UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Synthesis and testing of isofagomine and derivatives as inhibitors of human glucocerebrosidase Hill, Tara 2009

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

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

Item Metadata

Download

Media
24-ubc_2009_spring_hill_tara.pdf [ 1.83MB ]
Metadata
JSON: 24-1.0061636.json
JSON-LD: 24-1.0061636-ld.json
RDF/XML (Pretty): 24-1.0061636-rdf.xml
RDF/JSON: 24-1.0061636-rdf.json
Turtle: 24-1.0061636-turtle.txt
N-Triples: 24-1.0061636-rdf-ntriples.txt
Original Record: 24-1.0061636-source.json
Full Text
24-1.0061636-fulltext.txt
Citation
24-1.0061636.ris

Full Text

SYNTHESIS AND TESTING OF ISOFAGOMINE AND DERIVATIVES AS INHIBITORS OF HUMAN GLUCOCEREBROSIDASE by TARA L. HILL B.Sc., The University of Minnesota Duluth, 2005 A THESIS SUBMITI’ED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS OF THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Chemistry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) January 2009 © Tara L. Hill, 2009 Abstract Gaucher disease is a lysosomal storage disorder in which the activity of the enzyme glucocerebrosidase (GCase) is deficient. As a result, the substrate of GCase, glucosylceramide, accumulates in lysosomes of Gaucher patients, causing enlargement of the spleen and liver, bone deformity and neurodegeneration. Deficient GCase activity results from point mutations in the gene encoding for GCase, which cause the protein to be misfolded. The misfolded mutant form of GCase, although catalytically competent, is identified and degraded by the endoplasmic reticulum-associated degradation (ERAD) pathway; thereby never reaching the lysosome or its substrate, resulting in accumulation. A threshold GCase activity level of 11-15% is all that is needed in order to be asymptomatic, with most Gaucher patients retaining between 5-20% of GCase activity levels. Therefore, therapeutic treatment of Gaucher disease only needs to impart modest increases in GCase activity in order to see large clinical improvements. An emerging therapy for Gaucher disease uses competitive enzyme inhibitors as pharmacological chaperones (PCs), which act to increase GCase levels in the lysosome. The PC binds and stabilizes the native, folded form of GCase, thereby allowing the enzyme to escape ERAD and resume proper trafficking to the lysosome. Once the mutant GCase/PC complex is delivered to the lysosome, the relatively high concentration of accumulated glucosylceramide will compete for the active site of GCase, thereby displacing the PC as normal glycolipid degradation activity resumes. Isofagomine (IFG), a known competitive inhibitor of GCase, was synthesized along with five other C6-alkyl IFG derivatives; of which four lEG derivatives were novel compounds. All six compounds were shown to be potent competitive inhibitors of GCase with measured K, values ranging from 0.2 nM to 610 nM. As a strategy to suppress any inhibitory effects of the PCs after GCase reaches the lysosome, a potentially pH-labile acetal moiety was incorporated into IFG. Unfortunately, the synthesized acetal is very stable and unlikely to cleave under physiological conditions. All molecules synthesized, 11 have been, or will be, sent to our collaborators, Dr. Mike Tropak and Dr. Don Mahuran at the Hospital for Sick Children in Toronto for PC activity testing in Gaucher cell lines. 111 Table of Contents Abstract .11 Table of Contents.iv List of Tables vi List of Figures vii List of Schemes viii List of Abbreviations ix Acknowledl!ments xii 1. General Introduction 1 1.1 Glycosidases 1 1.1.1 Classification of Glycosidases 2 1.1.2 Catalytic Mechanism of Retaining f3-Glycosidases 3 1.1.3 Inhibitors of Glycosidases 6 1.1.3.1 Non-covalent Imino Sugar Competitive Inhibitors 7 1.2 Lysosomal Storage Disorders 10 1.2.1 Gaucher Disease 13 1.2.1.1 Glucocerebrosidase 13 1.2.1.2 The Cause of Deficient GCase Activity 17 1.2.2 Therapies for Lysosomal Storage Disorders 18 1.2.2.1 Enzyme Replacement Therapy (ERT) and Substrate Reduction Therapy (SRT) as Applied to Gaucher Disease 18 1.2.2.2 Enzyme Enhancement Therapy (EET) using Pharmacological Chaperones (PCs) as Applied to Gaucher Disease 19 1.3 AimsofthisThesis 22 2. Synthesis of Isofagomine and Derivatives as Inhibitors of Human GCase and as PC Candidates for Gaucher Disease 23 2.1 Synthesis 23 2.1.1 Synthesis of Isofagomine 23 2.1.2 Synthesis of C6-Alkyl IFG Derivatives 31 2.1.2.1 Synthesis of C6-n-Nonyl IFG 31 2.1.2.2 Synthesis of C6-n-Propyl IFG and C6,6-Di-n-propyl IFG 34 iv 2.1.2.3 Synthesis of C6-[9-Hydroxypropyl] IFG 37 2.1.3 Synthesis of Acetal-Containing IFG Derivatives as pH-Labile Linkers 38 2.1.3.1 Synthesis of C6-Benzyl Acetal IFG 38 2.1.3.2 Other Attempts to Synthesize Acetal-Containing IFG Derivatives 43 2.2 Inhibition Studies with Human GCase 46 2.3 Conclusions 54 3. Materials and Methods 56 3.1 Synthesis 56 3.1.1 General Material and Methods 56 3.1.2 Generous Gifts 57 3.1.3 (3R, 4R, 5R)-5-(Hydroxymethyl)piperidine-3,4-diol [Isofagomine. HC1 salt] (1.8) 57 3.1.4 (3R, 4R, 5R, 6S)-6-n-Nonyl-5-(hydroxymethyl)piperidine-3,4-diol [C6-n-Nonyl IFG•HCI salt] (1.9) 62 3.1.5 (3R, 4R, 5R, 6S)-6-n-Propyl-5-(hydroxymethyl)piperidine-3,4-diol [C6-n-Propyl TFG• HC1 salt] (2.17) 64 3.1.6 (3R, 4R, 5R)-6,6-Di-n-propyl-5-(hydroxymethyl)piperidine-3,4-diol [C-6,6-Di-n-propyl IFG• HCI salt] (2.18) 67 3.1.7 (3R, 4R, 5R, 6S)-6-[9-Hydroxypropyl]-5-(hydroxymethyl)piperidine-3,4-diol [C6-[9-hydroxypropyl] IFG•HC1 salt] (2.23) 68 3.1.8 (3R, 4R, 5R, 6S, 8S)-8-Benzyloxy[5,6-b]oxacyclohexane-5- piperidine-3,4-diol [C6-Benzyl acetal IFG] (2.32) 70 3.2 Enzymology 72 3.2.1 Human Glucocerebrosidase Kinetics 72 3.2.2 Inhibition Studies of Human Glucocerebrosidase 73 References 75 Appendix 81 V List of Tables Table 2.1 Summary of inhibitors synthesized and measured K1 values towards GCase 55 Table 3.1 Summary of inhibitor concentrations used in the assay for each mhi 74 vi List of Figures Figure 1.1 The reaction normally catalyzed by a glycosidase shown with the numbering scheme for a D-glucoside along with glycone and aglycone components 1 Figure 1.2 Stereochemical outcome of reactions catalyzed by inverting and retaining glycosidases 2 Figure 1.3 Half-chair conformation of oxocarbenium ion-like transition state with the co-planar arrangement needed for stabilization of the positive charge build up on Cl 4 Figure 1.4 The ion pair intermediate proposed by Phillips 4 Figure 1.5 Examples of iminosugars that are non-covalent competitive glycosidase inhibitors 8 Figure 1.6 Surface representation of GCase surrounding the active site 16 Figure 2.1 a) Desired product 2.3 of the isopropylidene reaction plus all possible side-products 25 Figure 2. ib) Representation of a TLC plate obtained from the isopropylidene reaction 25 Figure 2.2 ‘H-NMR spectra of ring protons before and after acetylation 27 Figure 2.3 Representation of a TLC plate obtained from the isopropylidene reaction with product identity assigned to each spot 28 Figure 2.4 Structure of the side-product identified from the IFG reaction (2.10) 31 Figure 2.5 General structure and numbering scheme for C6-alkyl IFG derivatives where R=alkyl 31 Figure 2.6 Structures of products 2.17 and 2.18 34 Figure 2.7a) Ideal structure for pH-labile linker shown here with alkyl groups 38 Figure 2.7b) General structure of synthetically attainable pH-labile linker 38 Figure 2.8 Undesired intra-molecular hemi-acetal formation with 04 (2.26) and 05’ (2.27) 39 Figure 2.9 Structure of Grignard reagent (2.33) 44 Figure 2.10 Compounds tested as inhibitors of GCase 46 Figure 2.11 Dixon plot for the inhibition of GCase by IFG (1.8) 48 Figure 2.12 Dixon plot for the inhibition of GCase by C6-n-propyl IFG (2.17) 49 Figure 2.13 Dixon plot for the inhibition of GCase by C6,6-di-n-propyl IFG (2.18) 50 Figure 2.14 Dixon plot for the inhibition of GCase by C6-[9-hydroxypropyl] IFG (2.23) 50 Figure 2.15 Dixon plot for the inhibition of GCase by C6-benzyl acetal IFG (2.32) 51 Figure 2.16 Dixon plot for the inhibition of GCase by C6-n-nonyl IFG (1.9) 53 Figure 2.17 Diagram of several inhibitors of GCase and their IC50 values 54 Figure A. 1 A plot showing the typical hyperbolic nature of the Michaelis-Menten equation (Equation A.6) 83 Figure A.2 A typical Lineweaver-Burk plot for an enzymatic reaction 85 vii List of Schemes Scheme 1.1 The mechanism of a retaining f3-glycosidase 3 Scheme 1.2 Inactivation of a n-retaining glycosidase by use of 2-F ‘Withers’ reagent 5 Scheme 1.3 The degradation of acid gangliosides by enzymes in the lysosome 12 Scheme 1.4 The reaction catalyzed by glucocerebrosidase 14 Scheme 2.1 Numbering scheme and synthetic route to IFG (1.8) 24 Scheme 2.2 Mechanisms and intermediates representing the transformations in the final step of the synthetic route to IFG (1.8) 29 Scheme 2.3 Synthetic route to C6-n-alkyl IFG derivatives 32 Scheme 2.4 Proposed mechanisms for formation of amine 2.16 33 Scheme 2.5 Synthetic route to C6-n-propyl IFG (2.17) and C6,6-di-n-propyl IFG (2.18) 33 Scheme 2.6 Synthetic route to C6-[9-hydroxypropyl] IFG (2.23) 38 Scheme 2.7 Products of ozonolysis reaction followed by reduction with use of either PPh3 (2.24) or DMS (2.25) 39 Scheme 2.8 Two possible products (2.28 and 2.29) resulting from acetal formation with 2-n-alkyl- 1,3-propane diol derivatives and aldehyde 2.24 40 Scheme 2.9 Possible products (2.30 and 2.31) from hydrogenation of 2.28 resulting in complex1H-NMR spectra 41 Scheme 2.10 Synthetic route to C6-benzyl acetal IFG (2.32) 42 Scheme 2.11 Product resulting from dimerization of Grignard precursor (2.36) 45 Scheme A. 1 General scheme for enzyme catalyzed conversion of a single substrate into a single product 81 Scheme A.2 General scheme for enzyme catalyzed conversion of a single substrate into a single product in the presence of a competitive inhibitor 86 viii List of Abbreviations A - Angstrom Abs - Absorbance AcOll - Acetic acid BnOH - Benzyl alcohol BnCOCC1 - Benzyl chloroformate BCA assay - Bicinchoninic acid assay C6 - The carbon at the six position in IFG numbering C6-benzyl acetal IFG - (3R, 4R, 5R, 6S, 8S)-8-Benzyloxy—[5,6-b] oxacyclohexane-5-piperidine-3,4-diol C6,6-di-n-propyl IFG - (3R, 4R, 5R)-6,6-Di-n-propyl-5- (hydroxymethyl)piperidine-3,4-diol C6-[9-hydroxypropyl] lEG - (3R, 4R, 5R, 6S)-6-[9-Hydroxypropyl]-5- (hydroxymethyl)piperidine-3,4-diol C6-n-nonyl lEG - (3R, 4R, 5S, 6S)-6-n-Nonyl-5- (hydroxymethyl)piperidine-3,4-diol C6-n-propyl lEG - (3R, 4R, 5R, 6S)-6-n-Propyl-5- (hydroxymethyl)piperidine-3,4-diol Cbz - Carboxybenzyl CHO - Chinese hamster ovary CNS - Central Nervous System - ci-deuteñum kinetic isotope effect 2,4-dNP-Glu - 2,4-Dinitrophenyl -D-glucopyranoside DMAP - Dimethylaminopyridine DMF - Dimethylformamide DMS - Dimethylsulfide DNJ - Deoxynojirimicin EET - Enzyme enhancement therapy ER - Endoplasmic reticulum ERAD - Endoplasmic reticulum-associated degradation ERT - Enzyme replacement therapy ESI - Electrospray ionization Et3N - Triethylamine ix EtOAc - Ethyl acetate EtOll - Ethanol GCase - Glucocerebrosidase GicCer - Glucosylceramide HTV - Human immunodeficiency virus HRMS - High resolution mass spectrometry Hz - Hertz I - Inhibitor 1C50 - Half maximal inhibitory concentration IFG - Isofagomine i-PrOH - Isopropanol Isofagomine - (3R, 4R, 5R)-5-(Hydroxymethyl)piperidine-3,4-diol kca/knon - Ratio of rate constants for catalyzed and non-catalyzed reactions kC(,t - Catalytic rate constant (turnover number) kDa - Kilodalton kH/kD - Ratio of catalytic rate constants for protio and deuteron substrates K, - Dissociation constant for an enzyme-inhibitor complex K, - Michaelis constant of a substrate LC-MS/MS - Liquid chromatography-tandem mass spectrometry LRMS - Low resolution mass spectrometry LSDs - Lysosomal storage disorders MeOH - Methanol 4-MU-Glu - 4-Methylumbelliferyl -D-g1ucopyranoside nilz - Mass to charge ratio NJ - Nojirimicin NMR - Nuclear magnetic resonance PC - Pharmacological chaperone PET - Positron emission tomography ppm - Parts per million psi - Pounds per square inch (unit of pressure) p-TsOH.H20 - para-Toluene sulfonic acid monohydrate py - Pyridine x QC - Quality control Rf - Retention factor RP - Reverse phase S - Substrate sapC - Saposin C SRT - Substrate reduction therapy TBDMSC1 - tert-Butyichiorodimethylsilane TBDPMSC1 - tert-Butylchlorodiphenylsilane Tf20 - Triflic anhydride THF - Tetrahydrofuran TLC - Thin layer chromatography TMS - Trimethylsilane UV-vis - Ultraviolet-visible v - Reaction velocity Vmax - Maximum reaction velocity xi Acknowledgements I would like to sincerely thank my supervisor, Dr. Steve Withers, for his patience, encouragement, guidance and friendly conversation over the last few years. I also want to extend much gratitude to the Withers group, past and present members, for creating a wonderful atmosphere where mentorship, friendship, laughter and smiles were in no short supply. Without the support system of the Withers group, this work would not have been possible. Special thanks go to my former lab mate, Dr. Sabrina Buchini, for her great friendship, constant advice and willingness to help. Many thanks to: Dr. Tom Morley for his clever synthetic tricks that almost always worked and for his patience with my thousands of questions; Dr. Bojana Rakic for her friendship and help with biological and synthetic work; Dr. Chris Tarling for guidance early on and for inspiring a love of the mountains; Dr. Emily Kwan for her help with Abg expression and purification; Dr. Hongming Chen for being a constant synthetic factory and providing me with substrates for kinetics; and Ms. Miranda Joyce for being our super-powered and all-knowing support. I would also like to acknowledge the NMR lab staff, especially Maria Ezhova, the MS lab staff, and all behind-the-scenes Departmental staff for their assistance. I also owe many thanks to those working on other lysosomal storage disorder projects, Dr. Chris Phenix and soon-to-be Dr. Brian Rempel. I greatly appreciate the time they spent optimizing conditions for work with GCase, for their technical advice and guidance with kinetics, and for their words of encouragement and understanding. To all my friends and fellow grad students in the Department, including but not limited to; the Tanner lab members, 3C floor A-wing residents and fellow TAs, many thanks for making the week as fun as the weekends. Last, but certainly not least, I would like to thank my wonderful husband, Austin, for his tremendous support and patience. I feel truly lucky to have you by my side. Without you this would not have been possible! xii General Introduction 1.1 Glycosidases It has been estimated that around 60% of the carbon in the biosphere is present in the form of carbohydrates. Due to this abundance, nature has developed extensive and varied uses for these molecules. To the cell, carbohydrates represent three essential components; energy, communication and structure. Examples highlighting the role of carbohydrates in each of these include: Energy: Glucose is at the very center of glycolysis; a process that fuels the cell. Communication: Antigens are often composed of oligosaccharide chains that are important in self vs. non-self recognition by the immune system. Structure: Cellulose and chitin, polymers of carbohydrates, give plants and crustaceans the structural rigidity needed for survival. These few examples show that carbohydrates are not simply of interest to the glyco scientist, but of central importance to the way life sustains and propagates itself. Nature needs a way to manipulate these all-important carbohydrates and it does so with enzymes that catalyze their synthesis, transfer, modification and hydrolysis. In particular, glycosidases are the enzymes that catalyze the hydrolysis of the carbon- oxygen glycosidic bond between the glycone and aglycone components (Figure 1.1). 6 OH Glycosidase + ROH OH OR H20 OH Glycone Aglycone Figure 1.1 The reaction normally catalyzed by a glycosidase shown with the numbering scheme for a D-glucoside along with glycone and aglycone components. OH 1 It is important to note that glycosidases achieve extraordinary rate enhancements (kcatlknon) of more than 1017 in the reactions they catalyze.1 In the absence of these powerful catalysts, a typical extremely stable glycosidic bond would take more than 5 million years to spontaneously hydrolyze.’ 1.1.1 Classification of Glycosidases Glycosidases can be classified by several criteria including: 1. The nature of the substrate/glycone. A specific glycosidase usually exhibits maximum specificity and activity for a specific glycone. For example, a glucosidase will more readily cleave a glucoside than a mannoside or a galactoside. 2. The anomeric configuration of the substrate. Usually a glycosidase will catalyze exclusive cleavage of either an o. or 1 glycosidic linkage. 3. Stereochemical outcome of the catalyzed reaction. A glycosidase will catalyze the cleavage of the glycosidic bond with either retention or inversion of configuration at the anomeric centre (Figure 1.2). 4. Sequence homology. Based on amino acid sequence similarities, glycosidases have been classified into some 113 different families. This system for classifying glycosidases is called CAZy and was developed by Henrissat.2 It can be accessed at (http://www.cazy.org/). Inverting OR Retaining OH Figure 1.2 Stereochemical outcome of reactions catalyzed by inverting and retaining glycosidases. 