Ar  Aryl group  B. halodurans  Bacillus halodurans  β-Xyl  B. halodurans GH 52 β-xylosidase  CAZy  Carbohydrate active enzyme  CBH I/II  Cellobiohydrolase I/II  C. fimi  Cellulomonas fimi  CenD  C. fimi endo-glucanase D  Cex  C. fimi endo-xylanase  DAO  7-Hydroxy-9H-(9,9-dimethylacridin-2-one)  DCM  Dichloromethane  DDAO  7-Hydroxy-9H-(1,3-dichloro-9,9-dimethylacridin-2-one)  DDAOR  7-Hydroxy-9H-(1,3-dichloro-9,9-dimethylacridin-2-one)  DDAOR-2FGal  9H-(1,3-dichloro-9,9-dimethylacridin-2-one-7-yl) 2-deoxy-2fluoro-β-D-galactopyranoside  DDAOR-2FGlc  9H-(1,3-dichloro-9,9-dimethylacridin-2-one-7-yl) 2-deoxy-2fluoro-β-D-glucopyranoside  DDAOY  2-Hydroxy-9H-(1,3-dichloro-9,9-dimethylacridin-7-one)  DDAOY-2FCel  9H-(1,3-dichloro-9,9-dimethylacridin-7-one-2-yl) 2-deoxy-2fluoro-β-D-cellobioside  DDAOY-2FGal  9H-(1,3-dichloro-9,9-dimethylacridin-7-one-2-yl) 2-deoxy-2fluoro-β-D-galactopyranoside  DDAOY-2FGlc  9H-(1,3-dichloro-9,9-dimethylacridin-7-one-2-yl) 2-deoxy-2fluoro-β-D-glucopyranoside  DDAOY-2FXyl  9H-(1,3-dichloro-9,9-dimethylacridin-7-one-2-yl) 2-deoxy-2fluoro-β-D-xylopyranoside  DMF  N, N-dimethylformamide  DMSO  Dimethyl sulfoxide  DNP  2,4-Dinitrophenol  DNP-2FCel  2,4-Dinitrophenyl 2-deoxy-2-fluoro-β-D-cellobioside  xiii  DNP-2FGal  2,4-Dinitrophenyl 2-deoxy-2-fluoro-β-D-galactopyranoside  DNP-2FGlc  2,4-Dinitrophenyl 2-deoxy-2-fluoro-β-D-glucopyranoside  DNP-2FXyl  2,4-Dinitrophenyl 2-deoxy-2-fluoro-β-D-xylopyranoside  DNP-Cel  2,4-Dinitrophenyl β-D-cellobioside  DNP-Glc  2,4-Dinitrophenyl β-D-glucopyranoside  ε  Molar absorptivity  E. coli  Escherichia coli  EG I/II  Endo-glucanase I/II  E  Free enzyme  E•I-X  Michaelis complex with inactivator  E•S  Michaelis complex  ESI MS  Electrospray ionization mass spectrometry  EtOAc  Ethyl acetate  EtOH  Ethanol  F2eq/ax  Equatorial/axial C-2 fluorine  2F  2-Deoxy-2-fluoro  2FGalF  2-Deoxy-2-fluoro-β-D-galactosyl fluoride  GH  Glycoside hydolase  HRMS  High resolution mass spectrometry  Hz  Hertz  Ka  Acid dissociation constant  Kb  Base dissociation constant  Kd  Dissociation constant  kcat  Catalytic rate constant  ki  Inactivation rate constant  Ki Ki  Inactivator dissociation constant app  Apparent inactivator dissociation constant  Km  Michaelis constant for a substrate  kobs  Observed rate constant  LacZ  E. coli β-galactosidase  MeCN  Acetonitrile  MeOD  Deuterated methanol  MeOH  Methanol  MeNO2  Nitromethane  xiv  NMR  Nuclear magnetic resonance  PE  Petroleum ether  PhMe  Toluene  PNP-Glc  4-Nitrophenyl β-D-glucopyranoside  PNS  Principle of Non-perfect Synchronization  RT SelectFluor  Room temperature ®  1-Chloromethyl-4-fluoro-1,4-diazoniabicyclo[2,2,2]octane bis(tetrafluoroborate)  T. reesei  Trichoderma reesei  TLC  Thin layer chromatography  THF  Tetrahydrofuran  Tris  Tris(hydroxymethyl)aminomethane  UV-Vis  Ultraviolet-visible light  Vmax  Maximum velocity of an enzyme-catalysed reaction  xv  Acknowledgements  I would like thank my supervisor, Dr. Stephen G. Withers, for his guidance, patience and excellent humour. 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. Ethan Goddard-Borger for our good discussions (and strange trivia), Dr. Tom Wennekes for his time in introducing me to all things chemical, Dr. Hong-Ming Chen for great synthesis advice, and Ms. Emily Kwan and Dr. David Kwan for their biochemical expertise. I am also grateful to Tony Ruzzini and Dr. Rahul Singh for access to and help with the Eltis lab’s stopped flow machine. Also thanks to the departmental NMR, Mass Spectrometry and X-ray Crystallography facilities 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.  xvi  1  Introduction  1.1  Glycosidases  Glycosidases, or glycoside hydrolases, make up an average of 1% of the genome of any organism and are active players in carbohydrate metabolism. Industrially, glycosidase mixtures have been used in a variety of applications, including brewing, pulp and paper, pharmaceuticals1 and bioethanol generation.2 Chemically, glycosidases catalyse the cleavage of the acetal (C-O) bond between a sugar (glycone) and its nucleofuge. The leaving group can be a second sugar or an aglycone such as a polypeptide or a lipid. A water molecule is directed to attack at the anomeric centre, displacing the leaving group. Glycosidases can speed up the hydrolysis of glycosidic bonds by up to 17 orders of magnitude.3  Figure 1: General glycosidase-catalysed hydrolysis reaction.  1.1.1 Glycosidase Classification  Glycosidases can be classified in several ways: 1.  Substrate specificity – Glycosidase active sites have often evolved to recognize and cleave one kind of sugar with maximum efficiency. However, their specificity is rarely absolute. For example, many β-glucosidases also act on β-galactosides and βxylosides, but this promiscuous activity generally occurs at a lower rate than for the preferred substrate.  2.  Anomeric specificity – Individual glycosidases usually cleave only α- or β-linked glycosides. 1  3.  Inverting/Retaining – This classification is based on the relationship between the anomeric configurations of the substrate and product. An inverting glycosidase will produce a product with the opposite anomeric configuration to that of the substrate, whereas a retaining glycosidase will produce a product with the same configuration as the substrate.  4.  Endo/Exo-glycosidases – Glycosidases that cleave at the termini of a long chain of polysaccharides are termed exo-glycosidases, those that do so in the middle are called endo-glycosidases. Endo-glycosidases usually have a long cleft-shaped active site that forms interactions with multiple glycosyl residues.  5.  Sequence homology – The current CAZy database (www.cazy.org) has classified glycoside hydrolases into 130 families based on primary sequence homology. Usually enzymes with high amino acid homology have similar secondary and tertiary structures and also act with the same mechanism, hence catalysing reactions with the same stereochemical outcomes. However, substrate specificity can be extremely variable.  1.1.2 The Catalytic Mechanism of Retaining Glycosidases  The enzymes focused on in this study are retaining glycosidases. Their double displacement catalytic mechanism, first suggested by Koshland4 in 1953, is illustrated in Figure 2. Two residues, often with carboxylic acid functional groups, function as general acid/base and catalytic nucleophile residues. In the first chemical step, glycosylation, the oxygen of the catalytic nucleophile attacks the anomeric centre of the glycone; in a concerted motion the acid/base residue side chain donates a proton to the nucleofuge to facilitate its departure. This reaction proceeds through an oxocarbenium ion-like transition state to form a glycosyl-enzyme intermediate, where a covalent bond is formed between the enzyme and the glycone. During deglycosylation, the general acid/base residue accepts a proton from the water medium to facilitate hydrolysis and release of the glycone through another oxocarbenium ion-like transition state. In this step, the nucleofuge is the catalytic nucleophile of the enzyme; the enzyme active site is regenerated and ready for further catalysis. 2  In the glycosyl-enzyme intermediate, the anomeric stereochemistry of the glycone is opposite to that of the starting material; another SN2-like attack on the anomeric carbon by water will result in net retention of anomeric configuration.  Figure 2: The mechanism for a retaining β-glucosidase. This is a double displacement mechanism showing an “exploded” transition state. Figure adapted from Withers, 1987.5  An alternative mechanism for retaining glycosidases involved an ion-pair intermediate instead of a glycosyl-enzyme intermediate. This was proposed by Phillips in 19676, and arose from crystallographic evidence from hen egg white lysozyme (HEWL), the first crystal structure to be solved of an enzyme. The 2.0 Å resolution data suggested that there would be a distance of 3 Å between the anomeric carbon and the nearest residue, too great to be a covalent bond between the enzyme and glycosyl intermediate. Phillips’ ion-pair intermediate mechanism was contradicted by strong evidence from kinetic isotope effect studies.7 A large secondary deuterium kinetic isotope effect (>1) suggested that the hybridization change of the transformation from intermediate to product was sp3  sp2. This was consistent with the hypothesized glycosyl-enzyme intermediate (sp3-hybridized) that 3  passes through a sp2-hybridized oxocarbenium ion transition state to form product. If the ion-pair intermediate hypothesis was true, then experimental data should show an inverse secondary kinetic isotope effect (<1) corresponding to an increase in hybridization from a sp2-hybridized intermediate to a sp3-hybridized product. Further evidence from mass spectroscopy and new Xray crystallographic data have shown HEWL does indeed form a covalent glycosyl-enzyme complex during its catalytic cycle; the original ion-pair intermediate mechanism has since been revised.8 Although the double displacement mechanism involving carboxyl side chains seems to be the most prevalent mechanism among retaining β-glycosidases, variations and alternative pathways have been identified and are worthy of discussion. Instead of a carboxylic acid-containing residue, the Trypanosoma cruzi trans-sialidase (GH family 33) uses a tyrosine residue as the catalytic nucleophile.9,10 In contrast to negatively charged carboxylates, the neutral tyrosine nucleophile (structure shown in Figure 3) is thought to minimize repulsion between the negatively charged sialic acid and the enzyme active site, facilitating binding and catalysis.11  Figure 3: Amino acid side chains that function as catalytic nucleophiles.  Some retaining glycosidases of GH family 20 catalyse hydrolysis of substrates containing 2-acetamido groups without formation of a covalent glycosyl-enzyme adduct; rather, the enzymes position the substrate to facilitate an intramolecular attack on the anomeric carbon by the C2 acetamido oxygen, forming an oxazoline intermediate and releasing the nucleofuge (Figure 4).12  4  Figure 4: Oxazoline intermediate formed through the hydrolysis of N-acetylglucosaminides and N-acetylgalactosaminides.  Enzymes in GH family 4 have been shown to be metal ion and NAD+ dependent and hydrolyse both α- and β-linked activated substrates through a process of redox-assisted elimination and addition with retention of configuration.13,14 In addition, a novel mechanism elucidated for GH family 88 unsaturated glucuronyl hydrolases describes the enzyme’s role as hydrating a vinyl ether group to allow intramolecular rearrangement to cleave the glycosidic bond.15,16 The diversity and “creativity” of glycosidases are appreciated.  1.1.3 Michaelis-Menten Kinetics  In their study of invertase, Michaelis and Menten proposed a two-step kinetic model to describe the behaviour of an enzyme in the presence of substrate (Scheme 1).17,18 The rapid formation of the Michaelis complex (E•S) from free enzyme (E) and substrate (S) is represented by competition between the rates of association (k1) and dissociation (k-1). The rate-limiting chemical step forms product (P) and regenerates E, with an associated rate constant k2.  Scheme 1: A simple reaction catalysed by an enzyme in the presence of substrate.  The initial rate of reaction (ν0) can be expressed as a function of product formation. ν0 =  𝑑𝑃 = 𝑘2 [𝐸•𝑆] 𝑑𝑡  Equation 1 5  Under steady-state conditions, the concentration change of the Michaelis complex [E•S] is negligible when compared to that of the substrate.19 Thus, the rates of E•S formation and depletion are identical. 𝑘1 [𝐸][𝑆] = 𝑘−1 [𝐸•𝑆] + 𝑘2 [𝐸•𝑆]  Equation 2 The total enzyme concentration, [E]0, is defined as the sum of the concentrations of free enzyme [E] and bound enzyme [E•S]. [𝐸] = [𝐸]0 − [𝐸•𝑆]  Equation 3 Substitution of Equation 3 into Equation 2 and solving for [E•S] gives:  [𝐸•𝑆] =  𝑘1 [𝐸 ]0 [𝑆] 𝑘1 [𝑆] + (𝑘2 + 𝑘−1 ) Equation 4  Substitution of Equation 4 into Equation 1 gives the Michaelis-Menten equation (Equation 5), where relationship between the initial rate of enzymatic reaction observed (ν0) is shown to be dependent on both the substrate concentration [S] and total enzyme concentration [E]0. ν0 =  𝑘𝑐𝑎𝑡 [𝑆][𝐸]0 𝑚 + [𝑆]  Equation 5 In this equation, kcat is the first-order rate constant – in this case k2 – and describes the turnover number of the enzyme; Km is the Michaelis constant and is an apparent binding constant, defined as  . Rearrangement of Equation 2 shows Km is inversely proportional to all bound  forms of enzyme (Equation 6). The true binding constant, or dissociation constant 𝑘 ⁄𝑘 , assumes the off rate (k-1) will occur much faster than the chemical step (k2). 𝑘  𝑘 𝑘  [𝐸][𝑆] [𝐸•𝑆] Equation 6 6  ½ Vmax  Figure 5: A typical Michaelis-Menten type curve illustrating the behaviour of initial rate of enzymatic reaction (ν0) as a function of substrate concentration, [S].  The initial rate of enzymatic reaction (ν0) varies non-linearly with substrate concentration (Figure 5). When enzyme is saturated with substrate ([S] >> Km), the value of Km becomes negligible with respect to substrate concentration and the initial rate approaches its maximum value (Vmax); ν0 simplifies to: ν0 = V𝑚𝑎𝑥 = 𝑘𝑐𝑎𝑡 [𝐸]0  Equation 7 When [S] = Km, the initial rate is half the Vmax value.  ν0 =  𝑘𝑐𝑎𝑡 [𝐸]0 2  Equation 8  When the substrate concentration is much less than Km (Km >> [S]), [S] becomes negligible with respect to Km and initial rate becomes directly proportional to [S].  ν0 =  𝑘𝑐𝑎𝑡 [𝑆][𝐸]0 𝑚  Equation 9  7  The behaviour of a retaining glycosidase in the presence of a substrate is more complex due to two chemical steps (Scheme 2). E again binds reversibly to S to form the Michaelis complex (E•S). The first chemical 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.  Scheme 2: A kinetic scheme of the reaction catalysed by a retaining glycosidase in the presence of substrate.  Under steady-state conditions:  𝑑[𝐸•𝑆] 𝑑[𝐸-𝑃] = =0 𝑑𝑡 𝑑𝑡 Equation 10 Thus, the rate of formation and rate of depletion of [E•S] and [E-P] are the same:  𝑘1 [𝐸][𝑆] = 𝑘2 [𝐸•𝑆] + 𝑘−1 [𝐸•𝑆] Equation 11  𝑘2 [𝐸•𝑆] = 𝑘3 [𝐸-𝑃] Equation 12 As before, [E]0 is equal to the sum of the concentrations of all enzyme species. Solving for [E] gives:  [𝐸 ] = [𝐸 ]0 − [𝐸•𝑆] − [𝐸-𝑃] Equation 13 Substitution into Equation 11 then rearrangement to solve for [E•S] yields:  [𝐸•𝑆] =  𝑘1 [𝑆][𝐸]0 − 𝑘1 [𝑆][𝐸-𝑃] 𝑘2 + 𝑘−1 + 𝑘1 [𝑆] Equation 14  8  Substitution of Equation 14 into Equation 12, then rearrangement to solve for [E-P] gives:  [𝐸-𝑃] =  𝑘1 𝑘2 [𝐸]0 [𝑆] 𝑘 𝑘 𝑘−1 + 𝑘2 + 𝑘1 [𝑆] + 1 2 [𝑆] 𝑘3 Equation 15  Therefore, the initial rate observed is:  ν0 =  𝑑𝑃 = 𝑘3 [𝐸-𝑃] = 𝑑𝑡  𝑘1 𝑘2 [𝐸]0 [𝑆] 𝑘 𝑘 𝑘−1 + 𝑘2 + 𝑘1 [𝑆] + 1 2 [𝑆] 𝑘3  Equation 16 Equation 16 can be rearranged in the form of a Michaelis-Menten equation: 𝑘2 𝑘3 [𝑆][𝐸]0 𝑘2 + 𝑘3 ν0 = 𝑘3 (𝑘−1 + 𝑘2 ) + [𝑆] 𝑘1 (𝑘3 + 𝑘2 )  Equation 17  In the case of an enzyme reaction with two chemical steps, kcat is represented by the constants  1.2  ; Km by  .  Small Molecule Inhibitors of Glycosidases and Their Applications  Small molecule inhibitors of glycosidases are used in a variety of applications, including use as diagnostic tools for locating and imaging glycosidases.20 Furthermore, since glycosidases play an essential role in many diseases and in the metabolism of disease-causing organisms, they may also function as therapeutic targets: glycosidase inhibitors have been used for treatment of hereditary diseases such as Gauchers disease21 and diabetes22, as well as diseases caused by viruses and protists (influenza23, Chagas disease9).  9  Inhibitors interfere with enzyme action in either a non-covalent or covalent manner. Covalent inhibitors, or inactivators, are further classified into affinity labels and mechanismbased inactivators. Affinity labels contain a reactive functionality in addition to a directing moiety to impart specificity onto a particular protein. Mechanism-based inactivators are latent until acted upon by the enzyme. A simple summary is seen in Figure 6.  Figure 6: Flowchart showing the relationship between different types of inhibition.  1.2.1 Non-covalent Glycosidase Inhibitors  The most well-studied type of glycosidase inhibitor binds non-covalently to enzymes. Many of these act as competitive inhibitors and bind to the active site of the enzyme; these tend to mimic the structure of natural substrates or transition states. The structures of several competitive inhibitors are shown in Figure 7. For example, acarbose is an effective α-amylase inhibitor and is used towards the treatment of diabetes.22 Its structure closely resembles that of starch, the natural substrate of α-amylase. Isofagomine24 and nojirimycin25 are classified as iminosugars, which contain a ring-nitrogen which becomes positively charged at physiological pH, mimicking the oxocarbenium ion-like transition state. The sp2-hybridized anomeric carbon of gluconolactone25 bends the sugar ring into a conformation that is thought to also mimic the transition state.  10  Figure 7: Examples of non-covalent glycosidase inhibitors. Pictured: acarbose (1.1), nojirimycin (1.2), isofagomine (1.3) and gluconolactone (1.4).  1.2.2 Covalent Inactivators of Glycosidases  Covalent inactivators act by forming a covalent bond between the inhibitory molecule and the target enzyme to destroy enzymatic activity. The enzyme is made non-functional by blockage of the active site or modification of an important amino acid residue. These have been termed “irreversible inhibitors” or “inactivators”, since activity cannot be easily restored with dilution or even extensive dialysis. Hence, irreversible covalent inhibitors have a potential to be effective activity-based probes. There have generally been two distinct directions in the design of activity-based probes – affinity labels and mechanism-based inactivators.  11  Affinity Labels  1.2.2.1  Affinity labels contain an active-site directed portion linked to a reactive moiety to covalently modify the target enzyme. They can be further sub-categorized into those that are inherently active, and those that require a certain external stimulus to be activated, e.g. UV irradiation. Inherently active affinity probes contain reactive functional groups such as bromoketones or triazenes26 (Figure 8). These are directed to the enzyme active site by the glycone and may react with a nearby nucleophile.  Figure 8: Some examples of inherently active affinity probes. Pictured: mannosyl bromoketone (1.5), N-bromoacetyl glucosamine (1.6), glucosyl triazene (1.7).  Photoreactive affinity labels are inert until activated by ultraviolet light (Figure 9). For example, diazirines release nitrogen gas upon UV irradiation and turn into electrophilic carbenes.27 Another interesting example of a photoreactive affinity label is a bifunctional diazide probe (1.9) developed by Gandy et al. (Figure 9).28 A tight-binding iminosugar was used as the basis for glycosidase recognition. UV irradiation only activated the aromatic azide, expelling dinitrogen and converting it into an electrophilic didehydroazepine intermediate which can be subsequently attacked by a nucleophile from the enzyme. The benzyl azide was unaffected by this stimulus, leaving it free for later conjugation with a suitable reporter molecule.28,29  12  Figure 9: Examples of photoreactive affinity labels.  1.2.2.2  Mechanism-based Inactivators Mechanism-based inactivators, also called “suicide inhibitors”, are often substrate  analogues that are inert until activated by the enzymatic machinery; hence, they only act on catalytically competent enzymes. This class of compounds may, in addition to forming a covalent bond, physically block or chemically modify the active site so that no further catalytic cycles can occur. Quinone methides are often incorporated into the design of mechanism-based inactivators as they are stable when caged, yet become a highly electrophilic species once activated. Many latent glycosyl quinone methides are large compounds linked to a fluorescent or biotinylated reporter for ease of detection. One such latent quinone-methide probe, developed by Tsai et al. is shown in Figure 10.30 When cleaved, the aglycone will form a quinone-methide species that will react with potential nucleophiles. However, after separation from the glycone, the quinone methide also relinquishes specificity for the enzyme and may diffuse before reacting.  13  Figure 10: An example of the mechanism of action of a latent quinone methide probe.  Specificity is better retained if the covalent bond is formed directly with the glycone. Some reactive glycones are illustrated in Figure 11.  Figure 11: Examples of mechanism-based inactivators. Pictured: conduritol-β-epoxide (1.10), conduritol aziridine (1.11), 2-deoxy-2-fluoro-β-D-glucopyranosyl fluoride (1.12), 2,4dinitrophenyl 2-deoxy-2-fluoro-β-D-glucopyranoside (1.13).  14  The Withers group has developed a unique class of mechanism-based inhibitors for retaining glycosidases. These are fluorine-containing compounds that selectively form a covalent bond with the catalytic nucleophile of the enzyme (e.g. 1.12, 1.13). Due to the high electronegativity of fluorine, transition state energies of glycosylation and deglycosylation are elevated. A good leaving group, i.e. those with low pKa values such as fluoride or DNP, is attached to the anomeric position of these compounds and lowers the activation energy of the first glycosylation step; however, it does not affect the rate of the second step, deglycosylation, so a highly stable glycosyl enzyme intermediate is formed, sometimes with half-lives on the order of days. Activated 2-deoxy-2-fluoro-glycosides (2Fglycosides) have been used in various applications, for example as probes to identify the catalytic nucleophile of retaining glycosidases31, in functional profiling methodology32 and X-ray crystallography studies8. In this work, 2Fglycosides will be employed as mechanism-based probes towards determining the absolute concentration of enzyme in a solution.  