UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Carbohydrate-bearing ligands for biologically active metal ions Green, David 2004

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

Item Metadata

Download

Media
831-ubc_2004-99418x.pdf [ 8.67MB ]
Metadata
JSON: 831-1.0061177.json
JSON-LD: 831-1.0061177-ld.json
RDF/XML (Pretty): 831-1.0061177-rdf.xml
RDF/JSON: 831-1.0061177-rdf.json
Turtle: 831-1.0061177-turtle.txt
N-Triples: 831-1.0061177-rdf-ntriples.txt
Original Record: 831-1.0061177-source.json
Full Text
831-1.0061177-fulltext.txt
Citation
831-1.0061177.ris

Full Text

CARBOHYDRATE-BEARING LIGANDS FOR BIOLOGICALLY ACTIVE METAL IONS by David E. Green B.Sc. (Hons.), Brock University, 1995 M.Sc, McMaster University, 1997 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 2004 © David E. Green, Abstract Two distinct projects involving carbohydrates in bioinorganic chemistry are discussed in this thesis. In one study, several gallium and indium complexes with pendant carbohydrates were prepared and examined for potential use in radiopharmaceutical applications. Carbohydrate-bearing 3-hydroxy-4-pyridinone ligand precursors and their tris(ligand)gallium(III) and indium(III) complexes were synthesized and characterized by mass spectrometry, elemental analysis, ] H and l 3 C NMR spectroscopy, and in the case of one intermediate, by X-ray crystallography. With three equivalents of ligand, neutral complexes formed with the pyridinone moiety (as expected) complexing the gallium(III) and indium(III) metal centers. In the other study, a series of 3-hydroxy-4-pyridinones and their glycoside pro-drugs (pyridinone glycosides) were prepared and studied. The pyridinones have the potential to be used in metal chelation therapy to aid in the treatment of neurodegenerative disorders such as Alzheimer's disease. The pyridinone glycosides were synthesized via a Mitsunobu coupling and characterized by mass spectrometry, elemental analysis, ! H and 1 3 C NMR spectroscopy, and in one case by X-ray crystallography. The glycoside pro-drugs mask the metal chelation portion of the pyridinones by protecting the 3-hydroxy position. The glycosidic cleavage of the pyridinone glycosides was studied (and kinetic parameters determined for one compound) with Agrobacterium sp. P-glucosidase. This enzyme cleaves the pyridinone glycosides unmasking the pyridinones and making them available for metal chelation. A series of the 3-hydroxy-4-pyridinones also displayed significant antioxidant protection in a radical quenching assay. ii T A B L E O F C O N T E N T S Abstract ii Table of Contents i i i List of Figures vi List of Schemes viii List of Tables ix List of Equations ix List of Abbreviations x Acknowledgments xv Chapter 1 Introduction 1.1 Coordination Complexes in Medicinal Chemistry 1 1.2 Carbohydrates and Glucose Transport 1 1.3 Nuclear Medicine 4 1.4 Gallium and Indium in Nuclear Medicine 6 1.5 3-Hydroxy-4-pyridinone Complexes of Gallium and Indium 7 1.6 Carbohydrates and 3-Hydroxy-4-pyridinones for Chelating Metals 9 1.7 Oxidative Stress in the Central Nervous System 9 1.7.1 General Principles 9 1.7.2 Alzheimer's Disease 10 1.8 Potential Pharmacotherapies for Alzheimer's Disease 19 1.9 Thesis Overview 23 1.10 References 24 iii Chapter 2 Ga(III) and In(III) Complexes of Carbohydrate-Bearing 3-Hydroxy-4-pyridinones 2.1 Introduction 33 2.2 Experimental 35 2.2.1 Materials 35 2.2.2 Instrumentation 36 2.2.3 Abbreviations for the 3-Hydroxy-4-pyridinones 38 2.2.4 Syntheses 39 2.3 Results and Discussion 75 2.3.1 Ligand Preparations 75 2.3.2 Tris(3-oxy-4-pyridinato)gallium(III) and Indium(III) Complexes 78 2.3.3 'H and l 3 C NMR Spectra 2.4 Conclusions 89 2.5 References 90 Chapter 3 Carbohydrate-Protected 3-Hydroxy-4-pyridinones 3.1 Introduction 92 3.2 Experimental 95 3.2.1 Materials and Instrumentation 95 3.2.2 Enzyme Kinetics 96 3.2.3 Trolox Equivalent Antioxidant Capacity (TEAC) Antioxidant Assay 98 3.2.4 Cell Studies and MTT Assay 99 3.2.5 Synthesis 100 iv 3.3 Results and Discussion 110 3.3.1 Synthesis and Characterization of Products 110 3.3.2 Enzymatic Cleavage of Pyridinone Glycosides 117 3.3.3 T E A C Values of Pyridinones 123 3.3.4 Cel l Studies 125 3.4 Conclusions 126 3.5 References 127 Chapter 4 Conclusions and Future Work 4.1 Gal l ium and Indium Complexes with Carbohydrate-Bearing Pyridinones 130 4.2 Gall ium and Indium Complexes Containing One Glucose 132 4.3 Experimental for Ga(OG6GP)(dpp)2 and Ga(OG6GP) 2(dpp) 138 4.4 Other Glucose-Bearing Pyridinones for Potential in Molecular Metabolic Imaging 141 4.5 Carbohydrate-Lectin Interactions 142 4.6 Metal Chelation of Pyridinone Glycosides and Alzheimer's Disease 145 4.7 Glycosyl Protection and Pharmokinetics 147 4.8 Determining the Effects of Pyridinone Glycosides on Alzheimer's Disease 15 0 4.9 Radioactive Analogues of the Pyridinones 151 4.10 Increasing the Antioxidant Potential of Chelators 154 4.11 Targeting P-Amyloid 160 4.12 References 163 Appendix I X-ray Crystallographic Data for 8 and 27p 167 v LIST O F FIGURES Figure 1.1 Cartoon of AD pathologies. Beta-amyloid (neuritic) plaques and 12 neurofibrillary tangles and their spatial arrangement within and around the neuron. Figure 1.2 Proposed metal binding regions of metal-Ap complexes. 17 Figure 1.3 Structures of metal chelators used to break metal-Ap interactions. 21 Figure 2.1 Ligands for gallium or indium in nuclear medicine applications. 34 Figure 2.2 *H NMR spectra (400 MHz) in 1:1 CD 3OD:D 20 for (a) 82 Ga(OGBAP) 3 (b) In(OGBAP) 3 (c) H O G B A P . Figure 2.3 1 3 C NMR spectra in 1:1 CD 3OD:D 20 for (a) Ga(OG6GP) 3 (b) 84 In(OG6GP) 3 (c) HOG6GP. Figure 2.4 ORTEP diagram of 8 (with some H-atoms omitted for clarity) 86 showing 50% thermal probability ellipsoids. Figure 2.5 Unit Cell/Packing diagram for 8. 88 Figure 3.1 Structures of some products and reagents in Chapter 3. 94 Figure 3.2 Representative example of integration used to determine the anomer 113 ratio for the pyridinone glycosides. Figure 3.3 ORTEP diagram of 27p showing 50% thermal probability ellipsoids. 116 Figure 3.4 Silica TLC monitoring of Abg enzyme reactions with pyridinone 118 glycosides. Figure 3.5 Saturation curve for initial rates (AA.U./s) versus substrate 120 concentrations (mM) for 27p and Abg in sodium phosphate buffer at 37°C. Figure 3.6 Initial rates versus substrate concentrations for 27p and Abg in 122 sodium phosphate buffer at 37°C. Figure 3.7 TEAC values at 1, 3, and 6 minutes for typical antioxidants and 124 various 3-hydroxy-4-pyridinones. Figure 3.8 MTT plots for Hsbp and cisplatin, with IC 5 0 values equal to 570 ± 90 126 vi and 35 ± 5 uM, respectively. Figure 4.1 Prepared in this work: gallium and indium complexes of 131 carbohydrate-bearing 3-hydroxy-4-pyridinones. Figure 4.2 Different possible isomers for mixed ligand complexes 135 Ga(OG6GP)(dpp) 2 and Ga(OG6GP)2(dpp). Figure 4.3 *H NMR spectra (400 MHz, 1:1 D 20:CD 3OD) for 136 Ga(OG6GP)(dpp) 2. Figure 4.4 Other glucose-pyridinone derivatives that could be synthesized with 141 potential for glucose metabolic imaging with gallium or indium complexes. Figure 4.5 A potential hexadentate tri(carbohydrate/pyridinone) chelator. 144 Figure 4.6 Pyridinone glycosides from this work. 146 Figure 4.7 Anthocyanin 3-glucosides and quercetin derivatives. 148 Figure 4.8 Other potential metal chelators for AD. 149 Figure 4.9 Phenolic pyridinone compounds with enhanced antioxidant capacity. 155 Figure 4.10 Potentially applicable quercetin derivatives. 156 Figure 4.11 Commercially available flavonols and their common names. 157 Figure 4.12 3-Hydroxy-4-quinolinone analogues of certain flavonols.51 159 Figure 4.13 Compounds known to bind to AP plaques: BTA-1, IMPY, and 161 FDDNP are potential PET analogues that bind to Ap plaques. Figure 4.14 Potential 3-hydroxy-4-pyridinone derivatives with dimethylamine 162 and/or a connected ring system to target AP peptides, as well as iron and copper ions. vii LIST O F S C H E M E S Scheme 1.1 3-Hydroxy-2-methyl-4-pyridinones and their chelation to trivalent metal ions such as Ga(III) or In(III). 8 Scheme 1.2 APP cleavage by the secretases to produce soluble APP fragments (APPsa and APPsP) as well as APi . 4 0 and AP1.42. • . 13 Scheme 1.3 Oxidative stress that mediates Ap neurotoxicity may be partially due to the ability of AP (AA) to shuttle redox-active transition metals such as iron to the cell membrane with consequent catalysis of redox-mediated alterations in membrane components, such as lipid peroxidation. Pretreatment of Ap with desferrioxamine (DFO) reduces oxidative damage when iron is added to the system. 16 Scheme 2.1 Synthesis ofHOG6GP. 75 Scheme 2.2 Synthesis of HOGBAP. 76 Scheme 2.3 Synthesis ofHOGBPP. 77 Scheme 2.4 Synthesis of HAG6GP. 78 Scheme 2.5 Synthesis of HAGBAP. 79 Scheme 2.6 Synthesis of Ga(OG6GP)3 (95 %), In(OG6GP)3 (89 %), Ga(AG6GP) 3 (83 %), In(AG6GP)3 (86 %), Ga(OGBAP) 3 (95 %), In(OGBAP) 3 (91 %), Ga(AGBAP) 3 (90 %), In(AGBAP) 3 (93 %), Ga(OGBPP) 3 (88 %), and In(OGBPP)3 (91 %). 79 Scheme 2.7 Synthesis of Ga(hpp)3. 89 Scheme 3.1 Synthesis of pyridinones. 111 Scheme 3.2 Synthesis of pyridinone glycosides. 112 Scheme 3.3 Proposed Mitsunobu mechanism. 115 Scheme 3.4 Abg enzyme cleavage of the pyridinone glycoside P-anomers to produce free pyridinone and glucose. 118 Scheme 4.1 134 Scheme 4.2 152 Scheme 4.3 viii 153 LIST O F T A B L E S Table 2.1 1 3 C NMR AS for the tris(3-oxy-4-pyridinato)gallium and indium 81 complexes versus the free ligands. Table 2.2 Selected bond lengths (A) and angles (°) in 8 87 Table 3.1 Bond lengths (A) and angles (°) in 27p. 116 Table 3.2 Parameters for the glycosidic cleavage of 27p by Abg at 37°C from 121 this work, compared to other Abg data42 for phenolic substrates with similar pKa values. Table 3.3 TEAC values ± SD for 1, 3, and 6 minutes. 125 Table A l Selected crystallographic data for 8. 167 Table A2 Torsion angles for 8. 167 Table A3 Selected crystallographic data for 27p. 168 Table A4 Torsion angles for 27p. 168 LIST OF EQUATIONS Equation 3.1 119 Equation 3.2 119 Equation 3.3 121 ix LIST OF ABBREVIATIONS A angstrom, 1 x 10"10 metre a-Toc a-tocopherol P+ positron 8 chemical shift in parts per million (ppm) from a standard (NMR) 7 gamma rays X wavelength v0 initial rate AP p-Amyloid A absorbance Abg Agrobacterium sp. P-glucosidase enzyme ABTS 2,2'-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) AcOEt ethyl acetate ACS American Chemical Society (grade for solvents) AD Alzheimer's disease ADDP azodicarboxylic acid dipiperide ALS amytrophic lateral sclerosis Anal analytical APOE apolipoprotein E APP amyloid precursor protein arom aromatic atm atmosphere AU absorption units x bar unit of pressure = 105 Pa BBB blood-brain barrier BC bathocuproine BHT butylatedhydroxytoluene Bn benzyl BP bathophenanthroline °C degrees Celsius Calcd calculated CNS central nervous system COSY correlation spectroscopy (NMR) C.S.H.A.W.G Canadian Study of Health and Aging Working Group 2D two-dimensional d doublet (NMR), day(s) dd doublet of doublets (NMR) ddd doublet of doublet of doublets (NMR) Da Dalton DCC dicyclohexylcarbodiimide D E A D diethyl azodicarboxylate DFO desferrioxamine DMF dimethylformamide D-pen D-penicillamine DTPA diethylenetriaminepentaacetic acid EC enzyme classification EGTA glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid EPR electron paramagnetic resonance ESI electrospray ionization EtOH ethanol eq equivalents (of reagent) eV electron volt FAD familial Alzheimer's disease FDA Food and Drug Administration (USA) FDG [18F]2-deoxy-2-fluoro-D-glucose g gram GalNAc N-acetylgalactosamine GLUT glucose transporter h hour(s) Hcq clioquinol HMBC heteronuclear multiple bond correlation (NMR) HMQC heteronuclear multiple quantum coherence (NMR) HPLC high-performance liquid chromatography Hz hertz (s"1) IC50 drug concentration at which 50% of cells are viable relative to the control J coupling constant (NMR) kcat turnover number (number of substrate molecules converted to product per enzyme molecule per unit time, when the enzyme is saturated with substrate) kcJKm catalytic efficiency xii Km Michaelis constant, the substrate concentration at half the maximum rate L Litre or ligand LC liquid chromatography LSIMS liquid secondary ion mass spectometry m metre M molarity or metal MeCN acetonitrile MeOH methanol min minutes mol mole (6.02 x 1023 molecules) Mp melting point MS mass spectrometry MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide NFT neurofibrillary tangles NHS N-hydroxysuccinimide NMR nuclear magnetic resonance NSAID non-steroidal anti-inflammatory drug ORTEP Oak Ridge Thermal Ellipsoid Program PBS phosphate buffered saline Pd/C palladium on carbon PET positron emission tomography pH -log[H+] pK a -log K a xiii ppm parts per million PS preseniline q quartet (NMR) RBF round bottom flask Rf retardation factor ROS reactive oxygen species RT room temperature s second, singlet (NMR) SD standard deviation SGLT sodium dependant glucose transporters SN2 second order or bimolecular nucleophilic substitution SOD superoxide dismutase SPECT single photon emission computed tomography t triplet (NMR) t./2 half-life TEAC Trolox Equivalent Antioxidant Capacity TETA triethylenetetraamine Tg transgenic TLC thin layer chromatography TPEN N,N,N',N'-tetrakis(2-pyridylmethyl)ethylenediamine UV ultraviolet V m a x the maximum rate of a reaction vis visible xiv A C K N O W L E D G M E N T S I would like to thank my family and friends for all their help over the years. It sure seemed like a long road to get this Ph.D. (1 y not enjoying engineering, 5 y Hon. B.Sc, 2 y M.Sc. 6.5 y Ph.D.) so all of the support and encouragement from Dad and Helen, Mom, Marion and Terry helped me persevere and finally finish this trip. I have to thank all the buds back in Ontario as well - Jerome, Jacko, Rob, and Gerry thanks for taking me out for beers over the years. To Kelly: I am grateful for all of your help, encouragement, and understanding over the years which helped me stay the course for this degree. I would like to acknowledge my research advisor, Dr. Chris Orvig, for giving me the freedom to venture in whatever direction seemed fit on this project, for his optimism and enthusiasm throughout the project, and for making sure the group gets together every now and then to celebrate their research. Thanks to Dr. Kathie Thompson for all her help, discussions, and effort over the years. I would also like to acknowledge Dr. Harvey Schugar for all his help and encouragement during his stay at UBC and thereafter. Dr. Robert Stick's help with carbohydrate synthetic details and discussion were greatly appreciated. Where would I be without all the help from the Orvig group? Thanks to everyone-Barry, Tim, Mike, Leon, Alex, Cara, Ika, Kathy, Vishakha, Cheri, Chuck, Sherwin, etc., for all the good times and fun over the years. I also would like to thank Ms. Liane Darge (NMR), Dr. Nick Burlinson (NMR), Dr. Brian Patrick (X-ray), Marshall Lapawa (MS), and Dr. Yun Ling (MS) for all their expertise and help. Dr. Dave Kennedy's assistance with cell studies was also much appreciated. xv C H A P T E R 1 INTRODUCTION 1.1 Coordination Complexes in Medicinal Chemistry One of the great challenges of medicinal inorganic chemistry involves understanding and altering the transport of metal ions in vivo. The transport of metal ions comprises many concepts including the introduction, removal, distribution, and targeting of metal ions (native and non-native) in biological systems in order to achieve a desired effect. To facilitate metal ion transport, ligands are often chelated to a metal and used to impart their physical properties to their metal complexes. Ligands can be designed to alter the lipophilicity, solubility, stability, size, charge, and biodistribution of the metal complexes. Functionalization of ligands with biologically active molecules can increase the specificity and localization of metal complexes to target organs or receptors in vivo.1 This thesis describes some possible roles of carbohydrates to direct either metal ions or pro-drugs of metal ion chelating agents. 1.2 Carbohydrates and Glucose Transport Carbohydrates, the most abundant class of organic molecules, are essential to many life functions, including their role as a primary fuel source. The roles of carbohydrates are still being elucidated and refined as the understanding of the 1 interactions of carbohydrate-binding proteins (lectins) with carbohydrates are further understood. These interactions play pivotal roles in immune response, cell adhesion, malignancy and apoptosis. Carbohydrates are precisely defined as hydrates of carbon with the formula C x(H20) y ; however, the term generally refers to polyhydroxy carbon-based compounds and their derivatives. The most basic subunits of carbohydrates in nature consist of 5-membered furanose or 6-membered pyranose rings that can be monomeric (examples shown below), oligomeric or polymeric. O H HO O H H O O H D-Glucopyranose D-Ribofuranose (D-Glucose) Of all the carbohydrates, glucose is the primary energy source of the human body. The glucose transport system is a vital pathway to deliver glucose across membranes and into cells. There are facilitative (GLUT) and sodium dependent (SGLT) glucose transport pathways. Up to 6 sodium dependent glucose transporters are known (SGLT1-6) whose main functions are involved with the intestinal absorption of glucose.4'5 There are 13 facilitative GLUT transporters known: GLUT1-GLUT12 and the proton/myoinositol co-transporter, HMIT. 4 , 6 ' 7 All 13 transporters consist of 12 membrane-spanning helices with most transporters being monosaccharide selective for the transport of D-glucose. The well known exceptions are the predominant transport of fructose by GLUT5 and HMIT's transport of myoinositol.7 2 Glucose is a polar molecule and relies on glucose transporters to cross lipid bilayers. The blood-brain barrier (BBB) utilizes these transporters for delivery of glucose to the brain as its energy source. GLUT1 represents essentially all the facilitative transporters at the BBB 8 ' 9 and 15-30%10 of the glucose transporters in the brain are of the GLUT 1-type. A number of drugs, for instance anti-human immunodeficiency virus (anti-HIV) drugs" such as azidothymidine (AZT), would be more effective if they had better access to the brain. To enter the brain, drugs are typically required to be neutral and lipophilic, with molecular weights of -700 or less. With these criteria the drugs may be able to passively diffuse across the blood-brain barrier (BBB) to enter the brain. Hydrophilic molecules are able to penetrate the BBB if they are actively or facilitatively transported into the brain. Linking glucose to form new drugs/prodrugs increases their water solubility and may lead to the resulting glycoconjugates' utilization of GLUT1 transport for delivery to the central nervous system (CNS) or across cellular membranes. The glycoconjugate approach has been used by various groups,13"19 with improved access to the CNS shown for a number of compounds that have been tested for this purpose. With a large density of GLUT receptors at the BBB, linking glucose to ligands or metal complexes may be one avenue to assist these compounds in crossing the BBB or cellular membranes. Once in the CNS, the glycoconjugate could be enzymatically cleaved or metabolized to form the desired compound. For instance, glucose is metabolized to glucose-6-phospate by the hexokinases in the first step of glycolysis. There is also (3-glucosidase activity in the brain and other organs20'21 that has the potential to remove the glucose moiety. 3 1.3 Nuclear Medicine Nuclear medicine is a specialized branch of radiology that uses small quantities of radioactive substances (radiopharmaceuticals) in the diagnosis and treatment of disease. The radiopharmaceutical is either injected or inhaled and, in the case of diagnostic imaging, an external radiation detector quantifies the amount of radiation emitted from the entire body, tissue or organs. Time is required for localization of the radiopharmaceutical, clearance from other organs or blood so as to not interfere with the images, and for the imaging itself. Depending on the radiopharmaceutical, diagnostic imaging can provide physiological and biochemical information of disease pathology with the goal of aiding in the mitigation or prevention of disease in humans. Diagnostic imaging studies include the evaluation of metabolism and receptor function, bodily functions in diseased states, and how effective certain drug treatments may be. Two techniques are commonly used to generate diagnostic images; single photon emission computed tomography (SPECT) and positron emission tomography (PET). In SPECT imaging y-rays are collimated whereas in PET imaging coincidence detection is used to detect two 180° opposed 511 keV y-rays generated from a positron emitting radiopharmaceutical (PER). The detected emissions are reconstructed by computer to form three dimensional images of the desired target. The most common isotope used in diagnostic imaging is 9 9 m Tc (t>/2 = 6.0 h, y = 140 keV). Technetium-based radiopharmaceuticals are used in more than 85% of the diagnostic nuclear medicine scans done each year. 2 2 The 99Mo(t. /2 = 66 h)/9 9 mTc couple allows for use of the generator for about a week. A generator is a purification column 4 (typically with lead shielding) that allows the desired isotope to elute free from the parent nuclide. The parent nuclide ["M0O4] 2 " is easily separated from [99mTc04]" with an alumina column in its generator system. The availability of this generator is one reason for the popular use of this radioisotope. While SPECT imaging satisfies many needs, PET imaging is preferred in instances where higher spatial resolution is required and less time is available for scans. One of the most common PET isotopes used is 1 8F (ty2 = 1.8 h), while U C (t>/2 = 20 min), 1 3 N (fc/2 = 10 min), and 1 5 0 (ty2 = 2 min) have also been extensively studied for PET imaging applications.23 The short 2-20 minute half-lives for some of these isotopes are extremely limiting and typically require an on-site cyclotron for production. 18 The most widely used PET radiopharmaceutical is [ F]2-deoxy-2-fluoro-D-glucose (FDG).2 4 It is recognized by, and utilizes, GLUT receptors for transport across the BBB and into cells. FDG is commonly used for oncology to detect and evaluate tumours,25 in cardiology to assess myocardial viability,26 and in neurology to diagnose a number of conditions.27 In 1999, the US Food and Drug Administration (FDA) conditionally approved FDG in select applications for PET imaging and this is still the only approved carbohydrate-based diagnostic imaging agent. With only one FDA approved carbohydrate radiopharmaceutical, research into other carbohydrate based imaging agents is certainly warranted. OH OH i 8 F OH OH OH FDG 5 D-Glucose The production of 1 8 F (t>/2 = 1.8 h) is done using a cyclotron. With the relatively 1 o short half-life of F a nearby cyclotron ensures less decay of this isotope. However, satellite-PET centres (PET centres without a cyclotron) are provided with FDG on a daily basis from cyclotron production sites at relatively far distances. To broaden the applicability of PET, it may be useful to tap into some of the generator-based systems that produce PET radionuclides.29 PET generators for certain isotopes, such as the 6 8Ge-6 8Ga generator (vide infra), may be more convenient for satellite PET centers and could provide an alternative to daily shipments of shorter-lived isotopes such as 18F-based radiopharmaceuticals. 1.4 Gallium and Indium in Nuclear Medicine Isotopes of gallium and indium have been used in diagnostic nuclear medicine for decades. The gamma-emitting isotopes 6 7Ga (t/2 = 78 h, y = 93, 185, 300 keV) and 1 1'in (ty2 = 68 h, y = 245, 172 keV) are the most common gallium and indium radionuclides used in nuclear medicine. 67Ga-citrate is the most extensively used gallium radiopharmaceutical, used in the diagnostic imaging of soft tissue tumours, inflammation, and infection.30 Transferrin is a two-sited iron transport protein with a high plasma concentration (apo-transferrin has 50 uM of empty sites) that readily competes for Ga(III) and In(III) metal centres. As a result, the gallium centre in gallium citrate is displaced and forms a mono- or di-gallium-transferrin complex, that appears to be at least partially 31 responsible for its uptake in soft tissue tumours. 6 A more promising gallium radionuclide with ideal characteristics for generator 68 *+• 68 68 technology is the positron emitting Ga (ty2 = 68 min, P ) produced from a Ge/ Ga couple. The long half-life of 6 8Ge (ty2 = 280 d) gives the generator a life of at least a year and 6 8 Ga has the benefit of being used for PET rather than SPECT imaging. Indium has had decades of use in nuclear medicine with imaging of white blood cells for inflammation and infection, as well as platelets in thrombosis, using tris(8-hydroxyquinoline) complexes of " ' i n (fr/2 = 68 h, y = 245, 172 keV). 3 0 More recently, tumours are also being imaged with somatostatin-analogues using l u In . 3 2 A short peptide fragment is linked to DTPA (diethylenetriaminepentaacetic acid, Figure 1.3, page 21) which chelates ' " in , and the resulting radiopharmaceutical, '"ln-pentetreotide (Octreoscan™), was approved by the FDA in 1994. For better spatial resolution and to detect smaller tumours, PET imaging studies with somastatin-analogues of " 0 m In (t/2 = 69 min, P+) as well as 6 8 Ga have shown promise.33'34 Phase II clinical trials have demonstrated that higher repeated doses of "'ln-somatostatin have a therapeutic effect with Auger electron irradiation of tumour cells. 1.5 3-Hydroxy-4-pyridinone Complexes of Gallium and Indium Gallium and indium radiopharmaceuticals typically require a thermodynamically stable coordination environment to avoid demetallation by transferrin. Naturally, apo-transferrin has a high affinity for trivalent metal ions such as iron, gallium and indium (log Ki = 20.3 for Ga(III)36 and 18.7 for In(III)37). A number of gallium and indium radiopharmaceuticals such as '"ln-Octreoscan use aminocarboxylates such as 7 diethylenetriaminepentaacetic acid (DTPA) or triethylenetetraamine (TETA) (Figure 1.3, page 21) for chelation to help prevent demetallation by transferrin. The Orvig group has previously used 3-hydroxy-4-pyridinones to chelate gallium and indium.38"41 These ligands form neutral tris(ligand)gallium(III) and indium(III) complexes (Scheme 1.1) that predominate from pH -4.5-9 and they have high thermodynamic stabilities (log p3 -36-38 for Ga(III) and -31-33 for In(ffl)).38'39 O -N. + M 3 + R - J OH C H 3 C H 3 M = Ga, In Scheme 1.1: 3-Hydroxy-2-methyl-4-pyridinones and their chelation to trivalent metal ions such as Ga(III) or In(III). Biodistribution studies of some lipophilic aryl-pyridinato-0/Ga complexes have been investigated by our group and found to accumulate to some degree in the heart (dog, rabbit), liver (rabbit, mice, rat) and to a much lesser extent in the brain (rabbit) depending on the gallium complex and animal.42 Besides these complexes, there are relatively few examples of BBB-penetrating Ga or In radiopharmaceuticals that exist in nuclear medicine and no other examples of FDA approved carbohydrate-based PET radiopharmaceuticals besides FDG. 8 1.6 Carbohydrates and 3-Hydroxy-4-pyridinones for Chelating Metals For the concept of combining carbohydrate targeting and transport with radionuclides for development of metal-based carbohydrate imaging agents, one might simply complex the carbohydrate directly to the metal. While carbohydrates can bind and interact with metals,12'43 complexes with trivalent gallium and indium have very low thermodynamic stabilities compared to those of transferrin and would not remain intact in vivo, making them unsuitable candidates for diagnostic imaging. Also, depending on the number of carbohydrates/hydroxyl groups bound to the metal, the charge may not be ideal for targeting certain organs or the directing ability of the carbohydrate may be compromised as hydrogen bonding interactions would disappear upon binding to a metal centre. Incorporation of the carbohydrate moiety into a ligand framework that has a strong affinity for metals such as gallium and indium is a highly viable alternative. Strategies for the linking of glucose to 3-hydroxy-4-pyridinones for use as radiopharmaceuticals are discussed in Chapter 2. 1.7 Oxidative Stress in the Central Nervous System 1.7.1 General Principles The central nervous system (CNS) requires energy to maintain its function as the command center of the body. Glucose provides this energy as it is metabolized. The high rate of metabolic activity can also consume a considerable amount of oxygen. Some 9 unavoidable by-products of this mitochondrial respiration (and other biological processes) include superoxide (O2""), hydroxyl radicals (*OH), and hydrogen peroxide (H2O2), which are collectively termed reactive oxygen species (ROS). ROS can also be generated via Fenton-type processes involving redox active metal ions such as iron and copper.44"47 ROS readily react with lipids, proteins, carbohydrates, and nucleic acids, all interactions of which can lead to significant damage. Oxidative stress is reduced by a number of defense mechanisms utilizing various antioxidants such as vitamin C , 4 8 vitamin E , 4 9 glutathione,48 thionein, superoxide dismutases (SODs), peroxidases, and catalases, all of which are used to neutralize the damage from ROS. 5 0 Increased exposure to ROS is thought to manifest itself in a number of CNS diseases. There is strong evidence that metal ion accumulation and oxidative stress both play roles in a number of chronic neurodegenerative diseases including Alzheimer's disease (AD) , 5 1 ' 5 2 amytrophic lateral sclerosis (ALS) , 5 3 ' 5 4 and prion diseases.55"57 The destruction of neurons is a . common feature of these presently incurable neurodegenerative disorders. 1.7.2 Alzheimer's Disease Approximately 5% of Canadians over the age of 65 have Alzheimer's Disease (AD) with an estimated 238 000 cases in 2001. 5 8 , 5 9 In 2001, approximately 83 200 people were newly diagnosed with dementia,58'60 about two-thirds of which were A D . The number of A D cases is expected to increase as the baby-boom population ages. Alzheimer's disease is a progressive neurodegenerative disorder that causes senile 10 dementia and eventually results in death. Symptoms of AD include the loss of memory, judgment and reasoning, difficulty with daily functions and changes in mood and behaviour. AD is often classified as early-onset AD (<65 years of age) or late-onset AD (over 65 years of age). Less than 5% of the total number of AD patients have early-onset AD with the remaining 95% of AD patients considered to have late-onset A D . 6 1 Early-onset AD is typically inherited as an autosomal dominant trait and is referred to as familial AD (FAD). Late-onset AD is more randomly developed, often without a clear family history, and can be referred to as sporadic AD. So far, three genes have been identified whose mutations are responsible for FAD: the amyloid precursor protein (APP) gene, the presenilin 1 (PSI) and presenilin 2 (PS2) genes. The exact roles of APP and the presenilins are currently being investigated, however, one role of the presenilins is in the cleavage of membrane-bound proteins. Numerous other genes are believed to increase the susceptibility of developing A D 6 1 with a major susceptibility gene being apolipoprotein E (APOE). APOE is mainly associated with sporadic late-onset AD and some of the possible functions of this protein include cholesterol transport and membrane repair. Two prominent histological features of AD include the accumulation of beta-amyloid plaques (neuritic plaques in Figure 1.1) and neurofibrillary tangles (NFT). These plaques are found in the extracellular space between neurons (neuropil) whereas NFT are found in the neuron itself and interfere with the function of the neuron. 