2 1.1.2 The Catalytic Mechanism of Retaining -Glycosidases First proposed in 1953 by Koshland3and now widely accepted, the mechanism for retaining -glycosidases consists of two steps that each proceed with inversion of stereochemistry leading to a net retention of stereochemistry. This process is referred to as a double displacement. Typically two carboxylic acid residues in the active site are essential for this mechanism. One plays the role of the catalytic nucleophile and leaving group, while the other is the catalytic acid/base (Scheme 1.1). In the first step of the mechanism (glycosylation), the catalytic nucleophile attacks the anomeric center, and with assistance from the catalytic acid, displaces the aglycone to form a covalent glycosyl-enzyme intermediate. In the second step of the mechanism (deglycosylation), an incoming water molecule or carbohydrate (in the case of transglycosylation) is deprotonated by the catalytic base as it attacks the anomeric center and displaces the enzyme nucleophile, regenerating free enzyme and product. HO66R OH 0 Jw Scheme 1.1 The mechanism of a retaining f3-glycosidase. Glycosylation t H20 OH HO t HOOH - 0:10 HO- vw OH HO H c H OH’ dgiycosyiation 3 Each step in this mechanism proceeds though an oxocarbenium-ion like transition state that must stabilize the developing positive charge on the anomeric carbon (Cl) as the aglycone departs. This stabilization is accomplished by efficient orbital overlap and electron delocalization, hence partial double bond character, between the p-like orbitals on the endocyclic oxygen (05) and Cl (Scheme 1.1 and Figure 1.3). A conformational change is necessary to line up the orbitals, which results in trigonal geometry at Ci and greater sp2 hybridization. The substrate must be distorted and take on a half-chair conformation with a coplanar arrangement between C-5, 0-5, C-i and C-2 atoms (Figure 1.3). Figure 1.3 Half-chair conformation of oxocarbenium ion-like transition state with the co-planar arrangement needed for stabilization of the positive charge build up on Cl. In 1967 Phillips4proposed an alternative to the glycosyl-enzyme intermediate in the double displacement mechanism. He suggested an sp2 hybridized ion-pair intermediate (Figure 1.4). After many experiments and much debate, the double displacement mechanism with a covalently linked glycosyl-enzyme intermediate is now widely accepted. Figure 1.4 The ion pair intermediate proposed by Phillips.4 The first evidence against the ion-pair intermediate came with Sinnot and Souchard’ s o-deuterium kinetic isotope effect (KIE) experiments5as defined by kHIkD. For each of the seven substrates used in this study, it had already been demonstrated that 4 the second step (deglycosylation) was rate limiting.6 The resulting (kH/kD)> 1 indicates that the second step involves a decrease in hybridization at the anomeric centre from sp3 to sp2, which is fully consistent with the double displacement mechanism and a glycosyl enzyme intermediate. If an ion-pair intermediate did exist then an inverse Kifi would be expected from the increase in hybridization (sp2 to sp3)needed to break down the intermediate en-route to product.5 More evidence in favor of the double displacement mechanism comes in the form of X-ray crystal structures that reveal the covalent intermediate. Modified substrates containing an electronegative fluorine substituent at either C2 or C5 are used to inductively destabilize the positive charge build up in the oxocarbenium-ion like transition state.7 These molecules are actually mechanism-based inactivators and have proven useful in many experiments — even garnering the name of ‘Withers’ reagents.8 The destabilizing effect of fluorine slows each step in the mechanism, but in order to accumulate the intermediate, glycosylation must proceed faster than deglycosylation. This tuning can be accomplished by using a good leaving group, typically fluoride or dinitrophenolate, as the aglycone of the substrate. A good leaving group mitigates the destabilizing effect of fluorine, speeding glycosylation while deglycosylation remains slow, thereby resulting in accumulation of the glycosyl-enzyme intermediate (Scheme 1.2). OHHA __ XH Slow deglycosylation X=F or 2,4-dinitrophenol Scheme 1.2 Inactivation of a 3-retaining glycosidase by use of 2-F ‘Withers’ reagent. The trapped species is long lived enough to obtain crystals that can be imaged using X-ray crystallography, confirming the covalent bond between the glycosyl moiety 5 and the enzyme.9”° This trapping approach can also be used to identify the catalytic nucleophile through proteolytic digestion of the labeled enzyme and subsequent LC MS/MS analysis1’(and as reviewed by Withers and Aebersold).12 Experimental results have also shown the active site carboxylates to be consistently positioned about 5 A apart from one another, leaving ample room for the proposed mechanism to take place.13’4 Attempts to determine which of the active site residues are responsible for general acid/base catalysis have used crystal structures along with mutagenesis and kinetic studies (as reviewed by Zechel and Withers with references therein).’0 As is expected, and observed, acid/base mutants exhibit rates that vary widely with the leaving group ability of the aglycone. For substrates with poor leaving groups (disaccharides, methyl glycosides), initial cleavage of the glycoside is greatly slowed when compared to substrates with good leaving groups. On the other hand, the deglycosylation step is greatly slowed for all substrates because base catalysis is unavailable for deprotonation of the incoming water nucleophile. As a result, the covalent intermediate accumulates for substrates with good leaving groups in a pre-steady state kinetic ‘burst’, accompanied by unusually low Km values. These trapped acid/base mutants can have their activity rescued by addition of nucleophilic anions such as azide, formate or acetate. These anions do not need general base assistance and turn over the glycosyl-enzyme intermediate to generate a new product with retained stereochemistry. With wild-type enzyme, the same anion rescue experiments yield normal hydrolyzed substrates. This is presumably due to charge repulsion between the general base carboxylate and the anion, making it more favorable for water to enter the active site and react. 1.1.3 Inhibitors of Glycosidases Glycosidase inhibitors fall into several categories because they can bind enzymes reversibly, as with non-covalent inhibitors, or irreversibly, as with most covalent inhibitors. A class within the covalent category is that of the aforementioned mechanism based inactivators (Scheme 1.2). Further classification within the non-covalent type includes competitive, non-competitive/mixed and uncompetitive inhibitors. Competitive inhibitors bind to the enzyme active site thereby competing with the natural substrate for 6 this space (See Appendix for theory on reversible competitive inhibition). This thesis will focus on competitive, non-covalent inhibitors of glycosidases. Glycosidases can achieve such great rate enhancements because their active sites lower the activation energy of a particular reaction by binding the transition state much better than the ground state.15 It is not surprising then, that the best competitive glycosidase inhibitors mimic aspects of the oxocarbenium ion-like transition state such as positive charge and/or planar geometry. These competitive inhibitors can be powerful tools in kinetic assays for probing mechanistic aspects of a reaction.16 They can also be used to gain insight into important active site binding interactions through crystallographic studies with enzyme/inhibitor complexes.17 Using inhibitors to better understand substrate binding and catalysis will not only lead to better inhibitors, but possibly better enzymes with increased activities18 and/or altered specificities.’9 Glycosidase inhibitors are used in a wide variety of settings, as described above, as well as in therapeutics. They are either proposed or shown to be useful in the treatment of diabetes20,H1V2,cancer22 and influenza.23 As the biological roles of carbohydrates and their interactions with glycosidases continue to be elucidated, inhibitors of these enzymes will enjoy much ‘job’ security. 1.1.3.1 Non-covalent Iminosugar Competitive Inhibitors Although there is a good degree of ambiguity in the literature surrounding nomenclature of these compounds, in this thesis the general term iminosugar will be used for sugar-like, polyhydroxylated molecules containing at least one nitrogen, either in an endocyclic or exocyclic position (Figure 1.5). Many of these compounds are natural products that collectively have been shown to be potent competitive inhibitors of a wide range of glycosidases.24’56 At physiological pH, the nitrogen in most iminosugars is protonated and carries a positive charge, thus mimicking the oxocarbenium ion-like transition state that glycosidase active sites have evolved to stabilize.27’8 7 HONH 1.1 R=OH Nojirimicin 1.2 R=H Deoxynojirimycin éH 1.4 Swainsonine OH I COOH HO—NH HOH HO-, NHAc 1.6 Siastatin B 1.7 Fagomine HO 1.8 Isofagomine Figure 1.5 Examples of iminosugars that are non-covalent competitive glycosidase inhibitors. Arguably the most prominent iminosugar, nojirimicin (NJ) (1.1) contains a nitrogen atom at the endocyclic oxygen position. Interestingly, its synthesis was accomplished ten years prior to its discovery in nature or knowledge of its powerful biological activity.29 The deoxygenated version, deoxynojirimicin (DNJ) (1.2), also inhibits many enzymatic targets including both cr and glucosidases and has attracted the attention of synthetic chemists looking to expand the scope of this type of molecule. Nearly every possible isomer of DNJ, along with hundreds of N-substituted and C- 1.3 Acarbose OH 1.5 Castanospermine 8 branched derivatives have been synthesized and tested for biological activity30 (and references 6-9 therein), Of particular success is the N-hydroxyethyl version of DNJ, which acts as an o -glucosidase inhibitor for the treatment of diabetes and is marketed as miglitol (Glyset®).3 Another iminosugar natural product used for treatment of diabetes is acarbose (sold in North America as PrecoseTM) (1.3).20 The key moiety in this metabolically stable pseudo-tetrasaccharide is the valienamine unit with an exocyclic nitrogen atom. The valienamine moiety also contains a double bond, which serves to flatten the ring, thereby further mimicking the oxocarbenium ion-like transition state. Quite a few bicyclic iminosugars have also been identified from biological sources such as swainsonine (1.4), an ct-mannosidase inhibitor,32 and castanospermine (1.5), a broad spectrum glucosidase inhibitor.25’33 The discovery of natural products siastatin B (1.6), where the nitrogen atom replaces the carbon at the anomeric centre, and fagomine (1.7), which is the 2-deoxy version of DNJ, spurred the synthesis of a semi- rationally designed, very potent 13-glucosidase inhibitor, isofagomine (IFG) (1.8).3536 Placing the nitrogen at the anomeric centre leads to dramatic increases in inhibitory power towards certain glycosidases. An interesting comparison between DNJ and lEG, both inhibitors of c and 13 retaining glucosidases, is that DNJ is a more potent inhibitor of ct glucosidases by roughly 3-fold, while lEG is more potent towards 13 glucosidases by nearly 500-fold.36 This phenomenon can be partially explained by the different transition state charge distributions in the reactions catalyzed by a. and 13 glucosidases37(and as reviewed by Zechel and Withers).’° Many derivatives38’9and isomers40’1 of IFG have also been synthesized. N alkylated derivatives of IFG generally exhibit lower levels of inhibition compared to the parent lEG nonetheless, there is still interest in these molecules.39 Along with IFG, N adamantyl and N-octyl IFGs have been investigated as potential pharmacological chaperones for treatment of the lysosomal storage disorder, Gaucher Disease,42with lEG currently in clinical trials (http://www.amicustherapeutics.comlclinicaltrials/at2101 .asap). Along with IFG, 9 structures of N-butyl DNJ and N-nonyl DNJ complexed with glucocerebrosidase, a lysosomal hydrolase, have been determined by X-ray crystallography in order to gain insights into important binding characteristics.17’45 One research group has taken on the synthetic challenge of making C6-branched alkyl IFG derivatives and honed in on the most potent glucocerebrosidase inhibitor to date, C6-n-nonyl IFG (1.9) with an 1C50 value of 0.6 nM.38 1.2 Lysosomal Storage Disorders (LSDs) Lysosomes are organelles within the cell that are essentially the recycling depots and which operate under acidic conditions at pH 5•46 Various enzymes within the lysosome are responsible for the proper degradation and recycling of cellular components such as glycoproteins and glycolipids.47 If there is a deficiency in the activity of a particular degradation enzyme, that enzyme’s substrate will accumulate. This physical storage, along with a perturbation of signaling pathways, leads to a disease state.4749 So far around 40 of these disorders have been characterized, each by itself rare, but taken together there is a prevalence of 1 in 7,700 in the general population.5°These are collectively known as lysosomal storage disorders. Amongst the glycolipids degraded in the lysosome there is a class of molecules known as gangliosides. These are glycosphingolipids that contain a ceramide moiety attached to an oligosaccharide chain (Scheme 1.3). Ceramide is a lipid that imbeds in the membranes of animal cells in order to display the oligosaccharide chains into the extracellular space. This display allows for cell-cell interactions that mediate signaling and differentiation.5’ The gangliosides are degraded in a step-wise fashion by specific enzymes in the lysosome that act on a specific glycone (Scheme 1.3). If there is a missing enzyme activity in this orderly line of degradation, an accumulation of its substrate will occur, which contributes to disease phenotypes. Each accumulation product and disease phenotype is specific to the deficient enzyme. Scheme 1.3 shows the degradation of acid 10 gangliosides by lysosomal enzymes as well as the diseases associated with the deficiencies. This thesis will focus on the enzyme glucocerebrosidase and the disease associated with it, Gaucher disease. 11 GM1 ganglioside OH OH 02C GM3 ganglioside OH HOHO OH HOOH H20 nglioside neuraminidase NANA Sialidosis OH Lactosyl ceramide OH H2O Gal HO_ocer Glucosylceramide OH H20j1cocerebrosidase Gaucher disease Glc 0 HN’(CH2)12CH3 Cer = Ceramide = OH Scheme 1.3 The degradation of acid gangliosides by enzymes in the lysosome. Enzymes are written in blue and the disease associated with the enzyme deficiency is written in red. G—gaLactosidase Gal G1gang1iosidosis f3—Hexosaminidase A Ga1NAc Tay Sachs disease 12 1.2.1 Gaucher Disease Gaucher disease is the most common of all lysosomal storage disorders with a prevalence of 1 in 40,000—60,000 in the general population and 1 in 800 among the Ashkenazi Jewish population.5°Like all other lysosomal storage disorders, it is a heritable disease and was first described in 1882 by Phillipe Gaucher in his medical thesis.52 In it he described a patient with abnormal spleen cells, but it wasn’t until 1907 that Aghion characterized that abnormality to be the storage of glucosylceramide.52 Since then, the phenotypes of Gaucher disease have been elucidated to include enlargement of the liver and spleen, bone deformity, anemia, neuronopathic/central nervous system (CNS) involvement and death.47’952 The severity of these symptoms can vary widely, as does the age of onset and degree of CNS involvement. As a result, Gaucher disease is classified into three types. The most common and mild form is type 1 where patients lack CNS involvement and have early childhood to adult onset. Type 2 patients experience infant onset with rapid and severe neuronopathic involvement often resulting in death before age 2. Type 3 patients experience a slower childhood onset with typically milder neuronopathic involvement.52 There is no cure for Gaucher disease but there are currently two types of therapies available to patients, with an additional therapy in clinical trials; all of which will be discussed in section 1.2.2. 1.2.1.1 Glucocerebrosidase Glucocerebrosidase (GCase) is a membrane-associated 67 kDa protein with 497 amino acids.52 As a 3-g1ucosidase from CAZy family 30,2 GCase catalyzes the cleavage of the glycosidic bond between glucose and ceramide in a retaining fashion as shown in Scheme 1.4. The proposed mechanism for this transformation is that of the retaining I glycosidases shown in Scheme 1.1 (page 3). The pH optimum for GCase is 5.5 and corresponds to the pH maintained in the lysosome. 13 0 o OH HN”(CH2)12CH3 Glucocerebrosidase ç..OH HN’L(CH2)12CH3 H20 HO.OH + HO._..(CH)iC Scheme 1.4 The reaction catalyzed by glucocerebrosidase. The substrate, glucosylceramide (GicCer), resides in the inner lysosomal membrane and is not water soluble. The question of how GCase, a soluble protein, accesses GicCer, an insoluble substrate, must be addressed.54 For full activity in vivo, GCase requires negatively charged phospholipids and the activator protein saposin C (sapC).52 The exact mechanisms of activation are still being elucidated but recent studies have observed conformational changes in both GCase and sapC that must take place in order for proper catalysis to occur.55’6 The “liftase” model, which is most supported in the literature, suggests that sapC binds and simply perturbs the surface of the membrane thereby exposing GlcCer. GCase recognizes this membrane-bound sapC complex and binds at its interface gaining access to the substrate.55 An alternative “solubilizer” model suggests a more active role for sapC where it not only perturbs the membrane, but actually extracts GicCer and presents it to GCase as a soluble lipid-protein complex, as is seen with Saposin D.55 X-ray crystal structures of GCase show many hydrophobic residues in close proximity to the active site which may facilitate the membrane and sapC associations.57 When GicCer is modeled into GCase using computational programs, there are ambiguous conclusions about where the hydrophobic tails reside in relation to GCase. In one simulation the tails point away from the protein, indicating interactions with either the membrane or sap C.57 The other simulations have modeled the tails residing in two valleys emerging from the active site.17’45 In the valley scenario, the hydrophobic tails can still interact with the membrane or sapC. Extensive in vitro work has shown that GCase, in the absence of negatively charged phospholipids and sapC, requires triton-X 100 and sodium taurocholate for activity.58 14 Several X-ray crystal structures of wild type GCase have been solved under different conditions including: with a covalent inhibitor59,with several non-covalent competitive inhibitors’7’45,without inhibitors45’57as well as at neutral and acidic pH values.45 The crystal structure solved at pH 7.5 is the first and only at that pH as well as the first truly apo-structure.45 The findings of all these reports reveal several important dynamic loops that flank the active site (residues 311-319 : loop 1, residues 342-354: loop 2, residues 393-396 : loop 3). These ioops adopt a variety of conformations in the free enzyme at both pH values but interestingly, when a competitive inhibitor was found in the active site (only at acidic pH), the ioops were always fixed in one preferred conformation with loop 1 being found in a helical conformation.’7’45 Specific hydrogen bond networks were also observed, which stabilize these preferred conformations and allow open access to the active site (Figure l.6b).45 This open active site conformation was always observed in inhibitor/enzyme complexes, whereas it was only observed once at neutral pH in the structures of free enzyme (Figure l.6c). For all other free enzyme structures, different hydrogen bond networks were observed and subsequently, the active sites in these structures were largely closed and inaccessible. (Figure 1 .6a,d). Attempts to computationally dock GicCer into the active sites of inhibitor/enzyme versus free enzyme structures yielded good scores only for bound structures with open active sites (Figure 1 .6e).45 Movements of the loops that are needed to open or close the active site may also correspond to changes needed in order to interact with lipids, sapC and/or the membrane. 