1.2.3 Determination of Protein Concentrations  The most accurate method of determining the concentration of a protein for which a sequence is available is by quantitative amino acid analysis. This method consists of completely hydrolysing the peptide bonds in the enzyme to produce its amino acid monomers. The amino acids are then subjected to derivatisation to allow UV/Vis detection. Analysis of the hydrolysed amino acid mixture is then done by HPLC in the presence of an internal standard, usually a known amount of an unnatural amino acid such as norleucine. The absolute and relative amounts of each amino acid can thus be calculated, and the total absolute concentration of the enzyme accurately determined. The molar absorptivity (ε) at 280 nm can also be experimentally determined for future reference. However, this method is impractical, as it is a slow and expensive process that also requires a large amount of highly pure protein. If the primary sequence is already known from the DNA sequence, the approximate extinction coefficient (ε) at 280 nm can be calculated using bioinformatic tools (e.g. www.expasy.org). At this wavelength, the main amino acid contributors towards molar absorptivity are phenylalanine, tyrosine and tryptophan, with cysteine making a minor 15  contribution. Protein concentration (c) can then be determined by measuring the absorbance (A) at 280 nm of a solution of pure enzyme and quantifying this using Beer’s Law (A = εbc).This method is fast and requires only dilute solutions of protein; however, this method may be inaccurate as the absorptivity of each amino acid could be influenced by other amino acids in close proximity. Furthermore, overestimations are common since any contaminating protein will also contribute to the absorbance measured at 280 nm. Thus, this method would not be useful in quantifying the concentration of an enzyme in a mixture. Active site titration is a technique which can measure the absolute concentration of one specific enzyme in the presence of other proteins or other enzymes. The concept involves treatment of the enzyme with a mechanism-based inactivator designed to target and react with the enzyme active site of interest, resulting in accumulation of a covalently linked enzyme species and release of one equivalent of a chromophore. Since one enzyme molecule can react with only one molecule of inactivator, the amount of inactivator consumed will denote the absolute enzyme concentration. The idea of active site titration was first reported by Hartley and Kilby33, who noticed in a reaction of chymotrypsin with nitrophenylethyl carbonate that there existed a period of rapid 4nitrophenyl (PNP) release (a “burst”) followed by slow PNP release. Because nitrophenylethyl carbonate is a poor substrate, the reaction occurred in two steps: a rapid first step that released PNP quickly to form a semi-stable intermediate, followed by a slow second step corresponding to the hydrolysis of the intermediate. The total amount of accumulated intermediate, correcting for turnover, could be found by extrapolating the linear portion back to the y-axis. The magnitude of the y-intercept is known as the “burst”, and was dependent on enzyme concentration. Since then, a variety of enzymes have been quantified by this method including lipases34, aminoacyl-tRNA synthetases35 and retaining glycosidases36. 2Fglucosides such as DNP-2FGlc, PNP-2FGlc, and 2FGlcF have been shown to be competent active site titrating agents of β-glucosidases.36, 37 When a molecule of inactivator reacts with β-glucosidase, a leaving group molecule is released into solution which can be detected and quantified using UV/Vis or potentiometric methods. The active site is then physically blocked by the fluorine-containing glycone from participating in further catalytic cycles. Since each enzyme active site reacts rapidly with one equivalent of inactivator, the signal detected will correlate directly to the concentration of enzyme in the solution. 16  To measure the effectiveness of a potential active site titrating agent, studies are done to gauge how well a given compound inactivates a given enzyme. For rapidly acting inactivators with half-lives under 20 seconds, this has been done by directly monitoring the reaction of enzyme and inactivator using stopped flow techniques.38 For slower inactivations, reactions are followed directly by UV/Vis spectrophotometry, or indirectly by measuring time-dependent loss of enzyme activity.37  1.2.4 Enzyme Kinetics with a Mechanism-based Inactivator  Scheme 3 shows the behaviour of a retaining glycosidase in the presence of an excess of mechanism-based inactivator. An inactivator I-X, which consists of a reactive glycone (I) conjugated to a leaving group (X), reversibly binds to enzyme (E) with dissociation constant k-1/k1, also known as Ki. The chemical glycosylation step forms a covalent glycosyl-enzyme intermediate E-I, with associated rate constant k2. The intermediate can be subsequently hydrolysed to form E and product I-OH.  Scheme 3: Kinetic scheme of a retaining glycosidase reacting with a mechanism-based inactivator.  When the mechanism-based inactivator is a 2Fglycoside such as DNP-2FGlc, k3 is very small (approaching zero), very little hydrolysis of E-I occurs, and I-OH is virtually undetectable. Thus, the apparent initial rate of inactivation is related to the rate of covalent glycosyl-enzyme intermediate formation: 𝑑[𝐸 ] 𝑑𝑡 Equation 18 This time-dependent inactivation can be described by the Michaelis-Menten equation for substrate hydrolysis: 17  𝑘 [𝐸][ [  ] ] Equation 19  The initial rate of inactivation is therefore related to the rate constant of inactivation (ki = k2) as well as the inactivator dissociation constant (Ki = k-1/k1). This equation can be rewritten as: [𝐸]  𝑘  Equation 20 where 𝑘  [- ] [- ]  is the apparent rate constant of inactivation.  When the inactivation concentration is much less than Ki, [I-X] becomes negligible with respect to Ki so the rate of inactivation (kobs) becomes directly proportional to inactivator concentration: 𝑘  𝑘[  ] Equation 21  1.2.5 Small Molecule Latent Fluorophores Latent fluorophores, also called “pro-fluorophores”, are molecules that disguise their fluorescence until modified by some chemical reaction. Latent fluorophores are valuable for improving the signal-to-noise ratio of quantitative measurements, as the background fluorescence is zero. They have also been useful as tools for high-throughput screening, enzymelinked immunosorbent assays (ELISA), double-stranded DNA recognition and quantification, as well as fluorescence imaging.39 Recently, much effort has gone into developing fluorescent biological probes that absorb and emit in the near infrared region. Light of longer wavelengths is able to penetrate more deeply into tissues than light at wavelengths below 600 nm, and can circumvent the autofluorescence problem exhibited by biomolecules in the cell or body, such as oxy- and deoxyhaemaglobin, melanin, NADH and flavin.40 The following section contains examples of several fluorophores and pro-fluorophores that have been used as probes for enzyme activity. 18  1.2.5.1  7-Hydroxycoumarin  Figure 12: 7-Hydroxycoumarin.  7-Hydroxycoumarin (umbelliferone, 1.14) is a simple molecule exhibiting strong but short wavelength fluorescence properties (360 nm excitation, 465 nm emission). Modification of umbelliferones by extending π-π conjugation can allow creation of red-shifted fluorophores, such as 1.15, a 7-hydroxycoumarin-hemicyanine synthesized by Richard et al. (Figure 13).41 The emission wavelength of 1.15 of 636 nm has been red-shifted by almost 200 nm compared to that of 1.14, and it retains a large Stokes shift (63 nm). However, the quantum efficiency was found to be low at 0.7% to 4.9% in water at 25°C, compared to 90% for rhodamine 6G under identical conditions. The authors postulate that this may be due to the flexible structure of 1.15 resulting in twisting of the tetramethine moiety resulting in a loss of conjugation, or static quenching from aggregation of hydrophobic molecules.41  Figure 13: A water-soluble 7-hydroxycoumarin derivatised by hemicyanine.  A mechanism-based probe of a 7-hydroxycoumarin derivative has been achieved by linking it onto a latent glycosyl quinone methide, such as one developed by Nagano et al., shown in Figure 14.42 In addition to being a covalent probe, 1.16 also served as a highly sensitive imaging agent in mammalian cells to visualize enzymatic activity. This molecule features several groups adjacent to and interacting with each other. The authors incorporated a latent galactosyl quinone methide into the design of the molecule to target and label glycosidases. In addition, a fluorescein moiety in close proximity to a 7-hydroxycoumarin quenches the inherent 19  fluorescence of the coumarin through FRET. In mammalian cells, several things occur as recombinant β-galactosidase cleaves the glycosidic bond: the fluorescein is released and diffuses away enabling a spectral shift, simultaneously generating a reactive quinone methide centre which can be attacked by nearby nucleophilic functional groups, forming a covalent bond. However, the caveat of this particular molecule is that it had to be physically injected into cells due to its large size.  Figure 14: A coumarin-based quinone methide precursor as an imaging agent of β-galactosidase activity.  Kwan et al. simplified this concept by incorporating the quinone methide precursor directly into the ring structure of 7-hydroxycoumarin and synthesising a form that only becomes reactive upon cleavage (1.17, Figure 15).43 Furthermore, the small structure of 1.17 allowed direct diffusion into the cell. E. coli cells expressing recombinant Agrobacterium sp. βglucosidase were able to hydrolyse 1.17 releasing fluorophore, which when attacked by a nearby nucleophile, trapped the fluorophore within the cell (Figure 15). Fluorescence-activated cell sorting (FACS) was then used to successfully sort and enrich in cells expressing the target enzyme.  20  Figure 15: A coumarin-derived fluorogenic latent quinone methide as a probe of β-glucosidase activity.  1.2.5.2  Resorufin  Figure 16: Resorufin (1.18), resorufin α-D-mannoside (1.19), resorufin β-D-cellobioside (1.20).  A far-red shifted fluorescent molecule, resorufin has a pKa value of 5.8 and a large extinction coefficient of ε = 56000 M-1 cm-1 at 571 nm, and is characterized by excitation/emission wavelengths maxima at 571 nm and 585 nm.44,45 This conferred several advantages over 7-hydroxycoumarin as a reporter molecule: its far-red-shifted fluorescence avoided the issue of autofluorescence in cells, especially in plant cells, and a low pKa allowed continuous fluorometric determination of enzyme activities in lower pH conditions. However, as the planar structure of resorufin encouraged hydrophobic interaction and crystallization, the low aqueous solubility of resorufin glycosides (250-500 μM) limited its utility in aqueous and biological systems.46 Nevertheless, several resorufin glycosides, including 1.19 and 1.20 (Figure 16), have been documented to be fluorogenic substrates of glycosidases, and have been used in 21  the detection and imaging of a variety of glycosidase activities, such as those of fungal cellulases47, α-mannosidases48, and plant xyloglucanases49.  1.2.5.3  7-Hydroxy-9H-(9,9-dimethylacridin-2-one) (DAO)  Figure 17: DAO and DAO-derivatives as probes of enzyme activity.  First synthesized in 1928 by Goldstein and Kopp, and later by Hill et al. in 197050, Corey et al. improved the synthesis of DAO (1.21) in 1991 and patented many applications, such as probes and substrates of glycosidases and proteases.46 The properties of DAO are advantageous in comparison to resorufins, making it a very attractive fluorophore. First, DAO fluoresces at a more red-shifted wavelength (634 nm). In addition, without sacrificing the high extinction coefficients characteristic of very planar structures such as resorufin (ε634 = 54000 M-1 cm-1), the water solubility of DAO-derivatives is greatly enhanced. For example, the maximum solubility of DAO β-D-galactopyranoside (1.22) in water (20-25 mM) is reported to be two orders of magnitude greater than resorufin β-D-galactopyranoside (0.25-0.5 mM).46 This solubility enhancement is likely due to the substitution of oxygen by a dimethyl group at position 9; this extra bulk limits packing and discourages crystallization. Despite these promising properties, recent literature has not seen DAO-derivatives used frequently as fluorophores. A report by Richard et al. describes an attempt to measure peptidase activity by attaching various fluorophores, including DAO, onto a peptide recognition sequence through a self-cleaving linker (1.23).41 However, the authors report decreased fluorescence of free DAO in the presence of thiols such as DTT, making DAO unsuitable for accurate quantification in reducing environments such as those present in the cytoplasm of cells.41  22  1.2.5.4  7-Hydroxy-9H-(1,3-dichloro-9,9-dimethylacridin-2-one) (DDAO) An intermediate on the synthetic route to DAO performed by Corey in 199146 was the  dichlorinated version 7-hydroxy-9H-(1,3-dichloro-9,9-dimethylacridin-2-one) (DDAO, 1.23). Despite its simple structure and straightforward synthesis from commercial starting materials, DDAO was not widely present in the literature until DDAO derivatives became commercially available as fluorescent substrates for reporter systems.  Figure 18: DDAO (1.23) and fluorogenic substrates DDAO phosphate (1.24) and DDAO β-Dgalactopyranoside (1.25).  The fluorescent DDAO molecule was chosen as the aglycone for active site titrations in this project due to its red-shifted emission wavelength and acceptable fluorescence brightness. This molecule is also able to tolerate a large range of excitation wavelengths, so the excitation and emission wavelengths can be chosen to reduce interference by Rayleigh scattering. In addition, the pKa value is reported as 5 in the literature51, indicating DDAO is approximately a thousand times more acidic than methylumbelliferone. As the structure of DDAO is more similar to DAO than to resorufin, DDAO might also be expected to be more water-soluble than resorufin. The fluorescence signal of DDAO has been shown to be measureable as well as reproducible both in vitro and in vivo. In a plate-based inhibition assay, DDAO phosphate (1.24) was shown to be a good fluorogenic substrate for protein phosphatase 2A (PP-2A), even at low (10 μM) concentrations of 1.24.52 The presence of okadiac acid, a shellfish-toxin and inhibitor of PP-2A, decreased DDAO-induced fluorescence. Although the assay was performed under a 23  constant, but large, concentration of β-mercaptoethanol (6 mM), no loss of DDAO fluorescence was reported. In vivo, DDAO β-D-galactopyranoside (1.25) has been successfully used as a substrate for a β-galactosidase used as a transgene-generated reporter to image tumours in live mice.53  1.2.6 Small Molecule Latent Fluorophores as Active Site Titrants  Fluorogenic molecules have been used in reporter systems, in vitro and in vivo, as sensitive probes to locate and quantify expression. In most cases, these molecules act as substrates; others act as covalent probes. Recently, much effort in the Withers group has been made to design plate-based enzyme assays for use in high-throughput screening for applications such as directed evolution. When screening directed evolution libraries by assay of expressed enzyme activity, the actual value measured reflects both the inherent activity of the mutant and the quantity of mutant expressed in that well. A simple and convenient method to measure expression levels is necessary and one approach is to use an active site titrating agent. The use of a fluorogenic, mechanism-based, active site titrating agent can not only sensitively quantify the absolute concentration of a given enzyme based on the fluorescence change, but can also directly assess the efficiency of a particular enzyme by continuously monitoring the reaction and determining ki/Ki. To be an effective active site titrating agent, several requirements must be met. For accurate titration, the hydrolysis of the covalently linked enzyme species should be minimal. In addition, the properties of the fluorophore should address the following issues: 1. Sensitivity a. Quantum yield – a molecule with a larger quantum yield would release more fluorescence per excitation photon absorbed b. pKa – many enzymatic reactions take place in acidic buffer; using a molecule with a low pKa allows direct monitoring without the need for a base “stop”  24  2. Lability – labile enough to vault the higher transition state barrier effected by the presence of an electronegative fluorine group in the reagent, yet stable enough to avoid spontaneous hydrolysis and therefore increased background 3. Water solubility – many far red emitting fluorophores, due to their extended hydrophobic π-π conjugated systems, are poorly soluble in aqueous media. To be an effective component of inactivator assays, the aglycone must be soluble enough in its latent form to meet concentration requirements of the enzyme, as well as in the fluorescent form to impart enough fluorescence  It was expected that DDAO may fulfil these requirements.  1.3  Aims of Thesis  The aim of this thesis is to explore DDAO as a potential leaving group for mechanismbased inhibitors and its ability to act as a reporter molecule for active site titration. This will be achieved by:  1. The synthesis of novel 2Fglycosides containing DDAO as an aglycone; 2. The enzymatic evaluation of the above as mechanism-based inactivators for retaining glycosidases; 3. The development of a method for active site titration of glycosidases by fluorescence.  25  2  Results and Discussion  2.1  Synthesis and Characterization  2.1.1 Synthesis of DDAO A facile preparation of DDAO was performed as reported by Corey et al. (Scheme 4).46 The synthesis started with 3-(2-hydroxypropan-2-yl) phenol (2.2), which is commercially scarce but could be conveniently synthesized from the readily available, inexpensive 3’hydroxyacetophenone (2.1) by the addition of a methylmagnesium halide. This phenol was then reacted with the chloroquinone-imide (2.3) in the presence of base to provide the coupled product 2.4. Reaction times exceeding two hours resulted in increased side products, perhaps resulting from polymerization or oxidation of starting material or product.  Scheme 4: Synthesis of DDAO (1.23) from commercially available starting materials. i) Mg0, CH3I, I2, Et2O; ii) NaOH, H2O, THF, 0°C; iii) Na2S2O4 (aq); iv) 2 M HCl, reflux; v) NaIO4 (aq).  26  To create the tricyclic ring structure, the molecule was first reduced to enhance the nucleophilicity of the chlorinated aromatic ring. This reduction occurred rapidly during an aqueous sodium dithionite “wash” step when freshly prepared sodium dithionite solution was shaken in the separatory funnel with the organic layer containing the compound. The reaction was deemed complete by TLC as well as by a change in solution colour from deep to light amber caused by loss of conjugation between the rings. Afterward, strongly acidic conditions and heat were used to promote electrophilic aromatic substitution and ring closure. The last step involved reoxidation to complete the conjugation between aromatic rings through the imine bridge. This was again done using a biphasic reaction system consisting of the DDAO precursor dissolved in ethyl acetate, and sodium periodate in distilled water. Rapid stirring resulted in a change in colour from light amber to dark red, indicative of increased conjugation between the two rings as a result of oxidation to the iminoquinone. Dark violet crystals of DDAO were obtained by crystallisation of the crude product from hot ethanol. The combined yield attained after two crops of crystallization was 69% (1.6 g), which was less than that reported by Corey et al. (82%) but still more than enough to use as starting material.  2.1.2 Characterization of DDAO A 1H NMR spectrum run in deuterated methanol at 300 MHz revealed proton peaks from 6.8 to 7.7 ppm correlating to the four aromatic protons of DDAO, as well as a singlet at 1.88 ppm with an integrated area of 6. Aromatic ortho (8.7 Hz) and meta couplings (2.4 Hz) were detected. These results suggest either that the two tautomeric forms of DDAO (Figure 19) are equilibrating too fast to be differentiated on the NMR timescale, or that equilibrium is slow and one tautomer is present in a vast excess of the other. Performing a variable temperature NMR experiment may distinguish the two: if an equilibrium exists, at lowered temperatures this may be slowed down so peak separation may be observed.  27  Figure 19: The two tautomeric forms of DDAO.  To determine if DDAO has the properties to be a component of a good active site titrant, the absorbance spectrum, extinction coefficient, pKa and the fluorescence properties were explored. The absorbance spectra of DDAO (Figure 20) matched those previously reported53, with maximum absorbance in the visible wavelength range at 414 nm for the protonated species at pH 2.0, and 650 nm for deprotonated DDAO, pH 12.0. DDAO is a heavily conjugated system, hence the energy gap between the HOMO and LUMO is small, causing the π - π* transition for the protonated species to be in the visible wavelength range. When DDAO is deprotonated, an extra electron pair is liberated that fills the non-bonding orbital, decreasing the energy required for transition between the HOMO and LUMO states, thus causing a bathochromic shift. The absorbance wavelength range for the anion is large and almost constant from 600 nm to 650 nm with an extinction coefficient value (ε) of 32100 ± 900 M-1 cm-1 at 600 nm.  Figure 20: Absorbance spectra (solid lines) of protonated (pH 2) and deprotonated (pH 12) DDAO; fluorescence emission spectrum at 600 nm excitation wavelength.  