11 Figure 1.1: Cartoon of AD pathologies. Beta-amyloid (neuritic) plaques and neurofibrillary tangles and their spatial arrangement within and around the neuron. There are three distinct morphologies of beta-amyloid (A/3) plaques: dense-cored, fibrillar, and diffuse.62 If a plaque is associated with clusters of abnormal neuronal processes (dystrophic neurites) the plaque is further designated as a neuritic plaque. Dense-cored plaques have a dense core of A/3 surrounded by a corona containing "wisps" of A/3 while fibrillar plaques resemble spoke-like fibril bundles directed from a central mass and diffuse plaques have diffuse, amorphous A/3 without distinct edges.62 There is some debate as to whether diffuse plaques are converted into fibrillar and dense-cored plaques or if each plaque type is independently formed. It appears that the proportion of diffuse plaques is higher in early stages of AD whereas fibrillar plaque forms predominate in late stages of the disease. The A/3 protein fragments originate from proteolytic degradation of APP that are found in normal brain tissue. 4 B- and 7-secretases cleave APP at different residues to produce Aft.40 and AB]A2 fragments (Scheme 1.2). ABU42 is considered the main 12 component of plaques followed by A(31.40 and there are smaller amounts of A[317.42 as well as C- 6 9 and N-truncated70 forms of the peptide. The "amyloid hypothesis"71"73 is a critical hypothesis describing a possible causal factor of AD. This hypothesis states that the deposition of AP protein (the main component of Ap plaques) is responsible for triggering a cascade of physiological events that contribute directly to the initiation and progression of AD. According to this hypothesis, other pathological markers such as the formation of NFTs, cell loss, and dementia are believed to result because of Ap deposition. There are numerous markers of ROS present in AD brain and the toxicity of AP is often considered to be due to oxidative stress.74 The generation of ROS could occur ., .... C-Terminal .asm i . ' " fragments Scheme 1.2:64 APP cleavage by the secretases to produce soluble APP fragments (APPsa and APPsP) as well as A p M 0 and APi . 4 2 . 13 because of Ap peptide interactions with redox active metal ions or perhaps via an indirect mechanism such as microglia activation75 that is triggered by Ap deposition in plaques.76,77 Determining which component(s) of Ap cause(s) the greatest toxicity/oxidative stress in AD is still a controversial issue. Potential toxic components can include one or any combination of the following: 1) any one of the varying length soluble A P monomers in the brain 2) soluble aggregates of Ap such as dimers and oligomers 3) precipitated/insoluble Ap in plaques 4) any combination of the above where a metal ion is coordinated in some fashion to the Ap peptide or oxidative modifications have taken place. In the absence of amyloid fibrils and monomeric Ap, soluble A P dimers and trimers were found to decrease long term potentiation at the synapse in rats whereas increases in monomeric A P reversed this effect.78 This result indicates that dimer and trimer (and perhaps oligomeric) A P components are toxic by themselves even in the absence of metal ions. However, there is also increasing evidence that ROS generated via redox metal ion-Ap peptide complexes are quite toxic leading to neurodegeneration in AD. Increased levels of iron, copper, and zinc are found in the amygdala neuropil of AD patients.79 Concentrations of approximately 1 mM for Fe and Zn, and 0.4 mM for Cu ions were estimated in Ap plaques compared to the age-matched control neuropil which had more than 2 times lower concentration of these metal ions.7 9'8 0 These same metal ions are able to avidly complex and induce aggregation of A P M O and especially AP1-42 81 83 which may explain their increased metal ion presence in AD brain. " Upon complexation and aggregation of Api . 4 0 or APi . 4 2 , redox active metal ions such as iron and copper have been shown'to be oxidized and generate ROS in the presence of 14 dioxygen in vitro.M'w Iron, and especially copper, aggregates of AP1.42 produced a larger amount of ROS compared to the corresponding AP1.40 aggregates, a significant finding considering AP1.42 is a main component of AP plaques. While there are discrepancies reported for the affinity between AP1.40/AP1.42 and metal ions, 8 1" 8 3 ' 9 1" 9 3 these are likely caused by confounding variables,8 1' 9 4' 9 5 and it is generally believed that physiologically relevant concentrations of copper, iron, and zinc present in A D brain is one pathway that leads to the aggregation of APi.4o/Api_4 2 9 1 Zinc-AP aggregates do not generate ROS; however, copper ions have been shown to displace zinc in its precipitates with AP1-40 or AP1.42, especially at a slightly acidic pH of 6.6,91 likely the case in an A D brain.9 6 The replacement of zinc for a redox active metal ion such as copper would only add to the oxidative burden found in A D brain. Zinc levels may be related to oxidative stress97 via apoptosis induction that is intertwined with ROS production9 8'9 9 and hence the homeostasis of this metal ion could be relevant to A D pathologies. Other hypotheses surrounding A D have recently surfaced that account for results 78 from a number of studies. The toxic nature of dimeric and trimeric AP has led to the "oligomeric A P " hypothesis 7 3' 1 0 0 which appears to implicate all non-monomeric soluble 78 forms of APi . 42 as causal factors of A D rather than the plaque forms of Ap. Lipid and cellular oxidative stress is increased if these oligomeric Ap complex metals and form soluble metal-AP complexes. Rather than being localized to specific areas such as Ap plaques, these soluble metal-AP complexes would be more mobile and perhaps better at generating oxidative damage in this manner. The potential for soluble metal-AP complexes to reside in membrane structures (Scheme 1.3) could enhance 15 © Lipid Peroxidation Cell Survival Cell Death Cell Survival Scheme 1.3:89 Oxidative stress that mediates Ap neurotoxicity may be partially due to the ability of Ap (AA) to shuttle redox-active transition metals such as iron to the cell membrane with consequent catalysis of redox-mediated alterations in membrane components, such as lipid peroxidation. Pretreatment of AP with desferrioxamine (DFO) reduces oxidative damage when iron is added to the system. membrane oxidative damage.89 The interaction with cellular membranes may also play a role in the process of removing excess metal ions from cells into the extracellular space. Determining the coordination environment of the metal ions and peptide interactions could give insight why AP aggregates and how this process may be disrupted. Typically, the His-6, Tyr-10, His-13, and His-14 are considered the main metal binding regions of Ap, however, other metal-binding amino acid side chains or the amino/carboxy terminus of AP could potentially bind to iron, copper, and zinc ions contributing to AP aggregation. Raman studies of in vitro synthesized metal-AP complexes have determined that insoluble Zn-AP (pH 5.8-7.4) and Cu-AP (pH 5.8-6.6) complexes are coordinated via a bridging His-M-His moiety101 and one proposed 16 arrangement is shown in Figure 1.2A. N M R and EPR studies concur with a metal-bridging histidine but propose 5 histidines and an oxygen atom, perhaps from a tyrosine (Figure 1.2B) or a water/hydroxo ligand, are also used to coordinate the two metal centres. Soluble monomeric Cu- and Zn-Ap complexes were proposed by Miura et al.101 to have one histidine and 3 deprotonated amides bind the metal centre at pH 6.6-7.4 (Figure 1.2C). Another soluble monomeric Cu-Ap complex proposed by Curtain, and coworkers102 (Figure 1.2D) has His-6, Tyr-10, His-13, and His-14 coordinating the metal centre. Figure 1.2A-D:1 0 1'1 0 2 Proposed metal binding regions of metal-Ap complexes. A) one of the potential forms of precipitated Cu(II)-APi.4o (pH 5.8-6.6) or Zn(II)-APi.4 0 (pH 5.8-7.4) using 2 histidines from 2 strands of AP1.40. Tyr-10 is also implicated in binding to Cu(II). B) precipitated Cu(II)-AP models based on N M R and EPR studies of Api . 28 at pH 6.9. C) soluble Cu(II)-APi.4o model with 1 histidine and 3 amido ligands coordinated to the metal centre D) soluble Cu(II)-Ap models based on N M R and EPR studies of Apt. 28 at pH 6.9. 17 Precipitated and soluble Fe(III)-A(3 complexes were proposed to bind the metal centre through Tyr-10 and Glu-22/Asp-23 side chains103 while others indicate that the 3 histidines play roles in Fe(IH)-AP complexes.102 In general, there appear to be several possible binding modes of Ap that allow different metal ions to be complexed forming soluble and precipitated metal-Ap species in vitro. Part of these metal-Ap complexes may resemble portions of the structures of Ap present in AD brain either as soluble metal-AP complexes or those precipitated in Ap plaques. The most recent hypothesis surrounding AD has been termed the "bioflocculant hypothesis" which is more detailed than the original amyloid hypothesis in that it clarifies the roles of Ap in A D . 1 0 4 ' 1 0 5 AP1.42 and, to a lesser extent, A P M O are considered to play a neuroprotective role by forming precipitates with neurotoxins such as metal ions, excitatory neurotransmitters, and pathogens such as bacteria or viruses. As a bioflocculant, Ap binds and removes these materials to the extracellular space where they are deposited as AP plaques. Under normal brain function, microglia and/or astrocytes remove the Ap plaques.106 In an AD state, the neuroprotective defenses of the brain become overwhelmed and the removal of AP plaques is hampered or prevented. The presence of the plaques may disrupt neurites, increase inflammation, and decrease the flow and inter-cellular exchange of metabolites resulting in increased neuronal degeneration in the brain. 18 1.8 Potential Pharmacotherapies for Alzheimer's Disease There is presently no cure for AD and drugs that are currently prescribed only slow the progression of the disease. The most common drugs on the market are cholinesterase enzyme inhibitors107 that slow the breakdown of acetylcholine, a neurotransmitter deficient in AD that is important for learning and memory. These drugs are reported to stabilize symptoms of mild to moderate AD for 12 months or longer. Other potential therapies for AD may include the use of non-steroidal anti-inflammatory drugs (NSAIDs)1 0 9 (such as Ibuprofen) that are aimed at reducing inflammation associated with AD. Antioxidants such as vitamin E 1 1 0 have also been found to slow the progression of the disease. Efforts are being made to inhibit the P- and y-secretases to slow the production of Ap from APP. 1 1 1 Immunization against Ap is another possible strategy to remove amyloid plaques.112 Active immunization initially improved cognitive function to near normal levels indicating the impairment caused by Ap plaques may be reversible. Unfortunately, this approach resulted in an increasing number of encephalitis cases (brain inflammation) with 17 of 300 patients at last report,112 and clinical trials were halted.113 New immunization strategies continue to be 112 investigated. If soluble AP is an essential neuroprotective peptide (as in the bioflocculant hypothesis), inhibiting its biosynthesis or the complete removal of all forms of AP would halt this process for removing excess metal ions or toxins that reside in the brain. Over the long term this would result in more oxidative damage and cognitive decline. A desirable therapeutic strategy for AD could involve the dissolution of AP deposits, 19 containment of oxidative damage, while still allowing some metal-free monomeric A(3 to persist in the brain. With oxidative insult being one of the first signs in AD pathology,74 therapeutics that reduce ROS could be useful in treating AD. While traditional antioxidants could help to quench ROS, metal chelation therapy may provide a means to passivate metal ions that catalyze ROS production. Removal of metals from Ap could be accomplished with chelators and this would allow their redistribution to other tissues or perhaps even to the systemic region for excretion. Appropriate metal chelators may break metal-Ap interactions solubilizing Ap plaques and still leave some soluble monomeric AP present for its neuroprotective function. Indications that metal chelation could solubilize AP plaques came from one of the first groups to purify Ap monomers from the cores of dense-cored AP plaques of AD brain (although it was not realized at the time). Repeated washing of Ap plaques with a solution containing 10% 2-mercaptoethanol (a bidentate chelator) serendipitously achieved partial solubilization of the of Ap plaques.114 Other evidence shows that metal ions play pivotal roles in the etiology of AD and chelating agents (Figure 1.3) may reduce harmful effects. The iron chelator desferrioxamine (DFO) was found to slow the clinical progression of dementia associated with A D . 1 1 5 DFO systemically depletes metal ions as it does not cross the BBB. Unfortunately, it requires painful intramuscular injections that make compliance difficult. The copper chelator D-penicillamine (D-pen) was found to reduce oxidative damage and systemic copper levels in AD patients"6,117 as well as to delay the onset and extend the 53 * 116 118 lifespan in an ALS mouse model." D -pen has a number of serious adverse effects. ' 20 Treatment of Wilson's disease and rheumatoid arthritis with D-pen is common; however, these patients have a high risk of worsening neurological symptoms119'120 which may exacerbate AD neurodegeneration. Other chelators and their effects on Ap plaques in vitro or synthetic metal-AP aggregates have been studied with promising results but some drawbacks. N,N,N',N'-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN), 1 2 1 ' 1 2 2 bathocuproine (BC), 1 2 1 ' 1 2 3 bathophenanthroline (BP), 1 2 3 TETA, 1 2 3 ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid (EGTA) and DTPA are chelators that have been considered (Figure 1.3). As little as 4 uM of BC, TPEN, or EGTA was found to resolubilize Ap levels in post-mortem AD tissue.121 However, TPEN is a highly toxic membrane D-Pen T E T A T P E N Hcq BC (R=CH 3) BP (R=H) Figure 1.3: Structures of metal chelators used to break metal-AP interactions. 21 permeable chelator.124,125 Bathocuproine and other phenanthroline based chelators are able to decrease A(3 levels in post-mortem AD brain; however, they form charged copper complexes which may make redistribution of metal ions difficult. Normally aggregation of AP1.42 occurs in the presence of Cu(II), but no aggregation occurred with the addition of the copper chelator, DTPA. 8 1 E G T A 1 2 1 , 1 2 6 and DTP A are also favourable calcium chelators and therefore are not selective for iron or copper ions and typically do not penetrate membranes well. The most promising metal chelator for AD therapy is clioquinol (5-chloro-7-iodo-8-hydroxyquinoline, Hcq), a bidentate chelator.127 Previously used as an antibiotic, Hcq has completed phase II clinical trials for AD treatment.128 Hcq is commonly referred to as a copper and zinc chelator; however, 8-hydroxyquinolines actually have a higher affinity for trivalent metal ions (e.g. Fe(III)) than divalent metal ions (e.g. Cu(II), Zn(II)). Although antibiotic doses are higher compared to those of an AD chelator (750 mg/day or more compared to 20 or 80 mg/day)130, use of Hcq has been linked to severe neurological degeneration perhaps related to vitamin B-12 deficiency.131"133 The cq-iron complex Fe^ in\cq)3 was found to be cytotoxic to retinal cells at 50 uM, perhaps a higher concentration than would be required for AD treatment.134 In another metal-clioquinol study, Hcq was complexed to zinc, copper, calcium, and magnesium ions with only the cq-Zn complexes (7.5 ug/mL Hcq, 50 uM ZnCb) being cytotoxic as a result of mitochondria damage in neural crest-derived melanoma cells.135 Despite these potential drawbacks, Hcq has fared well in the dissolution of Ap in post-mortem AD brain tissue and in an AD transgenic mouse model, likely via metal chelation and/or intercalation of Ap . 1 2 7 It appears that Hcq is better at reducing the size of diffuse AP 22 plaques (from mice studies), which predominate in early stages of AD, although there is some reduction of denser plaques as well. 1 2 8 In early clinical trials, treatment of 20 AD patients for 3 weeks showed no side-effects, and some improvements in cognitive decline 130 were observed. These are promising results and give hope that this and other types of compounds may be useful for metal chelation therapy for AD. With the potential drawbacks of some chelators mentioned above, the development of other types of chelators, such as 3-hydroxy-4-pyridinones, are warranted for strategies utilizing this therapy for neurodegenerative disorders. 1.9 Thesis Overview This introductory chapter described background related to radiopharmaceuticals for diagnostic imaging as well as a number of key players involved with the progression of AD and potential therapeutic strategies. In this thesis, carbohydrate-appended linkages of 3-hydroxy-4-pyridinones for potential use in directing radiopharmaceuticals of gallium and indium are discussed. Also, a series of pro-drugs of 3-hydroxy-4-pyridinones that may be useful for chelation therapy related to AD or neurodegenerative disorders are discussed. Both of these topics involve 3-hydroxy-4-pyridinones that are being altered by the addition of a carbohydrate moiety and their end-use would be in the chelation of metals and altering their biodistribution. This is predominantly a thesis in synthetic chemistry. The syntheses, purifications and characterizations were generally long and involved. Chapter 2 describes a number of carbohydrate-bearing gallium and indium complexes that were 23 synthesized and may have potential as radiopharmaceuticals. Chapter 3 documents the synthesis of a series of pro-drugs of 3-hydroxy-4-pyridinones. The removal of the pro-drug glucose moiety is also studied in attempts to discern whether the pro-drugs may be activated in vivo. Also, the ability of the pyridinones to quench free radicals is considered to determine potential benefits of this therapeutic strategy. The quenching of radicals may decrease oxidative damage associated with AD. Finally, Chapter 4 discusses where some future directions are headed with these projects. 1.10 References (1) Liu, S.; Edwards, D. S. Chem. Rev. 1999, 99, 2235-2268. (2) Nelson, D. L.; Cox, M. M. In Lehninger Principles of Biochemistry; 3rd ed.; Worth: New York, 2000, Chapter 9. (3) Gabius, H. J.; Andre, S.; Kaltner, H.; Siebert, H. C. Biochim. Biophys. Acta 2002, 1572, 165-177. (4) Wood, I. S.; Trayhurn, P. Br. J. Nutr. 2003, 89, 3-9. (5) Wright, E. M . Annu. Rev. Physiol. 1993, 55, 575-589. (6) Gould, G. W.; Holman, G. D. Biochem. J. 1993, 295, 329-341. (7) Joost, H. G.; Thorens, B. Mol. Membr. Biol. 2001,18, 247-256. (8) Pardridge, W. M. ; Boado, R. J.; Farrell, C. R. J. Biol. Chem. 1990, 265, 18035-18040. (9) Takata, K.; Kasahara, T.; Kasahara, M.; Ezaki, O.; Hirano, H. Biochem. Biophys. Res. Commun. 1990,173, 67-73. (10) McAllister, M . S.; Krizanac-Bengez, L.; Macchia, F.; Naftalin, R. J.; Pedley, K. C ; Mayberg, M. R.; Marroni, M. ; Leaman, S.; Stanness, K. A.; Janigro, D. Brain Res. 2001, 904, 20-30. 24 (11) Vlieghe, P.; Bihel, F.; Clerc, T.; Pannecouque, C ; Witvrouw, M. ; De Clercq, E.; Salles, J. P.; Chermann, J. C ; Kraus, J. L. J. Med. Chem. 2001, 44, Ill-lib. (12) Gyurcsik, B.; Nagy, L. Coord. Chem. Rev. 2000, 203, 81-149. (13) Battaglia, G.; La Russa, M. ; Bruno, V.; Arenare, L.; Ippolito, R.; Copani, A.; Bonina, F.; Nicoletti, F. Brain Res. 2000, 860, 149-156. (14) Rouquayrol, M. ; Gaucher, W.; Greiner, J.; Aubertin, A. M. ; Vierling, P.; Guedj, R. Carbohydr. Res. 2001, 336, 161-180. (15) Bilsky, E. J.; Egleton, R. D.; Mitchell, S. A.; Palian, M. M. ; Davis, P.; Huber, J. D.; Jones, H.; Yamamura, H. I.; Janders, J.; Davis, T. P.; Porreca, F.; Hruby, V. J.; Polt, R. J. Med. Chem. 2000, 43, 2586-2590. (16) Shimada, M. ; Kawamoto, S.; Hirose, Y.; Nakanishi, M. ; Watanabe, H.; Watanabe, M. Histochem. J. 1994, 26, 207-212. (17) Fernandez, C.; Nieto, O.; Rivas, E.; Montenegro, G.; Fontenla, J. A.; Fernandez-Mayoralas, A. Carbohydr. Res. 2000, 327, 353-365. (18) Dumas, C.; Schibli, R.; Schubiger, P. A. J. Org. Chem. 2003, 68, 512-518. (19) Dunsaed, C. B.; Dornish, J. M. ; Aastveit, T. E.; Nesland, J. M. ; Pettersen, E. O. Anti-Cancer Drugs 1995, 6, 456-464. (20) McCarter, J. D.; Adam, M. J.; Hartman, N. G.; Withers, S. G. Biochem. J. 1994, 301, 343-348. (21) Matern, H.; Boermans, H.; Lottspeich, F.; Matern, S. J. Biol. Chem. 2001, 276, 37929-37933. (22) Jurisson, S. S.; Lydon, J. D. Chem. Rev. 1999, 99, 2205-2218. (23) Fowler, J. S.; Wolf, A. P. Acc. Chem. Res. 1997, 30, 181-188. (24) Adam, M. J. J. Label. Compd. Radiopharm. 2002, 45, 167-180. (25) Czernin, J. Acta Med. Austriaca 2002, 29, 162-170. (26) Segall, G. Nucl. Med. Commun. 2002, 23, 323-330. (27) Barrington, S. F. Nucl Med. Commun. 2000, 21, 237-240. 25 (28) Preusche, S.; Fuchtner, F.; Steinbach, J.; Zessin, J.; Krug, H.; Neumann, W. Appl. Radial hot. 1999, 51, 625-630. (29) Welch, M . J.; McCarthy, T. J. J. Nucl. Med. 2000, 41, 315-317. (30) Anderson, C. J.; Welch, M . J. Chem. Rev. 1999, 99, 2219-2234. (31) Luttropp, C. A.; Jackson, J. A.; Jones, B. J.; Sohn, M. H.; Lynch, R. E.; Morton, K. A. J. Nucl. Med. 1998, 39, 1405. (32) de Jong, M. ; Breeman, W. A. P.; Bakker, W. H.; Kooij, P. P. M. ; Bernard, B. F.; Hofland, L. J.; Visser, T. J.; Srinivasan, A.; Schmidt, M. A.; Erion, J. L.; Bugaj, J. E.; Macke, H. R.; Krenning, E. P. Cancer Res. 1998, 58, 437-441. (33) Lubberink, M. ; Tolmachev, V.; Widstrom, C ; Bruskin, A.; Lundqvist, H.; Westlin, J. E. J. Nucl. Med. 2002, 43, 1391-1397. (34) Hofmann, M . M. ; Boerner, A. R.; Maecke, H. R.; Otto, D.; Kalbacher, H.; Knapp, W. J. Nucl. Med. 2002, 43, 1252. (35) Anthony, L. B.; Woltering, E. A.; Espenan, G. D.; Cronin, M . D.; Maloney, T. J.; McCarthy, K. E. Semin. Nucl. Med. 2002, 32, 123-132. (36) Harris, W. R.; Pecoraro, V. L. Biochemistry 1983, 22, 292-299. (37) Harris, W. R.; Chen, Y.; Wein, K. Inorg. Chem. 1994, 33, 4991-4998. (38) Zhang, Z.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1991, 30, 509-515. (39) Clevette, D. J.; Lyster, D. M. ; Nelson, W. O.; Rihela, T.; Webb, G. A.; Orvig, C. Inorg. Chem. 1990, 29, 667-672. (40) Nelson, W. O.; Karpishin, T. B.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1988, 27, 1045-1051. (41) Matsuba, C. A.; Nelson, W. O.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1988, 27, 3935-3939. (42) Zhang, Z.; Lyster, D. M. ; Orvig, C. Nucl. Med. Biol. 1992,19, 327-335. (43) Burger, K.; Nagy, L. In Biocoordination Chemistry: Coordination Equilibria in Biologically Active Systems; Burger, K., Ed.; Horwood: New York, 1990, Chapter 6, p 236. (44) Hensley, K.; Floyd, R. A. Arch. Biochem. Biophys. 2002, 397, 377-383. 26 (45) Stadtman, E. R. Ann. N.Y.Acad. Sci. 2000, 928, 22-38. (46) Stadtman, E. R. Free Radic. Biol. Med. 1990, 9, 315-325. (47) Halliwell, B.; Gutteridge, J. M . Biochem. J. 1984, 219, 1-14. (48) Paterniti, S.; Cambria, R.; Fiore, P.; Falcone, M. F.; Levita, A.; Gravina, G.; Costanzo, T.; Cambria, M . G. Neuropsicofarmacol. 2002, 24, 187-193. (49) Yatin, S. M. ; Varadarajan, S.; Butterfield, D. A. J. Ah. Dis. 2000, 2, 123-131. (50) Esposito, E.; Rotilio, D.; Di Matteo, V.; Di Giulio, C ; Cacchio, M. ; Algeri, S. Neurobiol. Aging 2002, 23, 719-735. (51) Christen, Y. Am. J. Clin. Nutr. 2000, 71, 621S-629S. (52) Perry, G.; Taddeo, M . A.; Petersen, R. B.; Castellani, R. J.; Harris, P. L. R.; Siedlak, S. L.; Cash, A. D.; Liu, Q.; Nunomura, A.; Atwood, C. S.; Smith, M . A. Biometals 2003, 76,77-81. (53) Hottinger, A. F.; Fine, E. G.; Gurney, M. E.; Zurn, A. D.; Aebischer, P. Eur. J. Neurosci. 1997, 9, 1548-1551. (54) Jung, C.; Rong, Y.; Doctrow, S.; Baudry, M. ; Malfroy, B.; Xu, Z. Neurosci. Lett. 2001,304, 157-160. (55) McMahon, H. E.; Mange, A.; Nishida, N.; Creminon, C.; Casanova, D.; Lehmann, S. J. Biol. Chem. 2001, 276, 2286-2291. (56) Wong, B.-S.; Chen, S. G.; Colucci, M.; Xie, Z.; Pan, T.; Liu, T.; Li , R.; Gambetti, P.; Sy, M.-S.; Brown, D. R. J. Neurochem. 2001, 78, 1400-1408. (57) Kim, N . H.; Park, S. J.; Jin, J. K.; Kwon, M. S.; Choi, E. K.; Carp, R. I.; Kim, Y. S. Brain Res. 2000, 884, 98-103. (58) www.alzheimer.ca. (59) C.S.H.A.W.G. Can. Med. Assoc. J. 1994,150, 899-913. (60) C.S.H.A.W.G. Neurology 2000, 55, 66-73. (61) Rocchi, A.; Pellegrini, S.; Siciliano, G.; Murri, L. Brain Res. Bull. 2003, 61, 1-24. (62) Dickson, T. C ; Vickers, J. C. Neuroscience 2001,105, 99-107. (63) Armstrong, R. A. Dement. Geriatr. Cogn. Disord. 1998, 9, 227-238. 27 (64) Annaert, W.; de Strooper, B. Annu. Rev. Cell Dev. Biol. 2002,18, 25-51. (65) Roher, A. E.; Lowenson, J. D.; Clarke, S.; Wolkow, C ; Wang, R.; Cotter, R. J.; Reardon, I. M. ; Zurcher-Neely, H. A.; Heinrikson, R. L.; et al. J. Biol. Chem. 1993, 268, 3072-3083. (66) Iwatsubo, T.; Odaka, A.; Suzuki, N. ; Mizusawa, H.; Nukina, N.; Ihara, Y. Neuron 1994,13, 45-53. (67) Iizuka, T.; Shoji, M. ; Harigaya, Y.; Kawarabayashi, T.; Watanabe, M. ; Kanai, M. ; Hirai, S. Brain Res. 1995, 702, 275-278. (68) Gowing, E.; Roher, A. E.; Woods, A. S.; Cotter, R. J.; Chaney, M. ; Little, S. P.; Ball, M. J. J. Biol. Chem. 1994, 269, 10987-10990. (69) Wiltfang, J.; Esselmann, H.; Bibl, M. ; Smirnov, A.; Otto, M. ; Paul, S.; Schmidt, B.; Klafki, H. W.; Maler, M. ; Dyrks, T.; Bienert, M. ; Beyermann, M. ; Ruther, E.; Kornhuber, J. J. Neurochem. 2002, 81, 481-496. (70) Vigo-Pelfrey, C ; Lee, D.; Leim, P.; Lieberburg, I.; Schenk, D. B. J. Neurochem. 1993, 61, 1965-1968. (71) Hardy, J. A.; Higgins, G. A. Science 1992, 256, 184-185. (72) Sommer, B. Curr. Opin. Pharmacol. 2002, 2, 87-92. (73) Hardy, J.; Selkoe, D. J. Science 2002, 297, 353-356. (74) Nunomura, A.; Perry, G.; Aliev, G.; Hirai, K.; Takeda, A.; Balraj, E. K.; Jones, P. K.; Ghanbari, H.; Wataya, T.; Shimohama, S.; Chiba, S.; Atwood, C. S.; Petersen, R. B.; Smith, M. A. J. Neuropathol. Exp. Neurol. 2001, 60, 759-767. (75) Colton, C. A.; Chernyshev, O. N.; Gilbert, D. L.; Vitek, M. P. Ann. NY. Acad. Sci. 2000, 899, 292-307. (76) Sayre, L. M. ; Zagorski, M . G.; Surewicz, W. K.; Krafft, G. A.; Perry, G. Chem. Res. Toxicol. 1997,10, 518-526. (77) Walsh, D. T.; Montero, R. M. ; Bresciani, L. G.; Jen, A. Y. T.; Leclercq, P. D.; Saunders, D.; El-Amir, A. N. ; Gbadamoshi, L.; Gentleman, S. M. ; Jen, L.-S. Neurobiol. Dis. 2002,10, 20-27. (78) Walsh, D. M. ; Klyubin, I.; Fadeeva, J. V.; Cullen, W. K.; Anwyl, R.; Wolfe, M. S.; Rowan, M. J.; Selkoe, D. J. Nature 2002, 416, 535-539. 28 (79) Lovell, M. A.; Robertson, J. D.; Teesdale, W. J.; Campbell, J. L.; Markesbery, W. R. J. Neurol. Sci. 1998,158, 47-52. (80) Cuajungco, M . P.; Faget, K. Y.; Huang, X.; Tanzi, R. E.; Bush, A. I. Ann. NY. Acad. Sci. 2000, 920, 292-304. (81) Atwood, C. S.; Scarpa, R. C ; Huang, X.; Moir, R. D.; Jones, W. D.; Fairlie, D. P.; Tanzi, R. E.; Bush, A. I. J. Neurochem. 2000, 75, 1219-1233. (82) Bush, A. I.; Pettingell, W. H.; Multhaup, G.; Paradis, M. d.; Vonsattel, J.-P.; Gusella, J. F.; Beyreuther, K.; Masters, C. L.; Tanzi, R. E. Science 1994, 265, 1464-1467. (83) Garzon-Rodriguez, W.; Yatsimirsky, A. K.; Glabe, C. G. Bioorg. Med. Chem. Lett. 1999, 9, 2243-2248. (84) Huang, X.; Cuajungco, M. P.; Atwood, C. S.; Hartshorn, M . A.; Tyndall, J. D. A.; Hanson, G. R.; Stokes, K. C.; Leopold, M.; Multhaup, G.; Goldstein, L. E.; Scarpa, R. C ; Saunders, A. J.; Lim, J.; Moir, R. D.; Glahe, C ; Bowden, E. F.; Masters, C. L.; Fairlie, D. P.; Tanzi, R. E.; Bush, A. I. J. Biol. Chem. 1999, 274, 37111-37116. (85) Huang, X.; Atwood, C. S.; Hartshorn, M. A.; Multhaup, G.; Goldstein, L. E.; Scarpa, R. C ; Cuajungco, M . P.; Gray, D. N. ; Lim, J.; Moir, R. D.; Tanzi, R. E.; Bush, A. I. Biochemistry 1999, 38, 7609-7616. (86) Lynch, T.; Cherny, R. A.; Bush, A. I. Exp. Gerontol. 2000, 35, 445-451. (87) Smith, M. A.; Harris, P. L. R.; Sayre, L. M. ; Perry, G. Proc. Natl. Acad. Sci. U. S. A. 1997, 94, 9866-9868. (88) Tabner, B. J.; Turnbull, S.; El-Agnaf, O. M. A.; Allsop, D. Free Radic. Biol. Med. 2002, 32, 1076-1083. (89) Rottkamp, C. A.; Raina, A. K.; Zhu, X.; Gaier, E.; Bush, A. I.; Atwood, C. S.; Chevion, M. ; Perry, G.; Smith, M . A. Free Radic. Biol. Med. 2001, 30, 447-450. (90) Schubert, D.; Chevion, M . Biochem. Biophys. Res. Commun. 1995, 216, 702-707. (91) Atwood, C. S.; Moir, R. D.; Huang, X.; Scarpa, R. C ; Bacarra, N . M. ; Romano, D. M.; Hartshorn, M. A.; Tanzi, R. E.; Bush, A. I. J. Biol. Chem. 1998, 273, 12817-12826. (92) Maggio, J. E.; Esler, W. P.; Stimson, E. R.; Jennings, J. M. ; Ghilardi, J. R.; Mantyh, P. W. Science 1995, 268, 1920-1921. (93) Esler, W. P.; Stimson, E. R.; Jennings, J. M. ; Ghilardi, J. R.; Mantyh, P. W.; Maggio, J. E. J. Neurochem. 1996, 66, 723-732. 29 (94) Turnbull, S.; Tabner, B. J.; El-Agnaf, O. M. A.; Twyman, L. J.; Allsop, D. Free Radic. Biol. Med. 2001, 30, 1154-1162. (95) Monji, A.; Utsumi, H.; Ueda, T.; Imoto, T.; Yoshida, I.; Hashioka, S.; Tashiro, K.-I.; Tashiro, N . J. Neurochem. 2001, 77, 1425-1432. (96) Yates, C. M. ; Butterworth, J.; Tennant, M. C ; Gordon, A. J. Neurochem. 1990, 55, 1624-1630. (97) Watjen, W.; Haase, H.; Biagioli, M. ; Beyersmann, D. Environ. Health Per sped. 2002,110, 865-867. (98) Provinciali, M. ; Donnini, A.; Argentati, K.; Di Stasio, G.; Bartozzi, B.; Bernardini, G. Free Radic. Biol. Med. 2002, 32, 431-445. (99) Pong, K.; Rong, Y. Q.; Doctrow, S. R.; Baudry, M. Brain Res. 2002, 950, 218-230. (100) Robinson, S. R.; Bishop, G. M . Science 2002,298, 962-963. (101) Miura, T.; Suzuki, K.; Kohata, N. ; Takeuchi, H. Biochemistry 2000, 39, 7024-7031. (102) Curtain, C. C ; Ali , F.; Volitakis, I.; Cherny, R. A.; Norton, R. S.; Beyreuther, K.; Barrow, C. J.; Masters, C. L.; Bush, A. I.; Barnham, K. J. J. Biol. Chem. 2001, 276, 20466-20473. (103) Miura, T.; Suzuki, K.; Takeuchi, H. J. Mol. Struct. 2001, 598, 79-84. (104) Bishop, G. M. ; Robinson, S. R. Neurobiol. Aging 2002, 23, 1101-1105. (105) Robinson, S. R.; Bishop, G. M. Neurobiol. Aging 2002, 23, 1051-1072. (106) Bishop, G. M. ; Robinson, S. R. J. Neural Transm. 2003,110, 381-400. (107) Gauthier, S. Can. Med. Assoc. J. 2002,166, 616-623. (108) Winblad, B.; Brodaty, H.; Gauthier, S.; Morris, J. C ; Orgogozo, J. M. ; Rockwood, K.; Schneider, L.; Takeda, M. ; Tariot, P.; Wilkinson, D. Int. J. Geriatr. Psych. 2001,16, 653-666. (109) Stewart, W.; Kawas, C ; Corrada, M. ; Metter, E. Neurology 1997, 48, 626-632. (110) Sano, M. ; Ernesto, C ; Thomas, R. G.; Klauber, M. R.; Schafer, K.; Grundman, M.; Woodbury, P.; Growdon, J.; Cotman, D. W.; Pfeiffer, E.; Schneider, L. S.; Thai, L. J. N. Engl. J. Med. 1997, 336, 1216-1222. 30 (111) Wolfe, M. S. Nat. Rev. DrugDiscov. 2002,7, 859-866. (112) Schenk, D. Nat. Rev. Neurosci. 2002, 3, 824-828. (113) Senior, K. Lancet Neurol. 2002, 7,3. (114) Masters, C. L.; Simms, G.; Weinman, N. A.; Multhaup, G.; McDonald, B. L.; Beyreuther, K. Proc. Natl. Acad. Sci. USA 1985, 82, 4245-4249. (115) Crapper McLachlan, D. R.; Dalton, A. J.; Kruck, T. P.; Bell, M. Y.; Smith, W. L.; Kalow, W.; Andrews, D. F. Lancet 1991, 337, 1304-1308. (116) Squitti, R.; Rossini, P. M. ; Cassetta, E.; Moffa, F.; Pasqualetti, P.; Cortesi, M. ; Colloca, A.; Rossi, L.; Finazzi-Agro, A. Eur. J. Clin. Invest. 2002, 32, 51-59. (117) Rossi, L.; Squitti, R.; Pasqualetti, P.; Marchese, E.; Cassetta, E.; Forastiere, E.; Rotilio, G.; Rossini, P. M. ; Finazzi-Agro, A. Neurosci. Lett. 2002, 329, 137-140. (118) Wright, G. D.; Wilson, C ; Bell, A. L. Clin. Rheumatol. 1994,13, 80-82. (119) Brewer, G. J. Drugs 1995, 50, 240-249. (120) Boger, C. A.; Kellner, H. Aktuelle Rheumatol. 2001, 26, 285-290. (121) Cherny, R. A.; Legg, J. T.; McLean, C. A.; Fairlie, D. P.; Huang, X.; Atwood, C. S.; Beyreuther, K.; Tanzi, R. E.; Masters, C. L.; Bush, A. I. J. Biol. Chem. 1999, 274, 23223-23228. (122) Fonte, J.; Miklossy, J.; Atwood, C ; Martins, R. J. Ah. Dis. 2001, 3, 209-219. (123) Cherny, R. A.; Barnham, K. J.; Lynch, T.; Volitakis, I.; L i , Q. X.; McLean, C. A.; Multhaup, G.; Beyreuther, K.; Tanzi, R. E.; Masters, C. L.; Bush, A. I. J. Struct. Biol. 2000,130, 209-216. (124) Adler, M. ; Dinterman, R. E.; Wannemacher, R. W. Toxicon 1997, 35, 1089-1100. (125) Chimienti, F.; Seve, M. ; Richard, S.; Mathieu, J.; Favier, A. Biochem. Pharmacol. 