15 bd Active sit -w Loop 2 Loop 1 Loopi — Loop 2 Loop I Figure 1.6 Surface representation of GCase surrounding the active site, a) pH = 4.5 with glycerol bound in active site. Active site closed. b) pH =7.5 with IFG bound in active site. Active site open. c) pH =7.5 free enzyme. Active site open. d) pH = 7.5 free enzyme. Active site closed. e) Truncated GicCer docked into structure b. Figure taken from Lieberman et. al. ‘ Reprinted with permission from MacMillan Publishers Ltd: [Nature Chemical Biology] (Lieberman et. al. Nat. Chem. Biol. 2007, 3, 101-107), copyright 2007 e 16 1.2.1.2 The Cause of Deficient GCase Activity Like many proteins, GCase is synthesized and folded in the endoplasmic reticulum (ER), which maintains neutral pH. Under normal conditions, the emerging polypeptide chain is folded properly (native form), secreted out of the ER, transported to the Golgi apparatus for further maturation, and finally trafficked to the lysosome where it degrades GlcCer. A point mutation in the gene encoding GCase can result in an unstable, slightly misfolded GCase variant (non-native form), which is subsequently identified and retained in the ER by the quality control (QC) system. QC targets the misfolded protein for degradation by ER-associated degradation (ERAD) and as a result, GCase never reaches its substrate (GicCer) in the lysosome and GicCer accumulates. For extensive reviews on ER QC and ERAD see Ellgaard et al.6° and Yoshida.48 One interesting feature to note is that most people suffering from this enzyme deficiency express catalytically active, yet unstable, GCase, making this mostly a problem of protein misfolding and subsequent degradation.61 If these enzymes could ‘fool’ the QC and ERAD mechanisms and maintain proper trafficking to reach the lysosome, they would be able to properly catalyze degradation of GicCer. The residues responsible for maintaining the hydrogen bond networks that surround the active site and act to stabilize the open conformation have been identified,17’45and correspond to some of the GCase mutants that lead to Gaucher disease. Notably N370, the most common mutation seen in Gaucher patients (N370S), has a direct role in stabilizing the preferred helical conformation of ioop 1. In the absence of inhibitor, it is known that GCase samples many conformations.45 It is possible that when any remote residues are mutated, the resulting interactions bias closed active site conformations of GCase or allow a loop to be disordered. It is not known what exact structural features cause GCase variants to be identified by QC and degraded by ERAD. It is, however, generally understood that quatemary structure elements like hydrophobic patches, mobile loops and lack of compactness are important in the differentiation of native versus non-native proteins by cellular processes.6° 17 1.2.2 Therapies for Lysosomal Storage Disorders The therapies about to be discussed are in specific reference to GCase and Gaucher Disease but there are examples of parallel therapies for many other LSDs.62 Each disease has its own special set of circumstances that need to be optimized for each therapeutic strategy. In theory, each strategy can be applied to any LSD, as all share common themes of misfolded proteins leading to enzymatic deficiencies. 1.2.2.1 Enzyme Replacement Therapy (ERT) and Substrate Reduction Therapy (SRT) as Applied to Gaucher Disease A threshold GCase activity level of 11-15% is all that is needed in order for Gaucher patients to be asymptomatic, with lower levels of activity corresponding to a worsening clinical course of disease. It has also been demonstrated that many type 1 patients exhibit residual GCase activity levels that are 5-20% of normal levels.63 This means that modest increases in GCase activity could prove to be life changing for some patients. Today there are two types of therapy available for Gaucher patients, with a third on the way. The first is called enzyme replacement therapy (ERT)64 and involves injection of the recombinant form of GCase (Cerezyme®)65directly into the patient’s bloodstream. ERT is only effective for type 1 patients without CNS involvement because the administered enzyme cannot cross the blood brain barrier. This treatment option, offered since 199 1,66 is very expensive with costs upwards of $200,000 per year, per patient and even more to administer it.67 Current research efforts surrounding this strategy focus on expression of GCase in systems other than Chinese hamster ovary (CHO) cells68 and elucidation of GCase bio-distribution through use of PET imaging. This information should help to tune the dose to an individual in order to minimize costs. The other therapy available to Gaucher patients is substrate reduction therapy (SRT). This involves inhibiting the enzyme responsible for GlcCer biosynthesis with N butyl deoxynojirimicin (NB-DNJ, Miglustat, Zavesca®).69 With a reduction in the 18 amount of GicCer being synthesized, less is stored in the lysosome and the clinical course of the disease can be improved. However, this therapy is accompanied by some serious side-effects.7°It is not known what long-term effects this may have on glycolipid distribution since these molecules and their roles are only partially understood. Both ERT and SRT address the GlcCer storage problem, but do not address downstream effects of the unfolded protein response in which ER stresses, such as ERAD activation by misfolded proteins, can also activate other inflammatory pathways and induce apotosis.48 1.2.2.2 Enzyme Enhancement Therapy (EET) using Pharmacological Chaperones (PCs) as Applied to Gaucher Disease. As previously stated, only modest increases in GCase activity are needed to alleviate the maladies of Gaucher Disease. An upcoming and very promising therapeutic strategy called enzyme enhancement therapy (EET) seeks to use small molecule pharmacological chaperones (PCs) in order to stabilize the native conformation of misfolded GCase mutants.61’72 This PC-induced stabilization ‘fools’ the QC and ERAD mechanisms, allowing the patient’s endogenous enzyme to be trafficked to the lysosome as normal. The majority of small molecules identified thus far as PC candidates for Gaucher Disease have been competitive inhibitors of GCase.6’ By definition, competitive inhibitors bind to the enzyme active site and it is through these interactions that the native conformation of GCase is stabilized.6’Also by definition, competitive inhibitors can be displaced from the active site if another inhibitor or substrate is present in high enough concentrations. Ideally, increased levels of mutant GCase reach the lysosome following PC treatment, and a high concentration of accumulated GicCer will compete for the active site, thereby displacing the PC and allowing normal degradation to take place. This whole therapeutic strategy leads to the apparent paradox that an inhibitor will increase cellular enzymatic activity. For this strategy to be viable, the PC dosing must be fine-tuned in order to maximize lysosomal GCase levels. At certain concentrations, which vary depending on 19 the PC and the mutation, these molecules will actually decrease mutant GCase activity; adding insult to injury for someone already suffering because of insufficient activity.73 Even with the precautions and apparent paradoxes, PC therapy is desirable because small molecules can often cross the blood-brain barrier, making this a potentially useful therapy for type 2 and 3 Gaucher patients with CNS involvement. Another benefit of small molecule therapies, as opposed to peptide and protein-based therapies, is the reduced cost for administration.67 A third benefit is PCs allow the endogenous mutant enzyme to traffic normally. Thus, EET has the ability to relieve downstream effects from the unfolded protein response.48 lEG is currently in clinical trials as a PC (sponsored by Amicus Therapeutics) for EET of Gaucher Disease (http://www.amicustherapeutics.com/clinicaltrials/at2101.asap). The exact mode of GCase stabilization and chaperoning by IFG has been proposed based on H/D exchange mass spectrometry74and crystallographic studies’7’435both with and without IFG bound. A sampling of many conformations occurs in a rugged energy landscape as the mutant protein is being folded in the ER. As previously stated in section 1.2.1.1, the native form of GCase requires that ioop 1 be in an ordered helical conformation for the active site to be open and accessible. It is this native conformation that IFG will preferentially bind, thereby stabilizing loop 1 and locking it in the preferred conformation. IFG binding is also thought to impart greater global stability to GCase mutants as observed by a decrease in H/D exchange74and an increase in melting temperatures.75 This lEG-induced, globally stable native form is no longer a marker for ERAD and normal trafficking to the lysosome will resume. In order for IFG to function as a PC according to this mechanism an assumption must be met; that mutant GCase can sample the native conformation during ER folding. It is possibly because of this that PCs exhibit chaperoning profiles that are highly dependent on the particular mutation. For example, IFG can act as a PC for the most common GCase mutant N370S, but for L444P mutants it shows no efficacy.73 20 Strategies to optimize selectivity and/or activity of Gaucher PCs include high throughput screening of large compound libraries,74’6as well as rationally designed small molecules.38’4277The hopes of these efforts are to find small molecules that are either non-active site directed PCs, or active site directed PCs that exhibit differential binding profiles under different conditions, particularly different pH conditions. Strong inhibition representative of low K1 values would be ideal at neutral pH (conditions of the ER) with weaker inhibition and higher K1 values being ideal at acidic pH (conditions of the lysosome). These sorts of PCs are desirable because of the concern that when using potent in vitro inhibitors, they will behave as such in vivo as well. One way to rationally design a molecule that has differential binding characteristics in different organelles is to build in a pH-labile, or local environment- labile linker. With the lysosome operating at pH 5 and home to many proteases, an acetal or amide linker may well be susceptible to hydrolysis under these conditions. This would allow the inclusion of a hydrophobic arm which, based on previous studies,38 should impart tighter binding to GCase. Once the linker is cleaved under the predicted conditions and the hydrophobic arm released, a less potent GCase binder would remain and concerns for the molecule inhibiting delivered GCase would be alleviated. The idea for this strategy came about after a surprising observation from a study that tested many different small molecules as PCs for GCase. An adamantyl group was incorporated, into two otherwise identical molecules, via an ether or amide linkage. The amide linked version exhibited a 150% increase in GCase activity while the ether-linked version was inhibitory at all concentrations tested.73 The authors commented that the result was unexpected, but did not attribute it to cleavage of the amide. However it seemed possible to us that cleavage of the amide (but not the ether) in the lysosome could give rise to this outcome. 21 1.3 Aims of this Thesis Isofagomine (1FG) has been shown to act as a pharmacological chaperone (PC) in the treatment of Gaucher Disease in which GCase activity is deficient. It has also been. shown that C6-alkyl derivatives of IFG are the most potent inhibitors of GCase to date. The goal of my work is to develop novel GCase inhibitors based on the C6-alkyl derivatives of JFG that also have the potential to behave as PCs for the treatment of Gaucher Disease. The specific aims of this thesis towards that goal are three-fold. The first aim is to use synthetic organic and carbohydrate chemistry methodologies in order to synthesize novel C6-alkylated versions of IFG. In particular, it is ideal if these IFG derivatives contain a pH-labile acetal linker, and/or have the potential to exhibit differential binding profiles under the conditions encountered in the ER versus the lysosome. The second aim is to measure the GCase inhibition constants (K, values) of the synthesized IFG derivatives and determine what kind of inhibition they exhibit. The third aim is to provide promising PC candidates to our colleagues Dr. Don Mahuran and Dr. Mike Tropak in Toronto at the Hospital for Sick Children for PC activity testing in Gaucher cell lines. 22 2 Synthesis of Isofagomine and Derivatives as Inhibitors of Human GCase and as PC candidates for Gaucher Disease 2.1 Synthesis 2.1.1 Synthesis of Isofagomine Isofagomine (IFG) (1.8) is not found in nature and therefore all the material used in research and medicine must come from a synthesis. As a result several routes have been developed to access IFG and derivatives. When making this seemingly very simple molecule there are several major challenges that need to be overcome. First, it is necessary to introduce the nitrogen atom selectively, which often requires the use of complex protecting group strategies, and second, the stereochemistry at the three stereocentres must be controlled. A final consideration is choosing starting materials that can provide some of these elements but without exorbitant costs. Carbohydrates with the desired stereochemistries, or alkaloids with the nitrogen in a desirable position, are attractive options. In 1994 the first route to IFG was reported by Jespersen et al.,35 wherein IFG was synthesized in 10 steps from levoglucosan. The stereochemistries throughout the synthesis were set within the starting material and via a selective epoxide opening with vinyl magnesium bromide. The nitrogen atom was introduced via reductive amination with NH3. Subsequent publications describing syntheses of IFG have commented on the length and complexity of this route.39 The second route to IFG was published in 1995 by Ichikawa et al.,78 and started from D-Lyxose. The nitrogen was installed via an azido substitution of a tosyloxy group with inversion of stereochemistry, followed by hydrogenation to reach IFG in seven steps. Stereochemistry throughout this synthesis was largely controlled by the starting material and a partially selective deoxygenation. This synthesis also suffers from lengthy protecting group manipulation and complex purification procedures to separate diastereomers. As a brief sampling, other groups have synthesized IFG in five steps from the alkaloid arecolin,79 in five steps from D-arabinose,8°in seven steps from (R)-2,3-O- 23 cyclohexylidene-glyceraldehyde,4°and in seven steps from L-xylose.8’ Following the appearance of the L-xylose synthetic route in the literature, many of the new syntheses published have used this general methodology but either have reduced the number of steps, or reduced the use of expensive reagents.82 There have also been efforts towards the development of divergent syntheses that can access many different piperidine/IFG isomers.41 However, these routes are lengthy in comparison to those developed to exclusively synthesize IFG. A short four step synthesis adapted from the L-xylose synthetic route was published in 2005 by Zhu et al.38 It seemed to provide a convenient platform for the development of C6-alkyl IFG derivatives, exactly along the lines of the project’s goal. For this reason, Zhu’s route and protecting groups were chosen for reproduction and exploration (Scheme 2.1). A year later in 2006, when the project was already underway, Goggard-Borger et al. published a modification of this method which made use of D-arabinose, a cheaper starting material than L-xylose.83 4 2-methoxypropene BnOH, BF3’Et20 p-TsOHH0,THF (39 ) OBn (49%) 2.1 2.2 1) py,Tf20 CHI,2 h 2) KCN, 18[crownl-6 DMF (80%) N o 1) H2, Pd(OH)2/C MeOH AcOH,MeOH, 50ps 1.8 -ici (89%) 2.15 3) BnOCOCI, OBnfr THF:MeOH:H20 2.9 NaHCO3 (35%) Scheme 2.1 Numbering scheme and synthetic route to IFG (1.8). Adapted from Zhu et. at. 38 For the portions of this thesis describing syntheses that were exact reproductions of Zhu’s work, I will focus on the steps that I had difficulty reproducing, and what was done to synthesize the target molecules. 24 Zhu’s published protocol begins with benzyl c-L-xyloside (2.2), so this needed to be made in large quantities from L-xylose (2.1) first. For initial reaction optimization, the much cheaper D-xylose was used as the starting material. Once good conditions had been established, it was reproduced with L-xylose (2.1). The benzyl xyloside was made under Fischer glycosylation conditions by refluxing the starting material and acid catalyst, BF3’Et20, in benzyl alcohol. The product was isolated by precipitation in diethyl ether and purified by recrystallization from hot EtOH. The trade-off for such an easy protocol is a poor reaction yield; as demonstrated in my hands and reported as such in the literature.83 After the anomeric centre was protected, hydroxyls 2 and 3 were protected by installation of an isopropylidene group (Scheme 2.1, page 24). Starting with triol 2.2, the addition of 2-methoxypropene and p-TsOH in THF at 00 C resulted in the formation of the desired product (2.3) along with several side-products (Figure 2.la). Minor spots— Major spot — 2.lb 2.la Figure 2.1 a) Desired product 2.3 of the isopropylidene reaction plus possible side-products. Structure 2.7 indicates incomplete installation of the protecting group at either the 2, 3 or 4 positions with the remaining two hydroxyls free. b) Representation of a TLC plate obtained from the isopropylidene reaction. The black spot indicates highest degree of staining by molybdate and the grey spots indicate staining to a lesser degree. 