28  The pKa was determined by measuring fluorescence at excitation and emission wavelengths of 600 nm and 655 nm of 0.5 μM of DDAO in buffers of various pH values, respectively. The fluorescence intensity was plotted and fitted to a pH titration curve as seen in Figure 21. The pKa was found to be 5.34 ± 0.02, in fair agreement with the previously reported value of 5.0.51  Figure 21: Determination of pKa of DDAO by fluorescence using excitation/emission wavelengths of 600 nm/655 nm. The data was fitted onto a titration curve by GraFit 5.  The dependence of fluorescence intensity upon the concentration of fluorophore was explored by measuring spectra at a series of concentrations. Below a concentration of 5 μM, a linear fluorescence response for DDAO at 600 nm excitation/ 655 nm emission was seen (Figure 22). Above this concentration, the relationship is non-linear and even produces a decrease in fluorescence at concentrations above 40 μM (Figure 23). It is hypothesized that a self-quenching phenomenon occurs because the emission wavelength is close to the absorbance wavelength range. Thus the photons emitted from one molecule of DDAO may be absorbed by an adjacent molecule of DDAO, resulting in a decrease of the number of photons that reach the photomultiplier tube. However, the range of linearity of up to 5 μM observed (Figure 22) is suitable for the determination of concentrations of most enzymes, and absorbance in the visible region could be used above this range.  29  Figure 22: The linear range of fluorescence response for DDAO at a concentration range of up to 5 μM. Conditions: 10 nm/10 nm excitation and emission slit width, 600V (left); 5 nm/5 nm excitation and emission slit width, 640 V (right).  Figure 23: Non-linearity of fluorescence intensity as DDAO concentration increases. A selfquenching phenomenon is proposed.  In summary, the absorbance and fluorescence properties of DDAO make it a suitable reporter for titration of enzyme active sites. In addition, its relatively low pKa (5.34) suggests that it will be a sufficiently good nucleofuge for this purpose.  30  2.1.3 Synthesis of 2-Deoxy-2-fluoro Sugars  Scheme 5 illustrates the general synthetic method used to make the 2-deoxy-2-fluoro-Dglucosyl, D-xylosyl, D-galactosyl and D-cellobiosyl bromides, the precursor for the KoenigsKnorr glycosylation (Section 2.1.4 below). The free sugar was protected with acetyl groups, brominated, and reduced with solid zinc to form the glycal, except for the gluco compound where 3,4,6-tri-O-acetyl-D-glucal was purchased. Fluorine was electrophilically inserted at the C2 position by the Selectfluor® reagent (1chloromethyl-4-fluoro-1,4-diazonia-bicyclo[2,2,2]octane bis(tetrafluoroborate)), a relatively nonhazardous, non-toxic electrophilic fluorine donor reagent.54 Reaction of Selectfluor® and tri-Oacetyl-D-galactal (4.4) led almost exclusively to the 2-deoxy-2-fluoro-D-galactopyranose (4.5) regioisomer due to steric hindrance of top face attack previously reported.55 Conversely, the reaction of Selectfluor® with tri-O-acetyl-D-glucal (3.4) preferentially produced the 2-deoxy-2fluoro-D-manno-hemiacetal (3.6). The regioselectivity towards the D-gluco-epimer (3.5) was reported to be improved by reaction in a 5:1 ratio of a polar aprotic solvent (acetonitrile or nitromethane) and water to produce up to 45% of the desired hemiacetal, but in this work only 20% of the desired gluco-epimer (3.7) was isolated.54,56 Similar reaction conditions were used for the other glycals in the series, and the other Selectfluor® reactions produced similar or better yields. After completion of the Selectfluor® reaction, the product mixtures were per-O-acetylated and purified by silica gel column chromatography to afford anomeric α/β mixtures of the desired monosaccharide products (3.7, 4.7, 5.7). The C2-epimers 6.7 and 6.8 could not be separated by column chromatography, so an attempt was made to crystallize 6.7 out from diethyl ether and chloroform, as reported by McCarter et al.57 This resulted in enrichment of the desired C2 epimer in a yield of 12% (250 mg), sacrificing yield for an increase in purity (95%). The following isolated yields were obtained: 20% for 3.7, 68% for the 4.7, 72% for 5.7, and a 12% recrystallized yield for 6.7.  31  Scheme 5: General scheme for the synthesis of 2Fglycosides. i) Ac2O, 0.1 eq NaOAc; ii) HBr, AcOH; iii) Zn, AcOH, H2O; iv) Selectfluor®, MeCN/H2O or MeNO2/H2O; v) Ac2O, pyridine, DCM; vi) HBr, AcOH.  The correct stereochemistry was verified by inspecting the 19F NMR spectra of 3.7, 4.7, 5.7, and 6.7, which were consistent with previously reported data.58,59 In addition, identities could be directly deduced from the coupling constants between fluorine and vicinal hydrogens in all cases (Figure 24). The F2-H2 geminal coupling constant is large (50 Hz); vicinal couplings between F2-H1 and F2-H3 are considerably smaller. The relative magnitudes of the vicinal 32  coupling between F2 and H3 can be compared. According to the Karplus equation, a 180° orientation (axial-axial) would give a larger coupling constant than an equatorial-axial coupling constant.60 In all cases, the C2 epimer containing the equatorial fluorine was the desired compound. Therefore, a large coupling constant between F2 and H3 (> 20 Hz) would indicate the fluorine is in the axial position and thus the undesired compound. In these cases, 19F NMR is a highly useful tool for easy analysis of a mixture of compounds.61  Figure 24: The magnitudes of 3JF2-H3 can be used to deduce the presence of axial or equatorial fluorines. The sizes of the blue arrows correspond to the relative magnitudes of their vicinal coupling constants.  Bromination of the anomeric centre generally proceeded smoothly, in quantitative yields. For 3.7, 4.7 and 5.7, the protected monosaccharide was dissolved in cold dichloromethane, and anywhere from 2-20 equivalents of hydrogen bromide in acetic acid was added. Owing to the anomeric effect, the thermodynamic product is the α-glycosyl bromide, which can then be subjected to displacement by DDAO to form the β-glycoside in a Koenigs-Knorr reaction. The bromination reactions were generally complete within one hour. Conversely, bromination of per-O-acetylated 2Fcellobiose was more difficult. Reaction of a mixture of 6.7 and 6.8 resulted in two products by TLC after 3 hours, a lower major product spot and a minor spot 0.2 RF units higher. At first, these were thought to be the α-2Fglycosyl bromides 6.9 and 6.10 (Figure 25), and that separation at this step could simplify the purification as well as minimize losses due to crystallization. However, after isolation, a slow conversion of the top spot to the bottom spot was observed. The top spot was concluded to be the kinetic product – a mixture of the β-2Fglycosyl bromides, which anomerised to the lower running, thermodynamically stable α-2Fglycosyl bromides over time. This was supported by 19F NMR data of the glycosyl bromide mixture, which revealed four sets of peaks: two sets of dd corresponding to the α-bromides 6.9 and 6.10, and another two sets of ddd indicative of the β-  33  bromides of the C2 epimers. Finally, bromination of the pure, crystallized compound 6.7 for 3 hours yielded an α/β-mixture of per-O-acetylated 2Fcellobiosyl bromide, 6.9.  Figure 25: The thermodynamic products of bromination of per-O-acetylated 2Fcellobiose.  2.1.4 Koenigs-Knorr Glycosylation of DDAO and Per-O-acetylated α-Glucosyl Bromide The formation of a glycosidic linkage between α-D-glycosyl bromides and DDAO was explored using 3.2 and DDAO. A first attempt by phase transfer using TBABr as a catalyst resulted in almost no product formation, so another glycosylation method was investigated.  Scheme 6: Koenigs-Knorr coupling of per-O-acetylated α-glucosyl bromide to DDAO to generate two glycoside products.  34  In the original synthetic paper, Corey et al. used the Koenigs-Knorr glycosylation method to make 9H-(9,9-dimethylacridin2-one7-yl) β-D-glycosides (DAO glycosides), since anchimeric assistance by the C2-acetyl would ensure selectivity for the β-glycoside.46 This method was next applied to couple 3.2 and DDAO together, with slight modifications in solvents and base as described below and shown in Scheme 6. Compound 3.2 and DDAO were dissolved in dry acetonitrile (Corey et al. used ethyl acetate). Instead of quinoline, another weak hindered base 2,6-lutidine was used (pKa 6.77 in nitromethane).62 2,6-Lutidine is strong enough to deprotonate DDAO while causing only limited elimination of the α-glycosyl bromide. Solid calcium sulphate was added to absorb fortuitous water. Finally, after the addition of a heavy-metal promoter, silver(I) oxide, the reaction mixture was left to stir overnight at room temperature and was protected from light. TLC analysis of the reaction revealed two new, coloured spots of unequal proportions. The less polar, minor product was bright red. Approximately 0.06 RF units lower was the major product, a brilliant yellow. These two spots were surmised to represent sugars coupled to DDAO, as the compounds had a mass commensurate with them being DDAO glycosides, in an approximate ratio of 85:15 (by NMR analysis of the crude mixture). In addition, after being dipped in 10% sulphuric acid in ethanol and heated, both spots turned violet, suggesting that the acid-labile glycosidic bond hydrolysed, releasing DDAO – also violet under identical staining conditions. Thus, the minor product was named the “Red” isomer (DDAOR), and the major product was called the “Yellow” isomer (DDAOY). Figure 26 shows the two major resonance forms of ionized DDAO in solution. Thus, it could be reasonable to assume that the two glycosylation products obtained (3.11, 3.12) were the result of attack on the α-glucosyl bromide from different nucleophilic oxygens of DDAO in the reaction conditions employed.  Figure 26: The two major resonance contributors of deprotonated DDAO.  35  0.87  0.97 7.5  1.94 7.5  1.96 7.0  4.09 6.5  6.0  5.5  1.00 1.00 7.0  2.07 5.0 4.5 Chemical Shift (ppm)  4.04 6.5  6.0  5.5  2.02  5.0 4.5 Chemical Shift (ppm)  1.00  4.0  12.35 6.00 3.5  3.0  2.5  1.00 4.0  2.0  11.99 6.00 3.5  3.0  2.5  2.0  Figure 27: 1H-NMR spectra of per-O-acetylated DDAOY-Glc (above) and per-O-acetylated DDAOR-Glc (below) in CDCl3. The connectivity of DDAO was later determined from 2D HMBC NMR data (discussed in Section 2.1.6).  36  1.5  1.5  The 1H-NMR spectra of the two compounds were very similar (Figure 27), with the same number of proton peaks split up into three groups: 1. An upfield region from 1.8 ppm to 2.1 ppm containing four groups of acetyl hydrogens plus two overlapping singlets with a combined area of 6, from the 9,9-dimethyl groups of DDAO; 2. A midfield region spanning 3.9 ppm to 5.4 ppm containing 7 glucose ring-protons; 3. A downfield region from 6.9 ppm to 7.7 ppm containing the 4 aromatic protons of DDAO.  From the 1D proton NMR spectrum, it was not possible to establish the regiochemistry of each structure; these were later determined from the 2D HMBC NMR spectra of each compound (not shown). A detailed discussion on the structure elucidation of two similar products, the DDAO 2-deoxy-2-fluoro-β-D-glucopyranosides 3.15 and 3.16, is presented in Section 2.1.7.  2.1.5 Koenigs-Knorr Glycosylation of DDAO and 2Fsugars  Scheme 7: General scheme of the Koenigs-Knorr coupling of 2Fsugars to DDAO to generate DDAO 2-deoxy-2-fluoro-β-D-glycosides.  The coupling of the protected α-2Fglucosyl bromide (3.9) and DDAO was achieved by using the same glycosylation conditions as used for the parent sugar – dry acetonitrile and 37  silver(I) oxide (Section 2.1.4). Since anchimeric assistance was not possible from the 2-position, it was not clear whether the β-glycoside would preferentially form. A TLC analysis of the reaction mixture revealed only two coloured products, 3.13 and 3.14, formed in ratios similar to that of the parent compound. 1H-NMR analysis confirmed that the products had the βconfiguration, apparent from the large coupling constants between H1 and H2 (3JH1-H2 = 7 - 8 Hz). The Koenigs-Knorr glycosylations for the remaining monosaccharides (4.9, 5.9) were executed under the same reaction conditions; the glycosylation of 6.9 was done with an equilibrium mixture of the α- and β-2Fcellobiosyl bromides in dry DCM, with no change in yield (35 - 40% isolated yields). Future work includes reaction optimization, possibly by using an excess of, instead of an equimolar dose of DDAO to the glycosyl bromide.  Scheme 8: Deacetylation of per-O-acetylated DDAO 2Fglycosides.  Initial attempts at deacetylation of per-O-acetylated DDAOY-2FGlc (3.13) used a small speck of solid sodium in dry methanol, but unfortunately yielded free DDAO and methyl 2Fglucoside. Mild acidic deprotection conditions (1% hydrogen chloride in dry methanol) were then tried, but TLC analysis of the reaction mixture showed a ladder of multi-coloured 38  degradation products both more polar and more non-polar than the starting material. Finally, conditions for base deacetylation were optimized. A low concentration of sodium methoxide (20 mM) in dry methanol yielded 3.15 with minimal solvolysis and this condition was used for deacetylation of the other per-O-acetylated monosaccharides. The per-O-acetylated DDAO 2Fcellobioside was found to be more stable so it was deprotected using 50 mM sodium methoxide in dry methanol. Detailed inspection of the 1H and HMBC NMR spectra for each synthesized compound verified the structures proposed.  2.1.6 Elucidation of the Structures of DDAOY-2FGlc and DDAOR-2FGlc  An HMBC (Heteronuclear Multiple-Bond Correlation) NMR experiment records a twodimensional spectrum that is optimized to detect 2- to 4-bond couplings from heteronuclei (such as carbon) to protons, and can be used to determine the connectivity of DDAO to the sugar in DDAOY-2FGlc and DDAOR-2FGlc. Carbonyl carbons are the most deshielded, hence the furthest downfield signals in a carbon spectrum. Since there is only one carbonyl in either isomer of DDAO 2Fglucopyranoside, determining the number and identities of the aromatic protons that couple to that carbon should reveal whether that compound has chlorines directly adjacent to the carbonyl as in 3.16, or if the chlorines are situated closer to the glycone, as in 3.15. The HMBC spectrum for DDAOR-2FGlc is seen in Figure 28. From this spectrum, we can see that the minor product DDAOR-2FGlc is compound 3.16, inferred from two pieces of evidence: 1. The most downfield carbon resonance (174 ppm) is the carbonyl of DDAO (C2(Ar)), which couples to the singlet proton at 7.62 ppm. This suggests that the singlet proton and C2(Ar) are 2 to 4 bonds apart, only possible if these two atoms are on the same ring. 2. The anomeric proton (H1(Glc) at 5.23 ppm) couples to one aromatic carbon resonance; this corresponds to C7 of DDAO, the carbon attached to the glycone through the glycosidic (acetal) bond. This carbon also couples to 3 other proton resonances: a doublet of doublets at 6.97 ppm, a doublet at 7.10 ppm and a doublet at 7.61 ppm.  39  These protons correspond to H6(Ar), H8(Ar) and H5(Ar) of the DDAO molecule, which were previously identified by their characteristic splitting patterns.  C7(Ar)  C2(Ar)  2  3  4  5  1  Figure 28: A portion of the HMBC spectrum of DDAOR-2FGlc (3.16) showing the sp2hybridized carbon range from 130 ppm to 180 ppm. The structure and numbering scheme of DDAOR-2FGlc are shown. Two carbon resonances, carbonyl C2(Ar) and C7(Ar), are labelled. The relevant cross-peaks are labelled: 1) C2(Ar)-H4(Ar); 2) C7(Ar)-H5(Ar); 3) C7(Ar)-H8(Ar); 4) C7(Ar)-H6(Ar); 5) C7(Ar)-H1(Glc).  Inspecting the HMBC spectrum of DDAOY-2FGlc (Figure 29) and using similar lines of reasoning, we can see that this product is compound 3.15: 1. The most downfield carbon resonance (189 ppm) is again the carbonyl carbon of DDAO (C7(Ar)), and is seen coupling strongly to the doublet corresponding to H5(Ar) of DDAO, a 3-bond coupling. Due to the HMBC optimization of 3- and 4-bond 40  couplings, the 2-bond couplings can be very weak, so those between C7(Ar) and H8(Ar) and between C7(Ar) and H6(Ar) are unobservable above background levels. 2. The anomeric proton (H1(Glc) at 5.30 ppm) couples to one aromatic carbon resonance, which corresponds to C2(Ar) of DDAO, the carbon attached to the glycone through the glycosidic (acetal) bond. This carbon also couples to the singlet proton at 7.49 ppm corresponding to H5(Ar) of the DDAO molecule.  125 130 135 140  150  C2(Ar)  3  2  155 160 165 170  F1 Chemical Shift (ppm)  145  175 180  C7(Ar)  185  1  190  8.0  7.5  7.0  6.5  6.0  5.5 5.0 4.5 F2 Chemical Shift (ppm)  4.0  3.5  3.0  2.5  2.0  1.5  Figure 29: A portion of the HMBC spectrum of DDAOY-2FGlc (3.15), showing the aromatic carbon range from 120 ppm to 195 ppm. The structure and numbering scheme of DDAOY-2FGlc (3.15) are shown. The two relevant carbon resonances, carbonyl C7(Ar) and C2(Ar), are labelled. The relevant cross-peaks are labelled: 1) C7(Ar)-H5(Ar); 2) C2(Ar)-H4(Ar); 3) C2(Ar)-H1(Glc).  An examination of the HMBC spectrum of the other DDAOY-conjugated glycosides revealed the same coupling patterns as those of DDAOY-2FGlc.  41  Finally, the bonding of DDAO in DDAOY- and DDAOR-glycosides was unequivocally demonstrated in the crystal structures of per-O-acetylated DDAOY 2Fgalactoside (per-Oacetylated DDAOY-2FGal, 4.13, Figure 30) and per-O-acetylated DDAOR 2Fgalactoside (perO-acetylated DDAOR-2FGal, Figure 31) obtained from Dr. Hong-Ming Chen. In Figure 30, the chlorines (green) of 4.13 are clearly situated adjacent to the oxygen forming the glycosidic bond; in Figure 31 the chlorines are located away from the glycone, indicating that the bonding of DDAO in per-O-acetylated DDAOR-2FGal is opposite to that of 4.13. Also of interest are the π-stacking interactions seen when two planar DDAO moieties crystallize on top of each other. Their 9,9-dimethyl groups face in opposite directions, conceivably minimizing steric interactions. One major difference between the crystal structures of per-O-acetylated DDAOR-2FGal and compound 4.13 is in the orientation of the DDAO plane relative to the plane of the monosaccharide ring. For per-O-acetylated DDAOR-2FGal, an angle of 50.1° was measured; for compound 4.13, this angle was found to be 75.1°, an angle much closer to being perpendicular. This may indicate a twisting of the glycosidic bond in order to accommodate the large chlorine groups. There was no significant difference in glycosidic bond length between the two molecules. The protecting groups in per-O-acetylated glycosides are electron-withdrawing and diminish the extent of oxocarbenium ion formation, so bond length differences may be too small to distinguish.63,64  42  Figure 30: Crystal structure of per-O-acetylated 9H-(1,3-dichloro-9,9-dimethylacridin-7-one-2-yl) 3,4,6-tri-O-acetyl-2-deoxy-2-fluoro-β-D-galactopyranoside (per-O-acetylated DDAOY-2FGal). Colour code: gray, carbon; red, oxygen; blue, nitrogen; yellow, fluorine; green, chlorine. Hydrogens have been omitted for clarity.  Figure 31: Crystal structure of per-O-acetylated 9H-(1,3-dichloro-9,9-dimethylacridin-2-one-7-yl) 3,4,6-tri-O-acetyl-2-deoxy-2-fluoro-β-D-galactopyranoside (per-O-acetylated DDAOR-2FGal). Colour code same as that of Figure 30.  43  2.1.7 UV/Vis Characterization of DDAOY and DDAOR 2FGlycosides  As mentioned before, DDAOY-Glc and DDAOR-Glc interact with light in different ways. Likewise, DDAOY-2FGlc (3.15) and DDAOR-2FGlc (3.16), as well as the rest of the DDAOYand DDAOR-linked glycosides, exhibited similar behaviour. The UV/Vis spectra of the representative compounds 3.15 and 3.16 are shown in Figure 32. For 3.15 dissolved in 50 mM sodium phosphate buffer at pH 6.8, the wavelength of maximum absorption is at 400 nm with a molar extinction coefficient of 11300 M-1cm-1; the observer perceives a yellow colour. This value was used to determine concentrations for the rest of the DDAOY-linked 2Fglycosides because the structures and electronics of each molecule are very similar. Under identical conditions, the maximum absorption of the orange-coloured DDAOR-2FGlc occurred at the longer wavelength of 458 nm with a molar absorptivity of 12500 M-1cm-1.  Figure 32: Absorbance spectra of DDAOY-2FGlc (3.15) and DDAOR-2FGlc (3.16). The λmax of DDAOY-2FGlc is 400 nm; λmax of DDAOR-2FGlc is 458 nm. Spectra were obtained in 50 mM sodium phosphate buffer, pH 6.8, 37°C. The concentrations used were not identical.  44  2.2  Enzymatic Evaluation  2.2.1 Testing DDAOY-2FGlc as an Inactivator of Abg Agrobacterium sp. β-glucosidase (Abg, EC 3.2.1.21) is a GH family 1 retaining exoglucosidase. Abg has been highly useful as a model enzyme in the Withers group due to its relatively small monomer molecular weight of 52 kDa, its straightforward expression and purification, its stability under diverse conditions, and its ability to accept many synthetic substrates.65 The catalytic mechanism of Abg has been meticulously analysed65, and studies have shown that Abg is efficiently inactivated by a variety of aryl 2Fglucopyranosides.36 Abg has therefore been used as a model enzyme in the evaluation of glycosidase inhibitors and inactivators, and again is used in this thesis as a model enzyme to evaluate DDAO 2Fglucosides as active site titrating agents.37,5,66,67 Initial analysis of the reaction of DDAOY-2FGlc (3.15) with Abg by standard inactivation methods involved incubation of Abg with 3.15 and monitoring time-dependent loss of activity by removal of aliquots at time intervals and measuring this activity by assay with substrate, 4-nitrophenyl β-D-glucopyranoside (PNP-Glc). Inactivation proved to be very rapid, even at very low concentrations of 3.15, as seen in Figure 34. Time-dependent inactivation was observed, but the highest concentration that could be studied was 39 nM since at higher concentrations the enzyme was almost completely inactivated at the shortest assay time possible. As a consequence, inactivation rates at concentrations of inactivator approaching saturation could not be monitored. A rough measure of the second order rate constant (ki/Ki) of 3.4 × 105 mM-1 min-1 was obtained from the slope of a plot of rate constant observed against inactivator concentration (Figure 34 b). As a comparison, the ki/Ki for DNP-2FGlc (3.17, Figure 33), the best previous inactivator of Abg, was reported to be 1160 mM-1 min-1.38 The difference in inactivation efficiency of 2 orders of magnitude between these two molecules was extremely encouraging.  