2001,62,51-62. (126) Riley, W., Jr; Pfeiffer, D. J. Biol. Chem. 1986, 261, 28-31. (127) Cherny, R. A.; Atwood, C. S.; Xilinas, M . E.; Gray, D. N. ; Jones, W. D.; McLean, C. A.; Barnham, K. J.; Volitakis, I.; Fraser, F. W.; Kim, Y. S.; Huang, X. D.; Goldstein, L. E.; Moir, R. D.; Lim, J. T.; Beyreuther, K ; Zheng, H.; Tanzi, R. E.; Masters, C. L.; Bush, A. I. Neuron 2001, 30, 665-676. 31 (128) Dorrell, S. Drug Discov. Today 2001, 6, 61-62. (129) Martell, A. E.; Smith, R. M . Critical Stability Constants, Vols. 1-6; Plenum: New York, NY, 1974-1989. (130) Regland, B.; Lehmann, W.; Abedini, I.; Blennow, K.; Jonsson, M. ; Karlsson, I.; Sjogren, M. ; Wallin, A.; Xilinas, M. ; Gottfries, C. G. Dement. Geriatr. Cogn. Disord. 2001,12, 408-414. (131) Tabira, T. Science 2001, 292, 2251-2251. (132) Bush, A. I.; Masters, C. L. Science 2001, 292, 2251-2252. (133) Yassin, M . S.; Ekblom, J.; Xilinas, M. ; Gottfries, C. G.; Oreland, L. J. Neurol. Sci. 2000,173, 40-44. (134) Ohtsuka, K.; Ohishi, N. ; Eguchi, G.; Yagi, K. Experientia 1982, 38, 120-122. (135) Arbiser, J. L . ; Kraeft, S. K.; van Leeuwen, R.; Hurwitz, S. J.; Selig, M. ; Dickersin, G. R.; Flint, A.; Byers, H. R.; Chen, L . B. Mol. Med. 1998, 4, 665-670. 32 C H A P T E R 2 GALLIUM(III) and INDIUM(III) C O M P L E X E S of C A R B O H Y D R A T E -B E A R I N G 3-HYDROXY-4-PYRIDINONES 2.1 Introduction To further expand on the types of gallium and indium radiopharmaceuticals, metal complexes stable under in vivo conditions are needed. Currently, gallium citrate is the most used gallium radiopharmaceutical for diagnostic imaging; however, the coordinating citrate ligands are readily displaced allowing transferrin to complex the gallium(III) ions, thus altering the biodistribution of the isotope 6 7Ga. For the metal complexes to remain intact, chelators with high stability constants such as 8-hydroxyquinoline (Figure 2.1) have previously been used to complex gallium and indium. More recent efforts include the use of a pendant peptide linked to DTPA (pentetreotide, Figure 2.1), which complexes gallium or indium. This is a bifunctional chelate approach - the DTPA portion of the molecule chelates the gallium or indium metal center, and the peptide portion helps to direct the metal complex to somatostatin receptors in vivo. A recently FDA approved 1 1 ' in radiopharmaceutical (OctreoScan™) is being used for diagnostic imaging of somatostatin receptors. These receptors are overexpressed on neuroendocrine tumours, for instance. Besides DPTA, other chelators such as 3-hydroxy-4-pyridinones also have high stability constants with gallium(III) and 33 N H 2 DTPA-pentetreotide / DTPA-D-Phe-Cys-Phe \ 1 f TI \ (ol)-Thr-Cys-Lys / Figure 2.1: Ligands for gallium or indium in nuclear medicine applications. indium(III) ions. The pyridinone moiety also has a number of sites where it can be bifunctionalized with the addition of biomoleucles such as carbohydrates. The goal of this project was to synthesize and characterize carbohydrate-bearing pyridinone complexes of gallium and indium. The glucose-bearing ligands will likely direct the metal complexes in vivo. The corresponding radioactive gallium or indium complexes may have applications in diagnostic nuclear medicine related to metabolic 34 glucose imaging such as tumour, myocardial, or brain imaging or perhaps in imaging of GLUT receptors. With three equivalents of pyridinone, thermodynamically stable and coordinatively saturated tris(pyridinato)gallium and indium complexes form and these characteristics protect against demetallation by transferrin in vivo. 2.2 Experimental 2.2.7 Materials Water was deionized and organics were removed (Barnstead D9802 and D9804 cartridges) and this was followed by distillation with a Corning MP-1 Mega-Pure Still. All solvents were purchased from Fisher except for dry pyridine (Aldrich, "anhydrous", <0.003% H 20). MeOH, MeCN, and AcOEt solvents were HPLC grade and all other solvents were ACS grade or higher. Dry DMF was obtained by stirring for at least 1 h with 4 A molecular sieves (flame dried under vacuum) followed by distillation at -40-50 °C (0.5 mbar). Other anhydrous solvents that were required were dried according to standard procedures.1 Chemicals were from Aldrich (Madison, WI) unless otherwise stated. 3-Hydroxy-2-methyl-4-pyranone (maltol, 1) was obtained from Pfizer (Kirkland, PQ). H 2 and N 2 gases were from Praxair. Palladium black was obtained from Alfa Aesar. Agrobacterium sp. (3-glucosidase2 (Abg) was generously provided by Prof. S.G. Withers (UBC). Thin layer chromatography (TLC) was performed using silica TLC plates with aluminum backing (Merck). Silica for column chromatography was from 35 Silicycle (Quebec City, PQ). Sephadex G10 was allowed to swell in the solvent system being used at least 2 h before column packing. TLC plates were developed with UV light or other selected reagents such as ninhydrin spray (-0.2% wt/wt in EtOH) followed by heating (heat gun), a solution of FeCl3 (FeCL/6H20 appropriately diluted for TLC visualization with water and adjusted to -pH 2 with 1 M HC1), and a sulfuric acid spray (5% vol/vol H2SO4 in EtOH) followed by charring (heat gun). Nylon filters were from Gelman (25 mm, 0.2 um). Al l 10% Pd/C was wetted first with water before contact with other solvents. MeOH and even EtOH can ignite on contact with dry Pd/C. Latex balloons (Aldrich) used for hydrogenation reactions were -11 L when filled to 80% capacity and were discarded after each reaction. 2.2.2 Instrumentation All volumes 2 mL and less that did not require syringing were typically measured with the appropriate adjustable Gilson Pipetman (PI000, P200, or P20 model). Mixtures of solvents such as 1:1 MeOH:water are volume:volume measurements unless otherwise stated. The pH was measured using an Accumet 950 pH/ion Meter with a Metrohm Porotrode micro-combination glass electrode. A Parr series 4750 (200 mL) split ring high pressure reactor was used for higher pressure hydrogenations (rated for use up to -200 atm.) with an external stir plate. A thick walled glass insert (test-tube shaped) was made (UBC glassblowing) that inserted into the Parr apparatus and held 100 mL of solvent when full. Blast shields were set in front of all hydrogenation apparatuses. A very robust Edwards diaphragm pump with a variable pressure controller was attached to 36 the rotary evaporator allowing reduced pressures from -10-1000 mbar to be selected. Vacuum line pressure typically read 0.2-0.5 mbar. A vacuum desiccator was frequently used that contained phosphorus pentoxide (Fisher) as a drying agent. The desiccator was heavily wrapped with electrical tape and placed behind a blast shield to minimize potential implosion damage. NMR spectra of samples were recorded on a Bruker Avance 300, Avance 400, or A M X 500 instrument at room temperature (RT). NMR spectra were calibrated with the deuterated solvent and in the case where mixtures were used CD3OD was chosen as the reference. 2D NMR techniques such as COSY, HMQC and HMBC were used to aid in the characterization of all the final products as well as a number of intermediates. X-ray data for 8 were collected and processed3 by Dr. B.O. Patrick at UBC using a Rigaku/ADSC CCD diffractometer and are shown in Appendix 1. The X-ray structure was solved using direct methods4 and expanded using Fourier techniques.5 The non-hydrogen atoms were refined anisotropically. Some hydrogen atoms were refined isotropically, the rest were included in fixed positions. Al l calculations were performed using the teXan crystallographic software package.6 Mass spectra were obtained with either a Kratos Concept II H32Q instrument (Cs+-LSIMS) or a Macromass LCT (electrospray ionization, ESI). For ESI-MS, approximately 1 mg of sample was dissolved in 1 mL of solvent, typically a 1:1 mixture of methanol:water with no other additives added. Microanalyses for C, H, and N were done in this department (Mr. P. Borda or Mr. M. Lakha) or by Delta Microanalytical Services (Delta, BC). 37 2.2.3 Abbreviations for 3-Hydroxy-4-pyridinones Abbreviations for the pyridinones are typically 3-6 letter acronyms that sometimes originate from the name of the pyridinone, as with Hhpp (below). The upper case " H " starts the acronym and upon deprotonation/metal complexation it is removed from the acronym to indicate the analogous anion. Smaller case letters follow in the acronym for relatively simple pyridinones. For pyridinones with a pendant carbohydrate, all upper case letters are used and the letters of the acronym follow from the sugar, to the linker and the pyridinone as illustrated below for HOGBPP. linkage 3 -Hydroxy -1 -(4-hydroxypheny 1)-2-methyl-4(l H)-p_yridinone Hhpp H O G B P P For the H O G B P P acronym, the "O" refers to a carbohydrate with predominantly "OH" or hydroxyl groups on the sugar, in this case "G" for glucose. An "A" is used for predominantly acetylated sugars. The next letter of the acronym is for the linkage of the sugar, given by "B" indicating the beta-glucose linkage. In this work " A " refers to an alpha-linkage and "6" indicates linkage at the 6-position of the sugar. Next, the linker 38 between the sugar and pyridinone is given, where "P" denotes a phenyl linker, and finally a "P" for pyridinone is traditionally placed at the end of the acronym. 2.2.4 Syntheses 3-Benzyloxy-2-methyl-4-pyranone (2) (benzyl maltol). This compound was prepared in n a manner similar to a previously published that was pre-dissolved in 120 mL water with sonication. Benzyl chloride (160 mL, 1.39 mol) was added and all materials dissolved upon heating. Near reflux temperature, a white precipitate formed (NaCl). The reaction mixture was refluxed 6 h and TLC in 9:1 AcOEt:MeOH indicated some unreacted maltol at a slightly lower retardation factor (Rf) than the benzyl maltol spot. Additional benzyl chloride (40 mL, 0.35 mmol) and NaOH (2.5 g, 0.062 mmol) (pre-dissolved in 10 mL water) were added to the reaction mixture and refluxed for an additional 12 h. TLC indicated less maltol than before. The bright yellow solution (still containing white precipitate) was removed using a rotary evaporator (50°C), 700 mL water and 350 mL CH 2C1 2 were then added to the RBF. The contents of the RBF were transferred to a 2 L separatory funnel, the water layer was discarded and the CH2C12 was washed with 3x250 mL of 5% NaOH and 100 mL water. The CH2C12 layer was dried (Na2S04), filtered, and the solution was removed using a rotary evaporator (40°C), which produced a yellow oil. EtOH (500 mL) was added, the solution procedure with some modifications. To a 2 L round bottom flask (RBF) was added maltol (1) (151.3 g, 1 2 1.2 mol), 1 L MeOH, and NaOH (50.4 g, 1.26 mol) 39 was heated until the oil dissolved, transferred to a thick walled Erlenmeyer flask and recrystallized in the freezer. The crystals were broken apart with a large spatula, filtered with a course frit, and washed with a minimum of cold E t O H to remove the smell of benzyl chloride. The volume of the filtrate was reduced to -200 m L using a rotary evaporator, and recrystallized as above with the addition of seed crystals. A third recrystallization from the mother liquor was done as above; all the crystals were combined and dried under vacuum 24 h to produce a white solid (183.0 g, 71% yield). M p 52-54 °C. ' H N M R (300 M H z , CDC1 3 ) : 5 = 2.078 (s, 3H, C 2 - C / / 3 ) , 5.146 (s, 2H, Bn CH2), 6.397 (d, I H , H5, V 5 , 6 = 5.6 Hz), 7.34 (m, 5H, B n C6H5), 7.595 (d, I H , H6, V 6 , 5 = 5.6 Hz). 3-BenzyIoxy-l-(carboxymethyI)-2-methyl-4(l//)-pyridinone (3). To a 500 mL R B F was added sodium glycinate (58.23 g, 600 mmol), benzyl maltol (2) o 0 B n (86.50 g, 400 mmol), water and M e O H (125 m L each). A l l starting materials dissolved upon heating. After 36 h of reflux, T L C in 1:1 A c O E t : M e O H showed appreciable product formation (Rf ~0.3), as well ^ as other spots including some unreacted benzyl maltol (Rf -0.7). The volume of the red-brown solution was reduced to -100 m L using a rotary evaporator (50°C). A n additional 200 m L of water was added, the solution was heated to dissolve all the material, cooled, and unreacted benzyl maltol was extracted with 4x l00mL CH2CI2. The water layer was transferred to a R B F , stirred, and the p H was adjusted from 10.2 to 2.2 by adding -250 m L of 2 M HC1. A fine light yellow precipitate started to form at pH - 5 . After cooling on ice, the precipitate was filtered with a course frit and 40 rinsed with 2x30 mL cold water. The light yellow solid was dried 30 h on the vacuum line and weighed 108.5 g. TLC in 1:1 AcOEt:MeOH followed by treatment with ninhydrin spray indicated mainly product with the presence of some glycine. This solid was recrystallized by dividing it between two 2 L RBFs, adding 950 mL MeOH and 200 mL of water to each RBF, followed by refluxing and cooling to RT. After some recrystallization had taken place, the RBFs were placed in the fridge for 12 h, then yellow crystalline solid was filtered with a course frit, rinsed with 2x30 mL cold water, and dried at least 48 h in the vacuum desiccator (52.32 g, 48 % yield based on 2). 'H NMR (300 MHz, CD3OD): 5 = 2.130 (s, 3H, C2-C/73), 4.819 (s, 2H, C// 2 C0 2 H), 5.097 (s, 2H, Bn CH2), 6.604 (d, 1H, 775, V 5 , 6 = 7.4 Hz), 7.39 (m, 5H, Bn C6H5), 7.786 (d, 1H, 776,3J6,5 = 7.4 Hz). Anal. Calcd (found) for C, 5 H 1 5 N04: C, 65.93 (65.71); H, 5.53 (5.65); N, 5.13 (5.08). MS (+LSIMS): m/z 274 [M+H]+. Methyl 6-deoxy-6-tosyl-a-D-glucopyranoside (5). This compound was prepared using 0 O H OTs a previously published procedure starting from / 4 " H O ^ ^ S ^ " H O ^ S ^ S ^ M E T H Y L a - D - 8 l u c ° p y r a n o s i d e (4) ( S i g m a ) -H f V 3 Hfi ° x O x Yield 30%. H NMR (500 MHz, CD3OD): 8 = 4 5 2.403 (s, 3H, Ts-C773), 3.305 (s, 3H, C l -OC/73), 3.430 (dd, 1H, 774, V 4 j 3 = 9.7 Hz, V 4 , 5 = 9.7 Hz), 3.458 (dd, 1H, HI, V 2 > , = 3.8 Hz, V 2 , 3 = 9.7 Hz), 3.685 (dd, 1H, H3, V 3 , 2 = 9.7 Hz, V 3 , 4 = 9.7 Hz), 3.704 (ddd, 1H, 775, V 5 , 4 = 9.7 Hz, V 5 , 6 a = 2.0 Hz, 3 J 5 , 6 b = 4.9 Hz), 4.215 (dd, 1H, H6b, 3J6bi5 = 2.0 Hz, V 6 b , 6 a = 11.9 Hz), 4.293 (dd, 1H, H6a, 3 J 6 a , 5 = 4.9 Hz, V 6 a , 6 b = 11.9 Hz), 4.656 (d, 1H, 771,3Jl2 = 3.8 Hz), 7.307 (d, 2H, aromH,J= 8.2 Hz), 7.766 (d, 2H, arom H, J = 8.2 Hz). , 3C{'H} 41 NMR (125 MHz, CD3OD): 5 = 21.61 (Ts-C//3), 55.35 (CI-OCH3), 69.20 (C6), 69.39 (C5), 69.56 (C4), 71.74 (C2), 74.08 (C3), 99.35 (Cl), 127.97 (arom Q , 129.86 (arom Q, 132.76 (arom Q, 144.92 (arom Q. Methyl 6-azido-6-deoxy-a-D-glucopyranoside (6). This compound was prepared from 5 using a previously published procedure.9 After the rotary N3 HO- I^A^ O evaporator removed the solvent, a light yellow syrup was produced 3 H O I that was dried 24 h in a vacuum desiccator (97% yield). *H NMR ° \ 6 (500 MHz, CD3OD): 5 = 3.246 (dd, IH, HA, 3 J 4 , 3 = 9.8 Hz, 3 J 4 , 5 = 9.8 Hz), 3.385 (dd, IH, //6b, 3 J 6 b , 5 = 6.6 Hz, 2 J 6 b , 6 a= 13.1 Hz), 3.407 (dd, IH, HI, 3J 2 , i = 3.8 Hz, V 2 i 3 = 9.8 Hz), 3.424 (s, 3H, CI-OC//3), 3.491 (dd, IH, //6a, 3 J 6 a , 5 = 2.3 Hz, 2J 6 a ,6b = 13.1 Hz), 3.587 (dd, IH, H3, V 3 > 2 = 9.8 Hz, V 3 , 4 = 9.8 Hz), 3.665 (ddd, IH, H5,3J5,4 = 9.8 Hz, V 5 ,6a = 2.3 Hz, 3J 5,6b = 6.6 Hz), 4.685 (d, IH, HI, 3 J I i 2 = 3.8 Hz). 1 3C{'H} NMR (125 MHz, CD3OD): 5 = 52.70 (C6), 55.71 (CI-OCH3), 72.51 (C4), 72.58 (C5), 73.37 (C2), 74.82 (C3), 101.25 (Cl). Methyl 6-amino-6-deoxy-a-D-glucopyranoside (7). Palladium black has been used previously for azide reductions on carbohydrates.10 Compound 6 (2.920 g, 13.3 mmol) was dissolved in 50 mL MeOH and transferred to a Parr reactor with glass insert. Palladium black (142 7 mg, 1.33 mmol) was wetted with -0.5 mL water and added to the reactor using -5 mL MeOH. A stir bar was added and the reaction mixture was stirred rapidly. The reactor was sealed and hydrogen gas was pressurized to 60 atm from a 42 cylinder. The pressure gauge read 52 atm after stirring at RT for 20 h, at which time the pressure was slowly released. TLC in 2:1 EtOAc:MeOH followed by ninhydrin spray indicated one red spot (Rf ~0) that charred following sulfuric acid spray of the TLC plate. The reaction mixture was filtered through a medium frit and the filtrate was removed using a rotary evaporator, which produced a viscous oil. A tacky white solid formed from most of the oil after drying 48 h in the vacuum desiccator. The RBF was cooled in the freezer and the white solid was scraped into a vial (1.412 g, 73% yield) and used without further purification. The remaining oil was dissolved in water, the pH was adjusted to ~5 (2 M HC1 solution), and the solvent was removed using a rotary evaporator. This was recrystallized (freezer) from a minimum amount of hot MeOH and the hydrochloride salt was characterized. ' H N M R (500 MHz, CD 3 OD): 5 = 3.014 (dd, 1H, H6b, 3J6bi5 = 6.8 Hz, 2J6b,6a = 13.1 Hz), 3.166 (dd, 1H, 774, 3 J 4 j 3 = 9.6 Hz, 3 J 4 , 5 = 9.6 Hz), 3.343 (dd, 1H, /76a, 3 . / 6 a ,5 = 3.0 Hz, V 6a,6b= 13.1 Hz), 3.414 (dd, 1H, 772, V 2 , i = 3.7 Hz, 3J 2,3 = 9.6 Hz), 3.439 (s, 3H, Cl-OC/7 3 ), 3.608 (dd, 1H, 773, 3J 3 > 2 = 9.6 Hz, 3 J 3 > 4 = 9.6 Hz), 3.732 (ddd, 1H, 775, 3J 5, 4 = 9.6 Hz, 3J 5 ,6a = 3.0 Hz, V 5 , 6 b = 6.8 Hz), 4.730 (d, 1H, 771, V , ; 2 = 3.7 Hz). 1 3 C{'H} N M R (125 MHz, CD 3 OD): 5 = 42.26 (C6), 56.26 (Cl-OCH 3 ) , 69.30 (C5), 73.20 (C2), 73.43 (C4), 74.54 (C3), 101.53 (Cl). MS (+LSIMS): m/z 194 [M+H]+. 3-Benzyloxy-l-{[((2R,3S,4S,5R,6S)-3,4,5-trihydroxy-6-methoxy-tetrahydropyran-2-yl-methyl)carbamoyl]methyl}-2-methyl-4(li/)-pyridinone (8). The reaction used was typical of activated ester reactions in the literature." To a 150 mL RBF was added N -hydroxysuccinimide (NHS) (575.5 mg, 5.0 mmol), compound 3 (1.3665 g, 5.0 mmol) 43 0 and a stir bar. Dicyclohexylcarbodiimide (DCC) (1.0830 g, 5.25 12 OBn mmol) was added to a 25 mL RBF. Both RBFs were fitted with 13 4 to both RBFs with a cannula (-10 mL for DCC and -65 mL for 3 septa, evacuated for 1 h and refilled with N 2 . Dry DMF was added N H HO O. and NHS). The DCC solution was added to the stirring reaction mixture in the 150 mL RBF (via cannula). A white precipitate 8 formed after 5 min and this reaction mixture was left to stir for 24 h at RT under positive N 2 pressure. A 250 mL RBF containing a stir bar and 7 (1.009 g, 5.25 mmol) was dried 24 h in vacuo and connected to a Schlenk medium frit (30 mm diameter, 15 cm length) that had been flame-dried. The activated ester was quickly added to the top of the Schlenk apparatus and filtered removing the white dicyclohexylurea (DCU) precipitate. The filtrate dissolved 7 with stirring and the Schlenk frit was replaced with a septum and stirred under N 2 for 4 h at RT. The reaction mixture was rotovaporated to dryness (55°C, -30 min) and dissolved in a minimum of warm MeOH, filtered through a course frit, and applied to a silica column (5x45 cm) run in 3:1 AcOEt:MeOH. TLC of the fractions (1:1 AcOEt:MeOH) followed by charring revealed isolated product free from D C U (remains white on charred TLC). The product fractions were collected and the solvent was removed using a rotary evaporator (40°C). The solid obtained was dried in a vacuum desiccator 48 hours to yield a white product (1.697 g, 76 % based on 3). X-ray quality crystals were grown from slow evaporation in -3:1 AcOEfMeOH. Anal. Calcd (found) for C 2 2 H 2 8 N 2 0 8 : C, 58.92 (58.65); H, 6.29 (6.32); N , 6.25 (6.11). *H N M R (500 MHz, CD 3 OD): 5 = 2.079 (s, 3H, C9-C/73), 3.105 (dd, 1H, HA, V 4 , 5 = 9.8 Hz, 3 J 4 , 3 = 9.8 Hz), 3.359 (s, 3H, C l - O C / / 3 ) , 3.371 (dd, partially 44 overlapped with Cl-OCH 3 , HI, 3J 2 , i = 3.7 Hz, V 2 , 3 = 9.8 Hz), 3.390 (dd, partially overlapped with CI-OCH3, H6b, 2J6bfia = 13.2 Hz, V6b,5 = 6.7 Hz), 3.56 (ddd, overlapped with H3, H5), 3.597 (dd, partially overlapped with H5, H3,3J3,2 = 9.8 Hz, V 3 , 4 = 9.8 Hz), 3.651 (dd, IH, H6a, 2 J 6 a , 6 b = 13.2 Hz, 3J 6 a,5 = 2.6 Hz), 4.637 (d, IH, H\, 3Ji,2 = 3.7 Hz), 4.723 (s, 2H, HS), 5.042 (s, 2H, Bn CH2), 6.466 (d, IH, H\2, 3 J , 2 , i 3 = 7.4 Hz), 7.37 (m, 5H. Bn C6H5), 7.606 (d, IH, #13, V ] 3 , i 2 = 7.4 Hz). MS (+LSIMS): m/z 449 [M+H]+. l-{[((2R,3S,4S,5R,6S)-3,4,5-Triacetoxy-6-methoxy-tetrahydropyran-2-yl-methyl)carbamoyl]methyl}-3-benzyIoxy-2-methyl-4(l//)-pyridinone hemihydratc (9-0.5H2O). To a 50 mL RBF was added compound 8 (448.5 mg, 1.0 mmol) and a stir bar; the RBF was then evacuated and refilled with N 2 . Dry 'li^jfio'"" pyridine (12 mL) was syringed into the RBF as was 1.7 mL acetic anhydride (18 mmol, 6 equiv. per OH). The reaction mixture was left to stir under a positive N 2 pressure for 8 h at RT. After spotting AcO^v^J^ O o n silica TLC, a heat gun was used to evaporate most of the 3 AcOn pyridine and the TLC was run in 1:1 AcOEt:MeOH with the \ 9 product having a higher Rf value than the starting material. The solvent was evaporated at reduced pressure on the vacuum line while being stirred. After ~1 h for solvent removal the RBF was left on the vacuum line for an additional 16 h. The solid was redissolved in a minimum of hot 1:1 EtOH.water and recrystallized at RT. After cooling on ice, the precipitate was filtered with a medium frit and rinsed with 2x3 mL cold water, and was dried in a vacuum desiccator for at least 48 hours to yield a white solid (446.6 mg, 77% yield). Anal. Calcd (found) for C28H35N2O11 .5: C, 57.63 (57.52); 45 H, 6.05 (6.12); N, 4.80 (4.92). 'H NMR (400 MHz, CD 3OD): 5 = 1.956 (s, 3H, COC/73), 1.977 (s, 3H, COC//3), 2.031 (s, 3H, COC//3), 2.100 (s, 3H, C9-C// 3), 3.32 (dd, partially overlapped with CD3OD, 776b), 3.388 (s, 3H, Cl-OC/7 3), 3.609 (dd, 1H, 776a, 2^6a,6b = 14.4 Hz, V 6 a , 5 = 5.1 Hz), 3.895 (ddd, 1H, 775, V 5 , 4 = 9.9 Hz, 3JSM = 5.1 Hz, V 5 , 6 b = 2.5 Hz), 4.714 (s, 2H, 778), 4.87 (overlapped with CD 3OH, 772), 4.919 (dd, partially overlapped with HI, 774, 3y 4, 3 = 9.8 Hz, 37 4, 5 = 9.8 Hz), 4.936 (d, partially overlapped with H4, 771, 37],2 = 3.7 Hz), 5.045 (s, 2H, Bn C772), 5.380 (dd, 1H, 773, 373,2 = 9.8 Hz, 3y3,4 = 9.8 Hz), 6.460 (d, 1H, 7712,3J [ 2, l 3 = 7.4 Hz), 7.38 (m, 5H, Bn C6775), 7.588 (d, 1H, 7713,3J13 i,2 = 7.4Hz). MS(+ESI): m/z 597 [M+Na]+, 575 [M+H]+. 4-Nitrophenyl 2,3,4,6-tetra-O-acetyl-p-D-glucopyranoside (11). This compound was N 0 prepared in a similar manner to other OKc ^ f T ^ ^ l previously published Koenigs-Knorr O. 4 H . T j . . . „J2 procedures12 with a number of AcO ' n ^ ^ T T ^ ° 3 AcO 1 modifications. To a 250 mL RBF was 10 11 added 4-nitrophenol (2.782 g, 20.0 mmol), silver carbonate (5.515 g, 20.0 mmol), activated 4 A molecular sieves (-10 g), 100 mL dry MeCN, 2,6-dimethylpyridine (2.5 mL, 21.0 mmol), and a stir bar. The RBF was fitted with a septum and the reaction mixture was stirred for a few minutes. To a 100 mL RBF was added 4.112 g (10.0 mmol) of 2,3,4,6-tetra-O-acetyl-a-D-glucopyranosyl bromide (10) (Aldrich) which was dissolved with 40 mL of dry MeCN and the solution was transferred to the 250 mL RBF via a cannula. The reaction mixture was stirred for 1 h at RT in the dark, filtered through Celite, followed by rinsing of the RBF. and Celite 46 with 30 mL CH2CI2. The solution was removed using a rotary evaporator (40 °C), which produced an oil that was placed on the vacuum line for 2 h. The resulting syrup was dissolved with 150 mL CH 2C1 2 and washed with water (100 mL), 5% NaOH (3 x 100 mL), 5% HC1 (100 mL), and water (2 x 100 mL). The CH 2C1 2 layer was dried over sodium sulphate, filtered, and the solvent of the filtrate was removed using a rotary evaporator (40 °C). The product was recrystallized from a minimum of hot ethyl acetate/hexanes, filtered through a course frit, and dried 16 h on the vacuum line to produce a white solid (3.202 g, 68 % yield). ] H NMR (300 MHz, CD3OD): 5 = 1.994 (s, 3H, COC/73), 2.028 (s, 3H, COC/73), 2.030 (s, 3H, COC/73), 2.036 (s, 3H, COC/73), 4.15 (ddd, partially overlapped with H6b, 775), 4.161 (dd, partially overlapped with H5, /76b, 2 J 6 b , 6 a = 12.7 Hz, V 6 b , 5 = 2.2 Hz), 4.309 (dd, 1H, /76a, 2 J 6 a > 6 b = 12.7 Hz, 3 J 6 a j 5 = 5.4 Hz), 5.126 (dd, 1H, 774, V 4 , 5 = 9.6 Hz, V 4 , 3 = 9.6 Hz), 5.209 (dd, 1H, 772,V 2 ; 1 = 7.8 Hz, V 2 , 3 = 9.6 Hz), 5.405 (dd, 1H, 773, 373,2 = 9.6 Hz, 3 J 3 , 4 = 9.6 Hz), 5.546 (d, 1H, 7/1, 3Jl2 = 7.8 Hz), 7.190 (d, 2H, arom H,J= 9.4 Hz), 7.524 (d, 2H, arom H,J= 9.4 Hz). 4-Nitrophenyl p-D-glucopyranoside (12). This deacetylation was typical of other deacetylations in the literature.13 To a 1 L RBF was added compound 11 (22.53 g, 48.0 mmol) and a stir bar. The RBF was evacuated and refilled with N 2 . Dry MeOH (-500 mL) and dry CH 2C1 2 (-200 mL) were added via a cannula to dissolve the starting material. Sodium methoxide (259 mg, 4.8 mmol) was added to the RBF and the septum was quickly replaced. The reaction mixture turned 47 bright yellow and was stirred for 3 h at RT. TLC of the clear yellow reaction mixture in 1:1 AcOEthexanes indicated one spot (Rf = 0) with no starting material (Rf -0.5). A white precipitate formed on standing for a few minutes. The CH2CI2 was removed using a rotary evaporator (35 °C), an additional 100 mL of MeOH was added, and the solution was heated to dissolve the precipitate. Upon cooling (-30-40 °C), strongly acidic cation exchange resin (Rexyn 101, Fisher) was added (5 g of resin was activated with 6 M HC1 and rinsed with water until the solution was neutral in pH). The resin was stirred for 30 min, filtered through a Buchner funnel, followed by crystallization from the filtrate initially at RT (2 h) and then in the freezer. The precipitate was filtered through a course frit, washed with cold MeOH, and dried 24 h on the vacuum line (12.293 g). A second crop of crystals was obtained from a minimum amount of hot MeOH as above (1.963 g) and these products were combined (14.26 g, 99 % yield). ] H NMR (300 MHz, CD3OD): 5 = 3.34 (dd, partially overlapped with CD30D, H4), 3.45 (m, 3H, HI, H3, and H5 overlapped), 3.693 (dd, IH, H6b, 2J6bM = 12.1 Hz, 3 J 6 b , 5 = 5.5 Hz), 3.901 (dd, IH, H6a, 2J6 a,6b ^ 12.1 Hz, V 6 a , 5 = 2.0 Hz), 4.914 (d, IH, H\, V , , 2 = 7.2 Hz), 7.231 (d, 2H, arom H, .7=9.1 Hz), 8.208 (d, 2H, arom//, J= 9.1 Hz). MS (+LSIMS): m/z 302 [M+H]+. 4-Aminophenyl P-D-glucopyranoside, hydrochloride salt (13). This hydrogenation setup is similar to that described for HOG6GP. To a 1 L RBF was added compound 12 (9.640 g, 32.0 mmol), 150 mL MeOH, 4 b < _ y 220 mL water, and a stir bar. The 10% Pd/C (3.405 g, 3.2 mmol 3 HO 1 Pd) was wetted and rinsed into the RBF with water (-30 mL). 13 About 5 drops of 0.5 M HC1 were added and the adaptor was 48 opened to the balloon containing hydrogen gas. After 3 h, 2 mL of 2M HC1 solution was added to the reaction mixture. An additional 2 mL of 2 M HC1 solution was added after 6 h and the hydrogen balloon was refilled and the reaction left to stir an additional 18 h. TLC in 1:1 AcOEt:MeOH showed one spot (UV-lamp) with a lower Rf than the starting material. The spot turned red with ninhydrin spray and charred following sulfuric acid spray. The reaction mixture was suction-filtered through a medium frit and the Pd/C was rinsed with 10 mL of water . The filtrate was light yellow in colour and 2 M HC1 solution was added with stirring until the pH meter read ~4. The solvent was removed using a rotary evaporator (45 °C, ~2 h) and placed on the vacuum line for 24 h. The precipitate was recrystallized from a minimum amount of hot MeOH. Filtering and rinsing the precipitate with cold MeOH (2x15 mL) resulted in a white solid. The mother liquor was reduced to a minimum amount using a rotary evaporator, and the remaining residue was recrystallized a second time as above. The white solids were combined and dried in a vacuum desiccator at least 48 h before weighing to determine the yield (8.547 g, 87% yield). 'H NMR (300 MHz, (CD3)2SO): 8 = 3.17 (m, 4H), 3.447 (m, 1H, H6b), 3.697 (m, 1H, 776b), 4.529 (dd, 1H, J= 5.2 Hz, 7= 5.2 Hz), 4.566 (d, 1H, 771,3JU2 = 7.4 Hz), 4.952 (d, 1H, J= 5.0 Hz), 5.014 (d, 1H, J= 4.4 Hz), 5.198 (d, 1H, J = 5.0 Hz), 6.475 (d, 2H, arom 77, J= 8.7 Hz), 6.756 (d, 2H, arom 77, J= 8.7 Hz). MS (+LSIMS): m/z 272 [M+H]+. 3-BenzyIoxy-l-{[4-(P-D-glucopyranosyloxy)phenylcarbamoyl]methyl}-2-methyI-4(l//)-pyridinone (14). This was prepared in a similar fashion as 8 with the following changes: 1.913 g (7.0 mmol) 3, 806 mgNHS, 1.475 g DCC (9.1 mmol), and 2.800 g (9.1 49 o mmol) 13-HC1 were used with 250 m L RBFs . Dry D M F OBn (-140 mL) dissolved the N H S and 8. The D C C was 17 dissolved in -15 m L dry D M F (-5 m L to rinse the N H cannula). A minimum amount of dry D M F was used to HO HO-3 HO 1 the addition of triethylamine (1.9 mL, 9.1 mmol) just prior dissolve 13HC1 (250 mL R B F with septum), followed by to Schlenk filtering of the activated ester. After 12 h at 14 50°C, the D M F was removed using a rotary evaporator (55°C, -1.5 h), the residue taken up in a minimum amount of warm M e O H , filtered through a course frit and applied on a column (5x45 cm) run in 4:1 A c O E t : M e O H . T L C of the fractions was conducted (1:1 A c O E t M e O H ) , and fractions with only product were combined and the solvent was removed using a rotary evaporator (40 °C). A small amount of product co-eluted with a slight red-brown colour. These fractions were recrystallized from hot E tOH, filtered through a course frit, and rinsed with 5 m L cold E tOH to afford a white product. The white solids were combined and dried in a vacuum desiccator at least 48 h (2.630 g, 71 % based on 3). Anal . Calcd (found) for C27H30N2O9: C, 61.59 (61.34); H , 5.74 (5.86); N , 5.32 (5.23). ' H N M R (400 M H z , 9:1 C D 3 O D : D 2 0 ) : 5 = 2.038 (s, 3H, C13-C7/ 3 ) , 3.410 (dd, I H , HA, V 4 , 5 = 9.8 Hz , V 4 , 3 = 9.8 Hz), 3.47 (m, partially overlapped with H5 and H3, H2), 3.47 (m, partially overlapped with H2 and H3, H5), 3.48 (m, partially overlapped with H2 and H5, H3), 3.705 (dd, I H , H6b, 2J6bM = 12.3 Hz, 3 J 6 b , 5 = 5.4 Hz), 3.886 (dd, I H , H6a, V 6 a , 6 b = 12.3 Hz , 2J6a,5 = 2.0 Hz), 4.855 (s, partially overlapped with water, H12), 4.907 (d, partially overlapped with water peak, H\, 3 J i , 2 = 7.2 Hz), 5.042 (s, partially overlapped with water peak, H\2), 6.532 (d, I H , 50 7/16, V i 6 ; i 7 = 7.4 Hz), 7.090 (d, 2H, 7/8, \ 9 = 9.0 Hz); 7.30 (m, 5H, Bn C6775), 7.441 (d, 2H, 7/9, V 9 , 8 = 9.0 Hz), 7.655 (d, 1H, 7717, V 1 7 , i 6 = 7.4 Hz). 1 3C{'H} NMR (100 MHz, CD3OD): 5 = 13.05, 57.31, 63.21, 71.09, 74.62, 74.89, 77.51, 77.80, 102.31, 117.13, 118.23, 122.85, 129.43, 129.52, 130.36, 133.45, 137.95, 142.82, 145.93, 146.56, 155.75, 166.39, 175.27. MS (+LSIMS): m/z 527 [M+H]+. l-{[4-(2,3,4,6-Tetra-0-acetyl-p-D-glucopyranosyloxy)phenylcarbamoyl]methyl}-3-benzyloxy-2-methyl-4(l//)-pyridinone hemihydrate (15-0.5 H2O). This reaction was carried out in a similar fashion as for 9 with the following O changes: 708.0 mg (1.34 mmol) 14 was dissolved in 15 mL dry pyridine and 2.0 mL acetic anhydride (21.5 mmol, 4 equiv. per OH) were used. TLC was run in 7:3 AcOEt:MeOH. After 12 h of stirring at RT, the reaction mixture was evaporated to dryness using a rotary evaporator (50 °C, <30 min) followed by 12 h on the vacuum line. After recrystallization and filtering, the product was dried in a vacuum desiccator for at least 48 hours to obtain a white solid (698.8 mg, 74% yield). Anal. Calcd (found) for C35H39N2O13 .5: C , 59.74 (60.07); H, 5.59 (5.51); N, 3.98 (4.04). *H NMR (400 MHz, CD3OD): 5 = 1.983 (s, 3H, COC//3), 2.016 (s, 3H, COC//3), 2.031 (s, 6H, 2xCOC//3), 2.342 (s, 3H, C13-C7/3), 4.055 (ddd, 1H, H5, 3J5,4 = 9.9 Hz, 3J5,6a = 5.1 Hz, 3J 5 i 6b = 2.3 Hz), 4.136 (dd, 1H, 776b, V 6 b > 6 a = 12.3 Hz, 37 6 b, 5 = 2.3 Hz), 4.296 (dd, 1H, 776a, 2 J 6 a , 6 b = 12.3 Hz, 37 6 a, 5 = 5.1 Hz), 4.845 (s, partially overlapped with CD 3OD, 7/12), 5.059 (s, 2H, Bn CH2), 5.089 (dd, 1H, /74, 3J 4, 5 = 9.9 Hz, 51 3 J 4 > 3 = 9.6 Hz), 5.135 (dd, IH, HI, 3J2,\ = 7.9 Hz, 3J 2 )3 = 9.6 Hz), 5.282 (d, IH, HI, 3 J , , 2 = 7.9 Hz), 5.362 (dd, IH, H3, %2 = 9.6 Hz, 3J 3,4 = 9.6 Hz), 6.481 (d, IH, #16, 3Ji 6,i7 = 7.4 Hz), 6.995 (d, 2H, H9, %9 = 9.0 Hz), 7.163 (m, 3H, Bn C6H5), 7.396 (m, 2H, Bn C6#5), 7.480 (d, 2H, #8,3J9,8 = 9.0 Hz), 7.653 (d, IH, #17, Vn.ie = 7.4 Hz). MS (+ESI): m/z 717[M+Na]+, 695 [M+H]+. 