2.5 OBn 2.4 0N26 OBn 2.7 25 Zhu reported yields of 53% for synthesis of 2.3, thus he was able to attain partial selectivity. In my hands, stoichiometry was important because when more or less than 2.5 equivalents of electrophile were used, the reproducibility was poor. Perhaps even more important was the amount of acid catalyst used. When the amount of p-TsOH was varied from 6.3 mol% to 2.5 mol%, there was an increase in selectivity for the desired product (2.3) from 42% in the crude mix to 66%. The TLC plate showed three spots when stained with molybdate; one major, two minor (Figure 2.lb). The separation on silica gel with 4:1 hexanes:EtOAc was difficult but eventually accomplished with the three spots isolated from one another. In Zhu’s protocol, there was no mention of how the isolated products were verified to be the 2,3-0-isopropylidene (2.3), 3,4-0- isopropylidene (2.4), or any of the other possible side-products (2.5, 2.6 and 2.7). More experiments were therefore needed before moving on to the next step. As identified by LRMS, the top spot contained a mix of compounds 2.5 and 2.6 and the middle and bottom spots contained 2.3, 2.4 and a mix of 2.7 isomers (Figure 2. ib). This narrowed down the field but still did not provide any conclusive individual identifications. In separate NMR experiments, ring protons were assigned for the middle and bottom spots based on data obtained from 2D ‘H-NMR COSY experiments; however, no distinguishing features were identified. This is because1H-NMR shifts of protons adjacent to unprotected hydroxyl groups are expected to fall within the same range as those adjacent to isopropylidene protecting groups (3-4 ppm). As well, spin systems of the ring protons are isolated from those of the protecting group protons, rendering COSY correlations ambiguous for the distinction between 2,3-0 protection and 3,4-0 protection. The coupling constants are expected to be the same for each product due to identical stereochemistry; so with this technique alone there was no unequivocal way to tell the difference between each isolated spot. The method I used to identify each spot was as follows: A portion of each isolate was acetylated with pyridine and acetic anhydride. Without purification,1H-NMR and COSY data were obtained and assignments of ring protons were made. When comparing the spectra obtained before and after acetylation, one would expect to detect a proton 26 whose chemical shift had moved downfield following acetylation. This would correspond to the proton now adjacent to the electron-withdrawing acetyl group. The isolate which corresponded to the bottom spot showed a downfield shift for H4 after acetylation (2.8), indicating a free hydroxyl at that position prior to acetylation (Figure 2.2). This was the desired product and attainable in pure form (Figure 2.3). The middle spot contained more than one compound, but for the major component a downfield shift of the H2 peak was observed after acetylation, identifying it as the undesired compound 2.4. Yields for the desired product (2.3) eventually reached those reported by Zhu. Chemical Shift (ppm) Figure 2.2 1H-NMR spectra of ring protons before and after acetylation. Top black spectrum was measured before acetylation (2.3) and bottom red spectrum was measured after acetylation (2.8) with H4 peak denoted by arrows. 5.5 5.0 4.5 4.0 3.5 3.0 27 2.5+2.6 2.4 + 2.7 Figure 2.3 Representation of a TLC plate obtained from the isopropylidene reaction with product identity assigned to each spot. The installation of the nitrogen was accomplished through activation of the 4- hydroxyl in 2.3 by triflation followed by SN2 displacement with cyanide anion, inverting the stereochemistry at that centre to yield nitrile 2.9 (Scheme 2.1, page 24). Interestingly, this transformation also changes the molecule from an L-xylo configuration to a arabino configuration. Yields continued to go up as I became better at handling this sensitive reaction and also eventually reached the yields reported by Zhu. The final one-pot hydrogenation under mild acidic conditions, followed by strong acid treatment, to reach IFG (1.8) is a very elegant reaction (Scheme 2.2). Six different transformations take place, several under reversible conditions with the reactive species being one of the tautomeric forms: 1) Reductive removal of benzyl glycoside reveals a hemi-acetal (2.10) in equilibrium with the open chain aldehyde. 2) Reduction of the nitrile to a primary amine (2.11). 3) Nucleophilic attack by free primary amine onto the open chain form of the aldehyde yielding a hemi-aminal (2.12). 4) Imine (2.13) formation by expulsion of water under acidic conditions. 5) Reduction of imine to form secondary amine (2.14). 6) Acidic conditions hydrolyze isopropylidene and reveal free diol (1.8). 28 N III OBnt N H2N H2 H2 -. — OS’’Otoluene 0/HOH 2.11 OH H HO acetone 1.8 Scheme 2.2 Mechanisms and intermediates representing the transformations in the final step of the synthetic route to IFG (1.8). Taking clues from the literature, it was thought that purification would be difficult. Several lengthy purification protocols have been published for isolating pure 1.8 including, cation exchange chromatography (NH4 form), silica gel chromatography under conditions of 7:2:1 (i-PrOH:110:7M NH4OH), as well as size exclusion chromatography plus combinations thereof.39’81 Zhu reported just one purification step using cation exchange chromatography (NH4form) and this method was tried first. The basic idea behind this chromatographic method for separating amines is as follows: The amine-containing sample is loaded under acidic aqueous conditions. This ensures the amines are positively charged and bind to the negatively charged carboxylate moieties immobilized on the resin beads that make up the stationary phase. In theory, a water wash elutes all anions and neutral compounds leaving all positively charged compounds bound to the resin. Once all undesired compounds have been eluted, an NH- containing eluent (NH4OH) is applied in an increasing gradient to displace the 0 2.13 4, 112 2.12 2.14 0 29 positively charged amines according to binding affinity, with the weakest binders eluting first. Zhu reported elution of IFG (1.8) upon using 0.05M NH4O , whereas I started with this concentration and increased it at 0.1 M intervals until the product eluted at 0.35 M NH4O . Unfortunately it was not in pure form. It was necessary to treat fractions that contained the desired product with benzyl chioroformate in 2:1:1 THF:H20:MeO and sodium bicarbonate to install a carboxybenzyl (Cbz) group on the nitrogen atom (2.15) (Scheme 2.1). This allowed easier separation on silica gel with conditions of 1:1 petroleum ether:EtOAc followed by 9:1 CHC13:MeOH. After the pure, derivatized product (2.15) was in hand, a simple 2 hour hydrogenation under atmospheric pressure followed by filtration yielded pure IFG (1.8) (Scheme 2.1, page 24). While Zhu reported a yield of 81% for the one-pot hydrogenation under atmospheric pressure, I was obtaining impure yields of only 30% prior to derivatization. The reaction was conducted under rigorously dry conditions, as well as with no special consideration for dryness, and at several different pH values ranging from pH 2—6, all to no avail. In an effort to solve the problem of poor yields, the reaction mixture composition was investigated to look for side products and/or partially reacted products. A major component besides the desired product was identified as a hemi-acetal which also contained a nitrile moiety 2.10 (Figure 2.4 and Scheme 2.2). Several forms of evidence support this conclusion. The compound was eluted from a cation exchange column during the water wash; indicating the absence of an amine functionality, which is consistent with the expected elution profile of 2.10. As well, the compound was analyzed by ‘H-NMR and two anomeric proton shifts were observed, consistent with the a- and 13- anomers present when an aldose/hemi-acetal undergoes mutarotation at the anomeric centre. The compound was also acetylated and the resulting syrup was prepared as an IR sample. The peak observed at 2551 cm’ was indicative of the presence of a nitrile, further supporting the proposed structure. Finally, LRMS data were consistent with the 30 structure of 2.10. N III HOS’ 2.10 Figure 2.4 Structure of the side-product identified from the IFG reaction. Upon realization that poor yields stemmed from incomplete reductions, 50 psi of H2 was applied in a special apparatus for high pressure reactions. This was required in order for the hydrogenation to proceed at a decent rate. Simply using a catalyst (Pd(OH)2/C20%) obtained from Alfa Aesar as opposed to Sigma Aldrich also improved impure yields to 80% prior to derivatization. After the reaction was performed under high pressure, followed by cation exchange chromatography, Cbz derivatization, silica gel purification and Cbz removal, pure yields of 31% were reached for the transformation of the nitrile (2.9) to lEG (1.8) (Scheme 2.1, page 24). 2.1.2 Synthesis of C6-Alkyl IFG Derivatives 2.1.2.1 Synthesis of C6-n-Nonyl IFG In the same paper that outlined a four step synthesis of IFG, Zhu et. al. reported the only synthetic route that accesses C6-alkyl lEG derivatives of the general structure shown in Figure 2.5.38 HO Figure 2.5 General structure and numbering scheme for C6-alkyl IFG derivatives where R=alkyl. 31 Zhu reported a series of C6-n-butyl through C6-n-nonyl WG compounds synthesized via addition of the corresponding n-alkyl Grignard reagents to the nitnie intermediate (2.9) (Scheme 2.3). As the carbon atom in a nitrile moiety is electrophilic, and in 2.9 that carbon atom corresponds to the C6 position of IFG, nucleophilic addition of an alkyl grignard is a seemingly straightforward way to build up a series of C6- alkylated IFG derivatives, all of which have been shown to be strongly inhibitory towards GCase.38 N 1) H2, Pd(OH)2/C 1) CH3(C28MgBr H2N (CH2)8C3 AcOH, MeOH 2) NBH4,MeOH 2)_iIHC1 OBflf (43°’0) OBnfr (72%) 2.9 2.16 1.9 Scheme 2.3 Synthetic route to C6-n-alkyl llG derivatives. Shown here with C6-n-nonyl IFG (1.9). Conditions listed are adapted from Zhu and yields shown are those obtained in my hands. It was thought that reproducing the entire series would be unnecessary, therefore the most potent inhibitor of GCase (1.9, 1C50=0.6 nM) was chosen for synthetic reproduction. It was also desirable to synthesize 1.9 because the only confirmation of inhibitory ability for the four longest alkylated versions of the published series were IC50 values.38 Zhu cited mixed type inhibition as the reason why K values were not measured as well. We hypothesized that these compounds were indeed competitive inhibitors but that their inhibition assay was not sufficiently sensitive to handle sub-nanomolar inhibitors. It was hoped that the extensive kinetic evaluation done in our lab would allow the measurement of a true K1 value for 1.9. In the preparation of primary amine 2.16 (Scheme 2.3), the Grignard reaction required heating to 350 C and use of up to 5 equivalents of Grignard reagent in order to push the reaction to completion. In the second step, wherein the imino-magnesium complex (2.17) is reduced to the primary amine (2.16) by NaBH4,exclusive hydride 32 attack from behind (Re face) was observed (Scheme 2.4). This yielded only the S configured diastereomer as confirmed with data obtained from Nuclear Overhauser Effect (NOE) NMR experiments of the final product (1.9). Strong correlations were observed between axial protons H4 and H2 when H6 was irradiated. This mechanism of stereoselective hydride delivery was first proposed by Zhu and the proof was in the form of NOESY NMR data from the final product.38 Br-. C o J RMgBr I 0 0 OBn 2.9 Scheme 2.4 Proposed mechanisms for formation of amine 2.16. Crossed out red arrow indicates the unfavoured Si face attack. Green arrow indicates the favoured Re face attack yielding amine 2.16 in a stereoselective manner. R= C9H19. Adapted from Zhu et. al. Purification of the free amine (2.16) was accomplished on silica gel by using 20:1 CH21:MeOH with 0.1% Et3N. A substantial amount of nonane was eluted from the column with 2.16 and it took two columns to obtain 2.16 in pure form. After confirmation of only single Grignard addition by1H-NMR and LRMS, one pot cyclization in the high pressure reactor under 50 psi of H2 afforded the desired C6-n- nonyl IFG (1.9) (Scheme 2.3, page 32). The final product was purified via C-18 RP silica gel. Originally I thought that the column yielded two distinct products because one compound eluted in 30% MeOH in H20 and another in 60% MeOH in H2O. LRMS data revealed the same masses for both compounds and ‘H-NMR and COSY data revealed the same number of protons and two very similar, yet distinct, looking spectra. The compound which eluted in 30% MeOH matched the analytical data published by Zhu, but the identity of the second compound was still unknown. At first, it was thought that the OBn OBn 2.17 2.16 33 extra compound was the result of epimerization during the Grignard reaction. The proton a to the nitrile (H4 of 2.9) (Scheme 2.1, page 24) is slightly acidic with an estimated PKa of 25-30, lower than the estimated pKa of a terminal alkyl proton of nonane at 45-50. Nonyl magnesium bromide could react as a base by abstracting H4 rather than acting as a nucleophile. Upon quenching with MeOH, that centre might be reprotonated from either face yielding a mix of epimers at C4. However, when NOE experiments for both compounds were similar with respect to the stereocentre in question (C5 in IFG numbering, Figure 2.5, page 31), it was thought that the distinct ‘H-NMR spectra might be reflective of different protonation states of the amine. The compound which eluted in 60% MeOH was treated with 1 M HC1 and concentrated several times to ensure protonation. The1H-NMR of the HC1 treated version revealed an identical spectrum to that of the compound eluted in 30% MeOH, as well as to the data published by Zhu for 1.9. The extra compound was merely the free base version of C6-n-nonyl IFG (1.9). The C-18 RP column distinguished between these protonation states with surprising clarity. 2.1.2.2 Synthesis of C6-n-Propyl IFG and C6,6-Di-n-propyl IFG A trend was established within the K, and/or IC50 values for the series of compounds synthesized by Zhu whereby an n-alkyl chain of longer than four carbons was needed in order to observe stronger inhibition than that afforded by IFG (unalkylated).38 The planned synthetic route to establish a pH-labile linker diverged to also access C6-n- propyl IFG (2.17) and C6,6-di-n-propyl IFG (2.18) (Figure 2.6), neither of which had been published or tested by Zhu. It was desirable to see where 2.17 would fit within the trend as well as to probe the tolerance of GCase for C6 axial substituents, as in 2.18. 2.17 HCI 2.18 HCI Figure 2.6 Structures of products 2.17 and 2.18. 34 Starting with the nitriTe intermediate (2.9), allyl magnesium bromide was used to make the allyl amine (2.19) (Scheme 2.5). It was assumed that double Grignard addition would not happen since it had not been observed in the nonyl system, thus many equivalents of allyl Grignard were added in the first attempt, only later to discover that double addition does indeed occur in this system. This reaction was much faster than the nonyl Grignard reaction and required reduced temperatures upon addition of reagent in order to minimize the double addition. Purification was simpler than with the nonyl version because the quenched Grignard reagent yielded propene, a gas at room temperature, rather than nonane, with a boiling point of 151° C. Subsequent installation of a Cbz group onto the nitrogen atom using benzyl chloroformate in pyridine, CH21 and DMAP served to protect the amine (Scheme 2.5) for further functional group manipulation, and simplified purification on silica gel. After the protected amine (2.20) was obtained in very pure form, a batch was hydrogenated in MeOH under atmospheric pressure overnight to access C6-n-propyl IFG (2.17) (Scheme 2.5). No purification was necessary as all protecting groups that were removed are volatile including; two equivalents of toluene and one equivalent of acetone, C02 and H20. 35 1) H2,Pd/C10% (80%) AcOH, MeOH 2) 1NHCI HO 2.17 HCI Scheme 2.5 Synthetic route to C6-n-propyl IFG (2.17) and C6,6-di-n-propyl IFG (2.18). After the protected amine (2.20) was eluted from the column in 20:1 CH21:EtOAc, the column was flushed with neat EtOAc in hopes that some unprotected amine 2.19 would be recovered for re-reaction. However the actual compound flushed from the column was unexpected and corresponded to the double Grignard addition H2Ni N III C 2.9 OBnt 1)C3H5MgCI, (72%)I Et20 2) NaBH4,MeOH + (64%)! BnOCOCI, pyridine, DMAP + OBn 2.19 OBn 2.21 CbzHN1 2.20 (56%)! 1) H2,Pd/C10% AcOH, MeOH 2) INHC1 HO OH 2.18 HCI 36 product, di-allyl amine (2.21) (Scheme 2.5) as confirmed by MS. ‘H-NMR and COSY data. Apparently 2.21 was too hindered around the nitrogen to install the bulky Cbz group. This provided a very convenient way to separate the very chromatographically similar primary amines (2.19 and 2.21). Compound 2.21 was hydrogenated at atmospheric pressure overnight under acidic conditions to yield C6,6-di-n-propyl IFG (2.18) (Scheme 2.5), which was purified via C- 18 RP silica gel. Access to 2.18 provided the opportunity to investigate how an axial substituent at the C6 position of IFG affects GCase binding. It was desirable to know this in order to provide a basis for any further diversification of the C6-alkyl IFG derivatives. 2.1.2.3 Synthesis of C6-[9-Hydroxypropyl] IFG Using protected alkene 2.20 as starting material, another IFG derivative was accessible that could be used to probe the effect on GCase binding of a hydroxyl group within the alkyl chain. Addition of water across the double bond in an anti-Markonikov fashion under hydroboration — oxidation conditions yielded the primary alcohol 2.22 (Scheme 2.6). The other possible product, a secondary alcohol, was not observed. Upon purification on silica gel, with an increasing gradient of 10% to 25% EtOAc in CH21, 2.22 eluted with the isopropylidene group cleaved, presumably during the course of column purification. This protecting group was going to be removed in the next step anyway so it was unnecessary to re-install it. The next step was the final hydrogenation at atmospheric pressure overnight under acidic conditions to reach C6-[9-hydroxypropyl] IFG (2.23) (Scheme 2.6). This product was purified via cation exchange chromatography in the same manner as IFG (1.8) and eluted in 0.25 M NH4O . Treatment with HC1 ensured that all the compound rested in one protonation state, yielding pure 2.23. 37 CbzHN 1) BH3THF, THF 2) NaBO3 4HO (54%) 1) H2,PdIC1O% AcOH, MeOH HO 2) 1NHCI HCI 2.23 Scheme 2.6 Synthetic route to C6-[9-hydroxypropyl] IFG (2.23). 