45  Figure 33: The structure of DNP-2FGlc, an efficient inactivator of Abg.  In carrying out such inactivation studies, it is preferable to keep the inactivator concentration at least five times that of the enzyme to maintain a pseudo-first order kinetic relationship. However, the concentration of enzyme must be high enough so that hydrolysis activities can still be detected after dilution into the substrate. The lowest concentration of Abg employed in the inactivation reaction was 4 nM, which gave a minimum initial activity of 0.03 A400/min in the activity assay.  a)  b)  Figure 34: Inactivation of Abg with 3.15. a) Plot of normalized residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation. b) Plot of observed inactivation rate constants vs. inactivator concentration.  A possible solution to this problem was to directly measure inactivation by incubating the enzyme with inactivator and substrate and directly monitoring the time-dependent decrease in enzyme activity by fitting the derivative of the resulting pseudo-first order curve to an 46  exponential decay equation. An additional advantage in this case is that the substrate acts as a competitive inhibitor against inactivation, “protecting” the active site from inactivator and thereby slowing the inactivation reaction down. Saturating amounts of substrate (at least 7 times the Km of PNP-Glc) were used so that the substrate concentration did not change significantly as the reaction progressed, with a maximum of 8% total PNP-Glc hydrolysed. The observed ki/Ki could then be corrected for substrate inhibition using the equation for competitive inhibition (Equation 22) where Ki is the true inactivator dissociation constant, Kiapp is the apparent inactivator dissociation constant, [I] is the concentration of inhibitor (PNP-Glc), and Kic is the dissociation constant of the competitive inhibitor (Km of PNP-Glc). [ ] Equation 22 The inactivation rate constant ki is unaffected by the presence of a competitive inhibitor, so the corrected ki/Ki can be obtained by multiplying the reciprocal of Equation 22 by ki on both sides, then solving for ki/Ki to obtain Equation 23. 𝑘  𝑘  []  (  ) Equation 23  The substrate protection assay for the inactivation of Abg with 3.15 shows that inactivation is occurring in a pseudo-first order fashion (Figure 35a-c). Inactivation is indeed mechanism-based, for an increase in substrate concentration caused a definite decrease in the observed second order rate constant.  47  a)  b)  c)  d)  Figure 35: Inactivation of Abg by 3.15 monitored in the presence of saturating amounts of PNPGlc. Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation, at concentrations of a) 0.556 mM, b) 1.35 mM, and c) 3.70 mM PNP-Glc. d) Plot of observed inactivation rate constants vs. inactivator concentration at the indicated substrate concentrations.  Inactivation was successfully slowed so that higher inactivator concentrations could be tested; however, concentrations above 230 nM resulted in inactivation reactions that were again too fast to follow, even when using a continuous assay such as this. Therefore, only second order rate constants could be determined, reported in Table 1 below. The results show that the  48  corrected ki/Ki values fall in approximately the same range, and exceeded that of DNP-2FGlc (3.17, 1160 mM-1min-1) by at least 400 times.38  Table 1: Kinetic parameters for inactivation of Abg by 3.15, in the presence and absence of saturating amounts of substrate.  Inactivator  app  corrected  ki/Ki  ki/Ki  [PNP-Glc] (mM)  ki (min )  Ki (mM)  0.556  -  -  54000  4.4 x 105  1.35  -  -  18500  3.4 x 105  3.70  -  -  6400  3.1 x 105  -  -  -  340000  3.4 x 105  -1  -1  -1  -1  -1  (mM min ) (mM min )  DDAOY-2FGlc  The authenticity of this “substrate protection” method was assessed by using the approach to monitor inactivation of Abg by DNP-2FGlc (3.17), a known inactivator of Abg (Figure 36). An estimate of the second-order rate constant (ki/Ki) was thereby determined.  49  a)  b)  c)  d)  Figure 36: Inactivation of Abg by 3.17 monitored in the presence of saturating amounts of PNPGlc. Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation, at a concentration of 0.56 mM (a), 0.83 mM (b), and 1.1 mM (c) of PNP-Glc. d) Plot of observed inactivation rate constants vs. inactivator concentration at the indicated substrate concentrations.  The observed and corrected values for ki/Ki are summarized in Table 2. Indeed the increase in substrate concentration resulted in the expected decrease in apparent ki/Ki values, yet there are large discrepancies in the corrected second order rate constants. However, since these corrected values are in a similar range to those previously published36,38, it was concluded that  50  the substrate protection assay could be used to determine a rough estimation of inactivation parameters.  Table 2: Kinetic parameters for inactivation of Abg by 3.17, in the presence and absence of substrate, and comparison to known inactivation parameters.  app  Inactivator  DNP-2FGlc  corrected  ki/Ki ki/Ki [PNP-Glc] ki (min-1) Ki (mM) -1 -1 (mM) (mM min ) (mM-1 min-1)  Ref  0  123  245  -  502  Street, 1992  0  283  245  -  1160  Namchuk, 1995  0.556  -  -  128  1040  -  0.834  -  -  59  690  -  1.11  -  -  55  840  -  To study the action of 3.15 on Abg in further detail, it was necessary to use higher concentrations of inactivator so that saturation could be reached to directly assess the contributions from each individual component, ki and Ki. Previously, the maximum concentration of inactivator assayed was determined by the speed of inactivation, limited by the speed of drawing aliquots and manual mixing. This issue can be avoided by using a stopped flow machine with automatic mixing to observe changes in absorbance from greater than 1-3 ms after mixing. One benefit of having a chromogenic leaving group on the inactivator is the ability to directly observe the inactivation reaction by monitoring the release of DDAO. Reaction of Abg with excess amounts of 3.15 resulted in a pseudo-first order reaction whose derivative was fitted to an exponential decay function using the Applied Photophysics software. As shown in Figure 37, the observed rate constants were plotted against the inactivator concentration to give a Michaelis-Menten type curve with kinetic parameters of ki = 132 ± 8 s-1 (7920 ± 480 min-1), Ki = 6 ± 1 μM, and ki/Ki = (1.3 ± 0.3) × 106 mM-1 min-1.  51  Figure 37: Inactivation of Abg by 3.15. Plot of the observed rate constants of inactivation vs. concentration of inactivator.  There is an approximately 3-fold difference between the second order rate constants found using this method and those of the previous methods. One possible reason for this is that the concentration of 3.15 used in the previous two methods was very low. Due to the hydrophobicity of the aglycone, inactivator may stick to the sides of the plastic cell during the inactivation reaction, decreasing the actual concentration of 3.15 in solution. This loss may be apparent for reactions requiring low nanomolar concentrations of inactivator, but may become negligible when a much larger concentration of inactivator is used, such as during a direct titration by stopped flow. A similar discrepancy was also seen when 3.17 was tested as an inactivator of Abg by stopped flow assay methods by Namchuk and Withers38 where the ki/Ki found (1160 mM-1 min-1) was 2.3-fold higher than previously observed (502 mM-1 min-1). Street and Withers found that the lower the pKa value of an aglycone, the more reactive the inactivator containing said aglycone would be.36 DNP-2FGlc (3.17) bears a highly activated aglycone with a pKa of 4.0, and 3.17 was indeed reported to inactivate Abg rapidly with a ki = 283 min-1.36 The pKa of DDAO was experimentally determined to be 5.3, which is roughly 25 times less acidic than DNP, thus the ki of 3.15 should be a similar or lower value than that of 3.17. Unexpectedly, the inactivation of Abg by DDAOY-2FGlc (3.15) proceeded with a ki that was 30 times greater than that of 3.17. One possibility is that the two tautomers “DDAOY” (2.7a) and “DDAOR” (2.7b) may have different effective pKa values, and a combination of those may 52  give the experimentally determined pKa value of 5.34. The individual pKa values of 2.7a and 2.7b could not be determined as the percentage of each species in solution, as well as the rate of tautomerisation, are both unknown. Pre-steady state analysis revealed that 3.15 bound to Abg fairly tightly with a Ki of 6 μM, which is approximately 40-fold lower than that of 3.17, indicating that Abg may favour binding to large aromatic structures such as the tricyclic DDAO over smaller moieties such as DNP. To date, the novel compound DDAOY-2FGlc (3.15) is the most efficient inactivator of Abg tested, with an inactivation efficiency 400-fold greater than that of DNP-2FGlc (3.17).  2.2.2 Testing DDAOR-2FGlc as an Inactivator of Abg The use of DDAOR-2FGlc (3.16) was found to require high concentrations (60-120 μM) and long reaction times (up to 60 min) for complete inactivation of Abg. Given that compounds 3.15 and 3.16 come from the same synthetic precursors, it is conceivable that there may be contamination of 3.16 by a small amount of the highly active compound 3.15 at levels beneath what can be detected by NMR. Contaminating species were removed by treating 3.16 with 300 nM Abg as a “reagent” for removal of the highly active contaminant, and monitoring for inactivation. After 60 min of further incubation, a UV/Vis spectrum of the remaining sample was collected, which verified the presence of 80 μM of 3.16 (A460 = 0.25, 10x dilution). Compound 3.16 was separated from Abg by loading the treated solution onto a C18 SepPak reversed phase column. The treated 3.16 was indeed shown to be mildly contaminated, as the purified material reacted with an approximately two-fold lower inactivation rate constant at comparable concentrations (kobs = 0.018 min-1 at 57 μM of treated 3.16). Kinetic parameters for inactivation of Abg by DDAOR-2FGlc (3.16) of ki = 2.9 ± 0.6 h-1, Ki = 100 ± 40 μM, and ki/Ki = 0.5 ± 0.4 mM-1 min-1 were determined (Figure 38).  53  a)  b) Figure 38: Inactivation of Abg by 3.16. a) Plot of normalized residual enzyme activity vs. time at a range of inactivator concentrations. b) Plot of observed inactivation rate constants vs. inactivator concentration, fit to a Michaelis-Menten-like equation.  This finding was extremely unexpected. Previously, log(ki) has been shown to be directly proportional to the pKa values of the aglycone.36 Kinetic isotope and linear free energy relationship studies have shown that the glycosylation reaction of Abg goes through a late transition state, indicating that the glycosidic bond is largely cleaved, thus that the leaving group is largely ionized (Figure 39).7,65,68 Once ionized, the DDAO moieties of 3.15 and 3.16 are identical through electron delocalization, and hence share the same physical properties, which includes an identical pKb value. Consequently, we expected compounds 3.15 and 3.16 to 54  inactivate Abg with similar ki values. However, the behaviours of these two DDAO 2Fglycosides are clearly different, with differences in ki/Ki of more than 6 orders of magnitude.  Figure 39: Glycosylation transition state of 3.15 (top left) and 3.16 (top right). The glycosylation of Abg proceeds via a late transition state. Once ionized, the DDAO moieties of 3.15 and 3.16 are identical through electron delocalization and can be represented by a resonance hybrid structure (below).  This trend was verified through an identical experiment performed using DDAOY-2FGal (4.15) and DDAOR-2FGal (4.16).  2.2.3 Testing DDAOY-2FGal and DDAOR-2FGal as Inactivators of Abg GH family 1 β-glucosidases such as Abg have the highest hydrolysis rates for βglucosides, but are also able to hydrolyse a wide variety of substrates, such as β-galactosides and β-xylosides, albeit at reduced rate and/or with worse binding. Thus, aryl 2Fgalactosides can also serve as inactivators of Abg. The trend observed with 3.15 and 3.16 was again seen with the galactoside inactivator analogues, DDAOY-2FGal (4.15) and DDAOR-2FGal (4.16). Figure 40 shows the time-dependent inactivation of Abg by 4.15, with a second order rate constant (ki/Ki) of 6180 mM-1 min-1. The inactivation efficiency of 4.15 exceeded that of 2FGalF 55  by over 7000 times (ki/Ki = 0.81 mM-1 min-1)37, and that of DNP-2FGlc (3.17) by 5-fold (ki/Ki = 1160 mM-1 min-1).38 However, it is still more than 2 orders of magnitude below the inactivation efficiency of the corresponding 2Fglucoside DDAOY-2FGlc (3.15) (1.3 × 106 mM-1 min-1).  a)  b)  Figure 40: Inactivation of Abg with 4.15 and 4.16. a) Plot of normalized residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an equation for exponential decay. b) Plot of observed inactivation rate constants vs. concentration of 4.15.  Compound 4.16, however, functioned as a poor inactivator. A trial inactivation of Abg showed the pseudo-first order decrease in Abg activity with respect to time, but again only at a high (134 μM) concentration of 4.16, a concentration a thousand-fold higher than that used for 4.15, with an accompanying observed rate constant of only 2.8 min-1 (Figure 40). Testing higher concentrations of 4.16 was not possible due to its limited water solubility. This is a very surprising result as it seems to mean that the placement of the two chlorines on the ring can affect the reaction rate by several orders of magnitude. It is possible that the two large chlorine groups at the distal end of the “DDAOR” aglycone may prevent binding to Abg altogether; if that is the case, then DDAOR-linked glycosides (without the C-2 fluorine) should also act as poor substrates for Abg. Binding of the “DDAOY” aglycone could also be simply favoured by the Abg active site over “DDAOR”; this is in some ways reminiscent of the published results with TCP-Glc (3.19), a substrate which also contains an aglycone with orthochlorine; this was a highly active substrate for Abg with a catalytic efficiency of 56  26000 mM-1 s-1.65 However, the corresponding 2Fglucoside was a very poor inactivator with an inactivation efficiency of only 14.5 mM-1 min-1.36 In order to probe this further, studies were performed with a series of related substrates.  2.2.4 Testing DDAOY-Glc, DDAOR-Glc and DAO-Glc as Substrates of Abg  Three acridinonyl glycosides, DDAOY-Glc (3.20), DDAOR-Glc (3.21) and DAO-Glc (3.22) whose structures are shown in Figure 41, were obtained from Dr. Hong-Ming Chen and tested as substrates of Abg. Kinetic parameters were also measured for PNP-Glc as a control under the same conditions and the results agreed with those found in the literature.65 The Michaelis-Menten parameters obtained are shown in Table 3 and are compared to those of two representative substrates of Abg for which deglycosylation is the rate limiting step, DNP-Glc (3.18)38 and TCP-Glc (3.19)65. Compounds 3.20, 3.21 and 3.22 were all found to be excellent substrates of Abg. Indeed with kcat/Km values over 107 M-1s-1, they likely are reacting at or near diffusion-limited rates, as discussed later.  Figure 41: Structures of selected aryl glycosides. Compounds 3.20, 3.21 and 3.22 were tested as substrates of Abg in this work.  57  Table 3: Michaelis-Menten parameters for the hydrolysis of aryl β-glucosides by Abg.  -1  kcat (s ) Km (mM) -1  -1  kcat /Km (mM s )  DDAOY-Glc (3.20)  DDAOR-Glc (3.21)  DAO-Glc (3.22)  DNP-Glc (3.16)  TCP-Glc (3.18)  103 ± 4  122 ± 8  100 ± 10  130  240  3.7 ± 0.5  7±1  6±2  22  9.2  28000  17000  17000  5960  26000  Kempton and Withers observed that similar log(kcat) values were obtained for the hydrolysis of any aryl glucoside containing an aglycone with a pKa under 7.5 (Figure 42), and from these and other data deduced that deglycosylation was the rate-determining step for such substrates.65 The pKa values of DDAO (5.34) and DAO (7.0) were both ascertained to be below this threshold pKa, thus deglycosylation is likely to be rate limiting in all three cases. The steadystate parameters for the hydrolysis of compounds 3.20, 3.21 and 3.22 show that all three have similar turnover numbers (log(kcat) = 2.01, 2.08, 2.00 respectively), and are similar to those of other deglycosylation rate-limited substrates of Abg, e.g. 3.18, log(kcat) = 2.11; 3.19, log(kcat) = 2.38.  Figure 42: Brønsted plots of log(kcat) vs. aglycone pKa for the hydrolysis of a series of aryl glucosides by Abg. Graph and data adapted from Kempton65 and Namchuk38.  58  A comparison of the Km values again appears to show that Abg has greater binding affinity to large aromatic substituents such as DAO and DDAO than to smaller aryl aglycones, such as DNP. The low Km of compound 3.20 may suggest that the presence of ortho-chlorines also positively contributes to binding to Abg.65 However, the Michaelis-constant, Km, is an apparent binding constant that is inversely related to the concentration of the covalent glycosylenzyme intermediate in addition to the Michaelis complex E•S (  [ ][ ] [ • ]  ). When k3 is small  relative to k2 as in the case of a deglycosylation-limited enzyme reaction, there is a build-up of glycosyl-enzyme intermediate which contributes towards a small Km value (recall Km= ). The dissociation constant, Kd, is a measure of the true enzyme-substrate affinity, and can only be determined through pre-steady state kinetic studies. In an attempt to determine the Kd of compounds 3.20 and 3.21, Abg was reacted with compounds 3.20 and 3.21 and aglycone release was monitored by stopped flow. However, no burst phase was seen for either compound, even at lowered temperatures (down to 5°C). This may be because the reaction of compounds 3.20 and 3.21 is diffusion-controlled – that is, the rate-determining step is the formation of the Michaelis complex. Since a requirement of observing a pre-steady state burst is the accumulation of the glycosyl-enzyme intermediate, if both the glycosylation and deglycosylation steps occur at a greater rate than binding, no accumulation of the glycosyl-enzyme intermediate is possible.  Figure 43: Brønsted plots of log(kcat/Km) vs. aglycone pKa for the hydrolysis of a series of aryl glucosides by Abg. Graph and data adapted from Kempton65 and Namchuk38.  59  Kempton and Withers had observed that there was a reasonable correlation of the log(kcat/Km) to the pKa of the phenol nucleofuge65 (as the hydrolysis efficiency kcat/Km directly reflects the rate of the first irreversible step, glycosylation), up to a log(kcat/Km) value of approximately 4 when a “downturn” in the dependence was observed (Figure 43). This decrease in log(kcat/Km) value is especially apparent for the hydrolysis of DNP-Glc (3.18), a highly activated substrate, whose value falls far below the predicted correlation. Previous data for Abg (and almond β-glucosidase69) suggest that the hydrolysis rate for 3.18 may be partially limited by diffusion.70 Thus, it is possible that the rates of hydrolysis of compounds 3.20, 3.21 and 3.22 are also reaching diffusion control. If so, then the true glycosylation rates would be “masked” by diffusion control, making any differences between the leaving group abilities of “DDAOY” and “DDAOR” unobservable. In an effort to gain further insight into this behaviour, the inherent reactivities of the “DDAOY”, “DDAOR” and “DAO” aglycones were explored through the non-enzymatic hydrolysis of 3.20, 3.21 and 3.22.  2.2.5 Measurement of Spontaneous Hydrolysis Rate Constants of DDAOY-Glc, DDAORGlc and DAO-Glc  The relative inherent (i.e. non-enzymatic) reactivities of these compounds were investigated by monitoring the spontaneous release of DDAO or DAO from 3.20, 3.21 and 3.22 in pH 6.80 buffered solution at 50°C in the absence of enzyme (Figure 44). Solute-solute interactions were minimized by carrying out the experiment at dilute solutions of glycoside (~10 μM). Compound 3.20 hydrolysed completely within 10 hours, while no significant release of aglycone was observed in the case of compounds 3.21 and 3.22, even after 17 hours. The rate constant for spontaneous hydrolysis of 3.20 at 50°C was determined by fitting the trace observed to a first-order equation in GraFit 7.0. Rate constants for spontaneous hydrolysis of 3.21 and 3.22 were measured by following initial rates of DDAO or DAO release at a range of substrate concentrations. Since contamination was possible from a small amount of 3.20 which would confound initial rates measurements, aglycone release was monitored for over 14 hours. During this time any contaminating 3.20 should have completely hydrolysed, based on 60  its half-life of 1.9 h. The results are shown in Table 4, and are compared to those of DNP-Glc (3.18).70  a)  b)  Figure 44: A plot of the spontaneous hydrolysis of several compounds over time in pH 6.80 sodium phosphate buffer, 50°C. a) Absorbance vs. time; b) Rate of hydrolysis (A/min) vs. time.  Table 4: Rate constants and half-lives for spontaneous hydrolysis of select aryl β-glucosides at 50°C.  Rate constant of -1  hydrolysis (kobs, h ) t1/2 (h)  DDAOY-Glc (3.20)  DDAOR-Glc (3.21)  DAO-Glc (3.22)  DNP-Glc68 (3.18)  0.36 ± 0.04  (5 ± 1) x 10-4  (1.0 ± 0.2) x 10-4  0.155  1.9 ± 0.2  1400 ± 300  (5 ± 1) x 103  4.4  Rate constants for hydrolysis of 3.20 were also measured at a series of pH values, as shown in Figure 45. The rate constant was pH-independent below pH 7.0, suggesting that in this range, the rate of hydrolysis was not dependent upon protonation.  61  Figure 45: Plot of log(kobs) for hydrolysis of 3.20 at 50°C at a range of pH values.  