3-Benzyloxy-l-(4-hydroxyphenyl)-2-methyl-4(l//)-pyridinone (17). To a 2 L RBF p was added 4-aminophenol (49.11 g, 450 mmol), benzyl maltol (2) (64.87 5(< 4 p g, 300 mmol), 1.2 L water, 250 mL MeOH, and a stir bar. The reaction mixture was heated (with strirring) in an oil-bath at 90 °C for 72 h to yield a white precipitate. The solution was filtered hot through a medium frit, rinsed with cold water and MeOH (25 mL each) and the white precipitate was dried in a vacuum desiccator for 16 h (13.207 g). The warm filtrate was left at RT to cool and a solid crystallized. Before filtering, the solution was cooled on ice for 30 min, the solid was filtered through a course frit and dried 3 h in a vacuum desiccator (22.4 g). This solid was added to a beaker, 50 mL MeOH was added and refluxed (heat gun); a red-brown solution was decanted. This was repeated 4 more times, with the last decant having a slight yellow colour. The remaining solid was recrystallized from refluxing MeOH (-800 mL). After cooling in the freezer for 16 h, the solution was filtered through a course frit and dried in a vacuum desiccator 24 h to produce a white solid (13.733 g). The *H NMR and TLC in 9:1 AcOEt:MeOH looked identical for the precipitated and recrystallized products; they were combined (26.937 g, 29% yield). Mp 240-242 °C. Anal. Calcd (found) for C 1 9 H i 7 N 0 3 : C, 74.25 (74.24); H, 52 5.58 (5.60); N , 4.56 (4.58). 'H NMR (300 MHz, CD3OD): 5 - 1.855 (s, 3H, C2-C//3), 5.112 (s, 2H, Bn CH2), 6.010 (d, 1H, 7/5, V 5 > 6 = 7.3 Hz), 6.883 (d, 2H, phenol arom H, J = 8.8 Hz), 7.091 (d, 2H, phenol arom H, J= 8.8 Hz), 7.38 (m, 5H, Bn C 6 / / 5 ), 7.622 (d, 1H, H6,3J6;5 = 7.3 Hz). MS (+LSIMS): m/z 308 [M+H]+. l-[4-(2,3,4,6-Tetra-0-benzyl-P-D-glucopyranosyloxy)phenyl]-3-benzyloxy-2-methyl-4(l//)-pyridinone (18). This reaction was typical of Mitsunobu reactions in the literature.14 To a 500 mL RBF was added 2,3,4,6-tetra-O-O O R benzyl-D-glucopyranose (16, Aldrich) (5.407 g, 10.0 1 4 ^ 1 3 ^ 2 mmol), 17 (3.688 g, 12.0 mmol), and a stir bar. Azodicarboxylic acid dipiperidide (ADDP) (3.785 g, 15 mmol) was placed in a 50 mL RBF, and both RBFs were evacuated and refilled with N 2 . Dry CH 2C1 2 (-240 mL) Ig was added directly from the still (ground glass joint) to the 500 mL RBF, the septum was replaced, and the solution was stirred for 5 min leaving an undissolved white solid (17). Tributylphosphine (3.74 mL, 15.0 mmol) was added via a syringe and stirred for 5 min. Dry CH 2C1 2 (20 mL) was added to the 50 mL RBF (syringe), and the ADDP solution was transferred to the 500 mL RBF using a syringe. The RBF and syringe that contained the ADDP were rinsed with 10 mL dry CH2C12, and transferred to the 500 mL RBF. The reaction mixture was light orange in colour. All material dissolved after -1 h. The flask was stirred for 10 h at RT and TLC in AcOEt indicated the presence of product (Rf -0.35). The CH 2C1 2 was removed using a rotary evaporator (45 °C) and the reaction mixture was redissolved in hot AcOEt (-120 mL), 53 and was cooled in the freezer. A white solid crystallized (ADDP byproduct), and was filtered with a course frit, and rinsed with 2x10 mL cold AcOEt. The AcOEt filtrate was removed using a rotary evaporator (45 °C) to form an orange-brown oil. A minimum of AcOEt was added to make the oil runny, and this was applied to a silica column (5x45 cm) run with AcOEt. A bright yellow band eluted just ahead of the product. TLC of the fractions was conducted (AcOEt), the product fractions were collected, the solvent was removed using a rotary evaporator (45 °C), and dried in vacuo 48 h to produce a yellow viscous oil (6.008 g, 66% yield). ' H N M R (300 MHz, CDC1 3): 5 = 2.342 (s, 3H, C13-C# 3), 3.73 (m, 6H), 4.56 (m, 3H), 4.86 (m, 3H), 4.99 (m, 3H), 5.238 (s, 2H, Bn CH2), 6.467 (d, IH, #14, Vi 4 , i s = 7.6 Hz), 7.043 (d, 2H, #8, 3J 8,9 = 8.9 Hz), 7.119 (d, 2H, #9, 3 J 9 i 8 = 8.9 Hz), 7.201 (m, 3H, Bn C6H5), 7.31 (m, 21H, Bn C 6# 5), 7.45 (m, IH, Bn C 6# 5), 7.460 (d, IH, #15, 3J, 5,i4 = 7.6 Hz). MS (+LSIMS): m/z 830 [M+H]+. OH 3-Hydroxy-l-{[((2R,3S,4S,5R,6S)-3,4,5-trihydroxy-6-methoxy-tetrahydropyran-2-yI-methyl)carbamoyl]methyl}-2-methyI-4(l#)-pyridinone (HOG6GP). To a 150 mL O RBF was added 8 (1.345 g, 3.0 mmol) which was dissolved in MeOH (60 mL) with stirring. Water (40 mL) was added to prevent the product from precipitating when formed. The 10% Pd/C (319 mg, 0.3 mmol Pd) was weighed into a vial, wetted with water (5 H O - ^ ^ J - - O. mL), poured into the RBF with the residual slurry being rinsed into H O O ^ the RBF with a MeOH squeeze bottle (~5 mL). An 11 L latex HOG6GP balloon was filled with H 2 gas and fitted to a stopcocked adaptor which was opened after placement in the RBF ground glass joint. After stirring at RT for 54 12 h, the H 2 atmosphere was removed and silica TLC (1:1 AcOEt:MeOH) produced one spot (turned purple with FeCl 3 spray) with slightly lower Rf than the starting material, indicative of a completely debenzylated product. The solution was suction-filtered through a medium frit, the Pd/C was stirred with an additional 10 mL of water in the frit, suction-filtered, and the combined filtrates were syringed through a Nylon filter (25 mm, 0.2 urn) to remove fine Pd/C. This light orange solution was removed using a rotary evaporator (50°C, <30 min) and recrystallized in a small RBF by adding 3 mL water and 4 mL methanol, heating until dissolved and cooling to RT. The precipitate was cooled on ice, filtered through a medium frit, and rinsed with 2x3 mL cold MeOH to produce a white solid and a red-orange coloured filtrate. The solid was dried in a vacuum desiccator at least 48 hours to yield 568 mg, 53% yield. Anal. Calcd (found) for Q5H22N2O8: C, 50.28 (50.02); H, 6.19 (6.13); N , 7.82 (7.82). 'H NMR (400 MHz, 1:1 CD 3OD:D 20): 5 = 2.314 (s, 3H, C9-C7/3), 3.216 (dd, 1H, H4,3J4,5 = 9.8 Hz, 3J4>3 = 9.8 Hz), 3.370 (s, 3H, Cl-OC773), 3.429 (dd, 1H, /76b, 2J6bM = 13.9 Hz, 3 J 6 b , 5 = 6.8 Hz), 3.481 (dd, 1H, HI,3J2,i = 3.7 Hz, 3 J 2 > 3 = 9.8 Hz), 3.61 (dd, overlapped with H5, 7/3,V3,2 = 9.8 Hz, 3 J 3 , 4 = 9.8 Hz), 3.62 (ddd, overlapped with H3, 775), 3.671 (dd, 1H, //6a, 2 J 6 a , 6b = 13.9 Hz, 3J6a,5 = 2.7 Hz), 4.727 (d, partially overlapped with water peak, 7/1, 3J ] > 2 = 3.7 Hz), 4.86 (s, partially overlapped with water peak, 7/8), 6.486 (d, 1H, 7/12, V i 2 j i 3 = 7.2 Hz), 7.579 (d, 1H, /713, 3J 1 3 ji 2 = 7.2 Hz). 1 3C{ lH} NMR (75 MHz, 1:1 CD 3OD:D 20): 5 = 12.30 (C9-CH3), 41.39 (C6), 55.89 ( C I - O C H 3 ) , 57.00 (CS), 70.92 (CS), 72.63 (C2), 72.63 (C4), 74.16 (C3), 100.55 (Cl), 113.13 (C12), 134.59 (C9), 140.62 (C13), 146.10 (C10), 169.26 (C7), 170.95 (Cl 1). MS (+LSIMS): m/z 359 [LH+H]+; (+ESI): m/z 359 [LH+H]+. 55 Tris(l-{[((2R,3S,4S,5R,6S)-3,4,5-trihydroxy-6-methoxy-tetrahydropyran-2-yl-methyl)carbamoyl]methyl}-2-methyl-3-oxy-4(l//)-pyridinonato)gaIlium(IH) tetrahydrate (Ga(OG6GP) 3-4H 20). Gallium nitrate nonahydrate (41.8 mg, 0.1 mmol) was dissolved in 3 mL Ga water in a 4 mL vial. This solution was added dropwise to a stirring solution of HOG6GP (358 mg, 0.315 mmol) that was dissolved in 4 mL water in a 20 ml vial. The vial for the gallium nitrate was rinsed with 2 mL water Ga(OG6GP) 3 and added to the 20 mL vial. The pH of the reaction mixture was 1.6. A 1% NaOH solution was added dropwise to the reaction mixture until the pH was 7.0 and the solvent was removed using a rotary evaporator (50 °C, <30 min) to produce an off-white precipitate. The precipitate was redissolved in 3 mL water and applied on a Sephadex G10 size exclusion column (0.8 x 40 cm) and eluted until the head volume reached the top of the column. This loading procedure was repeated with a 2 mL rinse of the residual reaction mixture before a larger volume of water was added to the column. After eluting for -5-10 min, 1-2 mL fractions were collected with the metal complex eluting first, followed by the free ligand and salts. Silica TLC of the fractions was conducted (1:1 AcOEfMeOH) and verification of isolated product was obtained with a UV-lamp and charring of a sulfuric acid sprayed TLC plate. Salts remained white on a charred TLC plate and the ligand had a higher Rf value than the metal complex. The fractions containing the desired compound were combined and the solvent was removed using a rotary evaporator (50 °C, <30 min) to produce an off-white solid. The solid was dried in a vacuum desiccator for at least 48 hours to yield 115.6 mg, 95% yield. Anal. 56 Calcd (found) for C 4 5 H 7 iN 6 0 2 8Ga: C, 45.89 (45.72); H, 5.73 (5.70); N , 7.14 (7.18). 'H NMR (500 MHz, 1:1 CD 3OD:D 20): 8 = 2.343 (s, 3H, C9-C7/3), 3.207 (dd, 1H, 774, V 4 , 5 = 9.8 Hz, V 4 > 3 = 9.8 Hz), 3.354 (s, 3H, Cl-OC// 3 ), 3.441 (dd, 1H, 776b, 2J 6 b,6a = 13.9 Hz, 3 J 6 b j 5 = 6.9 Hz), 3.471 (dd, 1H, HI, 3J2,\ = 3.7 Hz, 3 J 2 j 3 = 9.8 Hz), 3.60 (dd, overlapped with H5, Hi,3J3,2 = 9.8 Hz, 3J 3,4 = 9.8 Hz), 3.61 (ddd, overlapped with H3, 7/5), 4.711 (d, lH-partially overlapped with water peak, 7/1, 3Ji j 2 = 3.7 Hz), 3.658 (dd, 1H, 7/6a, 2J6a,6b = 13.9 Hz, 3 J 6 a i 5 = 2.3 Hz), 4.924 (s, 2H-partially overlapped with water peak, HS), 6.649 (d, 1H, #12, 3Ji 2 ;i3 = 6.7 Hz), 7.521 (d, 1H, H\3,3Ju,n = 6.7 Hz). 1 3C{'H} NMR (125 MHz, 1:1 CD 3OD:D 20): 8 = 13.00 (C9-CH3), 41.36 (C6), 55.94 (CI-OCH3), 58.11 (CS), 70.90 (C5), 72.57 (C2), 72.59 (C4), 74.12 (C3), 100.48 (Cl), 109.04 (C12), 134.43 (C9), 137.24 (C13), 152.57 (C10), 168.61 (Cll) , 169.24 (C7). MS (+ESI): m/z 1163/1165 [ 6 9 / 7 1GaL 3Na] + 1141/1143 [ 6 9 / 7 1 GaL 3 H] + (with similar isotopic distribution as that calculated for C 4 5 H 6 4 N 6 0 2 4 Ga), 783/785 [ 6 9 / 7 1GaL 2] + , 381 [LH+Na]+, 359 [LH+H]+; (+LSIMS): m/z 783/785 [ 6 9 / 7 1GaL 2] + , 359 [LH+H]+. Tris(l-{[((2R,3S,4S,5R,6S)-3,4,5-trihydroxy-6-methoxy-tetrahydropyran-2-yl-methyl)carbamoyl]methyl}-2-methyl-3-oxy-4(l#)-pyridinonato)indium(III) hydrate (In(OG6GP)3-lH 20). This reaction was carried out in a similar fashion as for the analogous gallium complex except that indium nitrate trihydrate was used (35.5 mg, 0.1 mmol). After drying for 48 h, an off-white solid was weighed (107.7 mg, 89% yield). Anal. Calcd (found) for In(OG6GP) 3 C 4 5 H 6 5 N 6 0 2 5 l n : C, 44.86 (45.14); H, 5.44 (5.40); N, 6.98 57 (6.89). ' H N M R (400 M H z , 1:1 CD 3 OD:D 2 0): 5 = 2.367 (s, 3H, C9-C# 3), 3.202 (dd, IH, #4, 3 J 4 > 5 = 9.8 Hz, V 4 > 3 = 9.8 Hz), 3.346 (s, 3H, Cl-OC# 3 ) , 3.429 (dd, IH, #6b, 2 J 6 b , 6 a = 13.8 Hz, V 6 b ,5 = 5.9 Hz), 3.465 (dd, IH, #2, V 2 , i = 3.7 Hz, 3 J 2 j 3 = 9.8 Hz), 3.60 (dd, overlapped with H5, H3,3J3j2 = 9.8 Hz, 3 J 3 > 4 = 9.8 Hz), 3.61 (ddd, overlapped with H3, H5), 3.657 (dd, I H , H6a, 2J6i>6b = 13.6 Hz, 3 J 6 a , 5 = 2.7 Hz), 4.701 (d, partially overlapped with water peak, H\, V i ; 2 = 3.7 Hz), 4.919 (s, 2H-partially overlapped with water peak, HS), 6.618 (d, I H , #12, Vi 2 , i3 = 6.7 Hz), 7.537 (d, I H , #13, 3 Ji 3 > i 2 = 6.7 Hz). 1 3 C{'H} N M R (125 M H z , 1:1 CD 3 OD:D 2 0): § = 13.18 (C9-CH3), 41.38 (C6), 55.94 (Cl-OCH 3 ) , 58.24 (CS), 70.91 (C5), 72.59 (C2), 72.62 (C4), 74.13 (C3), 100.51 (Cl), 110.59 (C12), 136.19 (C9), 136.65 (C13), 152.88 (CIO), 169.13 (C7), 169.13 (Cl 1). MS (+ESI): m/z 1225 [ 1 1 5 InL 3 +K] + (with similar isotopic distribution as that calculated for C 4 5 H 6 3 N 6 0 2 4 K I n ) , 1209 [ 1 , 5 InL 3 +Na] + , 867 [ , l 5 InL 2 K(-H)] + , 851 [ 1 1 5 InL 2 Na(-H)] + , 829 [ 1 1 5 InL 2 ] + , 397 [LH+K] +, 381 [LH+Na]+. l-{[4-(P-D-Glucopyranosyloxy)phenylcarbamoyl]methyl}-3-hydroxy-2-methyl-4(l//)-pyridinone hemihydrate (HOGBAP 0.5H2O). This reaction was carried out in a similar fashion as for HOG6GP with the following changes: the starting material (14, 790 mg, 1.5 mmol) was dissolved in MeOH (40 mL) followed by the addition of water (20 mL) and 160 mg (0.15 mmol) 10% Pd/C. After the solvent was removed using a rotary evaporator, a minimum amount of 2:1 MeOH:water (-10-15 mL) was added to redissolve the product under reflux, and a solid crystallized at RT. The yield after crystallization and drying was 59% (394.8 mg). Anal. Calcd (found) for C 2 oH 2 5 N 2 09. 5 : C, 53.93 (53.68); H, 5.43 (5.54); N , 6.29 (6.42). ' H N M R (400 M H z , 1:1 CD 3 OD:D 2 0): 58 o 5-2.187 (s, 3H, C13-C/73), 3.438 (dd, 1H, HA, V 4 > 5 = 9.8 Hz, 16^15 17 4 OH 3 3J 2 , i = 7.4 Hz, 3 J 2 , 3 = 9.8 Hz), 3.52 (m, partially overlapped V4,3 = 9.8 Hz), 3.49 (dd, partially overlapped with H5, 772, with H3 and H2, 7/5), 3.539 (dd, partially overlapped with NH HO-H5, 7/3, 3J 3 ,2 = 9.8 Hz, 3J 3 ,4 = 9.8 Hz), 3.716 (dd, 1H, //6b, 2J6bM = 12.4 Hz, 3 J 6 b > 5 = 5.6 Hz), 3.887 (dd, 1H, 776a, V 6 a ,6b = 12.4 Hz, 3 J 6 a j 5 = 2.2 Hz), 4.97 (d, partially overlapped with H12 and water peak, HI, 3 J i , 2 = 7.4 Hz), 4.99 (s, partially H O G B A P overlapped with HI and water peak, #12), 6.505 (d, 1H, 7/16, Vie, 17 = 7.3 Hz), 7.108 (d, 2H, 778,3J8,9 = 9.1 Hz), 7.443 (d, 2H, H9,3J9j8 = 9.1 Hz), 7.617 (d, 1H, 7/17, 3J 1 7 >i 6 = 7.3 Hz). 1 3C{'H} NMR (125 MHz, (CD3)2SO): 5 = 11.59 (C13-CH3), 55.52 (C12), 60.75 (C6), 69.76 (CA), 73.26 (C2), 76.63 (C3), 77.01 (C5), 100.76 (Cl), 110.41 (C16), 116.63 (C8), 120.58 (C9), 129.38 (C13), 132.64 (C10), 139.22 (C17), 145.05 (C14), 153.68 (Cl), 165.04 (Cll) , 169.41 (C15). MS (+ESI): m/z 475 [LH+K]+, 459 [LH+Na]+. Tris(l-{[4-(P-D-glucopyranosyloxy)phenylcarbamoyl]methyl}-2-methyl-3-oxy-4(l#)-pyridinonato)gallium(III) dihydrate (Ga(OGBAP) 3-2H 20). To a 100 mL RBF was added H O G B A P 0.5H2O (122.4 mg, 0.315 mmol) and water (15 mL). A solution of 1% HC1 was added dropwise with stirring until pH 1.0, at which time all the ligand dissolved. Gallium nitrate nonahydrate (41.8 mg, 0.1 mmol) was dissolved in 3 mL water in a 4 mL vial and added dropwise to the solution in the RBF, the vial was rinsed with 2 mL water, and added to the RBF; this solution was clear and colourless (pH 1.4). 1% NaOH solution was added dropwise with stirring. At pH~4 a white precipitate formed and was 59 redissolved upon warming (-50 °C). 16 H O H O 14 O O. 3 Ga the cooling solution (until pH~6.5) resulted in more precipitate formation. The pH was adjusted to Continued addition of 1% NaOH to 7.0 after cooling to RT and the Ga(OGBAP) 3 solution was left to stir for 15 min. The precipitate was filtered with a medium frit, washed with cold water (5 mL), and suction-filtered to dryness. The filtrate solvent was reduced to a few mL using a rotary evaporator (50 °C, <30 min), water was added to make a volume of -10 mL and this solution was warmed and applied to a Sephadex G10 column (0.8 x 40 cm, run with water) as 2 x -5 mL portions (-100 mL of water washed the column between runs). TLC of the fractions (1:1 AcOEtMeOH) was used to verify product fractions as described above. The fractions containing the product were combined and the solvent was removed using a rotary evaporator (50 °C, <30 min) to produce a white solid. *H NMR spectra were identical for both the precipitated and column purified product. The product was dried in a vacuum desiccator for at least 48 hours to yield precipitate: 56.6 mg; column: 60.2 mg (116.8 mg combined, 83% yield). 'H NMR (400 MHz, 1:1 CD 3OD:D 20): 5 = 2.371 (s, 3H, C13-C//3), 3.437 (dd, 1H, HA, 3 J 4 > 5 = 9.8 Hz, 3 J 4 > 3 = 9.8 Hz), 3.49 (dd, partially overlapped with H5, HI, 3 J 2 > i = 7.4 Hz, V 2 ; 3 = 9.8 Hz), 3.52 (m, partially overlapped with H3 and H2, 775), 3.537 (dd, partially overlapped with H5, 773,3J3 ; 2 = 9.3 Hz, 3 J 3 j 4 = 9.8 Hz), 3.712 (dd, 1H, H6b, 2J6bM = 12.4 Hz, 3 J 6 b , 5 = 5.6 Hz), 3.881 (dd, 1H, H6a, 2J 6 a ,6b = 1 2 - 4 H z > 3j6a,5 = 2.2 Hz), 4.97 (d, partially overlapped with H12 and water 60 peak, HI, 3 J ] ; 2 = 7.4 Hz), 5.01 (s, partially overlapped with HI and water peak, #12), 6.643 (d, IH, #16, 3J, 6,i 7 = 6.8 Hz), 7.098 (d, 2H, #8, V 8 , 9 = 9.0 Hz), 7.429 (d, 2H, #9, 3J 9,8 = 9.0 Hz), 7.567 (d, IH, #17, 3 J i 7 , i 6 = 6.8 Hz). 1 3C{'H} NMR (100 MHz, (CD3)2SO): 5 = 12.35 (C13-CH3), 56.90 (C12), 60.68 (C6), 69.71 (CA), 73.25 (C2), 76.60 (C3), 76.96 (C5), 100.73 (Cl), 106.14 (C16), 116.56 (C8), 120.48 (C9), 129.134 (C13), 132.71 (CIO), 134.78 (C17), 152.82 (C14), 153.62 (Cl), 164.61 (CH), 168.29 (C15). MS (+ESI): m/z 1397/1399 [6 9 / 7 1GaL3+Na]+ (consistent isotopic distribution), 961/963 [69/71GaL2+Na(-H)]+, 939/941 [ 6 9 / 7 1GaL 2] + , 777/779 [ 6 9 / 7 1 GaL 2 (-C 6 H 1 0 O 5 for 1 glucose molecule)]+, 459 [LH+Na]+. Anal. Calcd (found) for C 6 0 H 7 3 N 6 O 2 9 Ga: C, 51.04 (50.89); H, 5.21 (5.04); N , 5.95 (5.87). Tris(l-{[4-(P-D-gIucopyranosyloxy)phenylcarbamoyl]methyl}-2-methyl-3-oxy-4(l#)-pyridinonato)indium(IH) pentahydrate (In(OGBAP) 3-5H 20). This reaction was carried out in a similar fashion as for the analogous gallium complex except that indium nitrate trihydrate was used (35.5 mg, 0.1 mmol). The product was dried in a vacuum In(OGBAP) 3 desiccator for at least 48 h to yield precipitate: 61.1 mg; column: 68.2 (129.3 mg combined, 86% yield). Anal. Calcd (found) for C 6oH 7 9N 60 3 2In: C, 47.69 (47.37); H, 5.27 (5.03); N , 5.56 (5.30). 'H NMR (400 MHz, 1:1 CD 3OD:D 20): 5 = 2.407 (s, 3H, C13-C#3), 3.437 (dd, IH, #4,3J4,$ = 9.8 Hz, 3J4,3 = 9.8 Hz), 3.49 (dd, partially overlapped with H5, #2, 3J 2 i, = 7.4 Hz, V 2 j 3 = 9.8 61 Hz), 3.52 (m, partially overlapped with H3 and H2, H5), 3.537 (dd, partially overlapped with H5, m, %2 = 9.3 Hz, V 3 , 4 = 9.8 Hz), 3.713 (dd, IH, H6b, 2J6bM = 12.4 Hz, 3 J 6 b ; 5 = 5.6 Hz), 3.882 (dd, IH, 776a, 2J6a>6b = 12.4 Hz, 3 J 6 a ,5 = 2.2 Hz), 4.97 (d, partially overlapped with H12 and water peak, HI, 3 J i > 2 = 7.4 Hz), 5.03 (s, partially overlapped with HI and water peak, 7712), 6.691 (d, IH, 7716, V 1 6 , 1 7 = 6.9 Hz), 7.101 (d, 2H, 778, 3 J 8 j 9 = 9.1 Hz), 7.434 (d, 2H, 779, V 9 ,8 = 9.1 Hz), 7.562 (d, IH, 7717, 3 J , 7 , i 6 = 6.9 Hz). 1 3C{'H} NMR (100 MHz, (CD 3) 2SO): 5 = 12.42 (C13-CH3), 57 (C12), 60.79 (C6), 69.81 (CA), 73.31 (C2), 76.65 (C3), 77.01 (C5), 100.80 (Cl), 107.94 (C16), 116.70 (C8), 120.68 (C9), 131.72 (C13), 132.64 (CIO), 134.30 (C17), 153.31 (C14), 153.76 (C7), 164.61 (Cll) , 168.96 (C15). MS (+ESI): m/z 1443 [115InL3+Na]+ (consistent isotopic distribution), 1007 [ll5InL2+Na(-H)]+-, 985 [ 1 1 5InL 2] +, 823 [ 1 1 5 InL 2 (-C 6 H 1 0 O 5 for 1 glucose molecule)]+, 661 [ 1 1 5 InL 2 ( -C i 2 H 2 0 Oi 0 for 2 glucose molecules)]+, 459 [LH+Na]+. l-[4-(P-D-glucopyranosyloxy)phenyI]-3-hydroxy-2-methyl-4(l//)-pyridinone hydrate (HOGBPP*H 20). This reaction was carried out in a similar fashion as for HOG6GP with the following changes: 18 (1.656 g, 2.0 mmol) was dissolved with MeOH (40 mL). The 10% Pd/C (2.128 g, 2.0 mmol Pd) was wetted with water (10 mL) and the vial was rinsed with MeOH (~10 mL). The reaction mixture was stirred vigorously at RT for 24 h, and following 5 HO 1 filtering and filtrate solvent removal with a rotary H O G B P P evaporator, the product was dissolved in 50 mL water and extracted with 3x 15 mL CH 2 C1 2 . The water was removed using a rotovaporator (50°C) 62 to produce an off-white solid, which was dried in a vacuum desiccator at least 48 hours to yield 652.4 mg (86%). Anal. Calcd (found) for d 8 H 2 3 N 0 9 : C, 54.41 (54.36); H, 5.83 (5.84); N, 3.52 (3.49). 'H NMR (400 MHz, 1:1 CD 3OD:D 20): 5 = 2.097 (s, 3H, C l l -CH3), 3.470 (dd, IH, H4, %5 = 9.8 Hz, 37 4, 3 = 9.8 Hz), 3.57 (dd, partially overlapped with H5, H2), 3.60 (m, partially overlapped with H3 and H2, 775), 3.61 (dd, partially overlapped with H5, 773), 3.735 (dd, IH, H6b, 2J6bM = 12.3 Hz, 3 J 6 b , 5 = 5.7 Hz), 3.915 (dd, IH, H6a, 2J6a,6b = 12.3 Hz, 3J 6 A,5 = 2.0 Hz), 5.101 (d, partially overlapped with water peak, HI, 3 J U = 7.3 Hz), 6.543 (d, IH, 7714, V 1 4 , i 5 = 7.2 Hz), 7.264 (d, 2H, HS, \ 9 = 9.0 Hz), 7.328 (d, 2H, H9,3J9,8 = 9.0 Hz), 7.613 (d, IH, 7715,3J[5M = 7.2 Hz). 1 3C{'H} NMR (100 MHz, 1:1 CD 3OD:D 20): 8 = 14.04 (C11-CH3), 61.75 (C6), 70.63 (CA), 74.12 (Cl), 76.88 (Ci), 11 A3 (C5), 101.28 (Cl), 113.00 (C14), 118.44 (C8), 128.94 (C9), 135.03 (Cll) , 137.13 (C10), 139.84 (C15), 145.73 (C12), 158.67 (C7), 170.99 (C13). MS (+ESI): m/z 402 [LH+Na]+; (+LSIMS): m/z 380 [LH+H]+. Tris{l-[4-(P-D-glucopyranosyloxy)phenyl]-2-methyl-3-oxy-4(l//)-pyridinonato}gallium(III) dihydrate (Ga(OGBPP)3*2H 20). This reaction was carried out in a similar fashion as for ^ G a Ga(OG6GP) 3 with substitution of the ligand (HOGBPP-H 2 0, 125 mg, 0.315 mmol) and the pH after addition of the gallium nitrate solution was 1.4. After drying in the vacuum desiccator at least 48 h, 117.4 mg of an off-white solid was produced (95% yield). Anal. Calcd (found) for OH I |] 1  H O H ° - T 2 ' O H Ga(OGBPP) 3 63 C54H64N3026Ga: C, 52.27 (52.15); H, 5.20 (5.24); N , 3.39 (3.40). *H NMR (400 MHz, 1:1 CD 3OD:D 20): 5 = 2.114 (s, 3H, Cll-C# 3 ), 3.459 (dd, 1H, 7/4, V 4 , 5 = 9.7 Hz, 3 J 4 > 3 = 9.7 Hz), 3.56 (dd, partially overlapped with H5, HI), 3.59 (m, partially overlapped with H3 and H2, H5), 3.60 (dd, partially overlapped with H5, H3), 3.722 (dd, 1H, H6b, 2J6bM = 12.1 Hz, 3 J 6 b , 5 = 5.5 Hz), 3.908 (dd, 1H, H6a, V 6 a , 6 b = 12.1 Hz, V 6 a ,5 = 2.0 Hz), 5.090 3 3 (d, partially overlapped with water peak, H\, = 7.1 Hz), 6.683 (d, 1H, H14, ^14,15 = 6.8 Hz), 7.256 (d, 2H, 7/8, % 9 = 9.0 Hz), 7.337 (d, 2H, H9, %s = 9.0 Hz), 7.556 (d, 1H, 7715, 3 J 1 5 , i 4 = 6.8 Hz). 1 3C{'H} NMR (100 MHz, 1:1 CD 3OD:D 20): 5 = 14.80 ( C l l -CH3), 61.76 (C6), 70.64 (C4), 74.12 (C2), 76.87 (C3), 77.41 (C5), 101.29 (Cl), 108.80 (C14), 118.32 (C8), 128.73 (C9), 134.74 (Cll) , 136.43 (C15), 137.85 (CIO), 152.24 (C12), 158.61 (C7), 168.66 (C13). MS (+ESI): m/z 1226/1228 [6 9 / 7 1GaL3+Na]+ (consistent isotopic distribution), 825/827 [ 6 9 / 7 1GaL 2] + , 402 [LH+Na]+. ' 0 H i I  1 1 ^ i/°'^V9 / \10 N Tris{l-[4-(P-D-glucopyranosyloxy)phenyl]-2-methyl-3-oxy-4(l#)-pyridinonato}indium(III) tetrahydrate (In(OGBPP)3*4H 20). This reaction was carried out in a similar fashion as for the in analogous gallium complex, with the exception that indium nitrate trihydrate was used (35.5 mg, 0.1 mmol) and the pH after its addition to the ligand was 1.8. After drying in the vacuum desiccator for at least 48 h, an off-white solid was produced (120.1 mg, 91% yield). Anal. Calcd (found) for C 5 4 H 6 8 N 3 0 2 8 In: C, 49.06 (48.86); H, 5.18 (4.98); N , 3.18 (3.25). 'H NMR (400 MHz, H O H ° - T r O H In(OGBPP) 3 64 1:1 CD 3OD:D 20): 8-2.106 (s, 3H, Cll-C# 3 ) , 3.456 (dd, 1H, #4, V 4 , 5 = 9.8 Hz, 3 J 4 , 3 = 9.8 Hz), 3.56 (dd, partially overlapped with H5, HI), 3.59 (m, partially overlapped with H3 and H2, H5), 3.60 (dd, partially overlapped with H5, H3), 3.719 (dd, 1H, H6b, 2J6bM = 12.1 Hz, 3 J 6 b ; 5 = 5.6 Hz), 3.906 (dd, 1H, H6a, V 6 a,6b = 12.2 Hz, 3 J 6 a , 5 = 2.1 Hz), 5.086 (d, partially overlapped with water peak, HI, V i , 2 = 7.2 Hz), 6.709 (d, 1H, #14, 3 J i 4 , i 5 = 6.8 Hz), 7.246 (d, 2H, #8, 3J 8 ; 9 = 9.0 Hz), 7.338 (d, 2H, H9,3J9,8 = 9.0 Hz), 7.547 (d, 1H, #15, 3 Ji 5 ; i4 = 6.8 Hz). 1 3C{'H} NMR (100 MHz, 1:1 CD 3OD:D 20): 8 = 14.96 (Cl 1-CH3), 61.78 (C6), 70.67 (C4), 74.14 (C2), 76.91 (C3), 77.44 (C5), 101.34 (Cl), 110.38 (C14), 118.34 (CS), 128.68 (C9), 135.87 (C15), 136.34 (Cll) , 137.86 (CIO), 152.49 (C12), 158.65 (Cl), 169.17 (C13). MS (+ESI): m/z 1272 [115InL3+Na]+ (consistent isotopic distribution), 871 [ 1 1 5InL 2] +, 402 [LH+Na]+. l-{[((2R,3S,4S,5R,6S)-3,4,5-Triacetoxy-6-methoxy-tetrahydropyran-2-yl-methyl)carbamoyl]methyl}-3-hydroxy-2-methyl-4(l#)-pyridinone hydrate (HAG6GP'1H 2 0). This reaction was carried out in a similar fashion as for HOG6GP O with the following changes: 90.5H2O (950 mg, 1.63 mmol) was dissolved in MeOH (40 mL) and 175 mg of 10% Pd/C (0.164 mmol Pd) was used (wetted with 5 mL H 2 0, rinsed with 5 mL MeOH) and the reaction was stopped after 20 h at RT. The filtered Pd/C was rinsed and suction-filtered with 3x10 mL MeOH taking A c ° 0 ^ caution not to completely dry the Pd/C in the frit. After filtering HAG6GP through a 0.2 am filter, the solvent was removed using a rotary evaporator (40°C, <30 min). The solid was redissolved in a minimum amount of hot 1:1 65 water:EtOH and recrystallized at RT. Filtering the precipitate with a medium frit and rinsing with 2x 3 mL cold ethanol produced a white solid and an orange filtrate. The product was dried in a vacuum desiccator over phosphorus pentoxide for at least 48 hours before elemental analysis was performed and weighing to determine the yield 321.5 mg (71%). Anal. Calcd (found) for C21H30N2O12: C, 50.20 (50.52); H, 5.72 (6.02); N , 5.52 (5.58). 'H NMR (400 MHz, 1:1 CD 3OD:CD 3CN): 5 = 1.992 (s, 3H, COC//3), 2.000 (s, 3H, COC//3), 2.048 (s, 3H, COC/73), 2.314 (s, 3H, C9-C//3), 3.382 (dd, 1H, //6b, 2 J 6 b , 6 a = 14.5 Hz, 3 J 6 b , 5 = 5.5 Hz), 3.422 (s, 3H, CI-OC//3), 3.555 (dd, 1H, //6a, V 6 a ; 6 b = 14.5 Hz, 3 J 6 a , 5 = 5.5 Hz), 3.924 (ddd, 1H, HS, 3J5,4 = 9.9 Hz, 3J5.6a = 5.5 Hz, 3J5,6b = 2.6 Hz), 4.720 (s, 2H, HS), 4.918 (dd, partially overlapped with HI, HI, 3J 2 , i =3.6 Hz, V 2 ,3 = 9.9 Hz), 4.944 (d, partially overlapped with H2 and H3, HI, 3 J i , 2 = 3.6 Hz), 4.962 (dd, partially overlapped with HI, HA, 3 J 4 > 3 = 9.9 Hz, 3 J 4 , 5 = 9.9 Hz), 5.387 (dd, 1H, H3,3J3,2 = 9.9 Hz, 3 J 3 ] 4 = 9.9 Hz), 6.379 (d, 1H, 7712, 3 J ! 2 j i 3 = 7.3 Hz), 7.476 (d, 1H, 7/13, 3 J , 3 , i 2 = 7.3 Hz). I 3C{'H} NMR (100 MHz, 1:1 CD 3OD:D 20): 5 = 12.17 (C9-CH3), 20.71 (COCH3), 20.78 (2xCOCH3), 40.10 (C6), 55.92 (CI-OCH3), 56.60 (CS), 68.93 (CS), 70.13 (C4), 70.97 (Ci), 71.72 (C2), 97.92 (Cl), 112.01 (C12), 132.78 (C9), 140.40 (C13), 146.40 (C10), 168.25 (C7), 170.89 (Cl 1), 171.23 (C4-COCH3), 171.50 (C2-COCH3), 171.53 (C3-COCH3). MS (+ESI): m/z 507 [LH+Na]+, 485 [LH+H]+. Tris(l-{[((2R,3S,4S,5R,6S)-3,4,5-triacetoxy-6-methoxy-tetrahydropyran-2-yl-methyl)carbamoyl]methyl}-2-methyl-3-oxy-4(l#)-pyridinonato)gallium(III) hydrate (Ga(AG6GP) 3 -lH 2 0). This reaction was carried out in a similar fashion as for Ga(OG6GP) 3 with the following changes: HAG6GP 1H20 (157.6 mg, 0.315 mmol) 66 was dissolved in 4 mL 1:1 EtOH:water. Gallium nitrate A c O - " AcO-AcO Ga nonahydrate (41.8 mg, 0.1 mmol) was dissolved in 2 mL containing the ligand. The vial for the gallium nitrate 1:1 H 20:EtOH and added dropwise to the solution O was rinsed with 2 mL of 1:1 EtOH:water. The pH meter Ga(AG6GP) 3 read 2.4 for this reaction mixture and a 1% NaOH solution was added dropwise until the pH meter read 7.0. The solvent was removed using a rotary evaporator at 50 °C (<30 min) to produce a white precipitate. The precipitate was redissolved in 3 mL 1:1 EtOH:water and applied on a Sephadex G10 size exclusion column (0.8 x 40 cm) run in 1:1 EtOH:water. The relevant fractions were combined and the solvent was removed using a rotary evaporator (50 °C, <30 min). The resulting solid was dried in a vacuum desiccator for at least 48 h to yield 138.6 mg (90%). Anal. Calcd (found) for C 6 3 H 8 3 N 6 0 3 4 Ga: C, 49.20 (49.11); H, 5.44 (5.46); N, 5.46 (5.63). 'H NMR (400 MHz, 1:1 CD 3OD:CD 3CN): 5 = 1.988 (s, 3H, COC773), 1.998 (s, 3H, COC/73), 2.044 (s, 3H, COC/73), 2.328 (s, 3H, C9-C/73), 3.357 (dd, IH, 776b, 2J6bM = 14.5 Hz, 3 J 6 b , 5 = 5.5 Hz), 3.407 (s, 3H, Cl-OC/7 3), 3.520 (dd, IH, 776a, 2 J 6 a , 6 b = 14.5 Hz, 3 J 6 a , 5 = 5.5 Hz), 3.918 (ddd, IH, 775,3J5A = 9.8 Hz, 3J5M = 5.5 Hz, 3J5,6b = 2.6 Hz), 4.782 (s, 2H, ffl), 4.909 (dd, partially overlapped with HI, 772, 3J2,i = 3.6 Hz, V 2,3 = 9.8 Hz), 4.929 (d, partially overlapped with H2 and H3, HI, 3J\>2 = 3.6 Hz), 4.958 (dd, partially overlapped with HI, HA, 3J4,3 = 9.8 Hz, 3 J 4 , 5 = 9.8 Hz), 5.379 (dd, IH, 773, 3J3,2 = 9.8 Hz, 3 J 3 i 4 = 9.8 Hz), 6.498 (d, IH, 7712, 3J 1 2 , i3 = 6.9 Hz), 7.395 (d, IH, 7713, 3 7 1 3 > 1 2 = 6.9 Hz). 1 3C{'H} NMR (75 MHz, 1:1 CD 3OD:D 20): 5 = 12.72 (C9-CH3), 20.70 (COCH3), 20.79 (2xCOCH3), 40.27 (C6), 55.98 (Cl-OCH 3), 57.81 (CS), 68.98 67 (C5), 70.33 (C4), 71.02 (C3), 71.76 {Cl), 97.91 (Cl), 108.08 (C12), 132.79 (C9), 136.21 (C13), 153.88 (CIO), 168.02 (Cl), 169.57 (CU), 171.17 (C4-COCH3), 171.44 (C2-COCH3), 171.50 (C3-COCH3). MS (+ESI): m/z 1541/1543 [6 9 / 7 1GaL3+Na]+ (consistent isotopic distribution), 1035/1037 [ 6 9 / 7 1GaL 2] + , 507 [LH+Na]+, 485 [LH+H]+. Tris(l-{[((2R,3S,4S,5R,6S)-3,4,5-triacetoxy-6-methoxy-tetrahydropyran-2-yl-methyl)carbamoyl]methyl}-2-methyl-3-oxy-4(l//)-pyridinonato)indium(III) tetrahydrate (In(AG6GP)3'4H 20). This reaction was carried out in a similar fashion as for the analogous gallium complex, with the exception that indium nitrate trihydrate was used (35.5 mg, 0.1 mmol). After drying in the vacuum desiccator for at least 48 h, a white solid was produced (172.1 mg, 93% 3 yield). Anal. Calcd (found) for C^H^NeOsyln: C, 46.22 In(AG6GP) 3 (45.91); H, 5.48 (5.24); N , 5.13 (5.27). 'H NMR (400 MHz, 1:1 CD3ODCD3CN): 5 = 1.988 (s, 3H, COCf/3), 1-997 (s, 3H, COC/73), 2.043 (s, 3H, COC//3), 2.356 (s, 3H, C9-CH3), 3.360 (dd, IH, 776b, 2J6bM = 14.5 Hz, V 6 b , 5 = 5.7 Hz), 3.404 (s, 3H, Cl-OC/7 3), 3.522 (dd, IH, 776a, 2J6a,6b = 14.4 Hz, 376 a,5 = 5.7 Hz), 3.907 (ddd, IH, 775, 3 7 5 ) 4 = 9.8 Hz, 3J5M = 5.7 Hz, 375,6b = 2.5 Hz), 4.786 (s, 2H, 778), 4.910 (dd, partially overlapped with HI, 772, 3 J 2 ; i = 3.6 Hz, 3 J 2 j 3 = 9.8 Hz), 4.927 (d, partially overlapped with H2 and H3, 771,3Ji j 2 = 3.6 Hz), 4.956 (dd, partially overlapped with HI, 774,3J4,3 = 9.8 Hz, 37 4 > 5 = 9.8 Hz), 5.378 (dd, IH, 773, 373,2 = 9.8 Hz, 3 7 3 ; 4 = 9.8 Hz), 6.531 (d, IH, 7712, 3 J , 2 j l 3 = 7.0 Hz), 7.386 (d, IH, 7713,37i3;i2 = 7.0 Hz). 1 3C{'H} NMR (100 MHz, 1:1 CD 3OD: CD 3CN): 5 = 12.90 (C9-CH3), 20.70 (COCH3), 20.79 68 (2xCOCH3), 40.23 (C6), 55.96 (Cl-OCH 3), 57.93 (C8), 68.95 (C5), 70.28 (CA), 71.00 (C3), 71.74 (C2), 97.88 (Cl), 109.76 (C12), 134.79 (C9), 135.61 (C13), 154.15 (CIO), 168.00 (C7), 170.05 (Cll) , 171.17 (C4-COCH3), 171.45 (C2-COCH3), 171.51 (C3-COCH3). MS (+ESI): m/z 1587 [115InL3+Na]+ (consistent isotopic distribution), 1081 [ 1 1 5InL 2]+, 507 [LH+Na]+, 485 [LH+H]+. O H AcO AcO-l-{[4-(2,354,6-Tetra-0-acetyl-p-D-glucopyranosyloxy)phenylcarbamoyl]methyl}-3-hydroxy-2-methyl-4(l#)-pyridinone hydrate (HAGBAP-1H 2 0). This reaction was carried out in a similar fashion as for HAG6GP with the following changes: 11 (507.0 mg, 0.71 mmol) and 76 mg 10% Pd/C (0.071 mmol Pd) were used and the reaction was stopped after 24 h at RT. The product was recrystallized from hot 1:1 MeOFLwater (10 mL). After filtering out a white solid, an orange filtrate remained. After drying, the product weighed 321.5 mg (71%). Anal. Calcd (found) for C 2 8 H 3 2 N 2 O i 3 : C, 54.02 (54.12); H, 5.50 (5.45); N, 4.50 (4.59). 'H NMR (400 MHz, 1:1 CD 3OD:CD 3CN): 5 = 2.011 (s, 3H, COC/73), 2.038 (s, 3H, COC#3), 2.054 (s, 3H, COC#3), 2.055 (s, 3H, COC#3), 2.342 (s, 3H, C13-C#3), 4.123 (ddd, partially overlapped with CD 3OH/H 20 peak, #5, 3J 5,4 = 9.7 Hz, 3J5M = 5.3 Hz, 3J 5 >6b = 2.3 Hz), 4.167 (dd, 1H, #6b, 2J6b,6a = 12.3 Hz, V 6 b , 5 = 5.3 Hz), 4.291 (dd, 1H, #6a, 2J 6 a ,6b = 12.3 Hz, 3J 6 a ,5 = 2.3 Hz), 4.857 (s, 2H, #12), 5.136 (dd, 1H, #4, 3J 4 ; 5 = 9.7 Hz, V 4 > 3 = 9.7 Hz), 5.184 (dd, 1H, HI, 3J2,\ = 8.0 Hz, 3 J 2 ; 3 = 9.7 Hz), 5.287 (d, 1H, #1, 3 J 1 ; 2 = 8.0 Hz), 5.382 (dd, 1H, #3, 3 J 3 ; 2 = 9.8 Hz, 3 J 3 > 4 = 9.8 Hz), 6.402 (d, 1H, #16, A c O H A G B A P 69 V,6,i7 = 7.2 Hz), 7.031 (d, 2H, #8, 3 J 8 ; 9 = 9.0 Hz), 7.524 (d, 2H, H9,3J9,8 = 9.0 Hz), 7.535 (d, 1H, #17, 3Ji7,i6 = 7.2 Hz). l3C{lH} NMR (100 MHz, 1:1 CD 3OD:CD 3CN): 8 = 12.24 (C13-CH3), 20.75 (4xCOCH3), 57.08 (C12), 62.87 (C6), 69.40 (C4), 72.21 (C2), 72.82 (C5), 73.54 (C3), 99.82 (Cl), 112.02 (C16), 118.26 (C8), 122.49 (C9), 132.74 (C13), 134.48 (CIO), 140.53 (C17), 146.42 (C14), 154.65 (C7), 166.15 (Cl 1), 170.94 (C15), 170.94 (C2-COCH3), 171.07 (C4-COCH3), 171.41 (C3-COCH3), 171.93 (C6-COCH3). MS (+ESI): m/z 627 [LH+Na]+, 605 [LH+H]+. Tris(l-{[4-(2,3,4,6-tetra-(?