2.1.3 Synthesis of Acetal-Containing IFG Derivatives as pH-Labile Linkers 2.1.3.1 Synthesis of C6-Benzyl acetal IFG Many attempts were made to synthesize a C6 TFG derivative that contained an acetal moiety, which in theory would be pH labile. Ideally, the acetal would contain two alkyl or aryl arms (Figure 2.7a), which would mimic the hydrophobicity of ceramide and impart tight binding to GCase at neutral pH values (ER) while intact. Upon hydrolytic cleavage at acidic pH values (lysosome), a weaker GCase binder would be left, minimizing any inhibition of the delivered enzyme. After much trial and error, the only type of acetal-containing IFG derivative that could be synthesized and purified was of the general structure shown in Figure 2.7b. O—--- 2.7a OR HO 3 2.7b Figure 2.7 a) Ideal structure for pH-labile linker shown here with alkyl groups b) General structure of synthetically attainable pH-labile linker. Starting again with 2.20, ozonolysis of the alkene followed by reduction of the ozonide with triphenyiphosphine afforded the aldehyde 2.24 (Scheme 2.7) in good yield. Attempts to reduce the ozonide with dimethyl sulfide (DMS) resulted in cleavage of the isopropylidene group. This liberated the 3-hydroxyl group, which was situated six bonds 2.20 2.22 38 Scheme 2.7 Products of ozonolysis reaction followed by reduction with use of either PPh3 (2.24) or DMS (2.25). CI away from the carbonyl carbon of the aldehyde and easily formed the six-membered hemi-acetal (2.25) as shown in Scheme 2.7. 1) 03, CH21 (75%) 1) 03, CH2I 2) DMS OBn 2.24 2.20 OH 2.25 OBn It was discouraging to realize so late in the project that the molecule was set up for an undesired intra-molecular hemi-acetal formation upon deprotection, not only with 04 of the final product (2.26), but with 05’ as well (2.27) (Figure 2.8). OH 0 HO HO NH 2.26 2.27 Figure 2.8 Undesired intra-molecular hemi-acetal formation with 04 (2.26) and 05’ (2.27). HO 39 As was realized at the time, in order to access IFG derivatives of the general structure shown in Figure 2.7a (page 38), an inter-molecular reaction between an alcohol such as hexanol, and the aldehyde would have to proceed faster than, and be favoured over, an intra-molecular reaction. The fundamentals of reaction kinetics would suggest that this was impossible. The only hope would rest in use of a 1,3-propane diol derivative as the alcohol, which would force two intramolecular processes to compete for acetal formation (Scheme 2.8). OH OH p-TsOH, CHC13 + acetone Scheme 2.8 Two possible products (2.28 and 2.29) resulting from acetal formation with 2-n- alkyl-1,3-propane diol derivatives and aldehyde 2.24. A pure sample of aldehyde 2.24 was reacted with 2-n-butyl-1,3-propane diol and p-T5OH in CDC13 and the reaction monitored by1H-NMR every two minutes. Two regions of the1H-NMR spectra were helpfully diagnostic in monitoring this reaction. First, the aldehyde proton peak at 9.8 ppm disappeared as the hemi-acetal and acetal formed. Second, as the isopropylidene group cleaved to yield acetone, a peak at ö 2.05 ppm was observed. Other spectral regions contained multiple overlapping peaks that could not be used diagnostically. While the ‘H-NMR spectra did not allow product identification, it was apparent that equilibrium was established quickly because the spectra stopped changing after 12 minutes and remained the same after reaction overnight in the NMR tube. OBn 2.24 2.28 CbzHN1 -O 2.29 40 The reaction mixture was purified on silica gel with 1:1 hexanes:EtOAc, and a product with the correct mass was isolated. The ‘H-NMR spectrum of the isolated material was not helpful in distinguishing between the two possible products (2.28 and 2.29) (Scheme 2.8) so an attempt was made to chemically differentiate the products, if indeed a mixture was present in the column isolate. Only one of the two proposed products contains a primary hydroxyl group (2.29), which would react much faster with TBDMSCI than would the secondary hydroxyls of 2.28. This should alter the Rf value of the undesired product (2.29) and allow separation. However, when I attempted to perform the reaction, no new spots appeared on the TLC plate and the MS data did not change, yet the reagent was confirmed to be active by reaction with a model compound. As a result, it was concluded that the isolate contained only the desired product 2.28. Emboldened by this finding, the product 2.28 was subjected to standard hydrogenation conditions at atmospheric pressure, but this time, the pH was adjusted to 8 with 0.01 M NaOH to minimize any new acetal formation. Unfortunately the ‘H-NMR spectrum of the resulting product mixture was extremely complex and contained four triplet peaks between 64.5 and 4.7 ppm, all of equal intensity, presumably corresponding to the different stereochemistries of the acetal proton of each possible acetal product (Scheme 2.9). Apparently my attempts to suppress new acetal formation were unsuccessful and no useful approaches were found for separation of these products. H H2, Pd/C 10% 0.01 M NaOH, R MeOH 2.30 R HOOH 2.31 Scheme 2.9 Possible products (2.30 and 2.31) from hydrogenation of 2.28 resulting in complex ‘H-NMR spectra. Red bond indicates acetal proton observed as triplet peaks from 6 4.5 and 4.7 ppm. OBn 2.28 41 It was apparent that the goal of synthesizing molecules with the general structure shown in Figure 2.7a (page 38) was unattainable with this particular carbon skeleton. The carbon chain extending from C6 would need to be at least three carbons longer in order to suppress all undesired five and six-membered intramolecular reactions and cyclizations. Given the limited time remaining this was not feasible. Moving forward with what was available and known; it seemed that only a single hydrophobic substituent could be introduced to the acetal as shown in Figure 2.7b (page 38). After contemplation, it appeared that this type of structure might actually be beneficial for a PC. Due to the intramolecular hemi-acetal formation (Figure 2.8, page 39), liberation of a free aldehyde upon hydrolysis is avoided. This is desirable because a free aldehyde could possibly react with any lysine residues of GCase and form a Schiff base, which could have negative side-effects. The intra-molecular hemi-acetal formation (Figure 2.8) (2.26 and 2.27) upon acetal hydrolysis might mitigate these effects. Aldehyde 2.24 was hydrogenated under acidic conditions at atmospheric pressure and, without purification, the residue was stirred in benzyl alcohol and p-TsOH (Scheme 2.10). Due to limited reactant solubility, the reaction was continued for four days. The reaction mixture was neutralized with anion exchange resin OH form) and the protonated form of product 2.32 was extracted with aqueous ammonium acetate solution at pH 7. This was washed with petroleum ether to remove the excess high-boiling benzyl alcohol. The product was purified on C-18 RP silica gel and eluted in 30% MeOH in H20. The trans-decalin ring system rigidifies 2.32 and, from the coupling constant of 3.1 Hz between H7ax and H-8, it appears that only one anomer was formed wherein the —OR group is axial; presumably due to the anomeric effect. I) H2 Pd/C 10% OH O-% 2) HO TsOH, HO NH (46%) HO NH 2.27 2.32 Scheme 2.10 Synthetic route to C6-benzyl acetal IFG (2.32). 2.24 42 The stability of 2.32 was tested by dissolving an aliquot of the intact acetal in water, adjusting the pH to 2, 3, 4, 5, 6, and 7 and incubating each solution at 37°C. Aliquots at time points of two hours, four hours, 24 hours, and 5 days were taken and analyzed by LRMS. The data revealed that 2.32 is a very stable acetal indeed. Hydrolytic cleavage was not observed at any of the pH values for any time aliquots. Indeed hydrolysis within 2 days at 37°C required lowering the pH to 1. One factor contributing to the unusual stability of this acetal is that it is part of a trans-decalin system. In order for acetal cleavage to occur, the hybridization of the acetal carbon atom must go from sp3 to sp2, and this requires a flattening of the ring system. The rigid trans decalin system makes this unfavourable. This result and conclusion is also consistent with a study published which describes relative hydrolytic cleavage rates of acyclic and six-membered cyclic acetals.84 In general, there was an increased rate with an increase in the stability of the corresponding alkoxy carbenium ion intermediate. This translates into faster rates for compounds with a higher degree of substitution at the acetal carbon, and for acyclic acetals in comparison to their cyclic counterparts. More specifically, acyclic acetals that were closest in structure to 2.32 hydrolyzed some thousand-fold faster than the corresponding cyclic acetal closest in structure to 2.32. Therefore, it is not unreasonable to assume that if acyclic acetal-containing 1FG derivatives (Figure 2.7a, page 38) were accessible, their hydrolytic cleavage profiles would be more in line with the goals of designing a pH-labile linker. 2.1.3.2 Other Attempts to Synthesize Acetal-Containing IFG Derivatives In Zhu’s work, only n-alkyl Grignard reagents were used and at the beginning of this project, it was not apparent if this methodology could be applied with more complex Grignard reagents. The initial attempts to synthesize a pH-labile IFG derivative centered around incorporation of a pre-formed acetal by using a Grignard reagent such as 2.33 shown in Figure 2.9. After attempts to use a commercially available Grignard reagent had failed, I tried to synthesize it in the lab. Benzaldehyde was used as a model 43 electrophile in order to test whether an active Grignard reagent was being made from the bromide. In the model system it was confirmed that active Grignard reagent was being made however, inseparable mixtures resulted when 2.33 was reacted with nitrile 2.9 (Scheme 2.3, page 32). BrM 0 2.33 Figure 2.9 Structure of Grignard reagent (2.33). In an attempt to gain insight into the problems I was having with the Grignard reaction, I searched the literature for examples of Grignard reagent addition to nitriles followed by reduction to the amine; as opposed to hydrolysis to the ketone. The literature had a very scant representation of this type of reaction. A 1986 paper described tandem alkylation-reduction of nitriles. They tested various Grignard reagent additions to nitriles followed by reduction to the primary amine with anhydrous ammonia and lithium metal. The yields for these reactions were unreliable and ranged from 32%99%.85 In a follow- up paper they described copper (I) activated addition of Grignard reagents to nitriles with a little more success.86 However, when I replicated the procedure using copper (I) iodide with the cyclic acetal Grignard reagent (2.33) and nitrile (2.9), I observed no reaction. Continuing the literature search, I found another paper describing cerium (III) chloride- activated addition of Grignard reagents to various electrophiles including ketones, ci, 3- enones, esters, amides and nitriles.87 Of more than 100 reactions reported, the yields for addition to nitriles came in last; ranging from no reaction at all to 28% with 72% recovered starting material. Nevertheless, the protocol with anhydrous cerium chloride was followed with the cyclic acetal Grignard reagent (2.33) and nitrile (2.9), and only about 5% reaction was observed. Side reactions that resulted in inseparable mixtures in the absence of copper or cerium salts were clearly suppressed in their presence. Unfortunately the desired reaction was suppressed as well. The cyclic acetal Grignard reagent (2.33) was no longer pursued after these results. 44 The next strategy was to make a Grignard reagent with a protected alcohol (2.34) which could be oxidized at a later stage (Scheme 2.11). The synthesis of the Grignard precursor (2.35) was straightforward starting with 1,3-propane diol followed by mono- protection with TBDPSC1. Next was de-oxygenative iodination with triphenyiphosphine, imidazole and iodine. Unfortunately this route led to a dead end when the reagent simply dimerized (2.36) upon treatment with magnesium as shown in Scheme 2.11. TBDPSO’MgI ÷ 2.34 Et20 TBDPSO>) . TBDPSO...../ 3 TBDPSO I 2.35 2.36 Scheme 2.11 Product resulting from dimerization of Grignard precursor (2.36). Active Grignard (2.34) reacting with iodide (2.35) in SN2 fashion. What can be taken from the literature search and the less than ideal results from the Grignard reagent additions to nitrile (2.9), is that this is a difficult transformation and has limited applicability. In my experience the best results were obtained when using hydrocarbon and alkenyl Grignard reagents, which is also consistent with what has been published about this reaction in the literature. It was apparent that attempts to introduce a pre-formed acetal moiety into the molecule via a Grignard reaction were futile. The best strategy seemed to require the use of a simple alkenyl Grignard reagent followed by oxidation in order to develop the acetal linker afterwards. This lengthens the synthesis considerably due to the required protecting group manipulation, but it still provides the best route of those investigated towards C6-acetal containing IFG derivatives. 45 2.2 Inhibition Studies with Human GCase After the desired compounds of IFG and derivatives thereof (Figure 2.10) had been synthesized and characterized, inhibition studies with human GCase were performed. I started with inhibition studies because the Michaelis-Menten parameters have already been established for GCase with the substrate that was used, 2,4- dinitrophenyl 3-D-glucopyranoside (2,4-dNP-Glu).88 Within the inhibition data collected there was always a control with no inhibitor. These uninhibited data were used to generate the Km and Vmax values, which always matched with what has already been published for this system. 1.8 1.9 2.17 HOOH 2.23 HO Figure 2.10 Compounds tested as inhibitors of GCase. For all compounds except 1.9, a standard continuous UV-Vis spectrophotometric assay was performed in which GCase buffer, GCase and inhibitor were pre-incubated at 37°C for ten minutes. The reactions were initiated by addition of the substrate 2,4-dNP- Glu and the release of 2,4-dinitrophenolate was monitored at 400 nm. Steady state rates were measured in the linear region of each curve as Abs/mm readings. 2.18 2.32 46 The concentrations of substrate that were used were 0.5 mM, 1.1 mM, 3.0 mM and 5.0 mlvi. Each substrate concentration was tested with five or six different inhibitor concentrations based on the inhibitor’s estimated K, value, usually 1/3 K1 to 3 K1, plus data with no inhibitor. After the first series of inhibitor and substrate concentrations were assayed, the data were fit to several nonlinear regression models for different modes of inhibition in order to calculate K, values along with associated errors. For all inhibitors tested, the data fit best to the equation describing competitive inhibition as shown below. Vmax[S1 V = Equation 2.1 [S] + Km(l + [I]/K1) The first calculated K, value was often lower than expected from the 1C50 runs performed at a single substrate concentration. As a result, a second series was assayed in which the same substrate concentrations were used and inhibitor concentrations shifted to sandwich the true K, value. This also allowed duplicate measurements, and in all cases the second calculated K, value did not change considerably from the first. Special consideration with respect to the enzyme concentration needs to be exercised when the inhibitors assayed bind in the nanomolar and sub-nanomolar range. This is because the enzyme must be present at a concentration much less than those of substrate or inhibitor in order to fulfill all the assumptions of Michaelis-Menten kinetics and to thereby obtain useful data. In practice this means that inhibitor concentrations must be at least five times that of enzyme concentration, ideally ten times more. If very low enzyme concentrations are used, one needs to ensure that the spectrometer used is sensitive enough to measure a reliable rate. For measurements with lEG, I dropped the concentration of GCase to 0.9 nM so that this requirement was fulfilled and still managed to observe reliable rates. A GCase concentration of 2.2 nM was used for all the other inhibitors evaluated by the UV assay. Dixon plots were used to graphically represent the data and the theory describing this method, along with reversible competitive inhibition, is outlined in the Appendix. 47 ____ 0.5 mM 2,4- dNP-Glu 1.1 mM 2,4- —a-- dNP-Glu20 __ 3.0 mlvi 2,4-A 30 ‘I dNP-Glu 5.0 mlvi 2,4- 10 dNP-Glu 1/Vmax -40 -20 0 20 40 60 80 100 [IFG] nM Figure 2.11 Dixon plot for the inhibition of GCase by IFG (1.8). Zhu reported the K1 value for IFG (1.8) with GCase to be 25 nM whereas I measured it to be 23 nM ± 2, statistically the same. It was reassuring that this value agreed with what had been published.38 80 _ _ ____________ —0-- dNP-Glu60 _ 1.1 mlvi 2,4- dNP-Glu 40 ___________ ii’i)I_ 0.5 mM 2,4- 3.0 mlvi 2,4- -*- dNP-GIu 20 5.0 mM 2,4-A dNP-Gki ___________________________ 1/Vmax0 _ ililililili -4 -2 0 2 4 6 8 10 12 [C6-n-propyl IFG] p.M Figure 2.12 Dixon plot for the inhibition of GCase by C6-n-propyl IFG (2.17). 48 Although C6-n-propyl IFG (2.17) was a C6-n-alkyl IFG derivative of the kind Zhu had published, this particular derivative was not synthesized or investigated. It was desirable to see where 2.17 would fit within the trend observed by Zhu whereby five carbons or more were needed at C6 in order to observe stronger inhibition than that afforded by IFG (unailcylated). Compound 2.17 did fit within the trend with a measured K1 value of 0.61 jiM ± 0.075, almost six-fold higher than the butyl derivative at 0.12 1tM.38 40 20 0 Figure 2.13 Dixon plot for the inhibition of GCase by C6,6-di-n-propyl IFG (2.18). It was hypothesized that the tolerance of GCase for C6 axial substituents, such as in compound 2.18, would be poor. Surprisingly, this was not the case; the dipropyl compound (2.18) exhibited the same level of inhibition as that of 2.17 with a measured K, value of 0.61 jiM ± 0.10. This result is particularly interesting because it forms the basis for any future exploration of C6-axial IFG derivatives as it seems GCase has space in the active site to accommodate a C6-axial group of at least three carbons in length. • 0.5niM2,4- dNP-Glu 1.1 mM 2,4- dNP-Glu A 3.0 mM 2,4- dNP-GIu 50 mM 2,4- dNP-Glu lfVmax -6 -4 -2 0 2 4 6 8 10 [C6,6-di-n-propyl IFGJ JIM 49 The natural substrate hydrolyzed by GCase, glucosyl ceramide (GlcCer), is quite hydrophobic overall. However, it contains a secondary hyciroxyl group as well as an amide bond. Amongst the inhibitors tested to date, none of the C6-alkyl IFG derivatives have contained any hydrophilic moieties within their alkyl chain, such as the hydroxyl group in 2.23. It was not clear how this would affect GCase binding. When this compound was assayed as an inhibitor of GCase, the K value was measured to be 104 nM ± 19. This was lower than we had expected and perhaps suggests that a new hydrogen bond forms between a residue in GCase and the hydroxyl group. This certainly could account for the tighter binding of 2.23 relative to 2.17 or 2.18. 