Compounds 3.20 and 3.21 are attractive substrates for use in assays, as their hydrolysis efficiencies (kcat/Km) for Abg are similar to that of 3.18 (Section 2.2.4). However, the rate constants for hydrolysis reveal 3.20 to be extremely prone to decomposition, hence an impractical substrate. In contrast, the high stability of 3.21 makes it a more appropriate substrate. The rate constant of hydrolysis for 3.20 is almost 3 orders of magnitude greater than those for 3.21 and 3.22. The reason for this large rate difference is not clear, but does reflect what is seen in the inactivation rate constants of the fluorosugar derivatives. One may imagine the large ortho-chloro groups of 3.20 contributing to ground-state destabilization of the molecule through steric strain; if so then the glycosidic bond may be elongated in DDAOY glycosides compared DDAOR glycosides. The crystal structures of per-O-acetylated DDAOY-2FGal (4.13) and DDAOR-2FGal (4.14, Section 2.1.6) were examined for evidence of this by comparing the C1-O1 (glycosidic) and C1-O5 bond lengths between 4.13 and 4.14, but any differences were not discernible. However, the crystal structures do show that the DDAO moiety is oriented more perpendicular to the plane of the monosaccharide ring in 4.13 than in 4.14 (75° and 50° respectively), possibly generating stress on the glycosidic bond and promoting dissociation of the aglycone. To probe this further, substrates and inactivators with phenyl or acridinonyl nucleofuges containing ortho-dimethyl groups could be tested for activity against Abg, as methyl groups will mimic the steric bulk of chlorines yet should not significantly affect the acidity of the molecule. Some proposed compounds are shown in Figure 46. 62  Figure 46: Glucosides containing aryl nucleofuges as possible substrates (R = OH) or inactivators (R = F) of Abg to probe the contribution of steric strain to reaction rate.  Another explanation may be that the initial assumption that rapid electron delocalization of the transition-state DDAO anion was incorrect. In some cases, electron delocalization has been shown not to be completely synchronous with bond breakage; this follows the Principle of Non-perfect Synchronization (PNS).71 If applied here, the two DDAO anions would now contain electron density at the phenoxide, and thus be unequal structures (Figure 47). In 3.20, the presence of strongly electron-withdrawing dichloro-substituents adjacent to O1 may be capable of stabilizing a negative DDAOY phenoxide, thereby lowering the activation energy. Compound 3.21 lacks appropriately positioned ortho-chlorines, so the phenoxide it forms enjoys little inductive stabilization, which in turn increases activation energy, and is reflected by the poor leaving group capability of DDAOR.  63  Figure 47: Transition states for hydrolyses of DDAOY-Glc (3.20, top left) and DDAOR-Glc (3.21, top right) are depicted with the negative charge concentrated on the phenolate. Resonance of DDAOY and DDAOR into a hybrid structure (below) may be slow, hence the two transition states are unequal. The DDAOY anion may be more stabilized due to the inductive effects of nearby electron-withdrawing chloro groups (blue arrow).  A further possible hypothesis invokes halogen bonding, in which the σ* orbital of the chlorine-carbon bond could interact favourably with a neighbouring Lewis base, such as the DDAOY phenolate (Figure 48).72 This interaction could further stabilize the DDAOY anion and facilitate dissociation.  Figure 48: Transition state for the hydrolysis of DDAOY-Glc (3.20) showing halogen bonding interactions (hash lines) between the ortho-chlorines and the phenoxide.  Although the geometry of the C-Cl-O interaction is generally reported to be almost linear, there may be some angular dependence on the strength.73,74 Evidence from crystal structures of anionic DDAO may reveal whether halogen bonding plays a part in stabilizing the phenolate,  64  such as a characteristic shortening of the Cl-O distance below the sum of the radii of C and O-, and a lengthening of the Cl-C bond above the sum of their van der Waals radii.72 The spontaneous hydrolysis results suggest that compounds containing the DDAOY aglycone are inherently more activated than those containing DDAOR. Thus, the high inactivation efficiencies of 3.15 and 4.15 (ki/Ki = 1.3 × 106 mM-1 min-1 and 6180 mM-1 min-1 respectively) can be rationalized to be partially due to the high inherent reactivity of the DDAOY aglycone. Likewise, compounds 3.16 and 4.16 are poor inactivators of Abg because the DDAOR aglycone may not be activated enough to overcome the high glycosylation transition state energy barrier caused by the electron-withdrawing fluorine group of 2Fglycosides. This effect is not seen in the glucoside substrates, as binding of both substrates seems to be diffusion-controlled, thus the chemical steps are faster in both cases. However, non-enzymatic reactivity accounts for only three of the six orders of magnitude difference in inactivation efficiency between 3.15 and 3.16. Perhaps the exclusion of solvent inside the enzyme active site may augment effects of inductive stabilization and/or halogen bonding interactions, resulting in a larger difference in inactivation rates.  2.2.6 Testing DDAOY-2FGlc as an Inactivator of β-Glucocerebrosidase (GCase)  Given the success of DDAOY-2FGlc (3.15) as an inactivator of Abg, its ability to inactivate β-glucocerebrosidase, a lysosomal enzyme crucial in the cause and treatment of Gauchers disease, was also tested.  Figure 49: Glucosylceramide, a glycolipid and substrate of GCase.  Gauchers disease is a lysosomal storage disorder caused by the deficiency of βglucocerebrosidase (GCase, EC 3.2.1.45). This GH family 30 retaining β-glucosidase catalyses the hydrolysis of glucosylceramide – an intermediate in glycolipid metabolism – preventing the 65  toxic build-up of these molecules in cells. Individuals with activity levels of GCase below a certain threshold value are likely to present with the clinical symptoms of Gauchers disease.75 The natural substrate of GCase is a glycolipid, so it was thought that the incorporation of a hydrophobic aglycone like DDAO may encourage tight binding and good inactivation by decreasing the dissociation constant Ki. Figure 50 shows the time-dependent inactivation of GCase by 3.15.  a)  b) Figure 50: Inactivation of GCase by 3.15. a) Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation. b) Plot of observed inactivation rate constants vs. inactivator concentration fitted to a Michaelis-Menten equation.  66  Results show that GCase was inactivated by 3.15 only twice as efficiently (ki/Ki = 0.049 mM-1 min-1) as by the known inactivator 2FGlcF (ki/Ki = 0.023 mM-1 min-1)76, and four times as efficiently as by DNP-2FGlc (3.17, ki/Ki = 0.012 mM-1 min-1).77 Due to the limited solubility of 3.15 in buffer (approximately 700 μM), full saturation was not achieved so the individual ki and Ki values listed must be interpreted cautiously. However, the predicted inactivation rate constant for 3.15 (ki = 0.043 min-1) is similar to that found for 3.17 (ki = 0.030 min-1), while binding of 3.15 is slightly better (Ki = 0.9 ± 0.2 mM) than 3.17 (Ki = 2.5 mM) but both bound much better than 2FGlcF for which no saturation was observed, even at a 18 mM concentration.76 This strongly suggests that GCase binds larger, hydrophobic aglycones such as DDAO and DNP better than it does a simple fluoride. Rempel and co-workers’ phosphate-based aglycones (Figure 51) have up to 4300-fold greater inactivation efficiency for GCase over 2FGlcF.77 It is believed that the incorporation of the two alkyl chains make for a much more successful substrate mimic than a simple aryl aglycone.  Figure 51: Efficient inactivators of GCase containing phosphate-based aglycones with two alkyl chains.  Compound 3.15 could still function as an active site titrating agent for GCase. Even though its moderate inactivation efficiency for GCase requires use of high inactivator concentrations or longer reaction times, the release of a highly UV-active aglycone offers the advantage of greater ease and sensitivity of detection.  2.2.7 Testing DDAOY-2FGal as an Inactivator of LacZ E. coli β-galactosidase (LacZ, EC 3.2.1.23) is a GH family 2 glycosidase that hydrolyses lactose. Its mechanism and structure have been thoroughly studied and it has been employed as a reporter enzyme in many biological systems.53 Figure 52 shows the time-dependent inactivation 67  of LacZ when treated with DDAOY-2FGal (4.15). The kinetic parameters for inactivation of LacZ by 4.15 are summarized in Table 5 and compared to those of DNP-2FGal (4.17), a known inactivator of LacZ.78  a)  b)  Figure 52: Inactivation of LacZ by DDAOY-2FGal (4.15). a) Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation. b) Plot of observed inactivation rate constants vs. inactivator concentration fit to a Michaelis-Menten equation. Table 5: Kinetic parameters for the inactivation of LacZ by DDAOY-2FGal (4.15) and compared to that by DNP-2FGal (4.17).  Enzyme Tested LacZ  Inactivator  ki (min-1)  Ki (mM)  ki/Ki (mM-1 min-1)  Ref  DDAOY-2FGal  2.7  9.4  280  --  DNP-2FGal  2.4  7100  0.34  McCarter, 1992  Compound 4.15 was found to be a remarkable inactivator of LacZ with an 800-fold greater efficiency than 4.17, mainly attributed to the tight binding of the DDAOY moiety in the active site.  68  2.2.8 Testing DDAOY-2FXyl and DNP-2FXyl as Inactivators of B. halodurans C-125 βXylosidase (β-Xyl)  The exo-β-xylosidase from Bacillus halodurans (β-Xyl) is a member of GH family 52, a family that has not been studied in much detail. This family features glutamate as a catalytic nucleophile and aspartic acid as a catalytic acid/base.79 This xylosidase was recently converted into a glycosynthase80, a mutant glycosidase with no hydrolytic activity that instead catalyses glycosidic bond formation using an unnatural, activated glycosyl donor. DDAOY-2FXyl (5.15) was tested as an inactivator for β-Xyl; parallel experiments were performed with DNP-2FXyl (5.17). Inactivator (5.15 or 5.17) was incubated with β-Xyl and time-dependent loss of activity was monitored by removal of aliquots at time intervals and measuring residual enzyme activity by assay with substrate PNP-Xyl. Figure 53 shows the time-dependent inactivation of β-Xyl by (a) 5.15 and (b) 5.17 at a range of inactivator concentrations.  a)  b)  Figure 53: Inactivation of β-Xyl by a) 5.15 and b) 5.17. Plot of residual enzyme activity vs. time at a range of inactivator concentrations, fitted to an exponential decay equation.  69  Figure 54: Inactivation of β-Xyl by DDAOY-2FXyl (5.15) and DNP-2FXyl (5.17). Plot of observed inactivation rate constants vs. inactivator concentration fit to a Michaelis-Menten curve.  The rate constants for inactivation were plotted against inactivator concentration as seen in Figure 54. The kinetic parameters found are summarized in Table 6. Compound 5.17 appears to be the better inactivator with a 5-fold greater inactivation efficiency (ki/Ki) than that of compound 5.15. The inactivation rate constants (ki) of each inactivator are similar, indicating that the principal difference is the 4-fold lower dissociation constant of 5.17 at 25 ± 5 μM when compared to compound 5.15 (110 ± 30 μM). Replacement of nitro-groups with chlorines or aliphatic carbons may take away potential hydrogen-bond interactions with certain B. halodurans β-xylosidase active site residues. Table 6: Kinetic parameters for inactivation of β-Xyl by DDAOY-2FXyl (5.15) and DNP-2FXyl (5.17).  Enzyme Tested β-Xyl  -1  -1  -1  Inactivator  ki (min )  Ki (mM)  ki/Ki (mM min )  DDAOY-2FXyl  2.0 ± 0.4  110 ± 30  18 ± 5  DNP-2FXyl  2.4 ± 0.2  25 ± 5  100 ± 20  70  2.2.9 Testing DDAOY-2FCel as an Inactivator of Cellulases  2.2.9.1  Testing DDAOY-2FCel as an Inactivator of C. fimi Endo-xylanase (Cex)  Cellulomonas fimi is a Gram-positive soil bacterium. It expresses several different carbohydrate-degrading enzymes when grown on glycan substrates, including C. fimi endoxylanase, a GH family 10 retaining hydrolase with the highest activity on 1,4-xylans. Its shortened name, Cex, was designated due to its lower, but significant activity on aryl βcellobiosides, originally misidentifying it as an exo-cellulase. This enzyme was chosen as the model enzyme for DDAOY-2FCel (6.15) due to its known activity with DNP-2FCel (6.17). Reaction of 6.15 with C. fimi Cex was monitored by direct observation of DDAO release by following absorbance changes at 600 nm, practical in this case since the inactivation halflives at the concentrations of inactivator measured were on a timescale of minutes. Figure 55 shows the Michaelis-Menten type curves generated by plotting the observed rate constants (kobs) against inactivator concentration.  Figure 55: Inactivation of Cex by 6.15; plot of observed inactivation rate constants (kobs) vs. inactivator concentration.  Compound 6.15 indeed acted as a covalent inactivator of Cex. Kinetic parameters revealed 6.15 to be a 3-fold better inactivator than the known inactivator 6.17, based on ki/Ki values.81  71  Table 7: Kinetic parameters for the inactivation of Cex by DDAOY-2FCel (5.15) and DNP-2FCel (5.17).  Enzyme Tested Cex  2.2.9.2  Inactivator  ki (min-1)  Ki (mM)  ki/Ki (mM-1 min-1)  Ref  DDAOY-2FCel  0.17 ± 0.05  90 ± 40  2.0 ± 0.2  --  DNP-2FCel  0.067  110  0.612  Tull, 1994  Testing DDAOY-2FCel as an Inactivator of C. fimi Endo-glucanase D (CenD)  To investigate if the scope of 6.15 could be extended, compound 6.15 was tested as an inactivator for C. fimi endo-glucanase D (CenD/Cel5A, EC 3.2.1.4), a GH family 5 endoglucanase. Catalysis was shown to proceed in a retaining fashion as incubation with cellopentaose produced the β-anomer of cellotriose as the major product.82 However, CenD was not inactivated by 6.15.  2.2.9.3  Testing DDAOY-2FCel as an Inactivator of T. reesei Cellulases  The Trichoderma reesei fungal cellulolytic system secretes a large quantity of glycosidases and cellulases to degrade cellulose, and has been used in industrial processes for renewable bio-ethanol2 and glucose production.1 The T. reesei cellulase system consists of four principal enzymes: two cellobiohydrolases (CBH I and II) and two endo-glucanases (EG I and EG II).2 The characteristic properties of these four cellulases are listed in Table 8. An activitybased profiling approach was developed for these industrially important cellulases83; in this work, T. reesei cellulases were investigated as possible enzyme targets for the mechanism-based inactivator DDAOY-2FCel (6.15).  72  Table 8: Characteristic properties of the four chief T. reesei cellulases.  Family (CAZy) Stereochemistry Endo/exo MW (kDa)  CBH I/Cel7A GH7 Retaining Exo 54  CBH II/Cel6A GH6 Inverting Exo 50  EG I/Cel7B GH7 Retaining Endo 48  EG II/Cel5A GH5 Retaining Endo 44  Initial examination of the reaction of inactivator and enzyme was performed by direct titration of 500 μM DDAOY-2FCel with enzyme for 3 hours. As the optimal pH of T. reesei cellulases is 5.0, only a third of the total DDAO released is ionized, so to get a significant burst (ΔA = 0.1 - 0.2), enzyme concentrations of 11 μM to 18 μM were used. CBH II is an inverting enzyme, so no accumulation of covalent glycosyl enzyme intermediate was expected. Indeed, no change in absorbance was observed, indicating that CBH II was unable to react with 6.15. Reaction of the retaining enzymes EG II and CBH I with 6.15 resulted in no DDAO release; previous studies also described lack of inactivation of these enzymes by 6.17 and DNP-2FLac.84,83 Compound 6.17 was, however, reported to inactivate EG I with values of ki = 0.022 min-1, Ki = 3.1 mM and ki/Ki = 7.1 × 10-3 mM-1 min-1.83 EG I was likewise inactivated by 6.15, but with improved inactivation kinetic parameters of ki = 0.033 ± 0.002 min-1, Ki = 0.22 ± 0.03 mM, and ki/Ki = 0.15 mM-1 min-1, a 20-fold improvement over 6.17 (Figure 56).  Figure 56: Inactivation of T. reesei EG I by 6.15. Plot of initial rate of inactivation as a function of inactivator concentration, fit to a Michaelis-Menten type equation.  73  Compound 6.15 was thus able to efficiently inactivate one of the four chief T. reesei cellulases and may therefore prove valuable in our activity-based profiling work.  2.3  Active Site Titrations of Abg  To evaluate whether DDAOY 2Fglycosides can be usefully applied towards the measurement of low enzyme concentrations, a proof-of-concept active site titration was performed using the model glucosidase Abg and DDAOY-2FGlc (3.15). Figure 57 plots the concentration of DDAO released against the Abg concentration determined by absorbance at 280 nm (  = 2.20 cm-1) on a Cary 300 Bio UV/Vis spectrophotometer.65 A slope of 1.02 ± 0.02  was found after titration by 3.15. An active site titration of the same stock of Abg by DNP-2FGlc (3.17) was performed in parallel for verification, as this compound has been shown previously to be an accurate active site titrant of Abg.36 A slope of 1.01 ± 0.04 for 3.17 was found, demonstrating that the titration of Abg by 3.15 was valid. Using absorbance, amounts down to 70 nM (25 pmol) of Abg could be reliably determined (slope = 0.98 ± 0.02); however, measurement of amounts under 50 nM was accompanied by considerable noise (Figure 59).  Figure 57: Active site titrations of Abg by DDAOY-2FGlc (3.15) and DNP-2FGlc (3.17) by UV/Vis absorbance. The DDAO and DNP released were monitored by absorbance at 600 nm and 400 nm respectively, and plot as a function of Abg concentration.  74  Figure 58: Active site titration of Abg by 3.15 by UV/Vis absorbance. The DDAO released was monitored by absorbance at 600 nm and plotted as a function of Abg concentration.  Use of a fluorescent read-out should significantly improve sensitivity. Active site titrations with 3.15 were monitored by stopped flow (Applied Photophysics) or by using a standard fluorimeter (Cary Eclipse) at excitation and emission wavelengths of 600 nm and 656 nm, respectively.  Figure 59: Active site titration of Abg by 3.15 on stopped flow at excitation/emission wavelengths of 600 nm/656 nm, 435 V. a) DDAO fluorescence release as a function of time at a series of Abg concentrations. No significant turnover of the covalent glycosyl-enzyme intermediate is observed. b) Plot of fluorescence response at 656 nm vs. Abg concentration.  75  By stopped flow (Figure 59), one can follow the pre-steady state reaction between enzyme and inactivator, but due to the small volumes used (150 μl), error in measurement again arises at concentrations below 50 nM. To increase the precision of the measurement, larger reaction volumes were used (500 μl) and the amount of DDAO released was quantified with a Cary Eclipse fluorimeter to give a correlation between the calculated and experimentally determined enzyme concentrations of 0.86 ± 0.01 (Figure 60). Compared to UV/Vis absorbance measurements, detection by fluorescence can be sensitized a further ten-fold to determine Abg concentrations as low as 3 nM (2 pmol of Abg). This is a satisfactory range for use in the detection of enzyme concentrations during screening processes.  Figure 60: Active site titration of Abg by 3.15 on the Cary Eclipse fluorimeter. Fluorescence response was measured at excitation/emission wavelengths of 600 nm/656 nm at 600 V.  2.4  Summary of Inactivation Parameters  The kinetic parameters for inactivation of selected enzymes by DDAO 2Fglycosides and their DNP-analogues are displayed in Table 9.  76  Table 9: Kinetic parameters for inactivation of selected glycosidases by DDAO 2Fglycosides.  Enzyme Tested  Abg  -1  LacZ β-Xyl Cex EG I  2.5  -1  ki (min )  Ki (mM)  ki/Ki (mM min )  DDAOY-2FGlc  7800 ± 500  6±1  (1.3 ± 0.2) x 10  DDAOR-2FGlc  0.05 ± 0.01  100 ± 40  0.5 ± 0.4  --  DNP-2FGlc  123  245  502  Street, 1992  DNP-2FGlc  283  245  1160  DDAOY-2FGal  --  --  6180 ± 80  Namchuk, 1995 --  2FGalF  2.6  3200  0.81  Withers, 1988  900 ± 200  0.05 ± 0.01  --  DDAOY-2FGlc 0.042 ± 0.008 GCase  -1  Inactivator  6  Ref --  DNP-2FGlc  0.030  2.5  0.012  Rempel, 2003  2FGlcF  --  --  0.023  Miao, 1994  DDAOY-2FGal  2.58 ± 0.01  9.8 ± 0.9  280 ± 30  --  DNP-2FGal  2.4  7100  0.34  McCarter, 1992  DDAOY-2FXyl  2.0 ± 0.4  110 ± 30  18 ± 5  --  DNP-2FXyl  2.4 ± 0.2  25 ± 5  100 ± 20  --  DDAOY-2FCel  0.17 ± 0.04  90 ± 40  2.0 ± 0.2  --  DNP-2FCel  0.067  110  0.612  Tull, 1994  220 ± 30  0.15 ± 0.02  --  3100  0.0071  Chen, 2011  DDAOY-2FCel 0.033 ± 0.002 DNP-2FCel  0.022  Conclusions and Future Directions  This work has shown that DDAO glycosides of 2-deoxy-2-fluoro sugars can serve as active site titrating agents. Since DDAO is an asymmetrical molecule with two possible tautomeric forms, two isomeric glycosides are formed: a yellow one (DDAOY-glycoside) with the chlorines close to the sugar, and a red one (DDAOR-glycoside) with the chlorines distal. Six novel compounds were evaluated as potential inactivators for a variety of enzymes: DDAOY-2FGlc, DDAOR2FGlc, DDAOY-2FGal, DDAOR-2FGal, DDAOY-2FXyl and DDAOY-2FCel. Inactivation of  77  the model glucosidase Abg by DDAOY- and DDAOR- 2Fglucosides and 2Fgalactosides was probed in detail. Enzymatic evaluation revealed a surprisingly large difference between the inactivation rate constants for DDAOY-2FGlc (3.15) and DDAOR-2FGlc (3.16): Abg was inactivated by the former with an efficiency of 1.3 × 106 mM-1 min-1, six orders of magnitude greater than the latter (ki/Ki = 0.5 mM-1 min-1). Exploration of the differences in reactivity between the DDAOY and DDAOR aglycones led to enzymatic evaluation of DDAOY-Glc (3.20) and DDAOR-Glc (3.21), as well as the reduced version DAO-Glc (3.22) as possible substrates of Abg. All three compounds are indeed excellent substrates, reacting at diffusion-controlled rates. The difference in reactivity was partially attributed to the inherent lability of the DDAOY aglycone, as DDAOY-Glc was found to be 700 times more partial to hydrolysis than DDAORGlc. Further studies to explore the basis of differences in enzymatic reactivity between DDAOY and DDAOR could include measurement with an acid/base mutant of Abg with 3.20 and 3.21. With such mutants, it may be possible to slow the rates of chemical steps sufficiently so that rates are no longer diffusion-controlled, thus differences of leaving group capability of each DDAO tautomer may be directly observed. Another future development would be the synthesis of substrates and inactivators containing aglycones with ortho-dimethyl substituents in place of the chlorines to probe the effect of any steric strain imparted upon the “DDAOY” aglycone. Methyl groups are similar in size to chloro groups, yet would have opposite effects on the acidity of the phenol. DDAOY 2Fglycosides were also shown to be effective inactivators of a variety of retaining glycosidases, including E. coli β-galactosidase, T. reesei EG I, and C. fimi Cex, and B. halodurans β-xylosidase. Studies are currently underway to expand our current DDAO active site titrating agent repertoire to include neuraminic acid derivatives, as well as to improve current compounds to target a wider range of cellulases. Finally, DDAOY-2FGlc (3.15) was shown to be a valuable tool for determining the concentration of Abg in a proof-of-concept active site titration. Through fluorescence detection, enzyme concentrations down to 3 nM could be measured. Application of DDAOY 2Fglycoside inactivators and DDAOR-substrates in metagenomic and directed evolution screens is also in progress.  