-acetyl-P-D-glucopyranosyloxy)phenylcarbamoyl]methyl}-2-methyl-3-oxy-4(l#)-pyridinonato)gallium(III) tetrahydrate (Ga(AGBAP) 3-4H 20). This reaction was carried out in a similar fashion as for 11 13| OAc n" IN 3 OAc 8 Ga(AGBAP) 3 Ga(AG6GP) 3 with the following changes: H A G B A P 1H 20 (196.1 mg, 0.315 mmol) was dissolved 1:1 EtOH:water (10 mL). After the addition of gallium nitrate, the pH meter read 2.2. When dropwise addition of 1% NaOH produced a precipitate, the reaction mixture was heated to ~40°C. As the reaction mixture cooled, 1% NaOH solution was added until -neutral pH and more precipitate formed. The solution was cooled on ice 15 min, the precipitate was filtered with a medium frit, rinsed with 5 mL cold water, and filtered to dryness. The solid was dried in a vacuum desiccator for at least 48 hours to yield 172.1 mg (88%). Anal. Calcd (found) for C 8 4HioiN 604 3Ga: C, 70 51.67 (51.60); H, 5.21 (5.18); N , 4.30 (4.54). ] H NMR (400 MHz, 1:1 CD 3OD:CD 3CN): 5 = 2.010 (s, 3H, COC/73), 2.037 (s, 3H, COC//3), 2.044 (s, 3H, COC/73), 2.051 (s, 3H, C O C / / 3 ) , 2.340 (s, 3H, CI3 -C / /3 ) , 4.049 (ddd, partially overlapped with CD 3OH/H 20 peak, 775,3J5A = 9.6 Hz, V 5 , 6 a = 5.2 Hz, V 5 ; 6 b = 2.2 Hz), 4.163 (dd, IH, H6b, 2J6bM = 12.3 Hz, 3 J 6 b > 5 = 5.2 Hz), 4.288 (dd, IH, 776a, 2 J 6 a ; 6 b = 12.3 Hz, 376a,5 = 2.2 Hz), 4.895 (s, 2H, H12), 5.130 (dd, IH, 774, \ 5 = 9.6 Hz, 3 J 4 , 3 = 9.6 Hz), 5.167 (dd, IH, H2,3J2,i = 8.0 Hz, 3 J 2 j 3 = 9.6 Hz), 5.282 (d, IH, HI, 3 J U = 8.0 Hz), 5.376 (dd, IH, H3,373>2 = 9.6 Hz, 3 J 3 j 4 = 9.6 Hz), 6.509 (d, IH, 7716, V i 6 , 1 7 = 6.3 Hz), 7.014 (d, 2H, 778,378;9 = 8.7 Hz), 7.450 (d, IH, 7717, 3 J 1 7 , i6 = 6.3 Hz), 7.505 (d, 2H, 779,379)8 = 8.7 Hz). I 3C{'H} NMR (100 MHz, 1:1 CD 3OD:CD 3CN): 5 = 12.87 (C13-CH3), 20.78 (COCH3), 20.81 (2xCOCH3), 20.87 (COCH3), 58.13 (C12), 62.85 (C6), 69.39 (CA), 72.23 (Cl), 72.81 (CS), 73.57 (C3), 99.78 (Cl), 108.22 (C16), 118.20 (CS), 122.37 (C9), 133.08 (C13), 134.58 (CIO), 136.59 (C17), 153.73 (C14), 154.50 (Cl), 165.88 (Cll) , 169.45 (C15), 170.94 (C2-COCH3), 171.05 (C4-COCH3), 171.39 (C3-COCH3), 171.92 (C6-COCH3). MS (+ESI): m/z 1879/81 [ 6 9 / 7 1GaL 3+H] + (consistent isotopic distribution), 1275/1277 [ 6 9 / 7 I GaL 2 ] + , 605 [LH+H]+. Tris(l-{[4-(2,3?4,6-tetra-0-acetyl-p-D-glucopyranosyloxy)phenylcarbamoyl]methyl}-2-methyl-3-oxy-4(l#)-pyridinonato)indium(III) dihydrate (In(AGBAP) 3-2H 20). This reaction was carried out in a similar fashion as for the analogous gallium complex, with the exception that indium nitrate trihydrate was used (35.5 mg, 0.1 mmol). After drying, 178.3 mg of a white solid was produced (91% yield). Anal. Calcd (found) for C 8 4H 97N 604iIn: C, 51.44 (51.19); H, 4.98 (4.79); N , 4.28 (4.30). 'H NMR (300 MHz, 71 1:1 CD 3OD:CD 3CN): 6 = 2.010 (s, 16 17 O. 3H, COC#3), 2.007 (s, 3H, COC#3), O In 14 O 2.033 (s, 6H, 2xCOC#3), 2.044 (s, C#3), 4.046 (ddd, partially 3H, COC#3), 2.341 (s, 3H, C D -overlapped with CD3OH/H2O peak, In(AGBAP) 3 #5, 3J5A = 9.6 Hz, 3 J 5 ,6a = 5.2 Hz, 3 J 5 i 6b = 2.3 Hz), 4.158 (dd, IH, H6b, V 6 b ; 6 a = 12.3 Hz, 3 J 6 b , 5 = 2.3 Hz), 4.287 (dd, IH, H6a, 2 J 6 a ,6b = 12.3 Hz, 3 J 6 a , 5 = 5.2 Hz), 4.885 (s, 2H, #12), 5.128 (dd, IH, #4, 3J 4, 5 = 9.6 Hz, 3 J 4 ,3 = 9.6 Hz), 5.177 (dd, IH, H2, V 2 , i = 7.98.0 Hz, V 2 j 3 = 9.6 Hz), 5.280 (d, IH, HI, 3 J i > 2 = 7.9 Hz), 5.376 (dd, IH, H3, 3J3,2 = 9.6 Hz, 3 J 3 , 4 = 9.6 Hz), 6.532 (d, IH, #16, 3 J 1 6 > i 7 = 6.3 Hz), 7.001 (d, 2H, #8, 3J 8, 9 = 8.6 Hz), 7.424 (d, IH, #17, Vn.ie = 6.3 Hz), 7.488 (d, 2H, #9, 3J 9 ,8 = 8.6 Hz). 1 3C{'H} NMR (75 MHz, 1:1 CD 3OD:CD 3CN): 5 = 13.01 (CI3-CH3), 20.78 (3xCOCH3), 20.83 (COCH3), 58.43 (C12), 62.87 (C6), 69.42 (C4), 72.25 (C2), 72.83 (C5), 73.59 (C3), 99.82 (Cl), 109.85 (C16), 118.22 (C8), 122.45 (C9), 134.54 (CIO), 134.92 (C13), 135.87 (C17), 154.10 (C14), 154.58 (C7), 165.85 (Cll) , 170.08 (C15), 170.93 (C2-COCH3), 171.05 (C4-COCH3), 171.38 (C3-COCH3), 171.91 (C6-COCH3). MS (+ESI): m/z 1947 [115InL3+Na]+ (consistent isotopic distribution), 1925 [115InL3+H]+, 1321 [ 1 1 5InL 2] +, 627 [LH+Na]+, 605 [LH+H]+. 3-Hydroxy-l-(4-hydroxyphenyl)-2-methyl-4(l/T)-pyridinone (Hhpp). This reaction was carried out in a similar fashion as for HOG6GP with the following changes: the starting material (17, 1.537 g, 5.0 mmol) was dissolved in MeOH (50 mL) and DMF (50 72 O mL). The 10% Pd/C (532 mg, 0.50 mmol Pd) was wetted and transferred to the RBF using ~10 mL of water. The reaction was stopped after 12 h at RT. The filtered Pd/C was rinsed and suction-filtered with 3x10 mL warm MeOH (taking caution not to completely dry the Pd/C in the frit), and 3x10 mL warm DMF (-60 °C). The solvent of the light pink-orange filtrate was removed using a rotary evaporator (50 °C, ~1.5 h) and the solid dried in vacuo for 2 h. The resulting light pink solid was dissolved in a minimum of hot MeOH, -25% more MeOH was added, and a fine precipitate formed on cooling. The precipitate was filtered with a medium frit, and rinsed with 2x3 mL cold MeOH to produce a white solid and orange filtrate. A second crop of white precipitate was obtained from the orange filtrate. The precipitates were combined and dried in a vacuum desiccator for at least 48 hours to yield 381 mg (35%). 'H NMR (400 MHz, (CD3)2SO): 5 = 1.931 (s, 3H, C2-C/f3), 6.164 (d, 1H, H5, V 5 i 6 = 7.3 Hz), 6.859 (d, 2H, arom H,J = 8.7 Hz), 7.204 (d, 2H, arom H,J= 8.7 Hz), 7.473 (d, 1H, H6,3J6,5 = 7.3 Hz). MS (+ESI): m/z 218 [M+H]+. Anal Calcd (found) for Ci 2 Hi,N0 3 : C, 66.35 (65.89); H, 5.10 (4.98); N, 6.45 (6.33). Tris(l-(4-hydroxyphenyl)-2-methyl-3-oxy-4(l//)-pyridinonato)gallium(III)-sesquihydrate (Ga(hpp)3*1.5H20). Method A: Gallium nitrate nonahydrate (208.9 mg, 0.50 mmol) was dissolved in DMF (5 mL). This solution was added dropwise to a stirring solution of Hhpp (329.8 mg, 1.52 mmol) at RT in 40 mL of DMF. A 1.0 % NaOH Ga(hpp) 3 73 solution (6.2 mL) was added dropwise to the reaction mixture and the solvent was removed using a rotary evaporator at 55 °C. After drying on the vacuum line for 16 h, the precipitate was transferred to a medium frit, stirred with warm water and filtered (3x20 mL). The resulting white precipitate was dried in a vacuum desiccator for at least 48 h to yield 319.2 mg (86%). Anal. Calcd (found) for C36H30N3O9: C, 58.01 (57.99); H, 4.46 (4.25); N , 5.64 (5.73). ! H NMR (400 MHz, (CD3)2SO): 5= 1.832 (s, 3H, C2-Ctf3), 6.415 (d, 1H, H5,3J5,6 = 6.7 Hz), 6.887 (d, 2H, arom H,J= 8.5 Hz), 7.233 (d, 2H, arom H, J = 8.5 Hz), 7.474 (d, 1H, H6, 3J6,5 = 6.7 Hz). MS (+ESI): m/z 718/720 [ 6 9 / 7 1GaL 3+H]+ (consistent isotopic distribution), 501/503 [ 6 9 / 7 1GaL 2] + , 218 [LH+H]+. Method B: Enzymatic cleavage. To a 1 mL Eppendorf tube was added 30 uL from a 50 mM solution of Ga(OGBPP)3 and 65 uL of 75 mM sodium phosphate buffer (pH 6.8). The solution was mixed briefly (vortex mixer) and 5 uL of a buffered solution containing 3.98 mg/mL of Abg was added to start the reaction. A white precipitate formed within a few minutes. After 30 min at RT, the reaction mixture (with some white precipitate) was spotted a few times on a TLC plate (dried at RT in between spotting). TLC in 1:1 AcOEt:MeOH was run -45 min. after addition of enzyme. The TLC indicated a new spot (non-charring) with the same Rf value (0.55) as the Ga(hpp)3 precipitate from Method A. After -1 h, the white precipitate was filtered (2 mL fine frit). TLC of the supernatant and filtered precipitate indicated that glucose (Rf = 0.35) was the only charred spot, present in the supernatant. The precipitate was dried 2 h in a vacuum desiccator producing -1 mg of a white solid. MS (+ESI): m/z 740/742 [6 9 / 7 1GaL3+Na]+ (consistent isotopic distribution), 501/503 [ 6 9 / 7 1GaL 2]+,218 [LH+H]+. 74 2.3 Results and Discussion 2,3.1 Ligand Preparations The syntheses of all five ligand precursors HOG6GP, H O G B A P , HOGBPP, HAG6GP, and H A G B A P were prepared by various routes from maltol (1). The hydroxyl group of maltol was benzyl protected using benzyl chloride to produce 2. Glycine was reacted with 2 under basic conditions to produce the corresponding pyridinone 3. DCC was used to activate the carboxylic acid group of 3 as the NHS ester, that was then reacted in situ with amines 7 and 13 to produce the amide coupled compounds 8 (Scheme 2.1) and 14 (Scheme 2.2), respectively. Debenzylation of 8 and 14 was accomplished by hydrogenolysis using 10% Pd/C to produce HOG6GP and H O G B A P , respectively. o o o 4 5 6 7 Scheme 2.1: Synthesis of HOG6GP: (a) BnCl, MeOH/H 20, 71%; (b) Glycine, MeOH/H 20, 48%; (c) TsCl, pyridine, 30%; (d) NaN 3, acetone, 97%; (e) Pd black, 60 atm H 2 , MeOH, 73%; (f) DCC/NHS, DMF, 76%; (g) 10% Pd/C, 1 atm H 2 , MeOH/H 20, 53%. 75 A c O H O H O 11 12 13 Scheme 2.2: Synthesis of H O G B A P : (a) 4-nitrophenol, Ag 2 C0 3 , MeCN, 68%; (b) NaOMe, MeOH/ CH 2C1 2, 99%; (c) 10% Pd/C, H 2 , MeOH/H 20, 87%; (d) DCC/NHS, DMF, 71%; (e) 10% Pd/C, 1 atm H 2 , MeOH/H 20, 59%. Amine 7 (Scheme 2.1) was prepared starting from methyl a-D-glucopyranoside (4) that was tosylated8 to produce 5 and further reacted with sodium azide, according literature preparations,9 to afford 6. Hydrogenation of the azide was accomplished at 60 atm using palladium black as a catalyst producing amine 7. Amine 13 (Scheme 2.2) was prepared starting from 2,3,4,6-tetra-O-acetyl-a-D-12 glucopyranosyl bromide (10). Koenigs-Knorr conditions with silver carbonate were used to convert 10 to 4-nitrophenyl 2,3,4,6-tetra-O-acetyl-P-D-glucopyranoside (11). Deacetylation with sodium methoxide produced 4-nitrophenyl p-D-glucopyranoside (12) and the nitro group was hydrogenated with 10% Pd/C as a catalyst to produce the amine (13) that was recrystallized as the hydrochloride salt. 76 The synthesis of HOGBPP is shown in Scheme 2.3. 4-Aminophenol was reacted with benzyl protected maltol (2) under slightly acidic conditions to produce 3-benzyloxy-l-(4-hydroxyphenyl)-2-methyl-4(l#)-pyridinone (17). ADDP and tributylphosphine Scheme 2.3: Synthesis of HOGBPP: (a) 4-aminophenol, H 20/MeOH, 29%; (b) ADDP, Bu3P, CH2C12, 66%; (c) 10% Pd/C, 1 atm H 2 , MeOH/H 20, 86%. were used to couple the phenolic moiety of 17 to 2,3,4,6-tetra-O-benzyl-D-glucopyranose (16) under Mitsunobu conditions,14 which produced the P-glucoside (18) after column chromatography. Debenzylation of 18 with H 2 and 10% Pd/C afforded HOGBPP. The syntheses of HAG6GP and H A G B A P are shown in Schemes 2.4 and 2.5, respectively. Compounds 8 and 14 were acetylated using acetic anhydride in pyridine to produce 9 and 15, respectively. Debenzylation of 9 and 15 with H2 and 10% Pd/C afforded HAG6GP and H A G B A P , respectively. 77 2.3.2 Tris(3-oxy-4-pyridinato)gallium(III) and Indium(III) Complexes The neutral gallium(III) and indium(III) complexes were prepared by addition of gallium or indium nitrate solutions to solutions containing the precursor ligands. The acidic solutions containing the reaction mixtures were basified to neutral pH with sodium hydroxide and subsequent column purification (G10) or precipitate collection yielded the metal complexes (Scheme 2.6). o o o 8 9 H A G 6 G P Scheme 2.4: Synthesis of HAG6GP: (a) Ac 2 0, pyridine, 77%; (b); 10% Pd/C, 1 atm H 2 , MeOH/H 20, 71%. HO AcO AcO 14 15 H A G B A P Scheme 2.5: Synthesis of H A G B A P : (a) Ac 2 0, pyridine, 74%; (b); 10% Pd/C, 1 atm H 2 , MeOH/H 20, 71%. 78 Elemental analyses of all the metal complexes were consistent with the calculated values. All metal complexes had 1 to 5 waters of hydration that were not eliminated after at least 48 h in a vacuum desiccator at RT. These waters are consistent with other gallium and indium pyridinone complexes with 0.5-5.5 waters of hydration often being NH RO-RO RO OR R O ; RO- OR M OH H O H O to " O H M 04 M M = Ga, R = H M = Ga, R = Ac M = In, R = H M = In, R = Ac M = Ga, R = H M = Ga, R = Ac M = In, R = H M = In, R = Ac M = Ga, R = H M = In, R = H Ga(OG6GP)3 Ga(AG6GP)3 In(OG6GP)3 In(AG6GP)3 Ga(OGBAP)3 Ga(AGBAP)3 In(OGBAP) 3 In(AGBAP)3 Ga(OGBPP)3 In(OGBPP)3 HO Scheme 2.6: Synthesis of Ga(OG6GP) 3 (95 %)a, In(OG6GP) 3 (89 %)b, Ga(AG6GP) 3 (83 %)c, In(AG6GP) 3 (86 %)d, Ga(OGBAP) 3 (95 %)a, In(OGBAP) 3 (91 %)b, Ga(AGBAP) 3 (90 %)c, In(AGBAP) 3 (93 %)d, Ga(OGBPP) 3 (88 %)a, and In(OGBPP) 3 (91 %)b. aGa(N0 3) 3-9H 20, H 2 0; bIn(N0 3) 3-3H 20, H 2 0 ; 0 Ga(N03)3-9H20, H 2 0:EtOH; d In(N03)3-3H20, H 20:EtOH. 79 reported,15 although up to 12 waters of hydration have been reported in some solid state structures.16'17 Drying some gallium and indium pyridinone complexes at 85°C in vacuo for 24 h failed to remove all waters of hydration from these complexes.15 The mass spectra (+ESI) for the metal complexes revealed protonated, sodiated, or sometimes potassiated parent molecules. These parent ions had the expected fragmentation patterns and isotopic distributions. 2.3.3 'H and13C NMR Spectra H NMR indicated the pyridinone ring protons had the most prominent changes in chemical shift upon metal complexation. This is illustrated by a representative 'H NMR plot in Figure 2.2. The largest change was for H5 (see Table 2.1 for pyridinone ring numbering) where metal complexation resulted in downfield shifts of 0.11-0.19 ppm for the complexes studied. These are close to the 0.19-0.32 ppm downfield shifts observed for other (crystallographically characterized) gallium and indium pyridinone complexes in CDCI3 and (CD3)2S0.1 5 Significant changes were also observed in this work for H6 with upfield shifts from 0.04-0.11 ppm occurring upon metal complexation. Gallium(III) and indium(III) ions have the potential to coordinate alkoxy groups • * 18 19 20 from citric acid ' or D-gluconic acid. The overlap of the glucose protons in some of the 'H NMR spectra in this work (3.4-3.7 ppm) made it difficult to ascertain whether there were any interactions between the hydroxyl groups of glucose and the gallium or indium metal centres. 1 3 C NMR spectra were obtained and assigned using 2D 80 Table 2.1: 1 3 C NMR A8 for the tris(3-oxy-4-pyridinato)gallium and indium complexes versus the free ligands. Only the pyridinone and glucose moieties are considered. A positive/negative value indicates a downfield/upfield shift for the metal complexes compared to the free ligand. Pyridinone or glucose ring numbering for comparison Pyridinone or glucose ring carbon atoms,.-, 1 3 C A5 from HOG6GP for Ga(OG6GP) 3, In(OG6GP) 3 1 3 C A8 from H O G B A P for Ga(OG6GP) 3 , In(OG6GP) 3 1 3 C A8 from HL for HOGBPP GaL3, InL3 0 2 (pyridinone) -0.16, +1.60 +0.25, +2.34 -0.29,+1.31 3 +6.48, +6.77 +7.77, +8.26 +6.51,+6.76 4 -2.34, -1.81 -1.12,-0.45 -1.22,-0.71 5 -4.09, -2.54 -4.27, -2.47 -4.11,-2.53 6 -3.39, -3.97 .4.44, .4.92 -3.41,-4.02 2-CH 3 +0.70, +0.88 +0.76, +0.83 +0.76, +0.92 H O ~ - ^ ^ ^ J . 3 HO 1 1 (glucose) -0.07, -0.04 -0.03, +0.04 +0.01,+0.06 2 -0.06, -0.04 -0.01,+0.05 0.00, +0.02 3 -0.04, -0.03 -0.03, +0.02 -0.01,+0.03 4 -0.04, -0.01 -0.05, +0.05 +0.01,+0.04 5 -0.02, -0.01 -0.05, 0.00 -0.02, +0.01 6 -0.03, -0.01 -0.07, +0.04 +0.01,+0.03 techniques to aid in determining whether any glucose-metal direct interactions take place. Also, a comparison of the pyridinone 1 3 C NMR region upon complexation to gallium and indium has not been previously undertaken (to our knowledge). The 1 3 C NMR chemical shift differences upon metal complexation for HOG6GP, H O G B A P and HOGBPP are listed in Table 2.1. There are very little chemical shift differences for the glucose carbon atoms (all less than 0.1 ppm shifts) indicating that the hydroxyl groups of glucose do not coordinate to the gallium or indium metal centres. One would expect more dramatic shifts if the glucose hydroxyl groups coordinated gallium or indium. When D-gluconic acid complexes gallium(III), the C2-C4 hydroxyl groups deprotonate/coordinate to gallium and the resulting changes in 1 3 C chemical shifts for C2-C4 range from -0.3 ppm (upfield shift) to +0.6 ppm (downfield shift) in 9:1 H 2 0:D 2 0. 2 0 Citrate complexation of 81 82 gallium at neutral pH results in 1.1 ppm upfield C shift for the tertiary hydroxyl group that deprotonates/coordinates gallium(III) at neutral pH. 1 8 ' 1 9 While there are little changes in the glucose chemical shifts, all of the pyridinone ring carbon atoms shift ranging from approximately -5 ppm to +8 ppm indicating the pyridinone moiety is complexing gallium and indium. A representative C NMR plot is shown in Figure 2.3. The largest changes occur for the pyridinone C3 atom, where a downfield shift of 6.5-8.3 ppm occurs upon formation of the gallium and indium complexes. Each indium complex has a slightly larger downfield shift for C3 (-0.3-0.5 ppm) compared to the corresponding gallium complex. In contrast, upfield shifts are observed for C4 pyridinone carbonyl) ranging from -0.4-2.3 ppm upon metal complexation. Each gallium complex has a slightly larger shift for C4 (-0.5-0.7 ppm) compared to the corresponding indium complex. The C5 and C6 atoms have upfield shifts from 2.5-4.9 ppm after metal complexation while only the indium complexes show a consistent downfield shift of 1.3-2.3 ppm for the C2 atom. Some notable heteronuclear correlations from HMBC experiments, a number of which aided in characterizing the precursor ligands or metal complexes by NMR, are mentioned here. For HOG6GP, HAG6GP, and their metal complexes, the carbamoyl methyl CH 2 (H8) had 3 J C H couplings to the amide C7 and pyridinone ring C9 and C13 atoms. There were also 3 J C H couplings between C8 and H13, as well as correlations from the amide C7 to H6a and H6b of glucose. The pyridinone C H 3 (C9-CH3) displayed a larger crosspeak for the 2JQH coupling to C9 and a smaller 3 J C H coupling to C10. Other T 3 correlations of interest include the JQH coupling of H13 to C l l and C9, H12's JQH coupling to C10 and smaller crosspeaks for the 2JCH of H12 to C13 and C l l . 83 s • bX) 84 Similar HMBC correlations were found for the pyridinone region for H O G B A P , H A G B A P , HOGBPP, and their metal complexes. All these compounds contain |3-glycoside linkages and the anomeric proton (HI) had a 3 J C H correlation to the C 7 of the phenyl substituent in all cases. Additionally, HOGBPP and its metal complexes had a 3 J C H coupling between the phenyl CIO and HI 5 of the pyridinone ring. The acetyl carbonyl carbon atoms for HAG6GP, H A G B A P , and their metal complexes could be assigned from the 3JQH correlation to the corresponding protons on the glucose ring. The conditions for pyridinone synthesis from the corresponding pyranones often require heating for extended periods of time in slightly acidic or rather basic conditions. Larger amino substituents typically hinder pyridinone formation resulting in very low or no yields. For these reasons the strategy was to form the pyridinone-linker system first using relatively smaller amines, and then couple this linker to the glucose moiety. The linkers include the carboxylic acid substituent as in 3 and a phenolic substituent as in 17. The point of glucose attachment was picked to give some variety in structure that was relatively straightforward to synthesize. So far, attempts to recrystallize the metal complexes or the precursor ligands, have failed to produce X-ray quality crystals. Vapour diffusion of alcohols into water or slow evaporation of water or MeCN were the most common methods employed. Sugars are notoriously difficult to crystallize, however, X-ray quality crystals of a benzylated pyridinone-glucose compound (8) were grown by slow evaporation of a 3:1 AcOEt:MeOH solution. The structure of 8 is shown in Figure 2 . 4 and the unit cell is in Figure 2 . 5 . The bond lengths and angles (Table 2 . 2 ) are typical for pyridinones. The 8 5 torsional angles indicate the near-planarity of the pyridinone ring and the first carbon atom of the N2-substituent (C9) is also close to the plane of the pyridinone ring, both characteristics of other pyridinone structures. The a-anomer of glucose is also displayed in the structure of 8 which concurs with the coupling constant and chemical shift in the 'Fl NMR spectrum (see Experimental Section), the latter data indicating no anomerization has taken place. Intermolecular hydrogen bonds form a tight net in the unit cell. The amide H-atom (H26) is 1.931(19) A from 02 of the glucose ring, the amide O-atom (06) is 1.83(2) A from H23, and the pyridinone carbonyl O-atom (07) is 1.81(3) A from H24. H 2 4 Figure 2.4: ORTEP diagram of 8 (with some H-atoms omitted for clarity) showing 50% thermal probability ellipsoids. 86 One of the metal complexes that contained (3-glucose linkages (Ga(OGBPP)3) was randomly chosen as a substrate for a broad-specificity (3-glucosidase, Abg. After addition of Abg to a buffered solution containing Ga(OGBPP)3 a white precipitate started forming within a few minutes at RT. The mass spectrum of this precipitate is Table 2.2: Selected bond lengths (A) and angles (°) in 8. 0(1) -C(l) 1.404(2) N(l)-C(6) 0(1) -C(7) 1.435(2) N(l)-C(8) 0(2) -C(2) 1.420(2) N(2)-C(9) 0(3) -C(3) 1.423(2) N(2)-C(10) 0(4) -C(4) 1.429(2) N(2)-C(14) 0(5) -C(l) 1.413(2) C(l)-C(2) 0(5) -C(5) 1.434(2) C(2)-C(3) 0(6) -C(8) 1.242(2) C(3)-C(4) 0(7) -C(12) 1.257(2) C(4)-C(5) 0(8) -C(13) 1.376(2) C(5)-C(6) 0(8) -C(16) 1.439(2) C(8)-C(9) C(l C(l C(13) C(6 C(9 C(9 C(10) 0(1 0(1 0(5 0(2 0(2 C(l 0(3 0(3 C(2 0(4 0(4 C(3 0(5 0(5 C(4 N(l 0(6 •0(1)-C(7) 0(5)-C(5) 0(8)-C(16) >-N(l)-C(8) •N(2)-C(10) i-N(2)-C(14) •N(2)-C(14) C(l)-0(5) •C(l)-C(2) •C(l)-C(2) •C(2)-C(l) •C(2)-C(3) -C(2)-C(3) |-C(3)-C(2) i-C(3)-C(4) -C(3)-C(4) •C(4)-C(3) •C(4)-C(5) C(4)-C(5) •C(5)-C(4) C(5)-C(6) i-C(5)-C(6) i-C(6)-C(5) C(8)-N(l) 112.9(1 114.0(1 113.3(1 124.4(1 117.6(1 122.2(1 120.1(1 113.0(1 108.2(1 109.7(1 109.1(1 112.5(1 110.8(1 109.1(1 109.9(1 109.0(1 111.0(1 106.2(1 112.4(1 111.8(1 105.6(1 110.8(1 112.0(1 125.3(2 1.454(2) 1.327(2) 1.463(2) 1.363(2) 1.375(2) 1.513(2) 1.518(2) 1.518(2) 1.519(2) 1.514(2) 1.520(2) 0(6)-C(8)-C(9) N(l)-C(8)-C(9) N(2)-C(9)-C(8) N(2)-C(10)-C(ll) C(10)-C(ll)-C(12) 0(7)-C(12)-C(ll) 0(7)-C(12)-C(13) C(ll)-C(12)-C(13) 0(8)-C(13)-C(12) 0(8)-C(13)-C(14) C(12)-C(13)-C(14) N(2)-C(14)-C(13) N(2)-C(14)-C(15) C(13)-C(14)-C(15) 0(8)-C(16)-C(17) C(16)-C(17)-C(18) C(16)-C(17)-C(22) C(18)-C(17)-C(22) C(17)-C(18)-C(19) C(18)-C(19)-C(20) C(19)-C(20)-C(21) C(20)-C(21)-C(22) C(17)-C(22)-C(21) C(10)-C(ll) C(ll)-C(12) C(12)-C(13) C(13)-C(14) C(14)-C(15) C(16)-C(17) C(17)-C(18) C(17)-C(22) C(18)-C(19) C(19)-C(20) C(20)-C(21) C(21)-C(22) 120.3(1) 114.2(1) 111.2(1) 122.1(2) 121.5(2) 123.1(2) 122.9(2) 114.0(1) 118.6(1) 118.5(2) 122.7(2) 119.1(2) 118.1(1) 122.9(2) 110.1(1) 117.8(2) 122.9(2) 119.3(2) 120.6(2) 120.1(2) 119.9(2) 120.4(2) 119.7(2) 1.349(3) 1.424(3) 1.445(2) 1.369(2) 1.493(2) 1.503(3) 1.398(3) 1.386(3) 1.381(3) 1.378(3) 1.388(3) 1.394(3) 87 Figure 2.5: Unit cell/packing diagram for 8. quite similar to an authentic sample of Ga(hpp)3 indicating that three P-glucose molecules were cleaved producing Ga(hpp)3. If Ga(OGBPP) 3 were to come in contact with a similar specificity P-glucosidase enzyme in vivo, it may give an indication that the P-glucose linkages would be cleaved. 88 O H OH Ga(OGBPP)3 Ga(hpp)3 Hhpp 17 Scheme 2.7: Synthesis of Ga(hpp)3: (a) Abg, sodium phosphate buffer; (b) Ga(N03)3-9H20, DMF, 86%; (c) 10% Pd/C, 1 atm H 2 , MeOH/DMF/H 20, 35%. The water solubilities of Ga(OG6GP) 3 , In(OG6GP) 3, Ga(OGBPP) 3 , In(OGBPP) 3, Ga(AG6GP) 3 , and In(AG6GP) 3 were all greater than 10 mM at RT. Ga(OGBAP) 3 and In(OGBAP) 3 have water solubilities of -2.5-3 mM at RT while Ga(AGBAP) 3 and In(AGBAP) 3 have quite low water solubilities (<0.5 mM). The solubilities of Ga(AGBAP) 3 and In(AGBAP) 3 are improved with mixtures of water and EtOH. 2.4 Conclusions A number of glucose-bearing tris(pyridinonato)gallium and indium complexes were synthesized and characterized by mass spectrometry, H and C NMR spectroscopy. With three equivalents of ligand, neutral complexes formed with the pyridinone moiety complexing the gallium and indium metal centres. There are no indications that the glucose moiety coordinates to or interacts with the gallium or indium metal centres. How the glucose moieties interact in vivo will dictate the biodistribution of the metal complexes and their potential for use in diagnostic nuclear medicine. 89 2.5 References (1) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory Chemicals; Permagon Press: Oxford, 1980. (2) Kempton, J. B.; Withers, S. G. Biochemistry 1992, 31, 9961-9969. (3) Molecular Structure Corporation. d*Trek. Area Detector Software. Version 7.11 (2001): MSC, 3200 Research Forest Drive, The Woodlands, TX 77381, USA. (4) Altomare, A.; Burla, M . C ; Camalli, M.; Cascarano, G. L.; Giacovazzo, C ; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 1999, 32, 115-119. (5) Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W. P.; de Gelder, R.; Israel, R.; Smits, J. M. M. The DIRDIF-94 Program System University of Nijmegen, The Netherlands, 1994. (6) Molecular Structure Corporation. teXsan. Single Crystal Structure Analysis Software. Version 1.8 (1992-1997): MSC, 3200 Research Forest Drive, The Woodlands, TX 77381, USA. (7) Harris, R. L. N . Aust. J. Chem. 1976, 29, 1329-1334. (8) Cramer, F. D. In Methods In Carbohydrate Chemistry; Whistler, R. L., Wolfrom, M. L., Eds.; Academic Press: New York, 1962; Vol. I, pp 244-245. (9) Cramer, F. D. In Methods In Carbohydrate Chemistry; Whistler, R. L., Wolfrom, M. L., Eds.; Academic Press: New York, 1963; Vol. II, pp 242-246. (10) Matsuda, K.; Tsuchiya, T.; Torii, T.; Umezawa, S. Bull Chem. Soc. Jpn. 1986, 59, 1397-1401. (11) Rosowsky, A.; Yu, C.-S.; Uren, J.; Lazarus, H.; Wick, M. J. Med. Chem. 1981, 24, 559-567. (12) Conchie, J.; Levvy, G. A. In Methods In Carbohydrate Chemistry; Whistler, R. L., Wolfrom, M. L., Eds.; Academic Press: New York, 1963; Vol. II, pp 335-337. (13) Thompson, A.; Wolfrom, M . L.; Pacou, E. In Methods In Carbohydrate Chemistry; Whistler, R. L., Wolfrom, M . L., Eds.; Academic Press: New York, 1963; Vol. II, pp 215-220. (14) Tsunoda, T.; Yamamiya, Y.; Ito, S. Tetrahedron Lett. 1993, 34, 1639-1642. 90 (15) Zhang, Z.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1991, 30, 509-515. (16) Nelson, W. O.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1989, 28, 3153-3157. (17) Matsuba, C. A.; Nelson, W. O.; Rettig, S. J.; Orvig, C. Inorg Chem. 1988, 27, 3935-3939. (18) Chang, C. H. F.; Pitner, T. P.; Lenkinski, R. E.; Glickson, J. D. Inorg. Chem. 1977, 99, 5858-5863. (19) Hawkes, G. E.; O'Brien, P.; Salacinski, H.; Motevalli, M. ; Abrahams, I. Eur. J. Inorg Chem. 2001, 1005-1011. (20) Escandar, G. M. ; Olivieri, A. C ; Gonzalez-Sierra, M. ; Frutos, A. A.; Sala, L. F. J. Chem. Soc.-Dalton Trans. 1995, 799-804. 91 C H A P T E R 3 3-HYDROXY-4-PYRIDINONES AS P O T E N T I A L C H E L A T I N G A G E N T S IN A L Z H E I M E R ' S DISEASE 3.1 Introduction In an effort to reduce the harmful effects of excess metal ions and oxidative stress in Alzheimer's disease (AD), therapies involving metal ion chelation and/or removal are being investigated.1"4 One strategy might involve protecting the metal chelation portion of the molecule until it is at the site required for metal chelation (prodrug strategy). With AD, there is reported to be a systemic elevation in copper5 and iron6 levels; however, P-amyloid (A(3) plaques present in AD brain have higher levels of redox active iron and copper,7 making this the main target organ for AD chelation therapy. If a lipophilic chelator is used, some of the apo-chelator may diffuse across the blood-brain barrier (BBB) but the chelator could also systemically sequester metal ions, form neutrally charged complexes, and these metal chelates could then enter across the BBB. This would increase metal content in the brain and does occur for a number of lipophilic metal chelators.8 This also happens with certain metal complexes of clioquinol (Hcq) (Figure 3.1); oral administration of Zn(cq)2, Ni(cq)2, and Hg(cq)2 increased the levels of these metals in mice brain by a factor of -2.6, 1.6, and 1.4, respectively, after 24 h.9 To prevent premature systemic metal chelation prior to entry across the BBB, a prodrug that is deprotected (i.e., unmasked) after crossing the BBB might prevent redistribution of systemic metals into the brain. 92 Protection of the 3-hydroxy-4-pyridinones at the 3-hydroxy position prevents metal chelation, making this the simplest route for prodrug strategies. Ester prodrugs of pyridinones are a viable option because such compounds are lipophilic and should cross cell membranes relatively easily; however, the stability of ester protected pyridinones in plasma vary, and it can be difficult to find appreciable water-solubility for these derivatives.10 An alternative would be to utilize glucose as a prodrug as this would protect the pyridinone and impart water-solubility on the resulting compounds. A high density of GLUT1 transporters are present at the B B B 1 1 and this may allow glucose-conjugates to utilize GLUT1 and gain access to the brain. In order for the glucose prodrug to be cleaved, one strategy would be to synthesize pyridinone glycoside derivatives. The glycoside link has the potential to be cleaved in vivo by a number of P-12 glucosidases, some of which are found naturally in the brain. If the pyridinone glycosides were able to penetrate the BBB and if the prodrug were cleaved to produce the free pyridinonate anions, lipophilic metal complexes could form in the brain and might passively diffuse across the BBB, and be redistributed or (hopefully) excreted. The 3-hydroxy-4-pyridinones were chosen as the chelators because they have proven iron-removing abilities in humans.13 For instance, Hdpp (Figure 3.1), also known as Deferiprone or Ferriprox (and sometimes referred to as LI), has been approved for use in India since 1994 and conditionally approved in Europe since 1999 to excrete excess iron in thalassemia patients.13 Iron is sequestered by Hdpp as its iron complex Fe(dpp)3 which is thermodynamically stable.14'15 3-Hydroxy-4-pyridinones such as Hdpp also have a relatively high affinity for Cu(II) ions15"17 but less so for Zn(II),15 and very 93 (+)-a-Tocopherol A B T S , diammonium salt Figure 3.1: Structures of some products and reagents in Chapter 3. 94 little affinity for common biological electrolyte ions such as sodium, potassium, calcium or magnesium.18 Specific, rather than systemic, chelation of excess iron or copper in the brain of AD patients could remove some of these metals from the brain and obviate the reactive oxygen species (ROS) production associated with them. Metal chelators with additional antioxidant properties could also help to reduce the oxidative stress that accompanies AD. Some recent reports show that the presence of pyridinones or the chelation of iron with pyridinones are able to reduce oxidative stress.19"21 The pyridinones might bind copper and iron associated with AP plaques, help to break apart and dissolve the plaques already present, prevent further plaque deposition, and reduce oxidative stress in AD. The goal of this project was to synthesize and characterize pyridinone glycosides that might be suitable for removing excess Cu and Fe from the brain. A secondary goal was to assess the ability of the pyridinone glycosides to cleave to form the pyridinones under relevant conditions and to obtain data on the pyridinones' ability to reduce markers of oxidative stress. 3.2 Experimental 3.2.1 Materials and Instrumentation Most information related to this section is contained in Section 2.2. Additionally, X-ray crystallographic data for 27p (Figure 3.1) were solved using direct methods and the collected and.processed data.were handled as previously described (Section 2.2) by Dr. B.O. Patrick. The first crystal used for X-ray studies of 27p underwent a phase change 95 somewhere between -20 and -30°C and shattered the crystal. Thus the data were collected at a temperature of -20 ± 1°C. The (3:a ratios for glucose derivatives of pyridinones in solution were determined by integration of the doublets for the anomeric protons in the 'H NMR spectra. Baseline corrections were done prior to integration and a larger number of scans were acquired when there were smaller proportions of the a-anomer. The amines (20-22) used for the synthesis of pyridinones and 2,3,4,6-tetra-O-benzyl-D-glucopyranose (16) (Figure 3.1), as well as reagents for Trolox Equivalent Antioxidant Capacity (TEAC) assays (vide infra) were obtained from Aldrich or Sigma. There was always a higher a-anomer content for 16, as indicated by its 'H NMR spectrum. The 4-(2-butyl)aniline starting material (22) (Figure 3.1) was provided from Aldrich as a racemic mixture. 