20 I 1/ - 10 _ r7 II -1 0 1 [C6-[9-hydroxypropyl] IFG] j.tM Figure 2.14 Dixon plot for the inhibition of GCase by C6-[9-hydroxypropylj FG (2.23). The data for the Dixon plot associated with 2.23 showed some signs of mixed- type inhibition because the lines do not all intersect at the l/Vmax line. However, when the raw data were fit to the various nonlinear regression models of inhibition including competitive, mixed type and non-competitive; the best fit was achieved with the competitive model. • 0.5mM2,4- dNP-Gk -a-.dNP-Gk A 3.OmM2,4- dNP-Glu —*--.dNP-Glu lNnnx 50 Previous studies performed in our group suggested that the 5’ primary hydroxyl group (equivalent to 06 in glucose) was particularly important for inhibitory power. When this hydroxyl group was replaced with a fluorine atom, there was a substantial decrease in the inhibitory ability of the compound compared to the parent compound, presumably due to the deletion of an important hydrogen bond. In compound 2.32, the intramolecular acetal involves the 5’ hydroxyl group, reducing the opportunities for hydrogen bonding. However, since the inclusion of a benzyl group might improve affinity, it was not clear what the overall effect would be. In fact, the K value was measured to be 160 nM ±6, still quite a good inhibitor. 40 ____ I ___ 20 _ 0 I -1 0 1 [C6-benzyl acetal IFG] j.tM Figure 2.15 Dixon plot for the inhibition of GCase by C6-benzyl acetal IFG (2.32). Due to the limited amount of 2.32 that remained, inhibition studies could not be performed on the cleaved version of the acetal, 2.27 (Figure 2.7). As it happens, it is unlikely that this version would cleave under physiological conditions anyway. So this was not deemed to be a crucial measurement. 51 Measurement of the K, value of C6-n-nonyl IFG (1.9) was rendered much more challenging by its anticipated sub nanomolar K1 value. As mentioned previously, the inhibitor concentration must be at least five times that of the enzyme concentration in order to obtain reliable data. In attempting to satisfy this requirement when assaying C6- n-nonyl IFG (1.9) with the continuous UV-Vis assay, very low enzyme concentrations had to be used and the rates observed were not reliable when substrate concentrations were low and inhibitor concentrations high. This indicated that the level of sensitivity of the UV-Vis spectrophotometer had been reached and that this assay could not be used to measure the K, value of 1.9. A more sensitive instrument, a fluorimeter, was used in order to investigate the inhibitory properties of 1.9. The idea behind a fluorescent assay is the same as with the UV assay with respect to rate measurement at various substrate and inhibitor concentrations, however, the practicalities are different. The fluorescent substrate, 4- methylumbelliferryl 3-D-glucopyranoside (4-MU-Glu) was used and the rate of release of 4-methylumbelliferrone was measured; much in the same way as the UV-Vis assay measures the release of 2,4-dinitrophenolate. Since GCase operates in the lysosome and has a pH optimum of around 5.5, this is the pH at which the inhibition is measured. The cleaved substrate however does not emit sufficient fluorescence at this pH and therefore, the pH must be raised above pH 10 in order to fully observe the fluorescence of the cleaved moiety. This requires the assay to be stopped rather than continuous. Buffered solutions containing the substrate were pre-incubated at 37°C in Eppendorf vials in the presence or absence of inhibitor (1.9). The reaction was initiated by addition of GCase to a total concentration of 5 pM, then at fixed time intervals of 3, 6 and 9 minutes, 100 iL aliquots were removed and diluted into a cuvette containing 500 iL of glycine buffer at pH 10.8. This both stopped the enzyme reaction and ionized the 4-methylumbelliferrone product, increasing its fluorescence to an observable intensity. Rates were calculated by linear regression of the fluorescence intensity measured at each time point with each fluorescence value being an average of 15 individual readings. The data were fit to the inhibition models in the same manner as with the other compounds, 52 and the K, value of 1.9 was measured to be 0.2 nM ± 0.01. This is a remarkably potent inhibitor and the strongest towards GCase measured to date. It also clearly exhibits competitive inhibition, laying to rest the claims made by Zhu that this is a mixed-type inhibitor.38 0.4 Putting all the K, values that I measured into context with what has been published in the literature suggests that C6-alkyl IFG derivatives are among the most potent inhibitors of GCase. Many of the compounds assayed thus far as inhibitors of GCase have had their inhibitory power evaluated by IC50 values, which are not directly comparable to K1 values, but can offer a rough comparison. Figure 2.17 shows several different iminosugars as well as nitrogen-containing heterocycles and their respective IC50 values towards GCase. Most of these previously published inhibitors are in the mid to low micromolar range whereas all the compounds I tested are in the sub micromolar to sub nanomolar range. I 0.2 0 -0.2 _____ 0.5 mM 4-MU- Glu 1.1mM 4-MU-Glu _ 3.0mM * 4-MU-Gin l/Vmax -0.4 -0.2 0 0.2 0.4 0.6 0.8 [C6-n-nonyl IFG] (nM) Figure 2.16 Dixon plot for the inhibition of GCase by C6-n-nonyl IFG (1.9). 1 53 N-butyl DNJ IC50 = 270 jiM HON x- I -C-nonyl DNJ IC50 = 0.27 jiM OH N-adamantyl amide IFG HO H 1C50=18j.tM -i-to N OH N-octyl-3-valienamine IC50 = 0.5 jiM 2 O%NH Thio-quinazoline 1C50=7.8j.tM Figure 2.17 Diagram of several inhibitors of GCase and their IC50 values. Sulfonamide value obtained from Zheng et. al. thio-quinazoline value obtained from Tropak et. at. , and all other values obtained from Butters.72 2.3 Conclusions In summary, I was successful in synthesizing five IFG derivatives in addition to IFG. Four of these derivatives were novel compounds (2.17, 2.18, 2.23 and 2.32). All the synthesized compounds were evaluated for their inhibitory ability towards OH N-butyl IFG IC50 =44jiM OH HO HN NTh( 0 0 Sulfonamide IC50 = 6.5 jiM 54 human GCase by measuring their K, values, which are summarized in Table 2.1 below. Some compounds have been sent, with the remainder to be sent shortly, to our collaborators Dr. Don Mahuran and Dr. Mike Tropak (Hospital for Sick Children, Toronto) in order to test their ability to behave as PCs in Gaucher cell lines. Inhibitor K, OH 23nM±2 1.8 OH 6lOnM±75 2.17 6lOnM+ 100 OH - 9? 2.18 OH lO4nM±19 2.23 0N_— l6OnM±6 HOHO NH 2.32 0.2nM±0.01 OH 1.9 Table 2.1 Summary of inhibitors synthesized and measured K1 values towards GCase. 55 3 Materials and Methods 3.1 Synthesis 3.1.1 General materials and methods All reagents were purchased from commercial suppliers (Sigma, Aldrich, Fluka, Alfa Aesar and Reike Metals) and were used without further purification, unless otherwise stated. Solvents used were either reagent, certified or spectral grade. Anhydrous/dry solvents were prepared as follows: CH21 and pyridine were distilled over CaH2;THF and diethyl ether were distilled over sodium and benzophenone; methanol was distilled over magnesium and iodine; DMF was dried over 4 A molecular sieves for 2 days prior to use. Deionized water, purified with a Millipore DirectQTM 5 Ultrapure Water system, was used for all aqueous solutions. Melting points were determined using a Laboratory Devices Mel-Temp II melting point apparatus and are uncorrected. Thin layer chromatography (TLC) was used to follow all reactions. TLC separations were performed using Merck Kieselgel silica gel 60 F254 aluminum-backed analytical plates. Compounds were detected using ultra violet light (where applicable) and/or stained with 10 % ammonium molybdate in 2 M H2S04(polyhydroxylated compounds), silica gel impregnated with iodine (general use), or 0.3 % ninhydrin and 3% acetic acid in ethanol (amines). All flash column chromatography was performed under elevated pressure on Sili-Cycle silica gel, 230-400 mesh. All reverse phase (RP) column chromatography was performed under elevated pressure using 2 g Waters Sep-Pak C-18 RP cartridges. All 1H nuclear magnetic resonance (NMR) spectra were either recorded on a Bruker AV-400 (400 MHz) or a Bruker WH-400 (400 MHz) spectrometer and chemical shifts are given in parts per million (ppm) as referenced from tetramethylsilane (TMS). Samples were referenced internally to CD3O at 3.31 ppm, CDC13 at 7.27 ppm, D20 at 56 4.78 ppm and acetone-d6at 2.05 ppm. Abbreviations describing the multiplicity of signals are: s-singlet, bs-broad singlet, d-doublet, t-triplet and m-multiplet. All 13C NIVIR spectra are proton decoupled and were recorded on either a Bruker AV-400 (100 MHz) or a Bruker WH-400 (100 MHz) spectrometer. Samples were referenced internally to CD3O at 49.15 ppm, CDC13 at 77.16 ppm, acetone-d at 29.8 ppm, and when D20 was used, CD3O was added as an external reference. Low resolution mass spectra (LRMS) were acquired on an electrospray ionization (ESI) Waters liquid chromatography — mass spectrometer (LC-MS) and high resolution mass spectra (HRMS) were acquired on an ESI Micromass LCT spectrometer by the mass spectrometry laboratory at the University of British Columbia. In cases where the desired compound has already been reported, all analytical data were identical to that already reported and referenced as such. 3.1.2 Generous Gifts Dr. Hong-Ming Chen in the Withers laboratory synthesized 2,4-dinitrophenyl 13- D-glucopyranoside. 3.1.3 (3R, 4R, 5R)-5-(Hydroxymethyl)piperidine-3,4-diol [Isofagomine. HC1 salt] (1.8) Benzyl x-L-xylopyranoside (2.2)38 1OH OBn Boron trifluoride etherate (1.5 mL, 11.8 mmol) was added to a suspension of L xylose (2.1) (17.9 g, 119 mmol) in benzyl alcohol (90 mL). The reaction mixture was stirred at 105 °C for 3 hours and at room temperature for 16 hours. All contents of the reaction flask were poured into an Erlenmeyer flask of cold diethyl ether (1.5 L) where a white precipitate formed immediately. Crystallization at 0 °C took 72 hours to complete. Crystals were harvested by suction filtration and re-crystallized from hot ethanol to yield 57 2.2 as white prisms (11.3 g, 47 mmol, 39%). 1H NMR (acetone — d6) 6 7.28-7.43 (m, 5 H, (Ph)), 4.83 (d, Ji,2 = 3.6 Hz, 1 H, H(1)), 4.75 (d, JphdH2=12.1 Hz, 1 H, (PhCH2)), 4.50 (d, JphdH2= 12.1 Hz, 1 H, (PhCH2)), 4.05 (dd, J5ax,5eq = 12.6 Hz,J54=3.4 Hz, 1 H, H(5eq)), 3.48-3.66 (m, 3 H, H(3), H(5ax), H(4)), 3.39 (m, 1 H, H(2)). ‘3C NMR (acetone — d6) 6 128.8, 129.0, 129.6, 99.9 (C(l)), 75.9, 74.1, 71.8, 70.2, 63.7. ESI MS m/z 263.2 [M + NaT. Calculated forC12H6NaO5263.1. Benzyi 2,3-O-isopropylidene-u-L-xylopyranoside (2.3)38 OH 0 0 OBn A solution of p-toluenesulfonic acid monohydrate (210 mg, 1.22 mmol) and THF (3 mL) was added by syringe to a solution of the benzyl xyloside (2.2) (10.5 g, 43.7 mmol) and 2-methoxypropene (10.5 mL, 110 mmol) in anhydrous THF (100 mL) at 0°C. The reaction mixture was quenched with triethylamine (0.7 mL) after 1.5 hours, then diluted with EtOAc (280 mL) and washed with ice water (2 x 200 mL) and saturated aqueous NaC1 (200 mL). The organic portion was dried over anhydrous Mg504,filtered and concentrated in vacuo. The residue was purified by flash column chromatography on silica gel (3.5:1 hexanes : EtOAc with 0.1% Et3N) to yield 2.3 as a clear syrup (5.99 g, 21.4 mmol, 49%). 1H NMR (CD3OD) 67.27-7.39 (m, 5 H, (Ph)), 5.16 (d,J12= 3.0 Hz, 1 H, H(1)), 4.78 (d, JPhCH2 = 12.0 Hz, 1 H, (PhCH2)), 4.60 (d, JPhCH2 = 12.0 Hz, 1 H, (PhCH2)), 3.94 (dd, J3,2 = J3,4 = 9.5 Hz, 1 H, H(3)), 3.86 (ddd, J4,3 =J4,5ax = 9.5 Hz, = 5.2 Hz, 1 H, H(4)), 3.58 (dd,J5=11.2 Hz,J5 = 5.2 Hz, 1 H, H(5eq)), 3.36 (dd, J2,3 = 9.5 Hz, J,j = 3.0 Hz, 1 H, H(2)), 3.31 (masked by MeOD, H(Sax)), 1.33 (s, 3 H, (CH3)), 1.32 (s, 3 H, (CH3)). ‘3C NMR (CD3OD) 6 129.5, 128.9, 111.6, 97.6 (C(1)), 81.7, 78.4 (C(3)), 77.4 (C(2)), 71.1 (C(4)), 70.7, 64.4 (C(5)), 27.3, 26.9. ESI MS m/z 303.3 [M + Na1. Calculated forC15H20NaO5303.1. 58 Benzyl 4-O-acetyl-2,3-O-isopropylidene-a-L-xylopyranoside (2.8) OAc 0 0 OBn Acetic anhydride (3 mL) was added by syringe to a solution of the alcohol (2.3) (40 mg, 0.14 mmol) and pyridine (3 mL) at 0 °C. The reaction mixture was quenched with cold water (8 mL) after 6 hours, extracted with CH21 (2 x 20 mL) and washed with saturated aqueous NaHCO3(2 x 15 mL), water (20 mL) and saturated aqueous NaC1 (20 mL). The organic portion was dried over anhydrous MgSO4,filtered and concentrated in vacuo to yield 2.8 as a clear syrup. ‘H NIVIR (CDC13)6 7.37-7.38 (m, 5 H, (Ph)), 5.24 (d, J1,2:= 3.1 Hz, 1 H, H(1)), 5.06 (ddd,J43=J4,5ax= 9.6 Hz, J4,5eq 5.5 Hz, 1 H, H(4)), 4.80 (d, JPhCH2 12.1 Hz, 1 H, (PhCH2)), 4.65 (d, JPhCH2 12.1 Hz, 1 H, (PhCH2)), 4.19 (t, J32 = J34 = 9.6 Hz, 1 H, H(3)), 3.92 (dd, J5eq,5ax 10.8 Hz, J5eq4 = 5.5 Hz, 1 H, H(Seq)), 3.56 (dd, J2,3 = 9.6 Hz, J2,1 = 3.1 Hz, 1 H, 11(2)), 3.34 (dd,J5ax, = 10.8 Hz,J5ax,4= 9.6 Hz, 1 H, H(Sax)), 2.09 (s, 3 H), 1.50 (s, 3 H, (CH3)), 1.47 (s, 3 H, (CH3)). Benzyl 4-cyano-4-deoxy-2,3-O-isopropy1idene--D-arabinopyranoside (2.9)38 N UI OBn( Tf20 (1.3 mL, 7.7 mmol) was added dropwise by syringe to a solution of the alcohol (2.3) (1.44 g, 5.14 mmol) and dry pyridine (2 mL) in anhydrous CH21(20 mL) at -78 °C under an inert atmosphere. The reaction mixture was warmed to 0 °C after addition was complete and stirred for an additional 2 hours. Upon warming, a dark red color persisted. The reaction mixture was then diluted with EtOAc (140 mL), washed with ice water (2 x 120 mL), saturated aqueous NaC1 (1 x 120 mL) and re-extracted with EtOAc (100 mL). The organic layer was dried over anhydrous MgSO4,filtered and concentrated in vacuo to yield a dark red syrup that was used immediately without further 59 purification. The mixture of the crude triflate, KCN (3.3 g, 50.8 mmol), 18-crown-6 (1.5 g) and 3 A MS (3 g) in dry DMF (110 mL) was stirred at room temperature for 16 hours. The reaction mixture was diluted with EtOAc (160 mL), washed with water (2 x 160 mL), saturated aqueous NaCJ (1 x 160 mL) and re-extracted with EtOAc (100 mL). The organic portions were dried over anhydrous MgSO4,filtered and concentrated in vacuo. The resulting residue was purified by flash column chromatography on silica gel (CH2C1 with 0.1 % Et3N) to yield 2.9 as a pale yellow syrup (1.18 g, 4.08 mmol, 80%). ‘H NMR (CDC13)ö 7.3 1-7.37 (m, 5 H, (Ph)), 5.37 (d, J,,2= 2.7 Hz, 1 H, H(1)), 4.78 (d, JPhCH2 12.0 Hz, 1 H, (PhCH2)), 4.68 (d, JPhCH2 12.0 Hz, 1 H, (PhCH9)), 4.13 (dd,J32= 9.6 Hz, J3,4= 4.7 Hz, 1 H, H(3)), 3.89-3.96 (m, 2 H, H(2), H(5a)), 3.79 (dd,J5b,a 12.1 Hz, J5b,4 = 2.6 Hz, 1 H, H(5b)), 3.28-3.35 (m, 1 H, H(4)), 1.51 (s, 3 H, (CH3)), 1.50 (s, 3 H, (CH3)). ‘3C NMR (CDC13)ö 136.8, 128.3, 127.8, 127.5, 116.9 (C(CN)), 110.8, 97.5 (C(1)), 74.4 (C(2)), 70.0, 69.3 (C(3)), 59.1 (C(S)), 34.0 (C(4)), 26.5, 26.4. ESI MS m/z 312.2 [M + NaTt. Calculated forC,6H19NNaO4312.1. (3R, 4R, 5R)-N-Benzyloxycarbonyl-5-(hydroxymethyl)piperidine-3,4-dioi (2.15) HO-\ HO.L%I.III..NCbZ 2 1 Pd(OH)2/C(20% , 60 mg) was added to a solution of the nitrile (2.9) (150 mg, 0.52 mmol) and glacial acetic acid (8 pL) in dry methanol (6 mL) and the mixture was hydrogenated in a high pressure reactor at 50 psi for 72 hours. The catalyst was removed by suction filtration through Celite then the filter cake was washed with methanol (30 mL) and acetone (30 mL). The filtrates were concentrated in vacuo, dissolved in 1 M HCJ (20 mL) and stirred for 12 hours. The solution was concentrated in vacuo and partially purified by ion exchange chromatography (Amberlite CG-50 type I, NH4form) using aqueous 0.35 M NH4O as an eluent after the water wash. The fractions containing the desired product were concentrated in vacuo to yield impure isofagomine (1.8) as a pale yellow crystalline solid (68 mg, 0.46 mmol, 89%). Benzyl chioroformate (0.12 mL, 0.84 mmol) was then added to a solution of isofagomine (1.8) (90 mg, 0.61 60 mmol) and NaHCO3(156 mg, 1.86 mmol) inH2OIMeOH/THF (2:1:1, 10 mL) and the mixture was stirred at room temperature for 20 hours. The reaction was quenched by adding 1 M HC1 (1 mL), concentrated in vacuo, co-evaporating with toluene. The residue was then purified by flash column chromatography on silica gel (petroleum ether:EtOAc 1:1 then MeOH:CHC131:9) to yield (2.15) as a clear oil (59 mg, 0.21 mmol, 35%). 1H NMR (MeOD)6 7.28-7.39 (m, 5 H, (Ph)), 5.12 (s, 2 H, (COOCHPh)), 4.13- 4.31 (m, 2 H, H(6a), H(2a)), 3.82 (dd, J = 10.9 Hz, J 3.5 Hz, 1 H, H(5’a)), 3.46-3.64 (m, 1 H, H(5’b)), 3.32-3.38 (m, 1 H, H(3)), 3.25 (dd, J = 11.0 Hz, J = 9.0 Hz, 1 H, H(4)), 2.50-2.79 (m, 2 H, H(6b), H(2b)), 1.55-1.67 (m, 1 H, H(5)). 13C NMR (MeOD) 157.2, 138.4, 129.9, 129.5, 129.2, 75.9, 72.9, 68.8, 62.3, 49.8, 46.9, 46.2, 46.0. HRMS (ESI) m/z 304.1167 [M + Na]. Calculated forC14H9NNaO5304.1161. (3R, 4R, 5R)-5-(Hydroxymethyl)piperidine-3,4-diol [Isofagomine.HCI salt] (1.8) 4 6 HO HO NH2 1 •HCI PdJC (10%, 10 mg) was added to a solution of IFG-Cbz (2.15) (28 mg, 0.10 mmol) and glacial acetic acid (8 iL) in dry methanol (6 mL) and the mixture was hydrogenated at atmospheric pressure for 10 hours. The catalyst was removed by suction filtration through Celite then the filter cake was washed with methanol (30 mL) and water (30 mL). The filtrates were concentrated in vacuo, dissolved in 1 M HC1 (5 mL) and stirred for 12 hours. The solution was concentrated in vacuo and then lyophilized from water (2 mL) to yield 1.8 as a pale yellow solid (13 mg, 0.46 mmol, 89%). Analytical data matches that reported in Zhu et. al.38 1H NMR (D20) 8 3.85 (dd, J5’a, 5’b = 11.7 Hz, = 3.5 Hz, 1 H, H(5’a)), 3.73-3.82 (m, 2 H, H(5’b), H(3)), 3.50-3.59 (m, 3 H, H(2eq), H(4), H(6eq)), 2.85-3.04 (m, 2 H, H(6ax), H(2ax)), 1.92-2.04 (m, 1 H, H(5)). l3 NMR (MeOD) 672.3 (C(4)), 69.7 (C(3)), 60.1 (C(5’)), 47.8, 46.0 (C(2), C(6)), 42.3 (C(5)). HRMS (ESI) m/z 148.0972 [M + H]. Calculated forC6H14N03148.0974. 