78  3  Materials and Methods  3.1  Generous Gifts  T. reesei cellulases were obtained from Iogen Corporation. Other generous gifts include cellulases Cex and CenD from Dr. Emily Kwan, DNP-2FXyl (5.17) from Dr. Ethan GoddardBorger, and expression vectors containing the wild type Abg gene from Dr. David Kwan. Dr. Hong-Ming Chen provided the acridinonyl substrates DDAOY-Glc (3.20), DDAOR-Glc (3.21), DAO-Glc (3.22), as well as per-O-acetylated DDAOR-2FGal (4.14) used in single crystal X-ray diffraction studies.  3.2  General Materials and Methods All materials were obtained from commercial suppliers (Sigma-Aldrich®, Fisher  Scientific®). Methanol was distilled in the presence of Mg. Acetonitrile was distilled in the presence of CaH2. Deionized water was prepared using a Millipore-Directed QTM 5 Ultrapure Water System. TLC was performed on a Merck pre-coated 0.2 mm aluminum-backed sheets of Silica Gel 60F254. TLC was stained in 10% sulfuric acid in EtOH or 10% ammonium molybdate in 2 M H2SO4, followed by charring. Flash column chromatography was performed using 230400 mesh Silicycle silica gel. Proton NMR spectra were acquired on a Bruker 300 MHz, 400 MHz Inverse or 400 MHz Direct Spectrometer. Mass spectra were obtained using a Waters ZQ Mass Detector equipped with ESCI ion source. LacZ was commercially available from SigmaAldrich®. GCase was obtained as the commercially available recombinant form Cerezyme®.  3.2.1 General Procedure for Acetylation of Free Sugars  Acetylation of free sugars was performed in Ac2O containing 0.1 mole-equivalent of NaOMe. The mixture was refluxed for 1 hour, then poured onto ice and stirred overnight. The crude product, a white powder, was filtered and recrystallized from hot EtOH to give the per-Oacetylated sugar.  79  3.2.2 General Procedure for Deacetylation  The per-O-acetylated glycoside was dissolved in dry MeOH (1-2 ml MeOH per 10 mg sugar) and cooled to 0°C. Sodium methoxide in MeOH (380 mM) was added drop-wise to give a final concentration of 20 mM to 50 mM NaOMe. The reaction was stirred at RT and monitored by TLC; upon completion, the reaction was neutralized with acid resin (Amberlite® IR120 H+), filtered and concentrated under reduced pressure. If needed, further purification was carried out by flash chromatography on silica gel or crystallization from MeOH. 3.2.3 General Procedure for Synthesis of α-Glycosyl Bromides  The per-O-acetylated sugar was dissolved in DCM under argon and cooled in an ice bath. Hydrogen bromide in AcOH (33% w/w) was added drop-wise and the solution was stirred at RT and the reaction monitored by TLC. Upon completion of the reaction, the mixture was diluted with EtOAc and washed with saturated NaHCO3 until basic, followed by brine. The organic layer was dried with anhydrous MgSO4, filtered and the solvent evaporated in vacuo. These products were used immediately in further syntheses.  3.2.4 General Procedure for the Synthesis of 2Fglycosides  The per-O-acetylated glycal (3,4,6-tri-O-acetyl-D-glucal, 3,4-di-O-acetyl-D-xylal, 3,6,2’,3’,4’,6’-hexa-O-acetyl-D-cellobial, 3,4,6-tri-O-acetyl-D-galactal) was dissolved in the appropriate solvents, 20 ml MeNO2/H2O (9:1) or DMF/H2O (2:1) per gram of sugar. Selectfluor® (1.2-1.5 equivalents) was added and the solution stirred at RT for overnight before refluxing for 1 hour. After cooling, the mixture was concentrated and resuspended in excess amounts of pyridine and acetic anhydride and left to stir overnight at room temperature. Upon completion of the reaction, solvents were evaporated under vacuum, then co-evaporated twice with toluene to remove residual pyridine. The mixture was diluted with EtOAc or DCM, washed successively with 1 M HCl, H2O, saturated NaHCO3 until basic, brine, then dried over anhydrous MgSO4, filtered and concentrated in vacuo. Products were purified by flash chromatography, or recrystallized to homogeneity. 80  3.2.5 General Procedure for the Coupling of DDAO and Per-O-acetyl-2Fglycosyl Bromides  All glycoside couplings were done via the Koenigs-Knorr reaction. A 25 ml round bottom flask was charged with protected 2-deoxy-2-fluoro-α-glycosyl bromide, 1.0 equivalents of DDAO, dry MeCN or DCM as solvent (1 ml/100 mg of protected sugar), 200 mg of CaSO4, 0.5 ml 2,6-lutidine and 200 mg Ag2O. The reaction was protected from light and left to stir under argon at room temperature overnight. Upon completion, the reaction was filtered through a Celite cake and washed three times with saturated NaHCO3, once with brine, dried with anhydrous MgSO4, filtered and concentrated. The products were purified by flash chromatography and/or crystallized.  3.3  Synthesis and Characterization  3-(2-Hydroxypropan-2-yl)phenol (2.2)  3-Hydroxyacetophenone (1.51 g, 11.1 mmol, 2.1) was dissolved in anhydrous ether (30 ml) and added drop-wise to a mixture of methylmagnesium iodide, prepared from CH3I (29 mmol, 2.6 eq), Mg metal (1.0 g,42 mmol) and one crystal of I2 in diethyl ether (65 ml). The reaction was stirred for another 2.5 h, then quenched by the addition of saturated NH4Cl. The mixture was diluted into EtOAc and washed with water (50 ml), brine (2 × 50 ml) and dried with anhydrous MgSO4. The mixture was filtered and the solvent evaporated under reduced pressure. The residue was purified by flash chromatography (EtOAc/PE, 1:9) yielding a pale yellow solid. Recrystallization using EtOAc and PhMe gave 1.15 g (68%) of compound 2.2 as a white powder.85 The proton NMR spectrum agreed with those previously reported.86 1H NMR (300 MHz, CD3OD) δ ppm 1.50 (s, 6 H, 2 × CH3) 6.63 (ddd, 1 H, JH6-H5 8.1 JH6-H2 1.4 JH6-H4 0.9 Hz, H6) 6.93 (ddd, 1 H, JH4-H5 8.1 JH4-H2 1.8 JH4-H6 0.9 Hz, H4) 6.94 (dd, 1 H, JH2-H4 1.8 JH2-H6 1.4 Hz, H2) 7.11 (atd, 1 H, JH5-H4=JH5-H6 8.1 JH5-H2 0.9 Hz, H5).  81  9H-1,3-Dichloro-9,9-dimethylacridin-2-one (DDAO) (1.23)  A 10 ml round bottom flask containing 2.2 (1.21 g, 8.0 mmol) and 2,6-dichloro-4(chloroimino)cyclohexa-2,5-dienone (1.67 g, 8.0 mmol, 2.3) dissolved in THF (0.75 ml) and H2O (1.0 ml) was cooled in an ice bath for 20 min. Sodium hydroxide (2.0 M, 2.3 ml) was added drop-wise via syringe for 15 min; the solution immediately turned dark blue. The reaction was cooled and stirred for a further 1.5 h. After the disappearance of all starting material, the reaction was diluted into EtOAc (60 ml) and neutralized with saturated NH4Cl (2 × 50 ml). The intermediate was reduced by washing twice with Na2S2O4 (12.5 g in 125 ml H2O). The organic phase was washed with brine (50 ml) and dried with MgSO4. After filtration and solvent evaporation under reduced pressure, the resulting syrup was dissolved in MeOH (10 ml), added into rapidly stirring, deoxygenated 2 N HCl (100 ml), and was refluxed for 1.5 h under Ar(g). The reaction mixture was cooled and extracted with EtOAc (3 × 40 ml), washed with NaHCO3 until neutral, then poured into aqueous NaIO4 (3.5 g in 100 ml H2O) and stirred rapidly for 15 min. After oxidation was complete, the organic layer was separated from the aqueous layer, washed with brine (50 ml), dried with anhydrous MgSO4, filtered, and the solvent was evaporated in vacuo. The product was purified by crystallization from hot EtOH, yielding 1.6 g (5.2 mmol, 65%) over 4 steps. The proton NMR spectrum agreed with those previously reported46. 1H NMR (300 MHz, CD3OD) δ ppm 1.88 (s, 6 H, 2 × CH3) 6.83 (dd, 1 H, JH6-H5 8.9 JH6-H8 2.3 Hz, H6) 7.03 (d, 1 H, JH8-H6 2.3 Hz, H8) 7.53 (d, 1 H, JH5-H6 8.9 Hz, H5) 7.69 (s, 1 H, H4).  1,3,4,6-Tetra-O-acetyl-2-deoxy-2-fluoro-D-glucopyranose (3.7)  Compound 3.7 was synthesized according to the General Procedure for Synthesis of 2Fsugars. Tri-O-acetyl-D-glucal (0.914 g, 3.35 mmol) and Selectfluor® (4.20 g, 11.8 mmol, 3.5 eq) were dissolved in DMF (15 ml) and H2O (6 ml) and stirred overnight at room temperature. 82  The mixture was diluted with H2O and extracted with DCM (3 × 50 ml) and dried with anhydrous MgSO4. Filtration and solvent evaporation under reduced pressure yielded a crude mixture of 2-deoxy-2-fluoro-gluco- and manno-configured hemiacetals (3.5, 3.6). The residue was dissolved in DCM (20 ml) containing pyridine (1 ml) and Ac2O (1.5 g) and stirred at room temperature for 12 hours before refluxing for one hour. Upon completion of the reaction, the mixture was worked up to produce a viscous syrup. Flash chromatography was performed (EtOAc/PE, 1:9 to 1:4) to separate per-O-acetylated 2-deoxy-2-fluoro-D-glucopyranose (3.7, 0.25 g, 1.5 mmol, 20%) from 2-deoxy-2-fluoro-D-mannopyranose (3.8) and other byproducts. ESI MS calculated for C14H19FO9: 350.1; found: 373.1 [M+Na]+. 1H NMR (400 MHz, CDCl3) δ ppm 2.04, 2.05, 2.08, 2.09, 2.18, 2.20 (s, 6 × 3 H, C(O)CH3) 3.86 (ddd, 1 H, JH5β-H4β 9.9 JH5β-H6β 5.1 JH5β-H6’β 2.2 Hz, H5β) 4.06 (m, 2 H, H6β H6’β) 4.10 (ddd, 1 H, JH5α-H4α 9.4 JH5α-H6α 4.7 JH5αH6’α  2.3 Hz, H5α) 4.28 (m, 2 H, H6α H6’α) 4.45 (ddd, 1 H, JH2β-F2β 50.6 JH2β-H3β 9.4 JH2β-H1β 8.1  Hz, H2β) 4.65 (ddd, 2 H, JH2α-F2α 48.4 JH2α-H3α 9.4 JH2α-H1α 4.0 Hz, H2α) 5.07 (t, 1 H, JH4βH5β=JH4β-H3β  9.4 Hz, H4β) 5.09 (t, 1 H, JH4α-H3α=JH4α-H5α 9.4 Hz, H4α) 5.37 (dt, 1 H, JH3β-F2β 14.5  JH3β-H2β=JH3β-H4β 9.4 Hz, H3β) 5.55 (dt, 1 H, JH4α-F2α 12.1 JH3α-H2α=JH3α-H4α 9.4 Hz, H3α) 5.78 (dd, 1 H, JH1β-H2β 8.1 JH1β-F2β 3.1 Hz, H1β) 6.42 (d, 1 H, JH1α-H2α 4.0 Hz, H1α).  9H-(1,3-Dichloro-9,9-dimethylacridin-7-one-2-yl) 3,4,6-tri-O-acetyl-2-deoxy-2-fluoro-β-Dglucopyranoside (3.13)  3,4,6-Tri-O-acetyl-2-deoxy-2-fluoro-α-D-glucosyl bromide (3.9) was made according to the General Procedure for Synthesis of α-Glycosyl Bromides: 3.7 (448 mg,1.2 mmol) was reacted with HBr in AcOH (33% w/w, 4 ml, 23 mmol) in cold DCM (3 ml) under Ar(g). After stirring at room temperature for 1 hour, the reaction mixture was worked up to give the αglucosyl bromide (3.9). This was immediately dissolved in dry MeCN (4 ml) and coupled with DDAO (327 mg, 1.21 mmol, 1 eq) according to the General Procedure for the Coupling of DDAO and Per-O-acetyl-2Fglycosyl Bromides in the presence of 2,6-lutidine (0.5 ml, 4.3 mmol), Ag2O (200 mg, 870 mmol) and CaSO4 (200 mg) overnight at room temperature. After workup 83  and solvent evaporation under reduced pressure, the resulting syrup was flash chromatographed (acetone/PE, 10%-15%) to separate residual DDAO from glycosylation products; DDAO was collected and recycled. The mixture of glycosylation products (3.13 and 3.14) were columned again (PE/DCM/EtOAc, 2:2:1) to enrich in each isomer. Compound 3.13 was crystallized from hot acetone and PE to give yellow needles (230 mg, 0.38 mmol, 32%). HRMS mass calculated for C27H26Cl2FNO9: 597.10, found: 620.0880 [M+Na]+. 1H NMR (400 MHz, CDCl3) δ ppm 1.78 (s, 3 H, CH3(Ar)) 1.80 (s, 3 H, CH3(Ar)) 2.01 (s, 3 H, C(O)CH3) 2.04 (s, 3 H, C(O)CH3) 2.12 (s, 3 H, C(O)CH3) 3.72 (ddd, 1 H, JH5-H4 10.0 JH5-H6 4.7 JH5-H6’ 2.6 Hz, H5) 4.11 (dd, 1 H, JH6’-H6 12.2 JH6’-H5 2.6 Hz, H6’) 4.20 (dd, 1 H, JH6-H6’ 12.2 JH6-H5 4.7 Hz, H6) 4.74 (ddd, 1 H, JH2-F 50.7 JH2-H3 9.0 JH2-H1 7.6 Hz, H2) 5.17 (dd, 1 H, JH4-H3 9.7 JH4-H5 10.0 Hz, H4) 5.38 (dd, 1 H, JH1-H2 7.6, JH1F2  2.4 Hz, H1) 5.42 (ddd, 1 H, JH3-F2 14.6 JH3-H2 9.0 JH3-H4 9.7 Hz, H3) 6.67 (d, 1 H, JH8(Ar)-H6(Ar)  1.8 Hz, H8(Ar)) 6.67 (dd, 1 H, JH6(Ar)-H5(Ar)10.2 JH6(Ar)-H8(Ar) 1.8 Hz, H6(Ar)) 7.36 (d, 1 H, JH5(Ar)-H6(Ar) 10.2 Hz, H5(Ar)) 7.75 (s, 1 H, H4(Ar)) 19F NMR (282 MHz, CDCl3) δ ppm -199.25 (ddd, JF2-H2 50.7 JF2-H3 14.6 JF2-H1 2.4 Hz, F2) 13C NMR (101 MHz, CDCl3) δ ppm 20.86 (C(O)CH3), 20.93 (C(O)CH3), 20.97 (C(O)CH3), 29.10 (CH3(Ar)), 29.16 (CH3(Ar)), 38.41 (C9(Ar)), 61.72 (C6), 68.33 (d, JC4-F2 6.9 Hz, C4) 72.56 (C5), 72.86 (d, JC3-F2 19.9 Hz, C3), 90.04 (d, JC2-F2 194.7 Hz, C2), 100.69 (d, JC1-F2 23.0 Hz, C1), 128.54, 129.19 (C8(Ar)), 130.48 (C6(Ar)), 132.67 (C4(Ar)), 132.91, 133.59, 141.05, 141.20 (C5(Ar)), 148.55, 149.81 (C2(Ar)), 153.50, 169.77 (C(O)CH3), 170.31 (C(O)CH3), 170.62 (C(O)CH3), 187.54 (C=O(Ar)).  9H-(1,3-Dichloro-9,9-dimethylacridin-2-one-7-yl) 3,4,6-tri-O-acetyl-2-deoxy-2-fluoro-β-Dglucopyranoside (3.14)  Compound 3.14 was crystallized from hot ether and PE to give red needles (25 mg, 0.043 mmol, 2.5%). 1H NMR (400 MHz, CDCl3) δ ppm 1.88 (s, 3 H, CH3(Ar)) 1.89 (s, 3 H, CH3(Ar)) 2.08 (s, 3 H, C(O)CH3) 2.09 (s, 3 H, C(O)CH3) 2.13 (s, 3 H, C(O)CH3) 3.97 (ddd, 1 H, JH5-H4 9.9 JH5-H6 5.4 JH5-H6’ 2.3 Hz, H5) 4.20 (dd, 1 H, JH6’-H6 12.3 JH6’-H5 2.3 Hz, H6’) 4.29 (dd, 1 H, JH6-H6’ 12.3 JH6-H5 5.4 Hz, H6) 4.63 (ddd, 1 H, JH2-F2 50.8 JH2-H3 8.7 JH2-H1 7.6 Hz, H2) 5.12 (dd, 1 H, 84  JH4-H3 9.6 JH4-H5 9.9 Hz, H4) 5.28 (dd, 1 H, JH1-H2 7.6 JH1-F2 2.9 Hz, H1) 5.46 (ddd, 1 H, JH3-F2 14.5 JH3-H4 9.6 JH3-H2 8.7 Hz, H3) 7.03 (dd, 1 H, JH6(Ar)-H5(Ar) 8.7 JH6(Ar)-H8(Ar) 2.6 Hz, H6(Ar)) 7.17 (d, 1 H, JH8(Ar)-H6(Ar) 2.6 Hz, H8(Ar)) 7.63 (s, 1 H, H4(Ar)) 7.63 (d, 1 H, JH5(Ar)-H6(Ar) 8.7 Hz, H5(Ar)) 19  F NMR (282 MHz, CDCl3) δ ppm -199.57 (ddd, JF2-H2 50.8 JF2-H3 14.5 JF2-H1 2.9 Hz, F2) 13C  NMR (101 MHz, CDCl3) δ ppm 20.54 (C(O)CH3) 20.63 (C(O)CH3) 20.67 (C(O)CH3), 26.72 (2 × CH3(Ar)), 39.08 (C9(Ar)), 61.86 (C6), 67.90 (d, J42-F2 6.9 Hz, C4) 72.39 (d, JC3-F2 19.9 Hz, C3) 72.38 (C5), 88.88 (d, JC2-F2 192.4 Hz, C2), 97.87 (d, JC1-F2 23.7 Hz, C1), 115.39 (C6(Ar)) 115.72 (C8(Ar)) 133.66 (C5(Ar)) 135.15, 137.12, 137.15, 139.38 (C4(Ar)) 140.32, 148.80, 159.13 (C7(Ar)) 169.47 (C(O)CH3) 169.90 (C(O)CH3) 170.37 (C(O)CH3) 173.13 (C=O(Ar)).  9H-(6,8-Dichloro-9,9-dimethylacridin-7-one-2-yl) 2-deoxy-2-fluoro-β-D-glucopyranoside (DDAOY-2FGlc) (3.15)  Compound 3.13 (53 mg, 0.089 mmol) was dissolved in dry MeOH (20 ml) and deacetylated using NaOMe (20 mM) according to the procedure for General Deacetylation. The reaction was completed in 20 min as indicated by TLC, as well as by a colour change of the solution from yellow to green. After neutralization, the product was dry-loaded and flash chromatographed to remove hydrolysed byproduct (HPLC-grade EtOAc/HPLC-grade hexanes, 60%-80%). A yield of 91% (38 mg, 0.081 mmol) of product 3.15 was isolated. HRMS mass calculated for C21H20Cl2FNO6: 471.06, found: 494.0562 [M+Na]+. 1H NMR (400 MHz, CD3OD) δ ppm 1.83 (s, 6 H, 2 × CH3) 3.34 (m, H5) 3.44 (t, 1 H, JH4-H3=JH4-H5 9.3 Hz, H, H4) 3.65 (dd, 1 H, JH6-H6’ 12.0 JH6-H5 5.8 Hz, H6) 3.70 (ddd, 1 H, JH3-F2 16.0 JH3-H2 8.9 JH3-H4 9.3 Hz, H3) 3.80 (dd, 1 H, JH6’-H6 12.0 JH6’-H5 2.0 Hz, H6’) 4.40 (ddd, 1 H, JH2-F2 51.4 JH2-H3 8.9 JH2-H1 7.8 Hz, H2) 5.41 (dd, 1 H, JH1-H2 7.8 JH1-F2 2.1 Hz, H1) 6.67 (dd, 1 H, JH6(Ar)-H5(Ar) 9.9 JH6(Ar)-H8(Ar) 1.8 Hz, H6(Ar)) 6.79 (d, 1 H, JH8(Ar)-H6(Ar) 1.8 Hz, H8(Ar)) 7.43 (d, 1 H, JH5(Ar)-H6(Ar) 9.9 Hz, H5(Ar)) 7.77 (s, 1 H, H4(Ar)) 19F NMR (282 MHz, CD3OD) δ ppm -200.52 (ddd, JF2-H2 51.4 JF2-H3 16.0 JF2-H1 2.1 Hz, 93 F) 13C NMR (101 MHz, CD3OD) δ ppm 29.09 (CH3(Ar)) 29.20 (CH3(Ar)) 39.60 (C9(Ar)) 62.52 (C6) 71.19 (d, JC4-F2 7.7 Hz, C4) 76.35 (d, JC3-F2 17.6 Hz, C3) 78.94 (C5) 94.25 (d, JC2-F2 85  187.8 Hz, C2) 102.39 (d, JC1-F2 26.1 Hz, C1) 129.68, 129.99 (C8(Ar)) 131.60, 133.30 (C6(Ar)) 133.63 (C4(Ar)) 134.97, 141.86, 142.48 (C5(Ar)) 150.46, 151.74 (C2(Ar)) 154.43, 189.14 (C=O(Ar)).  9H-(1,3-Dichloro-9,9-dimethylacridin-2-one-7-yl) 2-deoxy-2-fluoro-β-D-glucopyranoside (DDAOR-2FGlc) (3.16)  Compound 3.14 (25 mg, 0.042 mmol) was dissolved in 20 ml of dry methanol and deacetylated with 20 mM NaOMe according to the procedure for General Deacetylation. The reaction was completed in 20 min as shown by TLC. The product was crystallized in dry MeOH and PE to yield pure compound 3.16 (18 mg, 0.038 mmol, 91%). HRMS mass calculated: 471.06, found: 470.0580 [M-H]-. 1H NMR (400 MHz, 30% CDCl3 70% CD3OD) δ ppm 1.89 (s, 3 H, CH3(Ar)) 1.90 (s, 3 H, CH3(Ar)) 3.44 (t, 1 H, JH4-H3=JH4-H5 9.7 Hz, H4) 3.59 (ddd, 1 H, JH5-H4 9.7 JH5-H6 6.1 JH5-H6’ 2.1 Hz, H5) 3.70 (dd, 1 H,JH6-H6’ 12.1 JH6-H5 6.1 Hz, H6) 3.77 (ddd, 1 H, JH3-F2 16.0 JH3-H2 8.8 JH3-H4 9.7 Hz, H3) 3.93 (dd, 1 H, JH6’-H6 12.1 JH6’-H5 2.1 Hz, H6’) 4.31 (ddd, 1 H, JH2-F2 51.7 JH2-H3 8.8 JH2-H1 7.6 Hz, H2) 5.34 (dd, 1 H, JH1-H2 7.6 JH1-F2 2.6 Hz, H1) 7.12 (dd, 1 H, JH6(Ar)-H5(Ar) 8.7 JH6(Ar)-H8(Ar) 2.5 Hz, H6(Ar)) 7.31 (d, 1 H, JH8(Ar)-H6(Ar) 2.5 Hz, H8(Ar)) 7.61 (d, 1 H, JH5(Ar)-H6(Ar) 8.7 Hz, H5(Ar)) 7.67 (s, 1 H, H4(Ar)) 19F NMR (282 MHz, 30% CDCl3 70% CD3OD) δ ppm -200.80 (ddd, JF2-H2 51.7 JF2-H3 16.0 JF2-H1 2.6 Hz, 1 F) 13C NMR (101 MHz, 30% CDCl3 70% CD3OD) δ ppm 26.98 (CH3(Ar)) 27.08 (CH3(Ar)) 40.55 (C9(Ar)) 62.28 (C6) 71.05 (d, JC4-F2 8.4 Hz, C4) 76.20 (d, JC3-F2 16.8 Hz, C3) 78.46 (H5) 93.09 (d, JC2-F2 192.5 Hz, C2) 98.97 (d, JC1-F2 23.7 Hz, C1) 116.11 (C8(Ar)) 117.25 (C6(Ar)) 134.83 (C5(Ar)) 135.73, 137.75, 137.85, 140.66 (C4(Ar)) 141.88, 142.51, 149.44, 161.73 (C7(Ar)) 174.48 (C=O(Ar)).  86  3,4,6-Tri-O-acetyl-D-galactal (4.4)  Galactose (20 g, 110 mmol, 4.1) was acetylated in Ac2O (220 ml) and sodium acetate (0.97 g), per General Procedure for Acetylation of Free Sugars, yielding crystalline per-O-acetylD-galactopyranose  (4.2, 18 g, 46 mmol, 42%). The dried product was dissolved in 50 ml of DCM,  HBr in AcOH (33% w/w, 20 ml) was added and reacted according to the General Procedure for Synthesis of α-Glycosyl Bromides to form 2,3,4,6-tetra-O-acetyl-α-D-galactosyl bromide (4.3) in quantitative yield. The crude product was dissolved in AcOH (30 ml) and added to a mixture of ice-cold glacial acetic acid (330 ml), water (190 ml) and zinc powder (150 g, 2.3 mol). Copper(II) in the form of CuSO4•5H2O (4.22 g, 26.4 mmol) was added to catalyse the reaction. The reaction mixture was allowed to warm to room temperature and left to react for 18 hours, after which it was filtered clean of solid particles, diluted with EtOAc (500 ml) and washed thoroughly with water (4 × 400 ml). The water layers were then extracted once with EtOAc (150 ml). The organic layers were combined, washed with saturated NaHCO3 until basic, brine (200 ml), dried with anhydrous MgSO4, filtered and the solvent was removed under reduced pressure. The crude 3,4,6-tri-O-acetyl-D-galactal (4.4, 6 g, 50%) was used in the next reaction without further purification. The proton spectrum agreed with those previously reported.87 1H NMR (400 MHz, CDCl3) δ ppm 2.00 (s, 3 H, C(O)CH3) 2.06 (s, 3 H, C(O)CH3) 2.09 (s, 3 H, C(O)CH3) 4.19-4.22 (m, 2 H, H6 H6’) 4.29 (m, 1 H, H5) 4.70 (m, 1 H, JH2-H1 6.2 Hz, H2) 5.40 (m, 1 H, H4) 5.52 (m, 1 H, H3) 6.43 (d, 1 H, JH1-H2 6.2 Hz, H1).  1,3,4,6-Tetra-O-acetyl-2-deoxy-2-fluoro-D-galactopyranose (4.7)  Compound 4.7 was made according to the General Procedure for Synthesis of 2Fsugars: the protected galactal 4.4 (2.3 g, 8.3 mmol) was dissolved in MeNO2 (45 ml) and water (5 ml). Selectfluor® (4.0 g, 11 mmol, 1.3 equivalents) was added and the reaction mixture stirred at room temperature for 2 hours, then refluxed for 1 hour to obtain the protected 2-deoxy-2-fluoro-  87  D-galacto-hemiacetal  (4.5) along with the 2-deoxy-2-fluoro-D-talo-epimer (4.6). The mixture  was concentrated by the removal of solvents under reduced pressure and then stirred overnight in Ac2O (30 ml) and pyridine (20 ml) to obtain the per-O-acetyl-2-deoxy-2-fluoro-D-galacto- and talo-pyranoses (4.7, 4.8, respectively). Flash chromatography (EtOAc/PE, 1:3) afforded mixture of the α- and β-anomers of compound 4.7 (2.0 g, 68%). 1H NMR (400 MHz, CDCl3) δ ppm 2.00-2.17 (m, 24 H, 8 × C(O)CH3) 4.00 - 4.17 (m, 5 H, H5α H5β H6β H6’α H6’β) 4.29 (t, 1 H, JH6α-H6’α=JH6-H5=6.5 Hz, H6α) 4.62 (dt, 1 H, JH2β-Fβ 51.4 JH2β-H1β 8.0 JH2β-H3β 8.9 Hz, H2β) 4.87 (ddd, 1 H, JH2α-Fα 48.9 JH2α-H3α 10.1 JH2α-H1α 3.9 Hz, H2α) 5.17 (ddd, 1 H, JH3β-Fβ 13.1 JH3β-H2β 8.9 JH3β-H4β 2.9 Hz, H3β) 5.38 (ddd, 1 H, JH3α-H2α 10.1 JH3α-Fα 10.7 JH3α-H4α 3.2 Hz, H3α) 5.43 (ddd, 1 H, JH4β-H3β 2.9 JH4β-H5β 3.2 Hz, H4β) 5.49 (t, 1 H, JH4α-H3α=JH4α-H5α=3.2 Hz, H4α) 5.78 (dd, 1 H, JH1β-H2β 8.0 JH1β-Fβ 3.9 Hz, H1β) 6.44 (d, 1 H, JH1α-H2α 3.9 Hz, H1α).  9H-(1,3-Dichloro-9,9-dimethylacridin-7-one-2-yl) 3,4,6-tri-O-acetyl-2-deoxy-2-fluoro-β-Dgalactopyranoside (4.13)  The α-galactosyl bromide was made according to the General Procedure for Synthesis of α-Glycosyl Bromides: 4.7 (430 mg, 1.23 mmol) was dissolved in cold DCM (5 ml) under Ar(g) and HBr in AcOH (33% w/w, 0.5 ml, 2.9 mmol, 2.4 eq) was added drop-wise. After stirring at room temperature for 1 hour, the reaction mixture was worked up and the crude galactosyl bromide (4.9) dissolved in 4 ml dry MeCN and coupled with DDAO (375 mg, 1.21 mmol, 0.98 eq) as per the General Procedure for the Coupling of DDAO and Per-O-acetyl-2Fglycosyl Bromides. 2,6-Lutidine (0.5 ml, 4.3 mmol), CaSO4 (200 mg) and Ag2O (200 mg, 870 mmol) were added to the reaction vessel and the reaction was stirred at room temperature for 12 hours in the dark. After workup, flash chromatography (EtOAc/PE, 3:7 to 1:1) and subsequent crystallization from hot acetone, ether and PE afforded pure 4.13 (130 mg, 0.22 mmol, 18%) as yellow needles. HRMS mass calculated for C27H26Cl2FNO9 597.10; found 620.0859 [M+Na]+. 