3.2.2 Enzyme Kinetics A Hewlett-Packard (HP) model 8453 diode array spectrophotometer, equipped with kinetics software package, was used to monitor enzyme reactions with Agrobacterium sp. P-glucosidase (Abg). The spectrophotometer cuvette holder was connected to a Fisher Isotemp 1050 circulating water-bath so that the temperature in the cuvette was 37.0 ± 0.2°C after equilibration for 15 minutes. Enzyme reactions took place in quartz cuvettes (semi-micro cell, 1.4 mL total volume, 10 mm pathlength). The background was taken for each empty cuvette. The cuvette was then filled (using appropriate pipetters or syringes) with 600 uL of 50 mM sodium phosphate buffer (pH 96 6.8), 80 uL of a 50 mM sodium phosphate buffered solution containing 1% bovine serum albumin, and 100 uL of substrate (27p) in water. Paraffin was used to cover the cuvette to prevent evaporation. The cuvette was equilibrated for 15 minutes in the cuvette holder before 20 uL of Abg was syringed through the paraffin to initiate the reaction (final [Abg] ~ 89 nM). The cuvette was quickly inverted and replaced in the cuvette holder. From 27p, the release of 3-hydroxy-l,2-dimethyl-4(l/f)-pyridinone (Hdpp) (Figure 3.1) was followed at 290 nm (kmax = 285 nm). The initial rates were typically followed up to 2 minutes after addition of Abg with data points being measured every 0.5 seconds. For each concentration of substrate (27p), the initial rates were determined in triplicate. The slopes for initial rates were calculated using on-board HP software. 22 Mathematical analyses for A:oat or Km values were done using GraFit software. The substrate used (27P) was column purified, recrystallized (see Section 3.2.5), and characterized by elemental analysis and *H NMR spectroscopy. The corresponding pyridinone (Hdpp) was received as a white semi-crystalline solid and was used without further purification (Aldrich, 99%). A recently calibrated four decimal place balance with ± 0.2 mg error was used to weigh at least 20 mg for both 27p and Hdpp, reducing the error from this measurement to less than 1%. Solutions of these compounds were appropriately diluted in volumetric flasks and the initial concentrations for 27p ranged from 0-3.0 mM in the cuvettes. Preliminary reactivities of substrates ( 2 7 P - 3 0 P ) with Abg were tested and monitored by TLC (see below). The reactions took place in 1 mL Eppendorf tubes at RT with an Abg concentration of 3.9 nM. Initial substrate concentrations were 7.5 mM for all pyridinone glycosides except 27p (15 mM). TLCs were spotted between 20-30 97 minutes after Abg initiated the reactions and were run at -30 minutes with 7:3 EtOAc:MeOH, except for 27p (1:1 EtOAc:MeOH). A UV lamp, FeCl 3 spray, and charring of TLC plates were used in the characterization of compounds (FeC^ has been 23 used for nearly a century to aid in characterizing pyridinones ). 3.2.3 Trolox Equivalent Antioxidant Capacity (TEAC) Antioxidant Assay The 3-hydroxy-4-pyridinones were tested using the TEAC antioxidant assay as a measure of their overall ability to scavenge free radicals compared to antioxidant standards such as Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid), a-tocopherol (a-Toc) and BHT (butylhydroxytoluene) (Figure 3.1). An improved ABTS*+ (2,2'-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) diammoniuim salt) (Figure 3.1) radical cation decolorization assay24 was used to determine relative TEAC values (Table 3.3, Figure 3.7). ABTS was dissolved in water (7 mM), was subsequently reacted with aqueous potassium persulfate (2.45 mM), and placed in the dark for 16 h before use. Thus ABTS oxidizes to the ABTS*+ radical cation. The ABTS*+ product solution, after equilibrating to 30°C (Fisher Isotemp circulating water bath), was diluted with EtOH to an absorbance of 0.70 (± 0.02) at 734 nm. Stock solutions of the pyridinones in EtOH were diluted so that addition of 20 uL to 2 mL of ABTS*+ solution caused a reduction of 20-80% in the absorbance as a result of the reduction to ABTS. To obtain this range, final concentrations for the pyridinones ranged from 2.5-15 uM. After the solutions were initially mixed, the A 7 3 4 readings were taken at 30°C after 1, 3, and 6 minutes. These 98 readings were done in triplicate. The percentage inhibition of absorbance at 734 nm was calculated and plotted as a function of pyridinone concentration. The slopes were then compared to the standard Trolox, with its TEAC value normalized to 1 (Table 3.3, Figure 3.7). 3.2.4 Cells Studies and MTT Assay Cells were received that had previously been cultured and prepared with the experimental details described elsewhere.25 For this assay, human breast cancer cells (MDA-MB-43 5 S) 2 6 were transferred to a 96-well plate by addition of 100 uL of the cell solution (1 x i o 4 cells) to 54 of the wells (9x6). Another 6 wells had growth medium added (100 uL) as blanks and all perimeter wells had deionized water added (200 uL). The plate was then incubated at 37°C for 24 h. Phosphate buffered saline (PBS) solutions (100 uL) of 300 (Figure 3.1) at 8 different concentrations (10-5000 uM) were then added to 48 wells (8x6) containing cells. The remaining 6 wells containing cells had' PBS added (100 uL) and served as the control. PBS (100 uL) was also added to the 6 blanks and the plate was incubated at 37 °C for 3 days. 97 A modified procedure of Mosmann was used for the MTT assay (MTT = (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (Figure 3.1). A PBS solution of MTT (50 uL, 2.5 mg/mL) was added to each well of the plate which was then incubated for 3 h, by which time a purple precipitate of formazan formed at the bottom of certain wells. The contents of each well were carefully aspirated off to leave the 99 formazan that was then dissolved in DMSO (150 uL); the plate was shaken and analyzed by a plate reader (Spectra Max 190 from Molecular Devices) to determine the absorbance of each well at 570 nm. The percentage cell viability was calculated by dividing the average absorbance of the cells treated with 30p by that of the control; % cell viability versus drug concentration (logarithmic scale) was plotted to determine the IC50 (drug concentration at which 50% of the cells are viable relative to the control), with its estimated error derived from the average of 3 trials (Figure 3.8). 3.2.5 Synthesis l-Hexyl-3-hydroxy-2-methyl-4(l//)-pyridinone (Hmhpp). This was prepared by a previously published procedure.28 After sublimation, TLC (AcOEt) of the pale yellow solid followed by ninhydrin treatment revealed a small amount of hexylamine (20) and the compound was recrystallized from a minimum of hot AcOEt. The solution containing recrystallized product was filtered using a course frit, the solid collected was rinsed with cold AcOEt and Hmhpp dried in vacuo for 24 h (51% yield based on 1). Anal. Calcd (found) for C 1 2 H 1 9 N O 2 : C, 68.87 (68.79); H, 9.15 (9.25); N, 6.69 (6.63). ! H NMR (300 MHz, CD3OD): 5 = 0.90 (m, 3H, hexyl CH3), 1.32 (m, 6H, hexyl CH2), 1.72 (m, 2H, hexyl CH2), 2.421 (s, 3H, pyrid. CH3), 4.027 (t, 2H, J= 7.6 Hz, N-C#2), 6.378 (d, 1H, pyrid. H5,J= 7.1 Hz), 7.46 (d, 1H, pyrid. H6,J= 7.1 Hz). MS (+LSIMS): m/z 732 [M+H]+. 100 3-Hydroxy-2-methyl-l-phenyI-4(l//)-pyridinone (Hppp). This was prepared by a previously published procedure.29 The compound recrystallized as large white crystalline chunks (45% yield based on 1). The 'H NMR spectrum is similar to that reported.29 1 3C{'H} NMR (75 MHz, CDC13): 8 = 13.60, 110.91, 113.11, 126.74, 129.52, 129.86, 137.34, 141.71, 145.64, 170.07. Hppp l-[4-(2-Butyl)phenyl]-3-hydroxy-2-methyl-4(l/f)-pyridinone (Hsbp). This pyridinone 29 was prepared similar to previously published procedures. To a 500 mL 5 l r ^ [ 3 ° H r o u n d b o t t o m flask (RBF) w a s a d d e d m a l t o 1 C1) ( 6 - 3 0 6 §> 5 0 - ° mmol), (±)-4-(2-butyl)aniline (22) (14.924 g, 100 mmol), a dilute HC1 solution (3 mL cone. HC1 and 100 mL water), and 40 mL MeOH. The mixture was refluxed under Ar for 72 h. The solvents were removed on the rotary Hsbp evaporator and the product was recrystallized from a minimum amount of refluxing MeOH. The solution containing recrystallized product was filtered using a course frit, the collected solid was washed with cold MeOH, and dried in vacuo for 24 h producing 2.942 g (23% yield based on 1) of an off-white solid that was presumably a mixture of enantiomers. Anal. Calcd (found) for C22H19NO2: C, 71.63 (71.41); H, 5.51 (5.44); N, 6.96 (6.91). ! H NMR (300 MHz, CD3OD): 5 = 0.848 (t, 3H, 2-butyl CH3, J = 7.1 Hz), 1.280 (d, 3H, 2-butyl CH3, J= 7.1 Hz), 1.656 (m, 2H, 2-butyl CH2), 2.105 (s, 3H, pyrid. CH3), 2.723 (m, IH, 2-butyl CH), 6.457 (d, IH, J= 7.3 Hz, pyrid. H5), 7.35 (m, 4H, arom. C6H5), 7.577 (d, IH, J= 7.3 Hz, pyrid. H6). MS (+LSIMS): m/z 258 [M+H]+. 101 3-(2',3',4',6'-Tetra-(?-benzyl-D-glucopyranosyloxy)-l,2-dimethyl-4(l#)-pyridinone (23). This Mitsunobu reaction30 was carried out in a manner similar to that of 18 (see Experimental Section, Chapter 2) with the following azodicarboxylic acid dipiperidide (ADDP) (3.785 g, 15 mmol) some material remained undissolved but eventually went into solution after ~1 h. The flask contents were stirred for 12 h at RT (as was the case for all analogues 23-26), the CH2CI2 volume was reduced to -30 mL using a rotary evaporator and crystallization occurred upon cooling (ADDP byproduct). The byproduct was removed on a course frit, and rinsed with 2x10 mL cold CH2CI2. After CH 2C1 2 removal using a rotary evaporator, a second crop of ADDP byproduct was recrystallized from a minimum of hot AcOEt, followed by filtering and rinsing with 2x10 mL cold AcOEt. Evaporation using a rotary evaporator produced an orange-brown oil that was taken up in a minimum amount of AcOEt and applied to a silica column (5x45 cm) run with AcOEt. The product was collected, rotoevaporated, and dried on the vacuum line for 48 h producing a viscous yellow oil (3.083 g, 47% yield based on 16). The 'H NMR spectrum indicated mainly P-anomer with the a-anomer as a minor product, similar to what was seen for all these analogues (23-26). A small amount of the oil was spread thinly in a vial and dried an additional 24 h on the vacuum line before elemental analysis. Anal. Calcd (found) for C 4 1 H 4 3 N O 7 : C, 74.41 (74.23); H, 6.55 (6.60); N, 2.12 (2.13). 'H NMR (300 MHz, CDC13): 5 = 2.293 (s, 3H, pyrid. CH3), 3.44 OBn modifications: dry CH2CI2 (-220 mL) was added to 2,3,4,6-tetra-O-benzyl-D-glucopyranose (16) (5.407 g, 10.0 mmol) and Hdpp (1.670 g, 12.0 mmol). After addition of tributylphosphine (3.74 mL, 15.0 mmol) and 102 (m, 1H), 3.443 (s, 3H, N-C// 3), 3.71 (m, 5H), 4.465 (m, 2H, Bn CH2), 4.570 (d, 1H, Bn CH2,J= 11.2 Hz), 4.79 (m, 3H, Bn CH2), 4.995 (d, 1H, Bn CH2,J= 10.9 Hz), 5.245 (d, 1H, Bn CH2,J= 11.2 Hz), 5.485 (d, 1H, anomeric H,J= 7.3 Hz), 6.419 (d, 1H, pyrid. H, J= 7.3 Hz), 7.25 (m, 19H, Bn C6H5), 7.46 (m, 2H). MS (+LSIMS): m/z 662 [M+H]+. 3-(2',3',4',6'-Tetra-0-benzyl-D-glucopyranosyloxy)-l-hexyl-2-methyl-4(l#)-pyridinone (24). This procedure was similar to that used for the synthesis of 23 with the following substitutions: dry CH2CI2 (-80 mL) was added B n O - ^ ^ Q to 16 (2.703 g, 5.0 mmol) and Hmhpp (1.256 g, 6.0 B n O ^ ^ O mmol) to dissolve these materials. Tributylphosphine (2.49 mL, 10.0 mmol) and ADDP (2.523 g, 10.0 mmol) were added to the reaction mixture. Column purification produced 1.946 g (53 % yield based on 16) of a viscous yellow oil. Anal. Calcd (found) for C46H53NO7: C, 75.49 (75.09); H, 7.30 (7.35); N , 1.91 (2.10). *H NMR (500 MHz, CDC13): § = 0.88 (m, 3H, hexyl CH3), 1.27 (m, 6H, hexyl CH2), 1.61 (m, 2H, hexyl CH2), 2.330 (s, 3H, pyrid. CH3\ 3.413 (m, 1H), 3.70 (m, 7H), 4.454 (m, 2H, Bn CH2), 4.561 (d, 1H, Bn CH2,J = 11.0 Hz), 4.755 (m, 2H, Bn CH2), 4.820 (d, 1H, Bn CH2,J= 11.0 Hz), 4.947 (d, 1H, Bn CH2 J= 11.0 Hz,), 5.252 (d, 1H, Bn CH2, J= 11.2 Hz), 5.497 (d, 1H, anomeric H, J= 7.6 Hz), 6.409 (d, 1H, pyrid. H, J = 7.6 Hz), 7.24 (m, 19H, Bn C6H5), 7.46 (m, 2H). MS (+LSIMS): m/z 732 [M+H]+. 103 3-(2',3',4',6'-Tetra-0-benzyl-D-glucopyranosyloxy)-2-methyl-l-phenyl-4(l/r)-pyridinone (25). This procedure was similar to that used for the synthesis of 23 with the \ ^ produced 2.028 g (56 % yield based on 16) of a viscous yellow oil. Anal. Calcd (found) for C46H 4 5 N0 7 : C, 76.33 (76.18); H, 6.27 (6.41); N , 1.93 (2.02). 'H NMR (500 MHz, CDC13): 5 = 2.068 (s, 3H, pyrid. C/73), 3.471 (m, IH), 3.66 (m, 4H), 3.786 (dd, IH, J= 9.0 Hz, J= 9.0 Hz), 4.460 (s, 2H, Bn CH2), 4.561 (d, IH, Bn CH2, J= 11.0 Hz), 4.80 (m, 3H, Bn CH2), 4.963 (d, IH, Bn CH2, 7= 11.0 Hz), 5.278 (d, IH, J= 11.1 Hz, Bn CH2), 5.563 (d, IH, J= 7.6 Hz, anomeric H), 6.475 (d, IH, J= 7.6 Hz, pyrid. H), 7.21 (m, 24H, arom. C6H5), 7.47 (m, 2H). MS (+LSIMS): m/z 724 [M+H]+. 3-(2',3,,4',6'-Tetra-0-benzyl-b-glucopyranosyloxy)-l-[4-(2-butyl)phenyl)]-2-methyl-4(l/f)-pyridinone (26). This procedure was similar to that used for the synthesis of 23 with the following substitutions: Dry CH2CI2 (-100 mL) following substitutions: dry CH2CI2 (-120 mL) was OBn added to 16 (4.325 g, 8.0 mmol) and Hppp (1.932 g, 9.6 mmol) to dissolve these materials. Tributylphosphine (3.0 mL, 12.0 mmol) and ADDP (3.028 g, 12.0 mmol) were added to the reaction mixture. Column purification was added to 16 (4.866 g, 9.0 mmol) and Hsbp (2.779 g, produced 4.248 g (61 % yield based on 16) of a viscous were added to the reaction mixture. Column purification (3.36 mL, 13.5 mmol) and ADDP (3.406 g, 13.5 mmol) 10.8 mmol) to dissolve these materials. Tributylphosphine 104 yellow oil. A presumed mixture of the 2-butyl stereogenic carbon atom would result in even a pure P-anomer being a diastereomeric mixture. Anal. Calcd (found) for C50H53NO7: C, 77.00 (76.83); H, 6.85 (6.82); N , 1.80 (2.00). 'H NMR (500 MHz, CDCI3): 8 = 0.842 (t, 3H, 2-butyl CH3, J = 7.3 Hz), 1.262 (d, 3H, 2-butyl CH3,J=6.0 Hz), 1.617 (m, 2H, 2-butyl CH2), 2.068 (s, 3H, pyrid. CH3), 2.665 (m, IH, 2-butyl CH), 3.446 (m, IH), 3.65 (m, 4H), 3.781 (dd, IH, J= 9.0 Hz, J= 9.0 Hz), 4.459 (s, 2H, Bn CH2), 4.557 (d, IH, Bn CH2, J = 11.0 Hz), 4.79 (m, 3H, Bn CH2), 4.958 (d, IH, Bn CH2, J= 11.0 Hz), 5.277 (d, 1H,J= 11.2 Hz, Bn CH2), 5.553 (d, 1H, .7=7.6 Hz, anomeric/f), 6.409 (d, IH, J = 7.6 Hz, pyrid. H), 7.25 (m, 23H, arom. C6H5), 1 Al (m, 2H). MS (+LSIMS): m/z 780 [M+H]+. 3-(P-D-Glucopyranosyloxy)-l ,2-dimethyI-4(l//)-pyridinone (27P). This hydrogenolysis reaction was carried out in a similar fashion as for HOG6GP (see Experimental Section, jj Chapter 2) with the following changes: 23 (1.078 g, 1.63 2 7 P I mmol) was dissolved in MeOH (50 mL). The 10% Pd/C (694 mg, 0.65 mmol Pd) was wetted with water (10 mL) and the vial was rinsed with MeOH (10 mL). For all analogues (27-30), the flask contents were stirred vigorously at RT for 24 hours, the reaction mixture was filtered, the filtrate was evaporated using a rotary evaporator (45°C), and the residue was dried in a vacuum desiccator for 24 h to give an off-white product (468 mg for the reaction above). TLC (1:1 AcOEtMeOH) indicated no starting material and FeCl3 spray indicated no deprotected a-hydroxyketone present. This solid was recrystallized from a minimum of hot MeOH (~20 mL). After 105 filtering, drying in a vacuum desiccator for 48 h, 335 mg (68% yield) of almost X-ray quality crystals were isolated and characterized. A small amount of this material was used to saturate a solution of MeOH; X-ray quality crystals were grown by slow evaporation (at RT) from a slightly opened vial. Characterization of 2 7 P : Anal. Calcd (found) for C i 3 H 1 9 N 0 7 : C, 51.82 (51.76); H, 6.36 (6.06); N , 4.65 (4.66). *H N M R (500 MHz, CD 3 OD): 5 = 2.535 (s, 3H, C7-C// 3 ), 3.232 (ddd, 1H, HS, 3J5A = 9.5 Hz, 3 J 5 j 6 a = 2.3 Hz, 3J 5 ) 6b = 5.5 Hz), 3.333 (dd, 1H, 774, 3J 4,5 = 9.5 Hz, 3 J 4 , 3 = 9.5 Hz), 3.39 (dd, partially overlapped with H2, 7/3), 3.42 (dd, partially overlapped with H3, HI), 3.674 (dd, 1H, //6b, V 6 b , 6 a = 11.9 Hz, 3 J 6 b j 5 = 5.5 Hz), 3.753 (s, 3H, N-C//3), 3.837 (dd, 1H, //6a, 2J 6 a ;6b =11.9 Hz, 3J 6 a,5 = 2.3 Hz), 4.621 (d, 1H, 7/1, 3 J l j 2 = 7.4 Hz), 6.442 (d, 1H, 7710, 3Jio,n = 7.4 Hz), 7.723 (d, 1H, HU, 3 J 1 U 0 = 7.4 Hz). 1 3 C{'H} N M R (125 MHz, CD 3 OD): 5 = 14.04 (C7-CH 3), 42.60 (N-CH 3), 62.64 (C6), 71.07 (C4), 75.50 (Ci), 78.49 (C2), 78.70 (CS), 107.19 (Cl) , 116.78 (C10), 142.63 ( C l l ) , 146.12 (CS), 147.40 (C7), 174.06 (C9). MS (+LSIMS): m/z 324 [M+Na]+, 302 [M+H]+. Partial characterization of 27a: ' H N M R (400 MHz, CD 3 OD): 5 = 2.522 (s, partially overlapped with p-anomer, C7-C// 3 ), 3.483 (dd, 1H, HI, V 2 , i = 3.7 Hz, 3 J 2 > 3 = 9.8 Hz), 4.003 (ddd, 1H, 775, 3 J 5 > 4 = 9.9 Hz, 3 J 5 , 6 a = 2.2 Hz, 3J5,6b = 5.9 Hz), 5.148 (d, 1H, 7 /1 , 3 J U 2 = 3.7 Hz). 3-(D-Glucopyranosyloxy)-l-hexyl-2-methyl-4(l//)-pyridinone (28), (97:3 p:a). This hydrogenolysis reaction was carried out in a similar fashion as for 27p with the following changes: 24 (1.934 g, 2.64 mmol) was dissolved in MeOH (60 mL) and 10% Pd/C (1.124 g, 1.06 mmol Pd) was used. The off-white solid obtained from this reaction (873 mg) was recrystallized from EtOH to give 635 mg (65% yield) of a white solid. Anal. 106 O H Calcd (found) for CgH^NOy: C, 58.21 (58.21); H, 7.87 H H O ^ A y O JL (7-96); N' 3>77 (3>87)- MS (+LSIMS): m/z 394 H° ^ L ^ H [M+Na]+, 372 [M+H]+. Characterization of 28p: J H 2 8 P ) NMR (400 MHz, CD3OD): S = 0.91 (m, 3H, hexyl C#3), ) 1.35 (m, 6H, hexyl CH2), 1.758 (m, 2H, hexyl CH2), / 2.565 (s, 3H, C7-C#3), 3.244 (ddd, 1H, #5, 3J5A = 9.5 Hz, V 5 ,6a = 2.2 Hz, 3J5,6b = 5.6 Hz), 3.333 (dd, 1H, #4, 3J4,5 = 9.5 Hz, 3 J 4 , 3 = 9.5 Hz), 3.39 (dd, partially overlapped with H2, H3), 3.42 (dd, partially overlapped with H3, HI), 3.673 (dd, 1H, H6b, 2 J 6 b , 6 a =11-9 Hz, V 6 b , 5 = 5.6 Hz), 3.842 (dd, 1H, H6a, 2J6 a,6b =11-9 Hz, 3 J 6 a , 5 = 2.2 Hz), 4.042 (t, 2H, N-C# 2, 3J = 7.7 Hz), 4.630 (d, 1H, H\, 3Jl2 = 7.5 Hz), 6.477 (d, 1H, #10,3Jio,n = 7.4 Hz), 7.759 (d, 1H, #11,3Jn,io = 7.4 Hz). 1 3C{'H} NMR (100 MHz, CD3OD): 5 = 13.61 (C7-CH3), 14.27 (hexyl CH 3), 23.55 (hexyl CH 2), 27.06 (hexyl CH2), 31.53 (hexyl CH 2), 32.44 (hexyl CH2), 55.56 (N-CH2), 62.67 (C6), 71.10 (C4), 75.54 (C3), 78.50 (C2), 78.67 (C5), 107.19 (Cl), 117.05 (C10), 141.95 (Cll) , 146.35 (C7), 146.48 (C8), 173.96 (C9). Partial characterization of 28a: [ H NMR (400 MHz, CD3OD): 5 = 2.553 (s, partially overlapped with p-anomer, C7-C#3), 3.491 (dd, 1H, #2,3J2,i = 3.7 Hz, 3 J 2 > 3 = 9.7 Hz), 5.150 (d, 1H, #1, 3Ji>2 = 3.7 Hz). 3-(D-Glucopyranosyloxy)-2-methyl-l-phenyl-4(l/T)-pyridinone (29), (96:4 P:a). This O H hydrogenolysis reaction was carried out in a similar fashion as for 27p with the following changes: 25 (1.998 10 11 g, 2.76 mmol) and 10% Pd/C (1.175 g, 1.10 mmol Pd) were used. The off-white solid obtained from this 107 reaction (930 mg) was recrystallized from MeOH to give 704 mg (70% yield) of a white solid. Anal. Calcd (found) for C i 8 H 2 iN0 7 : C, 59.50 (59.71); H, 5.82 (6.18); N , 3.85 (3.73). MS (+LSIMS): m/z 386 [M+Na]+, 364 [M+H]+. Characterization of 29p: *H NMR (400 MHz, CD3OD): 5 = 2.252 (s, 3H, C7-C/73), 3.270 (ddd, IH, 775, V 5 > 4 = 9.5 Hz, 3 J 5 , 6 a = 2.2 Hz, 3 J 5 ,6b = 5.5 Hz), 3.332 (dd, IH, 774,374j5 = 9.6 Hz, 3 7 4 j 3 = 9.6 Hz), 3.42 (dd, partially overlapped with H2, 773), 3.45 (dd, partially overlapped with H3, 772), 3.657 (dd, IH, H6b, 2J6bM = 11.9 Hz, 3 J 6 b > 5 = 5.5 Hz), 3.839 (dd, IH, H6a, 2J6a,6b = 11.9 Hz, 3 J6a ,5 = 2.2 Hz), 4.739 (d, IH, HI, 3 7 u = 7.6 Hz), 6.562 (d, IH, 7710, 3Jm i = 7.4 Hz), 7.43 (m, 2H, arom. C 6H 5), 7.59 (m, 3H, arom. C 6H 5), 7.740 (d, IH, 7711,3Ju,\o = 7.4 Hz). 1 3C{ !H} NMR (100 MHz, CD3OD): 5 = 15.68 (C7-CH3), 62.67 (C6), 71.13 (CA), 75.58 (C3), 78.50 (CT), 78.72 (CS), 107.05 (Cl), 116.85 (C10), 127.97 (C14), 131.11 (C15), 131.18 (C13), 142.22 (Cll) , 143.06 (C12), 145.90 (Ct), 146.69 (C7), 174.89 (C9). Partial characterization of 29a: *H NMR (400 MHz, CD3OD): 5 = 2.245 (s, partially overlapped with p-anomer, C7-C773), 3.522 (dd, IH, 772, 37 2 >i = 3.7 Hz, 3 7 2 ; 3 = 9.9 Hz), 3.994 (ddd, IH, 775,3J5>4 = 9.7 Hz, 3 J 5 ; 6 a = 2.2 Hz, 3 J 5 ,6b = 5.7 Hz), 5.273 (d, IH, 771, \ 2 = 3.7 Hz), 7.750 (d, overlapped with p-anomer, 7711, 3Ju,io = 7.3. Hz); 1 3 C NMR (100 MHz, CD3OD): 18.36 (C7-C/73), 104.9 (a-anomeric carbon, HMQC correlation-signal not seen above baseline in 1 3 C NMR spectrum). l-[4-(2-Butyl)phenyl)]-3-(P-D-glucopyranosyloxy)-2-methyl-4(l/T)-pyridinone (30P), mixture diastereomers resulting from a mixture of the 2-butyl stereogenic carbon atom. This hydrogenolysis reaction was carried out in a similar fashion as for 27p with the following changes: 26 (2.535 g, 3.25 mmol) was dissolved with MeOH (90 mL) and 108 OH 10% Pd/C (1.383 g, 1.30 mmol Pd) was used. The off-HO- white solid obtained from this reaction (1.335 g) was 10 n recrystallized from MeOH to give 991 mg (73% yield) of a white solid. Characterization of 30p: Anal. Calcd (found) for C22H29NO7: C, 62.99 (62.95); H, 6.97 (6.89); N, 3.34 (3.23). MS (+LSIMS): m/z 442 [M+Na]+, 420 [M+H]+. 'H NMR (400 MHz, CD3OD): 5 = 0.849 (t, 3H, 2-butyl C#3, J = 7.4 Hz), I. 281 (d, 3H, 2-butyl CH3, J= 6.9 Hz), 1.658 (m, 2H, 2-butyl CH2), 2.253 (s, 3H, C7-CH3), 2.730 (m, 1H, 2-butyl CH), 3.271 (ddd, 1H, H5,3J5A = 9.5 Hz, 3 J 5 j 6 a = 2.2 Hz, 3J5>6b = 5.5 Hz), 3.333 (dd, 1H, HA, %5 = 9.6 Hz, 3J 4 ;3 = 9.6 Hz), 3.41 (dd, partially overlapped with H2, H3), 3.44 (dd, partially overlapped with H3, HI), 3.658 (dd, 1H, H6b, 2J6hM = II. 9 Hz, 3 J 6 b j 5 = 5.5 Hz), 3.840 (dd, 1H, H6a, V 6 a,6b = 11.9 Hz, 3J 6 a,5 = 2.2 Hz), 4.731 (d, 1H, #1, 3 J , , 2 = 7.5 Hz), 6.552 (d, 1H, #10,3Ji0,n = 7.4 Hz), 7.338 (d, 2H, 7/14, 3 J 1 4 j l 3 = 7.6 Hz), 7.420 (d, 2H, H13, 3 J i 3 , i 4 = 7.6 Hz), 7.731 (d, 1H, #11, 3 J i U o = 7.4 Hz). 13C{'H} NMR (100 MHz, CD3OD): 5 = 12.48 (2-butyl CH 2CH 3), 15.68 (C7-CH3), 22.17 (2-butyl CHCH3), 32.08 (2-butyl CH2), 42.72 (2-butyl CH), 62.65 (C6), 71.12 (C4), 75.57 (C3), 78.48 (C2), 78.70 (C5), 107.07 (Cl), 116.80 (C10), 127.74 (C14), 129.72 (C13), 140.86, (C12), 142.35 (Cll), 145.88 (CS), 146.85 (Cl), 151.20 (C15) 174.81 (C9). Partial characterization of 30a: 'H NMR (400 MHz, CD3OD): 6 = 2.232 (s, partially overlapped with p-anomer, C7-C#3), 3.522 (dd, 1H, 772, 3 J 2 j i = 3.7 Hz, 372,3 = 9.9 Hz), 3.996 (ddd, 1H, #5, 3J 5 > 4 = 9.9 Hz, 2J5M = 2.1 Hz, 3J5 > 6b = 5.9 Hz), 5.264 (d, 1H, #1,3Ji,2 = 3.7 Hz), 7.748 (d, overlapped with p-anomer, #11, V u , i o = 7.3 Hz). 109 3.3 Results and Discussion 3.3.1 Synthesis and Characterization of Products The 3-hydroxy-4-pyridinones (Scheme 3.1) were synthesized from maltol (1) in low to moderate yields (23-51%) analogous to those observed in other pyridinone syntheses.28'29 The pyridinones were coupled to 2,3,4,6-tetra-O-benzyl-D-glucopyranose (16) under Mitsunobu conditions30 (Scheme 3.2) and 23-26 were column purified yielding the (3-anomers (23-26) as the major products with moderate yields (47-61%). The ratio of reagents was typically 1.5 eq. for ADDP and tributylphosphine, and 1.2 eq. of the pyridinone compared to the sugar (16) as the limiting reagent. Debenzylation of 23-26 was accomplished using H2 and 10% Pd/C; yields following recrystallization ranged from 65-73%. O O R = methyl 19 R = hexyl 20 R = phenyl 21 OH R = methyl Hdpp R = hexyl Hmhpp R = phenyl Hppp maltol, 1 R R = 4-(2-butyl)phenyl Hsbp (4-(2-butyl)phenyl) Scheme 3.1. Synthesis of pyridinones. 110 For the Mitsunobu syntheses of 23-26 (Scheme 3.2), the 'H NMR spectra of the reaction mixtures were overlapping in parts of the anomeric region, making it difficult to determine the ratio of P to a-glycosides at this stage. The reaction mixtures were column purified and eluted until there were no fractions containing the pyridinone glycosides, as indicated by TLC. Early fractions from the column purifications of 23-26 appeared to have very little or no a-anomer present, according to the 'H NMR spectra, with the P-anomer as the sole, or major, component. Later tailing fractions contained smaller amounts of material but had increasing proportions of the a-anomer. In general, there were only a few instances where later tailing fractions had over 10% of the a-anomer and these were always under 15%. For 23-26, sometimes small amounts of the late tailing fractions (<100 mg) were kept separate for hydrogenation and the higher a-anomer proportions made it easier to partially characterize these anomers (27a-30a) by NMR. Major Product: P-anomer O OH R = 4-(2-butyl)phenyl 26 Scheme 3.2: Synthesis of pyridinone glycosides, a) ADDP, J3u3P, CH2C12, yields for 23-26 range from 47-61% b) 10% Pd/C, 1 atm. H 2 , MeOH/H 20, yields for 27-30 range from 65-73%. I l l After hydrogenolysis to form 27-30, the NMR spectrum of the crude product indicated larger proportions of a-anomer than in the recrystallized product. Thus recrystallization nicely aided in the purification of the P-anomers for 27 and 30 and reduced the a-anomer content to 3-4% for 28-29. The ratios of P:a anomers were measured from integration of the anomeric proton peaks in the 'H NMR spectra (Figure 3.2). The p-anomeric proton signals appear as doublets with coupling constants of -7.5 Hz. The a-anomeric protons appear as doublets with smaller coupling constants (typically less than 4 Hz) and the chemical shifts are -0.5 ppm or more downfield from the p-anomers. For instance, the P-anomeric proton signals ranged from 4.62-4.74 ppm (J = 7.4-7.6 Hz) for 2 7 P - 3 O 0 while the signals for the <x-anomeric protons ranged from 5.15-5.27 ppm with a coupling constant of-3.7 Hz for 27a-30a. Partial 'H NMR spectral characterization of the a-anomers was accomplished in the mixture with the P-anomer. The largest chemical shift differences were for HI, H2, and H5. The COSY spectra showed correlations between HI and H2, as well as others for H2 and H5 that were consistent with coupling to other glucose protons such as H3, H4, H6a, and H6b. However, these signals were at least partially overlapped with the P-anomer. making complete NMR characterization difficult. A slight upfield shift of the pyridinone CH3 peak and slight downfield shift of the downfield pyridinone ring proton signal were also observed. Higher a:P anomeric ratios and an increased number of scans allowed HMQC to aid in some a-anomer assignments. For instance, in one sample of 29, the anomeric carbon of 29a was assigned (104.9 ppm) from the HMQC correlation with the anomeric proton and a signal for the pyridinone CH3 was also identified. 112 Figure 3.2: Representative example of integration used to determine the anomer ratio for the pyridinone glycosides. ' H N M R spectrum (400 M H z ) of the anomeric region for a mixture containing 30a and 30p in CD3OD. Integration reveals a 30P:30a ratio of 92:8 for this particular sample with expected coupling constants for the anomers. 113 The Mitsunobu reaction produced mainly the [3-anomers of the pyridinone glycosides (27P-30P). This was, in part, because the starting material (16) contained mainly the a-anomer (typically 8 0 - 9 0 % a-anomer) and the subsequent SN2 displacement of the Bu3P-0-a-glucoside (Scheme 3.3) by a pyridinone produces the P-anomer of the pyridinone glycosides. The choice of ADDP instead of diethyl azodicarboxylate (DEAD) as a Mitsunobu reagent was because the former forms a more basic intermediate30 (Scheme 3.3) and would likely be more effective in deprotonation of the relatively high pK a pyridinones (pKa - 9 .4 -9 .9 ) . Tributylphosphine is used because it is more nucleophilic than triphenylphosphine, which is typically used with DEAD. OBn P-anomer Bu 3 P -0 -a-Glucoside Scheme 3.3: Proposed Mitsunobu mechanism. 114 Other glycosidic bond-forming reactions such as the Koenigs-Knorr reaction31 with silver carbonate, and BF3-etherate32 with pentaacetylglucose were attempted in this work; however very low or no yields were obtained. Koenigs-Knorr coupling with flavonols has been reported to be low yielding for the pyrone glycosylation, and thermodynamically stable boron-pyridinone complexes have been synthesized in the Orvig group,34'35 possibly explaining why BF3-etherate may not be an ideal reagent. X-ray quality crystals of one of the pyridinone glycosides (27p) were grown by slow evaporation of a MeOH solution; the structure is shown in Figure 3.3 (crystal data shown in Appendix 1). The bond lengths and angles (Table 3.1) are typical for pyridinone structures.28'36 The torsional angles indicate the near-planarity of the pyridinone ring, and the N-CH3 carbon atom (C13) lies approximately in the same plane as the pyridinone ring, typical of other pyridinone structures.28,36 The shortest hydrogen bond in this structure is between H14 and 07 (keto-oxygen) with a distance of 1.76(3) A. The P-anomer of glucose is displayed in this solid state structure, in agreement with the coupling constant and chemical shift in the *H NMR solution spectrum (see Experimental Section) for purified 27p, the latter data indicating the presence of one isomer. 115 Figure 3.3: O R T E P diagram of 27p showing 50% thermal probability ellipsoids. Table 3.1: Bond lengths (A) and angles (°) in 27p. 0(1) -C( l ) 1.414(2) 0(7)-C(9) 1.257(3) C(4)-C(5) 1.524(3) 0(1) -C(5) 1.429(2) N(l ) -C(7) 1.378(3) C(5)-C(6) 1.517(3) 0(2) -C(2) 1.419(2) N ( l ) - C ( l l ) 1.356(3) C(7)-C(8) 1.366(3) 0(3) -C(3) 1.427(2) N(l)-C(13) 1.471(3) C(7)-C(12) 1.490(3) 0(4) -C(4) 1.417(2) C(l)-C(2) 1.518(3) C(8)-C(9) 1.432(3) 0(5) -C(6) 1.409(3) C(2)-C(3) 1.519(3) C(9)-C(10) 1.421(3) 0(6) -C(l) 1.407(2) C(3)-C(4) 1.515(2) C(10)-C(l l ) 1.348(4) 0(6) -C(8) 1.388(2) C(l) -0(1) -C(5) 112.9(1) C(3)-C(4)-C(5) 109.7(2) 0(3)-C(3)-C(4) 107.6(2) C(l) -0(6) -C(8) 115.0(2) 0(1)-C(5)-C(4) 110.5(2) C(2)-C(3)-C(4) 110.4(2) C(7) -N(l ) - C ( l l ) 119.6(2) 0(1)-C(5)-C(6) 106.2(2) 0(4)-C(4)-C(3) 112.9(2) C(7) -N(l ) -C(13) 120.7(2) C(4)-C(5)-C(6) 110.9(2) 0(4)-C(4)-C(5) 106.5(2) C ( l l ) - N ( l l)-C(13) 119.7(2) 0(5)-C(6)-C(5) 112.0(2) O(7)-C(9)-C(10) 124.0(2) 0(1} -C(l) -0(6) 106.0(1) N(l)-C(7)-C(8) 118.4(2) C(8)-C(9)-C(10) 113.5(2) oo: -C(l) -C(2) 110.0(1) N(l)-C(7)-C(12) 118.6(2) C(9)-C(10)-C(l l ) 121.4(2) o(6: -C(l) -C(2) 108.7(2) C(8)-C(7)-C(12) 122.9(2) N( l ) -C( l l ) -C(10) 122.9(2) 0(2: -C(2) -C(l) 111.5(2) 0(6)-C(8)-C(7) 119.3(2) 0(2: -C(2) -C(3) 109.8(2) 0(6)-C(8)-C(9) 116.5(2) C(l) -C(2) -C(3) 109.5(2) C(7)-C(8)-C(9) 124.1(2) 0(3; >-c(3: -C(2) 111.3(2) 0(7)-C(9)-C(8) 122.5(2) 116 3.3.2 Enzymatic Cleavage of Pyridinone Glycosides The pyridinone glycosides (27P-30P) require the glucose moiety to be cleaved before metal chelation by the pyridinone moiety is enabled. There are a number of enzymes that cleave (3-glycosides in humans.37 Some of these include glucocerebrosidase (EC 3.2.1.45), cytosolic P-glucosidase (EC 3.2.1.21), and lactase phlorizin hydrolase (EC 12 38 3.2.1.62, EC 3.2.1.108). ' Other enzymes such as P-galactosidases may have some activity toward P-glycosides. For instance, cytosolic P-glucosidase has catalytic activity toward P-D-glucosides and P-D-galactosides.39 Because more than one enzyme can act on a given substrate and the human enzyme may be membrane-bound or difficult to obtain, using rat tissue homogenates is one method to determine the overall P-glucosidase activity.40'41 Using this method, brain homogenates of male Wistar rats were determined to have glycosidase activities of approximately 4-4.5 nmol/h/mg brain homogenate.40'41 In lieu of homogenized tissue, an initial screening method for glycoside substrate reactivity with other P-glucosidases may give some indication of activity in vivo. For this purpose a broad specifity P-glucosidase, Agrobacterium sp. P-glucosidase (Abg), was chosen to determine if there was any activity toward the pyridinone glycosides (27p-30p) (Scheme 3.4). TLC monitoring of the Abg reactions was used as a relatively quick method to determine if each pyridinone glycoside showed any sign of being cleaved by Abg. Once this is determined more detailed enzyme studies can be initiated. With an Abg concentration of 3.9 nM and initial substrate concentrations of 7.5 or 15 mM (27P), some cleavage of the glucose moiety was observed for each pyridinone glycoside studied. These enzymatic reactions were followed using TLC (Figure 3.4). In 117 27P-30P Abg H HO R + OH HO OH R = methyl Hdpp R = hexyl Hmhpp R = phenyl Hppp R = 4-(2-butyl)phenyl Hsbp Scheme 3.4: Abg enzyme cleavage of the pyridinone glycoside /3-anomers to produce free pyridinone and glucose. T L C was used to follow the crude reactions with Abg and the formation of the free pyridinone (see Figure 3.4). all cases, a considerable amount of cleavage was observed after -20-30 minutes at RT. The free pyridinones (highest Rf spots) turned purple with FeCh spray and did not char, unlike the pyridinone glycosides and glucose which both char. (i) Q H 27P d P P (ii) 7 j o E 28 G M O E 29 G 3 0 H P P o ° U . E 30p G H s b P E 30p G H s b Figure 3.4: Silica TLC monitoring of Abg enzyme reactions with pyridinone glycosides. E = enzyme reaction with Abg added to initiate the reaction; G = Glucose. TLCs were run after reacting -30 min. Spots that were UV-active (254 nm lamp) are outlined. The highest Rf spots correspond to the free pyridinones that turn purple-blue in colour with FeCl3 spray. Glucose is only visualized after charring (not at 254 nm). The starting pyridinone glycosides have a lower Rf than the free pyridinones and turn pale yellow in colour with FeCl3 spray. A representative acid charred TLC plate is also shown (v). 