61 3.1.4 (3R, 4R, 5R, 6S)-6-n-Nonyl-5-(hydroxymethyl)piperidine-3,4-diol [C6-n-Nonyl IFG•HC1 salt] (1.9) Benzyl 4-[(S)-1 ‘-aminodecyl]-4-deoxy-2,3-O-isopropylidene-p-D-arabinopyranoside (2.16)38 OBn n-C9H,MgBr (3.4 mL of 1.0 M solution in diethyl ether) was added dropwise by syringe to a solution of the nitrile (2.9) (260 mg, 0.9 mmol) in dry diethyl ether (35 mL) at -78°C under an inert atmosphere and then slowly warmed to room temperature and stirred for 4 hours. The reaction mixture was heated at 35°C for 1.5 hours then cooled to room temperature before NaBH4 (200 mg, 5.3 mmol) was added in one portion. The reaction mixture was then cooled to 0°C and dry methanol (13 mL) was added by syringe for 5 minutes, then allowed to warm to room temperature and stirred overnight. Water (20 mL) and diethyl ether (20 mL) were added to the reaction mixture and the solids removed by suction filtration then washed with diethyl ether (15 mL). The filtrate was washed with water (2 x 20 mL), saturated aqueous NaCl (20 mL) and re-extracted from the aqueous portion with diethyl ether (20 mL). The organic portions were dried over anhydrous MgSO4,filtered and concentrated in vacuo. The resulting residue was purified by flash column chromatography on silica gel (20:1 CH21:MeOH with 0.1% Et3N) to yield 2.16 as a pale yellow syrup (163 mg, 0.39 mmol, 43%). ‘H NMR (CDC13)ö 7.19- 7.29 (m, 5 H, (Ph)), 5.17 (d,J12= 2.8 Hz, 1 H, H(1)), 4.68 (d, JphdH2= 12.2 Hz, 1 H, (PhCH2)), 4.54 (d, JPhCm = 12.2 Hz, 1 H, (PhCH2)), 4.25 (dd, J3,2 = 9.9 Hz, J3,4 = 4.6 Hz, 1 H, H(3)), 3.79 (dd, J23 = 9.9 Hz, J21 = 2.8 Hz, 1 H, H(2)), 3.57-3.67 (m, 2 H, H(5a), H(5b)), 3.13-3.22 (m, 1 H, H(1’)), 2.25-2.45 (bs, 2 H, (NH2)), 1.95-2.05 (m, 1 H, H(4)), 1.54-1.57 (m, 1 H, H(2’a)), 1.39 (s, 6 H, (2 x CH3)), 1.10-1.27 (m, 15 H, H(2’b-9’)), 0.81 (t, Jio’,9’a= J,O’,9’b= 6.3 Hz, 3 H, H(10’)). 13C NMR (CDC13)6 137.2, 128.0, 127.3, 127.2, 62 108.8, 97.2 (C(1)), 74.0 (C(3)), 72.4 (C(2)), 69.2, 59.5 (C(5)), 47.9 (C(1’)), 46.7 (C(4)), 34.7, 31.6, 29.5, 29.3, 29.2, 29.0, 26.7, 26.0, 25.0, 22.3, 13.8 (C(10’)). ESI MS m/z 420.3 [M + H]. Calculated forC25H42N0420.3. (3R, 4R, 5R, 6S)-6-n-Nonyl-5-(hydroxymethyl)piperidine-3,4-diol [C6-n -Nonyl IFG• HC1 salt] (1.9•HCI)38 3 1 •HCI Pd(OH)2/C(20%, 80 mg) was added to a solution of amine (2.16) (160 mg, 0.38 mmol) and glacial acetic acid (8 pL) in dry methanol (20 mL) and the mixture was hydrogenated in a high pressure reactor at 50 psi for 20 hours. The catalyst was removed by suction filtration through Celite and the filter cake washed with methanol (30 mL). The filtrate was concentrated in vacuo, dissolved in 1 M HC1 (25 mL), and stirred at room temperature for 12 hours. The solution was then concentrated in vacuo and purified by colunm chromatography using C-18 RP silica gel, eluted with 30 % methanol in water for the HC1 salt product and 60 % methanol in water for the neutral product. Each set of fractions was then lyophilized from water (15 mL) to yield a white foam of 1.9•HC1 (40 mg, 0.15 mmol) and 1.9 (35 mg, 0.13 mmol) with a combined yield of 72 %. 1H NMR (CD3OD) 4.02 (dd,J5’a,b 11.4 Hz,J5’a, 2.4 Hz, 1 H, H(5’a)), 3.64-3.70 (m, 2 H, H(5’b), H(3)), 3.54 (t, J4,3 J4,5 = 9.6 Hz, 1 H, H(4)), 3.35 (dd, J2eq,2ax = 12.2 Hz, J2eq,3 = 4.9 Hz, 1 H, H(2eq)), 3.25 (ddd, J = 11.0 Hz, J = 7.6 Hz, J = 3.4 Hz, 1 H, H(6)), 2.79 (t, J2ax,2eq J2ax,3 12.2 Hz, 1 H, H(2ax)), 1.88-1.98 (m, 1 H, H(7a)), 1.59-1.70 (m, 1 H, H(7b)), 1.20-1.59 (m, 15 H, H(5), H(8aIb), H(9-14)), 0.90 (t,J154a= J15,14b 6.9 Hz, 3 H, H(15)). ‘3C NMR (CD3OD) ö 72.2 (C(4)), 70.3 (C(3)), 58.2 (C(6)), 57.5 (C(2)), 48.1 (C(5’)), 47.1 (C(5)), 33.2, 31.4, 30.8, 30.7, 30.6, 30.5, 25.8, 23.9, 14.6 (C(15)). HRMS (ESI) in/z 274.2389 [M + H]. Calculated forC15H32N0274.2382. 63 (3R, 4R, 5S, 6S)-6-n-Nonyl-5-(hydroxymethyl)piperidine-3,4-diol [C6-n-Nonyl LEG] (1.9) HO% ‘H NMR (CD3OD) ö 3.93 (dd,J5’a,=2.7 Hz,J5’a,b 11.2 Hz, 1H, H(5’a)), 3.69 (dd, .15’b,5’a= 11.2 Hz, J5’b,5= 3.3 Hz 1H, H(5’b)), 3.36-3.46 (m, 2 H, H(4), H(3)), 3.09 (dd, J2eq,2ax = 11.3 Hz, J2eq, = 4.0 Hz 1 H, H(2eq)), 2.56-2.64 (m, 1 H, H(6)), 2.40 (t, J2ax,2eq=J2ax,3= 11.3 Hz, 1 H, H(2ax)), 1.69-1.8 (m, 1 H, H(7)), 1.10-1.52 (m, 16 H, H(5), H(7), H(814)), 0.90 (t,J154a= J,5,14b= 6.8 Hz, 3 H, H(15)). ‘3C NMR (CD3OD) ö 75.2 (C(4)), 74.0 (C(3)), 59.4 (C(5’)), 57.1 (C(6)), 51.5 (C(2)), 50.3 (C(5)), 34.1, 33.4, 31.5, 31.0. 30.8, 30.6, 26.7, 24.1, 14.8 (C(15)). 3.1.5 (3R, 4R, 5R, 6S)-6-n-Propyl-5-(hydroxymethyl)piperidine-3,4-diol [C6-n- Propyl IFG• HCI salt] (2.17) Benzyl 44(S)-i’ -amino-3’ -butenyl]-4-deoxy-2,3-O-isopropy1idene--D- arabinopyranoside (2.19) H2N 0 0 0 OBn Allyl magnesium chloride (0.63 mL of 2.0 M solution in THF) was added dropwise by syringe to a solution of the nitrile (2.9) (221 mg, 0.76 mmol) in dry diethyl ether (15 mL) at 0°C under an inert atmosphere and then slowly warmed to room temperature and stirred for 3 hours. The reaction mixture was cooled to 0°C, NaBH4 (140 mg, 3.6 mmol) was added in one portion and then dry methanol (4 mL) was added by syringe over 5 minutes. The reaction mixture was warmed to room temperature after the addition was complete and allowed to stir overnight. Water (8 mL) and diethyl ether (8 mL) were added to the reaction mixture, and then the solids were removed by suction 64 filtration and washed with diethyl ether (10 mL). The filtrate was washed with saturated aqueous NaHCO3(15 mL) then extracted with diethyl ether (2 x 20 mL) and EtOAc (20 mL). The organic portions were dried over anhydrous MgSO4,filtered and concentrated in vacuo. The residue was purified by flash column chromatography on silica gel (gradient of 5% to 8% to 10% EtOAc in CH21 with constant 0.1% Et3N) to yield 2.19 as a pale yellow oil (193 mg, 0.58 mmol, 72 %). 1H NMR (CDC13)ö 7.28-7.39 (m, 5 H, (Ph)), 5.78-5.91 (m, 1 H, H(3’)), 5.26 (d,J12= 3.1 Hz, I H, 11(1)), 5.07-5.20 (m, 2 H, H(4’alb)), 4.76 (d, JphdH2= 12.1 Hz, 1 H, (PhCH2)), 4.62 (d, JPhCH2 12.1 Hz, 1 H, H(PhCH2)),4.33 (dd,J3,= 9.9 Hz,J34= 4.8 Hz,1 H, H(3)), 3.88 (d,J23= 9.9 Hz, J2j = 3.1 Hz, 1 H, H(2)), 3.76 (dd,J5a,b 12.4 Hz,J5a,4 1.0 Hz, 1 H, H(5a)), 3.69 (dd,J5b,a= 12.4 Hz, J5b,4= 3.0 Hz, 1 H, H(5b)), 3.35 (ddd, Jl’ ‘= J1’,2’b= 9.1 Hz, J1’,4= 3.0 Hz, 1 H, H(1’)), 2.44-2.53 (m, 1 H, H(2’a)), 1.84-2.16 (m, 4 H, H(2’b), H(4), (NH2)), 1.47 (s, 6 H, (2 x CH3)). 13C NMR (CDC13)ö 137.4, 134.7 (C3’)), 128.3, 127.6, 127.4, 118.0 (C(4’)), 109.0, 97.5 (C(1)), 74.1 (C(3)), 72.7 (C(2)), 69.4 (C(PhCH2)),59.7 (C(5)), 47.4 (C(l’)), 46.5 (C(4)), 39.5 (C(2’)), 26.8 (C(CH3),26.3 (C(CH). HRMS (ESI) mlz 334.2014 [M + H]. Calculated forC19H28N04334.2018. Benzyl 4-[[(S)-1 ‘-N-Benzyloxycarbonyl]-3’ -butenyl]-4-deoxy-2,3-O-isopropylidene- -D-arabinopyranoside (2.20) Benzyl chloroformate (121 L, 0.86 mrnol) was added to a solution of the amino olefin (2.19) (150 mg, 0.45 mmol) in pyridine (2.5 mL), CH21 (2.5 mL) and DMAP (6 mg, 45 imol) at room temperature and allowed to stir for 60 hours. The reaction mixture was concentrated in vacuo, dissolved in CH21 (20 mL), washed with water (20 mL), and re-extracted with CH21 (2 x 20 mL). The organic portions were combined and washed with saturated aqueous NaCT (20 mL), dried over anhydrous MgSO4and concentrated in vacuo. The residue was purified by flash column chromatography on 65 silica gel (20:1 CH21:EtOAc with 0.1% Et3N) to yield 2.20 as a white crystalline solid (m.p. 115-117°C) (134 mg, 0.28 mmol, 64%). 1H NMR (CDC13) 7.27-7.40 (m, 5 H, (Ph)), 5.72-5.87 (m, 1 H, H(3’)), 5.45-5.59 (bs, 1 H, (NH)), 5.26 (d, J,,2= 2.9 Hz, 1 H, H(1)), 5.03-5.17 (m, 4 H, H(4’a), H(4’b), (COOCH2Ph)), 4.76 (d, JPhCH2 12.1 Hz, 1 H, (PhCH,)), 4.63 (d, JphdH2= 12.1 Hz, 1 H, (PhCH2)), 4.27 (dd,J32= 10.0 Hz,J34= 4.3 Hz 1 H, H(3)), 4.11-4.23 (m, 1 H, H(1’)), 3.88 (dd, J23 = 10.0 Hz, J21 = 2.9 Hz, 1 H, H(2)), 3.71 (dd,J5a,b = 12.3 Hz,J5a,4 = 2.7 Hz, 1 H, H(5a)), 3.63 (d,J5b,a= 12.3 Hz, 1 H, H(5b)), 2.61-2.70 (m, 1 H, H(2’a)), 2.19-2.36 (m, 2 H, H(4), H(2’b)), 1.42 (s, 3 H, (CH3)), 1.32 (s, 3 H, (CH3)). 13C NMR (CDC13) 133.5 (C(3’)), 128.6, 128.5, 128.2, 128.1, 127.8, 127.6, 118.6 (C(4’)), 114.6, 109.7, 97.9 (C(1)), 74.1 (C(3)), 72.6 C(2)), 69.8, 66.6, 60.4 (C(S)), 47.8 (Cl’)), 42.0 (C(4)), 36.1 (C(2’)), 26.7 (C(CH3)), 26.5 (C(CH3)). HRMS (ESI) m/z 490.2205 [M + Na]. Calculated for27H33NNaO 490.2206 (3R, 4R, 5R, 6S)-6-n-Propyl-5-(hydroxymethyl)piperidine-3,4-diol [C6-n-Propyl IFG•HC1 salt] (2.17) HO HCI Pd/C (10%, 20 mg) was added to a solution of the terminal olefin (2.20) (50 mg, 0.15 mmol) and glacial acetic acid (8 jiL) in dry methanol (3 mL) and the mixture was hydrogenated at atmospheric pressure for 16 hours. The catalyst was removed by suction filtration through Celite and the filter cake washed with methanol (30 mL). The filtrate was concentrated in vacuo, dissolved in 1 M HC1 (8 mL), and stirred at room temperature for 12 hours. The solution was concentrated in vacuo and then lyophilized from water (2 mL) to yield 2.17•HC1 as a pale yellow foam (26 mg, 0.12 mmol, 80%). ‘H NMR (020) ö 3.98 (dd, J5’a, 5Th = 12.3 Hz,J5’a, = 2.3 Hz, 1 H, H(5’a)), 3.70-3.80 (m, 2H, H(S’b), 11(3)), 3.59 (t, J4, = J4,3 = 10.0 Hz, 1 H, H(4)), 3.46 (dd, J2eq, 2ax 12.0 Hz, J2eq, 3 = 5.1 Hz, 1 H, H(2eq)), 3.27 (ddd, J = 11.3 Hz, J = 8.0 Hz, J6,7a 3.3 Hz, 1 H, H(6)), 2.86 (t, J2ax,3= J2ax,2eq = 12.0 Hz, 1 H, H(2eq)), 1.89 (dddd, J7a, 7b 15.5 Hz, J7a, 8a= 10.7 Hz, J7a, 66 8b= 5.2 Hz, J7a,6 3.3 Hz, 1 H, H(7a)), 1.58-1.69 (m, 2 H, H(7b), H(5)), 1.28-1.51 (m, 2 H, H(8aJb)), 0.93 (t, J9, 8a= J9,8b= 7.4 Hz, 3 H, H(9)). ‘3C NMR (D20) 71.5 (C(4)), 69.5 (C3)), 57.4 (C(6)), 57.1 (C(5’)), 47.1 (C(2)), 46.0 (C(5)), 32.6 (C(7)), 18.3 (C(8)), 14.1 (C(9)). HRMS (ESI) m/z 190. 1441 [M + H]. Calculated forC9H20N3190. 1443. 3.1.6 (3R, 4R, 5R)-6,6-Di-n-propyl-5-(hydroxymethyl)piperidine-3,4-diol [C6,6-Di- n-propyl IFG. HC1 salt] (2.18) 9 During the preparation of Cbz-protected terminal olefin (2.20), an additional product was accessed in the purification. After the desired single addition product (2.19) was eluted in 20:1 CH21:EtOAc, the polarity of the eluent was increased to neat EtOAc, then the unprotected di-allyl amine (2.21) was eluted. Without further purification, Pd/C (10%, 20 mg) was added to a solution of the di-allyl amine (2.21) (40 mg, 0.11 mmol) and glacial acetic acid (8 iL) in dry methanol (3 mL) and the mixture was hydrogenated at atmospheric pressure for 16 hours. The catalyst was removed by suction filtration through Celite and the filter cake washed with methanol (30 mL). The filtrate was concentrated in vacuo, dissolved in 1 M HC1 (25 mL), and stirred at room temperature for 12 hours. The solution was then concentrated in vacuo and purified by column chromatography using C-18 RP silica gel, eluted with 5 % methanol in water. The fractions containing the desired product were pooled and lyophilized from water (3 mL) to yield 2.18•HC1 as a pale yellow syrup (16 mg, 0.06 mmol, 56%). ‘H NMR (D20) 3.8-3.95 (m, 3 H, H(4), H(5’a), H(5’b)), 3.73-3.79 (m, 1 H, H(3)), 3.36 (dd,J2ax= 12.3 Hz, J, = 5.4 Hz, 1 H, H(2eq)), 3.02 (t, J2ax, 2eq =J2ax,3= 12.3 Hz, 1 H, H(2ax)), 1.64- 1.97 (m, 5 H, H(5), H(7aIb), H(7’alb)), 1.18-1.49 (m, 4 H, H(8aIb), H(8’alb)), 0.95 (t, J = 7.1 Hz, 2 x CH3,H(9), H(9’)). 13C NIVIR (D20) 71.4 (C(4)), 69.86 (C(3)), 64.8 (C(6)), 59.5 (C(5’)), 48.2 (C(5)), 43.0 (C(2)), 38.2 (C(7)), 34.0 (C(7’)), 16.2 (C(8)), 16.0 (C(8’)), H • HCI 67 14.5 (C(9)), 14.4 (C(9’)). HRMS (ESI) m/z 254.1738 [M + H]. Calculated for C12H25N03254.1732. 3.1.7 (3R, 4R, 5R, 6S)-6-[9-Hydroxypropyl]-5-(hydroxymethyl)piperidine-3,4-diol [C6-[9-hydroxypropyl] IFG•HC1 salt] (2.23) Benzyl 4-[[(S)-1 ‘-N-Benzyloxycarbonyl]-butan-4’-ol]-4-deoxy-2,3-O-isopropylidene- P-D-arabinopyranoside (2.22) OBn BH3THF (130 iL of a 1.0 M solution in THF) was added to a solution of the terminal olefin (2.20) (27 mg, 57.8 tmol) in dry THF (2.5 mL) at 0°C under an inert atmosphere. After the addition was complete the reaction mixture was allowed to warm to room temperature and stir for 3 hours or until the starting material disappeared (by TLC). Keeping the reaction mixture at 25°C, water (1.5 mL) was added dropwise. Once the hydrogen stopped evolving, NaBO3 4HO (15 mg, 97.4 imo1) was added in one portion and the reaction mixture stirred overnight at room temperature. The mixture was concentrated in vacuo, dissolved in diethyl ether (15 mL), washed with water (15 mL), re-extracted with diethyl ether (3 x 15 mL) and finally washed with saturated aqueous NaCl (15 mL). The organic portions were dried over anhydrous MgSO4and concentrated in vacuo. The residue was purified by flash column chromatography on silica gel (gradient of 10% to 25% to 50% EtOAc in CH21 with constant 0.1% Et3N) to yield 2.22 as a clear syrup (15 mg, 31 itmo1, 54%). 1H NMR (CDC13)ö 7.27-7.41 (m, 5 H, (Ph)), 5.29 (bs, 1 H, (NH)), 5.27 (d,J12= 3.0 Hz, 1 H, H(1)), 5.01-5. 15 (m, 2 H, (COOCH2Ph)), 4.76 (d, JPhCH2 11.8 Hz, 1 H, (PhCH2)), 4.64 (d, JPhdH2= 11.8 Hz, 1 H, (PhCH2)), 4.27 (dd, J= 10.1 Hz,J34=4.2 Hz, 1 H, H(3)), 4.06-4.17 (m, 1 H, H(1’)), 3.82-3.9 1 (m, 1 H, H(2)), 3.56-3.76 (m, 4 H, H(5a), H(5b), H(4’a), H(4’b)), 2.16-2.25 (m, 68 1 H, H(4)), 1.81-1.92 (m, 1 H, H(2’a)), 1.52-1.70 (m, 3 H, H(3’a), H(3’b), H(2’b)). HRMS (ESI) m/z 468.2000 [M + Na]. Calculated forC2431NNaO7468.1998. (3R, 4R, 5R, 6S)-6-[9-Hydroxypropyl]-5-(hydroxymethyl)piperidine-3,4-diol [C6-[9- hydroxypropyl] IFG HC1 salt] (2.23) HOOH HCI Pd/C (10%, 10 mg) was added to a solution of the terminal alcohol (2.22) (33 mg, 0.07 mmol) and glacial acetic acid (8 j.tL) in methanol (3 mL) and the mixture was hydrogenated at atmospheric pressure for 16 hours. The catalyst was removed by suction filtration through Celite and the filter cake washed with methanol (30 mL). The filtrate was concentrated in vacuo, dissolved in 1 M HC1 (25 mL), and stirred at room temperature for 12 hours. The solution was then concentrated in vacuo and purified by ion exchange chromatography (Amberlite CG-50 type I, NH4form) using aqueous 0.3 M NH4O as an eluent after the water wash. The fractions containing the desired product were concentrated in vacuo, treated with 0.01 M HC1 (3 mL) and lyophilized to yield 2.23•HC1 as a pale yellow syrup (7 mg, 0.03 mmol, 43%). 1H NMR (D20) 4.01 (dd, J5’a, 5’b 12.1 Hz,J5’a, 2.5 Hz, 1 H, H(5’a)), 3.74-3.80 (m, 2 H, H(5’b), H(3)), 3.62-3.67 (m, 3 H, H(9a/b), H(4)), 3.50 (dd, J2eq, 2ax= 12.3 Hz, J2,= 5.2 Hz, 1 H, H(2eq)), 3.33 (ddd, J= 11.2 Hz, J = 7.4 Hz, J= 3.7 Hz, 1 H, H(6)), 2.89 (t, J2ax,2eq J2ax,3= 12.3 Hz, 1 H, H(2ax)), 1.97-2.06 (m, 1 H, H(7a)), 1.57-1.80 (m, 4 H, H(5), H(7b), H(8a/b)). ‘3C NMR (D20) 70.1 (C(4)), 68.5 (C(3)), 61.1 (C(9)), 56.5 (C(6)), 56.1 (C(5’)), 46.2 (C(2)), 45.0 (C(5)), 26.6 (C(7)), 26.3 (C(8)). HRMS (ESI) m/z 206.1392 [M + H]. Calculated forC9H20N4206.1392. 69 3.1.8 (3R, 4R, 5R, 6S, 8S)-8-Benzyloxy[5,6-b]oxacyclohexane-5-piperidine -3,4-diol [C6-Benzyl acetal IFG] (2.32) Benzyl 4-[[(S)-1 ‘-N-Benzyloxycarbonyl]-propan-3’-al]-4-deoxy-2,3-O- isopropylidene-P-D-arabinopyranoside (2.24) H0 Ozone was bubbled through a solution of the terminal olefin (2.20) (73 mg, 0.16 mmol) in CH21 (10 mL) at -78°C until it turned deep blue. The excess ozone was purged from the resulting solution with argon gas until the solution became clear again. Triphenylphosphine (164 mg, 0.63 mmol) was then added in one portion and the solution was allowed to warm to room temperature and stir overnight. The reaction solution was concentrated in vacuo and purified by flash column chromatography on silica gel (5.5:4.5 hexanes:EtOAc, 0.1% Et3N) to yield 2.24 as a white solid (55 mg, 0.12 mmol, 75%). 1H NMR (CDC13) 9.82 (s, 1 H, H(3’)), 7.28-7.42 (m, 5 H, (Ph)), 5.76-5.96 (bs, 1 H, (NH)), 5.27 (d, Ji,2= 3.1 Hz, 1 H, H(1)), 5.08 (s, 2 H, (COOCHPh)), 4.76 (d, JphCH2= 12.2 Hz, 1 H, (PhCH2)), 4.64 (d, JphCH2= 12.2 Hz, 1 H, (PhCH2)), 4.45 (ddd, J= 10.5 Hz, J= 10.5 Hz, J= 6.0 Hz, 1 H, H(1’)), 4.29 (dd,J3,= 10.0 Hz,J34= 4.6 Hz, 1 H, H(3)), 3.86 (d, J23 = 10.0 Hz, J2,i 3.1 Hz, 1 H, H(2)), 3.74 (dd,J5a,b = 12.8 Hz, J5a,4 = 2.4 Hz, 1 H, H(5a)), 3.50 (d,J5b,a 12.8 Hz, 1 H, H(5b)), 2.90 (dd, J2’a,2’b 16.9 Hz, J= 3.1 Hz, 1 H, H(2’a)), 2.71 (ddd,J2’b,a 16.9 Hz, J= 6.0 Hz, J= 3.1 Hz, 1 H, H(2’b)), 2.44-2.53 (m, 1 H, H(4)), 1.44 (s, 3 H, (CH3)), 1.34 (s, 3 H, (CH3)). ‘3C NMR (CDC13) 200.9 (C(3’)), 137.5, 136.5, 128.6, 128.55, 128.54, 128.24, 128.2, 127.9, 127.8, 127.7, 109.9, 97.9 (C(1)), 73.9 (C(3)), 72.4 (C(2)), 70.0, 66.9, 60.2 (C(S)), 47.9 (C(2’)), 45.7 (C(1’)), 42.9 (C(4)), 26.7, 26.5 2x(C(CH3)). HRMS (ESI) m/z 492.2006 [M + Na]. Calculated for 26H8NNaO6492.1998. 70 (3R, 4R, 5R, 6S, 8S)-8-Benzyloxy[5,6-bjoxacyclohexane-5-piperidine-3,4-diol (2.32) HO7 L 3 21 Pd/C (10%, 20 mg) was added to a solution of aldehyde (2.24) (60 mg, 0.13 mmol) and concentrated HC1 (8 iL) in dry MeOH and the mixture was hydrogenated at atmospheric pressure for 16 hours. The catalyst was removed by suction filtration through Celite and the filter cake washed with methanol (30 mL). p-Toluenesulfonic acid monohydrate (70 mg, 0.37 mmol) was added to a solution of the impure syrup (31 mg) in benzyl alcohol (1.5 mL) and CHC13 (1.5 mL) and stirred for 96 hours. As the reaction mixture was stirring, Amberlite 1R410 strongly basic (0H form) resin was added until the pH of the reaction was neutral. The resin was removed by suction filtration and the filtrate was evaporated in vacuo until the residue was a thick syrup that was washed with petroleum ether (5 mL) and extracted with 0.8 M aqueous ammonium acetate (2 x 8 mL, pH 7). The aqueous portions were concentrated in vacuo and purified by column chromatography using C-18 RP silica gel, eluted with 40% methanol in water. The fractions containing the desired product were pooled and lyophilized from 0.01 M aqueous ammonium acetate (2 mL) to yield 2.32 as a clear syrup (16 mg, 0.06 mmol, 46%). ‘H NMR (CD3OD) ö 7.28-7.38 (m, 5 H, H(Ph)), 5.03 (d, JS,7ax 3.1 Hz, 1 H, H(8)), 4.69 (d, JPhCH2= 11.8 Hz, 1 H, (PhCH2)), 4.48 (d, JphCH2= 11.8 Hz, 1 H, (PhCH2)), 3.96 (dd, J5’eq, 5’ax 11.3 Hz, J5’eq, = 4.3 Hz, 1 H, H(5’eq)), 3.59 (t, J5’ax, 5’eq= JS’ax, 11.3 Hz, 1 H, H(5’ax)), 3.46-3.52 (m, 1 H, H(3)), 3.20 (dd, J2eq,2ax 11.7 Hz, J2eq,3 5.1 Hz, 1 H, H(2eq)), 3.13 (dd, J = 10.4 Hz, J = 8.8 Hz, 1 H, H(4)), 2.96 (ddd,J65=J6,7= 12.3 Hz, J6, 7eq = 4.3 Hz, 1 H, H(6)), 2.58 (t, J2ax, 2eq J2ax, 3 11.7 Hz, 1 H, H(2ax)), 2.02 (dd, = 12.3 Hz, J7eq,6= 4.3 Hz, 1 H, H(7eq)), 1.63 (ddd,J7ax,=J7ax,6 12.3 Hz, J7ax, 8= 3.1 Hz, 1 H, H(7ax)), 1.51-1.58 (m, 1 H, H(S)). 13C NMR (MeOD) ö 129.3, 128.9, 128.7, 97.3 (C(8)), 74.8 (C(4)), 72.7 (C(3)), 69.8 (C(PhCH2)), 61.9 (C(5’)), 52.9 (C(6)), 50.35 (C(2)), 46.7 (C(S)), 36.1 (C(7)). HRMS (ESI) m/z 280.1550 [M + H]. Calculated forC15H22N04280.1549. 