1  H NMR (400 MHz, CDCl3) δ ppm 1.80 (s, 3 H, CH3(Ar)) 1.82 (s, 3 H, CH3(Ar)) 1.98 (s, 3 H,  C(O)CH3) 2.10 (s, 3 H, C(O)CH3) 2.21 (s, 3 H, C(O)CH3) 3.94 (td, 1 H, JH5-H6=JH5-H6’ 7.0 JH5-H4 88  2.7 Hz, H5) 4.12 (m, 2 H, JH6-H6’=JH6’-H6 10.9, H6 H6’) 4.93 (ddd, 1 H, JH2-F2 51.4 JH2-H3 9.7 JH2H1  7.5 Hz, H2) 5.22 (ddd, 1 H, JH3-F2 13.1 JH3-H2 9.7 JH3-H4 3.5 Hz, H3) 5.36 (dd, 1 H, JH1-H2 7.5  JH1-F2 3.2 Hz, H1) 5.46 (dd, 1 H, JH4-H3 3.5 JH4-H5 2.7 Hz, H4) 6.69 (d, 1 H, JH8(Ar)-H6(Ar) 1.8 Hz, H8(Ar)) 6.69 (dd, 1 H, JH6(Ar)-H5(Ar) 9.9 JH6(Ar)-H8(Ar) 1.8 Hz, H6(Ar)) 7.38 (d, 1 H, JH5(Ar)-H6(Ar) 9.9 Hz, H5(Ar)) 7.77 (s, 1 H, H4(Ar)) 19F NMR (282 MHz, CDCl3) δ ppm -206.63 (ddd, JF2-H2 51.4 JF2-H3 13.1 JF2-H1 3.2 Hz, 1 F) 13C NMR (101 MHz, CDCl3) δ ppm 20.57 (3 C, 3 × C(O)CH3) 28.79 (CH3(Ar)) 28.86 (CH3(Ar)) 38.09 (C9(Ar)) 60.58 (C6) 67.27 (d, JC4-F2 8.4 Hz, C4) 70.80 (d, JC3-F2 18.3 Hz, C3) 71.28 (C5) 88.14 (d, JC2-F2 190.4 Hz, C2) 100.94 (d, JC1-F2 22.9 Hz, C1) 128.31, 128.88 (C8(Ar)) 130.14, 132.35 (C5(Ar)) 132.62 (C6(Ar)) 133.24, 140.70, 140.90 (C4(Ar)) 148.25, 149.67 (C2(Ar)) 153.16, 169.91 (C(O)CH3) 170.02 (C(O)CH3) 170.22 (C(O)CH3) 187.26 (C=O(Ar)).  9H-(1,3-Dichloro-9,9-dimethylacridin-7-one-2-yl) 2-deoxy-2-fluoro-β-D-galactopyranoside (DDAOY-2FGal) (4.15)  Removal of acetyl protecting groups was done according to the procedure for General Deacetylation: 4.13 (93 mg, 0.15 mmol) was deprotected using NaOMe (25 mM) in dry MeOH (12 ml). After flash chromatography on silica gel (EtOAc/PE, 4:1), 4.15 (61 mg, 0.13 mmol, 84%) was obtained. HRMS mass calculated for C21H20Cl2FNO6: 471.07; found: 494.0547 [M+Na]+. 1H NMR (400 MHz, 30% CDCl3 70% CD3OD) δ ppm 1.82 (s, 3 H, CH3(Ar)) 1.82 (s, 3 H, CH3(Ar)) 3.55 (m, 1 H, JH5-H6 6.7 JH5-H6’ 5.9, H5) 3.69 (dd, 1 H, JH6’-H6 11.1 JH6’-H5 5.9 Hz, H6’) 3.77 (dd, 1 H, JH6-H6’ 11.1 JH6’-H5 6.7 Hz, H6) 3.83 (ddd, 1 H, JH3-F2 14.3 JH3-H2 9.2 JH3-H4 3.2 Hz, H3) 3.98 (m, 1 H, JH4-H3 3.2 Hz, H4) 4.77 (ddd, 1 H, JH2-F2 52.2 JH2-H3 9.2 JH2-H1 7.4 Hz, H2) 5.30 (dd, 1 H, JH1-H2 7.4 JH1-F2 3.0 Hz, H1) 6.67 (dd, 1 H, JH6(Ar)-H5(Ar) 9.7 JH6(Ar)-H8(Ar) 1.9 Hz, H6(Ar)) 6.73 (d, 1 H, JH8(Ar)-H6(Ar) 1.9 Hz, H8(Ar)) 7.41 (d, 1 H, JH5(Ar)-H6(Ar) 9.7 Hz, H5(Ar)) 7.74 (s, 1 H, H4(Ar)) 19F NMR (282 MHz, 30% CDCl3 70% CD3OD) δ ppm -207.88 (ddd, JF2-H2 52.2 JF2-H3 14.3 JF2-H1 3.0 Hz, F2) 13C NMR (101 MHz, 30% CDCl3 70% CD3OD) δ ppm 29.13 (CH3(Ar)) 29.19 (CH3(Ar)) 39.14 (C9(Ar)) 61.30 (C6) 70.00 (d, JC4-F2 8.4 Hz, C4) 72.85 (d, JC3-F2 17.6 Hz, C3) 89  76.84 (C5) 92.93 (d, JC2-F2 179.4 Hz, C2) 102.58 (d, JC1-F2 27.6 Hz, C1) 129.30 (C8(Ar)) 129.53, 131.26, 132.91 (C6(Ar)) 133.36 (C4(Ar)) 134.3, 141.20, 142.07 (C5(Ar)) 150.02, 151.55 (C2(Ar)) 153.75, 188.74 (C=O(Ar)).  1,3,4-Tri-O-acetyl-2-deoxy-2-fluoro-D-xylopyranose (5.7)  This synthetic method is essentially the method used to make compound 4.7. For further details, please read below. Xylose (20 g, 130 mmol, 5.1) was acetylated in Ac2O (250 ml) and sodium acetate (1.05 g), per General Procedure for Acetylation of Free Sugars, yielding crystalline per-O-acetyl-Dxylopyranose (5.2, 20 g 63 mmol, 50%). The dried product was dissolved in DCM (50 ml) and HBr in AcOH (33% w/w, 20 ml) was added and reacted according to the General Procedure for Synthesis of α-Glycosyl Bromides to form 2,3,4-tri-O-acetyl-α-D-xylosyl bromide (5.3) in quantitative yield. The crude 5.3 was dissolved in AcOH (30 ml) and added to a mixture of icecold glacial acetic acid (330 ml), water (190 ml) and zinc powder (150 g, 2.3 mol). Copper(II) in the form of CuSO4•5H2O (4.2 g, 26 mmol) was added as a catalyst. The reaction mixture was allowed to warm to room temperature and left to react for 18 hours, after which it was filtered clean of solid particles, diluted with EtOAc (500 ml) and washed thoroughly with water (4 × 400 ml). The water layers were then extracted once with EtOAc (150 ml). The organic layers were combined, washed with saturated NaHCO3 until basic, brine (200 ml), dried with anhydrous MgSO4, filtered and the solvent was removed under reduced pressure. The crude 3,4-di-Oacetyl-D-xylal (5.4, 7.6 g, 38 mmol, 60%) was used in the next reaction without further purification. Compound 5.7 was made according to the General Procedure for Synthesis of 2Fsugars: 5.4 (2.3 g, 10 mmol) was dissolved in MeNO2 (45 ml) and H2O (5 ml). Selectfluor® (5.4 g, 15 mmol, 1.5 equivalents) was added and the reaction mixture stirred at room temperature for 12 hours, then refluxed for 1 hour to obtain the protected 2-deoxy-2-fluoro-D-xylo-hemiacetal (5.5), along with the 2-deoxy-2-fluoro-D-lyxo-epimer (5.6). The mixture was concentrated by removal of solvents under reduced pressure, then stirred overnight in Ac2O (50 ml) and pyridine (20 ml) to obtain the per-O-acetyl-2-deoxy-2-fluoro-D-xylo- and lyxo-pyranoses (5.7, 5.8 respectively). Flash chromatography (EtOAc/PE, 1:3) afforded a mixture of α- and β-anomers of 90  compound 5.7 (2.0 g, 7.2 mmol, 72%). 1H NMR (400 MHz, CDCl3) δ ppm 2.04-2.19 (m, 18 H, 6 × C(O)CH3) 3.52 (dd, 1 H, JH5β-H5’β 11.5 JH5β-H4β 8.7 Hz, H5β) 3.67 (t, 1 H, JH5α-H4α=JH5α-H5’α 10.5 Hz, H5α) 3.89 (ddd, 1 H, JH5’β-H5β 11.5 JH5’β-H4β 5.3 Hz, H5’β) 4.10 (ddd, 1 H, JH5’α-H5α 10.5 JH5’α-H4α 4.7 Hz, H5’α) 4.40 (ddd, 1 H, JH2β-F2β 49.9 JH2β-H3β 8.3 JH2β-H1β 7.0 Hz, H2β) 4.58 (ddd, 1 H, JH2α-F2α 48.5 JH2α-H3α 9.8 JH2α-H1α 3.8 Hz, H2α) 4.94 (ddd, 1 H, JH4β-H3β 8.3 JH4β-H5β 8.7 JH4β-H5’β 5.3, H4β) 4.97 (ddd, 1 H, JH4α-H3α 9.8 JH4α-H5α 10.5 JH4α-H5’α 4.7, H4α) 5.32 (dt, 1 H, JH3β-F2β 13.4 JH3β-H2β=JH3β-H4β 8.3 Hz, H3β) 5.54 (dt, 1 H, JH3α-F2α 11.9 JH3α-H2α=JH3α-H4α 9.8 Hz, H3α) 5.78 (dd, 1 H, JH1β-H2β 7.0 JH1β-F2β 5.1 Hz, H1β) 6.33 (d, 1 H, JH1α-H2α 3.8 Hz, H1α) 19F NMR (282 MHz, CDCl3) δ ppm -202.65 (dd, JF2α-H2α 48.5 JF2α-H3α 11.9 Hz, F2α) -199.86 (ddd, JF2β-H2β 49.9 JF2βH3β  13.4 JF2β-H1β 5.1 Hz, F2β).  9H-(1,3-Dichloro-9,9-dimethylacridin-7-one-2-yl) 3,4-di-O-acetyl-2-deoxy-2-fluoro-β-Dxylopyranoside (5.13)  This synthetic method is essentially the method used to make compound 4.13. For further details, please read below. The α-xylosyl bromide (5.9) was made according to the General Procedure for Synthesis of α-Glycosyl Bromides. Compound 5.7 (430 mg, 1.61 mmol) was dissolved in cold DCM (5 ml) under Ar(g) and reacted with of HBr in AcOH (33% w/w, 1.5 ml, 8.8 mmol) at room temperature for 1 hour. The xylosyl bromide (5.9) was dissolved in dry MeCN (15 ml) and coupled with DDAO (410 mg, 1.33 mmol) as per the General Procedure for the Coupling of DDAO and PerO-acetyl-2Fglycosyl Bromides in the presence of 2,6-lutidine (0.5 ml, 4.3 mmol), CaSO4 (200 mg) and Ag2O (60 mg, 230 mmol). The reaction was stirred at room temperature overnight in the dark. After workup, flash chromatography (EtOAc/PE, 3:7 to 1:1) and subsequent crystallization from warm EtOAc and PE afforded pure 5.13 (211 mg, 0.40 mmol, 31%) as yellow powder. HRMS mass calculated for C24H23Cl2FNO7: 525.07; found: 548.0654 [M+Na]+. 1H NMR (400 MHz, CDCl3) δ ppm 1.84 (s, 6 H, 2 × CH3) 2.13 (s, 3 H, C(O)CH3) 2.20 (s, 3 H, C(O)CH3) 3.62 (dd, 1 H, JH5ax-H5eq 12.56 JH5ax-H4 6.02 Hz, H5ax) 4.47 (dd, 1 H, JH5eq-H5ax 12.56 JH5eq-H4eq 4.19 Hz, 91  H5eq) 4.89 (ddd, 1 H, JH2-F2 47.97 JH2-H3 7.31 JH2-H1 5.03 Hz, H2) 5.07 (td, 1 H, JH4-H3=JH4-H5ax 6.28, JH4-H5eq 4.34 Hz, H4) 5.40 (dt, JH3-F2 12.79 JH3-H2=JH3-H4 7.00 Hz, 1 H) 5.55 (dd, 1 H, JH1-F2 8.15 JH1-H2 4.95 Hz, H1) 6.71 (d, 1 H, JH8(Ar)-H6(Ar) 1.68 Hz, H8(Ar)) 6.71 (dd, 1 H, JH6(Ar)-H5(Ar) 10.51 JH6(Ar)-H8(Ar) 1.83 Hz, H6(Ar)) 7.40 (d, 1 H, JH5(Ar)-H6(Ar) 10.20 Hz, H5(Ar)) 7.81 (s, 1 H, H4(Ar)) 19  F NMR (282 MHz, CDCl3) δ ppm –196.96 (ddd, JF2-H2 48.43 JF2-H3 13.82 JF2-H1 7.54 Hz, F2)  13  C NMR (101 MHz, CDCl3) δ ppm 20.92 (CH3 (Ar)) 28.92 (C(O)CH3) 29.04 (C(O)CH3) 38.25  (C9(Ar)) 62.62 (C5) 68.46 (d, JC4-F2 4.60 Hz, C4) 69.74 (d, JC3-F2 24.54 Hz, C3) 87.86 (d, JC2-F2 183.26 Hz, C2) 101.06 (d, JC1-F2 29.14 Hz, C1) 128.30, 129.00 (C8(Ar)) 130.00, 132.47 (C6(Ar)) 132.85 (C4(Ar)) 133.51, 140.75, 141.04 (C5(Ar)) 148.37, 150.48 (C2(Ar)) 153.23, 169.83 (C(O)CH3) 170.19 (C(O)CH3) 187.39 (C=O(Ar)).  9H-(1,3-Dichloro-9,9-dimethylacridin-7-one-2-yl) 2-deoxy-2-fluoro-β-D-xylopyranoside (DDAOY-2FXyl) (5.15)  Removal of acetyl protecting groups was done according to the procedure for General Deacetylation: compound 5.13 (88 mg, 0.17 mmol) was deprotected using 20 mM NaOMe in dry MeOH (20 ml) for 15 min. After neutralization, methanol was evaporated under reduced pressure and flushed through a column (EtOAc/PE, 4:1) to give 5.15 in quantitative yield (78 mg). HRMS mass calculated for C20H18Cl2FNO5: 441.05; found: 476.0237 [M+Cl]-. 1H NMR (400 MHz, CDCl3) δ ppm 1.81 (s, 3 H, CH3(Ar)) 1.81 (s, 3 H, CH3(Ar)) 2.48 (br. s., 1 H, OH) 2.75 (br. s., 1 H, OH) 3.43 (dd, 1 H, JH5ax-H5eq 11.9 JH5ax-H4 7.5 Hz, H5ax) 3.88 (td, 1 H, JH4-H3= JH4H5ax7.5  JH4-H5eq 4.6 Hz, H4) 3.94 (dt, 1 H, JH3-F2 13.5 JH3-H2=JH3-H4 7.5 Hz, H3) 4.28 (dd, 1 H,  JH5eq-H5ax 11.9 JH5eq-H4 4.6 Hz, H5eq) 4.71 (ddd, 1 H, JH2-F2 49.6 JH2-H3 7.5 JH2-H1 6.1 Hz, H2) 5.42 (dd, 1 H, JH1-H2 6.1 JH1-F2 4.6 Hz, H1) 6.68 (d, 1 H, JH8(Ar)-H6(Ar) 1.7 Hz, H8(Ar)) 6.69 (dd, 1 H, JH6(Ar)-H5(Ar) 9.9 JH6(Ar)-H8(Ar) 1.7 Hz, H6(Ar)) 7.37 (d, 1 H, JH5(Ar)-H6(Ar) 9.9 Hz, H5(Ar)) 7.78 (s, 1 H, H4(Ar)) 19F NMR (282 MHz, CDCl3) δ ppm -200.66 (ddd, JF2-H2 49.6 JF2-H3 13.5 JF2-H1 4.6 Hz, 1 F) 13  C NMR (101 MHz, CDCl3) δ ppm 28.93 (CH3 (Ar)) 29.04 (CH3 (Ar)) 38.26 (C9(Ar)) 65.22 (C5)  69.12 (d, JC4-F2 5.3 Hz, C4) 73.69 (d, JC3-F2 19.1 Hz, C3) 91.05 (d, JC2-F2 186.3 Hz, C2) 101.32 (d, 92  JC1-F2 26.0 Hz, C1) 128.28, 129.00 (C8(Ar)) 130.09, 132.46 (C4(Ar)) 132.84, 133.46, 140.75, 141.07 (C5(Ar)) 148.42, 150.32 (C2(Ar)), 153.20, 187.44 (C=O(Ar)).  3,6,2’3’4’6’-Hexa-O-acetyl-D-cellobial (6.4)  This synthetic method is essentially the method used to make compound 4.4. For further details, please read below. Octa-O-acetyl-D-cellobiose (6.2, 20 g, 30 mmol) was dissolved in DCM (100 ml) and cooled to 0°C. HBr in AcOH (33% w/w, 25 ml) was added and reacted according to the General Procedure for Synthesis of α-Glycosyl Bromides to form 2,3,6,2’,3’,4’,6’-hepta-O-acetyl-α-Dcellobiosyl bromide (6.3) in quantitative yield. After workup, the crude product was dissolved in glacial AcOH (30 ml) and added to a mixture of ice-cold glacial AcOH (300 ml), water (150 ml) and zinc powder (150 g, 2.3 mol). Copper(II) in the form of CuSO4•5H2O (3.92 g, 24.5 mmol) was added to catalyse the reaction. The reaction mixture was allowed to warm to room temperature and left to react for 18 hours. The reaction mixture was filtered through Celite to remove solid zinc and its salts, diluted with EtOAc (600 ml) and washed thoroughly with water (6 × 500 ml). The water layers were then extracted once with EtOAc (200 ml). The organic layers were combined, washed with saturated NaHCO3 until basic, brine (200 ml) and dried with anhydrous MgSO4. After filtration and solvent evaporation under reduced pressure, compound 6.4 was crystallized using EtOAc, acetone and PE to obtain a white powder (12 g, 71%). 1H NMR (400 MHz, CDCl3) δ ppm 1.97-2.09 (s, 18 H, 6 × C(O)CH3) 3.67 (ddd, 1 H, JH5’-H4’ 9.6 JH5’-H6’a 4.4 JH5’-H6’b 2.1 Hz, H5’) 3.96 (dd, 1 H, JH4-H5 7.4 JH4-H3 5.4 Hz, H4) 4.04 (dd, 1 H, JH6’bH6’a  12.3 JH6’b-H5 2.1 Hz, H6’b) 4.11 (ddd, 1 H, JH5-H4 7.4 JH5-H6a 5.8 JH5-H6b 2.1 Hz, H5) 4.12 (dd,  1 H, JH6a-H6b 11.5 JH6a-H5 4.4 Hz, H6a) 4.28 (dd, 1 H, JH’a-H6’b 12.3 JH6’a-H5’ 4.49 Hz, H6’a) 4.42 (dd, 1 H, JH6b-H6a 11.5 JH6b-H5 2.1 Hz, H6b) 4.67 (d, 1 H, JH1’-H2’ 7.9 Hz, H1’) 4.79 (dd, 1 H, JH2-H1 6.1 JH2-H3 3.4 Hz, H2) 4.95 (dd, 1 H, JH2’-H3’ 9.6 JH2’-H1’ 7.9 Hz, H2’) 5.06 (t, 1 H, JH4’-H3’=JH4’-H5’ 9.6 Hz, H4’) 5.16 (t, 1 H, JH3’-H2’=JH3’-H1’ 9.6 Hz, H3’) 5.39 (dd, 1 H, JH3-H4 5.4 JH3-H2 3.4 Hz, H3) 6.38 (d, 1 H, JH1-H2 6.1 Hz, H1). 93  1,3,6,2’,3’,4’,6’-Hepta-O-acetyl-2-deoxy-2-fluoro-α-D-cellobiose (6.7)  Compound 6.7 was made according to the General Procedure for Synthesis of 2Fsugars: the protected cellobial (6.4, 6.7 g, 12 mmol) and Selectfluor® (6.0 g, 18 mmol, 1.5 equivalents) was ground using mortar and pestle and dissolved in MeNO2 (100 ml) and water (20 ml). The reaction was warmed to ambient temperature and continued to stir for 72 h to obtain the protected 2-deoxy-2-fluoro cellobio-hemiacetal along with the β-D-glucopyranosyl-(14)-2deoxy-2-fluoro-β-D-manno-hemiacetal. After workup, a portion of the mixture (4 g, 7.2 mmol) was per-O-acetylated by overnight stirring in Ac2O (120 ml) and pyridine (25 ml). Selective crystallization and subsequent recrystallization from CHCl3 and diethyl ether gave an enriched mixture of 1,3,6,2’,3’,4’,6’-hepta-O-acetyl-2-deoxy-2-fluoro-α-D-cellobiose (6.7, 250 mg, 0.36 mmol, 12% yield, 80% purity). The other major component was its β-anomer (17%). The 1H and 19  F NMR peaks agree with those previously reported.57 1H NMR (300 MHz, CDCl3) δ ppm 2.06  (s, 3 H, C(O)CH3) 2.01 (s, 3 H, C(O)CH3) 2.04 (s, 3 H, C(O)CH3) 2.22 (s, 3 H, C(O)CH3) 2.14 (s, 3 H, C(O)CH3) 2.12 (s, 3 H, C(O)CH3) 3.69-5.54 (m, 13 H, H2-H6 and H1’-H6’) 6.37 (d, 1 H, JH1-H2 3.8 Hz, H1) 19F NMR (282 MHz, CDCl3) δ ppm -203.93 (ddd, JF2-H2 49.3 JF2-H3 28.6 JF2-H1 6.8 Hz, 3%, F2ax-β-impurity) -202.16 (dd, JF2-H2 48.7 JF2-H3 12.0 Hz, 80%, F2eq-α) -200.73 (ddd, JF2-H2 51.0 JF2-H3 14.3 JF2-H1 2.8 Hz, 17%, F2eq-β).  9H-(1,3-Dichloro-9,9-dimethylacridin-7-one-2-yl) 3,6,2’,3’,4’,6’-hexa-O-acetyl-2-deoxy-2fluoro-β-D-cellobioside (6.13)  The 2-deoxy-2-fluoro-D-cellobiosyl bromide (6.9) was made according to the General Procedure for Synthesis of α-Glycosyl Bromides: compound 6.7 (100 mg, 0.16 mmol) was dissolved in cold DCM (10 ml) under Ar(g), and HBr in AcOH (33% w/w, 3.0 ml, 15 mmol) was 94  added drop-wise. After stirring at room temperature for 6 hours, the reaction mixture was worked up to give the α- and β-anomers of the protected 2-deoxy-2-fluoro-D-cellobiosyl bromides. These were dissolved in 4 ml dry acetonitrile and coupled with 60 mg (0.19 mmol) of DDAO as per the General Procedure for the Coupling of DDAO and Per-O-acetyl-2Fglycosyl Bromides, along with 2,6-lutidine (0.5 ml, 4.3 mmol), 200 mg CaSO4 and 100 mg Ag2O (432 mmol). The reaction was protected from light and stirred at room temperature overnight. After workup, flash chromatography using 2:2:1 DCM:PhMe:EtOAc to give 12 mg (0.014 mmol, 8%) of pure cellobioside 6.13. HRMS: mass calculated for C39H42Cl2FNO17: 885.18, found: 908.1707, [M+Na]+. 1H NMR (400 MHz, CDCl3) δ ppm 1.79 (s, 3 H, CH3(Ar)) 1.80 (s, 3 H, CH3(Ar)) 1.99 (s, 3 H, C(O)CH3) 2.00 (s, 3 H, C(O)CH3) 2.02 (s, 3 H, C(O)CH3) 2.06 (s, 3 H, C(O)CH3) 2.11 (s, 3 H, C(O)CH3) 2.14 (s, 3 H, C(O)CH3) 3.61 (ddd, JH5-H4 9.7 JH5-H6a 4.8 JH5-H6b 2.1 Hz, 1 H, H5) 3.69 (ddd, JH5’-H4’ 9.9 JH5’-H6’a 4.2 JH5’-H6’b 2.0 Hz, 1 H, H5’) 3.88 (dd, JH4-H3 9.0 JH4-H5 9.7 Hz, 1 H, H4) 4.07 (dd, JH6a-H6b 12.1 JH6a-H5 4.8 Hz, 1 H, H6a) 4.08 (d, JH6’b-H6’a 12.4 JH6’b-H5’ 2.0 Hz, 1 H, H6’b) 4.39 (dd, JH6’a-H6’b 12.4 JH6’a-H5’ 4.2 Hz, 1 H, H6’a) 4.51 (dd, JH6b-H6a 12.1 JH6b-H5 2.1 Hz, 1 H, H6b) 4.55 (d, JH1’-H2’ 7.9 Hz, 1 H, H1’) 4.66 (ddd, JH2-F2 50.2 JH2-H3 9.0 JH2-H1 7.4 Hz, 1 H, H2) 4.94 (dd, JH2’-H3’ 9.2 JH2’-H1’ 7.9 Hz, 1 H, H2’) 5.09 (dd, JH4’-H3’ 9.2 JH4’-H5’ 9.9 Hz, 1 H, H4’) 5.16 (t, JH3’-H2’=JH3’-H4’ 9.2 Hz, 1 H, H3’) 5.35 (dd, JH1-H2 7.4 JH1-F2 2.2 Hz, 1 H, H1) 5.40 (dt, JH3-F2 14.7 JH3-H2=JH3-H4 9.0 Hz, 1 H, H3) 6.68 (d, JH8(Ar)-H6(Ar) 1.8 Hz, 1 H, H8(Ar)) 6.68 (dd, JH6(Ar)-H5(Ar) 10.0 JH6(Ar)-H8(Ar) 1.8 Hz, 1 H, H6(Ar)) 7.36 (d, JH5(Ar)-H6(Ar) 10.0 Hz, 1 H, H5(Ar)) 7.75 (s, 1 H, H4(Ar)) 19F NMR (282 MHz, CDCl3) δ ppm -198.87 (ddd, JF2-H2 50.2 JF2-H3 14.7 JF2-H1 2.2 Hz, F2) 13C NMR (101 MHz, CDCl3) δ ppm 20.47 (C(O)CH3) 20.51 (2 × C(O)CH3) 20.58 (C(O)CH3) 20.65 (C(O)CH3) 20.72 (C(O)CH3) 28.77 (CH3(Ar)) 28.82 (CH3(Ar)) 38.05 (C9(Ar)) 60.88 (C6) 61.58 (C6’) 67.73 (C4’) 71.53 (C2’) 72.05 (C5’) 71.99 (d, JC3-F2 19.1 Hz, C3) 72.79 (C3’) 73.18 (C5) 75.84 (d, JC4-F2 6.8 Hz, C4) 89.86 (d, JC2-F2 193.5 Hz, C2) 100.27 (d, JC1-F2 23.7 Hz, C1) 100.71 (C1’) 128.85 (C6(Ar)) 129.01, 130.07 (C8(Ar)) 132.34 (C4(Ar)) 132.56, 133.25, 140.69 (C6(Ar)) 140.88, 148.22, 149.47 (C2(Ar)) 153.15, 168.85 (C(O)CH3) 169.25 (C(O)CH3) 169.49 (C(O)CH3) 169.87 (C(O)CH3) 170.22 (C(O)CH3) 170.46 (C(O)CH3) 187.24 (C=O(Ar)).  95  9H-(1,3-Dichloro-9,9-dimethylacridin-7-one-2-yl) 2-deoxy-2-fluoro-β-D-cellobioside (DDAOY2FCel) (6.15)  Removal of acetyl protecting groups was done according to the procedure for General Deacetylation: 40 mg (0.46 mmol) of compound 6.13 was deprotected using 50 mM NaOMe in 20 ml dry MeOH for 3 hours. After neutralization, solvent methanol was evaporated under reduced pressure and the crude product crystallized in dry MeOH to give 15 mg of product 6.15 (0.024 mmol, 53%). HRMS: mass calculated for C27H30Cl2FNO11: 633.118; found: 656.1072 [M+Na]+. 1H NMR (400 MHz, CD3OD) δ ppm 1.84 (s, 3 H, CH3(Ar)) 1.85 (s, 3 H, CH3(Ar)) 3.37 3.94 (m, 11 H, H3-H6 and H2’-H6’) 4.47 (d, JH1’-H2’ 7.7 Hz, 1 H, H1’) 4.48 (ddd, JH2-F2 51.3 JH2H3  8.8 JH2-H1 7.6 Hz, 1 H, H2) 5.45 (dd, JH1-H2 7.6 JH1-F2 2.2 Hz, 1 H, H1) 6.69 (dd, JH6(Ar)-H5(Ar) 9.7  JH6(Ar)-H8(Ar) 1.8 Hz, 1 H, H6(Ar)) 6.81 (d, JH8(Ar)-H6(Ar) 1.8 Hz, 1 H, H8(Ar)) 7.45 (d, JH5(Ar)-H6(Ar) 9.7 Hz, 1 H, H5(Ar)) 7.80 (s, 1 H, H4(Ar)) 19F NMR (282 MHz, CD3OD) δ ppm -198.74 (ddd, JF2-H2 51.3 JF2-H3 13.19 JF2-H1 2.2 Hz, F2) 13C NMR (101 MHz, CD3OD) δ ppm 29.08 (CH3(Ar)) 29.19 (CH3(Ar)) 39.60 (C9(Ar)) 55.04, 61.69, 62.61, 71.55, 74.85 (d, JC4-F217.5 Hz, C4) 75.03, 77.38, 78.14 (d, JC3-F2 30.5 Hz, C3) 79.96, 93.86 (d, JC2-F2 188.9 Hz, C2) 102.33 (d, JC1-F2 22.9 Hz, C1) 104.66 (C1’) 129.65, 130.00 (C8(Ar)) 131.59, 133.31(C6(Ar)) 133.63 (C4(Ar)) 134.98, 141.92, 142.48 (C5(Ar)) 150.42, 151.70, 154.49, 189.12 (C=O(Ar)).  3.4  Crystallization  Per-O-acetylated DDAOY-2FGal (4.13) and per-O-acetylated DDAOR-2FGal (4.14) were crystallized using an ethanol-hexane solvent system (4.13) or ethanol-cyclohexane (4.14). Some water molecules were also present within the structure of 4.14 due to use of wet solvent.  96  3.5  Expression and Purification of Wild Type Abg  Plasmid DNA containing wild type Abg with N-terminal His6 tag was transformed into E. coli BL21(DE3) cells by electroporation. A single colony was grown overnight at 37°C in 10 ml of LB media containing 50 μg/ml of kanamycin; 5 ml of this culture was then used to inoculate 1 L of LB media containing 50 μg/ml of kanamycin and allowed to proliferate at 37°C. Protein expression was induced using 1 mM IPTG once cell density had reached OD600 of 0.6. The cells were grown overnight at 37°C. The cell culture was centrifuged for 20 min at 4000 rpm and resuspended in 45 ml of lysis buffer (20 mM Tris at pH 7.9, 0.5 M NaCl, and 0.02 M imidazole) containing an EDTAfree protease inhibitor. Cells were lysed by addition of 5 ml of BugBuster® (Merck) and shaking for 30 min at room temperature. The cell lysate was then centrifuged at 24000 g for 30 min. The supernatant was passed through a 1.0 mL HisTrapFF column (CE Healthcare) and unbound proteins were washed with lysis buffer containing 0.05 M imidazole. Abg was eluted using a gradient of 0.05-0.30 M imidazole by Fast Protein Liquid Chromatography (ÄKTA), monitored by UV/Vis at 280 nm. Fractions containing Abg were checked for purity by SDS-PAGE and concentrated using a 30K molecular weight cut-off centrifugal filter (Millipore). Concentration was determined using absorbance at 280 nm (ε0.1% = 2.18).65  3.6  Kinetic Analysis  Kinetic experiments were performed on a Cary 4000 or 300-Bio UV/Vis spectrophotometer equipped with a circulating water bath, unless otherwise specified. Extinction coefficients for all aglycones, except those for DDAO and DAO, were obtained from Kempton and Withers.65 All experiments were performed in buffer with DMSO content (v/v) not exceeding 5%. The assay conditions used for each enzyme are listed in Table 10.  97  Table 10: Reaction conditions used in enzymatic assays.  Enzyme  Conditions Used  Substrate (for indirect assays)  Buffer  pH  Temperature (°C)  Abg  PNP-Glc  50 mM sodium phosphate  6.8  37  GCase  DNP-Glc  50 mM citrate  6.5  30  LacZ  ONP-Gal  50 mM sodium phosphate, 1 mM MgCl2  6.8  37  β-Xyl  PNP-Xyl  50 mM sodium phosphate  6.8  37  EG I  --  50 mM citrate  5.0  30  EG II  --  50 mM citrate  5.0  30  CBH I  --  50 mM citrate  5.0  30  CBH II  --  50 mM citrate  5.0  30  CenD  --  50 mM sodium phosphate  6.8  37  Cex  --  50 mM sodium phosphate  6.8  37  3.6.1 Steady State Rate Determinations  Substrate was pre-incubated in buffer (50 mM sodium phosphate, pH 6.8) at 37°C for 5 min. The reaction was initiated by addition of Abg for a final concentration of 0.5 nM and followed via UV/Vis absorbance at 600 nm for DDAO or at 634 nm for DAO. Initial rates were plotted against substrate concentration and fit to a Michaelis-Menten equation on GraFit 5.0/7.0 to obtain Vmax (kcat) and Km.  98  3.6.2 Standard Procedure for Indirect Inactivation  Varying concentrations of the compound of interest were incubated with enzyme at the enzyme’s optimal temperature and buffer conditions. At appropriate time-points, aliquots of enzyme (15 μl) were taken out and diluted into a cuvette containing substrate at a concentration of 7-15 × Km that was pre-incubated in the appropriate buffer and temperature conditions, for a final volume of 200 μl. The inactivation reaction was stopped by dilution of inactivator as well as competition by high concentrations of substrate. The initial residual rate was monitored by UV/Vis and the values were plotted as a function of time. The resulting curve was fit to a pseudo-first order decay equation to obtain the rate constant of inactivation (kobs). The resulting kobs values were plotted against the corresponding inactivator concentration and fit to the equation below to obtain the kinetic parameters ki and Ki. 