118 To obtain more detailed information about Abg and pyridinone glycoside substrates, some kinetic parameters were obtained for one of the pyridinone glycosides (27P) . This pyridinone glycoside was chosen because it was one pure diastereomer. If the Michaelis-Menten model for enzyme kinetics is observed then the enzyme (E), substrate (S), enzyme-substrate complex (ES), and product (P) appear as in equation 3.1. ki k 2 = k c a t E + S ES *• E + P 3.1 k.i Based on equation 3.1 with assumptions that [ES] is constant and the formation of product is irreversible, the Michaelis-Menten equation is written as equation 3.2. Vmax [S] Vo = , 0 Km + [S] 3 2 The initial rate is defined as v 0, V m a x is the maximum rate with V m a x = koat [E]totai at high substrate concentrations, and Km is the Michaelis constant with Km = k\l(k.\ + kcai). If Michaelis-Menten kinetics are observed, a plot of initial rates versus substrate concentrations generates a hyperbolic curve and fitting of this curve is done to determine Km and V m a x values. If the total enzyme concentration is known, then &cat can be determined from V m a x -119 Figure 3.5: Saturation curve for initial rates (AA.U./s) versus substrate concentrations (mM) for 27p and Abg in sodium phosphate buffer at 37°C. The graph was fitted and Michaelis-Menten kinetic parameters V m a x = 0.0102 ± 0.0004 AA.U./s and Km = 1.5213 ± 0.0944 mM were determined using GraFit. Substrate concentrations above 1 mM were not used in determining these kinetic parameters. The units for V m a x convert to 1.72 ± 0.07 x 10"6 M/s. Inset is a Lineweaver-Burk reciprocal plot. For Abg and 27p, the initial rates (v0) were determined for substrate concentrations ranging from 20-3000 uM. From a plot of the initial rates versus substrate concentrations (Figure 3.5), it appeared that Michaelis-Menton kinetics were being observed at substrate concentrations from 20-1000 uM. At higher substrate concentrations (above -1000 uM), there were noticeable decreases in initial rates, indicating that the substrate inhibited enzyme activity; these points were not used in 120 Table 3.2: Parameters for the glycosidic cleavage of 27p by Abg at 37°C from this work, compared to other Abg data42 for phenolic substrates with similar pK a values. Substrate pK a of aglycone kcat Cs"1) (mM) (s-'mM"1) 4-bromophenyl-P-D-glucopyranoside 9.34 28.8 0.56 52 4-chlorophenyl-P-D-glucopyranoside 9.38 29.6 0.64 46 2-naphthyl-p-D-glucopyranoside 9.51 25.3 0.16 160 27p 9.86 19.8 1.52 13 phenyl-p-D-glucopyranoside 9.99 5.44 2.12 2.6 determining kinetic parameters. Assuming no substrate inhibition was taking place at substrate concentrations from 20-1000 uM, a number of kinetic parameters were determined for 27p as shown in Table 3.2. There does not appear to be substrate inhibition of enzyme at the lower substrate concentrations for 27p (Figure 3.5). The catalytic efficiency (kcat/Km) can be determined at low substrate concentrations ( [S ] « /C m ) where substrate inhibition does not appear to be a factor. For the regions with low substrate concentrations, the initial rates can be written as in equation 3.3. _ Acat[E][S] V o v 3-3 121 Using this method (Figure 3.6), the slope was determined and with the enzyme concentration known, kaJKm was calculated to be 12.2 ± 0.8 s''mM'' which agrees well with 13.0 ± 0.8 s''mM'' determined from the hyperbolic curve fitting method, but this is not surprising in view of the small curvature from the graph. There is less likely to be enzyme inhibition at low substrate concentration so the initial slope kcat/Km determination and its similarity to the curve fitting determination likely indicate that relatively no substrate inhibition was taking place from 0-1000 uM. [270] (mM) Figure 3.6: Initial rates versus substrate concentrations for 27p and Abg in sodium phosphate buffer at 37°C. The initial slope for lower substrate concentrations was determined to be 0.0063 ± 0.0004 using GraFit.22 From the initial slope, kcJKm was calculated to be 12.2 ± 0.8 s ' W . 122 Previous work with Abg has shown that substrates with relatively poor leaving groups (p/Ca > 8) have a significant dependence of kcat on the leaving group ability.42 A value of &cat = -20 s"1 for 270 does not seem out of place compared to phenolic substrates with similar pA'a values (Table 3.2), indicating that the kcal values would likely increase slightly for pyridinones with lower p/Ca values. This would include pyridinone glycosides 29 and 30 as their aglycones (pyridinones) have pKa values of -9.4. With a high Km value and relatively low kcat and kcJKm values, this pyridinone glycoside (27p) does not appear to be the perfect substrate for Abg. However, the pyridinone glycosides may behave differently with other glucosidases in vivo to provide enough free pyridinone for therapeutic metal chelation. The slow cleavage may provide a long-lived continuous release for chronic delivery of pyridinone. 3.3.3 TEAC Values of Pyridinones It can be difficult to monitor every oxidative/antioxidative marker in a given system. The TEAC assay is a simple method to quantify the total antioxidant activity of biological fluids, extracts, or pure compounds.24 This assay was used to measure the ability of the pyridinones to quench the ABTS*+ radical cation, by monitoring the disappearance of the ABTS , + signal. Trolox is used as the antioxidant standard in this assay with its values normalized to 1. Trolox is a more water-soluble analogue of a-tocopherol, as Trolox lacks the lipophilic phytyl tail. TEAC values are plotted in Figure 123 3.7 and listed in Table 3.3. Al l of the pyridinones showed equal or enhanced values compared to those of a-tocopherol and BHT, demonstrating the radical quenching ability of these compounds. In this assay, Hhpp (see Experimental Section, Chapter 2) showed the highest activity after 1,3, and 6 min. After 3 and 6 min, all aromatic pyridinones (Hhpp, Hsbp, and Hppp) displayed higher TEAC values than those of the alkyl pyridinones (Hdpp and Hmhpp). The higher values for Hhpp may be due to its phenol substituent which is generally associated with a more protective function against oxidative stress. a-Toc BHT Hdpp Hmhpp Hppp Hsbp Hhpp Figure 3.7: TEAC values at 1, 3, and 6 minutes for typical antioxidants and various 3-hydroxy-4-pyridinones. Error bars represent ± 1 SD above and below the average TEAC value. Each TEAC value was determined in triplicate except Hhpp (sextuplicate). 124 Table 3.3: TEAC values ± SD for 1, 3, and 6 minutes. Compound 1 min 3 min 6 min a-Toc BHT HDPP HMHPP HPP HSBP HHPP 0.72 ±0.02 0.13 ±0.01 0.67 ±0.02 0.73 ± 0.02 0.82 ±0.04 0.82 ±0.03 1.10 ± 0.03 0.72 ±0.03 0.23 ±0.01 0.71 ±0.02 0.82 ±0.02 1.04 ±0.04 0.98 ± 0.03 1.28 ±0.04 0.72 ±0.03 0.34 ±0.01 0.74 ±0.02 0.83 ±0.02 1.12 ±0.04 1.05 ±0.04 1.34 ±0.04 3.3.4 Cell Studies Human breast cancer cells (MDA-MB-435S)26 were used to study the biological activity of one of the pyridinone glycosides (30P). An MTT assay was used that has been described elsewhere: the MTT assay is a colorimetric determination of cell viability during in vitro treatment,43 in this case with 30p. The assay, developed as an initial stage of drug screening, measures the amount of MTT reduction by mitochondrial dehydrogenase and assumes that cell viability (corresponding to the reductive activity) is proportional to the production of purple formazan that is measured spectrophotometrically.44 A low IC50 implies cytotoxicity or antiproliferation at low drug concentrations. The MTT plot (Figure 3.8) reveals a high IC50 value for 30P (570 uM), indicating that it is much less cytotoxic than is cisplatin (IC50 -35 uM). (It is unknown if any 30P becomes deglycosylated to produce free pyridinone (Hsbp) in this assay.) Certain iron chelators have antiproliferative properties with some tumor cell lines and the free pyridinones might be expected to be more cytotoxic in this cell line. 4 5 ' 4 6 125 0.01 0.1 1 10 C oncentration (mM) Figure 3.8: MTT plots for 30p and cisplatin, with IC50 values equal to 570 ± 90 and 35 ± 5 /xM, respectively. The error bars indicate one standard deviation of the averaged cell percent viability. Error bars for each point represent ± 1 SD above and below the average. Cell viabilities for each concentration were done in sextuplicate (6 wells each). 3.4 Conclusions Several pyridinones and pyridinone glycosides were synthesized and characterized by mass spectrometry and ! H and 1 3 C NMR spectroscopies. The pyridinones appeared to have some antioxidant properties as indicated from their radical quenching abilities in the TEAC assay. The pyridinone glycosides have some propensity to be cleaved by a broad specificity /3-glucosidase (Abg), indicating they may be useful pro-drugs for the pyridinones. Studies of one of the pyridinone glycosides (30P) with human breast cancer cells demonstrated a relatively high IC50 value of -570 uM, indicating that relatively high concentrations of 30P could be used without cytotoxicity to these cells. 126 3.5 References (1) Crapper McLachlan, D. R.; Dalton, A. J.; Kruck, T. P.; Bell, M. Y.; Smith, W. L.; Kalow, W.; Andrews, D. F. Lancet 1991, 337, 1304-1308. (2) Cherny, R. A.; Legg, J. T.; McLean, C. A.; Fairlie, D. P.; Huang, X.; Atwood, C. S.; Beyreuther, K.; Tanzi, R. E.; Masters, C. L.; Bush, A. I. J. Biol. Chem. 1999, 274, 23223-23228. (3) Regland, B.; Lehmann, W.; Abedini, I.; Blennow, K.; Jonsson, M. ; Karlsson, I.; Sjogren, M. ; Wallin, A.; Xilinas, M. ; Gottfries, C. G. Dement. Geriatr. Cogn. Disord. 2001,12, 408-414. (4) Ritchie, C. W.; Bush, A. I.; Mackinnon, A.; Macfarlane, S.; Mastwyk, M. ; MacGregor, L.; Kiers, L.; Cherny, R.; Li , Q.-X.; Tammer, A.; Carrington, D.; Mavros, C ; Volitakis, I.; Xilinas, M. ; Ames, D.; Davis, S.; Beyreuther, K.; Tanzi, R. E.; Masters, C. L. Arch. Neurol. 2003, 60, 1685-1691. (5) Squitti, R.; Lupoi, D.; Pasqualetti, P.; Dal Forno, G.; Vernieri, F.; Chiovenda, P.; Rossi, L.; Cortesi, M. ; Cassetta, E.; Rossini, P. M. Neurology 2002, 59, 1153-1161. (6) Ozcankaya, R.; Delibas, N . Croat. Med. J. 2002, 43, 28-32. (7) Lovell, M. A.; Robertson, J. D.; Teesdale, W. J.; Campbell, J. L.; Markesbery, W. R. J. Neurol. Sci. 1998,158, 47-52. (8) Anderson, O. Chem. Rev. 1999, 99, 2683-2710. (9) Tjalve, H.; Stahl, K. Acta Pharmacol. Toxicol. 1984, 55, 65-72. (10) Bebbington, D.; Dawson, C. E.; Gaur, S.; Spencer, J. Bioorg. Med. Chem. Lett. 2002,12, 3297-3300. (11) Pardridge, W. M. ; Boado, R. J.; Farrell, C. R. J. Biol. Chem. 1990, 265, 18035-18040. (12) Matern, H.; Boermans, H.; Lottspeich, F.; Matern, S. J. Biol. Chem. 2001, 276, 37929-37933. (13) Kowoghiorghes, G. J.; Neocleous, K.; Koinagou, A. Drug Saf. 2003, 26, 553-584. (14) Kline, M. A.; Orvig, C. Clin. Chem. 1992, 38, 562-565. 127 (15) Clarke, E. T.; Martell, A. E. Inorg. Chim. Acta 1992,191, 57-63. (16) Pashalidis, I.; Kontoghiorghes, G. J. Drug Res. 2001, 51, 998-1003. (17) Eljammal, A.; Howell, P. L.; Turner, M. A.; Li , N . Y.; Templeton, D. M. J. Med. Chem. 1994,37,461-466. (18) Sheppard, L. N. ; Kontoghiorghes, G. J. Drug Res. 1993, 43, 659-663. (19) Chaston, T. B.; Richardson, D. R. J. Biol. Inorg. Chem. 2003, 8, 427-438. (20) Bondy, S. C ; Tseng, H.; Orvig, C. Neurotoxicol. Teratol. 1998, 20, 317-320. (21) Hasinoff, B.; Barnabe, N . Free Rad. Biol. Med. 2001, 31, S32-S32. (22) Erithacus Software Ltd. GraFit. Version 4.0 (2001): P.O. Box 274, Horley, Surrey RH6 9YJ, UK, 2001. (23) Peratoner, A. Gazz. Chim. Ital. 1912, 41(11), 619-685. (24) Re, R.; Pellegrini, N. ; Proteggente, A.; Pannala, A.; Yang, M. ; Rice-Evans, C. Free Rad. Biol. Med. 1999, 26, 1231-1237. (25) Wu, A.; Kennedy, D. C ; Patrick, B. O.; James, B. R. Inorg. Chem. 2003, 42, 7579-7586. (26) Brinkley, B.; Beall, P.; Wible, L.; Mace, M. ; Turner, D.; Cailleau, R. Cancer Res. 1980, 40,3118-3129. (27) Mosmann, T. J. Immunol. Methods 1983, 65, 55-63. (28) Nelson, W. O.; Karpishin, T. B.; Rettig, S. J.; Orvig, C. Can. J. Chem. 1988, 66, 123-131. (29) Zhang, Z.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1991, 30, 509-515. (30) Tsunoda, T.; Yamamiya, Y.; Ito, S. Tetrahedron Lett. 1993, 34, 1639-1642. (31) Conchie, J.; Levvy, G. A. In Methods In Carbohydrate Chemistry; Whistler, R. L., Wolfrom, M. L., Eds.; Academic Press: New York, 1963; Vol. II, pp 335-337. (32) Mikata, Y.; Yoneda, K ; Tanase, T.; Kinoshita, I.; Doe, M. ; Nishida, F.; Mochida, K.; Yano, S. Carbohydr. Res. 1998, 313, 175-179. (33) Tsushida, T.; Suzuki, M . J. Jpn. Soc. Food Sci. 1995, 42, 100-108. 128 (34) Nelson, W. O.; Orvig, C ; Rettig, S. J.; Trotter, J. Can. J. Chem. 1988, 66, 132-138. (35) Orvig, C ; Rettig, S. J.; Trotter, J.; Zhang, Z. H. Can. J. Chem. 1990, 68, 1803-1807. (36) Zhang, Z. H.; Rettig, S. J.; Orvig, C. Can. J. Chem. 1992, 70, 763-770. (37) Coutinho, P. M. ; Henrissat, B. Carbohydrate-Active Enzymes server at http://afmb. cnrs-mrs. fr/CAZY/. (38) Henrissat, B.; Callebaut, I.; Fabrega, S.; Lehn, P.; Mornon, J. P.; Davies, G. Proc. Natl. Acad. Sci. U. S. A. 1995, 92, 7090-7094. (39) De Graaf, M. ; Van Veen, I. C.; Van Der Meulen-Muileman, I. H.; Gerritsen, W. R.; Pinedo, H. M. ; Haisma, H. J. Biochem. J. 2001, 356, 907-910. (40) McCarter, J. D.; Adam, M . J.; Hartman, N . G.; Withers, S. G. Biochem. J. 1994, 301, 343-348. (41) Fernandez, C.; Nieto, O.; Rivas, E.; Montenegro, G.; Fontenla, J. A.; Fernandez-Mayoralas, A. Carbohydr. Res. 2000, 327, 353-365. (42) Kempton, J. B.; Withers, S. G. Biochemistry 1992, 31, 9961-9969. (43) Alley, M. ; Scudiero, D.; Monks, A.; Hursey, M. ; Czerwinski, M. ; Fine, D.; Abbott, B.; Mayo, J.; Shoemaker, R.; Boyd, M . Cancer Res. 1988, 48, 589-601. (44) Carmichael, J.; DeGraff, W.; Gazdar, A.; Minna, J.; Mitchell, J. Cancer Res. 1987, 47, 936-942. (45) Bergeron, R. J.; McManis, J. S.; Franklin, A. M. ; Yao, H.; Weimar, W. R. J. Med. Chem. 2003, 46, 5478 -5483. (46) Chua, A. C. G.; Ingram, H. A.; Raymond, K. N. ; Baker, E. Eur. J. Biochem. 2003, 270, 1689-1698. 129 C H A P T E R 4 CONCLUSIONS AND F U T U R E W O R K 4.1 Gallium and Indium Complexes with Carbohydrate-Bearing Pyridinones Several Ga(III) and In(III) complexes with carbohydrate-containing 3-hydroxy-4-pyridinone ligands have been synthesized and characterized in this thesis work (Figure 4.1). The pyridinone portion of the molecule chelates the gallium and indium metal ion centres forming stable /ra-ligand metal complexes. The glucose molecules offer biologically relevant moieties that may be recognized by a number of proteins and enzymes in vivo, and these interactions are expected to alter the biodistribution of these metal complexes. To further evaluate these compounds, PET imaging with Ga, or SPECT imaging with 6 7 G a or 1 1 ' i n in animals would provide information on the biodistribution of these metal complexes. The radiosynthesis and radiopurification of these metal complexes should be completed before any of the compounds are injected into mammals for biodistribution studies. A generator which can produce Ga lasts about a year. If a Ga metabolic glucose imaging agent were in widespread use, this generator might become a viable 18 option and would be an improvement over the daily shipments of [ F]2-deoxy-2-fluoro-D-glucose (FDG) currently provided to satellite-PET centres and hospitals. 130 N H RO RO o. o • I O y C H , L O ^ ^ ^ R 0 O C H 3 M M = Ga, R = H M = Ga, R = Ac Ga(OG6GP)3 Ga(AG6GP)3 M = In, R = H M = In, R = Ac In(OG6GP)3 In(AG6GP)3 M = Ga, R = H G a ( O G B A P ) 3 M = Ga, R = Ac G a ( A G B A P ) 3 M = In, R = H In (OGBAP) 3 M = In, R = A c In(AGBAP) 3 Figure 4.1: Prepared in this work: gallium and indium complexes of carbohydrate-bearing 3-hydroxy-4-pyridinones. A technetium-chelate complex containing a glucose conjugate has shown some initial promise in rats as a glucose metabolic imaging agent.1 This technetium complex has a molecular weight of -591 Dalton (Da) and likely utilizes GLUT receptors for transport. The gallium and indium complexes described in Chapter 2 have molecular 131 weights that range from ~1120-1420 Da for non-acetylated glucose derivatives and 1520-1925 Da for acetylated derivatives. With an upper limit of -700 Da for passive diffusion across the BBB, these complexes are likely too large to enter the brain via this route. With their polar nature and glucose moieties, these complexes may utilize GLUT transporters in order to pass through lipid bilayers or the BBB. Perhaps other carbohydrate recognition processes will be discovered such as an endocytosis uptake mechanism that may transport these complexes. If the Ga and In complexes were transported through the GLUT receptors and entered cells, and if the P-linked glucose units remained intact long enough, the 6-position of M(OGBAP)3 and M(OGBPP)3 complexes could become phosphorylated by hexokinases3 and have the potential for metabolic glucose imaging. 4.2 Gallium and Indium Complexes Containing One Glucose If only one glucose-containing 3-hydroxy-4-pyridinone ligand were complexed to gallium or indium, the resulting complex might better mimic glucose and would have a lower molecular weight. The other two ligands could be smaller pyridinones such as Hdpp or Hmpp (the N-H pyridinone). The molecular weights for the mixed ligand complexes 6 8Ga(OG6GP)(mpp) 2, 6 8Ga(OGBPP)(mpp) 2, and 6 8Ga(OGBAP)(mpp) 2 would then be 673, 694, and 751 Da, respectively (716, 737, and 794 Da for 1 1 ' in analogues). With the smaller molecular weights, the complexes might be better transported through GLUT receptors. 132 Preliminary results are presented below that describe the synthesis of mixed ligand Ga complexes of this type (Scheme 4.1). Hdpp was used as the smaller pyridinone because the methyl group is a convenient NMR probe, and HOG6GP was chosen as the carbohydrate containing ligand because its gallium and indium tris(ligand) complexes were the most water-soluble. With the lower molecular weight differences inherent in a mixture of complexes from this synthesis (vide infra), high water-solubility would ensure that a small band could be applied to a Sephadex G10 column for purification. To synthesize this set of complexes, Ga(N0*3)3 was added to a 2:1 ratio of HOG6GP:Hdpp in water (Scheme 4.1) and the pH was adjusted to 7. HOG6GP and Hdpp have similar pK a values and would likely have similar stability constants with gallium. If identical pK a and stability constants with gallium were assumed for HOG6GP and Hdpp, then this should provide a statistical distribution of metal complexes and should elute in the following theoretical order from a size exclusion column (theoretical yields/statistical distribution shown in parenthesis): Ga(OG6GP)3 (3.7%), Ga(OG6GP) 2(dpp) (22.2%), Ga(OG6GP)(dpp)2 (44.4%), Ga(dpp)3 (29.6%). The actual yields were 21% for Ga(OG6GP)2(dpp) and 43% for Ga(OG6GP)(dpp) 2 , which matched closely the expected statistical distribution. This synthesis has the drawback of lower yields because of the mixture of gallium complexes. While the yields based on gallium may not be ideal for radiopharmaceuticals, the method does provide a relatively straightforward method to obtain mixed pyridinone complexes. 133 Scheme 4.1 There are a number of possible isomers for the metal complexes mentioned above. For tris-chelate complexes with the same unsymmetric ligands (Ga(OG6GP)3 and Ga(dpp)3) there are 2 geometrical isomers ifac, mer) and their enantiomers (A, A), totalling four possible isomers. For tris-chelate complexes where one unsymmetric ligand is not identical, then there are 4 geometric isomers, and their 4 enantiomers totalling 8 possible isomers (Figure 4.2). At least some of the 8 possible isomers appear to be present according to the 'H NMR spectra for Ga(OG6GP)(dpp) 2 (Figure 4.3) and Ga(OG6GP) 2(dpp). The C9 methyls have more than one peak, or the peaks are slightly broadened, which is typical for these metal complexes. The aromatic region contains complex multiplets which are likely from the different isomers overlapping in this region. The 'H NMR spectra (1:1 134 mirror plajie Figure 4.2: Different possible isomers for mixed ligand complexes Ga(OG6GP)(dpp) 2 and Ga(OG6GP) 2(dpp). O—O' represents the hydroxyl (O) and keto (O') oxygen atoms of the unsymmetrical 3-hydroxy-4-pyridinonato ligands with A and B representing different pyridinone ligand substituents. 135 136 D 20:CD 3OD, RT) did not change after more than 48 h for either Ga(OG6GP)2(dpp) or Ga(OG6GP)(dpp)2. The closest analogue to the mixed ligand gallium complexes that has been studied by NMR spectroscopy is Ga(dpp)3. The coalescence temperature (Tc) for Ga(dpp)3 was previously determined to be -9°C (CD3OD). 4 There are four possible isomers for Ga(dpp)3, and monitoring of the methyl groups below the T c indicated all four isomers are observed (-2 ppm, Figure 4.3 inset).4 The 'H NMR patterns of the methyl groups and pyridinone protons (at RT) for Ga(OG6GP)(dpp)2 (Figure 4.3) resemble those of Ga(dpp)3 that was cooled below its Tc. Perhaps these features become apparent because Ga(OG6GP)(dpp)2 and Ga(OG6GP)2(dpp) are less symmetric than is Ga(dpp)3. The ligands around the mixed ligand Ga complexes could exchange intramolecularly or intermolecularly. Ga(III) metal centres are generally considered to be sluggishly labile, and their water exchange rates corroborate this.5 Ligand exchange rates often correlate with aqua ligand exchange rates6 and the bidentate pyridinones may exchange at a rate slow enough that the mixed ligand Ga complexes remain stable for radiopharmaceutical applications (-24 h). This may be the case as 'H NMR spectra of the mixed ligand complexes do not change over prolonged periods. For tris(pyridinone) gallium complexes, the pyridinone protons (-6-8 ppm) and methyl group (-2 ppm) appear slightly broadened, and addition of excess L to a solution of GaL3 generally results in sharp peaks for L and unchanged slightly broadened peaks for GaL3. This may be indicative of intramolecular ligand exchange about the Ga metal centre compared to intermolecular exchange, but further studies may be needed to unequivocally determine this. 137 4.3 Experimental for Ga(OG6GP)(dpp) 2 and Ga(OG6GP)2(dpp) (l-{[((2R,3S,4S,5R,6S)-3,4,5-Trihydroxy-6-methoxy-tetrahydropyran-2-yl-methyl)carbamoyl]methyl}-2-methyl-3-oxy-4(l//)-py"dinonato)bis(l,2-dimethyl-3-oxy-4(l#)-pyridinonato)galIium(III) (Ga(OG6GP)(dpp)2). This was synthesized in a manner similar to that for Ga(OG6GP) 3 with some modifications. Gallium nitrate 6" ^ " l 3 b ' 1 2 c 1 3 c ' 2 nonahydrate (20.9 mg, 50 umol) was ^ ^ ^ , ^ „ w . „ dissolved in 1 mL water in a 4 mL vial. Ga(OG6GP)(dpp)2 This solution was added dropwise to a stirred solution containing a mixture of HOG6GP (18.8 mg, 53 umol) and Hdpp (14.6 mg, 105 umol) dissolved in 3 mL water in a 20 mL vial. The vial for the gallium nitrate was rinsed with 1 mL water and these washings were added to the 20 mL vial. The pH of the reaction mixture was 2.4. A 1% NaOH solution was added dropwise to the reaction mixture until the pH = 7.0; the solvent was subsequently removed with a rotary evaporator (50 °C, <30 min) to produce an off-white precipitate. The precipitate was redissolved in 2 mL water and applied to a Sephadex G10 size exclusion column (0.8 x 40 cm) as described for Ga(OG6GP)3. Silica TLC of the 1-2 mL fractions was conducted (1:1 AcOEtMeOH) and fractions with only one spot (UV-lamp) that charred (sulfuric acid spray/heat gun) were combined. For the above compound, fractions from the second main component that eluted from the column had the solvent removed with a rotary evaporator (50 °C, <30 min) to produce an off-white solid. The solid was dried in 138 a vacuum desiccator for 48 h before the yield was determined (15.2 mg, 43% yield based on Ga with the assumption there are no waters of hydration). 'H NMR (400 MHz, 1:1 D20:CD3OD): 5 = 2.326 and 2.330 (at least 2 overlapped singlets, 3H-combined integration, C9a-C7/3), 2.391 and 2.395 (at least 2 overlapped singlets, 6H-combined integration, C9b and C9c-C//3), 3.206 (dd, 1H, 774, 3 J 4 , 5 = 9.8 Hz, 3J4,3 = 9.8 Hz), 3.347 (s, 3H, Cl-OC// 3 ) , 3.433 (dd, 1H, 776b, 2J6hM = 13.7 Hz, 3J6b,5 = 6.5 Hz), 3.469 (dd, 1H, H2, 3 J 2 j I = 3.7 Hz, 3 J 2 , 3 = 9.8 Hz), 3.60 (dd, overlapped with H5, 7/3), 3.61 (ddd, overlapped with H3, 7/5), 3.646 (dd, partially overlapped with H3, 7/6a, 2J6a,6b = 13.7 Hz, 3J 6a ;5 = 2.4 Hz), 3.796 (s, 6H, N-CH 3), 4.703 (d, lH-partially overlapped with water peak, 7/1, 3 J i > 2 = 3.7 Hz), 4.909 (s, partially overlapped with water peak, 7/8), 6.53 (m, 2H, //12b and //12c), 6.59 (m, 1H, //12a), 7.52 (m, 3H, //13a, /713b, and //13c). , 3C{'H} NMR (75 MHz, 1:1 D 20:CD 3OD): 8 = 12.99 (C9a-CH3), 13.12 (C9b and C9c-CH3), 41.37 (C6), 43.83 (N-CH3), 55.95 (CI-OCH3), 58.09 (CS), 70.89 (C5), 72.56 (C2), 72.60 (C4), 74.11 (C3), 100.48 (Cl), 108.62 (C12b and C12c), 108.97 (C12a), 134.43, 137.24, 134.23, 134.31, 134.94, 135.01, 135.91, 137.14, 152.36, 152.40, 152.63, 152.67, 167.38, 167.43, 168.66, 168.72, 169.27. MS (+ESI): m/z 725/727 [69/71Ga(OG6GP)(dpp)2Na]+, 703/705 [69/71Ga(OG6GP)(dpp)2H]+ (with similar isotopic distribution as that calculated for C 2 9 H 3 7 N 4 Oi 2 Ga), 564/566 [69/7lGa(OG6GP)(dpp)]+, 345/347 [69/7lGa(dpp)2]+. Bis(l-{[((2R,3S,4S,5R,6S)-3,4,5-trihydroxy-6-methoxy-tetrahydropyran-2-yl-methyI)carbamoyI]methyI}-2-methyI-3-oxy-4(l//)-pyridinonato)(l,2-dimethyl-3-oxy-4(l#)-pyridinonato)galliuni(III) (Ga(OG6GP)2(dpp)). 139 This was purified from the reaction described above, eluting ahead of (Ga(OG6GP)(dpp)2) and was handled in the same manner as above. H' Yield: 9.7 mg (21% yield based on Ga with the assumption there are no waters Ga(OG6GP) 2 (dpp) of hydration). 'H NMR (400 MHz, 1:1 D 20:CD 3OD): 5 = 2.336 (s, 6H, C9a and C9b-C/73), 2.402 (singlet, 3H, C9c-C/73), 3.206 (dd, 2H, HA, 3 J 4 , 5 = 9.8 Hz, 37 4, 3 = 9.8 Hz), 3.350 (s, 6H, Cl-OC/7 3), 3.439 (dd, 2H, /76b, 2J6bM = 13.6 Hz, 3 J 6 b ; 5 = 6.6 Hz), 3.450 (dd, 2H, HI, 3J2,i = 3.7 Hz, 37 2, 3 = 9.8 Hz), 3.60 (dd, overlapped with H5, H3), 3.61 (ddd, overlapped with H3, 7/5), 3.653 (dd, partially overlapped with H3, /76a, 2J6^6b = 13.6 Hz, 3 7 6 a ; 5 = 2.3 Hz), 3.808 (s, 3H, N-CH 3), 4.708 (d, partially overlapped with water peak, HI, J\i2 = 3.7 Hz), 4.918 (s, partially overlapped with water peak, 778), 6.54 (m, IH, /712c), 6.61 (m, 2H, /712a and /712b), 7.53 (m, 3H, /713a, /713b, and /713c). 1 3C{'H} NMR (75 MHz, 1:1 D 20:CD 3OD): 5 = 12.99 (C9a and C9b-CH3), 13.14 (C9c-CH3), 41.37 (C6), 43.84 (N-CH3), 55.94 (Cl-OCH3), 58.11 (C8), 70.89 (C5), 72.57 (C2), 72.60 (C4), 74.12 (C3), 100.49 (Cl), 108.66 (C12c), 108.97 (C12a and C12b), 134.33, 134.41, 135.03, 135.11, 135.96, 137.16, 152.33, 152.37, 152.60, 152.64, 167.34, 167.39, 168.63, 168.68, 169.25. MS (+ESI): m/z 922/924 [69/71Ga(OG6GP)2(dpp)H]+, (with similar isotopic distribution as that calculated for C 3 7 H5oN 5 0,8Ga) 783/785 [6 9 / 7 1Ga(OG6GP)2]+, 564/566 [69/71Ga(OG6GP)(dpp)]+, 359 [(HOG6GP)+H]+. 140 4.4 Other Glucose-Bearing Pyridinones for Potential in Molecular Metabolic Imaging Substitution - of the fluorine atom at the 2-position in 18-FDG suggests this position may be fruitful for substitution in other glucose compounds for use in metabolic imaging.7 For the 3-hydroxy-4-pyridinones, one compound substituted with a glucose moiety has recently been published8 and is shown in Figure 4.4 ("Feralex-G" or HOG2GP using our notation). This might be an ideal ligand for gallium and indium with 6 8Ga(OG2GP)(mpp) 2 having a molecular weight of 647 Da. The p-glucose molecules in this thesis work have potential to be enzymatically cleaved by a P-glucosidase enzyme prior to being phosphorylated by hexokinases. More robust C-glycosides could be synthesized (Figure 4.4) and these should be inert to enzymatic cleavage by P-glucosidases. O HOG2GP (Feralex-G) H O G B A P H O G B P P C-glycoside C-glycoside Figure 4.4: Other glucose-pyridinone derivatives that could be synthesized with potential for glucose metabolic imaging with gallium or indium complexes. 141 For the two pyridinone-carbohydrates with the glucose moieties acetylated (HAGBPP and HAGBAP) , it was hoped that the acetyl groups would render the metal complexes lipophilic to bypass the GLUT receptors. While the acetyl function makes compounds more lipophilic and may help some sugar derivative cross the BBB, it is not the most dependable route, as compounds such as acetylated FDG are not stable in plasma.9 The compounds would have to penetrate the BBB quickly to avoid deacetylation, and then perhaps could be retained by phosphorylation. 4.5 Carbohydrate-Lectin Interactions For the gallium and indium complexes bearing three glucose molecules (Figure 4.1), there is the potential to have mono-, bi-, or trivalent carbohydrate interactions.10 Multivalent or polyvalent carbohydrate interactions with their carbohydrate-binding proteins (lectins) play pivotal roles in biology.10'11 Cell-adhesion, virus-mediated endocytosis, and binding of bacteria to cell surfaces are just some of the many processes that involve carbohydrate-lectin interactions. An increase in the number of carbohydrates on a molecule can often lead to an increased avidity for the carbohydrates that bind to these lectins,11"13 as often there are numerous lectins on each cell's surface. For instance, when -39 N-acetylgalactosamine (GalNAc) units were bound to bovine serum albumin, the resulting molecule had a 140 000-fold increased potency towards a certain parasite than did monovalent GalNAc. 1 2 In another example, 120 glucose or galactose units bound to gold nanoparticles13 increased their avidity of the latter towards an HIV-related 142 lectin by up to 300 times compared to the bivalent carbohydrate interactions. While these are large numbers of sugars, it illustrates that an increase in multivalent carbohydrate interactions can clearly increase binding avidities. Potential applications for the Ga and In complexes may be related to instances where an increased multivalency is required. For instance, if a GalNAc specific lectin is overexpressed on certain tumour cells,11 an increased number of GalNAc residues on a molecule could increase its avidity toward the tumour cells. Having one GalNAc-bearing pyridinone would mean trivalent interactions are possible, or if two GalNAc were attached to each pyridinone then hexavalent carbohydrate-lectin interactions would be possible. With a gallium or indium metal centre, tumour imaging may be possible. If some of the metal complex is internalized in the cell, radiotherapy with 1 1 ' in may be practical. There are likely to be many other examples in normal or disease states where increased multivalent carbohydrate-lectin interactions may have potential for imaging or drug therapies. With the ability to synthesize mixed pyridinone ligand complexes as described in Section 4.2 and 4.3, one avenue may open to develop a library of carbohydrates to test different carbohydrate-lectin interactions. If three different carbohydrate-bearing pyridinones (A, B, C) were added to a gallium or indium metal centre (M), the following ten metal complexes could be formed (without counting enantiomers/isomers): M(A)(B)(C), M(A)2(B), M(A)2(C), M(A) 3, M(B)2(A), M(B)2(C), M(B) 3, M(C)2(A), M(C)2(B), and M(C) 3. From the total number of different ligands (x) such as A, B, C, etc., the total number of metal complexes generated (for x > 3) was found to follow the x-3 equation: x 2 + 1 + ^]3x. For 5, 7, and 10 different pyridinones, for instance, a o 143 maximum of 34, 80, and 185 different metal complexes could be generated, respectively (without counting enantiomers/isomers). These metal complexes could be separated on analytical HPLC and characterized by LC-MS (or LC-MS-MS). Lectins of interest may already be available in the form of a lectin affinity chromatography column or other lectins can be immoblized on particles14 for column use. While this initially seems like a separation nightmare, three ligands at a time should help narrow down particular complexes of interest. It may become clear that particular sugars have better avidities for certain lectins further reducing the number of complexes. Metal complexes with the strongest interactions for the lectin should elute last and if this is a complicated mixture then other columns (size exclusion, chiral) could be added in series. This may be a method to generate some lead carbohydrate-containing metal complexes that have specific interactions with certain lectins. If a particular complex has promising properties, further testing may be warranted, and perhaps then a connected hexadentate ligand could also be synthesized (Figure 4.5). Figure 4.5: A potential hexadentate tri(carbohydrate/pyridinone) chelator. 