71 3.2 Enzymology 3.2.1 Glucocerebrosidase Kinetics Pure human glucocerebrosidase was obtained from previously used vials of Cerezyme® (produced by Genzyme Inc.) donated from patients currently undergoing enzyme replacement therapy. The buffer used for kinetics was pH 5.5, 20 mM citric acid, 50 mM Na2HPO4,1 mM tetrasodium EDTA, 0.25% v/v Triton-X-100® and 0.25% w/v taurocholic acid. 2,4-Dinitrophenyl -D-glucopyranoside was used as the substrate for continuous UV spectrophotometric assays performed at 37°C on a Varian Cary 4000 or Varian Cary 300 UV-Vis spectrophotometer. The concentration of enzyme used, as determined by BCA assay,89 was 2.2 nM for all assays except for compound 1.8 (IFG), for which 0.9 nM enzyme was used. Upon initiation with substrate, the release of 2,4- dinitrophenolate was measured at 400 nm and rates calculated using an extinction coefficient of 10.7 mlVf1cm90 4-Methylumbelliferyl -D-glucopyranoside was used as the substrate for all stopped fluorescent assays. Buffered solutions containing substrate were pre-incubated in Eppendorf tubes at 37°C in the presence or absence of an inhibitor. The reaction was initiated by addition of enzyme to a total concentration of 5 pM, then at fixed time intervals of 3, 6 and 9 minutes 100 tL aliquots were removed and diluted into a cuvette containing 500 iL of glycine buffer (pH 10.8). Dilution into pH 10.8 buffer both stopped the enzyme reaction as well as ionized the cleaved 4-methylumbelliferrone product, increasing the fluorescence signal. The resulting fluorescence from the release of methylumbelliferone was measured on a Varian Cary Eclipse fluorimeter. The instrument parameters were set as: 72 Excitation Wavelength (nm) 355.00 Emission Wavelength (nm) 460.00 Excitation Slit (nm) 5 Emission Slit (nm) 5 Ave Time (sec) 0.1000 Excitation filter Auto Emission filter Open PMT Voltage (V) High Michaelis-Menten parameters for both assays were determined by direct fit to the Michaelis-Menten expression using Grafit version 5.0.13.91 3.2.2 Inhibition Studies All inhibition kinetics were performed by measuring the rate of reaction at fixed concentrations of artificial substrate and enzyme while varying the inhibitor concentration. K1 values were determined by directly fitting the data to various inhibition models using GraFit version 5.0.13. Best fits in each case were to the equation for competitive inhibition (Equation 2.1) as shown below. = Vmax[S] Equation 2.1 [S] + Km(l + [I]/K1) In order to graphically represent K1 values, the inverse of the reaction rate was plotted Egainst the concentration of inhibitor in what is known as a Dixon plot. The line for lI’Vmax will intersect the lines for varying substrate concentrations at K, = -[I] if the inhibitor is a competitive inhibitor. The theory describing this method of data representation is explained in the Appendix. The linear graphs, along with inhibition constants are presented in Section 2.2. The inhibitor concentrations used in all assays are listed in Table 3.1 below. 73 Inhibitor [I] 0, 10, 30, 50, 100 (nM) 1.8 OH 0, 300, 500, 1000, 5000, 9000 (nM) 2.17 0, 300, 500, 1000, ,OH 5000, 9000 (nM) 2.18 0, 20, 50, 100, 750, 1000 (nM) 2.23 0, 20, 50, 100, 750, 1000 (nM) HO NH 2.32 OH 0, 0.05, 0.1, 0.2, HO.(CH)BC 0.5, 0.75 (nM) 1.9 Table 3.1 Summary of inhibitor concentrations used in the assay of each inhibitor. 74 References (1) Wolfenden, R.; Lu, X. D.; Young, G. J. Am. Chem. Soc. 1998, 120, 6814- 68 15. (2) Henrissat, B.; Bairoch, A. Biochem. J 1996, 316, 695-696. (3) Koshland, D. E. Biol. Rev. 1953, 28, 4 16-436. (4) Phillips, D. C. Proc. Nati. Acad. Sci. U. S. A. 1967, 57, 484-&. (5) Sinnott, M. L.; Souchard, I. J. L. Biochem. J 1973, 133, 89-98. (6) Sinnott, M. L.; Viratell, 0. M. Biochem. J 1973, 133, 8 1-87. (7) Withers, S. G.; Street, I. P.; Bird, P.; Dolphin, D. H. J. Am. Chem. Soc. 1987, 109, 7530-753 1. (8) Davies, G. J.; Ducros, V. M. A.; Varrot, A.; Zechel, D. L. Biochem. Soc. Trans. 2003, 31, 523-527. (9) White, A.; Tull, D.; Johns, K.; Withers, S. G.; Rose. D. R. Nat. Struct. Biol. 1996, 3, 149-154. (10) Zechel, D. L.; Withers, S. G. Acc. Chem. Res. 2000, 33, 11-18. (11) Withers, S. G.; Warren, R. A. J.; Street, I. P.; Rupitz, K.; Kempton, J. B.; Aebersold, R. J. Am. Chem. Soc. 1990, 112, 5887-5889. (12) Withers, S. G.; Aebersold, R. Protein Sci. 1995, 4, 361-372. (13) McCarter, J. D.; Withers, S. G. Curr. Opin. Struct. Biol. 1994, 4, 885-892. (14) Wang, Q. P.; Graham, R. W.; Trimbur, D.; Warren, R. A. J.; Withers, S. G. .1. Am. Chem. Soc. 1994, 116, 11594-11595. (15) Wolfenden. R. Acc. Chem. Res. 1972, 5, 10-18. (16) Vasella, A.; Davies, G. J.; Bohm, M. Curr. Opin. Chem. Biol. 2002, 6, 619-629. (17) Brumshtein, B.; Greenblatt, H. M.; Butters, T. D.; Shaaltiel, Y.; Aviezer, D.; Silman, I.; Futerman, A. H.; Sussman, J. L. J. Biol. Chem. 2007, 282, 29052-29058. (18) Kim, Y. W.; Lee, S. S.; Warren, R. A. J.; Withers, S. G. J. Biol. Chem. 2004, 279, 42787-42793. 75 (19) Lairson, L. L.; Watts, A. 0.; Wakarchuk, W. W.; Withers, S. 6. Nat. Chem. Biol. 2006,2, 724-728. (20) Truscheit, E.; Frommer, W.; Junge, B.; Muller, L.; Schmidt, D. D.; Wingender, W. Angew. Chem. mt. Ed. 1981, 20, 744-761. (21) Gruters, R. A.; Neefjes, J. J.; Tersmette, M.; Degoede, R. E. Y.; Tuip, A.; Huisman, H. 6.; Miedema, F.; Ploegh, H. L. Nature 1987, 330, 74-77. (22) Goss, P. E.; Baker, M. A.; Carver, J. P.; Dennis, J. W. Clin. Cancer Res. 1995, 1, 935-44. (23) Gubareva, L. V.; Kaiser, L.; Hayden, F. G. Lancet 2000, 355, 827-835. (24) Ganem, B. Ace. Chem. Res. 1996, 29, 340-347. (25) Stutz, A. E. Iminosugars as glycosidase inhibitors : Nojirimycin and beyond; Wiley-VCH: Weinheim ; New York, 1999. (26) Look, G. C.; Fotsch, C. H.; Wong, C. H. Ace. Chem. Res. 1993, 26, 182- 190. (27) Jensen, H. H.; Lyngbye, L.; Bols, M. Angew. Chem. mt. Ed. 2001,40, 3447-3449. (28) Varrot, A.; Tarling, C. A.; Macdonald, J. M.; Stick, R. V.; Zechel, D. L.; Withers, S. 6.; Davies, G. J. J. Am. Chem. Soc. 2003, 125, 7496-7497. (29) Paulsen, H. I. Iminosugars as glycosidase inhibitors Nojirimycin and beyond; Stutz, A. E., Ed.; Wiley-VCH: Weinheim ; New York, 1999, pp.1-7. (30) Baxter, E. W.; Reitz, A. B. J. Org. Chem. 1994, 59, 3 175-3185. (31) Scott, L. J.; Spencer, C. M. Drugs 2000, 59, 521-549. (32) Molyneux, R. J.; James, L. F. Science 1982, 216, 190-19 1. (33) Hohenschutz, L. D.; Bell, E. A.; Jewess, P. J.; Leworthy, D. P.; Pryce, R. J.; Arnold, E.; Clardy, J. Phytochemistry 1981,20, 811-814. (34) Umezawa, H.; Aoyagi, T.; Komiyama, T.; Morishim.H; Hamada, M.; Takeuchi, T. J. Antibiot. 1974, 27, 963-969. (35) Jespersen, T. M.; Dong, W. L.; Sierks, M. R.; Skrydstrup, T.; Lundt, I.; Bols, M. Angew. Chem. mt. Ed. 1994, 33, 1778-1779. 76 (36) Dong, W. L.; Jespersen, T.; Bols, M.; Skrydstrup, T.; Sierks, M. R. Biochemistry 1996, 35, 2788-2795. (37) Bols, M. Acc. Chem. Res. 1998, 31, 1-8. (38) Zhu, X. X.; Sheth, K. A.; Li, S. H.; Chang, H. H.; Fan, J. Q. Angew. Chem. mt. Ed. 2005,44, 7450-7453. (39) Ichikawa, Y.; Igarashi, Y.; Ichikawa, M.; Suhara, Y. J. Am. Chem. Soc. 1998, 120, 3007-30 18. (40) Kim, Y. J.; Ichikawa, M.; Ichikawa, Y. J. Org. Chem. 2000, 65, 2599- 2602. (41) Pandey, G.; Bharadwaj, K. C.; Khan, M. I.; Shashidhara, K. S.; Puranik, V. G. Org. Biomol. Chem. 2008, 6, 2587-2595. (42) Chang, H. H.; Asano, N.;Ishii, S.; Ichikawa, Y.; Fan, J. Q. FEBS J. 2006, 273, 4082-4092. (43) Steet, R. A.; Chung, S.; Wustman, B.; Powe, A.; Do, H.; Kornfeld, S. A. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 13813-13818. (44) Yu, Z. Q.; Sawkar, A. R.; Whalen, L. J.; Wong, C. H.; Kelly, J. W. J. Med. Chem. 2007, 50, 94-100. (45) Lieberman, R. L.; Wustman, B. A.; Huertas, P.; Powe, A. C.; Pine, C. W.; Kbanna, R.; Schlossmacher, M. G.; Ringe, D.; Petsko, G. A. Nat. Chem. Biol. 2007, 3, 10 1-107. (46) Alberts, B. Essential Cell Biology : An Introduction to the Molecular Biology of the Cell; Garland: New York, 1998, PP. 476-477. (47) Futerman, A. H.; van Meer, G. Nature Rev. Mol. Cell Bio. 2004, 5, 554- 565. (48) Yoshida, H. FEBS J. 2007, 274, 630-658. (49) Wong, K. D.; Sidransky, E.; Verma, A.; Mixon, T. H.; Sandberg, G. D.; Wakefield, L. K.; Morrison, A.; Lwin, A.; Colegial, C.; Ailman, J. M.; Schiffmann, R. Mol. Genet. Metab. 2004, 82, 192-207. (50) Meikie, P. J.; Hopwood, J. J.; Clague, A. E.; Carey, W. F. J. Am. Med. Assoc. 1999, 281, 249-254. 77 (51) Varki, A. Essentials of Glycobiology; 2nd ed.; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, N.Y., 2009, pp.136-41. (52) Zhao, H.; Grabowski, G. A. Cell. Mol. Life Sci. 2002, 59, 694-707. (53) Pentchev, P. G.; Brady, R. 0.; Hibbert, S. R.; Gal, A. E.; Shapiro, D. J. Biol. Chem. 1973, 248, 5256-5261. (54) Forneris, F.; Mattevi, A. Science 2008, 321, 213-216. (55) Alattia, J.-R.; Shaw, J. E.; Yip, C. M.; Prive, G. G. Proc. Nail. Acad. Sci. U. S. A. 2007, 104, 17394-17399. (56) Rossmann, M.; Schultz-Heienbrok, R.; Behlke, J.; Remmel, N.; Alings, C.; Sandhoff, K.; Saenger, W.; Maier, T. Structure 2008, 16, 809-817. (57) Dvir, H.; Harel, M.; McCarthy, A. A.; Toker, L.; Silman, I.; Futerman, A. H.; Sussman, J. L. EMBO Reports 2003, 4, 704-709. (58) Blonder, E.; Klibansky, C.; Devries, A. Biochim. Biophys. Acta 1976, 431, 45-53. (59) Premkumar, L.; Sawkar, A. R.; Boldin-Adamsky, S.; Toker, L.; Silman, I.; Kelly, J. W.; Futerman, A. H.; Sussman, J. L. J. Biol. Chem. 2005, 280, 23815-238 19. (60) Ellgaard, L.; Molinari, M.; Helenius, A. Science 1999, 286, 1882-1888. (61) Fan, J. Q. Biol. Chem. 2008, 389, 1-11. (62) Beck, M. Hum. Genet. 2007, 121, 1-22. (63) Schueler, U. H.; Kolter, T.; Kaneski, C. R.; Zirzow, G. C.; Sandhoff, K.; Brady, R. 0. J. Inherited Metab. Dis. 2004, 27, 649-658. (64) Sawkar, A. R.; D’Haeze, W.; Kelly, J. W. Cell. Mol. Life Sci. 2006, 63, 1179-1192. (65) Grabowski, G. A.; Barton, N. W.; Pastores, G.; Dambrosia, J. M.; Banerjee, T. K.; Mckee, M. A.; Parker, C.; Schiffmann, R.; Hill, S. C.; Brady, R. 0. Ann. Intern. Med. 1995, 122, 33-39. (66) Barton, N. W.; Brady, R. 0.; Dambrosia, J. M.; Dibisceglie, A. M.; Doppelt, S. H.; Hill, S. C.; Mankin, H. J.; Murray, G. J.; Parker, R. I.; Argoff, C. E.; Grewal, R. P.; Yu, K. T. New Engi. J. Med. 1991,324, 1464-1470. 78 (67) Beutler, E. Mol. Genet. Metab. 2006, 88, 208-215. (68) Shaaltiel, Y.; Bartfeld, D.; Hashmueli, S.; Baum, 0.; Brill-Almon, E.; Galili, G.; Dym, 0.; Boldin-Adamsky, S. A.; Silman, I.; Sussman, J. L.; Futerman, A. H.; Aviezer, D. Plant Biotechnol. J. 2007, 5, 579-590. (69) Cox, T.; Lachmann, R.; Hollak, C.; Aerts, J.; van Weely, S.; Hrebicek, M.; Platt, F.; Butters, T.; Dwek, R.; Moyses, C.; Gow, I.; Elstein, D.; Zimran, A. Lancet 2000, 355, 148 1-1485. (70) Futerman, A. H.; Sussman, J. L.; Horowitz, M.; Silman, I.; Zimran, A. Trends Pharmacol. Sci. 2004, 25, 147-15 1. (71) Butters, T. D.; Dwek, R. A.; Platt, F. M. Glycobiology 2005, 15, R43-R52. (72) Butters, T. D. Curr. Opin. Chem. Biol. 2007, 1], 412-418. (73) Sawkar, A. R.; Adamski-Werner, S. L.; Cheng, W. C.; Wong, C. H.; Beutler, E.; Zimmer, K. P.; Kelly, J. W. Chem. Biol. 2005, 12, 1235-1244. (74) Tropak, M. B.; Kornhaber, G. J.; Rigat, B. A.; Maegawa, 0. H.; Buttner, J. D.; Blanchard, J. E.; Murphy, C.; Tuske, S. J.; Coales, S. J.; Hamuro, Y.; Brown, E. D.; Mahuran, D. J. ChemBioChem 2008, 9, 2650-62. (75) Kornhaber, G. J.; Tropak, M. B.; Maegawa, G. H.; Tuske, S. J.; Coales, S. J.; Mahuran, D. J.; Hamuro, Y. CheinBioChern 2008, 9, 2643-9. (76) Zheng, W.; Padia, I.; Urban, D. J.; Jadhav, A.; Goker-Alpan, 0.; Simeonov, A.; Goldin, E.; Auld, D.; LaMarca, M. B.; Inglese, J.; Austin, C. P.; Sidransky, E. Proc. Nati. Acad. Sci. U. S. A. 2007, 104, 13 192- 13 197. (77) Compain, P.; Martin, 0. R.; Boucheron, C.; Godin, 0.; Yu, L.; Ikeda, K.; Asano, N. ChemBioChem 2006, 7, 1356-1359. (78) Ichikawa, M.; Igarashi, Y.; Ichikawa, Y. Tetrahedron Lett. 1995, 36, 1767-1770. (79) Jakobsen, P.; Lundbeck, J. M.; Kristiansen, M.; Breinholt, J.; Demuth, H.; Pawlas, J.; Candela, M. P. T.; Andersen, B.; Westergaard, N.; Lundgren, K.; Asano, N. Biorg. Med. Chem. 2001, 9, 733-744. (80) Andersch, J.; Bols, M. Chem. Eur. J. 2001, 7, 3744-3747. 79 (81) Best, W. M.; Macdonald, J. M.; Skelton, B. W.; Stick, R. V.; Matthew, D.; Tilbrook, G.; White, A. H. Can. J. Chem. 2002, 80, 857-865. (82) Meloncelli, P. J.; Stick, R. V. Aust. J. Chem. 2006, 59, 827-833. (83) Goddard-Borger, E. D.; Stick, R. V. Aust. J. Chem. 2007, 60, 211-213. (84) Deslongchamps, P.; Dory, Y. L.; Li, S. G. Tetrahedron 2000, 56, 3533- 3537. (85) Weiberth, F. J.; Hall, S. S. J. Org. Chem. 1986, 51, 5338-5341. (86) Weiberth, F. J.; Hall, S. S. J. Org. Chem. 1987, 52, 3901-3904. (87) Imamoto, T.; Takiyama, N.; Nakamura, K.; Hatajima, T.; Kamiya, Y. J. Am. Chem. Soc. 1989, 111, 4392-4398. (88) Paal, K.; Ito, M.; Withers, S. G. Biochem. J2004, 378, 141-149. (89) Wiechelman, K. J.; Braun, R. D.; Fitzpatrick, J. D. Anal. Biochem. 1988, 175, 23 1-237. (90) Wong, A. W., Ph.D. Thesis; The University of British Columbia, 2001. (91) Leatherbarrow, R. I. GraFit; Version 5.0.13, Erithacus Software Ltd., 2006. (92) Segel, I. H. Enzyme Kinetics . Behavior and Analysis ofRapid Equilibrium and Steady-State Enzyme Systems; Wiley: New York, 1993. 80 Appendix A-i Fundamental Equations of Enzyme Kinetics The Michaelis-Menten treatment of enzyme kinetics, where a single substrate is converted to a single product, is based on the following reaction scheme (Scheme A.1) where E represents enzyme, S the substrate and P the product. k1 k2 E + S — ES E + P k..1 Scheme A.1 General scheme for enzyme-catalyzed conversion of a single substrate into a single product. In the first step of the reaction, the enzyme and substrate combine in a rapid, reversible process to produce an enzyme-substrate complex (ES). In the second step, the bound substrate is converted to product and subsequently released from the enzyme. Under steady state conditions, the rate of change of the concentration of ES complex is zero, as described in Equation A.l. 8[ESJ =k1[E][SJ - k..1{ES] -k2[ES] = 0 EquationA.1 As the total amount of enzyme [E]0 added to the reaction is quantifiable and is equal to the sum of the concentrations of both free enzyme [El and bound enzyme [ES], [E]0 can be represented by Equation A.2. [El0 = [E] + [ES] Equation A.2 81 Solving for [ES] in Equations A.1 and A.2, [E]0S [ES] = Equation A.3 [S] _____ k ) Assuming that the rate limiting step is breakdown of the ES complex to yield P, then the rate of the reaction (v) can be described by Equation A.4 [P] Equation A.4 =k2[ES] By substituting the expression for [ES] from Equation A.3 into Equation A.4, the result is Equation A.5 k2[E]0S v= /7 1 Equation A.5[S] +(‘c’+2\ k Defining the ratio of the rate constants (k1+2)/k as the Michaelis constant (K), and the rate constant k2 as the turnover number (k1), then Equation A.5 can be simplified to the more general form known as the Michaelis-Menten equation (Equation A.6), where the product of kcat and [El0 is also referred to as Vrnax. kcat[EIo[SI = [SI + Km Equation A.6 Equation A.6 describes a rectangular hyperbola such as is shown in Figure A. 1. 82 V Vmax[S]IKm v—k [El—V —- -. — cat = Km [Substrate] Figure A.1 A plot showing the typical hyperbolic nature of the Michaelis-Menten equation (Equation A.6) The Michaelis constant (Km) is defined as the substrate concentration at which the reaction rate (v) is half-maximal (V = VmaxI2). It may be treated as an apparent dissociation constant of all the enzyme bound species and is expressed in Equation A.7 — [E][S Km — 2’ [ES] Equation A.7 Therefore, the Km value reflects the stability of the bound enzyme-substrate complex. Lower values of Km indicate tighter binding of the substrate to the enzyme. At low [S] ([SI <<Km), the Michaelis-Menten equation reduces to Equation A.8 and V is linearly dependent upon substrate concentration (also refer to Figure A.l). Since most of kcat[El0[S] = Km Equation A.8 83 the enzyme is unbound, the total enzyme concentration [E]0 can be approximated to the concentration of the free enzyme [E]. Under these conditions, the Michaelis-Menten equation can now be expressed as Equation A.9 kcat = Km [El [SI Equation A.9 In Equation A.9, kcatlKm is an apparent second-order rate constant for the reaction of the free enzyme with free substrate. This kinetic parameter is also a measure of the overall efficiency of the enzyme. At high [S] ([SI >> Km), v approaches a limiting value, Vmax, and the Michaelis-Menten equation can now be expressed as Equation A. 10 (also refer to Figure A.1) Vmax = kcat[Elo Equation A.1O Prior to the advent of computers and for the purpose of graphical representation of the enzymatic rate data, the Michaelis-Menten equation was often rearranged into the linear form of Equation A. 11 1 Km 1 1 —+ v • s-’ V Equation A.11max i max As shown in Figure A.2, plotting 1/v versus 1/[S] (also known as a Lineweaver Burk plot) yields a straight line with a slope of KmfVmax, and a y-intercept of 1/Vmax, and a x-intercept of 1/Km. 84 1/v Figure A.2 A typical Lineweaver-Burk plot for an enzymatic reaction. A-2 Reversible Competitive Inhibition Substances that combine with an enzyme in a reversible manner to decrease the activity of that enzyme are referred to as inhibitors. A competitive inhibitor is a compound that competes directly with a normal substrate for binding to the enzyme active site. These compounds often resemble the substrate but are unreactive towards the enzyme. In the presence of such a compound, the equations for the simple enzyme- catalyzed reaction must be expanded to include a second equilibrium, K,, which describes the dissociation of the inhibitor from the enzyme-inhibitor complex (K, = [E] [I]/[EI]) (Scheme A.2). Slope = Km/Vmax 4/Km 1/[Substrate] 85 k1 k2 E + S ES E + P k1 + I El Scheme A.2 General scheme for enzyme-catalyzed conversion of a single substrate into a single product in the presence of a competitive inhibitor. The total concentration of enzyme is now given by Equation A.12 [E]0 [Ej + [ES] + [El] Equation A.12 Applying the steady state approach (Equation A. 1) and substituting into Equation A.6 gives Equation 2.1. Vmax [SI [SI + Km(l + [Il/K1) Equation 2.1 As can be seen from Equation 2.1, a competitive inhibitor only affects the Km term of the Michaelis-Menten equation, increasing it by a factor of (1 + [I]/K). The value of Vmax is unaffected since at high concentrations of substrate, the inhibitor is displaced from the enzyme active site. Just as a linear transformation of the data can elucidate uninhibited kinetic parameters (Figure A.2), a similar linear plot can be used to graphically represent K1 values. By plotting 1/v versus [I] (Dixon plot), the line for 1/Vmax intersects the lines for different substrate concentrations at K1 = -[I] if the inhibitor is a competitive inhibitor. The theory for this method of representing K, values is presented below, taken from Segel.92 86 For competitive inhibition, Vmax [S] [S] + Km(1 + [I]/K) The reciprocal of this equation is 1 Km[I] 1 Kin v = Vmax[S]Ki + Vmax (i + [SI ) When_L 1 V Vmax 1 Km[I] 1 Km Vmax = Vmax[SIKi + Vmax (i + [S] ) and Km[II ____ 1= +1+[S]K1 [5] so, Km[I] — Km - [S]K1 ES] and [I] = -K 87

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0061636/manifest

Comment

Related Items