𝑘[  𝑘  ] [  ]  For reactions at inactivator concentrations well below the Ki value, a ki/Ki was obtained by fit to: 𝑘  𝑘[  ]  3.6.3 Inactivation in the Presence of a Competitive Inhibitor  A solution of PNP-Glc at concentrations of 0.056 mM, 1.35 mM or 3.70 mM was preincubated in 50 mM sodium phosphate buffer, pH 6.8 at 37°C. Abg (final concentration 1.2 nM) was added and immediately afterwards, the appropriate volume of inactivator (3.15 or 3.17) was added for a total volume of 300 μl. PNP release was monitored continuously by absorbance at 400 nm. The residual rates at various time-points were obtained by measuring the average change in absorbance at that time-point, over a short time-frame of approximately 15 seconds. These values were plotted as a function of time and fit to a pseudo-first order decay equation to obtain the rate constant of inactivation (kobs); these values were then plotted against inactivator concentration and fit to Equation 21 to obtain ki/Ki. 99  3.6.4 Direct Assays for Inactivation on a Standard UV/Vis Spectrophotometer  Inactivator was pre-incubated at the appropriate temperature and buffer. The reaction was initiated by addition of enzyme. The final concentrations of enzyme were as follows: 2.5 μM Abg, 2.1 μM Cex, 1 μM CenD; 11 μM EG I, 18 μM EG II, 13 μM CBH I, 15.5 μM CBH II. DDAO release was monitored continuously by absorbance at 600 nm. The inactivation parameters (ki, Ki) were obtained either by fitting the initial rate of inactivation, or the observed rate constant of inactivation (kobs) to a Michaelis-Menten type equation (Equation 24). For the latter, the kobs values were obtained by fitting the derivative of the absorbance vs. time graph to an exponential decay curve on GraFit 7.0.  3.6.5 Direct Assays for Inactivation by Stopped Flow  Experiments were performed on an Applied Photophysics SX20 stopped flow machine coupled with a water bath. Two syringes were filled with 800 μl of 0.64 μM Abg or 3.15 (2.58 μM to 137 μM) and incubated at 37°C for 5 minutes. DDAO release was monitored by absorbance at 600 nm, and the resulting curves were averaged and fitted with the accompanying software (Pro-Data SX Viewer) using an equation for a single exponential to obtain the observed rate constant of inactivation, kobs. Inactivation kinetic parameters ki and Ki were obtained by fitting a plot of kobs vs. inactivator concentration to Equation 24 (above).  3.6.6 Rates of Spontaneous Hydrolysis  The methodology used to determine the rates of spontaneous hydrolysis of 3.20, 3.21 and 3.22 originated from the paper by Cocker and Sinnott.88 Rates were determined in 25 mM sodium phosphate, 0.40 M KCl buffer at 50°C. The concentration of 3.20 used was 8 μM; the concentrations used for 3.21 and 3.22 ranged from 40 to 600 μM. Buffered solutions (1200 μl) were placed in quartz cuvettes and sealed with Teflon plugs and allowed to equilibrate for 5 min at 50°C before data collection. Aglycone release was followed by absorbance readings at 600 nm and 634 nm for DDAO and DAO, respectively, using extinction coefficient values of 32100 M-1 cm-1 and 23800 M-1 cm-1. For 3.20, rates were determined by fitting the derivative of the first100  order absorbance vs. time curve to the exponential decay equation  𝑒  , where k is the  hydrolysis rate constant. For compounds 3.21 and 3.22, initial (i.e. not exceeding 5% of total substrate concentration) rates of aglycone release (A/min) were monitored and converted to M/min using Beer’s Law. Following this, rates were plotted as a function of substrate concentration and analysed by linear regression. The slope of the curve corresponded to the hydrolysis rate constant.  3.7  pKa Determinations pKa determinations for DDAO were performed in buffer containing 50 mM citrate, 50  mM sodium phosphate and 50 mM Tris. pH values were adjusted using NaOH or HCl. Solutions (1000 μl) of 2 μM DDAO were allowed to incubate at 37°C in a Cary Eclipse standard fluorimeter with circulating water bath for 5 min. The fluorescence response at 600 V was recorded and plotted as a function of pH, and fit to a pH titration curve in GraFit 5.0. A rough estimation of the pKa of DAO was carried out under similar conditions. Absorbances at 410 nm (protonated DAO) and 634 nm (deprotonated DAO) were monitored on a Cary 300-Bio spectrophotometer and plotted as a function of pH and fit to a pH titration curve (GraFit 5.0). The pKa value was then taken to be the pH value at which the two curves intersected.  3.8  Extinction Coefficient Determinations  DDAO (11 - 15 mg) was weighed out and dissolved in 1.000 ml of DMSO. Dilution (100-fold) into DMSO, then another dilution (20-fold) into 50 mM sodium phosphate buffer (pH 6.80) gave a final concentration of 17 to 24 μM DDAO. Solutions of 10, 20, 50 and 100-fold dilutions were allowed to equilibrate at 37°C for 5 min before absorbance values at 600 nm were measured. Values were plotted as a function of DDAO concentration to give good linear correlation (R2 > 0.97 in all cases). The final extinction coefficient value was an average of 7 replicates. Extinction coefficients of DAO, 3.15, and 3.16 were also determined in this manner.  101  3.9  Active Site Titrations of Abg  Solutions of Abg in pH 6.8 sodium phosphate buffer were left to equilibrate at 37°C for 5 min. A baseline absorbance (600 nm) or fluorescence was measured. The fluorimeter conditions were as follows: excitation/emission wavelengths 600 nm/656 nm, 600 V, 10nm/10nm excitation/emission slit width. Absorbance at 700 nm was also measured as an internal control. Active site titrant (DDAOY-2FGlc, 3.15, 50 μl) was pipetted in for a final concentration of approximately 70 μM, and the change in absorbance or fluorescence response was monitored. The final change in absorbance was calculated as follows: −  −  −  The change in DDAO concentration was then calculated by Beer’s Law  𝑐 where  ε = 32100 M-1cm-1, b = 1 cm. DDAO release by fluorescence was quantified using a calibration curve (Figure 61).  Figure 61: Calibration curve of DDAO in the standard fluorimeter at excitation/emission wavelengths of 600 nm/656 nm.  102  References 1. 2.  3. 4. 5. 6. 7.  8. 9. 10.  11. 12.  13.  14.  15. 16. 17. 18.  Nevalainen, H.; Suominen, P.; Taimisto, K., On the safety of Trichoderma-reesei. J. Biotechnol. 1994, 37 (3), 193-200. Kubicek, C. P.; Messner, R.; Gruber, F.; Mach, R. L.; Kubicekpranz, E. M., The Trichoderma cellulase regulatory puzzle - From the interior life of a secretory fungus. Enzyme Microb. Technol. 1993, 15 (2), 90-99. Wolfenden, R.; Lu, X. D.; Young, G., Spontaneous hydrolysis of glycosides. J. Am. Chem. Soc. 1998, 120 (27), 6814-6815. Koshland, D. E., Stereochemistry and the mechanism of enzymatic reactions. Biol. Rev. Camb. Philos. 1953, 28 (4), 416-436. Withers, S. G.; Street, I. P.; Bird, P.; Dolphin, D. H., 2-Deoxy-2-fluoroglucosides - A novel class of mechanism-based glucosidase inhibitors. J. Am. Chem. Soc. 1987, 109 (24), 7530-7531. Phillips, D. C., Hen egg-white lysozyme molecule. Proc. Natl. Acad. Sci. U.S.A. 1967, 57 (3), 484-&. Sinnott, M. L.; Souchard, I. J., Mechanism of action of beta-galactosidase - effect of aglycone nature and alpha-deuterium substitution on hydrolysis of aryl galactosides. Biochem. J. 1973, 133 (1), 89-98. Vocadlo, D. J.; Davies, G. J.; Laine, R.; Withers, S. G., Catalysis by hen egg-white lysozyme proceeds via a covalent intermediate. Nature 2001, 412 (6849), 835-838. Buchini, S.; Buschiazzo, A.; Withers, S. G., A new generation of specific Tryponosoma cruzi trans-sialidase inhibitors. Angew. Chem. Int. Ed. 2008, 47 (14), 2700-2703. Damager, I.; Buchini, S.; Amaya, M. F.; Buschiazzo, A.; Alzari, P.; Frasch, A. C.; Watts, A.; Withers, S. G., Kinetic and mechanistic analysis of Trypanosoma cruzi trans-sialidase reveals a classical ping-pong mechanism with acid/base catalysis. Biochemistry 2008, 47 (11), 3507-3512. Vocadlo, D. J.; Davies, G. J., Mechanistic insights into glycosidase chemistry. Curr. Opin. Chem. Biol. 2008, 12 (5), 539-555. Vocadlo, D. J.; Withers, S. G., Detailed comparative analysis of the catalytic mechanisms of betaN-acetylglucosaminidases from families 3 and 20 of glycoside hydrolases. Biochemistry 2005, 44 (38), 12809-12818. Yip, V. L. Y.; Withers, S. G., Family 4 glycosidases carry out efficient hydrolysis of thioglycosides by an alpha,beta-elimination mechanism. Angew. Chem. Int. Ed. 2006, 45 (37), 6179-6182. Yip, V. L. Y.; Withers, S. G., Mechanistic analysis of the unusual redox-elimination sequence employed by Thermotoga maritima BgIT: A 6-phospho-beta-glucosidase from glycoside hydrolase family 4. Biochemistry 2006, 45 (2), 571-580. Jongkees, S. A. K.; Withers, S. G., Glycoside Cleavage by a New Mechanism in Unsaturated Glucuronyl Hydrolases. J. Am. Chem. Soc. 2011, 133 (48), 19334-19337. Itoh, T.; Hashimoto, W.; Mikami, B.; Murata, K., Crystal structure of unsaturated glucuronyl hydrolase complexed with substrate. J. Biol. Chem. 2006, 281 (40), 29807-29816. Michaelis, L.; Menten, M. L.; Johnson, K. A.; Goody, R. S., The original Michaelis constant: translation of the 1913 Michaelis-Menten paper. Biochemistry 2011, 50 (39), 8264-9. Michaelis, L.; Menten, M. L., The kinetics of the inversion effect. Biochem. Z. 1913, 49.  103  19. 20.  21.  22.  23.  24. 25. 26. 27.  28. 29. 30. 31.  32.  33. 34.  35.  Briggs, G. E.; Haldane, J. B. S., A note on the kinetics of enzyme action. Biochem. J. 1925, 19 (2). Phenix, C. P.; Rempel, B. P.; Colobong, K.; Doudet, D. J.; Adam, M. J.; Clarke, L. A.; Withers, S. G., Imaging of enzyme replacement therapy using PET. Proc. Natl. Acad. Sci. U.S.A. 2010, 107 (24), 10842-10847. Diaz, L.; Casas, J.; Bujons, J.; Llebaria, A.; Delgado, A., New glucocerebrosidase inhibitors by exploration of chemical diversity of N-substituted aminocyclitols using click chemistry and in situ screening. J. Med. Chem. 2011, 54 (7), 2069-2079. Truscheit, E.; Frommer, W.; Junge, B.; Muller, L.; Schmidt, D. D.; Wingender, W., Chemistry and biochemistry of microbial alpha-glucosidase inhibitors. Angew. Chem. Int. Ed. 1981, 20 (9), 744-761. Vonitzstein, M.; Wu, W. Y.; Kok, G. B.; Pegg, M. S.; Dyason, J. C.; Jin, B.; Phan, T. V.; Smythe, M. L.; White, H. F.; Oliver, S. W.; Colman, P. M.; Varghese, J. N.; Ryan, D. M.; Woods, J. M.; Bethell, R. C.; Hotham, V. J.; Cameron, J. M.; Penn, C. R., Rational design of potent sialidasebased inhibitors of influenza-virus replication. Nature 1993, 363 (6428), 418-423. Winchester, B. G., Iminosugars: from botanical curiosities to licensed drugs. Tetrahedron: Asymmetry 2009, 20 (6-8), 645-651. Reese, E. T.; Parrish, F. W.; Ettlinge.M, Nojirimycin and D-glucono-1,5-lactone as inhibitors of carbohydrases. Carbohydr. Res. 1971, 18 (3), 381-&. Jager, S.; Kiss, L., Investigation of the active site of the extracellular beta-D-glucosidase from Aspergillus carbonarius. World J. Microbiol. Biotechnol. 2005, 21 (3), 337-343. Kuhn, C. S.; Lehmann, J. C.; Sandhoff, K., Efficient photoaffinity-labeling of human betahexosaminidase-A - synthesis and application of 3-azi-1- (2-acetamido-2-deoxy-1-beta-dglucopyranosyl)thio butane and 3-azi-1- (2-acetamido-2-deoxy-1-beta-d-galactopyranosyl)thio butane. Bioconjugate Chem. 1992, 3 (3), 230-233. Gandy, M. N.; Debowski, A. W.; Stubbs, K. A., A general method for affinity-based proteomic profiling of exo-alpha-glycosidases. Chem. Commun. 2011, 47 (17), 5037-5039. Scriven, E. F. V.; Turnbull, K., Azides - their preparation and synthetic uses. Chem. Rev. 1988, 88 (2), 297-368. Tsai, C. S.; Li, Y. K.; Lo, L. C., Design and synthesis of activity probes for glycosidases. Org. Lett. 2002, 4 (21), 3607-3610. 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. Hekmat, O.; Florizone, C.; Kim, Y. W.; Eltis, L. D.; Warren, R. A. J.; Withers, S. G., Specificity fingerprinting of retaining beta-1,4-glycanases in the Cellulomonas fimi secretome using two fluorescent mechanism-based probes. ChemBioChem 2007, 8 (17), 2125-2132. Hartley, B. S.; Kilby, B. A., The reaction of para-nitrophenyl esters with chymotrypsin and insulin. Biochem. J. 1954, 56 (2), 288-297. Fujii, R.; Utsunomiya, Y.; Hiratake, J.; Sogabe, A.; Sakata, K., Highly sensitive active-site titration of lipase in microscale culture media using fluorescent organophosphorus ester. Biochim. Biophys. Acta, Mol. Cell. Biol. Lipids 2003, 1631 (2), 197-205. Fersht, A. R.; Ashford, J. S.; Bruton, C. J.; Jakes, R.; Koch, G. L. E.; Hartley, B. S., Active-site titration and aminoacyl adenylate binding stoichiometry of aminoacyl-transfer-RNA synthetases. Biochemistry 1975, 14 (1), 1-4.  104  36.  37. 38.  39. 40. 41. 42.  43.  44.  45.  46.  47.  48.  49.  50. 51.  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. Withers, S. G.; Rupitz, K.; Street, I. P., 2-Deoxy-2-fluoro-D-glycosyl fluorides - A new class of specific mechanism-based glycosidase inhibitors. J. Biol. Chem. 1988, 263 (17), 7929-7932. Namchuk, M. N.; Withers, S. G., Mechanism of Agrobacterium beta-glucosidase: Kinetic analysis of the role of noncovalent enzyme/substrate interactions. Biochemistry 1995, 34 (49), 16194-16202. Goddard, J. P.; Reymond, J. L., Enzyme assays for high-throughput screening. Curr. Opin. Biotechnol. 2004, 15 (4), 314-322. Licha, K., Contrast agents for optical imaging. Contrast Agents II 2002, 222, 1-29. Richard, J.-A.; Massonneau, M.; Renard, P.-Y.; Romieu, A., 7-hydroxycoumarin-hemicyanine hybrids: A new class of far-red emitting fluorogenic dyes. Org. Lett. 2008, 10 (19), 4175-4178. Komatsu, T.; Kikuchi, K.; Takakusa, H.; Hanaoka, K.; Ueno, T.; Kamiya, M.; Urano, Y.; Nagano, T., Design and synthesis of an enzyme activity-based labeling molecule with fluorescence spectral change. J. Am. Chem. Soc. 2006, 128 (50), 15946-15947. Kwan, D. H.; Chen, H. M.; Ratananikom, K.; Hancock, S. M.; Watanabe, Y.; Kongsaeree, P. T.; Samuels, A. L.; Withers, S. G., Self-immobilizing fluorogenic imaging agents of enzyme activity. Angew. Chem. Int. Ed. 2011, 50 (1), 300-303. Bueno, C.; Villegas, M. L.; Bertolotti, S. G.; Previtali, C. M.; Neumann, M. G.; Encinas, M. V., The excited-state interaction of resazurin and resorufin with amines in aqueous solutions. Photophysics and photochemical reaction. Photochem. Photobiol. 2002, 76 (4), 385-390. Dreier, J.; Breitmaier, E. B.; Gocke, E.; Apfel, C. M.; Page, M. G. P., Direct influence of S9 liver homogenate on fluorescence signals: impact on practical applications in a bacterial genotoxicity assay. Mutat. Res.–Genetic Toxicol. Environ. Mutagen. 2002, 513 (1-2), 169-182. Corey, P. F.; Trimmer, R. W.; Biddlecom, W. G., A new chromogenic beta-galactosidase substrate - 7-beta-D-galactopyranosyloxy-9,9-dimethyl-9H-acridin-2-one. Angew. Chem. Int. Ed. 1991, 30 (12), 1646-1648. Coleman, D. J.; Studler, M. J.; Naleway, J. J., A long-wavelength fluorescent substrate for continuous fluorometric determination of cellulase activity: Resorufin-beta-D-cellobioside. Anal. Biochem. 2007, 371 (2), 146-153. Coleman, D. J.; Kuntz, D. A.; Venkatesan, M.; Cook, G. M.; Williamson, S. P.; Rose, D. R.; Naleway, J. J., A long-wavelength fluorescent substrate for continuous fluorometric determination of alpha-mannosidase activity: Resorufin alpha-D-mannopyranoside. Anal. Biochem. 2010, 399 (1), 7-12. Ibatullin, F. M.; Banasiak, A.; Baumann, M. J.; Greffe, L.; Takahashi, J.; Mellerowicz, E. J.; Brumer, H., A real-time fluorogenic assay for the visualization of glycoside hydrolase activity in planta. Plant Physiol. 2009, 151 (4), 1741-1750. Hill, R.; Bedford, G. R.; Webster, B. R., Phenol dyestuff of Liebermann as an acridan derivative. J. Chem. Soc. C 1970, (18), 2462-&. Warther, D.; Bolze, F.; Leonard, J.; Gug, S.; Specht, A.; Puliti, D.; Sun, X.-H.; Kessler, P.; Lutz, Y.; Vonesch, J.-L.; Winsor, B.; Nicoud, J.-F.; Goeldner, M., Live-Cell One- and Two-Photon Uncaging of a Far-Red Emitting Acridinone Fluorophore. J. Am. Chem. Soc. 2010, 132 (8), 25852590.  105  52.  53.  54.  55. 56.  57.  58. 59. 60. 61. 62. 63. 64.  65. 66. 67. 68. 69.  Leira, F.; Vieites, J. M.; Vieytes, M. R.; Botana, L. M., Characterization of 9H-(1,3-dichlor-9,9dimethylacridin-2-ona-7-yl)-phosphate (DDAO) as substrate of PP-2A in a fluorimetric microplate assay for diarrhetic shellfish toxins (DSP). Toxicon 2000, 38 (12), 1833-1844. Tung, C. H.; Zeng, Q.; Shah, K.; Kim, D. E.; Schellingerhout, D.; Weissleder, R., In vivo imaging of beta-galactosidase activity using far red fluorescent switch. Cancer Res. 2004, 64 (5), 15791583. Burkart, M. D.; Zhang, Z. Y.; Hung, S. C.; Wong, C. H., A new method for the synthesis of fluoro-carbohydrates and glycosides using Selectfluor. J. Am. Chem. Soc. 1997, 119 (49), 1174311746. Albert, M.; Dax, K.; Ortner, J., A novel direct route to 2-deoxy-2-fluoro-aldoses and their corresponding derivatives. Tetrahedron 1998, 54 (19), 4839-4848. Ortner, J.; Albert, M.; Weber, H.; Dax, K., Studies on the reaction of D-glucal and its derivatives with 1-chloromethyl-4-fluoro-1,4-diazoniabicyclo 2.2.2 octane salts. J. Carbohydr. Chem. 1999, 18 (3), 297-316. McCarter, J. D.; Adam, M. J.; Braun, C.; Namchuk, M.; Tull, D.; Withers, S. G., Syntheses of 2deoxy-2-fluoro monosaccharide and oligosaccharide glycosides from glycals and evaluation as glycosidase inhibitors. Carbohydr. Res. 1993, 249 (1), 77-90. Adamson, J.; Marcus, D. M., 2-Deoxy-2-fluoro-D-galactose. Carbohydr. Res. 1970, 13 (2), 314&. Adamson, J.; Marcus, D. M., Fluorinated carbohydrates .13. 2-Deoxy-2-fluoro-D-galactose. Carbohydr. Res. 1972, 22 (2), 257-&. Karplus, M., Vicinal proton coupling in nuclear magnetic resonance. J. Am. Chem. Soc. 1963, 85 (18), 2870-+. Michalik, M.; Hein, M.; Frank, M., NMR spectra of fluorinated carbohydrates. Carbohydrate Research 2000, 327 (1-2). Clarke, K. a. R., K., A kinetic study of the effect of substituents on the rate of formation of alkylpyridinium halides in nitromethane solution. J. Am. Chem. Soc. 1960, 1885-1895. Briggs, A. J.; Glenn, R.; Jones, P. G.; Kirby, A. J.; Ramaswamy, P., Bond length and reactivity Stereoelectronic effects on bonding in acetals and glucosides. J. Am. Chem. Soc. 1984, 106 (21). Jones, P. G.; Kirby, A. J., Simple correlation between bond length and reactivity - combined use of crystallographic and kinetic data to explore a reaction coordinate. J. Am. Chem. Soc. 1984, 106 (21). Kempton, J. B.; Withers, S. G., Mechanism of Agrobacterium beta-glucosidase - Kinetic-studies. Biochemistry 1992, 31 (41), 9961-9969. McCarter, J. D.; Withers, S. G., 5-fluoro glycosides: A new class of mechanism-based inhibitors of both alpha- and beta-glucosidases. J. Am. Chem. Soc. 1996, 118 (1), 241-242. Rempel, B. P.; Withers, S. G., Non-stick sugars: synthesis of difluorosugar fluorides as potential glycosidase inactivators. Australian Journal of Chemistry 2009, 62 (6), 590-599. Knier, B. L.; Jencks, W. P., Mechanism of reactions of N-(methoxymethyl)-N,Ndimethylanilinium ions with nucleophilic-reagents. J. Am. Chem. Soc. 1980, 102 (22). Dale, M. P.; Kopfler, W. P.; Chait, I.; Byers, L. D., Beta-glucosidase - Substrate, solvent, and viscosity variation as probes of the rate-limiting steps. Biochemistry 1986, 25 (9).  106  70.  71. 72. 73. 74. 75. 76.  77. 78.  79.  80. 81. 82.  83. 84.  85. 86.  87.  Namchuk, M. N. Substituted aryl glycosides as probes of the mechanism of spontaneous and enzyme-catalyzed glycoside hydrolysis. Diss. The University of British Columbia, Vancouver, 1993. Bernasconi, C. F., The principle of nonperfect synchronization - more than a qualitative concept. Acc. Chem. Res. 1992, 25 (1). Erdelyi, M., Halogen bonding in solution. Chemical Society Reviews 2012, 41 (9). Mukherjee, A.; Desiraju, G. R., Halogen Bonding and Structural Modularity in 2,3,4-and 3,4,5Trichlorophenol. Crystal Growth & Design 2011, 11 (9). Cavallo, G.; Metrangolo, P.; Pilati, T.; Resnati, G.; Sansotera, M.; Terraneo, G., Halogen bonding: a general route in anion recognition and coordination. Chemical Society Reviews 2010, 39 (10). Grabowski, G. A., Lysosomal storage disease 1 - Phenotype, diagnosis, and treatment of Gaucher's disease. Lancet 2008, 372 (9645). 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. Rempel, B. P. Synthesis and Enzymatic Evaluation of Activated Fluorosugars as Inactivators of Lysosomal Enzymes. Diss. The University of British Columbia, Vancouver, Canada, 2009. 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. Ben-David, A.; Bravman, T.; Balazs, Y. S.; Czjzek, M.; Schomburg, D.; Shoham, G.; Shoham, Y., Glycosynthase activity of Geobacillus stearothermophilus GH52 beta-xylosidase: Efficient synthesis of xylooligosaccharides from alpha-D-xylopyranosyl fluoride through a conjugated reaction. ChemBioChem 2007, 8 (17). Goddard-Borger, E. D.; Fiege, B.; Kwan, E. M.; Withers, S. G., Glycosynthase-mediated assembly of xylanase substrates and inhibitors. ChemBioChem 2011, 12 (11), 1703-1711. Tull, D.; Withers, S. G., Mechanisms of cellulases and xylanases - A detailed kinetic-study of the exo-beta-1,4-glycanase from Cellulomonas-fimi. Biochemistry 1994, 33 (20), 6363-6370. Meinke, A.; Gilkes, N. R.; Kilburn, D. G.; Miller, R. C.; Warren, R. A. J., Cellulose-binding polypeptides from Cellulomonas-fimi - endoglucanase-D (CenD), a family A beta-1,4-glucanase. J. Bacteriol. 1993, 175 (7). Chen, C. F. Activity-based Proteomics Profiling for Identification and Quantification of Trichoderma reesei Cellulases. Diss. University of British Columbia, Vancouver, 2011. Stahlberg, J.; Divne, C.; Koivula, A.; Piens, K.; Claeyssens, M.; Teeri, T. T.; Jones, T. A., Activity studies and crystal structures of catalytically deficient mutants of cellobiohydrolase I from Trichoderma reesei. J. Mol. Biol. 1996, 264 (2), 337-349. Bruce, J. M.; Knowles, P., Light-induced and related reactions of quinones: 3. Light-induced reactions of some monosubstituted 1,4-benzoquinones. J. Chem. Soc. C 1966, (18). Hartsel, J. A.; Craft, D. T.; Chen, Q.-H.; Ma, M.; Carlier, P. R., Access to "Friedel Craftsrestricted" tert-alkyl aromatics by activation/methylation of tertiary benzylic alcohols. J. Org. Chem. 2012, 77 (7). Zhao, J. Z.; Wei, S. Q.; Ma, X. F.; Shao, H. W., A simple and convenient method for the synthesis of pyranoid glycals. Carbohydr. Res. 2010, 345 (1), 168-171.  107  88.  Cocker, D.; Sinnott, M. L., Generation of glycopyranosyl cations in spontaneous hydrolyses of 2,4-dinitrophenyl glycopyranosides - Evidence for general intermediacy of glycopyranosyl cations in acid-catalyzed hydrolyses of methyl glycopyranosides. J. Chem. Soc., Perkin Trans. 2 1975, (13).  108  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073565/manifest

Comment

Related Items