144 The Orvig group has previously determined that two pyridinones connected between the N-atoms tends to polymerize forming (M 2 L3) n metal complexes.15 Recent evidence suggests that if three 3-hydroxy-4-pyridinones are connected at positions ortho to the 3-hydroxy or 4-keto position, then hexadentate tri(pyridinone) complexes form without polymerization.16 This may be a method to minimize the number of enantiomers/isomers and to increase the overall stability of gallium and indium pyridinone complexes even further. 4.6 Metal Chelation of Pyridinone Glycosides and Alzheimer's Disease Oxidative stress is known to play a role in Alzheimer's Disease (AD), and an increased amount of reactive oxygen species (ROS) may be produced by the excess iron and copper ions present in AD patients.17'18 Iron, copper and zinc are found in excess in beta-amyloid (Ap) plaques,19 one of the hallmarks of AD. The pyridinone glycosides (Figure 4.6) are designed to chelate excess iron and copper ions present in patients with AD, hoping that the coupling of glucose to the pyridinone results in better uptake and passage across the BBB without the potential for prematurely binding systemic metal ions. If the pyridinone glycosides reach the brain intact, the P-glucosidase activity in the brain may then deglycosylate the glucose moiety and allow iron and copper metal ions to be scavenged. Ap protein fragments are known to precipitate in vitro in the presence of iron, copper and zinc ions; chelation of these metals from AP plaques with 3-hydroxy-4-pyridinones may result in the dissolution and removal of these plaques. This therapy thus has the potential to remove Ap plaques already present which may return cognitive 145 OH HO-R = methyl 270 R = 4-(2-butyl)phenyl 30P R OH HO O O R = hexyl P:a 97:3 28 R R = phenyl p:a 96:4 29 Figure 4.6: Pyridinone glycosides from this work.function function to AD patients. Metal chelation could also halt future amyloid deposits from forming and halt the progression of the disease. Deferiprone (Ferriprox, L l , or Hdpp) is in the clinic in numerous countries, used for its ability to chelate and excrete excess iron ions in iron-overload disorders. The 3-hydroxy-4-pyridinones are strong, but only somewhat selective chelators. Their complexes with iron and copper are quite thermodynamically stable compared to those of common electrolyte ions such as sodium, potassium, calcium, and magnesium. The 3-hydroxy-4-pyridinones in this work have the same chelating moiety as Deferiprone, but with varied N-substituents. For the pyridinones synthesized in Chapter 3, tris-pyridinone chelates of Fe(III) and bis-pyridinone complexes of Cu(II) would be neutral and lipophilic. The molecular weights would range between -470-1040 Da for FeL3 and -340-720 Da for CuL 2. This may enable the smaller pyridinone-metal complexes to exit the brain by permeating the blood-brain barrier (BBB), after which they could be excreted. 146 4.7 Glycosyl Protection and Pharmacokinetics The pharmacokinetics of the glycosyl prodrug strategy need extensive investigation. Ideally the pyridinone glycosides would be administered orally. Many questions remain such as whether the glycosides remain intact after their uptake, are they able to permeate the BBB, and if they do reach the CNS will the pyridinone glycosides by cleaved to allow metal chelation? There are many glycosidic compounds found in food and beverages that are studied to determine their metabolic fate. In some cases the glycosides may be absorbed and relatively high portions remain intact in plasma. Examples of this include anthocyanin glycosides21"23 such as malvidin and cyanidin 3-glucoside (Figure 4.7). Early studies of quercetin glycosides indicated that quercetin-3-glycosides remained intact or were more robust than 7 or 4' glycosides (numbering in Figure 4.7) in the small intestine and liver.24 More recent studies indicate that all quercetin glycosides are 25 transported across the intestine by the sodium-dependent glucose transporter, SGLT1. 26 After transport, deglycosylation and/or glucuronidation may take place, and the 27 glucuronidated forms of quercetin may help with the crossing of the BBB. The pyridinone glycosides could remain intact across the BBB and then undergo deprotection to chelate iron or copper ions. If, on the other hand, the glycosides were deglycosylated prior to reaching the brain, the sole purpose of the glycosidic moiety would be to increase the water-solubility and bioavailability of the pyridinones. 147 O H 7 O H Cyanidin 3-glucoside ( R i = OFF R2 = H ) Malvidin 3-glucoside ( R j = R2 = OCH3) Quercetin 3-glucoside (R = GH2OH) Quercetin 3-glucoronide (R - CQ 2H) Figure 4.7: Anthocyanin 3-glucosides and quercetin derivatives. If there is partial or perhaps even full deglycosylation of the prodrug prior to delivery to the brain, many pyridinones are quite lipophilic and some metal-free chelator may be able to pass across the BBB to chelate metals. With a high concentration (5 mM) of pyridinone injected directly into the carotid artery entering the brain, some pyridinones were found to permeate the BBB well in rats. It is possible that these unprotected chelators could also carry systemic iron or copper to the brain. Other non-protected pyridinone chelators such as CP24 (l-rc-butyl-2-methyl-3-hydroxy-4-pyridinone) and CP94 (l,2-diethyl-3-hydroxy-4-pyridinone) removed excess iron in brain. Ferrocene-loaded rats had up to -30% of excess iron reduced in brain after 2 weeks of chelation therapy (30 mg/kg/day).29 These chelators could have removed the metals systemically or by entering the brain and chelating the excess metals before excretion. There are systemic increases of iron30 and copper31 ions in AD patients. If a portion of the 3-hydroxy-4-pyridinones remained in the systemic region, these excess 148 metal ions could be excreted, removing a potential source of redox active iron and copper ions. The level of metals in the brain and systemic region may equilibrate and result in a net removal of metal ions from an AD brain. This is believed to be how desferoxamine (DFO), a hexadentate iron chelator that does not penetrate the BBB, was able to slow the progression of dementia associated with AD patients. A recent study indicated that an unprotected pyridinone-glucose derivative ("Feralex-G" or HOG2GP) removed metals such as iron from neurofibrillary tangles (another common pathology associated with AD) in ex vivo AD brain.33 Certain pyridinones from Chapter 2 may be promising candidates for iron or copper chelation in AD if they were able to penetrate the BBB. HOG6GP and its acetylated derivative HAG6GP might be ideal candidates from this work (Figure 4.8). The reducing end of the sugar is protected (a-OCFD) which could protect it in an oxidative environment, and the 6-position is substituted and unable to be phosphorylated or easily metabolized via this position. O o o NH NH NH HOG2GP (Feralex-G) HOG6GP HAG6GP Figure 4.8: Other potential metal chelators for AD. 149 Several metabolic studies26 involved with flavonol glycosides utilized HPLC detection coupled with mass spectrometric (MS) or tandem MS (MS-MS) analysis, viable options to determine the metabolites of the pyridinone glycosides. MS analysis is an important tool for determining derivatives from the HPLC. Glucuronides and glucosides have very similar UV-Vis profiles and elute with similar retention times in acidic media; MS allows the differentiation of these compounds. 4.8 Determining the Effects of Pyridinone Glycosides on Alzheimer's Disease If the 3-hydroxy-4-pyridinones were able to alleviate oxidative stress by removal of metal ions, dissolving Ap plaques, and either halting the progression of AD or even returning some cognitive function, a transgenic animal model of AD would be useful for testing compounds in this work. One transgenic mouse model (Tg2576) overexpresses the amyloid precursor protein (APP) which results in many of the hallmarks of AD, one of which is Ap plaque formation. There are other transgenic models available, some that may better resemble AD pathologies than the Tg2576 mice.34 The Tg2576 model has been used with clioquinol studies, and after 9 weeks of treatment this bidentate chelator was shown to reduce the plaque surface area in these mice as well as to return some cognitive function.35 This mouse model would be useful for testing the 3-hydroxy-4-pyridinones in a similar manner. Instead of sacrificing the mice after 9 weeks, it would be useful to follow their progression by PET imaging of the AP plaques (vide infra -Section 4.11) at different time intervals in a small animal PET scanner (microPET). 150 It is imperative to develop screening methods for the pyridinone glycosides prior to animal testing. A number of relatively inexpensive avenues exist for this. Solubilization of Ap deposits in post-mortem AD brain tissue is one strategy.35 A different method involves the precipitation of synthetic AP1.42 or APi . 4 0 with iron and copper, or even mixtures of the metal ions, followed by addition of pyridinones to determine which are best at resolubilizing the precipitates with different concentrations of metal chelators, ' such as the 3-hydroxy-4-pyridinones. Another strategy would involve the use of different cell lines for testing metal chelators and antioxidant potential. For a cell line that had oxidative stress induced (e.g., by hydrogen peroxide), the antioxidant capacity of the pyridinones could be tested to determine cell viability.37 Cells that overproduce metal ions could also be used to test the metal chelating abilities of the pyridinones in cells, and how this affects cell viability.38 The biologically relevant tests above should give indications of which compounds would be most useful in a metal chelator treatment of AD. 4.9 Radioactive Analogues of the Pyridinones To complement the HPLC-MS metabolic studies and to better evaluate the pyridinone glycosides in vivo, a radioactive analogue would be useful for biodistribution studies. Labeling of the pyridinone ring with 1 4 C (ty2 = 5700 y) is one method to obtain labelled analogues. A labeled maltol, 14C-ethyl maltol (an Amersham product), is available with the radiolabel on the ethyl CH2 substituent, and a similar synthesis could provide the label on the CH3 of maltol. Custom radiolabeled syntheses described above 151 are typically expensive. A labeled maltol would also require further syntheses to obtain analogues of 27p-30p and there would be storage, handling, and disposal drawbacks associated with a long-lived isotope such as 1 4 C. For these reasons, shorter-lived PET radioisotopes may provide alternative labeling strategies. PET tracers may be able to provide some initial biodistribution results.39 To have essentially the same compound, one of the stable carbon or nitrogen atoms could be substituted with n C (t/2 = 20 min) or 1 3 N (t/2 = 10 min), respectively. This could provide routes to some of the simple pyridinone glycosides shown in Scheme 4.2. H Scheme 4.2 Smaller alkyl amines (and ammonia) tend to condense more readily with protected maltol derivatives and should provide enough material in a timely fashion to make the radiosyntheses in Scheme 4.2 viable. Under ideal conditions, 1 'C has maximum observation times of 100-120 min40 and this would be approximately halved for 1 3 N. The 152 maximum observation time would be shortened because of synthesis, but some biodistribution results should still be obtainable. With the short half-lives of U C and , 3 N , the radiosyntheses and time required to obtain images may mean a longer-lived PET isotope would be useful for determining 18 extended biodistribution of compounds. Longer-lived PET isotopes such as F (t>/2 =1.8 h) or 1 2 4 I (t'/2 = 4.2 d) could also be incorporated into pyridinone molecules. One potential method for radionuclide incorporation is outlined in Scheme 4.3. Scheme 4.3 The key reaction for the incorporation would involve the radical bromination of the 2-CH3 of the pyridinone ring (bottom of Scheme 4.3), to CH 2Br, and subsequent halide displacement with 1 8F or 1 2 4I (P+, t>/2 = 4.2 d). Other tested radical reactions41 found a 42% yield for the bromination step (top of Scheme 4.3) utilizing l,3-dibromo-5,5-dimethylhydantoin (DBDMH) and azoisobutyronitrile (AIBN) as a radical initiator. 153 Incorporation of one bromine atom is predominantly expected for the pyridinone glycoside 29p (at the C2-methyl); however, more than one bromine atom could be incorporated into 27p (2 methyl groups), 28p (2-methyl and hexyl moiety), and 30p (2-methyl and sec-butyl substituent). This may provide more than one brominated compound for these analogues that may require some purification before labeling. These new analogues would obviously have structures different from that of the parent compounds and may themselves have beneficial biodistribution and/or chelation properties. 4.10 Increasing the Antioxidant Potential of Chelators The pyridinone moiety appears to impart some antioxidant activity (high Tocopherol Equivalent Antioxidant Capacity, or TEAC, values) that would help combat the excess ROS produced in an AD brain. The 3-hydroxy-4-pyridinones quenched radicals (ABTS*+) as well or better than common antioxidants such as a-tocopherol and BHT (Chapter 3). The phenolic pyridinone (Hhpp) had the highest TEAC values of all pyridinones tested, likely because of the added antioxidant protection that the phenol moiety offers. Similar antioxidant effects have been observed for pyridinones with a pendant BHT moiety.42 The results with phenolic derivatives merit the inclusion of other phenolic antioxidants with pyridinones to increase their antioxidant capacity. Some potential compounds of interest include those described in Figure 4.9. The syntheses of a number of phenolic-pyridinone compounds are being undertaken in the Orvig group. These have electron-donating methyl, methoxy and t-butyl groups ortho to the phenol, which increase their antioxidant behaviour. 154 R = C H 3 R = O C H 3 R = /-butyl Figure 4.9: Phenolic pyridinone compounds with enhanced antioxidant capacity. The glycosylation, and other synthetic steps, do require some effort and certain steps are moderate to low yielding (i.e., pyridinone formation with aromatic amines). A promising approach might be to take advantage of commercially available compounds that have metal chelating abilities with antioxidant potential already incorporated into the structure. Numerous flavonols are available which have metal chelation and antioxidant potential.43'44 Quercetin and its derivatives (some shown in Figure 4.10) are examples of one type of flavonol. Derivatives of quercetin include glucosides, galactosides, xylosides, rhamnosides, arabinoglucosides, and rutinosides, usually found on the 3, 4', and 7 positions of quercetin. Many other flavonols are commercially available (Figure 4.11) and there are compounds with glycosidic substitutions similar to the quercetin derivatives. The flavonol aglycones are generally not water-soluble; glycoside derivatives would increase their water-solubility and bioavailability. 155 Quercetin-3-glucoside H Q H O Quercetin-3,4'-diglucoside Quercetin-3-rutinoside Quercetin-4'-glucoside Figure 4.10: Potentially applicable quercetin derivatives. 1 5 6 Flavonol Myricetin Myricetin trimethy 1 ether Tamarixetin Robinetin Morin Figure 4.11: Commercially available flavonols and their common names. 157 Exemplifying their antioxidant potential, high TEAC values are typically obtained for most flavonols. TEAC values for five flavonols that have been studied: quercetin (4.7 ± 0.1) > myricetin (3.1 ± 0.3) > morin (2.55 ± 0.02) > quercetin-3-rutinoside (2.4 ± 0.06) > kaempferol (1.34 ± 0.08).43 Most of these TEAC values are higher than those of the phenolic pyridinone Hhpp (-1.3 between 3 and 6 min) and the other 3-hydroxy-4-pyridinones tested in Chapter 3. While the total number of ring hydroxyl groups on the flavonols appears to be important for antioxidant potential, other key structural antioxidant features on the flavonols become apparent.43 One important feature is the 3',4'-dihydroxyphenyl (catechol) moiety on the B-ring. A trihydroxy moiety was found to have approximately the same antioxidant potential on this ring and meta-OH groups provided lower TEAC values. Important features for the C-ring include a C3-OH and the C2-C3 double bond that allows for conjugation between the B and C rings. Antioxidant potential for the flavonols is believed to increase the most if a para-OW arrangement (5-OH and 8-OH) is available on the A-ring which allows a more stable quinone to form.43 A 5-OH on the A-ring also increases derealization, which helps to stabilize radicals. With these structural features in mind, the most promising candidates for antioxidant activity would appear to be gossypetin, quercetagetin, quercetin, myricetin, and their methoxy or glycosyl derivatives. 158 The above-mentioned flavonols and their derivatives would be worthy of testing in an AD mouse model. Several recent studies have reported the antioxidant properties of flavonols when tested with oxidatively stressed cell lines. Some of these studies have used flavonoid extracts37'45'46 that may contain certain flavonols. Others have used specific flavonols such as kaempferol47 and quercetin48;49 to illustrate the protective effects of these flavonols; however, there do not appear to be any studies of the flavonols in the transgenic mice typically used in Alzheimer-type studies. This could test the antioxidant potential of the compounds, as well as how they may chelate metals and/or break apart Ap plaques. The flavonols have the capacity to reduce reactive oxygen species (ROS) but are also known to bind metals such as copper and iron,50 however, higher stability metal complexes with these metal ions would likely form with 3-hydroxy-4-quinolinones51 (Figure 4.12), sometimes referred to as aza-flavones. These compounds and their methoxy/glycosyl derivatives may have antioxidant properties similar to those of the. flavonols52 and the increased affinity for iron and copper could aid in chelating these metals involved in AD and dissolution of Ap plaques to improve cognitive function. HO' OH OH OH Quercetin analogue Myricetin analogue Kaempferol analogue Figure 4.12: 3-Hydroxy-4-quinolinone analogues of certain flavonols. 51 159 4.11 Targeting p-Amyloid The pyridinone glycosides will hopefully deliver a pyridinone to the brain, and once there, the unmasked 3-hydroxy-4-pyridinones could then chelate excess iron and copper to alleviate ongoing oxidative damage in AD brain. p-Amyloid (AP) plaques are found in the extracellular space and the chelators should interact with the plaques simply because there are excess metals ions present and the pyridinones are good scavengers for these. A stronger interaction between the pyridinone chelators and metal-containing Ap plaques may be required for the pyridinones to specifically target the Ap peptide fragments. For instance, chelation combined with intercalation of the Ap peptide could increase the pyridinones' potential to break apart AP and bind copper or iron ions. Typically dyes used for staining Ap plaques are considered to have high affinities for these peptides. The most common dyes include Congo Red, Chrysamine G, Thioflavin S (a mixture of sulphonated analogues of thioflavin T) and Thioflavin T (Figure 4.13). Sulphonated or charged dyes typically do not penetrate the BBB well so more lipophilic analogues are typically considered for binding Ap plaques in vivo. Recently emerged are other compounds (BTA-1, IMPY, and FDDNP in Figure 4.13) that target AP plaques for use in PET imaging.53 Nicotine (Figure 4.13) also appears to bind AP plaques and break them apart in vitro.54 160 BTA-1 FDDNP Figure 4.13: Compounds known to bind to A(5 plaques: BTA-1, IMPY, and FDDNP are potential PET analogues that bind to AP plaques. A common structural feature of a number of these compounds is the connection of two rings by a single bond. A study of Congo Red binding to AP1-40 revealed that torsional mobility between the two rings is important for binding this peptide.55 In the case of Congo Red, the torsional angle changes from -25° to near planar (torsional angle of 0°) upon binding to AP1.40. The torsional mobility is thought to allow the rings to manoeuvre through Ap plaques, become close to planar, and insert into the p-sheets of the peptide.55 161 The N-aryl-3-hydroxy-4-pyridinones (Figure 4.14) share this single bonded, two ring motif and have some torsional mobility allowing some propensity for binding to AP peptides. The solid state structures of aryl pyridinones reveal wide ranging torsional angles of -66-89° between the pyridinone and phenyl rings.56"58 The 'H and 1 3 C NMR spectra of the N-phenyl pyridinones in this work showed no signs of broadened peaks in solution, indicative of unhindered rotation of the two rings at room temperature, and a fair degree of torsional mobility at 37°C is expected. Another important structural feature appears to be the N-alkyl or N-dialkyl amine substituent. The addition of the methylamine or dimethylamine moiety to the aromatic ring of the pyridinones could increase their affinity for Ap plaques. Some potential compounds that add these additional features to the structure of the pyridinones are shown in Figure 4.14. Figure 4.14: Potential 3-hydroxy-4-pyridinone derivatives with dimethylamine and/or a connected ring system to target Ap peptides, as well as iron and copper ions. 162 These additions to the structure of the pyridinones might impart added affinity for Ap, which in turn may aid in metal chelation, dissolution of Ap plaques, and the return of cognitive function in patients with Alzheimer's Disease. 4.12 References (1) Yang, D. J.; Kim, C. G.; Schechter, N. R ; Azhdarinia, A.; Yu, D. F.; Oh, C. S.; Bryant, J. L.; Won, J. J.; Kim, E. E.; Podoloff, D. A. Radiology 2003, 226, 465-473. (2) Oldendorf, W. H. Exp. Eye Res. 1977, 25, 177-190. (3) Aleshin, A. E.; Zeng, C. B.; Bartunik, H. D.; Fromm, H. J.; Honzatko, R. B. J. Mol. Biol. 1998, 282, 345-357. (4) Nelson, W. O.; Ph.D. Thesis, Dept. of Chemistry, University of British Columbia: Vancouver, Canada, 1988. (5) De Vito, D.; Weber, J.; Merbach, A. E. Inorg. Chem. 2004, 43, 858-864. (6) Saito, K.; Nagasawa, A. Polyhedron 1990, 9, 215-222. (7) Chenault, H. K ; Mandes, R. F.; Hornberger, K. R. J. Org. Chem. 1997, 62, 331-336. (8) Kruck, T. P. A.; Burrow, T. E. J. Inorg. Biochem. 2002, 88, 19-24. (9) Xu, Y. M. ; Choi, S. R.; Kung, M. P.; Kung, H. F. Nucl. Med. Biol. 1999, 26, 833-839. (10) Gabius, H. J.; Andre, S.; Kaltner, H.; Siebert, H. C. Biochim. Biophys. Acta 2002, 1572, 165-177. (11) Mammen, M. ; Choi, S. K.; Whitesides, G. M. Angew. Chem. Int. Ed. 1998, 37, 2755-2794. (12) Adler, P.; Wood, S. J.; Lee, Y. C ; Lee, R. T.; Petri, W. A.; Schnaar, R. L. J. Biol. Chem. 1995,270, 5164-5171. 163 (13) Nolting, B.; Yu, J. J.; Liu, G. Y.; Cho, S. J.; Kauzlarich, S.; Gervay-Hague, J. Langmuir 2003,19, 6465-6473. (14) Li , X.; Andrews, D.; Regnier, F. J. Proteome Res. 2003, 2, 618-625. (15) Nelson, W. O.; Karpishin, T. B.; Rettig, S. J.; Orvig, C. Can. J. Chem. 1988, 66, 123-131. (16) Liu, Z. D.; Hider, R. C. Coord. Chem. Rev. 2002, 232, 151-171. (17) Perry, G.; Taddeo, M . A.; Petersen, R. B.; Castellani, R. J.; Harris, P. L. R.; Siedlak, S. L.; Cash, A. D.; Liu, Q.; Nunomura, A.; Atwood, C. S.; Smith, M. A. BioMetals 2003, 16, 77-81. (18) Christen, Y. Am. J. Clin. Nutr. 2000, 71, 621S-629S. (19) Lovell, M . A.; Robertson, J. D.; Teesdale, W. J.; Campbell, J. L.; Markesbery, W. R. J. Neurol. Sci. 1998,158, 47-52. (20) Kowoghiorghes, G. J.; Neocleous, K.; Koinagou, A. Drug Saf. 2003, 26, 553-584. (21) Passamonti, S.; Vrhovsek, U.; Vanzo, A.; Mattivi, F. FEBSLett. 2003, 544, 210-213. (22) Cao, G.; Muccitelli, H. U.; Sanchez-Moreno, C ; Prior, R. L. Am. J. Clin. Nutr. 2001, 73, 920-926. (23) Frank, T.; Netzel, M. ; Strass, G.; Bitsch, R.; Bitsch, I. Can. J. Physiol. Pharmacol. 2003, 81, 423-435. (24) Day, A. J.; DuPont, M. S.; Ridley, S.; Rhodes, M. ; Rhodes, M . J. C ; Morgan, M. R. A.; Williamson, G. FEBS Lett. 1998, 436, 71-75. (25) Wolffram, S.; Block, M. ; Ader, P. J. Nutr. 2002,132, 630-635. (26) Day, A. J.; Williamson, G. Br. J. Nutr. 2001, 86, S105-S110. (27) Juergenliemk, G.; Boje, K.; Huewel, S.; Lohmann, C ; Galla, H. J.; Nahrstedt, A. PlantaMed. 2003, 69, 1013-1017. (28) Habgood, M. D.; Liu, Z. D.; Dehkordi, L. S.; Khodr, H. H.; Abbott, J.; Hider, R. C. Biochem. Pharmacol 1999,57, 1305-1310. (29) Aouad, F.; Florence, A.; Zhang, Y.; Collins, F.; Henry, C ; Ward, R. J.; Crichton, R. R. Inorg. Chim. Acta 2002, 339, 470-480. 164 (30) Ozcankaya, R.; Delibas, N. Croat. Med. J. 2002, 43, 28-32. (31) Squitti, R.; Lupoi, D.; Pasqualetti, P.; Dal Forno, G.; Vernieri, F.; Chiovenda, P.; Rossi, L.; Cortesi, M. ; Cassetta, E.; Rossini, P. M. Neurology 2002, 59, 1153-1161. (32) Crapper McLachlan, D. R.; Dalton, A. J.; Kruck, T. P.; Bell, M . Y.; Smith, W. L.; Kalow, W.; Andrews, D. F. Lancet 1991, 337, 1304-1308. (33) Shin, R. W.; Kruck, T. P. A.; Murayama, H.; Kitamoto, T. Brain Res. 2003, 961, 139-146. (34) Robinson, S. R.; Bishop, G. M . Neurobiol. AginglOOl, 23, 1051-1072. (35) Cherny, R. A.; Atwood, C. S.; Xilinas, M. E.; Gray, D. N. ; Jones, W. D.; McLean, C. A.; Barnham, K. J.; Volitakis, I.; Fraser, F. W.; Kim, Y. S.; Huang, X. D.; Goldstein, L. E.; Moir, R. D.; Lim, J. T.; Beyreuther, K.; Zheng, H.; Tanzi, R. E.; Masters, C. L.; Bush, A. I. Neuron 2001, 30, 665-676. (36) Cherny, R. A.; Legg, J. T.; McLean, C. A.; Fairlie, D. P.; Huang, X.; Atwood, C. S.; Beyreuther, K ; Tanzi, R. E.; Masters, C. L.; Bush, A. I. J. Biol. Chem. 1999, 274, 23223-23228. (37) Horakova, L.; Licht, A.; Sandig, G.; Jakstadt, M. ; Durackova, Z.; Grune, T. Arch. Toxicol. 2003, 77, 22-29. (38) Fukuyama, R.; Nakayama, A.; Nakase, T.; Toba, H.; Mukainaka, T.; Sakaguchi, H.; Saiwaki, T.; Wada, M. ; Sakurai, H.; Fushiki, S. J. Biol. Chem. 2002, 277, 41455-41462. (39) Gatley, S. J.; Carroll, F. I. Drug Dev. Res. 2003, 59, 191-193. (40) Tewson, T. J. Drug Dev. Res. 2003, 59, 261-268. (41) Briggs, J. C ; Haines, A. H.; Taylor, R. J. K. J. Chem. Soc. Perkin Trans. 1995, 27-32. (42) Bebbington, D.; Monck, N . J. T.; Gaur, S.; Palmer, A. M. ; Benwell, K ; Harvey, V.; Malcolm, C. S.; Porter, R. H. P. J. Med. Chem. 2000, 43, 2779-2782. (43) RiceEvans, C. A.; Miller, N. J.; Paganga, G. Free Rad. Biol. Med. 1996, 20, 933-956. (44) Cos, P.; Ying, L.; Calomme, M. ; Hu, J. P.; Cimanga, K.; Van Poel, B.; Pieters, L.; Vlietinck, A. J.; Vanden Berghe, D. J. Nat. Prod. 1998, 61, 71-76. 165 (45) Ono, K.; Yoshiike, Y.; Takashima, A.; Hasegawa, K.; Naiki, H.; Yamada, M. J. Neurochem. 2003, 87, 172-181. (46) Bastianetto, S.; Quirion, R. Cell. Mol. Biol. 2002, 48, 693-697. (47) Wang, C. N. ; Chi, C. W.; Lin, Y. L.; Chen, C. F.; Shiao, Y. J. J. Biol. Chem. 2001, 276, 5287-5295. (48) Patil, C. S.; Singh, V. P.; Satyanarayan, P. S. V.; Jain, N . K.; Singh, A.; Kulkarni, S. K. Pharmacology 2003, 69, 59-67. (49) Bastianetto, S.; Zheng, W. H.; Quirion, R. Brit. J. Pharmacol. 2000,131, 711-720. (50) Jungbluth, G.; Ruhling, I.; Ternes, W. J. Chem. Soc, Perkin Trans. 2 2000, 1946-1952. (51) Sui, Z. H.; Nguyen, V. N. ; Altom, J.; Fernandez, J.; Hilliard, J. J.; Bernstein, J. I.; Barrett, J. F.; Ohemeng, K. A. Eur. J. Med. Chem. 1999, 34, 381-387. (52) Gao, F.; Johnson, K. F.; Schlenoff, J. B. J. Chem. Soc, Perkin Trans. 2 1996, 269-273. (53) Zhuang, Z. P.; Kung, M . P.; Wilson, A.; Lee, C. W.; Plossl, K.; Hou, C ; Holtzman, D. M.; Kung, H. F. J. Med. Chem. 2003, 46, 237-243. (54) Ono, K.; Hasegawa, K.; Yamada, M. ; Naiki, H. Biol. Psychiat. 2002, 52, 880-886. (55) Miura, T.; Yamamiya, C ; Sasaki, M. ; Suzuki, K.; Takeuchi, H. J. Raman Spectrosc. 2002, 33, 530-535. (56) Burgess, J.; Fawcett, J.; Russell, D. R.; Zaisheng, L. Acta Cryst. 1998, C54, 430-433. (57) Kojicprodic, B.; Ruzictoros, Z.; Horvatic, D. Acta Cryst. 1989, C45, 126-129. (58) Zhang, Z. H.; Rettig, S. J.; Orvig, C. Can. J. Chem. 1992, 70,163-710. 166 APPENDIX 1 X-ray Crystallographic Data for 8 and 27p Table A l : Selected Empirical Formula Formula Weight Crystal Dimensions Crystal System Lattice Parameters Space Group Z Calculated Density Wavelength T *1(F 0) crystallographic data for 8. C22H28N2O8 448.47 0.35X0.10X0.10 mm Orthorhombic a = 6.4378(3) A b= 11.9203(6) A c = 28.586(1) A V = 2193.7(2) A 3 P2i2,2, (#19) 4 1.358 g/cm3 0.71069 A 1.04 cm"1 173(1)K 0.057 0.075 Table A2: Torsion angles for 8. atom atom atom atom angle atom atom atom atom angle 0(1)C(1)0(6) C(8) 107.6(2) 0(1) C(l) C(2) 0(2) 179.6(2 0(1)C(1)C(2) C(3) 57.8(2) 0(1) C(5) C(4) 0(4) 177.8(2 0(1) C(5) C(4) C(3) 55.3(2) 0(1) C(5) C(6) 0(5) 59.5(3 0(2) C(2) C(l) 0(6) 64.7(2) 0(2) C(2) C(3) 0(3) 63.2(2 0(2) C(2) C(3) C(4) 177.4(2) 0(3) C(3) C(2) C(l) 174.1(2 0(3) C(3) C(4) 0(4) 66.3(2) 0(3) C(3) C(4) C(5) 175.0(2 0(4) C(4) C(3) C(2) 172.1(2) 0(4) C(4) C(5) C(6) 64.7(2 0(5) C(6) C(5) C(4) 179.6(2) 0(6) C(1)0(1) C(5) 179.3(1 0(6) C(1)C(2) C(3) 173.5(1) 0(6) C(8) C(7) N( l ) 178.2(2 0(6) C(8) C(7) C(12) 1.1(4) 0(6) C(8) C(9) 0(7) 3.0(3 0(6) C(8) C(9) C(10) 177.8(2) 0(7) C(9) C(8) C(7) 179.8(3 0(7) C(9) C(10)C(11) 178.4(3) N(l) C(7) C(8) C(9) 1.0(4 N(1)C(11)C(10) C(9) 1.7(4) C(l) 0(1) C(5) C(4) 60.7(2 C(1)0(1)C(5) C(6) 178.9(2) C(l) 0(6) C(8) C(7) 100.9(2 C(l)0(6) C(8) C(9) 81.7(2) C(l) C(2) C(3) C(4) 54.7(2 C(2) C(1)0(1)C(5) 61.9(2) C(2) C(l) 0(6) C(8) 134.2(2 C(2) C(3) C(4) C(5) 53.4(2) C(3) C(4) C(5) C(6) 172.8(2 C(7) N(1)C(11)C(10) 1.2(4) C(7) C(8) C(9) C(10) 0.5(4 C(8) C(7) N(1)C(11) 0.1(4) C(8) C(7) N(l) C(13) 179.3(3 C(8) C(9) C(10) C ( l l ) 0.9(4) C(9) C(8) C(7) C(12) 178.3(3 C(10) C(11)N(1)C(13) 177.9(3) C ( l l )N(1)C(7) C(12) 179.2(3 C(12) C(7) N(1)C(13) 0.0(4) 167 Table A3: Selected crystallographic data for 270. Empirical Formula C 1 3 H 1 9 N O 7 Formula Weight 301.30 Crystal Dimensions 0.50X0.25 X0.10 mm Crystal System Monoclinic Lattice Parameters a = 8.2508(8) A b = 8.0153(6) A c= 11.633(1) A P = 113.797(4)° V = 703.9(1) A 3 Space Group P2,(#4) Z 2 Calculated Density 1.421 g/cm3 Wavelength 0.71069 A 1.2 cm"1 T 253(1)K tfl(Fo) 0.073 wR2(F02) 0.099 Table A4: Torsion angles for 27p atom atom atom atom angle 0(1) C(l) 0(6) C(8) -107.6(2) 0(1) C(l) C(2) C(3) 57.8(2) 0(1) C(5) C(4) C(3) -55.3(2) 0(2) C(2) C(l) 0(6) -64.7(2) 0(2) C(2) C(3) C(4) -177.4(2) 0(3) C(3) C(4) 0(4) -66.3(2) 0(4) C(4) C(3) C(2) 172.1(2) 0(5) C(6) C(5) C(4) 179.6(2) 0(6) C(l) C(2) C(3) 173.5(1) 0(6) C(8) C(7) C(12) -1.1(4) O(6)C(8)C(9)C(10) -177.8(2) O(7)C(9)C(10)C(ll) 178.4(3) N(1)C(11)C(10)C(9) 1.7(4) C(l) 0(1) C(5) C(6) -178.9(2) C(l) 0(6) C(8) C(9) -81.7(2) C(2)C(1)0(1)C(5) -61.9(2) C(2) C(3) C(4) C(5) 53.4(2) C(7)N(1)C(11)C(10) -1.2(4) C(8)C(7)N(1)C(11) -0.1(4) C(8)C(9)C(10)C(11) -0.9(4) C(10) C(11)N(1)C(13) 177.9(3) C(12)C(7)N(1)C(13) 0.0(4) atom atom atom atom angle 0(1) C(l) C(2) 0(2) 179.6(2) 0(1) C(5) C(4) 0(4) -177.8(2) 0(1) C(5) C(6) 0(5) 59.5(3) 0(2) C(2) C(3) 0(3) 63.2(2) 0(3)C(3)C(2)C(1) -174.1(2) 0(3) C(3) C(4) C(5) 175.0(2) 0(4) C(4) C(5) C(6) 64.7(2) 0(6) C(l) 0(1) C(5) -179.3(1) 0(6) C(8) C(7) N(l) 178.2(2) 0(6) C(8) C(9) 0(7) 3.0(3) 0(7) C(9) C(8) C(7) -179.8(3) N(l) C(7) C(8) C(9) 1.0(4) C(l) 0(1) C(5) C(4) 60.7(2) C(l) 0(6) C(8) C(7) 100.9(2) C(l) C(2) C(3) C(4) -54.7(2) C(2) C(l) 0(6) C(8) 134.2(2) C(3) C(4) C(5) C(6) -172.8(2) C(7)C(8)C(9)C(10) -0.5(4) C(8)C(7)N(1)C(13) -179.3(3) C(9) C(8) C(7) C(12) -178.3(3) C(11)N(1)C(7)C(12) 179.2(3) 168 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0061177/manifest

Comment

Related Items